ETRICS - American Journal of Obstetrics & Gynecology

January 1996 in two parts, part 2 volume 174, number 1

¯

ETRICS AND GYNECOLOGY

Copyright © 1996 by Mosby-Year Book, Inc.

SOCIETY OF PERINATAL OBSTETRICIANS

1996 16TH ANNUAL MEETING Postgraduate course

(jointly sponsored by The American College of Obstetricians and Gynecologists)

Scientific, Clinical, and Business Meeting

February 4-10, 1996

Kamuela, Hawaii

Published by

MOSBY-YEAR BOOK, INC.

St. Louis, MO 63146-3318 ISSN 0002-9378

American Journal of OBSTETRICS AND GYNECOLOGY .Copyright -~ 1996 by Mosby-Year Book, Inc.

16th Annual Meeting

of the

Society of Perinatal Obstetricians

February 4-10, 1996

Kamuela, Hawaii

In accordance with the wishes of the majority of the SPO membership, we ask that our

members and guests refrain from smoking in the meeting rooms. This conforms to the spirit of our society and the written procedures of our parent organization.

Table of Contents

Organization .............................................................................. iii

Letter from the Program Chair ............................................... iv

Program Committee ................................................................. v

List of Reviewers ........................................................................ vi

Acknowledgments ..................................................................... vii

Map of Meeting Areas--Hilton Waikoloa Village .................. viii

Program ..................................................................................... ix

Special Interest Group Meetings ............................................. xiii

Scientific Sessions Program ...................................................... xiv

Oral Session I (Plenary) ........................................................... 301

Concurrent Oral Sessions Session A ............................................................................. 305

Session B ............................................................................. 309

Oral Session II (Plenary) .......................................................... 313

Concurrent Oral Sessions Session C ............................................................................ 317

Session D ............................................................................ 321

Concurrent Oral Sessions Session E ............................................................................. 325

Session F ............................................................................. 329

Poster Session I ......................................................................... 333

Poster Session II ....................................................................... 365

Poster Session III ...................................................................... 397

Poster Session IV ...................................................................... 429

Poster Session V .............................................. [ ......................... 461

Subject Index ............................................................................. 493

Author Index ............................................................................. 501

Academic Institution Index ...................................................... 519

ii January 1996 American Journal of Obstetrics and Gynecology

16th Annual Clinical, Scientific, & Business Meeting

of the

Society of Perinatal Obstetricians February 4-10, 1996

Officers Board of Directors

President: Gary D.V. Hankins

Vice President/President Elect: Donald R. Coustan

Secretary-Treasurer: Mary E. D’Alton

Assistant Secretary-Treasurer: J. Peter VanDorsten

Karin J. Blakemore W. Patrick Duff Roberto Romero Lawrence D. Devoe

John P. Elliott James N. Martin, Jr. Su~san M. Cox Timothy R.B. Johnson E. Albert Reece

’96 ’96 ’96 ’97 ’97 ’97 ’98 ’98 ’98

Past Presidents

William N. Spellacy Roy M. Pitkin James A. O’Leary Donald M. Shedine Loren P. Peterson Bruce A. Work Robert H. Hayashi Roy N. Petrie John C. Morrison Amelia C. Cruz Steven G. Gabbe Robert J. Sokol Richard H. Paul Frank C. Miller Thomas J. Garite Garland D. Anderson Larry C. Gilstrap Valerie M. Parisi

’77 ’78 ’79 ’80 ’81 ’82 ’83 ’84 ’85 ’86 ’87 ’88 ’89 ’90 ’91 ’92 ’93 ’94

American Journal of Obstetrics and Gynecology January 1996 iii

Society of Perinatal Obstetricians 409 12th STREET, SW

WASHINGTON, DC 20024 (202) 863-2476

Dear Colleagues:

This was another record year for abstracts -- 1058 were submitted. The quality of the abstracts was superb and we feel confident that several presentations will change the way most of us practice maternal-fetal medicine. There are also a number of fascinating basic science papers shedding additional light on several important pathophysiologic conditions.

Seventy-six papers were accepted for oral presentation, 600 for poster presentation, and 382 were not acceptable. A number of outstanding abstracts were submitted by fellows. Although one entire plenary session is devoted to fellow abstracts, there were more judged to be suitable for oral presentation than could be contained within this session; these have been distributed among the other oral sessions.

This year has also seen a significant increase in the number of abstracts submitted by colleagues from countries outside of North America. Some of these individuals are serving in the capacity of moderator or judges. Clearly, the SPO has grown to become the preeminent International Perinatal meeting in addition to its national prominence.

Our choice of site for this year’s meeting reflected the poll taken two years ago of SPO membership. We are confident that the travel and logistic difficulties presented by a Hawaii site are more than outweighed by the obvious advantages. Nevertheless, the Program Committee encourages your input as we consider Hawaii for a potential 2004 SPO meeting site.

We would like to thank not only those have submitted abstracts, but a number of individuals who have tirelessly worked to put this program together. Foremost among them is my assistant, Karylyn Bliss, who has devoted a good portion of her professional life this Fall to putting the meeting together. We would also like to extend a special thank you to Linda Oriti from Wilford Hall Medical Center in San Antonio, Texas, for writing a remarkable computer program to assist in setting the program and recognize her for her support in the abstract process. We would also like to thank Carolyn DeJesus, assistant to Dr. Mary D’Alton for her efforts with the abstract process. In addition, special thanks go to Jerm Woodruff, Karrie Allred, Kara Manwaring, Suzanne McAllister, and Kathy Field whose efforts are very much appreciated.

We acknowledge Drs. Mary D’Alton, Jim Martin, John Elliott, and Greigh Hirata who participated on the Program Committee and are confident that from a scientific and social standpoint, this will be a meeting to remember.

Sincerely,

~RK, M.D.

Program Chair, 1996 SPO Founded 1977

Education ¯ Service ° Research

iv January 1996 American Journal of Obstetrics and Gynecology

1996 PROGRAM COMMITTEE

Steven L. Clark, MD Program Chair

Mary E. D’Alton, MD Poster Chair

James N. Martin, Jr., MD Postgraduate Course Chair

Donald R. Coustan, MD Fundraising Chair

John P. Elliott, MD Coordinator, Special Interest Groups

Greigh Hiram, MD Local Arrangements Chair

American Joumal of Obstetrics and Gynecology January 1996 v

Reviewers for 1996 SPO Abstracts The Program Chair, on behalf of the Society, is most grateful to the followzng people who so consczentiously and promptly judged the abstracts for this meeting.

Regular Reviewers Manuel Alvarez, MD

Garland D. Anderson, MD

Robert L. Andres, MD

Joseph J. Apuzzio, MD

David A. Baker, MD

Jeffrey M. Barrett, MD

William H. Barth, Jr., MD

Laxmi V. Baxi, MD

Thomas J. Benedetti, MD

Richard L. Berkowitz, MD

Michael D. Berkus, MD

Jorge D. Blanco, MD

Frank H. Boehm, MD

RonaldJ. Bolognese, MD

Mlan T. Bombard, MD

Sidney F. Bottoms, MD

D. Ware Branch, MD

Paul C. Browne, MD

Cynthia G. Brumfield, MD

Eleanor Capeless, MD

Steve N. Caritis, MD

RobertJ. Carpenter, Jr., MD

Lony Christine Castro, MD

Robert C. Cefalo, MD

Curtis Cetrulo, MD

Frank Chervenak, MD

Nancy C. Chescheir, MD

Ronald A. Chez, MD

Joshua A. Copel, MD

Larry Cousins, MD

Donald R. Coustan, MD

Robert K. Creasy, MD

William R. Crombleholme, MD

Dwight P. Cruikshank, MD

F. Gary Cunningham, MD

Luis B. Curet, MD

Mary E. D’Alton, MD

Bonnie J. Dattel, MD

Richard O. Davis, MD

Richard Depp, MD

Lawrence D. Devoe, MD

LeroyJ. Dierker, MD

Gary A. Dildy, MD

Michael Divon, MD

Mitchell P. Dombrowski, MD

Sharon L. Dooley, MD

Donald Dudley, MD

W. Patrick Duff, MD

Gary Eglinton, MD

John Elliott, MD

Nancy L. Eriksen, MD

Hossam E. Fadel, MD

James E. Ferguson, II, MD

Harold E. Fox, MD

Harvey Gabert, MD

Stanley A. Gall, MD

Norman F. Gant, MD

Ronald S. Gibbs, MD

James D. Goldberg, MD

Robert L. Goldenberg, MD

John W. Goldkrand, MD

Phillip J. Goldstein, MD

Michael F. Greene, MD

John C. Hauth, MD

Robert H. Hayashi, MD

William N.P. Herbert, MD

L. Wayne Hess, MD

Washington C0 Hill, MD

Calvin J. Hobel, MD

R. Harold Holbrook, Jr., MD

Robert L. Jacobson, MD

John W.C. Johnson, MD

Timothy R.B. Johnson, Jr., MD

Helen H. Kay, MD

Mlen P. Killam, MD

G. Eric Knox, MD

Robert A. Knuppel, MD

Neil tC Kochenour, MD

Bruce Kovacs, MD

Russell Laros, Jr, MD

Kenneth Leveno, MD

Michael I~ Lindsay, MD

Michael J. Lucas, MD

Barbara Luke, MD

David A. Luthy, MD

Lauren Lynch, MD

Bill Clinton Mabie, MD

Elliott K. Main, MD

James Martin, Jr., MD

Arthur S. Maslow, MD

Brian Mason, MD

Paul J. Meis, MD Michael T. Mennuti, MD

Brian M. Mercer, MD

Joseph Miller, Jr, MD

Kenneth J. Moise, Jr., MD

John C. Morrison, MD

Eberhard Mueller-Heubach, MD

Stephen Myers, DO

Michael P. Nageotte, MD

Edward R. Newton, MD

Jennifer R. Niebyl, MD

Thomas Nolan, MD

William F. O’Brien, MD

Julian T. Parer, MD

Valerie Parisi, MD

Richard H. Paul, MD

Man M. Peaceman, MD

Kenneth G. Perry, Jr., MD

Jeffrey P. Phelan, MD

Elliot H. Philipson, MD

Lawrence D. Platt, MD

Richard P. Porreco, MD

T. Flint Porter, MD

J. Gerald Quirk, Jr., MD

Susan M. Ramin, MD

William F. Rayburn, MD

John A. Read, MD

Kathryn L. Reed, MD

Robert Resnik, MD

Roberto Romero, MD

Michael G. Ross, MD

David A. Sacks, MD

AndrewJ. Satin, MD

Harold Schulman, MD

Lee P. Shulman, MD

Baha M. Sibai, MD

Robert Silver, MD

Robert Sokol, MD

William N. Spellacy, MD

Joseph A. Spinnato, MD

Ronald Stoddard, MD

James E. Sumners, MD

Shiraz Sunderji, MD

Nergesh A. Tejani, MD

Gary R. Thurnau, MD

Paul Tomich, MD

Guillermo J. Valenzuela, MD

Michael W. Varner, MD

Gael P. Wager, MD

Steven L. Warsof, MD

Joseph R. Wax, MD

Paul Weinbaum, MD

Carl Weiner, MD

Stuart Weiner, MD

George Wendel, Jr., MD

Katharine Wenstrom, MD

Isabelle Wilkins, MD

John Williams, III, MD

Edward Wolf, MD

James Woods, MD

John Yeast, MD

Sze-ya Yeh, MD

Margaret Lynn Yonekura, MD

Senior Reviewers:

Karin Blakemore, MD

Steven L. Clark, MD

David Cotton, MD

Steven Gabbe, MD

Thomas Garite, MD

Larry Gilstrap, III, MD

Gary D. V. Hankins, MD

Durlin Hickok, MD

Robert P. Lorenz, MD

Denise Main, MD

Michael L. Socol, MD

vi January 1996 American Journal of Obstetrics and Gynecology

ACKNOWLEDGMENTS

The Society of Perinatal Obstetricians extends sincere thanks and appreciation

to the following organizations for their generous support. As per the SPO’s

Board of Directors policy that corporate contributions no longer go to directly

support the Annual Meeting, the funds donated by these organizations will be

used to support such SPO activities as the Foundation Fellowship Award and

the Fellows Retreat.

PATRON OF THE SPO Contributions of $10, 000 or more

A C USON

COROMETRICS MEDICAL SYSTEMS

TOKOS MEDICAL CORPORATION

I BENEFACTOROFTHESPO I Contributions of $5,000-9,999

ADVANCED TECHNOLOG Y LABORATORIES (ATL)

HEALTHD YNE

INTEGRATED GENETICS

NELLCOR

ON-TIME MEDICAL C ORPORATION

R OSS PROD UCTS DIVISION - ABBOTT LABORATORIES

WYETH-AYERST

FRIEND OF THE SPO [ Contributions of $2,500-4, 999

ADEZA BIOMEDICAL CORPORATION

ALFIGEN - THE GENETICS INSTITUTE

GENETRIX

ORTHO PHARMACE~ICAL CORPORATION

This list reflects only those contributors who have confirmed their level of support as of press deadline.

An updated Acknowledgments list will be available at the meeting.

American Journal of Obstetrics and Gynecology January 1996 vii

Kohala Ballroom

Kohala 4

Kohala 3

Kohala 2

Kohala i

HILTON WAIKOLOA VILLAGE Function and Meeting Space

I I I I

Water’s Edge Ballroo~m~

I I Grand Promenade

Lagoon Lanai

!Kona 2 I Kona 3

"~koloa Suites

Staircase

Lagoon

SOCIETY OF PERINATAL OBSTETRICIANS 16th Annual MeetingmFebruary 4-10, 1996 Hilton Waikoloa Village, Kamuela, Hawaii

Progr a m Sunday, February 4 6:00pm -8:00pm Registration Grand Promenade

Monday, February 5 6:00am-6:00pm

6:00am-6:00pm

6:00am-7:00am

7:00am-2:00pm

7:00am-2:00pm

8:50am-9:15am

l 1:00am-noon

3:00pm-6:00pm

Registration

Speaker Ready Room

Continental Breakfast

Postgraduate Course I "Perinatal Infectious Diseases Update"

Postgraduate Course II "Medical-Surgical Complications of Preffna n cy "

Coffee Break

Lunch

Board of Directors

MeefingI

Grand Promenade

Waikoloa Suite 2

Lagoon Lanai

Kohala Ballroom

Kona Ballroom

Grand Promenade

Lagoon Lanai

Boardroom

Tuesday, February 6 6:00am-6:00pm

6:00am-6:00pm

6:00am-7:00am

7:00am-2:00pm

7:00am-2:00pm

8:50am-9:15am

11:00am-noon

11:00am-2:00pm

2:00pm-6:00pm

Registration

Speaker Ready Room


Postgraduate Course III "Critical Care Obstetrics"

Postgraduate Course IV "Invasive Approaches to

Petal Diagnosis and

Therapy"

Coffee Break

Lunch

Board of Directors Meeting II

Special Interest Group Meetings

Grand Promenade

Waikoloa Suite 2

Lagoon Lanai

Kohala Ballroom

Kona Ballroom

Grand Promenade

Lagoon Lanai

Boardroom

(see page xiii for locations)

8:00am-6:00pm

8:00am-6:00pm

9:00am-noon

Wednesday, February 7 Registration

Speaker Ready Room

International Society of Perinatal Obstetricians

Special Interest Group

Grand Promenade

Waikoloa Suite 2

Queen’s 4

American Journal of Obstetrics and Gynecology January 1996 ix

8:00am-lO:OOam

Noon-4:30pm

l:00pm-7:00pm

5:00pm-6:00pm

7:00pm-9:00pm

Thursday, February 8 6:45am-5:00pm

6:45am-7:00pm

6:45am-7:30am

7:30am-7:45am

7:45am-10:00am

10:00am-noon

Noon-l:00pm

Noon-1:00pm

1:00pm-3:30pm

1:00pm-3:30pm

3:30pm-5:30pm

5:30pm-6:30pm

5:30pm-6:30pm

Friday, February 9 7:00am-5:00pm

Legislative/Practice

Issues for the Subspecialty Chair:

J. Peter VanDorsten, MD

Postgraduate Course V "Communications Skills

for the Pe~4natolog~st: An

Interactive Workshop"

Poster Session Set-Up

Program Directors Meeting

Opening Reception

Registration

Speaker Ready Room


Welcome &

Announcements Steven L. Clark, MD 1996 Program Chair

Welcome Gary D.V. Hankins, MD

President, SPO

Oral Plenary Session I Moderator:

Timothy R.B. Johnson, Jr., MD

Poster Session I & Coffee

Women in Perinatology Special Interest Group

Box Lunch Meeting

Lunch: Cash Buffet

Oral Concurrent Session A Prematurity Moderator:

John P. Elliott, MD

Oral Concurrent Session B Genetics and Prenatal Diagnosis Moderator:

Karin j. Blakemore, MD

Poster Session II & Coffee

Annual SPO Business Meeting (Regular Members Only)

Associate Members Meeting (Fellow-in-Training Members and other Associate Members

of SPO)

Registration

Kona 5

Konal-4

Kohala Ballroom

Kona5

Grand Ballroom

Grand Promenade

Waikoloa Suite 2

Lagoon Lanai

Grand Ballroom

Grand Ballroom

Kohala Ballroom

Water’s Edge

Lagoon Lanai

Grand Ballroom

Kona Ballroom

Kohala Ballroom

King’s Grand

Qneen’s Grand

Grand Promenade

x January 1996 . American Journal of Obstetrics and Gynecology

7:00am-7:OOpm

7:00am-8:OOam

7:00am-8 :OOam

7:00am-8:OOam

8:00am-10:OOam

lO:OOam-noon

Noon-1:00pro

1:00pm-3:30pm

1 :OOpm-5:3Opm

3:30pm-5:30pm

6:30pm-7:30pm

7:30pm-lO:OOpm

10:OOpm-11 :OOpm

Saturday, February 10 7:00am-noon

7:00am-1:00pro

7:00am-8:00am

8:00am-10:30am

8:00ran-10:30am

10:30am-12:30pm

12:30pro

Speaker Ready Room

Vendor greakthst

International Society of Perinatal Obstetricians

Special Interest Group Breakfast Business Meeting

Continental Break[:ast

Oral Plenary Session II (Fellows’ Plenar), Session) Moderator:

E. Albert Reece, MD

Poster Session & Coffee

Lunch: Cash Buffet

Oral Concurrent Session C Obstetric Risk Assessment

Moderator:

Lars Magnus Westgren, MD

Oral Concurrent Session D Maternal-Fetal and

Placental Physiology Moderator:

James E. Ferguson, III, MD

Poster Session 1V & Coffee

Cocktail Reception

Annual Banquet &

Awards Presentation

After-Dinner Drinks (cash bay)

&

Hawaiian Trio

Registration

Speaker Ready Room


Concurrent Oral Session E Clinical Obstetrics

Moderator:

Michael W. Varner, MD

Concurrent Oral Session F Medical Complications Moderator:

David B. Cotton, MD

Poster Session V & Coffee

Presentation of

Saturday Oral and Poster Awards

Waikoloa Suite 2

Boardroom

Kona 1-3

Lagoon Lanai

Grand Ballroom

Kohala Ballroom

Lagoon Lanai

Grand Ballroom

Kona Ballroom

Kohala Ballroom

Kona Ballroom

Grand Ballroom

Water’s Edge

Grand Promenade

Waikoloa Suite 2

Lagoon Lanai

Grand Ballroom

Kona Ballroom

Kohala Ballroom

Kohala Ballroom

ADJOURN

American Joumal of Obstetrics and Gynecology January 1996 xi

ACCME Accreditation

The American College of Obstetricians and Gynecologists is accredited by the Accreditation Council for Continuing

Medical Education (ACCME) to sponsor continuing medical education for physicians.

Continuing Medical Education Credits Postgraduate Conrses I ( Perinatal Infechous Diseases Update), 1I (Medical-Surgical Comphcations of Preg~tancy), 1II ( C’~Jticz~l

Care Obstetrics), and IV ([nvasive Approaches to Fetal Diagnosis and Therapy) have been approved for 6 cognate hours and 6 Category 1 hours. Postgraduate Course V (C0mmunicah0ns Shills for the Perinatolog4st) has been approved for 4

cognate hours and 4 Category 1 hours. All the courses meet the criteria of the ACOG Program for Continuing Professional Development and the Physician’s Recognition Award of the AMA.

Disclosure of Faculty and Industry Relationships

In accordance with ACOG policy, all faculty members have signed a conflict of interest statement in which they

have disclosed any significant financial interests or other relationships with industry relative to topics they will discuss at this program. At the beginning of the program, faculty members are expected to disclose any such information to participants. Such disclosure allows you to evaluate better the objectivity of the information

presented in lectures. Please report on your evaluation forrn any undisclosed conflict of interest you perceive. Thank you.

xii January 1996 American Journal of Obstetrics and Gynecology

SPECIAL INTEREST GROUP MEETINGS TUESDAY, FEBRUARY 6, 1996

COMMUNITY BASED PERINATOLOGISTS

COMPUTER USAGE IN PERINATAL MEDICINE

CRITICAL CARE IN PERINATAL MEDICINE

DIABETES MELLITUS IN PREGNANCY

GENETICS

HYPERTENSION IN PREGNANCY

INFECTIOUS DISEASES IN PERINATAL MEDICINE

PRETERM LABOR

RESEARCH SUPPORT PERSONNEL

ULTRASOUND IN PERINATAL MEDICINE

4 - 6 prn

2 - 5 pm

2 - 5 pm

2 - 5 pm

2 - 6 pm

2 - 4:30 pm

2 - 5:15 pm

2-4pm

2 - 4 pm

2 - 4 pm

Room

Kona 4

King’s 2

King’s 1

King’s 3

Queen’s 6

Kohala Ballroom

Queen’s 4

Monarchy

King’s 5

Queen’s 5

INTERNATIONAL SOCIETY

OF PERINATAL OBSTETRICIANS

(Scientific Presentations9

WEDNESDAY, FEBRUARY 7, 1996

9 - Noon Queen’s 4

WOMEN IN PERINATOLOGY (Box Lunch Meeting)

THURSDAY, FEBRUARY 8, 1996

Noon - 1 pm Water’s Edge

INTERNATIONAL SOCIETY

OF PERINATAL OBSTETRICIANS

(Breakfast Business Meeting)

FRIDAY, FEBRUARY 9, 1996

7 - 8 am Kona 1-3

American Joumal of Obstetrics and Gynecology January 1996 xiii

SOCIETY OF PERINATAL OBSTETRICIANS Kamuela, HI - February 4-10, 1996

Schedule of Oral Presentations

Thursday, February 8th

7:30 a.m. to 10:00 a.m. Oral Plenary Session I Moderator: Timothy R.B. Johnson, Jr., MD

Grand Ballroom

7:30-7:45 HrELCOME AND ANNOUNCEMENTS

7:45-8:00 SPO Foundation Fellowship Award

Winner for Aeademic Year 1994-95

REL&TIONSHIP BETWEEN PREECLAMPSIA, HYPOXIA AND PRODUCTION OF NITRIC

OXIDE BY THE PLACENTA

Sehgman SP, Kadner SS,

l~Tnlay TH, NYU Medical Center

8:00-8:15 1 DIETARYVITAMIN E PROPHYIA2KIS AND Sivan E, Wu YK, Homko C,

DIABETIC EMBRYOPATHY: MORPHOLOGICAL, Reece EA, Temple University BIOCHEMICAL, AND MOLECULAR ANALYSES

8:15-8:30 2 THE PRETERM PREDICTION STUDY: FETAL

FIBRONECTIN, BACTERIAL VAGINOSIS AND PERIPARTUM INFECTION

Goldenberg R, Iams J, Mercer B,

Meis P, Moawad A, Copper R,

Das A, Thorn E, Johnson F Roberts j, McNellis D, NICHD MFMU Network

8:30-8:45

8:45-9:00

9:00-9:15

9:15-9:30

FETAL FIBRONECTIN AS A PREDICTOR OF PRETERM BIRTH IN SYMPTOMATIC PATIENTS A MULTICENTER TRIAL

INTERNATIONAL TERMPROM TRIAL: A RCT OF INDUCTION OF LABOR FOR PRELABOR RUPTURE OF MEMBRANES AT TERM

AMNIOTIC FLUID EMBOLISM AND OXYGEN TRANSPORT PATTERNS

A MULTICENTER RANDOMIZED MASKED TRIAL OF ANTIBIOTIC VS. PLACEBO THERAPY AFTER PRETERM PREMATURE

RUPTURE OF THE MEMBRANES

Peaceman AM, Andrews Wig,,

Thorp JM, Cliver SP, Lukes A,

Fetal Fibronectin Study Group, Northwestern University, University of Alabama,

University of North Carolina

Hannah M, Ohlsson A, Fa,ine D,

Hewson & Hodnett ~ Myhr T,

Wang E, Weston J, Willan A, University of Toronto

Harvey C, Hankins GDV, Clark SL,

University of Texas, Galveston, University of Utah

Mercer B, Miodovnik M, Thurnau G, Goldenberg R, Das A,

Me~nstein G, Ramsey R,

Rabello Y, Thorn E, Roberts J, McNdlisD, NICHD MFMU Network

9:30-9:45 7 L-ARGININE REVERSES THE ADVERSE PREGNANCY CHANGES INDUCED BY NITRIC OXIDE SYNTHASE INHIBITION IN THE RAT

Helmbrecht GD, Farhat MY,

Yadgarova KT, Lochbaum L,

Brown HE, Eglinton G& Ramwell PW,, Georgetown Univ.

xiv January 1996 American Journal of Obstetrics and Gynecology

9:45-10:00 MANAGEMENT OF SYMPTOMATIC

PLACENTA PREVIA: A RANDOMIZED

CONTROLLED TRIAL OF IN-PATIENT VERSES OUT-PATIENT EXPECTANT

MANAGEMENT

Wing DA, Paul RH, Millar LK,

University of Southern California, Los Angeles


l:00 p.m. to 3:30 p.m. Oral Concurrent Session A

Prematurity

Moderator: John P. Elliott, MD

Grand Ballroom

1:00-1:15 9 FETAL FIBRONECTIN AND PRETERM

DELIVERY Cox & Little B, DaxJ, Leveno K, UT Southwestern, Dallas

1:15-1:30 10 A PROSPECTIVE RANDOMIZED CLINICAL TRIAL OF ANTIBIOTIC THERAPY FOR

PRETERM PREMATURE RUPTURE OF MEMBRANES

Lovett & Weiss j, Diogo M,

Williams P, Garite T, East Bay Perinatal Medical Associates,

Oakland, CA

1:30-1:45 11 PRETERM PREMATURE RUPTURE OF MEMBRANES VERSUS EXPECTANT MANAGEMENT

How H, Cook C, Cook V,

SpinnatoJ, University of Louisville, Marshall University

1:45-2:00 12 TOCOLYFIC EFFICACYOF NIFEDIPINE VERSUS RITODRINE: RESULTS OF A RANDOMIZED TRIAL

Papatsonis DNM, van Geijn HP,

Bleker OP, Lange FM, Ader HJ,

Dekker GA, Free University Hospital Amsterdam, University

of Amsterdam, Zuiderzee Hospital

Lelystad

2:00-2:15 13 INTERLEUKIN-6 (IL-6) CONCENTRATIONS IN CERVICAL SECRETIONS IDENTIFY INTRA-AMNIOTIC INFECTION IN PATIENTS

WITH PRETERM LABOR (PTL)

Rizzo G, Capponi A, Rinaldo D,

Arduini D, 7)deschi D, Romanini C, University of Rome

2:15-2:30 14 ELEVATED MIDTRIMESTER AMNIOTIC FLUID

TUMOR NECROSIS ALPHA LEVELS: A PREDICTOR OF PRETERM DELIVERY

Ghidini A, Eglinton GS, Spong G&

Jenkins CB, Pezzullo JC, Ossandon M, MilIJF, Georgetown

University

2:30-2:45 15 COMPARISON OF INTRAVENOUS MAGNESIUM Clavin DK, Bayhi DA, Nolan TE,

SULFATE AND NITROGLYCERIN FOR Rigby FB, Cork RC, MillerJM,

PRETERM LABOR: PRELIMINARY DATA LSUMC, New Orleans

2:45-3:00 16 DOES INDOMETHACIN PRETREATMENT OF MYOMETRIAL TISSUE ALTER THE TOCOLYFIC EFFECT OF ATOSIBAN?

McNamara H, Badaj R,

Hatangadi S, Nathanwls P, SmulianJ, Knuppel tL UMDNJ-

Robert Wood Johnson Medical School, Cornell University

3:00-3:15 17 PATHWAYS TO PREMATURITY Goldenberg RL, Iams fl), Mercer BM, Meis PJ, Moawad AH, Copper RL, Das A, Thorn E,

Johnson F, McNellis D, Robe, s J, NICHD MFMU Network

American Joumal of Obstetrics and Gynecology January 1996 xv

3:15-3:30 18 DETERMINING THE OPTIMAL GESTATIONAL Macones GA, Bader TJ, Asch DA,

AGE LIMIT FOR TOCOLYSIS: A DECISION University of Pennsylvania, ANALYSIS Jefferson Medical Center


1:00 p.m. to 3:30 p.m. Oral Concurrent Session B

Genetics and Prenatal Diagnosis Moderator: Karin J. glakemore, MD

Kona Ballroom

1:00-1:15 19 THE ANTIOXIDANT LIPOIC ACID-PREVENTS MALFORMATIONS IN OFFSPRING OF DIABETIC RATS

Wiznitz~ A, Hershkovitz R,

Mimon E, Mazor M, Leiberman JR,

Bashan N, Reece EA, Ben-Gurion University, Israel, Temple

University

1:15-1:30 20 LIMB REDUCTION DEFECTS (LRDs) ARE NOT INCREASED FOLLO~AqNG FIRST

TRIMESTER CHORIONIC VILLUS SAMPLING

(CVS)

Wapner R, Jackson L, Evans MI,

Johnson MP, Jefferson Medical College, Hutzel Hospital

1:30-1:45 21 THE USE OF SECOND TRIMESTER "GENETIC SONOGRAM" IN GUIDING CLINICAL MANAGEMENT OF PATIENTS AT INCREASED

RISK FOR FETAL TRISOMY 21

Vintzileos AM, Campbell WA,

Rodis fl~, Guzman ER, Smulian JC,

McLean DA, UMDNJ-Robert Wood

Johnson Medical School, University of Connecticut

1:45-2:00 22 GENETIC AMNIOCENTESIS MAY BE REASONABLY AVOIDED IN WOMEN WITH ABNORMAL SERUM SCREENING FOR

ANEUPLOIDY BUT NORMAL ULTRASOUND

Bahado-Singh R, Tan A, Deren O, Hunter D, CopelJ, Mahoney J,

Yale University, Norwalk Hospital

2:00-2:15 23 ENDOSCOPIC TRACHEAL PLUGGING USING AN INFLATABLE BALLOON IN THE FETAL LAMB

Deprest JA, Evrard VA,

Van Ballaer PP, Verbcken EA, Vandenbergtze K, Brosens IA,

Van Assche FA, Lerut T, Centre

for Surgical Technologies, Univ. Hospitals "Gasthuisberg’, Belgium

2:15-2:30 24 INTRA-AMNIOTIC PRESSURE REDUCTION IN Garu D, Lysikiewicz A, MaysJ, TWIN-TO-TWIN TRANSFUSION SYNDROME TejaniJ, New York Medical College

2:30-2:45

2:45-3:00

25 ELEVATED AMNIOTIC FLUID INTERLEUKIN-6 LEVELS AT GENETIC AMNIOCENTESIS

PREDICT SUBSEQUENT PREGNANCY LOSS

Wenstrom KD, Andrews W~,

Tamura T, DeBard M, Johnston KE, Hemstreet GP, University of Alabama, Birmingham

26 ALTERED EXPRESSION OF PLA, GENE Reece EA, Wu Y-K, Air-Allah A,

IMPLICATED IN MOLECULAR MECHANISM OF Salameh W,, Temple University DIABETES-INDUCED NEURAL TUBE DEFECTS

(NTDs) : A NEW REVELATION

xvi January 1996 American Journal of Obstetrics and Gynecology

3:00-3:15 27 ENGRAFTMENT FOLLOWING IN UTERO BONE Blahemore K, Bambach B,

MARROW TRANSPLANTATION FOR GLOBOID Moser H, Corson V, Griffin C, CELL LEUKODYSTROPHY Noga 5; Perlman E, Wenger D,

Zucherman R, Khouzami A,

Jones R, .Johns Hopkins University, Kennedy Krieger Institute, Thomas Jefferson University

3:15-3:30 28 FETAL BONE MARROW ORGAN CULTURE Ga~ne! ~, O’Donnell CO, Ulrich C,

Crombleholme T, New England Medical Center, Tufts University

Friday, February 9th

8:00 a.m. to 10:00 a.m. Oral PlenaD’ Session 11

(Fellows Plenary Session) Moderator: E. Albert Reece, MD

Grand Ballroom

8:00-8:15 29 THE EFFECT OF HIGH DIET~MRY n-3 FATTY ACID SUPPLEMENTATION ON ANGIOTENSIN

II PRESSOR RESPONSE IN HUMAN PREGNANCY

Adair CD, Sanchez-Ramos L,

Briones DL, University of Florida,

Jacksonville

8:15-8:30 30

8:30-8:45 31

8:45-9:00 32

ORAL TERBUTALINE AFTER PARENTERAL

TOCOLYSIS: A RANDOMIZED, DOUBLE-BLIND, PLACEBO-CONTROLLED TRIAL

BETAMETHASONE DELAYS ENDOTOXIN- INDUCED PRETERM LABOR IN THE MURINE MODEL

REGIONAL PRETERM BIRTH SUBSETS: IDENTIFICATION AND CLINICAL

IMPLICATIONS

Lewis R, Mercer B, Salama M, Walsh M, Sibai B, University of Tennessee, Memphis

Schwartz W, Christensen H, Carey J, Gonzalez C, Rayburn W,,

University of Oklahoma

Porter TF, Clark SL, Dzldy GA,

Toohe-Miller C, Minton SP,

Schlappy D, James B, University of Utah

9:00-9:15 33 CHOROID PLEXUS CYSTS AND TRISOMY 18: Gratton RJ, Hogge WA, Aston CE,

RISK MODIFICATION BASED ON MATERNAL University of Pittsburgh AGE AND MULTIPLE MARKER SCREENING

9:15-9:30 34

9:30-9:45 35

THE CLINICAL EFFICACY OF ORAL TOCOLYI’IC THERAPY

LACK OF APPARENT CORTICOSTEROID BENEFIT IN <1000 G INFANTS BORN AFTER PRETERM AMNION RUPTURE

Rust OA, BofilIJA, Andrew M, Arfiola R, Mm~ison JC, University

of Mississippi, Jackson

Chapman S, Hauth JC; Goldenberg RL, Owen J,

Bottoms SF, McNellis D,

MacPherson C, Thorn E, University

of Alabama, Birmingham, NICHD MFMU Network

9:45-10:00 36 PROGNOSTIC FACTORS FOR THE

DEVELOPMENT OF FETAL CEPHALHEMATOMA WITH VACUUM

EXTRACTION

Bof!llJA, Rust OA, Devidas M,

Pe~ry Jr KG, Morrison JC, MartinJrJN, University of

Mississippi, Jackson

American Journal of Obstetrics and Gynecology January 1996 xvii


1:00 p.m. to 3:30 p.m. Oral Concurrent Session C

Obstetric Risk Assessment

Moderator: Lars Magnus Westgren, MD

Grand Ballroom

1:00-1:15 37 THREE DIMENSIONAL ULTRASOUND Ludomirski A, Khandelzoal M, EVALUATION OF FETAL FACIAL AND SPINAL Uopairojhit B, Reece EA, ANATOMY Chan L, Temple University

1:15-1:30 38 IS INTRAPARTUM FETAL BRIAN INJURY IN THE TERM FETUS PREVENTABLE?

PhdanJP, Ahn MO, Korst L, Martin GI, Pomona Valley Hospital Medical Center, CA

1:30-1:45 39 IN PREMATURE RUPTURE OF MEMBRANES AND PRETERM LABOR NEONATAL NUCLEATED ERYTHROCY-FE NUMBER

(nRBCs) IS REI~kTED TO HISTOLOGIC ACUTE INFLAMIVL~TION AND NOT TO PLACENTAL M,~RKERS OF HYPOXIA

Salafia CM, Minior VK,

Pezzullo JC, Ghidini A,

Ernst LM, Sherer DM, Georgetown University, UCONN Med. Ctr,

Farmington, CT

1:45-2:00 40 NUCLEATED RED BLOOD CELLS: AN UPDATE Korst LM, Ahn MO, PhelanJP, ON THE MARKER FOR FETAL ASPHYXIA Pomona Valley Hospital Medical

Center, CA

2:00-2:15 41 FETAL BIOPHYSICAL SCORE AND CEREBRAL ManningF. Harman C, PALSY AT AGE 3 YEARS Menticoglou S, University of

Manitoba

2:15-2:30 42 PREECLAMPSIA: IS INDUCTION OF LABOR MORE SUCCESSFUL?

Xenakis EMJ, Piper j, Field N,

Conway D, University of Texas, San Antonio

2:30-2:45

2:45-3:00

3:00-3:15

43 RISK OF PERINATAL TRANSMISSION OF HUMAN PAPILLOMA’VIRUS (HPV) IS LOW: RESULTS FROM A PROSPECTIVE COHORT STUDY

44 COMPARATIVE STUDY OF A TWO DOSE

SCHEDULE OF THE PGE~ ANALOGUE MISOPROSTOL FOR LABOR INDUCTION IN

PATIENTS WITH AN UNFAVORABLE CERVIX

45 NON-INVASIVE ASSESSMENT OF THE MATERNAL CEREBRAL CIRCULATION BY

TRANSCRANIAL DOPPLER ULTRASOUND IN THE HYPERTENSIVE CRISES OF PREGNANCY

Watts DH, Kouts~f LA, Holmes KK, Lee S-K, Goldman D, KuypersJ,

Kiviat NB, Galloway, DA, University of Washington, Seattle

Sanchez-Ramos L, Farah L, Rosa C, Johnson J,, Delke I,

Dd Valle G, University- of Florida,

Jacksonville

Payne AJ, Naidu & Moodley J,

Hoffmann M, Gouws F, University

of Natal, South Africa

3:15-3:30 46 SINGLE AND MULTIPLE MARKERS OF ALTERED VAGINAL FLORA: EFFECT OF TREATMENT ON PRETERM BIRTH (PTB)

Andrews WW,, Hauth JC, Goldenbog RL, DuBard M,

Chandra G, Copper RL, University

of Alabama, BirIningham

xviii January 1996 American Journal of Obstetrics and Gynecology


1:00 p.m. to 3:30 p.m. Oral Concurrent Session D Maternal-Fetal and Placental Physiology Moderator: James E. Fergnson, IIi, MD

Kona Ballroom

1:00-1:15 47 L-ARGININE REVERSES THE HYPERTENSION INDUCED BY CHRONIC NITRIC OXIDE SYNTHESIS INHIBITION IN PREGNANT BUT NOT IN NONPREGNANT RATS

Ahohas RA, Friedman SA,

Sibai BM, University of Tennessee, Memphis

1:15-1:30

1:30-1:45

1:45-2:00

48 EFFECT OF CHRONIC TOCOLYI’IC THERAPY ON MATERNAL VENTRICULAR FUNCTION IN PREGNANT RABBITS

49 INTRAPULMONARYSHUNT (QS/QT) AND POSITION IN HEALTHY THIRD-TRIMESTER PREGNANCY

50 EXPRESSION OF OXYFOCIN AND OXSTOCIN RECEPTOR IN AMNION, CHORION, BASAL AND CHORIONIC PLATES AT TERM IN SPONTANEOUS, AND INDUCED LABOR

Russo LtL Besinger RE,

Tomich PG, Thomas JrJX,

Loyola University, Stritch School

of Medicine, Maywood, IL

Hankins GDV, Harvey C, Clark SL,

Uckan E, University of Texas, Galveston, University of Utah

MilIJF, Keasler S, Salafia CM, Starzyk KA, Vertmrg M,

Georgetown University

2:00-2:15

2:15-2:30

2:30-2:45

51 ROLE OF THE L-ARGININE/NITRIC OXIDE PATHWAY IN HYPOXIC FETOPLACENTAL

VASOCONSTRICTION (HFPV)

52 THE FETAL ECG:- PR WAVEFORM CHANGES

ASSOCIATED WITH METABOLIC ACIDEMIA

53 NITRIC OXIDE REGULATES OXYGEN TRANSFER INDEPENDENT OF FETOPLACENTAL VASCULAR RESISTANCE IN THE PERFUSED HUMAN COTYLEDON

Byrne BM, Adamson SL, Morrow RJ, Howard RB, Mount Sinai

Hospital, Toronto

Murray HG, University of Sidney at Westmead Hospital, Australia

Byrne BM, Adamson SL, Morrow RJ, Howard RB, Mount Sinai

Hospital, Toronto

2:45-3:00

3:00-3:15

54 THE ROLE OF THE PULMONARY

CIRCULATION IN THE DISTRIBUTION OF

HUMAN FETAL CARDIAC OUTPUT

55 VASCULAR ENDOTHELIAL GROWTH FACTOR

EXPRESSION IN THE CHRONICALLY ANEMIC OVINE FETAL HEART

Rasanen J, Wood DC,

Ludomirski A, Huhta J6;

Pennsylvania and Temple University

Hospitals

Martin C, Hohimer A, Davis L,

Oregon Health Sciences University, Portland, OR

3:15-3:30 56 UTERINE MYOMETRIUM CONTAINS A

SECOND GAP JUNCTION SEQUENCE

(CONNEXIN45) WITH DIFFERING EXPRESSION LEVELS FROM CONNEXIN43

Albrecht J, Atal N, Orsino A,

Lye & Saadovsky Y, Beyer EC,

Washington University, St. Louis, MO, Mount Sinai Hospital, Toronto

American Journal of Obstetrics and Gynecology January 1996 xix

Saturday, February l Oth

8:00 a.m. to 10:30 a.m. Oral Concurrent Session E Clinical Obstetrics

Moderator: Michael W. Varner, MD

Grand Ballroom

8:00-8:15 57 PERCUTANEOUS UMBII.ICAL-CORD LIGATION IN COMPLICATED

MONOCHORIONIC MUI.TIPI.E GESTATIONS

()_ui~dero IgA, God, calves L,

.]ohf~son MP, Rehh tL Romero R,

Carreno C, Evans MI, Wayne State University, Wyoming Valley OB/

G~ Associates, ~ngston, PA, NICHD MFMU Network

8:15-8:30 58 A PROSPECTIVE, RANDOMIZED, BLINDED COMPARISON OF TERBUTALINE VERSUS PLACEBO FOR SINGLETON, TERM EXTERNAL CEPHALIC VERSION

Fernandez CO, Bloom S, Wendel G,

University of Texas SW, Parkland

Hospital, Dallas, TX

8:30-8:45

8:45-9:00

9:00-9:15

59

60

61

INCREASED INCIDENCE OF PRETERM LABOR AND PRETERM DELIVERY ASSOCIATED WITH INCREASE SALIVARY ESTRIOL LEVEL

RANDOMIZED DOUBLE-BLIND STUDY COMPARING SULINDAC TO TERBUTM,INE: FETAL CARDIOVASCULAR EFFECTS

DOES THE USE OF A TOCOLYTIC AFFECT THE SUCCESS RATE OF EXTERNAL CEPHALIC VERSION

Goodwin 7WI, ./ack.son (;M, McGregcn JA, La(hdin CGL,

A~tal R, Dullier~ V, University of Sonthern California, University of

Utah, University of Colorado, University College, London, SUNY,

Syracuse, Biex Inc., Bonlder, CO

Kramer W, Saade G, Bel~brt M,

Dorman K, Mayes M, MoiseJr K,

Baylor College of Medicine, Houston

Mawuette G~, Boucher M, Theriault D, RinJket D, University

of Montreal, Saintez][ustine Hospital, Montreal

9:15-9:30 62 THE USE OF A SURGICAL ASSIST DEVICE DESIGNATED TO REDUCE GLOVE

PERFORATIONS: A RANDOMIZED TRIAL

T, vissman MJ, Bebbington MW,

University of British Columbia, gallconveF

9:30-9:45

9:45-10:00

10:00-10:15

63

64

65

24 HOUR MOTHER-INFANT DISCHARGE WITH Bru@eM C, Ndson K, StolzoD, A FOLLOW-UP HOME HEALTH VISIT: RESULTS Patterson P, Sp~ayberry N, IN A SELECTED MEDICAID POPULATION Shannon S, University of Alabama,

Birmingham

MISOPROSTOL AS A LABOR INDUCTION AGENT: A PILOT STUDY COMPARING EFFICACY, SAFETY, AND COST

Ma~ib(U P, Ogburn Jr P, Harris D, Suman V, Hamzn K, Mayo Medical Center, Rochester, MN

NEW TECHNOLOGY IN 3-DIMENSIONAL

OBSTETRICAL ULTRASONOGRAPHY:

TECHNIQUE, ADVANTAGES, AND LIMITATIONS

Ludomirski A, Uerpair(~k,t B, Whiteman VA, Reece EA, Chu GP,

Chan L, Temple University

10:15-10:30 66 THE INFLUENCE OF ULTRASOUND EXAMINATION AT TERM ON THE RATE OF CESAREAN SECTION

Adash& JA, Lagrew DC, lriye BK,

Carr MH, Porto M, Freeman RK,

University of California, Irvine, Long Beach Memorial Women’s Hospital, Saddleback Memorial

Women’s Hospital, Laguna Hills, CA

xx January 1996 American Journal of Obstetrics and Gynecology

Saturday, February 10th

8:00 a.m. to 10:30 a.m. Oral Concurrent Session F

Medical Complications of Pregnancy Moderator: David B. Cotton, MD

Kona Ballroom

8:00-8:15 67 ELEVATED AMNIOTIC FLUID INTERLEUKIN-6 PREDICTS NEONATAL PERIVENTRICULAR LEUKOMALACIA AND INTRAVENTRICULAR

HEMORRHAGE

Figueroa R, Martinez E, Sehgal P, Garry D, Patel K, Verma U, Visintainer P, Reale M, Klein S,

Tejani N, New York Medical College

8:15-8:30 68 AMNIOTIC FLUID CONCENTRATIONS OF INTERLEUKIN-6 IDENTIFY FETUSES AT RISK FOR THE DEVELOPMENT OF

PERIVENTRICULAR LEUKOMALACIA

Yoon BH, Romero R, Jun JK, Park KH, Yang SH, Kim I0,

Munoz H, ChoiJH, Seoul National University, Korea, NICHD MFMU

Network, Wayne State University, Hutzel Hospital, Detroit, MI

8:30-8:45 69 POSTNATAL GROWTH AND COGNITIVE

DEVELOPMENT IN THE VERY LOW BIRTH WEIGHT SMALL FOR GESTATIONAL AGE INFANT

Gdles.lM, Dzvon MY, McCarton CC,

Wallace ], Albert Einstein College of Medicine, Bronx, NY

8:45-9:00 70 A RANDOMIZED CONTROLLED TRIAL OF PREDNISONE AND ASA IN WOMEN WITH AUTOANTIBODIES AND UNEXPLAINED RECURRENT FETAL LOSS

Lashin C, Bombardier C,

Mandel F, Ritchie K, Hannah M, Farine D, Farewell V, Spitzer K,

Fielding L, Yeung M, University of Toronto, University of Waterloo, Ontario

9:00-9:15 71 ELECTIVE DELIVERY FOR MACROSOMIA IN THE DIABETIC PREGNANCY: A CLINICAL COST-BENEFIT ANALYSIS

Conway D, Langer O, University of Texas, San Antonio, TX

9:15-9:30 72 A RE-EVALUATION OF THE INDICATIONS FOR PULMONARY ARTERY CATHETERS IN OBSTETRICS: THE ROLE OF 2D ECHOCARDIOGRAPHY AND DOPPLER ULTRASOUND

BelJbrt M, Mares A, Saade G, Wen T, Rokey R, Baylor College of Medicine, Houston, TX, Marshfield

Clinic, WI

9:30-9:45 73 BACTERIAL VAGINOSIS (BV) IN A POPULATION OF 3600 PREGNANT WOMEN AND RELATIONS TO PRETERM BIRTH

EVALUATED FROM THE FIRST ANTENATAL VISIT

Thorsen P, Molsted K, .]ensen IP,

Arpi M, Bremmelgaard A, Jeune B,

Moller BR, University Hospital, Odense, Denmark

9:45-10:00 74 WOMEN WITH SICKLE CELL TRAIT ARE AT INCREASED RISK FOR PREECLAMPSIA

Larrabee KD, Monga M, University of Texas, Houston, TX

10:00-10:15 75 THE EFFECT OF MATERNAL POSITION ON (2MIDIAC OUTPUT WITH EPIDURAL

ANALGESIA IN LABOR

10:15-10:30 76 THE EFFECT OF DOBUTAMINE HYDROCHLORIDE ON VENTRICUIAR FUNCTION AND OXYGEN TRANSPORT IN PATIENTS WITH SEVERE PREECLAMPSIA

Danilenko-Dixon DR Tefft L, Haydon B, Cohen RA, Carpenter MW,

Brown University, Women & Infant’s

Hospital, Providence, RI

Graves C, Wheeler 7; Troiano N, Vanderbilt University,

Nashville, TN

American Journal of Obstetrics and Gynecology January 1996 xxi

SPO Foundation Fellowship Award Winner For Academic Year 1994-95

Thursday, February 8, 1996 7:45 a.m. - 8:00 a.m.

Grand Ballroom

Susan P. Seligman, MD

"Relationship between preeclampsia, hypoxia, and production of nitric oxide by the placenta"

xxii January 1996 American Journal of Obstetrics and Gynecology

RELATIONSHIP BE’RNEEN PREECLAMPSIA, HYPOX|A AND PRODUCTION OF NITRIC OXIDE BY THE PLACENTA Susan P. Seh~]man, Susan. S. Kadner × and Thomas H. Finlay,x Department of Ob/Gyn, NYU Medical Center, NY, NY. OBJECTIVE: Preeclampsia is associated with structural defects in the spiral arteries supplying the mtervillous space. As a consequence, the placental vascular bed in preeclampsia is relatively hypoxlc. Nitric oxide (NO) has been shown to maintain low vascular tone in ~ uteroplacental circulation and its synthesis may be decreased in preeclampsia. Because of potential relationships between NO, hypoxia and preeclampsia, we compared the expression of eNOS mRNA in trophoblast from normal placentas maintained under hypoxic and normoxic conditions and in trophoblast from preeclamptic placentas. STUDY DESIGN: Trophoblast were isolated from normal or preeclamptic placentas at term. Trophoblast cultures from normal placentas were maintained under normoxia (20% 02) or hypoxia (0-1% 02) for 48 h. Steady state eNOS mRNA levels were determined by northern blot analysis using a [32p]-Iabeled bowne eNOS cDNA probe. The relative amounts of eNOS mRNA were determined by dens~tometdc scanning and were normalized for levels of actin mRNA. RESULTS: Under hypoxia, trophoblast remain viable for at least 72 h. There is no sigmficant difference ~n overall protein synthesis in cells maintained under hypoxic or normoxic conditions and the banding pattem alter SDS-PAGE of 35S- labeled proteins from trophoblast maintained under hypoxia and normoxia is remarkably similar. Some specific differences in the hypox~c trophoblast are apparent: particularly a 5-fold reduction in the synthesis of ~hCG and a 3-4-fold increase in the synthesis of (zl-antitrypsin. As in other tissues, trophoblast

respond to hypoxia by synthesizing the DNA binding protein hypoxia-inducible factor 1 (HIF-1). Levels of eNOS mRNA were found to be 4-fold higher under normoxia than hypoxla. Similar results were obtained when NOS actiwty was determined directly by measuring the conversion of [3H] arginine to citrulline. In trophoblast from the severely preeclamptic placentas, expression of eNOS mRNA was as little as 10% of that in normals, while in trophoblast from the mdd preeclamptic placentas, levels were intermediate. CONCLUSIONS: These results are consistent w~th a model in which the hypoxia-inducad impairment of NO production by syncytiotrophoblast results in uteroplacental ~nsufficiency and the production of a toxin responsible for t~ climcal mamfestations of preeclampsia.

American Journal of Obstetrics and Gynecology January 1996 xxiii

Oral Plenary Session I

Thursday, February 8, 1996 8:00 a.m.- 10:00 a.m.

Grand Ballroom

Moderator: Timothy R.B. Johnson, Jr.., MD

Judges: Richard L Lowensohn, MD David A. Luthy, MD James A. McGregor, MD

ABSTRACT NUMBERS: 1-8

Volume 17t, Number 1, Part 2 SPO Abstracts 303 Am J Obstet Gynecol

DIETARY VITAMIN E PROPHYLAXIS AND DIABETIC

EMBRYOPATHY: MORPHOLOGICAL, BIOCHEMICAL, AND MOLECULAR ANALYSES. E. Sivan’, Y.K. Wux, C. Homkox, E.A. Recce, Dept. of OB/GYN & RS, Temple Univ. Sch of Med, Phil&, PA. OBJECTIVES: The present study sought to determine whether dietary sopplementatiou with Vitamin E, a known antioxidant, would reduce the incidence of diabetic embryopathy in an in-vive rat model. STUDY DESIGN: Eighty day old Sp~ague Dawley rats (n = 80) were

randomly assigned to one of four groups: two control groups (Group 1

and 2) and two diabetic groups (Groups 3 & 4). One group of controls (Group 2) and one group of diabetic rats (Group 4) received dietary supplements of vitamin E (400rag/day). The other two groups (Groups 1

& 3) received a n~mal diet only. On day 6 of gestation, diabetes was induced in Groups 3 & 4 with Streptozotocin (65mg/kg). Animals were sacrificed on day 12; embryos and yolk sacs were examined for evidence

of malformations and DNA/RNA content. Myoinositol and arachldonic

acid levels were analyzed by GC mass spectrophotometry. RESULTS: Both groups (Group 3 & 4) of diabetic rats had a signifi-

cantly higher mean blood glucose level than controls (375 vs 110

mg/dI.,). The unsupplemented diabetic group had a mural tube defect (NTD) rate of 21.8%; abr, orption rate of 19.6% as c~npared to the supplemented diabetic group 7.3% and 2.2% respectively, (p < 0o01). Both control groups had similar NTD rates (5.8% and 6.75, respectively;,

p > 0.05). Vitamin E levels, measured by HPLC, were sighiflcantly higher in the supplemented groups (37.28,7.79 vs. 9.76~1.99 Nl/ml; p <0.001) than in controls. Supemxide dismutase activity was reduced in

the diabetes group and restored with vitamin E therapy.

CONCLUSIONS: Supplementation with the matloxldant, vitamin E, confers a significant protective effect against diahetic embryopathy, and which may lmtentiaily s~rve as a dietary prophylaxis hi the

future. We postulate that this protective effect is mediated by a reduc- tiou in the oxidative load induced by hyperglycemia.

FETAL FIBRONECT1N AS A PREDICTOR OF PRETERM BIRTH 1N SYMPTOMATIC PATIENTS--A MULTICENTERTRIAL AM Peaceman WW Andrews, JM Thorp, SP Chver’, A Lukes~, and the 031 Fetal Fibmnectm Study Group. Depts. of Ob/Gyn, Northwestern Unlv Med School, Umv of Alabama, and Univ ofN Carohna, Cthcago, I1, Bwmingham, A1, and Chapel Hall, NC OBJECTIVE. To detenmne whether the presence of fetal fibronectm (IFN) in cerv~covaginal secrouons of patients w~th symptoms suggestive of preterm labor predicts preterm delivery. STUDY DESIGN’ Patients presentang to the hospital between 24 and 34 6/7 weeks’ gestation with singleton pregnancy, intact membranes, no prior tocolysis, symptoms suggestwe ofpreterm labor, and cervical ddatton < 3 cm were recruited at 10 sites. Swabs of the posterior fom~x were assayed for the presence of fFN by monoclonal antibody assay, with a pos~twe result defined as >50 ng/ml Results were not available to the managing physicians Tocolysis was employed when chnically re&cared after specimen collection. RESULTS: 725 patients had IFN results and pregnancy outcome data avadable

for anal~,s~s

Mean GA at sampling

Tocolysis

Delivery within 7d

Dehvery w~th~n 28d

Delivery < 37 weeks

+fin (n=142)

30.6

58

19

47

61

-fin (n=583)

30.3

139

2

48

78

1.7 (1.3, 2 2)

39 (9.1,165)

4.0 (2 8, 5.7)

3.2 (2.4, 4.3)

The negative predictive value for dehvery w~thin 7 days, wtth~n 28 days, and at <

37 weeks was 99 7%, 92°/’o, and 87%, respectwely Using multiple logistic regression analysis to control for potential confounding vermbles, only the

presence offFN (OR 31.2, CI 6 0-163) and tocolys~s (OR 5 7, CI 1.5-21.9) were

assocmted wth b~rth w~thin 7 days; cervical dilatation >1 cm (OR 2.7, CI 1 5-4 8),

tocolys~s (OR 4.8, CI 3.0-7 6), and fFN (OR 3 6, CI 2 2-5 9) were all independently associated w~th dehvery at <37 weeks.

CONCLUSION. In a population of symptomaUc patients, the presence offFN

m cerv~covaginal secretaons best defines a subgroup at increased risk for dehvery

w~th~n 7 days, the h~gh predictive value of negative fFN samphng supports less

intervention for patients with this result.

THE PRETERM PREDICTION STUDY: FETAL FIBRONECTIN, BACTERIAL VAGINOSIS AND PERIPARTUM INFECTION. R. Goldenbero, J. lares, B. Mercer, P Mois, A. Moawad, R. Copperx,

A. Das×, E. Thom×, F. Johnsonx, J. Roberts, D. McNellis×, and the NICHD MFMU Network, Betheeda, MD. OBJECTIVE: To determine the relationship between vaginal and upper genital tract infection and cervical-vaginal fetal fibronectin (FFN). STUDY DESIGN. 2,899 women at 10 centers were screened routinely every 2 wks from 23-24 to 30 weeks’ gestation for FFN A positive test was defined as a value >_ 50 ng/mL The odds rahos (OR) for a positive FFN test and bacterial vaginosis (BV) by Gram Stain at 23-24 wks predicting clinical (C) and histologic (H) chorioamnionitis (CA) at delivery and neonatal sepsis (NS) were determined. RESULTS: FFN was present in 4.0% of cervical and/or vaginal samples at 23-24 wks and was nearly twice as common in women with BV [OR 1 9 (1.2-2.7)]. Adjusting for the presence of BV, race and parity, women w~th a positive FFN were s~gnff~cantly more likely to have CCA (OR = 16.4 [7.1 - 37.8]) and NS (OR = 63 [2.0 - 20.0]) than those who were FFN negative. A positive cervical FFN (OR = 20 6 [8 7 - 48.7]) test was a better predictor of CCA than was vaginal FFN or a combination of vaginal and cervical FFN. In the same models, BV had an OR for CCA of 1.7 (0.7 - 3.9) and for NS an OR of 2.9 (1.1 - 7.8). In 40 women who delivered at less than 32 wks and had placental histology available for evaluation, 10 had a positive cervical and/or vaginal FFN test prior to delivery. All 10 (100°{,) had h~stologic evidence of CA compared to only 13 of 30 (43%) women who were FFN negative (p= 002). CONCLUSION: Women w~th BV were more likely to have a positive FFN. Women with a positive FFN who delivered at less than 32 wks

had a 16 - 20 fold increase ~n CCA, a 6-fold increase in NS, and significantly more HCA. There is strong evidence that upper genital tract Infection and cervical and/or vaginal FFN are closely linked.

INTERNATIONAL TERMPROM TRIAL: A RCT OF INDUCTION

OF LABOR FOR PRELABOR RUPTURE OF MEMBRANES AT TERM. M Hannah, A. Ohlssonx, D Farine, S. Hewsonx, E. Hodnettx, T Myhrx, E. WangX, J Westonx, A WillanX, for the TermPROM Study Group Umvers~ty of Toronto, Canada OBJECTIVES: To compare (l) mdunhon w~th IV oxytocm (I/o) vs

expectant care + oxytocm ff induction needed (E/o), (2) induction with vaginal PGE2 gel (I/p) vs expectant care + PGE2 gel ~f reduction needed (E/p), (3) ~nductmn w~th vaginal PGE2 gel (I/p) vs mductten w~th IV oxytocm (I/o), for women vc~th PROM at tema, m terms of neonatal

mfechon, cesarean seetmn (CS) rates, and sat~sfactmn wtth care STUDY DESIGN: Women w~th PROM at _>37wks were randomized to 1 of 4 groups E/o, I/o, Up, E/p Sample s~ze was 4960 women Infants had a CBC and blood culture within 24 hrs ofbrrth Neonatal mfect~on required chmcal s~gns and was determined bhnded to allocatmn group.

RESULTS: 72 centers from Canada, Britain, Australia, Sweden, Israel, and Denmark participated 5042 women were recruited to the study Data were received on 5041 (E/o N=1263, I/o N=1258, Up N=1259, E/p’ N=1261) Basehne data were similar ~n the 4 groups. There were no differences in rates of neonatal ~nfect~on (E/o 2 9%, I/o:2 0%, I/p.3 0%, E/p 2 7%) There were 4 infant deaths not due to anomahes E/o (n=2) [b~rth trauma(1 ), asphyxia ( 1)], E/p (n=2) [mfectmn (1), asphyyaa (I)] There were no thfferences ~n CS rates (E/o" 9 7%, I/o 10 1%, I/p.9.6%,

E/p 10 9%) Other fetal/maternal outcomes were

Me&an ttme to delivery (hrs)* 25E/°2 10I/°5 15I/P3 24E/P3 I [ Chonoammonlt~S# 8.6% 4 0% 6 2% 7 9% t 1~ 36% 1.9% 31% 30%

*i/o vs E/o, I/p vs E/p, I/p vs I/o. p<.000l, #I/o vs E/o p< 01 Sahsthction with care was highest m the I/o and I/p groups. CONCLUSIONS: Induction of labor w~th IV oxytocin is the preferred

optmn for women vath PROM at term

304 SPO Abstracts JanuatT 1996 Am J Obstet Gynecol

AMNIOTIC FLUID EMBOLISM AND OXYGEN TRANSPORT PAT- TERNS. C. Harvevx, G. Hankms, S. Clark. The Umversity of Texas Medical Branch at Galveston, TX. Ammotlc find embohsm (AFE) may be an accelerated physiologic process revolving cytol~nes, mcluding tumor necros~s factor and interleul~ns (IL-1, 2, 6, & 8), producing oxygen free ra&cals and t~ssue dysoxia, sirmlar to shock syndromes. OBJECTIVES: The physiological pattern of oxygen transport dunng the shock syndromes have been &wded into early, rmddle and late temporal patterns of cardiac mdex (CI), oxygen dehvery mdex (O2AVI), and oxygen consumption index (VO2I). The pattern of physiologic compensat:on has predictive value m detemuning the release of specific cytokines (IL-1, 6, 8, & TNF) and produces distract patterns of early, n’addle and late seps~s syndrome that predicts survx- vors The objectives of th~s study were to measure the patterns of oxygen transport and evaluate temporal patterns of response compared to shock syndrome in AFE. STUDY DESIGN: Saxteen (16) pregnant goats (Capra hircus) ~n the third trimester were given fresh autologous ammotic fluid w~tbout meconium in a volume of 2.5 ml/kg body weight. Seven recieved flutd contaimng mecomum. Six control aNNals received equal volumes of allanto~c find. Oxygen transport variables were measared at basehne, 10, 30, 60, 120, and 180 rmnutes after ~njectaon. Pulmonaty artery catheters and femoral artery lung water catheters were placed for speck- men/data collection. RESULTS: CI ~n both the AF and MEC groups followed !~l the temporal pattern of rmddle stage sepUc shock. The AF and MEC patterns were characteristic of"survivors" and "nonsurvivors", respectively. O~AVI and VO~I in the two groups followed patterns of late stage septic shock, with the AF and MEC groups maintalmng profiles of survivors and nonsurvivors, respectively. (See graph.) Middle stage patterns were present ~n the AF and MEC groups at the first sample collection at 10 nmnutes. CONCLUSIONS: Oxygen transport variables of AFE reproduced the temporal patterns of septic shock. The presence of mecomum altered the oxygen transport patterns of shock syndrome and accelerated the stages of the syndrome. Thts is the first study to report the staking s~milantles of oxygen transport patterns between AFE and both septac syndromes ~n any atoNal or human model.

L-ARGININE REVERSES THE ADVERSE PREGNANCY CHANGES INDUCED BY NITRIC OXIDE SYNTHASE INHIBITION IN THE RAT. *GD Helmbrecht. **MY Farhat, *KT Yadgarova, **L Lochbaum, **HE Brown, *GS Eglinton, **PW Ramwell Dept. *OtYGyn, and **Physiology, Georgetown University Medical School, Washington, DC OBJECTIVE: Inhibition of nitric oxide synthase with L-Nitro arginine methyl ester (LNAME) induces a preeclampsia-like syndrome of hypertension, proteinuria, IUGR, and renal glomerular capillary endothelial lesions in pregnant rats. We attempted to reverse these changes with late pregnancy administration of L-arginine. STIJD¥ DESIGN: Timed pregnant $prague Dawley rats received infusions of either saline (n=12)(SC) or LNAME (n=12)(LC)(160 mg/kg/day) on gestational day 10 through term. On gestational day 16, half of the saline group (SA) and half of the LNAME group (LA) received L-arginine (21mg/kg/day) through delivery. Systolic blood pressures (SBP) were determined via tail cuff on days 10, 16, and 21. Pup weights were assessed at delivery, urine was collected by cystocentesle, and renal tissue were processed for histology. Data were analyzed us~] the Newman-Keuls test for multiple comparisons. RESULTS: In the LNAME treated animals, L-arginine significantly lowered SBP at late pregnancy (121+2.84mmHg v 146+5.5 mmHg) (p<.01), increase in mean pup weight (LA:5.6+O.11gm v LC:5.0 +O.02gm) (p<.001), decreased the degree of proteinuria (2+ v trace), and the propaion of injured glomeruli (19% v 64%)(p<.001). CONCLUSIONS: Lesions induced by chronic inhibition of EDNO synthesis (hypertension, IUGR, proteinuria, renal glomerulus injury) are reversed by trealment with L-arginine. These findings lend support to the potential for use of nitric oxide donors in the treatment/prevention of preodam~a.

A MULTICENTER RANDOMIZED MASKED TRIAL OF ANTIBIOTIC VS. PLACEBO TH I~PY AFTER PRETERM p I~MATUR£ RUPTUR~ OF THE MFJ~IBRANE~.

~e~e M Miodovnik, G Thurnau, R Goldenberg, A DaNx, G Merensteinx,

, Y Rabellox, E Themx, J Koberts, D M’~Nellisx, and the NICHD twork, Bethesda, MD.

O8JECTIW: To determine if antibiotic therapy will reduce infant morbidity after expectant management of women with preterm premature rapture of the membranes (pPROM). - STUDY DESIGN: At l0 centers, 614 women with pPROM between 240 and 320

weeks’, with membrane rupture less than 72 hours duration, and no oddence of chorioamnionitis, fetal distress or labor were randomized to either antibiotics or a matching placebo regimen. The antibiotic group received 48 hours of

intravenous amp. icillin and e .ryth romycin followed by 5 days of oral amoxicillin and erythromyctn unless dehvery occurred. Patients were cultured for group B streptococcus (GBS). By protocol, GBS carders were treated and e+;aluaied separately. Tocolytics and steroids were not given. The prospectively defined composite primary outcome consisted of: death, RDS, grade HI-IV IVH, stage

2-3 necrotizinl~ enterocolitis (NEC), or eady sepsis (~ 72 hours). GBS carriers and the over-an population were secondarily evaluated based on intent to treat RESULTS: Of the 614 women, 300 were assigned to antibiotics, 314 were assigned to placebo. 118 were GBS carriers. For those without GBS, the mean gestations at entry were 29.0 and 28£ weeks’ in the antibiotic and placebo groups respectively(NS). The primary outcome, for those without GBS, was less frequent with antibioties (p = 0.030) as were RDS, pneumonh and sepsis (see Table) Latency was significantly prolonged with antibiotics(p < 0.001) with significant increases at 7,14, and 21 days. Survival was 95% in both groups. The over-all population had fewer primary outcomes, RDS, and NEC with

antibiotics. GBS carriers did not benefit from additional antimicmbial therapy.

I Antibiot4811cs (%)

Placebo (%) RR CI N

~9 257

Primary outcome 55 0 82 0.69-0.98 RDS 51 0 80 0.66-0.97 Sepsis (eadv or late) 15 0.54 0.33-0 88 Pneumonia 3 7 0.42 0.18-0.96 BPD 14 21 0.68 0.46-1.00 Chorioamnionitis 24 34 0.70 0.53-0.93

COANCL~JSlONS: Antimicrobial therapy of expectantly mana~d pPROM at 24u3- 2u weeks’ gestat’on s’gmficantl)~ reduces infectious and-non’-mfectious infant morbi&ty.

MANAGEMENT OF SYMPTOMATIC PLACENTA PREVIA: A RANDOMIZED, CONTROLLED TRIAL OF IN-PATIENT VERSUS OUT-PATIENT EXPECTANT MANAGEMENT DA Wing. RN Paul, LK Millar, Dept. of Ob-Gy~, Ualversity of Southern California School of Medicine, Los Angeles, CA. OBJECTIVE: To deterinine the safety, efficacy and c~ta of in-patient and out-patient managen+ent of symptomatic placenta previa. METHODS: Fifty-three patients with placenta previa at 24-36 weeks’ gestation requiring hospitahzation for vaginal bleeding were stabilized, then randomized to inpatient versus out-patient expectant management. In-patients were placed at bedrest with ntinJlnal al pJ0nlation" received weeldy steroids until 32 weeks of gestation, and underwent ultrasouographie exarmnattuns at 2 week intervals to assess fetal growth and placental location. Out-patients were discharged home after a mum of 72 hours. Each week, they received steroids (untd 32 weeks) and ultrasound evaluations. Out-p~ents with recurrent bleeding were readautted for evaluation. Subjects reacl~ing 36 weeks with persistent previa underwent anmiocentesis. When fetal lung n~uuty was present, Cesarean was performed+ Stattslacal analyses included chi- square, Student t, and Mann Whitney U t~ where appsopdate. RESULTS: In-patients and out-p~..’ents were similar with regards to age, parity, race, typ~ of prevla (complete or parttal), number of prior vaginal bleeding episodes and imtial hemoglobin

J IN-PATIENT (N=27) 0UT-PATIENT {N=26) P

EGA at entry (weeks) 29 1 ± 3.1 29.9 ± 3.1 0.35 Placental Migration 5 (18.5%) 3 (11.6%) 0.48 Crossovers 7 (25.9%) 7 (26 9%) 0.93 EGA at delivery (weeks) 34 5 ± 2.4 34.6 ± 2 3 0.90 Birthweight grams) 2413 7 ± 642 7 2607 8 ± 587 1 0.32 Neonatal Morb dfly 8 (29.6=/o) 7 (26 9%) 0 83 NICU Admissions 13 (48 1%) 11 (42 3%) 0 49

NICU days, it admltled 8 (3-62) 10 (3-67) 0.75 Maternal Hospital Days 21 (4-69) 8 (3-58) 0 0001 Maternal Transfusions 4 (14.8%) 1 (3.8%) 0.61

Thirly-three paaents (62.3%) experienced recurrent episodes of bleeding with 28 requiring expeditious Ce.~arean. There was no difference in neonatal morbidity (RDS, IVll, NEC, and cunfirmed selx~ls) between the two groups (RR 0.73, CI 0.33-1.62). There were no neonatal deaths. Cost analysis bas~J on maternal hospital days reveals a savings of $15,078 per patient if patients with symptomatic placenta prevla are treated as outpatients. CONCLUSIONS: Symptomatic placent~_previa is a senous complication of pregnancy that #aces the patient at significant risk for recurrent bleedmg and need for expedgmus cesarean delivery. Within the confines of this liImted investigation, outpatient managemeat of symptomatre placenta previa appears to be safe, effiCaClOLLS and co,,a effective.

Oral Concurrent Session A Prematurity

Thursday, February 8, 1996 1:00 p.m.- 3:30 p.m.

Grand Ballroom

Moderator: John P. Elliott, MD

Judges: Thomas J. Benedetti, MD Baha M. Sibai, MD Gary A. Dildy, MD


306 SPO Abstracts Januaiy 1996 Arn I Obstet Gynecol

PETAL PIBRONEC’rlN AND PRETERM DELIVERY. S Cox, B Littler,

J Daxx, K Leveno. Dept. Ob/Gyn, UT Southwestern, Dallas, "IX. OBJECTIVE: To determine if slgnificant concentrations of fetal flbmnectin In

the posterior vagina correlated w~th preterm birth In women diagnosed to have false preterm labor.

SI’UDY DESIGN: Women between 24 and 34 weeks’ gestation presenting to

an obstetrical emergency suite complaining of contraction-like discomfort but w~th intact fetal membranes and cervical dilation less than 3 cm were consented

for this investigation. Fetal fibronectin was assayed using an enzyme-hrlked monoclonal antibody test provided by Adeza Biomedical, Sunnyvale, Cahfornla

and values exceeding 50ng/ml were considered positive. RESULTS: A total of 193 women were examined and 175 were determined to

have false preterm labor and discharged. Of these, 25 (14%) had positive

fibronectm tests. The mean gestatioanl age at testing was not significantly

different between women with positive and negative fibronectin test (mean 31.0

± 2.4 wks vs 30.7 ± 2.7 wks). As shown in the table, a positive fibronectin test

was not predictive of delivery less than 35 weeks’ ge.tation (p = NS) whereas cervical dilation between 2 and 3 cm was predictive (p < 0.02). However, women

w~th normal fetal fibronectln levels never delivered within 7 day~ of sampling.

Negative Positive Total

Cervical Dilation N PTB N PTB N FIB

~1 cm 127 9 15 1 142 10 (T)

1-2 cm 21 1 9 2 30 3 (10)

2-3 cm 2 1 1 0 3 1 (33)

Totals 14 (8) PTB = delivery <35 w gestation CONCLUSION: Ceg, ical dilation is a better long-term predictor of preterm delivery (before 35 weeks’ gestation) than fetal fibronectm levels in women dlagnc~ed to have false preterm labor. However, women vath normal fibronectin levels avmded delivery v~thm 7 days of false preterm labor although this did not preclude them ultimately from preterm delivery.

11 Spinnato. Dept. of Ob/Gyn, University of Louisville, Louisville, ,KY and Marshall University, Huntington, WV. 0BJECTI~E! T?\deter~ine whether aggressive tocolysis (AT) of patients between 24 and 34 weeks’ gestation with preterm premature rupture of membranes (PPROM) improves neonatal outcome. STUDY DESI~! One hundred thirty-seven patients with documented PPROM between 24 and 34 weeks’ gestation were prospectively randomized to AT with intravenous magnesium sulfate (n=76) or no tocolysis (NT){n=61). The lecithin/ sphingomyelin ratio (L/S) was determined upon recruitment and every 48 to 96 hours until delivery. Both groups received weekly steroids and antibiotics pending culture results and were delivered promptly when chorioamnionitin, fetal distress or L/S of k 2.0 occurred. RES~LTSt No significant differences between the AT and NT groups were observed regarding demographic characteristics; gestational age at enrollment(30.7±2.4 vs. 30.7~2.5 weeks); gestational age at delivery (31.2±2.7 vs. 31.5±2.1 weeks); latent phase (3.9±6.8 vs. 5.3±8.2 days); development of clinical chorio- ~,nionitin (Ii/76[14.5%] vs. 4/6116.6%]); birth weight (1680~517 vs. 1773±559 grams); number of days in neonatal intensive care unit (median [interquartile range] 271261 vs. 27.51251); number of infants requiring oxygen (33/76143%1 vs. 25/61141%]} and ventilatory support (23/76130%| vs.12/61120%]); days on oxygen therapy (2127] vs. 61141 and ventilatory support (2.5114] vs. 515.5]); frequency of serious neonatal complications including hyaline membrane disease, necrotizing enterocolitis, intraventricular hemorrhage, neonatal sepsis {15/76120%] vs. 17/61128%]); and neonatal mortality (2/7612.6%] vs. 2/6113.2%]). Twenty percent(12/61) of the NT group were delivered within 24 hours of PPROM. CONCLUSI0~S* Our data suggest that AT in patients with PPROM does not significantly improve perinatal outcome. However, a possible fetal benefit from short-term tocolysis when necessary to complete steroid therapy was not excluded by this study.

10 A PROSPECTIVE RANDOMIZED CLINICAL TRIAL OF

ANTIBIOTIC THERAPY FOR PRETERM PREMATURE RUPTURE

OF MEMBRANES. S. LoveR. J Weiss, M Dlogox, P. Wflhams~, T Garite

East Bay Pennatal Methcal Associates, Oakland, CA

OBJECTIVE: To test the efficacy of two antibiotic therapies in preterm

premature rapture membranes (PPROM)

STUDY DESIGN: 112 women with PPROM between 23 and 35 weeks

gestation received one of three therapies ~n a prospective double-blind

randomizedtrlal 1) 38received 1 5 g ofampanllm-sulbactum every 6 hours

for 72 hours, followed by 500 mg amoxlcdhn elavulante orally every 8

hours (AS/AC), 2) 37 received 2 g ampicillin every 6 hours followed by oral

amoxxcdlin 500 mg every 8 hours (A/A), and 3) 37 received parenteral

placebo every 6 hours for 72 hours, followed by an oral placebo every 8

hours (P/P) Oral therapy was continued through dehvery Steroids and

tocolytle therapy were administered.

RESULTS: 49% of the neonates in the placebo group died or had sepsis,

respiratory distress syndrome or hronehopulmanary dysplasla versus 30% of

those m either antibiotic group (p=0.05 vs placebo), 26% in the AS/AC group

(p-0 05 vs placebo), and 32% in the A/A group (p=0.16) All three neonatal

deaths occurred in the placebo group (p<0 05 vs antibiotic therapy)

Neonates m the AS/AC group ~vere also s~gnlficantly heavier at birth versus

placebo (mean~- SE 1870 g ±101 vs 1543 ±95 g, p=0 01) When adjusted

for the estimated gestat~onal age at enroIlment, cigarette and drug use, and

time between PPROM and enrollment, relative to placebo the AS/AC group.

1) had sigmficantly less death, sepsis, respiratory distress syndrome and

hronehopulmonary dysplasia (p= 0.05), 2) weighed significantly more at

birth (p=0 002); and 3) significantIy postponed birth relative to both date of

rupture (mean ~= SE 6.3 ± 2 8 days longer, p=0 03) and date of eurollment

(mean =~ SE" 6 2 =: 2 5 days longer, p=0 03). Differences between the two

antibiotic therapies did not achieve significance for any of these variables.

CONCLUSIONS: Neonates of those mothers who were treated with

ampicdlm-sulbactungamoxicillm-clavulante have lower risk of death, sepsis

and respiratory distress syndrome, higher birth weight and increased latency

from PPROM to delivery

12 TOCOLYTIC EFFICACY OF NIFEDIPINE VERSUS RITODRINE; RESULTS OF A RANDOMIZED TRIAL. D.N.M. Papatsonist~, H.P. van Getjnt~, O.P. Bleker2x, F.M. Lange~, H.J. Ad~r4x and G.A. Dekker~L Departments of Obstetrics and Gynecology~, Epidemiology and Biostatistics~ Free University Hospital Amsterdam~, University of Amsterdam2, Zuiderzee Hospital Lelystad3, The Netherlands.

OBJECTIVE: To compare the tocolytic efficacy of nifedipine versus ritodrine in the treatment of preterm labor (PTL). STUDY DESIGN: 181 women with PTL were randomly assigned to either oral mfed~ptne (N= 93) or iv ritodrine (N=88). Endpoints; delay of delivery, side effects and perinatal outcome. RESULTS:

Clinical characteristics, ritodrme N=88 nifedlmne N=93 P value Gestat~onal age (wks) 29.6 (2.2) 28.8 (2.6) N.S

Nulhparlty (%) 56.8 53.8 N.S. ~

PPROM (%) 35.2 28 N.S.

Cervixdllatation(cm) 1 9 (2.1) 1.4 (2.1) N.S.

Outcome: ritodnne N =77" mfedlpme N= 93 Delivery: < 24 hours 22 (28.6%) I0 (10.8%) 0.003

< 48 hours 29 (37.7%) 19 (20.4%) 0.01

< 1 week 44 (57 1%) 31 (33.3%) 0 001

B~rth weight (gm) 1852 (836) 2137 (923) 0.03

AdmtssionNICU 64 (83.1%) 63 (67.7%) 0.02

S~deeffects 2.1 (1.6) 1.0 (034) <0.01

* R~todt ~ne administration was stopped in eleven patients because of severe maternal side effects. ( ) = SD CONCLUSIONS: Use of nifedipine in the management of PTL is associated with a longer postponement of delivery, less maternal side effects and less admissions of newborns to the NICU as compared with ntodrme.

Volume 174, Number 1, Part 2 SPO Abstracts 307

Am J Obstet Gynecol

13 INTERLEUKIN-6 (IL-6) CONCENTRATIONS IN CERVICAL

SECRETIONS IDENTIFY INTRA-AMNIOTIC INFECTION IN

PATIENTS WITH PRETERM LABOR (PTL). G Razz0x, A

I~appomx, D RinaldoX, D ArdmmX, D Tedeschlx, C Romammx

Fetal Medicine Center, Dept Ob/Gyn, Umversita’ dl Roma "Tor

Vergata", Roma Italy.

OBJECTIVE. Subclinical bacterial infectmn of the amniotlc cavity m

patients with PTL and intact membranes occurs m 10-30% of cases and

Is associated w~th ancreased levels of different cytokmes both an

amnionc fluid and at the chonon-decidual interface We hypothesized

that cervical cytokine concentratmn would also be increased an women

with PTL and subclinical bacterial infection of amniotic cavaty.

STUDY DESIGN Cervacal secretions were sampled immediately

before ammocentesis m 78 patients with singleton pregnancies and

intact membranes admired for PTL Women with concomitant lower

genital tract ~nfectlon were excluded Ammotic fluid was cultured and the

following cytok~nes measured in amniotic fired and cerwcal secretions

by ELISA’ interleukin-l[~ (IL-1I~), lnterleukln-1 receptor antagonist

(IL1-RA), tumor necrosis factor-a (TNFa) and IL-6.

RESULTS 20.5% of the amniot~c fired cultures were positive IL-I~

(p_<0.05), IL1-RA (p_<0 01), TNFa (p-<0.05) and IL-6 (p-<0 005) in

cervical secretions were each significantly increased in presence of

anfectaon There was a significant relationsh~p between IL-6 m

ammotic fired and in cervacal secretions (r=0.63, p_<0.001). An IL-6

level in the cervical secretaons above 410 pg/ml had a sensit~wty of

64 2% and a spemficity of 89,1% m the predmtion of intra-ammotlc

refection, higher than the other cytokines tested

CONCLUSION. Intra-amniotac mfect~on is associated with increased

levels of IL-IIL IL1-RA, TNFc~ and IL-6 concentraUons m cerwcal

secretions are related to IL-6 ammotic levels Measurement of IL-6 an

cervical secretions may help to non-mvas~vely identify intra-amniotic

infection among pregnancies with PTL, thus reducing the indmat~ons for

invasive procedure and improving the selection of women who may

benefit from antlblotacs

15 COMPARISON OF INTRAVENOUS MAGNESIUM SULFATE AND NITROGLYCERIN FOR PRETERM LABOR: PRELIMINARY DATA. DK Gavin, DA BayhP, TE NolanX,,FB Rigby, RC CorkX, JM Miller. Dept. of Ob/Gyn and Anestbesla, LSUMC, New Orleans, LA. OBJECnVE: Nitroglycerin has uterine relaxant properties and has been used for uterine inversion and retained placenta. TNs study compares the clinical efficacy and tolerance of intravenous nitroglycerin (NTG) with magnesium sulfate (MS) while also looking at side effects. STUDY DESIGN’ 34 of the subjects enrolled between 24-34 weeks gestation have delivered. Subjects were prospectively randomized to either IV NTG or IV MS for preterm labor. Atter successfid tooolysis, both groups then received oral mBglleslUnl ghlconate, Effectiveness of tocolytic therapy was determined by: delivery delay for 48 hours, one week, or until 37 weeks. Medication was changed to an alternate drug due to failure of the initial tocolytic agent or from severe side effects Data analyzed included gestational age at dehvety, delay of delivery at 48 hours, 7 days, and to term (_>37 wks), and total days to delivery. RESULTS: No sigNflcant differences exist when comparing patient age, gravidy, parity, gestational age at enrollment or imtml cervical dilatation. 19 patients were randomized to MS and 15 to NTG. Delivery delay was no different at any of the stated time intervals. 10.5% (2/19) patients failed MS tocolytic therapy: 1 due to toxicity and 1 with continued contractions wth resultant delivery. 46.7% (7/15) of NTG failed therapy: 3 due to severe hypotension (sustained diastolic blood pressure < 40 mmHg), and 4 after reaching maximum drug dose with no

decrease in enntractious (P < .05) AI! 7 v, ere switched to MS and none delivered acutely. Overall, the incidence of hypotension was greater in the NFG group : 10/15 versus 5/19 (P < .05). None of the MS patients had therapy discontinued

due to severe hypotension. Signific~mtly more patients on NTG complained of headache: 12/15 versus 9/19 (P < .05). 16/19 of the MS group compared with 5/15 NTG group experienced flushing (P < .05) Nausea was increased 47% versus 13°/~ along ~th palpitations 42% versus 0% in the MS group (P < .05). CONCLUSION: Arrest of preterm labor is no different between groups. Hypotension is a problem that may need to be addressed wth intravenous hydration and maternal position.

14 ELEVATED MIDTRIMESTER AMNIOTIC FLUID TUMOR NECROSIS ALPHA LEVELS: A PREDICTOR OF PRETERM DELIVERY. A Ghidini, G S. Eghnton, C.Y Spong*, C B Jenkins*, J.C Pezzullo*, M Ossandon,*J.F Mill*Permatal Research Facdity, Department of OB/GYN, Georgetown University Medical Center, Washington, D C OBJECTIVE: Placental lschemaa and mflammatmn are thought to be underlying mechanisms m the m~=lonty of cases of preterm delivery (PTD). Tumor necrosis factor-~ (TNF-a) as a cytokme released in both processes. We tested the hypothesas that amnlotlc fluid (AF) TNF-u levels in asymptomattc gestations at 15-20 weeks are a predictor of subsequent PTD STUDY DESIGN: We designed a case-control study of singleton gestations that had second trimester amniocentesis for standard indications between 1/93 and 12/94 Inclusion criteria were’ l) pregnancy outcome mfurmation available; 2) gestataonal age at ammocentesls 15-20 weeks, 3) no ewdence of fetal or neonatal structural or chromosomal anomabes, 4) absence of maternal conditions known to be associated with PTD, e g discolored AF, and hagh maternal serum u-fetoprutein. Cases delivering at _< 34 weeks (n=14) were m~tched with controls dehvermg at >_ 37 weeks (n=54) based on maternal age and parity AF samples were stored at -20C Levels of AF TNF-u were measured by lmmunoassay (R&D Systems) Statistical analysis utilized ANOVA after iog-transformataon, contingency tables, loglshc regression and receiver operator characteristic (ROC) curve analysis RESULTS: Median AF TNF-a level was significantly higher (10.2 vs 8 6 pg/ml, p=0 009) and African-American race more common (50% vs 79%, p=0.03) in women dehvermg at _<34 weeks than at term. Fetal sex and barth weaght percentile were similar in the two groups Logistic regression indicated that TNF-g was an Independent predictor of PTD after controlling for race (OR=8 1, 95% CI 1 01-65 9) ROC analysis indicated that the optamal AF TNF~ cut-off level of I 1 6 pg/ml, corresponding to 1 4 multiples of the median (MOM), had a sensitivity-36% and a specificity=93% in the prediction of PTD < 34 weeks. CONCLUSION: AF qNF-~t at 15-20 weeks identifies patients at risk for PTD at < 34 weeks, suggesting that prcexlstlng intrauterine ischemia or inflammatmn is an important risk factor for PTD A cut-off level of AF TNF-A > 1 4 MOM is the optamal predictor ofPTD < 34 weeks

16 DOES INDOMETNACIN PRETREATMENT OF MYOMETRIAL TISSUE ALTER THE TOCOLYTIC EFFECT OF ATOSIBAN? H Mc Namara×~, R Bajaj~:, S Hatangadl×~, P

Nathanlelsz,~, J Smutlan~, R Knuppel~ UMDNJ-Robert Wood Joh~see Medical School/St Peter’s Medical Center, New Brunswick, Ng Laborato~ for Pregnancy and Newborn Research, Cornell Umverslty, Ithaca, NY~ OBJECTIVE: Recent studies suggest that inhibition of myometnal contract=hty by an oxytocln antagonist may be overcome by prostaglandlns The objective of this study was to examine the effect of myometnal tissue pretreatment w~th a prostaglandln synthetase =nhlbltor, mdomethacln, on the tocolytlc effect of an oxytocm antagomst, atos~ban STUDY DESIGN. An in vitro model of uterine contractNy was used Myometnal tissue was obtained from thirteen pregnant Rambou~llet-Columbla ewes (term =145alGa) with singleton pregcancles of known gestatlonal age (GA) in the third trimester (117 v 126 dGa, = human 29 v 31 wks) The tissue was mounted in tissue baths containing Krebs-Hensellet medium aerated with 5% CO~ =n O~ at 34-36 °C and allowed to equilibrate for 60 minutes It was then stimulated with oxytomn and pretreated with fixed doses of mdomethacm Atoslban was added cumulatively In ascending doses and the dose of atoslban causing 50% i~hlbltlOn ofmyometnal contractd=[y (IC50) was calculated The log of the atoslban IC50 associated with each pretreatment dose was compared wdh the log tC50 of atoslban alone (control) using ANOVA with s=gn=ficance set at p<0 05 RESULTS’ Pretreatment with three different doses of mdomethacm d~d not s=gnlficantly effect the log IC50 of atoslba~ (p = 0 4021) When the group was split by gestatlonal age (117 v 126alGa), there was a s’~gnlficant difference in the log IC50 of atoslban between the groups (p = 0 0297)

INDOMETHACIN PRETREATMENT ATOSIBAN, LOG IC50

MEAN SEM

ALL CASES (n=13) None (control) -0 914 0 214

0 01mcg/ml ÷0 621 0 175

0 lmcg/ml -1 166 0 254

lmcg/ml -1,058 0 26

GESTATION 117 dGa (n=5) -0 651 0 122

126 dGa (n=8) -1 177 0 184

CONCLUSIONS: (1) Indomethac=n pretreatment of myometnal t~ssue d*d not alter the tocotyt~c effect of atos~ban {2) At earlier GA, a Ngher dose of atostban =s required to mhlb=t myometnal contractS[y,

308 SPO Abstracts JanumT 1996 A~n J Obstet Gynecol

17 PATHWAYS TO PREMATURITY. RL Goldenber_o, JD lame, BM Memer, PJ Mels, AH Moawad, RL Copper, A Das, ~: Them, F Johnson, D McNellis,

J Roberts, and the NICHD MFMU Network, Bethesda, MD.

OBJECTIVE: Fetal fibronechn (FFN), a short cervical length (CL) by

ultrasound and bactsnal vagmce~s (BV) were compared to estabhshed risk

factors (RF) m their abihty to predict spontaneous [ etsrm b~rth (SPB).

STUDY DESIGN: 2,929 women were screened at 2,~ wks gestabonal age

(GA) for more than 50 potential RFfor SPB ~ncludmg black race, previous

SPB, body mass ~ndex (BMI) <19.8, vaginal bleeding and perceived

centract~ons. Outcomes ~ncluded SPB <32, <35 and <37 wks GA.

RESULTS: Of the 9 factors with any s~gn~ficant association w~th SPB ~n

univadate analyses, the 3 best predictors of SPB <32, <35 and <37 wks ~n

black or white, nulhparous or mult~parous women were always FFN, a CL

-<25 mm and a h~story of SPB. Each of these 3 RF, plus vaginal bleeding and BV, were more strongly associated w~th SPB <32 wks than SPB <35

or <37 wks. BV was more common in, and a stronger RFfor SPB ~n black women. Women w~th BV had a 2-fold ~ncrease ~n FFN. Women vath a

prewous SPB were 3 t~mes more likely to have a CL -<25ram. Compared

to women with none of the 3 major RF, women w~th 2 RF had a 35-fold

~ncrease in SPB <32 wks and women with all 3 RF had a 100-fold ~ncreased nsk of SPB <32 wks Our analysis suggests a pathway leading

fi’om black race through BV and FFN to SPB. A prewous SPB is

associated w~th SPB through ashort CLand a positive FFN, but ~s also an

independent RF for SPB. A typical path d~agram (for SPB <32 wks ~n

multipares) show~ng the sNn~ficant odds rabos for each porhon of the path

ts presented below

Other I~lack Low BMI !,~ ~ infections = 22 R ce

Contractions 1,| ~ Vi 1.~ BV

CONCLUSIONS" FFN, a CL -<25mm, and prewous SPB are the strongest

RF of SPB and predict early SPB better than later SPB. Black race acts as a RF for SPB through increases ~n ~nfect~ons, BV and FFN.

18 DETERMINING THE OPTIMAL GESTATIONAL AGE LIMIT FOR TOCOLYSIS: A DECISION ANALYSIS. G A Macones. T.J. BadeP, D A AschX. Center for Clinical Epldemmlogy and Bmstatisttcs, Umversity of Pennsylvania School of Methclne and Department of Obstetrics and Gynecology, Jefferson Medical College, Phdadelphta, PA

OBJECTIVE: The upper gestatannal age limit for the use of beta-agonist tocolyhc agents for pretenn labor ~s controversml We sought to define this age limit, by companng maternal and fetal risk and benefit for three strategtes for the management of preterm labor at 32, 34 and 36 weeks

STUDY DESIGN: Our decision tree compared three strategies for the management of preterm labor" (]) tocolys~s with beta-agonist, (2) no toeelysis, and (3) ammocentesis for fetal lung maturity followed by tocolysis if immature, and no tocolysis ff mature Probabdities (wgh ranges) of adverse maternal drug events, fetal morbidlties at various delivery ages, tocolytic efficacy, and lung rnatunty test characteristics were obtatned from the hterature. The primary outcome for each strategy at each gestatmnal age was an average probabihty of "health," which equally weighted the chance of an adverse maternal or fetal outcome with that strategy.

Maternal-Fetal Health Probabihty

Tocolys~s No Tocolysis Amniocentas~s

32 weeks 0.929 0.825 0 911 34 weeks 0 971 0.967 0.965 36 weeks 0.989 0.991 0.992

At 32 weeks, tocolysis provided a statistically s~gnificant increase m the maternal- fetal health probaNhty and was the opUmal strategy. At gestational ages -> 34 weeks, all strategaes resulted m similar chmcal outcomes. Of note, our sens~tiwty analysis showed that amnmcentes~s was never the best strategy at 32 or 34 weeks, due to the lack of sensitwity of an L/S ratio m predicting RDS.

CONCLUSIONS’ Tocolys~s with a beta-agomst ymlds an improved likehhood of overall maternal-fetal health at 32 weeks Tlus ~mprovement in overall outcome is not seen with tocolys~s at ages > 34 weeks Also, the use of anm~ocentesis at 32 and 34 weeks to determine tocotytic use ~s of questionable value.

Oral Concurrent Session B Genetics and

Prenatal Diagnosis


Kona Ballroom

Moderator: KarinJ. Blakemore, MD

Judges: Mary E. D’Alton, MD W. Patrich Duff, MD

J. Peter Van Dorsten, MD


310 SPO Abstracts January 1996 Am J Obstet Gynecol

19 THE ANTIOXIDANT LIPOIC ACID - PREVENTS MALFORMATIONS IN OFFSPRING OF DIABETIC RATS. _A Wiznitze_P, R. Hershkovitzx, E. Mimou=, M. Mazor~, J.R. Leibermanx, N. Bashan=, E.A. Reeee. Depts, of OB/GYN & Pediat. Metab, Lab., Soroka Med. Cir., Ben-Guriou Univ., Israel and Dept. of OB/GYN & RS, Te~nple Univ Sch of Meal, Phila., PA, USA

OBJECTIVE: The purpose of the pre~nt investigation was to determine whether LA, an anfioxJdant, is an effective prophylactic agent against

diabetes-induced e~nb~opathy. STUDY DESIGN: Following conception, lipoic acid, in doses of 10, 30

and 100 mg/kg, was administered int~aperltoneally to 80-daymld Sprague-Dawiey rats, five thnes per week (pregnancy dayg 2-6). Strep- tozotocin was injected on pregnancy day 6 and glucose levels were allowed to remain above 350 mg/dL (normal=IS0 mg/kg). Fore groups of pregnant ra~s were analyzed f~ presence of emb~opathy: In Gronps 1 & 2, mothers were under euglycemic conditions (Glu. 150 mg/dL) with

and without LA supplementation. Group 3 & 4, were hyperglycemic {Gluo 450 mg/dL) with and without LA sut~lementatiou. On pregnancy day 16 conceptuses were examined for size, resorpti~, anomalies and DNA content. RESULTS: Anomaly rates were significantly highex in offspring of diabetic rats than am~mg controls (23.5% vs. 3.4%). LA supplementaton

of 30 mg/kg, resulted in reduction of malformations (23.5% vs. 6.9%) (p <0.001), and fetal loss (17.4% vs. 8.4%) (p <0.005). LA also protected against growth restriction in diabetic e~bryos (CRL - 3.6 cm vs. CRL -

3.2 cm). The placentas of LA group were essentially pro~ected from histolngi¢ changes.

CONCLUSIONS: Lip~ic acid confers a protective effect against diabetic

embryopathy, fetal l~s and embryonic growth restriction~ Our data lends further support to the hypothesis that free radicals are causally related to diabetes-associated maldevelopment,

21 THE USE OF SECOND TRIMESTER "GENETIC SONOGRAM" IN GUIDING CLINICAL MANAGEMENT OF PATIENTS AT iNCREASED RiSK FOR FETAL TRISOMY 21. AM Vmtzdees, WA Campbell, JF Rod,s, ER Guzman, JC Smuhan, DA McLean. UMDNJ-Robert Wood Johnson Med=cal School/St Peter’s Medmal Center, New Brunswick, NJ and Umversity of Connecticut Health Center, Farmlugton CT OBJECTIVE. To use ultrasound =n the dimcal management of pahents at ~ncreased risk for fetal tnsomy 21 and also to determ=ne the eff=cacy of this approach m detechog fetuses with Iesomy 21. STUDY DESIGN. From 11t!/92 to 6/30/95 a second trimester "genetic sonogram" was offered to all women with a singleton fetus at increased risk for tnsomy 21 (_>1.274) who e~ther had dechned genetic ammocentesls or chose to have a sonogram pnor to dec~ding whether to undergo an amniocentes~s In addition to standard fetal blometry the following aneuplo~dy markers were evaluated: structural anomalies [including face, hands, and cardiac (4-chamber view and outflow tracts)], short femur, shod humerus, pyelectas~s, nuchal fold thickening, echogen~c bowel, choro~d plexus cysts, hypoplashc mid phalanx of the fifth d~g~t, wide space between the 1st & 2nd toe, and 2-vessel umbilical cord Outcome ~nformation included the results of ped~atnc assessment and follow-up after b~rth RESULTS" A total of 457 pahents between 15-23 weeks (mean +_ SD:19_+1 7)were evaluated, 295 had advanced maternal age (> 35 years), 121 abnormal serum btochem=stry and 41 had both The majority (391 or 85.5%) had a normal genehc sonogram (absence of any abnormal ultrasound markers), 41 (9%) had 1 marker present and 25 (5.5%) had _>2 markers present Outcome was obtaned on 306 pat=outs [the remaining are ongoing pregnancies (n=137) or lost to follow-up (n=14)]. Eleven of 13 fetuses w=th tnsomy 21, 1 fetus with tnsomy 13 and 1 fetus w{th tnpJo~dy had ->2 abnormal ultrasound markers present, 1 fetus w=thtnsomy 21 had 1 abnormal marker and one had a completely normal ultrasound When >_1 abnormal ultrasound markers were present, the sensdlvlty, spec=ficity, positive and negative pred=chve values for tnsomy 21 were 92.3%, 85.2% 21.8% and 99 6%, respectively. The overall ammocentes~s rate was 13 3% and among cases w~th known outcome was 19.9%. CONCLUSION: In experienced hands, second trimester "genetic sonogram" of high risk fetuses may result ~n a h~gh detechon rate of tnsomy 21 (92.3%) w~th an amn~ocentes~s rate <20%

2O LIMB REDUCTION DEFECTS (LRDs) ARE NOT

INCREASED FOLLOWING FIRST TRIMESTER CHORIONIC VILLUS SAMPLING (CVS)R Wapner, L

Jacksonx, MI Evans, MP Johnson. Divisions of MFM and Reproductive Genetics, Jefferson Medical College, Philadelphia, PA and Hutzel Hospital, Detroit, MI.

OBJECTIVE: To compare the incidence of limb reduction defects

following chorionic villus sampling to the expected incidence in the general population of 5~6/10,000.

STUDY DESIGN: From 8/83-8/95, 19,938 paUents have undergone chorionic villus sampling at our institutions and have completed their

pregnancies. All patients or their physicians were contacted following their expected date of delivery to determine pregnancy outcome and the presence of congenital abnormalities. The

incidence of limb reduction defects was calculated for the entire patient population and segregated for each week of gestation. RESULTS: There were 6 LRD’s not associated with a known genelac syndrome for an mcidence of 3/10,000. The gestationul age

at sampling and the incidence of limb reduction were:

GA _<Swks 9wks 10wks llwks I 12wks J _>13wk~

# of CVS 1,025 3,875 7,764 5,685 1,380 240

# of LRD 0 1 3 2 0 0

LRD/10,000 0 2.6 3 9 3 5 0 0 births

CONCLUSION: In experienced centers, the incidence of LRDs following CVS is not increased above the expected for the general population. There is no association between the incidence of LRDs and the gestatmna~ age at which CVS is performed.

22 GENETIC AMNIOCENTESIS MAY BE REASONABLY AVOIDED IN WOMEN WITH ABNORMAL SERUM SCREENING FOR ANEUPLOIDY BUT NORMAL

ULTRASOUND. R. Bahado-Sin~h. A. Tanx, O. Derenx, D. Hunter,

}. Copel, J. Mahoneyx Yale University School of Medicine, New Haven CT. Norwalk Hospital, Norwalk, CT. OBJECTIVE: To prospectively study the use of ultrasound biometry to refine the risk estimates for both Down syndrome (DS) and any clinically significant chromosome defects (CSCD) in women with abnormal biochemical Triple Screen (TS). STUDY DESIGN: Expected values for humerus length (HL), feraur (FL), humerus plus femur (FL+HL), and abdominal circumference (AC) were generated based on bipanetal diameter obtained from a separate normal group. Threshold observed + expected (tIE) screening values of each measurement for DS and CSCD screening were determined using receiver operator characteristics curve. Using stepwise logistic regression analysis, the optimal screening test, including nuchal thickness (NT) (normal < 6mm) for DS and CSCD detection was determined. Tables of biometry - adjusted risk for DS and CSCD for pregnancies with abnormal TS were developed. RESULTS: There were 1034 cases wrth abnormal TS for DS (risk > 1/270) or trisomy 18 (T-18) with 11 cases of DS, 1 T-18 and 17 CSCD. Abnormal NT or O/E HL <0.92 was the most sensitive combination for DS detection Abnormal NT or O/E FL+HL <0.90 was the most sensitive for CSCD detection. With abnormal biometry or anatomy the DS risk was 8/127 vs 1/753 ~n normals, OR (95% CI) 50.4 (6.4-90.2), p <0 00001. W~th abnormal biometry and/or fetal anatomy the risk of CSCD was 11/90 vs 6/830 in normals, OR 19.3 (6.4-60.5) p<0.00001. In a pregnancy with 1/270 midtrimester DS risk based on TS, w~th normal biometry and anatomy the risk fails to 1/2,431. Only one of 609 cases with normal anatomy and biometry and TS risk < 1/50 had DS, (1/609 vs 8/256), p<0 0001. CONCLUSION: Normal anatomy and biometry sigmficantly reduces the risk of aneupinidy in abnormal TS cases and therefore the need for ammocentesls.

Volume 174, Number 1, Part 2 SPO Abstracts 311 Am,] Ob~tet Oynecol

23 ENDOSCOPIC TRACHEAL PLUGGING USING AN INFLATABLE BALLOON~

IN THE FETAL LAMB.

Deprest JA*, Evrard VA, Van Ballaer PP, Verbeken EA, Vandenberghe K, Brosens IA,

Van A~sehe FA, Lerm T Centre for Surgical TecImolog~es, KU Leuven, Depts Ob/Oyn &

Pathology, Umv Hospitals "Gaethuisberg", 3000 Leuver~ Belgium.

OBJECTIVE To evaluate the feasibility and pulmonary effects of mira-tracheal

obstnlction by a detachable balloon, endoscopleally positaened by fetal tracheoscopy

STUDY DESIGN. A case-controlled surgical t~aI was performed m 13 time-dated

pregnant ewes 15 fetuses were subjected to tracheoscople balloen obstruchon during a

progressively longer period (range 2-18 days) In ease of multiple pregnancies, other

fetuses (n=10) were used as negative controls A 1 2 mm mtmscope with a double lumen

sheat, allowing low flow irrigation and a coaxial catheter loaded with a dctachable

balloen, was used for tracheoseopie plugging The first nine fetuses (Gestataonal Age (GA)

95-120 days; tenn=I45 days) were used as a pdot group to develop the technique and to

assess the obstructive performance of this plug A second group ofslx nud-tnmestes fetuses

(GA 90-99 days) was allowed a longer follow-up (range 14-18 days) to assess pulmenary

effects, using lung-to-body weight ratio (LBWR) and morphometnc terminal brochial

density (MTBD) Feasibility of the technique was assessed in all ammals using operation

time, intra-oporative comphcations, and tracheal obstructton as outcome parameters

RESULTS: The first balloon failed to remain inflated due to an evttable valve failure In

the other 14 ammals, the trachea was successfully obstructed till dehvery Tracheoscopic

manipulation tame ranged from 3 to 14 minutes One lntra-operative death occurred, but

the eontralateral control fetus also died dunng the procedure In the second group of 6

treated nud.tnmester fetuses mean LBWR was 0060 _+ 001 (0051-0075) while in

controls 0.031 4- 0.01 (range0017-0039, P<00005) MTBD was 065_+059, as

compared to 130 4- 0 80 for eentrnls

CONCLUSION Using a simple and fast techmque of fetoscop~c tracheoscopy, the fetal

trachea was successfully obstructed with an inflatable balloon Pulmonary hyperplasia, of

potential value when treating Congenital Diaphragmatic Heroin, was demonstrated

25 ELEVATED AMNIOTIC FLUID INTERLEUKIN-6 LEVELS AT GENETIC AMNIOCENTESIS PREDICT SUBSEQUENT PREGNANCY LOSS, KD Wenstrom WW Andrews, T Tamurax, M DuBardx, KE Johnstonx, GP Hemstreet£. The Unwersdy of Alabama at Birmingham.

OBJECTIVE: To determine the proportion of pregnancy loss after genetic amn~ocentesis that is related to preexisting subclinical intrauterine ~nflammation STUDY DESIGN; We accessed our bank of identically stored second trimester amniotic fluid (AF) and maternal serum (MS) samples obtained from all women undergoing genetic amn=ocentesis at our ~nstltut~on from 1988 to 1995 (n=l 1,971). Interleukln-6 (IL-6) levels were measured by ELISA in samples from every case resulting ~n spontaneous post procedure loss (excluding aneuploidy and anomalies) within 30 days following the procedure (n=66), and 66 normal controls delivered at term and matched for year of test,

gestat=onal age, maternal age, and md~cat~on for amniocentesis.

RESULTS: Mean MS IL-6 levels were the same in each group (Cases=0.02 -+ 0 07ng/ml, Controls= 0 06 -+ 0 25 ng/ml, p=0 45) Mean AF IL-6 levels were h~gher m Cases (Cases=4 0 + 13.1 ng/ml, Controls= 0 5 -+ 0,7ng/ml, p=0,04), The highest AF I L-6 levels tended

to occur in Cases with the earhest loss: < 7 days = 8.9 ng/ml, 7-14 days = 4 5 ng/ml, 14-21 days = 3.5 ns~’ml, 21-30 days = 1 9 ng/ml. The higher AF mean in the Cases resulted from the inclusion of 8 very high values (>3 S.D. = >2.5 ng/ml). When these samples were excluded, the means and range of values were the same in each group (Cases = 0.4 -+ 04 ng/ml, Controls = 0 5 -+ 07ng/ml, p = 0 58) Twelve per cent (8/66) of the Cases and 3% (2/66) of the Controls had AF IL-6 values > 2.5 ng/ml (p = 0.048, Odds Ratio = 4 1,95% C I = 1.0 - 31 2) Although the overall correlation between MS IL-6 and AF IL-6 levels was good (r = 0 50, p = 0 0015),

only I of the 8 Cases with high AF IL-6 would have been ~dent~fied by aMSIL-6 > 3 SD (> 08ng/ml). CONCLUSIONS; Analysis of our complete, unselected group of post-amnlocentesis pregnancy losses ind~cates that up to 12% may result from preexist=ng subchn=cal intrauterine inflammation. This =nflammat~on is localized and not reflected by h=gh maternal serum IL- 6 levels, such cases cannot be identified by maternal serum testing prior to the procedure

24 INTRA-AMNIOTIC PRESSURE REDUCTION IN TWIN-TO-TWIN TRANSFUSION SYNDROME Garry D,~Lysiklewtcz A, Mays J,~Tejam N

New York Medical College, Valhalla, NY

OBJECFIVE: Serial amntocentesis has been performed in the treatment of

twin-to-twin transfusion syndrome (TTTS) Reduction in lntra-ammotic pressure

(IAP) has been proposed as the reason for favorable outcome although th:s has

never been demonstrated Our purpose was to evaluate/AP before and after

decompression amarocantests m TTTS

STUDY DESIGN: Deoompressmn ammocentesis for hydramnlos was performed

in the largest ammotic fluid (AF) pocket on 9 occasions m 3 TTTS patients The IAP was determined with a water manometer before and after drainage of AF

The manometer was referenced to the top of the maternal abdomen AF was

removed until the single pocket measured 8-10 em vertacally The procedure was repeated every 7-10 days until AF indexes equalized m the two sacs Patients

were followed through delivery. The IAP was a]so determined m the same manner

in 5 singleton gestalmns and measurements compared with reported values

Results were analyzed usmg t-tests

RESULTS: The mean lAP in normal singleton gestations was 8.01 cm H20

+ 3 8 The mean initial IAP in the TTTS panents was 19 2 cm H20 +_5 67 in the

hydrammotic sac AF pressure fell significantly (mean pressure change -5 02 cm

H20, 95% CI - 2 44 to - 7 62, p 001)m 8 of the 9 measurements. The mean endtrtg

pressure was 14 2 cm H20 +_5 08 which

~vas not different from normal singleton

measurements. The volume of AF removed

ranged from 500 to 1800 cc AF mdexes equahzed

m all cases including 1 "stack" twin 6 neonates ’

v, ath a mean birth weight of 1663 gm +_80 2 for

twin A and 1033 gm +_255 8 for twin B were

delivered at an average gestational age of 32 weeks --1 7

CONCLUSION: Repetltwe decompression ammocentesls for treatment of TTTS

significant!y reduces the L, kP The ending pressure may be used to guide AF

removal volume

26 ALTERED EXPRESSION OF PLA: GENE IMPLICATED IN MOLECULAR MECHANISM OF DIABETES-INDUCED NEURAL TUBE DEFECTS (NTDs): A NEW RELEVATION. E.A. Reece, Y- K. Wu~, A. Ait-Allahx, W. Salamsh~. Department of OB/GYN & RS, Temple Univ Sch of Med, Phila., PA OBJECTIVE: Hyperglycemia has been shown to induce NTDs and cytoa~hitectural changes in ~h¢ embt,jonie neutoepithelium and yolk sac. This was shown to be causally related to an arachidonic acid (AA) deficiemy state. The object of o~ studies was to detea-mine whether the decrease in AA resulted from altered PLA~ gene expression. STUDY DESIGN: 80-day old Sprague Dawley rats were mated, and following conception were randomly assigned to a diabetes and a control group. Diabetes was induced by streptozotocin (65 mg/kg). Serum levels of arachidonic acid (AA) were obtained. Animals were sacrificed on Day 12; conceptuscs (embryos and yolk sacs) were examined for malformations; mRNAs were extracted and Northern blot analysis peffo’med using radiolabelled PLA2 probe. After standard processing for in situ hybridization, yolk sac and embryo sections were hybridized with PLhn-digoxigenin probe. Signal was detected by alkaline phosphetase- tagged antidoxigeuln antibody.

RESULTS: As we have previously reported, diabetic rats had a significantly higher malformations than controls (22.4~g vs. 3.4%; p <0.001); lower AA levels (17.83±5.84/:g/ml vs. 14.18±2.58 /:g/ml; p <0.001); increased absorption and decreased conceptus size in comparison to controls. In situ hybridization revealed distribution of PLA~ mRNA in neuroepithelium and yolk sac of non-diahetic controls, but markedly reduced gone expression among diabetics, coincident with reduced serum AA levels and embryopathy. CONCLUSIONS: We have demonstrated for the Itrst time that hyperglyeeinia lnducas a reduced ¢xpression of PLA~ gen¢ activity, which rtsuits in an AA deficiency state and embryopathy. This novel finding advances a molecular basis for DM-induced NTDs.


27 ENGRAFrMENT FOLLO]ArlNG IN UTERO BONE MARROW

TRANSPLANTATION FOR GLOBOID CELL LEUKODYSTROPHY. K.~.

Blakemore. B. Bambach,x H. Mosor,x V. Corson," C. Griffin,¯ S. Noga," E.

Perlmun,z D. Wenger,~ R. Zuckerman,~ A. Khouzami,z R. Jones? D~ts. Gyn/Ob, Nenrul, Neurogenetics, Pathot, Aneathesiol, Onc~logy, Johns Hopkins

Univ. Sch. of Med. & Kennedy Kringer Inst., l~ltimore, MD and Dept.

Pediatr, Thomas Jefferson Univ., Philadelphia, PA.

OBJECTIVE: To study whether signit’~.~nt host engraftment can occur

following in stern bone marrow transplantation (BMT) during the first trimester

using CD34+-selectlon.

STUDY DESIGN: Following genetic counseling and informed consent under

an IRB/FDA-epproved study protocol, in mere BMT was performed at 13 3/7

weeks of gestation on a fetus with glohoid cell lenkodystrophy diagnosed by

chorionic villus sampling. 2.4 x 10~° nucleaW~ hone marrow cells were

harvested from the father, the mononu¢lcer cells concentrated, and the CD34+

cells isolated in a final volume of 0.8 ml following positive selection with the

CellPro Capcate SC Stem Cell Concentrator. Flow eytometry revealed 91%

purity for CD34+ cells (1.4 x 10’) with 5% (7 x 10~) T cells; these wzre

transfused into the fetal peritoneal cavity under ultrasound gnidanca using a 22-

gauge spinal needle.

RESULTS: Sonograms at0, 4 and 14 hours, and at 1 and 5 weeks after BMT

showed normal fetal activity, heart rate, and growth. At 20 weeks, fetal demise

was diagnosed. Following induction of labor, an autopsy and engraftment

studies were performed. Histology, revealed extensive and widespread

extramedullary hematopoiesis. Hemorrhage was present in the lungs and liver.

DN~,-RFLP atndle~ showed significant percentages of donor coils in the liver,

spleen, and skin (-95%, 50%, and 9% paternal in origin, respectively).

CONCLUSION: Our protocol is the first to enable substantlul engraftment of

a fetus without a primary immunodefieiuncy disorder following in mere BMT.

The enhanced donor engraftment likely resulted from the early gestatinnul age

and CD34+-selectiun allowing greater cell numbers in a very small volume. A

lO-fuld decrease in CD34+ cells is planned for future studies. In

has the potential to correct a variety of disorders dingnnsable prenatally.

28 FETALBONE MARROW ORGAN CULTURE S.Garmel~" C.O’Donnull,x

C. Ulrich,~ T.Crombleholme,x The Fetal Treatment Program, New England

Medical Center/Tufts University School of Medicine, Boston, MA

OBJECTIVE: We developed an in vitro model of fetal bone marrow organ

culture (BMOC) in order to study the normal interaction of the stem cell end

the developing stroma as well as provide an in vitro model of in mere stem cell

transplantation.

STUDY DESIGN: The fetuses of pregannt rabbits were hnrvested on day 27

of gestation end the long bones were dissented under sterile conditions. Thin

sections of fetsi hone were placed into 30ram culture wells on filter paper

scaffolding. Cultures were fed five times a week with IMDM, penicillin/

streptomycin (100 units each/ml), 20% fetal calf serum, end hydrocortisone

(lxl0~M). Each BMOC was cultured with either Interleukin-3 (IL-3, 25ng/ml)

and Granulocyte-Macrophage-CoionyStimulating Factor (GM-CSF, 25ng/ml)

or Interleukin-6 (IL-6, lng/ml). Progenitor assays were performed on the

effluent after 14 days and the presence of myeloid, erythroid, and lymphoid

precursors confirmed by histologio exg~aiention.

RESULTS: Fetal BMOC successfully supported fetal bone marrow eNls end

stroma in culture out to 9 weeks. The vast majority of colonies were of the

Colony Forming Unit, Granulocyte-Macrophnge (CFU-GM) type. The

maximal number of colonies were seen between five and seven weeks of

culture. This held true for BMOC cultured with IL-3 end GM-CSF as well as

BMOC cultured with IL-6.

CONCLUSIONS: BMOC is a viable in vitro model for the study of HSC-

stromal interactions. IMDM with growth factor additives was found to

support BMOC growth for at least 9 ~. This in vitro model will provide

a useful tool for the study of stem coll-stromal interactions, regulatory

mechanisms controlling hematopoiesis and engraRment, and in mere HSC

transpinnt~ion.

Oral Plenary Session II (Fellows Ple~iary Session)

Friday, February 9, 1996 8:00 a.m.- 10:00 a.m.

Grand Ballroom

Moderator: E. Albert Reece, MD

Judges: Susan M. Cox, MD James Martin, Jr., MD


Volume 174, Number 1, Part 2 SPO Abstracts 315 Am ] Obstet Gynecol

29 THE EFFECTS OF HIGH DIETARY n-3 FATTY ACID SUPPLEMEN* TATION ON ANGIOTENSIN II PRESSOR RESPONSE IN HUMAN PREGNANCY. C.D. Adair, L. Sanchez-Ramns, D.L. Briones. The University of Florida, Dept.’Ob-TG’,~-n-~"Jacksonville, FL. OB3ECTIVES: To evaluate the effects of n-3 fatty acid supplementation on vascular reactivity as measured by the angiotensin’ll sensitivity test (AST). METIIODS: Ten subjects experiencing uneventfi~t pregnancies who were free of any chronic medical illnesses, between 24-34 weeks gestation, participated. Each subject was provided with OMEGA-3 softgel capsules and instructed to take three tablets three times days (approximately 3.6 grams of eicosapentaeooic acid). The AST was performed poor to and 28 days post-supplementation. Compliance was assessed by analysis of computerized pill bottles. Statistical analyses of the data were performed using a paired t-test for evenly distributed continuous data. A sample size of eight patients was requiredto detect a difference in effective pressor dose of at least 10 ng/kg/min between the pre~ and post-supplement results, assuming an ~x of .05 and a II of .20 (80% power) RESULTS: Theeffectivepressor dose before treatment 13.6:i:6.3nglkg/min) (mean + SD) was significantly less (P=0.001) than after supplementation (35.8+ 15.9 ng/kg/mm).

AST Pressor Dose ~ (n~/kg/.m~) ~ ~upplemem ~tams

Gest. A~e at Patient Compliance % Entry (’~ks) Pre- Post-

1 100 34 8 2 92 32 16 20 3 90 26 12 28 4 83 26 16 24 5 79 25 8 28 6 100 24 24 50 7 89 31 8 >_50 8 90 28 24 ~ 50 9 100 28 12 >_50 10 28 30 8 8

MeanacSD 85.1+21.2 28.4:1:3.3 13.5+6 35.8+15.9"

"p =O . O0 l

CQNCLUSIONS: High dose n-3 fatty acid supplementation resulted m an elmancement of the pregnancy-acquire~l refractoriness to angiotensin If.

31 BETAMETHASONE DELAYS ENDOTOXIN-INDUCED PRETERM LABOR IN THE MURINE MODEL. W Schwarl2x, H. Christensenx, J Careyx, C. Gonzalezx, W. F/~yburn. Depts. of Ob,/G),n & Pharma/Toxicol, Univ of Okla:, Oklahoma Health Sci. Cir., Okla. City, OK OBJECTIVE: To deterrnme whether adminisiration of [~-methasone decreased the endotoxin-induced preterm parturition rate and inhibited the rise of cytokines in the murine model. STUDY DESlGN: C3H/HeOu mice were randomized to receive an

in~aperitoneal (i.p.) injection of either 50 mcg of lipopolysaccharide (LI~S) or phosphate buffered saline (PBS) on day 15 (70%) gestation. Mice were further stratified to receive either 0.05 mg I~-methaanne subcutaneously (s.c.) on day 13 & 14 (equivalent to a dose of 25 mg for a 60kg female) or 0.1 mg of I~-methasone s.c. on day 10-15 gestation. The study consisted of 2 phases: Phase I evaluated the time of i.p. injection to delivery of the mice.

Phase II analyzed the serum and amniofic fluid levels of IL-l-a, INF-~, and IL-6 at 4 hours after i.p. injection. RESULTS’ Phase I demonstrated a delay in preterm parturition in the two groups premedicated with [~-methasone prior to receiving LPS.

Delivery

Trealment Group Dam Day (+ SEM) p Value

l. FBS i.p., day 15 9 19.6 :t: 0.2 Control

2 LPS 50 mcg i.p., day 15 9 16.2 ± 0.4 0.001

3. I~-methasone, 0.05 mg, day 13,14x LPS 50 mcg .i.p., day 15 9 18.3 ± 0.6 NS

4. ~-methasone 0.1 rag, day 10-15x LPS 9 19.6 ± 0.5 NS

5. [~-raethasone 0.1 rag, day 10-15x PBS 9 19.7 ± 0.02 NS

Phase ll assigned 36 mice to the groups 1-4 as shown above. There was a marked increase in TNF-ct and IL-6 levels in the amniotic fluid of those

mice treated with LPS only when compared with the PBS grottp (p <001) or the I~-methasone and LPS groups, (p < .00I). CONCLUSION: Adminisiration of [~methasone will delay endotoxin-

induced preterm parturition and inhibit a rise in anartiotic fluid cytokine levels in the routine model.

3O ORAL TERBUTALINE AFTER PARENTERAL TOCOLYSIS: A RANDOMIZED DOUBLE-BLIND, PLACEBO-CONTROLLED TRI,qL. ’ F{ Lewis. ~ Mercer, M Satama,x M Walsh B Sibai. Department of Obstetrics and Gynecology, University of Tennessee, Memphis. OBJECTIVE: To determine if oral terbutaline after successful intravenous tocolysis will prolong pregnancy and prevent recurrent preterm labor (PTL).

s STUDY DESIGN: 203 women at 24-34 weeks’ gestation were randomized to receive either terbutaline (5 rng PO q4 hours) or a matching placebo ajter intravenous tocolysis of PTL. Treatmeht was discontinued at 37u wk. Recurrent PTL was treated with parenteral tocolysis, followed by the assigned study medication. Latency, recurrent PTL. maternal and neonatal outcomes were assessed. The primary outcome for this study was percent delivery at 1 week. Data analysis was oerformed based on intent to treat. RESULTS: ,&,total of 200 patients completed the study 13 lost to follow-up). Mean gestational age at randomization was s~milar for terbutaline and placebo groups (31.3 vs. 31.5 wk). No differences were seen in the incidence of either delivery at 1 week or recurrent PTU In addition, median latency (3,8 vs. 3.5 wk) and mea~ gestational ages at delivery (35.6 vs. 35.3 wk) were similar. Relevant outcomes are revmwed in the Table. No reduction in neonatal complications or NICU admissions were identified with terbutaline therapy. Post-hoc analysis of 96 women enrolled prior to 32 weeks’ gestation revealed significant pregnancy prolongation (P<0.01) with the use of terbutaline. Outcome {%) Terbutaline Placebo OR CI Oelivery within I week 18 24 0.93 0.80-1.07 Delivery within 2 weeks 35 33 0.94 0.64-1.39 Delivery within 3 weeks 43 44 1.02 0.75-1.40 Delivery within 4 weeks 55 53 1.18 0.89-1,57 Recurrent preterm ~abor 20 16 1.24 0.68-2.20 Preterm delivery < 37 wk, 63 63 1.0

CONCLUSION: Prophylactic oral terbutaline following successful tocolysis is not associated with pregnancy prolongation or reduction in recurrent preterm labor. However s=gnificant prolongation of gestation is seen in women treated prior to 32 weeks gestahon.

32 REGIONAL PRETERM BIRTH SUBSETS: IDENTIFICATION AND CLINICAL IMPLICATIONS. "IT Porter, SL Clark, GA Dlldy, C Tookc- Miller~, SP Minton~, D Schlappy~, BC James*, Department of OB/GYN, University of Utah, SLC, UT Objectives: To reexamme traditional assumptions regarding the effects of obstetric management on preterm birth in the context of modern surfactant based neonatal care. Methods: Data was collected on all preterm infants admitted to each of 5 neonatal intensive care units in one regional ret~rral area during 1992 Five subsets of preterm birth were exammed: iatrogenic (I), spontaneous preterm rupture of membranes (SPROM) with delivery < 48 hours (II) or > 48 hours (III), labor without SPROM and admissmn cervical dilatation > 4 cm (IV), or < 4 cm (V) Results: Table I

Con :lusinns: Conservative management of patients with SPROM did not result in improved permatal morbidity or mortality. However, the number of NICU days and cost per newborn was significandy greater in the group dehvered beyond.48 hours. The significantly higher rate of neonatal sepsis in the advanced cervical dilatation group suggests infection as the etiology of labor making the recognition of infection more important than earlier obstetric lnterventmn. Given the availability of artificial surfactant and modem neonatal care, neither medical nor financial benefits of conservative management of SPROM or early diagnosis of PTL can be demonstrated.

316 SPO Abstracts ]anuar~ 1996 Am ] Obstet (;yne~ol

33 CHOROID PLEXUS CYSTS AND TRISOMY 18: RISK MODIFICATION BASED ON MATERNAL AGE AND MULTIPLE MARKER SCREENING. R_RJ Gratton, WA Hogge, CE

AstonX. Dept. of Ob/Gyn, Univarsaty of Pittsburgh School of Medicine, and Dept. of Human Genetics, Graduate School of Public Health, Umvarslty of Pittsburgh, Pittsburgh, PA OBJECTIVE: to determine the effect of maternal age and multiple marker screening for chromosomal aneuploidy in the assessment of risk of trisomy 18 an a fetus with an isolated choroid plexus cyst. STUDY DESIGN: The risk of trisomy 18 was calculated using Bayesmn statistical modehng The prior probabilities were obtained from the second trimester age related (maternal age 20-45) risk for trisomy 18. Conditional variables included the incidence of choroid plexus cysts (30%) in fetuses with trisomy 18, the frequency of associated anomalies (80%) and multiple marker screening for tnsomy 18 (60% detection) RESULTS: The risk of trisomy 18 in a fetus with a choroid plexus cyst independent of assocmted anomalies (CPC), in a fetus with an isolated chormd plexus cyst (Isol. CPC) and the risks modified by normal multiple marker screemng (MMS) are shown for selected maternal ages

Al~e Related Risk of Trisom~ 18

..... Age CPC IsoI.CPC Isoi.CPC/MMS 20 1/153 1/725 1/1804 25 1/135 1/641 1/1595 30 1/92 1/432 1/1047 35 1/39 1/183 1/454 40 1/12 1/54 1/133

The risk of trlsomy 18 with an isolated chornid plexus cyst and normal multiple marker screen does not approach that of ammocentesis until a maternal age of 35 or greater. CONCLUSIONS: In a fetus with an isolated choroid plexus cyst, maternal age and biochemical screening for anueploidy are important factors m estimating the risk of trisomy 18. In the presence of an isolated choroid plexus cyst and normal multiple marker screen, amniocentesis as justified only in the advanced matemal age patient.

35 LACK OF APPARENT CORTICOSTEROID BENEFIT IN el000 G INFANTS BORN AFTER PRETERM AMNION RUPTURE. S. Chapman, J.C. Hauth, R.L. Goldenberg, J. (3wen, S.F. Bottoms, D. McNelhs, C. MacPherson×, E Them.* Dept. of Ob/Gyn, University of Alabama at B~rm~ngham, Birmingham AL and the NICHD MFMU Network, Bethesda, MD. QBJEO’rlVE. To determine the effect of antenatal maternal corticosteroid treatment on selected neonatal outcomes in <1000 g infants born after preterm rupture of membranes (PROM). STUDY DESIGN: In a one year (1992-1993) prospective observational study, the NICHD MFMU Network collected outcome data for 799 infants <1000 g (379 born after PROM). Only ~nfants who were deemed potentially viable by the obstetncian and would have received a cesarean delivery for fetal distress were included in our analysis. Neonatal outcomes were compared ~n mothers who d~ and d~ not receive antenatal corticosteroids. Logistic regression variables included birthweight (BVV), sex, race, chorioamn~onitis, MgSO4 tocolysm, mode of delivery and surfactant use. RESUL’I~: 214 infants were potentially viable of whom 62 mothers received antenatal steroids and 152 did not Groups were s]milar for gestation, BW, race, amnionitts and delivery mode. Women who received antenatal steroids were more hkely to have received MgSO4 tocolysis (p<0.001). Un~variate and regression analyses controlling for multiple confounders confirmed no neonatal benefits of maternal corticostero~d usage in this observat=onal study.

Steroids (n=62) Controls (n=152) P v~lue Survival 46 (74%) 107 (70%) 0.58 IVH grade Ill/IV* 12 (19%) 25 (16%) 0.81 NEG+ 6 (10%) 8 (5%) 0.25 Seizures 8 (13%) 13 (9%) 0.38 Rehnopathy grads Ill/IV~ 9 (15%) 26 (17%) 0.64 ~ntact survival 26 (42%) 51 (34%) 0.25 Days on ventdatlon 27.6 23 2 0.26 *IVH- mtraventncular hemorrhage, +NEC- necrohzmg enterocolit~s CONCLUSIONS: Maternal corticosteroid treatment ~n women with PROM showed no apparent benefit in newborns <1000 g.

34 THE CLINICAL EFFICACY OF ORAL TOCOLYTIC THERAPY. O.A.__..=

Rus~t, J.A. Bofill, M. Andrew¢, R. Arriola~, J.C. Morrison. Dcpts. Ob/Gyn and

Preventive Mad., Univ. of Mississippi, Jackson, MS.

OBJECTPgE: To determine if oral ,tocolytic therapy al~er successful treatment

of preterm labor (PTL) improves perinatal outcome.

STUDY DESIGN: Patients with documented PTL (n = 209) had labor arrested

with aggressive inpatient tocolysig and were prospectively randomized to 3

treatment groups: 1 - placebo, 2 - terbutaline 5 rag, 3 - magnesium chloride

128 mg (every 4 hours). All received a comprehensive system of preterm birth

(PTB) prevention including: PTL edueetlon, weekly clinic visits, home uterine

contraction assessment (HUCA), phone contact, and 24-hour access to perinatal

nurse providers. Patient and provider were blinded to treatment group.

RESULTS: Four patients were lost to follow-up, leaving 205 patients for final

analysis. Groups 1, 2 and 3 had similar demographics, PTB risk factors, in

patient therapy, maternal complications, emergency visits, re.admissions, and

compliance to medication. Findings related to pregnancy outcome included:

Variable Placebo Terb Mug

n = patients (infants) 68 (71) 72 (82) 65 (69) Gust age entry (wks) 30.3 29.9 30.3 Bishop score (at entry) 5. I 5.1 4.8 HUCA > 4/hr (%) 17.0 10.3 16.6 Gust age delivery (wks) 36.4 35.2 36.0 Delay of delivery (days) 36.5 34.4 34.2 Birth weight (gm) 2871.1 2382.4 2652.7

Overall, the three groups had only: 1) 15.6% delivering at < 34 weeks; 2)

20.4% NICU admissions; and 3) a neonatal length of stay of 6.3 days (no

difference batwecn groups); figures much less than usually rel~orted with such

high-risk pregaancies.

CONCLUSIONS: Oral tocolytic therapy after arrest of PTL does not

effectively inhibit uterine activity or extend gestation. The improvement in

perinatal outcome demonstrated in all groups can be attributed to a program of

eomprehenswe perinatal PTB prevention.

36 PROGNOSTIC FACTORS FOR THE DEVELOPMENT OF FETAL

CEPHALHEMATOMA WITH VACUUM EXTRACTION. J.A. Bofil!, O,A.

Rust, M. Deridesx, K,G. Perry, Jr,, J.C. Morrison, d.N. Martin, Jr.

Dept. Ob/Gyn, Umv, of Missiso~ppb Jackson, MS.

OBJECTIVE: To determine the factors Involved in the development of

fetal cephalhematoma with vacuum extraction (V).

STUDY DESIGN: Patmnts at ~> 34 weeks’ gestation with V deliveries

(n = 322) were randomized to continuous (CV) (n = 164) or intermittent (IV) (n = 158) technique. In the CV group, vacuum was

increased to 600 mmHg and maintained until delivery was effected, In

the IV group, the level of vacuum was decreased to 100 mmHg

between contractions. No attempt was made to prevent fetal loss-of-

station between contractions in the IV group. Instrument application-to-

dehvery t~me interval was recorded (sec) and other data pertinent to the

pregnancy and the dehvery itself were recorded. The diagnosis of

cephalhematoma was abstracted from newborn records. Statistical

analysis included X2, ANOVA, and multiple linear logistic regression.

RESULTS: Approximately equal numbers of cephalhematoma were

recorded in the CV (n = 20) and IV (n = 17) groups (p = 0.686).

Factors that did nut achieve statistical significance were gestatlonal age

(p ~ 0.765), birth weight (p = 0.982), instrumental rotation (p =

O.896), no prior vaginal dohvery (p = O.O51), increasing number of

vacuum pulls (p = 0.102), and increasingly difficult cup application (p

= 0.O89). Higher station (p = O.008), Increasing asynehtiam (p <

O.OO) and ~ncreaslng delivery time (p = 0.002) were significantly

correlated with cephalhematoma. Only the last two factors achieved

statistical significance after stepwlso multiple linear regression analysis,

The logistic model had a sens~twity of 6.2%, a specificity of 96.9%, a

false-positive rate of 60%, and u false-negative rate of 10.4%.

CONCLUSION: In th~s trial of V dehvery, the only predelivery factor

found to predispose to the development of cephalohematoma was the

degree of asynchtlsm at the time the cup was placed. Rotation was not

a s~gnlflcant association. Cephalhematoma are more likely to develop

as the vacuum time Interval increased.

Oral Concurrent Session C Obstetric Risk Assessment

Friday, February 9, 1996 1:00 p.m.- 3:30 p.m.

Grand Ballroom

Moderator: Lars Magnus Westgren, MD

Judges: Manuel Porto, MD Michael G. Ross, MD Durlin E. Hickok, MD


318 SPO Abstracts ]anuaiT 1996

Am J Obstet Gynecol

37 THREE DIMENSIONAL ULTRASOUND EVALUATION OF FETAL FACIAL AND SPINAL ANATOMY. Ludomirski A, Khandelwal M, Uerpairojkit Bx, Reece EA, Chan L. Depa~ment of OB/GYN/RS, Temple University School of Medicine, Philadelphia, PA OBJECTIVE: To assess the use of 3-D ultrasound in the visualization of the fetal face and spine. STUDY DESIGN: 24 examinations of the fetal face (gest age 22-40 weeks) and 25 examinations of the fetal spine (15-24 weeks) of pregnancies with elevated AFP (more than 2.5 MOM) had been performed using 3.5/5 MHZ abdominal transducer. Surface and volume rendering mode creates "Plastic" images of the fetal face and the transparent mode enables optimal views of the spine. The three perpendicular planes are displayed simultaneously on the U/S screen and can be rotated and evaluated in order to achieve reliable views required for diagnosis and volumetric measurements. The complete dala set are stored in the work station random -access memory for future evaluation. RESULTS: 21/24 optimal "plastic" images of the fetal face were achieved. 3/24 were unsuccessful due to fetal contact to the uterine wall and interposed on the umbilical cord: 17 were normal; 4/21 abnormal (ant cephalocele, anencephahis, cleft lip/palate - 2). Surface and transparent images of the 25 spinal examination revealed 4 abnormalities (spina bifida, meningomyelocele) and 21 normal anatomy. The possibility of evaluating the three different planes simultaneously within seconds creates an ideal mode for examinin~ each vertebrae. The time required for volume data acquisitton for each patient range between 5-18 seconds, surface rendering of the face, last between 7-12 minutes, and the "transparent" spinal reconstruction took 5-7 minutes. Technology and fetal anatomy illustrated by video. CONCLUSION: Three dimensional ultrusonography provides a more detailed and a~curate view of the fetal face and spine. The tr~’,sparent mode for spinal evaluation is quick and easy to perform. 3-D technology can be used in order to improve fetal anatomical evaluations.

39 IN PREMATURE RUPTURE OF MEMBRANES AND PRETERM LABOR NEONATAL NUCLEATED ERYTHROCYTE NUMBER (nRBCs) IS RELATED TO HISTOLOGIC ACUTE INFLAMMATION AND NOT TO PLACENTAL MARKERS OF HYPOXlA. ~M. Salafia*, V.K, Minior*, J.C. Pezzullo*, A. Ghidin=, L.M. Ernst*, D.M. Sharer. Perinatal Research Facdlty, Depts. Path & OB/ GYN, Georgetown Umv. Med. Ctr, Washington DC, UCONN Mad. Ctr, Farmington, CT. OBJECTIVE: nRBCs have been proposed as a marker of fetal hypoxm. We inveshgated relationships of nRBCs to maternal, neonatal and placental features in premature membrane rupture (PROM) and preterm labor (PTL). STUDY DESIGN" From a consecuhve set of 465 non-anomalous singleton hveb~r~hs without comphcat~ons of diabetes or chromc hypertension) delivered at 22-32 weeks gestational age GA), we retrieved 193 PROM and 161 PTL cases. Of these, 171/193 PROM (89%) and 143/f61 PTL (89%) had a CBC by 3 hours of {fie. Maternal and neonatal charts were rewewed Placental lesions were scored ~n 4 categones (1) histologlc acute intrautenne inflammation; (2) uteroplacental vascular lesions; (3) chronic inflammation and (4) ~oagulation related lesions. Normalized nRBCs/dl (=WBCs, nRBCs/100 WBC) was analyzed by ANOVA and regression (p<0.05 significant). NonparameVIC testing confirmed s~gnificant results. RESULTS: Mean nRBCs/dl in PROM was 2.3/dl (range 0.016.31.82,) and in PTL 2.5/dl (range 0.024-21.4 p N.S.}. nR, BCs/dl decreased wtth TGA for ~PROM (p=0.0005) and PTL (p=0,0006). TnRB,Cs/dl were associated with: Tmatemal WBCs (PROM p--0.004, PTL p=0.04), ~’maternal temperature < 24 hrs before b~r~b ~,PRO~ a~6 PTL p=O.O0~2~, ll~itia~ ~e~,t.~,l V,/~,C c~t (PROM p<0.0001.~ PTL p=0.0008), "l’~nitial neonatal WBC count (PROM and PTL, p<0.0001),Tband neutroph~l (PROM p=0.02, PTL p<00001), acute amnionitis (PROM and PTL p<0.0001) and acute umbilical vasculitls (PROM p=0.001 PTL, p.,:0.0001). Other plac.ental les~ons and cord blood gas values were not related to nRBCs/dl. A 10xT nRBCs/d would produce only a 0.004 reduction in UV pH (p=0.54, 5-95% Confidence L~mlts 0.015-0.008) CONCLUSIONS: In this population of prelerm PROM and PTL, increased nRBCs/dl was directly related to maternal and neonatal hematologic parameters and placental h~stologic markers of acute refection, but not to blood gas evtdence of telal hypoxla or placenta) lesions related to ~mpa~red fetal oxygenation. In cases of preterm PROM and PTL with h~stologic or chmcal evidence of acute ascending =nfection, nRBCs/dl may reflect a part of the fetal response to an ~nflamed environment and not necessarily fetal hypoxia.

38 IS INTRAPARTUM FETAL BRAIN INJURY IN THE TERM FETUS

PREVENTABLE? Jeffrey P Phelan, M,D, Myoung O Ahn, MD, PhD. MPH,

L~sa Korst, M D, Glthert I Martin, M.D, Depts of Ob/Gyn, Pomona Valley

Hospital Medical Center, Pomona, CA.

OBJECTIVE: To determine whether the outcome of 209 neurologically impaired

term neonates was preventable during the mtrapartum period

STUDY DESIGN: The obstetrical and neonatal records and fetal momtor str~ps

of 209 neurologically impaired neonates were retrospectivaly analyzed. We

excluded fetal demises, premature infants, traumatic births, and twin gestations

For purposes of thin study, a potantlally preventable intrapartum neurologlc injury

was defined as one which arose dunng labor as evidenced by a fetal heart rate

(FHR) tachycardia w~th decelerat]ons and a loss of variability If a fetus

manffasted a reactavc FHR pattern on admission and later had a prolonged FHR

deceleratmn lasting unt]l delivery, the injury was considered nonpreventable. A

nonprcventublc injury was also considered ~vhenever the fetus had evidence of a

prior central nervous system (CNS) injury such as a persistent noRreact]ve (NR)

FHR pattern, or a bradycerdm on admission to the hospital.

RESULTS: Of the 209 neonates, 156 (75%) were classified as nan-preventable

injuries for the following reasons: persistent NR-90, reactive w~th prolonged

decelcrat]on -46, persistent react]vity -14, FHR bradycartha on admission _6 In the 14 ~nfants with persistent reactive FHR patterns, no infant had evidence of

asphyxia. In the prewously reactavc group w~th a normal baseline rate and a

prolonged decelerat]on, fetal injury was cwdent in patients with a deceleration-

delivery interval of 15 minutes, and were considered nonpreventable. F~fty-three

(25%) pat]ents presented w~th a reactive admission FHR pattern and later

developed a tachycardla pattern with or without a prolonged FHR deceleration;

these were considered potant]ally preventable

CONCLUSION: A large percentage of infants with fetal brain damage are not

preventable in the mtrapartum period Our data suggest that many infants have

sustained CNS injury prior to hospitalizat]on, or ai~er birth Given the rapidity of

response required in the prolonged decelerat]on group to prevent neurnlogic

injuries, many hospitals, if not all, may not be technically capable of responding

m sufficient t~me to prevent CNS injury In contrast, the tachycartha group may represent the group with potentially preventable lntrapartum ~njury; but, further

study ~s necessary to clarify the actual timing of these ~njurles

4O NUCLEATED RED BLOOD CELLS: AN UPDATE ON THE MARKER

FOR FETAL ASPHYXIA. Lisa M. Korst. M.D., Myoung O Ahn, M D., Ph.D.,

MPH, Jeffrey P. Phelan, M.D. Dept of Ob/Gyn, Pomona Valley Hospital Medical

Center, Pomona, CA

OBJECTIVE: To update our expenenc~ vath nucleated red blood cells (NRBC)

as a marker for fetal asphyxia and to determine whether a relataoeship exists

between the presence of NRBC and long-term neurningic impalrmel~t

STUDY DESIGN: NRBC data fi’om 133 singleton term neurologically impaired

neonates (N[N) were compared with cord blood NRBC of 83 term nonasphyxiated

newborns Newbems v~th anemia, IUGR, and maternal diabetes wer¢ excluded

The NIN group was separated into the following subgroups I - persistent

nomeacfive (NR) fetal heart rate (FHR) pattern from admission to delivery (N =

63), II - reactive FHR on admission followed by tachycartha w~th decelerations

& absent variabthty (N = 38), II1 - reactive FHR on admission followed by an

acute prnlonged decelarataon (N = 32) The first and highest NRBC value and the

time of NRBC disappearance ~vere assessed

RESULTS: The NIN group exhibited a significantly higher mean nuinber initial

NRBC (35 9 + 84 1, range 0-732 per 100 WBC) than did the control group (3 4

+ 3 0) (P < 0 00001) When the N1N group was separated on the basis of the

neurologic impairment, distinct NRBC patterns were observed Significant

d~fferances were obtained between the three NIN groups and the normal group,

w~th respect to the maximum levels of NRBC [I - 67 3 + 128; II - 13 4 ± 10 2, III -

12 4 + 9.1 (P < 0 0005)] Group I also exhibited a longer clearance t]rne m hours

(160 5 + 132 7) than did either Group II (51.7 + 67 6) or Group 1II (61 8 + 63 2)

(P < 0 0001).

CONCLUSION: Our ongoing study of NRBC indicates that they identify the

presence of fetal asphyxia When fetal asphyxia is present, dlstinct NRBC

patterns are observed that relate to the timing of fetal injury In general, the closer

birth is to the asphyxia/event, the lower the number of NRBC Thits, our data

contmue to support the concept that NRBC levels could assist in the t]mlng of

fetal neurologlc injury

Volmne 174, Number l, Part 2 SPO Abstracts 319 Am J Obstet Gynecol

41 FETAL BIOPHYSICAL SCORE AND CEREBRAL PALSY

AT AGE 3 YEARS. F. Marmin~g, C. Harman, S. Menticoglou.

Dept. Ob/Gyn, Univ. of Manitoba, Winnipeg, Manitoba, Canada.

OBJECTIVE: To determine the relation, if any, between the

last fetal biophysical profile score (BPS) and the incidence of

cerebral palsy (CP) at age 3 years, and to contrast the incidence

of CP among tested and non-tested pregnancies.

STUDY DESIGN: A prospective controlled clinical study of the

incidence of CP among 22,336 high risk patients managed by

seria/ BPS testing and 30,224 mixed high risk/low risk non-

tested control patients.

RESULTS: Between 1989-1991 189 cases of CP were identified

in 52,560 patients (rate 3.66 per 1000): 27 CP cases (14.3% total

cases) occurred in the tested group (rate 1.2 per 1000) and 162

CP cases (85.7%) occurred in the non-tested controls (rate 5.36

per 1000). These differences were highly significant (p<0.001).

The relation of last BPS to the CP rate was inverse, exponential

and highly significant (p<0.0001); the CP rate ranged from 0.824

per 1000 with a normal BPS (10/10) to 250 per I000 with a very

abnormal BPS (0/10), a 300 fold difference.

CONCLUSION: A last normal BPS result does not exclude the

risk of subsequent CP. The risk of CP rises exponentially as the

BPS result falls. Management according to the BPS score yields

a significant reduction in the CP rate.

43 RISK OF PERINATAL TRANSMISSION OF HUMAN PAPILLOMAVIRUS (HPV) IS LOW: RESULTS FROM A PROSPECTIVE COHORT STUDY DH Watts. LA Koutsky,x KK

Holmes,x S-K Lee,x D Goldman,x J Kuypors,x NB Kiviat,x DA

Galloway.x University of Washington, Seattle, WA. Objective: To evaluate the risk of porinatal transmission of HPV among infants born to women with genital HPV infection. Methods: 146 infants born to women evaluated at < 20 weeks and 3~36 weeks gestation for genital HPV by questionnaire, clinical and colposcopic examination, polymerase chain reaction (PCR), and hybrid capture (HC) assays were evaluated at 6 weeks, 6, 12, 18, 24, and 36 months of age for detection ofHPV DNA by PCR from the mouth, external genitalia, and anus. Blood for HPV 6/11 and 16 serology was obtained at esch visit. PCR was pofformed using HPV LI consensus primers with PCR products hybridized with a generic HPV probe and probes to HPV types 6, 1 I, 16, 18, 31, 33, 35, 39, 45. Hybrid captme assay for HPV DNA using 14 probes was performed on specimens positive by PCR. Results: During pregnancy, 108 (74%) of 146 women had historical clinical, or DNA evidence of genital HPV infection. At 479 infant visits, HPV DNA was de~ected from 5 (1.5%) of 336 genital, 4 (1.2%) of 324 anal, and 0 of 338 oral specimens. A positive specimen was obtained from 3 (4%) of 80 infants born to women with HPV DNA detected at 34 weeks’ gestation and 5 (8%) of women without HPV DNA detected (p=0.47). All positive results in the infants were for nontypod HPV DNA, and all were preceded or followed by negative specimens. When tested further, maternal and infant specimens were of discordant I-IPV types. Result~ on a subset of specimens were confirmed in a second laboratory. No clinical manifestations of HPV were detected in any infanL Conclusions: Detection of HPV DNA from the infants was not correlated with maternal HPV states or type and was not persistent. The intermittently detected unclassified HPV types in the infants in this study may represent infection with non-genltal HPV or low-level background contamination. Although perinatal transmission of HPV is not ruled out by current data, the upper 95% confidence interval for perinatal transmission from women with any evidence of genital HPV was 2.8%.

42 PII~ECIAI~: IS INDUCTION 0F IA~0R MORE ~C~? E.M-J. Xen~, J. ~per, K ~el& D. Con~, 0. ~ger, DepL 0f 0b/G~, ~SC,

S~ ~o, ~. 0~ To ~st ~e h~ot~es~ ~t preecl~p~c women ~e more ready ~ducible ~ non-preecl~pflc women us~g ~ ~ffa~d ~duc~on protocol.

~ ~ ~ ~d ~duc~on pro~col us~g pros~ amnio~my ~d ~gh dose o~c~ w~ employed. Consecutive preecl~pflc women ~dergo~g ~ducflon of labor were prospectively comped ~ a conc~nt come.five coho~ of non-preecl~p~c women undergo~g ~duc~on of labor. To con~ol for po~nfl~ confo~g

effect, ~ pa~en~ were s~a~fled by ~lshop score at end, p~ ~d gesta~on~ age. 0u~ome v~ables ~cluded: success of ~ducflon

(de~ed ~ ac~evement of active ph~e of labor), mode of de~ve~, ma~ ~d ~e~ ~omp~eaflans.

~Y~ 1M preecl~pflc women ~d 461 non-preecl~pflc women entered ~e s~dy. ~ere w~ no si~flc~t $fference be~een ~e groups ~ Bishop score at end, o~c~ dose, Monte~deo ~ ~d

len~ of labor. For ~e women ~ Bishop score 0-3, ~e preecl~pflc

group had a five-fold ~er risk o[ f~ed ~duc~cn ~R ~.2 [1.4-21~. Overall, ~e ~sk of ces~e~ section w~ sl~iflc~fly ~gher ~ ~e

preecl~pflc ~oup ~R 2.2 [1.4-3.6~. To con~ol for ges~on~ age, p~, ~ Bishop score at end, ~o sep~a~ lo~sflc recession ~es we~ pe~o~ed. Independently, preecl~psia had a four-fold

~er ~k of f~ed ~cflon ~ 4.M [1.2-15.5~ ~d a ~ree-fold ~gher ~sk of ces~e~ sec~on ~R 3.~ [1.9-6.3~. g0N~ON: Con~ ~ ~e colony held belief ~at preecl~pflc women ~e more ready ~ducible, ~duc~on of labor ~ preecl~pflc women c~es a ~gher flsk of f~ed ~ducflon ~d ces~e~ section.

44 COMPARATIVE STUDY OF A TWO DOSE SCHEDULE OF THE PGE1 ANALOGUE MISOPROSTOL FOR LABOR INDUCTION IN PATIENTS WITH AN UNFAVORABLE CERVIX.

L. Sanchez-Ramos, L, FarahX, C. Rosa, J. Johnson, t, Delke, G. Del Valle. Dept. Obstetrics and Gynecology, University of Florida, Jacksonvdle, FL. OBJECTIVE: To determine safety and effectiveness of misoprostol tablets administered vaginally at doses of 25 I~g and 50 !.tg for labor induction in patients with an unfavorable cervix and medical indications for delivery. STUDY DESIGN: Two hundred patients were randomly assigned to receive either 25 #g or 50 ~.g of misoprostol. The med~cahon was placed ~ntravaginally in the posterior Iornix. The dose was repeated every 3 hours until adequate labor was achieved (at least 3 contrachons ~n 10 minutes). RESULTS: Among 174 patients evaluated, 88 were allocated to the 25 pg group and 86 to the 50 pg group. The incidence of cesarean delivery was higher tn the 50 p.g group: 15.1% vs. 3.4%;

(p=.02); RR 2.9; 95% CI 1.0-8.0, however, th~s was pnm.adly due to the high incidence of dystocia. Multiple logistic regression analysis of the patients undergoing cesarean delivery was performed, and the differences persisted. No differences were noted between the two groups in intrapartum complications including tachysystole, or neonatal and maternal adverse outcomes. The interval from induction to vaginal delivery was significantly shorter in the 50 p.g group (11 hours versus 15 hours; (P = .01). In 47.9% of patients in the 50 p_g group only one dose was required versus 30.6% in the 25 #.g group. P=.04; RR 1.6 95% CI 1.1-2.3. CONCLUSION: Induction of labor with 50 p.g of misoprostol significantly shortened the interval from the imtiahon of induction to vaginal delivery. The higher incidence of cesarean delivery associated with the use of 50 p.g d~d not increase the risk of maternal and neonatal adverse outcomes.

320 SPO Abstracts January 1996 Am ] Obstet Gynecol

45 NON-INVASIVE ASSESSMENT OF THE MATERNAL CEREBRAL

CIRCULATION BY TRANSCRANIAL DOPPLER ULTRASOUND IN THE

HYPERTENSIVE CRISES OF PREGNANCY. AJ Payne, S Naidu, J

Moodily, M. Hoffmann, F. Gouws, MRC/UN Pregnancy Hypertension

Research Unit, University of Natal, South Africa.

AIM: To assess maternal middle cerebral artery (MCA) flow velocity

patterns as measured by transcranial Doppler ultrasonography (TCD) in

eclempsie and imminent ectempala, and to investigate the effects of the

enticonvulsants magnesium sulphate (M GSO4) and phenytoin on cerebral

circulation.

STUDY DESIGN: A prospective randomized study was conducted in the

h~gh care obstetric unit, King Edward VIII Hospital, Durban. A total of

42 patients were recruited. Twenty-four were eclamptic, 13 receiving

MGSO4 and 11 phenytoin. The remaining 18 had imminent eclampsia,

9 of whom received MGSO4 and 9 phenytoin. Middle cerebral artery

flow velocity waveforms were measured using 2MHz pulsed Doppler

ultrasound via the trens-temporal approach, before and 15 minutes after

the loading dose of anticonvulsant. Three way analysis of variance

showed no difference between the eclamptic and imminent e~lamptic

groups which were therefore combined.

RESULTS: MGSO~ In = 22) significantly reduced the pulsatility index

(PI} in the middle oerebral artery, pro-infusion O. 78; post-infusion O. 73 -

p = 0.0001 and the mean flow velocity (MFV), pro-infusion 59, post-

rufus=on 63cm/seo - p = 0.0001. In contrast, phenytoin (n = 20) failed

to show sigmficant differences in PI, pro-infusion 0.70, post-infumon

0.69 - p > 0.05 end in the MFV, pro-infusion 65, post-infusion

64cm/sec.

CONCLUSIONS: Magnesium sulphate vasoddates the cerebral vessels

in patients with hypertensive crises of pregnancy, whereas phenytoin

fads to do so. This potential to relieve vasospasm and possible cerebral

ischaemia may help to explain why MGSO4 is a better anticonvulsant

than phenytoin, despite the success of the latter in non-pregnant

epileptic patients.

46 SINGLE AND MULTIPLE MARKERS OF ALTERED VAGINAL FLORA: EFFECT OF TREATMENT ON PRETERM BIRTH (PTB). WW Andrews, JC Hauth, RL Goldenberg, M DuBard~, G Chandra~, RL Copper~, Dept. of Ob/Gyn, University of Alabama at Birmingham, AL. OBJECTIVE: To determine the value of single versus multiple markers of altered vaginal flora to predict subsequent PTB (<37 weeks’ gestation, [wks GA]) and the efficacy of treatment w~th metronidazole plus erythromycin (M+E) to reduce PTB in these patients. STUDY DESIGN: Markers of altered vagina~ flora were determined before (22.9+2 5 wks GA) and after (27.6_+2.6 wks GA) treatment (a 2 to 1 double-blind randomization) with M+E (n=433) or placebo (n=191) ~n 624 women at risk for PTB. Markers selected for this analysis included, bactenal vaginosis (BV) by Gram stain, pH>4.5 (pH), succinate/lactate ratio >_0.4 (SLR), by gas-liquid chromatography and presence of organic ac=ds other than lactate (OA). RESULTS: Preterm Birth (%)

ye_olnal Markers Placebo M+E _p__

BV 54% 36% 0 03 pH 49% 39% 0 25 SLR 57% 34% 0.02 OA 59% 34% 0.016

~V + SLR 61% 35% 0.01 BV + OA 61% 54% 0.01 BV + pH 57% 42% 0.13 SLR + OA 67% 33% 0.007 SLR + pH 61% 42% 0.11 OA + pH 61% 40% 0.09 BV + SLR + pH 65% 42% 0.049 BV + SLR + OA 70% 33% 0.004 SLR + OA + pH 72% 40% 0.02 BV + SLR + OA + pH 76% 40% 0.01

CONCLUSIONS: Assessment of altered vaginal flora using multiple markers versus BV as a s~ngle marker identifies women at markedly increased nsk for PTB. Although M+E significantly reduced PTB in women with both single and multiple markers, the post-treatment risk remained similar and high in both of these groups.

Oral Concurrent Session D Maternal-Fetal. and

Placental Physiology


Kona Ballroom

Moderator: James E. Ferguson, III, MD

Judges: Richard K. Silver, MD

Kathleen A. Kennedy, MD Joshua A. Copel, MD


322 SPO Abstracts lanuary 1996 Am ] Obstet (;ynecol

47 L-ARGININE REVERSES THE HYPERTENSION INDUCED BY CHRONIC NITRIC OXIDE SYNTHESIS INHIBITION IN PREGNANT BUT NOT IN NONPREGNANT RATS. ~.A. Ahokas. S.A. Friedman, B.M. Sibai. Department of Obstetrics and Gynecology, University of Tennessee, Memphis, TN. OBJECTIVE: To assess the relative roles of the vascular endothelium and the autonomic nervous system in the hypertension induced by chronic nitric oxide (NO) synthesis inhibition in pregnant (PG) and non.~regnant (NP) rats STUDY DESIGN: The effect of L-arginine (50 mg/kg + 5 mg/kg/min 1V for 1 hr) on mean arterial pressure (MAP), heart rate (HR) and plasma levels of nitrare/nittke (NOx) was measured in NP and PG (day 20 of gestation) Wistar-Kyoto rats treated with NG-nitro-L-arginine methyl ester (L-NAME, 15

rag/day SC /’or 14 days). Chlorisondamine (2.5 mg/kg IV) was then administered to assess autonomic vascular tone.

RESULTS: L-NAME increased MAP and decreased plasma NOx levels in NP and PG rats (see Table). L-arginine abolished hypertension and increased plasma NOx concentration in PG but not in NP rats. The fall in MAP was accompanied by a baroreflex increase in HR which was greater in NP than in PG rats. Chlorisondamine eliminated the difference in MAP between NP

(58+7 mm Hg) and PG (46+1 mm Hg) rats following L-arginine.

MAP HR Plasma NOx N mm Hg bpm

N°~ntegnan~ rented 6 115+4 404+9 21.1:t:1.7

L-NAME 6 Baseline 157+4’ 408+18 9.8+1.0’ After L-arg. 147+5*t 485+ I1"~" 10.6+1.0’

Pregtmnt Untreated 6 91+2 428+8 35.7+2.2 L-NAME 6 Baseline 129+5" 425+7 147±2.6’

After L-arg. 96±5J" 455± 15j" 19.7+4 0*y

Mean ± SEM. * P < 0.05 vs Untreated ~" P < 0.05 vs Baseline

CONCLUSIONS: Chronic L-NAME-induced hypertension is due to endothelium-derived NO synthesis blockade in the PG rat, but is primarily centrally-mediated in the NP rat. Thus, the endothelium plays an even more prominent role in regulating vascular tone in PG than in N~ rats.

49 INTRAPULMONARY SHUNT (QS/QT) AND POSITION IN HEALTHY TI~IRD-TRIMESTER PREGNANCY. G. Haalctns. C. Harveyx, S. Clark, E U~’kanx. The Umversity of Texas Medical Branch at Galveston, TX, The Univer-

sity of Utah, Salt Lake City, UT. OBJECTIVE: The purpose of this study was to directly measure the effect of posttaon change on intrapulmonary shunting (Qs/Qt) in healthy normotensive primiparous patients at term. STUDY DESIGN: Ten normotanslve prirmparous patients between 36 and 38 weeks gestation underwent pulmonary artery catheterization via the subclavian ronte and radial arte~ canalization. The study was approved by the hospital Insti- tutional Review Board and an outside reviewer. Written mformeA consent was obtained. Baseline assessments were made with subjects in the left lateral (LL) recumbent position after a 30-wannte staMhzation period. Position changes were followed by a 10-minute premeasurement stabihzation period. Positions were the LL, right lateral (RL), supine (SLIp), knee chest (KC), sitting (SIT), and standing (STD). Carchac output was measured via thermodilutaon techmque. Blood samples were obtained simultaneously from the pulmonary and radial arteries and ana-

lyzed in duplicate for oxygen content on a blood gas analyzer (Cormng Model 168, Medfield, Mass.) Qs/Qt was calculated using the classic shunt equation. RESULTS: Subjects in the KC position had stgnlficantly lower shunts (p<0 05)when compared to the right and left lateral positions There was no sig- arficant difference within the other position groups, although the standing posi-

tion produced the greatest ranggOf Qs/Qt. Overall there was no slgnlficant correlation between Qs/Qt and cardiac output. [ Table I [

CONCLUSIONS: This is theI Qs/Qt ANDPOSIT1ON first report of directly measured I I~°sltmn ~ SD SD Error Qs/Qt in the healthy pregnant pa- [ LL 0 152 0 033 0 01

Uent It is important to note that [ RE 0 154 0 024 0 008

the posmons that most patients ~ sul, 0 138 0 016 0 005

with pulmonary compromise are [ KC 0 101 0 054 0 022

placed into ate associated with the [ SIT 0141 0019 0006

largest Qs/Qt measurements. The STD 0 135 0 069 0 024

moderate altitude of the study site is believed to contribute to the larger shunt values reported, but should not be

slgmilcant when comparing relative changes as responses to position.

48 EFFECT OF CHRONIC TOCOLYTIC THERAPY ON MATERNAL VENTRICULAR FUNCTION IN PREGNANT RABBITS. L R. RussoX, R. E. Besthger, P. G. Tomich, and J X. Thomas, ~rr x , Dept Ob/Gyne & Physiology, Loyola University-Chicago, Stnteh School of Medicine, Maywood, Illinois. OBJECTIVE: It has been suggested that penpartum cardiomyopathy may be associated with Ixqamhne~ toco~c therapy The null hypothesis of this study was that chronic infusion ofterbutaline in pregnant rabbits will diminish global maternal ventncular function STUDY DESIGN: On gestational day 20 (terms31 days), two 200pl Alzet mini-osmotic pumps were implanted in the subcutaneous tissue of pregnant New Zealand white rabbits Each pump was filled wtth terbutahne (20pg/pl, n=7) or saline (0 9%; n=7) and infused continuous yl~for 7 days The rabbits were menficed on the 2gth gestational day Maternal hearts were placed on a Langendorff (non-ejecting) perfusion apparatus for the assessment of cardiac function. At a constant perfusmn pressure and heart rate, LV diastolic pressure was vaned, while LV Developed Pressure and LV ~: dP/dt, and index of left ventncular contractility and relaxation were continuously recorded Comparisons between treatment groups at caeh preload were made using a Students T-test RESULTS Hearts taken from terbutahne treated rabbits exhibited periodic arrhythmlas and exhibited mechanical alternans in 5 / 7 hearts vs only 1 / 7 in the sahne group At a preload of 0 mmHg, both LV developed pressure (79 9 vs 37 5mmHg; p<0 001) and LV dP/dt (1397 vs 651mmHg/see; p<0.001) were less in terbu~aline treated rabbits At a preload of 10 mmHg, L~/ developed pressure (83 2 vs 59.4mmHg; p<0.01) and dP/dt (1424 vs 964mmHg/sec, p<0.001) were also significantly less in terbutaline treated rabbits Left ventrieular relaxation was also impaired at all preloads CONCLUSIONS,/n this model, chronic administration ofterbutahne during late pregnancy significantly depresses global maternal cardiac function

5O EXPRESSION OF OXYTOCIN AND OXYTOCIN RECEPTOR IN AMNION. CHORION, BASAL AND CHOR1ONIC PLATES AT TERM IN SPONTANEOUS. AND INDUCED LABOR. JF Mill* S Keasler*, CM Salafia*, K A. Starzyk*, M Verburg*. DeAPt oVOB/GYN, Georgetown University Medical Center, Washington, DL. OBJECTIVE: To test the hypothesis that oxytocln receptor (OTR) mRNA levels are depressed in term pregnancies undergmng fnduction as compared to spontaneous labor with and witlaout underlying histoloulcal signs of acute Inflammation. STUDY DESIGN: Placentas and assocmted membranes were collected from 15 women delivering at term 7 who were induced, and 7 with snontaneous labor, and 1 ilellvered by. ceasarean section without labor. Out of each group of 7, 4 cases had histological signs of acute inflammation w"hlle the remaining 3 did not. The tissues were dissected into amnlon (AM) and chono-g’ecidua (CD), and chononic (CP) and basal plate (BP) samples. RNA was prepared from the samples by a modified acid/phenol method utilizing the TrlzolTM reagent from BRL. Twenty microgram (ug) ahquots of RNA from CD, CP, and BP were electrophoresdl on denaturing formaldehyde gel prior to Northern blot analysis. The resultant filters were probeff WltE a radiolabeled cDNA for human OTR. Autoradiographic signals were analyzed by. dens~tome .t~ of digitized scans. One ug allquofs were reverse lranscribed for use in reverse transcrlptlon-polymerase chmn reaction (RT-PCR) utfllztng primers for born oxytocin (OT) and OTR Formahn~flxod parafin embedded sections ololacental membranes were examined by In SltU hybr~dlzation (ISH) utihzlng dlgoxlgenln labeled nboprobes for OTR. RESULTS: OTR mRNA was detected in all three tissue types examined both by Northern analys~s and RT-PCR. RT-PCR also ~dentified expression of OTR in the ammon. ISH confirmed the results of RT-PCR, demonstrating OTR expression in the amnlon eplthehum. Chorio- decidual (CD) tissue had the highest level of OTR expresston by Northern analysis, however amnion tissues were not examined. Inflammation elevated the levels of OTR expression. The patient without labor also showed elevated levels of OTR message. RT-PCR demonstrated OT synthesis in all tissues examined. CONCLUSIONS’: It is known that OTR expression increases during gestation and shows a marked rise prior to labor, and that OTR shows homologous desensitization, potentially down regulating ~ts transcription rate. The present results suggest that inpatmnts with labor a decrease is seen secondary to that downregulataon. In infected patients the levels of OTR are elevated. Th~s may be due to a shorter exposure to OT due to the action of other uterotomns such as cytokmes The actions of OT have been classically ascribed to maternal serum levels The active synthesis of OT by all tissues examined suggests an autoerine/paracnne role for this important hormone.

Volume 174, Nurnber l, Part 2 SPO Abstracts 323 Am.] Obstet Gynecol

51 ROLE OF THE L-ARGININE / NITRIC OXIDE PATHWAY IN

HYPOXIC FETOPLACENTAL VASOCONSTRICTION (HFPV). B M Bymex" S.L.Admnsonx, R J.Morrow, R B Howardx Dept of Ob/Gyn,

Mount Sinai Hospital, Toronto. OBJECTIVES To Investigate the hypothesis that HFPV is medmted by the inhibition of basal nitric oxide (NO) productmn m the placenta.

STUDY DESIGN: Using standard methods, 25 human placental cotyledons were perfused at constant flow with Earles Salt solutinr~ with dextran and L-Arginme (control period) In the first hr of the control period, maternal

02 dehve~y was 17+1 (x _+ SE) ml rain"1 kg"1 (normoxia) It was then

reduced to 1 5_+0 1 ml mm"l kg"1 for 20 rain (hypoxia) by reducing the 02

content ofperfusate Nornlox~a was restored for 40 min Cotyledons were

then perfused for lh normoxm, 20 rain hypoxia and 40 rain recovery (experimental period) vc~th a) control perfusate n=5, b) L-Argmme free perfusate n=5, c) perfusate containing the NO synthase ir~hibitor L-NAME

n=5, d) L-NAME and the NO donor Glyceryl Tfimtrate-OTN n=5 and e) GTN n=5 Angmtensm II 0 5 gg bolus was given at tlie end of the control

mid experimental periods RESULTS" Fetal perfusmn pressure (FPP) was increased with L-NAME

39 +_ 3 (x--~SE) vs 56 _+ 3 nun2Ig, p<O.OOl(patred totest), and ItFPV was

nflub~ted AFPP = 11 _+ 2 vs 0.8 --- 0 5 mmHg, p<O. 01. Adding GTN to L-

NAME prevented a rise m FPP but HFPV was still inh~bited: I8 + 2 vs 0.8

± 0 4 nnnHg; p<O 001. I-IFPV was partmlly inhibited by GTN alone 21___ 3

vs 6 __- 1 nm~Ig, p< 0. O1 and by L-Arginine free perfusate. 13 _+ 3 vs 7 +_ 2

nnmHg, p=O.02. AFPP in response to ang~otansm II did got differ with perfusate.

CONCLUSIONS The effects on HFPV of manipulating the L-Argmine/ NO pathway support the hypothesis that maternoplacantal hypox~a inhibits basal NO production in the fetoplacental vessels.

53 NITRIC OXIDE REGULATES OXYGEN TRANSFER INDEPENDENT OF FETOPLACENTAL VASCULAR RESISTANCE IN THE PERFUSED HUMAN COTYLEDON. B.M B~rncx, S.LAdamsonx, R J.Morrow, R B Howardx. Dept.of Ob/Gy~, Mount Sinai Hospital, Toronto

OBJECTIVE Nitrovasodflators may be of therapeutic benefit m PET and

HELLP syndrome, con&tmns in which the fetus may be growth restricted mid hypoxeunc. This has prompted us to examine tlie effects of NO released by mtrovasodfiators on placental 02 transfer

STUDY DESIGN Usmg standard methods, 20 human placental cotyledons were perfused for 2brs with Earles Salt Solution with dextran and L-Arginine

(control) Perfusate was gassed ~wth a mixture of 95% 02/5%CO2 (maternal) mid 94%N2/6%CO2 (fetal) Cotyledons were perfused for2 more hrs

(experiment) with a) control perfusate(n=5), b)perfusate with the NO synthase mlub~tor L-NAME(n=5), c) L-NAME and file NO donor Glyceryl Trinitrate- GTN(n=5) and d) GTN(n=5). Fetoplacental vascular resistance (PVR) and

02 delivery, constunptmn and transport were calculated.

RESULTS L-NAIv~ ~ncreased PVR from 0 20_--t-0 01 (x __ SE) to 0 29_-/-0 01 nmfl-lg kg mlq mm"1, p<O. 00! (bypaired t-test) and decreased nmternofetal

02 transfer from 1 0_+0.1 to 0 5_+0 1 ml 02.min’l.kg’l;p=O.01 Adding GTN

prevented the L-NAME ~nduced vasoconstriction but the impairment of O2

transfer renlanled, 1 3__-0 2 vs 0 5_--+0 1 ml 02 min’l.kg’l;p<0 0001 GTN

alone d~d not s~gnificantly alter vascular resistance but reduced 02 transfer

1 7_+0 2 vs 1 0_+0.2 ml 02 rain"1 kg"1, p<O.O01.02 transfer did not change

w~th control perfi~sate for 4lirs 02 constm~ptmn was unclmnged for all treatments CONCLUSIONS L-NAME constr~cts fetoplacantaI vessels and impmrs matemofetal 02 transfer. GTN alone reduces 02 transfer without altering

fetoplacental vascular resistance. This raises concern that the clinical use of nitrovaso&lators may unpmr 02 transfer to the fetus that is compromised

52 ]’HE FETAL ECG:- PR WAVEFORM CHANGES ASSOCIATED

WITH METABOLIC AcIDEMIA. HG. M~,.II~RAY*, TI~ University of Sydney at lArestmead Hospital - lArestmead NSIAr

2145 AUSTRALIA

OBJECTIVE: To inveshgate the relationship between fetal metabolic acidaemm and changes in the PR waveform of the fetal

elec’a’ocar diograrr~

STUDY DESIGN: Ten fetal lambs, gestation 125-135 days, were studied in a chromc preparation. At laparotomy the carotid artery

and jugular vein of the fetuses were cannulated and ECG electrodes placed on each upper forehmb and the central chest and the scalp. The cannulae and electrodes were extefiorised to the flank of the

ewe and the uterus closed. A balloon catheter was placed around the maternal aorta that could reversibly o(:dude the vessel, the degree of ocduston being measured by a carmular placed in the maternal femoral artery. Maternal condition was detected vm cannulae in the carotid artery and jugular veirt After 2 days maternal aortic occlusion over periods of 1 hour were used to develop metabolic aciderma m the fetus. Fetal pO~ pH, pCCh, hctate and catecholamine levels in the carotid artery were correlated

with PR interval changes in the fetal ECG waveform which was molated using a PC-based time coherent enhanced averaging.

RESULTS: Metabohc andemia in the fetus was assocmted with s~gmflcant shortening o~ the PR interv~ (p <0.001), and P wave

duration (p <0.02) and a rise in P wave height (p <0.01). The normal negative correlation between the PR interval and heart rate changed

to a positive correlation when lactate levels increased by rr~re than 5 retool/l/30 minutes in the fetus. T wave changes were found to correlate with c~techolamine levels and not metabolic aoderma.

CONCLLnaIONS~ PR waveform changes in the fetal sheep ECG correlate strongty with the development of metabolic acideaemia. The ECG waveform ~s therefore

54 THE ROLE OF THE PULMONARY CIRCULATION IN THE DISTRIBUTION OF HUMAN FETAL CARDIAC OUTPUT. L

Rasanenx, D. C. Wood,x A. Ludomirski*, J. C. Huhtax. Dept. Ob/Gyn,

Pennsylvania and * Temple Univ. Hospital, Philadelpkia, PA. OBJECTIVE: To detemine the distribution of the left (LVCO) and right (RVCO) ventficular cardiac outputs and the changes during the second half of the pregnancy. STUDY DESIGN: Sixty-three normal fetuses, aged 19 to 39 weeks gestation (median 28 weeks) were examined cross-sectionally by Doppler echocardiography. Heart rate (HR), vessel diameters and time-velocity- integrals (TVI) at the aortic (AV) and pulmonary (PV) valve annuli, fight (RPA) and left (LPA) branch pulmonary arteries and ductus arteriosas (DA) were measured and blood flows (Q) were calculated (TVI x area x HR). Total pulmonary blood flow (Qp) equals QRPA+QLPA. The blood flow across foramen ovale (QFO) was calculated by subtracting Qp from LVCO.

RESUL~S: The correlation between RVCO calculate6 ~rom QPV and RVCO calculated from QP+QDA was good (r=0.97; p<0.0001). At 38w of

gestation the proportion of RVCO (60%) was significantly higher than that of LVCO (40%) of the total combined cardiac output (CCO) (p<0.05) The table shows the distribution of fetal cardiac output expressed as a percentage of CCO at three gestational ages (compared to the 20w group; *p<0.05, ¯ *p<0.006). CCO is expressed as meanS:SD.

LVCO RVCO QDA QP QFO CCO(ml/min)

20w 47% 53% 40% 13% 34% 237+67 30w 43% 5~% 32%* 25%** 18%** 932_+225** 38w 40% 60% 39% 21%* 19%** 1681+225"* CONCLUSIONS: In the human fetus, the second half of pregnancy is characterized by increased pulmonary blood flow and its proportion of CCO is higher than suggested in animal studies. The fight ventdcular donainanee persists while both fetal pulmonary blood flow increases and the proportion of FO blood flow decreases.

324 SPO Abstracts January 1996 Am.] Obstet Gynecol

55 VASCULAR ENDOTHELIAL GROWTH FACTOR EXPRESSION IN THE

CHRONICALLY ANEMIC OVINE FETAL HEART. C. Mafdnx, A. Hohimerx,

L Davis Dept Ob/Gyn, Oregon Heslth Sciences University, Portland, Or Objective: In chronically anemic fetal sheep, mycoard~al blood flow

iP~raaeea six fold and coronary f~w reserve is maintained (Ma~n, SGI 19~5, P240). These adaptations are thought to be due in part to new

veeeel growth of myocardial resistance vessels. Vascular Endothelial Growth Factor (VEGF) is a unique growth factor known to augment

collateral coronary vascular growth (Banal, Cir 89:2183,1994). We sought

to determine if the concentration of VEGF protein in chronically anemic fotal ventricle is increased as compared to control heart tissue.

Study Design: Chronically catheterized fetal sheep wore made anemic by

daily isovolemic hemorrhage for 5-7 days. Six chronically anemic fetuses at

a mean EGA (+ SEM) of 132 3 :t: 1.7 days were studied and compared to

s~x age matched non-anemic controls (EGA 130 5 + 1.7 days) Venthcular

tissue was obtained from rapidly frozon heart specimens from each animal. Protein extraction and iysatee were quantified for total protein concentration

via the modified Lowry method. Protaln samples were then analyzed in non- reducing cond~ons by PAGE Western Blot analysis, loading equal amounts of total protein Into oach well, and a known concentration of human

recombinant VEGF standard (R&D Systems) on each gel. Genentech anti- Rh VEGF primary antibody was used for the protein identification followed

by development with an ECL detact~on system. Densitometric quantification

of the anemic and control VEGF concentrations were compared by t-test. Results: When compared to controls, the anemic fetuses had a lower

hematocrit (37 3 + 2 2% vs. 13 9 + 0 6%, p< 01), and carotid arterial

oxygen contont (7.1 + 2.4 ml/dl vs. 1,9 + 0.3 ml/dl, p<.01). The arterial pH

of control and anemic fetuses was not different (7.34:1:0.02 va. 7.33:1: 0,03). VEGF concentrations by quantitative dens~tometry wore 0,386 + 0.08

in controls and 2,172:1:0,47 in anemic hearts (p=0.004), expressed as ng

of VEGF/30~g total loaded protein

Conclusion’ Fetal vantricular VEGF concentrations were elevated noady

six fold in anemic fetal sheep hearts as compared to controls This increase

in VEGF concentration hkely represents a response to chronic hypoxia and

impl~catas VEGF as a mediator of coronary vascular growth dunng chronic

fetal anemia Supported by NHLBI grant HL45043.

56 UTERINE MYOMETRIUM CONTAINS A SECOND GAP JUNCTION SEQUENCE (CONNEXIN4$) WITH DIFFERING EXPRESSION LEVELS FROM CONNEXIN43. J.Albrech[, N. Atalz, A.Orsinoz, S.Lye~, Y.Sadovsky, E.C.Bcyerx. Dcpts. Ob/Gyn, Peds. Washington Univ., St. Louis, MO. and Mr. Sinai Hosp, Toronto, CANADA OBJECTIVE: Studies were performed to assess the expression of a recently characterized gap junction sequence, coanexin45 (Cx45), in mt myometrium and myometrial cell fines. Gap junction channels facilitate the electrical coupling between uterine myocytes. Only one connexin (Cx43) has previously been identified in this tissue. STUDY DESIGN: Myomelrium from pregnant rats was analyzed by RNA blotting. Syrian hamster myocytes (SHM) and human UT1 myometrlal cell lines were analyzed by transfer of microinjected Lucifer yellow, RNA blotting, and immunoflnorescence. RESULTS: Cx45 mRNA (2.2 kb) was detected at all gestationai ages examined. Cx45 mRNA levels remained relatively constant throughout gestation (days 15-23). Cx43 mRNA increased dramatically at term, whereas Cx45 did not. SHM and UT1 cells showed fanctionai gap junctional coupling. Both cell lines contained Cx43 and Cx45 mRNAs. Immunofluorescence showed similar distributions of Cx43 and Cx45 at appositional surfaces between these cells. CONCLUSIONS: Co-expression of Cx45 with Cx43 in uterine myometrium may regulate gap junctional coupling between these cells. Cx45 may facilitate basal coupling betwen these ceils, while Cx43 provides inducible increased coupling at tenn. SHM and UT1 cells may offer appropriate systems for the in vitro analysis of the myometrial gap junction proteins.

Oral Concurrent Session E Clinical Obstetrics

Saturday, February 10, 1996 8:00 a.m.- 10:30 a.m.

Grand Ballroom

Moderator: Michael W. Varner, MD

Judges: Marshall W. Carpenter, MD Garland D. Anderson, MD Thomas R. Easterling, MD


326 SPO Abstracts JanuaiT 1996 Atrr J Obstet Gyne~ol

57 PERCUTANEOUS UMBILICAL-CORD LIGATION IN COMPLICATED MONOCHORIONIC MULTIPLE GESTATIONS. Ruben A. Ouintero Lms Gun~alvesX, Mark P Johnson, Harry ReichX, Roberto Romero, Carlos C~efiox and Mark I. Evans Department of Obstetrics and Gynecology, Wayne Stat~ University, Detroit, MI, Wyoming Valley Ob/Gyn Associates, Kmgstun, PA, and the Permatology Branch, NICHD, Bethesda, MD. OBJECTIVE: The management of monochononic multiple gestations with a non- viable fetus remains a challenge Since vascular communications are virtually present in all cases, KCI selective fetlclde cannot be performed, and other mtravasctdar methods axe unreLiable We report our experience with percutaneous umblhcal-cord hgatlon (PUCL) STUDY DESIGN: Fifteen patients with pre-viable monochorionic multiple gestations m which one ferns was considered non-viable were assessed PUCL was offered if the normal twin had evidence of heart failure, a normal kalyotype no obvious anomalies, or ff polyhydrammos or uterine contractions were present. PUCL was performed under combined endoscopic and sonographlc guidance using 2-3 mm custom-designed ports, and under general anesthesia Permperative intravenous tocolysis and prophylactic antibiotics were given RESULTS: Two patients were excluded for mlsdiagnosls The mean GA at the time of the procedure was 21 weeks (range 16-25) Eleven patients had an acardlac twin, 1 patient had a discordant twin with acrania, and 1 patient had twin-to-twin trans.fumon with a severely hydrop~c fetus In a quadruplet pregnancy w~th an acardlac twin, saleCtlVe fetal reduction of an additional fetus was performed with PUCL PUCL was successfully performed m 11/13 (84%) cases Amnioinfuslon or annuolThexls was required ill all acardiac twin pregnancies Endoscopy was the prmaary visualization technique in 6/11 successful cases. In cases comphcated by ~ntraamniotlc bleeding, and after endoscopic identification, PUCL was accomplished under sonograph~c guidance alone. The average time gained after successful hgatiun was 7 weeks (range 0-17), and 4111 patients delivered after 30 weeks. Two patlants are still undelivered Premature rupture of membranes (PROM) within three weeks of the procedure occurred in 4/13 (30%) cases, and one resealed. Two of these patients had preterm labor prior to the procedure. Intraoperatwe bleeding occurred m 5113 (38%) of cases, but dxd not preclude tying of the cord in 4, nor was it associated with PROM Of the 2 unsuccessful cases, the placenta covered the entire anterior uterine wall in one, and the cord was inaccessible m the other. CONCLUSIONS: PUCL is a reliable technique for the management of abnormal monochoriomc multiple gestations In contrast to intravascular techniques, interruption of the blood flow is prethctabIe and under control of the operator. Pre- operative preterm labor appears to be the most important risk factor for subsequent PROM Though acarthac twins are the most common indication, PUCL may also be used prior to spontaneous death of a twin to prevent neurolo lc and other complications in the survivor Technologic advances should simplify tg~e procedure and improve the neonatal outcomes

59 INCREASEO INCIDENCE OF PRETERM LABOR AND PRETERM DELIVERY ASSOCIATED WITH INCREASED SALIVARY ESTRIOL LEVEL Goodwin TM_ Jackson GM, McGregor JA, Lachelin GCL*, Artal R, Duilien V*. LAC+USC Women’s & Children’s Hosp, Los Angeles; Univ. Utah, Salt Lake City; Univ. Colorado, Denver; Univ. College, London England; SUNY, Syracuse NY; Biex, Inc, Boulder CO. OBJECTIVE: To determine whether an elevated sadivary estriol (Sad E3) level is associated with an increased likelihood of preterm labor and delivery (PTL and PTD) at less than 37 weeks. STUDY DESIGN: Double-blinded study of Sad E3 levels obtained from 190 asymptomatic women, both high-risk and low-risk for PTD. Samples were obtained every two weeks beginning ad 26 weeks. Sad E3 was assayed using a sensitive EUSA (Biex, Inc). The incidence of PTL and PTD between women with and without a single Sad E3 level ~ 2.1 ng/ml at any time during pregnancy was compared using Fisher’s Exact test. RESULTS: Hie_h-Risk Patients PTL and P’ID NoPTL. Term Delivery No. (%) Pts with any 10145 (22%) 35/45 (78%) Sad E3 > 2.1 ng/ml No. (°/0) Pts with all 5/89 (6%) 84/89 (94%) Sad E3 < 2.1 ng/ml (P.0.007) Low-Risk Pati0pts PTL and PTD N~PTL. Term Delivery No. (%) Pts with any 6/24 (25%) 18/24 (75%) Sad E3 > 2.1 ng/ml No. (%) Pts with al! 1/32 (3%) 31/32 (97%) Sad E3 <2.1 ng/ml (P-0.03) CONCLUSION: The incidence of PTL and PTD is markedly increased in patients with a Salivary Estriol level _> 2.1 ng/ml at any time after 26 weeks gestation. Salivary estriol may prove useful as a screening test for PTL and PTD risk in both high-risk and low-risk obstetric populations.

58 A PROSPECTIVE, RANDOMIZED, BLINDED COMPARISON OF TERBUTALINE VERSUS PLACEBO FOR S]~IGLETON, TERM EX’TERNAL

CEPHALIC VERSION

CO FemandezM D..S Bloom,IvLD ,G Wendel,M D ,Dept of OB/GYN

U T S W M S -Parkland Hospital, Dallas, Texas

Objective: To evaluate the effi~ncy of subcutaueous (SQ) terbutalme therapy on the success rate of extemal cephahc vermon (PC’V) m term, smgieton breech

presentations

Study Design: Term smgteton breech gestations were offered exlrollmeilt mtu this

!RB- approved prospeeOve study Palaents were assigtied to receive eithe~

terbutalme 0 25 arg SQ or an ldantieal placebo SQ. Providers were bhnded to

the type of drug ~tven Randomization was done by a random number table m

the hospital pharmacy Fifteen to 30 minutes after the study drug was g~ven, ECV

was attempted The ECV was stopped after 3 attempts,pa~aent dlseomfott, fetal heart rate deeelerataon or success of version Patients were chseharged home and

allowed to enter spontaneous labor Outcomes measured were matial success of

ECV, presentation at dohvory and route of dehvery Analysis was by Ch~-square

and odds ratio (OR) v, ath 95% eanfidenee intervals Results: One hundred three women were entered rote the study from 1/94 to

6/95 Overall 41 (40%) lmtial ECVs were sueaessful, and 62 (60%) faded Fd’cy-

two women received terbutahne and 27 (52%) of these ECVs were sucoessful

Filth’-one women received placebo and 14 (27%) ECVs were successful (p =

0 02) (OR 2 85, 95% CI 1 17-7 08) Four of the 27 (15%) successful ECVs, m the terbutalmo group, had spontenotts reversion to breech and 3 of the 14 (21%)

sueaessful ECVs m the placebo group spontenously reverted to breech

presentation Ulttmately, m labor there were 24 (46%) eephahe presentataons an

the terbutalme group and 13 (25%) eephahe presentations in the placebo group (p=0.0475) (OR 2 51, 95% CI 1 01-6 32) Eleven of the 41 (27%) successful

ECVs were dehvered by cesarean section and 58 of the 62 (94%) failed ECVs

were dehverod by cesarean section (p<0 0001) (OR 0 03, 95% CI 0 01-0 10)

Conclusion: Terbotahne 0 25 mg given subeutanenusly to women at term vath

breech presentations significantly increased the mmal success rate of ECV

Furthermore, the successful ECVs ulkmately increased the rate of eephab, e

presentations m labor and decreased the rate of cesarean dehvary

( Supported m part by a grant from the March of D~mes - Dallas Chapter)

6O RANDOMIZED DOUBLE-BLIND STUDY COMPARING SULINDAC TO TERBUTALINE:

FETAL CARDIOVASCULAR EFFECTS. W. KramgL G. 8aade, M. Belfort, K. Dorman’, M.

Mayes’, K. Moise, Jr. Dept OB/GYN, Baylor College of Medicine, Houston, TX.

OBJECTIVE: To compare the effects of sulindac with those of terbutaline on the fetal

ductas arteriosus (DA), middle cerebral artery (MCA) and ~mbilical artery (UA).

STUDY DESIGN: Twenty patients with preterm labor and no evidence of fetal structural

anomalies or intra-amniotic infection were randomized to receive either sulindac 200rag

orally every 12 hrs, or terbutaline 5mg orally every 4 hrs. Patients in both groups received

opaque tablets wah glucose base every 4 boors for 68 hrs and than continued with oral

terbutaline. Doppler pulsatdzty indices (PI) were obtained in the DA, MCA and UA at baseline

and 5,12, 24, 48 end 72 hrs after the medication was started. Doppler data were analyzed

over hme within each group using raw values and between groups usmg the change in PI

from baseline (API). A negative API =ndicates a decrease from hasaline. Statistical analysis:

repeated measures ANOVA, Mann-Wh=tney and Student.t tests as appropriate. Stat=stical

sigdificance: P < 0,05, Results presented as mean (SD),

RESULTS: There was no difference in gestati0nal age between the two groups [32.2(1.2)

vs 31.7(1.5) wks]. Suhndac was stopped in 2 patients (32.7 and 34.4 wks) after severe

ductal constriction was noted at 24 hrs. When analyzed over tzme within groups, the PI in

the DA decreased significantly at 12 and 24 hrs in the sulindac but not terbutaline group. No

significant difference was noted in the MCA or UA within either group over time. When the

change in PI was compared between greups, a s~ificent difference in the API of the DA

between suhndac and terb~taline was noted at 5,12, 24 and 48 hrs. A similar effect was

noted in the API of the MCA at 48 and 72 hrs. The UA’s ~PI did not differ significantly.

~ 5 hrs 12 hrs 24 hrs 48 hrs 72 hrs

Suhndac -0,26 (0.27) -0.41 (0.56) -0.63 (0.56) -0.3 (0.62) -0.09 (0.44)

Terbotahne 0.0710.3) 0.15 (0.48) 0.15 (0.26) 0.22 (0.32) 0.2 (0.28)

CONCLUSIONS: Compared to terbutaline, sulindac results in a decrease in the PI of the DA

and MCA. In the DA, the effect occurs within 5 hrs of receiving sulindac and resolves within

48 hrs of discontinuing therapy. The effect on the MCA occurs later and persists ]enger

after discontinuing therapy. Suhndac has minimal effect on the UA PI. These findings ate

consistent with constriction of the fetal ductus artenosus in patients receiving sulindac.

Volume 174, Number 1, Part 2 SPO Abstracts 327 Am J Obstet Gynecol

61 DOES THE USE OF A TOCOLYTIC AFFECT THE

SUCCESS RATE OF EXTERNAL CEPtlALIC VERSION?

G.P. Marquette, M. Boucher, D. Th~riaultx, D. Rinfret~. Dept.

Ob/Gyn, University of Montreal and Sainte-Justine Hospital,

Montreal, Canada.

OBJgCTIVE: To study the effect of ritodrine tocolysis on the

success rate of external cephalic version (ECV) at > 36 weeks.

STUDY DESIGN: This is a double blinded prospective study.

All patients were at _> 36 weeks gestation, confirmed by early

ultrasound (U/S). ECV assessment included NST pre and post-

ECV and U/S evaluation of: type of breech, estimated fetal

weight, position of placenta and fetal spine and amniotic fluid

index (AFI). Patients were excluded ff the breech was not mobile,

or ff they had any contra-indications to tocolysis or ECV. After

randomisation, 217 patients received either rito&rine (111 ~g/min)

or identical placebo by I.V. infusion for 20 minutes. Up to three

attempts at ECV under U/S surveillance were performed.

Statistical analysis was performed using analysis of variance, chi-

square and multiple regression where appropriate.

RESULTS: There was a higher success rate in the group

receiving ritodrine tocolysis (59% vs 42%, p = 0.03). There were

no differences between study groups in terms of maternal age,

BMI, gestational age, AFI, position of fetal spine and placental

location. Statistical analysis controlled for parity and type of

breech.

CONCLUSION: Ritedrine tocolysis improves the success rate of

ECV performed at ~ 36 weeks.

63 24 HOUR MOTHER-INFANT DISCHARGE WITH A FOLLOW-UP HOME HEALTH VISIT: RESULTS IN A SELECTED MEDICAID POPULATION. C. Brumfield, K. Nelson,× D. Stotzer,X P Patterson,× N. Sprayberry,x S. Shannon x Dept. of OB/GYN, Un~vers~ of Alabama at Birmingham, Birmingham, AL. OBJECTIVE: To determine safety and cost effectiveness of 24 hour discharge ~n selected mothers and newborns. STUDY DESIGN: Women delivering at University Hospital were screened to determine their eligibility for 24 hour discharge. Mothers were ehg~ble if they had no serious medical problems, had an uncomplicated vaginal dehvery and postpartum course, and were 12 hours postoperative follow~ng bilateral tubal iigation (BTL). Newborns were ehgible if they were _>37 weeks, weighed >2500 g and had a normal examination at 24 hrs. At 48 hours postdelivery, each mother and infant was examined ~n their home by a home health nurse. Phone consultations with a staff physician were noted and outcomes were entered into a data base linked to hospital financial data, RESULTS: Of 5170 deliveries from October 1, 1993 - June 30, 1995, 812 mothers (16%) and 707 (14%) newborns were discharged home at 24 hours. Eight hundred of 812 mothers (98.5%) had a normal exam at the home visit. Twelve of 812 mothers (1.5%) had problems that required obstetncian phone consultation but only 3 required a physician visit of which 1 mother was readm~tted for tteatmsc~t of an infected BTL incision Six hundred fifty-three of 707 (92%) newborns had a normal exam Fifty-fore newborns (7.6%) had problems that required pediatrician phone consultation primanly for jaundice requiring blood for a bilirubin level (22), infant care questions (19), and a cardiac murmur (9) Seven ~nfants (1%) required a pediatric clinic visit No infant was rsadmitted to the hospital. Cost savings to the hospital for early discharge was $509,499 menus $79,940 cost for home visits = $434,559 net savings. CONCLUSIONS In a selected, low risk population mother-infant discharge at 24 hours postpartum with a home follow-up visit is safe an~l cost-effective.

62 THE USE OF A SURGICAL ASSIST DEVICE DESIGNED TO

REDUCE GLOVE PERFORATIONS: A RANDOMIZED

TRIAL. M.J. Trelssmanx, M.W. Bebbingtonx. Dept. of Ob/Gyn,

Univ. of British Columbia, Vancouver, B.C., Canada

OBJECTIVE: To test the effecttveness of a surgical assist device,

SuturemateTM, at protecting the surgeon from glove p.unctures during

postpartum vaginal repair.

STUDY DESIGN: FoBowing delivery, surgeons performing vaginal

repair were raadommed to use the surgical assist device, or to perform

the repair in the usual fashion. After the repair, gloves from all

members of the delivery team were collected and a standardtzed data

form was completed. The gloves were tested for perforations within 24

hours using the FDA approved hydrosufflation technique. Power

calculations determined that 250 glove sets in each arm would give a

power of 80%. Comparisons were made using chi-square statistics with

p<0.005 taken as statistically significant using a Bonferoai

adjusmaant for multiple comparisons.

RESULTS: A total of 476 pairs of gloves were evaluated. A

significant reduction in the overall glove perforation rate from 28.3%

in the control arm to 8.4% in the study arm occurred (p=0.0001). Rates

of perforation varied with level of expertise but fell in all groups using

the device. Family doctors had the highest perforation rate in the

control arm, 38% and a rate in the study arm of 6% (p=0.0001).

Obstetricians had a perforation rate of 29% in the control arm and 4%

in the study arm (p=0.0009). 76% of punctures were located in the first

two fingers and thumb of the nondomtnant hand of both right (n=418)

and left (n=26) handed operators. Punctures were recogniz~d in only

16% of submitted glove sets. The level of satisfaction with the device

was mixed but overall 50% of operators indicated that they were

satisfied or very satisfied with the device.

CONCLUSIONS: The rate of glove perforations during vaginal

repair is high. This surgical assist device significantly reduces the rate

of glove perforations during post-dehvery vaginal repair.

64 MISOPROSTOL AS A LABOR INDUCTION AGENT: A PILOT STUDY COMPARING EFFICACY, SAFETY, AND

COST P. Ma~tthayx, P. Ogburn, Jr., D. Harrisx, V. Sumanx*, K.

Ramin. Depts. Ob/Gyn and Biostatistics*, Mayo Medical Center,

Rochester, MN. OBJECTIVE: To compare the efficacy, safety, and cost of intravaginal misoprostol as a labor induction agent to a standard protocol using prostaglandin E2 (Prepidil) and intravenous oxytocin. STUDY DESIGN: A prospective, randomized clinical trial was done involving 38 gravidas with an indication for induction. Two patients randomized to Prepidil refused treatment. The principal investigator who performed admission and 12-hour Bishop’s scores was blinded to the induction agent used. Statistical analysis included Wilcoxon’s rank sum test and Fischer’s exact test. RESULTS: Seventeen women received misoprostol and 19 received Prepidil/oxytocin. The two groups did not differ significantly with respect to age, parity, gestational age, weight, height, race, reason for induction, or initial Bishop’s score. There was a significant difference in the median change of the Bishop’s score among those treated with misoprostol (4) and those of the control group (1) (p <.001). Fifteen (88%) of 17 receiving misoprostol delivered within 36 hours compared with 9 (47%) of 19 controls (p = .01). The time to complete dilatation for the 30 women who progressed in labor was significantly shorter for the misoprostol treated group (17 vs. 24 hrs.) (p = .01). While 26% of those women in the control arm underwent cesarean delivery compared with 18% in the misoprostol arm, this difference was not significant. The drug acquisition costs alone of Prepidil/oxytocin are almost 8 times the cost of Misoprostol ($76.66 vs. 82¢). The neonates who delivered within 36 hours did not differ significantly in regards to fetal heart rate abnormalities, Apgar scores, or short-term morbidity. CONCLUSION: Misoprostol is an effective and economical cervical ripening and labor inducing agent.

328 SPO Abstracts .]anua~y 1996 Am J Obstet Gynecol

65 NEW TECHNOLOGY IN 3-DIMENSIONAL OBSTETRICAL ULTRASONOGRAPHY: TECHNIQUE, ADVANTAGES AND LIMITATIONS. Ludomirski A, Uerpairojkit Bx, Whiteman VE, Reece EA, Chu GP~, Chart L. Department of OB/GYN/RS, Temple University School of Medicine, Philadelphia, PA OBJECTIVE: To evaluate the role of new technology of 3-D ultrasound in perinatal medicine. STUDY DESIGN: 92 Patients undarwent 107 Three Dimensional Ultrasound examinations between 11 and 40 weeks of gestation. 3.5/5 MHZ abdominal and 5/7.5 MHZ vaginal transducers were used. Three perpendicular planes are displayed simultaneously and can be rotated and evaluated in order to achieve reliable views required for diagnosis and volumetric measurements. Surface rendering provides "Plastic" images of a given organ and the transparent mode is used for structure with a higher echogenicity. Fetal organ volume (spleen, cerebellum) was calculated using the complete three dimensional data set. The technology will be illustrated by video. RESULTS: The major advantage of using surface rendering and transparent mode three dimensional ultrasound was found in evaluating fetal spine and face. 21 optimal 3-D facial views out of 24 examinations were achieved; 17 had normal face, 4 abnormal (anterior cephalocele, anencephalus, cleft lip palate). 25 examinations of fetal spine (gest age 15-24 weeks) of pregnancies with elevated AFP (more than 2.5 MOM) revealed 4 abnormalities (spina Bifida, meningomyelocele). The 3D technology enabled measurement of fetal organ dimensions and volumes during gestation. This new modality provides a clearer view of complex anatomical malformations. CONCLUSION: The new todmi~ue of three dimensional ultrasound allows detailed plastic unages of fetal anatomy, calculation of organ volume and evaluation of tissue density. Clinical applications of the new technology are important and advantageous in perinatal medicine.

66 THE INFLUENCE OF ULTRASOUND EXAMINATION AT TERM ON THE RATE OF CESAREAN SECTION. J.A Adasbek~ D.C Lagrew, B.K. Iriye, MM. Can’, M Porto, R.K Freeman Dept. Of OB/GYN, Univ. of Calif., Irvine, Orange, CA, Long Beach Memorial Women’s Hospital, Long Beach, CA, and Saddleback Memorial Women’s Hospital, Laguna Htlls, CA. OBJECTIVE: To determine the influence of an ultrasound examination obtained the last 4 weeks ofpreganncy on physictan behavior with respect to cesarean section rate (CSR) m relation to ultrasound estimated fetal weight and actual birth weight STUDY DESIGN: Th~s is a retrospective ease control study. All term patients, at three hospitals, over a 5 year period, with cephalic presentation who had a sonographic estimated fetal weight (EFW) ~ 3500 grams in the last four weeks of pregnancy comprised the study group (N=735) The control group comprised a group of patients dehvering term neonates 2 3500 grams who d~d not have a sonogram in the last 4 weeks of pregnancy (N=4617). RESULTS: 1. The scanned study group had a CSR of 55.2% (406/735) eompared to the non-seanned control group with a CSR of 18.6% (859/4617), p<0.0001, OR=5.05, 95% CI=4.31-5.91. 2 If the birth weight (BW) was 2 4000 grams in the scanned study group, the C/S rate was 62.1% (269/433), however, in this group, ffthe EFW was 24000 grams the C/S rate was 74.6% (206/276), and if the EFW was <4000 grams the C/S rate was 40.1% [(63/157), p<0.0001, OR=4 39, 95% CI=2.83-6.83] despite similar median birth weights in the 2 groups (4 38 vs 4.24 kg). 3. If the BW was <4000 grams in the scanned study group, the CSR was 44 4% (180/405), however, m this group, iftbe EFW was 24000 grams the CSR was 58.9% (56/95), and iftbe EFW was <4000 grams the CSR was 40 0% (124/310), p=0.001, OR=2.15, 95% CI=I 32-3.53 despite simdar median birth weights in the two groups (3.80 vs. 3.68 kg). 4. If the sonogram predicted a <4000 gram neonate, the CSR was not d~fferent between the 24000 gram BW and <4000 gram BW neonates [40 1% (63/157) vs 40.0% (124/310), p=0.98]. Interestingly, the median birth weights d~ffered by 560 grams (3 68 vs 4 24 kg) (p<o oool). CONCLUSION: These findings indicate that a sonogram obtained in the last4 weeks of gestation is associated with an mcreasod CS1L and the EFW by sonography appears to influence the CSR as much or more than the actual birth wetght

Oral Concurrent Session F Medical Complications

of Pregnancy

Saturday, February 10, 1996 8:00 a.m.- 10:30 a.m.

Kona Ballroom

Moderator: David B. Cotton, MD

Judges: Steve N. Caritis, MD Sharon L. Dooley, MD Maurice L. Druzin, MD



67 ELEVATED AMNIOTIC FLUID INTERLEUKIN-6 PREDICTS NEONATAL PERIVENTRICULAR LEUKOMALACIA AND

INTRAVENTR1CULAR HEMORRHAGE. R. Fi~ueroa~ E Martinez×, P Sehgalx, D Garryx, K. Patelx, U. Verma, P. Vis~ntainer~, M. Realex, S. Kleinx, N. Tejani. New York Medical College, Valhalla, NY OBJECTIVE: Our purpose was to (1) investigate the relationship between

elevated levels of ammohc fluid(AF) interleukin-6(IL-6) and the development of periventricular leukomalacia(PVL) and intraventncular hemorrhage(IVH) in the preterm neonate, and (2) compare the value of AFIL-6 to amniotic fluid culture(AFC) and histologic chorioamnionitis (HCA) in the predictmn of PVL and IVH. STUDY DESIGN: Fifty-eight women in preterm labor w~th intact membranes underwent transabdominal amniocentes~s. AF was cultured for aerobic and anaerobic bacteria and Mycoplasma species; IL-6 levels were determined by ELISA. The placentae were examined for histopathologic

evidence of inflammation. Neonates were evaluated with transfontanelle cranial sonography for diagnosis of PVL and IVH on days 3 & 7. Student’s Ftest, chi-square, ROC curve and logistic regression were used for analysis. RESULTS: Women with neonates who developed PVL and IVH(n-9) bad higher AFIL-6 levels(p=0.002) than women(n=49) who delivered neonates without PVL or IVH They also were at a lower gestational age(GA) at

admission(p=0 002), had a shorter median amniocentesis-to-delivery mtervai(p=0 03), a lower b~rth weight(p=0.00008), had more positive AFC (p=0.01), and more HCA(p-0.05). An AFIL-6 level _> 12,000 pg/ml had a sensitivlty(SENS) of 89% and specificity(SPEC) of 88% compared to a

SENS of 50% and SPEC of 92% for AFC, and a SENS of 56% and SPEC of 78% for HCA m the prediction of PVL and IVH. Regression analysis demonstrated that AFIL-6 was an independent risk factor for the development ofPVL and IVH(Odds Ratio 31.46, 95% CI 2 18 - 453 39; p=0.011) after controlhng for GA, AFC, and HCA. CONCLUSIONS: AFIL-6 predicts neonatal PVL and IVH in women in preterm labor w~th intact membranes.

69 POSTNATAL GROWTH AND COGNITIVE DEVELOPMENT IN THE VERY

LOW BIRTH WEIGHT SMALL FOR GESTATIONAL AGE INFANT

J M Gdlesx, M Y Divon, C C Mc Cartonx, I Wallacex Departments of O~/GYN and Pedlatncs, Albert Einstein College of Medicine, Bronx, NY

OBJECTIVE: To detsrmme whether morphometnc measurements at birth and

postnatal growth are predictive of cognffJve development In the preterm growth retarded infant

STUDY DESIGN: 695 very low blrLh weight infants ( BW < 1500gins) were

studied between 1978 and 1993 Head circumference, birth weight and height were measured Infants were classn~ed as e~ther Appropriate for gestatlonal age

(AGA) or Small for gestatJonal age ( SGA, BW < 10t~ pereentde for gestational

age) The SGA infants were further subdivided into Symmetric or Asymmetric SGA (sSGA and aSGA, respectively) All infants underwent serial evalua~ons

of cognltwe development by Mental development Index (MDI) or by Stanford Binet Intelhgent Quotient (IQ) Subsequent growth in the SGA infants was

categonzed by weight percentiles at 6, 12, 24 and at 36 months of age as: no change (<10=h, group I), mm~rnal increase 0e, 25=h > weight > 10=h, group II),

sizeable increase ( ~e ,weight > 25~h, group Ill) Data were stratified for

gestatlonal age at delivery as group A (24 to 29 weeks), group B (30 to 32 weeks) and group C (33 to 35 weeks) Analysis of Vananon and Chl Square

testing were used for statistical analys~s

RESULTS Wdhn groups B and C, on the first evaluation, AGA infants scored

higher than those who were sSGA on MD] tes*Jng (p < 0 01) AGA =nfants whose weights exceeded the 25th percentile at 12 month scored significantly

h~gher than SGA infants of comparable weight on MDI at 12 month (p = 0 04)

By 36 months postnatally there were no significant d~fferences among AGA, sSGA and aSGA ~nfants on IQ Relative to other SGA infants, group III infants

scored s~gnlficantly h~gher on IQ at 36 months (p <0 05)

CONCLUSIONS Initially AGA infants had a cognitive developmental

advantage over SGA infants By pre-school age, this advantage was lost

SGA infants who demonstrated sizeable postnatal growth scored slgmficantly higher on IQ Th~s implies that s~zeable postnatal growth ~mproves cogmbve development m growth restncted neonates

68 AMNIOTIC FLUID CONCENTRATIONS OF INTERLEUKIN-6 IDENTIFY FETUSES AT RISK FOR THE DEVELOPMENT OF PERIVENTRICULAR LEUKOMALACIA. B.H. Yoon, R. Romero, J.K. Junx, K.H. Parkx, S H Yang~, I.O. Kimx, H Mnnoz~, J.H. Chot~. Seoul National University, Seeul, Korea, the Permatology Research Branch, NICHD, Bethesda, MD and Wayne State University/Hutzel Hospital, Detroit, MI. OBJECTIVE: Penventricular leukomalacia (PVL), a common brain white matter lesion of preterm neonates, is a tnajor risk factor for cerebral palsy. Recently, cytokines have been implicated as mediators for neuronal injury in PVL. The purpose of this study was to determine whether anmiotic fluid concentrations of cytokines are of value in the identification of infants at risk for PVL. STUDY DESIGN: The relationship between amniotic fluid concentrations of mterleukin~i (IL-6), interleukin-ll~ (IL-11~), tumor necrosis factor-,~ (TNF-a) and interleukin-1 receptor antagonist (IL-lra) and the occurrence of PVL lesions was examined in 118 patients who delivered preterm neonates within 72 hours after amnioeontesis. Amniotic fluid was cultured for aerobic and anaerobic bacteria and Myeoplasmas. Cytokines were measured by ELISA. Multiple logistic regression and reeaiver-operating characteristic (ROC) curve were used for analysis. RESULTS: 1) PVL lesions were present in 25% (30/118) of newborns; 2) ROC curve analysis demonstrated that an elevated concentration of ammotie fluid IL-6, IL-11L and TNF-a but not of IL-lra were significantly asscoiated with the occurrence of PVL (p < 0.05 for each); 3) An IL-6 ~ 6.5 ng/ml had the highest sensitivity (79%) and highest specificity (78%) in the ~dentification of PVL of all tests performed; 4) Multivariate analysis indicated that anatgottc fluid IL-6 was an independent risk factor for PVL (odds ratio: 7.2, p < 0 02) after adjustment of other confounding "¢ariables (i.e , gestat~on_al age at birth, amniotie fluid culture, umbilical artery pH, etc.); 5) Moreover, IL-11~ and TNF-~ did not add significant information to that already provided by IL-6 (p > 0.1). CONCLUSIONS: 1) Amniotic fluid IL-6 ~s the best antenatal predictor of PVL; 2) Our data provide strong support for the view that PVL ~s the result of cytokine-mediated neuronal injury.

7O A RANDOMIZED CONTROLLED TRIAL OF PREDNISONE AND ASA IN WOMEN WITH AUTOANTIBODIES AND UNEXPLAINED RECURRENT

FETAL LOSS. C. Lask~n~, C. Bombardmr=, F. MandeP, K. Ritchle, M. ~annah, D. Fanna. Univ. Of Toronto, Toronto, ON, Canada. V. Farewell’. Univ. Of Watadoo, Waterloo, ON, Canada K. Spitzer=, L. Fielding, M. Yeungx. Univ. Of Toronto, Toronto, ON, Canada.

OBJECTIVE: The hypothesis under study is that prednisone and aspirin therapy will incra~se the dive birth rate in women wi~ autoanbbodle~ and unexplained recurrent fetal loss (U-RFL).

STUDY DESIGN: 789 women wd~ U-RFL (.~.2 fetal losses) were ecreer~d for the presence of autoantJbodiss including ant~nucfaar anbbodiss, an~-

DNA, anti-lymphocyte, anb-cardiohpin (aCL), and a lupus anticoagulant

(LA). From this, 202 pregnant women with at least one autoantibody, were randomized Into a double-bhnd, controlled trial comparing therapy with

prednisona (0.5-0,8 mg/kg) and asplnn (100 rag) (n=101) to placebo

(n=101) The primary outcome measure was a mul’0var~ats logistic regression analysis relating the probability of a successful pregnancy to the treatment effect, the pstJent~s age, and a previous history of sedy or late

fetal loss. The study had 80% power to detect a 33% ~mprovemont in outcome. Continuous variables were analyzed by 2-taded t-tests o~ �hi-

square as appropdam.

RESULTS: In women with U-RFL, 49.4% were found to have at least one autoantibody. Live births in the treatment group was 65% which was not

significantly different than 54.6% found In the control group (p--0.14).

Specific analysis of the groups w~th aCL and/or LA showed no benefit of

the t~eatment (T) over the control (C) group (47.8% vs 52 4% live births, p=0.50). Maternal side effects included an increase In hypertension

l T=12.9% ve C=4%; p~0.02) and diabetes melldus (T=14.9% vs C=5%; <0.02). Prematurity was significantly more frequent in the treatment group

than the control group, 58.5% vs 9.4% respectr~aly (p<0.003). CONCLUSIONS: In women with U-RFL, autoantibod~se are commonly

detected. Treatment of these women when pregnant, with predfilSOne and

aspirin, was no more effectJve than placebo in promoting live birth. Furthermore, th~s form of therapy is associated wdh significant maternal

and neonatal adverse effects.

Volume 174, Number 1, Part 2 SPO Abstracts 331 Am ] Ob’,tet (,ynecol

71 ELECTIVE DELIVERY FOR MACROSOMIA IN THE DIABETIC

PREGNANCY: A CLINICAL COST-BENEFIT ANALYSIS. D ~;onwav. O Langer, Dept. Ob-Gyn, UTHSC, San Autumn, TX.

OBJECTIVE: To test the hypothesis that elechve dehvery for ultrasound

diagnosis of macrosom~a m d~abehc pregnancies slgmficantly decreases the

rate of shoulder dystoc=a, without a s~gnificant increase m maternal

morbidity from cesarean sect=on or reduction

STUDY DESIGN: In a prospective study, we evaluated a protocol under

which d=abehc women with ultrasound EFW 242509 underwent elechve cesarean section; and those with EFW consistent with LGA, but <4250g,

underwent mducbon of labor Women were stratified into four groups ~n accordance w~th the degree of agreement between the EFW and the actual

b~rthwe~ght (eg EFW = macrosomm and actual b~rthweight = macrosomm)

Maternal and neonatal outcomes were analyzed and compared for the

benod pnor to and after mR~at~on of th~s study protocol Outcome vanables

included cesarean sect=on, shoulder dystocia, infechous morbidity, fetal

morbidity, length of stay and overall cost

RESULTS: 4011 dmbet=c patients were included ~n this prospective study,

wRh an overall rate of macrosomm of 8.5% The overall rata of shoulder dystocm for the benod pnor to m~hahon of the protocol (1990-1992) was 3-

fold higher than the period (1993-1995) following ~nst~tut~on of the protocol

(2.20% vs 0 73%, p<0 05) resulting in a 67% reduction ~n shoulder dystocm

rate Moreover, the overall C/S rate ~ncreased by only 1 6% (11 3% to

12.9% respechvely) No shoulder dystocia occurred in patients ~nduced for LGA or macrosom~a based on our protocol Only 1% of our d~abebc

population required ~nterventton for ultrasound diagnosis of LGA or

macrosom[a Of the patients who were dmgnosed as macrosom~c (EFW),

45% had BW <4250g However, only 15.4% of C/S were done for the

~nd=cation of EFW >4250g In contrast, only 4 3% of pabents were actually

macrosom~c when EFW was <4000g ~,

CONCLUSIONS: The use of a weight threshold for elective dehvery m

diabetic women wdl reduce the rate of shoulder dystocla without a

s~gmficant ~ncrease ~n C/S rate Th~s practice ~n conjunction w~th intensified

management approach m dmbetas wdl improve the outcome of these h~gh

risk women and their infants

73 BACTERIAL VAGINOSlS (BV) IN A POPULATION OF 3600 PREGNANT WOMEN AND RELATIONS TO PRETERM BIRTH EVALUATED FROM THE FIRST ANTENATAL VISIT. P. Thorsen K Molsted, I P. Jensen , M Arp~, A. Bremmelgaard, B. Jeune, B R. Mailer. Department of Obstetrics and Gynecology, Umvers~ Hospital, Odense, Denmark. OBJECTIVE: The aims of this study were to determine the effects of common urogendal tract infections and microbiological conditions among women m eady pregnancy in regard to preterm labor, preterm premature rupture of membranes (PPROM), preterm birth (PTB) in a large prospectively followed cohort of Danish women. STUDY DESIGN: Three thousand and s~x hundred pregnant women were asked to participate m the study before 24 gestational weeks Samples were collected from the cervical us or vaginal vault for Chlamydia trachomabs, gemtal mycoplasmas, TrJcho;nonas vagina/is, yeasts and aerobe and anaerobic bacteria Amsels’ cnteria (1983) were used for diagnosis of BV RESULTS: Mean gestabonal age at enrollment was 17+1 (seventeen full gestahonal weeks plus one day), [range 7+3 - 24+0] We found BV in the genital tract m 13 7% (402/2927) of the pregnant women. We analyzed data of women entenng labor spontaneously and/or w~th rupture of membranes who did not have evidence of severe congemtal malformations ~n prewous pregnancy, placenta prev[a, cerwcal cerclage, serious medical d~sease, preeclampsJa, multiple pregnancy, iso~mmunizabon or abrupt=o placentae. Women with and without BV prior to 24 weeks of gestation had a similar frequency of PPROM (before 37 full gestational weeks) [2 2% vs 2.1%; crude OR 1 0, 95% CI 0 4 - 2.3] and low birth weight (LBW) babies (< 2500 g) [3 6% vs 2 8%, crude OR 1 3, 95% CI 0 7 - 2 4] Rates of idiopathic PTB were similar among women with or without BV [3 4% vs 3 6%, OR 0 9, 95% CI 0 5 - 1.8] CONCLUSIONS: This prospective analysis of a stable Danish population shows low rates of PPROM, LBW, PTB and BV There were no drfferences for PPROM and LBW among pregnant women wtth or without BV prior to 24 full weeks of gestation

72 A RE-EVALUATION OF THE INDICATIONS FOR PULMONARY ARTERY CATHETERS IN

OBSTETRICS: THE ROLE OF 2D ECHOCARDIOGRAPHY AND DOPPLER ULTRASOUND.

M. Be{fort. A. M~es,x G. Saade, T. Wen, R. Bekeyx. Oepts. ObIGyn, Anesthesinlogy,

Medicine, Beylor College of Med., Houston, Tx, and Marshfield Climc, Marshfield, WI.

OBJECTIVE: Frequent indications for pulmonary artery catheterization (PA) in obstetrics

are hypertension unresponsive to ant~hypertensives, oliguria unresponsive to repeated fluid

boluses, and pulmonary edema. We prospectively tested the safety and utility of 2D

echocardiography (ECHO) and Doppler ultrasound in the managenmnt of such patients.

STUDY DESIGN: Pat=ents fulfilling standard criteria fur PA were studied using a technique

previously validated at our ~nst~tutmn. Patients thought to require cout~ouous ~ltoring had

PA, otherwise management dec=s=ons were based ou the nen-invasive examination.

Indications included 4 patmnts with persistant oligurio, 2 with recalc=trent hypertension and

2 with pulmonary edema. LV fdling pressure (LVFP), central venous pressure, cardiac output

(CO), LV/RV d=mensians, ejection fraction, wall motem and chamber function were

measured, calculated and/or assessed at the bedside.

RESULTS: Only one pat=out required PA catheterization, and outcome was good =n all

cases. Of these w=th oliguria all 4 had hyperdynam=c LV function and low LVFP prior to

volume expansion. One received 7.51 and one 41 of crystallmd before urine output was

adequate. None of the 4 required PA and none developed pulmonary edema. Two patients

failed to respond to 2 doses of hydralazlne (lOmg). One had a high CO, normal LVFP, LV

hypertrophy and a hyperdynamic LV - she responded well to labetalol. The second patient

was shown to have LV hypertrophy and intravascular contract=on. She responded well to

volume expansmn and further vaseddatatem. One patient with pulmonary edema was

diagnosed with hypertens=ve cardlomyopathy based on ECHO and she had PA. The second

pahent had terbutahne tox~citywith pulmonary edema and was managed without PA.

CONCLUSIONS: 2D ECHO and Doppler ultrasound can be safely used to evaluate patients

who may otherwise require PA catheterization. In many instances invas~ve procedures can

be avoided by (1) knowing the CO, LVFP and the potential of the patemt to tolerate volume

expansion (2) determining the cardiac performance profile which may indicate whether a

vasoddator or a J] blocker is more appropriate for lowering the blood pressure, and (3)

d~agnosing whetherLV function is impaired and ~f so whether or not there is systolic or

diastolic dysfunction, and whether or not continuous PA monitoring is needed

74 WOMEN WITH SICKLE CELL TRAIT ARE AT INCREASED RISK FOR PREECLAMPSlA. K.D. Larrabee=, M.Monga. DepL Oh. Gyn. & Reprod. SoL UTMSH, Houston, TX. OBJECTIVE: To determine the rate of preeclampsia in women who are posi6ve for sickle cell trait. STUDY DESIGN: A~I African Amedcan women were tested forsickle cell trait using the sickledex screen at their first prenatal visit and prospectively enrolled in this study from March, 1994 - June, 1995. Demographic data were cotiscted at the time of enrotiment. Outcome data, Including preedampsia (as defined by ACOG criteria), gestatJonal age at delivery, btrthwsight and postpadum endometdtis were collected Immediately postpartum. Assuming a 10% rate of positive sickle cell trait, 1199 patients were required to derno~strate a 10% difference in the rate of preedampsia with 80% power and p<0.05. Student’s-t, Mann Whitney U, Cht-squara and F’ishefs Exact tests were used for statistical ana.lysis. RESULTS: 1584 women were enrolled; 162 were positive for el~la cell trail Sickle cell trait positive (SC+) women were older than sickle cell trait negative (SC-) women (24.4 ± 4.6 vs 23.0 ± 4.4 years, p<O.001), but there was no significant dfference in padly (I [0-4] vs I [0-11]). The rate of preeclampsla was significantly increased in SC+ women (24.5% vs 9.7%, p<O.0001 ). There was no significant rifle rence In the rate of chronic hypertension, diabetes or smoking. Parous SC+ women more frequently gave a history of preeclampsla In a previous pregnancy (21.4% vs 9.1%, p<0.0001). Gestational age at delivery and bir~wsig~t were significantly decreased In the SC+ women (36.7 ± 2.7 vs 37.7 ± 3.0 weeks p<O.0001) and (3082 ± 591 vs 3369 ± 573 grams, p<0.0001). The rate of postpartum endometritis was elgnificantJy increased in the SC+ women (14.1% vs 5.3%, p<O.001) aRhough both groups had a similar cesarean sec~on rate (14.2% vs 14.4%, NS). CONCLUSION: This is the first prospective study to demonstrate that sickle cell trai.~t positive women are at significantly higher risk for developing pednatal complications which have tmdiitonally been associated with sickle disease.

332 SPO Abstracts lanuar~ 1996 Am I Obstet Gynecol

75 THI~ EFFECT OF MATERNAL POSITION ON CARDIAC oUTPUT WITH EPIDURAL ANALGESIA IN LABOR. DP, Danilenko-Dixon, L Tefftx, B Haydonx, RA Cohenx, MW Carpenter.

Brown University, Dept.s Ob/Gyn & Anesthesia, Women & Infants Hospital, Providence, RI. OBJECTIVE: To test the hypothesis that supine vs. lateral position is associated with greater decrement in cardiac index after epidural analgesia (EA) in labor. sTUDY DESIGN: Nineteen normal, term subjects were randomized to left lateral or supine position in early labor. Cardiac index (CI, measured by acetylene rebreathing), stroke volume index (SVI), heart rate (HR), mean arterial pressure (MAP), and systemic vascular rasistanco (SVR) were obtained at 5 minute intervals, beginning prior to 500 ml 1V crystalloid (baseline), and ending 45 minutes after EA. RESULTS: Mean baseline supine vs. lateral group differences were significant for 21% lower CI, 21% lower SVI, 19% higher MAP, 50% higher SVR (all p<0.05), and equivalent HR. Fluid bolus in the supine group resulted in increments in CI and SVI (p<0.01), decrements in MAP and SVR (p<0.03), and unchanged HR. Relative to postfluid values, EA in the supine group resulted in decrements in CI and SVI (p<0.05). The lateral group exhibited no hemodynamic alterations following fluid bolus or EA. CONCLUSIONS: The supine

position is associated with ~°1~ in CI, likely reflecting decreased preload volume. To our knowledge, this is the first direct evidence of positional effects on maternal hemodynamic homeostasis after EA. ~ ....

76 The Effect of Dobutamine Hydrochloride on Ventricular Function and Oxygen Transport in Patients with S~ver~ Preeclampsia. C. C~aves, T Wbeder, N. Trotano. I)¢pt Ob/Gyn, Vanderbdt Umversity, Nashville, TN. OBJECTIVE: Depressed left ventricular function has been previously reported in patients with severe preeclampsia. In critically ill patients, improvement of cardiac function may lcad to increased oxygen delivery. This study reports "the effect of dobutamine hydrochlofide in a subset of patients with severe preeclampsia. Its use has not beon previously reported in this population STUDY DESIGN: Pregnant patients in the third trimester w~th severe preeclampsm who required pulmona~ arte~ catheter(PAC) monitonng for ehnieal management were evaluated for the study. Entry criteria included depressed left ventrieular function after optimtzahon of volume and no previous h~stoW ofcardmc disease. Duting the study period, 40 patients were diagnosed wRh severe preeclampsia and required PAC. 8 patients met criteria for dobutamine administral~on Dobutamine was initiated at 3 mcg/kg/mm, then fitraled to the L¥SWI. Values were compared pre and post infusion Variables were analyzed using the t-test. RESULTS: The outcome data is listed below : Mean Pulmonary Capillary Wedge Pressure (PCWP) 18+ 5.5 Mean Hemoglobin 10.2+ 3 2

Premfusion Post~nfus~on Cardme Index (L’min’~.mz) 3.19 4 28 p=0.05 Left Ventneular Stroke Work Index (g.m/m~) 37.58 59 38 p=0 03 Oxygen Dehvery Index (ml.min~’m~) 445.63 615.38 p=0 05 CONCLUSIONS: Dobutamine hydroehlofide can be used to improve oxygen delivery in paaents w~th severe preeclampsm who have depressed left ventrieular function.

Poster Session I

Thursday, February 8, 1996 10:00 a.m.- 12:00 noon

Kohala Ballroom

CATEGORIES Antepartum Testing

Clinical/Operative Obstetrics Placental Physiology

POSTER NUMBERS: 77-196

Judges: Lawrence D. Devoe, MD Sharon L. Dooley, MD

J. Peter Van Dorsten, MD


77 TRANSCUTANEOUS AND ARTERIAL CARBONDIOXI- DE TENSION DURING INTERMITTENT UMBILICAL CORD OCCLUSION IN LAMBS. M.Bergmans*, G.Stevens*, H.Keunen*, T.Hasaart*. Dept. Ob/Gyn, Univ.of Limburg, Maastdcht, The Netherlands. OBJECTIVE: To what extent transcutaneous measurements reflect the actual intravascular acid base balance is not well known. We studied therefore the relationship between the transcutaneously measured carbondioxide tension (tcPCO2) and the arterial carbondioxlde tension (PCO2). STUDY DESIGN: Five fetal lambs (GA 120 days) were provided with a femoral artery catheter, an inflatable umbilical cord occluder and a transcutaneous electrochemical probe fixed to the abdominal skin. Heart rate, blood pressure and tcPCO2 were continuously monitored. Arterial acid base balance was measured at regular intervals. After a stabilization period of one hour, intermittent total umbilical cord occlusions of two minutes with a three minute interval were performed (n= 12). RESULTS: During each occlusion mean PCO2 increased 1.6 kPa and mean pH decreased 0.07. Partial recovery occurred between occlusion intervals resulting in the gradual development of mild acidosis (drop in pH from 7.27 to 7.12 and PCO2 rise from 6.0 to 8.1). Pearson correlations between tcPCO2 and the trends in PCO2 and pH were > 0.97 (p<0- .0001). CONCLUSIONS: Changes in tcPCO2 induced by intermittent cord occlusion correlated well with changes in PCO2, but only when the latter was expressed as a trend.

79 MECONIUM DRUG SCREENING OF STILLBORN INFANTS: A

FEASIBILITY STUDY MR Lauria., F Qureshix, SM Jacquesx, P Kurtzhaisx,

B Gonik. Y Sorokin, Dept Ob/Gyn and Path, Wayne State University,

Detroit, MI.

OBJECTIVE: Meconium drug testing of liveborn infants is highly

sensitive (87%) and specific (100%). Accurate knowledge of drug use

in mothers of stillborns improves our understanding of factors

contributing to psrinatal mortahW and permits more effective patient

counsehng and treatment. We determined the feambility of routine

meconium drug screening for opiates and cocaine in stillborns.

STUDY DESIGN: Stillborn infants delivered at our hospital had

meconium collected using a 4 mm spatula ~nserted into the anus.

Specimens were placed in methanol and frozen at -70°C until analysis

by gas chromatography. Charts were reviewed for conditions leading

to stillbirth, history of maternal drug use and maternal unne drug screen

(UDS) results.

RESULTS: During the study period, 30 meconium samples were

obtained. Despite our best efforts, 26/30 were below the m=nimal

amount needed (0.Sg) as determined by previous chnical trials.

Regardless, all samples were analyzed. Mean birth weight and

gestational age were 1581 ± 1442 g and 28.6 ± 7.3 wks,

respectively. No specimens were positive for opmtes; three were

posit=ve for cocaine (10%). Chrscal data were available ~n 2 of these

latter 3 cases, both of which were initially determined to have

"insufficient quantity" of sample. In one, the presumptive cause of

fetal demise was diabetes, with no add=tional factors suggesting

substance abuse. The other fetal loss was due to idiopathic preterm

labor at 21.5 wks, with e positive UDS.

CONCLUSION: In this pilot study, inability to obtain an adequate

volume of meconlum occurred frequently. Although this may have

limited the sensitivity of the test. in some patients, important and

unexpected laboratory data were generated even with "insufficient

quantity". This highlights the need to develop more rehned

methodologies for this screening tool.

78 THE EFFICACY OF A RAPID LAMELLAR BODY COUNT ASSAY IN

PREDICTING FETAL LUNG MATURITY. C. Weisch°, J. Woods’, M. Yancey, A. Same. Dept. Ob/Gyn, Tripisr Army Medical Center, Honolulu, HI.

OBJ ECTIVE: To determine the optimal value of the larnallar body count for the prediction of fetal lung matunty using a rapid, standardized assay method,

and compare the accuracy of this test against an accepted standard of maturity.

STUDY DESIGN: Amniotic fluid specimens obtained by transabdominal

amniocentesis over a 39 month period were evaluated for both lecithin/

sphingomyelin (US) ratio and lameller body count (LBC). Receiver-operator

curve analysis was used to detarmme a LBC pred,’dye of maturity a~ compared to the accepted standard of tasting, the L/S ratio. All neonatal

records for infants delivered within 3 days of amniesentasis were reviewed for evidence of respiratory distress syndrome.

RESULTS: There were 849 visually uncontaminated amniotic fluid

specimens available for analysis dudng the study period. The L/S ratio was

performed on a hmited basis by trained technicians and required

approximately four hours for completion, while the LBC was determined using

readily avsilab~ standardized equipment in less than 15 rni~. A lernal~ar

body count of 35,000/mL was determined as predictive of mature US ratio by

receiver-operator curve analysis. A mature US ratio was present in 93.8% of

specimens with an LBC ~ 35,000, while mature L/S values were found in

11.3% of samples with an LBC < 35,000. There were 238 infants with complete neonatal documentation delivered within 3 days of amniocantsais,

and 20 of thesa (8 4%) infants were identified with respiratory distress syndrome. For pred=cbon of the absence of neonatal complications of

pulmonary immaturity, the sensitivities of isrnallar body count and US ratio

were 86.7% and 89.9%, and the positive predictive values were 97.9% and 98%, respectively, all non-statistically significant differences. CONCLUSIONS: We determined the lamelisr body count of 35,000/mL is s

reliable predictor of fetal lung maturity as compared to the accepted standard of matuntT, the L/S ratio. The Pesi~va predictive value for the absence of

resp=ratory distress syndrome is equal to that of the L/S ratio. Considering

assay ~me, availability, simplicity, and validity of the ismelisr body count, we

feel it is a super’~r test for the assessment of fetal lung maturity.

8O ANTENATAL PREDICTION OF FETALpH IN INTRAUTERINE GROWTH RESTRICTED FETUSES USING COMPUTER ANALYSIS OF THE FETAL HEART RATE. E G~zman, A Vmtz=leos, J Egan, C Benito~, M Lake~, Y Laix.

UMDNJ Robert Wood Johnson Medical School/St. Pete(s Med=cal Center, New Brunswick NJ and Univers=ty of CT Health Center, Farmmgton, CT OBJECTIVE: To test the accuracy of a mathematical mode based on computer analysis (CA) of the fetal heart rate (FHR) tradng, in predicting umb~hcal artery (UA) pH at b~rth STUDY DESIGN: The study =ncluded three phases. In phase I a group of 37 mtrautenne growth restncted (IUGR) fetuses with abnormal Doppler valoc~metry were delivered by cesarean se~on before the onset of labor. In all cases 1 hour FHR traongs were obtained w=th=n 1 5 - 6 hours of dehvery and were analyzed by the System 8000 computer software package. The software calculated the number of accelerations of 10 and 15 beats per minute, ap=sodes of h=gh and low variation (Lo Epi) in minutes and long and short term vanat,on =n m$iseconds. The UA pH was obtained at b=rth Stepw;se and second order polynomial regress=one were used to determine ft~ best pred=c~on model (formula 1) In phase II of the study, formula 1 was prospectively apphed to a group of 22 I,UGR fetuses managed in the same manner and ~ts accuracy was determined In phase III the data from all 59 IUGR cases were combined to generate a new formula (formula 2) the accuracy of wh=ch was retrospectively assessed against the entire group of 59 lUGR fetuses Nonacidem~a was defined as UA pH>7 25, preac~demia as 7.20-7 25 and academia as <7.20 Values are reposed as median (range) and stat~sttcal compansons were made using Wflcoxon signed rank and 2 x 2 contingency tables w~th significance set at p<0 05. RESULTS: Lo Epi alone was ~e best pred=ctor of UA pH dudng phase l..’r~, e denved formula 1 was UA pH = 7.28+0 002(Lo Ep0-9.21~ (Lo ep0~ . In me prospectively studied 21 IUGR fetuses, there was no stalist~cal difference between the pred=cted and actual UA pH at b~rth [7 28(7.18-7.37) versus 7 28(7.1 - 7.29), p=0 79]. The accuracy of formula 1 ~n correctly categonzing the prospec~ve cases was: ac=demia 100% (1/1), preac=dem=a 100% (1/1) and nonac=demia 80% (16/20) p=0 003) Formula 2 was UA pH = 7 277 + 0.003(Lo Ep0 - 9 8961e-~(Lo Epi)~ Phase Ill (formula 2) results are: Actual

Normal Preacidos~s A~id0s~s

Nonacidemla (n=40) 32(80%) 8 (20%) 0 Pred=cted Preac=demia (n=6) 0 6 (100%) 0

Acldemta (n=13) 0 5 (38%) 8(62%) p<.0001 The median (range) dtfference (~) between the predicted and actual UA pH was -0 001 (-0.12 to 0 116). The frequency distdbu~on of the ~ UA pH was: 0 (n=5), ~ 01 (n=l 6), ~ 02 (n=22), _-H) 03 (n=35), <_4-0 05 (n=46), and >+ 0.05 (n=13) CONCLUSION: A mathematical model based on CA denved FHR Lo Ep~ can rehab[y predict UA pH at birth. TNs type of nonlnvasive mon~tonng may allow for antepartum est]mat~on, and perhaps, continuous tracking of fetal pH.

336 SPO Abstracts .]anuary 1996 Am | Ob~tet

81 THE ACUTE EFFECTS OF CIGARETTE USE DURING PREGNANCY ON FETAL ACTIVITY. B. Brost, M. Josephx, S.

Stratumx, D. Eller. Dept Ob/Gyn, MUSC, Charleston, SC OBJECTIVE: To evaluate the acute effects of maternal cigarette smoking on fetal activity in an otherwise low risk pregnancy. STUDY DESIGN: Fetal activity (FA) was prospectively evaluated in 30 smoking and 30 nonsmoking patients at 30 to 40 weeks gestation seen in the low-risk obstetric clinic. Patients were evaluated in a sem~- recumbent position for FA using a noninvasive Doppler fetal activity monitor. Baseline fetal activity was obtained for ten minutes in both groups. Smokers were monitored for fetal activity while smoking one of the paUent’s cigazTettes over a 3-5 minute period and for 50 addmonal minutes after smoking. The nonsmoking patients continued FA monitoring for 30 additional minutes after the baseline assessment Fetal actiwty was calculated as movements per m~nute for each one- minute interval during the study RESULTS: Dunng the initial 10-minute assessment, the baseline FA was significantly greater (p<0.05) in the smokers (FA = 9.0) than the nonsmokers (FA = 4.4). While FA was generally noted to increase m both groups with rest in the semi-recumbent position, an immediate decrease in FA was evident within the first 5 minutes of cigarette use in the smoking patients. FA decreased significantly (p< 0.05) from an average of 9.0 to 7.4 movements per minute. FA subsequently increased m the smoking group returning to greater than 9.0 movements per minute over the next 5 minutes. The FA in the smoking group returned to the presmoking baseline of fetal movement by 50 minutes after cigarette use compared to a similar return to baseline within thirty minutes in the nonsmoking group. CONCLUSIONS: Cigarette use in pregnant patients is associated w~th an ~mmed~ate decrease m FA. Of interest, the basehne FA of smoking patients is double that exhibited m fetuses of nonsmokmg patients. Th~s long term st~mulatory effect of cigarettes may increase fetal metabolic demand contributing to the lower fetal birth weight found m smoking pregnant patients.

83 SUPPRESSION OF FETAL BIOPHYSICAL ACTIVITY AND

FALSE DIAGNOSIS OF ASPHYXIA FOLLOWING ANTENATAL

STEROID ADMINISTRATION. S. Rotmen~£h, M. Liberati, Z.

Sahavix, S. Levx, Z. Efratx, M. Kobox, I. Bar-Have, Z. Ben-Rafaelx .

Depts. Ob/Gyn, Golda Mad Ctr, Tel Aviv Univ., Israel and Chieti Univ.,

Italy

OBJECTIVE: To examine the effect of maternal betamethasone

administration on fetal heart rate tracings (FHR-T), sonographic

biophysical profile (BPP) parameters, and Doppler flow indices of

middle cerebral (MCA) and umbilical arteries (UA).

STUDY DESIGN: 25 Patients at risk for preterm delivery received twice

weekly doses of 12rag bethamethasone. Morning and afternoon FHR-T

were recorded daily for 30 minutes. FHR-T analysis was performed by 4

interpreters blinded to timing of steroid administration. BPP parameters

(n=22), and DFI (n=lg) were examined at 48 hour intervals. Fetal

circadian rhythffl effects were controlled for by precise timing of

examinations. Statistical analysis was performed with student t-test and x~ test where appropriate.

RESULTS: FHR variability (score 2.3 +.72 vs 1.8 +.67, p=0.017), number

of accelerations (4.6+4.3 vs 2.7+--3.5, p=0.028), and maximal acceleration

amplitude (21,7+9.5 vs 16.9+7.2, p=0.001) were reduced 48h after steroid

administration and recovered within 24h. Number of breathing episodes

(6.6+3.3 vs 1,0+1.3, p<0.001) and total breathing time (778+487 vs

80+155 seconds, p<0.0001) were the most profoundly depressed

parameters at 48h. Fetal limb and body movements (14.7 +6.9 vs 7.0+5.0;

11.6+_5.6 vs 6.3+5.3, p<0.002) werealso reduced at48h. AllBPP

parameters returned to baseline values within 96h. DFI of MCA and UA

were unchanged. Three preterm patients were emergently delivered due

to antenatal testing suggestive of fetal asphyxia, but umbilical cord

blood gases at birth were normal.

CONCLUSIONS: Antenatal steroid administration transiently reduces

multiple fetal biophysical parameters and can lead to the false diagnosis

of fetal asphyxia.

82 A PROSPECTIVE STUDY OF GESTATIONAL EFFECTS ON MATERNAL PERCEPTION OF FETAL MOVEMENT IN POSTTERM PREGNANCY. S. SchweikertX, K.M. Placquadio, T.R. Moore, Dimsion of Patina:el Medicine, Umv. of Calif., San Dingo & Naval Hospital San D~ego. Objective: To deternmne the vali&ty of the "count to ten" system of of fetal movement momtonng in post:arm pregnancy.

Study Design: A cohort study was performed comparing daily hines to perce*ve 10 fetal movements ("lack times") before and at least 7 days after the EDC ~n well-dated pregnancies, using each patient (n=49) as her own control. ANOVA and linear regression were used to determine the effect of gestahonal age, placental pos~bon, amniotic fired voIume and other factors on fetal

movement times. Power analysis showed that 50 patients would detect a lack hmes decrease of 30% wtth (z=.05, [8=.8.

Results: The mean lack time in pre-EDC periods (8’:!:3 SD) was

not significantly different than those postterm (8’+3, p=ns). However, ANOVA demonstrated lack hmes from 34 to 41 weeks of gestation (11’+2 to

7’+-3, p=.008, Figure). I~ck t~mes were slightly longer wxth an anterior placenta, (9’+_3 versus 8’+2), low AFI (<7 cm) (18’+-3 vs 7"+2) than with control values (p<,01).

Obesity and parity had no effect.

Conclusions: We conclude

a progressive decrease in mean

that mean kick times decrease as pregnancy progressess ~nto the postterm period, suggesting that a decrease in fetal movements in postterm period is a sirmlarly valid and concermng finding as zt is preterm.

84 AMNIOTIC FLUID INDEX: WHICH CUTOFF TO USE? ~Kramer, E. Wagner’, M. Belfort,

K. Moise, Jr., G. Saade. Dept of OB/GYN, Baylor College of Medicine, Houston, TX.

OBJECTIVE: To dete’~ne whether an ammotic fluid index |AFI) of 5cm or ~ess is more

accurate in predicting adverse perinatal outcome when compared to one based on the 5th

percentde value for gestation.

STUDY DESIGN: A retrospective chart review of 455 high-risk patients who underwent

indicated bJophysJcal profile testing (BPP) and delivered within 7 days of the test. The AFI in

cm and the percentile for the gestation were recorded as well as the indication for the BPP.

Receiver operating characteristics (ROC) curves were constructed for adverse perinatal

outcomes using absolute values for AFI and amnlotic fluid percentile values for gestation. An

adverse outcome was defined as cesarean section (C/S), C/S for fetal distress, neonatal

resuscztation immediately post delivery, Apgar of 7 or less at 5 minutes of life and perinatal

death. Data were also analyzed by Fisher’s exact test or chi-squared analysis. A two-tailed P

< 0.05 was used to denote statistical significance.

RESULTS: The absolute number in cm for AFI and AFI percentile for gestatienal age were

not significant predictors of C15 or neonatal resuscitation at birth. As compared to an AFI

~< 5cm, an AFI ef ~ 8cm or ~< the 5th percentile for gestation had greater sensitivity

(8% vs 72% and 64%) wzth minimal decrease in specificity (92% vs 64% and 67%) in

identifying patients at risk for

having a C/S far fetal distress

end/or an Apgar score of < 7 at

5 minutes (hgure). There were no o s- petinatal deaths zn the study

group. ~, o a- CONCLUSIONS: A low AFI is not ~ a good predictor of C/S or need =~ o 4-

for neonatal resuscitation at birth. ~o

AnAFIof < 8cmor _< 5th o2-

percentile for gestation are better

than AFI _< 5 cm at zdentifying a o o-

patient at risk for having a C/S for

fetal distress and/or an Apgar

score of < 7 at 5 mins.

P<O001 ~

~AAF -- AFI PercenUl

AFI

0’0 " 0’2 0’4 " 0’6 " 0’8 1’0 1 - Specificity

Vohime 174, Number l, Part 2 SPO Abstracts 337 Am J Obstet Gynecol

85 THE RECEIVER O~ERATIN~ CI~a#J~"~RISTIC CURVE FOR ~IOTIC FLUID INDEX AS A DIN~O~TIC TEST FOR OLIGOHYDRA~IOS IN POSTI)ATE~ PATIENT~ CHANTS AS A FUNCTION OF TIME Christopher O’Reilly-Grmen Dept. of OB/GYN, Albert Einstein College of Medicine, Bronx, NY OBJEC’FIVE To evaluate the effect of time dependent changes in sensltlvity and specl ficity on the receiver operating cheracterist~c curve (ROCC) analysis for AFI. SllJDY DE~I(~ 622 postdates pat~ente with rupture of m~mbranes (ROM) 4 days or less after AFI measurement and a cl~n~cian’e evaluation of the amount and quallty of fluid at ROM were stratified according to the day of ROM. We constructed a ROCC for AFI using absent or decreased fluid at ROM as the gold standard for oligohydramnios for each of these groups. RESULT~ 295 patients had ROM between 2 and 4 days after AFI wee measured. 204 patients had ROM I day after, and 123 patients had ROM on the day of AFI measurement. The maximal change ~n slope in the group w~th ROM the same day as AFI occurred at a cutoff of 5, while in the group with ROM I day later, the maximal change was at a cutoff of 8. In the group with ROM 2 to 4 days later, tho maximal change in slope occurred at a cutoff of 9 for AFI. (XI~CLU~IONS The optlmal cutoff for AFI based on ROCC analysls changes as a function of the time interval between measurement of AFI and ROM, due to a decline ~n sensitivity and a r~se in speoificity as a function of time at each cutoff value for AFI. These ohanges in turn are due to increasing false negatives and decreasing false poslt~ves relative to true positives and true negatives as a functlon of time. The changes in false positives and false negatives may be due to time dependent decreases in amniotic fluid that occur after measurement. Thus as time elapses since measurement, a g~ven AFI appears to b~comes a better predictor of oligohydramnios. Consequently, the appropriate group te study for ROCC analysis of AFI as a d~agnostic test for oligohydramnios would be patients with AROM immediately after AFI measurement. The sensitivity and spocificity of AFI in thls group are mere likely to give an accurate reflection of the abllity of the test to detect oligohydrammos at the t~me AFI is measured.

87 NON-STRESS TESTING WITH ACOUSTIC STIMULATION IN TWINS. "IF

Por~er, GA Dddy, GM Jackson, SL Clark. Dept. of OB/GYN, University

of Utah, BLC, UT.

OBJECTIVE: To validate the use of acoustic stimulation (AS) with non-

stress tests (NST) performed in the antepartum aurvedlance of twin

gestation.

STUDY DESIGN: A retrospective observational study of twin pregnancies

seen in our antepartum testing center between 1985 and 1993.

Surveillance consisted of an NST and amniotic fluid volume assessment.

AS was routinely employed if the fetal heart rate was non-reactive after

10 minutes. Follow-up testing for non-reactive NSTs with AS included

contraction stress tests ICST) and biophysical profiles (BPP). Records

were examined for NST results, whether AS was used, the result of AS, amn=otlc fluid index (AFI) measurements, and pregnancy outcome

RESULTS: 405 women were tested Twin "A’: 1107 (56%) NSTs were

Initiafly reactive without AS; 801(41%) received AS; of these, 713

(89%) became reactive; and 88 (11%) were persistently non-reactive

(4% of the total number). Follow-up testing was performed in 25 cases

using CSTs (all negative) and 107 BPPs (median = 8, range = 2-8)).

Twin "B’: 1210 (63%} NSTs were ~nitially reactive without AS; 653

(34%) received AS; of these, 89% became reactive, and 74 (11%) were persistently non-reactive (4% of the total). Follow-up testing was

performed in 19 cases using CSTs (one poslttve) and 123 BPPs (median

= 8, range = 2-8). The mean (+\-SD) time of testing to reach reactivity

for all twins was 14 (+\- 8) minutes. The mean (+\-SD) AFI for each twin pair was 16.8 ( +\- 4.7) cm. Two intrauterine deaths occurred, both

comphcated by severe twin-to-twin transfusion. The first occurred at 34

weeks’ gestation after a reactive NST w~thout AS 3 days earlier. The

second occurred at 32 week’s gestation after a non-reactive NST with

AS and a BPP of 6/8 2 days earlier.

CONCLUSIONS: AS induced accelerations appear to be valid predictors

of fetal well-being in twin gestations. AS reduces the frequency of non-

reactive NSTs and shortens the time requ=red for testing.

86 88 COMPUTERIZED HEART RATE MONITORING OF GROWTH

RESTRICTED FETUSES. E. Ferrazzi~, S. Riganox, R Mezzopanex,

M.Ballottlx, A Barbara, G Pardx Dept Ob/Gyn ISBM San Paolo,

University of Mtlan, Italy.

Objective. To assess whether computertzed analysis of fetal heart rate

(c£HR) in growth-restricted fetuses, can identify fetuses at risk better and

earlier than visual analysis of fetal heart rate (vFHR) recordings

Study Design. Baseline FHR, long term (LT) and short term variainlity

(ST) during epochs of low (phase A) and high (phase B), variainhty, large

(>15 b.p.in.) and small (->10 b p m.) accelerations, duration of phase A and

B, were calculated by cFHR in 36 growth retarded fetuses, during i to 3

weeks before delivery. Fetal condition was monitored by means of fetal

binmetry, ammoUc fluid estimatinn, umbihcal artery and middle cerebral

artery Doppler veincrmetry, and vFHiL Timing of dehvery was decided by

the same clinical staffblmded to the results of computerized analys~s

Results: 21 fetuses had a normal and 25 fetuses had a persistently

abnormal uminlical Doppler velocimetry, vFHR was always normal m the

group w~th normal umbilical Doppler. Biophysical data, aLl the cFHR data,

except baseline FHR, and neonatal outenme were significantly worse m

fetuses with abnormal umbihcal Doppler. In the group with abnormal

umbilical Doppler, dehvery was prompted by a non-reactive vFHR in 11

fetuses. Neonatal outcome was significantly worse in these f~uses In

these 11 fetuses, before the last non-reactive vFHR, the cFHR mean values were phase A duration= 61%, LT variabihty= 7.01±0.4 b.p.m, ST

variability= 3.2~0 6 mmsec., large accelerations per 60 minutes= 1.5±1.

These data were already significantly lower than In the 14 fetuses who

were dehvered on the bas~s of biophysical data, without waxting for non-

reactive vFHR In these fetuses cFHR values were. phase A duration=

41%; ST variability= 74+1.1 b.p.m; LT variabihty= 3.8±0.8 mmsec ;

large acceleratinns per 60 minutes= 4 1+3.

Conclusion. These results show that computerized analysis of FHR can

~dentffy and measure abnormal characteristics of the FHR before the

appearance of non-reactive FHR at visual analysis


89 NONREACTIVE NST: PREDICTIVE VALUE FOR NEONATAL

ANEMIA IN THE ISOIMMUNIZED PREGNANCY J Ouzounian MD, H Monteiro MD,x O Alsulyman MD, and G Songster MD. University of Southern California School of Medicine, Los Angeles, CA. OBJECTIVE: TO assess the predictive value of fetal nonstress tests

(NST’s) in predicting neonatal anemia in pregnancies complicated by red cell isoimmumzatlon. STUDY DESIGNI The records of patients evaluated for lSolmmumzatlon in

pregnancy from 1/92 to 12/94 were l’eviewed retrospectively. In addition

to prenatal care, serial ultrasonography, and lnvasive testing when indicated, patients had twice weekly NST’s. NST’s were interpreted as

either reactive or nonreactive (NR) using standard criteria. Results of the last NST prior to delivery were analyzed. Neonatal outcome data were obtained prospectively and by chart review. RESULTS~ SIxty patients with lSolmmunization and complete NST records were identified. The mean (SD) maternal age and gestatlonal age

at delivery were 30 6 (5 4) years and 37 (2.6) weeks, respectively. Fifty- one patients (85%) had reactive NST’s until delivery and nine (15%) had NR NST’s which prompted delivery Twelve of 51 (23.5%) patients

with reactive NST’s and seven of nine (77 8%) patients with NR NST’s required neonatal transfusion (P=0.003, OR=ll.4 (1.7-120.2)) The mean (SD, range) hematocrit at birth was 38.9 (3.0, 21.3-52 0) in patients with reactive NST’s and 28.3 (3 8; 14.5-45.0) in those with NR NST’s (P<0.05). A NR NST had a 77.8% positive predictive value in identifying the need for neonatal transfusion All patients with reactive NST’s had normal ultrasound examinations. Of the seven patients with NR NST’s and neonatal transfusion, two had ultrasound findings consistent with early ascltes CONCLUSIONS: Within the constraints of this study, these findings

indicate that a NR NST is highly predictive of subsequent neonatal transfusion and may predict fetal or neonatal compromise prior to ultrasonographic evidence of anemia Antepartum fetal nonstress testing is a useful adjunct In the management of ismmmunlzed pregnancies

91 PROSPECTIVE ASSESSMENT OF THE ACCURACY OF FETAL RhD STATUS DETERMINATION FROM UNCULTURED AMNIOCYTES L-, Nelsonxl, G.M. Jackson1, K. Ward1,2. Depts. of Ob-Gyn1 and Human Genetics2, Univ. of Utah,

Salt Lake City, UT OBJECTIVE." To prospectively determine the clinical accuracy of PCR-based DNA analysis of fetal RhD status from uncultured amniocytes.

STUDY DESIGN-" 87 amniotic fluid (AF) samples from RhD- negative, RhD-sensitized women were referred to the DNA Diagnostic Laboratory for analysis for fetal RhD status and for whom neonata~ follow-up was available. Standard techniques and published primers (Bennett et al, NEJM, 1993) were used for DNA analysis. Results were compared to standard serologic typing of neonatal cord blood. RESULTS: Of the 87 samples tested, 84 (96.5%) were in agreement with serologic results. Of the 3 discrepant samples, 1 was interpreted as RhD-positive on PCR testing of AF but was RhD- negative by serology. Another AF sample tested as RhD-negative, with RhD-positive serology. This AF sample was small (<1 ml) and contaminated with maternal blood. The last discrepant AF sample

tested as RhD-negative; the pregnancy ended in fetal death (without hydrops). Cord blood was not available. An independent AF sample collected after the fetal death also tested RhD negative, however, renal tissue tested RhD positive by PCR. CONCLUSION: The technique of PCR-based DNA analysis of uncultured amniocytas for fetal RhD status is highly reliable, but the accuracy with current methods is less than 100%. The likelihood of a misidentification of fetal RhD status in our series was 3.4%; possible causes include a primer site polymorphism, specimen contamination, or tissue mosaicism. Further experience with this technique in clinical practice will be needed before its accuracy is confirmed.

9O FETAL DOWN SYNDROME DETECTION: A COST ANALYSIS OF ALTERNATE SCREENING PROGRAMS Cusick*, L Buttino*, AM. Vintzileos**, N,~ Gleiche~’, *Center for Human Reproduction, Chicago, IL. and **UMDNJ-Robert Wood Johnson/St Peter’s Medical Center, New Brunswick, OBJECTIVE: To compare the efficacy and cost effectiveness of different screening programs for fetal Down Syndrome (DS). STUDY DESIGN: Screening tools evaluated included maternal age, tdple screening (TS), and ultrasound (U/S) for fetal markers of DS. Sensitivities used were: TS:60% <35 yrs, 90% >-35 yrs; U/S:70O/o. Average regional fees used were: TS:$80, U/S:$200, ammecentesis (AM):$700. Five screemng programs were evaluated: 1) <35 yrs, no screening; >-35 yrs, AM; 2) <35 yrs, TS w=th AM for screen positive subjects; -> 35 yrs, AM; 3) all patients, TS with AM for screen pos=tive subjects; 4) <35 yrs, TS followed by U/S for screen positive women, AM for women with fetal markers of DS on U/S; _> 35 yrs TS with AM for screen positive subjects; 5) all women, TS followed by U/S -for screen positive women, AM for women with fetal markers of DS on U/S. The sens~iwty, total cost, costJcase DS detected (Cost/DS), AM losses, and residual r~sk (RR, undetected DS fetuses/women not receiving AM) were calculated for each screening program. Population analysis was performed using 1988 IL. delivery statistics. RESULTS: It was estimated that 260 cases of DS would occur in the population of 167,654 women. PROGRAMI #1 I #2 1#3 I #4 I #5 Sensitivity 30% 72% ~ 68.7% 56.1% 48.1%"’ CostlDS 126k 172k 145k 139k 151k AM losses 56 114 43 23 11 RR,<35 yrs 0.1% 0.05% i0.05% 0.07% 0.07% RR~>_35 yrs 0 0 ~ 0.08% 0.08% 0.22%

CONCLUSIONS: DS screening incorporating TS in all patients with (#4) and without (#3) selectwe U/S ~n women <35 yrs yield the best combination of sensitivity and cost effectiveness while minimizing the number of AM related losses.

92 INFLUENCE OF THE LATERAL DECUB1TUS POSITION ON THE AMNIOTIC FLUID INDEX

B Bootstaylor.x S Rigaud-Echols, J Barry, S Ohana, D Saltzman, Dept. of Ob/Gyn, Beth Israel Medical Center, New York, N.Y. OBJECTIVES: The amniotic fluid index (AFI) as described by Phelan J. et al., is obtained with the patient in the supine position, which may be impractical in some patients who develop supine hypotension syndrome. We proposed that an equally reliable AFI can be obtained in patients in the lateral decubitus position. STUDY DESIGN: Thirty-four patients (gestational age 28-41 weeks) referred for biophysical prof’fle were studied. Each received the standard AFI assessment with ultrasound. A "lateral" AFI was also assessed after positioning the patient in the lateral decubitus position for 1 minute. The observer performing the "lateral" AFI was blinded to the results of the standard AFI, and all observations were performed in triplicate with mean values reported. Patients were also asked if the lateral decubitus position caused more discomfort than the supine position. Linear regression analysis was used to compare the two AFI techniques. RESULTS: The "lateral" AFI has a significant correlation with the standard AFI. A correlation coefficient (R) of 0.94 was noted. The regression equation was: lateral AFI = 0.11 + 1.04x. The mean difference between the two techniques was 1.5 cm + 1.2 cm. Between the two techniques 27 (67%) patients had a difference of -< 1 cm, and 32 (94%) patients had a difference of-< 3 cm. Twenty- seven (67%) patients also felt more comfortable in the lateral decubitus position than the supine position. CONCLUSION: Assessing the AFI in the lateral decubitus position correlates well with AFI assessed in the standard supine position, and patient comfort is improved.

Volmne 174, Number 1, Part 2 SPO Abstracts 339 AanJ Obstet G~,necol

93 Is THE ROUTINE USE OF NON S’IRESS TESTING NECESSARY 1N T\MN

t,LSTATIONS? B Camr~belL r Fontenot, J Read, D Milhgan, L Tutt’, S

Crowe’, R Newman Dept Ob/Gyn, University of Kentucky, Lexington, KY

OBJECTIVE The purpose of this study was to assess whethel lOUtUle antenatal

ultrasonography in twin gestations may select the group of patients most likely

to benefit from antenatal fetal heart rate testing

STUDY DESIGN A review of 157 twin gestations followed m mlr antupartum

tWillS cinllC WaS performed Each patient received an ultrasound examination at

least every 4 weeks with more frequent examinations in the late third trunester

These patients also underwent routine non stress testing beginning at 32 to 34

weeks unless earlier testing was chnlcally indicated based on ultrasound results

Factors contributing to poor permatal outconle including mtrauteune glowth

restriction ([UGR), twin discordancy (greater than 25%). hydramnlos.

ohgohydramnlos, monoamnlonlclty, and structural anomalies were recorded

Non stress tests performed w~thln 1 week of delivery were reviewed

RESULTS The mean gestatlonal age at diagnosis and dehvcry were 18 5 + 6 8

wks and 35 5 _+ 2 8 wks, respectively Ninety-one percent of patients had an

ultrasound examination within 4 weeks of dehvery The ultrasound was

considered normal If there was no evidence of IUGR (<10th pelcentlle),

mtertwm discordance (_>25%) or fetal anmnahes and nm~nal amnlotlc fluid with

an intervening membrane were noted One hundred nineteen (76%) had nounal

ultrasound exmmnatlons and 38 (24%) had at least 1 abnonnal ultrasound

finding The incidence of fetal or neonatal compromise defined as 5 minute

Apgar of less than 7, cord arterial pH less than 7 10, mecomum aspnatlon,

neonatal smzures, or pcnnataI or neonatal death was 18 of 38 (47%) m the

abnomml ultrasound group compared to 7 of 119 (6%) In the normal ultrasound

group (p <0 05) in the normal ultrasound group 4 of 7 cases ~lth fetal or

neonatal compromise were associated ~vltll acute lntrapartom events that would

not have been predicted w~th antenatal testing ]’he modence ol preterm labor

and preterm rupture of membranes was not different

CONCLUSION l’hls study suggests that serial Ultlasound exmnmatlons may

effectively identify abnormal multiple gestations at risk for permatal/neonatal

compromise and therefore restrict the use of antcpartum testing to a smaller

population which may potentially benefit

95 A STATEWIDE PATTERN OF ACCESS TO PRENATAL

CARE. J_ Yankowltz, DM Howserx, JW Elyx Depts of Ob/Gyn and

Family Practice, University of Iowa College of Medtcme, Iowa City,

IA

OBJECTIVE: Reform of the health care system and graduate medical

education have emphasized the role of the primary care provider with

little attention to women’s access to prenatal care We studied the

demographics of prenatal care in one state and suggest an alternative

solution

STUDY DESIGN: Demographic data were obtained from the Iowa

Department of Pubhc Health, Iowa Birth Defects Registry, and the

College of Medlcme, University of Iowa.

RESULTS: The 1990 Iowa population of 2,776,755 was distributed

among 99 counties. From 1983-1990, the mean number of live births

per year was 39,965. All but 4 countres average over 100 deliveries per

year Seventeen percent of the counties tn Iowa (6 5% of the

populatmn) have no physician providing prenatal care Seventy-eight

counhes have at least one Family Physician (FP) Overall, 88.6% of the

population live In a county with at least one FP providing prenatal

care In contrast, 79 of 99 counhes had no obstetricians (OBs). Of the

160 OBs providing prenatal care in the 20 remaining counties, 157

practiced in the 18 most populated counhes which contmn 42 4% of the

populatmn, OBs performed slgnrficantly more dehvenes (p<0 01) and

of a higher risk (p<0.01) than FPs. Over 30% of the FPs report

performing less than 25 deliveries per year, versus only 2% of OBs

CONCLUSIONS: These findings may also apply to other rural states

where 50% or more of the population resides in nonmetropohtan areas.

Most rural communities have no obstetnmans and FPs are delivering all

low-risk obstetric care Innovative solutions to make obstetric

consultation more available to rural FPs or increase patient access to

OBs are needed New technologies, such as telemedicine, may offer such

a solution Impact on quality of care related to physician distribution or

patient load was not evaluated, but education at the resident level should focus on these types of collaborations

O4 MULTIVARIANT ANALYSIS OF ANTEPARTUM FETAL TESTS IN

PREDICTING NEONATAL OUTCOME OF GROWTH RETARDED

FETUSES. Z. Wether, M Y Dlvon, N Katz~, V.K MlnlO¢, A NasserP, B.

Glrz Dept of Ob?Gyu, Atbert Einstein College of Medicine, Bronx, NY

OBJECTIVE: To evaluate the efficacy of antepartum surveillance in

predicting neonatal outcome of growth retarded fetuses.

STUDY DESIGN: 135 growth retarded fetuses confirmed by a birth weight

(BW) < 10th percentile without chromosomal or structural abnormahtles were

included m the study Fetal tests performed within 3 days of delivery ~ncluded

NST, biophysical profile (BPP), and umbilical artery Doppler veloc~metry

Neonatal outcome was assessed by" umbilical artery blood gases, Apgar score

at 5 mln, length of stay in the neonatal intensive care unit (NICU),

lntraventrlcular hemorrhage, respiratory distress syndrome, and neonatal death

A stepwise logistic regression was used to analyze each of the outcome

parameters as a dependent variable The independent variables used m the

model were gestational age (GA), BW, and all fetal surveillance tests A

pH<7 15 with base deficit >9 defined metabohc acldosts. Late or severe

variable FHR decelerations and reduced varlablhty defined a non-reassuring

lntrapartum FHR

RESULTS: 26 fetuses had absent or reversed end-d~astohc velocity; 40

fetuses had BPP < 6, 42 had a non-reactive NST and 31 had antepartum FItR

decelerations The mean(~_SD) GA and BW at delivery were 36+3.8 weeks

and 1841+626 gins, respectively Five of 90 fetuses admitted to the NICU

died The presence of neonatal metabolic acidosis correlated only with

umbilical artery S/D ratios (p <0 0001) A non-reassuring lntrapartum FHR

was predicted by umbilical S/D ratios (p < 0 0001)and by BPP scores

(p=0 0013) All other adverse neonatal outcomes correlated only with BW or

GA (p < 0.0001)

CONCLUSIONS: Other than metabolic acidosis at bmh which was predicted

only by umbilical artery Doppler studies, morbidity and mortality in severe

IUGR was primarily determined by gestational age and birth weight and not by

abnormal fetal testing

96 THE IMPACT OF ALTERNATE PRENATAL CARE WITH REDUCED

FREQUENCY OF VISITS IN RESIDENCY TEACHING PROGRAM. N.

Ward*, S Baye~, B. Calhoun. Dept OB/GYN, Keesler Medical Center, Biloxi, MS.

OBJECTIVE: The null hypothesis is that routine, uncomplicated obstetrical

patients may undergo decreased prenetal visits without affecting outcome. STUDY DESIGN: A cohort study was designed to include 924 controlled

deliveries from 1 May 92-30 Apt 93 who had prenatal care per ACOG Committee

Opinion #79, Jannery, 1990, guidelines for uncomplicated obstetric care. The

second study cohort of 862 deliveries from 1 Oct 93-30 Sep 94 underwent prenatal

care with modified guidelines to include: fu’st visit 6-12 weeks to confu’m dating

and obtain initial laboratory, 2nd visit at 16-20 weeks to secure MSAFP screening,

3rd visit at 24-28 weeks for 28 week labs, 4th visit at 32 weeks, 5th visit at 36 weeks, 6th visit at 38 weeks, 7th visit at 40 weeks, and weekly thereafter.

Descriptive stetisti~s regarding number of clinic visits per patient, cesarean section rates, labor and delivery visits, antepartum admissions, and perinetal morality

were compared. RESULTS: We noted no decrease in the total number of visits for each patient

in our system in spite of the implementation of the above system. What was noted

is a shift in obstetrical visits from the urgent clinics in 1992 to the complicated

obstetrical clinics in 1993 and an increase in access for gynecologic appointments

by 56% up from 16,445 in 1992 to 25,586 in 1993 without a change in staffing number.

CONCLUSIONS: Decreased numbers of routine obstetrical visits do t~t cause

any increase in obstetrical mmbldlty and may not necessarily result in decreased numbere of tntal obstetrical vlsits. Decreased numbers of routlne visite may allow

more visits for focused problem solving in the obstetrical clinics. An apparent benefit of this clinic shifting is the ability to schedule more gynecologic patients

ns the need for urgent obstetrical clinics is met by the decrease in routine obstetrical visits.

340 SPO Abstracts ,]anua~7 1996 Am J Obstet Gynecol

97 ACCEPTANCE OF A COMPUTERIZED TELEPHONIC OBSTETRICAL HISTORY (PregNet~). HF Andersen and B Cummingsx. Dept OB/Gyn, Loma Linda Univ, Loma Linda, CA. OB,,rECTIVE: To determine patient and physician (MD) acceptance of a computerized telephomc obstetrical history (PregNet). STUDY DESIGN: 40 patients completed the PregNet history by telephone. The system asks 117 initial questions and up to 125 follow-up questions. Patients respond by pressing specific telephone keys. In the prenatal clinic patients were randomized to a standard MD interview using the ACOG prenatal form (n=22) or MD review of the PregNet history (n=18). After their initial visit patients were asked about ease of use, clarity of questions and explanations (scored 0=easy to 10=difficult) and preferences for MD or computer interview (scored 0=strongly prefer MD to 10=strongly prefer computer). RESULTS: Patients reported high levels of satisfaction with the ease of use of the system, and with the clarity of questions and explana- tious. Public clinic patients reported poorer understanding of explanations by the computer than private patients (private: 0.7 + .5; public: 1.5 _+_ 1.8; p=.04). Patients tended to prefer an interview with an MD rather than the computer and tended to understand explanations from the MD better than the computer, but the differences were not statistically significant. Patients were nearly neutral regarding sensitive questions (drug use, HIV, and domestic violence). IvlD’s using a computer or nurse generated history spent less time completing the initial prenatal interview than those not using assistance (with assistance: 24.5 min + 11.6; without: 35.0 min + 17.3; p=.08). CONCLUSIONS: The PregNet computerized telephonic interview for initial obstetric history is well accepted by patients and tends to reduce MD time for completion of the initial prenatal history.

99 TH~ IDEAL MATERNAL BODY VC~IGHT GAIN DURING

PREGNANCY FOR APPROPRIATE BIRTH WEIGHT. T T H=eh, C C

Hs~eh, T.C Chang Dept Ob/Gyn, Chang GungMemonalHos!~tal, Ta~pe~,

TA13,VAN.

OBJECTIVE: To define maternal body weight gem during pregnancy m a

group that dehvered smgloton~ wth appropriate bLrth w~ght for gostattonal ago

STUDY DESIGN: Th~s is a retrospecUve study of 9,599 pregnant woman who

dehvered at a medical cantor from July 1990 to June 1995 study subjects were

woman who dehvered un¢omphoated bye =ngiotons wth w~ght appropriate for

gostaUonal ago and wthout fetal anomaly Subjeots were chwded rote three

groups by pre~pregnam body mass index (BMI) as BlV~ ~ 19 (group I, n=

2,869), 19 < BMI-< 23 (group g, n=5,539), and Blvg > 23 (group 1~, n=1,191 )

The increment ofBMI m second- and third-trtmesters were analyzed by linear

regres=on wth weighted least square method

RESULTS: The increment of BMl were 0 314/wk, 0 182/wk, and 0 162/wk m

group I, g, Ig r~e~txvely (p< 0 001 ) For a woman x~th mean body height m

each group (158 9, 158 2, 157.6 cm) the increments of body weight were 3 17,

1 83, and 1 62 kg for every four weeks Adjustment of body weight increment

by 0 2, 0 12, and 0 1 kg for a 5 em 0~fforance m each group ~s necessary For

example, the xdeal body weight gem of women wth hotght of 155 cm m groups

I and I~I were 3 0 and I 5 k8 !aer gosta~o~al mo~th, respe~avely, wl~e tha~ of a

165 om height woman m group II was 2 0 kg

CONCLUSIONS: A =gmficant d~fferance was observed m maternal

weight gem during pregnancy between pre-pregnancy underw~ght (~roup

and overw~ght (group ~ women who dehvered smgletens wth appropriate

birth weight for gestaUonal ago Speofic nutrmonal consehngprogram for

different groups according to pro-pregnancy BIVE ~s mandatory

98 SHOULD SINGLETON IN VITRO FERTILIZATION PREGNANCIES BE MANAGED AS HIGH-RISK PREGNANCIES? ]~Ri?~, P Bfinsden~, JL A~but~, LJ G~ome. Boum-Hallam Clinic, Bourn, ~mbridge, I.~ University of South Alabama, Mobile, AL. OBJECI’IVE: While iris generally ~ that prvgnancie~ following in vit~o fertilization (IVF) are at increased risk bec~us~ of the frequent o~urrence of multiple gestations, less is known �onc~’ning the outcome of singleton IVF pregnancies. The purpose of this study was to determine if singleton pregnancies o~urring s~/VF should be oensidered hi~-risk because of an i~rense in the frequency of obstetrical ~omplkations. STUDY DESIGN: Between 6/1/85 and 12/31/89, 494 singleton pregnancies

outo3me of those p~gnen~s was enmpared with 978 enntrol singleton presnandes

matched for maternal M.e and parity. The eomplicatlona of vaginal bleeding and hyp~tenal?~ w~ ~i~ in.wom~ who ~d ho*~t~ien. ~s (RR) and 95% confidence inte~als (CI). RF~ULTS: ~mpated to the ~ontrol Stoup, vaginal bleeding ~ sisnif~,anfly more ot~m in IVF pregnm~ies (17% vs 11%; RR 1.66, C! 1~9-2.14). placenta ~ occmrod in 3% o~�~1~ �~ndy ~roup ~ in only 1.4% of ooaltols (RR 1.41, CI 1.01-1.92). Hyl~ian~ ~ in 14% of fi~l~o~ IrCF pro~m~i~ ~t~ 7.4% in tbe ennttol ~roup (RR 1.90 CI 1.40 - 2.59). ~ ~aon w~ l~f~mod in

47% of all ~ingloton IVF pro~i~ �oml~n’od to.24.% in dn~oton ~o~tto9 (RR It.7~, Cl 1.53-2.06~ Fwll~m~e, 14% ofpt~aos m the ~auty group deliv~l ~ 37 ~npla~l ~ ~ to 8% of singleton ¢ontrol~ (RR 1.78, el 1.30- 2.42). "1~ in¢idenc,¢ of low birt~ walght (<2500 g) wns 14% in ~ingl~n rvT ~ wrst~ 7% in ennt~ls (1.75, Cl 1.28-2.39). Tbe ~illbi~h ~ ~md the nennund, l~im~Ud, end infimt do~th na~ did not dit~ fi~m ~i~ oxl~t~l t=~

~ngmital malfonnatiom. EONCLI.JSION: C~nl~r~l to a ~ ~mlrol ~mp, IVF ~in~on lX~nnad~

~ ~ ~ow~.. ~ delivm,! be .f~e 37 ..cympleted week~ low infant birth w~s~" matm~l h~nsina requi~g uospitniizatioa, a~l dalivery by cesarean

100 THE "EMERGENT" CERCLAGE: A TEN-YEAR, SINGLE-INSTITUTION EXPERIENCE. S ChasenX, N Silverman. D=vision of Maternal-Fetal Medicine Jefferson Medical College, Philadelphia, PA. OBJECTIVE: To evaluate factors asseciated wth time success gained in utero and gestational age at delivery in patients with emergent cercleges. STUDY DESIGN: Retrospact~ve chart review of 43 patients who underwent non-prophylactic cerclages from 1984-1994. Gestational age (GA) at cerclege, GA at delivery, cervical dilatation at presentation, and presence or absence of bulging membranes (BMB) on admission were recorded for all patients. Noncontinuous data were analyzed using X2 or Fisher’s exact test; continuous data were compared with either student’s t or Mann- Whitney U tests. RESULTS: All but 2 cerclages were able to be placed, at a mean GA of 19.2~.3.8 weeks (median 20 wks, range 12-25), with a mean of 11.5+8.0 weeks gained in utero (median 12 wks). Clinically, only 16 patients (37%) presented with symptoms, while 22 (51%) had BMB and 13 (31%) were >2cm dilated. Overall, 63% of patients delivered at _> 28 weeks, with 28% delivering at _> 36 weeks. Patients with BMB were more likely to have had s .~rnptoms than those w~thout BMB (64% vs 10%; p<0.001), to be >2cm ddated (59% vs 0%; p=0.003), and to present at _> 20 wks GA (72% vs 29%; p=0.001). They also gained less time after cerdegeplacement (median 3.5 vs 16.6 wks; p<0.001) and were less likely to reach either 28 weeks (45% vs 81% p=0.04) or 32 weeks (27% vs 62%; p=0.05) at delivery. While patients with BMB who were a so >2era di ated d d not d ffer in GA at time of cerclege (median 21 wks) than those less dilated, they did gain less time in utero (median 2 vs 11 wks), with only 38% reaching a GA of > 28 wks. CONCLUSIONS: While emergent cerclage was a successful procedure for most patients in this cohort, the presence of either bulging membranes or advanced dilatation at presentation was associated w~th lower cerdage-to-delivery intervals as well as a lower chance of reaching at least 28 weeks gestation.


101 THE UTILITY OF PROPHYLACTIC CERCLAGE IN TRIPLET PREGNANCY. JM O’Bnen. JR Barton, NK Bergauer,x DL Jacques,* GJ Star~iano.x Central Baptist Hospital, Lexington, KY and Healthdyne Maternity Management, Marietta, GA. OBJECTIVE: To determine the efficacy of prophylactic cerclagc for prolonging the length of gestation m triplet pregnancies. STUDY DESIGN: A matched cohort study design was utahzed with data abstracted from a group of 463 triplet pregnancies enrolled ~n an intensive preterm labor identification program. Only patients undergoing prophylactic cerclage at -< 20 weeks’ gestatmn were ehglble for ~nclusion in the treatment group Control patients were matched for panty and maternal height using a computer-generated random number table to select ehgible patients from the remaining original cohort. Control patients were also paired at a 2 to l ratm to index cases. Statistical analysis was performed utilizrng McNemar’s test for dxscrete outcomes and the pmred t-test for continuous variables. RESULTS: Twenty-e~ght patients with a triplet gestatmn (6%) underwent

prophylactic carclage placement at a mean gestational age of 15.2 _+ 2.4

weeks. No s~gmficant differences were noted in maternal age (31.2 -+ 5.2

vs 30.1 + 5.1 yrs, p = 0.26), frequency of m~nonty race (i4.3% vs 10.7%,

p = 1.0), ~ncldence of tobacco abuse (3.6% vs 5.4%, p = 1.0), frequency of

mamage (92.9% vs 94.6%, p = 1 0), and history of preterm birth (7.1% vs

1.8%, p = 0.37) between the treatment and control groups, respectively.

The ind~cataons for home uterine actav~ty monltonng were also s~mflar

between groups w~th 6 women (21.4%) in the treatment group and 11

women (19.6%) m the control group beganning surveillance due to

preterm labor, p = 0.80. The remaining paUents m each group inmated

monitonng prophylactacally Outcome measures are shown ~n the Table

Cerclage Control

Obstetric parameter (n=28) (n=56) p

GA dehvery (weeks) 32 1 _+ 3.1 33.7 + 2.1 <0.01 Cesarean delivery (%) 100 98.2 1.0

Neonatal parameter (n=84) (n=168)

Birth weight (grams) 1685 _+ 585 1874 _+ 434 <0.01 NICU admission (%) 78.6 66.1 0.12

CONCLUSION: Prophylactic cerclage does not appear to prolong the length of gestation in triplet pregnancies and can not be recommended for routine management of these higher-order multiple gestations.

[03 PREDICTION OF FAILED MID-TRIMESTER MEDICAL TERMINATION OF PREGNANCY. AR GoeDfertx, d Owen, JC Hauth. Dept. of OB/GYN, University of Alabama at Birmingham, Birmingham, AL OBJECTIVE: To identify associations with failed and/or prolonged mduct~ons of labor using uteroton~c agents in women undergoing mid-trimester pregnancy terminations. METHODS: From March 1990 to June 1995, as part of 2

randomized chnical trials, 245 women underwent a mid-trimester termination with Prostaglandin E2 (PGE~), a concentrated oxytocin protocol (COP) or both COP and low-dose PGE2. PGE~ was administered as a vaginal suppository 20 mg Q4h or 10 mg Q6h when used in combination with COP. Patients who were not delivered (or in whom delivery was not imminent) after 24 hours were considered to have faded and received alternate therapy. RESULTS’ From the population of 245 women, 207 (85%) were successfully dehvered, while 38 (15%) comprised the Failed

group. There was no difference in the failure rates among methods PGE2 (17%), COP (19%), or both (11%), p=0.4. The two groups had sim=lar characteristics including maternal age, parity, and gestational age at induction, except that white women had an increased incidence of failure compared with non-whites (20% vs 10%, p=O.04). The 153 women w~th a fetal anomaly vs 92 women with other indications for termination had an increased risk of failure (22% vs 5%, p=0.001). In addition, the mean induction t~me ~n the successful group was s~gnificantly longer for women with fetal anomalies (15 + 6h vs 9 + 6h, p < 0 0001). CONCLUSIONS." Women with fetal anomalies are at increased risk for a fa=led and also longer mid-trimester medical induction

of labor.

102 SHIRODKAR CERCLAGE IS ASSOCIATED WITH LONGER PLACEMENT TO DELIVERY INTERVAL THAN IS MCDONALD CERCLAGE. G Eghnton, L Pisto~ax, T. Pmckert, J. Queenan, J King, J. Collea. Department of OB/GYN, Georgetown Unlvemty Med=cal Center, Washington, D.C. OBJECTIVE: Our Maternal Fetal Medicine division has a pnvate referral and acute maternal transport practice, =ncludmg a large number of patients referred for incompetent cervix, not n premature labor. These patients often present with a dilated cerv=x, w=th or w=thout a prior cerclage placement In recent years, we have performed a larger percentage of SNrodkar (S) cerclages. W e endeavored to determine =f we could detect d~fferences in outcomes between McDonald (McD) and S cerclages STUDY DESIGN: Pahents with cerclages were ~dent=fied from a database of MFM patients followed dunng 1992-94. Patients’ charts were rewewed to record h=story, ~nduding pnor cerclages in the current or previous pregnancies, =ndlcat=ons and types of cerclages, number and duration of hospitahzatlons, and dehvery =nformation. T- and Mann-WNney tests were apphed to continuous variables from the patients’ h=stones prior to our care, and from the current pregnancy. Multiple gestations were included RESULTS: We ~dent=fied 167 pahents with cerclages. Complete pregnancy and delivery data were available for 129 patients. 91 (70%) patlants had McD. Patients d~d not differ with respect to numbers of pnor full term, premature dehvenes, 1st or 2nd tnmester spontaneous or elective abortions, hwng children, cervical effacement or d=[atlon at t~me of cerclage (mean + standard deviation) (S = 0 5 + 0 9 centimeters, range 0-3, McD = 0.8 + 1.2, range 0-5, p = 0 18). Both techniques were used for emergency salvage operahons with the cervix open. The mean gestat~onal age at cerclage was lower for S (15.9 +_ 0.5 weeks, range 8-22 versus (v) 17 5 + 0.5, range 9-28, p< 0.05), with longer mean =ntervals to delivery for S (21.0 + 0.9 weeks v. 17.5 + 0 9, p < 0.03, and greater gestahonal age at dehvery for S (37 0 + 0.8 weeks v. 34.9 +_ 0 6, p = 0.06) . CONCLUSIONS~ For pahents with a d~agnosis of incompetent cervix ~n a current or pnor pregnancy, Sh=rodkar cerclage ~n the current pregnancy resulted =n greater prolongation of pregnancy, and a marginally sign=ficant =ncrease n gestatlonal age at delivery

104 NECROTIZING FA$CIITIS FOLLOWING CESAREAN SECTION. AR Goeofertx, DA Guinn, WW Andrews, JC Hauth. Dept. of OB/GYN, University of Alabama at Birmingham, B~rmingham AL. OBJECTIVE: To review our expenence with the diagnosis and management of necrohz~ng fesc~rtis (NF) following cesarean section (CS). STUDY DESIGN. We reviewed medical records of any women with

serious post cesarean wound infections at the University of Alabama at Birmingham between 1987 and 1994 to identify women with NF. The diagnosis of NF required intraoperahve ~dent~fication of necrotic fascia in febdle women undergoing post-cesarsan wound debndement. RESULTS: During the study period, 5048 women had cesarean deliveries of which 7 were complicated by NF. Selected demographic and obstetnc characteristios include:

Maternal age (years) 25.7 + 5.9 (mean _+. sd) Black race 4 (57%) Gestational age at delivery (weeks) 35.7 + 5.4 Weight (pounds) 182.4+ 68.1 (range I23-320) Preec~ampsia 2 (29°,6) Prolonged rupture of the membranes 1 (14%) Chonoamnionitis 2 (29%) Endometritis 2 (29%) Prophylactic antibiotics at section 7 (100°,6)

None of the patients were malnounshed, none had diabetes mellitus and there were no ~ntraoperative complications at CS. Four women had a Pfannenstiel and 3 a midline vertical skin incision for the CS. The mean duration of time from CS to the diagnosis of NF and reoperation was 8.6 + 4.3 days ranging from 4 to 14 days. All had surgical debridement upon consideration of the diagnosis and all received broad spectrum antimicrob~al therapy Four women required placement of synthetic mesh (Marlex) to effect fascia! closure following debridement. Wound cultures were positive in 3 women, 2 with polymicrobiat infections and 1 with Clostrid~a. Two women died: one as a result of metastatic breast cancer and another developed sepsis and ARDS and died 3 weeks following debndement. CONCLUSION: At our instituhon 1 ~n 1000 women developed NF following CS. NF results in appreciable morbidity and mortality and warrants diligent assessment of postoperative wounds to allow early diagnosis and primary surgical treatment.

342 SPO Abstracts ]anualy 1996 Am,] Obstet Gynecol

105 THE RISING INCIDENCE OF GASTROSCHISIS. J E Dickinson C R

Nicholsx, P PembertonX. Dept of Obstetrics, King Edward Memorial

Hospital for Women, and Dept of Neonatology, Princess Margaret Ilospital,

Perth. Western Australia OBJECTIVE- To develop a popalatton-based incidence of gastrosehisis

using the unique cbarastenstics of a geographically isolated state w~th a

single tertiary obstetrm and pediatric hospital

STUDY DESIGN A retrospective review using data from the Western

Australian Birth Deflects Registry, the Midwives’ Notification System and

case records of the tertmry referral hospitals from 1980-1993. RESULTS. Sixty-four cases of gastrosohisis were ~denfified dunng the

period 1980 to 1993 With 332,530 dehvenes in the 14 year period of the

review, the population incidence is 1 92 per 10,000 births (95% CI 1 51-

2 46). There has been a rise in me,dance from 0 48 per 10,000 births in 1980

to 3 16 In 1993 (NS) This alteration in incidence is an age-group specific

event. The age group 15-19 years, which accounts for a consistent 6 5% of

total deliveries, has 10 times the incidence of the age range 25-29 years A

sharp rise In the occurrence of gastrosohlsts In women 15-19 years was observed, with the incidence increasing from 4 0 to 26 5 per 10,000 births

over the period of review

Increased use of prenatal ultrasound has made antenatal diagnosis usual and

consequent referral to the tertiary referral hospital for dehvery There was a

46% incidence of smoking and a 19% use of recreation drug use A strong

association with preterm delivery and low birth weight was present with a

50% incidence of low birth weight, eight times the state incidence, (OR

14.82, 95% CI 8.97-24 51) The rate of preterm birth was 46 1% vs 6 6% in

the general population (OR 12 11, 95% CI 6 45-22 73) There was a high

incidence of perceived fetal distress, reflected in an increased rate of

emergency cesarean section (23% vs 8%, OR 4 31, 95% CI 1.91-9 74) The

early neonatal outcome was satisfactory with three neonatal deaths and a low

CONCLUSION’ The increase m incidence of gastroschisis in this

population-based study reveals a strong association ~mth young maternal age

Investigation into possible etiologic factors to explain this observation is

required.

107 COMPARISON OF URINE PROTEIN INDEX AND QUANTITATIVE 24-

HOUR URINARY PROTEIN EXCRETION. T Fomenot, L Lameaer×, D

MTflhgan, J Read, F iVfiller, W Porter×, A Mohapatra×, E Tuft×, and B

Campbell Department of Ob/Gyn, Univermty of Kantueky, Lexington, KY

OBJECTIVE: The purpose oftlms study is to determine if the unne protein

index (UPI) has precise, pomtive correlation with quantitative 24-houx ttrme

protem excretion

STUDY DESIGN: A prospective observaUonal trial comparing UPI to 24-hour

quantitative urinary protein excretion was performed on hospitalized patients

undergoing evaluataon for protomuna A 5 ml[hhter onne sample for creatmme

and protein quantita~aon was obtained mamedaataly liner to the 24-boor unne

collection UPI was calculated from thrs spot sample using the following

equations

Creatlmne (Cr) excrolaon (rag/d) = (140-agoXwetght m kg)(0 85) + 5

UPI (protein excret~_on, mg/d) = (Cr excretion, mg/dXspot protean, mg/dl)

spot ereatmine (mg/dl)

Forty-tee samples were obtained Precision of the UPI at pre&~g 24-ho~

prot~ oxero~ was evM~tod ~g eo~ela~ ~a~es S~m~ ~d

~ of~I m pr~&e~g pa~olo@e ~a (>400 re#d) ~d n~e

rm~ pr~a (>30~ re#d) were ~o eMedated

~TS: Two,-five (59%) of ~o samples tepid had pa~ulo~e

~ot~a ~d 12 (29%) had nop~e ran~ prot~a by 24-ho~

q~hve ~yms ~e eo~olafi~ eoeffi~ent (r) for ~I ~s 0 90

S~m~W ~d ~e~ for ~I m pre&~g pa~olo@e ~ot~a were 92

~d 88 pere~t, re~ee~vely S~m~mW ~d ~ee~ for pr~e~g

r~ ~ot~a were 1 ~ ~d 93 ~e~t, re~ee~vely CONCLUSIONS: ~e ~e prot~ index is a pre~se me~s of ~y~g ~y

~ prot~ exerefi~. C~c~a~ ~I from ~ ~ne ~mples p~des a

more ra~d assasam~t of prnt~ excre~, ~ m~g e~c~ dermis

e~t It is ~so a less m~bersome tee~q~ fer pewits, n~g

pers~d, ~d laboret~ tae~ci~s

106 THE ROLE OF CEPHALOCENTESIS IN THE MANAGEMENT OF SEVERE FETAL HYDROCEPHALUS (1985-95) TG Teoh x, G Ryan, P McParlund, N Sinlettox, S Grisarux, EN Kelly x, I Wiza x, D Farine.

University of Toronto Petinatal Complex, ON, CANADA & National Maternity Hospital, Dublin, IRELAND.

OBJECTIVE: To assess the role & outcome of ¢ephaloeantests in the Intrapartum management of severely hydrocephalic fetuses.

STUDY DESIGN: A descriptive series of 32 severly hydroeaphalic fetuses who underwent intrapartum cephaloeantesis, vath follow-up of all survivors. RESULTS: 12 fetuses had isolated severe hydrocephalus, 15 also had spina bifida & 5 had multiple anomalies The diagnosis was made in the third trimester m all cases, 4 clinically (2 in labour)Jail in 1985] & 28 on

ulfrasound (US). All parents were coun~selled as to the prognosis for the condition & probable lethality of cephalocentesis. There was one set of twins (1 uormal fetus) & 44% of fetuses were breech presentations. Labour was

induced in 9 cases A single drainage was sufficient to facilitate delivery of the head in 31/32 cases. 3 procedures were done trans-abdominally under US guidance, & 29 transvaginally. On average 410 nil were drained (30 - 2,200ml). There were 2 caesarean sections, 10 spontaneous vertex, 6

instrumental & 14 breech deliveries. Following cephalocentesis, there were

21 stillbirths, 9 neonatal deaths (5 min - 6 days) There are 2 survivors aged 9 months and 4 years One has a mild hemiplegia but is otherwise neurologically intact, the other ts severely developmentally delayed and institutiunahsed. In both cases, 21 or 22g needles were used for the

procedure, whereas in most other eases, much larger instruments were used.

CONCLUSION: It is taught that cephalocuntesis ~s invariably lethal. Our data show a 6% (2/32) survival rate with one infant having an almost normal

neurological outcome These results may relate to the technique & instruments used. Parents must be counselled as to the potential for survival with possible neurological impaimaent following cephalocentesis

108 RADIOGRAPHIC EVALUATION OF BREECH PRESENTATION: IS 1T

NECESSARY? T. FontenoL D. Lewis, B Campbell, D Ivralhgan, E Tutt×, J

Read, F Miller Depertments of Ob/G3m, Louamana State Umversaty Medacal

Center, Shrevsport and Umversaty of Kentucky Me&eel Center, Lexangton

OB.IECI’IVE: The purpose of thrs study IS to evaluate the preeisaon of

ultrasonography m tdent~fymg fetal pus~tion 0ower extrema~es and degree of eerwcel spine flemon) in the breech presentation

STUDY DESIGN: Patients beyond 35 weeks’ gestaUon being evaluated for external eephahc version or for symptoms of labor with a fetus m breech

presentation were offered enrollment Radingraphtc and Mtrasonograph~c

evaluations of the fetus were performed to determine breech position and degree

of cervmal spree flemon SenmU~ty and spe~tflcity were calculated to determine

the accuracy of ultrasonograpluc results

RESULTS: The following results were obtained from 48 subjects

Fetal Posation Frank Complete _Incomplete

Rachograph 0q) 22 9 17 Sonography (% correct) 91 44 71

Ceoaeal Spree Flenaon

Flexed Mflxtary Partially Deflexed Full Exteus~on

< 90° ~ 90° 90-105° > 105°

Ra&ograph (N) 13 22 9 4

Sonography (% correct) 85 59 56 25

Sensat~vity and speclficaty for detection of the Frauk breech postuon were 91

and 92 percent, respectavely Sensati,aty and speetfietty for detection of full

extenston (>105°) of the fetal cervmal spme were 25 and 100 percent,

respectively

CONCLUSIONS: Ultrasonography alone is precase m identification of a fetus

m the Frank breech po~on but is not sensatave m detection of a fully extended

fetal cer~cal spree Because of the sagmficant risk of injury to the fully extended

fetal cervacel spine during vaginal dehvery of a breech presentation, it is

paramount that sonograpbae findings are ventied by ra&ograpMc evaluation of

the fetal cer~eal spine pnor to a ~al of labor

Volume 174, Number l, Part 2 SPO Abstracts 343 Arn .10bstet Gynecol

109 THE CLINICAL USE OF THE KLEIHAUER-BETKE TEST

IN RH POSITIVE PATIENTS. J. Boyle×, J Klmx, H. Waleriusx, P. Samuels. Dept. of Ob/Gyn, The Ohio State University Methcal

Center, Columbus, OH. OBJECTIVE: Our objective was to detarm~ne the role of the

Klethauer-Betke (KB) test m managing Rh positive patients who are at risk for fetomatemal hemorrhage. STUDY DESIGN: A retrospective chart review was conducted on 655 Rh positive patients who had KB tests performed and were evaluated in our Isoimmumzat~on Program between January 1, 1991

and June 30, 1995. We analyzed indications for ordering the test, quantitative results, chmcal utility and pregnancy outcome. RESULTS: The most common indications for KB testing included maternal trauma (n=372), external versmn (n=95), and vaginal

bleethng (n=68). In 2 out of 3 cases, the patient was either discharged

or dehvered before the results were k~own. One hundred ten (16.89o) of the tests were performed at 10-23 weeks gestation. There were 127

(19.4%) positive test results ranging from 0.05-7.8% fetal cells with a mean of 0.3 + 0.9% SD. Only 8 (6.3%) patients had a test result >

1.0% fetal cells. The KB test was posttive tn 75 (20%) patients

experiencing maternal trauma, 14 (21%) experiencing vaginal

bleeding, 10 (11%) undergoing external version, 8 (47%) with fetal

hydrops, and 7 (17%) with an IUFD. In only 4 (0.6%) cases thd the result of the KB test affect clinical management. CONCLUSIONS: A sigmficant number of Klethauer-Betke tests are

being performed in Rh positive patients for maternal trauma, external version, and vaginal bleeding, many before fetal viability. In the

majortty of cases, clinical management is determined before the results are known or available. The number of significant fetomaternal

hemorrhages is small, and the results are of lirmted value in managing these patrents.

lll IN UTERO ALCOHOL EXPOSURE: EFFECT ON AMNIOTIC FLUID

VOLUME. S. Marticr. H.M. Wolfe* J. Ager,* R. Sokol*. Dept. Ob/Gyn.,

Wayne State Umv., Detroit, MI.

OBJECTIVE: Fetal exposure to alcohol has bean associated with impaired

growth and intellectual function and congenital anomahes after birth Specific

mechanisms for alcohol’s effects on the fetus have not been established. Our

purpose was to examine alcohol’s effect on ammotic fluid volume (AFV) whtch

has been associated with fetal structural defects and aberrant growth ~n utero.

STUDY DESIGN: 3887 patients with prospectively collected prenatal alcohol

exposure data were identified and records from the most recent ultrasound before

delivery rewewed for AFV following exclusion of fetal anomahes Discrimmant

function anaylses were performed for abnormalihes of AFV coded as increased

and decreased fluid and their association with alcohol exposure (absolute

alcohol/day, proportion drinking days and absolute alcohol/drinking day both

preconceptionally and during pregnancy) adjusted for birthweight percentile.

RESULTS: In the stepwise anaylses gestational age and ultrasound, birthweight

percentile and average absolute alcohol/day during pregnancy were all

significantly associated w~th increased AFV. Decreased AFV was not affected

by alcohol in this sample.

CONCLUSION: These preliminary results demonstrate an association of

prenatal alcohol on increased AFV. Further study ~s needed as a step toward

establishing mechantsms of alcohol effects.

110 SEVERE PREECLAMPSIA IS ASSOCIATED WITH THE FACTOR V LEIDEN MUTATION D. Dizon-Townsonl= L. NelsonTM, L. MolineTM, K. EastonTM, K. Ward1,2. Depts. of Ob-Gyn1 and Human Geneticsa, Univ. of Utah, Salt Lake City, UT. OBJECTIVE: Dekker et al described resistance to activated protein C in association with severe preeclampsia (SPO Abstract #72, 19951. A missense mutation in the factor V gene, the Leiden mutation, is the most common genetic cause of resistance to activated protein C. Our objective was to evaluate the frequency of this mutation in patients with severe preeclampsia compared to normotensive, gravid controls. STUDY DESIGN: DNA was extracted from whole blood of 158 gravidas meeting ACOG criteria for severe preeclampsia and 403 normotensive gravidas. The polymerase chain reaction was used to amplify exon 10 of the Factor V gene followed by allele-specific restriction with Mnl 1 for mutation detection. Results were analyzed with a X2 contingency table. R E S U LT S: Allele-Specific Restriction Analysis

Heterozygous Homozygous Normal

Severe 14 144 Preeclamptics n=158

Normotensive 17 386 Gravidas n=403

Severe preeclamphcs had a statistically significant higher frequency of the Leiden mutahon than the normotensive controls X2=4.686 p=0.03. CONCLUSIONS: Severe preeclampsia is associated with the factor V Leiden mutation. In the future, DNA analysis for the factor V Leiden mutation may serve as one component of a genetic screening test for preeclampsia.

112 PRETERM LABOR AND MATERNAL ABUSE IN PREGNANCY. J. Shumwa~, A. KhouzamP, P. O’Csmpo=, A. Glalan=, F. Wittor, K.

Blakemors. Depta. of GyrdOb, The Johns Hopkins U,, Sch, Mad. &

Hyg-Pub Hlth., Bairn., MD, and St. Louis U., St. Louis, Me.

OBJECTIVE: The prevalence of preterm labor (PTL) in prenatal

populations has been estimated to be from 6.9 to 10.0%. It has been

suggested that violence during pregnancy may be associated with an

incrsese in antenatal complications. The hypothesis la that physical

violence end verbal abuse in pregnancy lead to increased risk of PTL.

STUDY DESIGN: A cohort of 636 women attending the Adult

Obstetrical Clinic for their first prenatal visit from December 1989 to

September 1990 ware approached of which 667 enlisted as study

participants. Study participants were interviewed three times during

the course of their prenatal care and 401 successfully completed

their third prenatal interview. Violence data were obtained from these

interviews. Obstetrical end neonatal outcome data were obtained by

abstracting the maternal and neonatal medical records.

RESULTS: When stratified by levels of violence, those women who

experienced moderate or severe violence had significant entepurtum

complications of PTL of 18.4% and 17.2% respectively. Chi square

test for homogeneity revealed a significant difference among these

groups. n PTIJn Group %

Group A = no abuse 51134 3.7

Group 8 = verbal abuse 111144 7,6

Group C = moderate violence 10/65 154

Group D = severe violence 10/68 172.

Chisquarex= = 13.23 d.f = 3 p = 0.004

CONCLUSIONS: In our cohort of women, serious acts of verbal abuse

and physical violence occurred with significant frequency. PTL was

strongly correlated with increasing acts of violence with 4,1 times

greater risk of PTL in that cohort of women who experienced severe

viole0ce as compared to that cohort who experienced no maternal

abuse.


113 IN PRETERM FETUSES DECREASED AMNIOTIC FLUID VOLUME IS ASSOCIATED WITH DECREASED FETAL MOVEMENTS. ~, CY Sl)on~*, A Ghldlnl, CM Salafia*, VK Mnaor*. Perlnatal~research Facihlv, "Deoartments of OB/GYN and Pathology, Georgetown University MedlcarCenter, Washington DC and Umverslr¢ of Connecticut Medlcm Center, Farmlngton, CT OBJECTIVE: To examine the relatlonshm between amniotlC fluid volume (AFV) as assessed by biophi.~slcal profile (BPP) and the presence of fetal movements (FM) In preterm tetuses. STUDY DESIGN: From a database of 465 consecutive nonhypertanslve, nondlabetlc patmnts debvermg <32 weeks gestatmn, patients with singleton, nonanomalous fetuses with FM determined as part of a BPP assessment within 24 hours of delivery, irrespective of membrane status (Intact / ruotured), were studied AFV and FM were assessed sonographically. AFV was scored 0-2 according to the following criteria. largest l~ocket in vertical diameter < 1 cm = 0; < 2 but > 1 cm =’1, _> 2 cm = 2 FM were scored over 30 minutes: 0 if absent, 1 if 1-2 movements, 2 if >3 gross (limb/trunk) movements. Variables assessed included, fetal p.,r, esentation, gestatmnal age (GA), and chorloammonltlS. chorioammomtis was dlagnose~ by any 2 of the following criteria: maternal temperature > ~37.8°C maternal tachvcardla >1"20 bpm, leukocytosls _> 20,000/mm~ (withOut prior steroidsJ, fetal tachvcardla > 160 bpm, uterine tenderness or foul smelling amnlotic fluid. Statistical analyses ~ncluded contingency tables anal ANOVA with p<0.05 considered slKnlficant, RESULTS: 352 patients met the inclusion criteria Of these 181 (51%) had premature rupture of membranes and 171 !49%) had Intact memt~ranes. Of the 352 patients 80 (23%) bad AFV=0 60 (17%) had, AFV=I. and 212 (60%) had AFV=2. Of these oatlen~s 12 (3%) hau FM=0 30 (9%) had FM=I, and 310 (88%) had FIVr=2 (Table). There was a s_iP, n~ficant correlation between decreased AFV and decreased FM (p < 0 0C~I)

AFV

2 I ~ I 4~ 12oa Fetal 19resentatlon and GA were not significantly different between patlenfs based on score of FM. Loss of any points f6r either FM or AFV was independent of membrane status. As expected, the incidence of chonoamnlonitis was significantly greater in patients with FM=0 (p< 0.05). CONCLUSION: Decreased AFV is associated with decreased FM lrfes.~ective of fetal presentation, GA and membrane status. The decrease in I’M associated with decreased AFV in prematurlty may reflect modulation of fetal behavior in response to a restricted intrauterine environment.

115 AMNIOTIC FLUID VOLUME IN PRETERM RUPTURE OF MEMBRANES: ASSOCIATION WITH FETAL PRESENTATION AND INCIDENCE OF CLINICAL AND HISTOLOGIC EVIDENCE OF INFECTION. ~, CY Spong*, CM Salafia*, A Ghldlni, VK Minior*. Pennatal Research Facility, Depts. of OB/GYN and Pathology Georgetown University, Washington DC and UCONN, Farmmgton, CT. OBJECTIVE: To assess whether ammotic fluid volume (AFY) fgllowlng preterm rupture of the membranes ~PROM) is associated w.~th fetal presentation or the prevalence of clinical o.r hlstologlc evidence of infection in patients delivering at < 32 weeks gestation. STUDY DESIGN: From an estabbshed database of 465 deliveries < 32 weeks gestation, patients with singleton, non-anomalous fetuses with AFV assessment as part of fetal biophysical profile within 24 hours of delivery, were studied (n=146). Fetal presentation was confirmed by ultrasound wiIli 46 breech and 100 vertex-presentlng fetuses. AFV was scored 0-2 according to. largest l~ocket in vertical ularneter <1 cm = 0; <2 but >1 cm = 1" > 2 cm = 2. PROM was diagnosed by sterile speculum examination. ~limcal chonoamnlonitls was dial.nosed by any 2 of the following cfitena: maternal temperature > 37.8~C, maternal tachy.cardia > 120 bum, leukocytosls > 20 000 mm~ (without p.nor steroids), fetal tachycardla > "160 bpm, uterine tenderness or foul smelling amniotlc fluid. Hlstupathologg examination of amnlon (consldereo representative of maternal inflammation) and umblllcaI cord (reflecting fetal inflammation) were performed by a smgle pathologist blinded to clinical data. Outcome variables included: rupture-to-dehvery interval, g, estational age at delivery, neonatal morbidity parameters (1 ana 5 minute APGAR scores, Incidence of respiratory distress syndrome, bronchopnimonary dysplasia, retlnoEathy ofprematunt~,, neonatal sepsis, days of ventilation an.a hospitalization), anaplacentalnlstologic parameters of acute ascending ~nfection..Statlstical anal~,SlS included con’hngency tables and ANOVA (p <0.05 significant) RESULTS: AFV was significantly lower in breech vs. vert~ presentation ~AFV= 0~in 20 vs. 34; AFV=I in 19 v,s. 27; AFV=2 in 7 v,s.,. , p -S 0.014). No slgnmcant difference was notefl in the rupture-to-flenvery Interval, neonatal morbidity~arameters or histologic evidence of acute ascending refection (50% vs. 42%, p>0.2) between gestations with breech or.vertex oresentatlons The incidence ofclin.lcal clionoammonitis was s!gmflcantlv lower m breech vs. vertex presentlon (40% vs. 60%, p < 0.05), although gestatlonal age at delivery was significantly lower in breech vs. vertex presentation (27,7 + 2.5 vs. 28.5 + T.5 weeks, p=0.04) CONCLUSION: AFV in PROM is significantly lower in breech ~s, verte.x presenting fetuses Since amniotlc fluid at these gestatlona ages is composed~maln[y of fetal unne, we suspect this findlngmay, resulgfrom the proxlmlt~ ot the fetal urethra to the internal os, in nreech-p.resenting fetuses, leaning to increased loss of amniotlc fluid. Conversely,, vertex oresentatlon may enhance preservation of AFV by accumulation otflnid in higher pockets or relative sealing of the internal os by the fetal vertex. The decreased prevalence of clinical chon.’oamniomtls despite equal prevalence of histologic acute inflammation in breech presentation may reflect continued ross of lnflammed ammotic fluid through a "flushing effect".

114 IN DELIVERIES < 32 WEEKS, BREECH PRESENTATION IS ASSOCIATED WITH AN INCREASED INCIDENCE OF INTRAUTERINE GROWTH RESTRICTION. ~, CM Salafla*, CY Spong*, VK Mlmor *. Perinatal Research Facility, Departments of OB/GYN and Pathology, Georgetown University Medical Center Washington, DC and UCONN Medical Center, Farmlngton, CT. OBJECTIVE: Breech presentation and intrautenne growth restriction are each related independently with prematurlty. This study was designed to assess the possible relationship between breech presentation and lntrautenne growth restxlcuon in dehvenes < 32 weeks gestation STUDY DESIGN: From an established database of 465 consecutive dehveries < 32 weeks, of nonhypertensive, nondlabetlc patients with singleton non-anomalous fetuses, those in whom blrthweight, body length, fetal presentation (known to be either vertex or breech) and ammotIc fluid volume (assessed by ultrasound within 24 hours of delivery) had been recorded, were studied. Ammotic fluid volume was scored 0-2 according to: largest pocket in the vertical diameter < 1 cm = 0; < 2 but > 1 cm = 1; > 2 cm = 2. Growth restriction, defined as symmetric if both birthwelght and body length were < lOth cen61e and asymmemc if only the birthweight was < 10th centlle, was studied in relation to maternal age, parity, smoking, fetal presentatmn, amniotic fluid volume, membrane status (mpturedhntact) and gestational age at delivery. Statistical analyses Included contingency tables and ANOVA, with p < 0.05 considered slgmficant. RESULTS: 298 patients met the inclusion criteria In these patients 85 (28 5%) fetuses were breech and 213 (71.5%) vertex. A total of 56 (19%) fetuses were growth restricted Of these, 31 (10 4%) were symmetrically and 25 (8.3%) asymmetrically growth restricted The incidence of symmetric growth restriction ~n the breech-presenting fetuses was 16% (n=!4) vs 8% (n=17) in the vertex-presenting fetuses, and of asymmetric growth restriction 12% (n=10) vs. 7% (n=15), (overall p~-0.03) Intrauterine growth restriction was not associated by significant differences in maternal age, parity, smoking, amniotlc fluid volume, menabrane status, or gestatlonal age at delivery (each p > 0.10) CONCLUSION: At < 32 weeks gestation, intrauterine growth restriction is associated with an increased incidence of breech presentation, independent of clinical confounders. Although the underlying etiology of this clinical observation is unclear, these data suggest that in preterm pregnancies with impending delivery, identification of a breech-presenting fetus should alert physicians to the significantly Increased incidence of growth restriction

116 ELIGIBILITY CRITERIA FOR THE OUTPATIENT MANAGEMENT OF PRETERM PREMATURE RUPTURE OF MEMBRANES (PPRO_M). ,~, PGR..Seg. wsx..~d, D;,,.F__ayin~�, G. Ryan, E Kelly~ & JWK. Ritchie. Permatal Unit, Umversgy o[toronto, CANADA OBJECTIVE: To determine eligibility criteria for outpatient management of PPROM. STUDY METHODS: A retrospective review of all pat!ents with singleton pregnancies <37 wks geslational age [G,AI~n--440) .admitted to, our,center Mth PPROM i.3. 1991-93 was po~_.b.rmea, these ~ were an,aiyzea .using descriptive stalastics, ANOVA, Chl-square and regressmn analyses where appropriate. (*p<.05, **p<01, ***FL:_.00I)

SD RESULTS: TheGAat PPROMwas31.1+4wks(raenn+ ~). The interval from PPROM to the onset of labor-lPOL] was 129+ 248 hrs. Fil~ ~rcent of patients labored within 48 Ins, while 20% t~ma~ed undelivered after 1 wee~. "l~e POL interval was correlated wi.th the GA at PP.ROM, and cervical dilatation and leukocyte count on adwassmn (r = -0.38_*_*., -0 18.*_*,. -0 13"* respeqtivelv) The mean duration of labour was 7.6+_5.3 Ins witta 36 precipitous dellveffds (<2 hrs). Parity, a~d birthweight.were .as .s~jated with thedumtienoflabor(r=..Oll*,O.lg**) C/inica/chorioammomti s was suspected in 112 (25.5%) women of whom 89 (80%) labored spon.laneously. Cord prolapse oocurred in 9 patients Thyee had breech presentalmns agd .of the 5 viable fetuses in cephalic presentations, 4 prolapsed then- cords within 48 ltrs of PPROM. The only fetal loss follq ~w~ag .cora pro.la..pse, ,occt, a’r, ed at 21 weeks GA. The incidence of neonatal refection (positive mona or cerebrospinal fluid cultures withh148 lws of delivery) was low (3.4%) and was correlated with the POL tnterval (p--~0.02). The overall perinalal mortality rate was 6.1%. CONCLUSIONS: The following eligibility criteria for the outpatient management of PPROM are suggested. 1) a 48-72 hr inpatient slay [to obtain culture results and allow for those destined to de/iver <48 hi-s], 2) cephalic presentation,3) no co.rvica.l dilation o.n,stefile ~specul .ran, 4)..n9,, evidence of clinical chorioanmmnltlS, 5)an evlaenco oI m. ecom..ur9, b) u me patient is<34 wks GA she must live or have accommodatmn within a 15 minute drive of a Level 2 center, and 7) pataent competence in self- momtodng In retrospect, usmgcriteda I-5 might have incrensed the perinatal mortality from 6 1 to 6.4% (i.e. one additional loss due to cord prolapse) and potentially saved 598 hospital days. The above criteria are currently being evaluated in.a rand~omized controll .ed ~al of cemmunity based vs. inpatient antenatal care mr women experiencing pregnancy complications

Volume 174, Number 1, Part 2 SPO Abstracts 345 Ain J Obstet Gynecol

117 OUTCOME OF TWIN GESTATIONS COMPLICATED BY PRETERM PREMATURE RUPTURE OF MEMBRANES (PPROM). ~ P G.R. Seaward, D. Fanne, G. Ryan, E. Kelly~ & J.W.K. Rttchte. Vennatal Umt, University of Toronto, CANADA O .BJECTIVE: To compare the characteristics and outcomes of singleton and twin pregnancies complicated by PPROM STUDY DESIGN: In pregnancies with PPROM, twins (n=49) were

compared tot) all singletons (n=440) and, i!) 98 singleton controls (matched ior gestatinnai age [GA] at PPROM, and denvery year). All deliveries occurred from 1991-93. Flsher’s exact test or a two-tailed unpaired t-test were used where appropriate.

RESULTS: The GA at PPROM (mean + SD) for twins was 30.6+3 7 wks. This was similar to the entire singleton population (31.1+4 4 wks~. Selected ~ennatal oatcomes are shown below: -

Twins Singleton Singleto Results A(n--49) B(n--440), n A:B A:C

C (n=98) DelivewGA(wk) 31 2+3.3 31 8+3.6 31.7+3.3 NS NS" Latent Interval 93+202 129£_249 193+361 !0=.054 p<.001

Chorioamnionitis II (22%) 112(25%) 23 (23%) NS NS Cesarean section 18(37%) 91 (21%) 24 (24%) p<.05 NS Birthweight (gin} 1630+519 1848:t:708 1780!-609 p<.01 NS 5 rain Apiar<7 13 (13%) 67 (15%) 12 (13%) NS NS Neonatal Survival 92 (94%) 413 (94%) 92 (94%) NS NS Postpartum 11 (22%) 47(11%) 11 (12%) p<05 NS pyrexla >380C Respiratory 25(26%) 46(10%) 10(10%) p<001 p<.01 Distress (R’D S)

�s were comparex was di.fferent..inci_dence of cord prolapse or net: I infection (positive blood culture in the first 48 hrs) was not different. The overall perinatal mortality for each group was 6%. CONCLUSION: The natural history and outecane of twin and sh~letun pregnancies with PPROM were similar except for a shorter latent interval andan increased incidence of RDS in twillS, The differences in cesarean section rate, birthweight and incidence of postpurtum pyrexia between twins and smgletqns is consistent with the expected higher morbi&ty associated v, ath multiple pregnancy.

119 INCIDENCE OF PLACENTA PREVIA WITH PREVIOUS CESAREAN

David A Miller MD, F~deha G D~az MAx, Richard H Paul MD

University of Southern Cahforma School of Medtcme Los Angeles, CA

OBJECTIVE: To determine the relationship between the number of

previous cesarean deliveries and the risk of placenta prevla.

STUDY DES1GN: Data were gathered prospectively from labor and

dehvery records for the 10 1/2-year period beginning January 1, 1985.

Hospital records of all women with placenta previa were reviewed

RESULTS: Between January l, 1985 and June 30, 1995, there were

157,224 dehveries at LAC+USC Women’s Hospital, of which 624

(0 4%) were complicated by placenta prevxa There were 424 (0 3%)

cases of placenta previa among 140,017 women w~thout a previous

cesarean, and 190 (1 1%) among 17,207 women with at least one

previous cesarean The risk of placenta prevla with increasing numbers

of previous cesareans is presented below.

Incidence of placenta previa

41%

4%

2%

0% 0 1 2 3+

Previous cesareans

All differences statistically sigmficant (P< 0.01). CONCLUSION: The risk of placenta previa doubles with each additional prdvious cesarean delivery

118 WARM TUB BATH DURING LABOR IN WOMEN WITH PRELABOR

RUPTURE OF THE MEMBRANES AFTER 34 WEEKS OF GESTATION.

IS IT SAFE?

L Ladfors, L,A, Mattsson×, M EnkssonX, O Fall~ Dept Ob/Gyn ~stra

Hospital and Molndals Hospital, University of Goteborg, Sweden

OBJECTIVE: To evaluate the influence of bath dunng the first stage of

labor after prelabor rupture of the membranes (PROM) concerning

maternal and neonatal infections, operative del~ve~’ rates and the use of

analgesics

STUDY DESIGN: Healthy pregnant women w=th PROM (n = 1385)

dehvered1-89 (median 15) hours after PROM were managed without

dtgttal exammahons of the cerv=x unhl onset of active labor or at tnduct~on

Of the 1385 women 538 wanted a bath dunng labor (bath group) while 847

d=d not (reference group) Data were prospecttvely collected

RESULTS: The infectious morbidity was low and no differences could be

detected between the groups Chonoamnlomt~s occurred m 4 cases m each

group The frequency of neonates rece~wng antibiotics was 3 9 and 5 0 %

m the bath and reference group respectively Pnm~para in the bath group

used more nitrous oxide (p < 0 001) and paracervlcal block (p < 0 05)

compared to pnm~para m the reference group The use of ep~dural

anesthesia was s~mllar ~n the groups even when pnm~para and mult~para

were analyzed separately Furthermore oxytocm was used more frequently

among prlmlpara m the bath group (p < 0 05) compared to pnm~para m the

reference group The cesarean section rate was lower among pnmlpara

(2 4 %) in the bath group compared to pnm~para (5 1%) ~n the reference

group (p < 0 05) Mult~parous women ~n the bath group were more

frequently dehvered by ventouse (5 8 %) compared to mult~parous(1 8 %)

m the reference group (p < 0 05)

CONCLUSIONS: Bath dM not increase the risk of mfecUon for the mother

or the new-born after a prolonged latency penod and tub bath Although

these women were not randomly allocated to bath or not ~t may be

concluded that tub bath as a method to substitute other analgesics maybe

overstated today

120 RUPTURE OF THE UNSCARRED UTERUS

David A Miller MD, Pochard H Paul bAD

University of Southern California School of Medicine Los Angeles, CA

OBJECTIVE: To report a 12-year experience with uterine rupture in women

without previous uterme surgery.

STUDY DESIGN: All cases of uterine rupture were identified prospectively

Hospital records of women with uterine rupture without previous utenne

surgery were reviewed.

RESULTS: From January I, 1983 through December 31, 1994, 188,819

women delivered at LAC+USC Women’s Hospital, 20,328 (10.8%) of whom

had at least one previous cesarean delivery. The incidence of uterine rupture

in this group was 0.7%. Uterine rupture occurred in 11 (0.007%) of 168,491

women wllhout a previous cesarean. Complete hospital records were available

on 10 The mean age was 29, mean parity 2, and mean gestational age 38 4

weeks (range 28-43). Parity was greater than 3 in only one case None had

had previous uterine surgery. Three ruptures occurred following motor vehicle

accidents, two resulting in fetal expulsion into the maternal abdomen, and one

in fetal and maternal death. Seven ruptures occurred during labor, and are

summarized below Labor was spontaneous in three cases (one with twins),

and induced m four with prostaglandin E2 (1), oxytocm (2), or both (1)

Prostaglandln dosage did not exceed 3.5 mg, and mean duration of oxytocin

use was 4.5 hours (range 0 5 - 8 9 hours). No protraction or arrest disorders

were identified Four women underwent cesarean for fetal distress; 3 delivered

vaglnally (two with vacuum assistance). Among the latter, all underwent

laparotomy for post-partum hemorrhage Mean blood loss was 1985 co; 5

required transfusion (average 3 4 units PRBC). Five-minute Apgar scores

were < 7 in 2 cases, perlnatal asphyxia was diagnosed in one Mean

birthweight was 3697 gm (range 3320-4200gm) Fetal expulsion into the

maternal abdomen occurred in one case There were no maternal or neonatal

deaths in this group

CONCLUSIONS: Rupture of the unscarred uterus is a rare and potentially

catastrophic event, occurring in < 1 in I0,000 deliveries In this series, 5 of 10

women had no identifiable risk factors. Among the remaining 5, risk factors

included abdominal trauma, grand-multiparRy and multiple gestation.

346 SPO Abstracts January 1996 AmJ Obstet Gynecol

121 THE USEFULNESS OF COAGULATION STUDIES AND BLOOD BANKING IN THE SYMPTOMATIC PLACENTA PREVIA DAWinn. RH Paul, LK Millar, Dept of Ob-Gyn, University of Southern California School of Medicine, Los Angeles, CA OBJEDTIVE: To ascertain the usefulness of coagulation studies and blood banking utilization in the patient with symptomatic placenta previ~. METHODS AND MATERIALS: During the period July 1, 1993 through August 15, 1995, 87 patients with placenta previa ~24 weeks’ gestation were evaluated for vaginal bleeding on labor and delivery at Women’s & Children’s Hospital. Routine laboratory evaluation included complete blood count with platela count, fibrinogan, prothrombin time, and Kleihauer-Betke test. A blood type and screen was also sent to the blood bank, and blood availability was maintained for

all patients throughout the course of hospital*zafion. Analysis of the laboratory values was performed and need for immediate blood availability was asce~ained. Chaa review provided comp/ae data in 84/87 patients. RESULTS. There were no abnormal prothrombin times or positive Kleihauer- Bake tests in the 84 patients admitted with bleeding associated with placenta prevla. There was only 1 abnormal fibrinogen value in a patient in whom the complete blood count was within normal hmits The range of initial hematocrits was 21.1-40.0%, and the initial hemoglobins was 7.0-14 1 mg/dl. There were five patients with platela counts of ~150 k/ram3 (range 75-149 k/mrrd). Overall,

13 patients (14.9%) required transfusion with only 2 required antepactum. One sub.lext was transfused 1 unit of packed red blood cells prior to transfer to our facility. The second decision to transfuse was made after 6 days of observation. The other 11 tmusfusioas were made intra- or post-operatively, Two patients required massive transfusion, orm had a placenta accreta and the other underwent reexploration for hemopetitoneum and a rectus sheath hematoma. During this 2- year period, a single patient was taken for emergent deliver] within 1 hour of presentation to the hl~spital. However, this patient delayed seeking medical attention for >3 i~ours CONCLUSIONS: Routine evaluation of coagulation studies in the symptomatic bleeding placenta previa is unlikely to yield abnormal results. Massive blood loss would be required for a coagulopathy to he demonstrable, and coagulation studies shou/d not be obtained unless clinically indicated. Mandatory blend availabihty for the hospitalized patient with placenta previa appears unnecessary, as must patients did not require emergent blood tmnsfnsmn. When transfusion was necessary, there was usually ample tame to cross-match the blood products.

123 PREGNANCY RELATED COMPLICATIONS IN ’TWIN GESTATIONS RESULTINO~FI~OM

MULTIFETAL PREGNANCY REDUCT!ON AS COMPARED TO NON-REDUCED ’I~/IN AND TRIPLET PREGNANCIES A. Kow~ikx, M Smith-Levit~nX, F Moyx, M. H~tson, Z

Rosanwaksx, F. Chervenak OspL of Oh/Gyn, NYH-COr,’~, NewYork, NY.

OBJECTIVE: To determine whether mult~fetal pregnancy reducbon (MFPR) lowers the risk

o~ pregnancy re~ed complica~ons seen in ~iplet pregnancias to that normeily associated

~ ~n gasmeoas

STUDY DESIGN: The medical records of 214 pa~anto who conceived as a re~t of

ovule~on induc~on or in-vitro fer~liz~,ion an~ who de~vere~ beyoed 24 weeks ~esta~on

were reviewe~ to obtain ber~n~’~t medkat t~story The at~y popu~on con~sted of 60

pa~anto who under, ant MFPR to a twin pregnancy, 88 pa~ento v, ith a twin gascon and

54 p~ants vA~ ~ The rates of pretann labor (PTL), preterm dei~ve~ (PTD),

gantatJonal diabetes, and preedampsm (PEG) were compared. Stabs~cal analy~s was

performed u~ng ~ test and student t-last when appropnata.

RE~JLTS.’ The pe~ who underwent MFPR were less kkely to develop preedamps~n

end to deliver preterm than their t~plet counterparts. Trip~ were sta~s’dcally more ~dy to

ddiver between 28 to 32 weeks, There was no sta~s’~call’/s~n~cant difference in the rates

of preterrn lab=x and praterm delivery between MFPR and the t~n groups. There was no

difference in the rates of preterm premature rupture cf rnernbfanas or gastationd diabetes

between any # the ~udy 9roups.

{N)CATEGORY PEC PTL PTD PTD [ PTD 24-28~s 28-32~d~s I 32-36~s

MFPR (601 I 8113%) 11(18°/,) 2(3.3%) 315%) I 18130%)

T~"S I~) TRIPLETS (54]

Wp=0.02fortdpletsvs. MFPR ~ p=0003fortn#atsvs MFPR

~p=0001 forlnpletsvs t~ins ~p=0.001 forthpletsvs.MFPR and tvdns

CONCLUSION: MFPR lowers the rL~k of preedampsia ~ pretsrm labor and ddlvery in

higher order ge~oas to that norma~y seen in trans. Nthough the rains of extrame

prematunt~ ere not decreased, the

lowered.

122 RISK OF FETAL DEATH BY GESTATIONAL AGE IN TWIN AND

SINGLETON PREGNANCIES. AM Peaceman, J Bonebrake~, PM Gareia, S WaW Depts of Ob/G3,n and Pad, Northwestern Univ Mad School, Chmago, IL

OBdECTIVE: Elective delivery of twin pregnancies at term is commonly recommended because of concern for fetal death or deterioration occurring earlier

in twin gestations relative to singleton pregnancies. We examined national birth and fetal death data to determine if the risk of fetal death increased in twin relative

to singleton pregnancies at or near term METHODS: Natality data from National Center for Health Statistics tapes of all U.S live b~rths and fetal deaths aiter 20 weeks’ gestutmn an 1986 were obtained.

After separating singleton from twin pregnancies, fetal death rates for each group

were calculated for each gestatmnal age based on the number of fetal deaths divided by the number of live births at the same or greater gestatmnal ages.

RESULTS: During 1986, there were 3,448,221 live births and 22,960 fetal deaths

(6 6/1000 live btrths) reported for singleton pregnancies Fortwmgestations, there

were 75,070 live bu’ths and 1,868 fetal deaths (24 9/1000 live births) reported. For

both groups, fetal mortality reached its nadir by 30 weeks’ gestation, and began to

rise by 36 weeks (fig 1) When the data are expressed as a relative risk (RR) for

fetal death In twin compared to singleton pregnancies (fig 2), twin pregnancies were at increased ask at each gestatianal age However, the RR of fetal death for twins compared to singletons did not increase after 37 ~veeks’ gestation.

CONCLUSION: 1) A sigmficantnse m the rate of fetal death m twm pregnancies occurs attar 34 weeks’ gestation, while a rise in the rate of fetal death in singleton

pregnancies is not evident until 39 weeks; 2) While the risk of fetal death in twin gestations is greater than the ask in singleton pregnancies at all gestationnl ages, the RR does not increase further at term,

Fetal Death Rates

Gestatlonal Age (Weeks)

RR of Fetal Death

Twins vs Singletons ~

24 26 28 30 32 ~ 36 38 40 42

Gan~Bonal Age (weeks)

124 COMPARISON OF RIRTHWEIGHTS OF TWIN GESTATIONS

RESULTING FROM EMBRYO REDUCTION OF HIGHER ORDER

GESTATIONS TO BIRTHWEIGHTS OF TWIN AND TRIPLET

GESTATIONS USING A NOVEL WAY TO CORRECT FOR

GESTATIONAL AGE AT DELIVERY M Sreith-Levitinx, A

Kowahkx, J Blrnholzx, M. Hntson, D. Skupski, Z. Rosenwaksx, F

Chervanak Dept ofOb/Gyn NYH-Cornell, New York, NY.

OBJECTIVE: To evaluate the effects of embryo reduction (ER) on

birthweights (BW) of the remaining fetuses

STUDY DESIGN: The birthweights and gestational age at delivery of all

multiple gestations conceived with assisted reproductive technologies over a three and a half year Period at NYH were obtained from delivery logs.

Patients delivering anomalous fetuses, prior to 24 weeks, or at other

hospitals were excluded. All embryo reduetaans Performed prior to 12 weeks that left 2 fetuses were included. The birthweights were then

corrected for gestational age at delivery to the weight expected at 36 weeks,

which was obtainod from composite standardized growth curves. Student’s

T-test was then used to assess significance

RESULTS:

Category (N) EGA~_SD Actual BW.’tSD Corrected BW:t:SE

Twins (81) 36.1+27 2532+517 2461+344"

ER (55) 35 4± 3.1 2330± 607 2435± 386*

Triplets (45) 33 2±2.8 1906±449 2300±376*

*ER vs. Triplets p< 0 006 *Twins vs. Triplets p<0.0001

CONCLUSIONS: For multafatal pregnancies in which embryo reductlon is performed prior to 12 weeks, subsequent birthweight of the remaining twin

gestation is similar to twins and significantly larger than triplets This

suggests that factors, such as uterine crowding later in pregnancy, may play

a greater role in fetal growth of twins and triplets than early implantation and plan�elation.

Volume 174, Number 1, Part 2 SPO Abstracts 347 Am J Obstet Gyneco]

125 REDUCING FETAL DEATHS IN MULTIPLE BIRTHS: OPTIMAL BIRTH WEIGHTS AND GESTATIONAL AGE. B. Luke, Dept. Ob-Gyn, Univ,

Michigan Mad. School, Ann Arbor, Michigan.

OBJECTIVE: To estimate the birthweight and gestat~enal age associated w~th the lowest feta~ death rate for muff~p~e births.

STUDY DESIGN: Th~s study was a population-based analysis of all

births and fetal deaths (FD) in the US between 1983-86. Births and FDs were categonzed as singleton, twin, or tnplets and h~gher order (triplets).

From data on the FD and b~rth certificates, weeks’ gestation were

calculated from the last menstrual period and categorized into two-week intervals; b~rthwe~ght was categorized ~nto 300 g-~ntervals. The data was

cleaned to eliminate FDs and bvths w~th biologically-implausible

b~rthwe~ght-gestational age combinations. For twins and triplets, the lowest FD rates by gestat~onal age, by b~rthweight, and overall were

compared to those of singletons as relatives risks (RRs).

RESULTS: The fina~ study sample included 12,888,973 singleton i~ve births and 82,156 singleton FDs; 281,506 twin live births and 6,498 tw=n

FDs; and 6,222 tdplet hve b~rths and 197 triplet FDs. The overall FD rate per 1,000 conceptions was 6.3 for singletons, 22.6 for twins, and 30.7

for tnplets, w~th a RR of 1.0 for singletons, 3 6 for twins, and 4 9 for

triplets. The lowest FD rate for singletons was 40-41 weeks (1.8/1,000),

compared to 36-39 weeks for twins (8.9/1,000), and 34-37 weeks for

tnplets (9.1/1,000). At these optimal gestat)onal ages, the RR for FD was 1.0 for s~ngletons, 4.9 for twins, and 5.1 for triplets The lowest FD

rate for singletons was at 3400-4300 g (1.2/1,000), compared to 2800-

3700 g for twins (1.9/1,000), and 1900-2800 g for tnplets (3.3/1,000). At

these optima~ bir~hweights, the RR for FD was 1.0 for singletons, 1 6 for

twins, and 2.8 for triplets.

CONCLUSIONS: The FD rate for ~nfants of multiple gestations can be

greatly reduced with attainment of b~rthwe~ght and gestational age within

an optimal range. For twins, this optimal range is 2800-3700 g at 36-39 weeks, and for triplets it ~s 1900-2800 g at 34-37 weeks. Attainment of

b~rthwe~ght w~thm the optimal range appears to be of greater beneffl in narrowing the d~fference m FD rates between singletons and multiples

than does attainment of gestat~onal age.

127 OUTCOME OF TWIN PREGI~L~NCY. FOI.LOWI, NG IN VITRO FERTILIZATION. J.W. BernRsko~, Lauren Lynch, Department of Obstetrics, Gynecology and Reproductive Science. Mount Sinai Medical Center, New York, NY. OBJECTIVE: To determine whether twin pregnancies achieved after in-vitro fertilization were at increased risk for obstetncal complications or perinatal morbidity. STUDY DESIGN: The computerized perinatal database was reviewed for all twin pregnancies managed by private obstetricians and delivered at MSH between 1990-1994. The obstetric and neonatal characteristics of those conceived following IVF were compared to those who conceived spontaneously using Chi-square analysis. RESULTS: There were 108 twin births following IVF and 298 spontaneously occurring twin pregnancies. The following were statistically more frequent in the IVF group; Maternal age >35 years (47.2% vs 32%); Nulliparity (83.3% vs 54.6%); tocolysis for preterm labor (40.7% vs 25.4%) and discordant fetal growth (26.9% vs 13.8%). There was no difference in the frequency of medical complications in pregnancy, low birth weight, preterm rupture of membranes, preterm delivery (62.9%-IVF vs 57%-nonlVF), fetal growth retardation, mean gestational age at delivery (35 4 + 3 weeks -IVF vs 35.9 _ 3 weeks-non-IVF), intrapartum complications, NICU admissions, RDS, IVH, NEC,sepsis,hyperbilirubinemia or death.

CONCLUSION: Although twin pregnancies following IVF are treated more frequently for PTL the perinatal outcome is comparable to that of spontaneously occurring twin pregnancies.

126 CASE-CONTROL STUDY OF TWIN GESTATION SUBSEQUENT TO

MULTI-FETAL REDUCTION, R Silver, B Helfand=, T Russellx, A Raginx

J Shell. Dept. of Ob/Gyn, Northwestern University Medical School,

Evanston Hospital, Evanston, llhnms.

OBJECTIVE: To compare pregnancy outcome in twin gestations resulting from mu|ti-fetal reduction {MFR;’ to "primary" twin

pregnancies, derived from spontaneous conception 8rid artificial

reproductive techniques (ART).

STUDY DESIGN: Between 8/90 and 11/94, 18 multi-fetal pregnancies

(> quadruplets) underwent transabdominal MFR to twins (Group A).

Control twin pregnancies consisted of two groups: spontaneous twins

(Group B) and those achieved through ART (Group C). Groups B & C

were derived separately by identifying and including the next

spontaneous end the next ART twin gestations that delivered

following each Group A delwery date.

RESULTS: Comparing antepartum comphcations between Group A vs

Groups B & C revealed similar frequencies of preterm labor, membrane

rupture and hypertension. Pregnancy outcome comparisons are

shown: Group A Group B Group C p

N 18 18 18

G,A. @ DEL 33.1+__3.5 31.2+__5.O 36.0+__2.3 .002"

BW-Twin A 1959+__784 1731~+823 2602+__614 .003"

BW-Twin B 1714+_,.857 1606-t-812 2540+576 <.001"

IUFD 0/36 6/36 0/36 <.05#

(* Avs C; #A vs B)

Group B twins included 4 monochorionic sets (compared to 0 and 1

in Groups A & C, respectlvelv), that were associated with twin-twin

transfusion, IUFD and extreme prematurityo

CONCLUSIONS: MFR pregnancies were delivered at s=gniflcantly

earlier gestationat ages and of correspondingly smaller infants

compared to "primary" ART twins. We speculate that the res=dual

nonviable fetuses and placentas shorten the length of gestation in MFR patients.

128 PERINATAL OUTCOME IN TRIPLET PREGNANCIES G. Kaufman=, D. Chelmow, A. Penzias, C. Cetrulo M. D’ARon, New England Medical Center, Tufts University School of Medicine, Boston, MA

Objective: To describe the pednatal outcome for a cohort of triplets managed at a single maternal fetal medicine center. Study Design: Data from 43 triplet pregnancies delivering in a single center over a 31 month period was collected and analyzed. All triplet pregnancies reaching 12 weeks of gestation were included. Results: One hundred nine triplets resulting from 43 triplet pregnancies survived until hospital discharge. Twenty six (61%) of the pregnancies resulted from in vitro fertilization or gamete intrafallopian transfer, ten (23%) resulted from ovulation induction, and seven (16%) occurred following spontaneous conception. There were three intrauterine demises and seventeen neonatal demises. This corresponds to an 84.5% survival rate. Sixteen of the seventeen neonatal demises resulted from extreme prematurity (19 - 26 weeks). Of the surviving newborns, 10 (9.2%) of the suwivors had one or more significant complications of prematurity, and g9 (90.8%) of the survivors had no significant complications. The average gestational age at delivery was 30.8 weeks (range 19 - 37). Nine of the pregnancies delivered at < 28 weeks (21%), eleven delivered from 24 thru 27 weeks (26%) and 23 delwared at greater than 32 weeks (53%). They two of the fetuses received surfactant therapy. Conclusion: Outcome data from a large cohort of triplet pregnancies managed since the availability of surfactant therapy is useful when counseling patients about reproductive options of in vitro fertilization and mdltifetal reduction.

348 SPO Abstracts January 1996 Am J Obstet Oyneco]

129 FETAL LUNG MATURITY IN DIABETIC PATIENTS USING THE TDx FLM ASSAY. N, Erik.sen. A. Toy*, J. Prieto*, E. O’Sullivan*, S. Wong*, £ Blanco. Dept, Ob/Gyn, UT-Hoaston, LBJ Hospital, Houston,TX OBJECTIVE: To compare the accuracy in near-term diabetic patients of the TDx-FLM assay with the lecithin-sphingomyelin ratio (L/S) and phosphatidyl glycerol (PG) determinations in predicting respiratory distress syndrome (RDS). STUDY DESIGN: We prospectively studied 103 diabetic patients belween July 1993 and June 1995 who had an amniocentesis (94) or vaginal pool (9) tmniotic fluid (AF) col|~ted for clinical indications. The mean geslational

age was 38.1 ¯ 2.2 and all patients delivered within 72 hours. All AF were evaluated by the TDx FLM assay and chromatography for L/S and PG. Pulmonary malarily was defined as an FLM > 70 my/g, L/S > 2 and PG > 0.3. Neonatal outcome was analyzed for the rate of RDS. RESULTS: The predicted and actual eases of RDS along with the sensitivity (Sen), specificily (Spot), positive (PPV) and negative (NPV) predictive values for all 3 tests are listed below.

FLM L/S FG

Predicted/Found

RDS/RDS 1 1 1 RDS/no RDS 8 12 20 no RDS/RDS 2 2 2 no RDS/no RDS 92 g~ gO

Sen 33% 33% 33% Spec 92% 88% 80% NPV 98% 9g% 98% PPV ll% ~%

Three term infants developed RDS . In 2 of the 3 cans, all three tests wore mature and one infant had ~,11 3 tests immature. The ~t~ of RDS following

~ imma~re ~st was not signifi¢~fly ~iff~ent wh~ ¢~ng ~� FLM to

L/S (p~l.0) ~ FLM m PG (p~0.5).

~N~US[O~: ~e aegt~ve ~ictive valse of ~e TDx FLM test in

ne~-~ di~efics is ¢~p~able ~ ~e ~S ~d PG.

131 DOES ULTRASOUND PREDICT SUCCESS OE EXTERNAL CEPHALIC VERSION? ~AF Borgida. H Hardardottir, M Bork, and JFX Egan St Francis Hospital, Hartford, CT and Univ. of Connecticut Health Center, Farmington, CT. OBJECTIVE: To determine if ultrasonographic parameters are associated with successful external cephahc versmn (ECV). STUDY DESIGN: Women who were referred for ECV at or near term had an ultrasound, terbutaline tocolysm, and up to three versmn attempts during one visit. Data collected on each pregnancy included: estimated fetal wmght (EFW), arnniotic fluid index (AFI), placental positron, fetal spine position, breech type, and maternal characteristics. Data were analyzed by ANOV, K, Student t-test, and Chi-square w~th p<,05 significant. , RESLrLTS: Of I27 referred women, ECV was attempted in 110, ECV was successful in 41/110 (37%) and 85% of these resulted in vaginal deliveries, ECV was unsuccessful in 69/110 (63%) and 96% of these underwent cesarean. Frank breech and anterior placenta were assocmted w~th unsuccessful ECV.

successful unsuccessful EqV ECV P

Number of patients 41 69 Mean gest age (wks~D) 37.6+1.3 37.7+1.0 ns Mean AFI (c~n~D) 11+3.7 11+_3.9 ns Mean EFW (grns~D) 3014+644 3004+-376 ns Anterior placenta 19.4% 80.6% .02 Frank breech 18.9% 81.1% .001 Anterior fetal spine 32% 68% ns Vaginal delivery, rate 85% 4% .0001

CONCLUSIONS: Ultrasonography was not useful in evaluating which ECV attempts would be successful. An anterior placenta and frank breechpresantation were both significantly associated with unsuccessful I~CV which conflicts with previous reports. There was no association between amniotic fluid volume, EFW, and position of fetal spine and ECV outcome.

130 THE EFFECT OF BREECH PRESENTATION ON UMBILICAL ARTERY ACID-BASE STATUS AT BIRTH K Z nger×, E Guzman, W Scorza, A Vmtmleos, C Ben=to* O~ws~on of Maternal-Fetal Medicine, UMONd-Robett Wood Johnson Medmal School, St. Peter’s Medical Center, New Brunswick, NJ. OBJECTIVE: To determine the impact of breech presentation and length of tirst stage of labor on umb~hcal artery add-base status at btrth m term fetuses. STUDY DESIGN: The study population of th~s retrospective study consisted of three groups of term fetuses dehvered by cesarean section 1) Group I in=60) consisted of breech presenting fetuses delivered by electwe cesarean section pnor to the onset of labor, 2) Group II n=44) consisted of breech presenting fetuses de vered electlvey by cesarean section after 4 hours, between 4 and 8 I~urs, an6 a~ter 8 or trite hours of labor ar~ 3) Grou9 Ill (n=60) consisted of cephalic presenting fetuses dattvered by elective repeat section prior io ~he onset of labor Umbihcal artery cord blood was collected at birth for pH, pCO2, pC2, HCOa, and base excess Abdominal incision to detwery times were determined In breech presantlng fetuses expenenclng labor, the cervical dilatation preceding cesarean section was documented and the duramen of labor was d~v~ed into 3 groups <4 hours (hrs), 4-8 hrs and >8 hrs Stat~stica~ analys~s was performed ustng the ManrbWhltney U test, Speam~n rank correlation, and Fmher’s exact test where appropriate with statistical s[gn[ticance defined as p<0 05. RESULTS: There were no significant differences in type of anesthesia, gestational age, or skin recision to delivery times among the three groups. No elgniticant d~fferances were found among umbrilcal artery acid-base values between breech and cephahc fetuses delivered by ele~ve cesarean sec~on prior to the onset of labor. Also, there were no significant differences ~n umblhcal artery acid-base values between breech fetuses delivered by elective cesarean section and those dehvered during the firat stage of labor In group II fetuses, there were no significant differences In umblhca~ artery acid-base values in the 3 labor duration groups, nor was there a correlation between umbihcal artery ac~d-basa values and the degree of cervical dilatation CONCLUSIONS: Breech presentation at term, in and of ~tself, has no effect on umbilical artenal anld-base status. Neither the duration of nor the degree of cervical dilatation dunng the first stage of labor had a s~gnri~cant association with umb~hcal artery acid-base values in fetuses presenting as breech Prewousty reported d~fferencas ~n umbilical artery blood gas values after vaginal dehvery between breech and caphal~c fetuses may be due to events occurring dunP,9 ~ second stage of labor.

132 FETO-MATERNAL HEMORRHAGE AFTER EXTERNAL CEPHALIC VERSION: AN OVERESTIMATED RISK. M Bouche[, G.P. Marquette, D. Rinfret~, J. Varin*. Dept Ob/Gyn, UniversiW of Montreal and Sa{nte-Justine Hospital, Montreal, Canada. OBJECTIVE: TO assess the frequency and volume of fetomaternal hemorrhage (FMH) during external cephalic version (ECV) of term breech singleton fetuses.

STUDY DESIGN: From 11/87 to 01/94, ECV for term breech presentation was attempted on 737 patients (success rate: 53.9%). clirlical data was prospectively collected. Maternal blood was obtained in all but 36 patients for acid elution test

(Kleihauer-Betke) before and at the end of ECV. Qualitative and quantitative results were correlated with clinical data. statistics were done by x~ and t-test when appropriate.

RESULTS: Maternal blood was positive for fetal erythrocytes

(FRBC) in one or both tests on 85 occasions (12%). However 63/85 were already positive before ECV (74%). A negative pre- ECV test was followed by a positive post-ECV test in orlly 22 patients {3.1%). A post-ECV FMH volume > 1.0 ml was observed in 35 patients but 27/35 (77.1%) had a significant amount of circulating FRBC before ECV. Therefore, in only 1.1% (8/701) of cases was an Ecv-related FMH _> 1.0 ml observed (1.5 to 80 mi) No relationship CoUld be established between placentatJon ornumberofattempts and frequency

or volume of FMH. CONCLUSION: ECV-related FMH rates have been previously overestimated. We report a FMH frequency of 1.1%. A)SO, no clinical factor could be identified as influencing the frequency and/or volume of FMH. Despite a low frequency of post-ECV FMH, when it Occurs, significant volume can be transferred. ConsequenUy, Rh immunoprophylaxis remains indicated after ECV in Rh-negative women.

Volume 174, Number 1, Part 2 SPO Abstracts 349 Am J Ob~tet Gynecol

133 FAILED EXTERNAl. CEPHALIC VERSION AND TERM BREECH DELIVERY. L.Couttrtpa S Krmtlow~, V. Lupo tICMC, Mlnneapohs, MN

OBJECTIVE:To evaluate the maternal and fetal outcome according to

dehvery route in patients w~th a term breech delivery faihng external

cephahc version (ECV) and offered a trial of labor in our population.

STUDY DESIGN: 54 patients (36 to 42 weeks CA) with failed ECV

were counselled regarding a vaginal dehvery. CT pelvimetry was routinely

obtained m pts desiring a trml of labor. Maternal and neonatal charts

were reviewed for outcome data Normative data was assessed by the

student T test or Mann WMtney. Proportions were compared by the

Flsher~s exact test.

RESULTS: 8/34 (15%) with failed ECV spontaneously converted to

vertex prmr to dehvery. 8/54 (15%) dechned CT pelv~metry and elected

to proceed w~th csectton. 10/46 (22%) had inadequate pelv~metry; nine

underwent primary csection and one presented completely dilated and

delivered vagmally. 10/28 (36%) with adequate pelvlmetry and breech

upon presentatmn to L&D elected to have csection (patient preference

N=6, Non frank presentatmn N=4). 72% (13/18) of attempted trtal of

labor dehvered vagmally. 28% (5/18) underwent csection for abnormal

labor curves. EBL and hospital days were s~gnif~cantly greater m patients

w~th an abdominal delivery. There was no difference m 5 minute apgars, cord ph, b~rth weight, or endometr ttis. Two fetal comphcattons (bilateral

hip dtslocation and 2 cm buttock laceration) occurred in the csection

group (NS).

CONCLUSIONS: 41% of failed ECVs in our population will have a

successful vaginal delivery, 36% of which wall be secondary to

spontaneous verston to vertex. Although sigmf~cant morbidity ~s not

apparent in the select term vaginal breech dehvery, small sample size

limits firm conclusions

135 BIRTH WEIGHT PRED(CTIONS OF SHOULDER DYSTOCIA INJURY AND

NEONATAL DEPRESSION DR Bryantx, SF Bottoms, MR Leonardlx, JB

Landwehrx. Department of Obstetncs and Gynecology, Wayne State

University, Hutze~ Hospital, Detrott, Mtch~gsn.

OBJECTIVE: To determmne the rmsk of shoulder dystocla injuries and

neonatal birth depressmon according to birth weight.

STUDY DESIGN: We reviewed the records of 20,034 vaginal dehvenes,

excluding pregnanmes comphcated by diabetes, at our hospital from

1991 to June, 1995, The frequencies of clavicular fracture, neonatal

depression (5 minute Apgar score of less than 7), and bracNa’ plexus

injury were analyzed according to birth weight.

RESULTS: The incidence of shoulder dystoma mNry (0.2%) was similar

to other reported series.

We~.qht N_ Brach( % ~ ~ 5mlnAp9ar < 7

<3499 14996 3(0.0) 6(0.0) 190(1.3)

3500-3999 4050 8 (0.2) 6 (0.2) 11(0.3)

4000-4499 884 8 (0.9) 6 (0.7) 7(0.8)

4500-4999 84 3 (3.6)* 0 (0.0) 4(5.8)"

>5000 20 2(100)" O(O.~ 2(10

20034 24 18 214

"p<O.O01

CONCLUSIONS: Btrth weights greater than 4500 grams predict

increased risk of brachla~ p~exus injury and neonatal depression, but not

clavicular fracture. However, because 80% of brachlal plexus injuries

are transient, 139 cesarean deliveries would be required in the 4500-

49999 group to prevent one case of lasting Erb’s palsy. For greater

than 5000g, 50 cesarean delivenes would be required to prevent a

single Erb’s palsy. We believe these factors to argue against routine

elective cesarean delivery for infants 4500-49999,

134 CAN SHOULDER DYSTOCIA BE PREDICTED~ DF Lew=s., MS Edwards, T. Asrat, C. Lockhartx, MT Fontenot, SN London’, LSU-Medmal Center, Shreveport, LA, Long Beach Memorial, Long Beach, CA OBJECTIVE: Some authors have suggested that shoulder dystoc~a (SD) IS predictable; therefore, It can be prevented. This study wasdeslgned to evaluate the predictability of SD STUDY DESIGN: All singleton patients w~th cephahc presentatmns dehverlng vagmally during a 12 month period were analyzed. Various preconceptual, antepartum, and intrapartum factors prewously identified as risk factors were evaluated between cases (S D.) and controls. Macrosomm was defined as birth weight greater than 4500 g. RESULTS: 99 cases and 1523 controls were ~dent~fied. No differences in maternal age, EGA, race or use of epidural anesthesia were noted. Factors found not to be slgmhcantly different between the two groups Included obesity, multlparlty, hx. of diabetes, short maternal stature, postdates, forceps deliveries, advanced maternal age, and induction or augmentation of labor. Factors that were significant included:

Cases Controls p Value OR (95% CI) Preve. Macro > 4000 g 11 76 0.02 2.4 (4.8 to 41.9)

~-45OO g2 23 0.46 1 4 (0.3 to 6.0) Previous SD -- 3 4 O.OO7 11.8 (2.1 to 63) Dmbetes 5 16 0.007 5.O (1.6 to 15.O) Fundal Ht.(FH) > 4Ocm29 214 0.OOO1 25(1.6to4.11

~’38 cm 60 662 0.OO1 20 (1.3 to 3.1) Weight. Gain >---42 Ibs 25 248 O 02 1.7 (1.1 to 2.9) Prolonged 2nc]-Stage 10 62 0.,O~1 2 7 (1.2 to 5.6) Macrosomla > 4000 g 35 73 10 10.9 (6.6 to 17.9)

"~’4500g 8 10 10.8 13.3 (47to 37 6) If FH =s exctu~Tg’d, 35% (35 patients) of SD cases and 22% of controls (337 patients) had significant nsk factors, predicting SD m only 9,4% (35/372) of the patients at nsk CONCLUSION: Many factors beheved to be associated w~th SD were not found to be significant m our study. Macrosom~a appears to be the most significant factor assomated with SD. However, this vanable cannot be accurately estimated prior to dehvery. Factors that may be helpful include: prewous SD, dmbetes, large fundal height, excessive weight gain and prolonged second stage of labor. We beheve that th~s data suggests SD even in the presence of slgmhcant nsk factors remains largely unpredictable. If we were to attempt to prevent SD by prophylactic surgical Intervention In all patients with risk factors, 91% (337/372) of these patmnts would undergo unnecessary cesarean delivery.

136 PREDICTORS OF FETAL MACROSOMIA. CW Benitm, CF Tbayerx, , MF Lakex, IRA Knuppel, AM Vintzileos. UMDNJ-Robert Wood Johnson Medical School/St Petees Medical Center, New Brunswick, NJ. OBJECTIVE: The purpose of this study is to construct a model to predict fetal rnacrosomia based on maternal demographics, medical history and neonatal factors STUDY DESIGN’ Terrn singleton btrths occurring from 1/1/90 to 12/31/90 in New

Jersey (n=103,425) were analyzed. The nurnber of rnacrosorntc neonates (> 4500

Urn) was 2,132(2.1% prevalence. Birth certificate data was used to develop a frequency distribut)on of neonatal birth weight Stepwise regresston analysts was used to determine the best prediction rnodel for fetal macrosomia considering the following independent vanables: rnaternal age, educatton, gravity, parity, weight gain, previous macrosornic b~rth, diabetes, race, pregnancy induced hypertension, chronic hypertension, meternal heart disease, Kessner ~ndex(mathematlcal rneasure of prenatal care), gestational age at dehvery, infant gender, congenital anomahes and meconlum aspiration. RESULTS, F~fteen ~ndepeedent variables c~ntnbuted significantly {P<O 05) to this model wdh a 76.2% correct positive prediction. Curnulattve prediction of the s=gntficant independent variable are hsted below. Three variables: gestational age at delivery, prewous macrosomlc birth and rnaternal weight gain > 30 Ibs predicted 71.3% of macrosom=c b~rths Independent Variables % Correct Positive % Correct Negettve

Prediction Prediction

Gestabonal age 52.2 20 4 Prewous rnacrosomic birth (> 4500 gm) 55.1 19 3 Weight Gain (> 30 Ibs) 71.3 27 3 Parity 731 26 0 Tobacco Use 74.0 25.0 Diabetes 74.6 24.4 Maternal Age 751 24.3 CONCLUSION’ This model correctly identified over 70% of rnacrosornic fetuses using only three maternalffetal factors This may be helpful in identifying wornen at risk for macrosornic b~dhs

350 SPO Abstracts JanuatT 1996 Am J Obstet G},necol

137 PERSISTENT BIRTH INJURY IN MACROSOMIC INFANTS AND ASSOCIATION WITH MODE OF DELIVERY

L B Koldempx, q J Musct, R K Laros, Dept Ob/Gyn, Umvers~ty of Cahfomla, San Francisco, CA OBJECTIVE: Null Hypothesis There is no significant assoclatmn between the mode of delivery and persistent birth injury in macrosomlc infants (bmhwelght>4000 grams) STUDY DESIGN: Of the 30,000 debvery records m the UCSF pennatal database, 2818 were identified with infant blrthwelghts>4000 g, and adequate documentation regardmg traumatic injury 4% were complicated by GDM 1% by preexisting diabetes Delivery methods included spontaneous vagmal(SVD)(n=1599), forceps asststed(n=243), vacuum assisted(n=132), and cesarean secoon(n=844) Records were exammed for evidence of birth InJury, and follow-up was obtained through pe&amc charts Resulting propomons were compared using chl square analyses, Flsher’s exact test, and relative risk assessment RESULTS: Overall mcMence of b~rth Injury was 1 6%(n=46). hajunes included 19 clavicle fractures, 14 brach~al plexus injuries, 1 factal nerve injury, 10 complex injuries involving nerve palsy and fracture, 2 lntracramal hemorrhages, and 1 death atmbutable to the delivery process. 20 were clinically resolved at discharge from the nursery, 10 resolved by 6 month follow-up, 8 persisted beyond 6 months or were permanent, and 8 were present at discharge but lost to pediatric follow-up. The maximum rate of chmcally

~i~reSiStent injury (>6 mo ) was therefore between 3 and 6 per 1000 dehvenes re was a stattsncally significant relanonshtp between birth injury and method of dehvery(p<0 005) In the forceps assisted group, 5.8% of the infants were diagnosed with birth injuries, compared to 1 8% of infants delivered by SVD, 2 3% delivered by vacuum, and 0 1% of those delivered by c-secnon In ad&non, there was a significantly increased number of complex and persisting m~unes d~agnosed m lbe forceps group In companng forceps to SVD the relanve risk was 3 29 for initial diagnosis of birth ~njury, Th~s nsk mcrea~d to 6 58 for injury persisnng beyond discharge, and 8 77 for mjury documented at 6 months or more There was also an increased proportion of persistent injuries in the forceps group relative to the vacuum group, but this difference was not stanstlcally slgmflcant CONCLUSIONS" Even for macrosomic Infants, injury during dehvery is a rare event, and most of these mjunes will resolve Despite this low incidence, dehvery by forceps results m significantly more clinically persistent injuries than both spontaneous vaginal dehvery and cesarean section However, ~n order to prevent one of these persistent injuries 246 unnecessary cesarean sections would need to be performed.

139 POSITIVE AND NEGATIVE PREDICTIVE VALUE OF ESTIMATED FETAL WEIGHT FOR MACROSUMIA IN POSTDATES PATIENTS Christopher O’Rmlly-Green Dept. of OB/GYN, Albert E~nste~n College of Medicine, Bronx, NY OBJECTIVE Evaluate the positive predictive value (PPV) of the estimated fetal weight (EFW) for macrosomia ~n a cohort of postdates patients, taking birthweight over 4000 grams or 4500 grams within 4 days of EFW measurement as alternative gold standards for macrosomia. SIDDY DESIGN We evaluated the PPV of sonograph~cally estimated EFN at a cutoff of 4000 grams and 4500 grams ~n a cohort of 202 consecutive postdates patients who had a b]rthwe~ght re~rded within 4 days of the last sor~x~raphic measure~nt of EFW. RESULTS Predictive Value of Estimated Fetal Weight

TRUE TRUE FALSE FALSE MACRO MACRO MACRO EFW POS NEG ~OS NEG PPV NPV PREVALENCE 4000 30 141 8 23 O, 79 O. 86 O. 26 4500 2 190 4 6 O, 33 O. 97 O. 04 (POS = posit~ve: NEG = negative; MACRO = b~rthweight > 4000 gm or > 4500 gm depending on which EFW cutoff value was used.) There was no newborn merb~dlty tn th~s cohort of postdates patients. CONCLUSIONS B~rthweight over 4000 grams occurred ~n 26% of this cohort of postdates patients. A positive test correctly predicted macroscmia 791 of the time, with a false positive rate of 21%. Undertaking delivery because of a 79% chance of macrosomia has not been shown to avoid the morbidity of macrosomia. Using a cutoff for EFW of 4500 grams as an indication for cesarean section would result in a false positlve prediction 67~ of the time. Consequently we do not use EFN over 4500 grams as an ~nd~cat~on for cesarean del ~very. An EFW under 4000 grams mssed 14% of newborns over 4000 grams. Cesarean delivery for arrest disorders in labor and caution ~n use of forceps in patients w~th protrac~c~on d~sorders continues to be prudent,

138 LABOR INDUCTION WITH A PRENATAL DIAGNOSIS OF

FETAL MACROSOMIA. L~Leapbai% M Meyer, E Capeless Dept

O[¥GYN. Ulmers~tv of Vermont, Burhngton, VT

OBJECTIVE: Studies examtolng labor induction m pregnauc~es

comphcated b3, fetal macmsorma have demonstrated a h~gh cesarean

dellxetv (CS) rate m tins population We hypothesized that m an

mst~ttmon Xxlth a low backgrotmd CS rate (13%) there would be no

d~ren~ ~m’~o patients who underweut mduchon for macrosom~a and

pa0ents ~ ~th smniar bt~h weights who enter~ labor spoutaneously

STUDY DESIGN: A relrospect~ve analys~s performed over a 19 month

peuod flora 12/16/93 ~o 7/16/95 revealed 53 nondiabet~c patients who

underwent mductlou for fetal macrosomm These study pabents (~)

were lnatched to the next patient dehvernag a chdd of equal or greater

bmhx~e@at ~bo entered labor sponlaneously (SL) Both ~natemal

delllogtaphlc and pregnancy oulcome data were reviewed Data were

anaI3zed nsmg dn square w~th Yates correction, Studeut’s t-test, or

F~she~" s exact ~est ~ hen appropriate

RESULTS There ~ere no s~gmficaat differences m lnatemal age,

gesta~onal age. rote ofnulhpar~,, incidence of shoulder dyst~m, 1 aud 5-

lnmute Apgars, or successful vaginal bmh after poor cesarean Regional

anesthesia was adnmnstered more frequeotly m ~ (~ 83% vs

SL 53%, p<0 003) There was no CS performed for faded mduct~on

Cesarean rate was s*gmficantly htgher ~n ~ (~ 36%vs SL 17%,

p<0 05) despite the fact that bush x~e~glat m ~ was lmver (~ 4102~

374g, SL 4355~ 349g. p<0 001)

CONCLUSION: In tia~s case control study, contra~ to our bypothests,

an increased usk of cesarean dehve~ ~s observed tn subjects nndergo~ng

ltlduchoa of labo~ m assocmhon w~th the ~denttficatton of fetal

140 PREGNANCY OUTCOME IN PATIENTS UNDERGOING ELECTIVE INDUCTION AT 41 WEEKS’ GESTATION. A. H~ett. D Gleason’, H Brown, K Bntton~. Dept. Ob/Gyn, Indiana Umv., Indianapolis, IN OBJECTIVE: To compare the maternal and neonatal outcome in nulhparous and multlparous portents undergoing electtve reduction dunng the forty-first week of gestation STUDY DESIGN: We rewewed 2351 dehvenes occumng from July 1993 to June 1994. Patients _> 41 weeks’ gestation and < 42 weeks’ gestatton were identtfied. Only those m spontaneous labor or undergoing electtve induction were included. Statlsttcal analysts was performed using t-tests, chi-square, and Wilcoxon sign rank when appropriate. RESULTS: 101 women and 94 mult~parous women met the study criteria. Of the nulhparous women, 31 (30 6%) were induced and 70 (69.4%) were admitted m labor. In the mult~parous women, 19 (20 2%) were ~nduced and 75 (79 8%) were admitted in labor. In nulliparous pattents, induced labors resulted in more cesareans when compared to spontaneous labors, 41 9% and 22.9% respecttvely (p = 05). Btrth wetghts were similar m both tnduced and spontaneous labor. Meconlum staining was lower tn the induced labor group (p = 04). The modified Btshop cervtcal score was not predlcttve of successful tnductton in nulhparas In multlparous patients, there was no difference in cesarean dehvenes between induced and spontaneous labors, 5 3% and 9.3% respecttvely Birth weight, meconium staining, low Apgar score, and chonoammonitts were s~mdar for tnduced and spontaneous labors. Induced nulhparous patients had a sigmficantly higher cesarean delivery rate when compared to induced multiparous portents, 41 9% and 5 3% respectively (p < .01) Birth weight, meconmm stained amniottc flu)d, low Apgar score, and chonoamnionltis were similar for both groups. CONCLUSIONS: Elective induction of nulliparous patients during the forty-first week of gestation was associated wtth an increased rate of cesarean delivery when compared to nulhparous patients m spontaneous labor and all multtparous pattents. Elecnve inducnon was not associated with an increase in maternal or neonatal morbidity and dld result in a lower lIlcldence of meconlum staining in nulhparous patients

Vnlume 174, Number 1, Part 2 SPO Abstracts 351

Am J Obstet G~necol

141 PERINATAL OUTCOME IN A LARGE COHORT OF POSTDATES

PREGNANCIES. M Y Dlvon B Haglundx, H Nisellx, P Otterbladx, M.

Westgren Center for Epldemmlogy, National Board of Health and Welfare,

Huddmge Unlv Hosp , Karohnska Inst, Stockholm, Sweden and Albert

Einstein College of Medicine, Bronx, NY

OBJECTIVE: To identify predictors of perlnatal outcome in postdates

STUDY DESIGN: A retrospective analysis of the Swedish Computerized

National Birth Registry of the following outcome varmbles intrauterine fetal

death (IUFD), neonatal death at ~< 28 days of life (ND), and Apgar < 7 at 5

minutes The fullowmg variables were identified as potential predictors of

adverse outcome and entered into a logistic regression analysis gestatlonal age

(GA), parity, fetal sex, h~story of maternal smoking or drag abuse, blrthwelght

(BW) and small for gestatlonal age (SGA defined as BW < 2 standard

deviations below mean for GA) Only patients w~th accurate dates (by LMP

and sono) entered the statlsUcal analysis Study permd’ 1/1/87 - 12/30/92 RESULTS: The total number of dehverles >38 weeks during the study permd

was 607,843 Accurate dates were established in 328,635 pregnancies, of

these 15%, 27%, 30%, I7%, 6% and 5% were dehvered at 38, 39, 40, 41, 42 and 43 weeks, respectwely The overall incidence of SGA was 2 0 %

Outcome by GA and Fetal Growth (*p =0 01, O R =odds ratio)

IUFD NM Apgars <7 at 5 rain

O R CI O R CI O R CI

> 40 weeks non SGA SGA

>41 weeks nonSGA SGA

>42 weeks nonSGA SGA

1 1 1

74 43-126" 33 14-82" 32 22-44"

1 3 1 1-18" 1.2 09-16 1 2 i 1-1 4*

10 4 5 9-18 2* 7 1 3 3-15 2* 5 0 3 6-7 0*

1 8 1 3-2 5* 1 4 0 8-2.0 1 9 1 7-2 2*

73 30-178" 93 38-228" 50 3.2-7 9*

CONCLUSIONS: Adverse fetal outcome increases slgmficantly as GA

advances beyond 41 weeks’ gestation, however, its association with SGA is

much stronger than with GA Thus, future strategies for management of

postdates should focus on fetal growth rather than GA per se

143 Outpatient Management of Postdate Pregnancy with Intravaginal

Prostaglandin E2 and Membrane Stripping

Waleed Denny, M D, Departments of Obstetncs and Gynecology,

Un~vers{ty of Cahforn~a, Los Angeles, Ohve V~ew-U C L.A Med{cal

Center, Sylmar, CA

Objective The purpose of the study was to determine whether

the combination of outpatient administration of intravaglnal

prostaglandm E2 (PGE2) and membrane stnpp~ng at or beyond 41

weeks gestation age would reduce the interval to dehvery

Methods: TNs study was a double bhnd placebo controlled thai

~n which 150 patients were randomly assigned to one of four

treatment arms’ placebo gel, prostaglandm E2 gel, membrane

strtppmg followed by placebo gel adm~n~sttatmn, or membrane

stnppmg followed by prostaglandm E2 gel administration. Patients

received the assigned outpatient treatment at 41 weeks, 42

weeks, and every 3-4 days thereafter. Interval to dehvery,

dehvery outcome, and comphcat~ons were analyzed Student’s t

Test, F~scher’s exact test, Wflcoxon test and ANOVA were used

when appropnate. P < 0.05 dehned statistical s~gnff~cance

Results: Pattents recetwng both membrane stripping and PGE2

had s=gn=flcantly shorter durat=on to dehvery (Med=an 1 day, 25

75th percentile 0-4 days, P = 0 001) and fewer ws=ts for

antenatal fetal testing (23% more than one ws~t, P = 0 02) than

the control group (Med=an 7 days, 25-75th percentile 3.5-11 5

days; 61% more than one ws=t, respect=vely) No s~gn=flcant

maternal or neonatal comphcatlons occurred with any of the

treatment arms.

Conclusion: The combmat=on of mtravagmal PGE2 and membrane

stnppmg can be safely used as outpat=ent procedures to reduce

the mtervaI from 41 weeks gestat=onal age to dehvery

142 COST COMPARISON OF INDUCTION OF LABOR AT 41 WEEKS VERSUS EXPECTANT MANAGEMENT IN THE POST-TERM PREGNANCY M Gardner,D.Rouse,R Goldenberg,J Lannlng’,E Thorn’, J Zachary’,& NIH Maternal-Fetal Medicine Unit Network, Bethesda, MD, Unlv orAL ,Birm, AL, Univ ofNM, Athuq, NM, G W Univ Washmgton DC Objective:The management of uncomplicated pregnancies (~. 41 weeks) with an unfavorable cervix generally involves either induction of labor or employing antepartum testing while awaiting spontaneous labor A multlcenter postterm pregnancy study from the National Institute of Cfuld Health and Human Development (NICHD) Net~vork of Maternal-Fetal Methc~ne Umts did not demonstrate a significant difference in pennatal morbithty or mortality between these strategies Since the outcomes of the three strategies were not different, our objective was to identify the least costly management scheme Methods Hospital charges were applied to the three groups In the NICHD study I) Expectant management(EM), 2) oxytocln induction of labor at 41-42

weeks’gestation,and 3) oxytoc~n Induction of labor 12 hours after cervical ripening vflth prostaglandln (PG)E2 intracervtcal gel Estimated charges were assigned for maternal hospital days, hours in labor and delivery, antenatal tests including ultrasound examinations, and the charge for one dose of PGE2 gel Results The mean maternal charges were $2,266 in the EM group, $2,296 in the elective lnductmn group that did not receive PGE2 gel, and $3,312 in the group that underwent prostaglandln cervical ripening followed by oxytocin lnductmn of labor There was no significant difference in the rates of cesarean

b~rths between the three groups. There were no slgnlficant differences in total infant length of stay, days in the NICU or neonatal charges between the three groups

Maternal Variables and Estimated Hospital Charges

EM (N=175) PGE2 Gel (N=174) Placebo (N-91) Mean Charges* Mean Charges* Mean Charges*

MHD* 3 3 $156l 4 0 $1892 3 2 $1514 HmL&D* 12 7 $538 29,6 $1255 18 1 $767

NST* 07 $51 0 06 $4 0 06 $4

U/S* 07 $116 006 $11 006 $I1

PGE2 gel 0 doses $0 1 $15000 0 $0 * Maternal Hospital Days, Hours in L& D, Non-stress Tests, Ultrasounds Conclusion Based on the results of the NICHD study,there is no cost advantage

to either EM or ~nduchon of labor with oxytocln In the postterm pregnancy Use of PGE2 gel ~ncreased maternal hospital length of stay and hospital charges, without improving maternal or neonatal outcome

144 THE ROLE OF INFORMATION BIAS WHEN USING PELVIC

EXAMINATION TO CONFIRM GESTATIONAL AGE J Robertsx, C Pet~erkln, J Fle~sher, S McCalla, H M~nkoff SUNY Brooklyn, Bklyn, NY

Objective: The American College of Obstetricians and Gynecologists

recommends early pelvic examination be used to confirm gestatlonal age Prior

knowledge of last menstrual period (LMP) however, could theoretically bins

assessment and compromise the utility of the physical examulatlon as an

independent confirmation of gestatlonal age We therefore undertook a study to

determine if prior knowledge of a patient’s last menstrual period biases the

clinician’s estimate of gestatlonal age as determined by pelvic exam

Study Design: This was a prospective study of pregnant patients registered at

Kings CounP¢ Hospital and State Unlversl~ Hospital of Brooklyn At their first

prenatal visit, women were randomized, via a standard randomization table with numbers maintained in opaque envelopes, to be examined before or after

informing the examiner of their LMP The mean discrepancies between

calculated gestatlonal age by LMP and measurement of gestatlonal age b3 pelvic

examination were compared between the two groups, using 1" test procedures

We assessed the effect of obesity and prior surgery on the level of discordant

pelvic exams, and the relationship of level of physician training (PGY level) to

accuracy of the pelvic examloatlon

Results: A significant d~fi’erence was found between the mean discrepancies in

the two groups Among women examined prmr to knowledge of LMP (n 70)

the pelvic findings d~ffered from the anticipated s~ze based on LMP by 1 7 wks,

while for women examined after knowledge of LMP (n-69) the difference was

only 1 l wks (p=0 02) Rates of discordance (difference _> 2 wks) were 42% for

the obese patients examined when the LMP was not known vs 17% when the

LMP was known (ns) Similarly, anlong patients with prior pelvic surgery the

values were 43% vs 30% (ns) As PGY level advanced the mean d~screpancles

between examination with and exam without knowledge of LMP, diminished

Conclusions: Chnlclans who evaluate patients after learning the women’s LMP

tend to have findings which more closely approximate the historical data Also,

the more difficult the pelvic examination, the more history may bias the chnlclan

when recording pelvic findings, although accuracy appears to ~mprove with

experience These results suggest that information bias may be influencing the

pelvic examination, thus rendering it less useful as an independent gauge of

gestatlonal age

352 SPO Abstracts ,]anu4~?, !996 Am J Obstet Gynecol

145 THE EFFECT OF DIGITAL CERVICAL EXAM ON MEASURED LEVELS OF FETAL FIBRONECTIN.

D. Burrus,~ D. Adair,~ K. Patel,x J. Ernest, A Frye,x J. Vedle. Dept of Ob/Gyn, Bowman Gray School of

Medicine, Wake Forest University, Winston-Salem, NC.

OBJECTIVE: The null hypothesis is that there would be no difference in measured levels of cervical fetal fibronectin (FFN) between samples obtained before and

after digital cervical exam in either contracting

(symptomattc) or not contracting (asymptomatic) populations.

STUDY DESIGN: I) Symptomatic patients evaluated at term in early labor had cervical FFN sampled pre and post

digital exam. 2) Asymptomatlc patients arriving for 36 week routine screen for cervical/vaginal pathogens also had cervical FFN sampled pre and post digital exam. Results

were compared within groups by paired T test with p < .05

considered szgniflcant. RESULTS: Asymptomadc pre digital exam mean FFN

< 50, asymptomatic post digital exam mean FFN < 50. Symptomatic pre digital exam mean FFN <50,

symptomatic post digital exam mean FFN 159 (p< 05) CONCLUSIONS: Preliminary results suggest digital

cerwcal exam appears to increase measurable FFN in symptomatic but not asymptomatlc patients. The effects of digital cervical exam on cervical FFN as it correlates to the

diagnosis of labor needs to be considered.

147 QIIALITY D~LIV~EB.Y FOR CLINICAL PATHWAYS.

ILIL Ransona’,F.D McDonald’, J W Gellx, D P,. Cotton Dept OB/GYN, WayneState

Umvemty School of Methc~ne, Hutzel Hospital, Detroit, MI

OBJECTIVE The pressures of managed competmoix and health care reform ~ led

hospitals to determine ways to ~mprove quahty, and s~multaneously, reduce costs The

L.wlementa~on ofctlm’~l pathways was used on our permatal service to ~mprove quahty and

efficiency

STUDY I)ESIGN~ Hospital Adrn~nistrauon requested the Chairman of Obstetrics and

Gynecology to "thamp~on" a chnical pathways prelect, choose the DRGs to be stuthed based

on volume and appomt a work group, w~th a raalonty of pennatologista, to m’eate the

pathways The first DRGs completed were 370 (C-section w~th complications) and ~73

(normal vaginal dehvery) The pathways defined, by time, those events which the work

groups believed were required to dehver quahty The work group identified w~ys to

~mprove 9uahty and reduce unnecessary resource consumpnoa. The group developed a

model that described the appropriate care for ~ *poetic panent c~se type and produced

pathways that were valid and credible F~nancaal statements of 50 panents m each DRG were

.nalyzed from the prewous stx months as a control, and, by natural class of expense, the

variance* between the ideal expressed by the pathway and what was actually consumed m the

care of the average panent was computed m umts and dollars The variances included length

of ~tay w~th maprovements developed for infant thscharge (pedmmc exam~naUon and wait for

c~rcumcision), transportation needs of pauents, transfer of chnIc records to hospital for

redt~don of duphcation of services, laboratory uuhzatmn and pharnuacy efficiency Finally,

the ct~ef of the department cornmumcated these action plans as goals to the department for

urholementauon The ho*p~tal edrrumstrat~on fol]owed the ~mpaet of length of stay over the

subsequent two years

REStlLTS: A reducuon ~n length of stay was found after ~mplementatlon of the clinical

pa~way from the coatr~l group to the stud)’ period of Jauttary 199~ to luly 1995

Spedfically, the length of stay for DRG 373 (normal vaginal dehvery) dropped from 2 2 days

to 1 7 days and the length of stay for DRG 370 (C-section with comphcatmns) dropped from

6.6 days to 5 9 days The c~t savings related to the shorter length of stay wa~ computed and

found to be

CONCLUSIONS: The implementation of chmcal pathways for Permatal D~agnostic Related

Group* w~ found to mnprove qu~hty and eft~.~ency wath a roducnon m length of stay causing

a s~gmflcant cost savings for the hospital

146 A NEW METHOD TO ASSESS FETAL HEAD DESCENT IN LABOR

WITH TRANSPERINEAL ULTRASOUND (TPU) A Barbera~, X Pombar,^

G PeruginO, E Ferrazzl, G Pardi, JC Hobblns Dept s of Ob/Gyn, ISMB San

Paolo Milan, Italy, UCHSC Denver, CO

OBJECTIVE To compare ultrasound assessment of fetal station during labor

with a standard digital examination

STUDY DESIGN - When an ultrasound transducer Is applied to the permeum,

a sagtttal view is produced of the long axis of the symphysls along with the fetal skull and scalp, allowing the angle of variation along the pelvic axis to be

measured TPU and paired digital exam by "blinded" chmmans were

performed in 50 laboring patients with cephahc presentations (20 in Denver,

30 in Milan) Head descent was quantified by TPU by measuring w~th a

goniometer the angle between the Inferior point of the long axis of the

symphysls pubis and the leading part of the fetal skull at varying intervals

throughout labor A subgroup had two to three TPU exams by different

blinded Investigations to determine inter observer variation

RESULTS

1 The mean inter observer variation wlth TPU never exceeded 10%

2 There was a linear correlation between cervical ddatatlon and station by

TPU (r=0 66)

3 Despite the wide varlablhty between digital assessment of station and tree

station by TPU (r=0 56), an angle of 125° was always associated with

engagement of the head by clmtcal exam

4 AROM had no immediate effect on fetal station by TPU

5 When caput was present there was as much as a 31% difference in angle

between the caput and skull, a finding not appreciated by digital exam

CONCLUSION TPU provides a reproducible and objective way to assess

descent of the fetal presenting part, while the Inherent difficulty in determining

simultaneously the/eve/of the fetal head compared with lschtal spines makes

digital examination very imprecise It is hoped that this objective techmque

can aid the chnlclan in better selecting patients for CSx who have tree failure

of descent In the second stage of labor

148 SALINAS FORCEPS: A VERSATILE INSTRUMENT FOR OPERATIVE VAGINAL

DELIVERY. ~N. Kramer, C, Coronadox, G. Saade, K. Moise, Jr., M. Belfort. Dept of OB/GYN,

Baylor College of Medicine, Houston, TX.

OBJECTIVE: To evaluate the safety of Sahnas forceps as compared with Kielland’s forceps

when used for midcavity rotational delivery, and with Simpson’s forceps when used for low

or outlet forceps delivery.

STUDY DESIGN. Retrospective chart review of all infants dehvered by Salmas forceps

between September, 1992, through December, 1994. These deliveries were matched for

maternal age, panty, ethnic origin, gestational age and fetal weight with a similar number of

midcawty rotatmnal forceps deliveries using Kielland’s forceps, and low and outlet forceps

dehvenes using Simpson’s forceps. All procedures were pedormed by residents in traiNng,

Data were analyzed using the Mann-Whitney U, Wilcoxon rank-sum, Student’s t, chi-squared

and Fisher’s exact tests as appropriate. Statistzcal s=gmficance P < 0.05.

RESULTS: Data presented as median (range) or number (%). Neonatal injury includes

bruises, lacerations and hematoma. Neonatal trauma includes Erb’s palsy and skull fracture.

Rotational Forceps (24 pairs) Low or Outlet Forceps (51 pairs)

Salmas K=ellend’s P Salinas Simpson’s P

~,pgar at 1 minute 8 (4-9) 8 (4-9) NS 8 (6,9) 8 (4-9) < 0,01

~,pgar at 5 minutes 9 (8-9) 9 (8.10) NS 9 (8-101 9 (8-101 NS

~Jeenatal inlury 4 (17) 13 (54) < 0 02 0 (O) 39 (77) < 0 01

Neonatal trauma 0 (0) 1 (4) NS 0 (9) 3 (6) NS

Epiuotomy 23 (96) 21 (88) NS 48 (94) 43 1841 NS

Ep=s extensmn 11 (46) 13 (54) NS 18 (35) 27 (53) NS

:ervical laceration 1 (4) 1 (4) NS 0 (0) 0 (0) NS

~agmal laceration 2 (8) 8 (331 < 0.05 418J 141271 < 0.02

F~egmnal anesthesia 17 (71) 16 (67) NS 35 (69) 35 (69) NS

’.ONCLUSIONS: Salinas forceps can he used safely for both midcawty rotational and

IowJoutlet forceps delivery.

Volume 174, Number 1, Pa~t 2 SPO Abstracts 353 Am ] Obstet (,ynecol

149 FACIAL NERVE PALSY: FREQUENCIES ASSOCIATED WITH

SPONTANEOUS, FORCEPS, AND CESAREAN DELIVERIES. D.A~

White," E.K. Pressman," G.V. Haana," M F. Odom," N.A. Callan, K.

Blakemore. D~t. of Gyn/Ob, Johns Hopkins University, Balto, MD

OBJECTIVE: To evaluate the types of deliveries and other risk factors

associated with facial nerve palsy.

STUDY DESIGN: A five year retrospective analysis was conducted of all

cases of neonatal facial nerve palsy diagnosed at The Johns Hopkins Hospital

from July, 1990 to June, 1995. Data collected included maternal age, race,

parity, infant birth weight, gestational age, and mode of delivery. Statistical

analysis was performed using X2 analysis.

RESULTS: There were 18,538 deliveries with 27 cases of facial nerve palsy

diagnosed during the study period, yielding a rate of 1.4/1000 deliveries.

Forty-eight percent of the cases, versus 43% of all other deliveries, were

nulliparas (not significant). Eighty-eight percent of the cases, versus 70% of

all deliveries, were of African descent (not significant). The mean birth weight

for the cases was 3034 grams (range 1945-3936 grams). Mode of delivery was

significantly different between the two groups: 18.5% of cases, versus 5%

overall, were via forceps (p <0.01); fifty-two percent of cases, versus 74% of

all other deliveries, were spontaneous vaginal deliveries (13 <0.01). The

cesarean section rates of ! 8.5 %, versus 21% for all other deliveries, were not

statistically significant.

CONCLUSIONS: Our data support the previously reported association of

facial nerve palsy with forceps deliveries. The magnitude of this association,

however, is five-fold less than previously reported. Only 18.5% of our cas~s

were from forceps deliveries, thus the majority of cases resulted from

spontaneous or cesarean deliveries. In addition, there was no association of

facial nerve palsy with nulliparity or macrosomia.

151 A RANDOMIZED PROSPECTIVE TRIAL OF VACUUM EXTRACTION

TECHNIQUE. J.A Bofd!, O A. Rust, S.J. Schorr, R.C. Brownx, W.5.

Roberts, J C. Momson. Dept. Ob/Gyn, Umv. of Miss., Jackson, MS.

OBJECTIVE: To determine ~f two techmques of vacuum extraction

delivery, continuous vacuum (CV} and intermittent vacuum (IV), have

different effects on maternal-fetal outcomes.

STUDY OESIGN: Patients to be dehvered by vacuum extracUon were

randomized to delivery by CV or IV. All dehvenes were performed using

the M~tyvac M-cup. In the CV group, the vacuum was brought to 600

mmHg and maintained st that level untd dehvery. The level of vacuum

was not decreased between contractions and the operator prevented

fetal loss-of-station between contractions by maintaining traction.

the IV group, the level of vacuum was decreased (to 100 mmHg)

between contractions and no effort was made to prevent fetal loss-of-

station. Time was recorded (sec) from placement of instrument to

dehvery of infant.

RESULTS: 322 patients were randomized. Data are presented as N

and mean (+ SD). Statistics used werex2, ANOVA, and Student t test

as appropriate. There are no significant differences in the comparisons

below.

CV IV

(N = 164) (N = 158) P

Fadure to deliver 12 9 .72

T~me(sec) 166 (175) 167 (150) 07

Btrthwe~ght 3128.9 (457.2) 3167.0 (489.8) .47

Apgar (5 ram) 8.89 (0 42} 8.96 (0.24) .07

Cord arterial pH 7.25 (.07) 7.26 (.08) .31

Cephalhematoma 20 17 .82

Shoulder dystocm 6 9 .55

3rd/4th deg lac/ext 22 16 .46

Hyperbdlrublnemla 7 14 15

CONCLUSIONS: There are no differences in maternal or infant

outcomes if the level of vacuum IS decreased between contractions or if an attempt is made to prevent fetal loss-of-station.

150 A COMPARISON OF TWO METHODS OF DELIVERY FOR FAILURE

TO DESCEND IN NULLIPAROUS PATIENTS. A Parchment~ JW Bernaskox, L Lynch Dept Ob/Gyn, The Mt S~nal Hospital, NY, NY

OB3ECTIVE: To assess maternal and perinatal morbidity assocxated with

mldpelvic forcep dehvery performed ~n a teaching hospital traxmng program

STUDY DESIGN: Nulhparous paUents dehvenng infants with blrthwelghts

between 2500-4000gms who underwent a trial ofmtd pelvic forceps for failure

to descend were compared to an identical group who were dehvered by cesarean

sectmn for the same mdicatmn without a trial of forcep Chl-square analysis was

used and a p < 0 05 was considered significant

RESULTS’ 481 patients w~th the dmgnos~s of failure to descend after complete cervical dilatation fit the criteria for the study Mldpelwc forcep dehvery was

attempted on 314 (65%) and fmled in 41 (13%) thereby requiring a cesarean

sectmn 167 patients were dehvered by cesarean sectton w~thout an attempt at

forcep dehvery The average birthwe~ght for the group that failed mid forcep

delivery was 3489gms The average blrthweight for the c/s group was

3477gms There was no s~gmficant dlft~rence in length of labor, maternal

hemorrhage, amulonltlS, meconlum aspiration, cephalohematoma, admlsslotl tO

the NICU, 1VH, fractures, ventilator use, low 5 minute apgars, or seizures

between the successful mid forceps group and the cesarean sectton group There

was an mcr~as~ in post par~um endometrltls for the c/s group over the forceps

group (14 4 ,~s 6 4%, p<0 05) and an increase hospital stay (4 vs 2 days,

p<0 05) When comparing the ~aded forceps group with the group that

underwent c/s w~thout a trial of forceps there was a significant lncress~ m

admissions to the NICU (21 9 vs 8 9%, p<0 05 ), and low 1 and 5 m~nute apgars

(14 6 vs 3 6% and 4 9 vs 0 6%, p<0 05) Them was one mtrapartum f~tal

demise In each group There were no significant differences In the ~nmdence of

IVH, cephalohematoma, skull fracture, seizures, brach]al palsies, laceratmns,

pnemnothorax, or prolonged need for oxygen in rather group There was no

s~gnlficant &fference m the number of days to discharge for the neonates from

the two groups that underwent c/s

CONCLUSION: The majority of trials ofm~dpelvlc forcep delwenes m

nulhparous patients for fadure to descend are successful and associated wxth less maternal morbld~ty than cesarean section Judicious use of mid pelvic forceps is

not assomated with excessive neonatal morbidity as compared to cesarean

section

152 PROGNOSTIC FACTORS FOR MODERATE AND SEVERE MATERNAL

GENITAL TRACT LACERATION WITH OPERATIVE VAGINAL DELIVERY.

J.A Boffll, O.A. Rust, M. Dev~dasx, R.W. Martin, J.C. Morrmon, J N.

Mart~n, Jr. Dept. Ob/Gyn, Umv. of M~smsslppl, Jackson, MS.

OBJECTIVE: To determine the risk factors that predispose to maternal

gemtal tract laceratmn/extenslon (L/E) m associatmn with operative

vaginal dehvery (OPVD).

STUDY DESIGN: 637 paUents undergoing OPVD were randomized

between forceps (F) (n = 315) and vacuum (V) wRh M~tyvac M-Cup (n

= 322) and t~med from imtml placement of instrument to final delivery.

Data were gathered prior to and after F or V dehver¥. Associated

genital tract injury was described as MILD (minor second degree

permeal, vaginal s=dewall, or periurethral laceratmns), MODERATE (th=rd

degree L/E or m=dvagmal L/E), and SEVERE (fourth degree L/E or L/E rote

fornlx). Statistics included X2, ANOVA, and multiple hnear logistic

regresmon

RESULTS: There were 35 SEVERE injuries (30 fourth degree L/E, 5

formx L/E) and 149 MODERATE Inlur~es (98 th=rd degree L/E, 51

mldvagmal L/E). S=gnif=cant assomated parameters included: no prior

vaginal dehvery (p < 0.0OO), indicated OPVD (p < 0.000), rotation >

45° (p = 0 004), use of F (p < O.000), use of eplslotomy (p < O.000),

Increasing t=me (p = 0.OO01), higher staUon (p = 0.O01), and

increasing birth weight (p = 0.0123). After stepwise multiple logistic

regressmn analyms, the following factors remained slgmfmantly

associated w=th moderate to severe L/E: star=on (p = 0.0284), no prior

vaginal dehvery (p = O O002), use of forceps (p = O.OO01), increasing

time (p = 0.OOO1), use of epls~otomy (p = 0.O001), and =ncreas~ng

Nrth weight (p = 0.0012). The Iog=st~c regression model thus created

had a senslt=wty of 45.1%, a spemfic=ty of 92.3%, a false-pos=tlve rate

of 36 2%, and a false-negat=ve rate of 15.2%.

CONCLUSION: Moderate to severe maternal gemtal tract L/E occur

most often with dehverles initiated at other than outlet station, in the

nulhpara, with F, with requirement for eplsiotomy, with larger infants,

and when the t=me of OPVD ~s prolonged.

354 SPO Abstracts Jam~ar), 1996

Am J Obstet Gynecol

153 A RANDOMIZED PROSPECTIVE TRIAL OF THE OBSTETRIC FORCEPS VERSUS THE M-CUP. 3.A. Bofill, O.A. Rust, S.J Schorr, R.C. Brownx,

R.W. Martin, J.N. Martin, Jr., J.C. Mornson. Dept. Ob/Gyn, Univ. of

M=ss., Jackson, MS.

OBJECTIVE: To determine the efficacy of the obstetric forceps versus

the M~tyvac M-cup, a new vacuum extractor cup without previously

documented efficacy and comphcatlon rates {maternal and fetal).

STUDY DESIGN: Operative vaginal dehveries (OPVD) were randomized

to forceps iF) or vacuum extraction iV). The type of F was left to the

operator. Maternal demographics, md=catmns for OPVD, analgesm,

position, station, and degree of caput and molding were prospectively

recorded. Time from apphcatlon to delivery was recorded (sec).

Eplslotomy and extensions, lacerations, and reason for abandonmer~t of

randomized instrument were recorded. Fetal weight, Apgars, cord

arterial gases, hyperbdirubmemm, phototherapy, and any ewdence of

fetal trauma noted at delivery or in the nursery were documented

RESULTS:

FiN = 315) V(N = 322) P

Fad to dehver (corr) 25 18 .237

Time (sac) 244 (4-218) 167 (4-163) 000

3rd des lac/ext 67 31 .O000

4th des lac/ext 23 7 .004 Blrthwe=ght 3082 (4-521) 3147 (4-473) .10 Stat=on .40 Apgar (5 rain) 8.92 (4-.38) 8.92 (4- 34) 99 Cord arterial pH 7.25 (4-.06) 7.26 (4-.07) 054 Shoulder dystoc=a 6 15 .08

Cephalhematoma 19 37 .02 Hyperblhrubmemm 18 21 .79 Bru=sing 44 39 56

CONCLUSIONS: The M-cup appears to be as efficient (and faster) than

the obstetric forceps, but is associated with slgmflcantly more fetal

cephalhematomas. Maternal injuries are more common with forceps

155 RANDOMIZED TRIAL OF A VBAC EDUCATION AND SUPPORT

PROGRAM. W. Fraser, E. Maunsell, E. Hodnett, J-M. Moutquin and the

CAP Study Group, Dept.Ob/Gyn and Soc.Prev.Med., Univ. Laval., Qua,

and Fac. Nursing, Umv.Toronto, Ont.

OBJECTIVE: Many women consider themselves to be key decision

makers in the elective cesarean/VBAC choice. The goal of this study was

to assess if an indiv=dualized VBAC prenatal education and support

program increases the probability of vaginal birth.

STUDY DESIGN: In 11 centres, women with a single previous cesarean

were recruited prior to 28 weeks through doctor’s offices and pre-

admission registries. The woman’s self-assessed motivation to attempt

VBAC was measured on a 10 cm Visual Analogue Scale (VAS): Stratum

I (Low motivation) - 0.0 to 5.0 cm; Stratum II (High Motivation) - 5.1 to

10cm. Consenting women were randomized by stratum to one of two groups. Those in the "Verbal" group participated in an individualized

VBAC education and support program; those in the "Document" group

were simply provided a pamphlet detailing the benefits of planned VBAC.

RESULTS: Data are available on 1275 of 1301 women randomized The

propo~on of vaginal deliveries was slightly greater in the Verbal group,

but this difference did not achieve statistical significance: Verbal -

339/641 (53%); Document - 310/634 (49%); RR = 1.08, 95% CI = 0.97 - 1.21. There was no evidence of heterogeneity across centres or across

strata. Women in stratum I (low motivation for VI~AC) were more than 3

times as likely to undergo scheduled cesarean than women in Stratum

II (47% vs 13%). Exploratory analysis found that for women with ’Very

low motivation to attempt VBAC" (0.0 - 2.5 cm on the VAS) those who

were in the Verbal group had a higher rate of VBAC than those in the

Document group (RR = 1.8, 95% CI = 1.1 - 3.0).

CONCLUSIONS: For women who are motivated to attempt VBAC, an

individualized prenatal VBAC program does not significantly increase the

probability of vaginal delivery over written information. For women who

are very reticent to attempt VBAC, individualized counselling may

increase the likelihood of vaginal delivery. Confirmation of this hypothesis

is required in subsequent studies (MRC Grant No MT-11430)

154 VAGINAL BIRTH AFTER CESAREAN SECTION: COM- PARISON OF OUTPATIENT USE OF PROSTAGLANDIN GEL TO EXPECTANT MANAGEMENT. L. GIttens, C. Schenkel×, S Strassbergx, J. Apuzzio. Dept of OB/GYN, UMDNJ-New Jersey Medical School, Newark, New Jersey

07103 OBJECTIVES: To determine ff patients w~th one previous low segment transverse (LST) cesarean section (c/s) who receive outpatient PGE2 at term, have h=gher a VBAC rate and lower hospital costs than do controls STUDY DESIGN: Patients with one previous LST c/s, who were candidates for thai of labor, and had a Bishop’s score

of < 6 at 39 weeks, were prospectively randomized to a control or treatment group. The treatment group rece=ved 0 5 mg intracerv~cal PGE2 at 39 weeks. Th~s dose was repeated weekly unt=l adequate cervical ripening occurred.

The groups were otherwise managed s~milarly. " Rates of successful VBAC, and delivery costs were compared RESULTS: 14 (82%) of 17 patients that received PGE2

succeeded with VBAC; 10 (66%) of 15 controls succeeded No difference in rate of successful VBAC was noted When study and control patients who had VBAC were compared,

no d~fference in length of hosp~talizahon or d~rect costs to the pahents was noted. VBAC pahents in the PGE2 group, however, had shorter admission to delivery intervals (p< 01).

CONCLUSION: This protocol neither ~ncreased the successful VBAC rate, nor decreased costs to pabents Since outpahent use of PGE2 shortened labor and delivery stays, costs to the institution may be reduced.

156 PREDICTION OF THE SUCCESS OF VBAC USING ARTIFICIAL INTELLIGENCE A.Lysikiewicz, E.Leikin, &Dresden, N Tejani. Dept. Ob/Gyn New York Medical College, Valhalla, N.Y. OBJECTIVE: To predict the success of VBAC us=ng artificial intelligence (Neural Network). STUDY DESIGN: Data on 634 patients w=th a history of previous Cesarean section undergoing a trial of labor were used to predict successful vaginal delivery. Data included: patient’s age, gestational age, grawdity, parity, fetal weight, fetal length and head c=rcumference, presence of maternal antepartum comphcations and the physician In charge of making clinical decisions. Maternal complications were categonzed into advanced maternal age, diabetes, hypertension, anemia, infection, hemorrhage and others. Bra=nmaker (California Scientific Software) was used. The input data was entered into the neural network as continouos and categorical variables. After training was completed the Neural Network made predictions on a new set of cases. The predictions by the Neural Network were compared with the actual success of VBAC in clinical practice. The Chi-square test was used to compare the number of successful VBACs in predictions made by the Neural Network and the success rate in clinical practice. RESULTS: Sixty-seven percent of successful VBACs were correctly predicted by the neural network. In clinical practice the overall success of VBAC was 51%. The difference is statistically significant using Chi-square at P<0.05. CONCLUSION: The Neural Network can predict success of VBAC with 70% accuracy using information generally available to the managing physician This represents a statistically s=gnificant improvement over clinical practice and may be helpful m counseling patients regarding their attempt to undergo a thai of labor after Cesarean section.

Volun~e 17~t, Number 1, Part 2 SPO Abstracts 355 Am,] Obste~ (;)necol

157 CESAREAN DELIVERIES AT A UNIVERSITY HOSPITAL: ANALYSIS OF RATES AND INDICATIONS. J Pollard and E Capeiess Dept Ob/Gyn, Unw of Vermont, Burhngton, VT OBJECTIVE: ~n a tertiary care center, reliance on an mst~tut~onN cesarean (CS) rate does not reflect the general obstetnc practice ~n the communrty Therefore, nsk specific CS rates should be used for compansons STUDY DESIGN: All dellvenes (1227) at the University of Vermont from January 1 to June 30, 1995 were retrospectively analyzed for route of dehvery, rod=cat,on for CS, and practice group (to reflect risk status) RESULTS. The overall CS rate was 13 0% (160/1227) and the pnmary CS rate was 109% (122/1115) Overall CS rates by practice groups were Maternal-Fetal Med=cme (MFM), 33 3% (35/105), Pnvate obstetncs , 12 9% (103/796), Nurse m~dwlves, 7 3% (16/219), Ob/Gyn residents, 6 7% (3/45), Family practice, 4 8% (3/62) Indications for CS were abnormal presentation (306%, 49/160), arrest of dilation (20 0%, 32/160), arrest of descent (13 1%, 21/160), non-reassunng fetal assessment (8 1%, 13/160), faled ~nducflon (3 7%, 6/160), and other (24 4%, 39/160) The corrected CS rate (MFM excluded)was 11 1% (125/1122, pnmary rate 8 8%, 9111033) In tNs subgroup excluding MFM, abnormal presentation remained the most common ~nd~cat~on for cesarean (23 2%, 29/125) CONCLUSIONS: The ~nc~dence of CS at a university hospital can be compared wth commun~ based hospitals by d~scountmg MFM patients Additionally, abnormal presentation ~s the predominant indication for CS ~n all nsk groups

159 INTRAVENOUS GLUCOSE TOLERANCE TEST COMPARED TO A SINGLE ABNORMAL VALUE ON ORAL GLUCOSE TOLERANCE TESTING AS PREDICTORS OF CESAREAN DELIVERY B. Brost. S. Stramm, R. Newman. Dept. of Ob/Gyn, Medical University of South Carolina, Charleston, SC OBJECTIVE: Several ~nvestlgaUons have described an increased rate of macrosomia in patients w~th a single abnormal value (SAV) on an oral glucose tolerance test (OGTr) using the National D~abetes Data Group (NDDG) criteria. Intravenous glucose tolerance testing (IVGTr) was formerly thought to provide a more physiologic test of glucose intolerance Oral and intravenous glucose tolerance test results were compared to determine which provides a more sensitive predictor of cesarean delivery due to abnormal fetal growth METHODS: Patients 20 and 32 weeks gestation with a one-hour post 50-gm serum glucose value >_ 135 mg/dl were prospectively evaluated with both an IVGTT and a standard 3-hour OGTI’. Patients with a negative OGTT, a SAV on OGTT, or an abnormal (k < 1.13) or normal IVGTT were prospectively followed without dietary, medical, or obstetrical mterventmn Patients with a positive OGTT or dehvery < 37 weeks were excluded from the analysis. Macrosomia was defined as birthwe~ght > 4000 gms and infants large for gestatlonal age (LGA) were designated by Ballard assessment > 90%.

RESULTS: Of 97 women enrolled, 83 women delivered at > 37 weeks gestation and had < 2 abnormal values on their OGTI’.

Macrosomia LGA C-Section

SAV on OGTT (e=26) 19.2% 50.5% 46.2%

Pos IVGTT (n=21) 19.0% 429% 57.1% Neg 1VGTI’/OGTT (n=37) 10 8% 32.4% 10 8%

Women with both an abnormal IVGTT or a SAV on OGTT had sigmficantly elevated rates (p < 0.05) of macrosomia, LGA, and c-section compared to women with a negative IVGTT and OGTI’. CONCLUSIONS: A SAV on an OGTr yields comparable results to an abnormal IVGTI" in ~dentifying pregnancies at risk for cesarean section due to abnormal fetal growth.

158 THE INFLUENCE OF GESTATIONAL AGE AT ELECTIVE CESAREAN SECTION ON NEONATAL RESPIRATORY MORBIDITY Cochen D. Ash K, BoNn B, Division of Perinatology, University of Ottawa, Ottawa, Canada. OBJECTIVE: To evaluate the influence of gestational age at elective cesarean on incidence of neonatal respiratory morbidity. STUDY DESIGN: A retmspecfive review of all cesarean sections at our institution over an 8 year period was performed to determine the number of elective cesarean sections done in each gestational week (>37 weeks, no labour, no maternal or fetal indication prohibiting delay of delivery). The incidence of NICU admissions for uncomplicated respiratory distress syndrome or transient tachypnea of the newborn was determined by reviewing all NICU admissions following elective cesarean section over the same time period. Confidence intervals for the incidence of respiratory morbidity per gestational week were calculated using

a Poisson Distribution. RESULTS: The incidence of neonatal respiratory morbidity after elective cesarean section at 37 weeks gestation was 69/1000 (95% CI; 25-135), 38 weeks gestation was 1711000 (95% CI; 5.9-

32); 39 weeks was 8.611000 (95% Cl; 1.6-21); 40 weeks was 7.8/1000 (95% Ct; 0.3-30); and 41+ weeks was 011000 (95% CI; 0-14) CONCLUSIONS: A significant reduction in neonatal respiratory morbidity would be achieved by delaying elective cesarean sections from 37 weeks to 39 weeks gestational age.

160 SUBSEQUENT REPRODUCTIVE OUTCOME FOLLOWING CLASSICAL

CESAREAN SECTION AT 24 TO 28 WEEKS. MR Leonardlx SF

Bottoms, Department of Obstetncs and Gynecology, Wayne State Umverslty/Hutzel Hospital, Detrmt, MI

OBJECTIVE: To detarmme ~f classical cesarean dehvery of extremely premature infants adversely influences neonatal outcome In subsequent pregnancies STUDY DESIGN: We ~dent~fied atl women having more than one delfvery at our hosp=tal between 1985 and 1995 and stud=ed the subsequent pregnancy outcomes for pattents dehverlng between 24 and 28 weeks or greater than 37 weeks In the index pregnancy The =nfluence of classical cesarean section and extreme preterm b~rth on subsequent pregnancy outcome were evaluated using MANOVA

RESULTS:

Subsequent Pregnancy Outcome

Classical cesarean Vertex SVD

Index pregnancy 24-28 > 37 24-28 > 37

n 302 253 470 33,975

GA 35 6 38 4 34 8 38 4

Bwth weight 2596 2858 2387 3118

No s~gn~ficant effect of preterm classical cesarean sec~on on birth weight (p>0 7) or gestat=onal age (p>O 5) m the subsequent pregnancy was identified Extreme preterm b~rth ~n the tndex pregnancy had a Nghly sNn=ficant relat~onsNp to both b~rth we=ght (p<0 0001) and gestatmnal age (p<0 0001) =n the following pregnancy CONCLUSION: Extreme preterm dehvery, not classical cesarean section, accounts for poor reproductive performance In subsequent pregnancies Classical cesarean sect=on =s not w~thout morbidity In the subsequent

pregnancy, but our analysis does not support withholding otherwise indicated intervention for concerns of neonatal morbidity =n the subsequent pregnancy

356 SPO Abstracts JanualT 1996 Am J Obstet Gynecol

161 REPEAT CESAREXN DELIVERY: ~,~dAT ARE TIlE RISKS? K~S

CaseL RM Salver, G Jeng~, DW Branch, SL Clark Dept Ob/Gyn, Unlv

of Utah, Salt Lake City, UT, and DeN Blostatlsncs, Umv Iowa, Iowa City,

IA

OBJECTIVE," Many chmcrans feel that multiple repeat cesarean dehvenes

are fraught with serious morbidity However, such cases are often

overlooked or under-represented when evaluating morbidity from cesarean

birth The purpose of our study was to determine whether morbldw Is

increased with increasing numbers of cesarean births

STUDY DESIGN: Pregnancy outcome and surgical comphcauons tbr 1,146

women undergoing their first cesarean (control group) were compared to 636

women undergmng therr second or greater cesarean (study group) from 1 / 1/93

to 12/31/94 These Included 278 women with multiple repeat cesareans

Groups were compared using Ch>square analysis, Fishers exact test, the

Mann "Nhltney-U test, and multiple logistic regression as appropriate

RESULTS: Obstetric comphcauons and the indications for cesarean, such

as placenta prevla, abruptlon, and labor abnormalities, were highly correlated

with morbidity and were more predictive of surgical complications than the

number of previous cesareans After adjusting for these variables, women

undergmng repeat cesareans were mere hkeIy to have hysterectomies (RR

3 28, 95 % CI 1 41 - 7,65), re-operanon (RR 1 83, 95 % CI 1 t 5 - 2 91), and

wound mfecuons (RR 1 41,95% CI 1 04 - 1 91) than those having primary

cesareans The adjusted relative risks were similar among groups for blood

transfusion, bladder injury, re admrssron, endometrms, and Ileus Only the

risk for hysterectomy Increased with increasing numbers of previous cesareans

(p = 0 01) Placenta prevlas accounted for the majority of morbidity in

women with repeat cesareans, including 6 of 8 hysterectomies

CONCLUSIONS: Women wJrb muhJple previous cesareans who desire

subsequent pregnanc’,es should be advtsed of their increased risk for

complications associated with placenta prevla

163 COMPLICATIONS OF MID-TRIMESTER PREGNANCY TERMINATIONS: THE EFFECT OF pRIOR CESAREAN DELIVERY. ~, M A Grlspens~, d Owen, K Savagex, Dept. of OB/GYN Univ of Alabama at Birmingham, AL. . . OBdECT|VF.’. 3"0 deta~m~e the effect cff p~lor cesare~ (C/S) on the incidence of maternal comphcations =n women nav=ng a mid-tr rooster medical termination of pregnancy. STUDY DESIGN: Women undergoing mid-thmester medical terminations of pregnancy ,at,our institution from,January 1980 to July 1995 were rstrospectiveq identified and their charts were reviewed. Obstetric histories, uteroton~c a~lents and the presence of uterine rupture, blood transfusion or retained placenta were ascertained. The frequency of maternal c~o.rn_plications was compared in women with and w thout a previous ~/b. RESULTS: Our study popu, lation ~ncluded 606 women with a mean gestat~onal age of 21 1 weeks (range 11 to 28). The mean maternal acle was 26.3 + 7 years. Forty-eight percent were non-white and 58% of women-were porous. Seventy-nine (13%) had undergone a prior C/S Of those, 66 (83.5%) were confirmed low transverse ~ncislons 2 (2.5%) were low vertica~ incisions, 1 (1.3%) was a classical incision, and 10 (13.0%) were unknown scars. The nd cat on for delivery was fetal anomalies in 274 (45.2%), preterm rupture of membranes or oligohydramn~>s in 111 (18 3%), rr~. te~na~ d{sease in 107(180%), fe~al death in 100 (165%),and other indicat des in 14 (2 0%,). Induction method wa~..vag~na[ ,PG~E_.~n 3,22 (53.1%), concentratee oxytoc~n in 125 (20.6%), vaginal concentrated oxytocin in 89 (14.7%), standard oxytocin tn 38 and ~ntra-amnlotic urea and PGF~e’ in 32 (5..3°~). There was n.o significant difference in the need fdr curettage oe~een ~omen w~;n and without a pror C/S(21/79, 26.6% vs 134/527 25.4~, p=O.&3). There was an increased need for blood transfusions in women with ~ prior C/S (9/79 11.4% v~. 28/527~ 5 3%, p-~0.04}. The incidence of uterine rupture was significantly h~gher among women w~th a prior C/S (3/79, 3.8% vs.. 1/527, 0 2%, p=0aO08). .When comparisons w~ere mired only to muKiparous women me incidence OT uterine rupture

was st h gher among women with a prior C/S (3/79 3 8% vs. 0~7_0., 0% p=O.03). All four. uten.ne, rupt.ure.s, were in.pre.gn,ancies weeks’ gestation, haa oxy~ot c ~nduc[Ion agents, nod a prior lOW transverse or unknown scar, and requirbd surgery (repair or

h~/oSterectomv)- NCLUSION: Our data suggest that a p_rior C/S ~s a significant risk factor for utenne rupture an~blood transfusions in women having a mid-tnmester pregnancy termination by I~bor =nduct~on.

162 ANALYSIS OF REPEAT CESAREAN INDICATIONS" IMPLICATIONS OF

HETEROGENEITY. ML Haa~eyX, JC Smd~an, MF LakeX, AM V~tzdeos, DA McLean O~v of

Maternal-Fetal Medicine, UMDNJ-Robed Wood Johnson Medical School/St Peter’s Medical Center, New Brunswick, NJ OBJECTIVE: To descnbe the charactenshcs and relative contebuhons of three distinct groups of patients to the overall repeat cesarean dehvery rate, STUDY DESIGN. Hospital records of all women with a previous cesarean section (CS) who dehvered between 1/1194 - 6/30/94 were rewewed Those who had a repeat CS were categorized as follows (1) elective repeat cesarean (ERC), (2) "indicated" repeat cesarean 0RC) diagnoses inc)uded maternal condlhons, prevta, previous dehiscence, prewous classical incision, prevl0us T incision and/or extensive lacerahons, previous myomectomy, severe fern compromise, breech, -> 3 prewous cesareandehvenes, macrosomla [estimated fetal weight (EFW) _> 4500 gms], twins, and (3) failed vaginal birth after cesarean(FVBAC) Descnptlve and outcome data were collected Data were analyzed using conhngency tables, parametric and nonparametnc tests where appropnate, S~gmflcance was set at p< 0 O5 RESULTS: There were 376/406 records available for review Of these, 235 had a repeat CS 107 (45%)= ERC, 56 {24%)= }RC, and 72 {31%)= FVBAC The ERC and c0mpnsed a group of pahents not given a trial of labor. The remaining 141 patients had a successful VBAC dehvePj There were no slgndlcant differences with respect to maternal age or medical c0rnp/tcatlens of pregnancy among the 3 groups Sma~ but slgelhcant efterences were found among the groups regarding est~mat~ fester/peal age, grav~lt~, parity, and number of previous CS In 50 5% (103/204) of patients wdh prlvate/HMO insurance an elective procedure was performed, compared to 12 9% (~/31} of patients w~th Med~ca~dlself~ay, p= 0004 Pahent refusal of VBAC was noted tn 52 6% {56/107) of the ERC group AdCtlonalty, ~n 13 1% (141107) ot the ERC group, nonmacresom~c (<45g0 9ms) EPWs were {inked with tile decision to perform a repeat cesereae The rr~st common indication for a repeat CS in the FVBAC was dystocla 80 6% (58/72) Post op comphcahons were slgnlficantq greater (p=02) for the IRC group (13/56, 59 1%) as compared to both the END group (3f107, 13 6%} and the EgAC groups (6172, 27 3%} CONCLUSIONS: 1) Circumstances leading to repeat CS, as well as petlent charactenstlcs and outcomes are vaned, strategies to reduce repeat CS rates must address these ¢fferences and r~flect the poputahons o~ each hospital setting, 2} Not all pahents who undergo a scheduled repeat CS are candidates for a teal of labor (24% of patients had IRC), 3) WNe interventions aimed at reducing refusa~ rates and resolving ~ssues of macrosomla may be of beheld, tNs is unhkely to alter the 31% of repeat CS in the EVBAC group

164 ONE VS. TWO LAYER CLOSURE OF A LOW TRANSVERSE CESAREAN: THE NEXT PREGNANCY. ~LChanman, J Owen, JC Hauth, M DuBard~, K Savage× Dept. of OB/GYN, Universi~ of Alabama at B~rmingham, B~rmingham, AL. OBJECTIVE: To determine if low transverse cesarean closure method (one vs. two layers) affects subsequent pregnancy outcome. STUDY DESIGN: In a previously reported prospeohve trial, 906 women were assigned to either a one or two layer closure of their low transverse inc~s~n. Of these, 164 were identified as having a subsequent pregnancy and dehvery (>18 weeks) at our institution and comprised the study populahon. Maternal and neonatal outcomes were ascertained by medical chart review. RESULTS: Of the 164 subsequent pregnancies, 83 had been prewously closed in one layer, while 81 had undergone a two layer closure. The groups were similar with regard to age, race, par~y, and gestatlonal age at delivery of the subsequent pregnancy. Nineteen of 164 02%) underwent a repeat cesarean w’rthout labor, and the remaining 145 expenenced labor e~her with (n=85) or without (n=60) oxytocin shmulation; there was no difference between the two groups. Length of labor, mode of delivery (cesarean vs. vaginal) and the incidence of chorloamnion~t~s, postpartum metrrhs, hemorrhage, transfusion and abnormal placentatlon were also similar ~n women with 1 vs 2 layer closures (p value range = 0.12 - 1.00). Immediate neOr~tal outcomes (Apgar scores, cord pH, birthweight) were also similar Four pennatal deaths occurred ~n the 1 layer group: 2 stillboms and 2 neonatal deaths from extreme prematurity. Five deaths occurred in the 2 layer closure group: 2 stillboms and 3 neonatal deaths again from extreme prematurity One patient in the 1 layer group hada uterine scar dehiscence and required a cesarean for fetal heart rate abnormahty. This pahent had no catastrophic utenne bleeding requtnng transfusion or hysterectomy Th~s newborn had Apgars of 7~ 9~ and arterial cord pH=7.16. CONCLUSIONS," These findings suggest that closure of a low transverse cesarean in one continuous locking layer does not significantly affeCt the outcome of the next pregnancy.

Volume 174, Number 1, Palt 2 SPO Abstracts 357 Am J Obstet G),ne(ol

165 LOWERING THE CESAREAN SECTION RATE IN A PRIVATE HOSPITAL: ANALYSIS OF INDIVIDUAL PHYSICIAN PRACTICE. D C.Lagrew, and JAAdashek Saddleback Memorial Medical Center, Laguna Hdls, CA and Univ ofCahf, Irvine, Orange, CA

OBJECTIVE Our purpose was to compare the practice patterns and outcomes of physicians delivering at our restitution to identify risk factors and management techniques which could explain the differences in individual cesarean section rate

(CSR) STUDY DESIGN We retrospectively reviewed detaded computerized delivery records (n = 16,230) collected from 5/16/88 to 7/30/95. We excluded physicians who had less than 100 deliveries at our institution dunng the study period The physicians were then divided into two groups depen&ng on whether or not their individual CSR was greater than or less than 15 percent Various cesarean section rates, risk factors for abdominal dehvery, labor management techniques and neonatal outcome parameters were then calculated for each group The cesarean section rates oftbe two groups were analyzed by year to assess changes RESULTS As expected, by study design, the total CSRs were markedly different between the two groups (24.1 vs 13.8%) In addition, the primary, repeat, pnmagrav~da, and multtparous CSRs were all significantly less m the lower CSR group The rates of cesarean section for fetal distress (1 5 vs 3.3%) and cephalopelwc disproporhon (5 3 vs 8.4%) were also significantly less The rates of breech, tbard trimester bleeding, and active herpes cesarean sections were not lower The bagher CSR group had s~gmfieantly more post-term (8.3 vs !4 3%) and > 4000 gram ~nfants (12 0 vs 13.5%) but similar numbers of low barth weight, mull~ple gestalaon and preterm in fants Tbe lower CSR group used more epidural anesthesia, p~toc~n ~nductions, tnal vaginal births after cesarean (VBAC) and had more successful VBACs Over the study period the CSR in the lower group remained stable but the bather group had progresstve reduction in the numbers of

CONCLUSION: Individual physician’s lower casarcan sectmn rates are primarily achieved by labor techniques and attempting VBAC The practice patlems used by our physicians led to no increase in neonatal morbid~ly

167 IS VICRYL SUTURE CLOSURE OF UTERINE WOUNDS ASSOCIATED W’ITH MORE DEHISCENCE? L. Zuidema, R. Elderkln, C. Cook, R. Jelsema. Butterworth Hospital, Grand Rapids, MI, OBJECTIVE: To determlne a posslble associatlon of suture type (chromlc vs vlcryl) with uterine scar dehlscence. STUDY DESIGN: All patlents undergoing primary cesarean section durlng the years 1987-1989 were retrospectively identlfled. This was a time period during which the uterine incision closure technique was changing from a chromic closure to a vicryl closure, and durlng which patients were increaslngly being encouraged to undergo vaginal birth after cesarean. Charts were reviewed for subsequent obstetrical history. RESLrLTS: 1784 patients were identified as undergoing a primary cesarean. 621 charts were available for revlew. 84 had no further recorded pregnancies, leaving a study population of 537 patients. 237 patlents (44%) had chromic closures, while 302 (56%) had vicryl closures. There was a total of 17 uterlne scar separations (3.2%), 9 of which were noted at the time of electlve repeat cesarean and 8 dlagnosed at the time of failed VBAC. Three of the separations occurred in the chromic group, while 14 separations occurred in the vlcryl group. The odds ratlo of uterine scar separation occurring with a vlcryl closure compared to a chromic closure was 3.70 (1.01-16.8). CONCLUSION: V~cryl uterlne closure is associated with an increased incldence of uterine scar dehiscence.

166 TRIAL OF L~-BOR AFTER MULTIPLE CESAREAN BIRTH-A META- ANALYSIS. I.Kornfeld(X),K Amankwah,R.Kung (X) Women’s College,Hosp.,Dep. Ob/Gyn,Univ of Toronto,Canada OBJECTIVES:To examine the success rate and safety of trial of labor(TOL) after 2 and respectively 3 or more cesarean sections(CS) using as controls patlents with TOL after 1 CS. STUDY DESIGN:Data sources:Literature search 1950-1995 Search terms:TOL,VABC Study selection:cohort studies of patients with TOL after slngle and multiple CS.Out -come measures:vaginal delivery(VAD) and uterine rupture(UR) rates.Data synthesis:odds ratios wlth 95% cI for V~!9 and UR rates after 2 and 3 or more CS were calculated by Mantel-Haenszel. Results:15 cohort studies met the entry crlteria.2162 patients had TOL a~ter 2 CS and 304 patients had TOL after 3 or more CS~For those patients with 2 CS,the chance of VAD was 85% of that of patients with TOL af ter 1 CS(odds raLio-0 42,CI-0.36 0.45;RR=0 85,CI=0.8- -0.87) and the risk of UR was 3 times greater which is statistically signlficant(odds ratio=3.2,CI-2 06 4.97;RR=3.16,CI=2.09 4.78)No statistically slgnlfi-

cant difference was noticed when the same outcome measures were analysed in patients with 3 or more CS CONCLUSION:TOL after 2 CS has a statistically slgni ficantly lower VAD rate and higher UR rate than TOL after 1 CS.The efficacy and safety of TOL after 3 or more CS cannot be ascertalned in comparison to TOL after 1 CS as the number of patients who experienced this management is too small to assess statistlcal significance.

168 POSTOPERATIVE INCISION COMPLICATIONS FOLLOWING CESAREAN SECTION IN PATIENTS WITH ANTEPARTUM HELLP SYNDROME: DOES DELAYED PRIMARY CLOSURE MAKE A DIFFERENCE? R. Bri£,,s, R. S. Chari, B. Mercer, and B. Sibai. Dept. of Ob/Gyn, University of Tennessee, Memphis. OBJECTIVE: To evaluate wound complication after primary closure and delayed closure, as well as Pfannenstiel and midline skin incision after cesarean section of women with antepartum HELLP syndrome. STUDY DESIGN: Medical records of women with antepartum HELLP that were delivered by cesarean section at our institution were reviewed. The immediate (prior to hospital discharge) and late (2 week postoperative check) wound complications were analyzed with respect to wound closure, and type of skin incision. RESULTS: A total of 104 patients were identified; 75 patients had a primary skin closure and 29 patients had a delayed closure 48-72 hours post operatively. Immediate wound complications including wound infection, and hematoma, occurred in 18 (24%) patients with primary closure versus 8 (28%) patients with a delayed closure (p=0.900). A late wound breakdown was seen in 1 (1.3%) patient with primary closure versus 0 with delayed closure. There were no fascial wound dehiscences. Additionally, no benefit was identified with either type of skin incision. (see table).

Wound Breakdown Primary Delayed OR 95% CI

Pfannenstiel 13/53 4/22 1.5 0.4-6.2 Midline 5/17 4/12 0.8 0.1-5.4 OR 0.8 0.4 95% CI 0.2-3.1 0.1-2.9

CONCLUSION: In women with antepartum HELLP syndrome delivered by cesarean section, no difference in postoperative wound complications in primary versus delayed closure or type of skin incision was found.

358 SPO Abstracts Janualy 1996 Am J Obstet Gynecol

169 POSTOPERATIVE MANAGEMENT OF CESAREAN

SECTION PATIENTS: THE EFFECT OF IMMEDIATE

FEEDING ON THE INCIDENCE OF ILEUS. R, KramerLx J Van Someren,~’C Qual(s~ L Curet Dept Ob/Gyn, Umv of

N M, Albuquerque, NM

OBJECTIVE: To evaluate the effect of =mined=ate feeding on

gastrointestinal functton m pahents undergoing cesarean

section.

STUDY DESIGN: A prospecttve study was conducted m which

200 pahents were random=zed to e~ther early feeding or

feeding according to the institution’s current protocol

Questionaires were filled out by the subjects on the day of

dmcharge F~sher’s exact test was used to compare the two

groups and to compare the incidence of gastrointestinal

symptoms A one-sided binomial confidence ~nterval was

used to determine the upper bound of the hkehhood of

paralytic deus

RESULTS: There was no slgmficant d~fference between the

control and study groups There was no significant d=fference

~n the number of gastrointestinal symptoms between the two

groups. The incidence of postoperative paralytic fleus was zero in both groups

Symptom Controls S_.tudy Group p value None 43/109 45/91 .20 Abd pa~n 451109 37/91 1 0 D~stens~on 32/109 22/91 42 Nausea 16/109 13/91 1 0 VomJtmg 6/109 5/9! 1 0 CONCLUSIONS: Routine eady feeding of cesarean secbon

patients can be =mplemented w~thout an increase m

gastrointestinal symptoms or paralytm fleus

171 UTERINE RUPTURE AFTER PREVIOUS LOW TRANSVERSE CESAREAN DELIVERY (LTCS) J. Van Winter. M. Evans, K. Ramin, P. Ogburn, Jr., M. Webb, K. Podratz. Mayo Medical Center, Rochester, MN. OBJECTIVES: To identify the incidence of and risk factors associated with uterine rupture after previous cesarean delivery. To discuss the different presentations of uterine rupture following vaginal birth after cesarean (VBAC) attempt and assess maternal/neonatal outcome. STUDY DESIGN: A retrospective study with review of charts and monitoring strips. RESULTS: Between January 1, 1994 and July 31, 1995, 129 patients attempted a VBAC defivery. Ninety-seven (75%) of patients had a successful vaginal delivery and. the majority (95%) of infants delivered vaginally had a 5-minute Apgar score of 8 or greater. Oxytocin induction or augmentation was used in 90 (70%) of patients to effect vaginal delivery. Three (2.3%) of patients experienced complet~ uterine rupture resulting in 2 hysterectomies. One rupture was detected during labor and 2 postpartum following low forceps delivery. In 2 of the 3 cases, 1 or more spontaneous vaginal deliveries occurred between the initial cesarean delivery and subsequent delivery with uterine rupture. Maternal characteristics or intrapartum events were not predictive of the extent of uterine rupture. CONCLUSIONS: Normal spontaneous vaginal delivery following cesarean delivery does not appear to decrease the potential for uterine rupture in subsequent pregnancies. Therefore, a trial of labor should take place in a hospital setting where emergency cesarean section can be performed. Uterine rupt~e presenting several hours postpartum mandates against early dismissal of V’BAC patients.

170 HISTOLOGIC MYOMETRIAL FIBERS ADHERENT TO THE PLACENTA: IMPACT OF METHOD OF PLACENTAL REMOVAL. H. Hardardothr, A.F Borgida, M. M Sanders*, L. Ernst*, W. A.

Campbell University of Connecticut Health Center, Farm~ngton, CT. OBJECTIVE: Myometnal fibers (MF) adherent to the placenta may represent foci of placenta accreta. We investigated whether manual removal versus spontaneous delivery of the placenta affects the presence of MF on h~stologic examination. STUDY DESIGN: Over 3 months, consecuhve pahents undergoing cesarean dehvery were randomized to manual extraction or spontaneous expuls~or~ of the placenta. The pathologist was bl~nded to removal method. F~ve grades were used to tndtcate the extent of adherent MF: 0 (none), 1+ (1 focus m one t~ssue sect=on), 2+ (2 foc~ in one hssue section), 3+ (> 2 foci in multiple tissue sect*ons), 4+ (extens=ve), The pre- and post-operative Hgb/Hct, estimated blood loss, gestational age at dehvery, and h~story of previous cesarean sectmn were also recorded Mann-Wh=tney U test, t-test and test of proportions were used for analys~s, and p < 0 05 was s~gmficant. RESULTS: Twenty-e~ght pahents at a mean gestatmnal age of 33.3 and 32.4 weeks, respectwe~y, were randomized 1o manua~ vs. spont dehvery of the placenta (p= 0 6) Eight patients had a previous cesarean (5 manual, 3 spont p = 0.16) Removal Method Adherent MF Grading

O ->1+ 11+ 2+ 3+ 4+

Spontaneous del (~=14) ~ 6 4 1 1 0

I I

The mc{dence of adherent MF was no different in manual (7/14) vs spont. (6/14) dehvered placentas (p= 0.99) Adherent MF grading did not differ between manual and spont, dehvery (p= 0.5). In pat=ents w=th prewous cesarean (8/28), 3/8 had adherent MF, nit dehvered manually There was no s~gmficant d=fference m the pre- or post-operative Hgb/Hct or blood loss between the groups (p>0.10). CONCLUSION: Adherent MF to the placenta ~s a common finding at cesarean section and seems unrelated to method of removal.

172 UTERINE ACTIVITY PATTERNS IN UTERINE RUPTURE PkTIENTS:

A CASE- CONTROL STUDY. Jeffrey P Phelan, M.D, LJsa M Korst, M D,

Derma Setdes, RNC, Dept of Ob/Gyn, Pomona Valley Hospital Medical Center,

Pomona, CA

OBJECTIVE: To determine whether utenne actw~ty patterns or the use of

oxytocm ~s responsible for utenne rupture giwng rise to fetal brain injury.

STUDY DESIGN: Case-patients were women who had prior low transverse

cesarean dehverles (CS), and who underwent uterine rupture resulting m a

neurologically ~mpalred neonate (NIN) Two groups of control patmnts were

systematically selected from consecutive dehvenes One group underwent

successful vaginal birth after cesarean (VBAC), the other group underwent

spontaneous vaginal dehvery (SVD) with no history of prewous utenne scar The

uterine activity patterns of case-pahents were compared with the patterns of each

control group With each patient’s fetal monitor strip, the number of recorded

utenne contractions (CTX) per strip, CTX per hour, tetanm episodes (CTX > 90

seconds), oxytomn use, hyperstimulation events (5 or more CTX m a 10 minute

period) and the total FHR decelerations were compared

RESULTS: We studied 18 case-patients, 33 patients with SVD, and 35 patients

with VBAC W~th the exceptton of increased panty in the groups with prmr CS,

the demographic features among the 3 groups were comparable The utenne

actIVlW patterns revealed fewer CTX per hour m the uterine rupture group (15 g

+ 7 3) compared wtth both the SVD group ( 19 4 + 6 6) (P = 0 09) and the VBAC

group (19 7 + 5 5) (P = 0 03) At the same hme, the use of oxytocm, the number

oftetamc and hyperst~mulahon epxsodes were statmt~cally sxmflar

CONCLUSION: Neither uterine contraction patterns nor the use of oxytoc~n

appeared to correlate with subsequent uterine rapture

Vohune 174, Nuniber 1, Part 2 SPO Abstracts 359 AIn ] Ob~tet Gynecol

173 EPISIOTOMY AND PER1NEAL MORBIDITY: A NATURAL

EXPERIMENT. RK B~nsalx, WM Tanx, JL Eckerx, SJ Kzlpatnck

Dept, of Ob/Gyn, Umv of CA, San Francisco; San Francisco, CA

OBJECTIVE: To compare changes in maternal permea] morbidity

assocmted with a chronologic decrease m the use of epImotomy

STUDY DESIGN: We accessed records from 21,601 consecuhve,

term, s~ngleton vaginal dehvenes between 1976 and 1994 ~n the UCSF

perinatal database. We determined whether dehvery was assJsted or

spontaneous, if epIsmtomy was performed, and the degree of any

permeal laceratmns Rates and proportions over time were compared

using ebb-square and linear regression.

RESULTS: From 1976 to 1994 the rate of episiotomy m spontaneous

delivery at our institution fell from 86.8% to 10.4% (r=.96, p< 0001)

A concomitant fall in the rate of third and fourth lacerations from 9.0%

to 4.2% (r= 77, p< 0001) was correlated with the fall in eplslotomy rate

(r=.83, p<.0001) The propomon of women w~th ~ntact permeums after

delivery rose from 10.3% to 26 5% (r=.83, p< 0001) Although the rate

of ep~siotomy also fell from 1984 to 1994 for those undergoing

operahve vaginal dehvery (93 4% to 35 7%, r= 92, p<.0001), the

proportion of women w~th third and fourth degree laceratmns stayed the

same (32.0% to 32 4%, r= 24, p= 48) The number of women in the

operahve group w~th ~ntact perineums rose slightly (2 2% to 7 0%;

r=.61, p=.046), but this was not correlated with the decreasing

episiotomy rate (r= 53, p= 09) Over the interval studied, there was no

change m bxrth weight, % occ~put posterior, ethnacity, panty, or rate of

cesarean section that would explain these changes in penneal trauma

CONCLUSIONS: Over time, a signffmant decrease in ep~siotomy ~n

spontaneous deliveries was associated with a s~gnff~cant decrease ~n the

rate of comphcated penneal laceratmns Smailar changes were not seen

when women undergoing operative dehvery were separately stuthed

175 THE EFFECT OF EARLY NEONATAL DISCHARGE ON READMISS~ON RATE IN A TERTIARY MEDICAL CENTER E.J.Brau~, B.M.Rosenn, T.A.Siddlql, M.Miodovnik. Dept. Ob/Gyn., Unlv. of Cinti., Cinti., OH. OBJECTIVe: To test the hypothesis that early neonatal discharge from a tertiary medlcal center is associated with an increased rate of neonatal readmissions. STUDY DESIGN: An early discharge program was instituted at a tertiary medlcal center in July 1993, with the object of dlscharging mothers and babies 24 hours after a vaginal delivery, and 48 hours after a cesarean section. Maternal educational goals were achieved prlor to dlscharge, and all patients had at least one home visit by a qualified nurse to assess maternal and neonatal status. Rates of neonatal readmission durlng the early discharge program period were compared to the previous period of conventional postpartum discharge. RESULTS: From 1/1/92-6/30/93 (conventional discharge period) 50 out of 4341 neonates (1.15%) were readmLtted within i0 days of discharge. From 7/1/93-3/31/95 (early discharge period), 22 out of 4157 neonates (0.53%) were readmitted. Characteristics of readmitted neonates are presented in the following table:

Conventional Early discharge discharge

Readmlssion rate 1.15% 0.53% Days from discharge to 5(1-10) 4(1-10) readmisslon (median, range) Readmission length of 4(1-17) 2(1-15) stay(median, range) Readmission diagnosis Jaundice 22% 46% Infectlon 36% 9% Gastrointestinal 24% 27% Fever 10% 0% Other 8% 18%

CONCLUSIONS: A structured program for early neonatal discharge does not increase the risk of neonatal complications.

174 EPISIOTOMY AND PERINEAL TRAUMA IN

SPONTANEOUS VAGINAL DELIVERIES. Susan M Lanmx, Anthony Ambrose Penn State Umversity, Hershey, PA

OBJECTIVE’ To detenmne If midline ep~siotomy pre&sposes to

third and fourth degree (high degree) lacerations in spontaneous

vaginal deliveries at term.

STUDY DESIGN A retrospectave review of the 1992 delivery

log at the Penn State University Medical Center was performed,

including gestations from 37 to 42 completed weeks which resulted

in the spontaneous vaginal delivery of a fetus ~n the cephalic

presentataon (n=335). All episiotomies were midhne. A logistical

regressmn model was created to control for fetal birth weight, and

maternal panty, wnh the outcome variable being the presence of a

high degree laceration

RESULTS The overall episiotomy rate in the study population

was 35% In umvariate analysis, 24 of 116 (20 6%) women having

epIsiotomles sustained high degree lacerations, whtle the "no

episiotomy" group included 6 of 219 (2 7%). Nulliparous women

sustained 21 of 91 (18.7%) high degree lacerations, while

mult~parous women had 9 of 214 (4%) high degree lacerations

Multivariate analysis revealed episiotomy (odds ratio=7 07, 95%

Confidence interval=2 73 - 18 22) and maternal panty (odds ratio-0 23,

95% CI=0 096-0 565) were both strong predictors of anorectal

trauma Fetal hirthwe~ght failed to be a predictor of perineal trauma

in both m-and multivariate analysis.

CONCLUSION: Mldllne epIstotomy placed patients at 7 times

the risk for third or fourth degree lacerations, controlling for the

effects of maternal panty and fetal birth weight

176 DOES THE AVAILABILITY OF AMNIOINFUSION REDUCE THE INCIDENCE OF MECONIUM ASPIRATION SYNDROME (MAS)? R SlgInan, Dept Ob/Gyn, Washington Umv,

St Louis, MO

OBJECTIVE: The null hypothes~s was that there was no ddl’erence

m the incidence or seventy ofmeconium aspiration syndrome m the

years since Its Inceptmn compared to five years prior to ~ts use

STUDY DESIGN: All 94 infants d~scharged with the diagnosis of

MAS between the years 1983 and 1994 were ~denhfied and their

mother’s charts, the distillate of 49,454 dehvenes, were examined A

database was developed looking at all risk vmables as well as delivery

events, tins was correlated to neonatal outcomes Statistical evaluation

was performed by the SAS stattat~cal package Ammolnfusmn use

became prevalent m 1989 A comparison was made between the years

1983 to 1988 and 1989 to 1994

RESULTS: The rate of MAS pro-1989 was 001919, while the rate

post-1989 was 001855 This gave a difference of 0006 which was

not s~gmficant, 95% CI (-0 000711, 0 000839) The power to detect

a 50% change was 81 Of the 39 post-1989 infants, 13 received amnio~nfuslon The decision to use amnm~nfuslon was based on

knowledge of the presence ofmecomum, identification of vanable fetal heart rate decelerattons, or tane avadable before dehvery The severl/y

of &sense as indicated by cord ph, apgar score, and days of hospital

stay were sumlar The frequency of caesarean section for fetal distress

was 65% with no d~fference between groups

CONCLUSIONS: The avaflabdfiy of ammolnfus~on has not proven

successful m reducing the ~ncidence of MAS

360 SPO Abstracts JanuaiT 1996 Am ] Obstet Gynecol

177 NEL’~/IUM-STAJNED A~/IOTIC FLUID: IS IT ASSOCIATED ~ITH ELEVATED ERYTHROPO]ET[N LEVELS? S.Bloom, S. Ramin, S. Neyman,x B. Little,x L. Gitstrap, Dept. Ob/Gyn, Univ. TX Southwestern Medical Center, Dallas, TX OBJECTIVE: Elevated levels of erythropeietin have been measured in term infants with meconium-atained amniotic fluid in spontaneous labor possibly suggesting either chronic stress or hypoxia. ~e sought to determine whether erythropoietin levels were significantly different in infants with meconiuta-stained amniotic fluid whose mothers were no__t in tabor. STUQY DESIGII: Fifteen parturients at term, not in labor, were divided into two groups: Group I consisted of 5 wo~en who had meconium-stainedammotic fluid discovered either at amniocenteais for fetal lung maturity or at elective repeat cesarean delivery. Group 2 consisted of 10 women who had clear amniotic fluid. All 15 women underwent elective repeat cesarean deliveries. The meconiL~-stained am.~iotic fluid was centrifuged and the volume percent of particu[ate matter (i.e., meconium) measured. Umbilical artery blood and mixed arterial/venous cord blood were obtained at each delivery. Erythropoietin levels were measured by enzyme-linked irm~unosorbent assay. RESULTS: NO significant difference in maternal age, parity, estimated gestational age, birthweight, umbilical artery blood pH or 5-minute Apgar scores were found between the groups. All of the infants had an ufabilica[ artery blood pH > T.10. The volume percent of particutatematter in Group I ranged between 3 and 10%, consistent with "thin" meconium-stained amniotic fluid. The mean erythropoietin levels were significantly higher (P=O.04) [5T.5±40.2 mlU/mL (range 15.6-116.8)] for Group I compared with Group 2 [23.9~18.0 mlU/mL (range 7.7-61.5)]. CONCLUSION: Erythropoietin leveis were significantly elevated in newborns with thinmeconium-sta]ned am~iotic fluid delivered from mothers who were not in labor. These results are consistent with results obtained in women withmeconium-sta~ned~otic fluid who were in labor. Finally, whether meconium in such women is a marker for either chronic stress or an acutet but transient episode of hypoxia remote from delivery is unclear at this time.

179 THE FACTOR V LEIDEN MUTATION IS NOT ASSOCIATED WITH IDIOPATHIC RECURRENT MISCARRIAGE D. Dizon-Townson1, S. Kinneyix, D.W. Branch1, K. WardL2. Depts. of Ob-Gyn1 and Human Genetics2, Univ. of Utah School of Med=cine, Salt Lake City, UT. OBJECTIVE: Previous investigators have suggested that placental thrombosis and infarct=on can cause recurrent abortion (RAB). A common missense mutation in the factor V gene, the Leiden mutation, renders factor Va resistant to cleavage inactivation by activated protein C and predisposes to thrombosis. We have found that the Le=den mutation is associated with placental infamts at term. Our object=ve was to evaluate the frequency of the factor V Leiden mutation in a RAB population. STUDY DESIGN: DNA was’extracted from whole blood of 43 couples w=th a history of id=opathic recurrent miscarriage and 19 couples w~th a history of proven fertdity (8 or more live births). The polymerase chain reachon was used to amplify exon 10 of the Factor V gone followed by allele-specific restriction with Mnl 1 for mutation detection. Results were analyzed with a Z2 contingency table. R E S U LT S: Allele Frequencies

1691G 1691A (Mutation)

RAB Females (n=4.3) 83/86 3/86

RAB Males (n=43) 86/86 0/86

Control Females (n=19) 38/38 0/38

Control Ma/es (n=19) 35/38 3/38

CONCLUSIONS: Couples with a history of recurrent miscarriage do not show an increase in the frequency of the factor V Leiden mutation. DNA analysis for the Leiden mutation is not useful in the routine evaluation of these patients.

178 PROGRESSIVE STRUCTURAL CHANGES ALSO IN THE RADIAL ARTERY DURING GUINEA PIG PREGNANCY. _C Verkeste×, M Daemenx, B Slangen~, H van Straaten~, G

Kohnen~, L L H PeetersI Umverstty of Limburg, Dept Gyn!Obst~, Patho logy~, Anatomys, Maastncht, The Netherlands and RWTH, Dept Patholo gy4, Aachen, Germany.

OBJECTIVE: Pregnancy-induced mvasmn of trophoblast rote spiral arteries

leads to structural w~denmg of these vessels, a prereqmsIte for succesful fetal outcome We investigated whether structural changes also occur in

radial and arcuate arteries, which dilate dunng pregnancy and might be revolved m the regulatxon of placental blood flow STUDY DESIGN: The vascular system of non- (NP, n-4), m~d- (MP, n-4) and late-pregnant (LP, n=6) guinea pigs was perfuslon-fixed (10%

phosphate buffered formalin / Na-mtruprusslde) under general anaesthesia Uterine vascular segments were fixed overmght (10% phosphate buffered formalin) and embedded m paraffin Cross-sections of radial and arcuate arteries were stained with Hematoxylin and antibody against cytokeratln to

detect trophoblast (MNFll6, only LP) All sections were evaluated by hght microscopy RESULTS: In the course of pregnancy, progressive endothehal swelhng and vascular smooth muscle reducUon was observed in rathal arteries These

changes seemed to be correlated to the migratmn of multmucleus grant cells

(MGC) from the mtravasc space across the endothehum to the subendothe- hal space The MGC, which appeared to be of placental ongm, stained positively with the cytokeratln marker These phenomena were neither observed in arcnate arteries (MP & LP), nor in the radial and arcuate arteries of NP ammals

CONCLUSION: Structural physmlogical changes were observed in guinea

pig radial arteries These alterations might be a prereqmsltc to enable the pregnancy-related ~ncrease in placental blood flow and the regulation of placental blood flow by the radml artery

180 FETAL FIBRONECTLN: A NEW TOOL FOR PREDI(TrING SUCCESS OF LABOR INDUC’rION. TJ Game, D Casalx, A Garc~a*, G Jimmez×, JA Ayalax, M Vdlax, JC Izquaerdox, T Rem~buldx, UC Irvme, Orange, CA and

HoN:atal Ltas Castelazo Ayala, Mexico C~ty, Meyaeo

OBJECTI’VF,: The Bishop score ~s the only ava~lab!e reel for the pred!ctmn of

successfifl labor mduc’aon Vaginal Fetal fibroneetm (Ffn) has been shown to

prechet whmh pafaents wall enter labor spontaneously, usually vathm 2 weeks

Tins study was desagned to determine ff Ffia can also predaet successful labor mduetlou

STUDY DESIGN: Over a 1 year penod, palaeuts undergoing labor mductmn

at~r 36 weeks had a smgte sample from the va~na tested for Ffia pnor to the

tmaa~aon of cerwcal npe~m~g and/or oxytocm mductmn The swab was tested

wth a quahta~ve ~mmune assay for Ffn, w~th a point, re value set at >- 50 ng/ml

Ffn results were blinded to the chmclan

RESULTS: Of the overall 160 subjects, no sa~mficant d~fferenees were found m

duralaon or success of reduced labor The remainder of the results are restncled

to the subanalys~s of the chmcally ~mportant ~roup of nullipnrous pzlients with Bishop score <= 5. In th~s subgroup, 52 had a (+) and 20 a (-) Ffn The mean-

interval between the first dose of prostm and vaginal d~hve~¢ was substanl~ally

shorter m palaents wth a + Ffia (21 7 vs ~3 1 hrs, P-~) 0004) ns was duratmn of

the first stage of labor (17 8 vs 22 8 hrs, P=0 01) Whde heady all received a

first dose of pros~n, over half the patients m the - Ffia group (ll/20) were

undehvered at~er 24 hours aud were judged to requtre a second course of prostm compared to only 2/52 in the + Ffn group (P=0 000001) In ad&laon,

the mara~um d~se ~f oxytoem r~quxr~d was substar~aally less m the +Ffn group

(3 9 vs 6 7 mU/mm, ~ 0001) C-Seclaon rates &d not reach sta~stacal

Sl~mficance (11/52, 20% +Ffia vs 7/20, 35% -Ffia, ~ 23)

CONCLUSIONS: Va~al Ffia appears to be an efficacious new test v2mch predacts which parents ~ have a shorter and earner mduclaon of labor Since

the malonty of patients vath a -F~ test reqmred a second day of mdue~aon and

ver2¢ few wth a +Ffia d~d, this test may have great clinical utthty and reduce cost Stu&es wath larger samples may also show that th~s test may be used to

avoid UnlleCessary cesarean sections

Volume 17t, Nnmber 1, Part 2 SPO Abstracts 361 Am J Obstet Oynecol

181 UMBILICAL CORD TWlffr: RELATIONSHIP WITIt PLACENTAL

GROSS MORPIIOLOGY S.Shen-Schwarz=, E.King’, C.Benito’, E.Guzman,

J.Smullian, A.Vintzileoa UMDNJ-Robert Wood Johnson Medical School / St.

Peter’s Medical Center, New Brunswick, N J.

OBJECTIVE: Absent umbilical cord twist has been reported to be associated

with adverse permatal outcome. We analyze the relmionship between cord twist

and the gross morphology of cord, amnion-chorion and plaeen~ disc.

STUDY DESIGN: In 1993, placentas from all singleton deliveries, including late

second trimester miscarrtnges, were routinely examined by a pathologist. These

records were reviewed to ascertain cord twist (lefi, right, mixed and absent),

mtonse cord splrahng, true knot of cord, cord insertion (disc, marginal and

velamentous), two vessel cord, insertion of amnion-chorion to placental disc

(marginal, circummarglnate and circumvallat~) and abnormal shape of disc

(bilobed ~nd accessory lobe). 5846 singleton placentas with documented cord

twist were analy~ed using contingency tables, comparing the abov~ placental

finthngs among different types of cord twist.

RESULTS:

LEFT RIGHT MIXED ABSENT

Total: N= 5846 4876 (83%) 712 (12%) 112 (2%) 146 (2.5%)

%Normal insertion: N=5096 87.7 85.2 88.4 78.8

%Marginal cord: N= 669 11.1 12.8 8.9 17.8

%Velamantous cord: N = 78 1.2 2.0 2.7 3.4

%Abnormal disc: N = 145 2.2 3.4 6.3 3.4

Absent cord twist is associated with marginal and velamentous cord insertion

when compared with left (P=0.0017), left and right combined (P=0.0025), and

left, right and mixed combined (p =0.0024). Mixed cord twist is associated with

abnormal placental shape (p =0 013 |) Intense spiraling is more frequent in left

(5.8%) than right (3.5%) twist (p=0.0123). None of the cord twist correlated

with circummarginute and circumvallate membranous insertion, true knot and two

vessel cord.

CONCLUSION: (1) Absent cord twist is associated with marginal and

velamentous cord insertion. (2) Mixed cord twist is assecia~:l with abnormal

shape of placenta. (3) The ratio of left to right twist (6.8 to 1) is similar to that

in the literature.

183 ARGININE VASOPRESSIN (AVP) INDUCED CHANGES IN BLOOD FLOW TO THE OVINE CHORION, AMNION AND PLACENTA ACROSS GESTATION. H L Hednana, R.A Brace×, and W M Gilbert Divisions of Perinatal Medicine, UCSD, La Jolla and Maternal-Fetal Medicine, UC Dav=s, Sacramento, CA OBJECTIVE: To determine whether developmental changes =n blood flow rate to the owne fetal chonon and amnion are affected by physiologic Increases ~n fetal plasma AVP concentrahon STUDY DESIGN: Colored microspheres (15 10:1:0 02 u (SD)) were infused ~nto the superior vena cava before and at 30, 60, and 110 m~n of an AVP infusion (3 ng/m=n/kg) =n 9 chronically catheterized fetal sheep between 103 to 141 days gestation (term = 147 days). Chorion, ammon, and placental cotyledons were removed, and microspheres were counted to determ=ne blood flow rates. RESULTS: Fetal artenal pressure (rAP) increased (p<0 0001) and heart rate (FHR) decreased (p<0.0001) during the =nfusion, and these responses were greater =n the older fetuses (_>125 days, ANOVA, p<0 001) From control values, choriomc blood flow rate ~ncreased by 15% at 30 rain of the AVP rufus=on (p = ns), and declined by 20% (60 m=n, p<001) and 35% (110 re=n, p<0001). S=mflarly, placental blood flow rate ~ncreased by 20% at 30 mm of the infusion (p<0 01) and dechned, in parallel with chononic blood flow rates, by 5% (60 ram, p<0.04) and 18% (110 min,p<0 001) from control values Ammotic blood flow rate d~d not change sigmficantly dunng the infus=on Membranous and placental blood flow rate responses to AVP =nfusion did not correlate to advancing gestation CONCLUSION: Choriomc and placental blood flow rates undergo an early =ncrease followed by a steady, gradual decrease induced by phys=olog=c nses of fetal plasma AVP concentration Unhke rAP and FHR, these blood flow changes are not gestahon-dependent, and could play an =mportant role m determining abnormahties of amn~ohc fluid volume chnically observed in some stressed fetuses.

182 A RANDOMIZED CLINICAL TRIAL OF A SINGLE DOSE OF

ZITHROMYCIN IN THE TREATMENT OF CHLAMYDIA AMONG PREGN oA~ WOMEN. Wehbeh,H., Rugg ..... RX, Ah,Y ~,Lopez,GX,

Shahem,S ,Zarou,DX, Lutheran Medical Center, Brooklyn, N Y

OBJECTIVE: To assess the efficacy, the occurrence of side effects and patient

compliance when using a single dose of Zlthromymn in treating C TrachomaUs

refection among pregnant woman

STUDY DESIGN: Patients and their sexual partner were randomized together

into 1 of the following treatment groups’ (1) Both patient and sexual partner

received a single dose of Zlthromycln, (2) Patient given Erythromycin while

sexual partner given a standard course of Tetracychne (500mg taken 4 times a

day for 7 days, n=21) In order to assess the importance of treatment among

sexual partners on the efficacy of Zlthromycin among patients, a third treatment

group was included (patients given 1 dose of Zithromycln (1 gram,n = 10), sexual

partner given the standard course of Tetracycline as described above,n= 10)

RESULTS: Compliance rates among patients given Zlthromycln were 92 6%

vs 71 4% among patients given Erythromycin (p <0 0001), In addition, 57 9%

of pauents given Erythromycln reported suffering side effects vs only 14 8% of

patients gwen Zlthromycln (p=0.002) Among sexual partners 51.8% given

tetracycline reported suffering side effects vs 0.0% given Zlthromycln

(p=0 0003) With respect to cure rates only 4 2% of study subjects given

Zlthromycm showed positive cultures vs 20 0% of study subjects given

Erythromycln or Tetracycline (p=0 28) These findings do however,

demonstrate Znhromycm to be at least as effective as Erythromycln in treatment

of C Trachomatls among pregnant women

CONCLUSIONS: The use of Zithromycln in the treatment of C Trlchomatls

among pregnancy woman substantially reduces the side effects associated with

the current approved methods of treatment and substantially reduces the necessity

of medication change caused by such side effects The results reported herein

lurther show the efficacy of Zithromycln to be equal to or better than currently

approved treatment modahtles and that compliance with treatment significantly

improved among patients given Zlthromycln

184 DETECTION OF ADRENOMEDULLIN, A HYPOTENSIVE

PEPTIDE, IN AMNIOTIC FLUID BY RADIOIMMUNOASSAY

~, T. Moody Ph.D ,x M.J. IValler MD,x K Gray Ph D.,X

M. Gallagher MD, F. Cuttitta Ph.D ,x A Martmez Ph D x Departments

of Obstetrics ~md Gynecology, National Naval Medical Center, Uniformed Services Umversity of Health Sciences, and Biomarkers and

Prevention Research Branch, National Cancer Instatute, National

Institutes of Health, Bethesda, Maryland.

OBJECTIVE: Our purpose was to determine if Adrenomedullm

(AM), a multi-functional regulatory peptide involved m blood flow

regulation and growth stimulation, is expressed m the ammotic fluid

(AF) of second trimester human fetus.

STUDY DESIGN: A prospechve descriptive study design was

utilized. 134 patients undergoing armfiocentests were enrolled

following counseling, ultrasound, and informed consent.

Immunocytochemistry, Reverse Transcriptase Polymerase Chain

Reaction (RT-PCR) and m situ RT-PCR were used to confirm AM expression in fetal amniotic membranes and fetal epidermis Molecular

analysts of eytospms from select AF samples demonstrate AM message

expression To quantify the amount of AM in the AF we developed

and validated a rad~oimmunoassay (RIA) using a rabbtt polyelonal

antibody raised against the carboxy tern’finns of AM.

RESULTS: 134 amniotie fluid specimens were analyzed AM-hke

Immunoreactivity was found in all specmaens. Concentration of AM-

like Immunoreactiwty ranged from 10 fmoF25uL to 300 fmol/25 uL.

CONCLUSIONS: AM is found in second trimester human AF .The

role of AM in pregnancy is not known. However, AM expression has

been found to be decreased in pregnancies complicated by

preeclampsia This RIA may be a useful tool in monitoring AM

fluctuations in normal and hypertensive disorders of pregnancy.

362 SPO Abstracts January 1996 An’, J Obstet Gynecol

185 EFFECT OF EXTERNAL COMPRESSION, TWIST AND

LONGITUDINAL STRETCH ON VENOUS FLOW THROUGH THE

HUMAN UMBILICAL CORD G M Dado x, p B Dobnn x, p G Tom~ch, R Mrl~ckax Dept OB/GYN andDept Surgery, LoyolaUn~versJlyMedmal

Center, Maywood, IlhnoJs

OBJECTIVE: Strong, et al has demonstrated an association between

decreased umbthcal cord coiling and ~ncreased perinatal morNdlty In an effort

to examine thJs, we created an experimental model that measures venous flmv

m urnbthcal cords wtth a normal vs abnormal cothng ~ndex when external

compressive force, twist or Iongttu&nal stretch ~s apphed to the cord

STUDY DESIGN: Ten 18 2 + 0 9 em umbd~cal cord segments were obtained

at the tame of vaginal dehvery The number of coils in each segment was noted

and cords were categorized as normal (6 cords) vs abnormal cords (4 cords)

following the umbthcal cothng index descnbed by Strong The two umNhcal

artenes and smgle umbilical vem were catheterized, pressurized at physmlog~c

pressures of 60 mmHg and 20 mmHg respectively and placed into a warm

sahne bath at 37°C The cord was sublected to stepwlse eompresmon w~th a

5 cm long plate atlached to a Grass force gauge The cords were then

subjected to incremenN 180° clockwise and counterclockwise twists to 720°

Lastly, stepwlse cord lengthemng up to 15% using a non-rotating blade

micrometer was performed The effect of external compression, twJst, and

lengthening on venous flow was measured

RESULTS: Increasing eompresstve force produced a stepw~se decrease m

venous flow, but there was no difference between cords wJth < 0 1 cod/cm vs

those with 0 l-0 3 cods/em (p=NS) Incremental twist produced a reductmn

in flow, espectally at 540° But again, there was no difference between cords

wth < 0 1 co~l/cm vs those w~th 0 1-0 3 colls/cm (p-NS) F~nally, Iong~tuthnal

stretch had a neghg~ble effect on flow for both groups of cords (p=NS)

CONCLUSION: These fin&ngs suggest that differences in morbldlU

associated with umNhcal cord cothng may not be attributed to smaple

mecharacal factors

187 PHOSPHATIDYLSERINE-DEPENDENT ANTIGENS AS MARKERS FOR PLACENTAL PATHOLOGY. H. Awaloeff, T W. Lyden*, C.M. Salafia*, L. Sdberman*, N S Rote*. Pennatal Research Famhty, Depts. of M=crob=ology Immunology, Pathology and OB/GYN, Wnght State University, Dayton OH, Danbury Hosp=tal, Danbury CT, and Georgetown University Medical Center, Washington DC. OBJECTIVE: To appraise the use of phosphahdylserine (PS)- reactive monoclonal ant=bodies (aPLs) to evaluate trophoblast pathology in comphcated pregnancies. STUDY DESIGN: Ten third-trimester placentas from chnically comphcated pregnanmes with Nstolog~c chronic vdlitis and ten normal term placentas were studied. Comphcations of pregnancy ~ncluded diabetes (n=2), IUGR (n=2), fetal distress (n=4), and preterm labor and delivery (n=2) Immunoperox~dase Nstochemistry employed the PS-reachve monoclonal aPLs 3SB9b and BA3B5C4 w=th appropriate positive and negative controls. RESULTS: In normal term placentas, 3SB9b intensely labeled the apical surface of syncyt~otrophoblast, w~th little, =f any, villous cytotrophoblast labeling. BA3BSC4 strongly labeled wllous cytotrophoblast w~th a cytoplasmic pattern. Most of the comphcated pathologic samples showed no major differences in labeling patterns in lesional as compared to non-les~onat areas or as compared to normal placentas from the same gestational age. However, the two placentas from d=abetic mothers labeled d#ferently than normal placentas. Although 3SB9b labeled the syncytmtrophoblast of the d=abet~c placentas, BA3B5C4 labeled both the villous cytotrophoblast and syncyt=otrophoblast CONCLUSIONS: We have prev=ously reported that 3SB9b and BA3B5C4 ~dentify two PS-dependent antNemc determinants found on placental vdlous trophoblast and are expressed sequenhally m relationship to trophoblast=c dffferent~atmn and mtertrophoblastm fusmn. While aPL cross reactwRy w~th trophoblast has been associated with IUGR and chronic vilht=s m aPL syndrome, no difference in PS or CL could be detected in these cases of chronic vilhtis in the absence of aPL syndrome. In d~abetic placentas the co-expression of two different PS- dependent ep=topes on the outer layer of trophoblashc vdh suggests that trophoblast is incompletely d~fferentiated.

186 LIPOPROTEIN(a) IN DISTAL (BASAL PLATE) UTEROPLACENTAL VESSELS: HIGH LEVEL EXPRESSION ASSOCIATED WITH NON-HYPERTENSIVE PREMATURITY. C M. Salaha*, K.A. Starzyk*, M Ossandon*, L Vercruysse, V. Parkash, J. Lage, R Pqnenborg Departments of Pathology & OB!GYN, Georgetown Un~veraty Med=cal Center, WasNngton DC, Yale Un~veraty School of Medicine, New Haven CT, Un~veraty Hospital, Leuven Belgium. OBJECTIVE: Dpoprotem(a) (Lp(a)), a marker of atherosclerotlc vascular damage, is depoated rarely ~n the normal placental bed, commonly ~n the preeclampt~c placental bed, and umformly in ~nvolutmg artenes. We stud=ed Lp(a) depoatlon in uteroplacental artenes of the basal plate ~n preeclampt~c and non-preeclampt=c preterm delivenes STUDY DESIGN: From a consecutive senes of pregnancies dehvered between March and June, 1995, basal plate samples containing uteroplacental artenes were ~dent=hed for 7 cases of preeclamps~a dehvered 28-31 weeks, and 11 non- preeclampt~c cases (dehvenng for Ind=cat~ons of premature rupture of membranes or preterm labor at 24-32 weeks). Uteroplacental artenes In basal plate sections of formahn hxed matenal were stained for Lp(a) (Organon Technika, 1:500. Imrnunoreactlvlty was ~dentlfied as absent or present (+) with oca st pples or dense (++), and the involved artery c=rcumference and Iocahon of Lp(a) pos~t=vtty recorded by computer asasted ~mage analyszs RESULTS: Ten of 10 (100%) of uteroplacental arteries =n preterm preeclamps~a, and 49/56 (88%) of non-preeclampt~c prematunty (p >0 20). N=ne of the ten artenes in preterm preeclampsla had >75% mural (++) The least Lp(a) reacWity was seen m a 24 week cases of premature membrane rupture w~th no Lp(a) (+) artenes (0/3), and the most Lp(a) reactivity m a 27 week case of preterm labor (13/15 Lp(a) (++) artenes). Lp(a) reactivity occurred independently of the presence/extent of physzolog~c change CONCLUSIONS: Lp(a), assoc=ated w~th atheroma format=on and inhibition of hbnnolys=s, ~s depos=ted ~n all basal plate vessels =n preterm preeclampaa and =n some artenes in most cases of non-preeclamptlc prernatunty. These data support placental h~stolog=c and Doppler ewdence that uteroplacental vascular insufficiency underhes non-preeclarnpttc prematunty. We speculate that Lp(a) may mark a cascade of events of vascular damage or premature vascular ~nvolut=on wNch may be assoc=ated w~th e=ther preterm preedampsfa or spontaneous prematunty

188 CHRONIC VILLITIS: A PROCESS THROUGH WHICH UTEROPLACENTAL PATHOLOGY IS RELATED TO INTRAPLACENTAL VASCULAR DAMAGE? CM. Salafia*, M.Y. Divon, V.K. M=mor*. Depadments of Pathology & DB/GYN, Georgetown Univers=ty Med=cal Center, Washington DC, Albert Einste=n Medical Center, Bronx NY. OBJECTIVE: To study relat=onships of chronic vdl=tie to other placental les~ons In rntrauterme growth restnct=on (IUGR) with absent or reversed end-d=astolic flow (AEDF and REDF) on umbilical artery (UA) Doppler velocimetry. STUDY DESIGN: Of 57 consecutive Nrths of singleton infants <10th cent~le for b~rthwe~ght 0UGR) that were admitted to the NICU between 1/89-6/95, 48 (84%) had available placental histopathology. UA Doppler values were obtained by pulsed Doppler ultrasonography classified as normal (S/D<3 0, N=6), borderhne (S/D=3-4.5, N=10), abnormal (S/D>4.5, N=7), AEDF (N=20), or REDF (N=5). Chart review revealed no d=agnoses of congenital viral refection. Placental Nstologic shdes were scored 0-3 (bhnded to chmcal data) for 3 general categories of lesion: (1) uteroplacental vascular pathology (UPV) and related wllous lesions, (2) mtraplacental vascular pathology ([PV), and (3) lesions of chronic inflammation (chronic villit~s (CV), intervlllositis (CI), decidual eosmophils and plasma cells, and dense chronic basal and extraplacental membrane inflammation) Contingency tables considered p<0.05 s~gnificant. RESULTSt CV and/or CI was seen in 21 cases (44%) and CV alone ~n 14 (29%) Twelve of the 21 (57%) also had multlfocal UPV, and 11 had multifocal IPV (52%). Of the 23 cases w=th abnormal UA Doppler but present EDF, 16 (70%) had <1 of the 3 lesion categories present (e.g, UPV, IPV or CV and/or CI) Of the 25 AEDF/REDF, 19 (73%) had >2 lesion categones multlfocally present (p<0.004, OR 10 4). Of AEDF/REDF cases, 11 had CV and/or CI (p >0 20). Ten of the 11 (91%) also had either extensive IPV (n=7) or UPV (n=3). CONCLUSIONS: AEDF or REDF ~s not assoc=ated w=th CV and/or CI in the absence of UPV or IPV. While quantitat~on of IPV =n abnormal UA Doppler has been reported, we show that AEDF/REDF =s also associated w=th les~ons representing more than one pathophys~Noglc process. Severe UPV ~s associated w~th auto-~mrnune d=sorders wNch may cause chronic inflammation. Chromc wll=t~s may destroy placental vessels and lead to IPV In IUGR with abnormal UA Doppler, chronm villitis may reflect pathology related to UPV, and be a possible intermediate step by which UPV causes IPV.

Vohune 174, Number 1, Part 2 SPO Abstracts 363

Am J Obstet (,~necol

189 UMBILICAL ARTERY ANASTOMOSIS: REVISITING AN OLD AXIOM. C M Salaha*, M.Y Divon, JF Mill, C Y Spong*, D M. Sherer, C.B. Jenkins, V.K, Mtmor* Permatal Research Facllfty, Departments of Pathology and OB/GYN, Georgetown University Medical Center, Washington DC, and Albert Einstein College of Medicine, Bronx NY OBJECTIVE To determine the presence and nature of umblhcal arterial anastamoses within 15 cm of the chorlonlc insertion site STUDY DESIGN: Umbthcal cords and chorlonlc plates from uncomplicated term deliveries were dissected free from Wharton’s jelly to identify the structure and course of the umbilical arterles within 15 cm of the chorlonlc insertion Dlssectmn extended to the chorlonlc plate A 0 5 mm blunt probe was used to enter all vessels to determine patency Probln~g wasperformed from the umbilical, arteries towards the Insertion, and /rum air major arterial branches of the chormmc plate back into the ulnbdlcal arteries In selected cases, the absence of anastamoses was confirmed with injection of radio-opaque dye and fluoroscopy RESULTS Ten cases comprised a teaching set, which allowed optimization of the dlssectmn technique Placentas from the next 75 consecutive normal term dehvenes were retrieved for umblhcal arterial dissection Of these, 6 had detached coids and/or hematomas at the umblhcal cord base and were not suitable fur dissection. In 69 cases, complete dissections were performed In 28/69 (41%) either no anastamosls was ldentffmd (N=22), or a wslble anastomosis was trot probe-patent (N=6) In 9/69 (13%), the umblhcal arteries completely fused to form a common channel In 28 (41%) an "H’-type fusion, w~th a single vessel connecting the 2 arteries was seen, and in all cases it occurred within 2 cm of the chorlonlc insertion. Five cases (7%) demonstrated an anastamosis on the chorlonlc plate distal to a first generation umbilical arterial branch point

CONCLUSIONS: The notion that the umbilical arteries anastamose in the vicinity of their chorlonlc insertion has been accepted with little scientific evidence Variance in anastamotlc patterns may explain the variable extent of fetal compromise seen in cases of focal placental lesions such as calcifications, infarct and abruptlon. In addition this data suggests that placental perfusion and lmpedence may not be uniformly distributed between the 2 umbilical arteries

191 LIPOPROTEIN(a) DEPOSITION IN THE UTEROPLACENTAL BED DISTINGUISHES NORMAL AND PREECLAMPTIC PREGNANCIES. ¢ K. Starzyk*, M Ossandon*, L, Vercruysse*, V. Parkash*, J Lage*, R. P~nenborg*, Pennatal Research Facd=ty, Departments of Pathology & OB/GYN, Georgetown Un=versity Medical Center, Washington DC, Umvers=ty Hosp=tal, Leuven, Belg=um, Yale Un=vers~ty School of Medicine, New Haven GT OBJECTIVE: We stud=ed depos=bon cf Dpopratem(a) (Lp(a)), a marker of vascular wall damage and an mh=b=tor of fibnnolys~s and promoter of atheroma development, wltNn spiral adenes of placental bed b=ops=es (PBBs) m normal and protemunc preedampt~c pregnancies. STUDY DESIGN: From an established data set of PBB materials whmb met standard cnteria for charactenzat~on as placental bed samples, 6 PBBs from normal term pregnanc=es uncomplicated by fetal growth restncbon, maternal d~abetee mell=tus, chronic hypertens=on or auto-~mmune d=sease and 4 PBBs from cases of severe proteinunc preeclampsia w=thout underlying maternal medical d=eeases were stud=ed using a monoc]onal anhbody to Lp(a) (Organon-Techn=ka, 1 500) A computer =mage of each vessel lumen was captured and vessel wall area calculated using IP Lab Spectrum software, Vessels were scored as pes~t=ve ff any immunoreact=wty was detected w~thm the area of the vessel wall Contingency tables were used w~th p<0 05 s=gnihcant. RESULTS: A total of 56 spiral vessel lamina were ~denhhed m the normotenslve PBBs, and 66 spiral lumma m the preeclampt=c PBBs Lp(a) pos~tivlty was detected in 2/88 (2%) of normotens=ve sp=ral arteries, compared to 33/66 (50%) of sp=ral vessel arteries in preeclampI~c PSBs (p<0.OOOf, odds ratio 43, pos~twe pre&ct~ve value 97%). In the preeclamptlc PBBs, the percent of Lp(a) pos~tlwty ranged from 38% (8/21) to 71% (12/17). Lp(a) was observed ~n converted and unconverted spiral arteries and in basal artenes ~n 3 cases CONCLUSIONS: Lp(a) deposition in vessels of the normal placental bed is rare, and =s a Nghly select=re and Nghly predmt=ve marker of the placental bed vascular pathology generally assoc=ated w~th protemunc preeclamps=a. Basal arteries, beheved to be inert, may also demonstrate th=s marker of vascular damage. Detechon of Lp(a), wNch =s unrelated to presence of trophoblast or phys=ologm change, may allow d=agnos=s of vascular pathology ~n spec=mens wh=ch do not meet classic placental bed cntena.

190 CHRONIC PLACENTAL LESIONS IN CASES OF PRETERM LABOR DELIVERING AT TERM. CM Salaha*, L. S=lberman*, Permatal Research Facility, Departments of Pathology & OB/GYN, Georgetown Umverslty Medical Center, Wash=ngton DC, Danbury Hospital, Danbury CT OBJECTIVE. To determine ~f preterm labor (PTL) w~th and w~thout cerwcal changes del~venng at term =s associated w~th placental les~ons. STUDY DESIGN’ A commumty hospital obstetnc database =dent~hed 48 pahents w*th s=ngleton gestat=ons admitted for preterm labor w=th ~ntact membranes (PTL) who dehvered at >36 weeks from July 1992 to June 1993. Chart rewew conhrmed that 22 had cervical ddatat~on of _> 2 cm and/or cervical effacement >60%, ol these 21 (95%) had complete gross and hlstologlc placental examinations Of the 26 cases whmh d=d not meet cervical change cntena, 22 (85%) had placental examinations. Forty-three consecutive normal term dehvenes served as controls. Htstologlc diagnoses (bhnded to chmcal data) included villous dysmatunty, chromc inflammahon and intraplacental vascular damage w=th villous stromal erythrocyte fragmentat=on (hemorrhagic endovasculht~s, "HEV") Conbngency tables cons=dered p<0.05 slgmficant RESULTS Admission gestatlonal age (GA) for PTL was similar in PTL with and w~thout cervical change GA at dehvery and placental weNhts were s=mdar ~n the 2 PTL groups and controls Les~ons of chron=c mflammat=on (grade 3-4 wlht~s, dense mtervdlos=t=s, and/or mult~focal uteroplacental vascuhhs) were identified m 10/21 (48%) PTL with cervical change, 2/22 (9%) PTL without cervical changes and 5/43 (12%) controls (p<0 05) The 10 cases of PTL with cerv=cal change had _>2 types of chromc inflammatory les=ons, v 2/43 (5%) controls (p<0 001) Vdlous dysmatunty was seen in 8/21 (40%) PTL w=thout cervical change, 4/22 (20%) PTL with cervmal change, and 3/43 (7%) controls (p<0 05) Other hlstologlc features did not differ among the 3 groups In 2/43 term controls, "HEV" was seen In < 5 terminal wlh "HEV" involving >50 vdh in >3 different wllous samples were seen In 3/21 (14%) PTL w=th cerv=cal change CONCLUSIONS: Placentas of patients who present w=th PTL, cervtcal change, and dehver at term are more hkely to exh=b=t d]ftuse and mulbfocal chromc ~nflammat=on, wNle those admitted to rule out PTL, w=thout cervical change and dehver at term are more hkely to have abnormal placental maturahonal patterns than uncomphcated term dehvenes In th=s small sample, 14% of cases w=th PTL and cerwcal change had w~despread h~stolog~c lesmns of "HEV" wh=ch has been associated with stdlNrth and long-term morb=dlty

192 UTERINE "ALLERGY" IS NOT A CAUSE OF PRETERM BIRTH C M

Salafia*, A Ghldinl, V K Mlnior* Dept of Pathol & Ob Gyn, Perlnatal

Research Faclhty, Georgetown Univ Med Center, Washington, DC, U Corm

Med Ctr, Farmlngton, CT

OBJECTIVE Following the ldeutlfiCatlon of eosmophils in the amnIotlc fired

of women in preterm labor leading to preterm delivery (PTD), it has been

suggested that a maternal "allergy-bke" mechanism may uuderhe a fraction of

PTD cases If this hypothesis were true, the eoslnophds should be maternal xn

origin The aim of this study was to identity the origin of eoslnophxls in cases of

eoslnophll-assoclated PTD

STUDY DESIGN From an established set of non-anomalous singleton Infants

delivered at 22-32 weeks (n-465), we retrieved 161 cases of PTD following

spontaneous onset of preterm labor (PTL), 78 cases with maternal preeclampsla,

33 cases of abmptlon and 193 cases of premature rupture of membranes (PROM)

Maternal and neonatal charts were reviewed HlstoloDc examination of the

placenta, umbilical cord, and membranes was conducted In cases with evident

gradient of eosmophlls toward the amnlotlc cavity, the origin of the eoslnophlls

(fetal vs maternal) was determined by the proximity to fetal or maternal vessels

Cord blood obtained at delivery was analyzed for eosinophd count

RESULTS Hlstologlc evidence of a placental eoslnophlhc gradient towards the

ammotm cavity was present in 19% (90/465) of PTD cases and was significantly

more common In cases of PROM [30% (54/193)] and PTL [21% (34/161)] than

abruptmn [6% (2/33)] and preeclampsla (0/78)] (p<0 0001) In 93% (84/90) of

cases, the eosmophlhc gradmnt was present In the context of multiple hlstologlc

indicators of ascending intrauterine inflammation An eosmophlhc gradient was

present only ~n the fetal (umblhcal cord and chorlon) compartments Ewdence of

~solated eoslnophlha (l e not part of a generic myelold response) was found m

only 4/90 available cord blood at delivery

CONCLUSION An eosmophlhc gradient towards the arrmlotlc cavity is present

in 1/5 of cases of PTD, and ~s mostly associated wlth intrauterine inflammation

The origin of the eosinophlls is exclusively fetal, thereby ruling out a maternal

"allergy-like" mechamsm as the cause of PTD We hypothesize that eosinophlhc

gradient is due to chemoattractlon of fetal eoslnophiIs to the amnlottc tim&


193 EFFECT OF OXYHEMOGLOBIN, A NITRIC OXIDE SCAVENGER, ON THE RESPONSE

OF ISOLATED UMBiLiCAL VESSELS TO SEROTONIN AND SIN-I. G. Saade, R.

Aneedx, M. Belfort, Y. Vedernikovx. Dept 0B/GYN, Baylor College of Medicine, Houston, TX.

OBJECTIVE: To determine the effect of human oxyhemoglobin, a scavenger of nitric oxide

~NO), on seroton~n 15.HT/.induced contractions and SIN-fqnduced relaxation of the human

umbihcal artery and vein.

STUDY DESIGN: Rings of umbilical artery and vein (4 mm in width) were suspended in

Krebs-Hensetemt sdatmn at 370 C in organ baths for isometric tension recording and

bubbled with a gas mixture of 2.5% 02, 8% CO2 and balance nitrogen. Passive tension was

set at 2 g for the artery and 1 g for the vein. Care was taken to preserve the endothelium

After a stable reference contraction to 60 mM KCI was obtained, meclofenamm acid (10s

M) was added to the solution to block cyc~o-axygenase. Concentration-response

experiments were then performed for 5-HT (10 e to 105 M) in the presence or absence of

oxyhemoglobm (3 xlOe M). Similarly, concentration-response experiments were performed

for SIN-1 using a separate set of rinos contracted with 5-HT. The molar concentration

producing 50% of the maximal effect (EC~ for 5-HT and IC~ for SINd) was determined.

The Wilk-Shapiro test was used to check normalcy and the data were analyzed using the

Student t test Statistical significance: P < 0.05.

RESULTS: Incubation with human oxyhemoglobln did not significantly change the

concentration-response curve to 5-HT in either the umbilical artery or vein. Oxyhemoglobln

decreased the relaxation induced by SIN-1 in both artev and vein.

5-HT -log ECso

n SIN-1

-log ICso

Umbdmal Artery

Control 0xyhgb P

7 ,~ 7.3-+0.4 72--+0.2 NS

6 6 6.5_+0.9 6.1_+0.9 002

Un~ibcal ’~em

Control Oxyhgb P

5 5 78-+0.4 7.6_+05 NS

5 5 6.5_+0.5 5.9_+03 0.01

CONCLUSIONS: In isolated umb~l~ca~ vessels, basaIJ released NO did not change the

contractile activity of strong agomsts such as 5-HT. The NO scavenger however, resulted in

a slight decrease in the response to exogenous NO released by SIN-1. A low sensitivity of

the umbihcal vascular smooth muscle to NO may explain these findings.

195 HELLP SYNDROME ALTERS HUMAN PLACENTAL VASCULAR RELAXATION TO pROGESTERONE. Hatim

OmaV, R Ramlrez~, J Arsach~, D GloveP, S Lynchx and Mark

Gabsonx West Vlrglma Umverstty, Morgantown, WV

BACKGROUND We have recently described a dose-

dependent, endothehum-mdependent relaxation to progesterone

(P) m human placental arteries and vems (HPA&V) Tins

receptor-operated, cAMP-me&ated relaxation may be of value

m mamtawang adequate blood flow m the placental circulation

OBJECTIYE to investigate whether Progesterone-me&ated

placental vascular relaxation as altered m HELLP Syndrome

STUDY DESIGN zsolated HPA&V from 6 pregnancaes

comphcated by HELLP syndrome and 10 matched controls,

incubated m Krebs-bacarbonate under 5%02/5%C02/ balance

N2 (Po2 35-38tot;) and submaxlrnally precontracted with U46619, were exposed to cumulatwe doses of P (0 01-30gM)

ANOVA and Student t test startles were uttlmed

REgULT~ HELLP syndrome reduced the reiaxatien to P by

50-100% m both artenes & veins compared to com~ol (for

example relaxation to 10gM P was reduced from 50~7% m to

12.5a-8% m arteries and from 65d:8% to 30±10% m veins, n-=6-

10, P<0 05), wtule responses to other vasoactive ,agents sv_cfl as

serotonm and mtroglyeenn were unchanged

CONCLUSION Progesterone-me&ated relaxation of HPA&V

is markedly reduced m the presence of HELLP syndrome Thas

phenomenon may contribute to the increased placental vascular

resastance and dimtmshed placental perfus~on observed m these

pregnancxes

194 INTRACELLULAR FREE MAGNESIUM ([Mg2+]0 HOMEOSTASIS

IN THE PLACENTA IMPLICATION OF NdMg EXCHANGE PR

Standley~, M lbrahtm~, F Yehanx , CA Standley~ . Depts Internal

Mechcme & Ob/Gyn, Wayne State Umv/Hutzel Hospital, Detroit, MI.

OBJECTIVE: The placenta is revolved tn maternal-fetal magnesium

regulation and may be abnormal m some d~seases of pregnancy.

Therefore, we assessed cytotrophoblast [Mg~+) homeostatic

mechamsms that may be implicated in maternal-fetal ion gradients.

STUDY DESIGN: Human chonocarcmoma ceils (HCC) cultured on

glass covershps were loaded with MAG-Fura 2-AM (SHM x 30 mm) to

spectrofluorometncally assess kmetics of [Mg:÷]i Cells were

superfused with various concentrations ulNa, Mg, Ca and umpramme

(IMP, a putative blocker of erythrocyte Na/Mg exchange) [Mg~+]~

calibration was determined via Triton X-100 and EDTA.

RESULTS: Basal [Mg2+]i was 439+ 44 gM Sequential lowering of

extracellular Na ([Na]o) from 145 mM to 95, 45 and 0 mM caused

progressively larger transient increases m [lVlg2+]i When [Na]o was

lowered to 95 rnM, [Mg2+]l rose from 0 82 + 0 05 to 0 92 _+ 0 08

fluorescence ratto units (p=0 01, N=21, pa~red t-test), peaked in 145 +

24 sec, then returned to baseline after 265 + 21 sec These transients

were completely dependent upon [Mg]o, but independent from [Ca]o

Wh~le acute IMP (10 gM x 20 ram) did not alter basal [Mg~+]L ~

ehrnmatedthereturn-to-basalphase Innormal [Na]o, mcreasmg [Mg]o

from 0 to 1 and 15 mM caused stepw~se increases in [Mg~+]l

CONCLUSIONS: HCC appear to possess a putative Na/Mg

exchanger that funct~ons to maintain [Mg2+]t Additionally, [Mg2+]l is

acutely regulated by [Mg]o Since placental trophoblasts are sites of

maternal-fetal ion exchange, and [Mg]o as well as [Mg2+]a are altered

in preeclampsm, derangements in or modulation of tins exchanger may

contribute to complications such as preeclampsta and preterrn labor

196 TItROMBOMODULIN: A NOVEL MARKER FOR ABRUPTIO

PLACENTA. U. Maerivle#’, D.W. Chan~, J.A. Copol, C.D. Hsu. I~pt.

OB/GYN’, Yale Univ., New Haven, CT and Johns Hopkins Univ. School of

Med., Baltimore, MD.

OBJ-ECTIVE: Thrombomodulin (TM), an endothelial cell marker, is

predominantly located on the syncytiotrophoblast. We ~tudied whether maternal

TM is a clinically useful marker for abruptio placenta.

STUDY DESIGN: A prospective cohort atudy of twenty-five pregnant women

admitted with suspected placental abruptioa. Routine laboratory work included

coagulation profile, Klcihauer-Betke and fetal ultrasound. Women with

pre¢clampsia were excluded. Abruption was confirmed after delivery in eight

cases (Grogp l). Gro~o 2 conalsted of seventeen patients with no clinicat or

pathologic evidence of placental abruption after delivery. Plasma TM was

determined using a two-site enzyme-linked immunosorbent assay. Group~ were

compared by Student t-tests. Data are expressed as mean + SEM.

RESULTS: There was no statistically tfignificant difference in ~g¢, race,

gravidity, parity or gestutional age between the groups. "i’M was significandy

elevated in Group 1 (71..59 :t: 5.35 vs. 48.29 + 3.53 ng/ml, p=0.001). The

sensitivity and specificity of TM a~ a mark¢r for abruption (using a cut-off of

80 ng/ml) were 75 and g8%, respectively. All six of the true positives had

acute abruptio placente¢ on pathological analysis. Ther¢ were two false

negatives, both in patients with chronic abruptions. The sensitivity of an

abnormal coagulationprofile, maternat Kleihaner-Bc~eand ultrasound in Group

1 were 0, 16.7 and 28.6%, respectively.

CONCLUSIONS: TM is a highly sensitive and specific marker for acute

abruptio placenta in patients without evidence of preeclampsia.

Poster Session II


Kohala Ballroom

CATEGORIES Maternal-Fetal Physiology

Medical Complications of Pregnancy


Judges: Thomas J. Garite, MD Eleanor L. Capeless, MD Larry C. Gilstrap, III, MD


Am ~ Obstet G),ne~_ol

197 EARLY ONSET NEONATAL GROUP B STREP (EOGBS) INFECTIONS: DO RISK FACTORS EFFECT LONG TERM

OUTCOME? M Pearlman. R. Fmx, TRB Johnson Dept. Ob/Gyn and

Pediatrics, Unlv of M~ch Med Ctr (UMMC), Aon Arbor, MI. OBJECTIVE: Several risk factors have been ~dentified which increase the

hkelihood of acqms~t~on of EOGBS ~nfection, but httle information ~s

available about whether those risk factors effect early mortality and long

term morbidtty STUDY DESIGN A retrospective observanonal study of all infants born at

UMMC w~th culture proven EOGBS infectmns from February 1972 to February 1994 was conducted. Labor variables known to be assocmted with ~ncreased hkelihood of GBS infection were identified in the mothers prior

sib with GBS, fever in labor (>100.4 F ), prolonged ROM (> 18 hours), and dehvery < 37 weeks Mortahty and long term neurolog~c sequelae was

stratified based on risk factors RESULTS’ A total of 74 infants with EOGBS ~nfect~ons born at UMMC

during this time period with an average follow-up of 49 months (range 18-

168 months) Among these 74 ~nfants, there were 19 deaths (25 7%) Of

the 55 survivors, long term outcome was available for 44 The effect of preterm dehvery on outcome ~s noted ~n Table 1 Table I. Prematurity as a Predictor of Outcome in Early Onset GBS

Infection PRETERM DIED NEURO NORMAL

BEFORE D/C SEQUELAE

Yes 17142 (40 5%) 7/24 (29 2%) 17/24(70.8%)

No 2132 (6.3%) 1120 (5%) 19120 (95%)

p < 0.005 p =0043 p<0 1

Neither fever, prolonged ROM or a prior sib w~th EOGBS, e~ther alone or m combination, was s~gmficantly assomated w~th higher mortahty or long term morNd~ty.

CONCLUSION. Preterm delivery ~s the best predictive factor of both short

term and long term outcome in ~nfants born with EOGBS mfect~ons

199 REFERENCE VALUES AND INFLUENCE OF SMOKING ON MATERNAL MIDDLE CEREBRAL ARTERY BLOOD FLOW VELOCITIES. O. Irion°*x, JM. Moutqum°, K. Wilhams+, JC. Forest°. Dept. Ob-Gyn, °Lavai Univ., +Univ. of BC, Canada; *Univ. of Geneva, Switzerland OBJECTIVE: To determine the values of systolic, diastolic and mean maternal middle cerebral artery flow velocities (MCAV) at 18 and 26 pregnancy weeks and the potential effect of smoking on MCAV. Data available to answer these two questions are derived from small samples of women (<25). STUDY DESIGN: As part of a prospective cohort study carded out in normotensive nulliparas to determine performances of MCAV to predict preeclampsia, women were evaluated in left lateral decubitus, then sitting at 18 (n=305) and at 26 weeks (n=279); 212 were examined at both periods. Left and right MCAV were measured using a 2-MHz pulsed-wave Doppler (Medasonics CDS, CA) and averaged for report. Statistics included pau’ed and unpaired t-tests, and regression analysis. RESULTS. MCAV are reported in the table. Sitting resulted in a statisttcally although not clinically significant decrease in diastolic and mean velocities. Systolic velocities decreased significantly at 26 weeks, but diastolic velocities remained unchanged. Thirty % of women smoked 11 cigarettes per day, on average. Systolic, diastolic and mean MCAV were significantly higher in smokers in both positions at both gestational ages (p<0.01) The amount of cigarettes smoked correlated positively w~th higher MCAV. CONCLUSIONS: normal MCAV values appear to be higher than previously reported on small series of pregnant women. This could be due to the unintentionai measurement of carotid artery end-portion in an earlier study (Obstet Gynecol 1994;84:445-8), or in the larger proportion of smokers m our population (Br J Obstet Gynaecol 1993;100:85-91). Our data confirm an maportant increase of MCAV m smokers, exceeding changes due to posture or gestational age. TABLE: 18 weeks 26 weeks

MCAV cm/s (-I-SD) decubitos sitting decubitus sitting systolic 113 (18) 112 (18) 109 (17) 109 (18) diastolic 47 (8) 43 (7) 46 (8) 42 (8) mean 69 (11) 66 (11) 67 (10) 64 (11)

198 MATERNAL AND FETAL CARDIOVASCULAR EFFECTS OF A NICOTINE-PATCH VERSUS MATERNAL SMOKING. H Hardardottm C Oncken’, VR Lupo*, C Daragjat¢, R Chang*, JS Smeltzer. University of Connect=cut Health Center, Farmington, CT OBJECTIVE: To prospechvely evaluate mcohne levels and the maternal and fetal cardmvascular effects of a transdermal nicotine- patch compared to maternal smoking in a randomized cross-over study. STUDY DESIGN: Patients who were >18 years old, at 24-36 weeks gestabon, with an uncomplicated pregnancy, who smoked ->15 cigarettes/day for the past year were ehg=ble. All pahents had salivary levels of the nicotine metabohte cotmme (SC) measured prior to randomization, to confirm cigarette use. Subjects were eligible for the study ~f SC was >85ng/ml. They were randomized to e~ther wear a 21 mg nicotine patch or smokead Itbitum after 12 hours of abstinence from smoking One week later they were crossed over to the other modahty. Maternal v~tal signs, serum nicotine levels, fetal heart rate (FHR) and resistance ~ndex (RI) of the uterine artery (UA), umblhcal artery (UMB) and middle cerebral artery (MCA) were obtained at 0900 (baseline), 1100, 1200, 1300, 1500 and 1700 hours. Results were analyzed by ANOVA and pa~red t-tests for differences from basehne, w~th p<0.05 considered s~gmficant. RESULTS: Seven patients, of 13 enrolled, have completed the study. Maternal systohc, d=astolic, and mean arterial pressure (MAP) ~ncreased significantly dunng patch days when compared to basehne and to smoking days (p<0 01, n=7). The maternal heart rate, FHR, UA, UMB and MCA R~ d~d not d~ffer between smoking and patch days (n=7). Serum mcotine levels d~d not differ between the groups (p=0.08, n=5). CONCLUSION: These preliminary results show h=gher maternal BP’s dunng transdermal nicotine patch use when compared to smoking, at comparable n~cohne levels. No d~fferences were found in fetal card=ovascular effects when comparing patch use to smoking, as measured by FHR and RI of the UMB and MCA, at comparable nicot=ne levels.

200 ABSTRACT WITHDRAWN AT AUTHOR’S REQUEST

368 SPO Abstracts lanualT 1996 Am J Obstet (;ynecol

201 INTRAHEPATIC CHOLESTASIS OF PREGNANCY IS ASSOCIATED

WITH ABNORMAL ESTROGEN EXCRETION K. K Leshe, L Reznlkov

Dept of Ob/Gyn, Umv of Colorado Health Sciences Center, Denver, CO

OBJECTIVE Intrahepatlc cholestasls of pregnancy (ICP) defines a group of

dlseases which include the milder pruritus gravldamm and the more severe

recurrent jaundice of pregnancy ICP is an extreme example of the subclinical

cholestas~s experienced by most pregnant women during the third trimester The

disease affects as many as 1 in 100 women in the U S , and is associated w~th

relatively high perlnatal morbidity Numerous hnes of ewdence lmphcate estrogen

as a primary causative agent for cholestasls in these conditions The large

estrogen load produced by the fetal/placental unit is metabolized and conjugated

primarily in the maternal liver for final excretion in the urine We explored

potential abnormalities in liver-associated metabohsm and excretion of estrogens

In women with ICP

STUDY DESIGN Urine and plasma samples were collected from 85 normal

women at varying gestatlonal ages throughout pregnancy, and postpartum Plasma

and urine was also collected from women with mild and severe lntrahepatlc

cholestasls of pregnancy from the United States and from Chile (n=23),

representing the largest number of patients with ICP studied to date An array of

stermd hormones was measured xn each sample by gas chromatography mass

spectrometry (GCMS)

RESULTS Urinary excretion of the most active, cholestatlc estrogens was

slgmficantly decreased m the urine of women with ICP compared to normal

controls Plasma levels of estrogen precursors were not different between groups,

indicating that abnormally high estrogen production by the placenta and the fetus

does not occur in ICP However, the low concentrations of estrogens in the urine

of patients with ICP suggests that hepatocytes of affected women may be unable

to adequalely excrete the normal large estrogen load produced by the fetus and the

placenta during pregnancy

CONCLUSION A necessary step prior to urinary hormone steroid excretion is

conjugation in the hver Our working hypothesis is that conjugation of estrogemc

compounds Is impaired In ICP, lack of excretion of actlve estrogens from the liver

may result in a viscous cycle of worsening cholestasls set In play by high

lntracellular hepatocyte estrogen content

203 MATERNAL SERUM INTERLEUKIN-6 LEVELS ARE ELEVATED IN TERM AND PRETERM LABOR. P GreNxl, A Murtha1, C. Jimmerson1, W Herbert1, B. Ro~tman-Johnson2, J. Allen2, Dept Ob/Gyn, Duke University~, Durham, NC and R&D Systems2, Mpls., MN.

OBJECTIVE: Prewous work has found elevated ammotic fluid (AF) Interleukln-6 (IL-6) in patients with term labor and preterm labor (PTL) with intrauterine infection The aim of th~s study was to determine ~f maternal serum IL-6 also ~ncreased with these conditions. STUDY DESIGN: Serum samples were obtained from patients who were 22-34 weeks, not m labor (n=59); term, not in labor (n=62), term, in labor (n=149); ~n PTL at 22-34 weeks who dehvered after failed tocolysis (n=14); and 22-34 weeks in false labor who dehvered at term (n=26). Placentas from patients who delivered preterm were examined for Nstologic chonoamnion~hs, wNch was used to define infection. Maternal serum was measured in pg/ml using a specific ELISA kit (R&D systems). Comparisons of IL-6 levels between patient groups was performed using the Mann-WNtney U test RESULTS: At term, patients in labor had sigmflcantly higher serum IL-6 than without labor ( median- 4.7 vs 2.2, p<0.0001) Preterm patients who failed tocolys~s and dehvered early had s~gnificantly Ngher maternal IL-6 compared to those ~n false PTL who delivered at term (med~an=12.5 vs 1.9, p<0.0001). Patients in false PTL had s~mdar IL-6 levels to preterm patients w~thout labor (median-1.9 vs 1.6, p=.18). Of patients dehvenng preterm who had placentas available for study, 89% (8/9) had Nstologic chonoamniomt~s. A value of >7 pg/ml gave a sensitivity of 100%, specificity of 97%, PPV of 91% and NPV of 86% for infechon and tocolyhc failure CONCLUSION: Like AF IL-6, maternal serum IL-6 ~s elevated during term and preterm labor and may have a physiologic role in partunhon. Elevated serum IL-6 is a very sensitive and specific marker for patients in PTL who are infected and fail tocolys~s

202 CIRCULATING MATERNAL SERUM GNRH AND CRH IN NORMAL AND ABNORMAL PREGNANCIES. K A Sorem. C B Smlkle*, D K Spencer’, C Lunt~, E Yoder’, D. Braget~, J. Ramos’, M Gravson~, T M Sller- Khodr’ Departments of OB/GYN, University of Texas Health Smence Center, San Antonio, and Wllford Hall Medical Center, Lackland AFB, San Antonio, TX OBJECTIVE: GnRH and CRH are produced by the placenta and fetal tissues and have been measured in maternal circulation during pregnancy, ho~vever, a longitudinal study of GnRH and CRH in normal pregnancy and in early pregnancy loss has not been reported Our objective was to determine the normal values of these hormones throughout the pregnancy and to determine whether abnormal levels or ratios were predictive of abnormal pregnancies STUDY DESIGN: Fifty-one pregnancies were follmved prospectively, with levels of GnRH and CRH measured at 8,10,12,14,16,28,36 weeks gestation and during labor Specific and sensitive RIAs were used to determine the levels of hormones in the samples RESULTS: Thirty-three pregnancies, with samples at multiple time points and outcome data available, were completed to term without complication In the normal group, CRH increased from low or undetectable levels at 8 weeks to 32 5 + 2 3 pg/ml at 16 weeks Thereafter, there was a significant increase In inatemal CRH to 1609 -- 113 pg/ml In labor GnRH levels demonstrated a blmodal distribution increasing from 8 to 14 weeks, decreasing at 16 weeks, and lncreasxng from 28 weeks to term The ratio of CILH to GnRH in the normal group was 0 198 at 8 weeks, increased s~gnlficantly at 16 weeks, and peaked at 5 847 m labor In e~ght cases of early pregnancy loss, GnRH and CRH levels and ratios were normal at 8 weeks In two cases of premature dehvery and one case of severe preeclamps~a, these GnRH levels and ratios were within the normal range, CRH was rather normal (n-2) or elevated (n=l) CONCLUSIONS: Mammal levels of CRH and GnP, H m normal pregnancies and in labor at term were defined In thin group, neither concentrations of GnRH and CRH nor the ratio of CRH to GnRH were useful m predicting early pregnancy loss

204 THE EFFECT OF MATERNAL SERUM ON BONE MARROW HEMATOPOIESIS. T Nesb~ttx, H Kay, J Kurtzbergx Depts Ob/Gyn and Pediatncs, Duke University Medical Center, Durham, NC. OBJECTIVE: Pregnancy is associated with known hematologic alterations, but regulation of these changes are not well understood. Our objective was to determine if humoral factors in pregnant women directly affect hematopoies~s in human bone marrow progemtor cells STUDY DESIGN: Human Progenitor Cell Assays (HPCA) used m this study allow analys=s of hematopoies~s at an earher and more accurate level than that of standard morphological and hlstochemical techmques HPCA in tnplicate were performed on normal human bone marrow (n=6) to determine the hematopoiet~c response ~n the presence of 10% by volume maternal sere (n=40) for each bone marrow sample or 10% by volume growth media (controls) Colony Forming Units- Granulocyte/Macrophage(CFU-GM),Granulocyte/Erythroid/Monocyte/ Megakaryocyte (CFU-GEMM) and Burst Forming Units-Erythroid (BFU- E) were scored after a two week ~ncubation Data were analyzed w~th respect to trimester of pregnancy and the presence or absence of labor Continuous variables were compared by paired t-test, and categorical vanables were compared by F~scher’s exact test RESU LTS: Sera from non-labonng pregnant women had no effect on granulocyte or macrophage growth at any stage of gestahon A s=gmflcant shmulatory effect m erythro~d precursors was observed w~th hrst trimester sera (p=< 0001), tNs effect was not seen in later gestation Both myelold and erthyro~d HPC growth were profoundly ~nNblted by sera from labonng women, no erythrocyte, macrophage, granulocyte or lymphocyte growth was demonstrated m any culture (p=< 0001 ). CONCLUSIONS: Circulating factors present in maternal sere have marked effects on hematoporems First tnmester sera stimulated erthropoletm act=v~ty of human progemtor cells, whde sera from labonng patients, regardless of gestahonal age, demonstrated profound ~nh=bltory effects on both myeloid and erythro~d precursors Phys=olog~c hormonal changes, either direct or mediated through cytok~ne activity, are hkely to be responsible for these observations

Volume 174, Numbe~ 1, Part 2 SPO Abstracts 369 Am ] Obstet Gynecol

205 FETAL SERUM CYTOKINE LEVELS ARE CORRELATED WITH PRESENCE OR SEVERITY OF HISTOLOGIC ACUTE INTRAUTERINE INFLAMMATION, ¢ M Salah~*, J M Lage*, S Lenck¢, G.S Eghnton, V Parkash* Permatal Research Facility, Departments of Pathology & OB/GYN, Georgetown Unw. Medical Center, Washington, DC, Yale Umvers=ty School of Med=c=ne, New Haven, CT. OBJECTIVE: To study relahonsNps of Nsto!og~c acute mflammat=on to maternal and fetal serum levels of =nterleukms (IL)I-& 6 and IL 2-receptor (IL-2 R) STUDY DESIGN: In 1992, 32 consecuWe pahents at 20-36 weeks w~th achve progressive labor and failure of tocolys~s were recruited Maternal serum sampled dunng the acWe phase of labor, and fetal serum from the umbd=cal vein at birth were assayed by ELISA for levels ([pg/ml or IU/ml]), expressed as mean _+ S.E) of soluble IL-2 R, IL-6, and IL-113 T-Cell D~agnoshcs bhnded to chnlca data. Acute iaf ammat~on ot maternal or gin (e g., amnlon, A~} and fetal ongln (umbilical vascuhhs, UV) were scored by 2 Independent groups bhnded to chmcal data on a 0-4 scale Nonparametnc tests, conhngency tables, ANOVA and regression corrected for multiple comparisons w~th p<0.05 as stgnthcant RESULTS: Fetal [IL-113] were elevated =n the 14 cases w=th grade 3-4 UV compared to the 5 cases w~th no UV (114_+24pg/ml v 19.2+ 9 pg/mi, p=0.02) Fetal serum ILL-2 R] were elevated ~n the 17 w~th grade 3-4 AI 1203+166 U/ml versus e~ther the 3 with grade 1-2 A} 463_+101 U/ml) or the 7 cases-without Ai (685 _+104U/ml, p=0.02 Fetal [IL-2 R] were elevated =n the 16 w~th grade 3-4 UV (1247_+172 U/ml) vs the 7 w~th grade 1-2 and the 5 w=th no UV (686_+111 U/ml, and 590_+25 U/ml, respectively, p<0.002) Elevated total [IL-6] were correlated w=th Ngher maternal antepartum temperature (p=0 01), and a trend to increased rate of maternal tachycard~a (p=0.06) Increased maternal ILL-2 R were assoc=ated w~th increased incidence of maternal tachycard~a p=0 01. Chn~ca d=agnos~s of chonoammon~t~s was not related to presence/seventy of any histolog=c markers of acute =nflammat=on. Maternal cytokme levels were not associated with measures of Nstologlc Inflammation =n maternal or fetal tissues CONCLUSIONS: The lack of correlabon of maternal serum cytok~ne levels ~n preterm labor with any h=stolog=c markers of maternal or fetal =nflammat~on may explain the generally poor correlation between the chmcal diagnosis of chor~oammomt~s and Nstolog~c ~nflammat~on. If fetal cytokme levels are correlated wdh maternal chn=cal symptoms, fetal cytok=ne levels may be more stable and reflect=ve of the =nflammatory process than maternal levels.

207 MPOPROTEIN (a) IN SPIRAL ARTERIES AT TERM AND AFTER PLACENTAL DELIVERY: CAN INVOLUTION-RELATED PROCESSES BEGIN BEFORE PARTURITION? C.M. Salafia*, K. Starzyk*, J Lage*, M Ossandon*, L. Vercruysse*, V Parkash*, C Y Spong, R. P~lnenborg. Departments of Pathology & O8/GYN, Georgetown Unwerslty Me�cat Center, Washington DC, Yale University School of Medicine, New Haven CT, Unwerslty Hospital, Leuvan Belgium OBJECTIVE: Dpoproteln(a) (Lp(a)), a marker of vascular damage, is deposited rarely in the normal placental bed and commonly in the preeclampt~c placental bed, but basal plate and ~nvolut~ng artenes have not been stud~ed We stud=ed Lp(a) ~n basal plate uteroplacental arteees ~n term normal and preeclamphc placentas and ~n uteroplacental vessels w~th~n 5 months after placental dehvery STUDY DESIGN: From a consecut=ve senes of pregnancies dehvered =n March and June, 1995, basal plate sp=ral artenes were ~dent~hed for 13 cases of uncomphcated term Nrth of appropnately grown infants, 6 preeclamps~a cases delivered 36-40 weeks, and 4 cases of involutional ~mplantahon s=tes (2 removed by curettage at 2 weeks and 2 months after dehvery, 2 by hysterectomy at 3 and 5 months post partum) Spiral artenee ~danf~fied on hemaloxylm and eosm preparahons of formahn fixed matenal were stained for Lp(a) {Organon Technlka, 1.500). Immunoreactlvlty was identlhed as present (+) or dense (++) Image analysis recorded c=rcumference w=th reactmty and Its location (endothehal, =nt=mal, mural) was recorded. RESULTS: Of thirty-five artenes ~n the 13 uncomplicated term dehvenes, 14 (40%) had at least one a~tery with Lp(a) reactlwty (max=mum 5/7 (+) artenes), compared to 20/21 (95%) of artenes m term preedampsla, and 25/26 (96%) of involuting artenes (p<0 0001). No features of ~mmunoreact=v=ty were correlated w=th presence/extent of physiologic changes, or w~th term normal v. term preeclamps~a CONCLUSIONS: Lp(a), associated w=th atheroma formation and =nh~b=t=on of fibnnolys~s, ~s present =n basal plate artenes ~n many normal term Wths, and in 95% of basal plate artenes ~n term preeclamps~a and =nvolutmg artenes Processes cntlcal to normal utenne vascular involut=on may play a role in term preeclamps=a. Our data suggest that ~nvoluhon-related changes may develop before partunt~on even ~n normal term dehvenes and pers=st for months post partum This process may explan decreased late th=rd tnmester fetal growth, trophoblast ischem=a proposed as lhe cause of preec)amps~a, and the poor prognos~s associated w~th subseuqent pregnanc=es w=thln one year of partunt=on.

206 ~ UANTITATIVE VASCULAR CHANGES OF THE TEROPLACENTAL ANATOMY IN PLACENTAL BED BIOPSY

DEFINE PREECLAMPSIA K Starzyk*, ~, L Vemmysse*, J.M Lage*, J C Pezzullo*, R. P1jnenborg* Departments of Pathology and OB/GYN, Georgetown Umverslty Medical Center, Washmgton DC, University Hospltal, Leuven Belgium. OBJECTIVE: To determine quantitative markers to dlstlngmsh normal and nreeclamnUc (PE) uteroplacental vasculature STUDY DESIGN: From an estabhshed data set of placental bed biopsies (PBBs) 6 samples obtmned from uncomphcated ferm dehvenes and 4 samples from severe protelnurlc PE were selected Mvometrlal vessels were identified and characterized as having complete physlolog~c chaeges (cPC), partial physiologic changes (pPC)’, no physlolo~:lc change (nPC), andas splra’l or basarartenes lSv a single observer (RP) An Image of each vessel was analyzed (1P Lab Spectrum) to calculate mean walrthlckness a lumen area an effective diameter (the diameter of a circle of the same perimeter as the measured vessel), and a relative wall- to-lumen ratio. Elastic stains (Sigma Chemicals) were performed on matched tissue sections Elastic was assessed as intact, lragmented or absent. Contingency tables and ANOVA considered p<0.05 sl~Znfflcant RESULTS: 82 arteries were ldennfled in nomal P’BBS anal66 in PE PBBs 13 (16%) of the 82 normal PBB arteries showed cPC and 15(18%) had oPC, compared to 17/66 (26%) PE PBB arteries with cPCand 8/66 (12q~) PE PBI~ arteries w~th pPC (p=0.19) None of normal PPB arteries had fibrmoad necrosis or atherosls, versus 3/25 (12%) of PE PBB arteries (p=0 22). Lumen areas and effective dmmeters of all types of vessels were smaller m PE as compared to normal PBB (each p<g 05). Mean wall thicknesses of PE PBB arteries with either cPC or pPC(0 2_+ 0.09ram and 0 10_+ 0 08ram) were thinner than normal PBB arferms w~th cPC or pPC (0 3_+0,12ram and 0 17-+007ram p<0.05) In PE_ the wall-to-lumen ratio was also reduced compared’to normals (p<0 05) Of normal nPC spiral arteries 5 (38%) had ~ntact elast*c, whale no PE nPC smral arteries bad intact elastic (p=0 02). Elasuc was present ~n 22/35 (60%) normal basal arteries versus 2/24 (8%) of PE basarartenes CONCLUSIONS: We have tbund extensive quantitative differences between normal and PE PBBs (1) Whde s~mdar numbers of cPC and pPC PBB arteries are seen, their wall thickness is significantly thinner in PE compared to normal (2) PE spiral and basal arfenes are also smaller m cahber with relatively thinner walls (3) Elastic is more disrupted m nPC PBB arteries of PE compared to normms, despite the lack of’trophoblast Several hypotheses may be proposed. (I) An antomical variant of spiral and basal arterial anatomy with smaller cahber vessels may explain some cases of familial PE (2) Aspects of uteroplacental conversion (such as loss of elastic) may be independent of trophoblast (3) Abnormally thin walled sl~lral arteries may indicate a more fragile vasculature more susceptible to myometrlal and other hemodynamlc forces during gestation

208 COMPONENTS OF THE BIOPHYSICAL PROFILE SCORE IN THE PREDICTION OF HISTOLOGICAL ACUTE ASCENDING INFECTION IN PREMATURE RUPTURE OF THE MEMBRANES (PROM) DELIVERED AT <32 WEEKS GESTATION. _.C,M Salafm*, V K Mlmor*, J.C. Pezzullo*, A Ghldlm, D M. Sherer, L.M Ernst*. Perlnatal Research Facility, Deptartments of Pathology & OB/GYN, Georgetown Umv Medical Center, Washington DC, U-CONN Medical Center, Farmlngton CT. OBJECTIVE: To assess the performance of components of the biophysical profde score (BPP) performed within 24 hours of dehvery m the prediction of severe hlstologlc acute inflammation in premature rupture of the membranes (PROM) dehvered <32 weeksgestatlon. STUDY DESIGN: An estabhshed consecutwe series of non-anomalous hveborn singleton births at <32 weeks gestation contained 193 cases of PROM, of which 166 (86%) had a BPP within 24 hours of birth. Hastologac acute ~nflammahon (AI) an ammon (e.g maternal nflammatmn), and in umblhcal and chorrumc vessels (e g fetal

inflammation) was scored by a single pathologist on a numerical scale blinded to chmcal data "Severe" AI (SAI) was scored 3-4 on a scale of 0- 4 The diagnostic indxces of each component of the BPP (ammotlc fluid volume [A’FV<lcm], fetal breathing movements [FB<30sec/30man], fetal movements [FM <_2/30mini, and non-stress test [NST nonreacUve]) were calculated RESULTS: Prevalences of severe maternal AI, any fetal AI, and severe fetal AI were 18% (30/166), 33% (80/166), and 28% (471166), respectively. Data are presented as values with 5-95% confidence intervals

SAI, fetal

AFV Seas 65% {57-74)

AFV Spec 34% {27-40)

FB Seas 89% {65-7S)

FB Spec 24% (15-34)

FM Sons 86% (43-99)

FM Spec 29% (27-29) 69% (63-76) 24% [17-33)

SAI, maternal

81"/o (73-88) 17% (12-20)

81% (77-86)

17% (10-26) 86% (43-99) 18%,{16-19)

NST Seas 78~{o (74-84) NSTSI3ec 14’/o 8-21)

48% (38-58}

45% (38-52)

51% (45-57)

46% (35-67) 57%

48% {47.50}

52% {45-59)

47% (38-56) CONCLUSIONS: The sensitivities of BPP components are h~gher for maternal than fetal hlstologac mflammatmn, even though fetal A1 and SAI are more prevalent than maternal SAI in very preterm PROM Our data suggest that altered fetal behavior ~s more often a manifestation of maternal inflammation in very preterm PROM

370 SPO Abstracts JanuaD, 1996 Am J Obstet Gynecol

209 INTERLEUKIN-6: PLACENTAL IMMUNOREACTIVffY IS PRIMARILY LEUKOCYTE-ASSOCIATED IN ACUTE INFLAMMATION, BUT IS SYNTHESIZED BY CYTOTROPHOBLAST INDEPENDENT OF ACUTE INFLAMMATION ~ M Salaha*, J F M~II*, K.A. Starzyk*, M Ossandon*, D Spacer*, C. Lester*. Pennatal Research Facility, Departments of Pathology & OB/GYN, Georgetown University Medical Center, WasNngton DC. OBJECTIVE: To study Interleukm-6 (IL-6) synthesis and ~mmunoreactlvdy in placenta and their vanance with gestatlonal age (GA), labor, and hlstologlc evidence of acute ascending infection (AI) STUDY DESIGN: E~ghteen cases delwenng 37-40 weeks GA and 8 cases dehvenng singleton, nonanomalous ~nfants between 26-33 weeks GA were selected 10 at term and 3 preterm w~th spontaneous labor, 7 at term w~th tnduced labor, 3 preterm w~th spontaneous but augmented labor, and 1 term and 2 preterm w~thout labor for preeclamps~a D~abetic patients were excluded H~stolog~c AI was d~agnosed blinded to chn~cal data With antibody to IL-6 (B~osource Int’l, 1 100), ImmunoNstochemtcal (IHC) reactlwty in formahn-hxed extraplacental membranes, chononlc and basal plate, and chononlc vlll~ was graded as weak (+) or intense (++) In cases with AI, maternal and fetal leukocytes (WBCs) were also assessed. In s~tu hybndlzatlon (ISH) was performed on senal sections utlhzlng d~goxlgenm-labelled nboprobes from a human IL-6 cDNA, and s~m~lar mtenstty scale was used to score ISH staining. RESULTS: In all cases, apical aspects of ammon eplthehum were focally IHC (++). By ISH, amnlon was uniformly (++) and chonodecldua (+) at all GA, but decreased to 0/(+) in AI cases at all GA Subchononm and basal cytotrophoblast (CT) were IHC (+) reactive in an annular pattern dehmltmg cell membranes. ISH conhrmed presence of IL-6 message m these ceils. In AI cases at all GA, maternal WBCs in amnion, chonodec~dua, chonon~c plate and ~nterv~llous space, and fetal WBCs ~n chononlc vasculitls were uniformly IHC (++), but were nonreactive on ISH. ISH and IHC reactlvitles were independent of labor or labor type CONCLUSIONS: IL-6 synthesis (by ISH) and Iocahzabon (by IHC) in amnlon (and to a lesser extent in chonodec~dua) is down-regulated by AI. Non-villous (subchononm and basal plate) CT showed low levels of IL-6 message at all GA and no vanance w~th AI or labor In non-N, IL-6 assayed ~n amn~ot~c fluid may be placentally denved, but ~n AI IL-6 ~s hkely to be denved from WBCs Negative regulation of extraplacental membrane IL-6 expression may prowde a fetal defense against potential deletenous effects of excess IL-6.

211 INCREASED INTERLEUKIN-1-ALPHA (IL-I-~) IMMUNOREACTIVffY IN BASAL CYTOTROPHOBLAST RELATED TO HISTOLOGIC ACUTE INTRAAMNIOTIC INFECTION. C Salafia*, J M~II*, K. Starzyk*, M. Ossandon*, D. Splcer*, C Lester*. Pennatal Research Fac~hty, Departments of Pathology & OB/GYN, Georgetown University Medical Center, Washington DC OBJECTIVE: To examine ~f IL-l-et immunoreact~wty in placenta ~s gestat~onal age related and affected by histolog~c evidence of acute ascending infection. STUDY DESIGN: &xteen non-anomalous singleton, nondlabetlc cases were selected, including 8 delivered 37-40 weeks, and 8 dehvered between 26-33 weeks for pnnc~pal ~nd~cat~on of preterm labor/premature rupture of membranes. Four of each group had moderate hlstologic acute marg~nabng chonodec~du~t~s. Using anhbody to IL-I-o~ (Sigma Chemicals, 1’100), relative ~mmunoreact~wty of frozen tissue samples was graded as weak (+), moderate (++), and intense (+++). Cells assessed ~ncluded syncytlotrophoblast (ST), cytotrophoblast (CT, of basal plate, placental septae, subchononlc fibnn, and ~n perlvlllous fibrin) vascular smooth muscle, endothelial and stromal cells, and edematous and non-edematous w[h. RESULTS’. In preterm cases without acute inflammation, approximately 25% of basal plate and chodon~c plate CT were (+). Most endothehal cells and septal CT were (+++). Less than 50% of villous stromal cells, and scattered stromal and vascular smooth muscle cells in the chononm plate and fetal vessels were (++). Preterm cases w~th acute inflammation showed >75% basal CT (+++), and increased numbers of subchorion~c CT and chonomc plate cells (+++). In term non- ~nflamed cases, >75% of basal plate, septal and ~nterwllous CT were (+++), most endotheha and villous stroma and vascular smooth muscle cells were (++); but few subchononm CT were (++). In ~nflamed term cases, no consistent patterns of reactivity were seen. In basal plate, IL-I-~ react~wty was also seen m extracellular matnx Independent of gestabonal age or inflammation IL-l-o~ was not d~fferent ~n edematous wll~ compared to non-edematous villi. CONCLUSIONS: IL-l-ct reactivity in preterm placental endothellum ~s intense and may quantitatively decrease by term CT (and decidual extracellular matnx) ~s a major locus of IL-l-ot both preterm and at term. Intense IL-I-o~ react~wty was seen m preform inflamed cases, and in all term cases ~ndependent of acute inflammabon. IL-l-cq the pnmary species contaned ~n trophoblast~c cells, has been suggested to modulate placental development. IL-I-~ expression at the maternal/placental interface may contnbute to normal partunt~on, and may be a component of preterm inflammation-related partunt~on

210 INTERLEUKIN-6 RECEPTOR SYNTHESIS: DOWN-REGULATED BY ACUTE INFLAMMATION IN AMNION, CHORION, DECIDUA AND VILLI AND IN SPONTANEOUS LABOR IN THE BASAL PLATE. C M Salafla*, J F M~!I*, KA Starzyk*, M. Ossandon*, D Spacer*, C Lester*. Permatal Research Facility, Departments of Pathology & OB/GYN, Georgetown Umvers~ty Medical Center, Washington DC. OBJECTIVE’. To examine ~f placental mterleuk~n-6 receptor (IL-6R) synthesis and ~mmunoreact~wty ~s related to gestat~onal age at b~rth (GA), labor (spontaneous/no augmentation [SL], augmented [ALl, induced [IndL] w~th vaginal delivery) and acute ascend~nq ~nfect]on. STUDY DESIGN: 18 cases dehvenng 37-40 weeks GA and 8 cases dehvenn# s~nglefon nonanomalous ~nfants between 26-33 weeks GA were selected. 10 a’t term and 3 preterm w~th SL, 7 at term w~th IndL, 3 preterm w~th SL+AL and 1 term and 2 preterm w~thout labor for preeclamps~a. D~abet~c patients were excluded. Hlstolog~c acute inflammation (AI) was dlaqnosed (0-4 scale) With antibody to IL-6R (BioSource, 1 10% immunoNstohcemlcar(IHC) reachvlty ~n frozen tissue samples was graded as 0 weak (+) or ~ntense (++) In s~tu hybridlzatlon(ISH) was performed on adlacent t~ssue sections fixed ~n formalin ut~l~zinq d]qox~enin-labelled nboprobes from a human IL-6R cDNA, and scored on a s~mlFar mt’ens~y scale RESULTS: ISH demonstrated complex IL6-R patterns of expression in amn~on and ~ts sub acent stroma, chonon stroma, scattered chonon cytgtrophoNast CT) and decldual stromal and leukocytm cells. TNs pattern was abohshed to 0-focally(+)] by AI af alt GA ISH also demonstrated IL-6 R mRNA uniformly m syncytlotrophoblast ST) fetal endothella, many wllous stromal and vascular smooth muscle cells, and basal CT at all GA. In AI, endothe lal and smooth musc e staining was absent, subchononlc CT staining was increased and ST showed large areas of negative stalnln# associated with v~lous edema Basal CT ISH reactlw[y was ~ndependent of AI. Non-wllous CT were the only IHC (+) cells ~n all cases Subchononlc CT reactivity increased in AI, and did not vary wdh labor. In term SL and IndL with no AI, basal CT was 0- focally (+) by IHC, but was >50% (++) m AI, both m term without labor, preterm SL, and In preterm no labor without AI. Preterm SL+AL had negative <25%(++ basal CT In all cases, IL-6 R IHC react~wty was identified in extracellular sites. CONCLUSIONS’ IL-6 R expression m extraplacental membranes is complex and modulated by AI at all GA. In basal plate, CT IL-6 R may vary w~th labor type and GA. A phys~ctog~c effect of labor augmentation ~n preterm b~rth ~s modulation of IL-6 R function Many cells intensely ISH reactive (e g, ST, villous stroma and endothehal ceils) are IHC non-reactive. TNs comNned w~th IHC suggests that CT- associated IL-6 R may be soluble m the CT enwronment, and may be produced by CT or by another cell-type ~n the ~ntrautenne m~l~eu

212 NEONATAL NUCLEATED ERYTHROCYTES IN PRETERM PREECLAMPSIA MARK FETAL ACIDOSIS AND PLACENTAL ISCHEMIA. C M. Salafia*, V,K. Mimor*, J.C Pezzullo*, L.M. Ernst*, A. GNdm~, D M Sherer. Pennatal Research Fac~hty, Departments of Pathology & OB/GYN, Georgetown University Medmal Center, Washington DC UCONN Medical Center, Farmington CT. OBJECTIVE:To determine relationships among neonatal nRBCs, assessment of fetal well-being, umbihcalarterial and venous (UA, UV) blood gases and placental lesions ~n preterm preeclampsia (PE). STUDY DESIGN: From an established data set of non-anomalous s ng etch rebirths delivered 22-32 weeks gestational age (GA) w~thout cases of maternal d~abetes mellitus or chronic hypertension, there were 78 cases of PE and 70 (89°/o) had a complete blood count by 3 hours of hfe. The b~ophy~’~cal profile (BPP) co, mponent scor,es w~thin 24 hours,of birth, feta heart rate (FHR) abnormalities neonata~ anthropometnc oata and p acenta es on scores were collected. Placental les~ons in cate~lories of (1) uteroplacental vascular and related villous lesions mclud!ng h~stologic abrupt~on 2) chronic inflammatory lesions, and (3) coag,u~ahon related lesions as we as villous edema, were scored on a 0-4 sca~e, nRBC/dl (= WBCo nRBC/100WBC) normalized by log transformation, was analyzed by ANOVA and regression with p<0.05 s~gn~ficant. RESULTS’ Mean nRBCs was 5.4/dr(range 0.03-57.7). nRBCs/dl was ’ n d~’nt of GA 07 mean GA 29+2 wk ran e 22 8-32 wk The }~gPw~en~ were asso~Pa~e~l ~vlth I"nRBCs/~. ,l, toi’al B~P (p~0.04), ~’feta[

movement s.core on 8PP (.p=0,.005), greater ~nc~denc.e£,f ~FHR vanability (p=0.027), SUA and UVpH[p=0.02, p=0~0002)and 4,uA and UV base excess (p=0..008, p=0.006). Each 10-fold T~n nRBC/dl is associated with an average 4. m UV pH of 0.02. Maternal maximum blood pressure and degree ofproteinuna and infant we~g.ht or length cenNes were not related to nRBCsldl. P acental lesions of Tcytotropnoblast (X-cell) proliferation (p=0 04), abruption rela!ed histology(~=0.015), and severe villous edema {~=0 0002) were also related to 4. nRBCs/dl. CONGLUS ONS: In preterm PE, nRBCs/dl are correlated w~th 4, fetal well-be ng, aboratory evidence of fetal ac~,dosis chroni,c lcytotrophoblast proliferation) and more acute (abruption-re~ated lesionsI placental lesions ref ect ng abnorma uteroplacental perfus~,on. T,hls ,sugges!s that in preterm PE, nRBCs/d can be used as a marKer o~ aonorma~ uteroplacental perfus on surf c~ent to cause fetal t~ssue hypoxia. The association of severe wllous edema with nRBCs/dl suggests ~n certain cases vl ous edema may be a marker of fetal hypoxla.

Voltmte 174, Number l, Pall 2 SPO Abstracts 371 Am J ObsteI Gyneco

213 KINETICS OF NUCLEATED ERYTHROCYTE NUMBER (nRBCs) IN NEONATAL CIRCULATION IN THE FIRST DAY OF LIFE ARE RELATED TO UNDERLYING PATHOPHYSIOLOGY OF nRBC RELEASE IN PRETERM BIRTH, C M Sa!afla*, J.C. Pezzullo*, C. Spend, A. Gh dtnl Per natal Research Facthty Departments of Pathology & OB/GYN, Georgetown Umverslty Medical Center, Washington DC. OBJECTIVE: nRBCs have been proposed to mark fetal hypoxia Inflammatory cytokmes are related to release of mature and immature myelotd cells as well as nRBCs from hematopmettc sites Are processes of antenatal nRBC release distinguished by kmetlCS of nRBCnumber m the neonatal circulation9 STUDY DESIGN. From a data base of consecutive non-anomalous singleton llveblrths at 22-32 weeks tn 1989-94 (excludmg maternal diabetes meliitus, chrome hypertension and non-hy~pertenslve abruptton), 125 cases had a complete blood count and nRBC count obtamed within 3 hours of life, and a 2nd complete blood count and nRBC count within 24 hours after the first value [92(74%) premature membrane rupture {PROM} and preterm labor/Intact membranes {PTL}, 33 (26%) preeclampsla {PE}] Neonatal data regarding volume expansion and transfusion was recorded. Lesions in 4 categories (lesions el utero-placental vascular pathology, chrome mflammatlon, coagulation, and placental vase-occlusion) were scored and summed to give totalpathologlc burden nRBCs/dl was calculated as WBC. nRBC/100WBC, and a relative rate of decrease in nRBC/dl over the fime ~nterval of the 2 complete blood counts calculated (AnRBC). ANOVA and regression analyses corrected for multiple comparisons (p<0 05). RESULTS: No one placental lesion was related-to AnRBC m either PROM/PTL or PE In PROM/PTL increasing; total burden of uteroplacental vase-occlusion was inversely related to AnRBC (p=0.009) Increasing total rnyelold number in Initial complete blood count was positively correlated with AnRBC (p=0 007) Both factors remained independently significantly related to AnRBC m a multiple regression that included ueonatal volume therapy (overalI p=0.003). They were not reIated to AnRBC in PE (p>0.50) Other placental lesions or categories were not related to AnRBC in PROM/PTL or PE CONCLUSIONS: Uteroplacental vascular lesions (shown to impair fetal oxygenation) would be hypothesized to be related to AnRBC vxa chromc erythropoletln stamulatlon. Acute ~nflammatlon-related AnRBC via cytoktnes might be expected to be more transient. Our data are consistent with these hypotheses. Patterns of AuRBC tn the early newborn period vary with the underlytng associated fetal disease process(es), and are independent of neonatal ddutlonal effects.

215 INTERLEUKIN-10 IN MIDTRIMESTER AMNIOTIC FLUID OR SERUM DOES NOT PREDICT SUBSEQUENT INTRAUTERINE GROWTH RESTRICTION. Y~.~LQ~g*, DM Sherer, CB Jenkins*, A Ghidim, M Ossandon*, JC PezzuIlo*, FD SeydeI*, GS Eghnton Pennatal Research Fatality, Departinent of OB/GYN, Georgetown Unl’~ers~ty Medmal Center, Washington DC OBJECTIVE lnterleuk~n-10 (IL-10), a potent immunosappress~ve cytokmc, has been shown to be elevated in mldtnnlester ammonc fluid in patients w~th subsequent mtrautenne growth restnctmn (IUGR) compared w~th term controls This study was designed to evaluate IL-10 as a potentml predictor of subsequent IUGR m rather ammot[c fired or maternal serum compared with appropriate for-gestatmnal-age controls matched for gestational age (GA) ar dehvery STUDY DESIGN Patients who underwent mldtrunester ammocentesls between 1992 and 1995 wuh follow-up dehvery data were ~dent~fied IUGR was defined as btrthwclght < 10% for GA Control patients were matched for GA, maternal age, race and panty with at least two controls for each study patmnt Panents with maternal immune disease, chronic hypertension, diabetes, asthma, congenital heart disease, muluple gestation, and fetuses with structural or chromosomal anomalies were excluded Mldtrlmester amnlonc fluid and serum samples were assayed by EL|SA for IL-10 (Endogen Cmnbndge, MA) The ELISA senslt~wty for ammot~c fired was 5 6 pg/mL and for serum -<3 pg/mL IL-10 values were nommllzed using natural log transformarlon Potential confounding variables considered were maternal serum alpha-fetoproteln (MSAFP), smoking history, pregnancy induced hypertension, and neonatal gender Statlstmal analysis included X~ and ANOVA with p<0 05 considered s~gmficant RESULTS: From the amnmtlC fluid database, 18 patients (6%) delivered neonates w~th 1UGR and were matched with 46 controls (mean GA at samphng 17 3-+2 9 and 16 6-+2 9 weeks respecnvely, p=0 4) From the maternal serum database 13 panents debvered neonates wUh IUGR (7%) and were matched w~th 45 controls 0nean GA at samphng 16 2-+l 1 and 16 8_+1 5 weeks respectively, p-0 4) Amnlonc fired and maternal serum 1L-10 levels were not slgmficantly elevated m parmnts subsequently dehvenng IUGR neonates compared with controls (Table, mean _+ SD)

Ln IL-10 ,gghaL’ IUGR

I Co~xu:O’

I ~ Ammotm fired 3 0-+ 02 {n=18) 28-+ 05 {n=46) 0 ~ Maternal serum 29 -+ 04 (n=13) 30 -+ 07 (n=45) 060

No significant differences were Identified In maternal age, suloklng history, African American race, nulhpanty, GA, pregnancy ~nduced hypertension, or MSAFP >20 MOM m panents dehverlng IUGR neonates vs controls respectwely (each p value >0 05) As expected, b~rthweight was s~gmflcantly lower m patmnts dehvenng IUGR neonates compared with controls (p<0001) CONCLUSION After matchang for GA at dehvery, rmdtnmester ammonc fired or maternal serum IL-10 values are not predictive of subsequent development of I UGR

214 EVALUATION OF ANGIOGENIN IN MIDTRIMESTER AMNIOTIC FLUID AND SERUM AS A PREDICTOR OF INTRAUTERINE GROWTII RESTRICTION. CY~* DM Shcrer, A Ghld~nl, CB Jenkins*, M Ossandon*, JC Pezzullo*, FD Seydel*, GS Eghnton Pennatal Research Fac~hty, Departmem of OB/GYN, Georgetown Umversny Medical Center, Washington DC OBJECTIVE Prollferanon of villous capdlanes ~s seen in placentas exposed to chronic oxygen deficiency Markers of compensatory placental neovascularlzatlon may assist in the prediction of intrautenne growth restriction (IUGR) Angiogenm, a single-chain polypepnde, ~s a potent ~nducer of neovasculanzatmn Th~s study was designed to evaluate angiogenin as an antenatal marker for the pre&cnon of 1UGR m mldmmester amnlOtlC fluid and maternal serum compared wxth appropriate-for gestanonal-age controls matched for gestatlonal age (GA-) at dehvery STUDY DESIGN Patients who underwent m~dmmestei ammocentcs~s between 1992 and 1995 with follow-up dehvery data were ~denufied IUGR was defined as bmhwmght < 10% for GA Control patients were matched for GA, maternal age, race andparlty wuh at least two controIs for each study patient Panents with maternal immune d~sease, hypertension, diabetes, asthma, cohgemtal heart disease, muhlple gestatmn, and fetuses wnh structural or chromosomal anomalies were excluded Mldtrtmester ammotlc fluid and serum samples were assayed by ELISA for anglogenln (R&D Systein~, Mtnneapohs MN) The sensrtivlty for amnmtlc fluid was 0 078 nghnL and for serum 0 6 ng/mL Angm~emn values were normahzed using natura/Iog transforinanon Potential confounders considered were maternal serum alpha fetoprotmn (MSAFP), smoking history, pregnancy reduced hypertension, and neonatal gender Stansncal analysis ~ncluded Z~ and ANOVA w~rh p<0 05 considered significant _ RESULTS: From the amnlot~c fluid database 18 panents (6%) dehvered neonates wah IUGR and were tnatched with 46 controls (mean GA at sampling 17 3-+2 9 and 16 6+2 9 wks respectively p=0 4) From the maternal serum database 13 atmnts (7%) dehvered nebnates w~l{ IUGR and were matched with 45 controls (m~anP GA at samphng 16 2_+1 I and 16 8_+1 5 wks respectively p=0 4) Mean -+ SD amnlotlc fluid and matema selmn IL 10 levels were nor s~gmficandy different m IUGR panems compared w~th controls

Amnlotlc fired 34±07 (n=18) 3 1 ±05 (n=46) 006

Maternalserum 58±02(n=13) 59±02(n=45) 1 040

No significant differences were Identified belween maternal age, smoking history, African-American race, nulhpamy, GA, pregnancy ~nduced hypertension, or MSAFP >2 0 MOM m pauents dellvenng IUGR neonates compared wuh co~trols {each > value >0 05) As expected, blrthwe~ght was s~gmficantly lower in pauents debvenng IUGR neonates comp~ued wuh cmitrols (p<O 001) CONCLUSIONS: M~dtrlmester serum anglogenln is not a predictor for developtnent of IUGR However, elevated amniotlc fluid angmgemn levels do demonstrate a trend toward significance In the predlcnon of subsequent IUGR Addolonal studms are needed to clarify the relauonship

216 CORRELATION OF UMBILICAL ARTERY LEVELS OF INTERLEUKIN-6 (IL-6) AND SOLUBLE INTRACELLULAR ADHESION MOLECULE-I (SICAM-1) WITH UMBILICAL ARTERIAL BLOOD GAS MEASUREMENTS. *JC Sm~len, **WA Campbell, *AM Vmtzateos,

**JF ROdls *UMDNJ-Robert Wood Johnson Med=cal School/St Peter’s Med=cal Center, New Brunswick, New Jersey **Umversl~y of CT Heal~ Center, Farmlngton, CT OBJECTIVE Umblhcal arteff (UA) blood gas status may be compromised by advanced fetal sepsis This study was designed to determine the relahonshlp of fetal serologic markers of mfiammahon, UA levels of Ib6 and slCAM-1, to UA blood gas measurements STUDY DESIGN UA blood samples were collected at the time of dehvery Blood gas delermmahons [pH, pC2, pCO2, base excess (BE)] were performed on hepanmzed UA specimens Non-hepannlzed serum specimens were assayed for IL-6 and stCAM-1 using ELISA assays adlusted for minimum detectable levels of 7 pg/mL and 7 ng/mL, respectively The Spearman rank correlahon was used to determine relationships between UA levels of IL-6 and slCAM-1 w~th UA blood gas parameters Analyses were performed based on labor status, route of dehvery, and presence or absence of neonatal seps~s Blood gas results were compared between subgroups using the enpared t-test Significance p<0 05 RESULTS: UA specimens were obtained In 3 term labor, 2 term C-section, 4 preterm labor,

8 preterm rupture of membrane, and 2 indicated preterm delivery patients The median values of UA IL-6 and slCAM-1 were18 pg/ml (range, <7-45,174) and 159 ng/ml (range 81-473} IL-6 Correlahons pH BE

Total (19) 7 27 -0 389 0 098 -3 6 -0 475 0 044

Labor (11) 7 27 -0 697 0 028 -3,9 -0 761 0 016

No Labor (8) 7 28 0 047 O 902 -3 2 -O 171 O 650

SVD (8) 7 26 0 575 0 128 -4 8 -0 826 0 029

CS (11) 7 28 ~0 147 0 641 -2 8 -0 243 0 443 No Seps~s (13) 7 29 -0 608 0 035 -2 8 -0 373 0 196 Seps~s (6) 7 25 0 143 0 749 -5 3 -0 029 0 949

*Rho=Spearman’s coefficient for correlation of UA IL-6 levels and pH or BE UA pCO2 AND pC2 showed no slgmbcant correlations w~th UA IL-6 levels There were no s~gnlflcant differences between the mean values of the UA blood gas components for each analyzed subgroup UA slCAM-1 levels showed no s~gn~bcant correlations with UA blood gas measurements for any of the subgroups analyzed CONCLUSIONS. 1) UA levels of IL-6 have a significant inverse correlation with UA pH in the presence of labor 2) UA levels of IL-6 have s~gnlhcant inverse correlabons wdh UA BE for pahents undergoing labor and who have an SVD 3) Tbe lack of correlahon of UA tL-6 w~th pH or BE in patients with neonatal sepsis suggests that low pH and low BE in the presence of intraamnlobc ~nfectlon may be due to the effects of labor rather than advanced sepsis

372 sPa Abstracts .]an~a~ 1996 Am J Obsret Gynecol

217 THE RELATION OF UMBILICAL VEIN INTERLEUKIN-6 AND SOLUBLE

INTRACELLULAR ADHESION MOLECULE-I TO HISTOLOG~C PLACENTAL INFLAMMATION, *JC Smuhan, **WA Campbell, **L Ernst ,

*AM Vmtzdeos, **JF Rodls *UMDNJ-RWJ Medical SchooVSt Peter’s Medical Center, New Brunsw=ck, NJ **Umv of CT Health Center, Farm=ngton, CT OBJECTIVE Amn~obcfluid levels of mterleukln-6 (IL-6) and soluble mtrace}}u}ar adhesion molecule-1 (slCAM-1) have been shown to have significant assoclabons with acute and chromc ~nflammatlon of the placenta, respectively The study was designed to evaluate the relatlansNp of placental Inflammation (acute and chromc) to umb~hcal vein (UV) levels of IL-6 and slCAM-1 STUDY DESIGN UV samples were collected after cord clamping in 26 term and preterm dehverras The serum speomens were assayed in duphcste for IL~i and slCAM-1 using ELIBA assays adjusted for mlmmum detectable levels of 7 pg/mL and 7 ng/mL respecbvely Placental specimens were evaluated for maternal (MAI) and fetal (FAI) acute inflammatory responses (MAI was =dent=fled by the presence of chonodecldultls and/or amnlonlt=s FAI was identn~ed by the presence of umbilical vascuhtls and/or chonomc vascuhtls ) Both MAI and FAI were graded for severity using a score ofO-4 Chronic mflarnmabon (CI) was indicated by chromc lymphocytic and/or plasma cell infiltrations of the membranes, basal plate, vllh, or intervdlous spaces Mann-Whitney U was used to evaluate levels of IL-5 and slCAM-1 ~n the presence and absence of MAI, FAI, and CI KruskaI-Walhs test was used to evaluate fL-6 and sICAM-~ levels according to seventy of inflammation Significance p<O 05 RESULTS There were 20 preterm (13 preterm rupture of membranes, 5 preterm labors, 2 indicated preterm deSvenes) and 6 term gestabons 3-he mean gestatlona~ age at dehve~y was 32 3 _+ 4 1 weeks The median values of ~L-6 and slCAM-1 were <7pg/ml (range <7-21,384) and 149 ng/ml (range 62-1,223) IL~ levels were associated with both MAI (p=O 03) and FAI (p=0 02) IL-6 levels were

pos=bvely correlated w~h the severity of MAI response (p=O01) Soluble ICAM-1 levels w~re associated w~th FAI (pie O3), but d~ not reach statistical s~gn~ficance for k,1/k~ (pie 08) Levels of slCAM-1 were positively C~rre~ated w~th the sevent~ of MAI response (p=O 02) Levels of slCAM-1 were assocrated with CI (p=O 03), but IL-6 levels were not (p=O 37) CONCLUSION 1) UV ~evela of IL-6 are szgntficantly associated with ~ presence and seventy of hlstolog=c MAI and FAI responses 2) UV levels of slCAM-1 are s~gmficantly associated with both h=stolog~c FAI responses and CI, but also may have a weaker assooabon with MAI 3) UV IL-6 may be a better marker forMAl and FAI p~acental responses and slCAM-1 appears to be a better rnarker for CI placental lesions

219 CORRELATION OF UMBILICAL ARTERY AND VEIN LEVELS OF INTERLEUKIN-6 AND SOLUBLE INTRACELLULAR ADHESION MOLECULE-1 WITH NEONATAL HEMATOLOGIC INDICES AND EARLY SEPSIS. *JC Smuhan,

Campbell, *AM V~ntzdeos, **JF Rodrs *UMDNJ-Robert Wood Johnson Medical School/St Peter’s Medical Center, New Brunswick, NJ **Umv of CT Health Center, Farmlngton, CT OBJECTIVE The re~at~on ol umlN~ca~ cord blood markers of inflammation to coever~t~eoa( determinates of eady neonatal seps~s has not been descobed This study was designed to evaluate correlations of eady neonatal hematologic radices and sepsis status with umb=hcal artery (UA) and veto (UV) levels of mtedeuk=n-6 0L-6) and soluble mtracellular adhesleo melecule-1 (sICAM-1) STUDY DESIGN UA and UV samples were collected after cord clamping m preterm (<37 weeks) and term dehvenes The serum speomens were assayed in duplicate for IL-6 and slCAM-1 using ELISA assays adlusted for minimum detectable levels of 7 pg/mL and 7 ng/mL resbect~vely Neonates were categonzed as having edher suspected or confirmed sepsis (S) versus no seps~s (NS) within 3 days o1 birth by dm=cal and/or laboratory cntena Mann-Whitney U was used to evaluate levels of IL-6 and slCAM-1 based on sepsis status Neonatal hematologic studies including absolute neutropbd counts (ANC), absolute band counts [ABC), ~mmature/total neutropbll ratios (/K), and platetet counts (PC) were correlated with UA and UV levels of IL-6 and slCAM-1 using the Spearman rank correlation Slgmficance was set at p<0 05 RESULTS Therewere 17 preterm(S=7, NS=10) and 6 term (S=1, NS=5) dehvenes There

were significant differences of tL-6 levels between S and NS Infants for specimens from the UA (mean rank 17 8 vs 8 9, p=0 0017) and UV (mean rank 18 5 vs 8 5, p<0 0001)

IL-6 UA UV

(n=18) ~ho* p Rho* p

ANC -0 105 0 67 -0 203 0 40 ABC 0 668 0 006 0 640 0 008

lit 0 650 0 007 0 694 0 004

PC 0 006 0 98 -0 161 0 51

* Spearman’s coeff~:lent for corralaben of ~L-8 and hematologic results There were no s~mficant d~ftereeoes of slCAM-1 between S and NS infants for speolmans from either the UA (mean rank 133 vs 11 3, p=052) or UV (mean rank 138 vs 11 1, p=0 37) Neither UA nor UV levels of slCAM-1 correlated with neonatal ANC (p=0 7), ABC (p=O 8), UT (p=O B), or PC (p=0 8) CONCLUSIONS 1) UA and UV levels of IL-6 (but not slCAM-1) are slgnlflcanSy higher in neonates davelop~ng chnlcal and/or laboratory ewdence of eariy sepsis 2) UA and UV levels ef lL-6 (but not slCAM-1) correlate with neonatal ABC and I/T ratios 3) UmNical cord blood IL-6 ~s poter~tally useful as a marker for eady neonata~ sepsis

218 CORRELATION BETWEEN UMBILICAL ARTERY AND UMBILICAL VEIN LEVELS OF fNTERLEUKIN-6 AND SOLUBLE IN/RACELLULAR ADHESION MOLECULE-1. *JC Smuilan, *’WA Campbell, *AM Vmtzlleos, **JF Redls *UMDNJ-RWJ Medical Scheol/St Peters

Medical Center, New Brunswick, NJ, **Umv at CT Hea~lh C~nler, Farmlegton, CT OBJECTIVE The relationship between umb=ltcal artery (fetal compartment) and umbtl=cal vein (maternal/placental compartment) inflammatorY markers has not been descnbed previously This study was designed to determine if there are correlations between umbthcal artery (UA) and veto (UV) levels of either mtedeukm-6 (~L-6) or soluble ~ntracellutar adhesleo molecule-1 (sEAM-l) and to assess potenbal effects of gestatlrmal age (GA), labor, and route of delivery on these correlations. STUDY DESIGN UA and UV samples were separately collected after cord damping Serum speomans were assayed in duplicate for IL-6 and slCAM-1 using ELISA assays adjusted to have minimum detectable levels of ? pg/mL and 7 ng/mL respectwely The pared-sign test was used to compare UA to UV levels of IL-6 and slCAM-1 The Spearotan rank correlation was used to determine relationships between paired UA and UV levels of both substances for term gestahons (_>37 weeks), pretermgestations (<37 weeks), presence or absence of labor, and route of dehver~ A p<0,05 was considered slgmficant RESULTS The mean GA at deliverY was 32 4 _+ 4 3 weeks Themedlan values of )L-6 for UA and UV were 18 pg/ml (range <7-45,174) and <Tpg/ml (range <7-21,384), (pINS) The median values of slCAM-1 for UA and UV were 159 ng/ml (range 81-473) and 143 ng/ml (range 62-513), (pINS) UA levels were slgmhcantly greater than UV levels for both tL- 6 (p=0 039] and s/CAM-1 (p=0 035)

IL-6 slCAM-I

(n) Rho* p Rho* p

Total (23) 0 845"- <0 0001 0 806"- 0 000~

Preterm (17) 0 928 0 0002 0 818 0 001

Term (6) 0 786 0 079 0 714 0 11

SVD(12/ 0 762 0 012 0 764 0 0113

CS(11) 0 911 0 004 0 834 0 0084

Labor (15) 0 775 0 0037 0 789 0 0031

No Labor (8) 1 00 0 0082 0 881 0 02

*Rho=Spearman’s coefficient for corralatton of UA and UV levels CONCLUSIONS 1) UA levels of IL-6 and stCAM-1 correlate s~jnlf~cantly ruth UV levels regardless et the route of dehvery or the pres~ose of ~aber 2) The UA and UV correlations

are significant for preterm gestahons with a trend for s~grshcance with term gestahons 3) Either UA or UV specimens could be used for studies analysing cord IL-6 and slCAM-1, but the potential for fetal productleo should be considered If defining parameters for cl#n~ca] use

220 DISCORDANT/CONCORDANT ’I~VINS GROWTH, AND THE

LI~VEL$ OF INSULIN-LIKI~ GROWTH FACTORA (IGF-1)

INSULIN AND GROWTH HORMONE. A. Wi~itzer:, B. Furman~,

R. Gakman~, M. Mazor~, M. Marhach~, A. Koiplmaan~, L Levy~, LR.

Leibetmaa~, E.A. Reece. DelX. of OB/GYN & Clinical Laboratory of

Endocrine, Soroka Medical Center/BelvGurion Univ., Israel and Dept, of

OB/GYN & RS, Temple Univ Sch of Med, Philadelphia, PA, USA

OBJECTIVE: To investigate the role of IGF-I, insulin (IN), and growth

lx~mme (G~, in twin pairs, wi~ and without growth restriction. STUDY DESIGN: S~um samples were obtain~ fr(m~ 27 twin l~irs,

immediately after delivery. Assigameat to discordant twin group

w~s based m intertwin birthweighi difference >20~, and to concordant

twin group (hill), by intcrtwin birthweight difference <20~. Maternal

serum and c~d blood were analyzed for IGF-I, IN and GH. A receiver-

operator characteristic curve used intertwin IGF-I differences.

RESULTS: ~ larger twin had significantly higher (r=0.66, p <0.001)

IGF-I level in all cases of discordant twin growth compared with the

smaller twin. A mea~ intertwin difference in IGF-I cord blood levels of

69.9 ngtml was consistent with discordant twin gestation. "I"nere was

inteztwin differences in IGF-I levds in the concordant group (Figures).

Additionally, there was ~o ¢x~-relatio~ b~tw~ bkthweight ~ cord

Need level o~ ]~ a~ GH.

�ONCLUSIONS: Oar data deao~trate that IGF-! i~ ~a

in-~tero ~rowth promoter~ ~d seeas to ~]a)" ~, crucial role in ~orm~l

aad devlam fetal ~rowth.

Volume 174, Number 1, Part 2 SPO Abstracts 373 Am ] Obstet Gynecol

221 GLUCOSE AND AMINO ACID TURNOVER IN UNTREATED

GESTATIONAL DIABETICS. D. Zlmmer, A. GohchowskL

A Baronx, S. Dennex, Indmna UmversW Medmal Center, Indpls., IN OBJECTIVE: Although gestat~onal diabetes (GDM) affects as many

as 3% of all grawdas, ~ts specific effects on glucose and protmn metab-

ohsm have not been clearly dehneated It is proposed that GDM results ~n increased glucose production and proteolys~s dunng fasting STUDY DESIGN: The rates of appearance (Ra) of glucose (GLU). leucme (LEU) and phenylalanlne (PHE) were determined ~n 10 patients

with GDM within 2 weeks of dmgnosis and prior to imtmt]on of diet or

msuhn treatment Eight healthy, nondiabenc grawdas served as controls (C). GDM and C had a mean age of 25_+8 and 25_+6 yrs, gestatmnal age of 32-~3 and 32_+2 wks, weight of 85_+24 and 78_+9 kg

and body mass index of 33_+7 and 28_+3 kg/m2, respectively After an overnight fast, a prime constant intravenous infusion of L-[I-13C]-LEU, L-[nng-ds]-PHE, and I3-[6,6,d2]-GLU was admmtstered for 5 hrs.

RESULTS: (*p<0 05)

GLU Ra LEU Ra PHE Ra INS GLU

mg/kg/mrn p.moLe&g/hr ~tU/ml rag/all

GDM 2 2-+0 6 148+_29 40-+7* 26_+1" 76-+8*

Contml 2 4+-0.5 135-+18 47+_6 5_+2 70+_2 CONCLUSIONS: GLU concentrations throughout the study were

increased in GDM, but there was no increase m GLU Ra There was no difference in LEU Ra, and PHE Ra was actually decreased m GDM. Nor-

mal GLU turnover in the face of 5-fold h~gher prevaihng insulin (INS) concentrations suggest GDM ~s assomated with both hepatic

and peripheral INS reslstance. Elevated INS levels probably contribute

to the fact that proteolyms rates are not ~ncreased in GDM.

223 LONGITUDINAL STUDY OF cz-2 ADRENERGIC RECEPTOR NUMBER AND FUNCTION DURING PREGNANCY. R~chard M. Smfleyx, Carol B Pantuckx, Dept of Anesthes~olegy, Columbm Umv New York, NY OBJECTIVE To determine it c~-adrenergic receptor (c~AR) number or funchon ~s altered during human pregnancy STUDY DESIGN: Platelets were isolated from blood (30 ml) obta=ned from 21 healthy pregnant women at weeks 10, 20, 30, and 37 of gestation, and 10-12 weeks postpartum (PP). The func~onal status of ~Rs was assessed (m the presence of ptopranotel) by epinephrine inhibition of PGE-stimulated cAMP production (PGE/~2) and expressed in pmoles cAMP/IO6 cells/lO rain. c~2-inhibition of PGE- slJmulated cAMP production was also expressed as a percentage decrease (% a2 mhlb.). ~2AR num~r (Bmax, receptom per cell) and binding aflimty (KD, nM) were determined using H-yoh=mbme (1-20 riM) and 8catohard analysis. Data were analyzed using repeated measures ANOVA, RESULTS: The table includes mean data obtained thus far SEMs are omitted for readability. Data ~om 20 sublects should be available by the t~me of the meeting No statistically significant changes over the course of pregnancy have yet been detected

PLATELET ~AR CHARACTERISTICS

10 wks 20 wks 30 wks 37 wks P P n 15 19 14 11 9

PGE cAMP 2.0 2.8 3.1 3.7 3,9

PGEM~ cAMP 0.8 1 0 1 0 1 6 0 9

% ~2 inhib. 60 64 68 57 77

Bmax 227 229 231 253 253

KD 2.6 29 34 30 32

CONCLUSIONS c~ARs are involved in uterine contraction, vascular tone, and

neural signalling and modulation, Platalet cz~ARs appear to reflect utenne c~ARs

very closely (1). Knowledge regarding czAR function dudng normal pregnancy may lead to better understanding of the molecular physiology of pregnancy, improve management of hemodynamicelly compromised pregnant women, and contribu~ to understanding altera!]ons =n pare tolerance and other CNS changes m pregnancy Th~s work was supported in pert by a Climcal Schotar Grant from the Interna~onal Anesthesia Research SoQeiy. REFERENCES. 1 EurJ Pharmase1150: 403-404,1988.

222 COMPARISON OF FETAL LUNG MATURATION IN

WELL DATED DIABETIC AND NON-DIABETIC

PREGNANCIES K. Berkowitz. M.D. C Reyes, M.D., P.

Sadaat~ & S. Kjos, M D. University of Southern California

School of Medicine, Los Angeles, CA

OBaECTIVE: To compare fetal lung maturation in diabetic and

non-diabetic pregnancies and to evaluate the influences of

diabetic class, third trimester glycemic control and hypertension.

STUDY DESIGN: Prospective cohort study of well-dated

dtabetics (n=585) and non-diabetics (N=628) delivering between

1987-1992

RESULTS: 96% of diabetics and 90% of non-diabetics had

mature L/S ratios at 39 weeks. PG% was mature in 93% of

diabetic and 100% of non-diabetics at the same gestational age

Neither d~abetic classification, glycemic control or hypertension

affected the rate of maturation in the diabetic patients.

27 31 32 33 34 35 36 37 38 39 40 41

*p< 05

WEEKS GESTATIONAL AGE

CONCLUSIONS: Lung maturatton is not delayed in well-dated

drabetic pregnancies compared to non-diabetics. Neither diabetic

classfication, glycemic control or hypertension affect the rate of

fetal lung maturation among diabetic pregnancies

224 LONGffUDINAL STUDY OF ~2 ADRENERGIC RECEPTOR NUMBER AND FUNCTION DURING PREGNANCY Richard M Smileyx, Carol B Pantuckx, Dept. of Anesthes~olegy, ColumbB Univ. New York, NY. OBJECTIVE: To determine d g-adrenerglc receptor (BAR) number or funchon is altered during human pregnancy STUDY DESIGN BARs were studied on lymphocytas dunng pregnancy Blood (30 ml) was obtained from 21 healthy pregnant women at weeks 10, 20, 30, and 37 of gestation and 10-12 weeks pos’~a~m LPP) Lymphocytes were )sol.a, ted an~. cyclic AMP (cAMP) producf~on (pmoles/10~ cells/t0 rain J determineo m me basal state (BA8) and in response to ~’ea~’nent with isoproterenol (ISO), forskolin (FSK), and prostaglandm E1 (PGE1). BAR number (Bmax, receptors per cell) and

b~nding afhmty (KD, pM) were determined with 1251qodopindolol, and evaluated by Scatehard analys~s. Data were analyzed using repeated measures ANOVA RESULTS The table includes mean data obtained thus far. SEMs are omitted for readability Data from 20 subjects should be available by the time of the meeting No statishcally signilicant changes have yet been detected

LYMPHOCYTE BAR CHARACTERISTICS lOwks 20wks 30wks 37wks PP

n 15 19 14 11 9

BA8 cAMP 6 2 4 5 4.4 5.6 5 6

ISO cAMP 17.4 11.4 11 0 13.5 13.8

PGE CAMP 87.3 674 698 874 712

FSK cAMP 9.5 10 0 8.2 8 6 8.0

Bmax 1530 1265 1220 1162 1344

KD 42 41 21 16 32

CONCLLISIONS: The lymphocyte ~AR has been shown to be a good model for lIARs on other organs, and correlates particularly well ruth myome~la~ BARs Understanding BAR funchon dunng normal pregnancy may increase the understanding of the molecular phys~olegy of pregnancy, improve management of hemodynamically compromised pregnant women such as those with pre- ecfampeia or pre-exislJng cardiac valvular disease and rationahze the use of 82- agon~sts for preterm labor. This work was supported ~n part by a Chnlcal Scholar Award from the Intemat~onal Anesthesia Research Sodety REFERENCES’ 1 Chn Pharmacol Ther 1989, 45. 1-8

374 SPO Abstracts Jamia~y 1996 Am J Obstet (;ynecol

225 POST PARTUM CHANGES IN RAT UTERINE ARTERY VASCULAR TONE ARE DUE TO WITHDRAWAL OF NITRIC OXIDE. *GD Helmbrecht. **MY Farhat, **L kochbaum, **HE Brown, *GS Eglinton, "PW Ramwell Dept. *Ob/Gyn, and **Physiology, Georgetown University Medical Scho(W, Washington, DC OBJECTIVE: Endothelium derived nitric oxide (EDNO) is thought to contribute significantly to the decrease in vascular tone during pregnancy. We evaluated the effects of inhibition of EDNO synthesis on vascular reactivity during the post partum transition. TTUI)Y DESIGN: Pregnant Sprague-Dawley rats received chronic infusions of sahne (S) or the EDNO synthase inhibitor L-nitro arginine methyl ester (L)(LNAME 50 mg/day) from mid gestation to term. Uterine arteries (internal diameter 250 - 500m) were harvested on day 18-20 (P)(n=10) or 24 hours post partum (PP)(n=10). Vessels were mounted on a microvascular isometric myograph tor determination of vascular reactivity in response to prostaglandin-F2o~ (PGF2o0, acetylcheline (ACH), and sodium nitroprusside (SNP). Data were analyzed using two way ANOVA and Newman-Keuls test for multiple

RESULTS: The maximum tension developed (Tmax) in response to PGF2~ was greatest lot saline treated P compared to PP (3.25 v 1.60mN) (p<.01). LNAME attenuated the contractile force to a greater extent in P than in PP vessels (49% v 20%). Among the saline treated groups, relaxation response to ACH was greater for P than for PP (98% v 65%)(p<.01) reflecting greater EDNO activity in the former. LNAME significantly inhibited ACH induced relaxation. All vessels responded similarly to SNP. CONCLUSIONS: These data support a role for EDNO in the maintenance of the lowered vascular tone during pregnancy. Post partum changes in uterine artery vascular tone result from the abrupt withdrawal of the inlluence of EDNO on vascular smooth muscle

227 A PROSPECTIVE LONGITUDINAL STUDY OF TRANSFORMING GROWTH FACTOR - 6 1 (TGF- r~ 1) LEVELS DURING FIRST, SECOND AND THIRD TRIMESTER, LABOR, POST PARTUM AND WITH CORD BLOOD AND LABOR ANESTHESIA, F R~gby, E Bowen~, D Clawn, D Powersx, C. Campbellx, T Nolanx, S Greenberg*, Dept of OB/GYN, LSU Medical Center, New Orleans, LA Objective TGF- 6 1 ~s a multlfunctfonal polypeptlde growth factor for which the vest majority of human cell types have receptors It has been ~mphcated ~n the first tnmester decidual ~mmunoregulatory response Th~s ~s the first prospecbve longitudinal study of the levels of th~s growth factor ~n pregnancy Study Design" Peripheral venous blood was obtained dunng first, second and third trimester, m labor, w~th anesthesm administration and post partum Cord blood was also obtained. Results Please see graph Vertical ax~s is plasma TGF- 15 1 m ng/ml and horizontal ax~s is number of samples in each trimester.

80-

60

40-

,o. Iii I o

29 15 15 25 14 20 25

~ First Trimester

==Second Trimester

¯ Third Trimester

[] Labor

[] Anesthesia

¯ Cord Blood

[] Post Partum

Conclusion These values represent the first prospective longitudinal study of TGF- I~ 1 levels throughout pregnancy. The stabsbcally s~gnificant lower levels noted in second and third trimester and ~n labor may represent down regulat=on of th~s ~mportant ~mmuno-regulatory pepUde by the placenta.

226 CHARACTERIZATION OF AMBULATORY BLOOD PRESSURE IN NORMOTENS1VE PRENGNANCY. G Bmlyx, M

Ruddy~, E Malka×, F C~offi Section of Hypertension, LrMDNJ~Robcrt

Wood Johnson Medical School, New Brunswick, NJ

OBJECTIVE The purpose of the present study was to estabhsh the

normal ambulatory blood pressure (ABP) profile for normotenslve,

healthy pregnant women.

STUDY DESIGN A total of 84 pregnant subjects w~th normal clime

blood pressures were recruited for this study. Of these, 18 were first

trimester, 29 were second trimester, and 37 were third trimester Each subject underwent 24-hour Ambulatory Blood Pressure Monitormg

using a Spacelabs 90204 ABP apparatus There were no differences

among the groups with regard to age and height

RESULTS

Mean + SE 1st Trim 2nd Trim 3rd Trim

Clinic SBP mmHg 110+22 118"35 114±26 Clinic DBP 68 + 2 2 71 ± 2 7 70 a: 1 9

Awake SBP 1144. 1.5 113::1:1.6

Awake DBP 67 4- 1 3 66 4. 1,0 70 4- 1 2" Awake HR bpm 86+18 904-12 91:t:14

Sleep SBP 103 4- 2 4 103 4- 2 1 109 4- 2 5 Sleep DBP 58 4- 1 8 55 4- 1 0 61

Sleep HR bpm 74 4. 1.8 79 a: 1,5 79 4- 1 6

% Nocturnal Decline [= (awake-sleep BP) - awake BP) × 100)] SBP mmHg 914-13 914.12 764-0.9 DBP mmltg 1444-16 1614.13 1244-14

"Statistically significant difference by ANOVA (p _< 005) from 2rid trim CONCLUSIONS Ambulatory diastohc BP was highest m the third trimester, and lowest m the second, for both awake and sleep Awake SBP was also highest m the tlurd and lowest m the second trimester There was a sttmlar tendency in sleep SBP

228 ERYTHROCYTE INSULIN RECEPTOR BINDING IN PREECLAMPSIA: RS Eeermn~., ES Umstot,x RN Andersen,x

PR Casson,x and BM Sibai, Department of Obstetrics and Gynecology, University of Tennessee, Memphis OBJECTIVES: Insulin resistance may contribute to hypertension during pregnancy. Therefore, we compared insulin binding to erythrocytes by means of competitive radioreceptor assay in patients with preeclampsia and normotensive pregnancies. STUDY DESIGN: Blood was obtained after an overnight fast from eight patients with preeclampsia and from eight normotensive controls. All patients had normal glucose screening and no familial history of diabetes. Erythrocytes were used as an indicator of peripheral insulin receptor expression Erythrocytes were isolated by densit~ gradient centrifugation, washed, and reconstituted with buffer at 4uC, Serial

dilutions of biosynthetic human insulin were pre~p,K.ed and incubated overnight with an aliquot of red cells and an ~ ~’~’I insulin tracer. Erythrocytes were centrifuged through oil and counted in a gamma counter. Scatchard analysis determined maximal % binding, number of binding sites/cell, and affinity Ka RESULTS: There were no differences between the two groups regarding maternal age, BMI, gestational age, and fasting glucose and insulin levels at time of sampling (Table). As expected, preeclamptic patients had significantly higher mean arterial pressures (MAP) than controls (P=0.006). However, there were no differences regarding any insulin binding studies (Table).

PREECLAMPSIA NORMOTENSIVE

Maternal Age (yr) 21.8 ± 1.1 22.4 ± 1.2

Body mass index (kg/m2) 29 0 ± 1.8 28.1 ± 0.4

MAP(tmaHg) 99.7 ± 3.7 81.9 ± 4.8

GA at sampling (wk) 35.4 + 1 0 36.4 ± 0.6

Fasting glucose (mg/dl) 82.3 ± 3.6 78 0 ± 2 3

Fasting insulin (mu/ml) 7.5 + 1.6 10.5 ± 1.0

Maxima] binding (%) 15.2 + 2.3 14.3 ± 1.0

Receptor sites per cell 8.6 ± 0.8 7.5 ± 1.2

Affinity’ Constant (Ka xl09) 4.7 + 0.8 4.0 ± 0.5

CONCLUSION: Erythrocyte insulin receptor expressxon is not altered in preeclampsia. However, these findings need confirmation in target tissues of msulin activity such as adipocytes and myocytes.


229 IMMUNOHISTOCHEM|CAL LOCALIZATION OF PHOSPHOLIPASE A2

ISOFORMS IN HUMAN MYOMETRlUM DURING PREGNANCY AND

PARTURITION. D SkannaU A E~s,X D Brockman,~ T S~ddiql, L. Myatt.~

Dept. Ob/Gyn, Univ. of Cinti Coil Med., Clnonnat~, OH

OBJECTIVE; D=stinct =soforms of phospholipase As including 14kDa

secretory (sPLA~) and 85kDa cytosolic (cPLAz) have been demonstrated

The isoforms have d=fferent substrates spec~ficities and cellular Iocal~zations

which may define their roles in s~gnal ttansduchon pathways involving

arachidonlc acid mobilization and proetaglandin synthesis. The purpose of

this study was to ~dent~ and determine changes m expression and

incahzation of sPLA2 and cPLA2 in pregnant human myometnum w~th

gestational age or partunt~on

STUDY DESIGN: Myometnum was c~ected at cesarean section at term

(>37 weeks) or preterm (<37 weeks) from patients who were or were not

m labor /n = 5 each group), flash frozen ~n liquid nitrogen and t~ssues

sectioned on a cryostat at 7 i~m Sections were ~ncubated with spec=fic

monoclonal ant~bod=es against sPLA~ or cPLA~ and blotinylated anti-mouse

IgG second anhbody and vlsuahzed with the Vectastam ABC elite method

Control sections had no primary antibody Intensity of immunostain=ng in

different cellular Iocahzatlons was scored by an investigator blinded to tissue

ioentlty and compared between tissues using students ’t’ test

RESULTS: Secretory PLA; Immunostalmng was dispersed m the

perlnuclear region throughout the myometrlal smooth muscle fibers and m

vascular smooth muscle Variable intensity of staining was seen between

samples Cytosohc PLA~ tmmunostamlng was predominantly localized to

endothehal cells of myometrlal btood vessels and weakly throughout

myometrial fibers Overall sPLA~ immunostammg was more intense than

cPLA= No d~fferences m intensity or Iocahzat~on of sPLA2 or cPLA~ could

be determined with respect to gestat~onal age or partunt~on

CONCLUSION: Both sPLA= and cPLA= ~soforms are present ~n pregnant

human myometrium but at d~erent locations suggesting d~stinct physiologic

roles, The apparent lack of change in expression with gestation or labor

suggests changes in myometrial PLA= actiwty and, hence, local arachidoNc

acid mobilization and prostaglandm synthes~s may not mediate partunt~on

231 METABOLIC RESPONSE TO MEAL EATING AND EXTENDED OVERNIGHT FAST IN TWIN GESTATION. H Casele, S Dooley,

B Metzger Dept Ob/Gyn and Center for Endocrinology, Metabohsm and Nutrition, Northwestern Umvermty, Chicago, IL

ORIECTIVE: To compare the metabolic response to normal meal eating and the vulnerabthty to starvation ketosts m twin versus smgleton gestation STUDY DESIGN: 11 twin and 11 singleton non-dmbetic gestattons were enrolled into a 40-bout metabolic study Singletons were age (+ 5

yrs) and prepregnancy weight (+ 10% ideal Body Weight !IBW]) matched -onth the twins. The diet (35 kcal/kg IBW singletons; 40 kcaLtkg

IBW twins) was &stiabuted as 1/5 0800 hr, 2/5 1300 br, 2/5 1800 hr An overnight fast was extended untd noon the followmg day Glucose

and B-hydro~butyrate (B-OHB) measurements were made hourly except at oiglat when they were drawn every 2 hours Insulin values were obtained satrroundmg droner and on the day when breakfast was delayed RESULTS: ]qae glucose and insulin excursion m response to meal eating and fasting were snmlar m twins and singletons (ANOVA for repeated measures, p> 05) Dunng the day when the patients were fed, the ketone excursions are also stmilar However, during fasting, the ketone excursions are significantly different (ANOVA for repeated

measures, 13=.0001) At 8"00 AM after a |4-b, our fast, mean B-OHB was ,26 + 08 mmol/l for singletons and 28 + 12 retool/1 for twins (p> 05)

By noon, mean B-OHB was 46 _+ 10 mmo///for smgletorls and 76 + 36 mmol/l for t~a, ins (t=2 52, p< 05)

(7ONCLUSION: On the basis of these data, it avOe, urs that the metabohc response to meal eating xs smular m twins versus singletons despite the provismn of additional calories to the twins It also appears that twms are more vulnerable than smgletens to accelerated starvation When an overnight fast was extended by delaying breakfast, ketone

levels rapidly deviated We speculate that thts observed difference may be due to the increased metabohc demands of twin gestation

230 CHANGES IN CIRCULATING LONG CHAIN ESSENTIAL

FATTY ACIDS (LCEFA) IN NORMAL HUMAN

PREGNANCY, P. Ogbam, M Kanayamax, J. Van Wintcrx, K.

Schwarzx, R. Holmanx. Dept. Ob/Gyn, Mayo Chnic, Rochester, MN,

Dept. Pediatrics, Johns Hopkins, Baltimore, MD, Hormel Institute,

Austin, MN.

OBJECTIVES: The unll hypothesis is that patterns of circulating

serum phospholipid (PL) long chain essential fatty acids are unchanged

by pregnancy. STUDY DESIGN: Serum was obtained from groups of normal

pregnant women m each trimester of pregnancy (11-14 wks = 1st A

[n=16]; 26-30 wks = 2nd A [n=44]; 34-39 wks = 3rd A [n=12]). Fatty

acid (FA) ana!ysis of serum PL was done by lipid extraction, TLC,

methyl esterification, and capillary GLC. Mean + SEM results were

compared using Student’s t-test.

RESULTS: Percent of total fatty acids in PL of serum in each trimester

are given for each LCEFA.

Fatty Acid First A Second A Third A

Arach~donic acid 12.03 + 0 52 11 33 + 0.24 8.11 ± 0.33**

224w6 0.81+-007 0.62+--002* 040+0.01"*

22:5w6 0.82 + 0.06 0 80 + 0 03 0.79 _+_ 0 05

EPA (20 5w3) 0 60 + 0.03 0.27 +_. 0.02* 0 22 +- 0 02***

22 5w3 0 92 + 0.05 053 + 0 02* 0.37 + 0.02**

22.6w3 3.76+025 397+0.10 238+_016"*

(* = < .!301, 1st A vs 2nd A; ** = p < .001 2nd A vs. 3rd A,*** = p <

001 lstAvs 3rdA)

CONCLUSION: Patterns of circulating LCEFA in PL change as

pregnancy progresses Significant decreases are see in most LCEFA of

both omega-6 and omega-3 fatty acid families These decreases support

the cOncept that normal pregnancy transfers LCEFA to the fetus for

growth and development leaving the mothers’ fatty acid patterns

consistent with essential fatty acid deficiency.

232 OXYGEN TRANSPORT VARIABLES DURING NORMAL TItIRD-TRI- MESTER PREGNANCY C. Harveyx, G Hankans, S. Clark, E U~:kanx, D Cotton The Umverslty of Texas MeNcal Branch at Galveston, TX. The Umver-

s~ty of Utah, Salt Lake City, U’~. Wayne State Umverslty, Detroit, MI. OBJECTIVE: To directly measure oxygen transport variables at moderate altt- rude in healthy nonnotenslve pnrmparous pauents at term. STUDY DESIGN: Ten normotensive prianparous patients between 36 and 38 weeks gestation underwent pulmonary artery catheterization and radial artery canalization. The study was approved by the hospital Institutional Review Board and an outside reviewer. Written informed consent was obtaaned Basehne assessments were made m the left lateral recumbent posmon after a 30-rmnutes

stabilization period, cardiac output was measured via thermodflutlon. Simulta- neously blood obtmned from the pulmonary and radial arteries was analyzed for oxygen content on a blood gas analyzer (Cormng Model 168, Medfield, Mass ) All of the 10 subjects had complete oxygen transport profiles antepartum and

seven subjects had complete data at 12 weeks postpartum

RESULTS: In&vldual oxygen dehvery variables are hsted ~n Table 1.

CONCLUSIONS: This is the first report of directly measured oxygen transport variables in the healthy pregnant pataent. Although there were stansttcal differences in the antepartum (AP) and postpartum (PP) measurements, the difference is not chrucally slgmficant However, the difference in the AP and PP CaO2 and CvO2 are statistically s~gmficant and charcally relevant, as a reduced CaO2 and CvO2 in pregnancy must be addressed in medical management

376 SPO Abstracts ]anu,n~ 1996 Am J Obstet Gynccol

233 ARTERIAL BLOOD GAS ANALYSIS DURING NORMAL THIRD-TRI- MESTER PREGNANCYAND THE EFFECT OF POSITION CHANGES. ~, S. Clark, E U~kanx, C Harveyx, D. Cotton. The Uraversity of Texas Medical Branch at Galveston, TX, The Umverstty of Utah, Salt Lake City, UT, Wayne State Umverstty, Detroit, MI. OBJECTIVE: To establish normative data and evaluate the effect of position change on arterial blood gas and acid base status of normotenslve pnrmparous patients. STUDY DESIGN: Ten normotenslve prirmparous patients between 36 and 38 weeks gestanon volunteered to undergo radial artery canalization. The study was approved by the hospital Institutional Review Board and an outside reviewer Written informed consent was obtained. Basehne assessments were made in the left lateral recumbent posmon after a 30-minute stabthzaUon penod.A 10 minute

premeasurement stabilizatmn period was allowed between each position change. Measurements were obtained m the left lateral, right lateral, supine, sitting, standing, and knee-chest posltaons. Blood samples were analyzed m duphcate for oxygen content on a blood gas analyzer (Cormng Model 168, Medfield, Mass.) Statistical analysis was ANOVA with slgmficance defined at p<0.05 RESULTS: There was no slgmficant thfference in arterial blood gas variables

between any positions in this antepartum population of term healthy women (Table 1).

Po$1tlOfl Left lateral Right lateral Supine Sitting Standing Knee-chest

pn 746 ±002 746 5-006 745 ±002 746 5-003 746 ±001 746 5-002 pC02 26 6 5-27 27 4 ±2 0 267 5-21 264 ±22 25 5 ±2 4 26 7 ±12

p02 862 5-73 877 ±45 867 ±72 91 i ±73 888 ~78 895 5-59 HCO3 186 ±:9 183 5-14 188 ±14 184 5-08 179 ±07 173 5-16 Sub2 0 96 5-0 01 096 ±00 0 96 5-0 0 096 ±0 0 0 96 ±0 01 0 96 ±00

CONCLUSIONS: Artenal blood gas variables were not altered after positron change of the subject. These results were obtained at moderate altitude; however, chulcally significant changes in maternal oxygenation would have altered the hemoglobin saturation due to the low PaO2 of the arterial samples. Contrary to previous reports of thfferences between arterial Pub2 between sitlang and supine positions, we found no evidence of slgmficant arterial blood gas alter- atlons.

235 DEVELOPMENT OF FETAL CARDIAC COMPLIANCE

THROUGHOUT THE SECOND HALF OF pREGNANCY. Z_.

Wemeff.~.~, Z Efrat’,2, E Zlmmeff’2, J Itskovltz~’2, MY Dlvon1. Dept of

Ob/Gyn, Albert Emstear College of Medicine~, Bronx, N Y., and Rambam

Medical Center:, Halfa, Israel

OBJECTIVE: To study the development of the human fetal cardiac

compliance throughout the second half of pregnancy by measuring the 2

components of the ventricular filling: the rapid ventrlcular filling and the

atrial systole

STUDY DESIGN: A longitudinal study was performed on 25 low-rtsk

pregnant women from 24 weeks’ gestation until term Doppler studies of the

blood flow through the mitral and tricuspid valves were performed every 4

weeks using a pulsed wave Doppler ultrasound device (Acuson 128 XP1

The following indices were calculated from the flow velocity waveforms: 1)

The ratio between peak-velocity during the rapid ventricular filhng and the

atrial systole (E/A ratio), 2) The velocity time integral (VT1) of the atno-

venmcular blood flow (this integral correlates with volume flow), 3) The

ratio between VTI during the rapid ventricular filling and the atrial systole

(VTI ratio), 4) The FHR To improve the accuracy of these calculations we

accepted only measurements obtained with a beam angle < 20" Pearson’s

correlation was used to evaluate the effect of gestauonal age on these indices

RESULTS: All patients delivered at term and had an uncomplicated

pregnancy. Each patient had 4-5 tests Correlations between the Doppler

indxces and gestatlonal age are presented.

E/A ratio VTI VTI ratio

Mltral r=O 75** r=0 49* r=0.56" *p<0 01

Tricuspid r=0 8** r=0 55* r=0 61"* **p<0 001

There was a slight but nonsignificant decrease in FHR

CONCLUSIONS: These results indicate that, m the human fetus, the

relative contribution of the rapid ventncular filhng to the total ventricular

filhng increases as gestatlonal age advances. These findings could be

explained by an improvement in cardiac compliance and/or by a decrease in

peripheral resistance

234 HUMAN FETAL PULMONARY/SYSTEMIC VASCULAR

RESISTANCE. ,I. Rasanenx, D. C. Woodx, A. Ludomirakl*, J. C. Huhtax. Dept.Ob/Gyn, Pennsylvania and *Temple Univ. Hospital, Fhitadelphia, PA. OBJECTIVE: To establish the relation between human fetal weight- Indexed pulmonary (Rpi) and systemic (RsI) vascular resistances and the changes during the second half of the gestation. STUDY DESIGN: By using a cross-sectional study design 63 normal fetuses were examined between 19w and 39w (median 28w) by Doppler echocardiography. Heart rate (HR), diameters and time-velocity-integrals (TVI) at the aortic (AV) and pulmonary (PV) valve annuli, ductus arteriosus (DA), right (RPA) and left (LPA) pulmonary arteries were measured and weight-indexed blood flows (Qi) were calculated (TVI x area x HR/estmaated fetal weight). The systemic blood flow (Qsi) was calculated

by subtracting pulmonary blood flow (Qpi=QLPAi+QRPAi) from the fetal

combined cardiac, output (QAvi+QPvi). Fetal mean transpulmonary pressure gradient (P) was assumed to be equal to fetal mean systemic pressure gradient, which was assumed from values in premature newborns at the same gestational ages. Wmght-indexed vascular resistances were

calculated. R t (mmH g/mYmin/kg )=P(mmHg)/Qt (ml/min/kg) RESULTS: Mean£+SD values at three gestational ages (comparison to 20w group: *p<0.05, **p<0.000l; comparison to 30w group: ’~p<0.005, ~Ip<0.001).

P QPi Qsi Rpi RSi Rpi/Rsi 20w 30 91+33 496+160 0.35+0.08 0.07:L-0.03 5.9_+2.5 30w 42 188_+26"* 503-+74 0 23_+0.03* 0.08_40.01 2.7-+0.7* 38w 55 138_+18~ 470£-_99 0.40-2:0.05~1~ 0.12~+0.02" 3.4_+0.3 CONCLUSIONS: The decrease in Rpi between 20w and 30w reflects lung

growth and its increase later in gestation may be due to acquired vasoconstriction in the pulmonary circulation. The changes in Rpi/Rsi show the tnagmtude of the decrease in the Rpi and its importance in the regulation of the distribution of fetal cardiac output. We speculate that this regulation could be altered by therapeutic manipulation of fetal RPi/Rsl.

236 MYOGENIC ACTIVITY AND ENDOTIIEL1UM DEPENDENT DILATION

IN ISOLATED BLOOD VESSELS FROM PREGNANT WOMEN. H.

NiselP, K. Rasa-Kublickiene~, M.Yo Divon, B. Lindblom~, N. Olov LanelP, M.

Westgren Department of OB/GYN, Huddinge University Hosp.; Uppsala

University, Sweden & Albert Einstein College of Medicine, N.Y., U.S.A.

OBJECTIVE: To evaluate changes in vessel diameter in responsetu constant

and step-wlse intraluminal pressure alterations under no-flow conditions in

isolated resistance blood vessels from norrnal pregnant women. The vasoaetive

function of the endothelium was assessed by the response to aeetylcholine(Ach).

STUDY DESIGN: lntramyometrial and oraental arterioles were obtained during

C/S and pressurized in a superinfused vessel chamber that allowed the internal

diameter to be assessed continuously using video microscopic techniques.

RESULTS: At a constant intraluminal pressure of 70 mmHg in calcium free

solution w~th papaver~ne (10~M), the passive diameter of omental and

myometrial arterioles did not differ significantly (3114-20 ~m (mean :kSEM) and

305-+28, respectively). In Hepes-physiolog~calsaline (PSS) both binaural and

myometrlal vessels developed spontaneous tone which reduced lumen diameter

by 20-+7 % and 34-+7%, respectively, (p=NS). At an intraluminal pressure of

70 mmHg, Aeh (10 ~M) induced relaxation & was sigmficantly higher in

myometrial (24_+5%) compared with omental resistance vessels (9:k2%),

p<0.05. In the absence of calcium and presence or" papaverine step-wlse

increment in perfusion pressure from 20 to 120 mmHg evoked a continuous

s~mllar increase in diameter ~n both preparations. In Hepes-PSS change in

mtralumlnal pressure of myometrial arterioles from 40 to 60 and 60 to 80

mmHg, caused an initial dilation, followed by myogenic constriction that

returned the vessel diameter to the imtial value over a period of several minutes.

In omantal vessels, these changes in intraluminal pressure produced only a small

passive dilation.

CONCLUSIONS: Myogenic responses are more pronounced in myometrial

arterioles as compared with omental vessels. The results indicate an obligatory

role for calcium in pressure-dependenttone development of resistance vessels in

normal pregnancy. Ach receptor-mediated release of endothelial-derived

vasodltators in the uterine vascular bed is significantly enhanced relative to

omental vessels.

Volume 174, Number 1, Pall E SPO Abstracts 377 Am | Obstet (;~necol

237 Effects of advancing Pregnancy on the Umbilical Artery (UA) Pulsitility Index (PI) after Strenuous Non-Weight Bearing

(Bicycle) Exercise (Bex).

J C VeIlle, DR KitzmanX, K Tatumx, K Stewarff, P Mflsapsx Depts

Obstet/Gynecol & Medacine (cardiology), Bowman Gray School of

Medicine, Winston-Salem, NC

Introduction Previous studaes mdmate that strenuous Bex does not

significantly affect the UA Doppler waveforms (UADW) late m

pregnancy The present study attempts to determine the effects of Bex

iongitudmally on the PI of the UA

Study Design: Fifteen condattoned subjects were studaed 3 hmes dunng

the course of the pregnancy Subjects exermsed on an electncally-

braked bmycle (EBB) The remstanee of the EBB was increased by 25

watts Q10 rain untal the subject had reached an RQ of 1 I, their

maxtmal heart rate or had mgmtieant leg fatigue UADW were obtatned

prior to the exercise (Ex) and soon after the completion of the Ex A

mimmum of three UADW mete analysed and averaged PI were

compared pre and post Ex Results are reported as ~ 4- SEM

Stat~stmal analyses were done using

non parametric and ANOVA

Results: 1) PI decreased slgmficantly

ruth advancing GA (p< 0 002),

2) There was no significant changes

m the PI between pre and post Ex

3) Advancing OA dad not affect

UADW response to strenuous Bex

Conclusion: Strenuous Bex does not

seem to affect UADW at any of the

GA periods studied

239 Longitudinal M-Mode echocardiography (Ec) of the End Diastolic (EDD) and End Systolic (ESD) dimension of the right OtV) and

Left (LV) ventricles from early Fetallife to Year One. J C VeIIle, K Tatum~, L Steelex, S McNed~ Depts ObstetJOynecol & Pedlatrms (Cardtology), Bowman Gray School of Medicine, Winston-Salem, NC Objective: To evaluate RV and LV EDD and ESD using Ec m fetal, neonatal (Transitional) and infancy so as to understand hemodynam~e adaptatmn dunng these periods

Material & Methods: Ec were started at the 16th week of gestatmn m 52 normal fetuses and repeated every 4- 6 weeks until term Two more Ec were done dunng the "Transitional" period, and 3 more were done at 6 weeks, 6 months and at 12 months of age The M-Mode cursor was placed so as to transect the A-¥ valves All rettnrnlng Ec were recorded on a stup chart at a preset speed of 50mm/see EDD and ESD were measured according to published standards Mean and SEM for each of the EDD (graph) and ESD were tabulated for each of the 11 study periods Correlation coeffic~ant for LV and RV were made and ANOVA for repeated

measures were used for significance

Results: 1) LV was h~ghly correlated w~th advancing age tAG} (r~0 93; p<0 0001); 2) Although RV was eorrolated to AG (ta=0 361,p<0 05), RVEDD had a slower rate of growth after Nrth, 3) RVEDD was s~gmfieantly

larger in utero (p<0 01 ), and LVEDD was

sigmficantly larger after birth (p<0 0001)

Conclusion: Thts ~s the first s~ady to longitudmally follow the rate of growth of the RVEDD and LVEDD dunng the fetal to the infant pertod Changes m respective ventncular afterload most likely influence venhacular dimensions

238 Effect of Advancing Gestation on E!A Ratio of the Left and the

Right Ventricle after Strenuous Bicycle Exercise. JC Vclll~ DR

Kltzman×, K Tamm×,K Stewart~, P Mflsaps× Depts Obstet/Gynecol &

Medmine (Cardiology), Bowman Gray School of Medacme, Winston-

Salem, NC

Introdnction D~astohc ventricular filling (DVF) is assessed by

analysis of Early (E) to Atrial (A) ventrlcular inflow Doppler waveform

The purpose of tbas study was to determine the erect of strenuous btcycie

exermse (Bex) on the right (RV) and the left (LV) ventricle E/A ratio

with advancing pregnancy

Material & Method Fifteen healthy subjects were studaed three tLmes

dunng thear pregnancy Doppler waveforms (DW) were obtained from

the RV and LV just below the A-V valves prior to and immediately

after Bex A minimum of 3-5 DW were traced and averaged Results

are reported as X and :!: SEM One way ANOVA rank test were done

for sl!gnfleanoe

Results 1) There was a slgmficant decrease in the E/A ratto of RV

between rest and Bex only during the

1st study period (p<0 02) 2) There

was a s~graficant increase In the E/A

ratio w~th advancing GA (p < 0 02) 3)

Bex did not slgmficantly affect tins

ratio during the other study periods

Conclusion Overall, strenuous Bex

dad not significantly affect the DVF tn

human fetuses

240 Longitudinal Follow-up of the Descending Aorta (DA) Hemodynamics in Normally Grown and IUGR Fetuses From Early Intrauterine Life To Year One. J.C Veflle, K. Tatum~, L SteeleX,C Legault~, Depts Obstet./Gynecol & Pubhc Health, Bowman Gray School of

Medacine, Winston-Salem, NC Introduction: Studms have shown conflicting results regarding

blood flow through the DA m 1UGR human fetuses. No longitudanal stu&es followed these fetuses into childhood to

demonstrate residual effects from the poor in-utero environment.

Materials & Methods: Sixty-six (66) Normally grown fetuses

and tharteen (13) fetuses with 1UGR were followed every 4 weeks with Doppler U/S from the 18th week until delivery Two

studies were done after dehvery and one at

6 wks, 6 months and 12 months

Results: There was no overall sigmficant difference between the two groups durmg

fetal, transitmnal and newborn period for ~~ -

the peak flow velocity (Top figure), for the tame velodty integral (Lower figure)and for "’

the AC/ET ratio (no figure shown) between

the two populations. Conclusion: These preliminary results do not indicate major differences m the

vascular hemodynamic of the DA between

these two populations.

378 sPa Abstracts ,lanuaLy 1996 Am [ Ob~,tet

AN EVALUATION OF FETAL/MATERNAL PLASMA LEUCINE ENRICHMENTS IN NORMAL AND INTRAUTERINE GROWTH RETARDED PREGNANCIES. A.M.M~moni~, I.Cetinx, ILOo, voli~, C.Paolini*, S.Ronzonix, R.Fanellix, M.Buscagliax, G Pardi, F.C.Battaglia. Dept. of Ob/Gyn, Univ of Milano, Italy; M.Negn Inst. for Pharmacoh Res., Milano, Italy; Dept of Peal Univ of Co Sch of Meal, Denver (USA) OI~JECTIME: To assess fetal maternal leucine relationships in AGA and IUGR fetuses at the time of fetal blood sampling. STUDY DESIGN: A matema~ primed, constant infusion of L[1- ~3C]Leucine was given in 3 normal IAGAI and 6 intrauterine growth retarded (IUGR) pregnancies, IUGR pregnancies were divided according to velocimetry in the umbilical artery (PI) and to fetal heart rate (FHR} into: Group 1, normal FHR and PI, 2 cases; Group 2, normal FHR, abnormal PI, 3 cases and Group 3, abnormal FHR and PI, I case. Maternal arterialized samples were taken at time 0 and every 15 minutes for 110.6 -+ 18 min. Umbilical venous samples were obtained after 94.2 + t8 min from the start of the infusion. RESULTS: There was no difference in the fetal/maternal (F/M) leucine enrichment ratio between AGA and IUGR of Group 1 (0.79 vs 0.82). On the contrary, the F/M ratio decreased significantly (p<0.03J in IUGR of Groups 2 (0.721 and 3 (0.63~. In all patients, there was a significant linear relationship between maternal leucine disposal rate (DR) and concentration (Leu DR i~mol/kg/min= 0.44+10.8 Leu conc #mol/kg/min; r2= 0.67; p<0.01 ). CONCLUSIONS: This study shows that in IUGR pregnancies there is a progressive dilution of the fetal leucine enrichment, relative to the maternal plasma enrichment according to the severity of growth retardation. Since leucine ~s an essential amino acid, this dilution may reflect an increase of protein breakdown within the fetal and/or placental compartments and/or decreased transplacenta[ leucine flux.

243 CEREBRAL METABOLISM IN THE OVINE FETUS DURING HEART RATE DECELERATIONS WITH UMBILICAL CORD COMPRESSION. B. R~chardson.~, L. Carmichae~~, J. Homan*, L Johnstonx Dept Ob\Gyn and Physiology, Universit~ of Western Ontano, London, Ontado, Canada. OBJECTIVE: We sought to determine the extent to which cerebral blood flow (CBF) is =ncreased dudng variable-like fetal

heart rate (FHR) decelerations with umbilical cord compressmn, and whether cerebral oxygen dehvery (CDO2) and oxidative

metabohsm (CMRO~) are maintained. STUDY DESIGN: Nine near-term fetal sheep were stud~ed

immediately prior to, and again dunng the associated FHR decelerat=on w=th reduced umbihcal cord compression of -60 sec duration, and immed=ately afterwards. Cerebral arterial (a) venous (v) differences were analyzed for oxygen content, blood gases

and pH, CBF was measured w=th the m~crosphere technique. RESULTS: Umbilical cord compression with associated FHR deceleration from 158+_6 (SEM) to 67+_4 bpm resulted m a drop in fetal Pa02, from 21 +_1 to 14_+1 torr (p<O.01). CBF was vanably

changed, increasing from 184~_17 to 220~24 ml/lOOgm/min (NS) during cord compression and to 227 +- 12 ml/100gm/m~n (p<O.01) afterwards, w=th CDO~ thus decreased dudng cord compression, from 618+_48 to 470+_30 /~nol!lOOgm/min (p<O.02). CMRO~ remained little changed as cerebral fractional 02 extraction was

~ncreased from 0.28+_0.01 to 0,35+_0.03 (p<0.02) dudng umbilical cord compression, thereby contributing to the drop ~n sag=ttal sinus PvO~ from 17~1 to 11 +1 torr (p<0.01). CONCLUSIONS: The increase m CBF during moderate to

severe variable-like FHR decelerations with umbilical cord compression is insufficient to maintain CDO~ with a transient decrease in tissue oxygenation w=thin the brain occurring.

242 ANTEPARTUN ~IDDLE 14EAN CEREBI~L BLO(X) FLO!~ VELOCITY CORRELATIO~ WITH NATERNAL HENODYNARICS. Keith UtLID’II~, Susan Wilson, B.C.~o~en’s ~’lospita|, Dfv. of Raternet-Fetal. Red|tins, Van. B.C., Canada. V6N 5V5. 0BJECTWE: Rear maternal cerebral blood flo~ velocity has bee~ correlated in the Literature with cerebral vasospasm. No attempt has been ~ade to correlate =iddle cerebral blood flou velocity

with other ~aternaL hemo~ma~ic factors. ~e as=essed ~fddle cerebral blood flow velocity snd correlated with other ~aternal hemod~i c f a~tors, STUOY DESIGN: 5 normotensive patients ~ere assessed. Ratennat cerebral blood flow velocity was assessed using transcrantsl Doppler. Cardiac output, stroke voL~ne and other ~aternat hemodynamic factors were assessed using a non-invasive he~ody- nanfic monitor using thoracic electrical i~oedanee. All he~ody- namic parameters were sacpted four tl~es, assessing ~ean cerebral. blood ftou velocity, putsatitity index, heart rate, simultaneous- Ly mean arterial pressure, cardiac output, stroke vott.m~e and left ventricutsr end diastolic vol~e. Statistical analysis was then done using Pearson correlation coefficient and logistic

regnessfon analysis Mth stepuise regression. RESULTS: Near values for all of the parameters were developed. Both putsetik|ty index and mean middle cerebral blood flow velocity correlated significantly with meon arterial pressure (P<o05) but did not correlate with cardiac output, stroke volume or wfth any other ~=aternol he~odyna~ic parameters.

Cerdie¢ Output LIPId) 8.6 + 2.6; Cardiac Index (IPN/m2) 4.9 + 1.2;

Stroke Vot~r~ (N1) 10~ .9-1 32.7; Heart Rate (b/see) 83 :t: 4.9; Near Arterial Pressure (m~’lg) 8.19 9.7; Near Nfddte Cerebral

BLood Flo~ Velocity (caVeat) 59.6 ± 11; Pulset|tfty Index .85 ±

CONCLUSIONS: NJddte cerebral velocity correlates u|th mean arterial pressure but not with cardiac output or any other internal henxx:lyna~ic pare=star, In preectacpt|� patients at risk for cerebral vasospas~, control of ~ean arterial, pressure wilt result in icprove~ont in =can cerebral blood f|o~ veIoc|ty.

244 THE EFFECT OF DURATION OF LABOR ON THE

IMMUNOLOGICAL STATUS OF HEALTHY NEONATES.

W. Sch611{~!, P. Konstantiniuk, I. Kern, A Gmliani, B. Teuchner.

Dept. Ob/Gyn, Univ. of Graz, Austria

OBJECTIVE: Our purpose was to assess reference values of lympho-

cyte subpopulations in umbilical cord blood after normal spontaneous

deliveries. Is there a difference in immunological status with increa-

sing duration of labor~

STUDY DESIGN: In a prospectiv non randomized study the immu-

nological status of 120 full term healthy neonates of mothers with

uneventful pregnancies and spontaneous deliveries was determined.

Lymphocyte subpopulations were analysed by flow cytometry. The

full blood count and the differential blood count were performed 60

deliveries w~th durations of labor shorter than the median value were

compared to 60 exceeding this period. Data were analysed with either Students t test or Mann Whitney Wilcoxon U test or Spearman tes[.

RESULTS: S~gnificant elevations of total leukocyte count, total lymp-

hocyte count, T suppressor/cytotoxic cells, cytotoxic T cells and total

natural killer ceils were observed with prolonged labor. Total T cells,

T helper/inducer cells and the helper/suppressor ratio were significant-

ly lower after prolonged duration of labor. No difference between the

two groups were observed for neutrophiles, monocytes, total B cells,

activated T cells, T &/or NK cells, T helper (CD4+/Leu8-) cells, T

suppressor/inducer (CD4+/Leu8+) cells and DR+ Non T cells.

CONCLUSIONS: Prolonged labor influences immunological compe-

tent cells of the fetus with activation of unspecific immunity and de-

pression of specific immunity. Changes of the immunological status of

the fetus m~ght be a precursor to fetal compromzse.

Volume i74, Number 1, Part 2 SPO Abstracts 379 Am 10bstet (,~nec ~1

245 BREATH PENTANE CONCENTRATIONS DURING LABOR

AND THE EFFECT OF EPH)URAL ANALGESIA ON THE

PENTANE CONCENTRATION. Y. S._.~_.~_, J. Collea, Y. Kim’, S.

KimL Depts. Ob/Gyn/Anesthesia, Georgetown Univ., Washington, DC

OBJECTIVE: Free radicals cause tissue damage by fipid peroxidation

of cell membrane lipids. Increased fipid peroxidation has been observed in pregnancy and particularly in preeclampsia. Physical

stress is also known to increase the lipid peroxidadon. Pentane, a

product of lipid peroxidation, is a reliable index of lipid peroxidation

in vivo. We measured pentane in the breath of laboring women and

the effect of epidural analgesia on the level.

STUDY DESIGN: In thirty-six normotensive parturient women

~mdergoing induction of labor, the first end-expired breath samples

were collected into a bag (1 L) on admission. When epidural analgesia

for relief of labor pain was requested, breath samplings were repeated

before and after the epidural analgesia. The epidural analgesia was

performed using bupivacaine and fentanyl. Pentane was assayed by a

digital gas chromatograph, and its concentration (ppb) was converted

to a molar concentration. Data were compared by analysis of variance.

RESULTS: Pentane was detected from more than 95% of the breath

samples. The mean concentration of pentane on admission was

0.126+0A38 nmolfl (mean±SD), and the concentration was higher

(0.201+_0.204 omol/l, p<0.01) during labor. After epidural analgesia,

pentane levels decreased to 0.091+__0.106 nmol/l (p<0.001 compared to

the concentration before epidural analgesia).

CONCLUSIONS: This study demonstrates that pentane can be

measured from single breath samples in parturient women. The

increased breath pentane levels during labor suggest that labor may be

accompanied by a rise in lipid peroxidation. Epidural analgesia may

reverse the rise.

247 MANAGEMENT OF THE TWIN OLIGOHYDRAMNIOS-

POLYHYDRAMNIOS SEQUENCE AND TWIN-TO-TWIN

TRANSFUSION. ~ Brune[, T. Anderson~ Dept Ob/Gyn, Vanderbilt

University, Nashville, "12q"

OBJECTIVE. Evaluataon of serial decompression arnnincentesis in

Patients with the twin oligohydramnios-pelyhydramnins sequence (TOPS)

and twin-to-twin transfusion (TTT).

STUDY DESIGN’ Women fulfilling the standard ultrasound criteria for

the diagnosis of TOPS in the second trimester were assigned to TOPS or

true TTT groups after performance of sequential cordocentesis to

document presence absence or presence (respectively) of inter-twin

transfusion. All pregnancies were mmmged by deeompresston

amnioocntesm to relieve signs or symptoms of polyhydramnios.

RESULTS: Of 8 women in this study, true TTT was confirmed in 4. An

average of 2 5 amnioconteses was performed (range 0-6), removing 200-

3000 ml of amniotic fluid per procedure (menn=677 ml). Patients with

proven TTT requwed more amniocenteses (p=O.04), although less amniotic fluid was removed per procedure (p=0 005). All 8 suspected recipients were livebom, compared with only 2 donors; both donors were documented TTT Mean gestational age at delivery was 30 5/7 weeks (range 26 0/7-38 0/7 weeks) w~th no significant difference between groups CONCLUSIONS. These data illustrate TTT as a subset of TOPS, both of

which can be managed successfully by serial decompression of

polyhydrsmnic (reeipien0 twins Suspected recipients benefit from

prolongation of pregnancy in the absence of more serious comphcations of

fetal hypervolemia. Decompression amnioeentesis does not appear to

benefit ohgohydramnic (donor) twins. Although pregnancy outcome was

stmilar m both groups, the increased frequency of decompression required

in documented TTT ts likely related to patent anastomotic vessels, and

sugge~ that more aggressive surveillance or intervention may be requi~d

in these patients.

246 FETAL CARDIAC AND RENAL DOPPLER EVALUATION IN

PREGNANCIES WITH IDIOPATH!C POLYHYDRAMNIOS

J Rosne#, M Penry~, JC Veille. Dept. of Ob/Gyn, Bowman Gray

School of Medicine, Winston-Salem, NC

OBJ-ECTIVE: To compare fetal cardiac and renal blood flow in

euhydramnic and idiopathic polyhydramnic fetuses, the null

hypothesis is that fetal cardiac contribution to renal blood flow is

increased in pregnancies comphcated by idiopathic polyhydramnios.

STUDY DESIGN: Five fetuses in pregnancies complicated by

idiopathic polyhydrarnnlos (IP), mean AFI=26.7, mean gestational

age 28.4 weeks, were compared to 56 euhydranmic control fetuses,

mean gestational age 30.7 weeks (P=NS). Doppler waveforms

were recorded from the right and left ventricles just below the A-V

annulus, and from the renal artery at the renal hilus. Waveforms of

the right and left ventricles were analysed for time velocity integral

(TVI) during diastole, and dimensions for the fetal arteries were

obtained using the leading edge technique during systole. TVI and

area were used to provide estimation of total cardiac output (TCO)

and renal volume blood flow (RVBF). Mann-Whitney Rank Sum

Test was used to significance at P<0.05.

RESULTS: Fetuses with IP had significantly reduced TCO

(P=0.035), RVBF (P<0.0001), % of TCO perfusing the kidneys

(P=0.030) and renal artery diameter (P<0.001).

CONCLUSION: This preliminary study suggests that fetuses w~th

IP have a significantly decreased renal artery diameter, decreased

TCO and % TCO perfusing the kidneys, when compared to control

fetuses. This suggests that increased renal perfusion does not

appear to be responsible for IP.

248 HUMAN AMNIOTIC FLUID DYNAMICS: MATHEMATICAL MODEL OF FETAL SWALLOWING AND INTRAMEMBRANOUS FLOW S.._~.E Mann, MJ M Nijland’, M G Ross Dept. Ob/Gyn, Harbor-UCLA Med Ctr, Torrance, CA OBJECTIVE: AmniotJc fluid (AF) volume and composition is maintained by a balance of fetal fluid secretion (fetal urine, lung hquid) and resorplmn (fetal swallowing, mtramembranous flow) Among the s~tes, only AF volume and fetal urine flow can be accurately quan~ed in human pregnancies We sought to develop a mathematical model quantifying ¯ e relaWe contnb~ons of fetal swallowing and intramembranous flow to the maintenance of AF volume and compos=tion dunng human gastat~on STUDY DESIGN: Published data of human AF and fetal urine comp~on and volume (11-42 wks), and extrapolated data from ovine lung fluM production were utilized Modelling assumptions included: (1) da~ly changes in AF volume result from a net d~erence in AF production and resorption processes, (2) 50% of secreted lung fluid enters the ammot=c cavfty, (3) lung fluid =s isotomc to fetal plasma, (4) v~th the exception of swallowed lung fluid, swallowed fluid ~s isotonic to AF, and (5) intramembranous flow is free water diffusion. RESULTS’ Calculated fetal swallowed volume (S) and intramembranous flow (I) -’~* S are similar (7 to 230 ml/day)until 28-30

_=:’.,, ... ~;~......i.. 1"

weeks. Da~ly swallowed volume then ~ =, .:" exponentially increases to a maximum of " 1320 ml/day at term, while intramembranous flow continues on a linear trend to reach 336 ml/day at term CONCLUSIONS: The mathematical model indtcatas that the normal reductmn m AF volume beginning at 34 wks results from the marked increase =n swallowed volume dunng the third trimester. Modelhng of AF dynam~ea can predict normal changes in fetal fluid exchange and may atd in understanding etiologies of AF imbalances

380 SPO Abstracts January 1996

Am I Ob~tet (,ynecol

249 THE EFFECT OF INTRAVENOUS FLUID LOAD ON AMNIOTI4C FLUID

INDEX IN PATIENTS WITH OLIGOHYDRAMNIOS I. Bush,× H Minkoff,

S. McCalla S. Moy, H Chung Dept. Ob/Gyn SUNY-HSCB Brooklyn,NY

OBIECTIVE: To determine the effect of intravenous fluid load on the

amniotlc fluid index (AFI) of women with ohgohydrammos and to

determine factors that predict response.

MAIERIA[$ & METHODS: A prospective cohort study of patients wtth

ohgohydramnios (AFI < 5.0 cm) admitted for induction of labor was

performed at SUNY Brooklyn and Kings County Hospital Center.

Patients with rupture of membranes, congenital anomahes, &

contraind~cations to fluid challenge were excluded. Urine specific gravity

and serum BUN and creatnme were performed. 500cc of normal saline

or nngers lactate was infused over a thirty minute penod. The infusion

rate was then set at 125cc per hour, An AFI was performed prior to and,

30 minutes, 1 hour, and 2 hours after the initiation of the fluid load. To

assess reliability and inter-obsewer vanabihty, two examiners who were blinded to eaeh others fmdingmeasuredtheAFI m a subset of patients.

The statistical design used was the student t test.

RESULTS= Fourteen patients were included w~th an average gestational

age of 39.3 weeks. Twelve patients had an increase in their AFI 2 hours

after the ~mtiat~on of the fluid challenge. Seven had a 50% increase in

their AFI. Three patients had a 100% increase in AFI. Four had an

> 5 after the fluid load. No correlation between percent increase in AFI

and urine specific grawty was noted.

CONCLUSIONS: 1. Intravenous fluid challenge is associated with an

increase ~n AFI in a majority of patients with oligohydramnios , ~n

several cases resulting in normal values. 2. WKh th~s small sample size

we were unable to estabhsh a correlation between urine specific gravity

and AFI increase 3. Further research ~s needed to determine if known

nsks of morbidity associated with oligohydrammos are less severe if fluid

level increases with fluid load.

251 EFFECTS OF FETAL SEIZURE ACTIVITIES ON FETAL HEART RATE, BLOOD PRESSURE AND BREATHING MOVEMENTS IN FETAL LAMBS S. Dol x, y.

Murat& E.J. Quilligan, N. Nagatax, T Ikedax, 8. Park.x Dept. OB~SYN, Uni~, of Caltlorma, Ir’ane, Orange, CA OBJECTIVE: To determine the fetal biophysical responses to episodic intrauterine seizures after recovery f~om asphy’~c insults, STUDY DESIGN: Fetal heart rate (FHR), breathing movements (FBM), blood pressure {FBP) were observed during 325 convulsive episod.es.[n ejgh.t @ron~ca!ly.~nsi{u, m~nt~ near-term fetal lambs equ~ppod with ECG and ECou e~ec[, ro~., m~ralracneal ana caro~d arterial catheters. Four fetuses showed asphy’,dc evidence (pI--1<7.0,BE<-16) at the end of surgery, but survived. The remaining fetuses underwent the surgery without significant changes in add-base status and were. subseguently subje..~..d, to.s~, vere asphy’~a (pH<6.8, BE<-20) by umbilical cord ocousion tot 60rain All the re[uses shm,,ed complete recovery to normal aod-base status after the insults, but later developed ~ntrauterine seizure activities. Fetal seizure was ~dent~ed by a repetitive and thythmic discharge in ECoG conf~rmed by lerk’y body and extremity movements 14s&,a,lized ~g 2-dJnlensional ulfi’asound of the fetus Fetal b ood gas analys~s was pedormed every 2 hours after the onset of each sezure. RESULTS: Fetal bophysical changes auhng a seizure ac~’~bj

Nonaoidemic p1-1~7.20 Acidemic #-1<7.20 Number of seizure episodes 257 68 Fetal Blood Pressure MAP elevated >10rnmHg 220(85.6%) 57(83.8%) MAP decereased >10mmRg 0 3(4.4%) No Changes 37(14,4%) 8(11.8%) Fetal Breathing I~ovemer~ Repof~ive rapid and deep FBMs * 160(62.2°/=) 10(14.7%) Deep ~nspirafion * 43(16 7°/o) 53(77 9%) No Changes .54{21 0%) 5(7.4%)

Decolerat~on >10bpm * 33(12 8%) 59(86.8%) Acceleration >10bprn * 68(26.5%) 0 ~,ooe~’a~on tottowe~t by deceleration * t46(56.8%) 0 No Changes 10 (3.9%) 9(1&2%) MAP: mean arerial pressure * p<0.001 .t~ analysis or Fisher’s exact probability CONCLgSIONS: Fetal seizure activibes produced elevated blood pressure and abnormal breathing movements in both nonacademic and aodemic fetuses. The most common FHR pattern in nonacidemic fetuses was acceleration followed by decelera~on. FHR decaleralion wasthe most common pattern in aodem~c fetuses

250 THE EFFECTS OF DEVELOPING AUTONOMOUS NERVOUS SYSTEM ON FHR VARIABILITIES DETERMINED BY THE POWER SPECTRAL ANALYSIS IN RHESUS MONKEY FETUSES M Matsuurax, Y Murata, T H~ranox, N. Nagatax, S. Dotx, K Sudax Dept. of Ob/Gyn, Umv. of Califorma, I~ne, Orange, CA OBJECTIVE: The purpose of this study is to evaluate the development of f~tal autonomous nervous system(AN’S) and the effect of developing ANS on FHR vanabihtms(LTV, STV). STUDY DESIGN: E~ghteen chromcally instrumented fetal rhesus monke~swere used( ~i4- 152 days gestanon(term=i65 days)). R-R intervals of ECG were calculated(resolution of t msec) and resampled at equal time intervals of 35o msec. The power spectral analysts(PSA) by fast Fourmr transform was performed. The frequency bands were d~v~ded into low frequency (LF=o.o~ 0.2 Hz) and h~gh frequency (HF=o.4-Lo Hz). Studies using autonomic antagonists have indicated that these were mediated by the sympathetic(SYM) and parasympathet)c nervous systems(PSYM), respcctively(Akselrod S. AJ Phys. 249:11867,~98~). The power spectral density(PSD) of LF,HF and LF/HF ratm were then calculate& LTV and SqN were defined as PSD of FHR fluctuations m o.o5-o.1 Hz(equlvalent to 3-6 cycles/ram) and as the total PSD o~the difference of beat-to-beat vanabflity, respectively. The data were dwided mto twogroups according to fetal breathing movement(FBM) present or abscent. Statistical anabfsis was performed us~nglirmr regression and ANOVA. REgULTS: 0 Both LF and HF showed sigmficant mcreases(p<o.o~ r=o.46) with gestanonal age(GA). A rate of the increase in HF was 3 times more than that of LF. The LF/HF showed a rap~d ~ncrease >GA 0.8 term 2) LTV showed a sigmficant ~ncrease (p<o.ol r=o.42 ), but STV showed a sigmficant increase only dunng FBM(p<o.oot r=o.~2) 3) LTV presented a posUwe correlationwth LF and LF/HF (p<o.oom r=o.99, p<o.ooz STV showed a posmve correlation with HF(p <o.oom, r=o.97), but an reverse correlatmnw~th LF/HF (p<o.o~,r=o.32). CON(;LUSIONS: [) Both SYM and PSYM system developed as GA advanced, and the development rate of PSYM was about 3 rimes more than that of SYM by PSA of FHR fluctuatmns. However, SYM dominated PSYM acnwty, parncularly >o.8 term. 2) LTV and STV increased’ruth the development of SYM and PSYM, respecnvely. 3) At a ~ven GA, an increase in LTV or STV corresponded to a dominance of gYM or PSYM. 4) An increase in STV appeared responsible for an increase in Varlal31hty dunng FBM.

252 SPECTRAL ANALYSIS OF HEART RATE VARIABILITY

GIVES NEW INSIGHT INTO FETAL AUTONOMIC

RESPONSE TO VIBROACOUSTIC STIMULATION.

R.M. Lewinsk¥’, S. Degani*. Department of Obstetrics &

Gynecology, Bnai Zion Medical Center, Haifa, Israel.

OBJECTIVE: To characterize the fetal autonomic response to

vibroacoustic stimulation (VAS) with and without prior exposure

to narcotic analgesics.

STUDY DESIGN: Fetal ECGs from 24 uncomplicated term

pregnancies were analyzed. The power content of frequency bands

representing sympathetic, parasympathetic and respiratory activity

was determined. Measurements were taken during natural and

narcotic-induced sleep, before and up to 1 hour after VAS.

RESULTS: The immediate response to VAS was characterized by

a significant increase in total power (p<0.01) and shift towards

the very low frequency band (0-0.05 Hz) representing sympathetic

activity (from 46 _+ 17% to 87 -+ 24%; p<0.001). This response

was observed both during natural sleep or after narcotic-induced

sleep. The decrease in power in the low (0.05-0.15 Hz) and high

(0.5-0.9 Hz) frequencybands representing parasympathetic activity

and respiratory sinus arrythmia respectively, were proportionally

equal. While the increased sympathetic activity after VAS was

sustained for up to 1 hour during natural sleep, after narcotic-

induced sleep, a complete recovery to prestimulation values was

observed after 10 minutes.

CONCLUSIONS: Spectral analysis of FHR variability is effective

in quantitating the fetal response to VAS. The change in fetal

wake state, but not the immediate "startle response" is suppressed

by narcotic analgesia.


253 QUANTITATIVE ASPECTS OF IN VIVO 1H MR SPECTROSCOPY OF HUMAN FETAL BRAIN. Paul P. van den Berg~, Arend Heerschap~, Departments of Obste~cs and Gynaecology and Diagnostic Radiology,

University Hospital Nijmegen, The Netherlands OBJECTIVE: The principal possibility to use proton Magnetac Resonance Spectroscopy (1H MRS) as a new non-invaslve tool to wew a number of metabolltes in the human fetal brain has been recently demonstrated [Heerschap A, van den Berg PP, Am J Obstet Gyn 1994;170:1150-1]. The purpose of this study is to quantify the levels of some of the metabolites observed in the obtained spectra. STUDY DESIGN: Five healthy women with normally grown fetuses between 28 and 38 weeks of gestation were investagated after informed consent was obtained. Exammations were performed on a Siemens MR

system at 1.5 Tesla fieldstrength. A 10 cm diameter surface coll was used for MR s~gnal reception 1H MR spectra were obtained from a box shaped volume, set at a location guided by MR images using a double spin-echo pulse sequence (T~ = 75-135 ms, T~ = 1.6 s, 384 scans, about 10 mm scan tame). The T2 relaxation tame of the H20 spins were obtained from measurements at &fferent echo lames. RESULTS: Metabohte levels in a brain volume were estamated from its MR spectrum takang the signal of H20 of the same volume as a reference Published data obtained from the preterm neonatal brain [Kreis R et el.,

MRM 1993;30.424-437] were used to correct for relaxation effects on the H20 and metabolite proton spins. Employing these relaxation data and a

brain water content of 90%, we derived average concentration values (per kg brain tissue) of 1.9_+0.4 mM for choline compounds, 4.1_+1 0 mM for creatine and 2.8_+0.9 mM for acetyi compounds These values are in good agreement with prewous data on brain metabohte contents of preterms CONCLUSION: 1H MR spectroscopy is the first technique that enables the acquisition of objective parameters about fetal brain metabolism during pregnancy. Apart from a poss~bthty to characterize non-invasively certain metabolic aspects of the developing human fetal bran, a very m~portant potential would be the detection of signs of insufficmnt oxygen supply.

255 ASSESSING COGNITIVE PROCESSING IN THE HUMAN FETUS. LI

~ DM Mooney~, SB Hoilaud~, L8 l~tlZ’, JL Attsrbory~, ~ l~..l~a, .sa~., University of South Alabama, Mobile, AL; Arkansas Children’s nenptr~, Lime Ro~k, OBJECTIVE: Studies in neonates, infants, sad adults provi.de con.vinci~.g e .~.’d .ence that a heart nee (HR) decole~ato~y .~spons¢ to Iow-mtensay slmmli is a physiological index of ~guiS~e proceesm.~ Howev~, a crilical fi~. tor~. of this relationship is that relatively inteese stimuli should evoke a HR a~.lcrati.on. The pu .r~o. se of ~ study was to del~-rmine if human fotas~, like .newborn. m~, ts, exhiblt a decm~e in Hit when ~timnlatsd with low-intonsi~ speeen soun~ saq an incvea~ in HR when sound intensity i~ increased. The finding that low- and high-

intensity sounds evoke different ~ype~ o.f ~ responses would h~elp support .the argument that a HR deceleration i~ a physiological correlate of coSmt~ve proceesmg in the human fetus. STUDY DESIGN: 18 low-risk human fetuses at 37-40 weeks were exposed to a

30-s sound consisting of the re~. " .g phonemes/ee/and/ab/ .(50~. ~s/ee/, 500

of 1024 Hz ~nd fetal R-wa~ were est~cted usmg adapUw u~nm .pt~ce~.mg. teshnk~. Each fetes was in q~et ~t~:ep (QS) for 3~ m bef~e being rano.omtzen to ~ the sfimulas at sa intensity of either 80 dB or 90 dB. At 5-7 m ~llowing Simulas offs~ fetuses who had received the 80-dB (90-dB) sound were ~amulated in QS aseeoadtim~atanintensity of 90-dB (80-riB). Response magnitoda,an.d the~ effect of ~imubis intensity we~ e~..on~d by ~po~.ted measures ana~ysts o[ varis~ late~d effeets were teswd ~ath ennservaa~ degrees of freedom, sad the linear, quadrat~ sad enbie orthognnal polynomial trenda were computed. RESULTS: The~ w~ a ~y significant difference in the mesa HR between the 80-dB and 90-dB response etu~en, F(l,17)ffiS.31, P<0.011, sad trend analyses

showed that the quadratic ~eeponsea for the 80-dB and ~9(~dB~ ,~t~,~li were ~fi~atly different, F (1,17)~10.15, p<0.006. Follow-up te~ts o~me ~

o~ response trend~ from baseline .revVed. ¯ signifie~at d~eleratmy l~.sponse ~ stimulatien a~ an iat~ ffSO dB, quadratic .~�~n. d~ F (1,17)=7.60, p<0.012, anu eubi~ ~ F (.1,17)ffi6.10, p<0.025; and a. ugnificant a~eleratoty response to stimul~t~a et sa ~ of 90 all}, .q~ ~ad~ F (1,17)~..72, p<&020. The interaction between ~thnulu~ intensity ami HR w~ ~ ugnificamt, F (39, 663)ffi2.67, p<0.027. CONCLUSION: Low-risk huma~ fetmu~ exhibited a sastai~d deerenso in HR to

intemity w.as inca~eased to 90 dB. Our l~u. Its ar~. enn.slste.nt w~ta postnalal ~uolcs

earlyaspe~ of mgnitive po~easmS, This work was eup~ by~a Srant IR29HD32767 5~n the NZGHD.

254 PSYCHOSOCIAL FACTORS AS PREDICTORS OF LOW

BIRTH WEIGHT AND PRETERM DELIVERY

K.M. Paarlberg~, A.J.J.M. Vingerhoets~, J. Passchier~,

G.A. Dekkerx, A.G.J.J. Heinen~, H.P. van Geijn~. Department of Obstetrics, Free University Hospital Amsterdam, The Netherlands.

OBJECTIVE: To study the influence of psychosocial factors on

low birth weight (LBW) and preterm delivery (PD). STUDY DESIGN: 396 nulliparous women completed questionnaires on: general characteristics, daily stressors, psychological and mental well-being, social support, and work factors in each

gestational trimester. Outcome measures were: LBW (birth weight _< 10th percentile; n=40) versus normal birth weight, and

PD (gestational age < 37 wk; n=27) versus term delivery. Results were analysed by multivariate logistic regression and

expressed in odds ratio’s (OR) and their 95% confidence inter-

vals (CI). RESULTS: Independent of maternal weight and height, educati-

onal level and smoking, the followin~ predictors were found: trimester LBW I’D

CONCLUSIONS: Number of hours housekeeping per week in the first trimester demonstrated to be the most pronounced predmtor of low birth weight independent of biomedical risk. Psychological factors appear to have more influence on preterm delivery than on low birth weight.

256 LONGITUDINAL STUDY OF BEHAVIORAL STATE ORGANIZATION: CONTINUITY BETWEEN THE FETAL AND NEONATAL PERIODS. LI ~ MJ Swiber~, JL Altedm~y~, LS Bunts ~, SB Holland ~. University of Sonth Alabama, Mobile, AL. OBJEL~IVE: There is now a large body of data demonstrating that behavioral stste o~mlzatien is a marker of central nerv.ous system (CNS) d~ve!op .m.z~nt in bo~

fetuses and neonates. The purpose of thls study was to detemun.e ~t~al eta orgsaiza~on was predtettve of stato ors .a~.ization in the n .eon.atal period. STUDY DESIGN: 30 low-tisk human subjects were examined as fetuses at 38 - 40 weeks gestation and again as neonates at 2-weeks postnatal age We were pa~ esze£ul to control f~ potenlinl confounding variables: only non-smoking mothers with no medical or obsts~cad complicahons wer~ reeruited; pregnant mothers fasted after midnight and were given the same standard meal on an-ival to the tes6ng unit; behavioral states were assigned similarly for fetuses and neonates

in terms of heart rate (HR) pattern and the presence or absence of eye (EM) .and body movements; and fetal and neonatal studies were conducted at the same time of day in a quiet room. Data were analyzed using the Student t-test, Z~, sad linear regresalon. RESULTS: The fraction of time spent in quiet sleep (QS~ active sleep (AS), and tt~nsiUon l~rleds was virtually identical for fetuses and neonates (QS: 33.3 + 7.9% vs 34.5 + 7.8%; AS: 56.3 + 9.3% vs 57.7 + 7.2%; Trees: 10 3 + 8 7% vs. 7.8 + 3.5%). No relationship was found between the duration of AS epochs for fetuses

and ~e duration Of AS epochs for neo.na.tes (r=0.i.74, p>0.05); but fe~,ses who ~ m a .concrete bout of Q~ for a .cert~ .n length of Ume had QS bouts oxme same w~vs dum~e.wh~. _ examined again at 2-weeks pos~atel ~e.(r=0.500, p--0:007). Fetosee made s~gmficsatly fewer QS~AS and AS~QS transalons.~ 100 m~n. utes of sleep than neonat~ (2.9 + 0.9 vs 3.6 + 0..6, p<0.001) but .r~qu:re~ more tame to complete a state change (QS~AS: 2.7 + 3.4 mm vs 2.1 + 1.9 nun, pffi0..064; AS-QS : 3.4 + 5.3 min w 1.7 +_ 2.1 rain, p--0.006). For both fetuse~ sae neonat~ a significant difference was found b~wcon QS~AS and AS~QS transitions in the

ordering Of HR and EM fin QS~AS tnmsiSena, HR most oi~’n dumSed bo~_re EM; and for AS-QS trensition~, EM most often ebenged before HR (fetus: ~’=50.9, p<&001; neonate: X~ffi91.3, F<0.001). There was no diff~ence in the ord~." g of HR and EM between fetus~ and neonats~ for the s~te change QS~AS (Z~ffi4.5, p>0.05), but cededng of HR and EM for. AS~QS transitions was significantly more structured for neonates tlum fetuses (X -~7.1, p=0.028). .

CONCLUSION: Our result~ indicatte that, where~ ~CI~._S ~ regnlatinl~ QS, were fairly well developed by 38-40 w~. of fetal life, ~gnificant m.~..Uoonl ~hsages o~un~d over the 4-weeks ~mmg the ~ sad neoma~ pofi~h m tbe capacity ofindivid~ ~ubje~ts to regulate AS.

382 SPO Abstracts Janu,u,~ 1996

257 EXPECTANT MANAGEMENT IN PREGNANCIES COMPLICATED BY

GROWTH RETARDATION AND ACIDEMIA? S.~M B~rry, SJ Fm~dx,

MP Dombrowski, JM Lanouettex, CL BrownX, DB Cotton. Dept.

OB/Gyn, Wayne State Univ/Hutzel Hospital, Detro=t, MI.

OBJECTIVE: Academia, a frequent finding m ~ntrauterme growth

retardation (IUGR), has been used as a lust~flcat=on for framed=ate

dehvery. Our objectwe was to report our experience with

pregnancies comphcated by IUGR and fetal acidem=a.

STUDY DESIGN: Rap=d karyotype was the primary mdicat=on for

venous cordocentesis among 10 IUGR fetuses at 23 to 37 weeks.

Data were prospectively collected over a 3 year period.

RESULTS: All fetuses were karyotyp~cally normal. Estimated fetal

wmghts were < 5th%tfle ~n 9, and < 10th%tde ~n one fetus. Four

had absent arterial end diastohc flow prior to cordocentesis. Nine

fetuses had hemoglobin (Hgb) values > 951h%tde for gestat~onal

age (CA). Atl specimens had pH and oxygen content < 5th%tile for

CA. E=ght spec=mens had blood gas values cons=stent w~th a mixed

ac~dem=a. Two fetuses w=th a metabohc ac=dem=a were delivered

> 2 days post procedure. Six pregnanc=es continued > 2 days after

cordocentes=s (mean=24.3, range 2 to 112 days). The mean 5

minute Apgar score was 8.5, and 2 neonates had an umbilical

arterial pH <:7.2 at dehvery. There were 9 cesarean sections (CS),

8 for non-reessunng fetal heart rate tracings ~one for a procedure-

re~ated bradycard~a), and one for a breech presentation.

CONCLUSION: Fetal compensatory mechanisms for academia

appear to ~nclude an increase m Hgb. In our experience, feta~

ac=dem=a =s not an indication for immediate delivery in many cases.

Nonetheless, acidem=a is strongly assomated with subsequent CS

for non-reassunng fetal heart rate tracings

259 THE EFFECT OF ACIDEMIA ON FETAL HEART RATE VARIABILITY, ~), Chaff~x, J. Bernardx, T. Phernettor~x, A. Newmanx, K. Reed. Departments of O~GYN, Arizona Health Sciences Center, Tucson, AZ, University of Wisconsin, Madison, WI, and Robed C. Byrd Health Sciences CantedChadeston Division, Charleston, WV. OBJECTIVE: To test the hypothesis that acidemia with normoxia decreases heart rate variability. STUDY DESIGN: Seven time.bred ewes near term were surgically instrumented for FECG and fetal blood gas measurements. After a 5 day recovery period, baseline tetal ECG and blood gas tensions were measured. A 0.5M HCI solution was then infused at a rate of O,764ml/min into a fetal vein. FECG recordings and blood gas measurements were repeated at a fetal pH 7.15, 7.05 and 6.95. The RR intervals were obtained from the FECGs using a computer. Long and short term variability (LTV and STV) were calculated using the algorithm of Huey et at. Resume were compared using ANOVA wilh repeated measures, RESULTS: The results are shown in the table. LTV and STV are repoded as beats/100 beats. Fetal head rate and long and shod term variability increased with acidemia similar to the increase repoded with acute hypoxia. CONCLUSIONS: These results suggest that the decrease in heart rate variability seen wilh chronic asphyxia is not the result of acidemia alone.

~’arameter Co’r~trol pH 7.15 p~7,05 pH 6.95 p Value I

H~ad Rate (bpm) 15~--+16 177+20 177+_26 177£42 0.05

LTV 28£18 62£20 g5£40 67+25 0.0004

STV 93£37 202+85 275£118 241-+125 0.006

pC02 (mmHg) 52.7±1,8 55.9_+1.8 56.0+_2.5 60.0£4.3 <0.001

pO2/mmHg) 20.9£1.6 24.6±3.0 26.3+2.1 26.8-+2.0 <0.001

258 PI3kCEI’,WMb ARTERY NITRIC OXIDE SYNTHASE EXPRESSION INCREASES DURING THE THIRD TRIMESTER. C Sheppard, CE Shawx, ]M

Birdx, RR Magnessx. Pennatat Research Labs~ Dept Obstetrics and

Gynecology, U Wisconsin, Madison, WI. OBJECTIVE: To meet the metabolic demands of the growing fetas during

the third lnmester, placental and uterine blood flows increase

dramahcally. We hypothesized that ~ncreased expression of endothelial mtnc oxide synthase (eNOS) contributes to flus augmentahon of perfus~on

m both fetal-placental and uterane circulalaons.

DESIGN: Placental and uterine arteries (PA, UA) were collected from 20 ewes at 110, 120, 130, and 142 days’ gestataon (term=145+ 3d).

Endothehum was removed from the luminal surface and solubthzed.

eNOS expression was measured by Western analysis and ECL detec~on.

Densitometry data were normahzed to human urabthcal veto endothelial

cell (HUVEC) s~gnal and are reported as the mean percentage of HUVEC standard _+ SE Data were analyzed by ANOVA.

RESULTS: eNOS expression was localized to the endothehum (and not the denuded vessel) by Western

blot and car~]’med by

lmmunocytochem~stry. PA eNOS expression at 130 d was Z.5-folcl greater (*p=0.02) than at 110 d.

At term, PA eNOS expression had

dechned to levels s~mflar to those

observed at 110 and 120 d. In

contrast, UA eNOS expression was

tma/tered throughout the third

trimester.

60o ] ~ p~ ~

2O0

O~ 110 120 130 142

Days

CONCLUSION: Placental artery eNOS expression peaks at 130 d gestahon, but no increase in UA eNOS occurs. Tins supports our

hypothesis vnth respect to the fetal compartment, and suggests that the

increases in placental and uterine blood flows occur by independent, specffac mechamsras. Supported by NIH HL 49210 and HD 3:3255.

260 RELATIONSHIP OF ACID-BASE STATUS AND NEONATAL MORBIDITY IN _<1000 g INFANTS. D,F, Kimberhn, J.C. Hauth, R L. Goldenberg, C MecPherson~, E. Thornx. S F. Bottoms, D. McNellis Dept, of OB/GYN, Univ. of h, lab~ma at B~rmir~gham, AL and the NICHD MFMU Network, Bethesda, MD. OBJECTIVE: To evaluate the relationship between acid-base status at birth and neonatal morbidity ~n _<1000 g infants METHOD~: In a one year (1992-1993) prospectwe, observational study, the NICHD MFMU Network collected outcome data for 799 infants < 1000 g. Only fetuses who were deemed potentially viable by the obstetrician and who would have recerved a cesarean section for fetal indicat=ons were included in our analysis. We evaluated umbilical artery acid-base status and selected neOnatal outcomes. Logistic regression was used to control for the effect of confounding variables (b~rthweight, race, gender, mode of delivery, chorioamn~on~t~s, rnatemal MgSO4 or corticostero=d therapy and neonatal surfactant therapy). RESULTS: Umb~hcal artery (UA) pH and base deficit resuRs were ava=lable for I99 potentially wable infants. In this subgroup, the mean UA pH was 7.25 _+ .11 (<7.00, n=5; 7 01-7.05, n=7; 7.06-7.10, n=lO; 7.11-7.20, n=27; >7.20, n=150). The mean UA base deficitwas -6,51 -+ 438 (between 0 and -10, n=140; between -10 and -15, n=22, between -15 and -24, n=9). Univarlate analyses did not reveal a statistically significant assooiation between UA pH or base deficit and selected neonatal outcomevariables: seizure act~wty (p= 211, p=.818), grade Ill/IV WH (p=.568, p=,229), abnormal neurologic exam (p=.O79, p=.550), and intact survival (p=.332, p= 233). As UA pH decreased, there was a s~gnificant increase in the frequency of prolonged (>35 d) mechanical venti]ation (p= .009). Regression analysis controlling for multiple potential confounders confirmed the association between UA pH and prolonged mechanical ventilation. When arterial, venous, and unlabeled specimens were grouped (n=363), similar univariate and regression analyses results were obtained. CONCLUSION: I~ this series of _<1000 g infants, decreases in UA pH were associated with an increased dsk of protoPIged mechanical ventilation. There was no assoc=ation between UA pH or base deficit and grade Ill/IV IVH, seizure activity, abnormal neurologic exam, or =ntact survival.

Volume 174, Nmnber 1, Part 2 SPO Abstracts 383

Am J Obstet Gynecol

261 SKIN BLOOD FLOW RESPONSE TO STABLE ASPHYXIA IN THE

PREMATURE FETAL I~MB. P Bobb’~, MY Divon, E Yun*, A Santos*

Depts. of OB/GYN and Anesthemology, Albert Einstein College of Medicine,

BronX, NY OBJECTIVE: This study was designed to determine whether a prolonged period

of stable asphyyaa affects skm blood flow(SBF) m the premature fetal lamb METHODS: 9 chronically instrumented pregnant ewes were studied at 11 g-119 days of gestation After a contr01 period, fetal lind-base status was assessed and

regional blood flows were determined with dye-labelled mtcrospheres Moderate

fetal asphyraa was then reduced by partial umNhcal cord occlusmn and

maintained over 90 mmutes Evaluahon of cardiovascular and acid-base status

was then repeated Regional blood flows(ml/mm*g) to areas of skin over the fetal

vertex and breech were then compared Data were evaluated by Student’s t-test

RESULTS: mean±SD, *p<0 05 CONTROL ASPHYXIA

pH 7 43:~0 04 7 33-k0 03*

PO2 25±4 14±3 *

PCO2 41±4 48±2*

SBF-VERTEX 0 48x0 24 1 75±0 79*

SBF-BREECH 0 38~’0 23 1 29±1 23

CONCLUSIONS: Fetal skin is accessible for mtrapartum evaluation Prevlous

investigators have shown that blood flow to the fetal carcass (bone, skin and

muscle) increased m response to hypoxemm produced by umbdlcal cord occlusion of short (4.5 minute) duratmn~ In this study, umbdtcal cord compression which

produced a stable level of asphyma over a 90 minute permd resulted m a

stgnificant increase in SBF over the fetal vertex The mcrease m SBF to areas over

the fetal breech approached slgmficance In this small sample(p=0 07) These

results must be considered when uhllzmg SBF or any method oflntrapartum fetal

evaluation dependent upon SBF to demonstrate fetal well-being

XAmencan Journal of Phymology(19gT), v 257, H100

263 INTRA UTERINE FETAL ESSENTIAL FATTY ACID STATUS; COMPARISON WITH POSTNATAL VALUES OBTAINED AT COMPARABLE GESTATIONAL AGE (GA)., ~ M l,olelnan-v l)longeleu, M AI, (. llolustla, I Nic~ Inn. K Nlcolaldes Dept llum Blt~l, Lunbulg lhnv, M,t,,stllcht. The Nchcflauds OBJECTIVE Because a bcttm knowledge ul the phyMology matelnal-fetal essential ratty acM hanqe] ~s lelcvailt lOl lUdlellla]

status dmlug ollgOlllg gestation hy an,dyMug tile laity acid conlpositmu of phosphohplds llOnl plasma ohtamcd by letal hlool s,m~phng STYDY DESIGN: The study Colnp~xsed 86 total blood samples obtaltmd dmmg p~egnancy (18-37 weeks GA) and 26 uml~d~cdl b!ood ~amples. taken immediately allel p~ete~m bulb (28 5-37 week~ ( RESULTS: Tile total dlllOUll[ o[ phosphohlnd-as~oc~ated laity acids lU plasma d~d nol change during geMatmn The level ol Illlolelc acid (18 211- 6) shuwed a shghl increase (p=0 015) dmmg gcslatum Thus the decrease 111 letal hllulelc acid observed helo~e duHog [he lust t~m~estm

oi a~ach~doulc acid (2(I 4n-6) decreased with lct,d ~uahu,d~on (p<00I)l)l),

(22 6n-3) increased m,ukedly dmmg letal dcvelopn~enl [p<0 0001) When tomp,umg ll~e letal llltla-UleltUe value~ {(iA 28-37 weeks) wlih

neonatal lesults ol p~etmm ml,mts holu ,11 ,i Mll~llal (iA the lehd hnole~c acid status is lowe~ (p<O 003) than values ol plClelm IIl[.lll[s The letal ovmall EFA status ~s lowm (p<O 03) fl~.ul Ihat ol ptctetm ml,mls CONCLUSION: "Ihe "unb~nn" ielus ha~ a ~m~cwh,d [<~wm I;I:A status than the p~etmm infants dneclly alte~ bath al a comp,~ahle (;A llowevm,

dllleleuces ale obSClVed between lctal hlood samples taken at cesaleau sections and samples taken ,dim a vaginal bath Ikctctm btuu cht[dtetl do

Thus the Iuw EFA MdtllS at b~tth {,t pletclru llllalllS seems

physmlogical m~g,m

262 UMBILICAL CORD GAS ASSESSMENT IN HIGH RISK TWIN GESTATIONS. AF Borglda, F Eng*, JFX Egan, JF Rodis, JS Smeltzer, GW Turner, and WA Campbell. University of Connecticut Health Center, Farmington, CT. OBJECTIVE: To compare the differences in umbihcal cord blood gas (CBG) values m high risk twin gestations. STUDY DESIGN: Twins had CBG obtained at the time of delivery. Arterial (Art) and venous {Ven) CBG were analyzed for pH, pO2, pCO2, and HCO3. Birth order, pregnancy complications, delivery mode, and birth information were recorded for all twins. Data were analyzed as mean values and mean difference between twins A and B usingpaired t-test and ANOVA with p<.05 s~gnificant. . RESULTS: 110 sets of twins were evaluated. The mean gestational age at birth was 31.7 weeks, and the mean b~rth weight was 1688 grams. Pregnancy complications included: Eremature ruptured membranes-30%,preterm labor-45%, birth weight discordance-32%, preeclampsia-17%. Delivery was cesarean in 79%, vaginal in 11%, and combined methods in 10%. There were no differences in mean birth weights or median Apgar scores.

twin A twin B Art pH 7.30+ 06 7.30+.07 Ven pH 7.34+ 06 7.34+.06 Art pO2 18.8_+6.9 18.3_+7.3 Ven pO2 27.8_+9.5 25.3+9.5* Art pCO2 47.8_+7.5 48.0+9 6 Ven pCO2 41.4_+7.0 43.0+7.3 Art 1:1CO3 23.5_+2.6 23.1"+3.5 Ven HCO3 22.2"+2 5 22.8"+2.7

(All data mean _+ SD) *p=.04 Analysis by mode of delivery showed the Ven pO2 was significantly lower in twin B only when delivery was combined (A=29.8"+6.8, B=24.3_+3 1, p=.01), and the Art HCO3was less in twin B when both twins had a vaginal birth (A=23.0-+2.2, B=21.2_+3.1, p=.04). No differences were related to discordant growth or pregnancy complications. CONCLUSION: In this study, the largest of CBG in twins to date, many of the previously reported differences between twins A and B were not confirmed m our high risk population. The small differences found in this study were related to mode of dehvery.

264 MEcONIUM HAS NO L OR S BUT AFFECTS THE L/S RATIO. S. Logo, C. Towers, A. Stranss~, T. Asrat, R. Freeman, OB/GYN, Long Beach Memorial Medical Center, Long Beach, CA and U~ve~"s~ of California, Irvme, Orange, CA. OBJECTIVE: Although, there have been a few repots that

have evaluated the effect of meCouium on the L/S ratio for fetal lung malunty testing. To date, these studies have only assessed the L/S ratio of araniotic fluid contaminated with mecenium. The pmIxx~ of this study was to determine if meconinm by itself 1) produces an L/S ratio, 2) contains Lecithin and Sphiagomyelin, and 3) has values that are canstant STUDY DESIGN: A standard Thin Layer Chromatography (TLC) L/S ratio was performed on the f~st raeconium stool of 19 neonates between 31 weeks and t~rm. A quantitative assay was performed on a sample from each gestational age (7 samples, range 31 weeks to t~m) to con~mn the presence of Lecithin and Sphingomyelin RgSULTS: The 19 samples gave atypical TLC migratory patterns that had dots similar to the zones for Lecithin and Sphingomyelin. The presumed L/S values ranged from 1.1 to 3.6 with an correlation to gestattonaI age. However, the quantitative assay did not detect the presence of Lecithin or Sphingomyelin m any of the analyzed samples. CONCLUSIONS: Meconium does not contain Lecithin or Sphingomyehn but has an unidentified material whose migratory pattern using a standard TLC quahtative assay is similar to that of Lecithi~Sphingomyelin Therefore, the presence of mecaninm in anmiotm fluid may falsely raise or lower the L/S ratio and confuse fetal lung maturity interpretatton.

384 SPO Abstracts January 1996 Am J Obstet G)necol

265 CHRONIC ENDOTHELIN INFUSION IN THE RAT: EFFECTS ON REGIONAL PERFUSION AND PREGNANCY OUTCOME. L~G

Thacte,x M G Neerhof, R K Silver, and M S. Caplan x Departments of Obstetrics and Crynecology and Pediatrics, Northwestern Umverslty, Evanston Hospxtal, Evanston, IL.

OBJECTIVE: To evaluate the effects of chrome endothelm (ET-1) infusion m the tired-trimester on organ perfusmn and pregnancy outcome

m the rat STUDY DESIGN: Venous mid arterial catheters were placed on day 14 of gestataon m Sprague-Dawley rats. Suline or ET-I (0 2 or 0.5 nmol/kg/kr)

was refused mtravanously vta osmottc tmmpump from days 15 to 21 of gestation (term=22 days). Mean arterial pressure (MAP) was recorded on

days 15, 18, and 21 On day 21, radtolabeled rmerospheres were refused for evaluataon of organ perfusian Anesthesia was then admunsteled and a hysterotomy was performed Fetal and placental weights as well as the

incidence of stillbirths were recorded Gamma counts were obtained from

fl~e following samples for evaluataon of organ peffuaion: reference blood, placenta, uterus, kidney, and brain.

RESULTS: (Mean +_ SE) Saline# 02. nmol ET-1# 0 5 nmol ET-1# Organ Perfusion Placental perf (mYmuv’g) 2 2 £0 3 2 1 +_0 3 0 9£0 2* Uterineperf (ml/mm/g) 0.5£0.1 0.5£0.2 04+_0 l Kidney perf (ml/mm/g) 5.0£0.8 57+_06 35+_1 1 Brain peff (ml/rmn/g) 08+_02 10+_04 1.4+_0,1 Pregnancy Outcome MAP (day 18, mmHg) 114+_3 117+_3 133£6" Fetal weight (grmns) 4 64+_0.06 429+_0.04* 397£0.08* Placental weight (grams) 053+_001 050+001" 0.49£001" Fetal denth (#/litter) 060£022 121+_061 155£065

#n=10 for each group *p<0 05 vs Saline by ANOVA

CONCLUSIONS: Chromc ET-I infusion m the third trimester decreases placental perfusion resulting in impaired placental and fetal growlh

267 AGGRESSIVE UTILIZATION OF ANTENATAL STEROIDS IN VERY

LOW BIRTH WEIGHT INFANTS. Brian Sonta~, Marion StewarP, James

Mabel Cindy McEvo3:’, Susan Bowhng~,Gary Kletnman, Dept ofObstetrtcs & Gynecology, Dept of Pediatrics, Univ of Fl, Pensacola, FL

OBJECTIVES - We compared the incidence of steroid utlhzation in our

delmrtment to the nattonal average for steroid usage We examined the differences

between treated and untreated very low birth weight (VLBW) infants for several

outcome measures including hospital cost

STUDY DESIGN - The subjects were 174 infants weighing between 500 and 1250 grams hosp~talmed in our NICU from 1/92 to 12/94 Our department policy

Is to aggressively pursue co~costom~d therapy for lung maturatmn in pahonts at risk for preterm dehvery including attempts to delay delivery 48 hours whenever

necessary Slxtymfantswere fully treated and 21 wore partially treated pnor to

dehvery Continuous vanables were compared with student’s t-test, and categorical

variables were compared by X2 analysis.

RESULTS - Our 35% rate of completed treatment with Betamethasone is

significantly tugher than tbe pubhshed national average of 15-20% (p< 001 ) Birth

weight, gestatlonaI age, rare and gender were s~milar between treated and untreated

~nfants Steroid therapy was associated w~th increased survival compared to no

treatment 88 3% vs 68 8%, p= 005. Although the length of stay for surviving

infants was similar between treated and untreated infants, the cost per day was

sigmficantly lower in steretd t~eated infants, $1421/d vs. $1985/d, p <0 01 Steroid

treated infants had fewer severe (Grade 3 or 4) IVH 7 5% vs 21%, p< 05, higher

systolic blood pressure 55 vs 46 mmhg, p< 001, less need for surfactant therapy

41% vs 67%, p< 01, less need for dopamlne 2 1% vs 17.6%, p< 01, lower mean mrway pressure 2 4 vs 4 9 cm H~O and a lower peak F~O224% vs 34% at 24 hours

of age, p< 001

CONCLUSIONS - Although it is not possible to delay delivery m all preterm

mtknts to achieve optimal antenatal steroid therapy, by making a concerted effort

to treat whenever a pattant ts at risk for preterm delivery, it ts possible to improve

on the histoncal steroid utilization rate. VLBW infants who receive the benefits of antenatal steroid therapy show an improved cardiovascular and pulmonary status

Overall these infants have less morfadRy, lower mortality, and lower average

health care cost

266 UTERINE AND PLACENTAL ENDOTIIELIN (ET-1) RECEPTOR

SUBTYPES IN THE RAT: RELATIONSHIP TO FETAL AND

PLACENTAL GROWTH AND ORGAN PEP, FUSION. L.G Thaete,x S R.

Magnuson,x M G Nenrhof Dept of Obstetrics and Gynecology, Northwestern

Umversity, Evanston Hospital, Evanston, IL and Abbott Labs, Abbott Park, IL

OBJECTIVE: To investigate the impact of chronic ET-1 infusion on the

expresmon levels of ET-I receptor subtype (ETA & ETB) mRNAs m the uterus

and placenta of the pregnant rot, and to evaluate the relationshtp bet’aecn reenptor

mRNA status, uterine and placental perfusmn, and fetal and placental growth

STUDY DESIGN: Chrnmc indwelling arterial and venous catheters were placed

into 5 control and 5 experimental Sprague-Dawl~ rats on day 14 of 8estahon. Either saline or ET-1 (0,5 nmol/kg/hr) was infused intravenously vm osmotic

mm~pump from days 15-21 (term=22 days) On day 21, 57Co-labeled

miurospheres were infused for evaluation of organ perfus~on Anesthesia was

th~a admanstored and a hysterotomy was performed. Gamma counts were obtained from samples of reference blood, placentas, and uterus Uterine and

placental samples from 3 control and 4 ET-l-truated rats wore qmek-fi’ozen in

hqmd mtrogan for extzection of RNA~ Total RNA was ~solated, a Northern gel

electrophoresis and transfer were performed and 3~p-labeled cDNA probes

specific for ETa or ET~ mRNA were used to identify and quantify these species

RESULTS: ETa mRNA Is 12 times more abundant m the uterus than m the

placenta, and shov~ a slight (20%) decrease in response to ET-I infusion ET8

mRNA levels are s~mdar m the uterus and placenta and are unaffected by ET-I

infusion

Perfusfun and Growttt (Mean + SEM) Saline# 0.5 nmol ET-I # Plaenntal perfusion (mFmm/g) 2 24 _+ 0.25 0 93 L 0,21" Utenne paffuslon (ml!min/g) 050+012 0 40 + 0.08

Fetal weight (grams) 4 64 _+ 0.06 3 97 + 0.08"

Placental weight (gran~s) 0 53 + 0 01 0.49 + 0.01"

an=5 for each group *p<0 05 vs Saline

CONCLUSIONS: The gre~ad rat uterus contams 12 times more ETA receptor

than the placenta ET~ reenptor conc.¢ntrnttons in the uterus and plaeanta are

stmdar ChroniuinfasionofET-1 has little effect o. therelatwe¢oncentrationsof these r~¢ptors Vasocons~cfion of uterme amrioles, mediated at Inast m part by

ET-1, limits placental perfusion, and thereby placental and fetal growth

268 COMPARISON OF THE PATHWAYS OF ADENYLYL CYCLASE STIMULATION IN CULTURED HUMAN MYOMETRIUM AND MONONUCLEAR LEUKOCYTES. Yu-Li Liu~, Uchenna C. Nwosu, and Peter J. Rice×. Departments

of Pharmacology and Obstetrics & Gynecology, East Tennessee

State University Qu,llen College of Med=cine, Johnson C~ty, TN.

OBJECTIVE: Th~s study compares actwat~on of cAMP productton

in cultured human myometrium and peripheral blood mononuclear

leukocytes, another model for the f~-adrenoceptor-cAMP system.

METHODS: Human myometnum samples obtmned at cesarean

delivery or hysterectomy were cultured ~n DMEM/F12 with 10%

FCS. Leukocytes from healthy males were studied fresh and

mmntatned for 3 days in RPMI1640 with 20% FCS. cAMP

production was measured by RIA under basa~ conditions and

following stimulation by NaF/AlCl3 (10mM/10pM), Forskolin

(24pM), or PGE1 (10pM).

RESULTS: cAMP production tn leukocytes was =20-fold lower

than m myometnal cells. In both tissues cAMP production was

unaffected by G-protein stimulant NaF, but was maximally

enhanced by PGEp ForskohrL which directly activates adenylyl

cyclase, enhanced cAMP production maximally in myometnum,

but not ~n leukocytes maintained m culture. Stimulation of cAMP

production was unaffected by 3-day exposure to a high clinical

concentrations of tocolytic ~-agomst terbutahne (0.2pM).

CONCLUSIONS: Leukocytes should be used cautiously as a

model of the less accessible uterine 13-adrenoceptor-cAMP

system. Exposure to high climcal concentrations of terbutahne

does not affect the myometrium cAMP production system

beyond the f~-adrenoceptor. (This study was supported =n part by

a grant from the ETSU Research Development Committee.)


Am,] Obstet Gynecol

269 ENDOTHELIN-1 INCREASES PRETERM OVINE FETAL URINE FLOW SY

STIMULATION OF ATRIAL NATRIURETIC FACTOR (ANF) SECRETION.

L.K. Kullamax, M.E.G. Sdvax, L.Dayx, M.G. Ervinx, and M.G. Ross. Dept

Ob/Gyn Harbor-UCLA Mad Ctr, Torrance, CA.

OBJECTIVE: Endothehn-1 lET-l) m present m high concentrations ~n

fetal plasma and ET-1 ~nfusion (25 ng/kg/min) to nearterm (131 d)

ovine fetuses increases arteria~ blood pressure, plasma ANFand urine

flow. ET-1 induced ~ncreases ~n fetal urine flow have been primardy

attributed to pressure diureals. As ANF renal responsiveness is greater

in preterm vs. nearterm fetuses, we hypothesized that preterm fetuses

would exhibit ET-1 mediated diuresis primarily via ANF secretion. STUDY DESIGN: Six chromcally catheterized preform ovine fetuses

1116 ± 1 d) were continuously monitored for arterial blood pressure and heart rate and fetal urine flow during sequential 60 min control,

Intravenous ET-1 infusion (25 ng/kg/mln}, end recovery periods. Five

time-control ammals received equivalent infusions of O. 15 M NaCI.

RESULTS: In response to ET-1 infusion, fetal plasma ET levels

(109± 16 to 204±35 pg/ml) significantly increased, though there was

no change m fetal arterial blood pressure, heart rate, pH, pO=, or pCO2.

There was a significant linear correlation between ET-1 induced

changes in fetal urine flow and plasma ANF levels (r =0.92, p<O.01),

with 3 of 6 fetuses demonstrating increased plasma ANF (99± 19 to

204 ± 52 pg/ml) and urine flow (O. 11 ± 0.02 to 0.35 ± 0.09 ml/kg/min).

Fetal urinary sodium (Na; 46±6 to 86± 13 mEq/I) and chloride (34±4

to 61 ±9 mE1/I) concentrations, and fractional Na excretion (7.4:t: 1,5

to 11.6±2.7%) significantly increased, while GFR did not change.

Plasma ET-1, ANF and urine flow did not change in control fetuses,

CONCLUSIONS: (1) ET-I reduces preterm fetal urinary diuresis by

stimulation of ANF secretion, and (2) ET-1 ~nduced ANF secretory

responses develop prior to vasopressor responses.

271 ANGIOTENSIN II (AII) INDUCED UTERINE VASODILATION IS MEDIATED BY AII TYPE 2 (AT-2) RECEPTORS VIA NITRIC OXIDE IN NONPREGNANT SHEEP D S Lambers~, S.G Greenberg~,

K E Clark~ Dept Ob/Gyn Umv. of Cincinnati, Clnclnnatl, OH OBJECTIVE To determine ff Anglotensm II-type 2 (AT-2) receptor stimulation causes vasodllatlon in the nonpregnant uterine vasculature STUDY DESIGN. Uterine vasoconstrictor responses to AII are medmted

through AT-1 receptors m nonpregnant ewes (SGI 1995) and are s~gmficantly blunted rn the presence of AT-2 receptor stlmulatmn possibly due to the

release of a vasodflator To investigate th~s, mean arterial pressure (MAP),

heart rate (HR), mad uterine blood flow (UBF) were measured and uterine vascular reststance (UVR) was calculated m eight nonpregnant ewes S~nce no AT-2 agomst currently exists, we studmd the effect of AT-2 stlmulatmn by AII m the presence of AT-I blockade (L158809) Systemm and uterrne hemodynam~c responses to intrauterine artery (IUA) infusmn of AII (0 03 ug/m~n for 10 ram) were recorded before (basehne) and after IUA lnfusrons of L158809 (3 0 mg/rnin for 5 turn) To determine the mechamsm of the observed vasodflatmn we tested mhib~tors of prostaglan&ns 0ndomethacrn,

2 mg/kg l.V ), mtnc oxide synthetase (NOS, L-mtroargme methyl ester, L- NAME, 10 mg IUA) and AT-2 receptors (PD123319, 3 mg/mln for 5 mln ) RESULTS: As expected, IUA infusron of All (0 03ug/mln) decreased baseline UBF by 64+_3% and increased UVR by 206+35% However, after

AT-1 receptor blockade w~th L158809, AII infusion increased UBF by 172+51% (p<0 007) and UVR actually decreased by 40+7% (p<0.0001) AII- induced vasodflation was reversed by IUA PD123319 admlmstratmn, and

could be attenuated by pretreatment wrth PD123319 Indomethacln had no effect, while L-NAME also reversed the vasodfiatron. No s~gmficant changes were observed in any systemic parameters CONCLUSION: These data suggest that the vasodrtatron seen w~th AII m the presence of AT-I blockade is mediated by the AT-2 receptor through stimulatmnofNOS SmceNOS as elevated in pregnancy, st~mulatmnofAT- 2 receptors rn the pregnant ovrne uterus may m part explain the mechanism of uterine refractoriness to AII Supported by HL-49901 and HL- 52280

270 INCREASED FETAL COLONIC MUSCLE CONTRACTILITY FOLLOWING GLUCOCORTICOID AND THYROXINE (T4) THERAPY: IMPLICATIONS FOR MECONIUM PASSAGE. B. Ross, K. Bradley, L. KullamaX, M.J.M. Nqland’, M.G. Ross. Harbor-ULCA Med Ctr,

Torrance, CA. OBJECTIVE: The incidence of meconium stained amniotic fluid increases with advanced gestat~onal age and fetal stress, and meconium passage ~s hkely dependent on fetal colonic muscle maturation. Antenatal steroid and/or T4 exposure improves fetal pulmonary and cardiovascular function. We hypothesize that =n utero maturational agents effect an increase m fetal distal colonic muscle contractdity. STUDY DESIGN: In a randomized controlled study 126 d (term 145 d) ovine fetuses were treated with ultrasound guided intramuscular injections of 0.5 mg/kg betamethasone (n = 5), betamethasone plus 50 #g/kg thyroxine (n=5), or saline (n=7). After 48 h, fetuses (128 d) were delivered, distal colon segments were removed, and peak tension responses to bethanechol (1 O* to 10aM) characterized in in vitro organ baths RESULTS: Peak muscle tensions were significantly greater m combined betamethasone and thyroxine treated fetuses (989 ± 190g/cm2) than in betamethasone alone (559 ± 75g/cmz) or the saline treated animals (509 + 9 lg/cmZ). The bethanechol ED~ values (2.1 ±0.5x10"s M) were not different among the 3 groups. CONCLUSIONS: Antenatal fetal betamethasone and thyroid hormone treatment increases fetal colonic muscle contractility. We speculate that endogenous or exogenous fetal maturational agents may potentiate the passage of meconium.

272 EFFECTS OF ESTRADIOL-17J~ (EzJ~l TREATMENT ON UTERINE ARTERY [UA)

ENDOTHELIAL NITRIC OXIDE SYNTHASE (NOS) EXPRESSION. RR Mannessx, CE Shawx, TM Phernettonx, IM Btrdx 0apt QhIGyn Perinatal Research Labs, Univ of

Wisconsin, Madison WI.

OBJECTIVE: Exogenous Ez[~ treatment dramatically increases uterine blood flow (UBF)

and reproduces the systemic cardiovascular adaptations of normal pregnancy (Magness et

al. 1993). Since Ez[3 also increases UA N0S specific actlvtty, we hypotheuze that Ez[3

will locally augment expression of constitutive NOS, specifically in UA end0thelium,

STUI1Y OESI6N: N0npregnant ewes (n-14), received Vehicle (16% Et0R in sahne iv) or

E~[3 (luglkg). Venous plasma was obtained at 0, 90, & 120 min for measurement of

cGMP, which Is produced in response to NO; UA and omental (systemic) arteries were then

obtained. Using Western immunoblot analys=s/ECL detection, endothelial cell N0S [ecNOS)

and neuronal N0S (nNOS) constitutive is0f0rm expression were evaluated in intact and

denuded (vascular smooth muscle; VSM) arteries, as well as in endsthelium-enriched

proteins. The latter was obtained by isolating the tunica intlma into "Western lysls buffer".

RESULTS: UA express ecNOS, but not nNOS. ecNOS expression was ~abzed only to the

end0the0om, i.e. the intensity of the observed endothebum-eeriched ecNOS protein hands

was as much as 700-800 fold greater than that observed in intact UA; expression was not

observed in VSM. Cellular lecahzation and specificity of the ecNOS constitutive ~soform

expression also was conhrmed by ~cytochemistry. Ea~ given at a dose wNch

induces maximal steady-state rises in UBF by 90-120min, appeared to elevate ecNOS

expression in UA (48%), but not systemic artery endothelium, VSM NOS expression was

unaltered by EzJ~ Plasma cGMP levels (11_+ 1 pmul/ml) were unaltered by Vehicle (-8 _+

7%), hut were increased by 34 _+ 13% (p < 0 015) 90-120 rain post-Ez~ injection. In other

studies, unilateral mlection of EzJ3 (3ug) directly into the uterine circulation locally

increased UBF and ecNOS expression, but only in the ipsilateral uterine horn.

CONCLUSIONS: 1) The constitutive isoform of NOS expressed in UA endothehum, but not

VSM, is ecNOS; nNOS was not detected; 2) exogenous EzJ~ locally increases ecNOS

expression in association with increases in plasma cGMP levels. These data support the

hypothesis that endothelial ecNOS protein expression specifically mediates the EzJ~-Induced

increase in ORF and possibly other systemic cardiovascular edaptatmns of pregnancy.

Supported by NIH HL 49210 and HD 33255.


Am.] Obstet Gynelol

273 REPEATED USE OF BETAMETHASONE: EFFECTS ON ADRENAL FUNCTION AND BIRTH WEIGHT IN THE RABBIT L Pratt, RR Magness~, SK Hendncks, DH N:~bo~,T Phernettonx,IM Birdr Dept OB/GYN, Univ of Wisconsln, Madison, WI

OBJECTIVE: To determine whether repeated matornal treatment with

betamethasone (BETA) ~ fetal lung matunbj suppresses fetal or maternal adrenal

f~nct]on or affects b~rth weight (BW) in rabbits

RESEARCH DESIGN: 35 t~me-bred rabbits (tarm=31d) were assigned to five

groups no treatment, sahne (0 2ml IM), or 1, 2, or 3 courses of BETA A BETA

course = 0 lmg/kg IM q24h x2. BETA inject=ons began on d19&20 (1, 2, and 3

courses) and were repeated on d21&22 (2&3 courses) and d25&26 (3 courses) At

sacrifice (d27) sera for cortlsol levels (EIA) and adrenals for 17c~-hydroxylase

(I7OH) mRNA expression (Narthom analys=s) were collected Data were analyzed by ANOVA (cor~se~, BW) and t-test (17OH)

RESULTS: Are expressed as % of maternal saline con~’ol (~ISD) I sahne 11 ~oulse 12 coulses 13 courses

mat corhsol 166 9(42 8)~. 100+ 93 4(93.7}+ 55 2184 3)# 1(0 8)" fetal cortlso128 0(7 0)+ 30 4(19 5)+ 26 9(24 8}+ 23.9(21 0)÷ 0(0 98)# rnat 17OH 17 5(47 7)-~ 100+ 79 6(4 4)+# 43 4(19 0)#" 12 0(20 3)" letat17OH 571(477)+ 505(276)+504(159}+250(246)# 2.1(32)" BW (gms) 31 4(3 0)+ 31 8(4 2)+ 30 6(2 8)+ 26 6(3 8)# 19 6(3 4)^

Means w~th the same symbol are not s~gn~ficantly d=fferent from one anoth~

within the same row Symbols which differ within the same row d~note

s=gnllicance at p<0 05 Maternal cort~sel levels were unaffected by 1 course of

BETA, declined 50% with 2 courses and were suppressed by 3 courses. Fetal cortisel levels were suppressed alter 3 courses Expression of the mRNA

encoding 17OH, the rate hmltmg enzyme for cortisol synthes~s, paralled these results BW was adversely affected following 2 and 3 courses.

CONCLUSION: Whereas a single course of BETA does not affect fetal growth ar

suppress adrenal funcf~on, repeated use of this drug results m complete suppression of maternal and fetal adrenal funct=on and ~mpacts fetal size We are

now exploring the pess~ble relat=onsh=p between timing of repeated BETA use relative to negative feedback Induced adrenal suppression

275 CHRONIC ENDOTIIELIN INFUSION IN TIlE PREGNANT RAT: HEMODYNAMIC, BIOCHEMICAL, AND PLACENTAL HISTOLOGIC EFFECTS. L.G. Thact¢,x M G. Ncerho~ R.K Silver, M.S Caplan,x and A G. Bmdin x Depar~mants of Obstetrics and GFa~,ology, Pcdm~ics, and Pathology, Northwestern Umvcrs~ty, Evanston Hospital, Evanston, IL OBJECTIVE: To determine ff chroinc endothelin (ET-I) uffusion producas any of the hcmodynamic, biochemical, or placental h~stologic characteristics described m HELLP syndrome m the human or with mtnc oxide synthas¢ antagonism in the rat. STUDY DESIGN: Chronic indwelling arterial and venous catheters wcr~ placed on day 14 of gesta~on m Spraguc-Dawloy rats. Salme or ET-1 (0.2 or 0.5 nmol/kg/hr) was infused intravenously via osmotic minipump from days 15 to 21 ofgestauon (term=22 days). On days 14, 18, and 21 of gestatmn, mean arterial pressure (MAP) was measured On day 21, arterial blood was obtmned for a CBC, tiv¢~ fiu~inn ~sts, and blood gas analysis and a hys~otomy was performed In 4 rats treated with 0 5 nmol ET-1 and in 3 conh’ols, a uterine hera was removed and fixed en block for hislolo~c evainatmn

RESULTS: (Mean + SE) Saline# 0.2 nmol ET-1 # 0 5 nmol ET-1 # MAP (day 18, mm Hg) 114+_3 117+_3 133+_6" Hematocrit (%) 27.6 + 1 1 31.1 + 0.8 30 5 + 0.9

Platelet Count (103/ec) 828 + 36 728 + 57 789_+ 40 SCOT (IU/liter) 145 +_ 20 91 +_ 16 98 +- 25 SGPT (1U/liter) 59 + 15 54 _+ 10 49 + 8 Arterial pO2 (ram Hg) 90.5+-3.5 92.2+3.7 86.2+8.6 Arterial pH 7.42+-0.03 7.41+-001 7.41+0.02

#n=10 for each group *p<O.05 vs Saline by ANOVA Placantal hist~lo~ of rots treated chronically with ET-I exhibited pcnvascular inflammation, but there was no ¢vidanca of necrosis or infarction. CONCLUSIONS: Chronic ET-I infusion does not produce the biochemical characteristics of HELLP syndrome. Maternal hypertansion is olr~erved only with high-dose infusmn ET-I mfusmn does not produce the placental necms~s and mfamt~on described with chronic nitric oxide inhibition in the rat.

274 Inhibition of lnterleukln-10 During Pregnancy Results in Neonatal Growth Retardation. A. Riihsinghani, S. Bhatlax, K. Thompsonx, L.

Tygrettx and T. Waldschmidt~. Depts of Ob/Gyn and Pathology, Univ. of Iowa, Iowa City, IA.

OBJECTIVE: 1L-10 is a major immuno-rogulatoxy cytokine secreted at the maternal-fetal interface. Preliminary studies suggest it may play an important role in reproductive fitness. We hypothesized that neutralizing

IL-10 with an IgG antibody would lead to pregnancy losses and alter T- cell immunity. STUDY DESIGN: Pregnant Balb/c mice were randomized to receive

clther anti-IL-10 antibodies or saline every other day starting day I0 of gestation. Six litters wcrn studied for fetal outcome. A litter of eight

neonates exposed to anti-IL-10 in utero was compared to four litters of untreated control mice. T-cell development was studied in the anti-IL-10 treated mothers by examining the major T-cell subsets based on CD4 and

CD8 and the early precursors based on CD44 and CD25. RESULTS: Prolonged deprivation of IL-10 did not alter intrauterine growth or pregnancy outcome. However, all neonates born to mothers

given anti-IL-10 during pregnancy showed evidence of growth retardation by weaning.

Fetus/Neonate AntML-10 RX Control P value 18 d fetus wt. (g) 0 98 =k 0.3 0.84 + 0.2 NS CRL (cm) 2.03 :t: 0 27 1.96 + 0.2 NS Litter size (n) 6.8 :t: 2.3 6.4 + 1.9 NS 4wks neonatewt. (g) 7.3 :t:0.9 9.4 +1.4 <0.0001 5 wks neonate wt, (g) i1,0 + 1+1 14.0 ± 1,0 <0.000l 6 wks neonate wt. (g) 14.7:1:1.4 16.8 + 0.8 0.01 No significant differences were observed in maternal thymic involution

and T cell subsets between anti-IL-1O treated and control pregnant mice.

CONCLUSIONS: There is a significant correlation between deprivation of IL-10 in utero and neonatal growth and development.

276 PROPRANOLOL INHIBITS PHOSPHATIDYLCHO- LINE (PC) SYNTHESIS BY ALVEOLAR TYPE-II CELLS I.L. Kimx. F.H.C. Tsaox and C.B.Martin, Jr. Depts. of Ob-Gyn

and Pediatrics, University of Wisconsin, Madison, WI. OBJECTIVE: 13-adrenergic antagonists inhibit the secretion of pulmonary surfactant, but an effect on surfactant synthesis has not been clearly demonstrated. Therefore, we investigated the effect of g-adrenergic blockade with propranolol on surfactant synthesis by alveolar type-II epithelial cells. STUDY DESIGN: In vitro study using isolated adult rat alveolar type-II epithelial cells in culture. In short incubation experimentS, 3H-choline (a PC precursor) and test drug were added to the culture after an 18-h preincubation, and the cultures incubated a further 2 h. In other experiments the test drugs were added during the primary incubation (18 h). After the cells were washed, 3H- choline was added for an additional 2 h incubation period. PC synthesis was determined by quantification of 3H-PC.

RESULTS: Propranolol (50 pM) decreased PC synthesis to 75% of control after 2 h (P=0.01), and to 50% of control after the longer incubation (P=0.0048). The depression of PC synthesis was also dose-dependent over the range 1 to 50 ~tM. Addition of dexamethasone (0.1 pM) to cultures with 50 I.tM propranolol restored PC synthesis to control levels. CONCLUSIONS: These data demonstrate that propranolol inhibits PC synthesis by alveolar type-II epithelial ceils at a cellular level in a time- and dose-dependent manner. This inhibition is reversed by dexamethasone. These observations may have clinical relevance in the management of pregnancies complicated by maternal hypertension, when preterm delivery is contemplated.

Volume 174, Number 1, Part 2 SPO Abstracts 387 Am ] Obstet (;ynecol

277 PREGNANCY HAS A PROTECTIVE EFFECT AGAINST N-

METHYL-D-ASPARTATE-INDUCED~ SEIZURES IN RATS CA

Standley x, DB Cotton Department of Ob/Oyn, Wayne State Umversity/Hutzel Hospital, Detrmt, MI

OBJECTIVE: There exasts a panc~ty of data regarding the effects of

pregnancy on secure potentml We examined seizure act~vtty using the

eonvulsant N-methyl-D-aspartate (NMDA) m pregnant versus

nonpregnant rats

STUDY DESIGN 74 Long-Evans rats were anesthetized with

pentobarb~tal sodmm and a bipolar recording electrode was unplanted

mto the dorsal hippocampus, while a cannula was set into the lateral

cerebral ventricle. One week later, rats were mated, while others served

as nonpregnant controls Nonpregnant and pregnant rats (GA=20 days) were randomized to receive no drug or a single inject~un of NMDA (5,

10 or 20 gg) through their mdwellmg cannulae (9-12 rats/group)

Seizures were thereafter assessed for 20 minutes Pups were counted

and wmghed following delivery. Adult brains were processed for

histology Data were analyzed with ANOVA and Student’s t-test,

RESULTS: Total seizure duration and total number of seizures were

sxgruficantly reduced m pregnant versus nonpregnant rats, especially at

the 10 and 20 gg doses ofNMDA (p<0.05, respectavely). Onset to

seizure activity was not significantly affected by pregnancy. The

number of healthy pups at postnatal day 3 tended to be reduced in

mothers injected with higher doses of NMDA

CONCLUSIONS: These data demonstrate that seizure actwity

Involving NMDA receptors ~s reduced in pregnant rats compared to

nonpregnant rats. Usmg autorad~ography, we have previously shown

that NMDA receptors are altered during pregnancy. We therefore

suggest that pregnancy affords some protection against seizures induced

by an aetivatton of NMDA receptors m the brain

279 FETAL GROVv’TH RESTRICTION: REVERSAL BY DIETARY L-

ARGININE IN A RAT MODEL. KB Harvey-Wdkesx, RJ Vosatka.

D~vismn of Newborn Medicine, New England Medical Center/Tufts

Umvers~ty School of Medicine, Boston, MA.

OBJECTIVE: Nitric oxMe, a potent vasoddator, has been proposed to

have a role m regulating the uteroplacental c~rculation. Nitric oxide is

synthesized from L-arginme. We have studied the effect of increasing

the dietary supply of L-arginlne on fetal weight ~n a rat model of fetal

growth restrtet~on We hypothesized that ~ucreased dietary L-arg~mne

would lead to increased synthesis of nitrm ox~de Nitric ox~de could

then increase uteroplacental blood flow, thereby improving oxygen and

nutrient dehvery to the fetus and improving fetal growth.

STUDY DESIGN: 31 pregnant rats were exposed to hypobanc

atmosphere (380 tort) between fetal day 9 and 21 (term=22 days). 13

of the hypoxic rats had 2% L-argmme added to their drmking ~vater

(HYP-LARG). 3 had 2%D-arginine added (HYP-DARG) 15 were not

supplemented (HYP-H20). 11 pregnant rats served as room air controls

(RA). Fetuses were delivered by hysterotomy on day 21.

RESULTS: HYP-H20 fetuses were smaller than RA fetuses (HYP-

H20): 338+0.70g, n=101 fetuses, RA5 11+0.75g, n=65 fetuses,

p<0.01) HYP-LARG fetuses 94 78+0 56g, n=52 fetuses) were larger

than HYP-H20 fetuses (p<0.05) and not different from RA fetuses

There was no d~fference m protein retake between HYP-ARG mothers

and HYP-H20 mothers (HYP-ARG 9 49+0.92g, HYP-H20 9 55+0 92g)

The influence of arg~mne on fetal weight was stereospemfic HYP-

DARG fetuses were smaller than RA fetuses (HYP-DARG’4 32+0.83g,

n=31 fetuses, p<0 05 vs RA).

CONCLUSIONS: L-arglmne is a specific regulator of fetal growth,

possibly as a precursor of nitric ox~de

278 THE EFFECT OF MAGNESIUM SULFATE INFUSION ON THE ECG

IN PREGNANT WOMEN. M.E. Caine, R L. Thomas, D.H. MacKayx,

K. Comportx, Department of Obstetrics and Gynecology. The Western

Pennsylvania Hospital, Pittsburgh, PA.

OBJECTIVE: Magnesium sulfate 0VIgSO4) is infused intravenously (IV) to

treat preterm labor and as seizure prophylaxis for preeelampsia. MgSO4 acts

by slowing or blocking neuromusoular and cardiovascular conducting system

transmission. Past studies in animals have shown prolongatioh of the QT

interval with MgSO4 administration. Prolongation of rate corrected QT interval

(QT,) has been associated with malignant ventricular arrhythmias ~n humans

ECG (electrocardiogram) effects of MgSO4 in pregnancy have never been

reported. This study was designed to evaluate ECG changes in pregnant

women undergoing IV infusion of MgSO~.

STUDY DESIGN: Sixty-one ECG rhythm strips were evaluated in ten

pregnant women undergoing MgSO~ infusion for complications of pregnancy.

Baseline ECGs were performed prior to MgSO~ infusion and then hourly. PR,

QT, and QRS intervals, serum electrolytes, MgSO4 levels, maternal heart rate,

and maternal arrhythmia were recorded. QT was corrected for maternal heart

rate (QT,). Statistical analysis was performed using repeated measures analysis

of variance. (P < .05 was censidered statistically significant).

RESULTS: Mean MgSO~ levels obtained after one hour reflected a steady

state A slgnifieant iucrease was noted from baseline (1.72 meq/L vs 6.35

meq/L, P < .001). A significant prolongation of the QTo interval from

baseline during IV MgSO~ was noted. Baseline QT, = .4471s, steady state

QTo = .4845s(P < .001). No signifieantdiffarencein other parametarswas

noted. Two arrhythmlas were identified during IV MgSO,, premature atrial

contractions and trigeminy. Neither was considered dangerous.

CONCLUSIONS: TheQTo intervalduring administrationoftherapantiedosos

of IV MgSO~ is significantly prnlonged from baseline. Prolongation oftheQT,

interval did not result in a dangerous arrhythmia in this pilot study. ECG

monitoring during IV MgSO~ infusion is probably unnecessary.

280 A PROSPECTIVELY RANDOMIZED TRIAL OF TWO SOLUTIONS

FOR 1NTRAPARTUM AMNIO1NFUSION: EFFECTS ON FETAL

ELECTROLYTES, OSMOLALITY AND ACID-BASE STATUS.

E. Pressmanx K. Blakemore. Dept. of Gyn/Ob, The Johns Hopki~ Univ. Sch.

of Meal. Baltimore, Md.

ObJECTIVE: To compare the effects of intrapartum amnioinfusion with

normal saline (NS) vs. lactated Ringer’s solution plus physiologic gluenso

(LR+DtoW) on neonatal electrolytes, glucose, osmolallty, lactic acid

concentration, and acid-base balance.

STUDY DESIGN: Patients undergoing amnioinfusion for variable fetal heart

rate decelerations, oligohydramnios or thick meconium staining of the arnniotic

fluid were randomized to receive NS or LR+DtoW at standardized

amnioinfasion rates. Data were collected prospectively on maternal

demographics, course of labor and maternal and neonatal outcomes. Artarial

cord blood was obtained for analysis of electrolytes, glucose, oamolallty, lactic

acid and blood gases. Laboratory persounea were blinded to the solution used

for amnioinfusion. Control subjects with normal fetal heart rate patterns, dear

amnlotic fluid and not receiving aranioinfusion were studied concurrently.

Statistical analysis was performed using an analysis of variance, t test, Fisoher’s

exact t~t or Pearson’s correlation where appropriate.

RESULTS: Data were collected on 59 patients (21 NS, 18 LR+DtoW and 20

controls). No significant differences were noted in mammal demographics or

neonatal outcome. The indications for amnioinfusion, duration of labor and

volume of amnioinfusate were similar in the NS and LR+DtoW groups.

Cesarean sections were performed more often in the amniciafusion groups

(33.3% for NS, 38.9% for LR+DIoW) than in the control group (5.0%), p <

0.05. Cord arterial electrolytas, glucose, osmolallty, lactic acid and blood gases

were not altered by amnioinfuslon with either solution.

CONCLUSION: Intrapartum amnioinfusion with either NS or LR+DtoW has

no effect on neonatal electrolytes, glucose, eamolality, lactic acid or acld-base

baltmce.


281 MATERNAL MORTALITY AT A TERTIARY CENTER WITH CRITICAL

CARE OBSTETRICS. S RiaP.kwellX~ MW Tomhnson, B Gomk, BA Mason,

JE Wh=tty, DB Cotton Dept Ob/Gyn, Hutzel HospANayne State Umv,

Detroit, MI

OBJECTIVE: In contrast to nationally reported data, anecdotal experience

suggests that maternal mortality (MM) etiologies may d~ffer in an inner sty

tert~ar’j care center w~th an indigent population This study reviews a single

institution’s 5 yr expenence to better define th~s Issue

STUDY DESIGN: Maternal deaths from 1990-94 were ~denhfied from an

established pennatal data base, Charts were reviewed for demographics,

prenatal care, and underlying medical conditions VVhen autopsy data was

unavailable, cause of death was assigned chn~cally

RESULTS: 17 deaths occurred

~n 43,083 b~rths (MM rate

39 5/100,000 b~rths) Causes of

death are shown Mean age was

31 yrs, 88% were African

Ameocan, and 24% had

insurance 88% had some

prenatal care Underlying

medical cond~t=ons were present

m 65% 59% presented

antenatally, and 35% expired c.,e~== EmbeS= a*p=~= ~rm C~

wRh~n 24 hrs of adm=sslon Adult

respiratory d~stress syndrome

was the proximate cause of death ~n all seps~s cases

CONCLUSIONS: Card=ac d~sease accounts for a disproportionate number

of maternal deaths in th~s setting Underlying d~sease ~s a s~gmficant

contributing factor TradRmnal etiologies such as hemorrhage,

hypertens;on (HTN), and septic shock have a lesser ~mpact

283 EXCESS RISK OF FETAL DEATH IN BLACKS VARIES BY GESTATIONAL AGE. SN Wall~and JC Collin~. Dept. of

Pediatrics, Northwestern Univ Med School, Chicago, IL. OBJECTIVE--Fetal deaths, as well as other adverse perinatal

outcomes, are more common in pregnancies of black women compared

to white women. While socioeconomic disadvantage is associated

with adverse perinatal outcomes in blacks, the etiology of the racial

disparity in fetal mortality is not well understood. To determine if the

excess risk of fetal death in blacks varies over the course of gestation,

we analyzed national databases of live births and fetal deaths. METHODS--Using 1986 live birth and fetal death data from the

National Center for Health Statistics, the fetal mortality rate (FMR),

calculated as fetal deaths per 1000 live b~rths (LB), was determined for

black and white women. At each gestational age (GA) between 20 and

42 weeks, FMR was calculated as the number of fetal deaths divided by

all live births at the same or later gestational ages.

RESULTS--There were 3,309,818 singleton live births and 21,829

singleton fetal deaths to black or white women between 20 and 42

weeks’ gestation. Overall FMR for blacks and whites respectively was

11 and 5.7 deaths per 10O0 live births. The black and white FMR at

each GA (Fig. 1) and the relative risk (RR) of fetal death in blacks

(black FMR divided by white FMR) at each GA (Fig. 2) are shown

below.

Conclusions--The increased fetal death risk in black women is

highest in the second trimester and decreases as pregnancy approaches

term.

Figure I Figure 2

~1"~, --Black __ ~. ~31 -

,

Gestatia~d Age (weeks) Gestational Age (weeks)

282 GRADED-COMPRESSION ULTRASOUND(GCUS) FOR EARLY RECOGNITION OF APPENDICITIS IN PREGNANCY. JB Landwehr~, MR Leonardl~, DR Bryant~, SC Johnson~, and SF Bottoms Wayne State University, Hutzel HospRal, Detroit, MI.

OBJECTIVE: To determine the utd~ty of the GCUS of the appendix as an adlunct in the early recognlbon of acute appendicitis in pregnancy STUDY DESIGN: Retrespect~ve chart analysis was performed on all pregnant patients having a GCUS of the appendix at our mstltu~on

between the years 1991-95 All patients had vague abdominal pare with a low to moderate clinical suspicion for appendicitis Dlscrlmlnant function

analysis (DFA) was used to dehneate the most useful predmtors for acute appendmlbs RESULTS: Mean gestatmnal age at the t~me of graded-compression U/S was 20 1 + 8 2 weeks Graded-compression U/S successfully predicted appendicitis In 4 of 5 patients (80%) The sensltivRy, specificity, and negative predictive value were 80%, 97%, and 97%, respecbvely The results of the DFA were as follows’

Mean values Value appendicitis (n=5) no append[Qtis (n=43) p GCUS(+) 80% 2 6% <0 001 Temperature 37 3 37 5 NS VVBC 17 9 13 2 NS Constipation 0 0% 30% <0 01 Nausea/vomiting 100% 65% NS Diarrhea 67% 30% NS Anorexm 100% 33% NS Abdominal pain 80% 98% NS Fevedchdls 100% 45% NS

CONCLUSION: Graded-compressmn ultrasound is a useful adjunct in the

early recognition of acute appendlclbs In pregnancy, perhaps allowing for earlier surgical intervention in patients with a low to moderate suspicion for appendicitis GCUS proved to be far superior than any of the other chn~cal signs or symptoms ~n predlc~ng appendicitis In th~s selected group

of patients

284 TI~ SllV~LlrtvJJ A(IIIE ~ ~ ~ ID ACCURATELY PREDICTS MORTALITY IN AN OBSTETRIC ICU POPULATION. ~x, BM. SibaL and WC Mabie. Department of ObsteU’ics and

Gynecology, University of Tennessee, Memphis, TN. OBJECTIVE: Severity of illness elasalfication systems have been validated in multidisciplinary intensive care units(ICUs) and in various disease states. Little is known regarding their applicability in an obstetric

ICU. SAPS II represents the easiest of all ICU severity systems to use for obtaining the probability of hospital mortality. Our objective was to

determine the applicability of SAPS II in predicting maternal outcome in a critically ill obstetric population. STUDY DESIGN: We evaluated the validity of SAPS H scores in 251 patients with various complications managed in our obsteta’ic ICU. SAPS ]I scores were calculated using the 17 variable SAPS II scoring sheet which includes 12 physiologic variables in addition to age, type of admission (scheduled surgical, unscheduled surgical or medical complication) and three underlying disease variables (AIDS, metastatic cancer, and hematologic malignancy). The worst values within the first 24 hour observation period in the ICU were recorded. This score was then converted to the probability of hospital mortality and compared with the actual mortality yielding a mortality ratio. A Receiver Operating Characteristic (ROC) curve was used to derme the SAPS lI score that was best predictive of outcome. RESULTS: ROC analysis reveals that a SAPS H score of 38 or greeter has a sensitivity of 86% and a specificity of 97.% The positive predictive value of this score is 63% and the negative predictive value is 99%. In addition, when a SAPS H score of 38 is used, the mortality ratio (actuel/predicted mortality) is 1.01" which is not statistically different from the expected mortality ratio of 1.

OB patients No SAPS II Predicted Actual ] Mortality score moaalitv (%1 moa.4itv (%1 ratio

TOTAL 251 16.83 5.52 5.58

Survivors 237 14.97 3.21 0

Non survivors 14 48.43 44.70 100 1.01"

*p=0.962 CONCLUSION: SAPS II accurately predicts the probability of hospital mortality in an obstetric ICU population.


285 PREGNANCY COMPLICATION RATES IN AN INNER-CITY PRENATAL DRUG TREATMENT PROGRAM. E Cardomckx, M

Comfortx, K Kaltenbachx, N Sdverman. Depts of Ob/Gyn and Pediatrics, Jefferson Medacal College of Thomas Jefferson Umv Hosp., Phila, PA OBJECTIVE: To evaluate pregnancy complicatmn rates over a 6-year period an a sangle-mstatutmn cohort of substance-dependent women enrolled in a multidisciphnary prenatal treatment program STUDY DESIGN: Retrospective analysis of prospectwely recorded pregnancy outcome data for 182 women dehvered from 1988-1994 All self-reported drug use was confirmed by routine urine toxacology testing For analysis, subjects were grouped as (1) Methadone-treated, + any illicit substances other than cocaine; (2) Cocaine use, _+ other substances, ancluding methadone; and (3) Currently drug-free Comparisons used either X2 or Fisher’s exact test, where appropriate RESULTS: In the overall group, premature rupture of membranes

(PROM) occurred in 20% of pataents, 70% of whom delivered preterm Clinical abruptmn occurred an 8% of pregnancies, 62% of whach were preterm, while meconium-stmned fired was seen at 21% of dehvenes The overall cesarean sectmn (C/S) rate was 21%. Pregnancies comphcated by preterm delivery (28% overall) had significantly higher rates of both PROM (47% vs 9%; p < 0.001) and abruptmn (17% vs 4%, p = 0 04) than those delivered at term When the rates of these complications were examined by drug-use groups, the h~gher rates for both PROM and abruption seen in preterm defiveries persasted only an the cocaine-using group. In addition, trends toward lower rates for C/S (15% vs 23%), small-for-gestation newborns (13% vs 22%) and abruption (3% vs 11%) were seen among women either drug-free or usang methadone only, compared to those who used any fllicat substances CONCLUSIONS: High rates of pregnancy complications were seen an

this cohort of substance-addicted women, with most assocaated w;th the group that used cocmne Preterm pregnancaes xn these women were assocaated wath higher rates of abruption and PROM than those dehvered at term. Women who remained free of ilhcat drug use dunng pregnancy trended toward lower rates of the complicattons studied.

287 CO4MUNI~Y-ACQJIRED PNEUMCNIA IN PREGNANCY.. RG Brjaas. WC Mabie, BM Sibai. Department of Obstetrics and Gynecology, Univ. of Tennessee, Memp.his OBJECTIVE: To determine the e.tiology as well as the maternal and perinatal outcome ot community-acquired pneumonia (CAP) complicating pre~lnancy_. STUDY DESIGN: Hospital recoras of 34 antepartum patients diagnosed with CAP between January 1, 1988 and January 1, 1995 were reviewed. Fever exceeding 37.5°C cough and infiltrate on chest x-ray were nc usion cr ter a. RESULTS: Coexisting illness (asthma sickle cell anemia HIV infection and epilepsy) was present in 10 patients (29%); 11 (~32%) had predisposing factors such as smoking >10 cigarettes/day and/or cocaine use. Diagnosis frequently required multiple laboratory procedures including sputum gram stain, blood and/or sputum culture arterial bloodgas an.alysis, and chest, x-ray. Invasive procedures such as thoracen~.esis ana bronchoscopy were also performed. The etiology was established In only 12 (35%), with varicella and Streptococcus pneumoniae (25%), diagnosed most frequently. Nine patients developed acute respirator~ failure; 7 (78%) of these required intubatlon and mechanical ventilation the other 2 (22%) required continuous positive airway pressure ventilation (CPAP) by fa..ce mask. Five of these 9 develop~_d acute respiratory distress syndrome (ARDS) with 2 (22%) resultant maternal deaths. In addition, there were perinatal deaths (1 stillbirth at 2.2 weeks and 1 neonatal death at 28 weeks gestation). CONCLUSION: The mortality and morbidity for CAP in pre_qnancy remain high. Coexisting illness or predisposing factors were not prerequisites fo.r the develoPment of CAP. Since the .etiology IS frequently not established, empiric antibiotic therapy is appropriate.

286 EFFECT OF COCAINE ON NITRIC OXIDE PRODUCTION BY CULTURED HUMAN UMBILICAL VEIN ENDOTHELIAL CELLS (HUVEC) IN VITRO ~,D. Hsu, Y.K. Chungx, J.A. Copel. Department of Ob/Gyn, Yale University School of Medicine, New Haven, CT. OBJECTIVE: Cocaine can induce vasoconstriction. Nitric oxide (NO) is a potent vasodilator. We studied the effect of cocaine on nitric oxide production using the in vitro model of cultured HUVEC. STUDY DESIGN" HUVEC were incubated with different

concentrations of cocaine hydrochloride at 0, 10"6, 10"5,10-4 M (N=4 each). After 24 hours of incubation, the media was changed to Hank’s balanced salt solution supplemented with CaCI2 (1.3mM),

MgSO4 (0.6 mM), and arginine (100mM). Histamine (20gM) was added as stimulator to test the release of NO by the cultured HUVEC with or without exposure of cocaine. After 60 minutes of stimulation, the supernatant was collected for analysis of NO. NO was measured by monitoring the formation of total NOx (NO+nitrite+nitrate) by a

chemiluminescence detector after reduction of NOx to NO by acidic vanadium (III), and quantified using an integrator by reference to NaNO3 standard. Results were adjusted by the cellular protein in each well Kruskal-Wallis test and Mann-Whitney rank sum test were used for statistical analysis. Results are given as mean+ SEM. RESULTS : Cocaine

NO 0 10-6M 10-5M 10-4M P

H~st(+) !0.2+1.7 9.0&-_ 1.2 8.7+ 1.7 9.3+ 1.7 P=NS Baseline 5.4 _+1.7 6.2+1.4 5.0~_ 2.0 5.7+ 1.8 P=NS

Hist (+):Histamine-stimulated,NO:pmol/~gprotein,NS :not-significant CONCLUSIONS: Cocaine does not alter NO production by cultured HUVEC after 24 hours of exposure. Although NO is a potent vasoddator, our data suggest that cocaine-induced vasoconstriction may not be mediated through the pathway of an inhibition of NO production.

288 BACTEREMIA IN OBSTETRICS: THE EMORY UNIVERSITY/GRADY

MEMORIAL HOSPITAL EXPERIENCE. L. Nathan, B. Dozier, M. Spreuve, Dept. of Gynecology and Obstetrics, Emory University

School of Medicine, Atlanta, GA.

OBJECTIVE: Since the early 1970s, little attention bae been paid to

bacteremia in obatetrice. New practioe patterns and the widespread

u~e of broed-epectrum antibiotics has stimulated a renewed interest

in infections complicating pregnancy. This study was designed to

characterize the microbiology of infections complicated by bacteremis

on an urban obstetric ~ervice in the 199Oa. STUDY DESIGN: From pregnancies delivering between 10/1191 and

9/30/93, those complicated by bacteremia were identified

retrospecUvely. These charts were reviewed and a descriptive analysis

of outcomes compiled.

RESULTS: Bacteremia was diagnosed in 116 obstetric patients during

this period, during which 10,695 women were delivered.

Pyelonephritis (36%), chorioamnionitie (35%), and endomyometritis

(19%) were the most frequent diagnoses. Escherichia �oil,

Streptococcus agaiactiae and Staphylococcus aureue were the most frequently recovered aerobes, whereas anaerobes were isolated from

only 11 (9%) patients. There were no deaths among these patients. Antibiotic Sensitivities

Amp Gent Pcn Met Eryth Vanc Cefsz Cefal

E �oil (n=47) 40% 100% .... 100% 100% S egalactiae (n=23) - 100% -

S aureue (n= 7) 0% 100% 57% 100% -

CONCLUSION: Knowledge of the most common microorganisms

responsible for bacteremia in obstetrics and their usual antibiotic

sensitivities should assist in planning antimicrobial therapy.

390 SPO Abstracts January 1996 Am J Obstet Gyne(ol

289 ALVEOLAR-ARTERIAL OXYGEN GRADIENT IN ACUTE PULMONARY EMBOLISM IN PREGNANCY. R. Pownex, L Larsonx, K Rosene Montella Dept of Med & Ob/Gyn, Women & Infants’ Hospital, Brown Univ., Providence, RI OBJECTIVE: To determine the incidence of normal alveolar- arterial (A-a) gradients m pregnant patients with documented pulmonary embolism (PE). STUDY DESIGN: A retrospective chart review was performed on all pregnant women without complicating cardlopulmonar¥ disease who had a documented PE at 6ur instifution between 198’7 a.nd 1995. Patients with high probability ventilation-perfusion (VQ) scanS,opOSihve pulmonary angiograms, or intermediate probability VQ scans Associated with-documented deep venous thrombosis were included in the study. Room air blood gases (ABG) atpresentation were used to calculate A-a gradient by the formula A-a gradient = 150 - (1.25 PaC02) - Pa02. This calculated value was compared to established normals as defined by values < 20 m~Hg; 2) values < age in years/4+4; 3) Pa0~> 80

~LTS: Eleven patients with PE were identified who had room air A~3Gs drawn Four of 11 (36%) had A-a gradients < 20, 6’11(55%) pataents had room air P02>_ 80 mmHg, and 2 11 (18%) had normal A-a 8radients as predicted for age by the formula age/4+4 This varms from pubhshed data on nonpregnant patients with PE where the range of normal A-a gradients varies from 1.9% -20%. CONCLUSIONS’ In this study, a higher percentage of pregnant patients with PE had normal A-a gradients and Pa02 on room air than in manyprevmus, studies investigating A-a gradients in nonpregnant patients with PE. This suggests that a normal A-a gradmnt as calculated from room air ABG may not be adequate alone to rule out PE in the pregnant woman. -

291 DIFFERENCES IN THE CONCENTRATION OF A N ENDOTOXIN BINDING PROTEIN HELP EXPLAIN SENSITIVITY TO SEPTIC COMPLICATIONS IN PREGNANCY. BKIfive, PJ Rumney~, D Adhoot’, T Asrat, CV Towers, SF Carrollx, MfI

Can’, JA Adashek, M White1. Long Beach Memorial MeAical Center, Long Beach, CA, and University of California, at Irvine, Orange, CA. OBJECTIVE -Endotoxin, through the activation of cellular and humoral cascades, initiates the inflammatory response and multisystem organ changes associated with sepsis. Two recently described banding proteins, bacteriacidal permeability-increasing protein (BPI) and lipopolysaccharide binding protein (LBP), both link with endotuxin but result in markedly different actions. LBP, which is made by the liver, binds endotoxin and dehvers it to the macrophage thus stamulating cytokine production and a cascade of other inflammatory changes involved in the complications of sepsis. In contrast, BPI ~s produced b y neutrophils and competitively brads endotoxin and p~events LBP- reduced signaliing. Thus, BPI acts as an antagonist to the actmus of LBP. Due to effects of pregnancy on the s~tes of productmn of each protein, we hypothesized levels of these proteins would be changed in the gravid patient and thus increase sensitivity to endotoxin (by leading to increased LBP, decreased BPI, or increased LBP/BPI). BPI results will be presented at a future date STUDY DESIGN- 18 pregnant patients in their 3rd trhnester of gestation were matched by age and race to 18 non-pregnant controls. Patients were excluded for evidence of active infection Plasma was obtained and the level of LBP was assayed RESULTS

Pregnant Non-pregnant p (N=18) (N=lS)

Age (yrs) 32 (18-42) 31.5 (21-45) 0 80

AST (lUlL) 16 (7-33) 15.5 (8-28) 0.77

LBP (mcg/ml) 10.5 (4.6-22.7) 3 8 (1.7-7.7) <0 0001 CONCLUSIONS - These data display that levels of LBP are markedly increased with pregnancy possibly leading to enhanced cytokine release and detrhnental inflammatory responses This finding supports the theory that LBP may be an etiologic agent behind the enhanced sensitivity of the pregnant patient to sepsis and its complicatmns

290 APPLICABILITY OF A THIRD GENERATION THYROID

STIMULATING HORMONE (TSH) ASSAY IN PREGNANCY. __R Bobrowski, P Strelcher~, J Dz=eczkowskix, M Dombrowski, K Puder, B Gon=k, Depts. of Ob/Gyn and Path, Wayne State Univ., Detroit, MI OBJECTIVE: A new third generahon TSH assay is now available, but

its use in graved patients has not been validated. We sought to

determ=ne the apphcabflity of th=s ultrasensihve assay in pregnancy. STUDY DESIGN: We obtained serum from 93 gravidas with a

singleton gestation Women w=th symptoms or h=story of thyro=d d~sease were excluded. TSH was determined by a two-site

=mmunoenzymat~c ("sandwich") assay w=th a reportable range of 0.006-100 0 plU/mL (Sanofi Dmgnoshcs, Chaska, MN) Standard

enzyme =mmunoassays were employed for total T4 (tT4), free T4 (tT4) and T3 levels Reference ranges (RR) established by the kit

manufacturer were used for comparision Analysis was by MANOVA RESULTS: Mean and range values for pregnant subjects were

MEAN RANGE RR TSH (plU/mL) 1.2 0.1-3 4 0 5-5 6 tT4 (pg/dL) 79 23-11 6 50-12.0 f’f4 (ng/dL) 1.1 0.7-1.8 1.0-2 5 T3 (ng/dL) 221 144-312 80-200

There were no differences in mean values between tnmesters for

TSH, tT4, f-14, and T3. Thirteen of the 93 women (14%) had a TSH value below the lower lim=t of the RR, though none had climcal ewdence of hyperthyroidism or an elevated fT4

CONCLUSIONS: Currently available RRs for a third generation TSH assay may not be applicable to pregnant women Until additional data become available, isolated TSH measurements for screening or

mondonng hyperthyro=d=sm dunng pregnancy should be thscouraged

292 THE LACK OF ASSOCIATION BETWEEN SELECTED

HEMATOLOGIC PARAMATERS AND FREQUENCY OF

PAINFUL SICKLE CELL CRISIS DURING PREGNANCY.

A Anyaegbunam, MD, M.S. Mikhad, MD~’ D. Jadah, MD~ and H. Billett,MDL Albert Einstein College of Medicine, Bronx, New York.

OBJECTIVE: Painful vasoocluswe crisis is the most common cause of morbidity m sickle cell(ss) disease and pregnancy has been associated

with an increase m the frequency of painful crises. The present study evaluates the association between prepregnancy hematologic parameters

and the frequency 9f crisis during pregnancy in women with sickle cell

disease. STUDY DESIGN: The study population consisted of 37 homozygous

sickle cell disease women recruited from our non-pregnancy sickle cell disease program. All patients were followed from the first trimester of pregnancy until delivery.

RESULTS: The rate of painful crisis was 2.88 + 3.23 per pregnancy. Selected hematologic tests and their average pregrawd values Were;

hemoglobin level (8.51 + 1.35)g/dl, hemoglobin F concentration (6.13 +4.22)70, percentage of dense cells (13.43 +9.71)% and the retmulocyte

count (13.08 + 5.32)%. There was no relationshrp between frequency of smkle cell crisis during pregnancy and hemoglobin level (r=-.05;

p=0.81); hemoglobin F concentration (r=0.12,p=0.56), percentage of dense cells (r=0.06;p=0.67), or the reticulocytc count (r=-

0.08;p=0.67). CONCLUSIONS: The studied prepregnancy hematologic parameters are

not predictive of the frequency of painful sickle cell crisis during pregnancy. Further search is needed to detect marker(s) that may identify patients at risk for frequent sickle cell crisis during pregnancy. Such markers would be useful m prepregnaucy counseling and prenatal management of patients with sickle cell disease.


293 MATERNAL OUTCOME ASSOCIATED WITH ADULT RESPIRATORY

DISTRESS SYNDROME. K.G. Perry, Jr., R.W. Martin, P.G. BlakeX, W.E. R~berts, J.N. Martin, Jr. Dept. Ob/Gyn, Univ. of Mississippi, Jackson, MS. OBJECTIVE: The purpose of our study was to characterize pregnancy-related adult respiratory distress syndrome (ARDS) in a single tertiary care center, and identify any factors that might be predictive of maternal outcome. STUDY DESIGN: Records of all pregnant patiants diagnosed with ARDS and admitted to Intensive Care at a single tertiary care center over a 14-year period were reviewed. Data collected on each patient included demographic characteristics, precipitating factors, coexisting diseases, obstetric and non- obstetric complications, days in the ICU, and outcome. The cases were

stratified into survivors and non-survlvors in order to analyze any factors that might be predictive of outcome. RESULTS: Forty one cases of pregnancy-related ARDS were identified with an incidence of 0.7 per 1000 live births. Thirty one patients survived for a maternal mortality rate of 24.4%. ARDS was diagnosed in the antepartum period in 23 (56.1%) patients, most in the third trimester (73.9%). There was no statistically significant difference in demographic characteristics between the survivor and non-survivor groups (p > 0.05). Coexisting diseases were divided similarly between survivors (S) and non-survivors (Non-S).

Precipitating Factors S (n = 31) Non-S(n = 10)

Infection/sepsis 12 (38.7%) 3 00%) Preeclampsia/HELLP 10 (32.3 %) 3 00%) Preterm labor/tocelysis 4 (12.9%) 2 (20%) Aspiration 3 (9.7%) 1 (10%) Obstetric hemorrhage 2 (6.5 %) 1 (10%)

The cause of death among the non-survivors included multisystem organ failure (5 patients), sepsis (4 patients), and disseminated intravascular coagulation (1 patient). CONCLUSION: Pregnancy-related ARDS continues to be associated with a high maternal mortality rate although this appears to be lower than previously reported. The etiology for ARDS during pregnancy is most oRan due to an

obstetric condition or complication which is not predictive of maternal outcome.

295 PLACENTAL INSUFFICIENCY IS CHARAutI~cdZED BY PLATELET ACTIVATION IN F~’ILIS AND MOTHER. B.I. Tmdingex, 7_. Wu*, J. Song,* S~Rowlands*. The University of Sydney at Westmead Hospital - Weslmead Nb3V 2145 AUSTRALIA OBJECTIVE: Maternal endothelial cell injury and platelet consumption occur in l:n~chmpsia and placental insufficiency. Small vessel pathology erasts in both the uteroplacental bed and fetal placental vdh. Our objective was to mvestagate whether activation of ctrculatmg platolets xs present in both the fetal and metemal systems m these caseg

STUDY DESIGN: We stu&ed 6 normal pregnancies and 20 mother-fetus pairs w~th an abnormal umbilical Doppler study indicating placental pathology. At elective Caesarean delivery blood was collected m sodium oh’ate, dduted with Tyrodes buffer, incubated with monodonal ant~bo&es and fixed. A flow cytometer techntque was used. T~ platelet population was speofiod with a monodonal anti glycoprotem 111 a (CD61) whilst monodonal anti - P mloctin (CD62) identified activated platelets. Using LYSIS II software a discrimination gate was placed around the platelet population to molete them from rod/white cells and debris. Platelet response to thrombim (0.03 to 0.25 ug/ml) was assessed. RESULTS: In the normal palients there was no evidence of platelet activation (<1% platelet poptflalaon). In the study group platolet aclivation was pp~sent in mother ( 5.7 + 2.9%) and fetus (5.0 ± 2.9%) and an exaggerated response to thrombin ~. No difference was noted m the subgroup of mothers with preedarnpsia (8 cases) m maternal (6.0%) or fetal (4.8%) ~ults, and w~th fetal IUG1L CONCLUSIONS:. In placental insufficiency platelet activation is present m both fetus and mother and independent of the maternal syndrome of preechrnpm~. Thrombin sens~tivtty is enhanced in both. R was present to the same extent m early and advanced disease suggesting ~t was an ~trly feature of the disease

294 MASSIVE FETOMATERNAL HEMORRHAGE AND FETAL DEATH: IS IT PREDICTABLE? R Samad~ MD,~ D Mdler MD, R Settlage MD,x 1Gv~azda,x R Paul MD, and T M Goodwnl MD University of Southern Cahforma School of Medicine, Los Angeles, CA.

OBJECTIVE: To report the ~nmdence of massive fetomatemal

hemorrhage (FMH) ~n fetal death and to test the hypothes~s that th~s

finding ~s less hkely to be present ~n cases of fetal death with risk

factors for FMH than in those w~thout risk factors

STUDY DESIGN: All cases of spontaneous fetal death greater than

500g were reviewed retrospectively from 1/1/90 to 12131/94

Kle~hauer-Betke (K-B) testing was ordered upon adm~smon at the

discretion of the attending physician Massive FMH was defined as

greater than 1% fetal cells m the matemal circulation Risk factors for

FMH Included abruptlon, hypertensive disorders of pregnancy,

trauma, utenne rupture and selected fetal/placental anomalies

Women with risk factors were compared to those without risk factors

in regard to the occurrence of massive FlvIH

RESULTS: During the study period, 645 cases of spontaneous fetal

death above 500 grams were identified The K-B test was performed

in 319 (49 5%) patients Massive FMH was :dent~fied in 15/319

(4 7%) cases It was present xn 6/110 (5 5%) patients with risk

factors for FMJq and m 9/209 (4 3%) patients w~th no risk factors

(P=0 86) CONCLUSION: Massive FMH is an uncommon but not a rare

finding ~n cases of fetal death, and Its presence ~s not reliably

predicted by chmcal risk factors To our knowledge, our data

represents the largest series of fetal deaths systemahcally stu&ed for

FMH Based on these findings, we recommend that K-B testing be

performed in all cases of fetal death regardless of supposed risk

factors lbr FMH

296 IS THERE AN ANTIPHOSPHOLIPID ANTIBODY NEGATIVE ANTIPHOSPHOLIPID-LIKE SYNDROME? JR. Scottx, RM.

Silver, DW. Branch. Dept OB/GYN, U of Utah, SLC, UT. OBJECTIVE: To :denttfy women who fulfill clinical criteria for the

annphosphohpld syndrome (APS) but lack antlcardlohpln antlbothes

(aCL) and lupus anticoagulant (LA) and to characterize their methcal and obstetrical outcomes. STUDY DESIGN: The study group consisted of 12 patients who had

experienced a thrombotic episode and at least one fetal death. All women tested negative for IgG and lgM aCL anubo&es and LA RESULTS: These patients had 19 separate thrombouc episodes ~ncluding 11 DVTs, 6 pulmonary emboh, 1 CVA and 1 arterial thrombos~s. 26% of these thrombotic episodes occurred during

pregnancy. One woman also had auto~mmune thrombocytopenla, and several had severe preeclampsla or a growth retarded fetus in a prmr pregnancy. Of 56 pregnancies, 16 (29 %) were first trimester losses,

19 (34%) were fetal deaths, and 21 (37%) were live b~rths. Pregnancy outcomes and complications were similar to those of patients w~th APS and antiphosphohpld anubodles. CONCLUSIONS: This case series emphasizes that some patients w~th clinical features strongly suggestive of APS are negative for LA and aCL These patients may have an as yet uncharacterized autmmmune syndrome and pose a difficult chmcal dilemma. Their

apparently h~gh risk for thrombos~s, as well as their history of poor fetal outcome, call for a prospective evaluation of anticoagulant therapy in these women.


297 HEPARIN THERAPY REDUCES THE RISK OF FETAL DEATH AND IUGR AND IS EFFECTIVE IN THE TREATMENT OF THE ANTIPHOSPHOLIPID ANTIBODY SYNDROME AND OF

REPEATED PREGNANCY LOSSES. ~x, j HirshbergX, M C LeivaX, R Librizzi, J.E. Tolosa. Department of OB/GYN, Pennsylvania Hospital/Thomas Jefferson Umversity, Philadelplua, PA and the Perinatology Research Branch, NICHD, Bethesda, MD OBJECTIVE: To determine the safety and efficacy of heparin treatment in patients with antiphospholipid antibody syndrome (APAS). STUDY DESIGN: A case-control design was used to study 134 pregnancies m 72 women diagnosed with APAS and repeated pregnancy losses Three groups were defined and each woman served as her historic control: 1) Systemic Lupus Erythematosus and + antibodies (Ant~cardiolipin antibody, Lupus anticoagulant, Antinuclear antibodies) 6/72; 2) Poor obstetric bAstory (> 3 first trimester losses and/or abruptio placentae, IUGR, ohgohydramnios, placental infarctions, IUFD) and + antibodies, 50/72, 3) Poor obstetric history alone, 16/72. IUFD was defined as intrauterine death after 12 weeks gestational age (GA). Heparin was started at 12,500 umts by subcutaneous injechon twice daily, adjusting the dose through a trough PTT > 40 75 pregnancies were treated and 59 were not Conconntant use of aspirin occurred m 24% and of prednisone m 12% of the pregnancies Odds ratios (OR) and 95% confidence intervals (CI) were calculated for pregnancies progressing beyond 12 weeks GA. RESULTS: Heparm treatment reduced the rate of fetal demise, IUGR and placental infarction, OR and 95% CI: 0.02 (0 01, 0.09), p < 0 0001, 0.31 (. 10, .94), p=0.019 and .38 (.16, .89), p=0.014, respectively. Stratafied analysis indicated that neither aspirin nor predmsone treatment conferred any additional benefit Thrombocytopenia ( < 100,000) occurred m 8% of treated pregnancies; one patient had an abdominal wall hematoma which required operative drainage CONCLUSIONS: 1) Heparin txeatment of the APAS and of women with repeated fetal losses is effectave by reducing fetal mortality and morbidity 2) This therapeutic modality appears to be safe during pregnancy

299 MANAGEMENT OF IMMUNE THROMBOCYTOPENIA DURING

PREGNANCY, BY MEANS OF FOETAL BLOOD SAMPLING. Bartot

P__~, Reman O, Arrazola J, Levaltier X, Herhcoviez M, Leporrier M, Muller

G, Levy G, University Hospital Center, Caen, France

OBJECTIVE. Obstetric management of Immune "l~rombocytopema (ITP)

during pregnancy and its xnfluence on neonatal outcome is controversial. Our

goal was to determine the form of delivery for thrombocytopemc women using

a foetal blood sampling (FBS) technique.

STUDY DESIGN This is a prospective study approved by an ethical

committee, undertaken from May 1993 to May 1995. All women attempting

delivery at the maternity clinic of the University Hosp:tal Center (C.H.R.U.) of

Caen (6300 patients) had a platelet count of 36 weeks of gestation Aider

exclusion of all other thrombocytopenlc mechanisms (allo-immunologic,

infectious disease, Antlphosphohpidlc syndrome, preeclampsia,

thromhncytopathlc thromhncytopenia), we included m a FBS protocol, 20

patients featuring platelet counts ~nferior to 100 Giga/l., including all cases of

previously treated ITP, even vath normal platelet counts. Therapy of these

patients included corhcosterolds and/or lntravanous munnaoglobulins. FBS

was performed between 38-40 weeks gestation and delivery induced within 1

week.

RESULTS’ Of the 20 cases in this study, 8 cases were known as to be

previously affected by the illness. AVERAGE PLATELET COUNT

BEFORE DELIVERY AFTER DELIVERY

Panent~ 85 Gtga/l (range 44-225) 202 Gtgo/l (range 77-276) Pattents 90 Gtgo/l (range 56-150) 225 Gtga!l (range 20-313)

No maternal or foetal comphcattons after FBS were evtdenced

CONCLUSIONS" Current pregnancy care involves increasing diagnosis of

maternal thrombocytopenla through systematic prenatal haematologic

exammatmns Our knowledge of an increased incidence of cerebral lesions in l vaginal deliveries of infants presenting a thrombocytopema, has shown the

preventive importance of a Cesarean Section, when platelet counts are inferior

to 50 Glga!l. Cordocentesis m high risk patients can be performed vathout

any bleeding complications Further studies vail be needed to establish the

actual value, risks, and limitations of this approach in foetal medicine.

298 VERTEBRAL COMPRESSION FRACTURES IN PREGNANCIES COMPLICATED BY PROLONGED HEPARIN AND CORTICOSTEROID THERAPY. N Slnlletto, S Granovsky-Gnsaru, T O Teoh, K Spltzer, D Farme, C Laskm. Dept OB/GYN, Mt Shun Hosp, Dept of Med, The Toronto Hosp., Umv of Torento, ON, CANADA OBJECTIVE: To assess the impact of a history of heparin, prednisone and

aspmn therapy, alone ~r m combination, on the mcidence of ostcoporotic vertebral c~apresston fractures in wccmon with recurrent pregnancy loss

(RPL) STUDY DESIGN: In a rctrospactlve analysis of 150 woman evaluated at Mount Sinai Hospital for RPL, 101 patients with documented

autoantibo&es or pre-existing auto-munune dlsense were placed on one of

a variety of trea~mant protocols using low dose aspmn (ASA) 80 rag/day, predmsone 0 5 mg/kg/day and/or heparm 20,000 units s q/day The patient records of those undergoing therapy were reviewed for obstetric outcome and for the incidence of osteoporolac vertebral fractures

RESULTS: In 101 patients undergoing therapy, 96 pregnancies reached the third trnnester Of these, 2 (2 0%) were placed on heparm only, 2 (2.0%) on prednlsene only, 18 (18.7%) on ASA alone, 13 (13.5%) on hepann and ASA, 53 (55 2%) on predmsone and ASA and 8 (8 3%) on all three agents Of these, 3 patients suffered vertebral compresston fractures, two of whom had prolonged therapy w~th one agent prior to pregnancy ( one

predmsene, one heparm ). The tturd had a 20 pack year smoking history Seven of eight palaents with triple agent therapy begun during pregnancy had neither fractures nor symptomatac osteoparotic disease CONCLUSION: Pataents vath longterm single agent heparin or corticostermd therapy predating pregnancy are at markedly increased risk of vertebral fractures if heparin or corticosterold therapy is added to their treatment regimen

300 HEMOLFTIC UR£~C SY2~ROME/THRO~BOTIC THROMBOCI’~PI~IlC

PURPURA INPUTS" ~ OF ~ C~. RS Epe~an ~d BM Sib~.

~pm~nt of Ohte~ ~d ~b~, Oni~i~ of Tennis, MemphB. ~ LiRle inaction ~ re~g hem~c u~c syndm~ ~ ~d thiom~t~c throm~topenic purpura (~) during pregn~. Few c~es of ~tep~m HUS have ~en publ~hed. We ~fl the l~g~t ~fi~ of ~S~, ~ emph~ on d~t~ ~d ~a~ment di~ ~th~ ~ ~nditiom

~Y D~I~: ~een 19M ~d 19~, 9 worn ~th either ~ (n = 6) or HUS (n = 3) we~ ~l~ted. Clini~ ~d la~rato~ finding, m well ~ ~temfl ~d neomtal out~m~, ~re studied. ~e patien~ ~re ~her ~te~fl~d into ~o ffou~: Group I de~p~ ~e s~dro~ ~ound dehve~ or i~iately ~t~um (n =5); Group ~ ~ ~o~ to ~ the s~dm~ ~fo~ pmgn~ (n = 4). ~$~$: In Group I, 4 ~tien~ w~e deliv~d at 27-39 weeM with a dign~B of s~ere prussia. Subs~uent~, they had ~man~g d~i~ ~d l~oto~ findin~ that were cons~tent ~th ~ ~t~tum (~y 1-11); the re~ing ~tient had ~ electwe termination of pregn~ at 13 ~e~, w~ suM~uently ~admitt~ on ~to~rafive day 1 for endometfith, ~d found to ~ve s~ere renfl ~ure. She had clanc~ =d l~o~to~ findin~ cons~tent ~th HUS on posto~rative ~y 2; the dia~h ~ ~nfi~ by reml bto~y. ~ong the 4 ongoing pregn~d~, the~ were 3 bye biflhs (27, 33, ~d 37 wee~) ~d 1 stillbirth (39 w~). ~1 5 ~n required re~ated courses of fr~h frozen pl~ma ~P), and 3 required p~phe~ ~). gl 4 pat~en~ in Group II ~ in re~n prmr to pmgn~. ~e 2 patien~ ~th pre~ting ~S r~uired delive~ at 15 ~d 27 w~’ ~station ~me of se~re detefio~tion in their clini~ ~ndition. ~e 2 ~ning patien~ ruth pre~tmg ~ had m~tion of thek d~e r~uifing deli~ at 17 ~h ~me of fet~ deme ~d at 33 w~ ~me of ~r~nt hte d~eleraOom. ~ 4 women ~qu~r~ re~ated courses of ~, ~d 3 ~quired PHg Moafli~ or ~r mo~fli~ ~ s~gn~fly higher in ~up I ~ in G~up II (5/5 ~. U4, p = 0.~8).

GRO~ I ~O~ ~ Platelet nad~ x 1#/~ (rage) 28.8 (2~8) 22 (1~2) i~ ~H ~U~) (~) 2433 (1455~1~) 1189 (3~18~) RenM imu~c~n~ 5/5 3/4

Me~ ~tmme (m~dl) (m~) 6.88 (1.~13.9) 4.3 (1.~.5) R~idu~ neumbg~ defiot 1/5 0/4

Ch~mc dtfl~ 2/5 1/4

~ 2/5 0/4 CONCLUMO~. HUS~ that de~bps de ~ in p~gn~ ~ ~at~ ~ hi~ ~temfl ~d fe~ ~i~ ~d ~. ~pite intemive ~nitofing in thee ~ predating d~e~e, ~n~ com#i~fions during p~ never~el~s ~d ~d fetfl outco~ ~ ~r; bng-te~ m~ernfl s~uet~, hoover, w~e l~s seve~.

Volume 174, Number l, Part 2 SPO Abstracts 393

Am J Obstet Gynecol

301 CHOLESTASIS OF PREGNANCY: PERINATAL OUTCOME

ASSOCIATED WITH EXPECTANT MANAGEMENT

O Alsulvm~, J Ouzounian, M A Castro, R Paul, T M Goodwin

Univ. of Southern California School of Medicine, Los Angeles, CA

OBJECTIVE: To compare the pregnancy outcome of patients with

cholestasis of pregnancy managed expectantly with antepartum

testing to other patients followed with a smailar testing scheme

STUDY DESIGN: All cases of cholestasis of pregnancy followed

with antepartum testing at our institution from 1988-1994 were

reviewed Their pregnancy outcomes were compared to controls

(matched for age and parity) followed with the same testing scheme for

a history of prior stdlbtrth Both groups had a weekly nonstress test

and amniotic fluid assessment untd spontaneous labor or delivery for

standard obstetrical indications.

RESULTS: The two groups did not differ with respect to mean

gestatlonal age at delivery (38.5 vs. 38 8 weeks) or mean birth weight

(3239 vs. 3256 ~ms!. Other results are summartzed below

Cholestasis Control P

(n=74)

Spontaneous preterm 12 (16%) 7 (9.5%) 0.2

birth (< 37 weeks)

Mecomum stained 31 (42%) 8 (11%) <0 001

ammotic fluid

Mecomum aspiration 3 (4%) 0 0.08

Fetal death 2 (2.7%) 0 0.15

The two cases of fetal death occurred at 36-37 weeks of gestation

within 5 days of normal antepartum testing in the absence of other

pregnancy complications. Thick meconium and appropriate birth

weight were noted in both cases

CONCLUSION: Cholestasis of pregnancy is associated with

mcreased incidence of meconium passage at delivery and mecomum

aspiration syndrome There is a trend toward increased incidence of

fetal death not predicted by conventmnal fetal surveillance

303 PROSPECTIVE LONGITUDINAL STUDY USING STABLE ISOTOPES IN PREGNANCY TO ASSESS CHANGES IN INSULIN SENSITIVITY IN OBESE INSULIN REQUIRING DIABETIC WOMEN. E. Sivan.~, C. Hc~nkoz, X. Cbenx, G. Boden=, E.A. Rce~.

Depts. of OB/GYN & GCRC, Temple Univ Sch of Med, Phlia., PA. OBJECTIVE: The ~ of the current study was to longitudinally quantify f~ the first llme, insulin sensitivity in obese pregnant wonaen with diabetes mellitns as compared to non-diabetic controls.

STUDY DESIGN: Pregnant women (n-10) were evaluated using 4-honr hyperinsulinemic-englycemic clamp studies at 16-27 and 32-38 weeks’

gestation. Body composition was estimated by skinfold anthropometry

and dilution deuterium method. Basal endogenous glucose production

was estimated with a primed constant infusion of 6,6-2 I-~ glucose. Insulin sensitivity was determined by the glucose infusion rate needed to

maintain plasma glucose constant at a level of 85 mg/dL RESULTS: Blood glucose levels were maintained at 83.323.51 mE/alL

for the control group and 85.4+_2.8 mg/dL for the diabetic group during the 4-honr clamp study. The glucose infusion rate (GIR) required to maintain euglycemia was significantly lower in the diabetes group in comparison to healthy controls during both the second (2.5920.59 vs.

5.46+_1.8 mg/Kg/min; p <0.05) and third (2.45+_0.54 vs. 4.9521.69 mg/Kg/min; p <0.05) trimesters. However, ~ GIR did not decrease

between the second and third trimesters in either the diabetic or control groups. The mean body mass inde~ for the study group was 3~. 1 2 8.9

kg/ma and mean percentage body fat was 39.9 ± 6.2%.

CONCLUSIONS: Although insulin sensitivity is reduced in obese IDDM women during pregnancy, we observed minimal change in GIR between the second and third trimesters in our study population. We postulate that insulin resistance is already well established in obese women by the second trimester, and therefore little

change in insulin sensitivity was observed with advancing gestation.

302 OUTCOMES IN PREGNANCIES COMPLICATED BY DIABETES (CLASS B TO FR) VERSUS NON-DIABETIC CONTROLS. E. Sivan=, C. Homko~, E,A. Reece, DeperUnent of OB/GYN & RS, Temple Univ Sch of Med, Philadelphia, PA.

OBJECTIVE: The purpose of the cua’rent study was to evaluate the

impact of contemporary diabetes management on the outcome of pregnancies complicated by insuiin-dependent diabetes mellitns (IDDM). STUDY DESIGN: The study population consisted of two hundred and eighty-eight (288) women with IDDM and one hundred and fifty (150) healthy controls (Group 1). Diabetic women were grouped according to the presence (Group 2; n-103) or absence (Group 3; nffi185) of diabetic

vasculupathy. Data were collected regarding diabetes management, level

of glycemic control, and fetal and maternal outcomes. RESULTS: A significant difference was found between the combined diabetes groups (Group 2 and 3) and healthy controls in all maternal-fetal outcomes examined, except SGA and stillbirth. However, there was no significant difference between the two diabetes groups in tea’ms of pre-

term labor, polyhydramnlos, pyelonepkritis, or growth aberrations. Acute hypertensive complications and the rate of malformations were

significantly higher in women with vasculupathy: 51.6% vs 32.9 (p <0.05), and 6.8% vs 1.6% (p <0.05), respectively. A number of peripextum complications were significantly higher in IDDM women who

were poorly controlled vs. those controlled during the third trimester: preterm labor - 30.8% vs 11.4%; polyhydramnios - 17.3 vs 5.1%; and macresc~nia - 51.9 vs 33.0%, respectively; p <0.05.

CONCLUSIONS: Women with diabetes can be counseled to expect favorable pregnancy outcomes, although the incidence of maternal and fetal complications are increased above the general population. Furthermore, with the exception of hypertensive disorders and malformations, outcomes in IDDMs with vnsculopathy are comparable to diabetic women without mlcrovnscular disease.

304 INCREASED INCIDENCE OF LARGE FOR GESTATIONAL AGE INFANTS NOTATTRIBUTABLE TO GESTATIONAL DIABETES. MR Leonard~X, SF Bottoms Depa~ment of Obstetrics & Gynecology, Wayne State Umvers~y, Hutzel HosNtal, Detroit, M~chlgan OBJECTIVE: To determine the if maternal obesity or gestat~onal

d=abetes is responsible for increased frequenc=es of large for gestatlonal age (LGA) and macrosom~c ~nfants STUDY DESIGN: Gestat~onal dtabetes was dtagnosed by Nattonat Diabetes Data Group criteria m 117 grawd women They were compared to 3090 control patients w~hout gestational dmbetes Women with pregestabonal d~abetes and multiple gestations were excluded. LGA and macrosom~a were defined as dependant variables. Macrosom~a was defined as a Nrthwelght > 4000g Prepregnancy body mass index (BMI) was used as a measure of maternal obesity Stepwlse Iog=stic regression was used to ~dentffy the influence of multiple antenatal variables as predictors of LGA and macrosomla RESULTS: Women w~th gestat~onal d=abetes tended to be older, heav=er, and more parous WNle GDM was pred=cUve for LGA (p=0 001), the best predictor of an LGA ~nfant in the total population was BMI (p< 0001) Once maternal obes=ty was accounted for (p < 0 0001 ), GDM d=d not s~gnlflcantly affect the mctdence of macrosom~a In grav~das whose prepregnancy we=ght was less than 140 pounds (n=1596), neither BMI or GDM pred=cted LGA as an outcome (p >0.4). CONCLUSIONS: It appears that maternal obesity, rather than GDM itself, is responsible for the increased frequencies of LGA and macrocom~c infants m pregnancies comphcated by GDM. When interpreting these results, ~ is ~mportant to consider that most patients w~th GDM received treatment, potentially altering outcome Further study is needed to determine ff untreated GDM alters the frequencies of these outcomes

394 SPO Abstracts Janua~ 1996 Am J Obstet Gynecol

305 PREGNANCY OUTCOME IN WOMEN WITH TYPE II DIABETES

MELLITUS. B M Rosenn, M Miodovnik, J C Khouryx, T.A. Siddlqi. Dapt.

Ob/Gyn, Univ of Cinti, Cincinnati OH

OBJECTIVE: There is a paucity of data on outcome of pregnancy in

women with Type II d~abstes (NIDDM). In these pregnant women, duration

and severity of d~sease is generally assumed to be less than in women with

Type I d~abetes (IDDM). The purpose of th~s study was to test the

hypothes~s that women w~th NIDDM have a better outcome of pregnancy

than women with IDDM.

STUDY DESIGN: Women with diabetes were enrolled in our Diabetes in

Pregnancy Program between 1978-1995, and a({ data were comp{(ed (n a

computerized database. Classification of diabetes was based on clinical and

laboratory characteristics. Chi-square and t test were used as appropriate

to analyze outcomes of interest, p<.05 was considered significant.

RESULTS: There were 450 pregnancies in IDDM women and 84

pregnancies in NIDDM women Charactenst~cs of subjects were as follows.

IDDM(n=450) NIDDM(n=84) p

Age 25 6±5 1 29.1-5.5 <.0001

Race (White-Black) 84%-15% 66%-34% <.0001

Years of diabetes 12 8±6 6 5.1 .-4.8 <.0001

Age of d~sease onset 12.7±6.5 23.4±6.4 < 0001

Microvascular disease 27% 7% < 0001 Prepregnancy wt (Lbs) 135±24 204±52 < 0001

Glucose control was sl~ghtty better m women with NIDDM. However, there

were no differences between groups with respect to rates of preeclampsla,

preterm labor or delivery, weight gain, cesarean delivery, LGA, macrosomia,

or newborn hyperbilirubinamia, but ~nfants of IDDM women had a higher

rate of hypoglycemia and RDS.

CONCLUSIONS: Although d~abet~c d~sease appears to be less severe in

pregnant women w~th NIDDM, most complications of pregnancy are as

common as m women w~th IDDM.

307 PATIENTS WITH ELEVATED GLUCOSE SCREENING TESTS ARE AT HIGHER RISK FOR CESAREAN SECTION , MACROSOMIA AND BIRTH TRAUMA DESPITE A NORMAL 3 HOUR GLUCOSE TOLERANCE TEST. Carol A. Maior. MD, Bruce F. Cohen, MD and Tricla Reimbold, RNx. Department of Obstetncs and Gynecology. Universzty of California, Irvme Medical Center, Orange, CA. OBJECTIVE: Patients with an elevated 1 hour glucose screening test (GST) followed by a normal 3 hour glucose tolerance tests (GTT) are usually managed as =f they had normal glucose tolerance, The purpose of th~s study is to compare the birthweights, inczdence of macrosomia, rates of cesarean secbon (C/S) for cepha/opelvic disproportion (CPD) and rates of shoulder dystco~a =n pahents with elevated GST and normal GI-I to normal controls (normal GST) STUDY DESIGN: All patients w=th an elevated GST and a normal GTi" (n=130) were idenhfled by reviewing laboratory records. A control group of patients with normal glucose screening (n=130) was selected from patients dellvenng during the same time period. Charts were reviewed for demographic data and outcome variables including blrthweight, route of delivery and birth trauma. The glucose screening test (GST) consisted of a 50 gm oral glucose load followed by a 1 hour plasma glucose determination. A value of > 140 mg/dl was considered abnormal and indicated the need for a 3 hour 100 gm oral glucose tolerance test (GTT) An abnormal GTT required that two or more of the following glucose values be met or exceeded: fasting 105 mg/dl, 1 hour 190 mg/dl, 2 hour 165 mg/dl; 3 hour 145 mg/dL Macrosomla was defined as a birthwelght > 4000gin. RESULTS: Those patients with an elevated GST and normal GTT, when compared to those with normal glucose screening, had a higher birthwe=ght (3692+ 537gm vs. 3273_+502gm, p = 0.001), a higher rate of macrosomia (28/130 vs 11/130, OR 2.97 CI 1.34-6.71; p = 0 005), a higher rate of C/S for CPD (18/130 vs 8/130, OR 2.65 CI 1.03-6.96, p= 0.030), and a higher rate of shouJdar dystoc~a (9/130 vs. 21130, OR 4.76 CI 0.93-32 6; p= 0.04 ). There were no differences between the groups w=th regards to maternal age, parity, race, or gestational age at delwery. CONCLUSION: Patients with elevated 1 hour glucose screening tests are at h~gher risk than controls for C/S for CPD, macrosomia, and shoulder dystcoia despite normal 3 hour glucose tolerance testing

306 PLATELET ACTIVATION IN DIABETIC PREGNANCY. J ~tar, K Rosene, M W Carpenter, J FerlandX, V Rayx, G DiLeone×, L Larsonx, A Kestm× Brown University, Depts of Ob/Gyn & Medicine, Women and Infants/Rhode Island Hospital, Providence, Rhode Island OBJECTIVE: To determine ~fthe observed third trimester decrease ~n platelet activation, as measured by platelet glycoprotein (GP) membrane

receptor binding, is amplified by maternal d~abetes STUDY DESIGN’ F~ve cc of whole blood were obtained by atraumat~c venipuncture from 20 third trimester pregnant subjects (8 w~th d~abetes w~thout evidence of vascular d~sease (4 Type I, 4 Type II), and 12 with

normal glucose tolerance) and 13 nonpregnant controls. The platelets were exposed to thrombm at maximal and minimal doses m whole blood, in parallel with a buffer control The samples were fixed and incubated w~th an

indicator antibody (Ab) and three test antibodies. S12 (a granule/P-selectin), 7E3 (fibrinogen/GPIIb/IIIa) and 6DI (yon Willebrand factor/GPIb) Using a flow cytometric technique, mean fluorescence intensity was determined lbr 5000 platelets per sample

RESULTS: Dmbetic subjects showed significantly zncreased actwation by S12 binding prior to stimulation, followed by decreased acUvation at minimal thrombm doses, compared with nondiabetic subjects (p<.05). 7E3 b~ndlng was significantly decreased m all pregnant subjects tested at maxima! thrombm doses compared to nonpregnant controls, as previously reported (data not shown) No significant differences were noted with 6D1.

Mean fluorescence intensity w~th S12 (4- SD) Thrombin(U/mB Diabetic subiects Nondiabetic subjects 0 3 9 (±6.4) -1.4 (±3) 0.08 35 4 (±24 4) 65.8 (4-31) CONCLUSIONS. Third trimester subjects with diabetes show evidence of

increased activation in resting platelets, w~th subsequently decreased sensinv~ty to agonist stimulation F~brmogen receptor binding is decreased

~n dmbet~c and nondiabetic subjects compared with nonpregnant controls. These findings suggest a chronic activation state in platelets dunng late

pregnancy, which is accelerated by maternal diabetes.

308 EARLY SCREENING FOR GESTATIONAL DIABETES: IS THERE A ROLE? Mlchele A Gerber, MDx MPH, Carol A. Ma!or. MD and Bruce F. Cohen, MD. Dept. of Obstetrics and Gynecology, University of California, Irvine Medical Center, Orange, CA. OBJECTIVE: The timing of diabetic screening during pregnancy in patients at high nsk for development of gestational diabetes (GDM) is controversial. Many practit=oners rout=nely screen patients with a previous history of GDM, fetal macrosom~a, an unexplained stlllb=rth or glycosuna at the hrst prenatal visit. The objective of this study is to determine if there is benefit to early diabetic screening in these patients at high risk for gestational diabetes. STUDY DESIGN: We reviewed the medical records of women receiving prenatal care at our Public Health Department over a 1 year period. Inclusion criteria for the study included. 1) First prenatal visit prior to 20 weeks, 2) Participation in the Health Department’s protocol of glucose screening at the first prenatal visit and again at 24- 28 weeks, 3) a h~story ol pnor GDM, fetal macrosom~a, prewous stillbirth or persistent glycosuda. The glucose screening test (GST) consisted of a 50 g~t oral glucose load followed by a 1 hour plasma glucose determination. A value of _> 140 mg/dl was considered abnormal and ind=cated the need for a 3 hour 100 gm oral glucose tolerance test (GTT). An abnormal GTT or a definitive diagnosis of GDM required that two or more of the following glucose values be met or exceeded: fasting 105 mg/dl, 1 hour 190 mg/dl, 2 hour 165 mg/dl, 3 hour 145 mg/dl. RESULTS: 149 patients satisfied the inclusion criteria for the study. 14/149 (9 4%) had an abnormal GST prior to 20 weeks and only 1 of these 14 patients (7.1%) had an abnormal GTr and was subsequently treated for GDM At 24-28 weeks, repeat screening revealed that 21 out of 149 (14%) patients had an abnormal GST and that 5 of these 21 patients had an abnormal GTT Early glucose screening detected 1 out of 5 of these cases for a sensitiwty of 20%. The total cost of the early diabetic screening protocol was $9428. The cost of performing traditional glucose screemng at 24-28 weeks would have been $5274. Therefore, the cost of early identification of 1 case of GDM was $4154 CONCLUSION: Even in this population who was at high risk for developing GDM, the incidence of GDM was extremety tow and the sensitivity ol early screening was poor. Early GDM screening was not cost effective in our population.

Volume 174, Number l, Patt 2 SPO Abstracts 395 Am J Obstet Gynecol

309 LOWERING THE THRESHOLD FOR THE DIAGNOSIS OF

GESTATIONAL DIABETES. O.A. Rust, J.A. Bofill, M. Andrewx, T.

Kincad~, T. Stubbs, E. Millerx, J.C. Morrison. Depts. Ob/Gyn mad Prevemive

Med., Univ. of Mississippi, Jackson, MS and Dept. Ob/Gyn Carolinas Mad.

Ctr., Charlotte, NC.

OBJECTIVE: To determine if lowering the threshold for the diagnosis of

gestational diabetes mal|itus (GDM) will select a population at higher risk for

adverse pcriuatel outcome.

STUDY DESIGN: In this retrospectwe study, 434 patients with an abnormal

50-gm glucose screen underwent a sOmdardizad 3-hour oral glucose tolerance

test (GTT) and were stratified into 4 groups: group 1 (n = 102) had the

diagnosis of GDM by standard criteria, group 2 (n = 71) would have had GDM

if the threshold had been lowered (fasting > 95, 1-hour > 172, 2-hr > 151,

3-hr > 130). Group 3 (n = 78) had one abnormal value on GTF. Group 4 (n

= 183) had a normal GTT. Multiple variables assessing perinatal outcome were

analyzed.

RESULTS: Demographics, risk factors for abnormal testing, route of delivery,

and mitenatal complications were similar for all groups except for age.

Additional findings included.

Variable 1

Age (yr) 25.1

Prepreg BMl 26.6

(kg/m2)

Preg wt gain (lb) 36.8

Birth wt (gm) 3277.3

Neo hypogly (%) 27.5

*NS after Bonferronl correction

Group (mean)

2 3 4 p

25.7 23.7 22.7 0.0001 26.6 25.5 24.8 0.011"

36.4 37.9 35.4 NS

3284.1 3345.4 3314.1 NS

5.6 10.6 10.9 0 0009

All other maternal and neonatal outcome variables were similar for the 4 groups.

CONCLUSIONS: Our results indicate that lowering the absolute GTT

threshold or using one abnormal value of GTT would over-diagnose GDM

without improving perinatal outcome.

311 CAN Wig IMPROVE SCREENING FOR GESTATIONAL DIABETES?

M. SermeP, D. NayloP, D. Farine, K. Ritchie, D. Gore~, H. Cohen. Univ. of

Toronto, ON, Canada

Objective: To compare screening strategies for detection of gestational

diabetes (GDM).

Study Design: In a prospective analytic cohort study 3,836 patients were

each assessed wRh random plasma glucose, a 50 gm Glucose Challenge Test

(GCT), and a 100 ere Oral Glucose Tolerance Test (OGTT). Clinical and

historical risk factors were recorded.

Rt’~ults: In a nlnltlvarlate analysis, age, race, body mass index, and an

abnormal obstetrical history were all significantly and independently

associated with altered risk of GDM. For example, the odds ratio (OR)

were: 1.0, 1.4 and 7.4 for ages <30, 31-34, and >35 respectively. The OR

for BMI of > 25.1 was 3 overall with blacks (OR=I.8) and orientals

(OR=5.8) at increased risk. Receiver operating characteristic (ROC) curve

analys~s showed random plasma glucose was unhelpful, but screening

efficieucy was optimized by combining the above risk factors with the GCT.

Areas under the ROC curve are as follows: standard GCT, 0.791; GCT

adjusted for time since last meal as previously reported from this cohort,

0.804; and risk factors plus adjusted GCT, 0.873. A scoring system was

derived based on the chnical risk factors. Rising scores led to higher

incidences of GDM (p < 0.0001 for trend): among patients scoring 0-2,

1.6% had GDM versus 12.8% among those scoring 7.

Conclusions: Screening for GDM can be Improved by assessing risk factors

thus providing a rationale for altenng screening strategies ie no testiog,GC~l"

or OGnI~r without GCT. Furthermore, combinrag risk factors with the GCT

results wdl improve the risk assessment for individual patients.

310 HOW PREDICTIVE IS THE DEGREE OF ABNORMALITY OF THE GLUCOLA TEST FOR GESTATIONAL DIABETES ? T Svondsenx, A Abuhamad, M de Veoana, J Murganx, A T Evans Dept Ob/Gy~, Eastern Virginia Medical School, Norfolk, VA OBJECTIVE: To determine how well the degree of abnormahty on a 1 hour glucola test predicts the diagnosis of Gestatloual Diabetes [GDM] STUDY DESIGN: Over a 7 year period [1987-93], 26ll pregnant patients were screened for GDM using a 1 hour [50gm] glucola test Patients were usually screened at 24-29 weeks Those with risk factors for GDM were screened at their first prenatal visit Patients with a 1 hour plasma glucose (1 hr PG) >_ 140 mg/dl had 3 hours [100gm] OGTTs GDM was diagnosed if >_ 2 OGTT values were abnormal

(>_105/190/165/145 mg/dl) Senslttvity, spec~ficity and predictive values at various lhr PG value cutoffs were calculated

RESULTS: Mean maternal age was 27 2 +_ 5 8 )Ts [range 14-44] and gestahonal age at screening was 24 3 + 7 0 wks [range 5-37] 38912611 had an abnormal lhr PG [15%] 45/389 were lost to followup, 344/389

[88%] had 3hr OGTTs Statistical parameters for various 1 hr PG cutoffs are presented below 171344 [5%] had at least one value > 250 on their 3hr OGTT, of these, 8117 [47~] had a lhr PG > 220 mg/dl and 6/17 [35%] had a normal FBS No significant antepartum complications occurred m patients with at least one value >250 on OGTT

lhr PG mg/dl Sensitivity Specificity Pos P Val Nee P Val >140 99 3 0 39 0 0 >180 37 0 90 0 70 4 69 0 >190 27 4 95 7 80 4 67 2 >200 20 7 97 6 84 8 65 7 >220 9 6 99 5 92 9 63 1

CONCLUSIONS: A llar PG >190 mg/dl is commonly considered diagnostic of GDM In our population, 20 ~A ofpahents with 1 hr PG > 190 mg/dl had a normal OGTT A llar PG cut off>220 mg/dl appears to be more predictive of GDM m our populatmn

312 CESAREAN DELIVERY IN RELATION TO BIRTHWEIGHT AND GESTATIONAL GLUCOSE TOLERANCE: PATHOPHYSIOLOGY OR LABELING BIAS? M. SermeP, D. Naylor’, D. Farine, K. Ritehie, D. Gates,

H. Cohen. Dept. Ob/Gyn and ICES, Univ. of Toronto, Toronto Ont, Canada Objective: To compare the birth outcomes of women with unrecognized gestational diabetes (GDM), treated GDM, and normoglycemta, with particular reference to birthweight and cesarean section.

Study Design: A prospective analyOc cohort study of 3,778 gravtdus aged 24 years and over, with post hoc case-control comparisons. Subjects undei~vent a 3-hour 100 gm oral glucose tolerance test at 28 weeks gestation, regardless of screening test results. Those meeting the National Diabetes Data Group (NDDG) criteria for GDM (n - 143) received the usual care, consisting of plasma glucose momtoriog, &etsry modification, and insulin where indicated. Physic~aus and patients were blinded to glucose results for all other subjects, mcludieg 115 women with unrecognized GDM by the broader Carpenter and Coustan criteria. Crude and adjusted rates of cesarean section and neonatal macrosomia (4,000 gms) formed the main outcomes of interest. R~sults: Compared to nonnoglycemic coutrols, patients with unrecognized GDM had increased rates of macrosom~a (28.7% vs 13.7%, p < 0.001) and cesarean section (29.6% vs 20.2%, p = 0.016). Usual care of NDDG diabetes normahzed birthweights, but the increased rate of cesarean section among treated patients compared to controls persisted despite adjustment for maternal age, race, parity, body mass index, pre-eclampsia, and gestatioual age (adjusted odds ratio: 2.1, 95% confidence interval: 1.39-3.22). Other maternal-fetal outcomes were similar between treated and untreated diabetic gravidas. Conclusion: Untreated GDM women (based on the broader Carpenter and Coustan criteria) had an increased task of macrosomta and surgical dehvery. Detection and treatment ofgestational diabete s normalized birthweights, but did not otherwise improve maternal-fetal outcomes compared to women with unrecogmzed diabetes. Specifically, rates of cesarean section remained inexplicably high. Recognition of GDM may lead to a lower threshold for surgical delivery that mitigates the potential benefits of treatment.


313 DELAYED PULMONARY MATURATION IS ASSOCIATED WITH POOR GLUCOSE CONTROL IN DIABETIC PREGNANCIES, J P=per, O Langer,

Dept Ob/Gyn, UTHSC, San Antonio, TX

OBJECTIVE: Pragnanoes comphcated by d~abetes have been shown to have delayed fetal pulmonary maturation as measured by both the delayed

appearance of b~ochemlcal indicators of pulmonary matunty (phosphat~dyl

glycerol [PG] and leclthm/sphlngomyehn raho [US]) and the occurrence of hyahne membrane d~seese (HMD) even m term gestations We sought to test the hypothesm that maternal glucose control ~nfluences the occurrence

rate of delayed fetal pulmonary maturation METHODS: Consecutive dmbet~c pregnancies w~th documentahon of

maternal glycemic control and amn=ohc fluid analys=s for PG and IJS were analyzed Maternal glycem=c control was defined as good ff the mean blood

glucese (MBG) was ~105 mg/dL and poor ~f >105 mg/dL Amn=ohc flu=d PG was considered mature =f present HMD was defined by well estabhshed cntena

RESULTS: 621 dmbet=c pregnanc=es were analyzed (261 good control, 360 poor control ) Overall PG was absent ~n 21% of good control vs 31% of

poor control pregnanc=es (p<0 005) When strat=fied by gestat=onal age, the nsk of an =mmature PG was sign=ficently h~gher ~n the poor control group

% Immature PG Overall O R 1 83

<34 34-36 9 37-37 9 38-38 9 z 39 Wks

Poor Ctfl 92% 58% 34% 18% 17%

Good Ctrl 92% 36% 23% 13% 8%

n 24 90 172 213 122

At 36-37 9 weeks, the control pregnanoes had s=gmficantly Ngher rates of

immature PG [37% vs 22%, O R 2 04 (1 1-3 9)] All cases of HMD were ~n poor control pregnancies <37 weeks (3/69, 4 3%) There were no cases

of HMD beyond 37 weeks gestation CONCLUSION: Abnormal maternal glucose levels are assocmted w~th delayed appearance of PG m diabetm pregnancies, however, beyond

37 weeks gestahon no s~gnff~cant neonatal pulmonary d~sease occurred

315 3-YEAR EXPERIENCE WITH INSULIN PUMP THERAPY

DURING PREGNANCY. R. Silverman, R. Artal. Division of

Maternal-Fetal Medicine, SUNY Health Science Center, Syracuse, NY

OBJECTIVE: To determine whether therapy with constant

subcutaneous insulin infusion pump throughout pregnancy results in less

daily capillary glucose variablhty.

STUDY DESIGN: Eighteen pregnant patients (White’s Classification

Class B through R) that were placed on insulin pumps during their three

trmaesters of pregnancy were compared to 18 diabetic controls matched

for age, race, gravida, duration of diabetes and compliance to care

Premeal and hs capillary glucose were averaged per week of gestation

and compared across both groups.

RESULTS: In the first trimester, predinner/hs glucose were lower for

pump versus control patients (118 versus 154 mg/dl, 113 versus 146

mg/dl, p <0.05). In the second trimester, fasting and hs capillary

glucose were lower for pump versus control patients (110 versus 143

mg/dl and 125 versus 152 mg/dl, p <0.05). In the third trimester, fasting

capillary glucose were lower in pump versus control patients (109

versus 125 mg/dl, p <0.05). Averaging all trimesters, the fasting,

predmner and hs capilla£� glucose were lower for the pump versus

control patients (120 versus 132, 124 versus 135 and 121 versus 144

mg/dl, p <0.05). All other values were nonsignificant between insulin

pump and control patients. The mean HbA1C, was lower in the second

and third trimesters for pump patients (5.0 versus 6.6, and 4.6 versus

6.0, p <0.05).

CONCLUSION: Use of constant subcutaneous insulin infusion pumps

results in lower capillary daily glucose variability as reflected by: (1)

Lower mean capillary glucose (2) Lower mean HbA1C during the

second and third trimesters.

314 GLYCEMIC CONTROL IN PREGESTATIONAL DIABETICS

INFLUENCES THE INCIDENCE OF PREECLAMPSIA. A Mentak~s* J. Kuboshige*, C R. Brinkman IIl*, M.T. Cabalum, Dept. ofOb/Gyn, Harbor-UCLA

Medical Center, Torrance, CA OBJECTIVE: To assess whether glyeemJc control in the second and third

trimesters of pregnancy ~nfluences the inctdence of preeclamsia

STUDY DESIGN’ Medtcal records of 108 pregestational diabetms (PGDM) seen

from 1982-1994 were reviewed. Second and third trimester fasting (FBS) and two

hour postprandial (2pp) blood sugars were evaluated The dmgnoms of

preeclampsta (P/E) was made atter 20 weeks based on blood pressures (BP)

>140/90 and urine protein >1+ on a cathetermed specimen Good glycemtc control

was presumed to be present ff the FBS <105 mg/dl and the 2pp<120 mg/dl

Renal/vascular disease was defined as the presence of >300 mg of protein in a 24-

hour urine specimen, or creatimne clearanc.e <120 ml!m~n, or evidence of dtabettc

retmopathy, or history of chronic hypertensmn based on BP>140/90 before 20

weeks. The associatmn between glyeemlc control and preeelampsm was evaluated

by cht-square and FIsher’s exact analysis RESULTS The incidence of P/E In the 108 patients w~th PGDM was 20 6%

Glyeem~c control dunng the second (23.9% vs 21A%, p=0.45) or third trimester

(18 1% vs 37 5%, p=0.33) d~d not influence the incidence of P/E. These patients

were strattfied in two groups (1) Group A (N=58) with evidence of renal/vascular

disease, and (2) Group B (N=50) without evidence of renal/vascular involvement.

In Group A, glyeem~c control in either the second(42A% vs 23 3 %, p=0 33) or

third trimesters (42 8% vs 24 4%, p=0 15) did not influence the incidence of P/E

In Group B, glucose control in either the second (12.5% vs 10 5%,p-0 99) or third trimesters (23 3% vs 10.1%,p=0 09) was not associated with an increase in

the ~nctdence of P/E However, when the FBS and 2pp were analysed

independently, paUents m Group A had a slgmficanfly higher incidence ofP/E tf

the third trimester 2pp>120 mg/dl (53 3% vs 22 2,p<0 05) Addltionallly, patients

m Group B with FBS>105 mg/dl during the thtrd trimester had a stgmficantly

h~gher inctdence of preeclampsia (50% vs 4.4%,p<0 005)

CONCLUSION (1) Glycemlc control during the second and third trimesters does

not influence the incidence of P!E in PGDM with and without ewdence of vasculopathy (2) There ts an association between thtrd trimester mean FBS and

the occurrence ofP/E m PGDM wtthout vascular dteease, and third trimester mean

2pp sugars in PGDM patients with vasculopathy.

316 RECURRENCE OF GESTATIONAL DIABETES MELLITUS:

IDENTIFICATION OF RISK FACTORS. CY Spong, L. Guillermo’,J.

Kuboshigex, T. Cabalum. Dept ofOb/Gyno Ha~or-UCLA Medical Center,

Torrance, CA

Objective: To evaluate the influence of certain maternal and neonatal

factors on the recurrence of gestational diabetes (GDM).

Study Design: A study was conducted on 164 predominantly Hispanic

patients whose index pregnancy was complicaLed by GDM and whose

subsequent consecutive pregnancy was managed at our institution between

January 1988 and December 1992. The diagnosis of GDM was based on

the criteria recommended by the National Diabetes Data Group using a 100

g oral glucose tolerance test.

Results: One hundred eleven (68%) of the 164 women had recurrence of

GDM. Fifty-threo (32%) did not demonstrate recurrence in their

subsequent pregnancy. Patients with recurrence had GDM diagnosed

earlier (30.3 vs 32.5 weeks, p=0.03)), frequently required insulin (25%

vs 8%, p= <0.05) and had more hospital admissions (32% vs 10%,

p<0.05) in their index pregnancy compared to women who did not have

recurrence of GDM. Women who had recurrence had elevated mean third

trimester plasma glucose values: fasting 87.6 vs 83 mg/dl, (p =0.009) and

2-hour post-prandial 109.7 vs 102.2 mg/dl, (p=0.008). Neonates of

patients with recurrence were heavier (3656 vs 3373 gin, p=0.004) and

had increased incidence of macrosomia (26% vs 10%, p<0.05). No

significant differences were observed in maternal age, prepregnancy body

mass index, Hgb A1 C, second trimester plasma glucose levels, incidence

of shoulder dystocia, and apgar scores between the two groups of women.

Coneluslon: Patients with history of GDM have significant risk of

recurrence in their subsequent pregnancy. The risk for recurrence in

women is increased if GDM is diagnosed earlier, they require insulin,

have elevated third trimester plasma glucose level, and deliver

maernsomic infants in their index pregnancy.

Poster Session III

Friday, February 9, 1996 10:00 a.m.- 12:00 noon

Kohala Ballroom

CATEGORIES Neonatology

Infectious Disease Diagnostic Ultrasound


Judges: Brian M. Mercer, MD Richard L. Berkowitz, MD Ronald S. Gibbs, MD

Volume 174, Number l, Part 2 SPO Abstracts 399

Am J Obstet Gynecol

317 RELATIONSHIP BETWEEN NUCLEATED RED BLOOD

CELL (NRBC) COUNTS AND INTRAVENTRICULAR

H~MORRHAGE (1VH) AND PERIVENTRICULAR

LEUCOMALACIA 0aNtL) E Lefl~m. U. Verma, S. Kleinx,

N Tejani NY Med Coil Valhalla, N Y

OBJECTIVE: Recent reports have associated increased NRBC

counts wath intrauterine growth restriction (FUGR) and perinatal

asphyma. We exurnmed the relslaonslup between NRBC counts and

other markers of perinatal hypoxas, 1VH and PVL in first week of

life The null hypothesis is that there is no &fference in NRBC counts

m infants who do and in those who do not develop IVH and PVL. STUDY DlgSIGN: NRBC counts were obtained in 441 infants

weighing between 500 and 1750 grams Neonatal eramal funtanel

sonography was performed on the 3rd and 7th day of life and neonates

&vided into those with normal and those with abnormal findings.

NRBC counts were compared between the groups mad between

normal and abnormal groups matched for birthwmght per-e~lile (BW

%11e) and gestational age (CA). T-statistics were used for continuous

variables,the Cba-square or Fisher exact test for categorical variables

RESULTS: Mean CA, BW %de, Apgar score, re~ratory distress

syndrome and mortality were afferent between infants wath normal

and those with abnormal ultrasounds. The NRBC counts, however,

were not different (40 5_+126 8 vs 41.8 + 71 7 t =0.009 p= 0.9); even

when results were controlled for GA and BW, There were significant

elevalaons in NRBC counts m IUGR infants.

CONCLUSIONS: NRBC counts are not dflTerent in preterm Infants

with or without IVH and PVL Therefore early IVH and PVL may be

caused by results other than hypoxia Birthwmght percentiles must be

considered when using NRBC counts as a marker of intrauterine

hypoxi..

319 HIGH EXPRESSION OF INTERLEUKIN-6, INTERLEUKIN-I[L AND TUMOR NECROSIS FACTOR-~ IN PERIVENTRICULAR LEUKOMALACIA. B.H. Yoon, R Romere, C J I~n~, K H. Parkx, S.A. Ho~gx, J.K. Jun~, F. GhezzP, H.C. SynL Seoul National University, Seoul, Korea, and the Peri~atology Research Branch, NICHD, Bethesda, MD. OBJECTIVE: Perivenmcular leukomalacia (PVL), a common bram wtute matter lesion of pret~rm neonates, is a major risk factor for cerebral palsy. Recently, PVL has been associated with sub-clerical chonoamnionitis and cytokines have been ~mplicated as mediators for neuronal injury in tins canthtion (DMCN 1993;35:553). To elucidate the relatlonshipbetween the local expression of eytoklnes and PVL, we studied brain lesions to determine i/they over-express interleukin-6 (IL-6), interleukth-lll (IL-II~), and tumor necrosis factor-~ O’NF-n). STUDY DESIGN: Immtmolfistoehemical staining for eytokine expression (IL-6, IL-11~, and TNF-~) was performed in 10% formalin f~xed paraffin- embedded brain sections of 18 cases with PVL. Bram~ from autopsy specimens performed between 1987 and 1994 were studied. The median duration of survival was 5 days (range 1-160 days). Ten cases without white matter lesions who died during the perinatal period were used as "controls ". RESULTS: Cytoklne expression was demonstrated in the cytoplasm of hypertrophic astroeytes in 94% (16/18) of cases with PVL. The astrocytes were also strongly stained with anti-GFAP monoclonal antibody. Neurons, oligodendrogha, and microglial ceils were devoid of cytokme expression The expression of IL-6, IL-11L and TNF-= was identified in 78% (14/18), 28% (5/18), 83% (15/18)cases of PVL, respectively However, neither IL-6 nor IL-lt~ were detected in histologic sections of the control group. In contrast, TNF-n was weakly positive in 5 of 10 cases without PVL. CONCLUSIONS: 1) There is over-expression of IL-11~ and IL-6 in the astrocytes in PVL; 2) Tiffs finding provides strong support for the hypothesis that PVL ts the the result of cytokine-mediated neuronal injury.

318 ANTEPARTUM MATERNAL STEROIDS AND THE RISK OF INTRAVENTRICULAR H~MORRHAGE AND PERIVENTRICULAR LEUCOMALACIA. Cantermo jX,Verma U,Tejam N, Klein SX,Reale IVP,

Jeanty !vP New York Medical College, Valhalla NY. OBJECTIVE The NIH task force has shown that antenatal steroids reduce the incidence and seventy of RDS and mtraventncular hemorrhage (IVH) We have shown that preterm labor (PTL) and premature rupture of

membranes (PROM) mcrense the mcidence and seventy of IVH and penventncular leucomalacia (PVL) Tlus study evaluates the effects of antenatal steroids on the risk and severity of IVH] PVL STUDY DESIGN: 745 consecuhve mbom neonates wath Bthwt < 1750 gms were evaluated by cramal sonography on days 3 and 7 IVH was graded according to Papille. Major abnormalities included grades 3 and 4 IVH and PVL The neonates were drnded into 1 ) PTL/PROM and 2) MFI group consisting of patients delivered for maternal-fetal radicalness Groups were further subdivided based on steroid usage The subgroups were compared

for the risk and seventy of 1VH/PVL Finally neonates vnth and w~thout antenatal stenod exposure were compared,matclung for CA, Bthwt and maternal group Flsher’s test was used RESULTS: The PTL/PROM group had a s~gnlficantly lugher incidence and seventy of 1VIJdPVL P < 0001 CRANIAL SONOGRAPItY

STERIODS IVfUPVL MAJOR IVtUPVL # # #

PTL/PROM YES 78 16 4 # 564 NO 486 149" 67

MFI YES 33 6 1 # 181 NO 148 26 1

*p<05 Comparison of matched groups showed that stenods decreased the incidence of WH~VL

CONCLUSIONS: Antenatal stenods slgmficantly reduced the incidence of IVH!PVL m the PTL/PROM group This neuroprotective effect was not seen m the MFI group

320 THE EFFECT OF INTRAPARTUM STRESS IN NORMAL PREGNANCY ON NEWBORN LIPIDS. C R. Parker, Jr~ Dept. OB/GYN, Univ Alabama at Birmingham, Birmingham, AL OBJECTIVE: Infants of women with severe pregnancy compticattons often have increased total- and LOL-cholesterol (C) levels. Since such infants often have low dehydroep~androsterone sulfate (DS) levels, hypercholesterolemia could be due to reduced adrenal LDL-C assimdation for steroldogenes=s. Since chromcally compromised infants also often have acute peripartum stress, we sought to determine whether fetal stress dunng labor ~n normal pregnancies might alter newborn lipid proNes. STUDY DESIGN: We quantified total C, triglycendes (TG) and DS ~n umbilical cord blood of 102 normally developed term infants of women wRh otherwise uncomplicated pregnancies who were judged to have experienced severe stress (abnormal fetal heart rate patterns and/or academia) during labor/delivery These infants

were matched on the bas~s of gestafional age and delivery method to 102 nomstressed infants of women with normal pregnancies. RESULTS: TG levels of stressed infants (0.53 + 0 20 mmol/L, Mean ± SD) were higher (P=O.O08) than those of the controls (0.45 ± 0 16 mmol/L), whereas C levels in the stressed (1.89 + 0 47 mmol/L) and control =nfants (1.83 + 0.41 mmol/L) were similar. Although DS levels in stressed ~nfants were shghtly subnormal (P=0.06), DS levels (,umol/L) in both stressed (3.22 1.18) and control (3.93 ± 1.56) infants whose C levels were high (>75th%) were strikingly lower than those of the entire control group (4 75 + 2 02). DS levels were normal in infants of both groups having TG levels >75th%. CONCLUSION: Penpartum stress leads to increased TG levels but has no overall effect on plasma C levels. Also, relabve hypercholesterolemia ~n newborn ~nfants may frequently be due to reduced adrenal uptake of lipoproteins for use in steroidogenesis.

400 SPO Abstracts January 1996 An, J Obstet Gyneco/

321 ALCOHOL AND CIGARETTE USE INCREASES RISK OF PNEUMONIA

DURING GESTATION. J E. Wh=tty~ M P. Dombrowskb S.S. Martler*,

R J. Sokol. Department of Ob/Gyn, Wayne State Unwerslty School

of Med=cme, Detrmt, MI

OBJECTIVE: Alcohol abuse is reportedly associated with

immunosuppression. We sought to determine ~f women who

abused alcohol during gestatmn had ~ncreased infectious morNdity.

STUDY DESIGN: Gravidas enrolled in the Fetal Alcohol Center

(FARC) data base between 1986-1992 were examined .for the

incidence of HIV infection, pneumoma, endometritzs and

pyelonephnt=s. FARC controls were comprised of gravidas

without infectious complications. Alcohol consumptmn was

determined by screeners skdled at illic~tmg drug and alcohol

histories. Stepw~se discrimmant analysis was performed for each

infectious morbidity using 11 risk factors including alcohol, illicit

drug use, cigarette smoking and medical comphcations.

RESULTS: Of 21 25 gravidas examined, 22 had HIV ~nfection, 550

had endometnt~s, 225 had pyelonephritis and 93 had pneumonia.

Absolute alcohol use per dnnk~ng day and number of cigarettes

smoked per day were significant ~ndependent predictors of

pneumonia during gestation and the postpartum penod. (F(22741 =

8.37; (R2 = 6%; p < .OO1). There was no s~gnificant correlatmn

between HIV ~nfection, endometritis or pyelonephrit~s and alcohol

or any of the other vanables examined.

CONCLUSION: B~nge drinking and mgarette smoking appear to be

significant nsk factors for pneumonia during pregnancy. However,

our data do not support immunosuppression as a mechanism for

the increased risk. We speculate that the most hkely mechanism

is aspiration and/or pulmonary pathology.

323 IMMEDIATE POSTPARTUM BACTEREMIA FOLLOWING

VAGINAL DELIVERY R Minedd MD.x Dtt Watts MD,

KA Boggass MD,x TJ Benedetti MD, DA Eschenbach MD.x University of Washington, Seattie. OBJECTIVES: Current American HeM Association recommendations for prophylaxis for bacterial endocarditis do not include nse of antibiotics in patients with heart valve abnormalities undergoing vaginal delivery. Our objective was to evaluate the frequency, risk factors and microbiology of baeteremia following vaginal delivery and compare this rate to bacteremla associated with cesarean section following labor. METHODS: Eligible I~tients included all those delivering vaginally who signed informed consent. Blood enituras were obtained for facnitetive and anaerobic bacteri~ within f’ffteen minutes of placental delivery. Chorioamnintic membranes from placentas of these patients were also cnitared. Demographic, labor, delivery and postpartum chacaeteristics were abs~acted from the medical records. Comparisons between groups were made using chi square analysis. RESULTS: To date, bacteremia has been detected in 5(9%) of 54 p~ieats delivering vaginally. All were between 36-43 weeks, ranged in age from 13-40 years, and had a mean infant weight of 3318g. Isolates included Prey. bivia (1), group B Streptococcus (2), anaerobic Gram positive cocci (2). The r~k of postpa~m bacteremia was not significantly related to duration of ruptured membranes, vaginal GBS culture result~ placental cnimre or use of intrauterine monitoring. 6 (11%) of 54 placentas were cniture-poaltive, and only one of these was from patients with bacteremla. CONCLUSIONS: In this ongoing study, postpartum bacteremia was not uncommon among women undergoing uncomplicated vaginal delivery at term. The bacteremla rate of 9% is similar to that in our previous study of cesarean section alter labor(14%). Clinical characteristics were not predictive of bacteremi~. Thus prophylactic antibiotics should be considered in those at risk for endocarditis during vaginal delivery.

322 VAGINAL AND CERVICAL FLLqD SIALIDASE ACTMTY ASSOCIATED WITH

(~ERVICOVAGINAL MICROORGANISMS AND PRETERM LABOR L Pennacc/n~,

G Coata~, P DeDomentco’, A Sensini~°, M ~VIarangt~, C Tass~~, G C Dt Renzo Inst of Ob/Gyn and °Dept of Exp, Med & Blochem Science, Umv of Perugla. Perugm, Italy

LNTRODUCTION: Mucolyhc enzymes incluthng muclnases and slahdases are recngmzed virulence lhctors an’~ong enteropathogens and vaginal mlcrocrganlsms Alterations of maternal cervico-vagmal mucosal membrane host defenses by such enzyme producing organisms may increase the ask of subclinical chonoanmlomtls m pregnancy, and of preterm labor PATmm’S A~ M~THODS: We prospectively evaluate cervical and vagmal fluid slahdase activity and selected cervical, vaginal and ~trethral bacteria alung with pregnmacy outcome in 50 pregnant women Slahdase activity was evaluated separalely m cervical and vaginal fluids by an ongmal method modified fi’om Tassl et al (Clin Chum Acta, 206 231, 1992) utihzmg 2L(4-

methyhunbelhfervl) et-D-Nl acetylueurammic acid as fluorescent substrate Activity was nonnahzed to protem contant RESULTS Presence of cervlcovagmal bacteria was associated with lllcreased ask of preterm labor and/or PROM (R R 3,2, C I 1 2 to 6 1, p <0 001) S~ahdase specific activity esther in cerwcal or m vaginal fluid was associated with bacterial organisms (Gardnerella vagmahs, Chlamydia tracbomatls, Streptococcus agalact~ae, Ureaplasma urealyticmn and others) and was found to be slgnfficantly higher compared to negative cervical and vagmal specImans (vaginal 6 0~:2 1 vs 1 2 ± 1 0, cervical 6 8±2 2 vs 1 22±0 9 mnol/mln/pg-1 protein, mean ± SD, p

<0 001) CO,~CLtlSIOn Negative microbiological specimens with high s~ahdase activity were related to a higher risk of preterm labor compared to negatxve spec~meiis vnth low smhdase activity (cut off 2 5 nnml/mm/gg4 protem) Smhdase producmg cervlco-vaginal microorganisms and increased cervical and/or vaginal fluid s~alidase activity are ask ~actors for possible preventable subchmcal chonoanm~omtls and preterrn labor H~gh sIalidase achvity in cervical and/or vaginal fluid maybe a marker of increased ask of preterm labor independently from positive microbiological cervIco-vagmal smear and identification of specific mxcr~rganIsms

324 VAGINAL SECRETORY IgA LEVELS IN NORMAL NON- PREGNANT AND PREGNANT WOMEN. ~M, Karoel. CN. Codipfll~, A. Vomero~, RK. Jaekle, DA. Baker. Division of Maternal-

Fetal Medicine, University M~ical Center, SUNY Stony Brook, Stony Brook, NY OBJECTIVe: lh’ior work has demonstrated local production of Secretory IgA (SIgA) in tissues of the female genital tract. The goal of this study is to establish SIgA levels in normal non-pregnant and pregnant

STUDY DESIGN: Eleven non-pregnant, non-smoking, condom using women with regular menses, no hormonal therapy, no vaginitis were studied on day 7, 14, 28 of their meustmal cycle. Twelve pregnant, nonsmoking woman with no history of pre-term labor, no vaginifls and no recent intercourse were stu&ed once Sterile water (0.5 cc) was placed

into the vagina and then aspirated ~ the walls were tabbed with a

sterile swab SIgA levels were measured from the aspirate using a sendwich ELISA. Cultures for gonorrhea and chlamydia, whiff test, wet prep, vaginal pH end gram stain were obtained and were all negative Statistical analysis was performed using T-Test and A_NOVA as appropriate. The data are presented as mean + S E. and the null hypothesis was rejected at p <0.05. RESULTS: Mean SIgA levels in non-pregnant women: day 7 [0 311 + 0.059 pg/mL], day 14 [0 405 + 0.104 gg/mL], day 28 [0.336 _+ 0.074 pg/mL]. Mean SIgA level in pragnant women: 2.036 +0.454 pg/mL. There is no statistically significant difference between SIgA levels throughout the menstrual cycle. SIgA levels are significantly higher in

pregnant women compared to non-pregnant women at each point in the cycle (p <0.001).

CONCLUSION: Vaginal SIgA levels can be measured in pregnant and non-pregnent women using a sandwich ELISA technique. SIgA levels are significantly higher in pregnant women compared to non-pregnant

Volume 174, Nmnber 1, Part 2 SPO Abstracts 401 Am ] Obstet Gyneco]

325 MATERNAL SERUM INTERLEUKIN-6 LEVELS ARE ELEVATED IN PATIENTS WITH PRETERM PREMATURE RUPTURE OF MEMBRANES AND EVIDENCE OF INFECTION. A. MurthaXl, P Greig1, C. J~mmerson1, W. Herbert1, B. Roitman-Johnson2, J. Allen2, Dept Ob/Gyn, Duke Un~vers~ly1, Durham, NC and R&D Systems2, Mpls, MN. Objective: To determine ~f maternal serum Interleukln-6 (IL-6) levels are elevated m patients with preterm premature rupture of membranes (PPROM) and ewdence of intrauterine infection. Study Design: Serum samples were obtained from preterm patients (22-34 weeks) w~th the following conditions, nonlaboring controls (n=59}; PPROM w~th Nodal samples drawn >48 hours pnot

to delivery w~thout evidence of infection (n=50); and PPROM with evidence of infechon as defined by climcal criteria and/or histologic chorioamn~onitis who had samples drawn within 24 hours pnor to

delivery (n=23). Maternal serum IL-6 levels (pg/ml) were determined using a specific ELISA kit (R&D Systems), The Mann-Whitney U test was used for statistical analysis. Results: Serum ~L-6 was significantly higher m PPROM patients drawn on the day of dehvery with evidence of ~nfection compared to the preterm, nonlabonng controls ( median- 15 vs 2, p=.0002). These PPROM patients with ~nfection also had s~gnificantiy higher IL-6 than PPROM patients whose samples were drawn >48 hours prior to delivery and were w~thout signs of ~nfecUon (median-15 vs 2, p<0 0001). A serum IL-6 >8 pg/ml had a sensitivity of 82%, specificity of 98%, PPV of 95% and NPV of 92% for identifying intrautenne ~nfecUon in PPROM patients on the day of delivery Conclusions: Maternal serum IL-6 increases in the 24 hours prior to dehvery ~n PPROM patients w~th signs of clinical ~nfechon or histologic chorioamnion~tis An elevated serum IL-6 level is a sensitive and specific marker for the identification of intrauterine refection in patients with PPROM

327 PRETERM PREMATURE RUPTURE OF MEMBRANES. A PROSPECTIVE, RANDOMIZED, PLACEBO CONTROLLED TRIAL OF ANTIBIOTIC TREAqTviENT. A.Ovalle~, MA Martfnez~, R.Gomez~, O.Valderramax, P Lirax, R Rubiox, A.FuentesL F.GhezzP, R.Romero. Department of Obstetrics and Gynecology, Hospital San Borja Arriaran/University of Chile, Santiago, Chile, Perinatolgy Research Branch, NICHD, Bethesda, MD OBJECTIVE: The role of antibiotic administration in patients with preterm PROM remains controversial. The purpose of this study was to determine the effect of antibiotic administration on the duration of pregnancy and perinatal outcome in patients with preterm PROM. STUDY DESIGN: Eighty women with preterm PROM were randomly allocated to either antibiotic administration (clyndamycin + genmmicin) (n=38) or placebo (n=42) and outcome determined. Fisher’s test and non parametric statistics were used for analysis. RESULTS:

I Antibiotics I P~cebo Ipval"e (n=38) (n=42)

Randomization to 11 (0-41) 4 (0-32) < 0.05 delivery interval (days)

Maternal infe,_’tion 1 (2.6%) II (26.2%) < 0.01

related morbidity

Admission to NICU 21 (55.3%) 36 (85.7%) < 0.01

Respiratory Distress 4 (10.5%) 13 (31%) < 0.05

Syndrome

lntraventricular 3 (7.9%) 7 (16.7%) NS Hemorrhage

CONCLUSION Antibiotic administration to patients with preterm PROM prolongs the duration of pregnancy, reduces the rate of maternal infection- related morbidity, NICU admission rate and frequency of respiratory distress syndrome.

326 TH~ EFFICACY OF TREATING UREAPLASMA UREALYTICUM IN PATIENTS WITH PRETERM LABOR OR PRETERM PREMATURE RUPTURE OF MEMBRANES K K Ogasawara, T M Goodwin Umverslty of Southern Califorma School of Medicine, Los Angeles, CA

OBJECTIVE: To test the hypothes~s that prophylactic erythromycm will alter the vertical transrmssinn rate of Ureaplasma urealytacum and v~ll improve maternal and permatal outcome STUDY DESIGN: 51 pahents with singleton pregnancies between 20

and 36 weeks gestatmn v~th preterm premature rupture of membranes or preterm labor were prospectively stud~ed Each patient was randomly

assigned to receive either erythromycin base 500 mg orally every 6 hunts

for 7 days or no treatment. Vaginal and cervical U urealyticum and group B streptococcus cultures were obtained Repeat cultures were done at the tune of delivery or after completing the 7 days of antibiotics Other standard treatments for preterm labor were not changed Continuous variables were compared by parred t test and categorical variables were compared by chi square analysis and Fisher exact test

RESULTS: Both groups were similar in baseline characteristics Lower gemini colonization wth U. ltrealyt~cum was 65% and group B streptococcus was 9 8% Vertical transnusslon of U urealytlcuin was

27% (3/11) In the erythroinycin group and 23% (4/17) for the controls Of the 12 patients colonized with U urealyt~cum 11 were still coloinzed after treatment Although not statistically sigmficant, the mcxdence of

histologic chorioamnionitis in pataents colonized wath U urealyt~cuin was lower in the erythromycin group (25%, 3112) compared to the controls (59%, 10/17) p = 0.076. The latency period was s~gnifieantly prolonged in the erythromycin group (19 4 days) compared to the controls (3 4 days) p=0031.

CONCLUSIONS: Prophylactic use of erythroinycin does not decrease vertical transnusslon of U. urealyticinn However, ~t may decrease the incidence of histologic chorioamniomtis and increase the latency period to dehvery

328 A COMPARATIVE STUDY OF THE DIAGNOSTIC AND PROGNOSTIC

VALUE OF MATERNAL SERUM C-REACTIVE PROTEIN, WHITE

BLOOD CELL COUNT AND AMNIOTIC FLUID WHITE BLOOD CELL

COUNT IN PATIENTS WITH PRETERM PREMATURE RUPTURE OF MEMBRANES. B.H. Yoon, R. Romero, K.H. Park~, LK. Jun~, S.H. Yangx,

C.1. Kim~, ]~,j~llZ,.~*, H.C. Syn~ Seoul National University, SeouI, Korea,

and the Perinatology Research Branch, NICHD, Bethesda, MD.

OBJECTIVE: Several amniotic fluid (AF) tests have been proposed to be of

value in the identification of patients at risk for intrauterine infection and/or

adverse outcome. However, these tests require the performance of

amniocentesis. Our objective was to compare the diagnostic and prognostic

performance of less invaslve indicators of infection such as maternal blood C-

reactive protein (CRP), and wMte blood cell (WBC) count with that of AF-WI3C

count m patients with preterm premature rupture of membranes (PPROM).

STUDY" DESIGN: Maternal blood was obtained for the determination of CRP

and WBC count at the time ofamnioeentesis in 90 patients with PPROM AF

tests included culture and WI?,C count determinations. Receiver-operator

characteristic (ROC) curve, logistic regression, and survival techniques were

used for statistical analysis.

RESULTS: 1) Patients with positive AF cultures had significantly higher

median CRP concentration, WI?,C count, and amniotic fluid WBC count

patients with a negative culture (p < 0.05 for each); 2) ROC curve and survival

analysis demonstrated that the results of CRP, WBC count, or AF-WBC count

were strongly associated with the likelihood of positive AF culture, shorter

amn:ocentesis-to-delivery interval, clinical chorioammonitis, and neonatal

morbidity (19 < 0 05 for each); 3) Of all tests, AF-WBC (~ 20 cells]ram~)

determination was the only s~gnifieant independent predictor of acute histologic

chortoamnionitis (odds ratio: 74), positive AF culture (odds ratio: 22), interval-

to-delivery (hazards ratio: 2.4), and neonatal morbidity (odds ratio: 4.3) after

other confounding variables were adjusted (p < 0.05 for each).

CONCLUSIONS: 1) An elevated maternal serum CRP, WBC count, or AF- WBC count identified patients with intrauterine infection and those at risk for

adverse outcome; 2) AF-WBC count was the only significant independent

predictor of these outcomes in patients with PPROM.


329 THE FREQUENCY AND CLINICAL SIGNIFICANCE OF FETAL LEUKOCYTOSIS/LEUKOPENIA IN PATIENTS WITH PRETERM PROM. B H. Yoon, R. Romero. J.K. Jun’, K H. Park’, S.H Yang~, J.O Kim’, R.Gomez~, H.C. Syn¯ Seoul National University, Seoul, Korea and the Perinatology Research Branch, NICHD, Bethesda, MD

BACKGROUND /OBJECTIVE: A wbate blood eel1 (WBC) count is widely used test in adult medicine to assess the likehhood of health and disease (t.e., infection, stress, etc.). The purpose of this study was to determine the frequency and significance of leukopenia and leukocytosis in fetuses with preterm PROM STUDY DESIGN: Cordocenteses and anmiocenteses were performed m patients with preterm PROM and without labor. IRB approval and written informed consent from all patients were obtained. Ammottc fluid (AF) was cultured for aerobic and anaerobic bacteria and Mycoplasmas. Fetal leukocytosis was defined as a WBC count above the 95th centlle for gestational age and leukopenia as a WBC count below the 5th centile. Survival analysis and non-parametric statistics were used. RESULTS: 1)The prevalence of positive ammotic fluid cultures was 26 % (6/23); 2) Fetal leukocytosis was present in 22% (5/23) and leukopema in 9% (2/23). 3) Fetuses with abnormalWBC had a significantly lugher rate of positive AF culture than those with normal WBC [71% (5/7) vs 6% (1/16), relative risk = 11, CI = 1.6-80.7, p < 001]; 4) The cordocentesis-to-delivery interval was significantly shorter m patients with an abnormal fetal WBC count than in those with normal WBC count [median 24 hours; range 5-100 vs. median 130 hours; range 10-1586, respectively, p < 0.02]. CONCLUSIONS: Fetal leukocytosis/leukopeula are associated with lntraamniotic infection (Relative Risk: 11) and the impending onset of spontaneous preterm parturition.

331 MICROBIAL INVASION OF THE AMNIOTIC CAVITY IS ASSOCIATED WITH INCREASED CONCENTRATIONS OF AMNIOTIC FLUID INTERLEUKIN-6 SOLUBLE RECEPTOR. F. GhezzP, J. Cohenx, J.E. Tolosa, R. Come#, M MazoP, S. Berry,

R Romero. Department of Ob/Gyn, Wayne State University/Hutzel Hospital, Detroit, MI, and the Perinatology Research Branch, NICHD, Bethesda, MD. BACKGROUND/OBJECTIVE: Interleukin-6 (IL-6) is a pleiotropic

cytokine which plays a central role in the regulation of the inflammatory and immune response. Recently, a soluble form of the IL-6 receptor (slL-6R) has been identified. This form of the receptor may participate m the physiologic control of IL-6 bioactivity. This study was conducted to investigate the effect of microbial invasion of the amniotic cavity (MIAC)

on anmiotic fluid (AF) IL-6 and slL-6R concentrations. STUDY DESIGN: A cross sectional study was designed to include women with preterm labor with and without MIAC. Amniotic fluid (AF) was obtained by amniecentesis from patients in preterm labor (PTL) (n~44) and cultured for aerobic and anaerobic bacteria as well as Mycoplasmas. MIAC was defined as a positive AF culture for micro-organisms. AF IL-6 and slL-6R were measured by ELISA. RESULTS: 1) IL-6 and slL-6R were detectable in all AF samples 2) Patients with MIAC (n=21) had a significantly higher AF IL-6 and slL- 6R than those wtthout MIAC (n=23) (IL-6: MIAC, medtan 31500 pg/mi, range 7797-99000 vs. no MIAC, median 1961 pg/mi, range 802.9-4241,

p < 0 01; slL-6R: MIAC, median 1893 pg/ml, range 531-2090 vs. no MIAC, median 1095 pg/mi, range 488 2-2400, p < 0.05). 3) A significant correlation was found between AF IL-6 and slL-6R (Spearman’s r=0.49, p < 0.01). CONCLUSIONS. 1) Microbial invasion of the amniotic cavity is associated

with an increased concentration of both AF IL-6 and the slL-dR. 2) A significant correlation exists between the levels of this cytokine and its soluble receptor in women with preterm labor.

330 A COMPARATIVE STUDY OF THE DIAGNOSTIC AND PROGNOSTIC

VALUE OF CLINICAL SIGNS, MATERNAL WHITE BLOOD CELL

COUNT, C-REACTIVE PROTEIN AND AMNIOTIC FLUID ANALYSIS

INPRETERM LABOR. B.H. Yoon, S.H. YangL J.K JunL K H Park*, C J

Kim~, H.C Synx, J.Cohenx, R. Romero, Seoul National University, Seoul, Korea, and the Perinatology Research Branch, NICHD, Bethesda, MD

OBJECTIVE: Several amniotxc fluid (AF) tests are of value in the

identification of patients with intrauterine infection and/or at risk for adverse

perinatal outcome. However, these tests require the performance of an amniocentesls. Our objective was to compare the diagnostic and prognostic

performance of less invaswe indicators of infection such as maternal blood C-

reactive protein (CRP), white blood cell (WBC) count, and body temperature

with that of AF-WBC count in patients with preterm labor and intact

membranes.

STUDY DESIGN: 102 patients with preterm labor and intact membranes were

studied. Maternal blood was obtained for the determination of CRP and WBC

count and maternal temperature measured at the time of amnlocentusis AF

tests included culture and WBC count determinations. Receiver-operator

characteristic (ROC) curve, logistic regression, and survival techniques

(proportional hazards model) were used for statistical analysis. RESULTS: 1) Patients with acute hlstologic chorioamnionitis had significantly

higher median maternal blood CRP concentration, WBC count, body

temperature, and AF-WBC count than did patients without this lesion (p <

0.05), 2) ROC curve and survival analysis demonstrated that an elevated CRP, WBC count or AF-WBC count was strongly associated with the likelihood of

acute histoloDc chorioamnmnitis, shorter interval to delivery, clinical chono-

amnionitis, and neonatal morbidity (p < 0 05 for each); 3) Of all tests, AF-

WBC count (2 50 cells/mm3) was the most sigmficant independent predictor of

positive AF culture (odds ratio 16.8), interval to delwery (hazards ratio 5.7),

clinical chorioamnionitis (odds ratio 15.2), neonatal proven/probable sepsis

(odds ratio 16.8), and significant neonatal complications (odds ratio 7.4) after

other confounding variables were adjusted (p < 0 05 for each).

CONCLUSION: 1) An elevated CRP, WBC count, or AF-WBc count identified patients with intrauterine infection and adverse outcome; 2) However,

AF-WBC count was a better independent predictor of infection and adverse

perinatal outcome than CRP, WBC count, or temperature.

332 POST-TRANSLATIONAL EXPRESSION OF CYTOKINES IN A FIRST TRIMESTER TROPHOBLAST CELL LINE INDUCED BY LIPOPOLYSACCHARIDE. D,~, Svinanchx, O.M. B~tont~x, R. Romero and B. Gon=k. Dept. Ob/Gyn, Wayne State Un=v, Detro=t, MI., Pennatology Research Branch, NICHD, Bethesda, MD. OBJECTIVES: Response to =nfect=on by first tr=mester trophoblasts remains a poorly understood event. The mduct~on of certain cytok=nes by hpopolysaccharide (LPS) has previously been indicated through Northern analysis. Th~s study was undertaken to demonstrate the post-translattonal expresston of cytokmes and charactenze the=r in vitro regulation in a first tnmester trophoblast cell line when challenged w=th LPS. STUDY DESIGN: A first tnmester human trophobMst cell hne was exposed to LPS (1 ~g/ml) for 0, 2,’4, 6, 8 and 24h. Culture supernatants were collected and rendered free of cellular debns by centnfugat=on Supernatants from control cultures were examined for the presence of contaminating endotoxm. ELISA’s specific for e=ther IL-I~, IL-6, IL-8 or TGF-I~I, were conducted on the culture supernatants and the k=net=cs of cytokine express=on determ=ned. RESULTS; Induct=on of IL-I~=, IL-6 and IL-8 express~on occurred between 4 and 6h follow=ng LPS exposure and showed a progressive t~me dependant increase to 24h. Induct=on of TGFqll occurred 8h following LPS exposure and also showed a progressive t~me dependant =ncrease to 24h. A comparat=vely Ngh const=tut~ve level of IL-6 and TGF-I]I express~on was detected ~n non-LPS =nduced cultures. CONCLUSIONS: These data demonstrate the LPS ~nduct~on and t=me dependent post-translat=onal express=on of the pro- ~nflammatory cytak~nes IL-I=, IL-6 and IL-8 as well as TGF-Ill These results support the content=on that first tnrnester trophoblasts part~mpate ~n cytokine based ~mmune s=gnahng m response to =nfect~on.


333 CHEMOPROPHYLAXIS AGAINST ENDOMETRITIS AFTER POST-

PARTUM INTRAUTERINE MANIPULATION SE Esplln~, B Osh~ro, RM

Silver Dept Ob/Gyn, Umverslty of Utah, Salt Lake C~ty, UT

OBJECTIVE: The use ~f antibiotic chemoprophylaxls against pelvic

infections IS widely accepted for vaganal surgery and cesarean dehvery

However, there is little Information regarding chemoprophylaxls after

determine the frequency and efficacy of anUbtottc prophylaxis after

prophylaxis after postpartum D&C and/or manual placental extraction

RESULTS: Antibiotic prophylaxis was used in 66 of 150 cases (44%) of

extractions and 47 D&Cs The majority of women received a single dose of

cefazo/ln No antlb~oucs were used in 84 procedures (comrols), including 54

manual extractions and 30 D&Cs Three women recelwng prophylaxis

(4 5 %) developed endometrltlS after D&Cs compared to three controls (3 6 % )

resolved with therapy Women who did and did not receive prophylax~s were

duration of rupture of membranes, blood loss, method of dehvery, and

maximum temperature Of 149 responding physicians, anublot~c prophylaxis

is routinely used by 42 (28%) after manual placental extraction and 69 (46%)

after postpartum D&C

CONCLUSION: Antibiotics are routinely administered by many physicians

do not support the efficacy of chemoprophylaxls after these procedures

335 HUMAN PARVOVIRUS B19 RELATED HYDROPS Ah’D ELEVATED FETAL CREATINE KINASE. N Shillettox, JFR Barrett, L Allenx, G Ryan, RJ Morrow, D Farina. Perinatal Unit, Mt Sinai Hosp. University of Toronto, ON CANADA. OBJECTIVE: To measure the ereatine kinase myocardial-brain isoenzyme (CK-MB) level in human fetuses infected with human parvovirus B19 (HPV-Blg), causing non-immune hydrops fetalis (NIHF), in order to determine whether elevations in CK-MB could detect myocardial tissue injmy STUDY DESIGN: Fetal blood sampling (FBS) was performed on 9 fetuses with NIH]z. Investigations included, complete blood count, karyotype, polymerase chain reaction (PCR) for HPV-B19 genome & radioimmunoassay for CK-MB levels. As controls, CK-MB measuzements wer~ performed on senun obtained from 12 additional fetuses, 6 with anatomical defects who were undergoing FBS for karyotypi¢ analysis & 6 with Rhesus (Rh) disease who were not hydropi¢ and undergoing their f’~t FBS for investigation ofRh alloimmunizafion. RESULTS: 4 of the 9 hydropio fetuses were found to be HPV-BI9 PCR positive. The range of CK-MB in the fetuses with anatomical defects was 0 to 4.3 ng/1 (mean 1.3 ng/l). Of the other 15 fetuses sampled, 12 had CK-MB values within this range 3 fetuses demonstrated elevated values. An R/a-sensitized fetus who was aneraie (hemoglobin (Hb) of 6g/dl) but not hydropie had an isocnzyme level of 16.2 ug/l, while 2 fetuses with NIHF secondary to HPV-B 19 (Hb’s of 2g/dl and 4g/dl) had values of 11 ug/l and 22ug/l respectively The 2 fetuses with HPV-B19 infection and normal Hb’s of 10g/dl and llg/dl had CK-MB levels in the normal range. Fetal ultrasound showed poor myocardial contractility in fetuses with elevated isuenzymes, and normal contractility in those with CK-M~ values in the normal range. CONCLUSION: These preliminary results suggest that fetal myocardial damage may occur in severe anemia caused by HPV-B19 infection

334 ~Iq~C~RVICAL VS ~I~4~TRIAL CULTURE A~D TH~ DIA(~IOSIS OF POSTPArTuM ~%~)(I~I~ITIS. Edward R. Newton, M.D., Dept Ob/Gyn/Reprod Sclences UnlV Texas Med School at Houston, Texas. OBJECTIVE: To determine the correlatlon between simultaneous endocervical and endometrlal cultures at the dlagnosls of postpartum endometrltlS. METHODS: Palred endocervlcal (ENDOCX) and end~metrlal (ENDO) cultures were obtalned in 497 patlents between 1989 and 1994 at the diagnosis of endometritis. EndometrltlS was diagnosed with a postpartum ter~perature ~ 3SoC recorded twice, 6 hours apart and associated wxth locallzlng slgns and symptoms. Endometrial cultures were obtalned by a sheathed lnjection/aspxratlon technique. Aerobes and anaerobes were isolated zn standard microbiologlc technlq~es. RESULTS: Table 1 descrlbes the incidence of selected organlsms at each s~te and concordance between endocez-vlcal and endometrlal cultures.

Enterococcus

E. coli

K. pneumoniae

Prevotella bivia

*P < 0.01 vs Endo

11% 10% 94%

29% 25% 76%

20% 14%* 81%

15% 12% 88%

3% 23%* 77%

22% 38%* 73%

CONCLUSIONS: Aeroblc cultures of GBS, enterococcus and Gram negatlve rods show good concordance and endocervical cultures may reliably be interchanged with endometrial cultures.

336 AN EPIDEMIC OF PARVOVIRUS B19 IN A POPULATION OF 3600 PREGNANT WOMEN. A STUDY OF DEMOGRAPHIC RISK FACTORS. P Thorsen, I P Jansen, C F Jensen, 8 Jeune, 8 R Mzller, 13 F Vestergaard Dept of Ob/Gyn, Odense University Hospital, Denmark OBJECTIVE: Human Parvovlrus B19 (HPV B19) is an =ncreasingly recognized cause of reproductive wastage and feta[/permatal morbidity ~e fetal hydrops and aplastic anemia We fortuitously prospectively cherac- tenzed HPV B19 morbidity among pregnant women and their progeny dunng a large epidemic ~n Denmark, November 1992 - December 1994 STUDY DESIGN: A major HPV 1319 epidemic occurred during the conduct of a large study prospectively to evaluate r~sks of preterm birth in Denmark Three thousand and s~x hundred pregnant women were asked to par~clpate m the study before the 24th week of gesta~on and 3155 had prospective serology study for HPV B19 at enrolment using a standard ELISA Thlrt~our percent of these were leG-B19 negative, suggesting endemic HPV B19 infection ~n Denmark Serial bloodsamples were obtained from 2866 women, of these 101 (3,5%) pregnant women suffered HPV B19 ~nfectlon documented by IgG seroconversion and/or IgM seropos~tiwty RESULTS: Charactenstics of women w~th HPV B 19 ~nfectlon vs women w~thout ~ncluded. Children at home [47,5% vs 32 6%; OR 1 9, 95% CI 1 3-2,8], h~gher educational achievement (> 3 years) [85.2% vs 75 6%. OR 1 9, 95% CI 1 1-:3 3], higher enwronmental stress [18 1% vs 9 6%, OR 2 1, 95% CI 1 1-3 8], nonspec~c ~llness in pregnancy (unhl 34th week of gestation) [48 5% vs 38 3%, OR 1 5, 95% CI 1 0-2 3] and prewously del~ves, [66 3% vs 51 9%, OR 1.8, 95% CI 1.2-2 8] There were no d~fferences between the women with and w~thout HPV 819 infection dunng pregnancy concerning mean age (28 8 years vs 28 0 years), work w~th children, mantal status and social status CONCLUSIONS: We fortur~ously observed a major epidemic of HPV 819 among pregnant Danish women, women with children at home, higher educational achievement, h~gher environmental stress and who suffered nonspec~c ~llness in pregnancy were more I~kely infected w~th HPV 819.

404 SPO Abstracts January 1996 Atn.] Obstet Gynecol

DETERMINING MATERNAL VS FETAL IMMUNOLOGIC CONTRIBUTION TO CHORIOAMNIONITIS (CHORIO). M. McNamarax, T. Walhsx, F. Qureshff, S. Jacques~, B. Gonik, Depts. Ob/Gyn and Path, Wayne State Udiv Shc of Med, Detroit, MI, OBJECTWE: CHORIO describes an inflammatory response to infection within the placenta, the origin of which is poorly defined. This study examines maternal and fetal acute immunologic contributions to overtly presenting (CHORIOcl) and subchnical (CHORIOH~t) chorioamnionitis in the preterm gestation. STUDY DESIGN: Paraffin embedded tissue blocks were ,dent~fied from histologically severe CHORIO in preterm (31-33 wks) pregnancies. Chart review was used to determ,ne CHORIOcl (n = 4) vs CHORIO"~ (n=4[ status. Only placentas from male fetuses were studied. Thick paraffin sections were digested in a protease solution, cell suspensions were prepared, and cytospin slides were generated. Fluorescence in situ hybridization (FISH) of sex chromosome probes were used to quantify the number of maternal (XX) and fetal (XY) polymorphonuclear cells (PMNs) extracted from placental tissues. RESULTS: Bhnded examination of cytospin FISH preps demonstrated no significant difference in total mean (± SD) number of PMNs/shde between CHORIOcl (45.0 ± 16.7) and CHORIOH~t (27.7 ± 23) Cases (p=0.14). Maternal denved PMNs accounted for 92% and 95% of the acute inflammatory response, respectively, in the two study groups. CONCLUSION: No s~gnificant difference in acute inflammatory response (PMNs) can be appreciated based on cl{mcal presentation of CHORIO. Under both circumstances, the maternal (vs fetal) acute inflammatory response predominates within the placenta.

339 FETAL BREATHING MOVEMENTS WITHIN 24 HOURS OF DELIVERY IN PREMATURITY ARE RELATED TO HISTOLOGIC AND CLINICAL EVIDENCE OF AMNIONITIS. CY

~p_ o_~g~, DM Sherer, CM Salafia*, A Ghidini, VK Mmlor*. Perinatal esearch Facdlty, Depts. of OB/GYN and Pathology, Georgetown

University Medical Center, Washington DC and UCONNMedical Center, Farmington, CT. OBJECTIVE: To detemune the association of fetal breathing movements (FBM) within 24 hours of delivery, with histologic evidence of acute ammonltlS and clinical chonoammoniUs (CA), in patients delivenng < 32 weeks gestation. STUDY DESIGN: From an established database of 465 consecutive nonhypertensive, nondiabetic patients delivering < 32 weeks gestation, 257 patients with non-anomalous singleton hve births and sonographmally assessed FBM within 24 hours of delivery were identified. Patients with suspicion but not diagnosis of CA were excluded. Two groups were defined: FBM present and FBM absent. CA was diagnosed by any 2 of the following criteria: maternal temperature~ > 37 8°C, matemal tachycardla >120 bpm, leukocytosis > 20,O00/mm~ (without prior steroids), fetal tachycardla > 160 bpm, uterine tenderness or foul smelling amniotic fluid. Histolog~c presence and extent of acute ascending intrauterine Infection was assessed In amnton, choriodenidua, chorlomc plate, umbilical, and chorlonlc vessels and scored by a single pathologist bhnded to chnlcal information. Statistical analyses consisted of contingency tables, uni- and multivariate analyses with p < 0.05 considered significant. RESULTS: FBM were absent in 138 cases (54%) andApresent in 119 cases (46%). The rate of histolog~c ~nflammation was slgmtlcantly higher in the absence vs. presence of FBM [51% (711138) vs. 37% (44/119), RR=I.8, 95% CI 1.1-3.1)]. Similarly, the prevalence of clinical lnfectmn was significantly higher in the absence vs. presence of FBM [49% (67/138) vs. (2-8/119), RR=3.1 95% CI 1.7 - 5.5)]. When cases of CA were excluded FBM were no longer related to measures of hlstolglC acute inflammataon (each p>0.05). Chmcal diagnosis of chonoammomtls was apoorpredlctor of histologic evidence of acute amnionitis [PPV=65% (61/94), NPV=33% (541162), p<0.01]. The above findings were independent of gestatlortai age at delivery, status of membranes (ruptured vs intact), presence of labor, mode of delivery and fetal presentation. CONCLUSIO~N: Removing the potential bins and imprec~sion of clinically diagnosed CA, the apphcatmn of a definitive method (histologlc assessment), demonstrates a significant association between the presence of lntraammotic infection and decreased FBM in preterm fetuses. However, once CA is diagnosed, presence or absence of FBM is not a predictor of histologic mflammatmn.

338 THE HISTOLOGIC DIAGNOSES OF MATERNAL AND FETAL MARKERS OF ACUTE ASCENDING INTRAUTERINE INFECTION ARE HIGHLY REPRODUCIBLE. J.M. Lage* C.M ~*, E. Marle~d ~ mP~kash*, ,.C. Pezzul,o* A. Ghiaini. Pdr~ Research Facdlty, ep t eats of Pathology and ~)B/GYN Georgetown Umverslty Medical Center, Washington DC, Yale Univers’lty School of Medicine, New Haven CT

OBJECTIVE: To determine the reproduclNhty of histologlc dlagnos~s and grading of maternal and fetal inflammatory lesions indicating acute

ascending infection. STUDY DESIGN: Histology shdes from 26 cases of preterm birth retrieved from pathology arc~iives included 15 sections of amnion 15 sections of umbilical cord, 26 sections of chorlodecldua, and 21 sections of chorlonlc plate. Guidelines for diagnosing and scoring acute ammomtls, chorlodecldums, chorlomtls and umbilical and chorionic vascuhtis (Obstet Gynecol, 1989 73:383-9) were provided without additional instructions to 2 separate groups of investigators who scored the histologlc slides. Since tissue necrosis may obscure tissue detail and comphcatediagnosls, the presence and extent of tissue necrosis was recorded. The scores were returned to the gmdeline author and entered into a data base. A weighted Kappa statistic was performed to consider not only exact concordance of scores, but the extent of d~scordance between scores (e.g., a discordance between grade l and grade 4 counted more than a discordance between grade 1 and grade 2). RESULTS: Kappa values, reflecting extent of ~nterobserver agreement, were as follows: amnion 0 84; chonodecidua 0.84; umbdmal cord 0.85, chonomc plate 0 73 Seventy-five of 77 diagnoses (97 4%) were concordant for the diagnosis of presence of acute inflammation. Twenty- six % (20/77) of the total scores were d~scordant for inflammation grade. Eleven (50%) of the dxscordant scores involved the severe end of the lnflammatmn scale (Grades 3 and 4) and would not have modified the diagnosis of high grade inflammation. In all of the 1I cases, multifocaI or diffuse tissue necrosis was present.

CONCLUSIONS: A scoring s/stem of hlstologlc acute inflammation can be learned--from a printed set of guidelines alone--with a high level of reproducibility. The grading system we used was devised for study_ of uncomplicated_ term deliveries. In the presence of severe and necrotlzlng inflammatory lesions, such as those seen in some cases of preterm birth, a modlflcatmn of this hlstolog~c grading to ~nclude the extent of tissue necrosis (which hampers the assessment of neutrophll pre~euce and number) will further increase the reproducibility of mstolog~c grading.

340 Labor-related intm-ammotic infection - Is Postpartum Treatment Necessary When Delivery Occu~ Vagmally9 C. LaBellax, IL ~andber~’.x D. Edelstone. Dept of Ob/Gyn, Umv of Pittsburgh, Magee-Womens Hospital, Pittsburgh, PA OBJECTIVE: We tested the hypothes~s that antibiotic treatment postpartum of an intm-ammotic infection that develops during labor is unnecessary, because materlm/mogo~&ty ~s low, and the source of mfectxon ~s removed at the time of delive~�. Patients not treated postpartum were predicted to have no difference ~n the rate ff maternal morbidity from infectxon compared with those palaents treated with antibiotics followmg dehveD, STUDY DESIGN: We performed a prospective, randonuzed, placebo-controlled, double-blinded trial of antibrotic treatment postpartum versus no antibiolac treatment postpartum in women whose labor was coraplicated by mtm-amniolic infection and who delivered vag~nally. Standard entry criteria for intra-amniotic uffedimn were used. All pataents received ampicillin, gentamicin, and clindamycm during labor. Following delwe~y, palaents were then randomized to continue anubiotic therapy or recewe placebo. RESULTS: To date we have enrolled 36 patients ~n the study. Between the treatment and placebo groups, there were no differences m postpartum endometritis, standard puerperal morbxdfly, bacteremia or other major infeclron-related complicaUous, or length of hospital stay The overall incidence of postpamma infection and associated coraplications was less than 5% m each group. CONCLUSIONS: Our data show that women with clinically- apparent mtra-arnniotic refection who deliver vaginally do not reqmre ant~brotic treatment postpartum


341 PyELONEPHRITIS IN PReGNANCy, ORAL VERSUS IV ANTIBIOTICS. V. Cook, M. Herzog,X S. Hughes,~ D. Voss, J. Spinnato, $. Gall. University of Louisville, Louisville, KY OBJECTI%~, This is an inpatient prospective randomized trial of oral versus IV antibiotics for the treatment of acute pyelonephritis (pyelo) during pregnancy. STUDY DESIGN, On the basis of history, physical, and laboratory findings, pregnant patients hospitalized for acute pyelo were assigned by last digit of their record number to treatment with oral (trimethoprim/sulfamethoxazole DS (TMS) every 12 hours or if <14 weeks, oral cephalexin 500mg every 6 hours) or IV antibiotics (aztreonam 2g every 8 hours). Bacteremic patients on oral antibiotics were changed to IV antibiotic. RESOLTS* From June 92 to August, 93 admissions were randomized into the study. Twelve were excluded for incorrect diagnosis (8) and recurrent pyelo (4) leaking 81 charts for analysis, 43 oral and 38 IV. No significant differences between the oral and IV groups were observed regarding demographic characteristics, admission to discharge interva! (3.6±1.7 vs. 3.4±1.1 days), ~aximum WBC count (14.0±4.4 vs. 15.8±7.6), maximum fever (I01.6±1.6 vs. 101.3±1.5), fever duratlon (45.8±25.4 vs. 43.1±26.7, p=.20), Ledger Degree Hours of Fever (LDF) (46±38 vs. 35±26, p=.14), or doses of antlbiotic (9.2±5.2 vs 9.7±3.6). Of the 43 patients in the oral arm, bacteremia (6), persistent fever (I), tachypnea (2), and persistent emeslS (2) prompted completion of treatment with IV antibiotic. Cost of antibiotic to the pharmacy was significantly less for oral than IV antibiotic whether calculated for aztreonam ($112±164 vs. $265±96, p<.0801) or for use of cefazolin 2g every 8 hours, ($48.8±66 vs. $101±36, p<.0001). CONCLUSIONSt Recovery from pyelonephritis occurred with equal rapldity in the two groups though treatment was completed with IV antibiotics in 26% of the oral group. Oral therapy alone was efficacious and offers financial savings. Further study of oral therapy is necessary prior to initiating outpatient treatment, particularly for those with bacteremla or fever > 102F.

343 THE BABOON AS A PRIMATE MODEL TO EVALUATE THE EFFECTS OF VACCINATION DURING PREGNANCY. MH Shearerx, JR Stanleyx, TC Chanhx, RC Kennedyx Dept

M~crob~o and Immun. & OB/GYN, Umv of Okla Health Sc= Ctr., Okla. City, OK and Dept. V~rol & Immun, Southwest Found. for Brained Res, San Antomo, TX OBJECTIVE: Maternal vaccmatton may offer a means of

prevent=ng the pennatal acqu=s~t~on of =nfections such as group B streptococcus and HIV The baboon ~s s~milar to man ~n placentation and immunology (4 IgG subclasses) The purpose of th~s study is to evaluate the baboon as a roodel for vaccination dunng pregnancy STUDY DESIGN: Three baboons were vaccinated =ntramuscularly

w~th 10 t~g of Haemoph=lus influenza type b (H~b) polynbosylrlb~tol phosphate (PRP) conjugate vaccine approximately 40 days prior to dehvery Serum was obtained from the mothers prior to vaccination and at delivery, and from the ~nfant within 24 hours of b#rth. Baboon anUbodles to Hmb PRP were detected using a radioantmgen binding

assay and an ELISA IgG and IgM anti-Hmb PRP reactiwty were also determined RESULTS: All mothers had greater than a three-fold mncrease in the ant~-Htb PRP antmbody levels after vaccination increasing 3 2, 54 3, and 46.3 t#mes baseline. Antm-H=b PRP IgG, but not IgM anUbodmes, were detected m the newborn ammals The amount of maternal antibody to H#b PRP transferred to the infant was not proportm~al to the maternal concentratmns CONCLUSIONS: The response of the pregnant baboon to Hmb PRP vaccmatmon ~s smmilar to that expected in humans While IgG anti-

Hib PRP antmbody crossed the placenta, IgM anti-Phb PRP did not TNs study lends support for the use of the baboon as a model for the study of vaccmatmon during pregnancy. Supported by NIH contract AI-35156

342 A RANDOMIZED CLINICAL TRIAL OF ANTIBIOTIC THERAPY FOR ANTENATAL PYELONEPHRITIS B. B~, B. Campbell, S. Stratumx, D. Eller, R. Newman. Dept. of Ob/Gyn, Medical University of South Carolina, Charleston, SC. OBJECTIVE: To prospectively evaluate the efficacy of a therapeutic course of intravenous antibiotics versus intravenous therapy followed by oral antibiotics for the treatment of antenatal pyelonephrius. METHODS: Pyelonephritis was documented by strict criteria m 67 pregnant women and treated with appropriate intravenous antibiohcs until afebrile and asymptomatic for 48 hours. Patients were than randomized to completion of a 10-day course of oral Macrodantin 100 mg QID or no further therapy. Antibiotic suppression was not used. Urine cultures (UC) were obtained 2 and 6 weeks after discharge, and at delivery. A posit*ve UC or readmission for pyelonephntls were endpoints for participation in the study. RESULTS: Readmissmn for pyelonephritis prior to the 2 week follow-up visit occurred in 2/36 (5 6%) women randomized to oral therapy compared to 4/31 (12.9%) in the no treatment group (p=0.03). At the 2 week follow-up visit, the number of pahents w~th a positive UC in the oral therapy (6/31) and the no treatment group (8126) were not significantly different (p=0.11). During the remainder of pregnancy, posinve UC developed in 6 and 2 women and pyelonephtitis in 2 and 1 women in the oral therapy and no treatment groups respectively. CONCLUSIONS: Ten days of oral antibiotics after appropriate antibiotic therapy reduces the risk of recurrent pyelonephritis immediately post treatment. Patients in the no treatment group also had a non-sigu~ficant increase in the frequency of positive UC at the 2-week follow-up visit, The h~gh rates of recurrent urinary tract refection by the end of pregnancy in both groups underscores the need for regular UC and possible need for oral antibmtic suppression even after a single episode of antenatal pyelonephntls

344 Effectiveness of the ACOG Guidelines in Reducing the Incidence of Early-

Onset Neonatal Group B Streptococcal Sepsis. Penny Clark=, Patrick Duff,

Carolann Ristey!, University of Florida, Gainesville, FL.

Obiective: To determine the effectiveness of the ACOG guidelines in

reducing the incidence of early-onset neonatal group B streptococcal (GBS)

sepsis.

Study Design: During the period 8/91-8/93, we screened for maternal GBS

colonization only in women who had the following risk factors: preterm

labor, preterm PROM, cborioamnionitis, and history of a prior infant with

GBS infection. Colonized women were treated intrapartum with

intravenous ampicillin, 2 g Q 6h. Using this selective screening protocol,

<3% of women received prophylactic antibiotics. In 9/93 our department

implemented the ACOG guidelines for prevention of neonatal GBS

infection. Specifically, all women with unknown colonization status and a

recognized risk factor for GBS infection received intrapartum antibiotics.

With this prot~o|, al~pruximately 20% of patients, most of whom were at

term, received intrapurtum antibiotics. In this retrospective survey, we

compared the incidence of invasive early-onset neonatal GBS infection

(bacteremia and meningitis) during the time periods 8/91-8/93 and 9/93-

7/95.

Results: In the initial 24 month period, 23 of 7942 liveborn infants

developed early-onset GBS sepsis (2.9/1000). After implementation of the

ACOG guidelines, 13 of 5831 livebirths developed early-onset sepsis; one

infant also had meningitis (2.2/1000, NS). The case fatality rate was 1/23

vs. 0/13 (NS). In the initial study period, 16/23 (70%) mothers had

recognized risk factors for neonatal GBS infection, while in the second

period only 5/13 (38%) had risk factors (P < .05). In infected neonates, the

mean durations of antibiotic administration were ILl and 12.0 d,

respectively (INS). The mean durations of hospitalization were 10.6 and

11.1 d, respectively (NS).

Conclusion: Implementation of the ACOG guidelines for prevention of

GBS infection did not decrease the incidence, nor lessen the morbidity, of

neonatal infection. It did result in a significant increase in the frequency

of maternal antibiotic administration.

406 SPO Abstracts .JanuaD’ 1996 Am.l Obstet Gynecol

345 ~ THE ACOG RISK FACTORS FOR GBS PROPHYLAXIS PRACTICAL? C Towers, P Rumneyx, S Posnerx, T Asra~ M Nageotte Long Beach Memorial Methcal Center, Long Beach, CA, Hoag Memorial Hospital, Newport Beach. CA, Little Company of Mary Hospital, Torrance, CA OBJEL-q’iVE: Presently, ACOG d~es not recommend roufine antenatal cultunng for GBS colonization lnstead, they recommend intrapartum antibiotic treatment for all pregnant women who have risk factors (RF) which are delivery < 37 weeks, membrane rupture for >lg hours, intrapartum fever >37.5 C, or a prior child affected by GBS. To date, no data are available which identify the number of pregnancies that present in labor wth GBS risk factors Our objective was to identify the number patients with risk factors and determine how many received appropriate antibiotic treatment. STUDY DESIGN: Over a 5 month period, every delivery over 24 weeks gestation at 3 hospitals was prospectively evaluated for gestattonal age at delivery, duration of membrane rupture prior to delivery, highest temperature prior to delivery, and the use ofmtrapartum antibiotics. RESULTS: 3958 patients dehvered at these 3 hospitals 343 (9%) pafients were < 37 weeks, 317 (8%) patients had ROM > lg hours, and 666 (17%) patients had temp > 37.5 C. Several patients had more than 1 RF, therefore, 1107 (28%) total patients had at least 1 risk factor. 349 (only 32%) of the 1107 risk factor paticots actually received lntrapartum antibiotic treatment During the course of this study, 3 cases of neonatal GBS sepsis occurred 2 had risk factors, 1 did not, and none of the 3 received antibiotic prophylaxis HOSP. # OF PTS # PTS WITH RF _#pTS W/RF TREATED A 1563 496 (31%) 194(40%) B 1589 415 (26%) 118 (28%) C 806 206 (26%) 37(18%) TOTAL 39~ 1107 (28%) 349 (32%) CONCLUSIONS: 28% of delivering women had a risk factor in labor that would reqmre antibiotic treatment if ACOG recommendations were adhered to smct/y Of conceru, many obstotncians are either unaware of or inattentive to these gmdelinns since 2 out of 3 were not treated. Therefore, in a practical chnical semng, the implementation of the ACOG guidehnes fell far short of the mark

347 PERFORMANCE OF A GROUP-B STREP (GBS) SCREENING PROTOCOL IN A LARGE COMMUNITY HOSPITAL. D_~P Reisner, MJ Haasx, RW Zingheimx, MA Williams×, DA Luthy. Swedish Medical Center, Seattle, WA. OBJECTIVE: To evaluate the effectiveness of an intrapartum GBS screening protocol. S~JDY DESIGN: Preterm and laboring patients were screened for GBS by colorimetric enzyme immunoassay test in 1994. Selective media cultures were done for screen-negatives. Screen-positive patients were treated with antibiotics. Neonatal morbidity and mortality were evaluated. Immediate costs were compared. RESULTS: 86% of 3469 deliveries were screened; 17.7% were positive by rapid screen or culture. Rapid screen sensitivity was 77% for any GBS. There were NO cases of neonatal GBS sepsis or maternal anaphylaxis.

Year I Detiveri~ Preterm ~ ~ [

Costs

~-~ I ~ I ~ I 0 I 3 I 2!,955 1~3 I 4070 I 8.8% I 6 I 4 I 141,341

A significant decrease was seen in case of sepsis in 1994 vs. 1993 (p=0.006) and 1994 vs. 1992 (p=0.007). 1994 costs for screening and treatment were $127,178. Intrapartum rapid screening would cost $76,583 vs. antepartum cultures at $50,300. CONCLUSIONSt Treatment of high-risk patients, plus rapid screening of low-risk laboring patients is a cost-effective method of minimizing neonatal GBS.

346 NEITHER UNIVERSAL SCREENING FOR HEPATITIS B OR REPEAT SCREENING OF H!GH-RISK PATIENTS IS COST EFFECTIVE: G Webb~, J Huddleston, D. Vroonx, A Gohrx Depts of Ob/Gyn end Pathology, Emory Onivers,ty School of Med,cine, Atlenta,

GA OBJECTIVE: To evaluate the efficacy and cost effectiveness of the CDC and ACOG recommendattuns for universal screening for HBsAg in pregnency and for repeat screening m the third trimester for patients at particularly lugh risk for contracting flus infection STUDY DESIGN: This retrospective study reviewed our experience at Grady Memorial Hospital, an indigent care facility in Atlanta, Georgia, from Jan l, 1994 through Dec 31, 1994

RESULTS: During the one year study period, 9679 Hl3sAg screening tests were performed on 7790 pregnant patients. S~xty-five patients were found to be H!3sAg positive, for a prevalence of 0 85%. Of those 4418 patients who completed therr entire obstetrical course during the study year, 1679 (38%) had at least one repeat test. Only one of these 1679 converted from negative to positive. That patient experienced en acute dlness consistent with hepatms All but two of the 66 total HBsAg- posture pahents had standard risk factors for hepatitis refection explicitly recorded m their prenatal charts There were red,cations of tlhctt drug use m the prevmus medical records of beth these patients CONCLUSIONS: For our at-risk population, the HBsAg conversion rote during pregnency ts extremely low (1/1679), end the rate of unrecograzed conversion is even lower W~th appropriate interviewing end me&cal record rewew, all HBsAg positive tests among the 7790 pregnant patients encountered would have been detected if screening had been based on risk

factors alone. Our data do not support e~ther universal screening or repeat screening of those patients initially tested negative, but rather suggest that screemng based on risk factors alone would be efficacious and more cost effective

348 THE SEROPREVALENCE OF H1V, HEPATITIS B, HEPATITIS C, AND SYPHILIS IN A LARGE URBAN PREGNANT POPULATION -- A FIVE YEAR ANALYSIS. D RINDFUSZx, F SEYDEL, J PEZZULLO, S PETERS, T PINCKERT Dept ofMatermd/Fetal Medm~ne, Georgetown Umvers~ty Medical Center, Washington DC OBJECTIVE: To determine the true seroprevalence for HIV, Syphd~s, Hepaans B and C within a large high-risk pregnant population and delineate any assoctaoon or interaction amongst these disease states STUDY DESIGN: 800 specimens were randondy selected from second mmester sera collected from prenatal chnlcs within the Washington DC area over a five year period from 1991 to 1995 and were assayed to determine the changing seroprevalence of HIV, Syph~hs, and Hepant~s B & C Multivariate logistic regression, analysis of variance, and contingency table analyses were used to evaluate the changes of seroprevalence rates over t~me and to assess the strength of associations among the different disease states Odds ratios and their confidence mterva/s were calculated from each year for H1V g~ven each of these STDs. RESULTS:

CONCLUSIONS: The seroprevalence ~s shown in HIV, Syphilis, and Hepatu~s B

syphlhs was found However, no significant assocmtlons between ttlV and either HepatulS B & C were identified These unanticipated findings could be reflective

pregnancy outcome and neonatal morbidity is yet to be determined Aggressive

population which would most benefit from treatment could then be better defined

Volume 174, Number 1, Pa~t 2 SPO Abstracts 407 Am J Obstet Gynecol

349 ANALYSIS OF POSITIVE BLOOD CULTURES AND ANTIBIOTIC SUSCEPTIBILITY ON AN OBSTETRIC SERVICE. M. Martens MD_, and B. Raybon MD, Hennepln County Medical Center, Minneapolls, Mlnnesota Bacteremla in pregnant and postpartum patients is a serious condltlon with potent~al±y l~fe-threatenlng consequences for both mother and infant. Much of the concern has recently centered around the Group B Beta- hemolytic streptococcus, however, there may be several organls~S which play a role in perlpartum bacteremla. Also, most antlblOt~c therapy is started emplrlcally, therefore ~t would be helpful if the most common organisms and their antlblotic susceptlbllltles are known prior to culture and sensltzv~ty results. Thls study revlewed all bacteremlc patlents and the organlsm’s susceptlblllty from the OB-OYN servlce at a slngle indlgent caze hospltal. From a two-year period there were 30,154 OB-GYN admlsslons with 387 posxtlve blood cultures reported, for an incidence of 7/1000 OB-GYN patlents. The most common obstetrlcal organisms recovered were separated by the cl~nlc slte where the cultures were taken. intepartum Unlt Labor & Delivery Postpartum

n = 14 n = 88 n = 35 E. coli (57%) Grp.B strep (27%) Grp.B strep (26%) S. aureus (21%) E. COll (25%) E. co11 (11%) K. pneumonla (7%) Alpha Strep (ii%] S. aureus (9%) S. pyogenes (7%) S. aureus (iO%) gnterococcus (9%)

The sensitivities for the most common isolates were: E. coll 100% sensltive to ceftlzoxlme, cefotaxime, and imlpene~, 98% sensltlve to gentamlcln, cefazolin, and cefuroxlme. Staph. aureus: 100% to Vancomycln, 86% to Ciindamycln, 82% to cephalexln, 16% to penicillin. Therefore, while it appears that Group B strep does play a major role in perlpartum bacteremla, E. COll is commonly isolated in patients on the antepartum unit and L & D along wlth S. aureus. Therefore, these organlsms should be consldered and covered in patients at high rlsk for the serlous consequences of bacteremla.

351 TINING OF l NTRAPARTUM N~ICILLIN INFUSlO~ FO~ ~(1~ B

STREPT~ (~S) ~HYL~IS. S.L. B[o~, K.J. Leveno, L.C.

Gitstrap, S.M. Cox, Dept. Ob/Gyn, Univ. TX Southwestern Medical Center, Dallas, TX ~JECTI~: To determine the mini~ ti~ r~uired to achieve MICsQ a~ m n ~ ~cteric da concentrations for GBS in a~on c flu a; cord b[o~, a~ ~ternat b[o~ following an infusion of 2 g of a~c~ll~n. STaY DESI~: 30 w~n schedut~ to u~ergo elective re.at

cesareans consent~ to patriciate. An ~ntravenous infusion of a~icitLJn (2 g over 1-3 min) was given at varying tJ~ intervals ~tween 3 a~ 70 min prior to delivery. Sautes of a~fonic fluid were obtained ~ a~iocentesis just prior to the hysterot~

incision. U~iticat artery a~ ~ternat bto~ were obtained at the ti~ of delivery. A~icitlin concentrations were then ~asur~ by HPLC. RESETS: A~ionic fluid levels ranged fr~ 0.10 ~g/ml to 11.62 ~g/mL; at[ revers were greater than the MICro for GBS (0.03 ~g/mL) a~ were achieved as soon as I rain 42 se6after a~Jcitlin infusion. MfnJ~ ~cteric~dat concentrations (range 0,2-2.5 ~g/mL) were first achieved 5 min 8 sec after infusion a~ were first exceeded 9 man 45 sec after a~iciltin a~inistrat]on. At[ ~terna[ bto~ a~ ~i[icat artery bto~ levels exceeded 2.5 ~Q/mL re~argtess of the ti~ interval ~tween infusion a~ col tection.

[

~O#CLUSIO#: Bactericidal concentrations of ampiciltin are achieved ~n a~qionlc fluid and cord blood within 5 min of administration.

350 CLINICAL EFFICACY AND COST EFFECTIVENESS OF ROUTINE GROUP B BETA- STREPTOCOCCUS GENITAL CULTURES AT 36 WEEK GESTATION. K. Demaslo, C Antoine Dept Ob/Gyn, NYUMed Ctr, NY, NY OBJECTIVE To determine the efl~cacy and cost effectiveness of routine Group B Streptococcus cervical cultures (GBSC) at 36 wk gestattonal age (GA) STUDY DESIGN Grawdas rn apnvate pracfice were followed prospecfively from 1993-1994 to evaluate GBS colomzatmn at 34-38 wk gestatmn, mean = 36 wks 308 consecufive GBSC were obtmned GA was confirmed by first and second trimester sonography, Group A comststed of 40 patients (13%)with antepartum (36wks) positive GBSC who received mtrapartum anfibmfic therapy ( Amp~cdhn or Vancomyc*n) on adm{sslon ~n labor Group B-c-ons*ste-d of 268 pafients who had negahve GBSC at 36wks Group C consmted of 52 pahents w~th posmve GBSC obtmned *ntrapartum or postpartum subsequent to maternal or neonate *n&e~-ons There were no differences *n maternal age, panty, ethmmty, soc*oeconom*c status, or marital status between the three groups Cost analys*s ~ncludes NICU charge per day = $2140, mulhphed by average length of stay, cost of antenatal GBSC, and anhb*ofic therapy costs (Group A only) Physician cost and procedure cost not ~nc~luded RESULTS: GROUP A GROUP B GROUP C

N 40 268 52 Avg GA Dehvery (wks) 39 39 39 Neonatal GBS Seps~s 1 (2 5%) 0 16(30.7%) * Cost of Care per group ($) 26,980 26,800 719,040 Cost per pahent ($) 15,080 100 45,040 + Rx

¯ P=< 001 forAvs C CONCLUSION: Screemng all grawdas at 36 wksprowdes, 1 Accurate mformahon for ~ntrapartum treatment of GBS colomzed women 2 Sxgmficantly reduced the number of neonates admitted to N I C U for seps~s 3. Further reduced the cost of treatment winch ~s s~gmficantly more than the cost of screemng antenatally

352 AUTO CRASH SIMULATION USING THE FIRST PREGNANT CRASH TESTDUMMY M Pearlma~D V~ano~Dept Ob/Gyn, Unlv of MicA, Ann

Arbor, MI and General Motors CorporaUon, Warren, MI

OBJECTIVE To develop a pregnancy insert for the Hybrid III crash dummy

allowing evaluation of the effect of various restraint condltmns on energy

traasm~ssmn to both the fetal and maternal compartments

STUDY DESIGN A pregnancy insert w~th a rubberized uterine shell, simulated

sthcon ammohc fired and 28 week s~mulated fetus was fitted to a 5th percentlle

Hybrid III female crash dummy The fetus was Instrumented w~th accelerometers

in the head and thorax and a transducer to measure force transm~tUon through the

uterus Thffty-nine crash tests were run in the driver and front passenger positron

on a Hyge sled under s~x different restrmnt conthtmns at three different speeds

Fetal responses were compared for thfferent restraint con&Uons

RESULTS. Table 1 demonstrates the fetal response to a 20 MPH crash under

various restraint conditions

RestraintoC~°YlOfi°~n"o:’: ,’i, HIC H-*ifL6~g~: AbdForce ~(~a: i,,~,,, ~’~

tJNtLB,~ndSB,’ ."; 0:;~ .~,2~ 45 ~6:2~7,,’~,/~1 0.79

Restraint conthtlon. Nl=Normal; LB=lap belt; SB=shoulder belt

HIC=Head Injury Criteria. H-T Long =Differential acceleratmn of head and

thorax m the fetal AP ax~s (max).

CONCLUSIONS Energy transmission can be measured both m terms of

abdominal force (kN) as well as rotatmnal or shearing effect (H - T long ares) on

the fetus (g) This new dummy demonstrates substantml transmlss~on of energy

which are accentuated in some restraint conthtlons commonly used by pregnant

women The use of the 3-point restraint system appears to reduce the hkehhood

of injury m tats model

408 SPO Abstracts Jamtal7 1996 Am ,] Obstet Gynecol

353 EVALUATION OF A RAPI~ OPTICAL/Z~IMUNOASSAY BASED TEST FOR GROUP B STREPTOCOCCUS COLONqT~.T1ON IN INTRAPARTUM PATIENTS

Ramm Samadl MD,x Ahce Stek MD, Jeffrey S Greenspoon MD

University of Southern Cahforma School of Medicine, Los Angeles, CA

OIKIECTIVE: To compare an opucal immunoassay (OIA) rapid k~t to standard

culture for diagnosis of vagmal colomzatlon with Group B Streptococcus (GBS)

STUDY DESIGN: 301 pauents ~n labor from 12/94 to 3/95 were selected at

random and provided with Informed consent for participation Exclusmn criteria

were any antibiotic treatmeut dunng the previous 7 days, vaginal bleeding, or

placenta prevla Vaginal cultures were obtained slmultsneously using a rouUne

Dacron tip culture swab (Starplex Scientific, Etoblcoke, ONT) and a rayon tip

swab for OIA Rectal cu}tures were oblamed by routine culture s~vab The

swabs were separately plated directly on sheep blood agar containing cohstln

and nahdlxac acid S agalachae strmns were confirmed using the CAMP test

The FDA approved STREP B OIA k~t (Bmstar, Boulder, CO) was used for rap~d

determination of the vaginal GBS colomzatton according to manufacturer’s

mstrnctlons OIA results were compared to results of routme culture

RESULTS: By standard culture 33 (11 0%) patients were vaginal GBS(+) and

42 (13 9%) were rectal GBS(+) Fifty tour (17 9 %) oflhe 301 patients had GBS

colomzatton at either site By OIA rap~d kit, 56 (18 6%) patients had vaginal

colomzatton with GBS Thirteen (4 3%) had umnterpretable results and were

excluded l"rom analys~s The OIA rap~d ka had a senslUv~ty and specificity of

63 6 % and 86 3%, respeclwdy, for the determination of vaginal GBS

colonization The positive predictive value and the negatwe predictive value for

the OIA were 37 5% and 94 8%, respecUvely

CONCLUSION: The sensitivity of the OIA rapid kit for detectmn of vaginal G]3S exceeded thai of other commercially available rapid kits These findings

indicate that this kit may hold promise as a rapid method to detect vaginal GBS

colomzatlon and warrants further clinical evaluation

355 IS RISK FACTOR SCREENING FOR HEPATITIS C SUFFICIENT? U.

Magriple#, P. Bernsteinx, B. Shneider~, E. Snyder~, J.A. Copol. Depts of

OB/GYN, Pediatrics and Blood Bank, Yale Univ., New Haven, CT.

OBJECTIVE: Prevalence studies of Hepatitis C virus (HCV) in obstetrical

populations have been limited by the inaccuracy of first generation testing and

the inclusion of patients at high risk for medical complkr, ations. We studied the

prevalence of HCV seropositivity in a low income urban ob~etrical population

using a highly accurate second generation assay and the ©ffeetiveness of risk

factor-guided screening.

STUDY DESIGN: We performed a prospective study of women attending

prenatal clinic at Yale-New Haven Hospital. Patients with HIV and medical

complications of pregnancy were excluded. Risk factors (RF) examined

included history of sexually transmitted disease (STD) or hepatitis, drug use,

transfusion, positive hepatitis B (HB) serology and gestational age at entry into

prenatal care. A~r informed consent, HCV antibody was measured by second

generationenzyme-linkedimmunosorbentassay (ELISA). Positiveresultswere

confirmed by radioirnmunoblot assay (RIBA).

RESULTS: We tested 886 women over 12 months. Forty-three women

declined testing (40 with no RF, 3 whose only RF was STD). Forty-eight

percent of women had at least 1 RF; 22.9% had RF other than STD. Eighteen

percent h~d a history of drug use (2.2% intravenously). Thirty-four women

tested positive for ~ICV, 32 were confirmed by RIBA (prevalence 3.6%).

History of drug use, hepatitis, at least 1 RF including or excluding STD, and

positive HB serolog)" strongly correlated with HCV (p =0.13001). Twenty-four

percent of woman with HCV had no RF. HCV+ women were older (30.3 +

1. I, 23.8 + 0.2, p < 0.0001). Fifty percent ofwomce with HCV had abnormal

liver function tests. The absence of any RF had a negative predictive value of

98.7% The positive predictive v~lue of having at leost 1 RFother than STD

was 12% with a sensitivity mad specificity of 73.3 and 78.6%, respectively.

CONCLUSIONS: Risk factor screening is effective in identifying patients at

risk for HCV. In populations with a high prevalence of HCV, patients with risk

factors should be tested and counseled for long-term follow-up.

354 LONGITUDINAL STUDY OF GROUP B STREPTOCOCCUS CARRIAGE

IN PREGNANCY. J.R~cc~-Goodman. R.Berg~, R Gnbbfe, P.Me=er, S Fee.

Dept ObtGyn, Marshfle~d Clinic, Marshf~e~d Wisconsin.

OBJECTIVE: Th~s prospective study was designed to (1) determine the

prevalence of Group B Streptococcus (GBS) {n our obstetric populabon

and (2) evaluate the predictive value of lower vag~nal/perlanal GBS

cultures obtained m each trimester of pregnancy relahve to GBS culture

status at del~ve@

STUDY DESIGN: Lower vaginal/penanal GBS cultures were obtained

the first trimester, at 26-28 weeks, at 37 weeks and on admission

labor.The investigators were blinded to the results of all cultures except

that obtained at 37 weeks The sensitivity, specificity, pos~twe and

negative predictive value of each group of cultures with respect to culture

status at dehvery were determined and the pattern of GBS carriage =n our

patients dehneated

RESULTS: 984 patients participated ~n th~s Iong~tudtnal study with data

ava=lable on 881 thus far The prevalence of GBS carnage was 13.7% =n

the first trimester, 14 0% at 26-28 weeks, 12 6% at 37 weeks, and 12.8%

at dehvery GBS carnage was conhnuous (all 4 cultures pos~bve) m only

4.2% and ~denttfied on a single culture only in 6 1%. Sensltlvtty

specificity ($2), positive predictive value (PPD), and negative predictive

value (NPD) for each set of antepartum cultures w~th respect to culture

status at delivery are as tabled.

DELIVERY GBS STATUS

$1 $2 PPD NPD FIRST TRIMESTER 51 0% 91.8% 47 6% 92.8% 26-28 WEEKS 67.7% 93 8% 61.2% 95.3% 37 WEEKS 64 0% 94.9% 64.7% 94.8%

CONCLUSION: The pattern of GBS carriage in pregnancy ~s highly

variable Regardless of when antenatal GBS cultures are done, they serve

as poor predictors of maternal GBS carriage at dehver¥

356 GESTATIONAL DIABETES DOES NOT INCREASE PUERPERAL

INFECTION RISK FROM GROUP B STREPTOCOCCUS. J Piper,

S Georg~eux, O Langer Dept Ob/Gyn, UTHSC, San Antonio, TX

OBJECTIVE: To test the hypothesis that Group B StreptoCoccal colomzatlon (GBS) ~s associated with Ngher maternal and neonatal

morbidity in diabetic pregnancies compared to nondlabetlc pregnancies

STUDY DESIGN: Diabetic (DM) and nondmbet~c (NonDM) women were enrolled following vagmal/anogen=tal culture for GBS colomzat=on =n pregnancy Major outcome parameters were mtraamnlotlc infection 0AI),

endometntls (ENDO) and neonatal sepsis (NNS) Other maternal and neonatal factors analyzed included dehvery mode, maternal glycam=c

control, need for augmentation/reduction of labor, and need for admission

to the special care nursery Antibiotic prophylaxis was not routinely performed

RESULTS: 1,499 pregnancies were analyzed (453 DM, 1,046 NonDM) with GBS colomzatlon found =n 12 1% (126% DM vs 12% NonDM, pins) Infection rates by GBS and DM status were

Infection GBS+ GBS- Overall

Rates DM NonDM DM NonDM p

IAI (%) 16 12 10 11 ns

ENDO (%) 8 7 6 6 ns

NNS (%) 2 3 1 1 ns

n 57 125 396 921

There was a two4old higher NNS rate in GBS+ pregnancies overall [2 7% vs 1%, OR 2 34(1 1-5 0)] but IAI rates (13 2% vs 10 8%, pins) and ENDO

(7 6% vs 6 2%, pins) did not differ When controlled for other labor factors, neither GBS nor DM status was an ~ndependent predictor of IAI, ENDO or

NNS CONCLUSION: The presence of gestahonal dmbetes =n no way alters the

nsks of infectious morbidity associated with GBS colonization in pregnancy

Volume 174, Numbm l, Part 2 SPO Abstracts 409 Am ,J Obstet Gyne~ol

357 DO PREGNANT WOMEN HAVE FEWER SEXUALLY TRANSMITTED DISEASE SYMPTOMS? J .P=per, E Newton, R Sham, S Perdue, J D~mm=tt Dept Ob/Gyn, UTHSC, San Antonio, TX OBJECTIVE: To test the hypothesis that sexually transmitted d=sease (STD) symptoms are less dlst~nctJdlscernable dunng pregnancy due to the h=gher basehne gen~tounnary symptomatology Decreased recogmhon of STD symptoms could lead to delayed d~agnos~s and d~mmlsh impetus to alter h~gh risk behawor STUDY DESIGN: 581 women w~th an active STD (gonorrhea, chlamydla, trichomcnas) entered a prospechve, randommed study to reduce STD recurrence Each underwent detaded questioning regarding current symptomatology of the gen~tounnary tract, w~th physical exam~nahon and comprehensive STD testing at entry and again at 6 and 12 months following therapy Infected women (entry) and proven umnfected women (6 month follow up) were strahfied by pregnancy status for symptomatology comparison RESULTS: 180 pregnant and 401 nonpregnant women w~th an actwe STD were enrolled Unnary symptoms were s~gmficantly more common ~n pregnant infected women than nonpregnant ~nfected women [frequency RR 6 02 (4. 0-9 0), urgency RR 2 76 (1 94. 1), low back pa~n RR 2.52 (1 7-3 7)], but complaints of vaginal symptoms (d~scharge, rtcNng, odor) and pare (abdominal, groin, dyspareun~a, dysuna) did not d~ffer In contrast, when the pregnant infected women (n=180) were compared to pregnant nonmfected women {n=95) sigmficantly Ngher rates of vaginal symptoms [d~scharge RR 2 82 (1.6-4.9)] and pare [dyspareun~a RR 2 90 (1 2-7 5)] were reported by the pregnant =nfected women whde urinary symptoms were frequent =n both groups and d=d not d~ffer CONCLUSION: W~th the exclus=on of urinary complaints, recogn=hon of STD symptoms is not d=m~mshed ~n pregnancy desp=te background sympk~matology Appropriate solicitation and pursuit of STD symptems may positively ~mpact STD d=agnos=s in pregnancy

359 EFFECT OF PREGNANCY STATUS ON EFFECTIVENESS OF A

SEXUALLY TRANSMITTED DISEASE RISK-REDUCTION

INTERVENTION AT SIX MONTHS’ FOLLOW-UP: A RANDOMIZED TRIAL. R ShaiW, E. Newton, ]M. Piper, S. Perdue, R.

Ramos, J. Dimmitt, F. Guerra. Dept. Ob/Gyn, UTHSC, San Antonio, TX. OBJECTIVE: To determine if a theory-driven, culturally relevant behavior

modiflcatMn intervention is more effective if provided during pregnancy.

STUDY DESIGN: A behavioral intervention, based on integration of

extensive qualitative data (belie% values, behaviors) with the AIDS Risk

Reduction Model, was evaluated for high-risk, pregnant and non-pregnant,

Mexican- and A~ican-Amedcan women via a controlled randomized trial.

Ha~ng a baselin~ STD (gonorrhea, ch~amydia, syphilis or Trichomonas) was

a criterion for eligibility. Maior outcome variables were reinfection rates and

change~ in sexual ~havior at 6 and 12 months’ ~ollow-up. Pregnancy status

was a major independent variable because of It~ possible influence on

intervention effectiveness (pregnant women may be less affected because of

diminished sexual activity; conversely, they may be more affected because

of heightened concerns for the fetus that may act as a trigger for behavior

change). The Mantel-Haenszel Chi-Square Statistic was used to determine

intervention effects, controlling for baseline pregnancy status.

RESu£’r$: 314 women were randomly a~igned to the study group and JOG

served as controls; 27% and 33% of these groups, respectively, were

pregnant. Attendance at intervention sessions was high: 90% of the assigned

women attended the l~t se~ien; 75% attended all 3 sessions (3 hours each) and 82% attended 2 of the 3. The sample retention rate at 6 months’

follow-up was 82%. Preliminary analysis indicates that the crude reinfection

rate with either gonorrhea or chlamydia among non-pregnant study-group

women was 9% w 15.5% for controls; the rate among pregnant study-~roup

women was 8% vs 14% for controls (P==.04 for study vs control across

both pregnancy status groups). CONCLUSIONS: A culturally relevant risk-reduction intervention was

equally effective in preventing reinfection among pregnant and non-pregnant

minority women.

358 EVALUATION OF THE CURRENT CDC RECOMMENDED

TREATMENT GUIDELINES FOR GONORRHEA IN PREGNANCY.

R_R Ramus, J Mayfieldx, O Wendel Dept of Ob/G~m, Uinv of Texas

Southwestern, and Dallas County Health Dept, Dallas, TX

OBJECTIVE: No mfurmation has been published on the efficacy of the

1993 Centers for Disease Control (CDC) recommendations for rite treatment

of uncomplicated gonorrhea m pregnancy This study is designed to

prospectively evaluate those gardelmes

STUDY DESIGN: Sixty-two women referred with probable endocerv~cal

gonorrhea since April 1994 had pre-treatment endocervtcal, anal, and oral

cultures for Neisser~a gonorrhoeae They were assigned by a random

number table to receive ceftnaxone 125 mg intramuscularly (IM) or

cefixume 400 mg orally Treatment was unbhnded and m a 1 I dlstrlbutmn

Test of cure cultures were performed 4-10 days after treatment

RI~SULTS: The success in eradlcahng Netssena gonorrhoeae according to

site of Infection and treatment rei

Site

Cerwx Pharynx Anus Cervix and/or anus (mmgemtal)

Overall 95% CI

tmen is summarized m the table below

Ceftnaxone l

Cefix~me

25/27 (93%) 26/27 (96%) 5/5 (100%) 4/4 (100%)

14/14 (100%) 11/11 (100%) 27129 (93%) 31/32 (97%)

28/30 (93%) 31/32 (97%) 77 9%-99 2% 83 8%-99 9%

Two of the three treatment t’atlures (one m each group) a&mtted to unprotected ~ntercourse prior to thear test of cure culture Etght women

(13%) had I]-lactamase producing strains of N gonorrhoeae, four in each treatment group All of these women were successfully cured

CONCLUSIONS: Both IM ceftnaxone at the 125 mg dose and oral cefixmle appear to be effective for the treatment of gouorrhea In pregnancy

No conclusions comparing the two regimens can be made at tilts tinle wtth the current number of subjects, patient recrmtinent is ongoing

360 DETECTION OF CHLAMYDIA TRACHOMATIS AND

TRICHOMONAS VAGINALIS DURING PREGNANCY BY

INTROITAL SAMPLING. SS W=tkmx, SR Inghs, M.

Polaneczkyx, Depts. Ob/Gyn, Cornell Unw. Med. College and

Jersey Qty bled, Center, New York, NY and Jersey Qty,

NJ. OBJECTIVE: Many pregnant women are reluctant or unable

to undergo a speculum examination for detection of genital pathoger~s. We evaluated whether detection of C. trachomat~s

and T. vaq=nahs m the mtro~tus by polymerase cha=n

react=on (PCR) was comparable to PCR analys~s of vaginal

and endocerwcal samples.

STUDY DESIGN: Endocerwcal, posterior vag=nal and Introltal samples from pregnant women were tested for C.

trachomat=s and T. vagmahs by PCR.

RESULTS: C. trachomat~s was detected ~n the endocervices

of 36 of 300 women (12.0%). In all but one of the positive

women, and in none of the negative cases, the mtro~tal

specimen was also pos~twe. T. va~ina!!s was detected m the

vaginas of 22 of 219 women (10.0%). All but one of the

positive women and none of the vagina PCR-negat~ve women

also y~elded pos~twe mtroffal samples. Compared to the

endocervical and vaginal PCRs, the mtro~tal testing had a 100% spec~fic=ty and a 97.2% and 95.5% sensmv=ty for

detecting C. trachomat~s and T_. vaqmabs, respectively.

CONCLUSIONS: Introita~ samphng and PCR ana~ys~s can

greatly increase the numbers of women at risk for infect=on

who can be tested for genital pathogens.

410 SPO Abstracts .]anumT 1996 Arn J Obstet Gynetol

36l DETECTION OF GENITOURINARY TRACT CHLAMYDIA INFECTION IN PREGNANT WOMEN BY LIGASE CHAIN REACTION ASSAY. WW Andrews H. Leex, WJ Rodenx, CW Moltx. Oept OB/GYN, Ur~iv. Alabama at B~rmingham, AL and Abbott Laboratories, Abbott Park, IL. OBJECTIVE; To determine the sensit~wty and spec~hcity (sans/spas) of a ligase chain reaction assay (LCR) performed in cervical swabs and voided urine to detect genitourinary (GU) tract Chlamydia trachomati~ (CT) infection in pregnant women. STUDY DESIGN; Matched cervical swab and voided urine samples were collected from 462 women at routine prenatal visits for evaluation using CT culture (cervix) and a LCR assay specific for CT-plasmid DNA (cervix and urine). The standard for GU refection against which each test was compared included all women w~th a positive culture plus those w~th a negative culture but a pos}tive LCR in either the urine or cervix that was confirmed by supplemental teshng w~th a direct fluorescence antibody or LCR assay specific for the chlamydia major outer membrane protein gene. RESULTS." The prevalence of genitourinary CT infection was 6.1% by cervical culture (28/462; sens/spec=30.1%/100%), 18.2% by cervical LCR (84/462; sens/spec=90.3%/100%), and 16.9% by urine LCR (78/462 sens/spec=84.9%/99.7%). The respective sens/spec to detect cervical CT infection was 82.8%/97.9% for LCR ~n voided urine and 96.6%/100% for cervical LCR. Cervical LCR detected 89.3% and urine LCR detected 82.1% of women with a positive cervix culture. CONCLUSIONS: In pregnant women, LCR ~n the cerwx or unne is superior to cerwx culture in the detection of GU chlamydia infection. Urine LCR is a simple and effective screening test for GU tract chlamydla ~n pregnant women.

363 VARICELLA-ZOSTER IMMUNE GLOBULIN (VZIG) FOR THE AT-RISK GRAVIDA? A DECISION ANALYSIS. DJ Rouse, M Gardner, S Allenx The Unwerslty of Alabama at B~rmmgham, AL OBJECTIVE: To compare three strategies for the management of the vartcel~a (VZ} exposed gray,de with no prior history of VZ: 1 ) Observation; 2) Immune Testing and admmmtratlon of VZIG to those who test non-immune; 3) Universal Administration of VZ]G without immune testing. METHODS: Cost-effectweness/cost-beneflt analysis. Baseline assumphons: 80% VZ immunity rate, 95% test sens~hvlty and 84% specificity, 90% transmission rate # exposed and susceptible, test cost-$25, VZIG cost-S400. I~ecause precise data are unavallaNe about VZ mortality rates (VZMR’s) and severe sequelae rates in pregnancy, e~ther with or w~thout VZIG, a range of potential VZMR’s was evaluated, from 31/100,000 (non-pregnant adult rate) to >5,000/100,000. The potential efficacy of VZIG In preventing death and severe sequelae was varied from 1%-99%. A strategy was defined as cost-effective If it cost <$50,000/hfe-year gamed. RESULTS: At 99% VZJG efficacy, Immune Testing was cost effechve compared to Observation when the VZMR was >70/100,000 0e., 2.3 X the non-pregnant adult rate) At 50% efficacy, Immune Testing was cost-effechve at VZMR’~ of >150/100,000, and at 1% efficacy, at VZMR’s of >7,640/100,000. Universal Adm~mstrat~on was h~ghly cost ineffective (>$700,000/hfe- year gained) when compared incrementally to Immune Teshng However, ff ~mmune testing ~s not available, Unwersal Administration is cost-effectwe compared to Observation under relahvely more ophm~st~c VZIG efhcacy and more d~re ZV sequelae assumptions The analys~s was sens~hve to the VZ transmission rate and the d~scount rate S~mdar results were obtained in a parallel cost- benefit analysis. CONCLUSION: Even though VZIG efficacy and VZ morb~d~ty/mortahty rates m pregnancy are imprecisely defined, from a sost-effect~veness/cost-beneht standpmnt, management based on Immune testing ~s preferable over a wide range of assumptions to either observation or umversal VZIG administration when canng for the VZ exposed grawda w~th a negahve Nstory of VZ

362 POLYMERASE CHAIN REACTION (PCR) TECHNOLOGY FOR

DIAG NOSIS OF CHLAMYDIA TRACHOMATIS DU RING PREG NANCY

IN URINE AND CERVICAL SWABS. M Chatteqee, A Humphrey Dept

Ob/Gyn, Meharry Medical College, Nashvdle, TN

OBJECTIVE: The goal of this study was to find out If PCR ~s more

sensitive than Genprobe for diagnosis of Chlamydla trachomatls in urine

and cervical swabs dunng pregnancy

STUDY DESIGN: This prospectJve studywas IR6 apt~roved A total of

127 pregnant patJents at Metro Nashwlle General Hosp,tal and Meharry

OB/Gyn Clmsc had 471 taet~ performed Each patient had the fo{Iowlng

5 tests performed at the first prenatal wslt 1 Genprobe (GP), 2 PCR

cervical swab (PCR-S), 3 PCR-urme (PCR-U), 4 Chlam;~d~a culture

(CC) and 5 Unne leukocyte ssterase (ULE) Clinicians were bhnded for

PCR-S and PCR-U. All five data sets were avadable m 50 patients at the

tJme of th~s report

RESULTS: In all eases of CC pos~twe samples, leukocyte e,,~terase

results were greater than 3+

PCR-S+ PCR-S- PCR-U+ PCR-U- GP+ GP-

CC+ 3 0 2 1" 3 0

CC - 1" 46 0 47 0 47

CONCLUSIONS: Our data suggest that PCR-S and GP are equally

sensitive for diagnosis of Chlamydla using culture as the gold standard

Unne PCR tasting can detaet (67%) of infections PCR being a sens~twe

test, study ~s underway for dmgnosls of ch{amyd{a by self collected

intrmtal swabs GenProbe ~s more cost effective than PCR

364 A RANDOMIZED TRIAL OF ERYTHROMYCIN AND

AZITHROMYCIN FOR THE TREATMENT OF CHLAMYDIA

INFECTION IN PREGNANCY

M Rosenn, M.D, GA Macones, M D and N Silverman. M D D~wsion of

Maternal-Fetal Medicine, Thomas Jefferson Umvers~ty, Phfla, PA

OBJECTIVE: To compare erythromycm and az~thromyc~n for the

antepartum treatment of chlamy&a cerv~citas

METHODS: In a prospective manner, pregnant patients with cerwcal

chlamydla rnfections as dlagaosed by rout*ne screening tests were

randomly assigned to recmve either erythromycm, 500 mg four tmaes

dady for seven days or azzthromycm, 1 gm as a one tune dose All

partners were g~vee prescnpnons for doxycychne, 100 mg twice daily

for seven days Treatment efficacy was assessed by follow-up chlamyd~a

testing, performed three to four weeks after therapy was completed. Side

effects, mtolerance to therapy and overall compliance, were evaluated by

means of a standar&zed post-treatment questmnnatre

RESULTS: There was no sigmflcant d~fference ~n cure rates noted

between the erythromycm group and the azithromycm group, 77% vs

91% respectwely (p=0.24) GastrmatesUnal side effects were reported

more frequently among pataents treated w~th erythromyc~n compared w~th

patients treated w~th azithromycln, 45% vs 17% (p=0 004). Patients

who received erythromycin reported intolerance to therapy, secondary to

side effects, more frequently than d~d patients who received az~thromyc~n,

23% vs 4% (p=0 07) Furthermore, patients in the az~thromyczn group

were more hkely to complete their course of therapy as prescribed, than

were patients rn the erythromyc~n group, 100% vs 61% respectively

(p=0 002)

CONCLUSIONS: Azzthromycln is efficacious and well tolerated for the

treatment of chlamyd~a cervicit~s ~n pregnancy Erythromycln, though

efficacious, ~s poorly tolerated as demonstrated by the number of patients

reporting sigmflcant side effects during the course of therapy The

present study supports the use of az~thromycln as an alternative to

erythromycm for the t]’eatment of chlamy&a cerwcitls zn pregnancy

Voh|mc 174, Numher 1, Part 2 SPO Abstracts 411 Am ] Obstet Gynecol

365 ATTITUDES TOWARD HEALTH CARE AND HUMAN IMMUNODEFICIENCY VIRUS INFECTION AMONG PREGNANT INNER-CITY WOMEN. N Sdverma~_,n D Rohnerx, L

Marksonx, B Turnerx Depts of Ob/Gyn & Medicine, Jefferson Medical College of Thomas Jefferson University, Phda, PA OB_lECTIVE:Permatal treatment Wl~rt zldovudme (ZDV) m women infected with the human immunodeficiency virus (HIV) has been shown to lower the risk of neOnatal HIV infection, leading to calls for expanded I-I~V testing m pregnant women Our objective was to explore actual attitudes toward health care, I-~V refection, and

~ renatal therapms m a cohort of pregnaJqt umer-qty women TUDY DESIGN: A 23-item questionnaire was administered as a

structured one-on-one interview to 75 women enrolled In one hospital’s prenatal chnic Answers were evaluated for associations witt~ both reported drug use and elicited HIV-speclfic attitudes ]KESULTS: Half of the women surveyed acknowledged recent illicit drug use, 65% smoked cigarettes Drug users were more hkely than non-drug users to defer prenatal care urinal later In pregnancy (24% vs 5%; p = 0047), and to say that maternal drug or alcohol use presented a low or moderate risk to the pregnancy Overall, only I5% of women said that pregnancy was "OK" for an I~IIV+ woman, 70% stated they would abort a pregnancy it they were diagnosed as H1V-mfected Non-drug users were twice as likely to say that they would corttmue such a pregnancy (40% vs 18%, p=0 06) Two-thirds of women said that maternal--neonatal H’IV transmission occurred "always" or "most tunes," only 20% had heard of medication that might lower the risk of neonatal lnfechon Still, 92% of women said they would take such a medicine during pregnancy, even In the absence of clear maternal benefit A woman’s having heard of ZDV increased the acceptablhty of prenatal anhwral therapies (100% vs 82%, p=0 007) ~ONCLUSIO~NS: The women m th~s study showed deficmnoes m use of health care services, and had limited understanding of the ~eonatal rinks o~ matet’nal ~-~V mIect~on Still, potentzal treatments during pregnancy to lower those risks were overwhelmingly acceptable, regardless of drug use status or other health-care attitudes These findings warrant cons~dera~aon m the formulation of pubhc pohcy regarding prenatal HIV screening and management

367 LONGITUDINAL QUALITY OF LIFE ASSESSMENT IN PREGNANT WOMEN WITH HIV. K.D.Larrabee’, M.Monga,

N.LEriksen, A.Helfgotr. Dept.ObGyn & Reprod Sci,UTMSH, Houston, TX.

OBJECTIVE: TO describe the perceived quality of life and functional

status of women wi~ HIV durin$ the antenatal, l~inatal and

STUDY DESIGN: Medical Outcome Su~ey-Sh~t Form (MOS-SF) questionnaires were completed during antenatal visits, 24 hours after delivery and 6 months ix~stpartum by 21 HIV-positive women and 21 HW- negative controls matched f~ age, race, perity, and ed~catim. The MOS-

SF measures subject perception of overall Imalth, pain, physical role, social and cognitive function, mental health, craggy/fatigue, health distress ~ality of life and health transition. Median steres between 0 and 100, (0 indicating poorest health), were compared using the Ma~m-Whltney U test. RESULTS: All HIV-posidve patients were asymptenmtic; mean CD4 count was 386 on entry into the study. HIV-positive patients had an i~re~ed smse of social function (70 vs ~7, p-0.02) but relxrted i~cre~ed health distress (35 vs 66, p<0.001) during antonatal visits. During the lzrimtal period, HlV-positive patients had decreased s~se of overall health (40 vs 80, p<0.001) and physical function (50 vs 67, p-0.006). Six months lmStpmaun, the HlV-positive women relxxted decreased parceived mental health ~50 vs 77, p<0.001) m~l worse social function (33 vs 83, po0.015). CONCLUSIONS: This is the first longitudinal evaluation of parceived quality of life in HW-positive pregmm subjects. We ce~clude that parceived quality of life differs between HlV-pesitive md HW-negative Ixegnant women. These differences may t~t manifest dmlng initial antenatal visits but may develop as pregnancy, the disease process and other Ule evems SlX~ to a~ivery ~a~ the pos~p~mm period ime~am

overall lxrceived quality of life. Longitudinal evaluation of quality of life issues may be important in the compreMnsive cam of HIV-~osltive women

366 HIV-1 TRANSMISSION FROM MOTHER TO CHILD: ANALYSIS OF

OBSTETRICAL, MEDICAL AND IMMUNOLOGICAL DETERMINANTS. M. Bouchel~, N Lapolnte~, J. Samsorl×, M. Fauve]×, T. Tran×, c.

HanklnSx Centre maternel et infantile sur le SIDA, $alnte.Justlne biospltal, Laborato~re de sante publlque du Quebec, Centre for

AIDS studies, Montreal General Hospital, Montreal, canada.

OBJECTIVE: TO assess the role of obstetrical, medical and Immunological factors on mother-to-child HIV-1 transmission in

a prospective cohort of ~nfected women.

STUDY DESIGN¯ This studies involves 108 women and their 121 children Antepartum z~dovudloe (AZTI was taken by 28 women.

Clinical/immunological data were collected at each trimester and

at delivery Statistical analysis was done by Wilcoxon rank sum,

median and Student’s t- tests. Significance was set at p < 0.05.

RESULTS: Transmission In AZT treated patients was reduced by

?4.9% (30 7% to 7.7%). In v~ew of the major effect of AZT on

transmission, other possible determinants were analyzed after

exclusion of these patients. No relationship could be established

between maternal age, parity, gestatlonal age, clinical or

Immunological status, duration of ruptured membranes, internal

FHR momtorlng or dehvery mode. An unreported association

was found between the duration of the 2n~ stage of labor and

HIV transmission (66.5 + 26 S minutes in transmitters vs 16.7 +

12.0 minutes In non-transmitters, p=o.0009). This observation

was valid in pnmlparous as well as ~n mUltlparous patients

CONCLUSION Our study supports previous reports of a

tremendous decrease in maternal-fetal transmission of HIV with

AZT Our observation of a significant relationship between the

duration of the 2n~ stage of labor and transmission suggests a

need for further studies on the possible role of shortening the

2no stage of ~abor or illtrapartum decontamination of the

cerwco-vaglnal canal. Furthermore, this ~s concordant with the

current hypothesis that the majority of children are infected at

or around birth

368 LOCALIZATION OF HUMAN IMMUNODEFICIENCY VIRUS TYPE 1 (HIV-1)

IN SPECIFIC PLACENTAL CELLS BY DUAL IMMUNOPHENOTYPING AND

IN SITU POLYMERASE CHAJN REACTION (IS-PCR) AMPLIFICATION. PM Garcia, D Jiyamapa~, D Stenislawsk~, BK Patterson~. Oepts of OB/GYN and

MEn, Northwestern Univeisity, Ch=cago, IL

OBJECTIVE To describe a method of arnpl(rylng and detecting HIV-1 within

speofic placenta~ ceils

STUDY DESIGN I METHODS: Placentas of four consecuhve HW*I ~nfectsd

women dehvenng at term were fixed =n a water soluble fixat=ve (Streck T~ssue F=xahve~) and paraffin embedded Multiple sect=ons of each placenta were

screened for the presence of HW-1 DNA by phenol-chloroform extracbon of nucleic acK~s and ~ wflh HIV-1 gag pnmers (SK38/SK39) Amphfied

product was detected non-lsotoplcally with a chomllum~nescent probe hybr~zat]on resctx)n (GeoProbe~) Placentas which screened posPdve for HIV-

1 DNA underwent IS-PCR Placental bssue secbons were adhered to s~lan~zed skies (Sigma®), dgested w~th proteinase K and arnplff~od using a hot-start m

s~tu PCR protocol developed by one of the authors (BKP) The amphficabon d~rec~y ilxx)q)oratad a fluoceecent (5-carbexyfluorescem) ohgonucleobde which

could be detected on fluorescence confocai ~mag~r~9 w~th argon laser ~luminabon a~k)w~g d}rect Iocabzal)on of amphfied product within the nucie) of

ce~ts (fig1) Pnor to fixation, d]gesfion and thermocychng, ~mmunophenotyp~ng

of placental cells was carned out with b~ohn-coniugated anfJ-CD4 and CD68 antib<x~es

RESULTS: HIV-I DNA was detected ~n one of four ~

term placentas HIV-I DNA was predommataly ~ kxmhzed in CD4 and CD68 pos~hve pMcental cells ~ Fiu~resence was also detected ~n ce~s which ~ resembled ac~vated endothehat cells ~

CONCLUSIONS In sltu PCR, by its abihty to ~ localize ampl~ed s~gnal wthm the nucle~ of cells, ~s ~

potentJally an accurate and eens~hve means of ~ detsrm~ng placental mfecbon ~ HIV-1 This ~ fig t technK~ue should help ~uo~date the role of placental confocal microscope image,

mfect~on ~n bennata~ H~V-1 transmission 630×, p~acenia, 37 weeks

412 SPO Abstracts .January 1996 Am l Obstet Gynecol

369 DIFFERENTIAL RATE OF TROPHOBLAST INFECTION AMONG HIV

QUASISPECIES A S Goustm~, R. A. Thomas×1, V. Zachar×1, V.

Zacharova×~, T B Jones2. Center for Molecular Medm~ne~ and

Department of Ob/Gyn~, Wayne State Unwermty School of Medm~ne,

Detroit, MI

OBJECTIVE: To evaluate the range of HIV-1 quasmpemes ~n a

delivering mother and determine what degree of variation exists in

their abd~ty to ~nfect trophoblasts cells.

STUDY DESIGN: A matched set of mother’s blood (peripheral blood

mononuclear cells[P~3MCs]), purified placental trophoblast cells, and

umbfimal cord blood (PBMCs and plasma) was obtained at the t~me

of dehvery. Amphflcatmn of the C2V3 env regmn of all three cell

sources was performed umng polymerase chain reactmn (PCR). The

PCR products were cloned =nto plasm~ds wh=ch were then sequenced

using the Taq cycle sequenmng method.

RESULTS; 19 cord sequences, 16 trophoblast sequences, and 22

maternal sequences were ~dentff~ed and represented substantial

mtrasample vanat~on in quasJspec~es. However, the resulting

phylogenetm trees revealed that maternal sequence variation was

greater than rather trophoblast or cord blood provwus: the most

divergent maternal sequences were 7.9% different, whde the

numbers were 3.1% and 1.4% for cord blood and trophoblast,

respectively.

CONCLUSION; These data are suggeshve of a genetm "bottleneck"

=n the trophoblast Th=s phenomenon may aid in narrowing the

sequence d~vermty among maternal HIV quas=specms and may allow

only a I~m~ted number of maternal v=ral quasmpecms to estabhsh

=nfection (prowrus) in trophoblast cells. Thin prewously unreported

observatmn appears to be another mechan=sm by which verUcal

transmmsmn may be reduced by the host.

371 QUALITY ASSURANCE OF A PERINATAL ULTRASOUND LABORA-

TORY: RATE OF SUB-OPTIMAL ORGAN VISUALIZATION VERSUS

WORK LOAD. IE Zador~, L Chtk*, V Salan x, M Treadwell, R J Sokol Dept Ob/Gyn, Hutzel Hosp/Wayne State Umv, Detro=t, MI

OBJECTIVE: To ~nvesbgate the relat=onsNp between the number of scans

performed per day by each sonographer (work load) m a busy chnlcal

practice and the rate of wsuahzat=on of fetal organs

STUDY DESIGN: A bed-side mterachve data management and report

generation system was introduced m the ultrasound laboratory m July 1994

The system was des=gned to increase producUv=ty and enhance quahty of

sonograph~c work As a process of quafity assurance, the da~ly work load and

the rate of subopt=mal wsuahzat=on of 6 fetal organs (face, postenor fossa, 4

chamber heart wew, outflow tract, hand and lower spree) were revtewed for

a consecut=ve penod of 359 work days The organs chosen were an

=mportant subset of the 36 organs w~th=n a standard=zeal departmental

ultrasound screemng protocol Patients and type of ultrasound hardware

(Acuson, ATL, Siemens) were randomly assigned to 9 registered sonographers (2 to 7 years of expenence) supervised by 7 pennatolog~sts Over

1000 fetal scans per month were performed on average Correlation analys~s

was apphed to compare the rate of subophmal wsuahzahon of these organs

to the work load

RESULTS: There had been a s~gmficant ~ncrease m work load ~n the study

penod (r=-0 64, p = 000) The mean work load was 8 2 :t 2 2 pahents per day

per sonographer The subopt~mat wsuahzahon (SV) rates for each organ

decreased with work load (WL) and days of computer s stem use (DCSU)

Face P fossa chamb Outflow Hand Spine

SV rate % 188 09 141 244 219 192

correlatlonwlthWL -035 -011" -028 -035 -030 -040

c~rrelat~on wRh DCSU -0 50 -0 06* -0 46 -g 58 -0 44 -0 6f I

Correlation coefficient (r= 135 for p=0 1,*=NS)

CONCLUSION: Gwen a pool of well-trained sonographers w~th adequate

supervision and avallabd~ty of modern equipment, supported by computerized

data management, the aNhty to wsuahze fetal organs has been enhanced

At least at tNs level of pahent flow, ~t does not constitute a challenge to the

quahty of sonographers’ performance

370 APPLICATION OF RADIUS STUDY CRITERIA TO PRIVATE AND CLINIC OBSTETRIC POPULATIONS. Mwabde JL~, CL Chert~, RK Jaelde, BA Meyer, VM £arisi. Dept, Ob/Gyn, University Medxcal Canter, SUNy @ Stony Brook, Stony Brook, OBJECTIVE: Our hypothesis is that the ehgibd~ty and inclusion criteria of the RADIUS study are not representatave of either a private office or lower socmecononue clinic populatmn and therefore has Incited applicability. METHODS: Retrospective chart review was performed on 406 consecutave obstetrical patients from the private faculty obstetric (P) and a predominantly Medicaid funded resident clinic (C) population Deraograptuc data was extracted and applied to both eligibility and exclusion criteria as defined by the RADIUS study (R). One-way ANOVA and p<O05) RESULTS: After apphcahon of RADIUS inclusion/exclusion criteria, significantly fewer patients met ehgibility (R 39 4%, P’ 26 8%*, C’ 13.1%*), In contrast to RADIUS findings, our populations hnd signifieamtly greater numbers of women <20 years old/> 35 years old (R: 5 0%, P 22.5%*, C: 24 6%*), minoritzes (R: 7 0%, P. 169%*, C ~39 5%*), and current substance abuse (R. 35 0%, P 19 9%*, C 64.0%*) There were s~gnificant &fferences between the private and clinic populations with respect to race, current substance use, gestational age < lg weeks (P 56 0%vs. C’ 22 0%*) and eligibility (P 26 8% vs C: 13.1%*). CONCLUSION: RADIUS study conclusions are not applicable to private or clinic patient population at our restitution, The small numbers ofpaUents meeting ehgibility and i~cluslon cntena severely lmuts RADIUS study conclusions m both private and clinic populations

372 VOLUMETRIC DISPLAY OF TItREE-DIMENSIONAL

ULTRASOUND DATA FOR TELEMEDICINE

APPLICATIONS IN PERINATOLOGY. C M,uccdoma ~

L~tt[efm[d~, D Collinsz ~Ocp, utmcn/ of O~/GYN, Mad~gan A~my

Mcdlca~ (TcntcL "lucoma, WA 2Pat’din Noltawcst Labo~aRa-=cs,

R~d~hmd, WA

OBJECTIVE Ophm~zc the ~olumelr~c d~spla3 s3stcms and hie

J ~l le}emc(hcme ,Ippl~cuium m pci

STUDY DESIGN. 24 gmv~d volunteers vxem eng~l!ed m the

stmly F.*ch had coevcntum,d 2D actccnmg tt[h-aaound at [6-20

wceks gcslatlon Using ou! previously denned continuous linear

,tcqtustUon mcdx~d, we obMmcd nnagc s{acks containing muir+pie

pa=allcl 21) sonoglaph~c h-ames Volomc v~suahTat~on was Ihcn

accx*mphsh~l uszng scvc~v~l pubhc domain soltwm-c packages

wclc succcsslul[~ tlanqommd into TIPP, P1CT, and

RESULTS We were ant to succcsshdly demonstrate thai

combination w=th w~dely avadable w~uallzalioa ellglnes m~d

pcrlnaR)loglsts

CONCLUSION Th~s ~es~ch m 3D ultms~und

Ihdt mcxpcns~vc mod~hcat~ons to slanda~d 2~ u!Imsound cqtupmcnt

Volume 174, Numbei l, Part 2 SPO Abstracts 413 Am ] ()betel Gynecol

373 NEUROLOGICAL TESTING IN .U.LTRASOUND EXPOSED INFANTS

K Belangg_~ JC Hobhms, IP Muller’, S Howardr’Dept of Ob/Gyn, Yale-New

Haven Umverslty Medical Center, New Haven, CT, University of Colorado

Health Sciences Center, Denver, CO, Gessler Clinic, Wlnterhaven, FL

OBJECTIVE: To assess the effect of in utero ultrasound exposure on

neurological development in childhood

STUDY DESIGN: A randomized chmcaI trial was conducted in Wmterhaven,

Florida (1986 90) to study the satety of ultrasound Low risk patients were

randomized to have 2 scans (16-20 weeks and 30-36 weeks) or, in the control

group, scans only when clmtcally indicated A subset of 286 singleton infants

born to mothers in the study were given Bayley evaluations at 6 and 18 months,

each test yielding 2 scores, Mental DeveIopmental Index (MDI) and

Psychomotor DeveIopmental (PDI) Repeated measures analysis of variance

was used to compare the two groups Two analyses were conducted, an

"Intention to treat" analysis and a comparlsoo of children exposed to ultrasound

vs unexposed.

RESULTS: I) There was no difference m MDI or PDI scores for children

whose mothers were in the scan vs the non scan group 2) Exclu~,mg 3

children in the scan group whose mothers did not have scans and 36 children

In the control group whose mothers needed scans, did not change the results

3) MDI was influenced by materoal education (p= 02) and infant sex (p= 01)

Girls and children whose mothers had more education scored higher

CONCLUSION: The analysis confirms that ultrasound used twice in

pregnancy at diagnostic intensities does not have any measurable effect on MDI

or PDI scores at 6 and 18 months of age

375 .50ROC-P~R1C £~LURTIO~ OF C£RVlCRL L£RGTH: DOES IT

pP~DICT 5~C£5~FUL OVI’COR£ OF IJJ~OR I/~DI~TIO~? ~_..

Watso___..__~n, D. Stevens, S. Welter, O. Day. Sioux Valley

Hospital, Sioux Fails,

OBJECTIVE: The purpose of this study was to sonograph[caIly evaluate cervical length in patients undergoing induction of labor and to compare this with the Bishop’s score factors and parity in predicting the number of hours to reach active phase labor. STIIDSDESIGH: In ~09 patients undergoing tabor induction we prospectively studied cervical dilatation, effacement, station, position, consistency, parity, and sonographic cervical length. A multiple regression mode[ was used to

determine which of these factors determine the number of hours from the beginning of induction to the onset of

RE~ULT~: A model using all of these 7 factors was predictive in determining the number of hours to reach the active phase 1F=32,1, P < ,000ll, Stepwise regression indicated that only cervical d[latation was independently predictive of time to reach active phase. There was a significant correlation between the clinical assessment of cervical effacement and the sonographic estimation of cervical length, P < .001, CONCLt~IONS: Only cervica! dilatation appeared to be predictive of the number o£ hours to reaqh active phase, The sonographic evaluation of cervical length did not add significant independent informetion,

374 YOLK SAC DIAMETERS: HOW HELPFUL ARE THEY?

CL Brown~, MC Treadwell, HM Wolfe. Dept. Ob/Gyn, Hutzel

Hospital/Wayne State University, Detroit, MI.

OBJECTIVE: Enlarged yolk sacs have been associated with

chromosomal derangements and embryonic demise. However,

ranges for abnormal or macro yolk sac diameters are inconsistently

reported. Our purpose was to better define normal yolk sac size and

examine ultrasound findings associated with enlarged yolk sac

diameters.

STUDY DESIGN: 986 first trimester sonograms from 4/1/94

through 6/30/95 were analyzed to deternune ranges and percentiles

of yolk sac measurements for gestational age based on crown-rump

length.

RESULTS: The median yolk sac measurement was 4ram (mean

3.Tmm) with a range of 2-10mm and showed no association with

gestational age. Abnormal or macro yolk sac was defined as

>5.4ram (two S.D, above the mean). 61197 available karyotypes

were abnormal including 4 cases of aneuploidy (T13, TI8 and 2

cases Tri 21) and 2 translocafions. None of the abnormal

karyotypes had an enlarged yolk sac diameter. 12/21 patients with

a yolk sac diameter >5.4mm were abnormal: 10 embryonic

demises and 2 with structural abnormalities. The remaining 9

normal pregnancies included 1 multiple gestation.

CONCLUSIONS: We did not demonstrate an association between

yolk sac diameter and abnormal karyotype. There was an

association between macro yolk sac and adverse pregnancy

outcome. At this time further study is needed to determine

appropriate counselling of patients with normal appearing

pregnancies and macro yolk sacs.

376 CERVICAL LENGTH BETWEEN 14-24 WEEKS AND RISK OF

PRETERM B[RTI~I.L Allbert, N.Davls~, Department of OB/GYN, Carolinas

Medical Center, Charlotte, NC

OBJECTIVE:Determine if shortened cervical length (CL) between 14-24 weeks

m patients with a previous delivery between 12-28 weeks predicts spontaneous

preterm b~rth (SPTB) or cervical funneling (CF)

STUDY DESIGN:A retrospective review of all patients who had transvagmal

cervical sonography (TVCS) performed over a one year period was undertaken

Twenty-five patients who had a previous second trimester delivery (12-28 weeks)

and who did not have a prophylactic cerclage placed were selected for review.

The last TVCS CL recorded between 14-24 weeks was correlated with the

hkekhood of SPTB or CF requmng rescue cerclage.

RESULTS:Of the 25 patients, 12 delivered preterm and 4 had a cerclage for

cervical funneling The mean CL was 27.8:t:9 0 mm. Of the 3 patients with

premature rupture of membranes (PROM), the mean CL was 23.3+0.58mm

compared to those without PROM 28 4+9.48 (p < 0.05).

CERVICAL LENGTH

-<30mm >30mm p value

n=17 n=8

SPTB 11 (65%) 1 (13%) 0.015

SPTB or Cerclage 14 (82%) 2 (25%) 0,005

PROM 3 (18%) 0 (0%) NS

CONCLUSION:Cervical shertnnmg (< 30ram) is predictive of preterm delivery

or cervical changes warranting placement of a rescue cerclage Subsequent

prospective studies are warranted to further evaluate these findings

414 SPO Abstracts .[anual’y 1996 Am J Obstet Gynecol

377 ULTRASONOGRAPHIC FOLLOW-UP OF CERVICAL CERCLAGE PLACEMENT. A Lysikiewicz, J. Canter=no, R.P Robinson, N.Tejani. Dept. Ob/Gyn New York Medical College, Valhalla, N Y OBJECTIVE: To determine whether the therapeutic benefit of cerclage correlates with the location of the cervical suture. STUDY DESIGN: 28 patients with a documented history of cervical incompetence were sonographically assessed following cerclage placement for the location of cerclage. Location of the cerclage was described as the distance from the external os and expressed as the ratio to the total length of the cervix. Outcome was defined by gestatJonal age at dehvery. PafJents dehvered for maternal and fetal indications, unrelated to cervical incompetence, were excluded. Gestational age at delivery was compared to the site of cerclage placement RESULTS: 67 % of cerclages were placed in a distat half" of the cervix, close to the external os (Fig.l) Despite this location, 85 % of the patients delivered after 34 weeks (Fig.2)

t.2 LOCATION OF THE CERCLAGE * SPONT LABOR

t o MD INDUCED ~ *

(~ONCLUSION: The site of cervical cerclage placement does not

alter length of gestation. The success of the cerclage in a suboptimal location suggests mechanisms other than mechanical support.

379 ASSOCIATIONS OF NON-DOPPLER-FLOW PLACENTAL CYSTIC LESION WITH SMOKING, ELEVATED MATERNAL SERUM ALPHA- FETOPROTEIN AND PREGNANCY OUTCOMES. M.G Pinette. Y Pan

S.C. Pinette, J. Blackstone, R Chard, Dept Ob/Gyn, Marne Medical Center, PorUand, ME OBJECTIVE: To search for the associations w~th non-Doppler*flow placental cystic lesion and its influence to pregnancy outcomes. STUDY DESIGN: Data was prospectively collected on 171 cases of placental cystic lesions without Doppler flow ~dentified by second and eady third trimester ultrasound. Smoking, maternal dtsease dunng pregnancies, reasons for ultrasound, and pregnancy outcomes were compared with a control group of 229 pregnant women RESULTS: There were 43.3% women who smoked during pregnancies (741171) compared with 21.4% (491229) in control group (P < 0.01) Elevated maternal serum alpha-fetoprotein was also higher m study group (34/171 - 19.88%) than in control group (12/229 = 5 2%) (P < 0 01). Pregnancy complications occurred significantly higher in study group than in control group (89/171 - 52.05% in study group vs. 601229 = 26 20% in control group, P < 0.005). Individually oligohydrammos (10.53% vs. 3.93%), intrauterine growth restriction (23.98% vs 5.24%), and pregnancy induced hypertension (14 62% vs. 3.06%) showed statistical differences. CONCLUSION: Smoking during pregnancy and maternal serum alpha- fetoprotein may be associated with the presence of non-Doppler-flow placental cystic lesion m pregnancy which may be associated with higher prevalence of adverse pregnancy outcomes than normal appearing placentas

378 CERVICAL CHANGES DETECTED BY TRANS-VAGINAL SONOGRAPHY

ASSOCIATED WITH POOR PERINATAL OUTCOME.

R,~:~ynA.~ dA R~t, J.V, Fayx, J.C. Goldbergx, M.A. Josephx, H.L.

Symeckox, R.J. Stiller, G. Dunston-Boone, and J.E, Collinsx, Yale

University School of Medicine, Bridgeport Hospital, Bridgeport, CT

OBJECTIVE: To identify, using trans-vaginal sonogrsphy (tvs), specific

types of cervical change that could predict poor permatal outcome. STUDY DESIGN: Pstmnts referred for tvs due to risk for incompetent

cerw< or cerwcel change from January 1, 1989 - December 31, 1994

were identified. 277 patients (out of 27,440 scans) matched a keyword

search; once charts were excluded due to lack of delivery or delivery

outcome, e study sample of 196 patients resulted. Cervical length,

degree of d~mphng, presence of funneling/basking and dynamic changes

were matched to perinatal risk factors and dehvery outcomes.

Connnuous dependent variables were analyzed by Spearman Correlation

or Wdcoxon Rank Sum analysis and categorical variables were assessed by x= analysis. Odds Ratios (OR), Confidence Intervals (CI), and

predictive values were calculated for dependent variables.

RESULTS: Gestatlonal age was correlated with length of intact cervix

I P =.01). Mean age of gestation decreased with every umt increase ~n dimple size (p =.0023). When modeled together, size of dimple was a more significant predictor of gestational age. When compared to

patmnte w~thout cerwcal change, newborn weight was significantly

lower by 870 gms ff funneling/basking was present, by 480 gms if any dimple was present, by 760 gms if dynamic change was present, and

by 630 gms if less than 20ms of intact cervix was present; all were assocmted with praters dehvery (p<.05). Patients with dynamic

changes (and no dlmplel had OR 4,67 (CI 1.6-14.2) for preterm birth,

and OR 5.22 (CI 1.4-20.1) for premature rupture of membranes.

Dynamic vs. dimple patients revealed dynamic changes having 80%

PPV, 91% specificSy, and OR 10.9 ICI 1.5-120.01 for low birth weight.

CONCLUSIONS: Quantitative and quahtative cervical changes can be

detected on tvs that predict poor perinatal outcome.

380 PLACENTAL LAKES FOUND ON ROUTINE ULTRASOUND DO NOT PREDICT OBSTETRIC OUTCOME. U. Magriplesx, W Shatterx, J.A. Copel Dept OB/GYN, Yale Univ., New Haven, CT. OBJECTIVE: Placental lakes have been reported to be associated with poor obstetric outcome. We studied whether placental lakes found on routine ultrasound were cbnically significant when maternal risk factors are taken into account STUDY DESIGN: We prospectively studmd 350 consecutive women attending the prenatal clinic at Yale-New Haven Hospaal undergoing routine ultrasound for anatomic survey or dating over eighteen months Women with vaginal bleeding, placenta previa and those who subsequently underwent elective termination were excluded Placental sonolucencies measuring I cm or greater in their largest dxmension were noted. Maternal demographics and obstetric outcomes were compared between the group w~th lakes (L) and the group w~th no lakes (NL) using chi-square and Student t-tests. RESULTS: Ftfty-seven women (16 3%) had placental lakes ~denttfied. There was no d~fference ~n maternal age, race, gravity, parity, history of preterm delivery, prevalence of smoking or drug abuse or medical risk factors for intrauterine growth retardation between the groups In a subset of patmnts (n=159) who had maternal serum alpha-fetoprotein (MSAFP) screening, there was no difference in the incidence of MSAFP elevations between the groups (L, 2 2% vs NL, 2.7%). Obstetric outcomes are presented below.

LAKES NO LAKES GA at delivery 38.9 +__ 0.29 38.9 + 0.14 NS

Birthweight (gm) 3245.9 _+ 97.6 3229.9 _+_ 34.4 NS

SGA 2 t % 5.6% NS Fetal Distress 8.2% 7 5% NS Placental Physiology 14.0 12.6% NS Cesarean SecUon 8 2% l 1.1% NS MSAFP (mean MoMs) 0 88 0.96 NS CONCLUSIONS: Placental lakes found on routine ultrasound, in the absence of maternal complications, are not associated with an increased incidence of obstetric complications.

\%lume /74, Number 1, Part 2 SPO Abstracts 415

Am [ Obstet G)necol

381 RELATIONSHIP OF FETAL URINE PRODUCTION RATE TO

BIOPHYSICAL SCORE IN PATIENTS WITH PREMATURE

RUPTURE OF MEMBRANES. T Ngnyen’, T Myles, A Bieniarz, W

Meyer, D Gauthier, B Nuwayhid~’. Dept. of OB/GYN, Umv. of

Illinois, Chicago, IL

OBJECTIVE: To evaluate possible relationship between fetal urine

productmn rate (FUPR) and biophysical profile (BPP) in patients with

premature rupture of membranes (PROM).

STUDY DESIGN: FUPR was determined on 26 PROM patients

between 24 and 39 weeks of gestatlonal age. Fetal bladder measurements

were performed every 3-5 minutes for up to 90 minutes. Bladder

volumes were calculated using the method described by Moore. FUPR

was corrected for gestational age prior to analysis BPP scores were

obtained using Mznning’s criteria. Fetal unne production rate in the

group of patients with BPP score of ~> 8 was compared to that of the

group with BPP < 8 Statistical analysis was done using t-test

Significant level was set at p < .05.

RESULTS: A total of 42 paired observatmns of FUPR and BPP were

made and used for analysis. The results are noted in the table below.

BPP < 8 BPP~> 8 N 18 24 Average FUPR (ml/hr) 16.7 33.6 Standard deviatmn 12 0 24.0

A statist~cally significant difference in FUPR between the two groups

was noted ( p < .009).

CONCLUSION: FUPR in patients with PROM appears to be

significantly decreased if the BPP score is less than 8. This

observation seems to confirm the established association between

BPP, and perfusion of non-vital organs (fetal kidneys). ChnJcal

usefulness of FUPR assessment in supplementing BPP for

determining fetal well-being remains to be determined.

383 ACCURACY OF AMNIOTIC FLUID VOLUME EVALUATION IN

DIAMNIOTIC TWIN PREGNANCIES: THE EFFECT OF

SONOGRAPHIC INTERPRETIVE TECHNIQUE AND OPERATOR

EXPERIENCE. E.F. Magann, KG Perry, Jr., M,B. McClurkanx, PJ.

Anfangerx, N.S. Whitworthx, J C. Mornson Dept. Ob/Gyn, Univ. of

Mississippi, Jackson, MS.

OBJECTIVE: To determine the accuracy of amniotic fluid volume (AFV)

assessment in diamniotic twin pregnancms by sonographic estimate stratified by

level of operator experience and currently used ultrasound techniques.

STUDY DESIGN: In this prospective study, the AFV of each twin of 16 sets

of diamniotic twins was subjectwely estimated as oligohydrammos, normal, or

hydramnios using ultrasound by a second-year obstetric resident, nurse

sonographer, maternal-fetal medicine (MFM) fellow, and MFM staffphys~cian

AFV of individual sacs was evaluated using the largest vertical pocket (LVP)

technique of Chamberlain, amniott¢ fluid index (AFI) by Phelan, and the two-

dmmeter (2-D) pocket technique of Magann. The actual AFV was determined

using amniocentesis and dye-dilution techniques

RESULTS: Based on normal AFV changes throughout pregnancy for singletons

(Brace and Wolfe), 8 amniotic sacs were oligohydramnie, 22 normal, and 2

hydramnic. The accuracy of subjective sonograph~c assessment (visual

interpretation w~thout ultrasonic measurement) to identify normal AFV by the

four sonogrmphers ranged from 91 to 100% (p = 0.506), while precisiot~ of

actual ultrasound measurements ranged from 82% by the 2-D pocket to 100%

by the AFI and LVP (p = 0.014). Ohgohydramnios was poorly idenhfied

subjectively, 125 to 37.5% (p = 0570) by the four sonographers, but

confirmed significantly more often by the 2-D pocket 75 %, (p = 0.009) than by

the AFI and LVP (both 12.5%). None of the sonographers or any of the three

ultrasonic measurements identified the two sacs with hydranmms.

CONCLUSIONS: The evaluation of normal indlviduaI ammot~e sac "�olumes

was simdar (82-100 %) regardless of operator experience but significantly less

accurate using the 2-D pocket versus the LVP and AFI The 2-D pocket

technique was sigmficantly more accurate in identifying ohgohydramnios than

the AFt and LVP Hydramnios was not identified by any techniques in these

twin pregnanmes.

382 MATERNAL AND FETAL COMPLICATIONS ASSOCIATED WITH AMNIOTIC SHEETS. N, Lazebmk. L.M. ~11, A. Many, J G. Martin Dept Ob/Gyn, MageeoWomens Hosp=tal, Umv.ef Pitt., Pittsborgh, PA OBJECTIVE: To evaluate the relationsNp between the onentatmn of ammohc sheets to the placenta and fetal and maternal comphcat~ons. METHODS: An ammotlc sheet was visualized m 59 patients The sonograpNc criteria for an amniotic sheet included. 1) the presence of a reflecWe membrane w~th measurable thickness that traversed the ammottc cavity, and 2) the fetus was not attached to the membrane. Amniohc sheets were defined according to thetr onentatton to the placenta, as parallel, obhque or vertical RESULTS: Amnlotic sheets were parallel, obhque or vertical to the placenta in 24, 13, and 22 cases, respectively. Verttcal ammottc sheets were assomated with a significantly higher incidence of prior uterine trauma or infection (p < 0 05) and breech

presentation at delivery (p < 0.001) than were oblique or parallel amniotic sheets. 24 patients (42%) had a Nstory of first or second trimester vaginal bleeding. The =nctdence of vaginal bleeding was not s~gmficantly d~fferent tn the three sub groups. CONCLUSIONS: Vertical ammottc sheets, like uterine synechtae, are associated w~th a Nstory of utenne manipulation or tnfectton, and abnormal presentation at term. There is a higher incidence of first and second trimester vaginal Needing with amniot~c sheets, regardless of their orientahon

384 AMNIOTIC FLUID INDEX: cOMPARISON OF CURVILINEAR AND LINEAR TRANSDUCERS. FT Callan~, RK ~aelde, BM Karpel, BA Meyer Dept Ob/GYN and Antepartum Testing Umt, SUNY @ Stony Brook, Stony Brook, NY.

OBJECTIVE: To confirm that araniotic fluid index (AF1) measurements obtained with a eurvilinear array correlate with results obtained with a linear array.

METHODS: A prospective randomized double blinded study approved by the IRE of 137 patients between 28 and 42 weeks of

gestation with singleton pregnancies was performed. AFFs were measured wltit both a 3.5 MHz linear array and a 5 MHz curvilinear array The order of the AFI measurement was determined by coin toss. The sonographers involved were blinded to the In:st AFI and both AFI measurements were completed within 15 minutes. Data were analyzed by ANOVA (*p< 0 05). RESULTS: The correlation between eurvilinear and linear array was good with an ra 0.63 and a slope of 0.97. The intercept of the

regression line however was 2 85, demons’aating that the cur~line,x array overestimated the amniotic fluid index obtained by linear array by

approximately 2.9 eros* Subgroup analysis of patients with AFI’s < 10 (r~. 0.36, slope’ 0.99, intercept 2.5) and gestational ages of 37 - 42 weeks (r2:0 75, slope: 0 92, intercept: 2.6) confirraed a systematic overestimation of ammotie fluid index by the eurvilinear array*.

CONCLUSIONS: 1) Excellent correlation exists between AFI determinations obtained by either a curvilinear array or linear axray. 2)

AFI determinations by curvilmear array consistently overestimate the amniotic fired index despite adherence to the strictly defmed techniques for the determination of armliotic flui4 mdex. 3) We

speculate the angular configuration of the ultrasound crystals and the curvilmear array results in an overestimation of the ammotic fluid by

simple geometry

416 SPO Abstracts Janrla~3, 1996 Am J Obstet Gynecol

385 MATERNAL HYDRATION AND ITS EFFECT ON THE AMNIOTIC FLUID INDEX. J.J_ Kerr*. AF Borg~da, H Hardardott~r, S Calhoun*, J Galetta*, JFX Egan St Franc=s Hosp=tal, Hartford, CT and Umv of Connechcut Health Center, Farmmgton, CT. OBJECTIVE:To determine ff maternal hydrahon increases the amniofic fluid index (AFI) ~n women with comphcated pregnancies STUDY DESIGN: Th~s prospechve, randomized chmcal trial enrolled pregnant women requmng antenatal testing from 12/94 to 4/95. Ent~ criteria were’ s~ngleton; >- 24 weeks, intact membranes; and absence of’ preterm labor, fetal d=stress, polyhydrammos, or maternal cardiac d~seaee. The hydrat=on group drank one hter of water over one hour. Control subjects drank 150 ml of water. Each subject had a unne specific grawty and AFI performed before and after treatment The AFI was measured by a single bhnded examiner, using Phelan’s techmque. Delta AFI (A AFI) was the difference between the post and pretreatment AFI Data was analyzed by the Student t test and ANOVA. A p-value of < 05 was s=gmficant RESULTS: F~fty pahents were randomized w=th a mean gestat=onal age of 35 weeks (range 26-41weeks). Ind~cat=ons for testing were similar between groups and =ncluded IUGR, HTN, postdates, bleeding, diabetes, and ohgohydramn~os. The hydration group (n=26) showed a s~gn=ficant decrease in the post-treatment specific grawty compared to the controls (n=24) (1 006 vs. 1.014, p = 0002) Hydration was net assoc=ated w=th a s~gn=ficant change ~n the AFI between the hydrat=on and control groups (A AFI, 0.57 vs. 0.92 cm, NS) In pat=ents w=th an AFI < 8 cm, response m the hydrahon group was double that of the controls (A AFI, 1.8 vs 0 87 cm, NS). There was an =ncrease ~n the AFI m hydrated pahents w~th a pre-treatment AFI < 8 cm when compared to those with a pre-treatment AFI > 8 cm (1.83 vs. - 0.61 cm, p = .0087.) Two of three hydrated patients with a pre-treatment AFI < 5 cm increased their AFI by > 60% CONCLUSION: Maternal hydrahon was not associated w=th our s=gmficant change =n the AFI between the hydrat=on and control groups. Hydration ~ncreases the AFI ~n patients with an AFI < 8 cm ~n a h~gh nsk populafion although =t is not statlshcally s~gn~ficant

387 CLINICAL SIGNIFICANCE OF PRENATAL

SONOGRAPHIC INTESTINAL DILATATION

IN FETUSES WITH GASTROSCHISIS

Q Alsulvman. H Montoro,x J Ouzouman, L Barton,x G Songster,

B Kovacs. University of Southern California, Los Angeles, CA

OBJECTIVE: To evaluate the clinical significance of intestinal

dilatation detected by prenatal sonographlc examination in fetuses

with gastroschisis

STUDY DESIGN: A retrospective review was performed of all

patients managed at our medical center with prenatal diagnosis of

gastroschisis Patients were divided into two groups based on the

extent of the sonographlcally measured fetal intestinal dilatation

Postnatal data were obtained on an on going basis and augmented by

chart review.

RI~SULTS: Twenty one patients with prenatally diagnosed

gastroschlslS were identified. E~ght fetuses had maximal intestinal

dll.atation greater than 17 mm (mean 37.2, SD 17 2, range 17-61 mm),

and thirteen fetuses had maximal intestinal dilatation less than 17

mm (mean 7.6, SD 3 7, range 4 16 ram) There were no statistically

significant differences in the mean gestational age at delivery (34 3

vs. 34.2 wks), birth weight (2163 vs. 2081 gms), days to full oral

feeding (46 vs 53 ), length of initml hospital stay (51 vs. 56 days) or

method and ttmxng of gastroschlsls repair between the two groups

Two newborns underwent bowel resection for intestinal atresla

Prenatal sonographic examination of these fetuses did not show

significant intestinal dilatation

CONCLUSION: In this study fetuses with significant intestinal

dilatation detected on prenatal sonography were smadar to those

without significant dilatation with respect to neonatal morbidity and

hospital course. These results suggest that prenatal evidence of

intestinal dilatation may not be an indication for delivery in the

absence of fetal lung maturity or other evidence of fetal compromise

386 ULTRASOLND EVALUATION OF AMNIOTIC tLUID INDEX IN TWIN PRFX2qANCY. ~,x B Mercer, R Lewis, SA Friedman, V Mosby, BM Sibai. Dept. of Ob/Gyn, Univ. of Tennessee, Memphis OBJECTIVE: To establish normal values of amniotic fluid index (AFI) in twin pregnancy, and to quantify the normal inter-sac difference. STUDY DESIGN: Seventy-six patients with uncomplicated twin pregnancy had serial measurement of AFI in both sacs, for a total of 404 measurements (202 for each twin). Each gestational sac was separately divided into four quadrants. AFI was measured as the sum of the maximum vertical fluid pocket in each of the 4 quadrants. The measurements were performe¢{ between 23 and 37 weeks’ gestation. For eachgestational age, the mean, 5th and 95th percentile AFI were plotted for twin A and B. The inter-sac difference in AFI wa~ calculated at each visit. The mean and standard deviation of the absolute inter-sac difference at each week were determined. RESULTS: The mean, 5th, and 95th~ercenfile AFI values in twins A and B were nearly identical, as depicteoin the nornogram. The mean + SD absolute inter-sac difference was 2,7+2.4 cm. A trend toward decreasing mean absolute inter-sac difference was seen with advancing gestafional age (-1 cm from 23 to 37 weeks).

24- o Twin A

~12-

~

4-

0 ~ ~ ~

22 24 26 28 30 32 34 36 38

Weeks’ Gestation

CONCLUSION: We provide a nomogram for the evaluation of amniotic fluid, in each individual fetus, in twin gestations.

388 SUPERIOR MESENTERIC ARTERY DOPPLER VELOCIMETRY AND

ULTRASOUND ASSESSMENT OF FETAL BOWEL IN GASTROSCHISIS:

A PROSPECTIVE LONGITUDINAL STUDY A Abuhama~, R Cortmax, G Manx, D Croitorux, A Evans Depts Ob/Gyn at Eastern Virginia Medical School,

Norfolk VA & Yale Medical School, New Haven, CT OB3ECTIVE: To determine whether Doppler velocanetry of the superior

mesentenc artery (SMA), its mesentenc branches (MB), mammal bowel diameter

(BD) and ma.,amal bowel wall thickness (BT) can predict postnatal outcome m fetuses with gastroschisls

STUDY DESIGN: Normal reference range for the SMA pulsahhty index (PI)

was determined by studytng 161 normal fetuses Over a 24 month period,

SMA PI, MB S/D ratio, BD, and BT were prospectively and longitudinally obtained ~om 17 fetuses with gastroschlsls Poor neonatal outcome(PNO) v, as

defined by bowel resection, staged repair, and/or a hospital stay > 50 days

RESULTS: 16 pregnancies were available for analysis (1 fetal demise) 6/16 (37

5%) neonates had PNO Measurements were obtained in 15/16 fetuses within 1 week of dehvery, SMA PI was abnormal (below 95 % CI) in 8/16 with no

difference between the groups (p 0 99) A trend towards a h~gher MB S/D ratio

was noted m the PNO group (p 0 11) Mean BT and BD were stgmficantly greater m the PNO group (p-0 06, p-0 03) A cutoffofBT > 3 5 mm and BD >

20 mm had the best sensitivity and specificity for PNO A cutoffof BD > 10 mm

at 28 - 32 weeks was h~ghly predictwe of neonatal outcome Longitudinal data analysis showed a trend of increasing BD with advancing gestation (p < 0 0001 )

When compared to the good outcome group, a greater rate of increase was noted

m the PNO group (p-0 01 ) There wc, s no difference in SMA PI with advancing

gestation between the 2 groups (p 0 6)

-Variable ...................... Sensitivity ........... Specificity ................ p ....

BT > 3 5 mm 4/6 (67%) 8/10 (80%) 0 12

BD > 20 mm 4/6 (67%) 5/10 (50%) 0 63

BD > 10 mm (28-32wks) 5/6 (83%) 7/8 (88%) 0 02

CONCLUSIONS: To our knowledge, this represents the only prospective

longitudinal study of fetal gastroschis~s SMA Doppler veloclmetry is not

pred~ctive ofPNO A BD cutoff > 10 mm between 28 and 32 weeks had the

h~ghest senmtwlty and specificity for predicting poor neonatal outcome This

lnformatlotl may be useful in counsel~ag and management of pregnanctes

conaphcated by fetal gastrosch~sls

Vohnlm 174, Numbe~ l, Pau 2 SPO Abstracts 417 Am.J ()bstet Gynecol

389 RENAL DUPLICATION ANOMALIES IN THE FETUS: CLUES FOR PRENATAL DIAGNOSIS. A AbuhaInad, C Horton×, S Horton×, A Evans Dept Ob / Gyn & Urology, Eastern Virginia Medical School, Norfolk, VA OBJECTIVE Duplex kldne~ (DK) ~s one of the most co,ninon major congemtal abnormahtms of the urnlaD tract The antenatal d~agnosls of

DK and ats associated ureterocele is ~nfrequent The objective of this study was to determine the ullrasonographi¢ (U/S) characteristics of DK ~n the fetus STUDY DESIGN U/S charactenshcs were prospectavely collected on

all fetuses wath renal duphcation anomalms d~agnosed at our ultrasound uint over a 24 month period U/S charactenstacs ~ncluded renal sagattal length, and the presence or absence of the followang dilated ~psdateral ureter, "cyst lake" stractnre an the upper renal pole, and a ureterocele In

the urinary bladder Postnatal genitounnary evaluatmn was obtained on all neonates RESULTS DK was &agnosed prenatally m 8 fetuses over the study

permd, and confirmed postnatally an all neonates The sagattal length of the DK was >~ 95 percentile for gestatmnal age In 8/8 fetttses A "cyst lake" structure ~n the upper pole of the DK and a ureterocele an the unnar~ bladder was present an 7/8 fetuses An ~psilateml dilated ureter was seen an 6/8 fetuses CONCLUSIONS Fetal renal findings of sag~ttal lengthF 95 percentde for gestatmnal age, upper pole "cyst hke" appearance of the kadney, a ureterncele an the urinary.’ bladder and ipstlateral ureteral &iatatmn are U/S characteristics ofDK m the fetus To our ktmwledge, this represents the first report of U/S characterastacs of fetal DK Increased faunhanty of the prenatal sonographer wath this enhty will allow for ats antenatal recogmtmn thus enabling more accurate prenatal counsehng and early postnatal evaluatmn and treatment

391 Is Prenatal Mild Renal Collecting System Dilatation of Pediatric Urological Significance? W H Persutte×, Martin Koyle~, Roger Lenke, Jam Klas~, Cheryl Chainberlain~, John Hobblns, Depa~s of Obstetrics and Gynecology. Surgeay/Urology, UniversW of Colorado Health Sciences Center and Pedaatnc Urology, The Chfldreffs Hospital, Denvar, Colorado, and St Vincent Hospatal, In&anapolls, Indiana OBJECTIVE Isolated bilateral or umlateral "mdd renal collectlng system &latation" (MRCSD) is a commonprenatal sonographic findmg While an assocaatmn between MRCSD and fetal aneuplmdy has been established, m the absence of concomitant anomaly MRCSD is thought to be benign and of no chmcal conseqnence To test this, we investigated the relationship between MRCSD and (1) progression to hydronephrosls,

(2) postnatal veslcoureteral reflux (VUR); and (3) postnatal surgery STUDY DESIGN Between lanuary I, 1992 and January 1, 1995, we performed 453 prenatal sonographic examinations on 306 patients with MRCSD and no other anmnahes MRCSD was defined as a pelvocalyceal fluid-filled space with the smallest of two transverse perpendicular sonographac measurements of ;,4ram and <10mm Hydronephrosas was defined similarly, but with a measurement of a=10mm Postnatal urologic assessment was obtained in 251 (83%) of these patients using ultrasound

and VCUG routinely, and with DPTA, MAG3, and IVP, as necessary. RESULTS One-hundred twelve patients (45%) had two or more prenatal examinations Thirty (27%) of these cases progressed to a meas~trement >10mm (hydronephrosis) Only 6 (5%) of the cases followed prenatally dnnmlshed to a measurement of <4ram Postnatal evalnatlon of the 241 infants showed slgmficant VIJR (diagnosed by VCUG) in 14 3% (36/25l) of cases. One case reqmred surgery An addmonal three pataents required urologic surgery to correct other pathologic conditions CONCLUSIONS Whereas the prenatal occurrence ofMRCSD amproved

in only a small nnmber of cases, we found that progression occurred in 27% Postnatal evaluation lbund sagnlficant renal pathology an at least 15 5% of cases All patients wath significant VUR are currently on antibiotic prophylaxis Finally, postnatal surgery was necessary in only a small number of these cases

390 What Is the Best Method to Assess Mild Renal Collecting System

Ddatation? W H Persutte× R R Lenkc, Departments of Obstetrics and

Gynecology, Umversaty of Co[orado Health Scmnces Cegter, Denvcr,

Colorado, and St Vmcant Hospital, Indianapolis, Indiana

OBJECTIVES "Mtld renal collecttng system ddatatmn ~CSD)" in

the Ictus has recently been associated w~th veslcoureteral refinx (VUR)

Unlortunately, the optmml method for assessing MRCSD has yet to be

determined The purpose of tins report ~s to determane winch prenatal

sonographIc method as best an predictang neonatal VUR, including (I)

the smallest of two transverse-perpendicular measuremeilts of the

pelvocalyceal fluid [SM], (2) the mean of two perpendicular-transverse

measurements [MN], and (3) a volumetric assessment [VOL]

STUDY DESIGN We obtained three nmasurements of the pelvocalyceai

fluid collection an fetuses wflh MRCSD, including (a) the smallest

transverse diameter [sm], (b) the largest transverse diameter [lg], and the

longest saglttal diameter [sag] TIae volume of the pelvocalyceal fluid

was calculated using ((sin x lg x sag)x 5). Each inthnt was evaluated

using veslcoureterography (VCUG) lbllowmg delivery. We

retrospectively carnpared the sensR~vlty and specificity of the three

methods described above for VIJR We sequentially compared the

predactabthty of the SM measurement at lmm increments from a4-

~10mm and the MN from a4-M6mm Volumetric measuremants were

obtained m 250ram ~ncrements from 500-2500mm We attempted to

&scern the threshold with the greatest sensitivity and high specificity

RESULTS Between July 1, 1993 and Jamlary 1, 1995, oiae hnndred-

thirty six fetuses were adentified to have MRCSD and were ancluded an

th~s study Nineteen of these cases were found to have significant

postnatal VUR We fmmd the optimal sensitivities and speclficitles to be

83% and 61%, respectavely, using the SM method at a threshold of

a8mm, 83% and 69%, respechvely, using the MN method at al0mm,

and, 88% and 77% respectively, with the VOL method, at a1250ram

CONCLUSIONS These data suggest that the volumetric method is

supermr to the single ~r mean diameter methods in predicting VUR This

method was 88% sensthve and 77% specific at a threshold of 1250rmn

392 DEVELOPMENT OF FETAL GYRI, SULCI AND FISSURES: AN

ULTRASONOGRAPHIC STUDY. A.Monteagudo, I.E. Timor-Tritsch

Dept. Ob/Gyn, Columbia University, NY NY.

OBJECTIVE: To determine the feasibifity of imaging specific sulci, gyri

and fissures using transvaginal sonography and to correlate their

development with gestafional age

STUDY DESIGN: Two hundred and sixty one fetal brain scans were

selected from 346 scans reviewed retrospectively. Scans were selected if

any of the following structures were seen: in the coronal plane the lateral

and callosal sulcus and the cingulate gyrus; in the median plane the parieto-

occipital sulcus and calcarine fissure, and the eingulate gyrus and suleus;

in an obfique section the lateral sulcus. The gestatmnal age when the

fissures, sulci, and gyrl were fu’st imaged was recorded and subsequently

compared with similar anatomic studies from the fiterature.

RESULTS: All of the targeted structures were unaged. The gestataonal age

at which the structures were first imaged is: the eallosal sulcus from 14

weeks, the lateral sulcus from 18 weeks, the parleto-occipital sulcus and

ealcarine fissures from 18 weeks and the cingulate gyrus from 26 weeks.

CONCLUSIONS: The developmetal maturation of the fetal bram follows

a predictable time-table, which can be followed with sonography.

Neurupathologist have found that the developmental maturation of the brain

is a more refiable estimate of age than biometry. This is specifically found

to be true between 22 and 30 postmenstrual weeks. The s_onographic

recognition of the fissures, gyri and sulci lagged behind those observed by

anatomlo studies. The greatest discrepancy was the first appearance of the

�regulate gyrus which in the anatomic studies could be seen by 18 weeks

and in our study only aRer 26 weeks. The one exception was the eallosal

sulcus in which the sonographic and anatomic studies it was fu:st seen at 14

weeks. In eonelusion, the recognition specific structures of the cortical

map is possible. The role of sonography to image the developing cortical

surface may be use for both estimation of gestational age and possible

detection of pathology affecting the central nercoua system.


393 ILIAC WING ANGLE AS A MARKER FOR TRISOMY 21 IN SECOND TRIMESTER FETUSES. B0rk MD,~ Egan JFX, Cus~ck

W, Borg~da AF, Hardardott~r H, Rod~s JF, Campbell WA. Univ of CT Hea~th Center, Farmmgton, CT and Oakwood Hospital,~ Dearborn M~ OBJECTIVE: The pelvic lilac wing angle is widened m newborns w~th tnsomy 21. Our oblechve was to evaluate the efficacy of fetal lilac w~ng angle to detect tnsomy 21 =n the second trimester of pregnancy. STUDY DESIGN’ Utilizing an axial view of the fetal pelws, ultrasonograph=c measurement of the angle between the right and left ~hac w~ngs 0hac w~ng angle) was prospechvely obtained at time of m~dtnmester ultrasound or genetic ammocentes~s. Trisemy 21 was diagnosed by karyotype results or newborn exam=nation for phenotypm ewdence of tnsomy 21. Sens~hwty, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated using mulhple cut-off points A receiver operabng characterishc (ROC) curve was used to ~denhfy the opt=mum ihac wing angle Descnphve statistics and Student t-test were utd~zed for analyses w~th p < .05 s~gn~hcant. RESULTS’ A total of 377 fetuses were examined The average gestabcnal age was 18.8 weeks (range 13-32) Karyotypes were available in 128 fetuses. The overall prevalence of trisomy 21 was 11/377 (2.9%). The ~l~ac wing angle measurements (mean + SD) in the

normal fetuses was 68.2°+_ 15.4° versus 98.5° +_ 11.3° in fetuses w~th tnsomy 21 {p<0.001). Using an ROC denved absolute cut-off of >90°, an abnormal ~hac wing angle had a sensitivity of 90.9% (10/11), spec=hcally 94.5%, (346/366), NPV 99 7% (346/347) and a PPV of 33 3% (10/30) to detect trisomy 21. CONCLUSION: Compared to normal fetuses, fetuses w~th tnsomy 21 have a slgmhcantly increased ihac wing angle. Using an ROC derived absolute cut-off of > 90° we could detect 90 9% of fetuses with trisomy 21 wRh a PPV of 33%. Th~s study suggests that d=ac w~ng angle =s a useful marker =n antenatal screening for tnsomy 21.

395 THE PROSPECTIVE EVALUATION OF FETAL CARDIAC ANATOMY: ADDITION OF THE PULMONARY ARTERY VIEW DOES NOT INCREASE DETECTION RATE. Kirk JS 1, Comstock CH ~, Smith RS ~, Lee W ~, Riggs TW~, Weinhouse E xz. , Division of Fetal Imaging, Dept. OB-GYN and 2 Division of Pediatric Cardiology, Dept. Pediatrics, William Beaumont Hospital, Royal Oak, MI. OBJECTIVE: To determine if the addition of the pulmonary artery view to the four-chamber and aortic root views increased prenatal detection of abnormal fetal hearts. STUDY DESIGN. Beginning in 5/93, we routinely attempted to obtain in every patient a view of the pulmonary artery in addition to the four-chamber and aortic root views. Of all consecutive patients at 14 weeks or more gestation scanned in our unit 5/93 to 7/95, the pulmonary artery could be evaluated in 3313 fetuses. When any view was abnormal, a fetal echocardiogram was performed to define the abnormality. Only singletons who delivered at our hospital were included. Pediatric echocardiograms and autopsies were reviewed to determine all abnormal hea~ls confirmed or discovered after delivery. Sensitivity was compared to our 2/90 to 6/92 study when our routine procedure was to attempt four- chamber and aortic root views (Obstet. Gynecol. 94;427-431). RESULTS: In the current study, there were 40 abnormal fetal hearts; twenty-six were detected prenatally. We found no difference (chi square, p = 0.73) in sensitivity (26/40 = 65%) for ’93 - ’95 when compared to our ’90 - ’92 study (36151 = 71%). CONCLUSIONS: While evaluation of the pulmonary artery is necessary in defining cardiac abnormalities, no increase in the detection rate was obtained by the routine addition of the pulmonary artery view. The majority of heart defects are already detected by screening with the four-chamber and aortic root views.

394 SONOGRAPHICALLY MEASURED FETAL CARDIAC AXIS AS A MARKER FOR CONGENITAL HEART DISEASE. Bork MD, Egan JFX, Borgida AF, Hardardott=r H, Fabbn EL*, Feeney LD*, Smeltzer JS, Campbell WA. Umv, of Connectmut Health Center, Farmlngton, CT and Oakwood Hospital, Dearborn, MI OBJECTIVE: To prospectively evaluate the efficacy of fetal cardiac axis as a marker for congemtal heart disease (CHD). STUDY DESIGN. Patients referred for a targeted ultrasound between 14-40 weeks age of gestation from 9/94 to 1/95 were prospectively studied. Patients specifically referred for fetal echocardlography (FE) were excluded. On screen measurement of fetal cardiac axis was determined using a prewously described method FE was then performed on all fetuses Data recorded included’ age of gestation, indication for exam, fetal b~ometry, fetal cardiac axis, type of CHD 0f present) and neonatal outcome. Previously estabhshed norms for fetal cardiac axis were utilized, with an abnormal axis defined as the mean + 2SD’ (43° ±16). Abnormal antenatal hndlngs were confirmed by neonatal echocardiography or postnatal autopsy The data were analyzed using descnphve statistics and Ch~ square with p<.05 s~gmficant. RESULTS. A total of 500 fetuses were evaluated dunng the study period. The mean (+SD) gestat~onal age was 22 8 (_+5.5) weeks. Antenatally, 11/600 (2 2%) of the fetuses had CHD detected by FE. These lesions included 8 ventncular septal defects (VSD), 2 complex cardiac lesions and one hypoplastlc left heart (HLH). There were lesions associated with an abnormal cardiac axis’ 3 VSDs; 1 HLH and 2 complex cardiac les~ons. All five tes~ons which were m~ssed by the axis were VSDs. The sensitivity of an abnormal fetal cardiac axis to idenhfy fetuses at risk for CHD was 54.4% (6/11), spec~hcity 95 1% (485/489), PPV 20% (6/30) and NPV 98 9% (465/470). CONCLUSION: Fetal cardiac ax~s measurement has a sens~t~wty of 54.5% to ~dentify fetuses with underlying CHD Our study suggests that cardiac axis determ=natmn may be a useful screening tool for CHD.

396 MATER.NAL DIABETES MELLITUS: IS FETAL ECHOCARDIOGRAPHY NECESSARY? RS Smith, CH Comstcok, RP Lorenz, JS Kirk, W Leo. William Beaumont Hospital, Divisions of Fetsl Imaging and Maternal-Fetal Medicine, Department of Obstetrics and Gynecolog3’, Royal Oak, Michigan OBJECTIVE: To determine what sonagrephic views of the heart detect the majority of ear&at defects in fetuses of insulin-requiring diabetic mothers. STUDY DESIGN: Fetal �~hocardiogrephy results from matsmal insulin-requiring diabetic patients who delivered at our institution were reviewed from 2/90 to 2/95. Components of the exam were analyzed for the detection of fetsl congenital heart defects. The exam consisted of evaluation of the 4 ¢hamber (chb) view, outflow tracts, aortic and ductsl arches, pulmonary veins, measurements of all chambers and curlew tracts at tbe level of the valves and Doppler velcoim¢l~% Sensitivity (sens), specificity (spec), positive predictive value (ppv) and negative predictive value (npv) were calculated. Multiple gestations and patients with additional risk fagtors for congenital heart defects were excluded. RESULTS: Two hundred and twenty-five insulin-requiring diabetic patients received fetal ¢chcoardiagrams during the 5 year period; 161 had post delivery follow-up. The mean gcstational age was 21 weeks (range 18-35 wks). Ther= were 5 (3 1%) congenital heart defects, of which 4 were conotranoal: transposition ofth© gz~at a.,’~s (1 ~ aorficopulmonary window (1), trancus artefiosus (1), pulmona~ am~ia with a veatneular septal defect ( 1 ) and a ventrlcular septsl defect with mild aortic stcnoms (1). Four hearts were detected as abnormal antcantally by the 4 chb view or outt]ow tracts The fifth case was missed due to morbid maternal obesity. Whon the 4 ehb view and outflow tracts appeared normal, add=fional views did not detect a cardiac defact. There wers no false positives. Views Abnormal sans svc¢ ppv ~apv 4 chb 3 60°/$ 100% 100% 98.ll%

4 chb or aortic outflow tract 4 g0°/, 100% 100% 99%

4 chb or aortic or pulmonary outflow tz’acts 4 80% 100% 100% 99%

CONCLUSIONS: A four-~hamber view together with outflow tracts will d~tc~t most car&ac defects in fetuses of insulin-requiring diabetic mothers. Additional cardiac evaluation m this high-risk group may not be neccssmT if thes~ views appoac normal.

Volume 174, Number 1, Part 2 SPO Abstracts 419 Am ] Obstet

397 ACCURACY OF PRENATAL DIAGNOSIS OF CONGENITAL HEART DEFECTS.V Berahellax, L Pagottd~, M.Kaufmanx, J HuN& R Wapner Maternal-Fetal Medicine, Thomas Jefferson Medical College and Perinatal Cardiology, Pennsylvania Hospital, Phdadelphia, PA. OBJECTIVE. ~1) Evaluate the accuracy of the initial prenatal ullrasound for the dragnosls (preND) of congenital heart defects (CHD) (2) Compare diagnostic accuracy of maternal-fetal medicine/radiologists (group A) to that of pennatal/ped~atnc cardiologists (group B) METHODS: Retrospective rewew of records was performed for 135 pahents who presented from 1/91-1/95 w~th CHD conhrmed poslnatal~y The postnatal defects were categorized ~nto 8 physiologic subgroups according to whether an error ~n d~agnos~s would alter postnatal management Ch~-square analys~s was used to determine stat~stlcel significance of the sens~t~wty of correct prenatal d~agnos~s between the two groups

RESULTS: PoND PND PND PNDNoI PND Total Correct Incorrect Detected False Pc

Group A B A B A I B A B A B H~po L Heart 7 32 "~ 31 ’~’} 1 ~ 0 " "~’ 0 AVCanal 5 4 4 4 0 0 t 0 2 0 Ebstein 10 2 ~ 2 i’}~ii1 0 ~, 0 :’"0 0 VSD 6 11 4 7 I ~ 1 t 3 ~ 0 ASD 6 6 ~ 6 O ~ 0 4 o t 0 C0adation ~ 3 2 3 ~""~ 0 "~" 0 ~ 1

@he~ ~ 9 1 10

0 ~ ’ 1 ’t 0 A~ Dx ~ 86 ~ 80 2 ~ 4 7 1

]’he overall accuracy was 83%. Group B physicians correctly classified the abnormahhes ~n 93% (80/86 of cases wh~le group A physicians were correct in 65% (32/49) (p>0 001) Of the 7 undetected cases in group A, 6 did not have formal fetal echocard~ograrns CONCLUSION: The ~nvolvement of a pennatal cardiologist in the prenatal diagnos~s of CHD improves diagnostic accuracy

399 FETAL CARDIAC AXIS IN TWIN PREGNANCY: A MARKER FOR FUNCTIONAL AND STRUCTURAL CARDIAC COMPROMISE Fink N., Ash K, Desjardins C. Division of Pennatology, Un~vers=ty of Ottawa, Ontado, Canada. OBJECTIVES: To establish the value of felal cerd~ac ax~s (CA} in twin pregnancy and to determ=ne =ts value in detecting acquired funcbenal cerd=ac compromise as well as congenital structural defects. STUDY DESIGN: A nomegram for CA was established in 1 t 0 twins (55 pairs) between 17 and 39 weeks gestation. Four chamber views were prospectively obtained in all ~ns and CA determined by the measured angle between a line drawn from spine to antedor chest cell (transverse view) and line drawn through the ~nterventocular septum. Inter-observer variation was determined by two observers in 36 twins. CA was then determined ~n twins d~scordant for congenital (anatomic) cerdiac defects, functional cardiac and hemodynamic compromise (i.e, IUGR and twin-twin transfusion syndrome (TITS) and twin reversed arterial perfusion (TRAP sequence). RESULTS: Mean CA ~n normal twins was 43 _+ 8.80 (range 24-62°) and was ~ndependent of gestatJonal age. Inter-observer variation was 1 9o. By 22 weeks the 4..chamber view was consistently obtained at the same visit in both twins in 65% of ceses. Functional Cardiac Compromise: abnormal CAwes seen in 100% of recipient’twins with severe TTTS (perinatal mortahty rate 83%) and in 100% of TRAP sequence pregnancies (710 and 680) in the pump twins i.e. left axis deviation. Abnormal CA with dght axis deviation was seen in 16% (1/6) of dichodonic twins with inter-twin growth discordance of > 20% (22~). Congenital Cardiac Defects: Abnormal CA (mean + 2SD) was seen ~n 100% of clinicelly signiticent structural cerdiac defects (Tetralogy of Fallot and large PDA). CONCLUSIONS: f ) CA in normal twins (430 + 9o) does not d~ffer signilicently from that raporteq in s~ogletons 2) CA may be useful as a marker for hemodynamic and functional cerd~ac compromise as in TTTS and may assist in d~scdmination of IUGR of other ebology. 3) 4) CA should be included ~n the routine sonegraphic evaluation of t~an pregnancy at each visit.

398 SONOGRAPHIC CRITERIA FOR CERVICAL FUNNELING PREDICTIVE OF PRETERM DELIVERY (PTD), V.Berghella×, K Kuhlman, S Welner, L Telxerax, R Wapner Dlwslon of

MFM, Jefferson Medical College and Pennsylvama Hospital, Philadelphia, PA

sonographxc cervxcal ex~ predictive of PTD STUDY DESIGN: Pahents w~th cerwcal funnehng on translabial/transvagmal ultrasound were retrospechvely ~dentlfied from 1194-4/95 Exclusion criteria were women w~th cerclages a~eady m place, preterm labor, cervical change first detected by manual exam, or delivery for maternal-fetal lnd~cahons Sonograph~c cervical v~ables stud~ed were funnel length (A), functional cervxcal length (B), percent funnehng (MA+B), and funnel w~dth (C) RESULTS: Of 81 patients wath cerwcal funnehng, 34 met study criteria Funnehng was detected at a mean gestat~onal age of 21 weeks [range 16-28) R~sk factors for P~ were ~dent~fied m 22 paUen~ (65%) Of 27 patxents wxth an tmmed~ate manual ce~cal exam following the ultrasound. 20/27 (74%) had a closed~ong cervix ~e overall rate of PTD was 47% (16/34), 50% m the risk group and 42% m the no-risk group (NS) By unlv~late analys~s and recexver-operator ch~acterxst~c c~ves. funchonal length ~1 8 cm, funnehng ~50% and funnel w~dth 21 5 cm were s~gnlficantly co.elated w~th PTD rates of 71%, 79%, and 100%, respecnvely Of patients not meeting any of the above criteria, 14115 (93%) dehvered at term Of patients meeting one, two, or all three criteria, 4/6 (67%), 5/7(71%) , and 616 (1~%) dehvered preterm CONCLUSION: Sonographlc cerwcal criteria which are h~ghly co.elated w~th PTD ~e func~onal length ~1 8 cm, funnehng ~50%, and funnel w~dth 21 5 cm

400 A FIVE-YEAR EXPERIENCE WITH 111 ABNORMAL FETAL HEARTS: DETECTED VERSUS UNDETECTED. Kirk JS Comstock CH ~, Lee W ~, Smith RS ~Divis~on of Fetal Imaging, Dept. OB-GYN and =Division of Pediatric Cardiology, Dept. Pediatrics, William Beaumont Hospital, Royal Oak, MI. OBJECTIVE’ To determine if there is a difference bet’,veen the type of cardiac lesions which are recognized prenatally as abnormal and the ones which are not detected. STUDY DESIGN.: The study consisted of consecutive patients at 14 weeks or more gestation scanned in our unit 2/90 to 7/95. Only singletons who delivered at our hospital were included. Pediatric echocardiograms and autopsies were reviewed to determine all abnormal heads confirmed or discovered after delivery. Our initial routine procedure was to attempt four chamber and aortic root views in all patients. Beginning 5/93, we also attempted to obtain a pulmonary artery view. Patients with risk factors for cardiac abnormalities, such as family history, underwent a fetal echocardiogram. No patients were excluded from the study because of oligohydramnios, fetal position, or maternal obesity. RESULTS: There was a total of 111 abnormal hearts. Sixty-five per cent (N = 72) of these were correctly recognized as abnormal prenatally. Sensitivities for the most common cardiac lesions were: 89% of atrioventricular septal (endocardia[ cushion} defects, 79% tetralogy of Fallot, 63% transposition of the great arteries, 63% aortic coarctation, 38% isolated ventncular septal defects (VSDs). The most frequent lesions that went undetected were isolated VSDs (N = 16) and aortic coarctation (N = 6), Most undetected VSDs were small or moderate size. No isolated atrial septal defect of the secundum type (N = 3) was detected prenatally. CONCLUSIONS: Most major cardiac abnormalities were recognized as abnormal prenatally. The most difficult lesions to detect were septal defects and aortic coarctation.

420 SPO Abstracts .January 1996 Am J Obstet Gynecol

401 ADJUSTED POSITIVE PREDICTIVE VALUE OFROUTINE VIEWS FOR THE DETECTION OF CONGENITAL HEART DISEASE

(CHD) JM Lanouette* HM Wolfe, KL DeVnesx, KS Puder, J

Gurczynsk~~ Dept Ob/Gyn, Hutzel Hospital, Dept Ped Cardiology,

Children’s Hospital of MI, Wayne State University, Detroit, MI

OBJECTIVE: To assess prediction of CHD by wsualization of abnormal

cardiac rhythm (CR), axis (CA), four-chamber view (4CV) or outflow

tracts (OFT) on routine ultrasound

STUDY DESIGN: Data on 10,232 women undergoing a first obstetric

ultrasound at greater than 15 weeks’ gestation was searched for

abnormahnes noted in any of four aspects of the fetal cardiac exam

Neonatal records for ~hose patients delivering at our mst~tution were

reviewed for accuracy of prenatal dlagnos~s.

RESULTS: Sixty cases were ~dentified with the following abnormalities,

CR n= 44, CA n=8, 4CV n=14, OFT n=6. Positive predictive values

(PPV) were calculated for each v~ew. Rased on rates of adequate

wsuahzation (AV) of these structures on all ~nit~al scans during the

same t~me period, an adjusted predictive value (APPV-AVxPPV)

was calculated.

PPV(%) AV(%) APPV(%) Rhythm (5/44) 11 1 O0 11

Ax~s (1/8) 12 5 99 12

Four-Chamber (10/14) 71 79 56

Outflow Tracts (6/6) 100 55 55

CONCLUSION: The prewous~¥ reported accuracy of four and f~ve

chamber views for prenatal dmgnos~s of CHD must be adjusted for

the difficulty in consistently obtaining these v~ews on a single

routine ultrasound. Rhythm and ax~s, although less predictive of

CHD are more easdy assessed. In the managed Care setting where

ultrasound utdizatmn ~s hm~ted, their contrlbutlo~ to prenata~

d~agnos~s may play a larger role.

403 CARDIAC AND SOMATIC GROWTH IN INFANTS OF INSULIN

CONTROLLED DIABETICS. J GandhL X Zhang~, J Maldman, Dept

OB/GYN, The Long Island College Hospital, SUNY-HSCB, New York

OBJECTIVE: To assess the relationship between somatic and myocardml

growth m fetuses of msuhn stabd~zed dmbet~c pregnancms

STUDY DESIGN: Fetal Cardmc/Chest area, Interventncular Septal and

bdateral Ventrmular Wall Thmkness and Diastohc Dtmensmns (CA~TA, IVS,

LVWT, RVWT, LVDD, RVDD) were measured by B and M-mode

Echocardiography in 34 insulin stabilized diabetic pregnancies and 31

controls(C) Bipanetal dmmeter (BPD) was used as common denominator to

express size and growth Fetuses at 20 to 23, 24 to 28, 29 to 31, and 32 to 36

weeks (1,2,3,4) and at b~rth wmght (BWT) percentiles of < 25, 26 to 75, 76

to 90 and > 90 (A, B, C, D) were compared to similar controls Mean dady

blood sugars (Bs) at fasting (FBS)and postprandial tmae (PBS) and HbA1 c

dunng testing week were recorded Insulin was used to maintain euglycem~a

RESULTS DMs were euglycemtc by BS and HbAlc and were comparable

to Cs ~n all age groups Mean !SWTs were hxgher m DMs overall (3652 vs

3137 gmsp < 0.05) and m each of groups 1,2, 3, 4 1) CA/TA, 2) IVs, 3)

RVWT, 4) LVWT, 5) RVDD/BPD, 6) LVDD/BPD, 7) RVDD/LVDD

progressively increased as GA advanced m in all fetuses DM heart size and

Wall thickness "~ In Cs al GA of> 24 wks, lhe dllt’erenee being most

s~gmficant ~n DM4 vs C4s (1)0.34 vs 0.32,2) 4.4 vs 3.73 4) 4.80 vs 4.61ram

and 5)0.16 vs 0.15, 6)0.15 vs 0.11 and 7)1.08 vs 1.07 p<0.05) However,

these modalmes were stahshcally similar in DMs and Cs at similar BWT

percentiles e.g. 1) 0.29 vs 0.29, 2) 3.88 vs 3.80, 3)3.82 vs 3.40, 4) 4.24 vs

4.45 mm of wall thickness NS and 5) 0.16 vs 0.19, 6) 0.15 vs 0.15 &7)1.08

vs 1.65 respectively NS at > 90th percentile of BWT.)

CONCLUSION: 1) Global car&ac enlargement and Myoear&al hypertrophy

affecting RV LV occurred more frequently in DM fetuses at later gestation

2) DM BWTs were > Cs despite maternal euglycem~a 3) Cantmc growth

patterns were slmdar m DMs and Cs of s~mllar BWT percentdes 4) Stimulus

to cardiac and somatic growth appears to be s’unilar in infants of insulin

controlled metabolically stable diabetics

402 AUSCI~LTATED FETAL ARR~YTHMIA: ANALYSIS BY EC~OCARDIOGRAM. P. Browne, L. Harsher III, R. Molina Atlanta Maternal-Fetal Mediclne, PC. and Dept. Gyrt/Ob, Emory Univ., Atlanta, Ga OBOECTIVE: This is a descriptive study of the frequency of arrhythmla patterns and structural defects found when patlents are referred from outpatient setting with auscultated arrhythmia. STDDy DESIGN: 119 patients referred durlng a 7-year period 7/1/88 through 7/i/95 for analysls of ar~hyth~ia constltuted the study populatlon. 36 patients (30~3%) had no abnormal findings of rhythm or cardlac anatomy. The frequency of abnormal flnd~ngs for 83 patients with abnormalities detected at echocardlogram are reported. RESULTS: Premature atrlal contractlons (PAC’S} were the most frequent positive findlng (60/83~ 72.2%). Atrial bigeminy (24/83, 28 9%) and atrial pauses (28/83, 24 I%1 were also frequent findings. Fetal su~raventricular tachycardia ISVT) was found in 14.4% (12/83). Complete heart block was rare (2/83, 2.4%). Structural cardiac abnormalities were found in eight patients (9.6%) and xncluded atrial septal defect (2), ventricular septal defect (2), L-transposition /2), b~ventrlcular hypertrophy {2), rlght atrial enlargement/right ventricular enlargement (ll Cardiomegaly (21/83, 25 3%) and pericardial effusions (5/83, 6 0%) were frequently associated wlth the presence of arrhythmia. CONCLUSIONS: Fetal arrhythmia is a rare condltlon, based on infrequent referral. The most common arrhythmla patterns found in associatlo~ with auscultated arrhythmla (PAC’S, blgeminy, pauses), are benign fetal rhythm patterns. Structural cardiac abnormalities were found in 9 6% of patients with documented arrhythmza. Fetal echocardlogram is indicated when a fetal arrhythmia is auscultated.

404 DEVELOPMENT AND APPLICATION OF A GENERALIZED THIRD TRIMESTER GROWTH CURVE

J C Bmahol& M D x, M Smlth-Lev~tm, M D x, F Chervenak,

M.D. Dept of Ob/Gyn New York Hospital-Comell Umv Med

Ctr., New York, NY

OBJECTIVE: To xdenttfy a generalized description of third

trimester fetal growth wbach can be used to assess fetal growth

patterns and to correct for variation m gestational age at dehvery STUDY DESIGN: The fraction of observed 40 week svmght at

weekly intervals (24-42 weeks) is thsplayed and curve fit for 10

population studies from the world literature and for 775 ultrasound

studies of 542 pregnancies w~th established gestational age

RESULTS: The fractional ;vmght growth curves all superimpose

The pattem is s~gmoldal with the 50% point at 3 l 2 weeks, and the maximum velocity of 29 9 gwdday at 34 8 weeks

FracWt = (0 050420237-0 00315645GA + 0 000125378GAa/(1-

0 0464331 IGA + 0 000613737GA~) r2=0 99

With an observed weight and gestational age, the fractmnal weight relation permits determination of weight at another age 0 e, at 40

weeks for singletons or at 36 weeks for multiples)

CONCLUSIONS: Fractional weight increments in the third

trimester have a sigrno~dal (logistic) relationship. This appears to be a generalized (normal) huma~ growth pattern The fractional

weight model can be applied to population studies or to individual

cases

Volume 174, Number 1, Part 2 SPO Abstracts 421 Am J Obstet Gyi~eco]

405 THE UTILITY OF LrLTRASOUND IN THE PREDICTION OF NEONATAL OUTCOME AFTER DELIVERY N 1,000 GRAMS. e~L~r~S Bottoms, R Paul, ] lares, C MacPbersonx, P Jonesx, G Norman ,r~E~omx, J Roberts, D McNdlisx and tbeNICHD MFMU network, Bethesda Obiectiv¢: to determine the accuracy of ultrasound prior to deliver,,, for infants < 1,000grams (ELBW) and compare ultrasound and best’estimate of gcstational-a~ estimation in the prediction of neonatal outcome. Study desigff: This analysis in&ides concurrently collected data re rdin 259 ’ve’" .- ¯

. _ ga g . It horn mfants, delivered at 11 permatal centers, who had an obstetnc ultrasound performed withm one wdek of ddivery and who had a documented estimated f~tal wetght (EFW) and estimated gestational age (EGA) based on ultrasound and best EGA based on dating andrultrasound criteria. Infants were included if they were considered =via"ble~ prior to delivery and if the obstetrician wo61d intervene for fetal distresS. Study paramdt~rs included estimated EFW, esnmated EGA based on ultrasound within one week of delivery, and best EGA Outcome parameters included: birth we ght surv gal mtact survival, 1VH, ROP, seizures, NEC, and 02 dependence. Results: The mean bmh weight was 769~: 143 grams." The mean error between EFW and actual birthweig[ft (mean absolute error) was 10 6% + 8.3%. By. center, the mean error varied between 6.8 and 14.3%. UltraSound accuracy dia not vary significantly wtth birthweight or best EGA The Table describes average’err~r from actual weight ("mean percent error + SD) at various birthwe ghts and gestat ons

Birth weight Error SD ] Gestation Error SD < 500grams 12

/113/ < 24 weeks 11

189/ 5 ~ 600" grams 13 24--25 weeks 11 6-700 grams 10 (7) 26-27 weeks 10 (8) 7- 800 grams 10 (8) 28-29 weeks 10 (8) 8-1,000 grams 11 (8) ~ 30 weeks 11 (10) Total 11 (8) Total 11 (8)

Multivariate analysis revealed best EGA to be more closely associated with intact survival d/an either EGA by recent ultrasound or’EFW. Neonatal morbidMes (p~rcent) based on best EGA are reviewed in the Table.

24-25 weeks [ 69 26 22 20 20 26.27 weeks ] 81 48 12 12 22 2g-29 weeks [ 88 42 9 18 18 > 30 weeks 100 89 0 5 5 Total 77 40 14 17 20

Condu*ion*: The accuracy of p~enatal ultrasound of the ELBW infant does not vary greatly with birthweight The obstetrician’s best EGA is mote closely associated with neonatal outcomes than EFW or EGA by late ultrasound.

407 ONTOGENY OF ISOLATED ULTRASOUND MARKERS

(IUM) FOR FETAL ANEUPLOIDY. M Bronshteinx. A

Drugan~, J Itskovicz-Eldor*, MP Johnson, MI Evans. Depts.

of Ob/Gyn, Rambarn Medical Center, Israel, Ob/Gyn,

Genetics & Pathology, WSU, Detroit. OBJECTIVE To assess the transtent nature of IUM for fetal aneuploidy. STUDY DESIGN: Transvaginal ultrasound (TVS) for detectaon of fetal structural anomalies was performed prior to aliLrtioc~qteslS in 76 patients later found to carry an aneuploid fetus Amniocentes~s was performed at 14 to 16 weeks gestation for advanced maternal age or detected ultrasound abnormalities Markers for anenploidy included nuchal ~!erna (NE), non septated cystac hygroma (NSCH), pyelectasis (P), echog~ruc bowel (EB), or choro~d plexus cysts (CPC) RESULTS: 8 patients had no ultrasound abnormalities, while multiple or major ultrasound anomalies were observed in 46 IUM were observed on TVS in 22 pahents (table), but remained in only 2 patients (9%) on follow t~ ultrasound after the karyotype was known Only 6 of these patients were older than 35 years

Isolated ultrasound markers and fetal aneuplmdy

Nuchal Edema/NSCH 1 %21 (12); sex (4); mar (3)

Echo~enic Bowel T-21; trans 3 15 (de novo)

Pyelectasis T-21

CONCLUSIONS: Reported IUM for fetal aneuploidy appear to be transient and may disappear later in gestataon TVS at 14-16 weeks gestation mayprovide the best ~ window for deteetaon of these markers which may d~sappear by 18-19 weeks gestation when most sonographic screening for structural abnormahties is

406 PRENATAL CYTOGENETIC ABNORMALITY RATES AMONG CASES WITH ULTRASOUND DETECTED FETAL ANOMALIES: A FIVE YEAR EXPERIENCE. RF Hurne. IE Zador’, MP Johnson, J Lampinen~, SAD Ebrahim~, ~ MI Evans Depts of Ob/Gyn, Genetics & Pathology, WSU, Detroit, MI. OBJECTIVE To determine karyotypic abnormality detection rates at prenatal diagnosis among cases with sonographic fetal dysmorphisms. STUDY DESIGN" All abnormal prenatal ultrasounds evaluated m our program from 1/1/90 - 12/31/94 were categorized accordmg to type of fetal anomaly and cytogenetic aberration RESULTS: 1,120 cases of abnormal fetal ultrasounds were identified from the experience of 14,725 reproductive genetic cases and classified as minor, major, or multaple congemtal anomalies. 139 (12 5%) had aneuploidy or other cytogenetic aberration. 39 (35% of total and 28% of aneuploidy) had a previously unsuspected genetic syndrome to explain the fetal dysmorphology 90/139 aneu ~loidy cases were maternal age less than 35 rears.

US Class Aneuploidy Total AMA >35 ~" 23/184 = 26%

= Minor 43 (8%) 550 <35 "~ 20/936 = 2"/o

Major 60 (13°/,) 461 NTD (6%) CNS (18"/,) Multiple 36 (33"1o) 109 CH 40%

CONCLUSIONS: 1)F( rural anomahes detected at prenatal ultrasound examination constitute an increasingly significant proportion of reproduciwe ge~tic practice, especially for those less than 35 years of age 2)The risk of aneuploidy increases with the severity of the fetal anomaly classff~cahon 3)Maternal age alone does not explain the h~gh detection rates. 4)Patients referred for isolated minor fetal anomalies have a significant risk for aneuploidy, associated major anomalies, and other genetic disease 5)Dd~gence m the search for associated anomahes, aneuplo~dy, pedigree analysis, and syndromology remain critical components of the differential diagnostic process

408 THE MULTIDISCIPLINARY APPROACH TO THE PRENATAL DIAGNOSIS OF GENETIC SYNDROMES NOT ASSOCIATED WITH ANEUPLOIDY. E Guzman D Day-Salvatore×, S Shan-Schwarz~, A V~ntzfleos, C Benltox,

R Knuppel D~vlslons of MFM, Ganetlcs and Pathotogy, UMDNJ-Robert Wood Johnson Medmal School/St. Peter’s Medical Center, New Brunswick, NJ OBJECTIVE: To descnbe a collaboratwe team approach and ~ts (fiagnost~c accuracy in the prenatal detection of genetic syndromes not associated with aneuploldy STUDY DESIGN: Dunng the study penod, 8/92 to 5/95, cases demonstrating more than one congenital abnormahty detected on ultrasound evaluation raising the susplclan of a genetic syndrome were reviewed by a team of Maternal-Fetal Med~c~nespeclahsts, chnlcalgenetlclstandpedlatncpathologlst In addition tothe usual components of a targeted ultrasound exammatmn, the fetal face, ears, hands, feet, and gen~taha were examined in the presence of the chnmal geneticist A skeletal survey was performed if deemed necessary A differential diagnosis was generated and d~scussed in a multidlsclphnary conference Autopsy specimen or hveborn were examined for definitive diagnosis All other non-aneuplold genetic syndromes identified postnatally were reviewed to determine whether a prenatal ultrasound examination was performed ~n our unit RESULTS Th rtyseven cases were ~dant~fied with tt~ follow~ng final d~agnost~c categones skeletal dysplas~a (n=15), sporadic syndrome (n=8), autosomal recessive syndrome (n=3) teratogenlc embryopathy (n=4), mlcrode[etlon syndrome (nil) cranlosynostosis syndrome (n=2), branchial arch d~sorder (nil), unbalanced translocation (nil), autosoma dominant (nil) and false posltwe (nil) In only two cases was there a previous family history of the diagnosed condition. The speofic prenatal d~agnos~s was correct in 20/37 [(54 1%), e g Jarcho-Levin, heterozygous achondroplasia (n=3), Goldenhar, Pena Shoke~r, osteogenes~s imperfecta types II lIB and IV Beckw~th-W~edemann (n=2), VATER, Fryns, Pentalogy of Cantrell Ivemark achondrogenesls II, limb body wall complex (n=2), d~abetlc embryopathy, and f~tal hydento~n syndromes There were 12 (32 4%) cases where the specific dtagnos~s was not made but were correctly categonzed and/or part of the differential d~agnos=s and not associated with any errorsln counsehng and management Sixofthese were skeletal dysplas=as, eg. short nb polydactyly-Majewskl type, osteogenes~s =mperfecta type Ill, kyphomehc dysplasia, femoral hypoplasla-unusual faces syndrome, Jeune asphyxiating thoraoc dysplasia and one that was one of several lethal forms Dunng the study penod, four (10.8%) cases were missed on prenatal ultrasound and there was 1 (2 7%) false positive Thus, in 86.5% of the cases either the speofic syndrome was diagnosed, was =n the dflferent~al or was appropnately categonzed Conclusion: A multld~sciphnary approach to the prenatal d~agnosls of diverse and complex genetic disorders not associated with aneuplo~dy can achieve a reasonable degree of d~agnost=c aptitude leading to appropriate patient counsehng and expectation and obstetric deos~on making

422 sPa Abstracts Janum~/1996 Am J Obstet Gynecol

409 A COMPARISON OF TRANSABDOMINAL VERSUS TRANSVAGINAL SONOGRAPHIC MEASUREMENTS OF FETAL NUCHAL TRANSLUCENCY IN THE FIRST TRIMESTER. JC Smuban, AM V~ntzlleos, DA Mclean, JC Rosenberg, ME Martins, DL Day~ UMDNJ-Robert Wood Johnson

Medical SchoollSt Peter’s Medical Center, New Brunswick, NJ OBJECTIVE. The first tnmester sonograpNc evaluat=on of fetal nuchat translucency (NT) has been proposed as a method to screen for tnsomy 21 This study was designed to compare transabdommal (TA) versus transvag=nal (TV) sonagraphlc measurements of fetal nuchal translucency =n the first trimester STUDY DESIGN: TA and/or TV measurements of NT and crown-rump lengths (CRL) were obtained in patients from 10-13 5 wks presenbng for first tnmester sonographlc screening for trlsomy 21 NT measurements were obtained in a saglttal plane using the maximal thickness of subcutaneous translucency between skin and soft tissue overlying the fetal spree Up to 3 TA and/or TV measurements were obtained for each fetus Mean values of TA-NT were compared to mean values of TV-NT for each fetus by paired t-test The relationship of TA to TV measurements and the effect of fetal size on TA and TV-NT was assessed by linear regression The intraobse~ver vanablhty for TA and TV measurements was calculated using atl fetuses with 3 TA or 3 TV values and is expressed as percent devlabon from the mean Other vanables were evaluated with descnpbve statistics

and unpa=red t-test S~gmficance p<0 05 RESULTS. There were 101 smgletoa fetuses evaluated of wh=ch 86 had TA and 71 had TV measurements There were 59 having both TA and TV measurements Two fetuses had a TA-NT>2 5mm and two had TV-NT_>2 5ram A single fetus had a NT>~ 5 by both methods There were no slgmficant differences between pa~red TA-NT (mean 1 22_+0 51mrdrange 0 2-3 5mm) and TV-NT (mean 1 33_+0 56ram/range 0-3 3ram) measurements (p=O 95) There was a s=gmficant correlabon between TA-NT and TV-NT (r2=0 578, p<00001) TA-NT measurements showed a significant relationship w=th CRL (r~=O 16, p=0 0001) TV-NT showed no significant correlation w=th CRL (r2=0 022, p=0 22) Sixty fetuses had 3 TA-NT measurements w=th an mtraobserver vanabdlty of 8 3+6 1% wNch was similar to the mtraobserver varlablhty of 8 4+6 2% for 49 fetuses havtng 3 TV-NT measurements (p=093) The mtraobserver vanaNhty of the measurements did not vary with CRL for edher TA-NT (r~=0 001, p=0 81) or TV- NT (r2=0 004, p=O 68)

CONCLUSIONS. 1) TA-NT measurements correlate well w~th TV-NT measurements from 10-135 weeks 2) Differences between TA-NT and TV-NT measurements are not significant 3) The intraobserver varlabdlty is similar for TA and TV evaluations of NT irrespective of CRL

411 ADJUSTING THE RISK FOR TRJSOMY 21 BY A SIMPLE ULTRASOUND METHOD USING FETAL LONG BONE BIOMETRY AM Vlntzlleos, JFX Egan, JC Smuhan, WA Campbell, ER Guzman, JF Rodls UMDNJ-Robert Wood Johnson Medical School/St Peter’s MeOcal Center, New Brunsw=ck, NJ and Un=vers~ty of Connecticut Health Center, Farmlngton CT OBJECTIVE’ To establish the efl=cacy of second tnmester fetal !ong bone blometry including femur (FL), humerus (HL), hbla (TL), and fibula (FIL) bone length, m detect=ng tnsomy 21 and to generate tables that would allow adjusbng the nsk for tnsomy 21 accord=ng to long bone b~ometry STUDY DESIGN. From 7/1/92 to 6/30/94 four long bones4nclud~ng FL, HL, TL and RL-were ultrasomcally measured =n s~ngleton fetuses pnor to genebc ammocentesls Regress=on equabons descnbmg predicted FL, HL, TL, and RL on the bas=s of the measured BPD were dehned from euplo~d fetuses. Measured to expected (M/E) tong bone ratios were calculated and the <10%lie was used to define a short bone Thereafter, sens=hwty (SN), specificity (SP), positive predictive value (PPV) and negative pred=ctive vaiue (NPV) were calculated. After confirm=ng that SN and SP were =ndependent of maternal age, tables were generated to adjust the nsk for tnsomy 21 by using Bayes’ theorem Outcome =nformabon mctuded the results of fetal karyotypes obtaned by genetic amn=ocentes=s RESULTS: A total of 515 fetuses (14-23 weeks gestat=on) were =ncluded, 493 normal and 22 wdh tnsomy 21 The SN, SP, PPV and NPV are shown below

Abn FL HL TL FiL 21 short bones

(M/E) (<088) (<089) (<086) (<086)

SN 22 7% 45 5% 27 3% 18 2% 63 3%

SP 89 9% 90% 91 3% 91% 78 5%

PPV 9% 17% 122% 83% 1!7%

NPV 96 3% 97 4% 96 8% 96 2% 98%

The adjusted nsk for tnsomy 21 of a 40 year old when no shor~ bones are =denbhed =s equal to that of a 35 year old CONCLUSION: Genebc amnlocentesls may not be recommended for women <40 year old when all four bones are normal Second tnmester fetal long bone b=ometry may be used to adjust the a pnon nsk for tnsomy 21 and therefore counsehng regard=ng genebc ammocentes=s

410 ANTENATAL ULTRASOUND EVALUATION OF THE FETAL THYROID..A_0_C ~, J C Smui~an, M Kungx, A Vu~tzlleos UMDNJ-Robert Wood Johnson

Medical School/St Peter’s Medical Center, New Brunswick, NJ OBJECTIVE Fetal thyroid enlargement is associated wgh maternal use of thyroid

med~cabon There Is only one prewous report addressing normal thyroid

measurements during pregnancy which ~ncluded only 31 fetuses of euthyro~d

pabents Th~s study was designed to generate normal fetal thyroid measurements In

order to evaluate fetuses at risk for thyroid enlargement

STUDY DESIGN Measurements of the w~dest d~ameter of the fetal thyroid were obtained in a transverse wew through the fetal neck at the level of the thyroid

Fetuses of pahents w~th a Nstory of thyroid disease or on thyroid medication were

excluded S~mple regression was used to describe a relationship between femur

length (FL), blpanetal diameter (BPD) and thyroid diameter (TD) The 10th, 50th, and 90th percenbles were determined and s~mple regression was used to develop the best fit equation for them

RESULTS One hundred tNrty-s~x pabents were evaluated Gestabonal age ranged from 15 8 to 39 3 weeks wdh a median of 25 4 weeks The equations for the BPD vs

TD percenbles are

Percentile Intercept Coefficient R2 10th -0 2840 0,9830 0"~9 50th 0 0734 0 9862 0 999 90th 0 29t0 1 0141 0 997

The equahon for the FL vs TD ~s

Pementfle intercept Coefficient R2 10th -0 3017 0 9855 1~0 50th 0 0024 1 0019 0 997 90th 0 2910 1 0135 0 998

CONCLUSIONS E~ther FL or BPD may be used for gestabonal age-independent

pred~cbon of fetal thyroid s~ze The relabonsh~p of e~ther BPD or FL to thyroid s~ze ~s

linear

412 ASSESSMENT OF THE DEGREE OF FEMUR LENGTH SHORTENING IN THE DIAGNOSIS OF SKELETAL DYSPLASIAS. D Day-Salvatore’, E Guzman, C Rodnguez~, A VlntzAeos, S Shen-Schwarz*, R Knuppel D]wslons of Genetics and Maternal-Fetal

Me¢clne, UMDNJ-Robett Wood Johnson Medical School/St PeteCs Medical Center, New Brunswick, NJ OBJECTIVE: To evaluate the accuracy of the degree of femur shortening, as proposed by Goncalves and Jeanty, In the prenatal diagnosis of skeletal dyspfaslas STUDY DESIGN’ Ultrasound data from the study penod, 8/92 to 5/95, were retrospectively reviewed on 15 cases of skeletal dysplaslas diagnosed postnatally Autopsy specimens or hveborns were examined by an expenenced d~nlcal geneticist (DDS) and pennatal pathologist (SSS) In all cases a radiographic skeletal swvey was obtained for definitive diagnosis The ultrasound femur length obtained at the first ultrasound examination was plotted on the curve developed by Goncaives and Jeanty (J Ultrasound

Meal 13 767-775, 1994) TNs curve assigns a specific (primary) d~agnos~s, as well as a I~st of differential d~agnoses to the degree of femur length shortening The degree of femur length shortening was expressed as 10% decremental decreases from the mean femur length for gestatlonal age The accuracy of the curve was evaluated by determining whether the dlagnos~s was 1) number one In the differential list, 2) included as part of the dlfferenhal diagnosis, or 3) missed or not included RESULTS’ Of the 15 skeletal dysplaslas the following 4 were correctly diagnosed (the final diagnosis was #1 In the dlfferenbal hst) heterozygous achondroplase (n=2), osteogenesls imperfecta lib & III (n=2) In the following 4 cases the final diagnosis was not #1 in the list but was included as part of the differential heterozygous achondroplasla (n=2), achondrogenesls (n=l), and osteogenes~s imperfecta lib (n=l) In the following 6 cases the diagnosis was missed or not hsted ~n the dlfferenhal diagnosis by the curve osteogenesls Imperfecta IV (n=l), Jeune asphyxiating thoracic dysplasla (n=l), Jarco- Levm (n=l), femoral hypoplasla*unusual facies (n=l), kyphomehc dysplasla (n=l), and unspec#led dystrophy (n=l) In one case (short-rib polydactyly, Majewskl type) the curve was not useful because of mab~hty to prenatally visualize the femur Thus, the prenatal dlagnos~s was the #1 diagnosis in only 27% (4/15) of the cases, it was part of the ddferentlal In 53% (8/15%) and it was missed or not hsted as part of the dlflerenhal in 40% (6/15) CONCLUSION Inth~s study populatlonthe usefulness offemurlength assessment alone m the specific prenatal diagnosis of skeletal dysplaslas was hmlted

Volume 174, Number 1, Part 2 SPO Abstracts 423 Am J Obstet Gvne(ol

413 THREE DIMENSIONAL ULTRASOUND OF THE FETAL

SPINE. D. D. JohnsonX, M. Riccabon#, N.E. BudorickX,T.R.

Moore, D.H. Pretoriusx, and T.R. Nelso#, Departments of

Reproductive Medicine and Radiology, Univ of Calif San Diego,

La Jolla, CA.

OBJECTIVE: To compare visualization of the fetal spme of

second trimester fetuses utilizing conventional two dimensional

(2D) and three dimensional (3D) ultrasound.

STUDY DESIGN: Sonographic examinations were performed

on conventional 2D and commercially available 3D equipment

(Combison 530, Kretz Technik AG, Austria). Twenty normal

fetuses and 5 abnormal fetuses (mean gestational age 21 weeks)

were included. Two investigators compared the anatomy

displayed on 2D images to the 3D images.

RESULTS: Pathological findings, which included neural tube

defect, scohosls, and hemivertabrae were depicted on both 2D

and 3D scans. 3D images were superior to 2D images in

displaying the relationship between the fetal spine and the

clavicles, scapulaes, ribs and dlum. In addition, pairing of the

posterior elements were easier to demonstrate in 3D views.

Finally, scoliosis was more clearly defined utilizing 3D.

CONCLUSIONS: 3D provides additional information about

fetal spinal anatomy. Thus, 3D has the potential to provide

useful chmcal information.

415 ELEVATED MATERNAL SERUM ALPHA-FETOPROTEIN AND PLACENTAL SONOLUCENCIES. R~ Ralnus, L Mar~mx, T Dowd, M Lucas, R Santos-Ramos, D Twickler. Dept of Ob/Gyn, Umversity of

Texas Southwestem Medical Center, Dallas, TX. OBJECTIVE: Patients with an unexplained elevation of maternal serum

alpha-fetoprotein (MSAFP) are known to be at an increased risk for adverse pregnancy outcomes. Placental sonolucencies are frequently seen m flus population, and It is unclear whether they independently contribute to

these adverse outcomes. Tins study exanunes the independent effects of

elevated MSAFP and placental sonohicencies STUDY DESIGN: Data was obtained from the MSAFP screening program

from 1990 to 1994 in women from three distinct groups 1) Unexplamed high MSAFP(>2 5 MOM) with a normal placenta on ultrasound(US). 2)

I-hgh MSAFP with placental sonolucencies; 3) Normal MSAFP with sonolucencles on US Outcomes were obtained on all women from our obstetrical database. Chi square analysis and ANOVA were usexl to compare outcomes in each group.

RESULTS: Selected outcomes for the 367 women are presented below. Outcome

N

Delivery < 37 weeks (%)

Small for gest. age (%)

Cesarean section (%)

Birthweight (gm)

Preeclampsla (%)

Anomalies (%)

Group 1 Group 2 Group 3 P

203 38 124

21.5 47.2* 10 7 <.001

145 83 74 134

30.1 22.2 19.0 .083

2665 2688 3182* < 001

17.7 13 9 8 3 065

13.3 7.9 8.1 .277

Preterm delivery was significantly more frequent m the high MSAFP with

sonolucencies group, and the birthwelghts were higher in the normal MSAFP group

CONCLUSIONS: Poor outcomes were increased in the high MSAFP groups compared to the sonohicency group When both risk factors are present there may be additive effects in regard to preterm delivery

414 SONOGRAPHIC PREDICTION OF FETAL OUTCOME IN SUSPECTED SKELETAL DYSPLASIAS USING THE FEMUR LENGTH TO ABDOMINAL CIRCUMFERENCE RATIO. R Ramus L MartinX, D. Twickler Dept of Ob/Gyn, Umversity of Texas Southwestern Medical Center, Dallas, TX. OBJECTIVE: To dctemline if the femur length (FL) to abdommal

circumference (AC) ratio can be utilized antenatally to predict a lethal skeletal dysplasla

STUDY DESIGN: All obstetrical sonograms performed from 1990 to 1994 were reviewed (36,731 studies) to fmd those scans suggestive of a skeletal dysplasia Twenty-eight patients were identified The FL/AC ratio was then calculated from each patient’s lmtlal sonogram Based on sonographlc fmdmgs three pattents elected to telminate the pregnancy, and one stdlbom had a vem of Galen aneurysm with high output failure and a normal

skeleton Btrth outcomes were obtained on the 24 remalmng patients RESULTS: The FL/AC ratio is stratified mto subeategories and matched with outcomes m the table below

<0.16 016to 018to >020 Outcome <0 18 <020

Stillbirth or neonatal death 10 0 0 0 Nonlethal skeletal dysplasia 1 2 4 0 No evidence of skeletal 0 0 3 4

dysplasia m the uursery The one fetus with a ratio < 0 16 who survived the neonatal period has

osteogenems lmperfecta and was lost to follow-up after a prolonged stay in the intensive care nursery

CONCLUSIONS: A stlllb~la or neonatal death occurred in 10 of 11 patients with a FL/AC ratio of less than 0 16, mdependent of gestational age Conversely, no fetus with a FL/AC greater than 0 16 was found to have a lethal skeletal dysplasia This mformation may be useful in counseling women when sonography suggests the diagnosis of a skeletal dysplama

416 ETHNIC DIFFERENCES IN FETAL BIOMETRIC NARKERS OF DOMN SYNDRC~E. $. Bloom, R. Ramus, D. Twickler, M. Lucas, Dept. Ob/Gyn, Univ. TX Southwestern Medical Center, Dallas, TX OBJECTIVE: To evaluate whether the ratios of measured to expected femur and humeral lengths, which have been reported to bemerkers for Down syndrome fetuses, vary by ethnicity. STUOY DESIGN: Between January 1 and July 31~ 1995, we evaluated

1491 women who underwent ultrasound examinations for indications between 14 and 24 weeks gestation. The mean maternal age was 24.8±6.5 years. The ratios of measured to expected femur and humeral Lengths were calculated for each fetus based upon previously published formulas ($enacerraf: J Ultrasound Med 11:449, 1992):

Expected femur length = -9.3105 ÷ 0.9028 x BP0 Expected humeral Length = -7.9404 + 0.8492 x BPD

The ratios were then stratified by maternal race, and the means were compared using an ANOVA and student Newman Keuls. RESULTS:

Hispanic Black Caucasian

Number 876 400 215

Mean femur ratio 1.020±0.072 1.033,0.091 1.00620.081

% femur ratio 6.6% 5% 13~ 5 0.91

Mean humeral 0.96420.082 0.981±0.089 0.95020.088 ratio 2 SD

% humeral ratio 17.4% 12% 27.4~ < 0.90

Both femur and humeral mean ratios for the three ethnic groups differed significantly from each other (P ¯ 0.05). These differences, In turn, affected the percentage of each group identified as at risk for aneuploidy when a single cutoff threshold was applied. A higher than expected percentage of abnormal humeral ratio was observed in all groups. COBCLUSION: Ethnic differences are an important consideration when utilizing biometrics for Down syndrome screening, ge suggest that population-specific formulas may be appropriate.

424 SPO Abstracts January 1996 Am J Obstet Gyne~ol

417 NORMATIVE DATA FOR FETAL RIB LENGTH:THORACIC

CIRCUMFERENCE MEASUREMENT AND APPLICATION TO

PRENATAL DIAGNOSIS OF SELECT SKELETAL DYSPLASIAS.

L DugofP, J C Hobbms Dept Ob/Gyn, Umv ot CO Health Sciences

Center, Denver, CO

OBJECTIVE: To esmhhsh new, normative data tbr the fetal r~b

length thoracac circumference ratm (RL.TC) and to evaluate the apphcatlon

of this measurement in the diagnosis of short rib and other skeletal

dysplaslas

STUDY DESIGN: RL and TC measurements were obtained on 75 patients

with normal, well-dated pregnancies between 14 and 38 weeks gestation

A single, cross sectmnal measurement of the fetal thorax was performed at

the level of a four-chamber view of the fetal heart RL was measured by

tracing from the most anterior point of the nb on one side, posteriorly

around the rib cage, to the most anterior point on the opposlte side TC

was obtained by tracing around the fetal thorax at the same level The

relatmnship between gestatmnal age and I~L TC was analyzed for patients

in three gestatlonal age groups’ 14-19 (n=27), 20-29 (n=26), and 30-38

weeks (n=22) usxng one-way ANOVA and the Bonferronl mulhpIe

comparisons procedure RL and TC measurements were also obtained

fl’om 6 cases of skeletal dysplas~a

RESULTS: The mean RL TC (_+ S E ) for each ot the 3 gestational age

groups was 0 66 + 0 009, 0 67 _+ 0 007, and 0 67 _+ 0 007, respectively

(p=NS) RL TC for the 6 cases of skeletal dysplas~a was 0 32 for short

nb-polydactyly syndrome (Sald~no Noonan), 0 54 for thanatophorlc

dysplasla, 0.59 for Jeune syndrome, 0 59 for atelosteogenes~s type 1I, 0 52

for achondrogenesls, and 0 82 for osteogenesls lmpeffecta(O1)

CONCLUSIONS: RL TC m normal fetuses ~s constant regardless of

gestationalage RL TC was decreased, as expected, m Saldino Noonan,

but was also unexpectedly decreased In several of the other skeletal

dysplasms. Conversely, RL TC was increased m the case of O1 We

conclude that the RL TC ~s independent of gestatlonal age, and may prove

to be helpful m d~agnosmg select types of skeletal dysplas~as

419 PRENATAL DIAGNOSIS AND MANAGEMENT OF FETUSES WITH INTRACRANIAL HEMORRHAGE. J. Notes z, A. Roberts, S. Carr. Division of Maternal-Fetal Medicine, New England Medical Ccoter/TuP~ University School of Medicine, Boston, MA; Dept. Ob/Gyn, National Womens’ Hospital, Auckland, NZ; Dept. Ob/Gyn, Brown Univeraity, Provident, RI. OBJECTIVE: To describe different presentations of prenatally diagnosed iatracranial hemorrhage and ira impact on subsequent nmaagement of pregnaneiea complicated by this condition. STUDY DESIGN: We reviewed the sonograms, medical records, neonatal studies and pathology reports in 11 selected pregnancies with intracranial hemorrhage diagnosed in utero between the period 1990-1995, and correlated ~onographic findings with postnatal studies in an effort to assess ultrasound’s diagnostic aecuraey and theroforo base counseling and management accordingly. RESULTS: Ul~end wa~ able to correctly identify the site of intraeranial hemorrhage (subdural, paronohymal or intraventrieuiar) and the development of ~.xondary ventricuhr dilatation. Significant fetal anemia was documented in 8 cases. Five eases resulted in intrauterine demise, five had an early neonatal demise (3 of which had labor induced between 26-35 weela in view of their extremely poor prognosis), and the only survivor had ~vero developmental delays. An insight into the natural history of iatraventricuhr hemorrhagein utere i~ provided, and enrrohtion with clinical presentation and outcome is made allowing for a more rational approach to the management of this condition. CONCLUSIONS: Ultr~ound can effectively diagno~ different types of ~ntracranial heanorrhage in utero as well as provide essential information regarding th~ extent o f fetal enmprom~, therefore becoming an invaluable tool in the management of these prognaneles.

418 ULTRASOUND ACCURACY IN THE PRENATAL

DIAGNOSIS OF SKELETAL DISORDERS. J. Notes x, S. Cam Division of Maternal-Fetal Medicine, New England Medical

Center/Tufts University School of Medicine, Boston, MA.; Dept.

Ob/Gyn, Div. of Maternal-Fetal Medicine, Brown University School of Medicine, Providence, RI.

OBJECTIVE: To evaluate the accuracy of ultrasound in the

prenatal diagnosis of skeletal disorders. STUDY DESIGN: Retrospective review of all consecutive prenatal sonograms performed over a two year period (5/92-5/94) at the

Prenatal Diagnosis Center in all patients referred with a family bdstory or a suspicion of skeletal abnormality. Prenatal sonographic

diagnosis was compared to f’mal postnatal diagnosis in all cases.

RESULTS: 63 patients were evaluated, 48 with a suspicion of

skeletal abnormality, 15 with a family history of a skeletal

disorder. There were 13 skeletal disorders, 4 chromosomal

anomalies, 2 dysmorphic syndromes, 2 cases of growth restriction

and 42 with normal outcome. Of those 13 skeletal disorders

encountered, 12 were diagnosed prenatally as a specific disorder, with only 5/12 correctly diagnosed with a positive predictive value

(PPV) of 42%. In the family history group the PPV was 100%

(313), while in the group with an incidental finding of a skeletal

abnormality the PPV was 22 % (2/9). In the patients diagnosed as

normal no skeletal disorders were encountered (n=48).

CONCLUSIONS: Prenatal sonographic diagnosis of a specific skeletal entity is seldom accurate, except in that subgroup of

patients with a family history of a skeletal disorder.

420 TIlE SIGNIFICI~NCE OF C’EREBRAL DOPPLER VEIA)(’IMETRY IN NORMAl, :LND

GROWfH RFIARDED FETUSES (~111~t7]~e LHZI~, Graztano (’lertc~~, (hart (’arlo

I)~ Renzo Insl of Ob/Gvn, Cenlre of PcrmataI Medicine llmv of Perug~a,

Pcrn~m I~,~l~

OnlEcrI~ES WC studied lhe varlatlou of the fetal cerebral circulation (as

mdlcated b’~ the pulsatlhlV ~ndex. PI) m nonnall3 developing tktuscs and m

IUGR t~tuses m tbc second hall of prcgnmacy STraY D~ZSl~;n" We recorded

watb a pulsed Doppler w~lh CFI ( B&K model no 7575) the F~ and

cnlculaicd the P1 ol the m~ddle cerebral arlerv (MCA), aulenor cerebra!

ariel3 (ACA) posterior cerebral nrlely (PCA) and umbthcal artery (UA) ~n

20 uncomphcated plegnancles every 2 weeks titan 24 ~ks ol geslallon

cstnnated t~lal body lIqabb ~ndcx (FBMI) ~EFW/ (cshmated tetal lenght):1

u a~ ~alculated m lhc ~mne preguaucxc~ I he percenlagc w~rlahon (~%) of lhc

1~ and F~3M[ at the dtlteren~ gestationM age ~as calculated We stud~cd the

same vc~scls m 20 IUGR ictuses ( 12 w:h gro~h markedl3 redoced. I JA end

thastohc/low present, and 8 w~th no grm~,~h, UA end dtastohc llox~, absent,

A]:I) m rever~c chdstohc Ilow, RFD) ,ard calculalcd the FBMI and A% ol PI

aud ol FBM1 m colnp~sm~ lo the normally developing ieluseb at comparablc

gestalmnal age RFMmFs, hi nonnall~ developing l~tnses, we observed >

50% recreate of the I’13~ betwceu 26 weeks of gcstallon and

prcguanc3 Ihc Pl ol cclcbral vessels and the PI ol umblhcal arte~ (UA)

decreased approxmmtclv 2t)% bclx~een 26 weeks otpregmmcv ut~d toni1 The

decrease of PI ~as ~elaled to gcstahonal age and to lq3~ m the nonnall~

dcvclopmg lctuse~ In the nnld ~J(}R we observed n decrease ol PI m all

cerebral vebscls al values approx 20% less than the PI ol normally

developing i~h/ses at the same ge~lahonal age. whde the PI of mnblhcal

artcr~ increased over 40% In severe ~JGR uc observed an increase oflhe

m ACA and PCA m presence of AEI) flow m umbd~cul mtety, while the PI

ol MCA m MI remained Ion m case of RED flou, also Ibe PI of MCA

mctabohc dcmnnds lhe maxmmm decrease ofPi ~n cerebral vessels resnlt~

approxmmtclv 20% al am gestahonal age

Vohnne 174, Numbel l, Part 2 SPO Abstracts 425 Am,] Obstet Gynecol

421 THE UTILITY OF THE CISTERNA MAGNA TO

CEREBELLAR DIAMETER RATIO AS A SCREENING

TEST FOR TRISOMY 18. R. Steiger, D, Lagrew, R, Randall.

Dept, Ob/Gyn, Univ, of Caltf,, Irvme, Orange, CA

OBJECTIVE: The uuhty of the ctsterna magna to cerebellar

dtameter (CM/CB) m the detection of msomy 18 was compared to

use of measurements of the cistema magna (CM).

STUDY DESIGN: A sample of 18 fetuses wtth an antenatal

thagnos~s of trisomy 18 and measurable cerebellums and cistema

magnas was mixed with a prospective cross sectional sample of

572 normal controls Measurements were done tn the standard

plane for measurement of cerebellar d~ameter. Results were

compared by hnear regression analysis and receiver operalor curve

(ROC) analysis.

RESULI"S: The normal range of the measurement of the c~stema

magna changes over the course of gestalton whereas the CM/CB

remains constant. ROC analysis showed the CM/CB to be an

overall superior thagnostic test in comparison to the measured CM.

In the range of high specifictty where a cutoff value would be

chosen there was htlle difference in the two tests. At 2.0 standard

devtations the sensitivtty, specifictty, posttive and negative

predtctive values of the CM/CB were 27.8%, 95.1%, 15 2% and

97.7% and for the CM tt was 26.3%, 96 2%, 18.5% and 97.5%.

CONCLUSION: The CM/CB has the advantage of having a

constant range over the course of pregnancy (0.09 to 0 39 at 2.5

SD) and techmcally is a superior diagnostic test by ROC curve

analysis but in the range where a cutoff value would be chosen it

offers little advantage over the measured CM.

423 FETAL CEREBRAL VENTRICULOMEGALY: CLINICAL SIGNIFICANCE AND PERINATAL OUTCOME. AM Adra XM Zeng~, SN Bcydoun. Department of Obstetrics & Gynecology, University of Miami, Miami, Florida.

OBJECTIVE: To assess the clinical significance, natural hi~tnry sod

outcome of fetal lateral ventricuinmegaly (VM).

STUDY DESIGN: 59 fetuses scanned serially between 16 and 38

weeks’ gestation to assess the natural history of VM. Periuatal outcome

with follow-up information up to 5 yoars of ago were obtained from

obstetric, neonatal, genetics and neurosurgery consultant charts.

RESULTS: 10 cases were excluded. The remaining 49 were divided

into 4 groups, IA-lsolated, stable (< 15 ram); IlA-VM & Meningomyelocele

IB-lsolated, progressive VM;IIB-VM & other CNS pathology

IA IB IIA lib

(n=4) 1n=12) 1n=24) (n=9)

VP Shunt (Postnatal) 0 66% 87% 37%

Severe Development Delay* 0 83% 13% N/A

Abnormal Karyotype 8.5% 12.5% 11%

Perinatal Mortality (/1000) 0 250 SS 110

* p < 0.05 - Chi Square test

The two most common postnatal diagnoses in group IB were intrauterine

infections (CMV & toxoplasmosis) in 4 cases and major CNS anomalies

in 3 cases.

CONCLUSIONS: 1) The prognosis for fetuses with isolated, mild

(.~.15 ram) VM is good. 2) Fetuses with apparently isolaW, d but

progressive VM have a much less favorable prognosis n.nd parents should

be counseled about the increased risk of intrauterin© infections and

undetected CNS anomalies. 3) Fetuses with VM associated with

meningomyelocele have better mental development than those with

isolated, but progressive VM.

422 PRENATAL COURSE OF CONGENITAL CYSTIC ADENOMATOID

MALFORMATIONS MC Treadwell MW Tomhnson, HM Wolfe, A Flake

Depts Ob/Gyn and Pediatric Surgeu¢, Hutzel Hospital/Wayne State Umverstty,

Detroit, MI

OBJECTIVE: The current hterature reflects a poor prognosis for prenatally

diagnosed microcysuc (Type III) congenital cystic adenomatold malformations

(CCAM). Abnormal cardiac axis and presence of hydrops have been

associated with a worse outcome. We report our experience

STUDY DESIGN: The ultrasound database ~denufted patients diagnosed with

CCAM between 1992 and 1995 Lesions containing cysts >5ram were

defined as macrocystlC, mlcrocystlc lesion cysts were <5ram diameter

Prenatal and ~mmediate postnatal follow up was obtained

RESULTS: Fou~een cases of CCAM were ~dentified and are summarized

below There was one case of hydrops in a fetus with macrocystic CCAM and

normal cardiac axis that terminated at 18 wks gestation. A second patient with

mlcrocysnc lesion, normal cardiac axis, and no hydrops terminated at 18 wks

Macrecystic Microcystic Total

N 6 8 14

Mean GA at dlagnosls(wks) 21 20 9 21

Axis dewatlon 4 6 10

Improvement on U/S 2 6 8

Mean GA del(wks)* 37 3 38 6 37 9

’excludes terminations of pregnancms

Only one neonate was (mildly) symptomanc at birth and dld well after surgery

pertbrmed at 2 days age

CONCLUSIONS: These findings support a better prognos~s than has been

previously reported in the literature. Mtcrocystlc les~ons and abnormal cardiac

axis did not seem to carry a worse prognosis Hydrops occurs infrequently

No patients required extensive support Immediately after dehvery

424 SECOND TRIMESTER CYSTIC HYGROMA AND A NORMAL KARYOTYPE: ASSOCIATION WITH SKELETAL ABNORMALITIES. JE Solomon: JA Notes,~ ME D’Akon, DW Bianehi:, Divisions of Maternal Fetal Medicine and Genetics, New England Medical Center/TuftS University School of Medicine, Boston, MA.

OBJECTIVE: Individual reports have notod the presence of cystic hygromas in louses affected with skeletal abnormalities (SA). To d~te, thorn is no series documenting the magnitude of this association. We sought to determine if the presenc� of a cystic hygmm~t in karyotypieally normal 2nd trimester fetuses is associated with an ineruased incidence of SA. STUDY DESIGN: We retrospectively reviewed 1753 sonograms entered into a prenatal anomaly database over a 3 yr period. Twenty-seven fetuses with a second trimester cystic hygroma were identified and karyotyped; 13 (48 %) had normal ehrntnosomes and comprised the study population. Post- delivery follow-up was obtained in all eases.

RESULTS: The incidence of SA was 61% (8 of 13). Seven fetuses were terminated. The following abnormalities were noted’ severe long bone

shortening (3), bowing of the long bones (1), hypomelia (1), polydaetyly (1), bihteral club feet (1), and fusion of the ribs (1). In the remaining 6 eases, long bone shoaening (2), polydaetyly (1), and hypomelia (1) were observed. All fetuses carried to term had resolution of the hygroma, but 2 have severe long bone shortening that only became evident in 3rd trimester

scans. CONCLUSIONS: The incidence of SA is significantly increased in 2rid trimester fetuses with cystic hygrnmas and a normal karyotyp¢. A specific pattern of anomahes was not found, but long bone growth deficiency was the most common, seen in 5 fetuses. A detailed skeletal survey of such fetuses is indicated, as well as follow-up scans in the 3rd trimester for assessment o f long bonc growth. Patients should be counsded regarding an

association with skeletal anomalies that may only become apparent later in gestation.

426 SPO Abstracts Janualy 1996 Am J Obstet Gynecol

425 SONOGRAPH1C IMAGING OF THE FETAL FACE AND NECK AND THE ESTABLISHMENT OF NORMATIVE

DIMENSIONS. E. Slvanx, L Chan, A Mallozzi-Eberie, E A Reece

Depts Ob/Gyn, Temple Umv Sch of Med & Einstein Med Ctr, Phila, PA OBJECTIVE. Sooographic imaging of the fetal face is important since a

number of chromosomal abberataons are assooated wath facial malformatmns. Prenatal diagnoses of abnormalities of the chin, hps, and neck have been based on subjective assessment. The purpose of the

current study was to establish normative damens~ons for the objective assessment of the fetal face STUDY DESIGN: A cross-sectional study was performed in 170 normal

healthy singleton pregnancies between 16 and 36 weeks gestatmn. Inclusion criteria included performance of a I st or early 2nd trimester

scan to confirm gestational age (GA), Exclusion criteria included birth weight <10th or >90th percentile for GA, and medical complications of

pregnancy Ultrasound measurements were made in the mldsagittal plane and included chin length (CL), upper hp width (ULW), and the

transverse diameter of the neck (TDN) RESULTS: A 1st degree hnear growth function was observed throughout pregnancy between GA and ULW (r=0 88), CL (r=0 91); and TND

(r=0.94). All measured data are expressed as mean .+ 2 S.D.

GA (wks) N ULW (mm) CL (mm) TDN (mm)

!6-20 42 43+_1 8 62+_30 20 1+_6 4 21-24 37 6 0_+2.0 9 6_+2.1 28 3-+4 8 25-28 23 7 8+1 4 11 2-+2 1 39 2+-6 2 29-32 33 8 8_+2 0 15 3_+2 0 42 7+6 6 32-36 35 10.5+-3.2 17.7+-4.3 47.2_+4.4

CONCLUSIONS: These results provide normative damens~ons throughout gestation for the fetal upper lip, chin and neck These data are expected to serve as a basis for the objective assessment of the fetal face and neck in high risk conditions.

427 A QUANTITATIVE EVALUATION OF THE FETAL FOREHEAD USING THE NEW THREE DIMENSIONAL (3-D) ULTRASOUND TECHNOLOGY. L. Char,. E Sivanx, B.

Uerpalrojkltx, G P Chux, E.A Reece, A. LudonurskL Dept of Ob/Gyn, Temple Umv Sch of Med, Phfla, PA.

OBJECTIVE: We sought to evaluate the fetal forehead and to generate normative data throughout gestation. These data are expected to permit the objective assessment of the normal fetal forehead as well as frontal

bossing in dysmorphology syndromes STUDY DESIGN: 67 normal fetuses were prospectively assessed using 3- D ultrasonography with 3.5/5 MHz volume transducer. The face was

~dentff~ed ~n the B-mode, the volume scanned and the data computer stored for subsequent analyses. A rmd-sag~ttal face profile was obtmned

and a hne connecting the apex of the philtmm and the nasmn was drawn across the anterior forehead which dehneated the area for analyses. The forehead length (FL), height (FH), and the area (FA) under the curve were

measured (mean +- SD) and the forehead ~ndex (FH/FL) was calculated.

RESULTS: All data followed a 2nd order pol)~nomial curve w~th GA.

GA (wks) N FH (mm) 1%, (mm) FA (cm2) FHIFL

16-20 19 7.3+-1 9 27.5+_64 1 2_+0.6 027.+004

21-25 18 9.1+_2.9 35 1-+6.4 1.8.+0.9 0.26_+006

26-30 16 12 3_+4.1 43 1_+9.8 3.3+-1 7 0.28-+0.05

31-35 10 11.8_+2.4 46 2_+7.4 3 2+_1 3 0.26_+0.03

36-40 4 9.6_+2.7 44 1_+7 5 2.1+_1.1 0.22_+0.05

CONCLUSIONS: 1)The growth of the fetal forehead (FH, FL, and FA)

increases until the early third trimester and is relatively unchanged

thereafter, 2) The fetal forehead ~ndex remmns relatively constant throughout gestation. Our established nomoerams of the fet,al

forehead index and area are exnected to serve as a basis

a~ainst which facial dysmornhglogy syndromes may b~

objectively evaluated.

426 CEREBELLAR VERMIAN DEFECT AND PRENATAL DIAGNOSIS OF DANDY-WALKER SYNDROME. I. Goldsteinz, E.A. Reece. Department of OB/GYN, Rambam Medical Center, Halfa, Israel and the Department of OB/GYN & RS, Temple Univ Sch of Med, Philadelphia, PA, USA.

OBJECTIVE: To determine the main anatomical defects observed sonographicaily in early gestation in fetuses with Dandy-Walker malfc~mation_

STUDY DESIGN: Sonographic examination of the posterior fossa of the fetal head was conducted in fetuses during early gestation. These exams included measurements of the transverse cerebellar diameter, the depth of the cistema magna, evaluation of the cerebellar hemispheres and the superior and inferior cerebellar vermis. An attempt was also made to

correlate characteristic posterior fossa abnormalities with an early diagnesis of Dandy-Walker malformation.

RESULTS: Transvaginal ultrasonography significantly improved prea~atal evaluanon of the posterior fossa. In this series of cases, the ultrasound findings can be subdivided into two groups: Group A, included

gestational ages 14 and 15.5 weeks’: the cerebellar hemispheres were widely separated by a U-shaped defect, and there was absentee of the inferior cerebellar vermis. Free communication between the cistema magna and the fourth ventricle was present. There was also normal

measurements of the transverse cerebellar diameter, the atrial width of the lateral ventricles, and the depth of the cisterua magm. ha the second group, gestatioual ages 16.5 and 18.5 weeks’, the cerebeliar hemispheres were separated, the inferior cerebdlar vermis was absent and associated with ventriculomegaly and/or cistema magna measurements above 9 ram.

CONCLUSIONS: The main anatomical lesion In Dandy-Walker

syndrome detectable by early sonography appears to be a partial or

complete absence of the cerebellar vermis, resulting In a p~terlor foasa defect and giving rise to "cyst" formation and eventually ventriculomegaly in later gestation.

428 VOLUMETRIC ASSESSMENT OF FETAL CEREBELLAR GROWTH AND DEVELOPMENT DURING PREGNANCY

USING THE NEW THREE DIMENSIONAL (3-D) ULTRASOUND TECHNOLOGY. L_Chal~ B. Uerpa~rojkltx, W,M.

Lmx, G P Chux, M Khandelwalx, E A. Reece, A Ludomlrskl. Dept of

Ob/Gyn, Temple Umversity School of Medicine, Phiia, PA OBJECTIVEg Using 3-D ultrasonography, we sought to evaluate volumetric growth of the fetal cerebellum and to estabhsh normative

dlmensmns throughout pregnancy. Such dimensions are expected to be useful for the m utero assessment of deviant fetal growth.

STUDY DESIGNI A prospective evaluation was conducted in 126 appropriate for GA fetuses, between the 10th and the 90th percentile for

all parameters. Multiple bmmetnc parameters were obtained including the estimated fetal weight (EFW), TCD and the total cerebellar volume (TCV). Transabdormnal scan were performed nslng either a 3 5 or 5 MHz volume transducer with bulltqn 3-D software The cerebellum was

identified ~n the B-mode, the volume of interest scanned and the data computer stored for subsequent analyses. Data below are expressed as mean +_ S D and regression analyses w~th curve fitting were performed. RESULTS: The fetal TCD and cerebellar volume follows a 2nd order

polynormal dlsmbunon, with R2=0877 and R2=0.836, respect~vel)~

GA (wks) N TCD (mm) TCV (cm3) I

I 116-20

33 18.4_+3 9 0.87+_0 34

]21-25 41 22.8_+3 0 1.80+_0 34

126-30 25 31.2_+4 3 4.77+_1 98

~31-35 21 39 8+4.3 8 45_+2.22

136-40 6 45 3_+5 6 14 71_+3 44

CONCLUSIONS: The fetal TCD and total cerebellar volume (TCV) have a high correlation wlthGA. Our established nomogram of fetal

cerebellar volume is exoected to serve as a basis against

which abberrant fetal erowth can be comnared.

Vohune 174, Number 1, Part 2 SPO Abstracts 427 AmJ Obstet G~,necol

429 THREE DIMENSIONAL (3-D) ULTRASOUND ASSESSMENT

OF CERCLAGE PLACEMENT AND CERVICAL

INCOMPETENCE. L Chan, B Uerpmrojkltx, FL Gomezx, EA Reece,

A Ludormrskl Dept Ob/Gyn, Temple Umv Sch of Med, Phlla, PA

OBJECTIVE: To evaluate the role of 3-D ultrasound in the assessment of

cervical changes in women Wlth cerclages for cervical incompetence

STUDY DESIGN: All pregnant women with cerclages were recruited for

transvaglnal scan using a 7 5 MHz volume transducer w~th bmlt-ln 3-D

software (n=9, 16 studies) The cervix was identified in the B-mode, the

volume of interest was scanned and the data computer stored for

subsequent analyses. The cervix was measured for dilation, length and

volume The cerclage was visualized in all 3 perpendicular planes and its

relatmnsh~p to the internal os was also noted Cerclage d~ameter and

circumference were measured Data were reported as mean _+ S.D

RESULTS: 3-D ultrasonography allowed the Vlsuahzatlon and

evaluation of the entire cerclage and cervix with its precise contours

appreciated Two patients had cervical dilation of 2mm and 32mm,

respectively One patient had persistent cervical funnehng and

dehvered at 37 weeks gestation. Cerclage dmmeter and circumference

were relatively constant over the course of pregnancy Cervical volume

changes were dependent on cervical length and effacement

Ges~ational Age (wks) Cervix Cerclage (mm)

At Cerclage 17 1_+3.0 Length 26 0±16 8 mm Dla 22.7-+8.7

Entry to study 20 4_+4 8 Volume 28 7-+25 7 mm3 Clrc 73 6+26 5

At de lvery 33 7-+4 4

CONCLUSIONS: 3-D ultrasound permitted the actual of information on

cervical volume, contour, cerclage diameter, circumference and its

precise location relative to the internal os and Vlsuahzatlon of the

cervical cerclage in its entirety 3-D ultrasound technology i s a

useful adjunct in the assessment of cervical incompetence

and the adequacy of cerclage placement.

431 IMPACT OF ULTRASOUND SCREENING ON THE PREVALENCE OF CONGENITAL ANOMALIES AT BIRTH. xE.Z. Zimmer, xz. Avraham, xp. Sujov, I. Goldstein, XM. Bronshtein. Dept. Ob/Gyn and NeonatoIogy, Rambam Medical Center, Haffa, tsrael. OBJECTIVES: 1. to assess whether ultrasound screening affected the prevalence of congenital anomalies at birth and the incidence of termination of pregnancy for fetal anomalies. 2. to assess the accuracy of prenatal ultrasonography. STUDY DESIGN: We studied all births and fetal deaths with congenital defects and all terminations of pregnancy for fetal anomalies in one Medical Center during a five-year period (1989-1993) and correlated it to ultrasound findings. RESULTS: There were 23.439 births during the study period. The percentage of newborns with anomalies decreased from 1.95% to 1.34% (p<0.01). The percentage of terminations of pregnancy increased from 0.35% to 0.83% (p<0.003) and the detection rate of malformations increased from 53.94% to 82.60% (p<0.001). CONCLUSION: Termination of pregnancy after ultrasound examination had an impact on the prevalence of anomalies in newborns. There is significant improvement in the accuracy of ultrasound.

430 GROWTH OF THE FETAL ORBITS AND LENS IN NORMAL PREGNANCIES. Israel Goldstein Ada Tamir x Dept. OB/GYN, Rambam Medical Center, The Brace Rappaport Faculty of Medicine, Haifa, Israel OBJECTIVES: To obtain dimensions of the fetal orbits and lens for comparison with deviations in growth. STUDY DESIGN: The study group included 170 normal pregnant women from 14-40 weeks’ gesta~on. Routine biometric measurements were obtained on all the padicipants, including biparietal diameter (BPD), head circumference (HC) and abdominal circumference (AC), and the measurements of the femoral length (FL) and the two diameters of the fetal orbits and lens. The orbital and lental circumferences were calculated for each gestatJonal age (GA). RESULTS: A linear grow~ function was observed between GA and the antero-posterior (r=0.952; p<0.00001; y---0.14967+ 0.05239xGA) and fransverse (r=0.91; p<0.00001; y----0.12877+ 0.0508xGA) diameters of t~e fetal orbits. A significant correlaSon

was also found between orbital circumference and GA (r=0.945; p<0.00001), BPD (r=0.944; p<0.00001) HC (r=950; p<0.00001), FL (r=0.904; p<0.00001) and outer orbital diameter (r=917; p<0.00001). A linear growth function was also observed between GA and lental diameter (r=0.885; p<0.00001; y=0.45532+0.16847, xGA) and the lental circumference (r=0.91688; p<0.00001). A first degree correlation was also found between lental circumference and FL (r=0.904; p<0.00001; y=5.852+2.05xFL). CONCLUSIONS: These results provide normative data of the fetal orbits and lens in various dimensions and across gestatJonal age. In addition, the data offer the potential for prenatal diagnosis of microlens.

432 FETAL LUNG MATURITY IN DIABETICS: THE PREDICTIVE

VALUE OF BIPARIETAL DIAMETER.

R Samadi,-~, M D, K. Berkowitz, M D. & S Kjos, M.D. University of

Southem California School of Medicine, Los Angeles, CA

OBJECTIVE: To determine the reliability of a fetal BPD in prediction

of fetal lung maturity (FLM) in diabetic patients.

STUDY DESIGN: A cohort study of all patients undergoing

amnlocentesis was performed from 1987-1994 In addition to

ultrasound and demographic data, information regarding maternal, fetal

and obstetrical complications was collected prospectively Ammotic

fluid was assayed for L!S, PI, PE and PG A BPD>9 2 cm was used as a

predictor of FLM Neonatal outcome was obtained by chart review

RESULTS: Of 3173 patients w~th amnlocentesls, 1299 were diabetic.

208 &abetlcs and 60 non-diabetics had a BPD_> 9 2 cm 22 6% (47/208)

of the diabetics and 26 7% (16/60) of the non-diabetics had an L/S<2 0

despite a BPD>9.2 (OR 1 3, CI 0.6-2.5)

DIABETIC NON-DIABETIC

L/S <2.0 _>2.0 <2.0 >2.0

(N=47) (N=I61) (n=16) (N=44)

RDS 0 (0%) 0 (0%) 1 (6.3%) 0 (0%)

TTN 1 (2.1%) 4 (2.5%) 0 (0%) 1 (2.3%)

The rates of mature L/S ratios or occurrence of RDS or TIN between

groups were similar In term infants (>_ 37wks) with BPD>9 2, only 13%

had L/S<2 0, whde 45% of preterm infants with BPD>9 2 had L/S<2.0

(p<.05) The overall rate of RDS in the study population was 0 3%

(11268) and none of the infants born to diabetic mothers exhibited RDS CONCLUSIONS: Prediction of a mature L/S ratio by a BPD>9.2 is as

reliable in diabetics as in non-diabetics Both groups exhibit slgmficant

numbers of patients with an L/S<2.0, but lhls "biochemical" false (+)

rarely correlates with clinical expression of RDS Stratification of false

(+) rates by gestatlonal age reveals significant differences in term versus

preterm gestations At term, diabetic fetuses with BPD_>9 2 do not

experience RDS and should not undergo arrmiocentesls.

428 SPO Abstracts January 1996 Am J Obstet Gyne(ol

433 ASYMMETRIC MACROSOMIA PREDICTS SHOULDER DYSTOCIA IN BORDERLINE MACROSOMIC INFANTS OF DIABETIC MOTHERS

Steehan~e Pen~x, Bruce Cohenx Deborah Ansleyx, Manuel Porto, Joseph Adashek and Thomas Garde University of Cahforn~a, Irv~ne, Orange, Cahforn~a OBJECTIVE’. Anthropomorphic measurements ~n macrosom~c ~nfants of d~abetic mothers suggest a tendency toward hgh truncal d~str~butlon of weight (asymmetric macrosom~a) For th~s reason, shoulder dystoc~a ~s of particular concern ~n the d~abet~c pregnancy and the opbmal route of dehvery of borderline macrosomic ~nfants (3800-4200 gm) ~s controversial The oblect~ve of th~s study ~s to determine ~f the difference between the abdominal d~ameter and b~panetal d~ameter (AD BPD difference), as measured on ultrasound, ~s predictive of shoulder dystocla m borderhne macrosom~c ~nfants of d~abet]c mothers STLIDY DESIGN: A hstoncal cohort study was performed of b~rths occurnng from January, 1990 - June, 1995, using b~rth logs, computer stored ultrasound data and medical records. Ehglb~hty requirements Included 1) Diabetic pregnancy, 2) Ultrasound within 2 weeks of delivery or 500 gm of blrthwelght, 3) Estimated fetal weght between 3800 and 4200 gm, and 4) Vaginal delivery Shoulder dystocla was defined as the need for Wood’s corkscrew maneuver and/or dehvery of the postenor arm to accomphsh dehvery RESULTS: Over the five year study penod, there were 32 pahents who met the inclusion cntena for the study S~gm£cant shoulder dystocia occurred in 6/20 patients (30%) when the AD BPD difference was >= 2 5 cm and in 0/12 patients (0%) when the AD BPD difference was <= 2 5 cm (P< 05, RR= 1 86, Cl = 1 30 - 2 65) Three of the six cases of shoulder dystocla resulted in b]rth Injury (1 transient Erbs palsy and 2 humoral fractures) No significant d~fferences in average blrthwe~ght (3928 vs 3974gm p= 74), maternal weight (174 vs 185 Ibs, p= 58) or gestatlonal age (39 2 vs 39 1 weeks, p= 8) were found between the groups with an AD BPD d~fference of >=2 5cm or <2 5cm, respecWely CONCLUSION: The ultrasound measured AD’BPD difference ~dentd~es those borderline macrosom~c ~nfants of d~abet~c mothers who are at hgh nsk for shoulder dystocla and birth injury

435 SONOGRAPHIC FETAL WEIGHT ESTIMATION: THE INFLUENCE OF RACE-SPECIFIC FORMULAS ON PREDICTIVE ACCURACY J Owen, R, Goldenberg, M. DuBard×, J. Hsu×, Dept of OB/GYN and Biostatistics, University of Alabama at B~rmmgham, B~rm~ngham, AL. OBJECTIMEr Since ethnic d~fferences (black vs white) ~n gestational age specific fetal b~ometry have been w~dely recognized, we sought to determine if regression formulas for estimating fetal weight could be optimized by using racially homogeneous populations for their derivation STUDY DESIGN: 2329 singleton pregnancies (1448 black, 881 white) w~th complete biometnc information were dehvered within 7 days of their sonograph~c evaluation. Regression formulas were derived, first from the collectwe population, and then separately for both black and white gravidas Both first and second order terms and first-order interaction terms were included ~n the model with logic birthwe~ght (BWT) as the dependent variable All three were optimized using the same stepwise se!ection procedure, RESULTS: The mean gestational age at delivery was 38.8 +_ 4 wks (range 20-44 wks) and the mean BWT was 3109 +_ 837 g (range 330- 6390 g) Using the "best-fit" formula for the collective population (r==0.93), the mean predicted-minus-actual BWTs were significantly d~fferent between the black and white groups (p=O.02), Additionally, the best-fit formulas derived from the collecWe population and both sub-populations (r2=093) contained unique sets of biometric parameters, Nevertheless, when the two race-specific formulas were evaluated in their respective populations, and then compared to their performance in the collecWe population formula, there were negligible differences in the mean absolute errors, the mean percent errors and the upper and lower 5~h percentile percent errors. CONCLUSIONS: Race-specific formulas for estimating fetal weight do not improve the clinical pred]cWe accuracy as compared to a formula derived from a racially heterogeneous population,

434 COMPARISON OF FOUR FORMUI.J~S FOR TH~ SONOGRAPHIC

ESTIMATION OF I~TAL WEIGHT. IMP O’Dayx, P Prom~;onex, IA Willkm% Dept of Ob/Cry~ Unlv of Texas Medical School and Baylor

College of Medicine, Houat~ TX.

OBJECTIVE: Two r~w formuL~ tO estin~te fetal weight have been

recently reporl~. Combs et al, de,.~nbed a f~mula based on fetal volume,

..~i.~ AC, HC & FL. F~e~o et ~1. r~lc~l~ed a regre..~i~ fm~mula

AC & FL. The purpose of this study was to test the acctu~y of these

formulas against those of Shcpurd ~d Hadlock.

STUDY DESIGN: Patiemts presenting at greate~ than 24 weeks gestation,

with no fetal anomalies and expected to deliver within 7 days wm’e

p~ospectively tutored. Fet~d n~.asurem~ts recorded were HC, BPD, AC,

~d FL ~d estimated fetal weight (EPNV) w~ calculated using the

Hadlock Shcpurd, Combs and Fe~e~o formulas. Based on established

gu~dalmes, EFWs within 10% of ~tuul birth weight (BW) were deemed

accurate.

RESULTS: Of 280 patients ¢nte~d, 24 deliv=ed l~yond ene week,

leaving 256 ultrasonnds for a~lyms. For all fetuses, 73% of EFWs

wthm 10% of BW u~ing Hadlock% formula; �omlmr~d to 68% with

Shcpard, 69% vath Combs g~d ~1% with Ferre~o (p=0.0Og). I~ ~e

subgroup of fetuses ~ than 200~ grmn~ (e=3~), EFWs using H~tlock’s

formula were a~cura~ 64%, comp~ed to 47% with Shupa~d, 39% with

Combs, and 36% with Ferrm’o (p>0.05). For ~ fetuses inure thin

grants (n=31), I-I#~ock wag m~curate 74% ofth~ time, compared to 74% f~

Shcpurd, 61% for Combs and 93% f~ Ferre~o (p=0.00~).

CONCLUSION: Across aft weight ranges, the a~curacy of fetal weight estimation is not different using the Hadlock, Shepard m-M Combs fe~nula~

However, these fennulas s~e mgnificantly mo~e accurate than Ferrero’$ for

fetuses < 4000 grams and less a~curate at >4000 grams

436 FETAL ABDOmINaL FAT LINE: MACROSOMIA BE DIAGNOSED? B. Petrikovsky, N. Gelertner, C. Oleschuk North Shore University Hospital Manhasset, New York OBJECTIVE: To determine the usefulness of fetal abdominal fat line (FAFL) for diagnosing macrosomia. STUD~ DESIGN: FAFL was measured in 133 fetuses, who delivered within 78 hours after the assessment. FAFL was measured from the outer to the inner anterior edge of the hyperechoic rim on the level of abdominal circumference. Macrosomia was defined as infants weight > 4000 g. Twenty fetuses were macrosomic. Results: Mean FAFL was 8.4 + 2.7 mm. FAFL correlated significantly with birth weight (R = 0.67,P < 0.001). Using cut point of 9 mm, specificity of FAFL to diagnose macrosomia was 66%, sensitivity 85%, negative predictive value 96% and positive predictive value 30%. CONCLUSIONS: Abdominal fat line thickness correlates with birth weight. It can be used as an adjunct to EFW to diagnose fetal macrosomia. In particular it can readily identify fetuses expected to have normal birth weight.

Poster Session IV


Kohala Ballroom

CATEGORIES OB Anesthesia/Pharmacology

Genetics and Teratology Hypertensive Disease In Pregnancy

Doppler


Judges: Ronald J. Wapner, MD Norman E Gant, MD Kathryn L. Reed, MD

Vohrnre 174, Numbe~ 1, Pa~t 2 SPO Abstracts 431 Am J Obstet Gynecol

437 I HE EFFECTS OF BUPIVACAINE AND INIIIBITION OF NITRIC

OXIDE IN THE ASPIIYXIATED IMMATURE FETAL IAMB

P Bobby~, MY Dwon, E Yun~, A Santos~ Depts of OB/GYN and

Anesthesiology, Albert Einstein College of Medxclne, Bronx, NY OB3ECTIVE: Fetal cardiovascular adaptatmns to asphyxia increase blood flow

to vital organs The purpose of thxs study was to determine the effects of

buplvacame(BUP) and mtrtc oxide synthase mhlbatlon wxth N-mtro-L-argmme

methyl ester(L-NAME) on regional blood flow and acid-base status m the

~mmature asphymated fetal lamb

METHODS: Chromcally instrumented pregnant ewes were studied at 118-119

days of gestahon Al]~er a control period, fetal asphyma was reduced by parhal

umbilical cord occlusmn After 90 minutes of stable asphyma(ASP) the effects of

BUP(n-13), L-NAME(n 12) and phenylephnne(PltE, n-7) on fetal heart rate,

mean arterml pressure, am&base status(arterial pH, PCO~, PC=) and regional

blood flows(ml/mm*g) to bram(BRN), myocar&um(MYO) and adrenal(ADR)

were studied Data were evaluated by ANOVA RESULTS. mean±SE, *p<0 05

AS._._~P BUmP AS~P L-NAME ASP PHE HR 175±9 172±5 186~*14 169±9 196:e4 189±14 MAP 52±2 50±2 51±2 60±2* 53±2 65±2* pH 728±02 733±03 728±03 730±03 733:=01 732±01

PCOz 52~:1 51±2 50±2 47±1 48:~1 46~1

PC2 15=05 16:~1 15-*05 18±1" 14±1 19±05* BRN 23:~04 12±02" 22±04 15±02" 30±08 11±05" MYO 45±05 29±04* 53±12 40±06* 103±40 57==12" ADR 49±08 54:=09 46~09 23±03* 84~23 32±03*

CONCLUSIONS: Exposure to chmcally relevant plasma concentrations of BUP

&d not worsen car&ovascular or acid-base status m the ~mmature asphyxiated fetal lamb but &d reduce blood flow to the BRN and MYO Reductions in organ

blood flow were also noted with L-NAME and PHE However, admtmstratmn of L-NAME and PHE dunng asphyma were associated w~th further increases m fetal

MAP and an improvement m fetal PO~

439 PREVENTION OF MATERNAL HYPOTENSION BY EPIDURAL

ADMINISTRATION OF EPHEDRINE SULFATE DURING LUMBAR

EPIDURAL ANESTHESIA FOR CESAREAN SECTION.

J. Fong, E, Gurewltsf~b~h× M Gomflhon, R. Press and L Volpe.

Departments of Anesthesiology and Obstetncs and Gynecology,

New York Hosp~taI-Cornell Medical Center, New York, NY

OBJECTIVE: To determine whether epldural adm~mstrat~on of

ephednne sulfate s~multaneous[y w~th induction of lumbar eDdural

anesthesia for non-emergency cesarean section reduces the

modence of maternal hypotens~on.

STUDY DESIGN: In a double-bhnded, placebo-controlled trial, 40

normotenswe, nonJabonng, American SoQety of Anesthes~olo#sts’

Class I or II women w~th term, uncomphcated, singleton

pregnancies were randomized to have either normal saline or

ephedrine sulfate admlmstered eDdurally and colncldentally with

mduct~on of lumbar epldural anesthesia for non-emergency cesarean

section All sublects were prehydrated with 20cc/kg crystallold and

placed in supine position on a right-sided wedge both before and

after induction. Serial blood pressures taken from time of ~nduchon

to debvery of the infant were compared to a basehne blood

pressure, chosen as the average of three pre-mductlon blood

pressures Hypotenslon was defined as a decline In systolic blood

pressure to _<90mmHg or _<70 percent of basehne Differences

between groups were analyzed by chl-square analys~s.

RESULTS: The Inodence of hypotenslon in the placebo group was

36 84%, which was not significantly different from the ~nodence of

38 10% in the treated group

CONCLUSION: Prophylactic eDdural adm~mstratlon of ephedrine

sulfate does not reduce the ~ncldence of maternal hypotenslon

following lumbar ep~dural anesthesia for non-emergency cesarean

section.

438 THE EFFECTS OF NARCOTIC AND NON-NARCOTIC CONTINUOUS EPIDURAL ANESTHESIA ON INTRAPARTUM FETAL HEART RATE TRACINGS AS MEASURED BY COMPUTER ANALYSIS. _(3 H0ffman Ill×, E Guzman, M Richardsonx, A V~ntzlleos, C Houhhanx, C Benlto× Depts of Maternal-Fetal Medicine and Anesthesiology, UMDNJ-Robert Wood Johnson Medical School, St Peter’s Medical Center, New Brunswick, N J OBJECTIVE: To evaluate ~e effec~ of narcotic and non-narcotm continuous epldoral anesthesia on ~ntrapartum fetal heart rate tracings as measured by computer analys~s STUDY DESIGN: We studied 28 women with uncomplmated pregnancies at termwlth reactive fetal heart rate (FHR) tracings in ear[y labor, One hour FHR tracings underwent on-hne computer analysis using the System 8000 software The software measured FHR baseline, number of accelerations of 10 and 15 beats per minute, eDsodes of low and high variation In minutes, and short and long term variation in mllhseconds The women were randomized, using a computer generated random number table, to receive ep~dural anesthesia with either buplwcalne with fentanyl or buplvlcalne alone The narcotic epldural consisted of 10-12 cc of 0 125% bup~wcalne with 50pg of fentanyl, followed by a continuous infusion of a solution composed of 60 cc of 0 125% buDvlcalne with 100pg of fentanyl at a rate of 8-12 cc per hour The non-narcotic ep~dural consisted of an ~n~tlal bolus of 10-12 cc of 025% buDwca~ne followed by an infusion of 0 125% buDwcalne at a rate of 8-12 cc per hour A repeat computer analysis of a one hour FHR tracing was obtained 30 minutes after the initial bolus of anesthetic agent A power calculation using a power of 90% and an alpha error of 0 05 revealed that 28 patients would be needed to detect a change from the 50th percant~le to the 5th percentde for long and short term vanat~on and number of accalerat~ons of 10 and 15 beats per m~nute Median values were compared using the Mann-Whitney U and Wdcoxon signed rank tests, and Chl-square analysis Statistical s~gnlficance was set at p <0 05 RESULTS: There was no difference ~n maternal weight, duration of rupture of membranes on admission, cervical d~latatlon at pre- and post -epldural FHR recording, time Interval between FHR recordings, gestat]onal age, birth weight, umbdlcal artery pH at b~dh, and Apgar scores at 1 and 5 m~nutes between the two study groups In the eady first stage of labor, there was no significant dlfferance in the pre-and post-epldural anesthesia FHR basehne, number of accelerations of 10 and 15 beats per minute, eDsodes of Ngh and low vanabon, and short and long term vanatlon when using e~ther narcotic or non-narcotic anesthetic agents CONCLUSIONS: The chnlclan can consider the use of narcotics as well as non- narcotic continuous eDdural anesthesia in the dosages used in our study with its attendant advantages without fear of obscunng the ~ntrapartum FHRT tracing

440 DOES STATION AT EPIDURAL PLACEMENT AFFECT THE INCIDENCE OF MALPOSITION OF THE FETAL VERTEX AT DELIVERY? Hosp of the Univ. of PA, Phila, PA. C A.

Robinson., N. Rnthx, G.A. Macones, M.A. Morgan. OBJECTIVE: To determine if high station of the fetal vertex at the time of epidural placement increases the incidence of occiput transverse or occiput posterior position at delwery. STUDY DESIGN: We performed a retrospective cohort study to ascertain the relationship between station at epidural placement and malposgion. One hundred forty charts were reviewed and 123 records of patients in labor (either spontaneous, induced, or augmented) who received eDdural anesthesm were included for analysis. Patients with a contraindication to labor, antepartum fetal demise, or twins were excluded. Patient demographics, initial cervical exam, epidural medication characteristics, and labor and delivery data were abstracted from me&cat records. Station was characterized as "high" (>-1) or "low" (5=1). RESULTS: The incidence of occiput posterior and transverse posmons at dehvery were significantly increased in patients who recmved epldural anesthesia when the fetal vertex was at high station. There were no differences in maternal demographics between the two groups.

*Relative risk=2.33(95%CI 1.16-4.7)

CONCLUSIONS: Epidural placement before engagement of the fetal head is associated with an increased incidence of occiput posterior posmon. This may provide an explanation for the empiric observation that operative delivery is variably increased in patients laboring under epldural anesthesia.

432 SPO Abstracts JanuatT 1996 Am J Obstet Gynecol

441 CORRELATION OF URINE AND SERUM BENZOYLECGONINE

LEVELS IN PREGNANT WOMEN. M, Monqa, S. Wong~, K.D.

Larrabee’, R. L Andres. Dept. Ob.Gyn. & Reprod.Sci. & Pathology,

UTMSH, Houston, TX,

Objective: To deten~ine ~ae correlation between udne and serum

banzoylecgonlne levels in pregnant women using cocaine.

Study Design: Women receiving care in a prenatal substance abuse

clinic were recruited for this study. Urine and serum were obtained

concurrently for quantitative determination of benzoylecgonine (BE) (the

major metabolite of cocaine) using mass spectrometry. Pearson

correlation was used for statistical analysis.

Results: 10 women who had recently smoked crack cocaine were

recruRed to ~{s study. The median reported intental since most recent

crack use was 35 hours (range, 9.5-63 hours). Urine and serum BE

levels are shown below.

Urine BE (ng/ml) Serum BE (ng/ml)

1 1038 27.6 2 1062 13.9 3 1150 10.9

4 2827 2.5 5 8083 10.1 6 9333 75.4 7 22728 92,9

8 76621 435.2

9 112470 1445.6 10 1950563 3074.3 Pearson correlation for urine and serum BE levels = 0.92 (p<O.O01).

Conclusions: This is the first study to report the con’elation between

udne and serum benzoylecgonine levels in pregnant women who use

crack cocaine. This information will be useful for future in vitro and

animal studies of the effect of cocaine on pregnancy.

443 THE EFFECT OF GLUCAGON ON SPONTANEOUS

CONTRACTILITY OF ISOLATED HUMAN UTERINE

MUSCLE Y. Shinx, J. Collea, Y. Kimx. Depts. Ob/Gyn/Anesthesia,

Georgetown Univ., Washington, D.C.

OBJECTIVE: Intravenous glucagon administration has been effective

in the treatment of gastrointestinal and uterotuhal spasms during

endoscopic or radiologic procedures. With the relaxant action of

smooth muscle and antiarrhythmic effect on the heart, glucagon could

play a role as a potential tocolytic agent.

STUDY DESIGN: A segment of the myometrium was excised from

six parturients undergoing elective cesarean section. After establishing

rhythmic spontaneous contractions in a tissue bath containing Krebs

solution aerated with oxygen, glucagon (Eli Lilly) reconstituted with

the accompanying diluent (glycerin 1.6% with 0.2% phenol) was

added to the bath in a cumulative manner resulting in the final

concentrations of 5.7x10~ M and 1.1xlOs M, respectively. In the

second phase of the study, glucagon reconstituted with water or the

diluent alone was added to the bath. Inhibition of contractility was

determined by a percentage decrease in the contraction amplitude.

Data were analyzed by analysis of variance and a paired t-test.

RESULTS: Glucagon, when reconstituted with the accompanying

diluent, suppressed the amplitude by 24_+9%(mean_+SD, p>0.1) and

65+_14%(p<0.01), respectively. However, glucagon reconstituted with

water failed m suppress the contractions. The diluent alone produced inhibition (25_+14, p>0.1; 70_+15%, p<0.01) similar to that observed

with glucagon reconstituted with the diluent.

CONCLUSION: Our results suggest that the inhibition of uterine

contractions was likely caused by the diluent instead of glucagon. We

conclude that glucagon has no relaxant effect on spontaneous uterine

.contractions in an isolated preparation.

442 THE INFLUENCE OF EPIDURAL ANESTHESIA ON CESAREAN SECTION RATES: A PROSPECTIVE, RANDOMIZED, CLINICAL TRIAL. A Clark, D Barry,* G Loyd,* V Cook, J Spinnato. Depts. of OB/GYN and Anesthesia, Univ. of Louisville, Louisville, KY, 40292. OBJECTIVE: To determine the effect of narcotic ap~dural anesthesia on labor and delivery managed according to the principles of active management of labor (AML). STUDY DESIGN: From January 1995 to August 1995, 145 spontaneously laboring, term, nullip~rous patients were randomized to receive either intravenous (IV) narcotics or narcotic epidural anesthesia for pain ralieL Analgesia was given upon patient request. Labor was managed according to the principles of AML Forceps and C/S deliveries were performed for obstetrical indications. Data analysis was conducted on an intentmn-to-treat basis. The Mann-Whitney U, x2 with Yates continuity correctmn, and t- tests were employed for statistical analysis when appropriate. RESULTS: 80 patients were randomized to IV narcotics and 65 patients to epidural anesthesia. Maternal age, gestational age, dilo/.ation on admission and infant birthwelght did not differ significantly between the two groups. Outcomes are presented m the table below.

Ep~dural Narcottc Variable (n=65) (n=80) P Value Dilatatmn @ 1st dose(cm) 4.31 + 1.8 402__+ 1.6 NS Pttocin reqmred #(%) 49 (~.4) 57 (71.3) NS 5 mtn APGAR 8.7 + 0.5 8 6 + 0.7 NS Venous pH 7 29 __+ 0 5 7 28 +__ 0.5 NS Venous base excess (meq/L) -3.3 __+ 3.1 -4.2 __+ 2.6 NS Active phase labor (min) 278 __+ 146 274 __+ 132 NS Stage II labor (rain) 56.7 + 39.4 63.4 __+ 52.0 NS Cesarean delivery #(%) 7 (10.8) 12 (15.0) NS (Data are reported as mean __+ SD or proportmns.) Dystocm requiring C/S occurred in 3 and 9 cases in the ep~dural and narcotic groups, respecuvely. 34 patients crossed from IV narcotics to epidural anesthesia with 8 requiring C/S for dystocia. All 8 patients demonstrated protracted labor before epidural placement Of the 65 patients imtially randomized to epidural, 35 were placed ~< 4 cm and 30 were placed at ~> 5 cm. There were 4 and 3 C/S in each group, respectively (NS, p=.834). Study enrollment continues. CONCLUSIONS: Epidural anesthesia does not increase the length of labor or need for cesarean delivery when AML ~s utilized Delay of epldural placement until an arbitrary degree of cerwcal dilatation does not appear warranted

444 CENTRAL HEMOOYNAMIC EFFECTS OF RECOMBINANT HUMAN RELAXlN IN THE ISOLATED, PERFUSED RAT HEART. C. Coulson, J M Thorp, Jr., D. C. Mayer’, R. C. Cefalo, B Eucker*. Dept. Ob/Gyn, Univ. of NC, Chapel Hill, NC. OSJECTIVE: To determine the cardiac effects of rek~xin in the isolated, p~fused rat head model and to see ~f pregnancy modems the hormone’e actions. STUDY DESIGN: Heads were exc~sed from 18 female Spregue- Dawley rats (10 pregnant, 8 non-pregnant) and attached to a Langendorff apparatus. Left ventricular systolic pressure, heart rate, and contractility were measured. Heads were exposed senally to 0.5, 1.0, 2.0, 4.0, 8.0, and 16 0 ng/ml concentrations of recombinant human relaxin RESULTS: Hearts from pregnant rats had lower heart rates than nonpregnant heads. Relaxin increased heart rate, left ventricular systolic pressure, and contrac~ in a do--dent fashion. Pregnancy did not modify this response CONCLUSIONS: Recombinant human relaxin is a potent inotropiC and chronotnpic agent. Thase propertJas cause one to wonder If the hormone may be involved in the cardmvascular changes seen in sedy pregnancy.

Vohlme 174, Number l, Part 2 SPO Abstracts 433 Am J Obstet Gynecol

445 THE 8-15 WKS GESTATIONAL AGE RELATED SONOGRAPHIC

PHENOTYPE OF 33~, ANEUPLOIDIES. S. Rottemx, M Bronshtem, D

Nyberg, J. van Vugg M Cullen, K. Nlcolmdes, SUNY-HSC at Brooklyn,

NY and the Internahonal Registry of the Onset of Fetal Anomahes.

OBJFCTIVE: To define the 8-15 wks sonomarkers of aneuploldles.

STUDY DESIGN: From a mult~ center data base of 1014 fetuses w~th an

abnormal sonogram be~veen 8-15 wks., we report those w~th aneuplo~dy

(n=334)

RESULTS: The d~stnbuhon of sonomarkers among 334 patients w~th

abnormal scans and aneuplo~dles ~s shown below

Genetic i I 10-11 I 12-13 I 14 <16

Disorder 8-9 Weegs

Weeks Weeks Weeks

21st/7NI~I 10st / 4cv

Trisomy 21 5~ 1cv / 1sr~ 3~1

Trisomy 18

Turner Syndrome

Trisomy 13

NF=Neck/Face NIH=Non-lmmune Hydrops, etc,; CV=Cardiovascular;

UT=Urinary; Gl=Gastro; SKL=Skeletal; CNS=Central Nervous System

CONCLUSIONS: The early sonograpNc phenotype and ~ts vanat~ons w~th

gestat~onal age should be consldered m genetlc counseling.

447 Assessment of Facial and Intracranial Biomet~" Using Prenatal Ultrasound in Identifying Fetal Alcohol Syndrome/Effects; A Pdot Study. W H. Persutte×, Tara Wass, John Hobbins, Department of Obstetrics and Gynecology, University of Colorado Health Sciences Center, and University of Denver, Denver, Colorado OBJECTIVE A strong theoretic basis exists to support the use of prenatal ultrasonography as a screening tool to identify fetuses who marafest the shgmata ofFetalAlcoholgyndrome/Effects (FAS/E) in women with a history of alcohol use We attempted to test this theory. METHODS AND MATERIALS Between January 1, 1992 and January 1, 1995, patients in our obstetrical clinic were administered a screening questionnaire (T-ACE) to assess alcohol use. ~Ilae results of the ultrasonographic measurement of (1) the fetal frontal lobe of the cerebrum (the frontal calvanum to each of the posterior thalami, the posterior cavum septum pellucldum, and the anterior cavtun), (2) the transcerebellar diameter, (3) the onter and nmer orbital diameters, (4) the mid-facial length, and (5) the phdtrnm, where compared between patients identified to be at risk for as FAS/E and controls RESULT During the study period, we performed 396 detailed ultrasound examinations on 229 patients Three hundred thirty four ultrasonographlc exam~nahons were conducted on 167 pahents at r~sk for FAS/E and these data were compared with that from 62 patients who abstained from using alcohol during the pregnancy and these served as controls Of the 167 patients at risk for FAS/E, 127 (77%) were mild to moderate alcohol users and 40 (23%) were heavy users Gestatmnal age specific nmnograms of the mtracranial and facial bmmetry were created. We found no significant ultrasonographlcal/y detectable differences in the facial and mtracranlal biometry between patients w~th moderate to severe alcohol use compared w~th controls CONCLUSION Although lhrther study limited to women wuth heavy, early, and/or binge use should be completed, we found no significant correlatmn between alcohol use and ultrasonographlcally detectable

blometrlC manifestations We are contnming the pursuit of an ongoing study of heavy alcohol users

446 THE USE OF THE FEMUR+HUMERUS/FOOT RATIO TO DETECT FETAL ANEUPLOIDY. A Johnson. B Endicottx, V Twayx, K Mollx" J

Kosterx, K Teidekenx Dept OB/Gyn, Jefferson Medical College,

Philadelphia PA & Div MFM Mercer Medical Center, Trenton NJ

OBJECTIVE: Johnson MP, et al prevmusly suggested that the raUo (femur

+ humeral)/foot may be an effective ultrasonic marker for fetuses affected

with trlsomy 21 (AJOG 172 1229,1995) The purpose of this study was to

further evaluate the utility of this raUo In the detection of tnsomy 21 as well

as other potentially viable fetal aneuploldles

STUDY DESIGN: Over a 16 month period a prospective seines of 2,081

consecutive fetuses were scanned at 14-24 weeks gestation Blometrlc

parameters obtained included the blpanetal diameter (BPD), abdonunal

circumference (AC), femur length (FL), humeral length (HL), and foot length (FT) Reference curves foi these parameters were established from

the analysis of 144 normal contxols Data was analyzed with unpaired T

test and Wllcoxon test for comparison of independent groups

RESULTS: There were 11 (0 5%) cytogenetlc abnormallhes detected, 4 trl

21, 3 tr1-8, 2-45X, and 2-other The mean, +/SD, (FL+HL)/FF ratios in the

control and abnormal cases were 1 989, +/- 0 165, and 1 955, +/- 0 245,

respectively Isolating on the tn-21 cases the mean, +/-SD, ratio was 1 810,

+/-0 127 When compared to controls this difference was staUstlcally

significant, p=0 03 Using a threshold value of _< 1 75 for the ratio

sensitivity, specificity, and positive predictive values for fetal aneuploxdy

were 36 4%, 94 3% and 33 3% respectively At this cut off, 3 of the 4 In-21

affected cases were screen positive However in these 3 cases there were

addxtlonal sonographlc findings suggeshve of an increased risk of fetal

aneuploidy

CONCLUSIONS: A (FL+HL)/FT ratio <__1 75 in the second tinmester is

useful In the detection of trl-21, but not other potentially v~able fetal

aneuploldies However, further testing will be necessary to determine the

significance of this sonographlc marker as an isolated finding

448 FETAL CHOROID PLEXUS CYSTS: AN ASSOCIATION

WITH TRISOMY 21 ? RJ Gratton. WA Hogge, CE AstonX, N

Lazebnlk Dept of Ob/Gyn, Umverslty of Pittsburgh School of

Medicine, and Dept of Human Genehcs. Graduate School of Public

Health, Umverslty of P~ttsburgh, Pittsburgh, PA

OBJECTIVE: The purpose of this study was to determine the incidence

of choro~d plexus cysts m fetuses w~th tnsomy 21 and to calculate the

age related risk of trlsomy 21 ~n the presence of an isolated choroid

plexus cyst

STUDY DESIGN: The second mmester (14 7-22 3 wks) sonograms

of 53 consecutive fetuses ~dentlfied prenatally w~th tnsomy 21 from

January 1990 to July 1995 were rewewed The indications for prenatal

d~agnos~s were advanced maternal age (24), sonographic abnormalities

(11), advanced maternal age and anomahes (6), and abnormal multiple

marker screen (12) All ultrasounds were performed at one institution and

were rewewed ~ndependently of the initial report R~sk estimates based

on the age related risk for tnsomy 21 and cond~tmnal variables

including the observed incidence of chorold plexus cysts in fetuses with

msomy 21, the frequency of assooated anomahes and multiple marker

screening for trlsomy 21 were calculated

RESULTS: The mean gestatmnal age (+SD) at ultrasound assessment

was 17.1 (+1 7) weeks Overall 17 of 53 (32.1%) of the fetuses with

tnsomy 2l had ~dentff]ed sonograph~c abnormalities. Chormd plexus

cysts were ~dent~fied in 2 (3 8%) of the trisomy 21 fetuses. Both of these

pregnancies were comphcated by advanced maternal age and both fetuses

had addltmnal abnormaht~es suggestive of a karotyplc abnormahty The

calculated risk of tnsomy 21 w~th an isolated choroid plexus cyst and

normal multiple marker screen does not approach that of amnlocentesls

until a maternal age of 35 or greater

CONCLUSIONS: The incidence of chormd plexus cysts m fetuses

wath trlsomy 21 was shghtly h~gher than that in the general obstetrical

population (1-2%) However, the risk of trisomy 21 in the presence of

an isolated chormd plexus cyst and a normal multlple marker screen,

was s~mdar to the age related risk

434 SPO Abstracts january 1996 Am J Obstet Gynecol

449 INTRAUTERINE FETAL GROWTH IN FETUSES WITH BECKWITH-WIEDEMANN (EXOMPHALOS-MACROGLOSSIA-GIGANTtSM) SYNDROME, A C Renzlm, D Day-

Salvatore~,T Turnerx, JC Smullan, A Vintzlleos Divmons of Maternal-Fetal Medicine and Human Genetics, UMDNJ-Robert Wood Johnson Medical School/St Peter’s Medical Center, New Brunswick, NJ OBJECTIVE. Intrauterine growth assessment of fetuses w@l Beckw~th-Wledemann syndrome(BWS) has previously been confined to isolated case reports We descnbe the intrautenne growth of a senes of fetuses diagnosed with (BWS) STUDY DESIGN Infants postnatally confirmed to have BWS were identified from records in the Division of Human Genetics Antenatal ultrasound and bdh records were evaluated EFW percenNes were based on the nomogram of Hadlock HC and brthwelght (Wt) data measured after dehvery were converted to percentiles based on the work of Lubchenko and Arbuckle respe@vely Polyhydramnlos (poly) was noted if the ammotlc fired index exceeded the 95th percenhle for gestatlonal age RESULTS: Seven fetuses wdh BWS who had antenatal ultrasound examinations were identdled Two fetuses were suspected to have BWS in utero (#3,4) S~gndlcant ultrasound hndlngs included omphalocele (crop) (#1,2,3) and enlarged liver and kidneys (#4) Fetal tongue protrusion on ultrasound was not ideetlfled in any fetus All 6ve fetuses with ultrasound examinations between 25-36 weeks gestation had poly The percenhles for ultrasound measurements of HC, AC and EFW at each of several gestatlonal ages and the HC and Wt percenNes at birth are shown below

HC/AC/EFW Birth (36-41 weeks) 18-25 wks 25-30 wks 30-36 wks HC Wt

(percentiles) (percentiles) 1’ 25/50/50 25/50/66 25/80/61’* >90 >99 2* !0/25/39 10/25/63"’90/5@70"* -- >99 3* 75/ /50 9@95/95 9@90180" >90 95 4 75/75/50 90/95199"* 90/95/80"* >90 >97 5 50 -- -- >90 >99 6 50/75/50 -- -- 80 >90

- 90/95/889"* -- *omphalocele **polyhydrammos 9mtrautenne demise at 27 weeks CONCLUSIONS: Ompwas present in only 3/7 (45%) patients Poly was identgled in 5/7 (70%) of fetuses wdh BWS Fetuses with BWS may begin to exbbd accelerated growth as early as 25-30 weeks gestation but become macrosomlc (>90%) after 36 weeks Fetuses with omph, poly and an AC between 25-95% may st~ll have BWS Poly and progressive macrosomla begmmng between 25-36 weeks gestation even wdhout omp should alert the physician to the posslblgy of BWS so hypoglycemia may be avoided at brth

451 SCREENING. T. Hallahanx2, J. Larsen2, P. Buchanan3, D. KrantzI, J. MacrlI. INTD Laboratories, Inc., Huntlng[on Station, NY, 2George Washlngton University Medical Center, Washington, DC, 3GeneCare Medical Genetics Center, Chapel H±II, NC. OBJZCT1-V~: To determine if second trimester serologic levels of dlmerlc inhlbln-A (DIA) are elevated in Down syndrome (DS). STUDY DZSIGN: DIA, free Beta hCG (FB) and AFP in maternal serum samples from 19 cases of DS, each matched with 3 unaffected controls for gestational age in menstrual weeks (GA), maternal age and length of freezer storage tlme were evaluated. Multiples of the GA-spec!fic medians (MoM’s) were determlned for each analyte. Detection efficiency and false positive rates were estimated by modeling observed likelihood ratios with the age distribution of live births. RESULTS: The GA-speciflc medians for DIA were:

Controls (pg/ml) 78 137 211 156 -- 149 233 Controls(Number) 3 12 24 9 0 6 3 DS (pg/m!) 451 297 334 402 -- 420 363 DS (Number) I 4 8 3 O 2 i

~/1 insignificant association of DIA wlth GA was observed (r 0.216, P .107). Overall medlan MoM’s of DS cases for DIA, FB and AFP were 2.14, 2.89 and 0.79, respectively. Correlation of DIA with FB was 0.344 in DS cases and 0.349 in controls. Correlation of DIA wlth AFP was 0.425 in DS cases and 0.307 in controls. At a 5% false posltlve rate, lndlvldual detection efficiency for DIA, FB, and AFP was 44%, 64%, and 31%, respectively . CONCLUSIONS: DIA is elevated in cases of DS. Additlonal data will determlne if DIA, when colr~ined with the current second trimester protocol of free beta hCG and AFP, wlll enhance detection of DS.

450 FIRST TRI~STER SEROLOGIC SCREENING: ADDITIONAL DETECTION OF TRISOM~ES 18 ~ 13. K~k~[ZAiLixI,B. Brm~atl2,L Tulul2, T. Hallahanl,p. Buchanan3,j. Larsen4,j. MacrlI. INTD Laboratories, Inc., Huntington Station, NY, 2First Institute of OBiGYN, Univ. of Milan, Italy, 3GeneCare Medical Genetics Center, Chapel Hmll, NC, 4George Washington Univ. Ned. Ctr., Wash±ngton, DC. OBJECTIONS: To determine if Trlsomy 18 (T-18) and Trlsomy 13 (T-13) can be detected within the scope of a first trimester Do~n s~zdrome (DS) screening protocol using free Beta hCG (FB) and PAPP-A. STUDY DESIGN: Flrst trimester samples from 1355 unaffected, 59 DS, 9 T-18, and 3 T-13 cases were evaluated retrospectively. Day-speclflc multiples of the median (MOM) were calculated for analytes. DS detectlon efficiency was determined by mo@eling observed likelihood ratios with the U.S. live b±rth age distribution. T-18 and T-13 detection and false positive rates were determined by atlrpical combinations of leve]s of both analytes. ~1 atyp~callty index based on Mahalanobls’ Squared Distance of >9.21 with each analyte MOM value <2.0 was used as the decision crlterla. Results: In DS, the median FB and PAPB-A HOM was 1.8 and 0.5 respecrlvely. At a 5% false posztive rate, detection efficiency for DS was 60%. In T-18, medlan FB and PAPP-A MOM was 0.12 and 0.5 respectively. A trlsomy 18 detectlon efficiency of 89% (8 of 9 cases) at a 0.4% false positive rate was achieved. In T-13, medlan FB and PAPP-A MOM was 0.19 and 0.67 respectively. One case of T-13 was detectable. Coneluszons: First trzmester screening with FB and PAPP-A may detect a medically slgnlficant number of cases of DS and T-18. The value of thls method for T-13 remains uncertain.

452 THE RISK OF PREGNANCY LOSS IS HIGHER ( 2.5 FOLD) WITH EARLY AMNIOCENTESIS AND CVS THAN WITH

MIDTRIMESTER AMNIOCENTESIS L. GodmflowL J E Tolosa, M.C Le~va", S. Weiner, R.J. Lthnzzi Pennsylvama HospKal/Thomas

Jefferson Universxty, Umversity of Pennsylvama School of Methcme,

Philadelphia, PA and the Perinatology Research Branch (NICHD), Bethesda, MD.

OBJECTIVE: To compare the rate of pregnancy loss associated wxth

early ammocentcsxs (EA) to that of chonomc villous sampling (CVS) and

madmmester amniocentesis (MTA) performed for geneUc mthcations

STUDY DESIGN: A cohort study of women undergoing prenatal

diagnoms from 1986 to 1994 was designed. Transcervical (TC-CVS; n=

2123), or transabdommal chorioinc villus sampling (TAB-CVS, n = 318)

were performed between 10 and 13 weeks gestational age (GA), early

ammocentesis (n=1895)between 11 and 14 GA, complete follow-up was

accomphshed m > 97% of parhc~pants. The rate of pregnancy loss for

each procedure TC-CVS=3 95%, TAB-CVS=3.I4% and EA:3.3%

was compared to the m~e of pregnancy loss for the group of patients who

underwent MTA (n=2882) from 15 to 20 weeks GA, between 1992 and

1994 and who had 91% complete follow-up. Patients undergoing

ammoeentems because of elevated maternal serum alphafetoprotem, or

abnormal ultrasound findings were excluded, as were twin pregnancies.

Relative risk (R R.) and 95% C I. were calculated for fetal loss after the

procedure up to 28 weeks GA

RESULTS: 1) The rate of pregnancy loss was significantly h~gher m

women who had an EA than m those having a MTA [ RR. 252

(1 69,3.76), p <0 001] 2) The lower the GA at wl~ch EA was loss (X for trend:26.6, p < performed, the higher the risk of pregnancy z

0 0001) 3) No difference m the rate of pregnancy wastage was found

between EA and CVS, [R.R. 0.86 (0.63, 1 18), p=.35] CONCLUSION: 1)Early ammocentesis is associated w~th a 2 5 fold l~gher risk of pregnancy loss than midmmester amniocentes~s 2) There

~s no difference m risk for pregnancy loss between and EA and CVS.

Voinme 174, Number 1, Part 2 SPO Abstracts 435 AmJ Obstet G},necol

453 AN ELEVATED MIDTRIMESTER AMNIOTIC FLUID

INTERLEUKIN-6 SOLUBLE RECEPTOR LEVEL IS A RISK

FACTOR FOR SUBSEQUENT FETAL DEATH. _A Donnenfeld. J.E. Tolosa, J Cohen~, R Gomez~, F Ghezzlx, V. Suryx, M. MazoP,

R Romero, P Sehgalx Departments of OB/GYN~ Pennsylvama

Hospital/Thomas Jefferson Umversity, Philadelphia, PA, Pennatology

Research Branch, NICHD, Bethesda, MD and New York Medical College, Valhalla, NY

BACKGROUND!OBJECTIVE: Sub-clinical lntrautenne mflammatmn has been recently identified as an important cause of pregnancy loss after

midtrimester amm~centesls. The rapid and prospective identification of

patients with tkus pathology has important chmcal and therapeutic

lmplrcatlons The objective of ttus study was to deternnne If elevated

ammotic fluid (AF) concentrations of the soluble interleukin-6 receptor

(slL-6R) can identify these patxents

STUDY DESIGN: A case-control study was designed by ~dent~fy~ng 12

patients with fetal demises and 52 controls who had a normal pregnancy outcome. Patients in both groups had undergone m~dtrimester genetic

amniocentesis. Those with anatomical congenital anomahes, aneuploldy

or vaginal bleeding were excluded slL-6R in AF was measured by

ELISA

RESULTS: An AF slL-6R ~ 716 pg/ml was associated with an odds ratio of 5.7 (95% comfidence interval’ 1 18-30 64; p< 0.03) for fetal death.

The mechan AF slL~R was significantly lugher m patients w~th pregnancy

loss than in those with normal pregnancy outcome (median 836 pg/ml,

range 320-2042 vs. median 533.5 pg/ml, range: 136-1551, p <0.05) An AF slL-6R concentration ~ 716 pg/ml had a sensitlwty of 75 % and a

specificity of 65 % for fetal death

CONCLUSIONS: Patients with an elevated mtdtrtmester AF level of

slL-6R are at increased risk of subsequent fetal death.

455 ABSENCE OF LEFT VERSUS RIGHT UMBILICAL

ARTERY DOES NOT PREDICT ASSOCIATED

CHROMOSOMAL ANOMALIES. Cosmas JM van de Ven,

Clark E Nugent. Richard A Bowermanx, Mason Barr Jrx. Dept

Ob/Gyn and Radiology, University of Michigan, Ann Arbor, MI.

BACKGROUND: Smgle umbilical artery (SUA) ~s found in 1%

of all deliveries. When SUA ts an isolated finding, perinatal

outcome does not seem to be affected. However, SUA is

associated with other congenital and chromosomal anomalies. At

the 1994 SPO meeting, data were presented (abstract # 10)

suggesting that "cytogenetic and complex fetal abnormalities

were found exclusively with absence of the left umbthcal artery"

STUDY DESIGN: Autopsy reports of 174 infants with SUA

were retrospectively reviewed to assess whether a correlation

exists between the absence of the left or right umbilical artery

(UA) and abnormal karyotype or phenotype. Karyotype was

assessed by cytogenetics and phenotype through detailed autopsy,

by one pathologist (author MB).

RESULTS: absent right UA absent left UA

normal karyotype 27 21

abnormal karyotype 28 27 X2=0.30, p<0.6

absent right UA absent left UA

no malformation 13 17

single malformation 10 11

multiple malformations 64 59 Z2=0.79, p<0.7

CONCLUSION: Review of 174 infants with SUA did not reveal

any evidence of a correlation between which, the right or the left,

umbthcal artery is missing and chromosomal or phenotyp~c

anomalies.

454 SELECTED PARAMETER OF F1BRINOLYSIS IN SECOND

TRIMESTER AMN1OTIC FLUID. A A Saleh S VatsX, J.R. Robersonx,

D L Van Dykex, E F Mammen~ Dept of OB/Gyn and Medical Genetics, St

Joseph and Henry Ford Hospitals, Wayne State Umv, Detroit and Pontiac, M[

OBJEC’FIVE: Tissue plasmmogen activator (t-PA), urokinase (u-PA) and

plasmmogen activator inhibitor 1 and 2 (PAI-I and PAI-2, respectively) are

synthesized by the ~rophoblast The relationship between ammodc fluid level of

these factors and abnormal fetal karyotype and possibly abnormal trophoblast

is unknown

STUDY DESIGN: t-PA, u-PA, PAI-1 and PAI-2 were measured by ELISA in

the second lrmaester ammotic fluid from 31 normal and 13 abnormals [tnsomy

21 (N-8) and s~x chromosome tnsom~es (N=5)] Both groups had comparable

ages (38~ 2 vs 39~- 3 years) and similar gestatlonal ages 16± 1 weeks Mann-

Wh~tneyU test was used for statistical analysts (p<0.05 considered significant).

RESULTS: Dala are expressed as Median (Range)

Normal Abnormal P

t-PA 24 1 9 NS

(n~/ml) (1 3-13) (1 2-4 4)

u-PA 0 7 0 9 N.S.

(n~ml) (0-2 3) (0 2-4 7)

PAl-1 0 9 6 3 <0 0001

(U/ml) (0-9 7) (3 4-9 8)

PAl-2 73 84 N S

(u~/ml) (11-159) (47-138)

CONCLUSIONS: 1 )Abnormal trophoblasts release increased amounts of PAI-

l in the ammohc fluid possibly m response to high levels of chonomc

gonadotropms 2)The utlhty of using ammotte fired PAI-I tn prenatal diagnosis

merits further mveshgatton

456 UMBILICAL CORD BLOOD TRANSPLANTATION: CLINICAL FACTORS AFFECTING SAMPLE UTILITY

x x x KG Schlecht , DT Harris , A Booth , HS Miller Departments of Obstetrics & Gynecology, Microbiology & Immunology, University of Arizona OBJECTIVE: To determine the significance of birth weight, gestational age and gender on cord blood volume, mononuclear cell (MNC) count and concentration(cells/cc). METHODS: Human umbilical cord blood was obtained from 211 normal vaginal deliveries at the University Medical Center. Cord blood was collected via syringes from the umbilical vein after hgation of the cord while the placenta was still in utero. Mononuclear cells were then separated using a Ficoll-Hypaque density gradient. Birth weight, gestational age and gender were then compared to cord blood volume, MNC count and concentration (cells/co) using an unpaired student’s t-test with p< 0 05 considered statistically significant. RESULTS: Birth weight was positively correlated with cord blood volume (p=0.0001) and MNC count (p=0.01). Gestatlonal age and gender did not correlate with cord blood volume or mononuclear cell counts. No correlation between birth weight, gestational age or gender and cell concentration was demonstrated. Among infants weighing <2800 gins (n=26), 19 2% (5/26) had fewer than 100 x106 MNC compared to 8 6% (16/185) for infants weighing >2800 gins. CONCLUSION: Birth weight is positively correlated with cord blood volume and MNC count. Since no correlation between birth weight and cell concentration was found, increased cell counts seen with higher birth weight infants are probably secondary to the larger volumes of cord blood obtained from larger infants This emphasizes the importance of improving retrieval techniques to harvest as much cord blood as possible at delivery. Infants <2800 gins are more likely to have cell counts less than the accepted minimum for successful adult transplantation.

436 SPO Abstracts January 1996 Am,] Obstet Gynecol

457 UNEXPLAINED INCREASED MATERNAL SERUM ALPHA-

FETOPROTEIN: THE VALUE OF MULTIPLE MARKER ANALYSIS

TRENDING IN PREDICTING ADVERSE PREGNANCY OUTCOME. E.E.E~ H Cardonlck~, A T Bombard, S J Gross×, S M Carter~, M Abruzzo-Fogarassy=,

D H BaradX, H M Nltowsky~ Montefiore Medical Center, Jacobi Medical Center,

Albert Emsntem College of Medicine of Yeshiva Univetsity, Bronx, NY, Jefferson

Medical College, Philadelphia, PA

OBJECTIVE: To determine whether or not serial measurements oF MSAF~,

MShCG, and MSUE3 are additional predictors of adverse pregnancy outcome in

patients with an initially increased, unexplained MSAFP

STUDY DESIGN: After validalang MSAFP as a predictor of adverse perlnatal

outcome by companng 34 cases having an lmtaal, unexplained MSAFP >2 0 MoM

matched wath the next screened woman of similar age and race, urban patients in

tbas prospective, longitudinal cohort study had monthly measurements of MSAFP,

MShCG, and MSUE3 to evaluate the ufihty of multiple markers in improving ask

assessment for 1UGR, P]H, and PTB Statastacal analysis was performed using Chi

square (X2) and relallve ask 0~,P~) ratios

RESULTS: 34 of 66 patients meeting the study criteria were matched with

controls Outcome data analysis revealed 1UGR 8 (35%) eases vs 2 (6 25%)

controls [p <05, RR 1 72 (1 13-262)LPTB 4 (11 7%) cases vs 1 (3 13%)

controls [p-ns, RR 1 63 (0 98-2 7) Pltt 3 (8 8%) cases vs 2 (6 25%) controls

[p-ns, R~ 1 18 (0 55-2 52)], CombinedPerinatalMorbidity 14 (41 2%) cases

vs 5 (14 7%) controls [p-0 2, RR 1 1 (1 12-2 65)] Analyte data analysis

revealed (1) persistence of increased MSAFP throughout pregnancy, (2) an

Increase in median MSAFP values in affected vs unaffected pts, (3) decrease m

median MSUE3 in affected vs unaffected pts, and (4) lack of correlation of

MShCG 0ncreased or decreased) in pataents with adverse outcomes, MShCG,

however, was not predictive (AFP log ANOVA f 4 11, p< 05 at power 0 51)

CONCLUSIONS: MSAFP (1) and MSUE3 (1) screening in the late 2nd and

earl), 3rd trimester are useful In predicting adverse pennatal outcome, however,

optimal gestatmn age for screening remains undefined

459 UTILILTY OF ROUTINE FETAL KARYOTYPING FOR PATIENTS UNDERGOING

AMNIOCENTESIS FOR ELEVATED MATERNAL SERUM ALPHA-

FETOPROTEIN D GonzalezX, T Barrel, J.Apuzzlo Dept Ob/Gyn, UMDNJ-New

Jersey Medical School, Newark, NJ

Obleetive: To determine the necessity for routine fetal karyotyping m patients

undergoing amnlocentes~s for elevated MSAFP

Study Design: Data was collected in a retrospective manner for all patients under

age 35 who unde~ent amn~ocentes~s for elevated MSAFP at UMDNJ-Ne’,~ Jersey Medical School between January 1,1986 and March 31,1995. Five hundred and

thlriy seven patients w~th MSAFP values greater than 2.5 mut~ples of the median

after correction for maternal weight, race, the presence of maternal ~nsuhn

dependent d~abetes mel[~tus or multiple gestation were included In the study

Patients were divided into two groups Patients weth normal amn~otlc fluid ~-

fetoproteln (n=509) and those with elevated values (n=28) Karyotype results Were

rewewed ~n each group ~n order to cieterm~ne the ~nodence of chromosomal

abnormalities,

Results" There were hve patients with an abnormal karyotype tn the group of 537

patients studied Four of the five patients w~th an abnormal karyotype had normal

amnlotm fluid ~x-fetoprote~n Five hundred and five of 509 patients ~n the group with

normal amniot~c fluid e~-fetoprotem had a normal karyotype (specificity 94,9%,

negative predictive value 99 2%) One of 28 patients in the group with an elevated

amn~ot~c fluid c~-fetoprotem demonstrated an abnormal karyotype (sensitivity 20%,

positive predictive value 3,6%)

Conclusions: The probabd~ty of a major fetal chromosomal abnormahty occurmg

m patients with an elevated MSAFP and a normal amnlohc fluid o:-fetoproteln ~s

extremely small Routine fetal chromosomal analysis of amnlot~c t]uld {s

unneccessary ~n tt~ese cases The incidence of chromosomal abnnrmal~t~es in the

group with elevated amn~otic fluid ~-fetoprotem is greater than f~ve t~mes the

background risk for the general population. A fetal karyotype ~s indicated In tNs

group

458 IS THERE AN EFFECT OF COCAINE ON MATERNAL SERUM ALPHA

FI~2F(/PROTEIN,Chr~srmc Edwards, Zohra Ah Khan~ MS, Department

OB/GYN, Sinai Hospaal of Baltanore, Baltanore, MD

OBJECTIVE: Cocaine ts known to disrupt the uterine and placental vascular

bed Transfer ofalphafatoprotem (AFP)from fetal to the maternal compartment

is Influenced by the integrity of the placental circulation We hypothesized that

maternal cocaine use m~ght Influence results of maternal serum alphafetoprotem

(MSAFP), hy altering transfer of AFP to the maternal compartment

STUDY DESIGN Between April 1994 and June 1995, we idenufied 43 women

who had positive drug screens ou routine mld-truuester reMing. Twenty women

tested passive Iklr cocaine, 9 Womeu tested positive for cocaine and other drugs

and 14 women tested posnlve for drugs other than cocaine One hundred

patients who tested negative for cocaine and other drugs were used as controls

MSAFP results were not ~tvmlable for 7 of the patients who only tested posmve

for cocaine Patanlts with fetal anomalies were excluded from the analysis

Dais was z~aJyzed by the a~aly~s at" variance

RESULTS: The data demoustrated a trend of higher levels of AFP in cocaine

users, but this trend was not stahstlcally slgmficm~t Patients who tested

posinve for cocmne had the highest levels ofalphafetoprotem, followed by those

who tested positive for cocaine and other drugs, and thirdly by those who only

tested pos~twe for other drugs Panems who tested negative for all drugs had

the lowest mean MSAFP

CONCLUSIONS: Cocaine use does hal seem to have a slgnlficoaat influence on

the amount of AFP found m the maternal blood stream However, a trend

other drugs Further study employing a larger sample site may demonstrate a

slgntficmit difference

460 AFP AND HCG VALUES FOR 304 TWIN GESTATIONS N. Ginsberg, G Zuckerman, Y Verlinsky, C. Strom, Reproductive Genetics Inst, Illinois Masonic Medical Center, Chicago, IL OBJECTIVE: To determine the normal values of maternal serum AFP and hCG in twin pregnancies. STUDY DESIGN: Medians of maternal serum AFP and hCG were determined from 304 twin pregnancies and compared to the medians for singleton pregnancies. RESULT: Twenty or more values per week were available for twin gestations from 15 weeks to 19 weeks gestation. The ratios of mean AFP for twin vs singleton pregnancies were consistent at all gestational ages, varying from 2.5 - 2 7. In contrast the ratio of mean hCG for twin vs singleton pregnancies began at 3 4 at 15 weeks and decreased continuously to 2 5 by 19 weeks Even when averaged over all gestational weeks, our observed hCG values for twin gestations were significantly higher than those observed by Wald et al. CONCLUSIONS: The ratio of hCG in twin vs singleton pregnancies is higher than originally published and decreases with advancing gestational age. These data must be considered if maternal serum analyte levels are used to predict aneuploidies in twin gestations

Volume 174, Number I, Part 2 SPO Abstracts 437 Am J Obstet Gyne(ol

461 MSAFP/HCG SCREENING FOR DOWN’S SYNDROME IN TWIN

GESTATIONS. J Stone*, L Lynch, R Laplnsk~, M Alvarez Dept of

Ob/Gyn. Mt S~nm School of Medicine, New York, NY

OBJECTIVE: To determine whether the MSAFP/HCG ratio is a

valuable screening test for Down’s Syndrome (DS) m twin gestations

and to determine whether correcting for twin median HCG and AFP

values increases the detection rate

STUDY DESIGN: 96 patients delivered of normal twins and 10

patients with twins discordant for DS who underwent MSAFP/HCG

screening at 15-20 weeks constituted the control and study

populations Median values of AFP and HCG were determined for both

groups Senslhvlty (SN), specificity (SP) and false positive (FP) rates

for DS screening were determined ROC analysis was also performed to

determine the SN and SP at variable FP rates The data was then

analyzed by correcting for median twin AFP and HCG values and again

SN, SP, FP rates determined

RESULTS: The following table demonstrates median values for both

groups Normal twins T.wlns discordant for DS

median AFP (MOM) 2 26 2 03 median HCG (MOMI 2 00 2 87

The Down’s ratio correctly identified 9/10 patients discordant for DS

using a cut-off risk of 1/270 for singletons with a FP rate of 0 18 The

following table reports the different SN, SP, and FP rates before and

after correcting for median twin values

A(~I R,sk SN SP FP

uncorrected 1/270 0 90 0 81 0 18 1/138 0 60 0 90 0 10

1/91 0 30 0 95 0 05

corrected 1/270 0 70 0 90 0 10 1/138 0 30 0 95 0 05

CONCLUSIONS: The MSAFP/HCG ratm is a useful screening test for

DS ~n twin gestations Using a cut-off risk of 1/270, 90% of twins

discordant for DS will be ~dentlfled However, this is associated with a

high FP rate Correcting for twins does not appreciably alter the

detection rate and results in lower SN (0 70) and lower FP rate (0 10)

463 SECOND-TR/MESTER MATERNAL SERUM SCREENING FOR DOWN’S SYNDROME: COMPARISON OF FREE BETA-hCG AND ALPHA-FETOPROTEIN WITH TOTAL hCG, ALPHA- FETOPROTEIN AND UNCONJUGATED ESTRIOL.

Ph Extcrmann x, p B~schof x, F. B~gum. Dept of Ob/Gyn, Umv of Geneva, Geneva, Switzerland OBJECTIVE: To compare two protocols for second-trimester maternal serum scr~*nmg for Down’s syndrome m the same serum samples, using a triple test (total hCG, alpha-fetoprotam, unconjugated ¢stnol) and a double test (free beta-hCG, alpha-fetoprotein) STUDY DESIGN: E~ghteen singleton pregnancies v~ath fetal Down’s syndrome and 2522 pregnant women receiving routine antenatal care m Geneva were the subjects of this study RESULTS: Among the 18 affected cases, at a cut-off r~sk of ] 380, the detectmn rate of Down’s syndrome was h~gher w~th the double test (94%, 17/18) than with the triple test (67%, 12/18) (p > 005). In the cohort of 2522 pregnant women screened between 15 and 18 week’s gestation, for a stat~shcally snmlar detection rate of affected cases, the double test achieved a significant reduction (p < 00001) m the number of false pos~hve cases among patients under 35 years of age, ~rrespect~ve of the cut- offnsk selected (from 1 190 to 1 380) CONCLUSIONS: Compared to total hCG, alpha-fetoprotem and unconjugated estriol, use of free beta-hCG and alpha-fetoprotem represents a better second-~nmester screening test for Down’s syndrome, because it s~gnificantly decre,mes the false positive rate at a lower ntunmg cost

462 COMPARING THE RATE OF DETECTION OF TRANSLOCATION-

TRISOMY 21 W13~I THE RATE OF DETECTION OF TRISOMY 21 WITH

DECREASED MSAFP. J Robertsx, A Bombard, S Gross~ , M Laksx , V

Puhjaal~, H Nltowsk3i’ Montefiore Medical Center, Bronx, NY 10461

OBJECTIVE: The association between decreased Maternal Serum Alpha

FetoProtmn (MSAFP) and fetal trmomy 21 is well known, however, an association

between decreased MSAFP and fetal translocatlons has not been reported Two

index cases of transloca~on-lnsomy 21 were discovered prenatally, through analysis of ammotlc fired, following counseling for decreased MSAFP These findings

ignited interest In whether other etiologies of Down Syndrome are as likely to be detected by a poslhve serum marker screen As a result, we sought to determine

whether the rate of detection oftranslocation Down syndrome was as great as the detection rate for trisomy 21

STUDY DESIGN: Data regarding the number of new cases of tnsomy 21 and

translocatmns revolving chromosome 21 in 1990-1993, were obtained from the New

York State Department of Health Chromosome Registry, courtesy of Drs Carolyn

L Olsen and Phihp Cross Information regarding mode ofdetectmn was also

available The translocatlon karyotypes were reviewed by an expenenced

cytogenetlclst, to determine which ones contained extra copies of chromosome 21

(translocation-trlsomy 21) The percentage of tnsomy 21 detected by decreased

MSAFP was compared to that of the translocatmn-trlsomies using a Chi Square

analysis with Yates correction

RESULTS: There were 888 cases oftnsomy 21, of these, the mode of deteehon

was available for 775 and 120 were detected by decreased MSAFP There were 24

translocatlons revolving chromosome 21, of which 19 were translecatlon-tnsomles

and 7 of these were detected by a decreased MSAFP X~ p = 0 012 (Yates

correction p-0 028)

CONCLUSIONS: The rate of detection of translocatlon-tnsomy 21 is mgmficantly

greater than the detection oftnsomy 21 The fact that the detection of translocat~on

Down syndrome is at least as good as for Down syndrome due to tnsomlc Down

syndrome, provides further evidence that the alteratmn of MSAFP ~s rdated to a

phenotype common to these two different chromosomal abnormahtles, versus a

cytogenetm feature umque to the tnsomy (e g, late repbcation, or a supernumerary

p arm) We acknowledge that we did not have information as to the number of

mothers who underwent MSAFP screening and, consequently, the true screen

positive rate for translocatlon Down syndrome

464 AMNIOTIC FLUID ALPHA FETOPROTEIN (AF-AFP) AT THE TIME OF GENETIC AMNIOCENTESIS: HAS IT OUT LIVED ITS USEFULNESS ? L.E. Shields. MD, S.B. Uhrich×, MS, CGC, S.E. Rutherford, MD, R. Zachariasx, MS, T.C, Winterx, MD Divisions of Perinatal Medicine and Diagnostic Imaging, University of Washington, Seattle WA 98195. OBJECTIVE: To evaluate the utility of routine measurement of AF-AFP at the time of second trimester

~ enetic amniocentesis. TUDY DESIGN: Retrospective 7 year review of pre-

amniocentesis ultrasound findings and AF-AFP levels (MoM) in all patients undergoing genetic amniocentesis. The utility of the AF-AFP test was assessed by finding an abnormal (high) AF-AFP with a normal pre- amniocentesis ultrasound exam confirmed by a normal newborn outcome. RESULTS: During the study period 6,876 patients underwent second trimester genetic amniocentesis. Outcome data was available in all cases. In 78 (1.0%) the AF-AFP was > 2.0 MoM. Twenty-six of the 71 (37%) had normal ultrasounds, and in 25/26 the AF-AFP was 2.0-3.0 MoM. None of the ultrasound screen negative fetuses had detectsble abnormalities at birth. Abnormal ultrasound findings were noted in 45/71 (63%) and 37 of these (82%) were neural tube defects, abdominal wall defects, or cystic hygroma. CONCLUSIONS: In this study, with close to 7,000 patients, AF-AFP did not increase the detection of fetal abnormalities. Based on these results, routine measurement of AF-AFP at the time of genetic amniocentesis does not appear justified.

438 SPO Abstracts .lanum), 1996 Am J Obstet Gynecol

465 NORMAL MID-TRIMESTER "GENETIC" ULTRASOUND REDUCES

’i~iE RISK OF ANEUPLOIDY IN PREGNANCIES WITH A POSITIVE TRIPLE SCREEN. S Longo. T Asrat, D Waitersx, C Towers, M Nageotte,

R Freeman, D Lagrew Dept. OB/GYN, Long Beach Memorial Medical

Center, Long Beach, CA. Saddloback Memorial Medical Center, Laguna Halls,

CA ~nd Umversity of Califomm, Irvme, Orange, CA.

OBJECTIVE. To ~k*ta, the issue of whether a normal targeted mid-trimester

ultrasound reduces the risk of aneuploidy in a patient wath an abnormal triple

screen has not been resolved Therefore, the purpose of our study was to

determine the sensitivtty and specafictty of combined multaple ultrasound

phenotypic markers and baometric measurements for detecting fetal aneuploady m a cohort of portents with a posatave triple marker screen (TMS)

STUDY DESIGN. Over a three year period, 8017 patients had TMS drawn, of

which 423 patients had abnormal results [ Down Syndrome (DS) risk > 1/250]

Prior to undergoing an amniocentesis, each patient had a targeted "genetic"

ultrasound consisting of a highly detailed fetal anatomical survey and

measurements of various biometnc parameters (BPD, OFD, HC, AC, FL, HL-

humeral length, CM--cisterna magna, NF-nuchal fold). A scan was considered

positive if biometdc parameters (NF’~6mm, CM>10mm, short FL, short HL,

IUGR) were abnormal structural malformations were present, or fetal survey

was inconclusive

RESULTS’ The overall prevalence of fetal aneuploidy was 14/423 (10-DS, 1-

Trisomy 13, 2-Tnsomy lg, 1-Mosaic Tumor’s Syndrome) A normal ultrasound had a specificity of 94% (383/409) and a angat~ve predactivo value

of 99% (383/386) An abnormal ultrasound had a sensitivity of 71°/, (10/14)

wRh a positive predictive value of 28% (10/36). With an abnormal ultrasound the risk of DS was 1/3.6 as compared to 1/97 with a normal ultrasound (F<lffs,

a RRof 26.88,9~% CI of 8.87-81.39).

CONCLUSIONS: A normal, targeted, mid-trimester "genetic" ultrasound in

patients with abnormal TMS reduces the risk of aneuploidy by 27-fold as

compared to cases vath a positive ultrasound. This data can be used to adjust

risk estimates of Down Syndrome based on biochemical screemng and thereby

allow better selection ofpataents for genetic amniocentesis.

467 CHOROID PLEXUS CYSTS AND RISK OF ANEUPLOIDY. MR Leonardi*., HM Wolfe, A Grebx, MP Johnson, JM Lanouette*, JB Landwehr*, MI Evans. Departments of OB/GYN, Molecular Medicine & Genetics, and Pathology, Wayne State University, Detroit, MI.

OBJECTIVEt The risk of aneuptoidy attributable to choroid plexus cysts (CPCs) is controversial. We compared the incidence of aneuploidy in cases of isolated CPCs to that of CPCs with associated major (MAJ) and minor (MIN) anomalies. STUDY DESIGN: Ultrasound and genetics databases were rev=ewed for all patients with CPCs and known fetal karyotype. CPCs were categorized as isolated, M{N, or MAJ. Minor anomalies included abnormalities of AFV, single umbilical artery, and pyelectasis. RESULTS: 149 CPCs at a mean gestational age of 19 weeks were identified. 18 of 149 (12%) of CPCs had associated anomalies, 10 of these were minor anomalies of which 2 had abnormal karyotypes. 8 had major anomalies of which 4 were aneuploid.

Total n (%)

Aneup~oid n (%)

Isolated MIN MAJ Total

131 (88) 10 (6.7) 8 (5.4) 149

0 (0l 2 {20) 4 150) 6 (4)

CONCLUSIONS: CPCs appear to be associated with an increase in the incidence of major and minor anomalies. The risk for isolated CPC remains controversial and cannot be settled here. While the presence of isolated CPCs may not place the patient at risk for aneuploidy, ~t should prompt a diligent search for other anomalies. CPC in the presence of any other anomaly warrants karyotypic evaluation.

466 THE CLINICAL IMPORTANCE OF PERICENTR1C INVERSION OF CHROMOSOME 9 IN PRENATAL DIAGNOSIS. A. kale-Allah’, P. MinE’, H. Salemx, O. Mohamed’, E.A. Reece. Dept. of OB/GYN & RS, Temple Univ Sch of Mad, Philadelphia, PA

OBJECTIVE: Pericentric inversion of chromosome 9, a structural chromosomal variant, is the most common type of inversion, with an overall incidence of 1 98%. Since this phen~nenon occurs m~t fi’e- quenfly in African-Americans (3.59%), we examined a relatively large

datahase of predominantly African-Americans to assess the perinatal

outcome and clinical profiles such fetuses. STUDY DESIGN: We have reviewed the results of 652 cytogenetic

studies on mid-trimester anmiocentesis (16-18 weeks) during a 10-year

pen(x1: January, 1985 to February, 1995; done following studies; and analyzed the data.

RESULTS: Of tl~ 652 cases reviewed, pericentzic inversion of chzomo- some 9 was found in 27, resulting in an incidence of 4.1%. The indica-

tions for genetic anmiocentesis among these 27 patients included ad-

vanced maternal age (16/27; 59.3%); abnormal ultzasoond findings (5117; 18 5%); abnormal maternal serum screening tests (4/27; 14.8%), and family history of an abnormal child (2/17; 7.4%) Immediate neonatal

evaluation was normal in 24 (88 9%) and abnormal in three (11.1%); one had low APGAR score at 1 and 5 minutes, one had bilateral capillary hemangioma over the eye lids and one had a heart murm~r and an irregular cardiac rhythm. Most patients (24/27) were mu~tigravid, nine of whom had a history of 2 or more previous spontaneous abortions.

CONCLUSIONS: These data demonstrate that although a normal outcome occurs in the majority of chromosome 9 inversion fetuses, some are associated with abnormal outcomes. Under such circumstances, parental karyotyping may be recommended since pericentric inversion is heritable.

468 ULTRASOUND MARKERS OF ANEUPLOIDY (A): GESTATIONAL AGE (GA) VARIATION BY SPECIFIC KARYOTYPE. ~, MP Johnson, RF

Hmne, RJ Sokol, Y Yaron, MI Ev,’ms. Depts. Ob/Gyn, Genetics, & Pathology, Wayne Skate University, Detroit, MI Objective: To determine the likelihood of different aneuploidies,

by GA, in patients with confirmed karyotypes having nuchul folds(NF), simplc(SIM) and seplated(SEP) cystic hygromas. Study Design: A retrospective analysis of 158 consecutive

patients with NF, SIM, or SEP on either abdominM or vaginal ultras(rand. Results: 38 patients had NF (21%A), 65 SIM (22%A), and 55

SEP (58%A). SEP had the h~ghest incidence of abnormalities (>50%). A peak of trisomy 21’s (27%) was found in the early midtrimester, leveling off at 11% m the late mid-trimester. A high incidcnce of trisomy 18 (22%) was noted in the late first trimester.

45,~ was found later in estalion ,’rod onl with SEP. GA NL % ’121 % 118 °~o 45,X % Other %

SIM 9-12 27 77 4 11 2 6 0 0 2 6 13-15 15 75 5 25 0 0 0 0 0 0

~6-20 9 90 0 0 "b" 0 0 0 1 10

SEP 9-12 4 44 1 11 3 22 1 11 1 11

13-15 6 40 4 27 1 6 4 27 0 0

16-20 13 42 3 10 1 3 11 35 3 10

NF 15 ,4 60 1 17 d 0 0 0 1 17

16-20 26 81 5 16 I 1 3 0 0 0 0

Tot 104 66 23 15I 7 4 16 10 8 5

Other = trisomy 13 and other translocations mid deletions Conclusion: 1) Aneuploidy rates vary by marker from 21-58%. 2)

Each marker varies for actual risk (% and type of aneuploidy) at dilfcrent GA’s. 3) Terminology must be used consistently lor proper patient counseling.

Volume 174, Number 1, Part 2 SPO Abstracts 439 Am.] Obstet (;ynecol

469 POSITIVE OR FAINT AMNIOTIC FLUID ACETYL-

CHOLINESTERASE BAND WITH NORMAL

ULTRASOUND. CL Brown~, KA Coldenx, RF Hume, MP

Johnson, MC TreadweIl, A Druganx, ] Lampinenx, MI Evans.

Depts. Ob/Gyn, Molecular Medicine & Pathology,

WSU/Hutzel Hospital, Detroit, MI. OBJECTIVE: The association of fetal abnormalities in the presence of a faint AF-ACHE band following amniocentesis was reported in the 1980’s. We now evaluate the risk for fetal abnormalities or poor outcome in the presence of a nl US and positive or faint AF-ACHE DESIGN: Between 1/1/89 and 2/1/95, 4,859 amnios were performed for AMA, abn. MSAFP, elevated AFAFP, posmve ACHE, history of an anomalous child, or abn US. 23 cases ofabn ACHE and nl US were identified and pregnancy outcome was determined. RESULTS: Only 1 of 23 abn AF-ACHE (4%) was aneuploid (45,X/47,XXX mosaic). Amnios were done before 16 wks in 9, between 16-20 wks In 8, and after 20 wks in 6. 17/23 (73%) fetuses were born without abnormalitms at term. No VWD or NTD were seen m infants with an abn ACHE and nl US.

I N I nl inlanls I Abn in~ants I IUFD I ETW I Pos ~nd 4 3 (75°/o) 0 0 1 (25°to)"

*=Termination for 45,X/47,XXX mosa cism, **=1 term infant with small bowel obstruction and tracheal occlusion; 1 36 week infant with brain hernatoma, ***=27 week twins with twin B (VSD, duodenal atresia, CNS abn on autopsy); PROM at 22 weeks; and stillborn at 20 weeks. CONCLUSIONS. 1)Advances xn US have lead to ~mproved detection of fetal abnormalities, 2)ln the presence of a nl US and karyotype, repeat invasive testing to confirm a suspicious AF-ACHE may not be necessary, 3)Since the incidence of 1UFD appears to be increased with abn ACHE, increased antepartum surveillance may be warranted.

471 THE IMPACT OF MATERNAL SERUM SCREENING TESTS ON MID- TRIMESTER PRENATAL DIAGNOSIS. RO Day=s, P Cosper~, M DuBardx, CG Brumfield, S Finleyx, J Owen, KWenstrom Dept OB/GYN, Umversity of Alabama at B~rmmgham, B=rmmgham, AL OBJECTIVE: To determine the effect of m=d-tnmester maternal serum screening (MSS) tests on our prenatal diagnosis program. STUDY DESIGN: We utihzed our computerized data base of women referred for prenatal diagnosis at 14-22 weeks’ from 1989-1994. Indications for referral were catogonzed as maternal age (MA), h=gh or low serum AFP (T AFP or ~. AFP), pos=ttve multiple marker screen (MMS), and other We compared numbers and types of referrals, rate of amn=ocentes=s, and the number of abnormal fetat karyotypes detected by amnlocentes~s RESULTS: Data on 11,222 women was collected in the six year study period The proportion of women referred for MA decreased from 64% to

50%, while referrals for abnormal MSS tests increased from 17% to 32% The proportion of women who underwent ammocentes~s decreased from 83% to 67% (Table)

Year Maternal Age Abnormal Other Amnio MSC

n (%) n (%) n (%) n (%)

1989 1117 64 312 17 323 18 1462 83

1990 1209 66 274 15 346 19 1481 81

1991 1089 64 312 18 292 17 1307 77

1992 1019 5~ 368 21 367 21 1282 73

1993 1113 54 514 25 430 21 1483 72

1994 1071 50 686 32 380 18 1441 67

Over the s~x year periqd, the rate of ammocentes=s after referral for MA

was 6211/6619 (94%), for t AFP 306/1193 (26%), for ~. AFP 744/886 (84%) and 311/387 (80%) for a MMS Using our screemng protocol, the number of amnloce~teses performed for each fetal sex chromosome abnormahty or serious autosomal abnormahty detected was 1/59 for MA, 1/62 for I" AFP, 1/91 for I AFP and 1/29 for a MMS CONCLUSION: The use of m=dtnmester MSS has increased referrals for prenatal d~agnos~s and is now the second most common ind~cat~on for referrals Compared to MA, i’ AFP and I AFP, ammocentesls is more efficient in detecting fetal chromosome abnormaht~es m women referred for a pos=t~ve MMS

470 DECREASED AMNIOTIC FLUID VOLUME: SUGGESTIVE BUT NOT

PREDICTIVE OF FAS. S Martier. H M Wolfe,* J Ager,* R Sokol*. Dept.

Ob/Gyn, Wayne State Umv, Detrmt, MI

OBJECTIVE: Decreased amnmtm fluid volume (AFV) is associated with

decreased growth consistent w~th IUGR. Decreased growth is a defining feature

of Fetal Alanhol Syndrome (FAS) It ts hypotheslzed that prenatal alcohol

exposure is associated with decreased AFV. The purpose ofth~s study is to

examine AFV ~n FAS, non-FAS infants

STUDY DESIGN: Fifty-nine FAS ~nfants were ~dent~fied over 7 years as part of

a large prospective study on alcohol and drugs in whmh 14,707 women were

screened for alcohol use. FAS infants, matched with controls on birth date, were

examined for AFV, categorized as ~ncreased or decreased A stepwlse

dlscrlmlnant analysis was conducted w~th FAS, non-FAS as the dependent

variable Continuous alcohol variables and AFV were entered into the equation

as predmtions Gestat~onal age at time of ultrasound examination was similar for

both groups and not entered into the equatmn.

RESULTS: As expected, alcohol (absolute alcohol per drinking day) prethcted

FAS (F (1,188)=6 046, R = 3%, p< 05)). Decreased AFV was seen more in FAS

cases, but it was not a significant predictor.

CONCLUSIONS: It is concluded that placental insufficiency is not a major

determinant of IUGR in FAS infants

472 FETAL CYSTIC HYGROMA--PROGNOSIS OF SEPTATEDVERSUS NONSEPTATED LESIONS. C G Brumfield, KD Wenstrom, RO Davis, J Owen, P Cosper. Dept of OB/GYN, Univ. of Alabama at Birm=ngham, B=rmingham, AL OBJECTIVE: To compare sonographic, karyotypic, and prognostic features between septated versus nonseptated les~ons METHODS: A computerized ultrasound database was used to identify fetuses at 14-26 weeks d~agnosed as hawng a cystic hygroma. Photographs from the =n~t~al ultrasound were retrospechvely rewewed, w=th the reviewer being blinded to the results, to record the lesion type (septated vs nonseptated) and any abnormal structural findings Fetal karyotypes were obtained from amn~ot~c fluid or fetal tissue culture. Pregnancy outcome data was obtained from hospital charts and physician office records. The ultrasound findings were then compared to fetal karyotype results and pregnancy outcome data. RESULTS: From 1990-1995, 61 fetuses w~th cystic hygroma were ~dentff~ed. Karyotypes were obtained in 55 and pregnancy outcome data was available in 53, An abnormal karyotype was present in 42 (76%) fetuses with 45,X being the most common In 30/42 (71%).

Septated (N=39) Nonseptated (N=16) P Values

Edema 12 (32%) 4 (25%) .63

Hydrops 23 (59%) 3 (19%) .007

Other Defects 4 (11%) 1 (6%) 1 0

Abnormal Karyotype 33 (87%) 9 (56%) ,03

Induced Abortion 21 (55%) 8 (53%) .90

Fetal Death 5 (41%) 3 (20%) .16

Neonatal Death 1 (3%) 0 .52

Surviving Infants 1 (3%) 4 (27%) .008

CONCLUSIONS: Fetuses with a septated cystic hygroma as compared to those w~th nonseptated lesions are more likely to be chromosomally abnormal and/or to develop hydrops and thus are less hkely to survive.

440 SPO Abstracts JanuaiT 1996 Am J Obstet Gynecol

473 SPONTANEOUS PREGNANCY LOSSES IN WOMEN REFERRED FOR MID-TRIMESTER GENETIC COUNSELING RO Daws., P Cosperx, M DuBardx, CG Brumfield, K Wenstrom, J Owen Dept OB/GYN, University of Alabama at B~rrnlngham, B~rmmgham, AL OBJECTIVE: To evaluate pregnancy losses m women referred m the m=d-tnmester for counsehng because of ~ncreased risk for a fetal abnormahty Since amnlocentes=s might effect pregnancy loss, women who underwent counseling only were compared to those who underwent ammocentesls after counsehng STUDY DESIGN: We utfl=zed a computerized data base to ~dentff’f women who had m=d-tnmester (14-22 weeks’) counsehng for prenatal d~agnos~s Indications for counseling included maternal age, elevated serum AFP, low serum AFP, poslhve mulhple marker screen (MMS) and other Pregnancy loss was defined as a fetal death, spontaneous aborton, stdlb~rth or neonatal death, and was characterized as either < 30 days or > 30 days after the m~t~al visit The study group was

I~-n~ted to singleton pregnancies Therapeutic abortions, known fetal anomahes, and abnormal fetal karyotype were excluded RESULTS: 8807/11,971 (74%) ofpregnanoes met the ~nclus~on criteria There were 195 early and late pregnancy losses (22%). The spontaneous loss rates at _< 30 days and > 30 days were 0 8% and 1 4%, respectively (p< 001) There were no slgn=ficant d=fferences in pregnancy loss rates at e~ther < 30 or > 30 days, regardless of md~cat~ons, ~n women who d=d or ~d not have ammocentes=s (Table)

Pregnancy Losses After Referral for Mid-Trimester Counseling Loss < 30 Days Loss > 30 Days

Amino No Amino Amino No Amnlo

A 42/5178(0 8%) 5/270(1 8%) 68/5178(1 3%) 4/270(1 41%) B 3/198 (1 5%) 8/605(1 3%) 10/198(5%) 17/605(2 8%) C 3/844 (04%) 1/142 (07%) 2/844(02%) 0/142(0%) D 2/625 (0 3%) 5/945 (0 5%) 8/625(1 3%) 17/945 (1 8%) A= Maternal Age, B=Elevated AFP, C=Low AFP/MMS, D=Other

CONCLUSION: In a large unselected populat=on of women w=th md~cat~ons for genetic counseling, the decision to proceed w=th amn~ocentes~s d=d not have a sNn=ficant =mpact on early or later pennatal loss

475 VERY LOW UNCONJUGATED ESTRIOL: A MARKER FOR INCORRECT DATING AND PREGNANCY ABNORMALITIES. C.Berman1, B. Ma~hak~, I. P. Rossiter’, E. Prassma~P, K. Blskemore. Dept. Gyn/Ob, Johns Hopkins Univ., Baltimore, MD. OBSECTIVE: To explore the relationship between vcr~ low ( .<0.5 multiples of the median (MoM)) matermfl serum unconjugat~ estriol (vI-MSu~) and pregnancy outcome. STUDY DESIGN: 28 patients referred between 1/92 and 6/95 had vl-MSuEs. Maternal ages, estimations of gestetional age, fetal karyotypes, sonographic

findings, and pregmmcy outcomes were compiled by chart review. RESULTS: Of the 28 patients, 26 had an incre.ased risk for trisomy 21, 1 for trisomy 18 and I elevated maternal serum alphafetoprotein. Incorrect eatimation of gestational age (> 10 days discrepancy between sonogrephic and menstrual age) explained 7 cases (25%). Of the remaining 21, no fetuses had trisomies 18 or 21, though 2 sex chromosome abnormalities were detected. One with hydrops, n cystic hygroma and a 45,X karyotyp© resulted in fetal death in utero (FDIU). The other had 5 % amniocytc mosaicism, 45,X/46,XY, confirmed by cord blood at term delivery of a phenotypically normal male. 45,X is suspected

in another FDIU (hydrops, cystic hygroma), but karyotype failed. Two isolated birth defects were identified-- Tetralogy of Fallot and a unilateral pelvic kidney. 3 patients had placental findings on ultrasound: one was large, cystic, and lobular; one ~ multiple cysts at the cord insertion site; and one lind a single hypoechoic region. Of the remaining 13 pregnancies, 5 carried to term with normal outcomes, 2 with normal karyotypes are ongoing, and 1 with no identified abnormalities was alectively aborted. Theother 5 pregnancies had normal karyotypes and no sonographic abnormalities, but pregnancy outcome data is unavailable. Of the 8 patients with estriol ’<0.4 MoM with correct dates, there were 3 placental and possibly 3 sex chromosome abnormalities. CONCLUSION: Low MSuEs appears to be an excellent marker for incorrect dating. In addition, in our series, "trun" vl-MSu~ is associated with pregnancy abnormalities, including FDIU and placental, Ioiryotypic, and congenital anomalies. Further studies are needed to confirm the value of MSuEs as a predictor of pregnancy outcome.

474 THE EFFECT OF PARITY CORRECTION ON DOWN SYNDROME DETECTION USING THE MULTIPLE MARKER SCREENING TEST (MMST). K.D. Wenstrom J Owen, L Bootsx The Un~vers=ty of Alabama at B=rmlngham OBJECTIVE: 1) To determine tf parity affects analyte levels in the MMST for Down Syndrome (DS) and, if so, 2) to denve a correction factor and determine Its effect on DS detection and screen positive rates STUDY DESIGN: Our data base consisted of 3039 MMST results and corresponding fetal karyotypes (2983 euploid / 56 DS), Cases were grouped by maternal parity’ Panty=0, 1, or :> 2 The mean multiple of the median (MOM) of MSAFP, estnol, and hCG was determined for each group A panty correction factor was then denved for each panty group Corrected MQM = Sample MOM + (Mean Panty group MOM/Mean Population MOM) The correction factors were apphed to the database and parity- corrected DS detection rates and screen positive rates were determined RESULTS: Panty slgmflcantly affected the mean MOM of hCG (Para 0=1 0916, 1=1 0134, 2=09329, p=0 0001), but did not affect estrlol or MSAFP (p=NS) Application of a panty correction factor for hCG ~ncreased the DS detechon rate ~n multips without ~ncreas~ng the overall screen positive rate

UNCORRECTED CORRECTED PARITY AGE SCREEN (+) SENS SCREEN (+) SENS

ALL 35 6 24.0% 68% 23.6% 71% 0 34 4 22 6% 72% 19 5% 72% 1 35 6 24 0% 67% 23,9% 67%

>2 36 7 25 0% 65% 26 6% 77% CONCLUSION: hCG levels, but not MSAFP or estnol levels, are slgmflcantly lower in multlparas Correction of hCG for maternal panty increases DS detect=on by 18% ~n women w~th 2 or more prewous pregnancies, w=th no d=scemable effect on the overall screen positive rate

476 QUANTITATION OF PERINATAL RISK IN THE NON-

ANOMALOUS FETUS BASED ON MATERNAL AFP, hCG

and uE3. R. Bahado-Singh. O.Derenx, A Tanx, W. Reguerox, J.

Zhangx J. Copel, R.A. Ehrenkranzx’ I. Mahoneyx, A. Baumgartenx

Yale University School of Medicine New Haven, CT.

OBJECTIVE: To develop probability curves for estimating the risk

of perinatal complications based on of mid-trimester maternal AFP,

hCG and uE3

STUDY DESIGN: We ascertained outcome in 1111 non-anomalous

singleton pregnancies undergoing screening and dehvered at our

hospital Multivariate logistic regression analysis was used to

identify maternal risk factors and the serum analytes significantly

correlated with adverse outcome. Equations were developed for

predicting the risk of perruatal complications: [prematurity, birth

weight <2500 gins or <10th percentile, NICU stay, PIH,

fetal/neonatal death, PROM, and abrupUon] Probabdity curves

estxmatmg the risk of comphcations for women with and without

significant maternal risk factors were plotted.

RESULTS: There were 214 (19.3%) pregnancies with

comphcations. There were 137 instances ofAFP >2MOM, 462 with

hCG >2MOM, 35 with both analytes elevated and 277 with uE3 <0.60

MOM Sigmficant risk factors in the logistic regression ’~ere race

(non white), OR 1 95, (1 42-2.66) p=0 0001, medical history, OR

1.94, (1 34-2 79) p=0 004, elevated AFP, OR l 90 (1.24-2 89),

p=0.002 and low E3, OR 1 92 (1.37-2 69) p=0.0001. Although low

hCG (<0.75 MOM) unhke elevated hCG predicted adverse outcome,

(p=0 004 and p=0 8 respecuvely), neither were significant in the regression analysis The probability of permatal complications was

expressed by, po=l/{ l+e’[-1 3482+AFPx(0.3942)-uE3x(0 6316)]}

CONCLUSION: Elevated AFP and low uE3 ,unhke hCG levels

independently predict perinatal comphcations Quantitative risk

estimates based on analyte levels and maternal factors, are now

available for counsehng.

Volume 174, Nulnber l, Part 9 SPO Abstracts 441 Ant,l Obstet Gynecol

477 NORMAL ULTRASOUND BIOMETRY REDUCES THE NEED FOR AMNIOCENTESIS IN ADVANCED MATERNAL AGE.

R Bahado-Smgh, A TanX,O Derenx, W Reguerox, J Zhangx, J.

Copel, J Mahoneyx, A Baumgartenx. Yale Uni’~ers~ty School of Medicine, St. Raphael Hospital, New Haven, CT OBJECTIVE: To develop biometry adjusted risk estimates for chromosome anomaly in women undergoing ammocentesis for advanced maternal age (AMA) STUDY DESIGN Ultrasound blometry was performed in 4279 women undergoing amnlocentes~s for AMA. Only widely utihzed parameters such as bipanetal dmmeter (BPD), head circumference (HC), transcerebellar diameter (TCD), femur length (FL), humerus length (HL), both combined (FLHL) and abdominal circumference (AC) were considered, Expected dimension of each parameter based on BPD was

established using regression equations Observed = expected values

for each parameter were subsequently calculated. Using receiver operator curves, the most efficient threshold O/E values for screening for Down syndrome (DS), and clinically significant chromosome defects (CSCD) were obtained Using stepwise logistic regression, the combinatmn of blometnc parameters with the highest detection rate for DS, and CSCD was determined. Risk tables for DS and CSCD in the sonographically normal fetus for AMA pregnancies were developed RESULTS: There were 40 cases of DS, and 136 CSCD. The optimal screening parameter for DS was O/E HL <0.90 and for any CSCD was O/E HL <0 90 or AC O/E <0.92. Of those with complete biometry, with either abnormal biometry or anatomy, DS occurred In 16/488 vs 21/3545 in normals, OR (95% CI) 5 69(2.8-11 46) p<0.00001. The values for CSCD were 531771 vs 74/3298, OR 3 22(2.2-4 69) p<0 00001 With both normal biometry and a fetal anatomy the mi~d

trimester DS risk for a 35 year old female fell from 1/270 to 1/476 while the risk for CSCD fell from 11115 to 11206. CONCLUSION: Normal ultrasound blometry and anatomy Slglnflcantly reduces the risk of chromosome abnormality in AMA. This information should be used for counsehng.

479 PROSPECTIVE EVALUATION OF PRENATAL SCREENING FOR TRISOMY 18. A Fultonx, J Yankowltz, R Wdhamsonx, S Grantx, WT Budeherx Depts Ob]Gyn & Preventive Med, Umv of Iowa

College of Medicine, Iowa City, IA OBJECTIVE: To evaluate the performance of prenatal serum screening for tnsomy 18 (T18) STUDY DESIGN: We included all cases of maternal serum testing (August 1991-June 1994) with a T18 positive screen (AFP <0 75 MuM, UE3 5.0 60 MuM, HCG <0 55 MuM) Results of all ammocenteses, ultrasounds, and birth or Lath reformation were obtained from the Iowa Expanded Serum Screening Program, the Iowa Dept of PuNIc Health, and the Iowa Birth Defects Registry RESULTS: Of 40,762 women tested, 175 were screen positive for T18 (043%) Fourteen samples from outside the state were excluded, leavrug 161 cases completely followed-up through amnlocentesls (n=121), birth certificate data (n=34), telephone contact (n=2), or an ultrasound lndlcatrug non-viable gestation (n=4) Of 121 that had ammocentesls, 119 had a normal karyotype, and 2 had an abnormal karyotype 69XXY

and 47XY+18, Of 36 that declined amnlocentesls, none had a chromosome abnormality on neonatal examination Twenty-seven of the 103 patients who had an ultrasound at the Umverslty had a subtle fetal abnormality or growth alteration, Both cases with abnormal karyotypes were In this group An addltlonal 7 cases without the T18 pattern were diagnosed either by amnlocentesls performed for increased Down risk by serum screening (n=l), elevated AFP (n=l), advanced maternal age (n=2) with serum screening drawn colncldentally, or postnatally (n=3) Triple marker values for these 7 cases are shown below Three of the 7 cases had early second trimester ultrasounds and all 3 showed abnormahtles

Patient No--~ 1 2 3 4 5 6 7

AFP(MoM) 0 23 5 60 0 78 0 98 0 92 0 64 0 90

uE3 (MUM) 0 4l 0 52 0 24 0 60 0 13 0 22 0 98

HCG (MUM) 1 20 0 10 0 76 0 34 0 15 1 00 0 34

CONCLUSIONS: The detection rate of T18 among patients offered amnlocentesls was significantly lower (p<0 05) than the expected rate

(10/161 based on pubhshed data) Combining serum screening with ultrasound to detect subtle abnormalities may improve predictive value by more precisely targeting ammocentesis for those at highest risk of T18

478 THE ROLE OF FETAL KARYOTYPING FROM UNCONVENTIONAL SOURCES

TG Teoh~, G Ryan, JM Johnson, EJT Winsor’, RJ Morrow University of Toronto Perinatal Complex, Toronto, ON, CANADA.

OBJECTIVE: To determine whether fetal specimens including Neural, ascitic, pericardial & cystic hygroma fluids or urine are suitable sources for accurate, rap~d chromosomal analysis. STUDY DESIGN: Twenty-eight samples of fetal fluid: pleural (7), ascetic (4), pericard~al (1), cyshc hygroma (7) and urine (9)were cultured and analyzed using standard techniques for peripheral blood [ymphocytes (pleural, hygroma & ascitic fluids) orfor amniotic fluid (fetal urine & hygroma fluids). These samples were obtained as part of the obstetric investigation & management of these cases. Back-up samples from conventional sources (blood, CVS & amniotic fluid) were also obtained on most patients. RESULTS: A karyotype was successfully obtained in 25/28 samples. Cells were harvested from all pleural, ascitic and 1 hygroma fluid in 2 - 4 days and from the fetal urine and remaining 6 hygroma samples in 7 - 12 days. Three cultures were unsuccessful: pericardial (1), ascitic (1) & urine (1), but all of these samples were suboptimal. Samples with high lymphocyte counts yielded results as quickly & reliably as fetal blood. CONCLUSION: The use of’alternative’ fetal fluids for karyotyping may be considered when amniotic fluid or fetal blood is difficult to obtain. In selected cases this approach avoids the unnecessan] risk of additional invasive procedures, whose sole purpose is to obtain a karyotype.

480 PCR DETERMINATION OF RH C/c/E BLOOD TYPE: AN

EVALUATION OF ACCURACY. J Yankowltz Dept of OB/GYN,

University of Iowa College of Medicine, Iowa City, Iowa

Objective Polymerase chain reaction (PCR) amphflcatlon of a

portion of the RhC/c/E/e gene could lead to a rapid, accurate determlnatmn of fetal RhC/c/E status by amnlocentesls Extensive

evaluation of PCR to determine RhC/c/E type has not been reported The

purpose of this study was to evaluate the accuracy of this technique by

testing a large number of DNA samples derived from individuals whose

RhC/c/E status was estabhshed by standard serologic method

Methods Samples were obtained from Centre d’Etude du

Polymorphlsme Humaln (CEPH) families used for studies of genetic

variation (n=659) DNA was extracted by standard techmques With few

modifications, published primers, reaction and electrophoresls

conditions were used

Results Rh C-positive (n=479), Rh C-negative (n=176), Rh C-

positive (n=524), Rh c-negative (n=131), Rh E positive (n=131) and Rh

E negative (n=528) samples were evaluated The sensitivity and

speclflcy of PCR for ~dentlf}qn~g the RhC/c/E antlt~ens as shown below

Antlgen~ RhC Rhc RhE

Sensitivity 98 3 98 1 96 9

Specificity 91 5 94 7 99 2

Conclusions Based on our data it would appear that use of PCR to

establish RhC/c/E type could aid in evaluation of RhC/c/E sensltazatlon

We are concerned about the instances of antigen positive lndlwduals

characterized as antigen negative This could be due to a polymorphlsm,

mutation or a data coding error The CEPH database is known to contain

such errors at a rate that may surpass the error rate of our testing A

second molecular technique could be used to achieve better accuracy In

the ascertainment of Rh Cle/E type These results are consistent with

pubhshed data evaluating the accuracy of serologic typing by blood

banks Based on review of our patient population, molecular DNA

techniques now available can aid the management of erythrocyte

allotmmumzatlon In pregnancy in over 70% of cases

442 SPO Abstracts .January 1996 Am J Obstet Gynecol

481 IS AMNIOCENTESIS NEEDED TO DETECT SEX CHROMOSOME ABNORMALITIES WHEN MATERNAL

SERUM ALPHA FETOPROTEIN IS ELEVATED? A Fultonx, J_

Yankowttz, R Wilhamsonx, S Grantx, A Rijhs~nghanl University of

Iowa College of Medicine, Iowa City, Iowa OBJECTIVE: To determrue whether elevated maternal serum alphafetoproteln (MSAFP) ~s predlcttve of fetal sex chromosome abnormahtles (SCA), and to determrue whether ammocentests ts necessary with a normal ultrasound and an elevated MSAFP STUDY DESIGN: We evaluated all cases (n=25) of non-Turner, non- mosaic SCA (47, XXX (n=9), 47, XXY (n-~), 47, XYY (n=6), 48, XXXY (n=l), 48, XXYY (n=l)) determined antenatally or postnatally and whose mother had e~ther MSAFP or triple analyte teshng Data were obtained through the Iowa B~rth Defects Regtstry, the Iowa Expanded Serum Screemng Program, the Umv of Iowa Depts of Cytogenet~cs, Genehcs and prenatal testing done in the Dept of OB/GYN (1982-1994) RESULTS: When a karyotype obtained for reasons not related to serum testing revealed a SCA, the mean MSAFP was 1.05 MoM (advanced maternal age-AMA (n=12), postnatal karyotype (n=l), family history of Down syndrome (n=l), polyhydrammos (n=l), previous Down syndrome (n=l)) One of these 16 cases had an elevated MSAFP (3 03 MoM) SCAs were found in 1817290 amniocenteses done for AMA (1/405). F~ve SCA were detected in the 1062 amniocenteses done for elevated MSAFP and a normal ultrasound (1/202). The mean MSAFP was 2 56 MoM Four were detected in 1938 amniocenteses for an increased Downs risk on serum screening (1/485) The mean MSAFP was 1 30 MoM The mean age of the pataents who had an amnlocentes~s for AMA and for elevated MSAFP was 37 4 years and 26 6 respectively CONCLUSION: Based on the mean age, the expected rate of the three most common SCA would be 1/714 m the elevated MSAFP group The actual vs expected rate was not statistically different for the elevated MSAFP (p>0.2) group. Though there ts a trend toward a higher rate of SCA w~th elevated MSAFP, it is a poor predictor of SCA and therefore an ammocentes~s is not reqmred if the ultrasound is normal

483 GENETIC COUNSELING IN A LOW LITERACY, LOW SOCIO- ECONOMIC SETriNG DECREASES MATERNAL STRESS AND IN- CREASES MATERNAL KNOWLEDGE, INDEPENDENT OF THE MOTHER’S DECISION TO ACCEPT OR DECLINE PETAL TESTING. V. Ranoaoort. Z Tatsugawa*, M. Broder*, K. Blazer*, Dept ofOb/Gyn, Olive View- UCLA Medical Center, Sylmar, California OBJECTIVE-The role of genetic counseling m the prenatal setting is to impart complicated technologic and genetic information so the patient can make a personal and informed decision about risks and benefits of fetal testing. This study addresses the efficacy and benefits of genetic counseling in meeting these goals in a population with low literacy and m which fetal testing is often declined due to cultural or religious beliefs regardless of risk. STUDY DESIGN- 90 women referred for prenatal diagnosis due to maternal age were enrolled. All women were low literacy,(less than high school education), primarily Hispanic. Of this group 65% accepted amniocentesis. Three structured interviews were conducted, precounseling (PreQ), immediately post counseling (PostQ) and at 1-2 months following the visit (F/U), The interview assessed knowledge of birth defects, chromosomes, Down syndrome, ultrasound and amnioccntesis, attitudes towards decision making and fetal testing and perccptiun of the counseling experience. Stress levels were assessed by the Spielberger test as well as quantitative stress rating by patient report. RESULTS-Knowledge scores significantly increased from the PreQ to the Post Q. Post Q scores were equivalent to a control group of prenatal care providers. Knowledge gain persisted on FAL Stress scores decreased significantly from the PreQ to F/U both for women who accepted amnioccntesis as well as those that declined. Women declining tested reported decreased pregnancy related stress after counseling. Women accepting testing reported increased stress until receiving results then decreased stress. All women in this study perceived the counseling session as beneficial and valuable. All would recommend genetic counseling to a friend. CONCLUSION- Genetic counseling is effective in imparting information in a low literacy, low socioeconomic population. Patients perceived the counseling as beneficial and had decreased pregnancy related stress levels whether or not they accepted fetal testing.

482 GENETIC AMNIOCENTESIS (AMN) FOLLOWING

MULTIFETAL PREGNANCY REDUCTION (MFPR) DOES

NOT INCREASE THE RISK OF PREGNANCY LOSS. LK

McLean. MI Evans, RJ Carpenter, MP Johnson, JD Goldberg.

Perinatal Associates N. CA, Sacramento, OB/Gyn,

Hutzel/WSU, Detroit, St. Lukes, Houston, UC, San

Francisco. OBJECTIVE: A previous publication suggested the risks of AMN were increased following MFPR. We sought to verify in a mud~ larger study. DESIGN’ A collaborative, retrospective study of patients who had both AMN and MFPR at 3 large centers for gestations that had been reduced to twins. MFPR was performed from 8-13 weeks. Amnios were performed between 13 and 20 wks. Collaborative data of MFPR patients who, to the best of our knowledge, did not have AMN were used as controls Pregnancy losses were defined as <24 wks. RESULTS: 76 patients had MFPR and subsequent AMN (1-sext, 3- quint, 23-quad, 49-triplet) Our collaborative group of 1789 pregnancies (SGI 95) minus th~se patients gave controls who were divided by all patients and for those reduced to twins onl

Total Losses <24 wks Loss %’ MFPH + Amn~o 76 4 5.26% MFPR oniy 1713 205 11.97% MFPR to twios anly 1359 151 11.11%

x~=1.3 p=hIS

D[SCUSSION: M~PR pts are probably the appropriate control group, not uncomplicated twins or sL~g]etons Therefore, our data suggest ])A3V’LN following MFPR does not have an increased risk of loss as compaz~ to other MFPR patients; 2)AMN can be sMe]y offere~ to MFPR patients as appropriate for their genetic indications, irrespective of ]VLPPR status.

484 PRENATAL DETERMINATION OF FETAL RhD TYPE BY

DNA AMPLIFICATION FROM TRANSCERVICAL SWABS

A Karimi, B Shahbahram~ x, j. Felix~, B W. Kovacs

University of Southern California School of Medicine, Los Angeles, CA

OBJECTIVE. Our primary goal was to develop a stmple non ~nvasive

method for early prenatal diagnosis. We sought to utthze deganeratmg vilh

that are shed into the endocervix as a source of fetal DNA for purpose of

molecular genetic analys~s.

STUDY DESIGN’ Fifteen RhD negative pregnant women at vataous

gestational ages presented to prenatal clinic of LAC+USC medical center

were enrolled in thin study. A method similar to pap smear was utdized to

collect mid cervical secretion within 2 cm of external Os. Cervical secretions

collected wtth cotton swab were fractionated by using a method to eliminate

spermatozoa DNA contamination, by separating spermatozoa from epithehal

fraction The PCR method was used on eptthelial cell fraction to analyze the

fetal RhD gone and a Y specific sequence These results were tabulated

against actual blood group and chnical gender of neonates after delivery for

accuracy specifietty and sensitivity of this method

RESULTS: We were able to obtmn trophoblastic DNA evidenced by RhD

positive results from RhD negative pregnant mothers. The identification of Y

specific sequence from the epithelial fraction also confirmed the

trophoblasttc origin of the DNA There were 8 RhD positive fetus by PCR

(53%), sensitivity 50% with zero false postttve results, specificity 100%.

Fourteen samples were identified as being Y positive by PCR and six of

these were confirmed male newborns (53% false positive) Collectavely five

of the 15 samples were correctly identified for both RhD and Y positiwty

(33%).The non mvasivenass of this method was confirmed by lack of

appearance of RbD antibody in our patients.

CONCLUSIONS: Fetal spectfic DNA sequence can be identified through

simple cervical sampling by utilization of PCR technology. TMs study

demonstrates the feasibihty for this method,

Volume 174, Number 1, Part 2 SPO Abstracts 443 Ain J Obstet Gynecol

485 PRGSPECT|VE EVALUATION OF EARLY MID-TRIMESTER

AMNIOCENTESIS. R S~lvar., T Russell’, T Mullenx, M Kembiehx, E

Leeth~, B Helfandx, S MacGregor, J Shell Dept. Oh/Gyn, Northwestern

University Medical School, Evanston Hospital, Evanston, Ilhnois

OBJECTIVE: To sva~uate the effmacy and safety of amniocentems

performed prior to 16 weeks of gestation among physicians with

vaned annual procedure volurfle.

STUDY DESIGN: Genetm ammocenteses performed between 1/92

and 12/94 were ascertained prospectwely, then divtded into 13-14

week tEA) and 15-20 week (SA) cases. Details of each procedure

were recorded and pregnancy outcomes were retrieved via a

questionnaire completed by the dehvering physician. Sampling efficacy

and pregnancy outcome were evaluated for EA end

relation to operator volume (defined by a breakpoint of < or

procedures during the study ~terva~l.

RESULTS: 900 of the 1025 amniocenteses were analyzed (lost to

follow-up, n = 25; undehvered, n = 100). Only 6 of the 46 physicians

exceeded 50 procedures during the study (Group A), compared to 40

with lesser experience (Group B). Sampling efficacy and rate of

spontaneous abortmn (SAB) stratified by physician experience and

gestational age at sampling are shown below:

Group A ~roup B

EA SA EA SA Amnio # 148 412 45 295 Single-pass (%1 98.0 94.1 88.3 86.2* SAIl t%l 0.4 0.6 2,2 0.3#

* p<.01 for both EA & SA comparing Group A vs B # p=O.1 for EA vs SA In Group B

More frequent use of continuous ultrasound guidance occurred in

Group A compared to Group 8, (64.6% vs 37.4%, p<.O01).

CONCLUSIONS: Operator experience directly influences sampling

efficacy for both EA and SA. Evaluation of SAB rate among less

experienced physicians will require a larger EA sample size.

487 EARLY AMNIOCENTESIS: OUTCOME, RISKS AND

COMPLICATIONS. H. Brown, L Padilla, K Palmerx ,M. Daveex

Department of Obstetrics and Gynecolgy and Medical and Molecular

Genetics. Indiana University Medical Cemer, Indtanapo[is, IN~

OBJECTIVE: The purpose of this report is to evaluate comphcations

and pregnancy outcome experience in early genetic amniocentesls.

STUDY DESIGN: We compared amniocentesis related compbcafions

and pregnancy outcome in singleton gestation of 348 women

undergmng genetm ammocentesis at 11 to 13 9 weeks (Group I) to 379 women undergoing anmiocentesis at 14 to 14 9 weeks (Group II) All

procedures were performed w~th ultrasound guidance using a 22 guage

needle betwee, n June 1989 and March 1995 The primary indication

for anmiocentesis was advanced maternal age or prior aneuploidy.

Comphcat~ons evaluated included cramping, amniotic fluid leakage,

bleeding and fetal loss prtor to 20 weeks. Analys~s was by X2 w~th

significance at the 0,05 level.

RESULTS: Early Group I procedures included 56 at 11 weeks, 106 at

12 weeks and 186 at 13 weeks. Leakage of fluid within 24 to 48 hours

occurred in 2a- (6.8%) of Group I and 18 (4.7%) of Group II

procedures(p~NS). Procedure related fetal loss (spontaneous abortion

or demise) occurred m 9 (2.6%) in Group I versus 7 (1.5%) of Group II

(p=NS). Of women with leakage of fired, 2 (,08 %) in Group I and 1 in

Group I1 (06%) experienced losses. Twenty-six women m both groups

experienced cramping. No woman w~th cramping had a loss. The rate

of bleeding, preterm dehvery and aneuploidy was s~milar for both gmups. CONCLUSIONS: Counsethng for early ammocentesis should reflect

a higher procedure related complication rate than commonly quoted for procedures preformed after 15 weeks.

486 EARLY AMNIOCENTESIS IN T~VIN AND TRIPLET PATIENTS CONSIDERING SELECTIVE EMBRYO REDUCTION. J. Nev. K. Kinnebergx, D. Brandtx, S. Baldinger’, B. Hatten". Abbott- Northwestern Hospital. Minneapolis, MN Obj, e~tive: T~ de~.ermi~ if early amniocentesis is a prac’6cat option to reliably choose abnormal embryos for selective embryo reduction (SER}. Study Design: Between August, 1992 and June, 1995, 16 consecutive twin and 22 triplet patients considering genetic studies for advanced maternal age or for an abnormal ultrasound were also considering embryo reduct=on. The study group was offered eady amniocentesis. The success of sampling, amount of amniotic fluid obtained, culture success, accuracy of results, and complications were recorded, Results: In all 97 attempted ammocenteses (32 twins, 65

triplets) at least 8 cc’s of amniotic fluid were obtained (range 8- 27). There were no culture failures~ Three fetuses with tosomles were detected and all were confirmed at SER. One twin patient with normal results choose termination of pregnancy. One patient had a failed cerclage at 21 weeks. One twin and one triplet d~ed. One patient chose SER of a twin with a large omphalocele. All the other patients with abnormal ultrasounds and normal karyotypes continued the multiple gestation. Eleven triplets w~th normal results reduced to twins, eight continued. The mean gestational age of the 25 ongoing twins was 37 weeks (range of 30-39) and of the 7 ongoing triplets was 32 weeks (range of 27- 36L One triplet patient experienced leakage of fluid after amniocentesis, declined SER and miscarried. No chromosomal discordances were present at birth. Conclusions: Early genetic amniocentesis in twins and triplets can accurately guide SER decisions.

488 IS T~ERE ~l OPIIMUM TIME FOR EARLY AMNIOCENTESIS? ~1 Blacks~ne.x M G Pme~e, Y. Pan, S G Pme~e, J M~chaud Dept 0NGyn, Marne Med=cal Cent, P~nd, ME OBJECTIVE: To determine rates of comphcahons assooated with early ~s ~ o~ ~e~nal ~ a~ ewluate ~ ~uen~ of re-a~ts and r~¢=ng STUDY DEIGN: Fr~ Ju~, 1989 b Dec~ 1993, 927 early amm~nt~s (1~.7 ~s ~ 15.9 weeks gesb~onal age) w~e ~rfo~ed and 903 wegnan~

RESULTS: ~e w~e 19 ~r~al abn~ahbes (2.1%) ~t~ed In ~e 903 pabents. Pregnancy complica~ons associated Wl~ amniocente~s ~curred ~n ~ ~ks a~ amm~ent~s =n 14 pahenb (1 ~%) The ~an~n of comph~bons occurring =n ~ch ges~onal week showed ~at the 12 0-12 ~ ~eks group h~ ~pli~on ra~ of 1 20% (~), the ~3 0-13 9 w~ had the h=ghe~ mc=~nce of c~ph=~ons at 2 68% (~298, P<0.01), ~e 14 0-14 9 ~ks had ~e least ~=~n~ of 0 5% (1/1 ~, P<0 01), and ~e 15 0-15.9 ~ks had an ~nc=~nce ra~ of 1 20% (~166) C~ph~ons less ~an 14 0 ~ks ~e 1~ (2,16%) ~. ~9 (0 ~o/~) ~ ~ ~ 14 0 ~ks (P<G,05) Oa~ ~ om o~ e~rience c~b~n~ and c~par~ to ~at already pubh~ed =s ~nmstent ~th ~e lowest ~ce of we~nancy loses ~o w~ ~st-wo~dure and ~re 28 0 w~ks in ~e 14 0.14 9 w~k group (1 4% ~ 2 6% ~ 2.1%) Thee ~re 25 re-a~t W~dwes out of ~e 903 amn~nte~s (2 77%) ~ 10~56 (3 90%) ~fue 13 0 ~ks of ge~a~onal age, 1 ~298 (4.03%) dunng 13.0-13 9 ~ks, and ~ (1.~%) ~fing 14 0-1~ 9 ~ks (P>0 05) Thi~-~ Wo~ ~e affempt~, but re-~heduled (3 70%), due ~ e=~ o~rlymg maternal ~1 ~ bla~, or un~ m~anes Thee ~re 2~ (8 98%) (P<0 01) ~f~e 13 0 weeks of get,anal age, 8~98 (2 68%) during 13 0-13 9 weeks, and 2/183 (1,09%) d~ing ~4 0-14 9 ~ CONCLUSION: Early amn=~nte~s ~s a ~fe and valuable w~dure, and ~y ~ ~fely ~fl=m~ as early as t~hm~lly posmble Al~ough ~e d~ff~ences ~re ~all, ~ese resul~ suggest ~at amm~entes=s may have few~ ~mphca~ons ~en ~ afl~ 14.0 ~eks Like.e, there =s a ~end ~r~ ~e ~an one punc~e prior to 14.0 ~eks F=nally, ~e ne~ to re-schedule ~e procedure =s m~ea~d ~fue 13 0 ~eks

444 SPO Abstracts ]anuar~ 1996 Arn J Obstet Gynecol

489 GENETIC DIAGNOSIS FROM FORMALIN-FIXED FETAL TISSUE USING FISH: A NEW TOOL FOR

GENETIC COUNSELING IN SUBSEQUENT PREGNANCIES. MD Fejgm~_D Kadron×, R Tepper~, Y

Beyth×, A Armelx Depts OB-GYN, Pathology & Genetxcs, Sapxr Med. Cntr, Kfar Saba, Israel

OBJECTIVE: The feasihihty of retrospective genetic

testing for numencal chromosomal aberrations was evaluated by applying FISH techmque to formahn-fixed fetal and placental tissues

STUDY DESIGN: Following a process of

deparaffiruzatmn and preparation, fetal or placental tissue

from rune old cases with known anueploldy and from 13 cases w~th known fetal malformatmns conunon m

aneuplo~dy, were tested w~th specific DNA probes for pencentromenc repeat re~ons of chromosomes 13/21, 18, X

and Y

RESULTS: In 8 of 9 cases with sufficaent cells, FISH

&agnos~s concurred with the known karyotype Numerical

aberratmn was &agnosed m 6 of 13 suspicaous cases wtth

fetal malformatmns (3 cases with XO and 3 cases w~th

+13/21)

CONCLUSIONS: We have shown that FISH techruque can

be utibzed to re~rospechvely evaluate fonnahn-fixed fetal

and placental tissue in cases in which karyotypmg was not

performed and when genetxc reformation Is desired

491 OBSERVED DETECTION RATES FOR IDENTIFICATION OF CHROMOSOME ABNORMALITIES USING CHROMOSOME SPECIFIC

PROBES WITH FLUORESCENCE IN SITU HYBRIDIZATION B~ ~Ward*. M Wright*, MP Carelh*. Integrated Genetics, Frarmngham, MA. OBJECTIVE: Rap~d ~dentification of numerical abnormalities for chromosomes 13,18,21,X&Y in uncultured amniocytes ~s possible using

FISH with chromosome specific DNA probes. However, this rapid detecUon procedure is not designed to detect other chromosomal abnormalines. In an

international, multicenter, prenatal diagnos~s cytogenetics survey, the estimated proportlon of cytogenetic abnormalmes which could have been detected by FISH ranged from 53% to 84%, averaging 66.8%. We here present the actual proportion of chromosome abnormalitxes identified by

FISH in our Molecular Genetics Laboratoff. STUDY DESIGN: Detection rates were calculated from 5,000 consecutive samples referred for rapid aneuploidy analysis by FISH performed as an adjunctive test to cytogenetics. The detection rate for aneuploidies by FISH was compared to the rate of aneuploidy diagnosed by cytogenetics and to the rate of all cytogenetlc abnormahties, including balanced and unbalanced structural abnormalities. RESULTS: A total of 287 specimens were chromosomally abnormal

(5 7%). There were 28 unbalanced structural chromosome abnormalities, !5 balanced abnormalities and 244 aneuploidies FISH analysis identified and reported 217 of 244 aneuploidtes (89% aneuploid detection). The remaimng 27 aneuploidies were "Uninformative" by FISH as they did not meet

reporting criteria. For the total population of chromosome abnormahties, the FISH detection rate was 76%. For the subpopulation of unbalanced chromosome abnormalihes (aneuplo~dies plus unbalanced structural

abnonnahties), the FISH detection rate was 80%. CONCLUSION: Over three-quarters of all chromosomal abnormalities

were detected by FISH using probes for chromosomes 13,18,21,X&Y. Although FISH cannot replace complete cytogenetic analys~s, it is useful for the rapid identification of chromosomal abnormalilies m high risk silnations.

490 THE USE OF FISH FOR ABNORMAL PRENATAL ULTRASOUND

FINDINGS: A FOUR-YEAR EXPERIENCE. Kirk j~l, Harper CEx~, SIoan CS=, Lorenz RP=, Comstock CH~, Lee Wt, Smith RS~, Yon Oeyen PI"=. DMsions oU Fetal Imaging and = Reproductive Genetics, Dept. OB- GYN, W~liam Beaumont Hospital, Royal Oak, MI. ONect]ve: To determine the usefulness of fluorescence in situ hybridization (FISH) analysis for the rapid detection of common chromosomal aneupioidies after abnormaICdes were detected by prenatal ultrasound. Study Design: From 9/91 to 7/95, we had 94 patients who underwent rapid aneupioidy analysis of chromosomes 13, 18, 21, X, and Y by FISH (Integrated Genetics, Framingham, MA) because an abnormal~ was detected by prenatal ultrasound. Analysis was performed on amniotic fluid except one case of cystic hygroma fluid. Records were reviewed to determine ultrasound findings, FISH results, standard cytogenetics analysis, and pregnancy outcomes. Results: FISH was performed at 15 3/7 to 39 2/7 weeks with 84% of patients less than 24 weeks. The most common ultrasound finding prompting a FISH was an abnormal heart (N = 42). FISH was usually performed within 1 day of the ultrasound diagnosis. FISH results were usually reported to us within 2 or 3 days of amnlocentesis. No numerical abnotma~es for chromosomes 13, 18, 21, K or Y were detected by FISH in 57 samples; karyotypes confirmed 30 male and 27 female fetuses were chromosomally normal. FISH correctb/iden~ed 27 aneuploidies: 9 trisomy 18 (1 also had XXX), 6 bisomy 21,4 monosomy X, 4 triploidy, and 4 trisomy 13. Nine specimens were uninformative. Katyotype analysis showed 1 case not detected by FISH. 46,XX/46,XX,-1,+isod=c(1)(1qter- >1p36.3:1p13->lqter) Thirty-five patients chose to terminate after FISH results were reported. Conclusion: Although standard cytogenet~c analysis is necessary to detect mosaJcism, deletions, and translocations, FISH provides a rapid and reliable determination of the most common chromosome aneupioidies associatedwlth abnormal prenatal ultrasound findings. This inform~on can be used to counsel patients m a timely manner regarding fetal prognosis and pregnancy management.

492 PROSPECTIVE EVALUATION OF THE ANTENATAL INCIDENCE

AND POSTNATAL SIGNIFICANCE OF FETAL ECHOGENIC

CARDIAC PAPILLARY MUSCLES, GA Dildy, VE Judd~, SL Clark. Departments of OB/GYN and Pediatrics, Umvemity of Utah School of Medicine, and IHC P~anatal Centers, Salt Lake City, Utah.

Objective: To evaluate the antenatal incidence and postnatal s~gnificance of echogemc cardmc papillary muscle (ECPM) observed during antenatal ultrasonography (U/S)

Materials and Methods: Between 12/6/94 and 4114/95, all women undergoing "routine" U/S between 16.0 and 24.9 weeks at one refelral center (UVRMC) were prospectively evaluated for ECPM of the left ventricle (LV) or right ventricle (RV). Referrals for maternal or fetal

comphcattons were excluded. Postnatal echocar&ography was performed for those identified with ECPM.

Results: 506 consecutive fetuses were evaluated at a mean (~: SD) gestatxonal age (GA) of 20.6 (:t: 1.6) weeks. There were 25 (4.9%) fetuses found to have ECPM (LV = 19 and RV = 6). Echocardmgraphy was

performed between 0 3 and 20.1 weeks postdehvery. Postnatally, 12 LV echogemc papillary muscles and 3 echogenic LV chordee were identified; there were no postnatal RV findings. There were no cases of mtracardmc

tumor or myocardial dysfunction; one neonate had minor structural malformations. C~n~ols (n = 50) were selected from t~ormais in the general

group for comparison to the study group. There were no signtfi.cant differences in maternal age, grav~d~ty, parity, GA at U/S, GA at delivery,

or 5-minute Apgars. A s~gmficant difference was observed in birthweight (BW) between the control and study groups (3.5 vs 3.0 Kg, Mann-Whitney U test, P = 0.002). Tins difference persisted after correcting for GA, although all infants m both groups born > 37 weeks weighed > 2.5 Kg. Conclusions: The incidence of ECPM during routine madtnmester U/S ~s

4.9%. As an isolated finding, ECPM may be associated with a statistically signnqcant but chmcally msigmficant decrease in BW, but should be considered a variant of normal not warranting foilowup clinical evaluation.

\:olmne 174, Number 1, Part :2 SPO Abstracts 445 Ant.] Obstet Gynecol

493 METABOLISM OF COCAINE BY RAT FETAL BRAIN IN VITRO. B.__~_B

Little,XA. Barneax, N Aguda-Mansilla,~ D K Ramsey,~ P Kemp x Dept Ob/gyn, LIT Southwestern Medical Center, Dallas, Tx and Chief Medical Examiner’s

Office, Oklahoma, Oklahoma City, OK

Objective: To analyze the ability of the rat fetal brain at 20 days gestation to

hydrolyze cocaine to bcnzoylecgonme (BZE) and ecgonlne methyl ester (EVE) m

vitro

Study Design: Rat fetal brains were obtained fresh and cultured m medium

Metabolism of Cocaine by Fetal Rat Brain 4

s

.~_

o 72 Time [hours)

Methods: Media was analyzed by GC/MS for cocaine and the drug’s metabohtes

Results: Cocaine was hydroIyzed primarily to BZE, wth only trace amoants of

EVE detected

Conclusions: Cocaine was transported into rat fetal neurons, and metabohzed to

BZE Therefore, near term (20 days gestation) rat fetal brains ha~e the ability to

uptake and btotmnsform cocaine, and secrete BZE m vitro

495 INCREASED FAMILIAL INCIDENCE OF MULTIPLE

SCLEROSIS: GENETICS AND EPIDEMIOLOGY IN

PREGNANCY. Schneider .1", Blea C, Hendricks SK. U Wisconsin,

Madison.

BACKGROUND/OBIECTIVE: Prior studies on multiple sclerosis (MS)

and pregnancy have focused on the neurologJc impact of pregnancy on

MS, Our objectives. 1) to define the effects of MS on pregnancy and 2)

to define the genetic risk to family members of MS patients.

STUDY DESIGN: 39 MS patients had 85 pregnancies. Complications

(PTL, PIH, SABs, TABs, infertility, neonatal deaths {ND}, stillbirths

{SB}) were compiled and compared to National and State of Wisconsin

statistics The incidence of ante- (AP), intra- and postpartum (PP)

exacerbations and remissions was evaluated A family pedigree was

taken Fisher’s Exact and Xz with Yates correction were used for analysis.

RESULTS: In 85 pregnancies, there was a 20% SAB rate (n=17; p<0 05)

with 68 ongoing pregnancies Of 37 remissions, 33 were antepartum

(p<0.01) 47 patients had exacerbations, PP exacerbations (33) were

>AP (14) (p<0 05) Rate of PTD (13 6%) and congenital anomalies (7%)

were elevated in comparison to national and state statistics. (p<0.01)

There were no significant increases in infertlhty, PIH, ND, TABs or SB

14 of 39 patients (35 8%) had strong family histories of MS; 11 of 14

(79%) were first or second degree relatives Onset and severity of

mamfestations were ~imdar m first and second degree relatives; less

concordance was seen with increasing distance of kinship.

CONCLUSIONS: MS in pregnancy is associated with a siguificant

increase in the pregnancy-related complications of preterm labor,

preterm birth, and spontaneous abortions. The incidence of congenital

anomalies is greater in infants born of MS mothers. Family members,

specifically first- and second- degree relatives, have an increased (but

currently undefineable) risk of developing the disorder The incidence

seen m our population is consistent w~th a polygemc disorder and xs

likely hnked to the HLA system Genetic counseling should include a

discussion of these risks and available diagnoses and interventions

494 A PRIMARY MALFORMATION OF THE AMNIOTIC SAC MAY CAUSE LIMB BODY WALL COMPLEX C. Gr~tv~n_~x, K. Ward1,2. Depts. of Ob-Gyn~ and Human Genetics2, University of Utah School of Medicine, Salt Lake City, UT. OBJECTIVE: While some cases of Limb Body Wail Complex (LBWC) are caused by amniotic bands or vascular disruption, we describe a d=shnct subset of LBWC, define its associated anomalies, and propose a pathogenesis based on a malformation of the amniotic membranes. STUDY DESIGN: A fetus wtth LBWC was included in thie study

if it had abdominosch~sis, a limb defect, and amnion continuous with the skin at the s~te of the abdominal wall defect. Four cases were selected from our inshtution and 15 were selected from the Merature. RESULTS: All 19 cases showed agenesis of the umbilical cord with displacement of the abdominal viscera into the extra-

embryonic coelomic space. All had malrotation ot the bowel, and 72% had anal atresia. Urinary abnormalities were frequent: bladder agenes~s ~n 50% and extrophy of the cloaca in 44%. Genital malformations were seen in 74%. Nine of 19 had a meningocele. CONCLUSION: In these cases, malformations were fimited to the structures whtch develop in the area of the yolk sac, cloaca and connecting stalk. There appears to be a primary failure of the membranes to grow and envelop the embryo. The membranes did not adhere to the connecting stalk to form a distinct umbilical cord.

Agenesis of the cord suggests that this malformation occurred early in gastrulation, between days 22 to 28.

496 AN INCREASE IN CONFINED PLACENTAL MOSAICISM IN PREGNANCIES EXPOSED TOSMOKE.~K |’R&TT, S DR(/STI’,, l’, PRI’~NFI(’I’~*, SQ WU* {t WISCONSIN MAI)~SON BA(’KGROUND/OIHIZ’TIVE Maltreat t~garcne ~mokmg h,t,

SI UI)Y I)h’SI(;N 12 paucut, ~ho ~mokc and 18 non,making ctmtrol, wc~c


497 THE ASSOCIATION OF MATERNAL OBESITY AND

ISOLATED MAJOR FETAL CONGENITAL CARDIAC

ANOMALIES IN AFRICAN-AMERICAN WOMEN. Mikha~ll LN×, Mittendorf R, Walker CK×. Department of Obstetrics and

Gynecology, The Umversity of Chicago, Chicago, IL

OBJECTIVE: To determine whether or not an unconfounded

association exists between maternal obesity and congemtal anomahes

among non-diabettc African-American women

STUDY DESIGN: We compared the body mass indices (BMI = kg/m2) of paras w~th anomalous fetuses (cases) with BMIs of

randomly selected paras with normal fetuses (controls) born from

1981-1994 at the Chicago Lying-in Hospital. Women were excluded because of" non-African-American race, age >35 years, d~abetes of

any type, muhfetal gestation, maternal seizure or psychiatric

disorders, radiation exposure, TORCH infection, or ethanol abuse.

Obesity was defined as BMI >_ 27.

RESULTS: Of the 130 cases, 63 were designated major anomalies

and 67 were minor. There were 144 controls. Compared to non-

obese women without diabetes, obese women without diabetes were s~gnificantly more likely to have a fetus with an isolated major fetal

cardiac anomaly (OR=6 5, 95% Confidence Interval 1.2, 34.9,

p=0.025). Obese women without diabetes were also significantly more likely to have fetuses with minor anomalies (OR=2.2, 95% CI

1.2-4,1, p=0 008).

CONCLUSIONS: Our data support the hypothesis of an association between maternal obesity in non-diabetic African-American women

and isolated major fetal cardiac anomalies Prenatal care standards

may need to be altered to include a second-trimester level II

ultrasound for obese African-American women.

499 FEMALE SENSITIZATION WITH ANTIBODIES KNOWN TO CAUSE HEMOLYTIC DISEASE. Ossie Ge~fman-Holtzman Elem Kosmasx, Martha Wojtowycz~ Genetics & Maternal-Fetal Medicine Division, Department of Ob/Gyn, SUNY Health Science Center, Syracuse, New York OBJECTIVE Developments in molecular technology enable prenatal determination of fetal Rh status by DNA analysis, potentially avoiding multiple invas~ve procedures. We sought to determine the current frequency of rod blood cell (RBC) antJgen sensrdzat~ons which are capable of causing hemolytJc disease and would be candidates for prenatal DNA studies STUDY DESIGN" We rewewed blood bank and medical records at a single large tertiary center for ~den~cation of patients with a pos~t=ve anti]body screen between January 1993 and June 1995. Data were registered using Epi Info (version 6) and analyzed based on gender and spec~c blood group sensitizat=ons. RESULTS We identified 452 (73 4%) females out of 615 pabents (pts) who had a positive antibody scroen E~ghty three (18.3%) of these subjects had an~bodies to more than one blood group antigen The frequency of specific RBC iso~mmun~zation rolevant to the development of fetal hemoly~c d=sease was an~ D (101 pts, 22.3%), anti E (78 pts,17.3%), ant~ Kell (73 pts,16 2%), ant~ c (32 pts,7 1%), anti C (27 pts,6 0%), ant~ MNS (26 pts,5 8%), ant~ (Duffy) (21 pts,4.6%) and anti Jk" (8 pts,1.8%). Thero wero 134 subjects w~th RBC anbbodies not known to cause fetal hemolytic d=sease. CONCLUSIONS: Desp~e the use of rhogam, ant~ D ~s still the most common antibody identical in women presenting to a tertiary care center. However, anti D account for less than 25% of the identified RBC antibodies. Research in prenatal determma’don of fetal antigen E,C,c,Kelt and others should continue as sensi’0zahon to these antigens ~s common

498 MATERNAL URINE SCREENING FOR FETAL DOWN SYNDROME: COMPARISON OF BETA-CORE FRAGMENT, FREE-BETA SUBUNIT, AND NCG. LH. KellnerTM, JA. Canick2x, G.E Palomaka3x, M NeuerTM, D.N. Sailer, Jr.4, R.P WalkeP5x, R. ~)sathanondh6x, A.T Bombard7 1Winthrop-Univ. Hosp, Mmeola, NY, 2Wornen & Infants Hosp, Brown Univ., Providence, RI, 3Foundation for Blood Res, Scarberough, ME, ;$Univ. Rochester Sch. Mad. Dent, Rochester, NY, 5Ciba Corning Diagn, Alameda, CA, 6Brigham & Women’s Hosp., Boston, MA, 7Montefiore Mad. Ctr, Albert Einstein Col. Mad, Bronx, NY. OBJECTIVE: Second trimester measurement of maternal urinary beta-core fragment (urinary gonadotropin peptJde, UGP), the major urinary metabolite of hCG, has been reported as a possible improvement in prenatal screening for fetal Down syndrome. The objective of the present study was to determine whether UGP is a better marker for fetal Down syndmma screening thar~ is urinary free beta subtmlt of hCG or urinary hCG METHOD: We analyzed maternal urine from 91 singleton (control) and 14 Down syndrome (case) pregnancies between 15 and 21 weeks gestation using five different assays one highly specific for UGP Ciba Coming Diagn ; UGP CCD); one wh ch measures beth UGP and free beta subunlt (Wake Dlagn.; UGP Wake), one highly spec#ic for flee beta subunlt of hCG (Bioclona); one which measures intact hCG plus free beta subunit (Abbott, total p hCG); one which measures intact hCG only (Abbott) Gestation-specific control medians for each of the five analytes were determined by weighted non-linear regression, and multiples of the median (MoMs) calculated for cases and controls. Our results on UGP CCD (Prenat. Dtagn. 15 739-44, 1995) are presented here for comparison. RESULTS:

Controls Cases Oetact~on Rate (mean, SD (mean, SD Case at

LA~alyte of log MoM) of !o~ MoM) Median MoM 90th cantile UGP CCD -0 011, 0.344 0.748, 0.195 5 34 93% UGP Wak.o -0 003, 0 335 0 534~ 0 299 3 62 64% free beta 0 025, 0.335 0 382, 0.342 2 61 36%

i totaJ p hCG -0.013, 0 312 0 363, 0 300 2 34 50% intact hCG 0 047, 0 341 0 395, 0 378 2.14 43% ~NCLUSlONS: In this series, UGP was more predictive of fetal Down

syndrome than were other urinary excretion products of hCG. The specific UGP assay provided the best results; measurement of free beta, total hCG, and intact hCG provided results s~m~lar to each other, consistent w~th their effectiveness as serum markers The assay which measured a combination of UGP and free beta provided results Intermediate to those of UGP and the other markers studied These results support the hnding [hat specific measurement of UGP may lead to a major improvement In prenatal Screening for fetal Down syndrome

500 URINARY GONADOTROPIN PEPTIDE (UGP) LEVELS IN PREECLAMPTICANDNORMOTENSIVE WOMEN. MA Williar~sx, DA Luthy, RW Zingheimx, AM ~x, TK Sorensenx, RG Restax. Swedish Medical Center and Dynacare/Laboratory of Pathology, Seattle, WA. OBJECTIVE~ Elevated maternal serum human chorionlc gonadotropin (hCG) is an important marker of Down syndrome. Notably, women with unexplained elevated serum hCG experience a 2 to 4-fold increased risk of preeclampsia (PE). UGP, also known as h-core hCG and thought by some to be a degradation product of hCG, has recently been shown to be elevated in Down syndrome pregnancies. We examined UGP levels in PE and normotensive pregnant women. STUDY DESIGN: We measured UGP levels in urine collected during the third trimester from 5 women with PE and 20 controls. UGP levels were determined usin~an enzyme immunoassay (Ciba-Corning Triton’~) and were standardized by UGP/creatinine ratio. Statistlcal significance testing was done using the Wilcoxon rank sum test statistic. RESULTS~ There was statistically significant elevation in UGP levels among PE cases, as compared to normotensive control subjects (p=0.03). Median UGP levels were 76.7 and 18.8 pmol/mg creatinine for PE cases and controls, respectively. CONCLffSION= These early findings suggest that elevations in UGP may be a risk marker for PE. Additional larger studies are in progress and should further clarify the relation between DGP levels and adverse pregnancy outcomes.

Volmne 174, Number 1, Part 2 SPO Abstracts 447 Am ] Obstet Gynecol

501 APOLIPROTEIN E (apo E) PHENOTYPES AND RISK OF PREECI.~dlPSIA. MA Williamsx, RW Zingheimx, IB King×, TK Sorensen, SM Marcovinax, DA Luthy. Swedish Medical Center, Fred Hutchinson Cancer Research Center, and University of Washington, Seattle, WA. OBJECTIVE: Lipid profiles of pregnant women who subsequently develop preeclampsia (PE) are characterized by elevated triglycerides and cholesterol levels. Apo ~2 and ~4 alleles are markers of hypertriglyceridemia and hypercholesterolemia, respectively; therefore, we sought to examine the association between maternal apo E phenotypes and risk of PE. STUDY DESIGN: Apo E phenotypes were determined by isoelectric focusing in 57 women with PE (meeting ACOG criteria) and 56 pregnant controls. Logistic regression procedures were used to estimate odds ratios (OR) and 95% confidence intervals (CI). RESULTSt The ~2 and ~4 alleles, respectively, were more frequent among PE cases as compared to controls. Women with at least one ~2 allele experienced a 4.4-fold increased risk of PE, compared to women with the ~3/~3 phenotype (OR = 4.4; 95% CI 1.6 - 12.0). PE risk was also positively associated with the ~4 allele (OR = 2.2; 95% CI 0.9 - 5.6). The aseociations remained after adjustments for possible confounding factors. CONCLUSIONS: Apo E alleles, established genetic markers for dyslipidemia and coronary heart disease in men and non-pregnant women, may also be a risk factor for preeclampsia.

503 A PROMOTER MUTATION IN THE TUMOR NECROSIS FACTOR ~ GENE IS NOT ASSOCIATED WITH PREECLAMPSIA. H. MajorTM, D. Dizon-Townson1, K. Ward1,2. Depts. of Ob-Gyn~ and Human Genetics2, Univ. of Utah School of Medicine, Salt Lake C=ty, UT. OBJECTIVE: Preeclampsia (PE) is characterized by reduced placental invasion. Tumor necrosis factor ~ (TNF st) is thought to play a role in implantation and trophoblast invasion. A mutation in the TNF c~ promoter has recently been described which alters transcription of the TNF c~ gene. We sought to determine whether this mutation is associated with preeclampsia. STUDY DESIGN: DNA was extracted from blood of 162 women with severe preeclampsia, 75 women with HELLP syndrome, and 41 women without pregnancy complications. The polymerase chain reaction (PCR) was used to amphfy the promoter region of the TNF o~ gene. The resulting PCR product was subjected to allele-specific restriction by Nco/. Fragments were sizs-fractionated on a 3% Metaphor agarose gel and detected with ethidium bromide. Results were analyzed by a 3(2 contingency table. RESULTS: No significant differences were found between patients with severe PE, HELLP syndrome patients and healthy controls.

Allele Frequencies

T1 T2

Severe PE (n = 162) 0.85 0.15

HELLP (n =75) 0.88 0.12

Control (n = 41) 0 85 0.15

Pubhshed Controls (n=40) 0.84 0.16

CONCLUSIONS: The frequency of this TNF c~ promoter mutation is not increased in patients with preeclampsia or HELLP syndrome.

502 DO MALE FETUSES PROTECT AGAINST PREECLAMPSIA? M Leavittl, K. Wardt2. Depts. Ob-Gynl and Human Genet~s2, Univ. of Utah, Salt Lake City, UT. OBJECTIVE: S=nce polymorphisms ~n an estrogen-responsive gene, angiotens=nogen (AGT), have been associated with the development of preeclampsia (PE), we hypothesized that the fetal sex would influence a mother’s PE risk. STUDY DESIGN: Medica~ records of 259 PE patients were reviewed for fetal sex and PE seventy. Fetal sexes for controls were obtained from the delivery log recording 3179 consecutive deliveries at the same instituhon. AGT T-235 genotypes were determined us=ng an allele-spec=fic polymerase chain reaction for 220 PE patients and 270 controls. Statistical significance was estabfished by chi square analysis. RESULTS: A female:male (f:m) ratio of 1.27 was seen in the 259 PE patients compared to 0.92 in the 3179 controls (Z2=6.2; p=0.01). The sex ratio shows the greatest distortion when the preeclamptic mother is homozygous for the T-235 allele of AGT: (*=p>0.01)

PEGroup(n) MM(f:m) ] MT(f:m) I TT (f:m)

A, PE<220) I 0.86 I 1.ta I ,.66. Para 0 (105) I 0.75 ! 0.96 I 2.0"

Severe PE (129) I 0.83 I 1.15 ] 1.92"

Controls (270) I 0.95 I 0.80 I 0.93*

CONCLUSION: Contrary to previous reports we found that s=gmflcantly more females are born to women with PE. We postulate that female fetuses produce estrogens which stimulate decidual AGT production, thereby ~ncreasing the AGT-relsted risk of PE. Conversely, male fetal sex blunts the effect of the T-235 AGT allele, and is relatively protective against PE.

504 INTRACELLULAR MAGNESIUM LEVELS ARE REDUCED IN

PREECLAMPSIA - ECLAMPSIA CA Standle~. PR Standley *, L

Bogex, E Russell" , DB Cotton. Department of Ob/Gyn and Internal

Medlc~ne, Wayne State UmversW/Hutzel Hosp*tal, Detrmt, MI

OBJECTIVE: Symptoms of preeclampsm include hypertensmn,

protemuna, edema and when severe, se*zures A severe reductmn of

magnesium leads to cerebral abnormallt~es in humans. We have shown

prewously that preeclampsla was paradoxacally assocxated with an

elevation m extracellular magnesium that increased w~th the seventy of

the &sease We speculated that mtracellular free magnesium,

as opposed to extracellular, might be more relevant to the disease state

We therefore investigated whether women wath preeclampsm have altered levels of UMg~+]~ In addmon, we determined fasting and

nonfasting [Mg;÷] 1 m nonpregnant women.

STUDY DESIGN: We obtained venous serum and plasma samples

from nonpregnant (n=lS), normal pregnant (n=15) and preeclamptac

women (n=13) at term Isolated red blood cells from hepanmzed

plasma were analyzed for [Mg>]i using aT-NMR spectroscopy. Serum

samples were also analyzed for extracellular concentratmns of glucose

(Nova Nucleus Chemistry analyzer) and insulin (ra&ormmunoassay).

Data were analyzed with ANOVA and least squares linear regressmn.

RESULTS: [MgV]i was reduced during pregnancy (nonpregnant’

171 0_+5 35 !.tM vs pregnant. 155 14+7 76 pM) and was significantly

decreased m preeclamptac women (126 36+4.9 pM, p<0 05). We found

no correlation between glucose and [Mg~+]~ or insulin and [Mg~+ ]*.

However, msulm and glucose were correlated (r= 47, p<0.01)

CONCLUSIONS These results demonstrate that [Mg~*]l ~s

s~gnfficantly reduced m preeclampsia We suggest that preeclampsia

may be assocmted with a [Mg>]z deficiency and thas trnbalance may

contribute to the neurologic manffestatmns of this dtsease.

448 SPO Abstracts JanuaIT 1996 Am J Obstet Gynecol

505 URINARY EXCRETION OF MAONESIUIvl IS REDUCED IN

WOMEN WHO DEVELOP PREECLAMPSIA CA Standley x, BA

Mason, J WNtty, R Bobrowsk~, L Bogex, DB Cotton Department of

Ob/Gyn, Wayne State Umverslty/Hutzel Hospital, Detroit, MI

OBJECTIVE We have prevxously presented data demonstrating that

severe preeclamps~a ~s associated wath tugher extracellular levels of total

magnesium when compared to normal pregnancy In the present study,

we examined urinary magnesium excretxon m women who developed

mild and severe preeclampsla m order to determine ~f magnesium xs

retained by the ktdney m this disease

STUDY DESIGN: A 24-hour urine analysis and a venous sertun

sample (within the analysis period) were collected early m the thtrd

tnmester from 25 women Urine protein (UP), urine output (UO), total

magnesmrn excretion (UTMg) and total calcium excretion (UTCa)

were determined m urine, while total magnesium (TMg) and calcium

were determined m serum Data were analyzed via Student’s t-test

RESULTS: 7 women developed preeclampsla (4 severe, 3 mad) while

6 women had normal pregnancies Others were excluded on the basis

of additional disorders Data are presented as mean + S E M

Control Preeclampsm UTMg(g/day) 0 13 + 01 007 + 02** UTCa (g/day) 039 +.04 0.11 + 03**

UO (cc) 22875+2577 2150 + 5379 UP (mg/day) 134.5 + 21 03 2158.5 +1063 5

TMg (mg/dl) 156 + 05 165 + 04

CONCLUSION: UTMg and also UTCa were significantly reduced in

those women who developed preeclampsxa (p<0 01, respectwely)

Serum TMg tended to be higher m preeclamptlcs Serum calcium was

not affected In this small series ofpatxents, a reduced UTMg occurs

prior to the manifestations of overt preeclampsla

507 A POTENTIAL MARKER FOR PREECLAMPSIA: PLASMA PLATELET-DERIVED GROWTH FACTOR-A 1S INCREASED IN PREECLAMPTIC WOMEN. S Krue~er~, M GurskP, E Brown~, D Lewis, G Brookg, M Edwards, Depts. of Pediatrics and OB/GYN, Lou~stana State University Me&cat Center, Shreveport, LA OBJECTIVE: The expression of mRNA for platelet-derived growth factor- A (PDGI:-A) has been shown by our laboratory to be ~ncreased ~n the placenta of women with prceclamps~a (Jurcov~cova et at., 1995) We reasoned that PDGF-A also might be increased in the plasma of preeclanaptic patients and serve as a potential marker for the disease STUDY DESIGN: Blood was obtained from 21 preeclampt~c and 4 normotensive women, w~th gestational ages z 36 wks Patmnts with

&abetes or other hypertensive disorders were excluded Four non-pregnant women were also included in the study. Thirteen of the preeclampt~c patients received Mg~ treatment during labor. The blood was collected in tubes specially prepared to inhibit platelet activation. Plasma was separated from the blood by centnfugation and stored at -70"C. An ELISA assay for PDGF-A was performed s~multaneously on all samples. Data were analyzed using a one-way analysis of variance, followed by a Newman-Keuls test. Data from preeclamptic patients, separated on the bas~s ofMg~ treatment were analyzed with a one-tailed t-test. RESULTS: The ELISA assay demonstrated that preeclamptic patients w~thout Mg~ treatment had increased PDGF-A, (,2 = 2 3 la:s e 0.23) when compared to both normotensive pregnant controls (,2 = 0.71±0.22) and nonpregnant females (,2 =1.36±0,37, p<0.05). Within the preeclamptic group, those patients not receiving Mg- therapy had a sigmficantly higher level of PDGF-A in their plasma relative to preeclamptic patients on Mg~ therapy (p<0.05). CONCLUSIONS: Patients already identified as preeclamptic demonstrate PDGF-A as a potential marker for th~s disease. How early ~n gestation this marker can be detected to predict preeclampsia remains to be studied. The advantage of the ELISA assay for PDGF-A Is that it can be readily performed in a chnical setting

506 EFFECT OF MAGNESFSM THERAPY ON INTRACELLULAR

MAGNESIUM LEVELS CA Standlev~. LV Perlex, DB Cotton Dept

of Ob/Gyn, Wayne State Umverslty/Hutzel Hospital, Detroit, MI.

OBJECTIVE: Hypertenswe women have been reported to have low

levels of mtracelhilar flee magnesmm [Mg~+]l m red blood cells We

have found tins to be true m women with preeclampsla In the present

study, we examined whether magnes:um therapy for preeclampsia could

readily alleviate the deficxt in

STUDY DESIGN: 8 women diagnosed w~th preeclampsla were

enrolled in the study 6 of the women had a venous plasma sample

drawn prior to magnesium sulfate admimstrahon, while 7 women had a

sample drawn 4 hours after wataatmg therapy. The red cells were

~solated and [Mg~+]l levels were analyzed via n P-Nuclear Magnetic

Resonance (~P-NMR) maaging In a second experiment, a group of 8

nonpregnant women had 2 venous plasma samples drawn Red blood

cells were isolated and then exposed to a con~ol buffer (1.0mM

magnesium) or a high magnesium buffer (4 8mM) for 4 hours before

analys~s v~a ~aP-NMR Comparisons were made using Student’s t-test

RESULTS: Magnesium sulfate therapy m women w~th preeclampsm

led to a s~gmficant increase m [Mg~+]l (pre, 129 48+ 9 31; 4 hours,

174 10 + 19 66, p<0 05) This effect was evident w~thm 4 hours and

lMg~+]~ approached previously reported normal levels hiterestmgly, the lower the autxal [Mg~+]l, the higher the 4 hottr [M~+]I value. In the

nonpregnant women, m wtro exposure of red blood ceils to high

magnesium did not produce a change m

CONCLUSIONS: Magnesium m vivo is a t~ghfly regulated ion

However, women w~th preeclampsla who have mltaally low levels of

[Mg~]~ show a dramalac tmprovement with standard magnesium sulfate

therapy This change may contribute to a reduction or the resolution of

preeclampt~c symptomatology

508 DETECTION OF PLATELET-DERIVED GROWTH FACTOR AND ITS DOWN-REGULATION BY MAGNESIUM IN THE PREECLAMPTIC PLACENTA M GurskP. S Kmegeff, J Jurcovicova×, E Browa~, E Gonzalez~, D Lewis, G Brooks~, M Edwards, Depts of Pediatrics, Pathology, & OB/GYN, Louisiana State University Medical Center, Shreveport, LA. OBJECTIVE: To detenmne if: 1) platelet-denved growth factor (PDGF), a mitogen found in vascular smooth muscle of hypertensive disease and known to he inhibited by Ca++ channel blockers, is presem in the ceils and vessels of preeclamptic placentas; and 2) Mg÷+, which has a function similar to Ca** channel blockers, inhth~ts PDGF expression thus provithng a specific pharmacologic approach to the molecular events involved in PDGF activity

STUDY DESIGN: Placental tissue was obtained from 10 preeclamptlc and 10 normotens~ve women. The criteria for preeclampsia included. proteinuna, edema and matemal diastolic pressure z 90ram Hg. Tissue samples were analyzed for PDGF by immunocytochemistry, in sttu hybrlthzation and mRNA solution hybridizanan RNase protection assay

The effects of MgSO, on PDGF mRNA were quantified by solutmn hybridization Data were analyzed using an analysis of covariance and linear regression. RESULTS Immunoreachve staimng revealed PDGF-A and the a receptor ~n

vessels and extra villous cytotrophoblastic (EVC) cells. Immunoreactivity of EVC cells indicated the presence of two populations, one exhibiting intracellular reactivity and the other surface reactivity. In sttu hybridizatmn on fetal villi confirmed the presence ofPDGF-A mRNA in the vascular musculature The results of the solutmn hybridization demonstrated

~ncreased expression ofmRNA for PDGF-A in the placental cotyledons of preeclampt~c women relative to placentas of normotensive women (p<0.01 ). Furthermore, for preeclampt~c women, PDGF-A expression was correlated with the duration of Mg÷÷ therapy (r = - 0.796, p<0.01). CONCLUSIONS: PDGF-A is present m both control and preeclamptic placentas. PDGF-A expression is increased with preeclampsia and this expression is reduced by Mg~ therapy. This down-regulation of PDGF-A, an important vascular growth factor, has therapeutic anplicatmns for preeclampsia, separate from its ant~-convulsive effects.


509 PLACENTAL II[3-HYDROXYSTEROID DEHYDROGENASE

ACTIVITY AND UMBILICAL CORD BLOOD CORTISOL

LEVELS IN PREECLAMPTIC AND NORMOTENSIVE

PREGNANCIES. C McCalla, V Lakshm~x Dept Ob/Gyn, Kings

County Hospital, SUNY HSC at Brooklyn,NY.

OBJECTIVE: Apparent mlneralocorticoid excess and Licorice

induced hypertension, both hypertensive dtsorders similar to that

seen in preeclampsia, have been attributed to a defect in the enzyme

11 [3-hydroxysteroid dehydrogenase (11 [3-HSD), which

~nterconverts cortlsol to cortisone. We therefore undertook this

study to determine the role of human placental 11 ~3-HSD activity in

preeclampsia.

STUDY DESIGN: 11~3-HSD activity was determined in placentas

of 16 normotensive and 12 preeclamptic patients matched for age,

at 36-42 weeks of gestation. Cortisol levels m umbilical venous

blood were also determined for both groups. Preeclampsia was

defined as blood pressure of at least 140ram Hg systolic and!or~ 90

mm Hg diastohc plus proteinuna ( _> 300mg/24 hr or _> 2+ by

dipstick) on 2 occasions at least 6 hours apart Statistical analysis

was performed using Students t-test, significance at p<0.05

RESULTS: 11[3-HSD acnvity was significantly lower in placentas

of preeclampt~c compared to normotensive pattents (0.37+_0.14 vs

0.49_+0.14nmoles/min./g placenta, p=003). Cortisol level in

umbilical venous blood was significantly higher in the preeclamptic

group (22.82 + 14 38 vs 10.53 _+ 9 9 gg/dl, p=0.027).

CONCLUSIONS: There appears to be decrease in ll[~-HSD

activity in the preeclamptic placenta. The decreased 1 I[~-HSD

actiwty was accompanied by an expected concomitant increase in

umbilical cord blood cortlsol level This enzyme may play an

important role in the pathogenesis of preeclampsia.

511 SEVERE PREECLAMPSIA: RBC TRANSFUSION IMPROVES

OXYGEN TRANSPORT C Graves, T Wheeler, M Welsslnger~, N

Trolanox Dept. Ob/Gytt, Vanderbdt University, Nashville, TN,

OB3ECTIVE: Preeclampt~cs may suffer hypoxic tissue ~nju~ due to their

inabd~ty to ~ncrease tissue oxygen extraction m response to metabohc demand

and ~npaired dehvery Improving the arterial oxygen content could thus serve

as a m~ns to augment Ussue delive~� and limit end-organ injury. Following

¢ryatalloid "~olume reeu~lta~a~n, we sought to determine the hemodynamlc

effect of postpartum red blood cell transfusion.

STUDY DESIGN: 22 singleton gestations w~th severe preeelampsia and

HELLP syndrome met prescribed criteria for mvastve cardiopdmonaty

momtoring Pat~eats received crystallo~d infusion tttrated to PCWP >8 mm Hg

pnor to retrying 2 units of packed red blood cells Netther diuretic nor presser

flrempy were employed Patients served as their own control group for 12 hour

study p~fiods preeedmg and foltowmg transfusion Indexed values are d~v~ded

Post-Transfusion % change

SaO: .98 1

VO21 165 35*

DOzl 664 66 3*

Cardmc Index 5.01 26*

SVR 1000 -20 2*

P<0 05

by BSA

IESULTS:

Prc-Transfuston

97

128

427

3.98

1212

CONCLUSIONS: Packed red blood cell transfuston favorably influences

postp~tum hemodymanic indices m pregnancies comphcated by preeclampsm.

510 GLOMERULAR FILTRATION IS A FUNCTION OF

OXYGEN DEBT IN PREECLAMPSIA T WheeleL C Graves,

A Kang~, N Tro~an0 Dept of Ob/Gyn, Vanderbilt Umvers~ty,

Nashvdle, TN

OBJECTIVE: Glomerular dysfunctton, fi-equently recogmzed in

severe preeclampsia, has been attributed to metabohc aciderma

Regional t~ssue hypoxaa due to tmpa~red tissue oxygen transport

creates such an oxygen debt. We compared parameters of tissue

oxygen dehvery (DO~I and calculated base deficit) with those of

glomerular filtratmn (serum urate and creatinme)

STUDY DESIGN: Forty-seven consecutive singleton gestations

wth severe preeclampsm met criteria for invas~ve cardiopulmonary

momtonng (refractory hypertensmn, oliguna, pulmonary edema)

Baseline arterial blood gas samples and hemodynatrac calculations

were obtained by ox~metnc pulmonary artery catheter Oxygen

dehvery md~ces and calculated base deficits were compared w~th

s~multaneously obtained values for serum creatinme and uric acid

RESULTS: Linear regressmn analys~s demonstrated a consistent

correlatmn between parameters of oxygen dehvery and those of

glomemlar filtration Uric acid and creat~mne levels rise as the DO~I

dechnes r=-43" and r=41* respectively Similarly, uric actd

(r= 57) and creatmine (r=0 41) rise w~th an elevated base deficit ¯ denotes P< 05.

CONCLUSION: Impared tissue oxygen transport results m

s~gnificant glomerular dysfunctmn Management &rected towards

reducing oxygen debt may ~mprove glomerular filtration in patients

w~th preeclampsta

512 THE PROCOAGULANT ACTIVITY OF RED BLOOD CELLS FROM

PATIENTS WITH SEVERE PREECLAMPSIA. D. C-rtsarux, E Zwangx*,

M.R Peyserx, A. Elderx*. Depts of Ob/Gyn "A" and *Hematology, Tel Aviv

Sourasky Medical Center, Saekler Faculty of Med~ctne, Tel Av~v Universtty,

Israel

OBJECTIVE: The working hypothesis is that red blood cells (RBC) from

patients with severe preeelampsia may show increased membrane exposure of

procoagulant phosphohpids (i.e. phosphatidylserine) whteh may initiate

tntravaseular clotting and platelet aetivatmn. STUDY DESIGN: The study group comprised, rune women with severn

preeelampsm in the 3rd trimester of pregnancy, ten term, normotens~ves w~th

uncomphcated pregnancy, and nine matched, nonpregnant healthy women. The

exposure of phosphatidylsefine on the outer phospholipid layer of the RBC was

analyzed, usin8 isolated and washed RBC added as a source of phosphohp~ds to a

"prothrombinase" complex. The resultant thrembin formed was m~sured by an

amidolytic assay (Chromozym TH, Boehringer Mannhe~m, Germany).

RESULTS: Thrombm formation was sigmficanfly higher with the add~tmn of

RBC from pre~lamptte p~ttents than wtth the additton of RBC from nor/llotenslve, pregnant and nonpre~nant women

Severe Normoteasive Non-prcgnant

preeelampsia pregnant

Age(years) 26.6+_.4.6 296_+62 290_+_35

Thrombln formation 202 + 36 115 + 62** 86 + 73***

(ASU/min)*

*ASU = arbitrary spectrometer units; **P < 0.002; ***P < 0 0005

CONCLUSIONS: To the b~st of our knowledge, this study ~s the first to prowde

evidence that RBC from preaclampfic pallents provide a significant prcooagulant surfase that may trigger thrembin formation and thus play a role in the

hypercoagulab|e state m severe pr~lampsia.

450 SPO Abstracts January 1996 Am.] Obstet Gynecol

513 INSULIN RESISTANCE AND PREECLAMPSIA Arjpn A. Kraayenbrink MD’, Rqk O.B. Gans MD PhD~’, Herman P. van Geqn MD PhD’, Gustaaf A. Dekker MD PhD’*, Depts. of ObstetncsI and Internal Medicine2, Free University Hospital, Amsterdam, The Netherlands.

OBJECTIVE: Insuhn resistance is known to be related to hypertension, its relation to preeclampsia (PE) is less clear. As

a secondary aim in a double-blind randomized placebo-controlled prospective study assessing acetylsahcylic acid (ASA) versus Allylestrenol (AE) in the prevention of PE, we studied the relationship between insulin sensitivity and the subsequent development of PE. STUDY DESIGN: 64 primiparous women with a Resistance Index

of the uteroplacental circulation >_ 0.58, measured with continuous

wave doppler ultrasound at 24 weeks’ gestational age, were randomized to receive dady AE 25 mg (18), ASA 60 mg (25), or placebo (21) until delivery. At 24 weeks (before medication) Insulin

Sensitivity Index (SI) was calculated according to the Minimal Model method. Results were analysed by Student’s t-test.

RESULTS: 24 women developed PE (37.5%), ASA-group 8 (32%), AE-group 8 (44.4%), placebo group 8 (38.1%). At 24 weeks women who developed PE had Ngher body weight: BM128.5 ___ 3.4 kg/m2

(mean 4- 2sd) vs 25.8 ± 5.2 (/3 = 0.012), were insuhn res=stant: SI 3.8 ± 1.9 vs 5.4 ± 3,7 (p = 0.026) and had higher diastohc blood pressure: 77.1 ± 8.2 mmHg vs 66.5 _+ 5.5 (/3 < 0.0001). In addition, a significant correlation between plasma ~nsulin and Mean Arterial Pressure existed: r = 0.37 (p = 0,003). CONCLUSIONS: Insuhn resistance appears to be revolved in the causation of the vasoconstriction associated w~th preeclampsla Further studies are needed to examine the pathophysiological mechanism which is responsible for this phenomenon,

515 DECREASED URINARY NITRIC OXIDE/

CREATININE RATIO IN PREECLAMPTIC PREGNANCY C.D. HSU, S.F. Hongx, Y.K. Chungx,

J.A. Copel, Department of Ob/Gyn, Yale University School of Medicine, New Haven, CT. OBJECTIVE: To determine whether urinary nitric oxide and

mmc oxide/creatinine (U-NO/C0 ratio are altered in preeclampsia. STUDY DESIGN Sixty-eight pregnant women were stu&ed. Thirty-four patients were diagnosed with preeclampsta, and 34 were normotensive. Preeclampsia was defined by ACOG criteria

and hyperuricemia(> 5 5 mg/dl). Fresh urine samples were collected and treated with 100% ethyl alcohol to precipitate urine

proteins which could confound the measurement of nitric oxide. Nitric oxide was determined by a chemduminescence detector after

reduction by acidic vanadium. The value was quantitated using an integrator by reference to NaNO3 standards. Student t-test was

used for statistical analysis. Data are expressed as meanY.+ SE. RESULTS: There were no significant differences in maternal

age, parity, race, or gestational age between the two groups. Urinary creatmine and nitric oxide levels, and U-NO/Cr ratio were

sigmficantly lower in preeclamptics than in the normotensive group. Urine Preeclamotics(N=34) Normotensive(N=34) P value

NO (mmol/ml) 7,1+ 1.0 15.4 + 1.7 <0.0001

Cr (mg/dl) 76.6 + 8.5 109.8 +10.7 0.02

NO/Cr 0.11 + 0.02 0.16 _+0.02 0.03

CONCLUSIONS: Significantly lower unnary creatinme and

nitric oxide levels, possibly due to impaired renal excretion, occurred in preeclamptic pregnancies. The reduced U-NO/Cr ratio may indicate impaired renal production of nitric oxide in

preeclampsia.

514 EFFECT OF MAGNESIUM SULFATE ON NITRIC OXIDE LEVELS IN PREECLAMPTIC PREGNANCY C.D. Hsu, S.F.

HongX, Y.K. ChungX, J.A. Copel. Department of Ob/Gyn, Yale Medical School, New Haven, CT. OBJECTIVE" To determine whether the intravenous administration of magnesium sulfate alters nitric oxide levels in preeclampsia STUDY DESIGN: We studied ten women with preeclampsia comphcating singleton pregnancies in the third trimester. Preeclampsla was defined by ACOG criteria and hyperuncemia (> 5 5 mg/dl). An intravenous magnesium sulfate bolus of 6 gm was given, followed by a 2 gm/hour maintenance dose. Blood samples were collected before and dunng administration of Intravenous magnesium sulfate for the determmination of nitric oxide levels. In addition, postpartum blood samples were obtained Serum magnesium level was determined during magnesium admimstration Ethyl alcohol 100% was used to precipitate serum proteins which could confound the measurement of nitric oxide Nitric oxide was theq determined by a chemdummescence detector after reductmn by acidic vanadium The value was quantitated using an integrator by reference to NaNO3 standards. Analysis of variance and Student-Newman-Keuls test,and student t-test were used for ststistical analysis Data are expressed as mean + SEM RESULTS’ There was no significant difference in nitric oxide levels before and during treatment with magnesium sulfate ( 28 4 + 5.4 vs 27 3 + 2.9 mmol/ml, P >0.05).Treatment serum magnesium sulfate levels ranged from 4.8 to 6 2 mg/dl. The nitric oxide level from postpartum was significantly lower than that before and dunng magnesmm administration (14 9 + 1.1 vs 28.4 + 5 4 and 14.9 -+ 1.1 vs 27 3 + 2 9 mmol/ml, respectively; P<0 02). CONCLUSIONS Intravenous magnesium sulfate administration did not alter the nitric oxide levels in preeclamptlc women. This may provide an indirect evidence that magnesium sulfate does not play an important role in reduction of nitric oxide-mediated vasodllatatlon effect in preeclampsla. Reduced postpartum nitric oxide levels may simply reflect a physiological change after delivery.

516 ORAL TOLERIZATION TO PATERNAL ANTIGENS AND

PREECLAMPSIA

Dekker GAx, Department of Obstetrics and Gynecology,

Free University Hospital, PO Box 7057, 1007 MB, Amsterdam,

The Netherlands.

Objective: Oral administration of myelin and collagen has a signifi-

cant beneficial effect in pat:ents with respectively muluple sclerosis

and rheumatoid arthritis. This effect (oral tolerizatmn) is probably

related to the specific way in which the antigen is processed by the

digestive tract immune system. Immune maladaptation has been

suggested to play a major role in the etiology of preeclampsm.

Recent studies demonstrated that increased sperm exposure is asso-

ciated with a decreased incidence of preeclampsm. In this retrospec-

tive study we evaluated ff oral sex (i.e. intraoral ejaculation) is

associated with a decreased incidence of preeclampsia.

Study Design: 41 primiparous patients with a history of proteinunc

preeclampsm and a control group of 44 women were asked if they

had had oral sex with their parmer before the index pregnancy.

Results: Preeclampsia; n = 41; oral sex 18 (44%); no oral sex 23 (56%)

Controls; n = 44; oral sex 36 (82%); no oral sex 8 (18%)

(Fisher’s exact test p= 0.0003; OR 5.7, 95% CI 2.150-15.378)

Conclusions: Oral sex before the first pregnancy ~s assoctated with

a sigmficantly decreased mcidence of preeclampsia. Further studies

are mdtcated in order to assess whether this reflects oral tolenza-

tion to paternal antigens or if oral sex is associated with increased

genital tract exposure to sperm. The hypothesis that preeclampsia ~s

caused by immune maladaptation is supFgrted by these results.

Volume 174, Number 1, Pall 2 SPO Abstracts 451 Am J Obstet Gynecol

517 INTRAVENOUS NIMODIPINE IN THE MANAGEMENT OF SEVERE PREECLAMPSIA: DOUBLE BLIND, RANDOMIZED, CONTROLLED CLINICAL TRIAL. M Belfort, O Taskm,×A Buhur,* G. Saade, A Yalcmoglu~ Depts. of Obstetrics and Gynecology, and Anesthesiology, Baylor College of Medicine, Houston, TX and Dept of Obstetrics and Gynecology, Inonu UmversW, Turkey OBJECTIVE: To compare the utdity of 1V mmodipme and IV magnesium sulfate in the management of severe preeclampsia STUDY DESIGN: Patients with severe preeclampsia (ACOG criterm) were randomized to receive a 6 gram IV MgSO4 bolus (in 100 cc crystalloid over

20 minutes) followed by 2g/hr mfusmn (n = 15) or a 100cc crystallmd bolus over 20 minutes followed by 30 gg/kg/hr infusion of IV nimod~pme (n = 18) All bolus solutions and drags were packaged s~mllarly and infusion rates were

identical for both groups All patients underwent induction of labor w~th oxytocin Initial blood pressure response, need for additmnal antlhypertensive medicahon, incidence of seizures, drug related side effects, urine output, adm,ss,on to mduct~on interval, induct,on to delivery interval, mean oxytomn dose, route of dehvery, blood loss, postpartum complication rate, and neonatal outcome were compared. Statistical s~gnificance p < 0 05. RESULTS: The two groups were demograph,cally eqarvalent The ~mtial MAP in the nimod~pine group was 143+/-7 mmHg and in the MgSO~ group it was 138+/-8 mmHg There were mmilar initial decreases in MAP (24%) in both groups (p<0.05) More patients in the nimo&pine group experienced symptomatic improvement (p<0.05) Two pahents in the MgSO4 group, and none in the nimo&pme group, had seizures whde recew~ng therapy (p=NS). There was no difference ~n the ~nductmn to dehvery interval between the 2

groups but the mmodpme group required a significantly lower mean oxytocm dosage 12+/-4 vs 9+/-3 IU/min S,x patients m the MgSO4 group (5/6 for fetal d~stress) and 7 in the nlmod,plne group (3/7 for fetal distress) had a cesarean sectmn There were no differences in neonatal (APGAR and cord blood gases) or maternal outcomes CONCLUSIONS: Intravenous mmodiplne is an effective and safe drug for the management of pattents with severe preeelampsm

519 INCREASED INCIDENCE OF GROWTH RESTRICTION IN

INFANTS BORN TO NORMOTENSIVE WOMEN WITH A

HISTORY OF SEVERE PREECLAMPSIA.

P. Samuels, L. CorderoX, M.B. Landon, S.G. Gabbe. Dept Ob/Gyn,

The Ohio State University, Columbus, Ohio

Objective: To determine the outcome of the next pregnancy in

women who were severely preeclamptic in their first pregnancy.

Study Design: Over a 5 year period, we admitted 437 severely

preeclamptic nulhparas. Of these, 282 (64.5%) were maternal

transports, and 239 (54.7%) were <36 weeks gestation We obtained

follow-up on 30l (68.9%) 196 (65.1%) had a subsequent pregnancy,

and these were compared with a gestatlonal age matched control

group for analysis. 94 (48%) of the study group developed

preeclampsia prior to 36 weeks gestation in their first pregnancy.

Results: Of the 196 previously preeclamptic women, 83 (42.2%)

again became preeclamptic, and 19 (22.9% of the preeclamptics) were classified as severe. The mean gestational age at delivery for all

patients was 36.6 +_ 3 1 weeks with a range of 29 to 41 weeks. The

mean gestational age at delivery for the 83 women developing

preeclampsia was 34 6 + 4.2 weeks vs 38.7 :!: 2.3 weeks for the 113

women who remained normotenslve (p <0.05). SGA infants were

born to 11/83 (13.3%) current preeclamptics, compared with 1 1/113

(9 7%) (p = NS) normotenstive women who had previously had

preeclampsia, and 9/196 (4.6%) (p<0.01) controls. The difference

between the incidence of SGA infants in normotenslve women who

had been previously preeclamptic and controls was also statistically

significant [(11/113) vs (9/196) (p<0.01)]. When analyzing only

the 94 women who previously had been severely preeclamptic prior

to 36 weeks gestation, all differences were more pronounced.

Conclusions: Women with a history of severe preeclampsia who

are normotensive in a subsequent pregnancy have a higher than expected incidence of growth restriction in this next pregnancy. The

etiology of this growth disturbance warrants further investigation.

518 RISlK FAffFOP, S ~ THE D~/~LOPlIHIT OF SUP~III(POS~ PlII~"IJ~(PSIA IN G~AVIDAS WI~II C~IONIC HY?~TI~NSION. E. Charles Lampley,x S.A. Myers, Dept of OB/GY~, Chicago Medical School, Mt. Sinai Hospital, Chicago, IL Olk]~e~IVg: To Identify factors affecting the risk of superimposed pre- eclaapsia (P~) in gravidas with chronic hypertension (CHTN). 6’IqIDY NSIGN: The computerized data base information was analyzed in 316 patients with CHTL Maternal age, parity, race, presence of IUGR, excessive weight gain, smoking history, and presence of PE or CHTN in prior preqnancy were examined to ascertain their relationship to t~e development of superimposed PE in the current pre~nansy. Odds ratios and ~ultiple logistic regression analysis were performed. I~SUL~S: Fifty-five of 316 patients (17.4%) developed superimposed Nulliparity, OR 3.31 (CI:I.6-6.9), PE in a prior pregnancy, OR 4.8 (CI: 1.9-12.3), exnessive weight gain, OR 6.8 (CI:I.4-29.7), and fetal ~rowth retardation, OR 2.7 (CI:i.4-5.4) were associated with a significant risk of developing superimposed P~. CHTN in a previous pregnancy was associated with a decreased incidence of superimposed preeela~psia OR 0.15 (CI:0.4-0.66). Following multivariate analysis, O~LY nullip~rity, OR 4.0 (Cl 1.9-B.4), PE in prior pregnancy, 6.1 (CI:2.4-15.2), and excessive weight gain, 8.1(Chi.9-35.0) were associated with the development of superiaposed PE. CHTN in a previous pregnancy was still ansoeiated with a lower incidence of preecla~psia 0.17(¢I:0.04-0.7B) CONCLUSION: Sibai, et al (AJOG, 172; 648, 1995) have shown that the risk of development of mE in healthy nulliparous patients is i~reased with primi~ravidity, elevated pre-pregnancy weight and elevated systolic blood pressure but decreases with a smoking history, This study demonstrates that the risk of superilposed PE in nhronlcally hypertensive gravities increases with nulliparity, PE in a prior pre~mancy, and excessive weight gain and decreases with a history of CHiN in a prior pregnaneL

520 EVALUATION OF NITRIC OXIDE AS A MEDIATOR OF SEVERE

PREECLAMPSIA. M Kupfermmc, R Sdver, T Russell*, L Adler*, T

Mullen* M Caplan* Depts. of Ob/Gyn and Peds, Northwestern

University Medical School, Evanston Hospital, Evanston, Ilhno=s.

OBJECTIVE: To determ=ne ff a reduction in endogenous

mtrovasodilator synthes~s plays a role in the pathophysiology of

severe preeclampsia (SPEL

STUDY DESIGN: Orculating nitrite and nitrate levels were

compared between women who met standard clinical critena for

SPE (n = 23) and normotensive controls (NC, n = 23). End-products

of mtric oxide synthes~s were measured from venous blood

samples us=ng chemiluminescence and n~trate reduction. To detect

m wtro suppression of nitnc oxide synthesis, human umbd~cal vein

endothehal cell (HUVEC) monolayers were grown to confluence

and randomly exposed to culture media containing either 20%

SPE or NC sera. Nitnte and mtrate production were compared at

1, 12 and 24 hours m duphcate monolayers for each cond=t~on and

expressed as means 4- SE in pmol/10~ cells.

RESULTS: C~rculating n=tnte + n=trate levels were similar in SPE

and NC groups (48.2 + 2.6 vs 50.2 4- 2.5 pmol/ml, p =0.7). As

expected, serum-contaimng med~a increased HUVEC n=tnte

production consistently during ~ncubation. However, supernatants

from HUVEC exposed to SPE or NC faded to show different

patterns of nitrite production at 1 and 12 hours and resulted ~n

equ=valent levels of synthes=s at 24 hours (SPE, 14.5 __+ 2.7 vs

NC, 13.3 4- 2.9 pmol/10~ cells, p=0.8),

CONCLUSIONS= Orculating nitnte and nitrate levels are not

reduced ~n pat=ents w~th SPE compared to NC. SPE sera do not

alter endothehal cell n=tnc oxide synthes=s in vitro. A contnbutory

role for n~tnc oxide as a mediator of SPE is not supported by these

data.

452 SPO Abstracts Janua~T 1996 Am J Obstet Gynecol

521 HUMAN NEUTROPHIL DEFENSINS ARE ELEVATED IN

PLASMA OF PREECLAMPTIC PATIENTS

J.A~P_Iit~*, A.V. Panyut~eh*, R.P. Heine*. Magee-Womens Research

Instttute, University of Pittsburgh, Pittsburgh, Pennsylvania and University of Califorma at Los Angeles, Los Angeles, California. OBJECTIVE: To determine if human defensins, specific markers of

neatrophil aetivatxen, are elevated in the plasma of patients with preeclampsia. STUDY DESIGN: Blood samples were obtained from 18

preeclamptic and 29 normal pregnant patients in the third trimester. Demographtc and clinical data were obtained from the medical record. No patient had evidence of labor and/or infection at the time of the

blood ckaw. Preeclamptic patients were defined by nulliparity, elevated

blood pressure of 140/90 mmHg, proteinuria of >300 mg in a 24 hour collecUon and hyperuricemia. Human defensins were measured by enzyme immunoassay of plasma samples &luted 1:100. Standard curve values ranged from 0.25 ng/ml to 16 rig/m1, thereby malang the lower

hmit of detection in plasma 25 ng/ml. Stastistical analysis included

Mann Whitney U test for comparison of absolute values between

groups and Fisher’s Exact test to compare the number of pattents in each group who were above the low point on the standard curve.

RESULTS: There was no difference in age or estunated gestational age between the two groups. There were more nulliparous patients in the

preeclampt~c group. Human defensin levels were significantly elevated

(13=0.005) in preeclamptic patients (25.1 ng/ml + 16.2) compared to normal controls (9.0 ng]ml + 8.9). Nine of 18 (50%) preeclamptic patients and 2 of 29 (7%) normal controls had defensin levels above

the low point on the standard curve (p=0.001). CONCLUSION: Ore" results suggest that preclampsia is associated wRh nentrophd activation. The bmlogical effect of elevated human defensins in the plasma of patients with preeclampsia remains to be determined.

523 COMPARISON OF METHODS OF BLOOD VOLUME MEASUREMENT IN NORMOTENSIVE AND PREECLAMPTIC PREGNANCIES. H Sdver., M Sesbeckx. Dept. Ob/Gyn, Brown Univ, Providence, R.I. OBJECTWE: To verify reduced blood volume in preeclampsm using two independent measures to improve accuracy, and to determine the appropriate F cell ratios. To accurately measure total blood volume (TBV), measurements of red cell mass (RCM) and plasma volume (PV) should be by independent methods, not by ex~apolation from peripheral hematocrit (PH) PH is not equiv~ent to whole body hematccrft (VVBH). It is frequently assumed that the WBH isthe PH multiplied by a constant of 0.9, the F call rat~o, determined by compadacn of TBV measurements by dtract measures and by extrapolation. The F cell rat~o has not been reported in presclampsia. Add~onally, due to =noraased Iosa, the vsiiddy of PV measurement in presclampeia with albumin labels such as Evan’s blue (EB) has been questioned STUDY DESIGN. RBCs labelled wrth an enriched stable isotope of chromium, S3Cr, and EB wore infused. Blood was sampled at 5,10,15,20,30,40,50,60 minutes for analysis of EB concentration and at 30 minutes for S3Cr analysis. RESULTS: Resu/ts of measurements are displa)~ed m the table. / Group / RCM I PV IMsas. TBVlExtrapoI. TBV*I FcallI pontro~ (n=20)p~7~4e 140~5~15722t:821 I 5760"~895 I0 89 +.051 prescl(n=20) II,~eg-~eT1339z, ee91 49~1*991 1 4876±870 094+.08[ ~value* / 0.43 I 0.0025 [ 0.008 I 0.003 001 I * Students T-test t Extrapolated TBV = PV by EB+(1-hematocnt x 0 9) The percent difference between measured and extrapolated TBV was 3 2 + 3 1 (range -14.1 to 2.9) ~n the presolampt~cs, and 1 9 + 1 5 (range -5 8 to 3.4) =n the controls. The rate of disappearance of EB was 7.7 + 13.1% per hour in the presc|amptica versus 2.7 ± 13.8% in the controls (p=0.25). CONCLUSIONS: We have confirmed reducad TBV in presclampsia using independent measures of PV and RCM. There is a significant d~erance in mean F cell ratio in controls and presclamptica, suggesting alteration ~n blood disthbution in presclampsia. This introduces a small but systematic error =n use of ox~ TBV m wesdampsia. The large ind=v=dual variatmn in F cell ratio can lead to significant error for a subject, Because the error both under and overeshmates TBV, extrapolated TBV is accurate for a cmss-esc~onal study. In a longitudinal st~udy, measurement of PV and RCM may have sign$cant advantage.

522 IN PRETERM PREECLAMPSIA, PATTERNS OF PLACENTAL PATHOLOGY ARE ASSOCIATED WITH SPECIFIC MATERNAL AND FETAL/NEONATAL CHARACTERISTICS. C. Salaha*, J Lopez-Zeno*, K Starzyk*, S. WNttmgton*, V. M=n=or*, J Pezzullo* Departments of Pathology & OB/GYN, Georgetown Un=versity Medical Center, Washington DC, University of Ponce, Ponce PR OBJECTIVE: To determine If spec~hc patterns of placental pathology were assooated w=th clinical features of preterm preeclampsla {PE), STUDY DESIGN: From a database of consecuWe non-anomalous singleton livab=rths dehvered at 22-32 weeks gestahon (GA) (excluding maternal diabetes melhtus and chronic hypertension) over five years (1989-93), we =dent=fled 78 cases of PE From charts were extracted mldtnmester maternal serum o~-fetoprotem (MSAFP), maternal blood pressures (BP), protemuna, platelet count, dlagnos~s of HELLP, parity, race, GA at dehvery, cord blood gas values, Nrthwelght (scaled to cent=le for GA), and m=tlal neonatal hematocnt and nucleated erythrocyte (nRBC) count Cases were graded for 36 h~stolog=c les~ons (utero-placental vascular and related villous les~ons, and les=ons of chronic inflammation, coagulation, and Intraplacental vase-occlusion) Factor analyses y=elded 13 significant les=on patterns (we=ghted combinations of les~on grades) These factor scores were used as =ndependent vanables ~n ANOVA and regression analyses (p<O 05 s=gnificant). RESULTS: ~Maternal platelet count, ~-weight cenhle, gplacenta] we=ght, ,~,umbilical venous and artenal pH, £nRBC’s/dl and !~hematocnt were each related to a pattern of chronic and acute Intrauterine bleeding, infarct, and uteroplacental thrombos=s (each p<O 03). £ Maternal prote~nuda was related to Nstologic abrupt=on (p=O 03), and to basal chronic Inflammation (p=O 005) Multivariate analyses showed independent effects of hlstolog~c abrupt=on (p=O.O02), a pattern of villous stromal hemorrhage/edema (p=O.04), and u(eropiacental thrombosis (p=O ~32) on abnon~at m=dtnmester MSAFP (overall p=O.O006, R=O 479) Advanced maternal age was related to chronic vllhtls (p=O.03) Features associated with actlva placental growth were related to # weight centlle (p=) 03). Other features were not correlated with lesion patterns CONCLUSIONS: Identlhable clinical parameters (such as degree of maternal protemuna, abnormal m~dtnmester MSAFP, fetal growth restriction and neonatal evidence of hypoxlaJacldosis) are related to speclhc placentakl lesions. These clinical findings may be used antenatally to predict and poss=bly treat potent=al patholog=c complications..

524 PREECLAMPSIA AND THE RISK OF INTRAVENTRICULAR HEMORRHAGE AND PERIVENTRICULAR LEUCOMALACIA

IN THE PRETERM NEONATE.Verma U. Cantenno jx, Tejam N,

Beneck gY, Klein Sx, Jeanty IvP NYMC Valhalla, NY

OBJECTIVE: Preeclamps~a (PRE) and intrauterine growth restnchon

(IUGR) often coexist IUGR sxgmficantly decreases the risk and seventy

of mtraventncular hemorrhage (IVH) and penventncular leucomalac~a (PVL) We hypothesize that PRE decreases the risk and seventy of

IVH/PVL

STUDY DESIGN: 716 consecuttve reborn neonates w~th Bthwt

_< 1750gm and GA <34 wks were evaluated by transfontanelle cramal

sonography done on days 3 & 7 IVH was graded according to Papflle

PVL was diagnosed by the presence of any hyper or hypoechotc lesmns m

the penventncular areas by day 7 Major abnormahty included Grades III

and IV IVH and PVL Neonates were diwded tnto groups based on

presence of PRE and compared for the risk and seventy of IVH~VL The

confounding effect of 1UGR was controlled by excluding all neonates

w~th 1UGR The remaining AGA neonates were strahfied mto GA and

PRE subsets These subsets were compared for the risk and seventy of

IVH/PVL F~sher’s and Students t- tests were used

RESULTS: TOTAL GROUP AGA GROUP

PRE # IVH/PVL MAJOR # IVH/PVL MAJOR

YES 143 27 3 98 19 3

NO 573 205 124 509 191 115

P value 0001 0000 0003 0002

PILE sxgmficantly reduced the risk and seventy of IVH/PVL m all the

GA subsets p < 05

CONCLUSIONS: PRE mgmficantly reduces the ~ncldence and seventy

of IVH/PVL even m the absence of 1UGR


.ann J Obstet Gynecol

525 URINARY CYCLIC GMP, ENDOTHELIN AND PROSTAGLANDIN E2 ARE DECREASED IN PREECLAPMSIA. J Ludmlrl, BA Clark~, JG Alvarezix and AG

Ludrmr~x Depts of Ob/Gyn1 and Med~clne:, Beth Israel Hospital,

Harvard Medical School, Boston, and San Marcos Umverslty3, Lima,

Peril

OBJECTIVE Cychc GMP, endothehn and prostaglandln E2 are

known vasoactlve factors (with cychc GMP acting as a second

messenger of mtnc oxade) They are natnuret~cs and their urinary levels

reflect mtrarenal production The oblechve of this study was to compare

the urinary levels of cyclic GMP, endothehn and prostaglandln E2 an

normal pregnancies vs pregnancies comphcated by preeclarnpsia

STUDY DESIGN Unne samples were collected from forty-five

pregnant women Group I (n =13) were normotensave pregnant women

at term, and Group II (n = 32) were pregnant women w~th preeclarnpsla

(16 had todd preeclampsaa and 16 had severe preeclampsm) The

presence or absence of prenalal care, history of hypertension, dietary

intake of milk and meat, use of vitamins and calcium intake were all

recorded for each patient The urinary levels of cyclic GMP, endothehn

and prostaglandln E2 were determined by radlolmmunoassay and

normahzed for creatlnlne levels in urine Values were compared by t test

and analysis of variance

RESULTS: No slgmficant dafferences were found in dmt, vitamin, or

calcium intake between the two groups (P> 0 2) None of the patients

had prior history of chronic hypertension The mean _+ SD dlastohc

pressures m Groups I and II were 67 ± 2 and 101 _+ 2 mmHg, respectively

(P< 0 05) Preeclamptlc women had sagmflcantly lower levels of urinary

cyclic GMP (0 67 ± 0 12 vs 2 1_+ 0 5 nmol/gm creatmme) endothehn

(0 88 + 0 09 vs 3 75 ± 1 4 ng/gm creat~mne) and PGE2 (26 ± 4 vs 47 ±

9 ng/gm creatlnlne) as compared to normotenslve women (P< 0 05) CONCLUSION: The urinary levels of cychc GMP, endothehn and

prostaglandln E2 are decreased In preeclampsla and may contrabute to

the sodium retentlon and hypertension observed in preeclampsla

Further slu&es should clanfy whether these changes precede or are

secondary to preeclampsla

527 PLATELET COUNTS AT DELIVERY IN PATIENTS WITH SEVERE PREECLAMPSIA AND/OR HELLP SYNDROME DO NOT PREDICT PERIOPERATIVE COMPLICATIONS S Y Sitx, B Chen, M.L.Druzln. Dept Ob/Gyn, Stanford Umverslty,

Stanford, CA OBJECTIVE: 1) To study the relationship of platelet counts at dehvery to perloperatave comphcatmns an patients with severe preeclamps~a (SPE) and/or HELLP syndrome undergoing cesarean section (CS) 2) To detemnne whether prophylactac preoperative or lntraoperat~ve platelet transfusion is indicated STUDY DESIGN: A retrospective analysis of 60 patients with SPE (1972 ACOG) and/or HELLP syndrome who underwent CS was performed The incidence ofpenoperatlve comphcalmns was ~ompared

between patients with platelet counts >100,000 at dehvery (Group A n=46) and patmnts with platelet counts >50,000 and <100,000 at dehvery (Group B n=14) Obstetrical demographic data including maternal age, weight, parity, weeks of gestation were similar between the groups No preoperative or intraoperatave platelet transfusions were administered to either group RESULTS: (n=46) (n=14) * p < 0 005 Grouo A Group B Platelet counts at delivery 197.7K * 74.1 K

Estimated blood loss (cc) 863 0 I 878 0 Time of operation (mln) 56.3 I 59.4 Change m hematocrlt (%) -7 3 p NS -8.3 Post-partum platelet transfusion 1 I 1 Post-partum pRBC transfusion 3

[

3 Wound complications 3 2 CONCLUSION: 1) There was no difference in penoperatlve and postpartum comphcatlons between groups. 2) In patients with SPE/HELLP syndrome (platelet>50,000 and<100,000) prophylactic preoperative or mtraoperatlve platelet transfusion is not indicated 3) The risk of transfusion-related disease may therefore be avoided in these patlents~

526 LIPOPROTEIN A SERUM LEVELS ARE SIGNIFICANTLY ELEVATED IN PREECLAMPSIA. D. Lasser, L. Baxi. Dept of Ob/Gyn, Columbia U., College of Physicians and Surgeons, New York, NY. OBJECTIVE: It is hypothesized that lipoprotein A (Lp(a)), an established predictor of atherosclerosis with thrombogenic properties, xs elevated in preeclamptic women. STUDY DESIGN: Serum was obtained from fifteen preeclamptic and fiIteen non-preeclamptic women at term Lp(a) concentratmn was measured by sandwich ELISA using human monoclonal anBbody MPA1 in a peroxidase chromogenic-substrate assay. Assay was normalized using recombinant human Lp(a) reference standard. Statistical comparisons were made using the Kruskal- Wallis One-Way ANOVA. Patient characteristics were compared using Student’s t-test and Chi-square analysis where appropriate. RESULTS: Study and control subjects did not differ significantly in age, weight and panty. Preeclamptic women at term had significantly elevated serum Lp(a) levels when compared to term uncomplicated pregnant women (p<0.05). Mean Lp(a) surpassed the critical atherosclerosis risk threshold of 35 mg/dl in the study group (45.35 mg/dl) but not ~n the control group (18.57 mg/dl). CONCLUSION: Elevated serum Lp(a) is a well established predictor of atherosclerotic disease. Its role in the pathogenesls of preeclampsia may arise from its properties as a fibrinloysis inhibitor. Lp(a) has 80% homology with plasminogen and competitively reduces conversion to plasmin A relation between lipeprotein metabolism, fibrinolysis and preeclampsia is suggested which may reflect an underlying predisposition to future atherosclerotic disease

528 MATERNAL OUTCOME IN WOMEN ~7ITH ECLAMPSIA. A.Y. Frangieh.~ S.A. Friedman, F. Audibert, I. Usta, B.M. Sibai Department of Obstetrics-and Gynecology University of Tennessee, Memphis, TN OBJECTIVE: To compare maternal outcome in pregnancies complicated by eclampsia occurring before vs. after delivery and before vs after 34 weeks. STUffY DESIGN: We reviewed the medical charts of 368 consecutive

women with edampsia managed at our perinata! center. For one analysis, women were dividedinto two groups: edampsia before delivery and edampsia a~er delivery A second analysis was performed on two other groups: eclampsia at <34 weeks and eclampsia at >34 weeks. Results are expressed as odds ratio (OR) and 95% confidence interval (CI) and compared using the Z~ test. RESULTS: There were 129,993 ddlveries during the study period, yielding an

incidence of eclampsia of 1 in 353 deliveries. Compared to women with eclampsia after dehvety (n= 108), women with eclampsia before delivery (n= 260) had a hi her incidence of abruptio placentae (13% vs. 6%,p= 0.04) and HELLP syndrome (13% vs. 4%, p= 0.006). Occurrence of other major

outcome variables did not differ between the two groups. The role of gestational age is described in the table below.

Abtuptio placentae (%) * HELLP syndrome (%) ~" DIC (%) Ncumlogic deficit (%) Aspiration pneumonia (%) Pulmonary edema (%) Cardiopulmonary arrest (%) Acute renal fa,lure (%) * Death (%)

~-34 weeks >34 weeks Odds 95%

(n =14 Z) (n=221) Ratio CI

15 8 2.1 1.03-4.4

17 6 3 0 1.5-6.4 8 5 1.6 0.6-3.8

5 6 0.9 0.3-2.5 1 9 0 2 0 0-0.7

7 4 1.7 0.6-4 7

3 5 0 6 0.2-1 9

7 2 4 0 1.1-15 3 1 1 1.5 0.0-55.5

HELLP, hemolysis, elevated liver enzymes, low platelets; DIC,

disseminated intravascular coagulopathy. * P <0.05 ~ P <0.005

CONCLUSIONS: The timing of eclampsia in relation to delivery and the gestational age at which eclampsia occurs both affect maternal outcome. Serious adverse maternal outcomes were more frequent in women whose

convulsions occurred before delivery (abruptio placentae and HELLP syndrome) and at earlier gestationalages (abruptio placentae, HELLP syndrome, and acute renal failure).


529 SF.RIM..~’qIOTIC FLUID INDET. IN SEVEI~ PREECIAMPSIA: A IK)OR

PREDICTOR OF ADVERSE OUTCOME. IL Schuckerx, B Mercer, F hudibert,x R Lewis, SA Friedman, BM Sibai, University of Tennessee, Memphis. OBJECTIVE: To determine the relationship between low AH, and IUGR and

STUDY DESIGN: We reviewed the medical records of I~4 women with severe preeclamps:a managed conservatively for at least ~ hours. Patients were followed with a daily nonstress test and AFL We evaluated AFI -<5 and -<7, measured either on admission or just prior to delivery ("Final’), and auempted to corrdate these findings with the incidence of fetal distress requiting cesarean section or IUGR (birth weight -<10th percenule). RESULTS: 97 patients had a cesarean section but only 35 (36%) of these were for fetal distress and 48 (36%) of the pregnancies resulted in IUGR infants. During expectant management, the AFI worsened in 47 (35%) patients and improved or remained the same in 87 (65%). Table below describes the association between adverse pregnancy oucome and AFI. In those patients with an AFI -<7 on admission, there was a signicandy higher incidence of IUGR as compared to those w:th a normal AFI (OR=6.0, 95% CI 1.8-21.2) and there was a trend toward an assoc:ation between IUGR and an AFI -<5 on admission (OR=3.95, 95% CI 0.82-21.17). However, there was no association between IUGR and an AFI-<5 or -<7 measured prior to delivery. Moreover, there was no difference in the frequency of fetal distress (OR=0.8, 95% CI 0.30-2.05) or 1UGR (OR=I.1, 95% CI 0.45-2.58) based on AFI at any time.

Fetal Di~tr--~ (%) IUGR (%) AFI N SE~S Plr¢ NPV SEN$ Pl’V NPV Adn’dsslon -<5 9 3 11 73 13 67 66 Admission -<7 18 11 22 73 27 72 70 Final -<5 20 9 15 72 21 50 67 Final -<7 35 17 17 71 31 43 67

SENS=sensitivity PPV/NPV=positivednegative predictive value

CONCLUSIONS: In women with severe preeclampsia, remote from term, AFI -<7 on admission is predictive of IUGR but lacks sensitivity, There is no association between the AFI status and frequency of cesarean section for fetal distress.

531 DO PLASMA LEVELS OF VASCULAR ENDOTHELIAL CELL ADHESION MOLECULE-I PREDICT PREECLAMPSIA OR PETAL GROWTH RESTRICTION? S.A. Friedman, E. Schiff, E.S. Umstor,x

L. Kao,x R.N. Andersen, B.M. Sibai. Department of Obstetrics and Gynecology, University of Tennessee, Memphis, TN OBJECTIVE: To determine whether circulating levels of vascular endothelial cell adhesion molecule-1 (VCAM), a marker of endothelial cell activation, are elevated in patients several weeks before they develop preeclampsia or idiopathic fetal growth restriction (FGK). STUDY DESIGN: A nested case-control study design was used. Blood was drawn between 23 and 34 weeks’ gestation from healthy nulli.parous women and stored as plasma at -70°C. After delivery, 10 women with preeclampsia (late-pregnancy hypertension, proreinuria, and hyperuricemia) and 10 women with idiopathic FGR (birth weight < 10th percentile without apparent etiology) were identified. Controls were matched for race, maternal age (+ 3 years), and gestational age at blood collection (:i: 2 weeks). VCAM was measured using a sensitive and specific enzyme immunoassay kit (R & D Systems, Minneapolis). RESULTS: All women had detectable plasma VCAM levels. Horizontal bars represent mean values. Differences between the groups were not significant (p values >0.3).

Plasma

VCAlvl

(rig/ml)

1100 -

1000-

900 -

800 -

700-

600-

500"

400-

o

300 Preeclampsh Control FGR Control

CONCLUSION: Because of the large degree of overlap, plasma VCAM levels are unable to predict the occurrence in 2"3 months ofpreeclampsia or FGR.

530 CONSUMPTION AND PLASMA LEVELS OF VITAMIN E IN PREECLAMPTIC PREGNANCIES. E. Schiff, S.A. Friedman, L. Kao,x P.H. Barrett,x B.M. Sibai. Department of Obstetrics and Gynecology, University of Tennessee, Memphis, TN OBJECTIVE: Vitamin E, a potent lipid-soluble antioxidant, has been suggested to play an important pathophysiologic role in preeclampsia. Our aim was to determine whether consumption and plasma levels of vitamin E were lower in preedampdc than in normal women. STUDY DESIGN: A case-control study design was used. Forty-eight preedamptic women were identified (late-pregnancy hypertension, proteinuria. , and hyperuricemia). Ninety normal women at similar gestanonal ages served as the control group. Blood was drawn from all women in the third trimester and stored as plasma at -70°C until assayed. Plasma vitamin E concentrations were determined using high pressure liquid chromatography. Vitamin E consumption was estimated at the time of blooddrawing using a previously validated dietary recall questionnaire (Stampfer et al., N Engl J Med, 1993), administered to 43 of the preeclamptic women and all 90 of the control women by a single trained research nurse. Results are expressed as mean + SD and compared using the two-tailed Student’s ttest. RESULTS: The mean plasma vitamin E concentration was significandy higher in preedampfic than in control women (1.41 + 0.39 vs. 1.15+0.32 mg/dl, p< 0.001). Dietary consumption of vitamin E was similar for both groups (16.82+13.72 vs. 16.07+10.49 IU/24 h, p = 0.73). Total vitamin E consumption, including that contributed by prenatal vitamins, was significantly higher in those who developed

~reeclampsia than in controls (42.28 +23.41 vs. 27.03-+28.28 IU/24 n, p= 0.003). In both groups of patients, there was a poor correlation between individual vitamin E consumption and plasma levels (preeclampsia group, r = 0.22, p= 0.2; control group, r= 0.08, p= 0.7). CONCLUSION: Plasma vitamin E concentrations were unexpectedly higher in preeclampdc than in normal women, possibly suggesting a response to oxidative stress. In addition, increased vitamin E consumption was noted in the preeclamptic women, suggesting that it is not protective against the development ofpreeclampsia.

532 DIAGNOSTIC CRITERIA FOR HELLP SYNDROME: TEDIOUS OR ,ItEI.LPFLrL. ? ~x, S.A. Friedman, A.Y. Frangiehx, B.M. Sibai.

Department of obstemcs and Gynecolpgy, Universi_ty of Tennessee, Memph~., OBJ--: To compare the maternal outcome of pregnancies complicatea HELLP (hemolysis elevated hver enzymes, low phtelets) syndrome, patthl HELLP s~aglmme, dr severe preeclampsh. . , STUDY DESIGN: From Ju/y 1, 1992 through June 30, 1995, 316 patients wim HELLP syndrome or severe preeelampsia were manag_~ at our pennatsl center. HELLP syndrome was stric_tl~, defined by previously p~blished hboratory criteria. Severe preeclampsia was ddfmed by ACOG criteria. Patients were divided into 3 groups: HELLP syndrome; partial HELLP syndrome (1 or 2 but not all 3 features of HELLP syndrome); and sevve preechmpsia (no feature oI’HELLP syndrome). Results are compared using Zz analysis. ¯ R~ULTS: Mean gestatioifiI age at delivery in the HELl.P, "partial HEI~ anu severe preeclamps:a groups were res~ctively 31.7, 32.7 and M.5 weeks’ (p<0.001 between HELLP and severe}. There was 1 maternal death from intracerebral hemorrhage ~n the HEL]~P group. The incidence of cesarean section, DIC, and need’for transfusion was significantly higher in the group than in the 2 other groups (p<0.001). Major maternal complications are nresented in the followin~ able.

(n = 67) Cesarean delivery (%) 74~" 52 48 DIC (%) 15, 0 0 Blood products trans fusion(%) 25"~ 4 3 Wound hematom.Vin fection (%) * 14.2, 10.8"* 2.3** Pleural effusion (%) 6.J: 0 ! Acute renal failure (%) 3:1: 0 0 Eclampsia (%) 9 7 9 Abruptio placentae (%) 9 4 5 Pulmonary edema (~ 8 4 ~

J’p<0.001 for HELLP vs. partial HELLP and Severe;, p<0.05 for I-IEILP vs. Severe. *among patients who had cesarean section only. **p=O.04 for partial HELLP vs. Severe.

CONCLUSIONS: Higher incidences of maternal complications stress the importance of strict cntetia for the definition of HELLP syndrome. Patients wit.h partial HELLP should be studied and managed differendy ~om patients witia complete HELLP syndrome.


533 GLYCERYLTRINITRATE LOWERS BLOOD PRESSURE IN PATIENTS WITH GESTATIONAL HYPERTENSION F.Facchinettlx~ I. Nerix, A Volpex Dept of Ob/Gyn, Umverstty of Modena, Italy. OBJECTIVE: To evaluate the effects of nitric oxide donor glyceryltrinitrate (GTN) on blood pressure changes in

pregnancy. STUDY DESIGN: Twenty-five patients with Gestatiorral Hypertension (GH, 8 of them with preeclampsm) and 11 Normotensive controls (N) were treated once with a patch

releasing 10 mg GTN over 24 hours (Deponit, Schwarzpharma). Systolic (SBP) and Diastolic Blood Pressure (DBP) were recorded automatically, every 30 rain, for 24 h, for t~vo consecutive days. Data were analyzed with the Cosinor algorythm. Serum and urine nitrites were measured.

RESULTS: GTN significantly lowered the 24-h mean SBP in both groups (5% and 4%, for (3H and N, respectively) while DBP was decreased to a major extent, but only in patients with GH (7%).These latter showed an altered rhythm with a delayed acrophase (the nadir of blood pressure values) of SBP in

respect to N’ GTN restored the rhythm to normal. Serum nitrites were low in the subgroup of preeelamptie patients. Urinary nitrites increased after treatment, mainly in GH patients. CONCLUSIONS: The transdermal application of GTN represents a useful means for managing patients with GH. Both quantitative and qualitative changes of blood pressure were

observed Such changes are mediated through mtric oxide release.

535 MULTIPL~ GESTATIONS FROM IN VITRO FERTILIZATION:

INCIDENCE OF PREECLAMPSIA IN REDUCED VERSUS NON-

REDUCED TRIPLETS. D Skupsk~, S Nelson,~ A. Kowalik,’ M

Pofaneczky,x J. Hutson, M. Smith-Levitin,~ Z. Rosenwaks~ Dept Ob/Gyn,

The New York Hosp~tal-Coraell Univ Meal Ctr, New York, NY

OBJECTIVE: To evaluate the contribution of successfuhmplantatmn to the

rate ofpreeclampsia and perinatal outcome m triplet pregnancies conceived

by in vitro fertilization (IVF).

STUDY DESIGN: IVF triplet pregnancies (n=38) were matched for

maternal age, parity, race and delivery date with IVF twin pregnancies

(n=38) resulung from a single fetal reduction (spontaneously or by

mulnfetal pregnancy reduction) after successful ~mplantatmn of triplets

Pates of preeclampsia, rates of other maternal complications, factors

affecting implantation, a~d perinatai outcomes were compared.

Preeclampsla and severe preeclampsia were defined by ACOG criteria

Student’s t-test and chl-square were used for statistical analysis

RF~IJLTS:The triplet group had an increased rate of severe preeclampsia

(263%) compared to the twin (reduced triplet) group (7.9%)

Preeclampsia (both m~ld and severe combined) was also increased in the

triplet group (44 7%) compared to the twin group (15.8%). There was

no difference in other maternal complications of pregnancy or ~n factors

affecung implantation such as micromanipulation, Mean fetal weight was

lower m the triplet group, but gestational age at dehvery was not

significantly different

CONCLUSIONS: The rate ofpreeclampsia lsincreased in IVF triplets when

compared to IVF triplets reduced to twins. This suggests that fetal number,

placental mass, or some factors unrelated to the success of implantation are

more important to the development of preeclampsla than successful

implantation alone.

534 ASSOCIATION OF PRE-PREGNANCY OBESITY WITH THE DEVELOPMENT OF PREGNANCY INDUCED HYPERTENSION. S. Crane, T Dyd’, V Patel, R Aubry, R Artal Dlvtston of Maternal-Fetal Medicine, SUNY Health Sc:ence Center, Syracuse, NY. OBJECTIVE: To deternune the significance and magnitude of the assocmt~on between pre-pregnancy obesity and the development of pregnancy induced hypertensmn (P STUDY DESIGN: The population studied included all women dehvermg a live b~.h m Central New York State (CNY) between June 1, 1994 and May 3t, 1995. The data was obtained from the CNY Regional Permatal Data System Pre-pregnancy obesity was defined as a body mass index expressed as kalogram/meter~ (BMI) > 29 as per the Institute of Med~cme standards Pre-pregnant obese women were compared with non-obese women with regard to the development of PIH, defined as ~ increase m blood pressure of at least 30 mm Hg systohc or 15 mm Hg daastolic on 2 measurements taken at least 6 hours apart after 20 weeks ofgestatxon Ststa~cal analys~s mcluded clu-square test, crude odds ratio, and adjusted odds ratio using logasttc regressmn to control for possible confounders including maternal demograpluc variables and coexistent medical comphcahons RESULTS: 20,130 women were tncluded m the study of wtuch 19,707 had vahd pre-pregnancy BIVI! values 3,537 women ( 17 9%) were obese (BMIs > 29) 6 9°/6 of obese women developed PIH as compared to 2 0°/6 ofnon-obese ~vomen (p < 0 0001) The crude odds rat:o for obese wom~ to develop P1H was 3 6 (95% CI 3.1 to 4 3 ). Log~stac regresston was performed to control for potential confounders, which included grawdity, panty, gestat~onal &abetes, renal disease, thyroid dlness, and heart d~sease and y~elded an adjusted odds ratio of 3 9 (95% CI 3.3 to 46,P <0.01) CONCLUSIONS: Pre-pregnant obese women are at a s~gun’icantly incrensed risk of developmg PIH as compared to non-obese women, even after controlling for confounding variables, and should receive mtensffied prenatal, care

536 EXPRESSION OF THE TWO ISOFORMS OF THE INSULIN RECEPTOR IS ALTERED IN PLACENTA OF GESTATIONAL HYPERTENSION AND GESTATIONAL DIABETES

MELLITUS PATIENTS. H Valensiscx, D Dell’Annax, Y Yu Liux, R.

Laarox, C Romanml x and G Sest:x Dept Ob/Gyn and Internal Medscine, Tor Vergata University Rome - Italy OBJECTIVE: It has been speculated that alteration ra the relative expression of the two isoforms of insulin receptor (H!R-A and HIR-B) may contribute to m vivo insuhn resistance both m gestatlonal hypertension (GI~ and gestattonal drahetes melhtus (GDM). STUDY DESIGN: Insulin binding, receptor affinity and expressmn of the two receptor lsoforms in trophoblast plasma membranes from normal individuals (n=5), GDM patients (n=5), GH panents (n=5), and GH patient w:th intrauterine fetal growth retardation (IUGR) (n=5) were characterized Plasma membranes, prepared by dffferenhal

centrtfugation, were incubated wtth 50 pM [125[]-lnsuhn ~n the presellce or absence of Increasing concentration of native lllSulln All

immunological assay based on the ability of a human autoannbody (1-2 lgG) to mlub~t msuhn binding to the H1R-A lsoform but not to the HIR- B lsoform was used to measure the relative steady state expressxon of the two receptor protern tsoforms RESULTS: Insulin binding did not differ among the four groups of subjects (BIT = 0 26±0 6, 036_+0.7, 036:~0.7 and 0 34-~0.9 respectively for normal, GDM, GH and GH IUGR subjects) On the contrary, receptor affinity was slgmficanty lower m GH a/ld GH-IUGR

patients (half-maximal rahlb~tlon occurring at 1 6x 10-9 and lxlO"9 M

native insulin respectively) but not ~n GDM patients (ECS0 = 7x10-10

M rasuhn) compared to the normal control (EC50 = 4 x 10-10 M insulin) Low affinity HIR-B form is significantly increased in GDM, GH and GH-IUGR subjects (HIR-B/total HIR = 49_+9%, 53:~4% and 49±2%, respectively, P<O 05) compared with normal controls (HIR- B/total HIR =35~6%) CONCLUSIONS: An alteration in expression of the two insulin receptor lsofomas ~n peripheral tissues may play a pathogenelac role in cellular insulin resistance in GDM and GH

456 SPO Abstracts .]anua~y 1996 Am J Obstet Gynecol

537 CALCIUM AND MAGNESIUM ALTERATIONS FOLLOWING

NIMODIPINE TREATMENT IN BLOOD AND BRAIN OF RATS. M~

Tomht~s_o~, CA Standleyx , BA Mason, MA Belfort, DB Cotton Dept of

Ob/Gyn, Hutzel HospNVayne State Umvers~ty, Detroit, MI, and Beylor

College of Medicine, Houston TX

OBJECTIVE: The dlhydropyndme calcium channel blocker, mmod~pme has

been proposed as an alternative to magnesium sulfate in the treatment of

pr~eclamps~a To study the pharmacodynam~ca of thts agent we examined

dose related changes of central and serum mmod~p~ne levels and ~ts

mfluenca on total and iomzed calc=um (TCa, ICa), and magnesium (TMg,

IMg) m serum and brain samples

STUDY DESIGN: 50 female rats randomly reca=ved an =ntrapentoneal

~nlecben of one of five soluhons (N~mod~pme 0 5, 1 0, 5 0, 10 0 mg/kg or

vehicle control) Rats were sacnficad 15 or 30 mms after mlechon

N~mod~plne levels were measured ~n cerebral spinal fluid (CSF) and serum

TCa, ICa, TMg, and IMg were measured ~n serum, CSF, and various bra~n

regions Two way ANOVA was used for data analys~s

RESULTS: N~modipme was detected ~n 0%, 5%, 70%, 100% of CSF

samples at 0 5, 1 0, 50, 10 0 mg~kg doses respechvely. Interestingly, CSF

levels were an order of magnitude h~gher than serum levels Only the

h~ghest dose showed an increased level between 15 and 30 m~ns At

15mln IMg was increased in a dose dependent fashion (p<0 01), but no

d~fference was seen at 30 m~nutes TCa increased whde TMg decreased

in a dose dependent fashion by 30 m~ns (p<0.01) No d~fferences were

seen ~n Ca or Mg ~n any brain reg=ons

CONCLUSIONS: N~mod~p~ne appears to concentrate centrally at h~gher

doses There =s a s~gmficant rise =n serum TCa and fall ~n TMg by 30

m=nutes w=th no change ~n the =onized fract=ons Th~s study suggests that

mmodipine can alter extracellular calc=um and magnesium concantrahons

Its role as a vascact~ve agent warrants further study

539 UMBILICAL VENOUS FLOW MAY BE DEPENDENT ON UMBILICAL COILING. R.M. Lcwinskv *, S. Deganl *, H. Berger’, D. Spiegel*. Department of Obstetrics and Gynecology, Ultrasound Unit, Bnai-Zion Medical Center, Tcchnion-Israel Institute of Technology, Halfa, Israel. OBJECTIVE: To quantitate umbilical vascular coiling in thc antepartum period and to correlate the coifing index with Doppler flow parameters in umbilical vessels. STUDY DESIGN: We studied 45 normal term fetuses within 24 hours prior to delivery. The umbilical coiling index was calculated using sonographic longitudinal vlcws of cord vessels from several segments antcnatally, and by dividing the total number of helices by the cord length (cm) postnatally. Doppler flow velocities were obtained from umbilical arteries and vein in each cord. Flow parameters were correlated with umbilical coiling index. RESULTS: The mcan -+ (S.D.) sonographic and actual umbilical coiling index were 0.44 +. 0.11 and 0.28 + 0.08 coils/cm respectively, with a significant linear trend (r=0.71, p<0.001). No correlation was found between coiling index and umbilical arterial Doppler flow parameters. Coiling index significantly correlated with time averaged velocity (r=0.47, p<0.006) and flow (r=0.59, p < 0.001) in the umbilical vein. CONCLUSIONS: An antenatal umbilical coiling index which correlates with the actual index at birth can be obtained by ultrasound. The correlation between umbilical coiling and flow parameters in the umbilical vein may explain the previously reported association between decreased coiling and adverse pregnancy outcome.

538 NIMODIPINE IS NOT EFFECTIVE IN REDUCING N-METHYL-D- ASPARTATE-INDUCED SEIZURES IN THE RAT. MW TomlinsoD, CA Standleyx, WJ Kupskyx, BA Mason, MA Belfort, DB Cotton. Depts of Ob/Gyn and Pathology, Hutzel HospitalNVayne State University, Detrmt, MI, and Baylor College of Medic=ne, Houston TX OBJECTIVE: Nimodipine, a dlhydropyrid~ne calcium channel blocker, has been proposed as an alternative to magnesium sulfate ~n the treatment of preeclampsla. We evaluated the anticonvulsant activ~ty of nimodipine in an N-methyI-D-aspartate (NMDA) induced seizure model STUDY DESIGN: 45 female Long Evans rats were stereotaxically =replanted w=th a bipolar recording electrode ~n the hippocampus, and an rejection cannula in the lateral cerebral ventricle Rats were randomly assigned to receive an intrapentoneal (IP) injection of one of five solutions (nimodipine 0.05, 1.0, 5.0, or 10 0 mg/kg or vehicle control). 30 rains after IP injection rats were given 20pg NMDA through the ventdcular cannula. Behavior was observed throughout the experiment Electroencephalogrephlc activity was recorded for 20 rains following ~ntraventricular NMDA injection. Surviving rats were sacrificed 7 days later and processed for histology Data were analyzed with one way ANOVA RESULTS: Pnor to NMDA injection, increased somnolence was noted in rats rece=wng the highest nimod=p~ne dose. Following NMDA treatment there were no significant d~fferences between the nimodipine

doses or the control m: time to seizure onset, total seizure number, total seizure durahon, or mortality. Moderate to severe ddahon of the cerebral ventncles, particularly the 3rd ventricle, was noted histologically. CONCLUSIONS: Central NMDA receptors are known to be involved ~n seizure genesis. Nimodipine is not an effective ant~convulsant in th=s NMDA =nduced-seizure model and leads to acute hydrocephalus.

540 COMPARISON OF DUCTUS VENOSUS AND UMBILICAL

ARTERY DOPPLER WAVEFORM INDICES IN THE

PREDICTION OF ADVERSE PER1NATAL OUTCOME. T.C.~.~.

Chang, G.S.H. yecx, Y.L. YanX, Dept. of Maternal Fetal Medicine, Kandang Karbau Hospital, Singapore.

OBJECTIVE : To compare ductus venosus Doppler waveform indices

with absent/reversed diastolic flow in the umbilical artery (AREDF-UA)

in the prediction of adverse perlnatal outcome.

STUDY DESIGN : One hundred eighty high-risk obstetric patients

referred for fetal growth and Doppler assessment were recruited into the

study. Peak forward velocities during ventricular systole ($), early diastole

(D) mad occurrence of reverse flow during atrial contraction in the ductus

venosus waveforms were assessed. Referenced standards for S and D were

derived from a separate group of 80 healthy fetuses. The occurrence of

AREDF in the umbilical artery was also noted. Outcome measures were

a ponderal index < 3rd centile at birth and admission to NICU. RESULTS : Twenty seven infants (15%) had a low pondaral index at

birth. Ten fetuses (5.5%) had AREDF-UA; 17 (9.4%), 2(0.1%) and 1

(0.05%) had ductus vanosus S > 2 SD, D > 2 SD and reverse flow

during atrial contraction respectively. The sensitivity and PPV of AREDF-

UA in the prediction of a low pondaral index were 33% and 92%

respectively. Corresponding values for ductus vanosus S > 2 SD were

21% and 45% respectively. All ten fetuses with AREDF-UA required

admission to NICU; only four of these fetuses (40%) had an abnormal

duetus venosus waveform.

CONCLUSION : Assessment of AREDF-UA is superior to assessment of

duetus venosus flow in the prediction of adverse perinatal outcome in

fetuses at high risk of IUGR.


541 USE OF COLOR DOPPLER TO STUDY THE BREATHING

PATTERNS OF FETUSES OF DIABETIC MOTHERS. L. Wyse,

B Petnkovsky, E. Schneider, D. Jornsay, S. Baig.

North Shore Univerqty Hospital, Manhasset, NY.

OBJECIIVE: To assess the breathing patterns of the fetus of

the well-controlled diabetic mothers utd~z~ng color and

spectral Doppler technique.

$IUD¥ DESIGN: Fetal breathing was studied m 33 well-

controlled diabetic pregnancies and 62 uncomplicated

pregnancies at 2840 weeks by ultrasonography combined

w~th color-flow and spectral Doppler analys~s. The following

parameters were studied’ insp~ratory peak and duration;

expiratory peak and duration; and resptratory rate The Mann

Whitney and Spearman rank correlation tests were used for

statistical analysis.

RESULTS: The mean values of the breathing parameters of

both the d~abehc and control groups were similar. There was

a negative correlahon between respiratory rate and gestational

age ~n both groups (r = -0 45; p < 0.005). A positive

correlation between gestat~onal age and expiratory peak and

duration was observed in the diabehc group only (r = +0.45;

p < 0.005).

CONCLUSION: The fetal breathing patterns: ~nsp~ratory phase

and resDratory rates of both groups were s~mdar. However,

the expiratory phase ~ncreased in depth and duration w~th

advancing gestatton in the diabetic group only. Th~s d~fference

is {ntriguing and warrants further ~nvest~gatmn to determine

what possible factors or substrates other than glucose may be

influencing fetal breathing act~wty.

543 THE FETAL ECHOCARDIOGRAPHIC AND SONOGRAPHIC FINDINGS IN HYDROPIC FETUSES WITH BART HEMOGLOBINOPATHY. Peter C. Hsieh., D.M. Kuox, T.H. Chlux, T.T. Hsiehx, Dept. OblGyn,

Chang Gung Memorial Hospital, Taipei, TAIWAN. OBJECTIVE: To study the hemodynamJcs in hydrops fetalis ~ith Bart hemoglobinopathy (Bart’s hydrops), We used t~o-dimensional ultrasonography with the pulsed Doppler technique to evaluate the

hydropic and Doppler echocardiographic changes of fetuses in utero. STUDY DESIGN: Intracardiac transvalvular Doppler measurements

over pulmonary artery and aorta as well as two-dimensional fetal echocardiography (Acuson 128, 3.5 MHz, Mountain View) performed in 63 cases of hydropic fetuses y~th non-immune origin and 332 cases of fetal risk factors for prenatal cardiovascular evaluation. Among them, 32 Bart’s hydrops proved later and 104 normal fetuses were enrolled in this study for comparison. The ultrasenographic findings of Bart’s hydrops and their hemodynamic changes Were analyzed. The characteristics of two groups Were compared by using the unpaired Student t test. RESULTS: All of the Bart’s hydrope older than 24 gestational Week

had transvalvular velocity over 100 cm/s in both great arteries, but after 34th gestational ~eek no significant increase or even Io~er velocity ~as detected associated ~ith dilatation of both vessels.

Abnormal ultrasonogrsphic pictures found at mid-second trimester

Were pericardial effusion, polyhydramnios, and placental thickening. During the third trimester, prenatal sonography sho~ed appearence of the pleural effusion, ascites, scalp edema, associated ,~ith oligohydre~nnios and bulky placenta. CONCLUSIONS: Bart’s hydrops accommodated themselves to hemodynemic status Well before 34th gestational Week, and then the

cardiac output declined. This finding along ~th the oligohydramnios late in gestation suggested progressive cardiac decompensation. Intrauterine fetal death might be preceded by mitral and tricuspid

regurgitation, bradycardia, and poor ventricular performance. The differences between early and late hydropic changes depended upon the severity of the fetal anemia, which played the major role in the development of congestive heart failure.

542 COLOR AND PULSED DOPPLER OF THE SPLENIC

ARTERY IN THE DIAGNOSIS OF POLYSP]cENIA / ASPLENIA SYNDROME IN THE FETUS A. Abuhamad, D. Schneider~, C. Bullaboy~, D. Trippelx, D. BogdanX, A. Evans

Depts. Ob / Gyn and Pediatrics Eastern Virginia Medical

School, Norfolk, VA

OBJECTIVE’ To determine whether color and pulsed Doppler

of the splenic artery is helpful in the prenatal diagnosis of

polysplenia / asplenia syndromes (PAS).

STUDY DESIGN" Over a 24 month period, an attempt to image

the splenic artery by color and pulsed Doppler was performed on

all fetuses wxth the diagnosis of PAS. Postnatal follow-up was

obtained on all neonates.

RESULTS The diagnosis of PAS was performed on 7 fetuses

over the study period Mean maternal age was 22.3 +/- 6.02

years. Mean gestational age at diagnosis was 20 +/- 3 2 weeks

with a range of 17 - 25 weeks All fetuses had complex cardiac

abnormalities. Abdominal situs abnormalities was noted in 5/7

fetuses The splenic artery was imaged by color and pulsed

Doppler on 5/7 fetuses, all with polysplenia syndrome confirmed

postnatally Splenic artery could not be imaged on 2/7 fetuses,

both with asplenia syndrome confirmed postnatally Only 2/7

infants are currently alive.

CONCLUSIONS Color and pulsed Doppler of the splemc

a:tery can assist xn the prenatal diagnosis of PAS. This

information is of value in the prenatal counseling of such

pregnancies

544 UMBILICAL VENOUS VELOCITY PULSATIONS ARE RELATED TO ATRIAL CONTRACTION PRESSURE WAVEFORMS. K.L. Reed, D.G. Chaffin, C.F. Anderson~, A.T. Newman~, Dept OblGyn, AZ Health Sciences Center, Tucson, AZ OBJECTIVE: To identify the source of umbilical venous velocity pulsations, times of transmission from the atrial contraction pressure waveform to velocity waves in the infenor vena cava (IVC), ductus

venosus (DV), intraabdominal umbilical vein (UV) and intraamn~otic UV were examined in six instrumented sheep fetuses. STUDY DESIGN: Lamb fetuses at 125-135 days gestation were instrumented with solid state pressure transducers in the IVC, fluid- filled catheters in the IVC and descending aorta, and epicardial pacemakers. Three days postoperatively, IVC, DV, and UV veloc~t=es were examined with Doppler ultrasound. Normal saline was admin=stered until UV velocity pulsations developed (180 + 54 ml). In three fetuses, premature atrial contractions (PACs) were induced under baseline conditions and after UV pulsations developed. RESULTS: Time from atrial contraction pressure waveform (n=6) untd reverse flow in the IVC velocity waveform (n =4), reverse flow in the DV waveform (n=6), pulsation in the intraabdominal UV waveform (n = 3), and pulsation in the intraamniotic UV waveform

(n=6) increased as the distance from the atrium increased (R = 0.99, P<0.01). In fetuses with induced PACs, IVC pressure increased with fluid administration from 3.7+4.7 mmHg to 9.27+2.34 mmHg (P<0.01). Time from increased pressure waveforms with post-extrasystolic atrial contractions to the nadir of subsequent UV velocity waves decreased from 0.123 + 0.05 sec at baseline to 0.074:1:0.004 sec {P<O.01 ). CONCLUSIONS: Transmission time of atrial pressure into the venous circulation increases w~th distance from the atrium, and decreases with volume loading. Umbilical venous velocity pulsations derive from atrial pressure changes transmitted in a retrograde fashion.


545 ACUTE INCREASE OF FETAL HEMATOCRIT AND DOPPLER FLOW VELOCITY OF THE MIDDLE

CEREBRAL ARTERY.

G. Marl.x F. Rahman,x P. Olofsson,x A. Hamdi,x L. Arefi,x

M A1 Ammar,x R. Khan,x J. Copel. Depts Ob/Gyn, Yale

University School of Medicine New Haven, CT, and King Faisal

Specialist Hospital, Payadh (Saudi Arabia).

OBJECTIVE: The middle cerebral artery peak systolic velocity (MCA-PV) is increased in fetal anemia. Our hypo~esis for this study was that an acute increase of the fetal hematocrit decreases the

middle cerebral artery peak systolic velocity. Therefore, we studied

the fetal middle cerebral artery peak systolic velocity in Rh isoimmunized pregnancies prior to and following intrauterine transfusion.

MATERIAL AND METHODS: Fetal middle cerebral artery peak systolic velocity was studied by Doppler ultrasound in 10 pregnancies complicated by Rh-isoimmunization prior to and following intrauterine transfusion with packed red blood cells (Hematocrit = 80%). Mean gestational age was 27.4 weeks

(Standard deviation = 5.9 weeks). Paired t-test was used for statistical comparison. A p value < 0.05 was considered statistically significant.

RESULTS: The fetal hematocrit prior to the procedure ranged

from 7.3% to 40%, and it was between 24.8% and 53.4%, following the procedure. Mean middle cerebral artery peak systolic velocity was 50.2 + 12.2 cm/s prior to the procedure and it

decreased to 34.7 + 12.1 cm/s following the procedure (p<0.0001).

CONCLUSION: The increase of fetal hematocrit decreases the middle cerebral artery peak systolic velocity supporting data that this Doppler measurement may be useful for the diagnosis of fetal anemia.

547 ADRENAL ARTERY VELOCITY WAVEFORMS IN THE APPROPRIATE AND SMALL-FOR-GESTATIONAL AGE FETUS.

G. Marl.x B. Uerpairojkit,x A. Abuhamad, J. Copel. Depts Ob/Gyn,

Yale University School of Medicine and Eastern Virginia Medical School, New Haven, CT and Norfolk, VA. OBJECTIVE: To describe adrenal artery flow velocity waveforms in the appropriately grown (AGA) and the small-for÷gestat~onal-age (SGA) fetus. STUDY DESIGN: Adrenal artery flow velocity waveforms were prospectively obtained from 131 appropriate-for-gestational-age fetuses, 21 small-for-gestational-age fetuses, and 2 anencephalic fetuses with pulsed Doppler ultrasonography. The pulsatility index was used to quantify the waveforms. RESULTS: In the appropriate-for-gestational-age fetuses, the pulsatility index decreased linearly with advancing gestational age (Pulsatility index = 1.7 - 0.016 x gestational age; R= 46; p< 0.001). In 10 of 21 small-for-gestational age fetuses, the adrenal artery pulsatility index was below the 95% confidence intervals. In small- for-gestational-age fetuses, the presence of abnormal pulsatility index was associated with a higher incidence of late decelerations (6/10 vs 1/11; Fisher’s exact testp: p < 0.03), preterm deliveries (10/10 vs 4/11; Fisher’s exact test: p < 0.004) and cesarean sections (8/10 vs 4/11; Fisher’s exact test. p = 0.08). The two anencephalic fetuses had a adrenal artery pulsatility index above the 95% confidence intervals. CONCLUSIONS: Our data indicate that the pulsatifity index of the adrenal artery in the normal fetus decreases with advancirtg The results suggest that in small-for-gestational-age fetuses there is an increased adrenal blood flow (adrenal sparing effect). The small-for- gestational-age fetus with a normal adrenal artery pulsatfllty index appears to be at lower risk for pregnancy complications when compared to the small-for-gestational-age fetus with abnormal pulsatility index values.

546 CAN THE MIDDLE CEREBRAL ARTERY PEAK SYSTOLIC VELOCITY BE USED IN THE DIAGNOSIS OF ANEMIA IN FETUSES UNDERGOING REPEATED INTRAUTERINE TRANSFUSIONS?

G. Marl,x F. Rahman,x P. Olofsson,x M. AI Ammar,x A Hamd~,x

L. Arefl,x R. Khan, x A. Ludom~rsky, J. Copel. Depts. Ob/Gyn, Yale University School of Medicine New Haven, CT, Temple University, Philadelphia, PA, and King Faisal Specialist Hospital, Riyadh (Saudi Arabia). OBJECTIVE: The middle cerebral artery peak systolic velocity (MCA- PV) is increased in fetal anemia However, the effect of repeated intrauterine transfusion on the MCA-PV has not been evaluated. We studied the MCA-PV and the hematocrit in fetuses undergoing serial intrauterine transfusions. MATERIAL AND METHODS: MCA-PV was studied in 53 fetuses at risk for Rh-isoimmunizat~on who underwent 114 cordocenteses. There were 4 groups based on the number of transfusions previously performed: A) fetuses who were not previously transfused (n=44 cordocenteses); B) transfused once (n=25 cordocenteses); C) transfused twice (n=20 cordocenteses), D) transfused three or more times (n=25 cordocenteses). Multiple standard deviation of the mean (MSDM) for the hematocrit and multiple standard error of the mean (MSEE) for the MCA-PV were correlated by regressmn analysis. RESULTS: There was an inverse linear correlation between hematocrit MSDM and MCA-PV MSEE in fetuses prior to their first transfusion (p<0.05), second (p<0.05), and third transfusion (p<0 05). Following the third transfusion there was no correlation between MSDM and MSEE (p>0.05). CONCLUSION: MCA-PV was inversely correlated to the fetal hematocrit. Following three or more transfusions, the MCA-PV was not correlated to the fetal hematocrit. This needs to be considered when the MCA-PV is used to diagnose fetal anemia

548 EVIDENCES OF MAJOR CHANGES OF FLOW VELOCITIES ALONG THE DUCTUS VENOSUS. G Pennatix*, M. Bellottix, E Ferrazzix, A. Garbierix, G Pardi. *Dept

of Bioengeneering, Politechnics of Milan, and Dept of Ob/Gyn ISBM San Paolo, Umversity of Milan, Italy Objective. To assess the possible changes of flow velocities and angle independent indexes along the ductus venosus. Methods. A three dimensional fired-dynamic model (Flmte Element Method, F.E.M.) was developed for the intrahepatic umbiltcal vein and for the ductus This model t~kes into account and solves the complete motion equation, and calculates velocity, pressure, and pulsatile flow for every segment of the vessels. Doppler measurements were obtained on the same vessels in human fetuses at 34 weeks gestation Results. According to the F.E.M. model, the trumpet shape of the ductus determines a progressive reduction of 51% of the velocity profile from the isthmus to the outlet, and a variation of the S/A ratio This predictive model was in good agreement with the Doppler velocimetry data of the human fetus. Pressure changes vary reversely to velocity

Conclusions. Fluid-dynanuc fimte element model and Doppler measurements demonstrate that flow velocity varies from the

isthmus to the outlet of the ductus venosus Clinical use of

Doppler veloc~metry must take into account these differences.

uml~hcal vein

Vdo~ty 70

(cm/s) 60

ductus 3

0 6 Ume (s)

Volurue 174, Number 1, Part 2 ~qPO Abstracts 459

Pun J Obstet Gynecol

549 FLOW VELOCITY WAVEFORMS OF THE JUGULAR VEIN IN THE

APPROPRIATE-FOR-GESTATIONAL-AGE FETUS AND IN THE

FETUS AT RISK FOR CARDIAC FAILURE. R. Gonzalez~, G. MariX, R. Gomez~, M. Mondion~, A. lnsunza~, I. Rojas~, E. Oyarzun. Sotero del Rio

Hospital, Santiago, Chile; Universidad Catolica de Chile., Yale University

School of Medicine, New Haven, CT. OBJECTIVE: To determine flow velocity waveforms of the jugular vein

in appropriate-f0r-gestational-age fetuses (AGA) and in fetuses at risk for

cardiac failure.

STUDY DESIGN: Jugular vein flow velocity waveforms were

prospectively obtained from 59 appropriate-for-gestational-age fetuses.

(Group A), 6 severe intra-uterine-growth-restricted (IUGR) fetuses, and 2

sets of twins in two cases of twin-twin-transfusion syndrome (Group B).

Gestationai age in the AGA fetuses ranged fi’om 26 to 41 weeks (mean ¯ SD

= 35.6±3.8 weeks); it was between 29 and 37 weeks in the severe IUGR

fetuses 02.4±3.4 weeks). The two sets of twins were studied at 29 and 30

weeks gestations, respectively. The peak systolic, peak diastolic, mean

velocity, and the minimum velocity were measured.

RESULTS: A polynomial function expressed the changes of the peak

systolic, peak diastolic, mean velocity, and minimum velocity with

advancing gestation. The fetuses of Group B were matched for gestational

age with the 59 AGA fetuses. A t-unpaired test showed a significant lower

value for the minimum diastolic velocity in the fetuses of Group B (p <

0.05) when they were compared to the AGA fetuses. In 2 fetuses of Group

B, reverse flow was noted during atrial contractions; a finding never seen

in the AGA fetuses.

CONCLUSION: Blood velocity of the jugular vein increases with

advancing gestation. In fetuses at risk for cardiac failure there is either

reverse flow or a lower velocity value during atrial contraction than AGA

fetuses.

551 TI~ SIGNIFICANCE OF A LOW UMBILICAL ARTERY RESISTANCE I~DEX ~ A HIGH RISK POPULATION. DE Cludfig. RW GilP, PS Warrenx. Royal Hospital for Women and CSIRO

Division of Radiuphysics, Sydney, Australia.

OBJECTIVE: The finding era low umbilical artery (UA) resistance index

(RI) (below the 10th centile for gestation) on prenatal Doppler

velocitometty is usually regarded as a reassuring or normal result. We

reviewed our perinatal Doppler database to determine if this was true.

STUDY DESIGN: Fallow-up data was obtained on 808 referred high risk

patients who underwent Doppler studies within 14 days prior to delivery.

Of these, 43 (4.9%) were found to have an RI below the 10th centile.

RESULTS: The finding era low UA RI was not associated with an overall

increased risk of." stillbirth, neonatal death, emergency Caesarean section,

low Apgar scores, low birthweight, maternal diabetes or fetal abnormality.

A low RI was, however, significantly associated with the follovAng factors:

Factor Low RI Normal Relative p value n=43 RIn=570 Risk

Antepertum 23.2% 12.1% 1 9 .03

hemorrhage Preterm ruptured 23.2% 11.0% 2 1 02

membranes Preterm labour 25.6% 10.0% 2.6 .002 High umbilical 37.5% 9.7% 3.8 .0001

venous flow (>95%) Days in intensive 12.3 4.5 .0001

care nursery (tqlCU) Gestation at delivery~ 34.6 wks 36.8 wks I .004

CONCLUSIONS: In our experience low UA RI is asso~inted with increased perinatal morbidity in the form of earlier gestation at delivery and

longer NICU admissions. We propose that a low UA RI is seen in those conditions which may lead to a hyperdynamic fetal c~rculatiou.

55O CORRELATION OF COMPREHENSIVE PLACENTAL PATHOLOGY WITH UMBILICAL ARTERY DOPPLER STUDIES iN GROWTH-RETARDED FETUSES.

~n *’ M. Divon*, V. Mmior*.J. Pezzullo*. Pednatal Research Facility, ts of Pathology & OB/GYN, Georqetown University Medical Center,,

Washington DC, Albert E~nstem CoJlege of Medicine, Bronx NY. OBJECTIVE: To iden~y histopatholo~c features ass~ctated with different types of abnormal umbilical artery (UA) Doppler veI~cimetry in intrauterine growth restriction

avascul~r terminal wllh ’hemorrhag=c endovasculitis’, fetal stem vessel thro~b=’~ complete vascular oblrterabon, severe mural hyperplasia and disorganizat=on) were graded as absent, <1 les~on/10x field or >1 lesfon/10x field. Ind=vidual lesion scores were summed togive a total les=on burden. RESULTS: Fetal stem vessel obhteration, "hemorrhagic endovascuhtis", aN avascular term=nal villi were related to abnormal UA Doppler vs. normal UA Doppler (each p<0.05). D=stributtons of summed intraplacental vase-occlusive lesions and uteroplacental vascular lesions in the 5 UA Doppler groups are shown.

~ONCLUSIONS: VHIous pathology: (1) is associated with a wide range of ~ntraplacental vascular lesions; ¢2) 15ecornes clearly visible one] a~ter EDF is alSsent; aN ~3) d~s ,net sh?w a deaf"dose-dependence" on pathology related to abnormal uterine vascu=ar perrusion.

552 THE VALUE OF UMBILICAL ARTERY BLOOD FLOW VELOCITY IN PREGNANCIES COMPLICATED BY SYSTEMIC LUPUS ERYTHEMATOSUS (SLE) D. Farine, S. .~- " ~, T.G. Teoh~, G. Ryan, G. Seaward, J.W.K. Ritehie. Div. of Perinatology, Dept. Ob/Gyn Mt. Sinai Hospital, U~v. of Tomato, Ontario, CANADA Objective: To assesss the role of umbilical artery waveforms in the surveillance of pregnancies complicated by SLE. Study design: A retrospective study of 56 patients with SLE managed between 1988 and 1994. Fisher’s exact test was used. Results: Absent end-diastolic velocity (AEDV) was detected in 6/56 pregnancies. No other forms of abnormal blood flow were observed. Outcomes were: Diagnosis AEDV Normal Doppler P-

n=6 n=50 value IUGR 4 12 0.001 Preterm 5 12 0.001 Delivery Hypertension 4 6 0.0 ! Cesarean Section 5 9 0.0002 NICU-Admissions 6 9 0.0002 Perinatal deaths 0 1 * 1"I.S. "Maternal death due m TrP at 31 weeks. Conclusions: A high incidence of AEDV (11%) was noted in

patients with SLE. AEDV identifies at risk pregnancies requiting

intense fetal surveillance.


553 CHANGES IN RIGHT ATRIAL SYSTOLIC VELOCITY IN DISCORDANT TWINS I. Forouzan. E. Graham: University of Pennsylvania Medical Center, Philadelphia, PA. OBJECTIVE: The objective of this study was to study the changes in the flow velocities of the tricuspid valve among intrauterine growth restricted twins, STUDY DESIGN: During past 4 years, discordant twins referred for evaluation to our medical center were evaluated by Doppler flow velocity measurements. Tricuspid valve (TV) flow velocities were measured in both twins. Those twins with one Intrauterine growth restricted (IUGR), and one appropriately grown fetus were included in this analysis. Those with chromosomal, or structural abnormalities, and twin-twin transfusion syndrome were excluded. We selected these twins for study in order to eliminate any maternal condition influencing the fetal circulation. Eighteen pairs met these criteria. TV flow velocities were compared using t-test. RESULTS: Mean gestational age at time of study was 305:2.4 weeks gestation, and at the time delivery was 34+2.6 weeks gestation. Mean early diastolic, and peak systolic velocities through the tricuspid valve in both IUGR and normally grown fetuses are presented in the table:

Early Diastolic Flow Peak Systolic Flow Velocity Velocity

(mean±SO) ImeanS:SD) Appropriately, Grown 34.05:6.7 50.3±1.2 1UGR 35.8±4.43 42.15:1.2

Sa/~nificance p=0.35 . p=0.049 Velocities are shown as centimeters per second SD: Standard deviation CONCLUSIONS: The velocity of blood flow through the TV during atrial systole is lower in growth restricted twins compared to appropriately grown pairs when growth restriction is secondary to uteroplacantal insufficiency. Its potential in differentiating discordancy due to the other causes remains to be explored.

555 CEREBELLRR DOPPLER UELO£1METRY I N THE

APPROPRIATE RNO SMRLL-FOR-GESTATIONAL-AGE FETUS. B. UerpaJro_ikitx, L. Chan, A. Ludomarska, J.A. Copel, E.

Martinez, E.A. Reece, G. Man. Depts. of Ob/Gyn, Temple Umv Sch of

Med, Phila., PA. and Yale Umv Sch of Med, New Haven, CT OBJECTIVE: TO compare superior cerebellar artery flow velocaty

waveforms in the appropriately grown and the small-for-gestatlonal-age

fetus. STUDY DESIGN: Superior cerebellar artery velocrty waveforms were

prospecttvely obtained from 172 apprnprrate-for-gestataonal-age

fetuses between 17 and 41 weeks gestation. The pulsatility rndex was used to quantify the waveforms. Superior cerebellar artery velocity

waveforms were also obtained from 30 small-for-gestat~onal-age fetuses

divided ~nto group A (n=15) with normal umbilical artery pulsatihty index and group B (n=15) with abnormal umhihcal artery pulsatd~ty

index. The transverse cerebellar diameter was measured in all small-for- gestat~onal-age fetuses. RESULTS: The superior cerebellar artery pulsatility index was best represented by a second order polynomial equation [Pulsatrixty index =

0.145 + 0.101 x (Gestatlonal Age) ~ 0.00197 (Gestational Age)2]

Small-for-gestat~onal-age fetuses of group A had a superaor cerebellar

artery pulsatdity ~ndex in the normal range; whereas 13/15 fetuses of

group B (86.7%) had a pulsatihty index value below the 5th percentile. The transverse cerebellar diameter ~n fetuses of group A was in the normal range an 10/15 cases; whereas ~t was in the normal range in all

the fetuses of group B. CONCLUSIONS: Small-for-gestat~onal-age fetuses w~th abnormal

umbilical artery pulsatality ~ndex have "cerebellar sparing effect" as suggested by a superior cerebellar artery pulsatihty index below the normal range and a normal transverse cerebellar diameter.

554 FETAL AND UTERINE HEMODYNAMICS DURING AND AFTER MATERNAL CARDIOPULMONARY BYPASS(CPB)

M. Khandelwal, J. Rasanenx, A. Ludormirski, P. AddomzinX, E.A.

Reece. Dept. Ob/Gyn, Temple Univ. School of Medicine and Pennsylvania Hospital, Phila, PA. OBJECTIVE. We report the first study descrthmg fetal hemodynam~cs

and uterine blood flow before, during and after maternal CPB, and compared to normal reference values an the second mmester. STUDY DESIGN: A 33-year old female with severe aortic

regurgitation underwent aortic valve replacement at 19 weeks gestation A non-pulsatile centrifugal Bao-medacus cardiac pump and Maxima-plus oxygenator was used for CPB. Blood velocity waveforms were recorded

by Doppler ultrasound at the level of maternal main utenne artery, fetal umbihcal artery (UA) and vern (UV), muddle cerebral artery (MCA) and descending aorta (DAn), ductus venosus (DV) and ~nfer~or vena cava

(IVC). Pulsaullty index (PI) values were calculated. Total cardiac bypass t~me was 74 m~nutes. The mean arterial pressure (MAP) was 77-90 mmHg, peak flow rate of 3 5-4.0L/mm/m2, core temperature of 34-35°C.

RESULTS. Preoperatively, fetal hemodynam~c parameters were within normal hmits Intraoperatwely, fetal bradycard~a ensued after aortic clamping (120 to 75 beats/minute), with a use in UA PI (1.7 to 7.1), and d~sappearance of dlastohc flow. The MCA PI (2.0 to 0.92) and Dan PI (2.22 to 1.97) decreased despite fall an FHR while the PI an 1VC (2.7 to

9.3) and DV (0.6 to 1.7) ~ncreased. Uterine artery PI before surgery was

3 9 (normal=0.5-1.5 at 20w). Durrng bypass after aortic clamping no~pulsanle utenne flow was noted. Postoperatively, the PI amproved

to 1.0. Two days postoperatively, hydrocephalus and hydrops were

noted. Four weeks later, the UA had absent end-daastohc flow. CONCLUSIONS: Desprte high peak flow rates, MAP and normotherrma, fetal outcome was dismal. Effects of nonpulsatlle versus

pulsat~le maternal CPB, fern-placental oxygen consumption and extractmn under normo-and hypothermic cond~hons need further study.

556 PROSPECTIVE EVALUATION OF THE EFFECT OF MATERNAL COCAINE USE ON FETAL CEREBRAL BLOOD FLOW Whiteman VE, Stanell Wx, Reece EA, Dept.

OB/GYN/RS, Temple University School of Medicine, Philadelphia, PA

OBJECTIVE: To evaluate the effect of recent maternal cocaine use on the fetal cerebral circulation using Doppler Flow analysis

STUDY DESIGN: Fifty (55) normal singleton fetuses were evaluated prospectively between the 24 and 36 weeks gestational

age. After an informed consent was obtained, all patients provided a urine specimen for cocaine metabolites. Following isolation of the Circle of Willis by Color Doppler enhancement, Doppler flow analyses were performed obtaining the Pulsatility Index (PI) of the internal carotid artery (IC-P1). The study groups were categorized as cocaine positive or negative. Using multiple logistic regression analysis the data were analyzed; Statistical significance at p<0.05. RESULTS: Our preliminary results are shown below:

Cocaine Positive Cocaine Negative P-value

Subject’s Cocaine Status N = 14 N = 41

Maternal Ago 31.7 + 5.7 22.4 + 4.6 0.00008

IC-PI 1.94 + 0.17 1.73 +0.41 0.00193

Through multivariate analyses, only cocaine use was found to be

associated with an elevated IC-PI (Odds Ratio = 11.35 with 95% Confidence Interval 86.9 to 1.48.)

CONCLUSIONS: The preliminary results of our .study demonstrate ~treduetion in the fetal c.erebral blood flow associated

with maternal cocaine use in vre~nanev. This could votentiallv

have adverse long-term consequeneezs for such infants.

Poster Session V

Saturday, February 10, 1996 10:30 a.m.- 12:30 p.m.

Kohala Ballroom

CATEGORIES Prematurity

Labor Fetal Therapy

Intrapartum Fetal Evaluation


Judges: Jennifer R. Niebyl, MD Richard 14. Paul, MD ~oshua A. Copel, MD


557 THE NATURAL HISTORY OF A POSITIVE RESPONSE TO TRANSFUNDAL PRESSURE IN WOMEN AT RISK FOR CERVICAL INCOMPETENCE. E Guzman. A Vlntzfleos, D McClean, M Martins, C Banltox, M Hanleyx D=v~slon of Maternal-Fetal Medicine, UMDNJ-Robert Wood Johnson Med=cal School/St Peter’s Medical Center, New Brunswick, NJ OBJECTIVE: To observe the evolution of the endooervical canal length (ECL)In women at risk for cerv=cal incompetence (CI) after a pos=tlve response to transfundal pressure (TFP) STUDY DESIGN: Ten women ~ risk for CI had a mldtnmestar cervical eva~uabon us=ngtransvaginalso~graphyand transfundal pressure (TFP) Us=ng a transvag=nal probe, the endocervical canal length (ECL) was measured. Transfundal pressure was apphed and the ECL was remeasured All patients had a positive response to TFP as defined as a decrease in ECL The d@tal exam=nabon of the cervix revealed a closed and long ce~x =n all 10 eases In 9 of the 10 patients, repeat exammabons were performed until 1) the ECL progressively shortened to <10 millimeters or 2) dlglta~ examina~on revealed a dilated cervix. These endpoints were md~cat~ons for placement of a cervical cerclage At each examlnatian TFP was appled if the membranes were not at the external cervical os. The ECL after the first response to TFP was compared to the ECL at the last exam~cation If at the last exam=nation the membranes were at the extemal os, no TFP was apphed and the ECL was zero One patient was lost to follow up, but the obstetnca] outcome was available Data are reported as med=an (range) millimeters with statistical anaysls performed using WiIcoxon s=gned rank test. Statistical s=gn=ficance was set at p<0 05 RESULTS: The median tJme interval between the first and final examination was 1 week (1-3 weeks) in 9 of the patients

First Examination Final Exam~nati0n Gesta’oona] Age(weeks) 19.0 (15-22), n=10 20.5 (18-24), n=9* ECLafterTFP(mm) 122 (4-10), n=10 0.0(0-95), n=9* p=00077 * One pa’0ent was lost to sonograph=c follow up Six patJents had membranes at the external cervical os before apphcat~on of TFP at the last examination. At the initial exam=nahon, despite ECL shortan=ng ~n response to TFP, there was no e~dance of CI on digital examination In all cases N~ne patients were eventually treated with cervical cerclage due to progressive cervical changes. Of these, six dehvered _>36 weeks, two at 27 & 34 weeks, and one miscarried the day after the procedure at 18 weeks. The ore patient lost to sonograph~c follow up mlscamed at 23 weeks of gestation, 6 weeks after a response to TFP CONCLUSION: In patients at risk for CI, a TFP response of shorten=ng of ~e ECL requires treatment with a cervical cerclage as It is associated w=th progressive cervical changes over 1-3 weeks This informat=on calls for as =nvestlgat]on on the s~gnificance of a pos=tlve response to TFP in women not at risk for pregnancy loss

559 IMPACT OF OUGOHYDRAMNIOS ON OUTCOME OF PROM AT 18~.,.

28 WKS. A, AL-Malt, S. Amini=, M. AbboucP, B. Cohlanx, E. Amon,

H. Winn. Div. of MFM and Neonatalogy, St. Louis U., St. Louis, MO.

OBJECTIVE: To determine the impact of oligohydramnios (oligo) on

pregnancy outcome after PROM st 18-28 wke.

STUDY DESIGN: Patients admitted to one tertiary care center (1993-

1994) with documented PROM, no labor at 18-28 weeks were

studied. Inclusion criteria: singleton, no fetal malformation and NL

fluid prior to PROM. Amniotic fluid index (AFI) ~ E cm was considered oligo. Antibiotics were given pending GBS culture results.

Steroids were used in absence of infection. Tocolytics were given for

subsequent preterm labor.

RESULTS: 118 patients had PROM at 18-28 wks; 69 had oligo and

49 had NL fluid within 1 wk of delivery. There was no significant

difference (NS) between the 2 groups (oligo ve NL) in maternal age,

race, parity, gestational age (GA) at PROM (172 + 17 vs 171 + 14

d}, latency period (9 ± 14 vs 8 ± 13 d), deliveries < 24 wks

(21.7% vs 22.4%), antibiotics use (75% vs 69%), clinical chorio

(48% ve 39%), histological chorio (65% vs 66%), or neonatal (NN)

sepsis (29% vs 30%). C-section (36% vs 20%) end abnormal fetal

heart rate (FHR) (56% ve 38%) were marginally higher (p = 0.06)

for oligo group. Differences between the 2 groups wrth respect to

stillbirth (7% vs 8%) end NN mortahty (30% vs 20%) were NS. The

use of steroids and tocolytics (30% vs 49%; and 48% vs 74%; p <

O.OS) was significantly lower in the oligo group. After adjusting for

confounding variables, oligo patients were 2.3 times more likely to

experience NN death. However, this difference remained statistically

NS (p = O.16). Multivariant analysis showed that GA at PROM was

the most important predictor of perinatal mortality followed by

latency period.

CONCLUSION: Oligo following PROM at 18-28 wks has no

significant effect on latency period or NN sepsis but marginally

increases the risk for abnormal FHR and C-section. Although

statistically NS, the risk for NN death may double.

558 PHYSICAL VIOLENCE AND VERBAL ABUSE IN PREGNANCY.

Shumwey=~ P. O’Campox, A. Gialenx, A. KhouzemP, F. Witter, K.

Blakemore. Depts. Gyn/Ob & Mat. Child Hifh., The Johns Hopkins

Univ. Sch. Mad. & Hyg-Pub Hlth., Balto., MD, St. Louis Univ., St.

Louis, MO.

OBJECTIVE: The prevalence of physical abuse in prenatal populations

has been estimated to be from 3,9 to 20,4%, It has been suggested

that violence during pregnancy may be associated with an increase

in perinatal morbidity with one recent report of low birth weight. The

hypothesis is that physical violence and verbal abuse in pregnancy

ere more likely in younger women and that more severe suboptimal

fetal growth is seen in increasing levels of abuse.

STUDY DESIGN: A cohort of 636 women attending our Adult Obstetrical Clinic for their first prenatal visit from December 1989 to

September 1990 were approached of which 567 enlisted as study

participants. 401 participants successfully completed ell prenatal

interviews. Violence data were obtained from these interviews end

the women were grouped by degree of violence. Obstetrical and

neonatal outcome data were obtained by abstracting the maternal

and neonatal medical records.

RESULTS:

Type of Women’s age Birthweight Gestational

Violence (years) (grams) age (weeks)

None (n = 134) 24.47 3237 38.68

Negative verbal 23.05 3176 38.82

interaction (n = 144)

Moderate violence (n=66) 22.04 3143 38.56

Severe violence (n = 68) 22.67 3376 38.90

t-test, 3 d.f. p = 0.006 p = 0.298 p = 0.892

CONCLUSIONS: Gestetional age at delivery and birth weight were

not associated with verbal or physical abuse in this intercity cohort

of women contrary to other recent reports in the literature. Younger

women and unmarried women were at risk for more severe violence.

560 OUTCOME AND COST ANALYSIS OF PRETERM PREMATURE

RUPTURE OF MEMBRANES IN AN OUTPATIENT SETTING.

MarOn~, M Gardner, K. HowelP, M Griffinx, J.L. Gonzale#, L. Curet. Dept

of Ob/Gyn, Univ of New Mexico.

OBJECTIVE: To analyze the cost of outpatient management of premature

preterm rupture of membranes (PPROM) compared to long term hospltahzatinn

STUDY DESIGN: We reviewed maternal and neonatal records from January

1992 to June 1995. Our analysis was hmited to those patients with PPROM <37

weeks gestation Maternal records were evaluated for gestational age (GA) at

ROM, GA at delivery and maternal infection. Neonatal records were evaluated

for NICU admission, diagnosis of respiratory distress syndrome (RDS),

necrotlzlng enterocohtis (NEC), mtraventricolar hemorrhage (IVH) and seps~s

Patients managed m the outpataent setting with daily home fetal heart rate

momtormg and weekly clinic visits and anmiotic fluid volume evaluation.

RESULTS: Over the study period 62 patients wth PPROM were managed as

outpatients. Of these mothers, 3.2% had chormammourtis or postpartum metritts

Of the 64 newborns (2 pairs of twins), there were 2 neonatal deaths The

neonatal sepsis rate was 6.3%, the rate of RDS-22.2%, IVH-4.8%, and NEC-

3.2%. This data is comparable to nationally published rates of neonatal

morbidity The mean latent period from PPROM until delivery was 29 5 days

The mean number of hospital days prior to delivery was 5 7 The mean costs

for outpatient management was $11,36g/patient The cost would have been

$33,517/patient if these patients had remained hospltalmed until delivery

CONCLUSIONS: Outpatient management of patients wth PPROM offers a

marked reduction in healthcare costs w~thout any apparent increase in the

incidence of maternal or neonatal morbidity Prospective trials should be

undertaken to estabhsh the uuhty and the ideal frequency of home fetal

monitoring in the outpatient management of PPROM


561 PRETERM AND TERM PREMATURE RUPTURE OF

MEMBRANES: SUBSEQUENT PREGNANCY OUTCOME.

RK Bansalx, RK Laros. Dept. Ob/Gyn; Univ. of Ca, San Francisco, San

Francisco, CA

OBJECTIVE: In two prevmus studies, rates of recurrence of preterm

premature rupture of membranes (PPROM) were reported as 21% and

32.2%. We studied our own instituhon’s recurrence rate and whether term

premature ropture of membranes (TPROM) is associated with adverse

outcome in the subsequent pregnancy.

STUDY DESIGN: 3501 patients (pts) had both prenatal care and

dehvery in 2 consecutive, singleton pregnancies at UCSF between 1975-

1994 We determaned the occurrence of preterm delivery (I?TD), PPROM,

and TPROM for all pregnancies. Three groups were formed based on the

outcome of their imtial pregnancy: PPROM, TPROM, and term delivery

without PROM. Proportions were analyzed using the z-statastic

RESULTS: In the 100 pts with PPROM in their initial pregnancy. 8%

had recurrence of PPROM, which is significantly less than the recurrence

rates reported previously (p<.001); 17% had PTD without PPROM for a

total PTD rate of 25%; and of those deh~’ering at term in their subsequent

pregnancy, 20.3% had TPROM. Of 761 pts with TPROM in the ~nitial

pregnancy, 2.6% had PPROM in the next pregnancy, and a total of 7 1%

had PTD, the recurrence rate of TPROM was 17.2% In 2640 pts w~th

term deliveries without PROM, the subsequent pregnancies had PPROM,

PTD, and TPROM rates of 1.8%, 4.6%, and 11.9%, respectively. The

rate of subsequent PPROM and PTD were s~gnificantly higher (p’s< 01)

in the PPROM group. PTD was more common in the TPROM group than

in the without PROM group, 7.1% vs 4 6% (p= 009). Both the PPROM

and TPROM groups had higher rates of subsequent TPROM than tile

without PROM group (p=.045 and p< 001, respectwely).

CONCLUSIONS: In our populatmn, PPROM recurred sigmficantl¥

less often than previously reported, but was still high as were the rates of

PTD and TPROM in the subsequent pregnancy TPROM in the ~mt~al

pregnancy was also associated w~th increased rates of P’rD and TPROM in

the next pregnancy. These findings suggest commonalttes m the

etiologies of TPROM and PPROM.

563 THE PREVALENCE AND CLINICAL

SIGNIFICANCE OF AMNIOT1C FLUID

COLONIZATION WITH U. UREALYTICUM AND M.

HOMINIS IN PATIENTS WITH PRETERM LABOR

AND/OR PREMATURE RUPTURE OF THE

MEMBRANES. L. Coultrip, C. lverson~ T. Tra~ R. I~

Cifuentes. Dept. Ob/Gyn, HCMC., Minueapolis, MN.

OBJECTIVE: To determine the clinical significance of

positive AF cultures for U.Urealyticum and M. Hominis in

patients with PTL and/or PPROM.

STUDY DESIGN: Maternal and neonatal outcome data in

patients with positive AF cultures for M. Hominis or U.

Urealyticum (N=18) were compared to patients with

negative AF cuhures (N = 241) and positive AF cultures for

other organisms (N=12). Proportions were compared

Fisher’s exact test or X2. Continuous data was compared by

student T test.

RESULTS: In contrast to pts with AF cultures positive for

other organisms, GA at amnio, AF wbc, maternal/neonatal

infection, were not different. Patients with cultures positive

for only U.urealyticum/M.Hominis were less apt to deliver

within 24 hrs [Odds 0.07, Cl 0.001 & 0.7], histologic CA less

frequent (p=0.08) and AF glucose was not decreased

(median 23 mg/dl).

CONCLUSIONS: A frequent AF isolate, isolated U.

urealyticum/M. Hominis may not portend immediate

delivery. AF glucose may be useful in determining clinically

significant microbial AF invasion.

562 MATERNAL HYDRATION IN PATIENTS WITH PRETERM PREMATURE RUPTURE OF MEMBRANES. J. Boy!~x, J. lams, S Gabbe Dept. of Ob/Gyn, The Oh=o State University Medical Center, Columbus, OH. Objective: To determine the effect of maternal Intravenous hydration on the ammot=c fluid index (AFI) in patients w~th preterm premature rupture of membranes (PPROM). Study Design: A prospective, randomized bhnded trial was conducted m 23 patients w~th PPROM managed expectantly between May, 1993 and August, 1995. The hydration group received 3L of DsW intravenously over three hours. The control group

received no IV hydration. Groups were compared for differences m pre- and post-infusion AFI, BUN, creat=nme, serum osmolahty, and urine spemflc grawty. One sonographer performed all AFI measurements without knowledge of group assignment. Data were analyzed w~th Student’s t-tests. Resulte: Thirteen pahents were randomized to the hydration group, and 10 to the control group The groups were s~mdar m age, grawty, panty, gestat~onal age, and pre-study AFI. There was no d=fference m mean AFI after hydration between the hydration group (2.1 _+ 2.4cm) and the control group (2.4 + 3,8cm). Despite aggresmve parenteral hydration, there were no laboratory parameters that reflected a change m maternal volume status compared to controls. Creatmme, BUN, serum osmolalnty, and unne specff=c grawty were s~mdar between groups¯ Conclusion: Maternal intravenous hydration does not increase the ammot~c fluid index compared to controls =n patients w~th PPROM.

564 THERE ARE DIFFEREHT HISTOLOGIC GROUPS WITH DISTINCT CLINICAL CHARACTEEISTICS A~ONG PATIENTS WITH PRETERM PRENATUEE RUPTURE OF .MEMBRANES. Fernando Arias MD, Alejandro Victoria ND , Karen Cho NDW~ and Frederick Kraus MD*. St. John’s Mercy Medical Center, St. Louis, MO. ObJectzve: To determine zf patlents with preterm premature rupture of the membranes (PPROM) before 34 weeks may be classified on the basis of the hlstologic examination of the placenta and whether these groups have dzfferent clinical characteristics and pregnancy outcomes. S~dy Dessert: Prospective cohort study. Methods: 235 patlsnts with PPRO~, 215 wath slngleton and 20 with twin pregnancies, were classifled accordin~ to the findings on the placental histologic examination and their prenatal and intrapartum course and their perinatal mortality and morbidity were analyzed and compared. Results: 102 patients (43.4~) exhiblted acute infl~mmatory lesions; 48 (20.4~) had vascular leslons; 48 (20,4~) had both inflammatory and Vascular lesions; 31 (13.1) had no pathologic findings; 4 had vxllous edema and 2 (0.8~) had chronlc villitis. Patlents with acute inflsmmatory leslons slngle or m~xed were slgnlficantly different from patients with vascular lesions and with no pathologlc f~ndlngs with respect to their gestational age at the time o~ PPROM (p = .00005)~ gestational age at dellvery (p = .000001), perinatal mortality (p = .02) and incidence of culture proven

may be classlfled accordin~ to the placental histologlc fzndings in groups that have different clinical manifestations, prognoses and outcome.


Am J Obstet Gynecol

565 GROUP B BETA-HEMOLYTIC STREPTOCOCCUS AND PRETERM PREMATURE RUPTURE OF MEMBRANES: A RANDOMIZED, DOUBLE-BLIND CLINICAL TRIAL OF THE EFFICACY OF AMPICILLIN. I. Grablex, P. G-arcia, D. Pentx, M. Socol Depaffment of Obstetrics and Gynecology, Northwestern Umversity Medical School, Chicago, IL. OBJECTIVE: To determine whether ampicillin prolongs the latency period after preterm premature rupture of membranes (PPROM) in patients colonized with group B beta-hemolytic streptococcus (GBBS) STUDY DESIGN: Sixty patients presenting at <35 weeks’ gestation with PPROM were included in the study. Participants were randonuzed to receive either intravenous empicillin for 24 hours followed by oral empicillin until delivery or intravenous and oral placebo Cervical, vaginal, and rectal cultures for GBBS were obtained. All patients were managed without the use of tocolytac drugs. To detect a 50% increase in the GBBS posilave patients undehvered at 48 hours after PPROM with a power of80%, ~t was determined that 10 patients would be required m each group An interim analysis using the O’Bnen and Fleming Multiple Testing Procedure and a conditional power analysis was conducted by a blinded statistician. The F~sher Exact Test, student t-test, Wflcoxon S~gned-Rank test, and chi square were used for statistical analysis. RESULTS: Fifteen patients had cultures positive for GBBS. GBBS positive patients who received empicillin (n=8) were significantly more likely to be undelivered at 48 hours after PPROM compared to those paUents who received placebo (n=7) (100% vs. 43%; p=0 01) At 7 days the~e was no differance m the percentage of GBBS positive patmnts (63% vs 29%; p=0 19) who reanamed undelivered GBBS negative patients who received ampicfllin were not more likely to be undellvered at 48 hours (87°/* vs. 64%; p=0.07) or 7 days (39% vs. 27%, p--0.40) after PPROM comlxued to placebo. There were no differences in the incidence of cesarean section, chorioammooitis, and postpartum endometritis CONCLUSION: Antibiotics will increase the percentage of GBBS culture pomttve patients who remain undehvered at 48 hours at~ PPROM. Antibiotic therapy may provide a window of oppo~umty for maternal treatment w~th corticosterolds to decrease the risk of neonatal morbidity in these preterm gestations.

567

24 ~ 34 WEEK~’ GESTATION. H. How, C. Cook, V. Cook, J. Spinnato, Dept. of 0b/Gyn, University of Louisville, Louisville, KY and Marshall University, Huntington, WV.

OBJEt’TZ~ES To report the pattern of change of the lecithinlsphingomyelin ratio (L/S) in patients with pretermpremature rupture of membranes (PPROM) between 24 and 34 weeks’ gestation. ST~ DESIG~t The L/S using transvaginally (n = 112) and transabdominally (n = 6) collected amniotic fluid in 49 patients with PPROM between 24 and 34 weeks’ gestation was determined prospectively on admission and every 48 to 96 hours until fetal lung maturity (FLM~ was achieved, i.e. L/S Z 2.0. Ali patients received weekly intramuscular steroids (12 mg of betamethasone every 12 hours for 2 doses). RES~LTSt Among patients at h 29 weeks gestation with an initial L/S of h 1.5 and < 2.0 (n = 17), 15/17 (88%) reached maturity at a mean of 3.1 ± 1.7 days. With an initial L/S of Z 1.0 and < 1.5 (n = 15), 13/15 (87%) reached maturity at a mean of 4.1 ~ 2.0 days. With an initial L/S of < 1.0 (n = i0), 5/10 (50%) reached maturity at a mean of 4.2 ! 1.7 days. At < 29 weeks, no infant had an L/S of Z 1.5 on initial sampling. The infrequency of patients with paired samples (n = 8) limits comment. Only 3/8 reached an L/S of Z 2.0 before delivery. Hyaline membrane disease occurred in 2/33 (6%) patients with an L/S of ~ 2.0 (26.4 and 31 weeks’ gestation). CONCL~BIONSt Our data document a dramatic acceleration of fetal lung maturation among patients treated with betamethasone with PPROM at h 29 weeks. This phenomena may explain the apparent lack of fetal benefit of prolonged latency among these patients.

566 PROSPECTIVE COMPARISON OF FETAL LUNG MATURITY

PROFILES IN VAGINAL POOL AND AMNIOCENTESIS

SPECIMENS IN PATIENTS WITH PRETERM R~ OF

MEMBRANES.T.KopczynskiL J.Allbert, S.LaurenP.Depat~nent of OB/GYN,

Carolinas Medical Center, Charlotte, NC OBJECTIVE:To determine which of the feral lung maturity tests ~n vaginal pool

(VP) amniotic fluid reflect the values obtained by amnmcentes~s (AC) in the same patient with preterm anmiorrhex~s (PROM). STUDY DESIGN:The VP was compared to AC specimens from 33 women w~th

PROM between 25-35 weeks. In determining lung maturity, the standard cut-offs for phosphatidyl glycerol (PG), the foam staMhty index (FSI), the lecithin/sphingomyelin (L/S) ratio and the surfactant/alburnin ratio (S/A) were used. RESULTS:Of the 33 panents, 11 delivered g2 days, 22 dehvered <7 days and

4 neonates developed gDS. Agreement of VP to AC was: S/A L/S PG FSI

n=29 n=21 n=30 n=25

Lung Immaturity 100 % 91% 89 % 82 %

Lung Maturity 77% 50% 75% 88%

Predict S/A S/A L/S L/S PG PG FSI FSI

RDS -<2d <7d -<2 -<7 -<2 -<7 -<2 _<7

Seas % 100 100 100 100 100 100 83 67

Spec% 78 50 50 36 44 29 63 53

CONCLUSION:Of the fetal lung maturity tests we evaluated, the S/A was the

least affected by vaginal secretions Of all the tests on VP specimens, the S/A was the only one that never contradicted an immature AC. Also, S/A appears to be the most sensitive and specific test for VP anmiotic fluid in predicting RDS.

568 THE PREDICTIVE VALUE OF AMNIOTIC FLUID ASSESSMENT IN THE MANAGEMENT OF PRETERM PROM (pPROM). B Mercer, R L¢wis, A. Frangiehx, J. Frickex, B. Sibai. DeFartment of Obstetrics and Gynecol6gy, University of Tennessee, Memphis OBJECTIVE: To determine the value of ammonc fluid assessment (AFA) m the predicnon of latency and adverse outcome, in women with pPROM METHODS: Women undergoing expectant management ofpPROM between 24-36 weeks gestation underwent initial evaluation of the amniotic fluid usmga modified amniotic fluid index (AFI). AFI was measured as the sum of the maximum vertical fluid pockets in each of the 4 quadrants of the uterus. The largest maximum vertical pocket (MVP) was also noted. Multivariate analys~s wa~ performed to determine factors (e.g. gestational age, race, and AFA) affecting latency and/or adverse outcome (c~horioamnionitis, abruptio placentae, and cesarean section). RESULTS: The study group included 318 women (70 5% African-American) The mean ± SD gestational age at pPROM was 29,9 ± 3.4weeks. 198 women (62%) had an AFI gS, whereas 103 (32%) had an MVP <2 Median latency was 4.8 days; 15% delivered within 48 hours, 60% within 1 week, and 80% within 2 weeks The incidence of chorioammonins was 31%, and that of abruptio placentae was 6 3% Multivariate analysis demonstrated that both latency and frequency of delivery within 1 week were independently associated with race, AFA, and gestational age at pPROM (P<0 001) In addition, delivery within 48 hours or within 2 weeks was associated with AFA only Latency outcomes based on AFI <5 are reviewed in the Table. Simdar latency findings were obtained based on MVP <2. However, the relatively infrequent occurrence ofMVP ~ 2 was associated with lower sensitivity (51%, 38%, and 35% for latency of 48 hr., 1 week, and 2 weeks, respectively) as compared to AFI <5. There was no association between AFA and either chorioamniomtis or abru~ptio placentae. However, chorioamnionitis was assoclatcd with both gestationa~ age atpPROM and latency, whereas abruptio placentae was only associated w~tb gestational a_~e at pPROM. Larenq, AFI>~ AFI ~i OR CI SENS SPEC PPV NPV

48 hr (%) 8 20 3.1 1.5-6.5 82 41 20 93 1 wk (%) 50 66 1.9 1.2-3 0 68 47 66 50 2 wks {%) 71 86 2 6 1 5-4.5 67 57 86 ~_ CONCLUSIONS: During expectant management of pPROM, initial amniotm fluid assessment is useful in predicting latency, but not in the prediction of chorioamnionitis or abruptio placentae The latter complications are better predicted by gestational age at pPROM

466 SPO Abstracts JanuatT 1996 Aru J Obstet Gynecol

569 NEONATAL OUTCOME B,~ED ON RACE~ GENDERAND INTRAPARTUM FACTORS PRIOR TO DELIVERY ~ | 000 GRAMS ]~.P,~.~S BO’ITOMS~.J IAM8, R PAUL, C

MACPHERSONx, G NORMANx, ~, E ’PHOM^, J ROBERTS, D MCNELH8X, AND THE NICHD MFMU NE:’rWORK, BETHESDA MD. Objective: To identi~i factors associated with altered neonatal outcome after dellvety at S 1,000 grams (ELBW). Study design: This analysis includes concurrently collected data regarding 349 liveboru infants, delivered at 11 perinatal centers, who had an obstetric ultrasound performed within one week of ddivery. Infants were included if they were considered %iable" hy their obstetrician prior to delivery, and if the

obstetrician would intervene for fetal distress. Study parameters included: race, center, steroid and tocoly~ic use, antibiotic therapy within 12 hours of delivery, estimated fetal weight, estimated ~estational a~e birthwei~ht ~ender and mode of ddn, ery. Outcomes included: s~rvival, intact survival, ~¢q-l, ROP, NEC, and 02 dependence, as well as NICU stay and duration of ventilation. Results: The population was 49.3% black and 46.4% male. Intervent ons included: steroids (44%), tocolysis (53%), antibiotics (44%), and cesarean section (61%) Multivariate analysis revealed steroid use to increase with

increasing gestational age (p=0.012)and to be more commonly given to non- black women (p = 0.003). Significant differences in steroid use occurred between centers. Similarly variability in tocolyzic use antibiotic therapy, and cesarean de ivery was also identified. Over-all survival was 76.2% with (range 60-86% between c~nters). Multivariate analysis, based on demographic characteristics, center, obstetric interventions and gestational age revealed significant associations with neonatal course. Survival was most significantly predicted by race, gender, and EFW (p<0.0001) with blacks having improved outcome despite less intervention w th steroids, tocolysis or operative delivery (OR 2.8, CI 1.5-5.0). Blacks, and females had an advaniage regarding intact survival, which increased with gestational age and varied between centers. NICU stay, ventilatory requirements, 02 dependence, and ROP among survivors were significantly associated with the various parameters while seizure activity, IVH and neurologic status were not. Steroid and tocolytic use, antibiot c treatment in labor, and cesarean section were not associated with increased survival or reduced morbidity among survivoss. Conclusions: Survival and infant morbidity vary with estimated gestarional age, race and infant gender. Infant morbidity varied by center. Desp te wide variation in obstetric practice, differences in ELBW infant outcome based on these interventions prior to delivery could not be identified.

571 INDOMETHACIN FOR THE ACUTE TOCOLYSIS OF

PRETERM LABOR: A META-ANALYSIS OF RCTs

Brenda Tanx Kate Panterx, Mary Hannah, Umv of Toronto, ON, Canada.

OBJECTIVE: To ldentlfiy all RCTs which have compared mdomethacm with

placebo or no treatment for the acute tocolysm (<72 hrs) of preterm labor m

women with intact membranes and to undertake a me~a-analys~s of the results

STUDY DESIGN: The mcluston cnterm used to select studtes were randomized controlled trtals, women m preterm labor with intact membranes who were treated

for _< 72 hours Computer searches of MEDLINE, EMBASE, CDSR, and manual

searches ofkeyjoumals and SPO abstracts were conducted Data collection forms

were constructed with chosen variables relating to dehvery delay, maternal s~de

effects, permatal mortahty and neonatal morbidity Two assessors then abstracted

the data independently Results were pooled and checked for d~screpanc~es

Authors were contacted for additional data For each outcome varmble, a typical

odds ratio (TOR) and 95% confidence interval (CI) were calculated using the Peto

method

RESULTS: 5 articles were ldenhfied, of which 3 met the inclusion criteria (1, 2,

3) The quahty of these arhcles was high There was no d~fference ln maternal slde effects between the two groups Permatal outcomes are reported in the table

below

Perina|al outcomes Trials lndomethacin/Betamimetic TOR 95% CI (n) (n/N) (n/N) low high

Dehv Delay <48hrs* 1 1/18 14/18 006 002 021

Pretenn Bl~h < 37w * I 3/18 14/18 0 09 0 03 0 34 Mortahty/Stdlb~rths 3 3/94 4/93 0 70 0 15 3 30 B~rthwe~ght < 25008* 2 18/78 36/78 0 36 0 18 0 69

Resp D~stress Synd 3 5/94 7/93 0 64 0 19 2 21

Bronchopulm D~spl 1 0/16 2/15 0 12 0 01 1 98

[*denotes s~gmficance, n - number of trials, n!N = total no events/tolal no enteredI

CONCLUSIONS: Indomethacln appears to delay delivery by 48hrs, to reduce the risk of birthweight <25008 and the risk of preterm birth These RCTs are too

small to assess the effects of lndomethacxn on neonatal morbidity This wdl

require larger RCTs References I Zuckerman H, et al J Permat Med

1984,12 25 2 N~ebyl J, et al Am J Obstet Gynecol 1980,136 1014 3 Katz Z, et

al Int J GynaecolObstet 1983,21 337-342

570 Is RESPIRATORY DISTRESS SYNDROME REDUCED WHEN DELIVERY OCCURS WITHIN 48 HOURS OF STEROID THERAPY ? B Mercer D Beaslcy R Briggs, J Frickex, B Sibai. Dept. of Ob/Gyn, University of Tennessee, Memphis. OBJECTIVe: A recent consensus report indicated infrequent use of corticosremids prior to preterm delivery. This may reflect, in part, a belief that at least 48 hou& are req~.tired to obrain’a benefit fiom treatmefit. Our ob~etive was to determine the effect of corticosseroid therapy on the inc dence of respiratory distress syndrome (RDS) when delivery occurs less than 4-8 hours after initial steroid administration. STUDY DE~IGN: This study included women adm rted with sponraneous preterrn labor or premature rupture of the membranes at 24-~4 wee~ gestation who received corticossemids f~r the ioduction of fetal pulmonary maturity and subsequently delivered within 7 days of treatment. Women were excluded f there was vaginal bleeding hypertension, diabetes, suspected fetal I~rowth restriction, or-abnormal fetal te’s~ing requiring immediate ~eliverg. UnTvarhte analysis was.performed based on del~ve~ at d~ferent rime intervals after steroid administration (<12, 12-23, 24-47 88-71, 72-168 hours). Multivariate analysis regarding the fre.quency of RDS was based on race gesrational age, latency from initial steroid therapy to delivery, and subsequent ~levelopmcot of chorloamnionitis. RESULTS: A total of 234 women with singleton gestations met the indus on criteria. The population was 73.4 % Afhcan-Amer can, with a mean + SD gestational age of 29.9 + 2.8 weeks. The mean + SD birth weight was 1,512 + 506 grams; ~1.2 % were delivered by cesarean section. A total’of 124 infants

(53 %) developed RDS. Multivariate’analysis revealed the incidence of RDS to deer.ease with increasing gestarional age (p < 0.001) and increasing latency after corticosteroid administration (p < 0.001). The highest incidence of RDS occurred among infants born before 12 hours after corticosteruid administration (73.6%). Compared with this group, the frequency of RDS was lower after delivery in each subsequent time interval(see Table). RDS d~xzreased progressively up to 48-71 hours.

1~-23 hours 36 ~2.g 0.03 0.40 0.18-0.92 24.-~7 hours 39 ~3.6 0.009 0.34 0.1~-0.76 48-71 hours 41 34.2 <0.001 0.19 0.08-0.42 72-168 hours 46 45 7 0.03 0,43 0.20-0.93

*compared with those delivering at < 12 hours after cort~zosteroid~.

.CQ.NCLUSI.ON: The frequency of RDS is reduced with increasing time after initial corticosteroid administration. The effect is evident within the first 24 hours of initial treatment. Corticosseroid administration should be considered when preterm delivery is imminent and fetal pulmona immaturity anticiplted, regardless of whether a 48 hour latency is ~xpected. ry

572 INDOMETHACIN VS I%MIMETICS FOR THE TOCOL¥SIS

OF PRETERM LABOR: A META-ANALYSIS OF RCTs Kate Panterx, Brenda Tanx, Mary Hannah, Umv of Toronto, ON, Canada.

OBJECTIVE: To identify all RCTs which have compared lndomethacin wtth

betam~mehcs for the acute tocolysls (<72 hrs) of preterm labor in women wtth

intact membranes and to undertake a meta-analysrs of the results

STUDY DESIGN: The inclusion criteria used to select studies were randomized controlled trials, women m preterm labor with intact membranes who were treated

for _<72 hours Computer searches of MEDLINE, EMBASE, CDSIL and, manual

searches ofkeyiournals and SPO abstracts were conducted Data collechon forms

were constructed with chosen variables relating to dehvery delay, maternal stde

effects, perlnatal mortahty and neonatal morbidity Two assessors abstracted the data independently Results were pooled and checked for dlscrepanctes Authors

were contacted for additional data For each outcome variable, a typical odds ratio

(TOR) and 95% confidence interval (CI) were calculated (Peto method)

RESULTS: 8 articles were rewewed, of which 4 met the inclusion criteria (1, 2,

3, 4) The quah~" of these articles was h~gh Maternal s~de effects such as

tachycard~a, angina, hypotenslon and the necessity to stop therapy due to s~de effects, were s~gmficantly reduced with ~ndomethac~n Permatal outcomes which

were s~smficantly dtfferent are reported in the table below Perinatal outcomes Trials lndometharin/Betamimefic TOR 95°/o CI

(n) (n/N) (n/N) low high Dehv Delay <48hrs 2 4/82 16/84 027 0 I1 067 P~elenn Bwth < 37w 1 9/30 17/30 0 34 0 12 0 95 PDAReq Surg Dg 2 4/72 0/75 808 1 12 5845 Necrohz Enterocoht~s 3 4/80 0188 90l 1 24 65 32

Bronchopulm DyspI 1 10/42 2/45 5 00 1 49 16 82

[n = number of trials, n/N = total no events/total no entered]

CONCLUSIONS: lndomethacm may be more hkely to delay dehvery by 48hrs,

have fewer maternal s~de effects and reduce the risk of preterm b~rth than

betamtmetlcs lndomethac~n may also increase the risk of neonatal morbtdlty

Indomethacm needs to be studted further before it ~s used as a routine tocolyhc

References 1. Eronen et al J Peds 1994,124782-8 2. Kurkl et al Obstet

Gynecol 1991,78(6) 1093-7 3. Morales et al Obstet Gynecol 1989,74(4) 567-

72 4. Newls et al Am J Obstet Gyneco11994,170 378


573 A RETROSPECTIVE ANALYSIS OF T~E EFFECT OF ANTENATAL STEROID ADMINISTRATION ON THE INCIDENCE OF RESPIRATORY DISTRESS SYNDROME IN pRETERM TWIN PREGNANCIES. M.A. Turrentinex, P. Dupra-Wilson=, I. Wilkins. Dept.Ob/Gyn, Oni~. of Texas, ~ouston, TX. OBJeCTIVE: To detez1~lne if antenatal steroid adminlstration reduces the incidence and/or severlty of respiratory distress syndrome(RDS) In twin pregnancies. STUDY DESIGN: Maternal and neonatal medical records were reviewed from twins delivered between 24 - 34 weeks gestatlon over the period of January i, 1990 - December 31, 1994. Optimal sterold therapy was defined as the use of two doses of betamethasone, wxth blrth occurrlng between 24 hours and 7 days after the first dose. The diagnosis of RDS was based on standard

~ESULTS: 98 pairs of twlns were identlfled of whlch 14 received a partial course of sterolds and were excluded from analysis. Of the remalning palrs, 21 received optimal steroid treatment (OST) and 63 received no treatment (NT). The mean gestatlonal age at delivery was 29.9 ± 2.5 weeks. The sample size achieved had an 80% chance of demonstrating a 50% reduction of RDS with P < 0.05. No statistlcal differences were ~een between the OST compared to the NT group for the covarlates of maternal race, infant gender, mean birth weight, mean gestatlonal age, preterm rupture of membranes, use of tocolytics, presence of labor, or use of surfactant therapy. No decrease was seen in the lncldence of RDS [OST 70.7% versus NT 68.0% unad3usted odds ratio (OR) 1.14, 95% confldence interval (C~) 0.49-2.65]. Multivariate loglstlc regresslon analysis for the above covariates showed no statlstical dlfference in the incldence of P~DS in the OST compared to NT group (ad3usted OR 0.63, 95% CI 0.2-1.95). There was no statistlcal difference in the incidence of mechanical ventilation (OST 58.6% versus NT 55.4%), median days of intubatlon (5.0 versus 5.0 days), or peak insplratory

CONCLUSION: The current regimen of antenatal sterolds utillzed in twin pregnancies did not affect the incldence or severlty of RDS in thls study.

575 THE EFFECT OF ANTENATAL STEROID ADMINISTRATION ON FETAL

WHITE BLOOD CELL COUNT. MR Leonard,S., SM Berry, HM Wolfe, JB Landwehr~, DB Cotton, and R Romeroo Department of O8/GYN, Wayne State Umversity/Hutzel Hospital, Detroit, MI, and Perinatology Research

Branch, NICHD, Bethesda, MD.

OBJECTIVE: The effect of antenatal steroids on the fetal and neonatal immune system Is controversial. We evaluated the impact of steroid use on fetal white blood cell (WBC) counts in preterm labor. STUDY DESIGN: As part of an approved protocol, 64 preterm labor patients with singleton gestations, intact membranes, and no evidence of infection between 20 and 35 weeks underwent cordocenteais. Fetal WBC counts were measured and confirmed by peripheral smear. Time intervals from maternal administration of steroids and B-m~metic agents to procedure were determined. AWBC counts were calculated by subtracting the established 50th centlle for gestational age from the observed value. Stepwlse multiple regression analysis was used to evaluate the effect of steroids, B-mlmetlcs, and interval to procedure

on AWBC values. RESULTS: 34 of 64 (53%) patients received steroids and 20 of 64 {31%) received B-mimetm agents prior to cordocentesls. Fetal WBC count was ~ncressed (AWBC = 1.09x10~ ± 1.86| in patients not

receiving steroids. Steroid use within 24 hours of cordocentesis was associated with a significant decrease in fetal WBC counts compared to fetuses ~n preterm labor (AWBC = -0.167x10° :t 1.2, p < 0.008|. This effect was not apparent when fetal blood sampling was performed more than 24 hours after steroid administration. Regression analysis for AWBC found a s~gnlflcant effect from steroid use with no contribution from B-mimetic agents (R=0.364, F=7.81, p < 0.008). CONCLUSIONS: Although transient, the observation that treatment

with steroids suppresses or potent~a~|y prevents a physiologic response to preterm labor has imphcat~ons for fetal and neonatal immune competence. Confirmation of this finding and ItS climcal significance may have imphcatlons for neonatal care.

574 INTRODUCTION OF CORTICOSTEROIDS FOR FETAL

MATURATION IN A LARGE OBSTETRICS SERVICE: A 2 YEAR

BEFORE*AFTER COMPARISON OF OUTCOMES. S.M. Cox, K.L

Leveno, F.G. Cunningham, W. Eagle’, C.R. Rosenfeld~, M.A. Kelly=,

L.M. Sherman=. D~t. Ob/Gyn, Univ. of Taxas, Southwestern Med. Or.,

OBJECTIVE: To measure pefmatal r~ults after introduction of

dexamathamn¢ therapy for fetal maturation.

STUDY DESIGN: Dexametheson~, 5 mg IM q 12 hours x 4 doses q 7

days was begun between 24 and 34 weeks in our obstetrics service on 1

May, 1994. Prior to this date �ortiensteroidswer© not used. Outcomes of

pregnancies ending between 24 and 34 weeks during the 12 months before

dexamcthasone (Dex) were compared to results during the 12 months aRer

introduction.

RESULTS:

l~liveries

Births 24-34 wks.

Outcome Clinical HMD Surfactant Rx Vent, 1st 72 hrs. Max. FiO~ ~> .65 CPAP (no yen0

BPD IVH, Ill or IV

l~roro Dex (%) After Dex (%)

14581 13633

662 (100) 567

D~x Rx No D~x

N = 272(100) N = 295(100)

155 (23) 49 (18) 68 (23)

10~ (~5) 4O (15) 59 (20)

176 (27) 89 (36) 96 (33)

121 (18) 60 (22) 49 (17)

89 (13) 33 (12) 29 (10)

27(4) 20(7) 20(7)

21 (3.7) 7 (2.6) 4 (1.2)

The incidenc� of grade IlI/IV IVH in neonates with birth weights 500-999g was 19% (13/68) before Dex and 27% (17/62) aRer Dex. CONCLUSIONS: Introductionofdexamathasonefor fetal maturationdoes not appear to improve neonata~ re~pirato~ ou~.,om~ or prevent in~xe~ranial hemorrhage. We conelade that although m&ny benefits have beoa recently attributod to ¢oRienstorolds for enhancing fetal maturation, they were not apparent in this unprecedented population based study.

576 USE OF ANTENATAL CORTICOSTEROIDS (ANCS) IN USA. Be, Planer’, RA. Bailard, Pt. Ballard, C. Cobum. C. Boardman, A. Cnaan, M. Morgan, J. Parer. Dept. Neonatology, Ob/Gyn and Biostatistics, Univ. of Pennsylvania and Child. Heap. of Philadelphia, Philadelphia, PA and Dept. Ob/Gyn, UCSF, San Francisco, CA. OBJECTIVE: To determine the prevalence and nature of ANCS use in clinical practice in the US, following the NIH Consensus Conference on ANCS. STUDY DESIGN: Survey (questionnaire) of self-reported practice of all US members of the SPO (N=1420). R E S g LTS: After 2 mailings there was a 6(Y/O response rate (N=846), of which 64% were board certified in MFM. All use ANCS. Of those responding 88% use Betamethasone, mean dose 12+.8 mg, given q12 (28.7%) or q24 (59.8%), to fetuses of gastationai age 24 to 34 weeks. The rest use Dexamethasone. Twenty eight % would treat if deliven/ is expected to occur within 6h, 56% ff within 6-12h, and 97% if within 24.48h. ANCS are used in the presence of PPROM (84%), PIH (95%), gast. diabetes treated with insulin (85%), and in mu~pie gestation without changing the dose. Ninety five % give mere than one course of ANCS, usually weekly (91%), with 69% retreating when active labor persists or recurs, and at least 28% retreating even after PTL has stopped and the patient has been discharged homo. 58% would give a maximum of 6+ courses, 23% 4-5 courses, 12% 3 courses, and 6% only 2 courses. CONCLUSIONS: As of March 1995, the ma~orlfy of US members of SPO are c~t with most recommendations of the NIH Consensus Conference. Since there are no data on the effects of multiple courses of ANCS, this practice clearly needs flJrther study.


577 ANTENATAL STEROID INFLUENCE ON THE PREMATURE INFANTS’ ADRENAL RESPONSE TO ACTH. C Colex, C Santos- Ocampo, M Malloy, A Athanass=ou, L Wolkoff, Wm Rand, A Dubey, H Nielsen, M D’Alton Dept of Peds and Ob/Gyn, New Eng Med Ctr, Tufts Univ Sch of Med, Boston MA OBJECTIVE: To assess the adrenal response of premature infants following multiple doses of the antenatal stern=d, betamethasone (ANS) STUDY DESIGN: 40 infants born to 32 mothers at < 35 weeks gestat=on were enrolled An ACTH stlmulatmn test was administered at 36 to 72 hrs of age Pretest serum cort=sol (C) levels (To) were obtained prior to ACTH stimulation (3 5 #g/kg IV) Cort~sol responses were measured 60 m~nutes (Ten) later Variables measured were gestatlonal age (GA), ANS exposure, cummulat~ve dosage (CD) and the Ume interval after the last dose (TLD) RESULTS: The average GA was 29 8 + 3 1 SD wks (BW 1444 + 567 gms) Mothers received CD of ANS from 12 to 144 mg given over a period of 1 to 38 days (ANS exposure) The increase in serum C levels (Tro - To) ~n response to ACTH (AC) was 14 8 + 8 3 #g/dl (Pa~red t test, p < 0 001) Using one way ANOVA comparing three ANS treatment groups, <24 mg, 24-48 rag, and >48 mg, no d~fference in cortisol response was observed (p = 0 12) Correlational studies of cortlsol response m fact suggest a trend toward an increase in AC w~th ~ncreasmg CD (r = 0 28, p = 0 085) rather than a decrease m AC AC correlates w~th ANS exposure (r = 042, p = 0 008), but not w=th GA (r = -004, p = 081) nor w~th TLDpnortob=rth(r=001, p=097) Of note, ten Infants hadAC <10 I~g/dl not related to GA, CD or TLD CONCLUSION: These data rod=care that multiple doses of prenatal betamethasone up to 144 mg do not ~nhlb=t the adrenal response to ACTH =n infants <35 weeks gestat=on

579 DILTIAZEM (D) FOR THE MAINTENANCE TOCOLYSIS OF PRETERM LABOR (PTL): A PROSPECTIVE RANDOMIZED TRIAL. Y. EI-Sayed, D.Baba*, R.H. Holbrook Jr., Dept. of GYN/OB Stanford University Medical Center, Stanford CA. OBJECTIVE: The successful use of nifedlpine (N), a calcium entry blocker, for PTL tocolys=s has been previously reported. Diltiazem (D), another calcium entry blocker, has been shown to provide potent tocolysis in animal models. To our knowledge, this is the first prospective randomized trial comparing the efficacy and safety of D to N for tocolysis. STUDY DESIGN: 45 patients had documented PTL before 35 wks. gestation (gest), and had received =ntravenous (IV) tocolysis with magnesium sulfate (MgSO4) pnor to random assignment to oral D or N for maintenance tocolysis. The oral dose was D 30- 60 rag. q 4-6 hrs. or N 20 mg. q 4- 6 hrs. and was stopped at 37 wks. Baseline (BL) and serial maternal blood pressure (BP) and pulse (P), and serial fetal antepartum testing were recorded on oral tocolys=s. Outcome variables for D vs. N were compared. RESULTS: 22 patients were randomized to D and 23 to N. There were no sigmficant differences in the # of spontaneous preterm deliveries (D=17; N=13), gest. age at dehvery (D: 33.5 wks. _+ 3.6 SD ; N: 34.9 wks. + 3.8 SD), days gained in utero (D: 22.3 + 16.4 SD ; N: 25 + 20.5 SD), antepartum test=ng results or b~rth weights (D: 2284g + 761 SD ; N: 2339g _+ 893 SD). No differences in either group were noted ~n BP or P. CONCLUSIONS: In this study D was as effective as N in pregnancy prolongahon and had similar maternal card=ovascular effects without adverse fetal effects. Continued evaluat=on of diltiazem as a tocolytic is indicated.

578 MAGNESIUM TOCOLYSIS AND NEONATAL RADIOGRAPtlIC BONE

ABNORMALITIES. S MacGregor, R Portx, J Leex, S Restockx, R Sdver Depts

of Ob/Gyn & Radtology, Northwestern Umv Med School Evanston, IL

OBffECTIVE: To evaluate the relationship between prolonged intravenous

mag~esmm sulfate (MgSO4) tooolytac therapy and radiographic abnormahltes of the neonatal long bones

STUDY DESIGN: We reviewed maternal and neonatal records from all

pregnancies treated ruth toeolyfie agents for preterm labor (.~ 34 weeks’ gestation) between l/l/g8 and 6/1/93. Those cases m which neonates had chest radiographs

obtained ~4thm 24 hours of delivery were considered study candidates, The

remaining subjects were divided into three groups Group I reccaved MgSO4 > 5 days, Group 2 reecived MgSO~ < 5 days, and Group 3 received toeolyti¢ agents

other than MgSO& Neonatal chest radtographs were randoraly reviewed by a

single pcdia~c ra&olog~st blinded to chmcal dam. The proximal humerus was

evaluated for radiolucent transverse m~aphyseal bands (RTMB), cortteal thmmng (CT), and sclerotic transverse bands (STB). The freqoencies of the redlographic

abnotmahttes m the three groups were oompared

RESULTS: There were 105 pregnancies resulting in dalivery of 142 neonates in

the study. The number of neonates m Group 1 was 53, Group 2 was 57 and

Group 3 was 32 The mean days of MgSO4 in Group I was 14.5 + 108 days

(range 5-,18) The mean gestational age at delivery and birth wetghts were sffmlar

in the 3 groups. Cortieosteroid exposure, sex and race of neonates were siradar in

the 3 groups RTMB and CT were commonly observed and the frequancies were

similar in the 3 groups. STB were less common and were observed s~gnificanfly

more often m Group I compared to Groups 2 & 3 (p < 01, 2X3 Chl2)

Group RTMB (n,%) CT (n,%) STB (~%) 1 (n=53) 35 (66%) 35 (66%) 8 (15%) 2 (n=57) 33 (58%) 33 (58%) 1 (2°/6) 3 (n-32) 15 (47’/0) 15 (47%) 1 (3%)

CONCLUSION: Rathographlc abnormalmes of the long bones are eommonly

observed among preterm neonates rognrdless of t)~e or duratton of too)lyric

exposure. Howe~er. sc|ero,ac transverse metaphyseal bands were more fi’equantly

observed in those neonates exposed in utero to prolonged MgSO4 therapy The

chmcal significance and need for follow-up evaluation needs to be eluctdated

580 THE EFFECT OF DUAL TOCOLYSIS ON THE INCIDENCE OF

SEVERE INTRAVENTRICULAR HEMORRHAGE IN THE

EXTREMELY LOW BIRTH WEIGHT INFANT. TA Iannucei, RE Besmger, S Fisherx, JG Gianopoulos, PG Tomtch. Division of Maternal-Fetal

Medicine, Loyola University Medical Center, Maywood, IL

OBJECTIVE: The null hypothesis ~s that dual tocolysm with magnesium

sulfate/indomethacitt does not increase the rate of grade I]I / IV

STUDY DESIGN: Fifty- six neonates weighing between 500-800 gins from

mothers tocolyzed with either magnesium sulfate alone or in combination with

~ndomethaein were the subjeota of this retrospective study. Most eramal

ultrasounds were performed on day ofhfe 3-4 and repeated 1 week later. Fetal

age, birth weight, umbilical cord arterial pH, Apgars, neonatal sepsis and

survival were not found to be statistically &fferent between the two groups

Surfaetant use was universal. Continuous variables were evaluated with a

students t-test, categorical variables by X~ analysis or Fisher’s exact test and

median Apgars with Mantel-Haenszel X2

RESULTS: The groups were significantly thfferent wtth a h~gher mcldence

of grade ~I / IV IVH in paUents treated with dual tocolytm therapy (p~0 02)

Factor ~ (n=22) Mg only (n=34) p value

Fetal age(wks) 24.4 24.9 NS

Birth weight(gins) 656.3 659 2 NS

Steroids 66 7% 38.2% 0 04*

Apgars(l/5m~n.) 4/6 4/6 NS

Arterial pH 7.33 7.30 NS

Sepsis 63 6% 69 7% NS

Grade I]]]IV 1VH 545% 23.5% 002*

Logistm regression analys~s showed that fetal age was an independent

prognostic factor for severe/VH (p=O 01), while lack or’steroid use was not

a factor (p=0.51). After controlhng for these factors, dual toeolys~s with

indomethaein remained a significant mdependant prngnoslac faetur (p~0 03)

CONCLUSION: Results of our study in&cate that dual tocolysis w~th

magnesmm sulfate and ~ndomethacin may place extremely low birth weight

infants at significantly increased risk of grade 1]~ / IV

Volume 174, Number l, Part 2 SPO Abstracts 469 Am J Obstet Gynecol

581 SUCCESSFUL TOCOLYSIS: DOES CERVICAL CHANGE AFFECT TIME

TO DELIVERY? DF Lewis., MS Edwards, MT Fontenot, RD Burhson~, J ImselsX, LSU-Medlcal Center, Shreveport, LA OBJECTIVE’ It ~s generally preferable to tocolyze patients with idiopathic prsterm labor. Little information is available about ultimate outcomes after successful tosolysls Our oblectlve is to evaluate the

reJatmnshlp between cerwca~ change aftertbs~n~t~atmn of tocoJys~s and the delay =n time to dehvery in patients with preterm labor. STUDY DESIGN: A h~stoncal analysis of all patmnts w~th successful tocolysls after preterm labor between 1/93 and 12/94 was underta ken. Patients were then placed in 1 of 3 categories based on cervical

change after the in~tlatmn of tocolys~s (Regression, Unchanged or Progression). Various demographic charactenstlcs, tocolytlcs used, labor characteristics and pregnancy outcomes were compared and analysed with the Flsher’s exact test, Chl-square analysis, analysis of variance, and Duncar~’s multiple range test

RESULTS: 126 patients had successful tocolysls and met cntena for admission into the study. There was no difference in age, race, panty,

EGA on admission or types of tocolytlc used between the three groups. Vadable Regression Unchanged Progression

Number 44 58 24 EGA @ Admission (wk) 30 5~+2 9 30 7~+2 5 30.5~+2 3 Admit ddatatlon(cm) 2.5+1.0" 2 0+0.8 1 9+ 1.0 Admission Effacement 60 3+_.26 58 6+_.21 53.~--+ 24 Delay toDehvery (d) 24.3+ 17.8 25 3+ 18.5 804+6.8" EGA @ Dehvery (wk) 34+~T7 34.1~___3.1 31.7~-3.1* # Delivery < 35 22(50%) 32(55%) 21(8~’~/o)* Birth Weight (gm) 2,260+815 2259~+642 1695~+555"

¯ = p ~< 0,01 CONCLUSION: Patients who have cervical progression after the inltlanor~ of toeotyslS are more likely to dehver prematurely, have a shorter delay to dehvery t~me and lower bwth weight than patients whose cervix regress or remains unchanged In our population, patients who have successful tocolysls have a h~gh rate of perterm delivery with an overall rate of 59.5% dehvermg before 35th of gestatten

583 PLASMA ALKALINE PHOSPHATASE AS A PREDICTOR OF PRETERM

BIRTH R. Goldenberg. T Tamura, K. Johnston, M. DuBard, R. Copper,

Y. Neggars. Dept of OB/GYN, Univ. of Alabama at B~rm~ngham, AL.

OBJECTIVE: To determtne the relationship between alkahne phcephatase

(AP), an enzyme produced by multiple maternal organs and the placenta,

birthwe=ght (BWT) and preterm birth (PTB).

STUDY DESIGN: 590 Afficen-Amencen women had plasma AP act=rely

measured at 19, 24, and 36 wks gestat=onal age (GA) wth the results

compared to BWT and PTB <32 wks and <37 wks GA.

RESULTS: Plasma AP act=~ty rose linearly dunng pregnancy from a mean

of 38 i~kat/L at 19 wke to 107 i*kat/L at 36 wks. In ~ndiwdual women, AP

activity was consistently high or low as confirmed by correlation coefficmnts

=n adjacent time periods ranging from 0.63 to 0.87. AP at 19 wks was not

s~gn~ficantiy aasoc~ated wth any outcome measure. However, at 24 wks,

AP ~n the h~ghast quartde was aasoc~ated wth 15.8% PTB <37 wks

compared to 4.8% in the ~ower quarries (p=.01), and for PTB <32 wks, 7.2

vs 0.8% {p <.001). When women ~n the h~ghast quartde of increase ~n AP

from 19 to 24 wks were compared to those ~n the lower quarbles, PTB >37

wks was 13.9 vs 5.9% (p=.001), and the PTB <32 wks, was 7.3 vs 0%,

(p=.0001). The mean BWT for the highest vs the lower 3 quartiles was

3080 ve 3253 g (p=.01) and the mean GA, 38.2 vs 39.0 wks (p=.003); the

top quartde of change from 19 to 24 wks vs the other quartiles was

associated with a mean BWT of 3017 vs 3290 g (p=.0001) and a mean GA

of 38.0 vs 39.2 wks (p=.0001). Regreas=on analyses adjus~ng for multiple

confounders confirmed the assoc=at=oc between h~gh AP at 24 wks and

PTB <37 wks [OR 2.4 (1.3 - 4.4)] and PTB < 32 wks [OR 3.5 (1.3 - 9.3)]. S~m=lar results were found for a high ~ncrease ~n AP between 19 and 24

wks. For PTB <37 wks, the OR was 2.7 (1.5 - 5.1) and for PTB < 32 wks,

the OR was 4.8 (1.7 - 13.8).

CONCLUSION: H{gh or {ncreas{ng AP act{vity at 24 wks, but not 19 wke,

was s~gnlficantly associated v~th subsequent PTB and a lower BWT. Since

increases =n AP have been reported ~n association v~th ~nflammatlon or

infection, these data suggest that the increase =n AP may be an acute

phase reaction related to genital tract ~nfect~on.

582 THE EFFECT OF MATERNAL MgSO,~ TREATMENT ON NEONATAL MORBIDITY IN -<1000 G NEONATES. D.F. Klmberlin, J.C. Hauth, R.L. Goldenberg, C. MacPhetsonx , E. Themx, S.F. Bottoms, D. McNellis. Dept. of OB/GYN, Univ. of Alabama at Birmingham, AL, and the NICHD MFMU Network, Bethesda, MD. OBJECTIVE: To determine the effect of maternal MgSO4 treatment on selected neonatal outcomes in -<1000 g infants. STUDY DESIGN: In a one year (1992-1993) prospective, observational study, the NICHD MFMU Network collected outcome data for 799 infants with birthweights <_1000 g. Only singleton, liveborns, without major anomalies, who survived >2 days and who were not the product of an induced abortion were included. Analysis was limited further to those fetuses/infants who were deemed potentially viable by the obstetricians and who would have received a cesarean delivery for fetal indications. Selected neonatal outcomes were evaluated =n this subgroup of potentially viable infants whose mothers did and did not receive MgSO4 tocolysis. Logistic regression analysis was used to control for the effect of confounding variables (maternal corticosteroid therapy, birthweight, race, gender, mode of delivery, chonoamnionitis, neonatal surfactant therapy). RESULTS: Dudng the study period, 308 women delivered infants deemed potentially viable; 124 of these women received MgSO4 for labor tocolysis. In ~nfants delivered to women who did and did not receive MgSO~ tocolytic therapy, the frequency of grade Ill/IV intraventncular hemorrhage (16% vs 20%, p=.34), seizure activity (7% vs 10%, p=.35), grade Ill/IV retinopathy of prematurity (21% vs 18%, p=.59), abnormal neumlogic exam (28% vs 28%, p=.91) and intact survival to 120 days or to dtscharge (48% vs 44%, p=.54) was similar. Regress=on analysis controlling for multiple potential confounders confirmed no association between maternal MgSO~ treatment and these selected neonatal outcomes. CONCLUSION: Maternal MgSO4 tocoiytic therapy was not associated with an improvement in selected neonatal outcomes in <_1000 g infants.

584 THE PRETERM PREDICTION STUDY: RISK FACTORS FOR PRETERM

BIRTH IN TWINS. R. Goldenberg. J. lares, B. Mercer, P. Meis, A. Moawad, R. Copperx, A. Dasx, E Themx, F Johnsonx, J. Roberts,

D. McNelhs, and the NICHD MFMU Network, Bethesda, MD. OBJECTIVE: To determine the assoc=a~on between various nsk factors

(RF) and tests and the nsk of spontaneous preterm b~rth (SPB) ~n t~ns.

STUDY DESIGN: 147 women ~th lyons were prospectively screened at

24 and 28 wks’ gestat=onal age (GA) for more than 50 potential RF for SPB as well as cerwcal length {CL) by u~trasound and bactar~al vagmes~s {BV).

Fetal tibronectln (FFN) was evaluated every 2 wks from 24-30 wks.

Outcomes ~nduded SPB <32 wks, <35 wks and <37 wks.

RESULTS: At 24 wks, compared to singletons, only a CL <-25ram and

perceived contract=one were s=gmficantly more common =n twins. Of the

RF evaluated by un~variate analysis at 24 wks for an association with SPB

<32 wks, only a CL -<25mm [OR 6.9 (2.0-24.2)] was significantly predictive.

A positive FFN had a non-s~gnificant OR of 5.6 (0.7-33.0) for SPB <32 wks. For SPB <35 wks, only a CL ~25mm [OR 3.2 (1.9-7.9)] and a unnary tract

~nfact~on [OR 3.2 {1.3-8.0)] were s~gn~ficant predictors. For SPB <37 wks,

the only significant RF were a CL <25mm [OR 2.8 (1.1-7.7)] and a previous

SPB [OR 4.0 (1.1-18.2)]. At 28 wks, a CL -<25mm was not a strong

predictor of SPB <32 wks, OR 3.8 (0.8-20 3). Both at 28 wks [OR 9.4 (1.0-

67.7), p=.05] and at 30 wks [OR 46.1 (4.2-1381), p=.002], a positive FFN

was significantly associated ~th SPB <32 wks. BV at 24 or 28 wks was

not associated ~th SPB in t~ns, Mult~vanate analyses confirmed the

assocmtions between CL _<25mm at the 24 wk visit and SPB <32 wks [OR

7.7 (1.7.34 4)]; v~th SPB <35 wks [OR 3.4 (1.2-92)]; and with SPB <37 wks [2.8 (1.0-7.8)], and also confirmed that at 24 wks the other RF were

less consistently and often not s~gmficantly associated ~nth SPB. Using survival analyses and evaluating all RF ~n both brae penods, only a positive

FFN at 28 wks was associated wah a s~gn~ficantty shorter time to delivery. CONCLUSION: Most known RFfor SPB were not s~gnificantly associated

~th SPB in t~nns. At 24 wks, CL -<25mm was the best predictor of SPB

<32 wke, <35 wks and <37 wks. Ofthe RF evaluated at 28 wks, FFN was the only s~gnificant predictor of SPB <32 wks

470 SPO Abstracts Janua~7 1996 Am J Obstet Gynecol

585 INDICATED VERSUS SPONTANEOUS PRETERM DELIVERY: AN EVALUATION OF NEONATAL MORBIDITY IN <_1000 G INFANTS. D Kimber!!=n, J Hauth, D Guinn, J Owen, E Thom,× SF Bottoms, D McNellis. Un~v of Alabama at Birmingham, AL and the NICHD MFMU Network, Bethesda, MD OBJECTIVE; To determine whether infants born to women who undergo an indicated preterm dehvery have different neonatal outcomes than ~nfants born to women w~th spontaneous preterm labor (PTL) or preterm premature rupture of the membranes (PPROM). STUDY DESIGN: In a one year (1992-1993) prospective, observabonal study, the NICHD MFMU Network collected outcome data for 799 infants with bir~hweights <_1000 g Only singleton, hveborns, without major anomalies, who survived >2 days and who were not the product of an, induced abortion were included. Analysis was limited further to those fetuses/infants who were deemed potentially viable by the obstetrician and who would have received a cesarean delivery for fetal indications (N=411). Selected neonatal outcome measures, Including neurologlc status, seizure activity, =ntraventricular hemorrhage (IVH) and retlnopathy of prematunty (ROP) were evaluated in potentially wable infants delivered to women ~n each of these groups Logistic regression was used to control for the effect of confound ng vanab es (race, b[rthweight, gender, mode of delivery, chonoamnionitis and maternal MgSO4 or corticostero~d therapy). RESULTS: 156 women had an ~ndicated preterm dehvery, 160 had PTL, and 95 had PPROM. Selected outcomes are listed.

Indicated Spontaneous PTL PPROM P

N (%) N (%) N (%) Intact survival 83/156 (53.2) 72/160 (450) 38/95 (40.0) .103 Neur. abnormahty 11/127 (8.7) 20/131 (15.3) 12/78 (15.4) .209 Seizures 8/!54 (5.2) 13/158 (8.2) 14/95 (14.7) .0.33 Grade Ill!IV IVH 14/155 (9.0) 26/156 (16.7) 20/95 (21.1) 024 Grade Ill/IV ROP 16/145 (11.0) 27/146 (18.5) 24/86 (27.9) .005 Regression analyses controlling for multiple confounders confirmed the =mproved neonatal outcome In the ~nd~cated group. CONCLUSION: In this series, infants born after an indicated preterm delivery had less IVH, ROP and seizure acbvity than infants born after PTL or PPROM.

587 INTERMITTENT UTERINE ACTIVITY MONITORING DURING THE 24-26TH WEEK OF GESTATION AS A PREDICTOR OF PREGNANCY OUTCOME. J Jones, R Wapner, N Bergauerx, D Jacquesx, G Stanzianox. J~n Medical College, Philadelphia, PA and Healthdyne Maternity Management, Mariet~., GA. OBJECTIVE: To determine ff uterine activity (UA) during the 24-26th week of gestation differs in patients that deliver term (T) versus those that deliver preterm (-<. 37 w,e~ks) (PT). STUDY DESIGN: 723 patients receiving home uterine activity monitoring (HUAM) prophylactically at 24-26 weeks’ gestational age (G-A) were identified from a large obstetrical patient database. All patients met the following inclusion criteria: HUAM started at 24 weeks GA, UA data transmitted during each week studied, no preterm labor at study start, and no history of preterm delivery. Two groups were identified: singleton (N=475) and twin (N=248) gestation, and were analyzed for T vs. PT delivery at each GA. Data were analyzed using independent Student’s t test. RESULTS: 13,075 hour-long sessions of uterine monitoring were analyzed for the mean number of contractions recorded per patient per monitoring session. There were significant differences in mean UA between T and PT patients within each group for each week analyzed (see table below). SINGLETONS Term Preterm P-Value 24 Weeks (1382) 1.1_+1.7 (424) 1.7±2.2 <0.001 25 Weeks (2449) 1.0_+1.3 (748) 1.4___1.7 <0.001 26 Weeks (2432) 1.1+1.3 (732) 1.6±i.7 <0.001 TWINS Term Preterrn P-Value 24Weeks 0.055) 1.5_+1.8 (522)2.1_+2.4 <0.001 25 Weeks (1114) 1.4_+1.4 (552) 1.9_+2.1 <0.001 26 Weeks (1127) 1.5+1.5 (538) 2.0_+2.0 <0.001 (# of sessions) mean + SD CONCLUSIONS: Patients delivering PT have a higher baseline UA level at 24-26 weeks’ GA than those delivering T. This data may be important when evaluating treatment modalities for the high-risk patient. UA at early gestational ages may be predictive of ultimate pregnancy outcome.

586 OBSTETRICAL DETERMINANTS OF NEONATAL NEUROLOGICAL MORBIDrTY AR Goeofertx, RL Goldenberg, JC Hauth, J Owen,

C MacPhersonx, E "Fnomx, S Bottoms, D McNelhs. Un=versdy of Alabama

at B~rm=ngham, AL and the NICHD, MFMU Network, Betheeda, MD.

OBJECTIVE. To =dent=fy obstetnc factors assoc=ated vath 3 adverse

neurological outcomes =n -<1000 g =nfants.

STUDY DESIGN: In a one year (1992-1993) observational study, the

NICHD MFMU Network collected obstetnc nsk factors for 799 infants

-<1000 g (486 survwed >2 days) infants’ records were evaluated at

d=scharge for se=zures, Grade Ill/IV IVH and abnormal neurolog=cal

exam=nat~on (ANE).

RESULTS: Of the 486 =nfants, 79 (16%) had IVH, 46 (9%) seizures and

57 (12%) ANE Both lower b~rthwe~ght (BW) and early gestational age

(GA) correlated (P < 0.01) vath an increasing ~nc~dence of all 3 outcomes. Seizures, IVH and ANE respectively ranged fTOm 14, 29 and 24% at<24

wks; 12, 21 and 16% at 24-25 wks; 8,12 and 13% at 26-27 wk8; 3, 7 and

7% at 28-29 wks, to 3, 3 and 7% at >-30 wks Obstetric assoQat~ons vath =ncreased IVH included maternal age <20 yrs (33 vs 22%) p = 0.03,

vaginal dehvery (21%) vs C/S (12%) p=0.01, and PPROM (21 vs 12°4) p=0008. Severe preeclamps~a (PE) (6 vs 19%; p=0.001), IUGR (8 vs

18%; p=0.032), and MD’s valhngness to perform a C/S for d~stress (15 vs

25% p = .03) were protect=ve Chonoamn~on=t~s (CA) was associated vath

an =ncrease ~n ANE (22 ve 11%; p=0 006). Smoking, drug use, lack of

prenatal care, the use of/~-m=met=cs, MgSO4 and co~costeroids (CS)

were not predictive of outcome. Logistic regress=on analyses confirmed

that BW and GA were h=gNy pred~ctwe of each outcome. Black race was

proteet=ve against IVH, OR = 0.6 (0.3 - 1.0) and ANE, OR = 0.6 (0.4 - 0.9)

but assoc=ated vath increased seizures, OR = 2.0 (1 0 - 4.0). Increasing

maternal age and PE, OR = 0.3 (0.1 - 0.9) were protective against IVH,

but not the other outcomes.

CONCLUSION: Low BW and early GA were the only significant predictors

of all 3 neurolog~c outcomes Black race, =ncreas~ng maternal age and PE were protective against IVH; other factors were attnbutable to BW and GA.

Measures of maternal behawor and CA as well as matemal treatment w~tfl

CS, ~-m~met~cs, or MgSO,~ d~d not predict outcome ~n th~s data set.

588 PRETERM LABOR IS NOT JUST EARLY TERM LABOR.

S. R In~s, H Yun, S C Presser, I Udom-Rme, S S

Witkin, Jersey City Medical Center, Jersey C~ty, N J, New

York Hosp~taI-Comell Medical Center, New York, NY

OBJECTIVE: Determine whether the ebology of preterm

labor and dehvery =s different from term labor and dehvery.

METHOD: TNrty-one women presenting w~th spontaneous

preterm labor and delivery (<36 weeks gestation) and 35

w~th spontaneous term labor and delivery (>37 weeks

gestation) were studied Swabs from the endocervIX were

assayed for fetal fibronecbn (FFN) and mtedeuk=n-6 by

enzyme-hnked =mmunosorbent assays, tumor necrosis

factor-alpha (TNF) by b~oassay. In all cases delivery

occurred within 48 hours of admission

RESULTS: There was no difference in the Incidence of

FFN m the endocerv~x ~n women having pratarm or term

dehvenes FFN was present =n 67% (16/24) of women

having prstarm dehvery and 87% (27/31, ns) of women

having a term dehvery. Cytokmes (TNF or IL6) were

present in 55% (17/31) of women hav=ng preterm delivery

but only 29% (10/35, p <0 03) of women having a term

delivery

CONCLUSIONS: The fact that fetal fibmnect~n was found

in both preterm and term tabor indicates that it Is a general

marker for labor That cytokmas were more often found in

preterm labor suggests that they may be markers of a

mechanism specific to preterm labor

Volume 174, Number l, Part 2 SPO Abstracts 471 Am J Obste[ Gynecol

589 FETAL FIBRONECTIN FOR PREDICTION OF PRETERM DEUVERY:

A META-ANALYSIS. G Faronx÷, M. Boulvainx°, O. Irion*’. Dept. Ob/Gyn,

÷Laval Univ., Quebec, CANADA and "Univ. of Geneva, SWITZERLAND.

OBJECTIVE: To assess the performance of cervicoovaginal fetal

tibronectin tes~ng to predict preterm delivery.

STUD~( DESIGN; A computerized search was carried out in an attempt

to identify all relevant prospective cohort studies. In order to obtain

unltorm informatmn and unpublished data, personal communication with

the authors was sought. Studies with uninterpretable data were

excluded only if no response was received to a letter sent to the author

requesting clarification. Studies were classified having been conducted

in a high risk group d the prevalence of preterm delivery {<37 weeks)

was Ngher than 15%. Results of ~nd=vidual studies were combined as

weighted likelihood ratios (LHR). Summary sensitivity, specificity and

95% confidence interval (CI) were computed, taking into account inter-

as well as intra-study variance.

RESULTS; Fourteen prospective studies were included, 10 conducted

in high risk and 4 in low risk cohorts. Five studies were excluded. A

total of 630 women were included in "high risk" studies and 3656 in "low

risk" studies. In studies conducted in high risk women, fetal fibronectin

had a summary sensthvity of 72% (95%C1: 59-86%), a summary

specificity of 77% (95%C1’66-89%), and a weighted LHR of 3.0

(95%CI 2.5-3.5). In studies conducted in unselected women, repeated

testing resulted in a higher sens~hvlty (60% vs 11%), but a lower

specificity (75% vs 98%), as compared w~th a single test policy.

Repeated testing had a tower LHR than single testing (2.3;95%C1:1.8-3.0

vs 5.2,’95%C1:3.5-7.7).

CONCLUSIONS; Fetal fibronectin detection in cervico-vaginal secretions

is associated with preterm delivery, both in symptomatic and

asymptomatic women. Women identified by this test could potentially benefit from preventive or therapeutic interventions. A pos=hve test result

could constitute an inclusion criterion for thals of interventions designed

to decrease the preterm birth rate.

59[ AMNIOTIC FLUID TUMOR NECROSIS FACTOR-ALPHA AND RISK OF RESPIRATORY DISTRESS SYNDROME AMONG PRETERM INFANTS. LIJj~"x, MA Krohnx, DL

Pattonx, SL Hillierx, EM Cassenx, P Tarczy-Hornochx, DA EschenbachXo University of Washington, Seattle. OBJECTIVE: Tumor necrosis factor-alpha ONF-a) has been found in the amniotic fluid of some women in preterm labor, and also in tracheal secretions of infants with chronic lung disease. We examined the effect of exposure to amniotic fluid TNF-a on pulmonary outcome in preterm infants. STUDY DESIGN: We followed a prospective cohort of 101 infants born at< 34 weeks to 215 women admitted in premature labor. Amniotic fluid TNF-a and cultures were obtained by amnioceutesis at maternal admission. TNF-a levels of > 20 pg/ml were considered positive. Chorioamnion cultures and histopathology were obtained at delivery. Infants were followed for the occurrence of respiratory distress syndrome (RDS) and length of ventilator and oxygen support. Logistic regression and Cox proportional hazards regression were used to adjust for birthweight and other confounding factors. RESULTS: 53 (52%)of 101 infants developedRDS. Amniotic fluid TNF-a was significantly associated with RDS (adjusted odds ratio 6.1, 95% confidence interval 1.6 - 23.9). Infants exposed to TNF-a were more likely to remain on supplemental oxygen (adjusted hazard ratio 1.9, 95% confidence interval 1.1 - 3.4) and assisted ventilation (adjusted hazard ratio 2.7, 95% confidence interva/1.4 - 5.3), compared to non-exposed infants. CONCLUSIONS: Exposure to amniotic fluid TNF-a appears to be a risk factor for lIDS and prolonged ventilation and oxygen support.

590 Correlation Between Low Systemic Creatme Kinase Levels and Delayed Labor

In Women With Premature Contractions Lavte O, Kremdler S, Hammerman C, Eylath U, Mllwltsky U, Shen O, Sodovsky E, Dlamam Y, Aboulafia Y, Dept of

OB/GYN, Shaare Zedek Methcal Center, Jerusalem, Israel

OBJECTIVES Creatlne kmase (CK) Is an mtramyocyte enzyme consisting of

several lsoenzymas which are found to be systematically elevated during muscu)ar

efforts or muscular damage We have determined the CK rises progressively

during the course of normal labor We therefore sought to evaluate whether

systemic levels of CK can be correlated with the time of dehvery in patients with

premature contractions (PMC)

STUDY DESIGN Serial systemic CK levels were prospectively monitored m

pregnant women with premature contractions documented by a minimum of 2

contractions/10 minutes on monitor Levels were taken at weekIy intervals from

the t~me of initial presentation until dehvery For purposes of analysis, the initial

CK level was selected for each patient and correlated with t~me until dehvery All

tests for CK levels were performed on Kodak Ektachek SD shdes which measure

CK actwi~ by reflectance spectrophot~)metry of chromophone production which is

hmlted by the activity of CK CK-MB was also performed on the samples to rule

out a myocardial source for the elevated CK Data analysls was pertbrmed using

two laded Student’s t-test, sigmficance was accepted at p<0 05

RESULTS A total of 48 patients were stuthed None of the women received any

ntedlcatlons prior to the first CK sample The women were d~vided into two

groups those who presented with low CK levels (<30 IU/L,n=27) and those who

presented with high CK levels (.~.30 IU/L,n=21) Time from presentation until

dehvery was compared between these two groups and was found to be significantly

longer among those with the lower CK levels (50 5+30 2 vs 34 5+17 3 days

respectwely, p-0 036) There were no significant differences in gestattonal age or

number of contractions at presentatmn bet~veen the two groups In addition, ~t was

noted that CK levels on admission were higher In women with multiple pregnancies

as compared with singleton pregnancies (60 3_+47 8 vs 33 3+34 7 [U/L, p=0 05)

CONCLUSIONS Low systemic CK levels in women wlth premature contracOons

have been shown to correlate with delayed delivery as compared to women with

high CK levels Furthermore, in women w~th PMC’s and overthstended uterus, CK

was found to be significantly elevated, possibly due to excessive muscular

stretching We suggest, therefore, that women presenting w~th PMC’s and high CK

levels should be aggressively monitored for possible imminent dehvery

592 THE VALUE OF FETAL FIBRONECTIN AS A PREDICTOR OF

PRETERMDELI~,’~,RYIN LOW RISK WOMEN. JB Greenhagent~,

J VanWagoner~L D Dudley~, C Hunter~x, M Mitchell~L D Casal:L M

VarneP ~Dept Ob/Gyn, Umv Utah, SLC, UT and ZAdeza Biomedical,

Sunnyvale, CA

OBJECTIVE: To evaluate the C[lmcal value of cervical fetal fibronectm

(FFN) detection by a quantltauve enzyme-hnked mmmnosorbent assay

(EL1SA) (PTDcheck, Adeza Bmmedlcal, Sunnyvale, CA, USA) as a

predictor of preterm dehvery in a population of pregnant women considered

to be at low risk for preterm delivery (N = 111)

STUDY DESIGN: A prospective study in which FFN samples were taken

fi’om the posterior formx every two weeks between 24 and 34 weeks

gestation

RESULTS: Twenty-two (20%) patients had at least one positive FFN

result Eleven women (10%) delivered spontaneously at < 37 weeks, of

whom seven had at least one posm’~e FFN test result (PPV = 31 8%,

Sensmvuy - 63 6) An addmonal three women dehvered prematurely

because of other obstetric indications and all bad negative FFN tests The

remaining 15 patients with at least one positive FFN test all dehvered at

term (~_ 37 weeks) Of the seven women with posmve FFNs who

dehvered prematurely, five delivered within two weeks of a positive result

However, there were no obvious chnlcal dlscrllillnators between true-

posuwe anti false posture FFN results Eighty-rune women tested negatn~e

and 85 ot tlmse women dehvered at term (Specificuy - 82 0%) The

negative predlctrve value of FFN as a predictor ot term dehvery in this low-

risk population is 95 5% with OR = 9 9 (3.1, 32 1), RR = 7 1 (2 3, 22 0)

and Fisher Exact p < 0 001

CONCLUSIONS: Although negative biweekly fetal fibronecUn

determinations for prediction of preterm dehvery in this low-risk obstetric

population correlate well with the absence of preterm delivery, they are of

hmlted chnlcal value for the actual predlctlon of preterm birth


593 PRETERM BIRTH IS ASSOCIATED WITH A DECREASED RATE OF FETAL GROWTH A L Medearls*, Moraye B Bearx** & Calvm J. Hobel** *Department of Gynecology/Obstetrics, Loma Lmda Umverslty School of Medicine, Loma L~nda, CA & **The Sctent~fic Data Center & Department of Obstetrics/Gynecology, Cedars Smal Medical Center, Los Angeles, CA OBJECTIVE: The purpose of this mvest~gaUon ~s to compare intrauterine fetal weight curves calculated from ultrasound measurements w~th the neonatal blrthwelght curves from preterm deliveries, to test ff there are s~gmficant differences in the rate of growth of these two populations STUDY DESIGN: A cross sectional database of ultrasound measurements from 216 normal patients between 20 and 37 weeks gestation was used to develop hnear and polynomml regression models of calculated intrauterine weight versus gestat~onal age A separate population of 871 infants dehvered between 20 weeks gestation and 37 weeks gestation was utlhzed to develop comparable regression models for the neonatal evaluation of fetal growth All models were developed from distinct populations In the same restitution. RESULTS: L~near models for each study population were tested for equahty of the slope and the intercept and were statistically different (p value < 0 0001) Quadrahc models best deptcted the climcal sigmficance of this finding for gestahonal age The weight and percent weight d~fference predicted for each gestattonal age from the quadrauc models for calculated intrauterine and dehvery weight are seen m the following table

Gestat~onal Calculated weight (gms) Difference by Age (weeks) Intrauterine Dehvery Weight Percent

20 301 226 75 25 24 728 609 119 16 28 1287 1164 123 9 5 32 1978 1891 87 4.4 36 2801 2789 12 0 4

CONCLUSIONS: These findings are further evidence to support the conclusions that preterm b~rth is associated with decreased fetal growth Traditional neonatal b~rthwe~ght curves consistently underestimate the s~gmficance of abnormal fetal growth prmr to 34 weeks gestation Th~s b~as ~s greatest prior to 28 weeks gestation

595 SURVIVAL OF PREMATURE BABIES WITHOUT MAJOR NEUROLOGIC INJURY. D. Battonx, C. Robertsx, T. Swadsx. Department of Pediatrics William Beaumont Hospital, Royal Oak, MI (Sponsored by R. Lorenz) OBJECTIVE: To determine the gestational age-specific incidence of survival without documented major neurologlc injury at the time of hospital discharge. STUDY DESIGN: All babies from 23 to 32 weeks of gestation born in our hospital from 1990-1994 were =n~t=ally included. A major neurologic injury was defined as a large intraventricular hemorrhage, any intraparenchymat hemorrhage, or cyst=c penventricular leukomalacia. Cranial ultrasounds were performed daily for the first four days, weekly for a month, and bimonthly until discharge. RESULTS: 97% of bab=es (678/698) had detaded information about the presence or absence of intracranial pathology by serial cranial ultrasound or autopsy. The overall incidence of neurolog~c =nlury was 5.8% (40/678). The surwval rate was 88.5% {600/678). The gestat=onal age-spec~hc survival rates and percent of survivors without neurologic ~n)ury were as follows:

Weeks of Gestation

2_33 2~4 25 2_~6 27 28 29 30 31 32 Total#babies 25 53 42 42 52 61 65 87 108 143 % Surv=val 44 60 71 74 85 93 92 94 96 97

% of survivors w=thoutinjury 10088 93 90 93 93 97 98 100 99 CONCLUSION: Although advanc=ng gestat=onal age has a major impact on survival rates of premature babies, the effect on the rate of major neurologic injury in survivors is much less pronounced. The vast malonty of survivors even at the border of viabd~ty are free major neurolog~c =nlury at the time hosp=tal d=scharge.

594 ETHNIC DIFFERENCES IN RATES OF IDIOPATHIC PRETERM BIRTH:

EFFECTS OF STRESS AND AVAILABILITY OF ECONOMIC SUPPORT. R Shaln×, A Holden, E Newton, S Perdue, S Hoppe, R G~bbs Dept Ob/Gyn, UTHSC, San Antonio, TX OBJECTIVE: To determine why poor Aft=can-American women have higher rates of idiopathic preterm labor culminating ~n preterm b~rth than do poor Mexican- and Anglo-Amencen women

STUDY DESIGN: 328 African-, 491 Mexican-, and 396 Anglo-Amer~can women were prospectively ~nterv~ewed at 2 or 3 t~me points dunng pregnancy (1st prenatal visit, 28th week when poss=ble and shortly after dehvery) and their medical charts were abstracted Rates of ~d~opathlc preterm b=rth were 104%, 5 1%, and 43%, respect=vely (OR = 232, P= 0004, Blacks vs others) After Identifying s~gnlficant medical, soc~o- demograph=c and psychosoc=al predictors of ~d~opath~c preterm iabor, the best predictor set was selected using multivariate logistic regrosslon Prehm~nary analysis of two compos=te psychosoc~al vanables from th~s broader set are presented Stress ~s a 3-level measure based on a factor-derived scale, addressing acute and chronic stressors, comb=ned w=th a variable assessing sat~sfact=on w~th hwng arrangements Ava=labll=ty of consistent economic support was operat=onahzed as hawng the same person throughout pregnancy who would help ff needed and d=d

not require pay back, also y~eld~ng a 3-level measure RESULTS: More Black (20 1%) than H=spamc (12 6%) or Anglo (11 9%) women experienced high levels of stress (P= 003) and fewer (36 6%) had access to consistent econom=c support than d~d women =n the other groups (50 1% and 51%, respectively, P= 0002) The =dlopathlc preterm

birth rate was 63% (62%, 76% and 50% for the 3 groups) for the 71 women =n the highest stress, lowest support cell, compared to 0 5% (2 3%, 0% and 0% for the 3 groups) for the 207 women =n the no stress, h=ghest support cell (P< 0001) Intermediate stress/support cells yielded =ntermed=ate outcome rates The odds ratio of ~d=opath=c preterm b=rth ~n Blacks after adjust=ng for the combined effect of these two predictors =s

1 34 (P= 36) CONCLUSIONS: Accounting for stress and ava=lab=l=ty of consistent econom=c support reduced the odds ratio of ~d~opathtc preterm b=rth in African-Americans from 2 32 (P= 0004) to 1 34 (P= 36) in th~s sample

596 CAN PAPANICOLAOU SMEARS PREDICT PRETERM DELIVERY? JM LanouetteX, KS Puder, SM Berry, MP Dombrowsk~, Dept Ob/Gyn, Hutzel Hospital/Wayne State University, Detroit, MI OBJECTIVE: Inflammation on Papanicolaou (Pap) smear has been

associated with a 30-50% incidence of bactenal vaginos~s (BV), a recogn=zed risk factor for preterm dehvery (PTD). We determ=ned whether ~nflammat=on on Pap smear ~s associated w~th PTD

STUDY DESIGN: 5348 cases were studied w~th complete prenatal data including the potenbal confounder oftroatment with ant=b=obcs Cases were categonzed by presence (n=1139) or absence (n=4209)

of inflammatton on Pap smear RESULTS: The proportion of Afncan Amencans was lower (66.9%

v 74 5%, p<0 000) in the inflammabon group, but thero were no s=gn~ficant d=fferences (t-test) for materna~ age, grawd=ty, h=story of

PTD, or gestabonal age at dehvery between groups Stepw=se regress=on analys~s was performed with gestabonal age as the dependent vanable Selected variables are presented’

F p

Maternal age 50 38 0 000 H~story of PTD 22 28 0 000

Afncan Amencan 6.73 0.000

Metron~dazole 5 40 0 000

Erythromyc~n 0 14 NS

Gonorrhea 0 98 NS

Inflammabon 0.08 NS

This sample s=ze was sufficient to detect a 2 day d~fference =n gestabonal age at dehvery between groups w~th an 80% power and p<0.05 CONCLUSION: Unhke bacterial vaginosis, inflammabon on roubne

Pap =s not assoc=ated w=th preterm birth.

Volume 174, Number l, Part 2 SPO Abstracts 473 Am J Obstet Gyneco]

597 TIlE EFFECT OF MODE OF DELIVERY ON EXTREMELY LOW

BIRTH WEIGHT BREECH INFANTS: CESAREAN DELIVERY

DECREASES NEONATAL MORTALITY. C. Lyons{ J. Bienstock~, E.

Preasmsaz. Dept. Gyn/Ob, Johns Hopkins Univ. Sch. of Meal., Baltimore, MD.

OBJECTIVE: To determine effeot of mode of delivery on neonatal outcome

for extremely low birth weight breech infants.

STUDY DESIGN: Perinatal and neonatal data bases as well as log books sad

hospital ¢.harte were reviewed, The study included all singleton broth

pregnancie~ at our institution betweon July 1988 and July 1995 of infsat~

weighing 500-1000 gm sad greater than 24 we~k~ (wi~) gestation. Neonatal

morbidity in the form of length of stay and intraventricalar hemorrhage (IVH),

perinatal mortality, sad mammal morbidity were analyzed. Statistical methods

included Fisher exact test, t-test, Chi-~quare, logistic sad linear regre~ian.

RESULTS: 58 women sad infants were ~tudied. 38 infants (66%) were

delivered by elective cesarean ~ction (C/S) and 20 04%) delivered vaginally.

Aa additional 9 patients were delivered by emergent C/S and not included in the

analysis. There were significant differences betwesa the groups as far as

gestational age (vaginal 25.1 wks; C/S 26.2 wks; p=.0001) sad birth weight

(vaginal 721.8 gin; C/S 827.9 gin; p=.001M). There were no algnifica~t

differencea in racial make-up or maternal age between groups (p > .05). There

were significantly more neonatal deaths in the vaginal group (40%) as compared

to the C/S group where there were no deaths (p < .0001). This cffeot persi~a~d

even when gestational age and birth weight were controlled for. Gastational

age, not birth weight or mode of delivery, was the main predictor of length of

~tay. There was no difference in the incideoceof IVH (vaginal 50%, C/S 34%,

p=.26). There was a significantly lower incid~neaof andomyon~tritis in the

vaginal group ( 15 %) versus the C/S group (63 %), p < .0001.

CONCLUSIONS: The route of delivery significsady influenced neonatal

mortality even when adjusted for gestational age and birth weight with a 40%

mortality rate in the vaginally delivered infants sad no nansatel deaths in the

C/S group. However, infant morbidity (IVH and length of slay) was similar in

the two groups. The incidence of endomyometritis was significantly lower in the

vaginally delivered group.

599 THE NOVEL DIPALMITOYL PHOSPHATIDYLCHOLINE TEST IS A RELIABLE TEST IN THE ASSESSMENT OF FETAL LUNG MATURITY IN CONTAMINATED SAMPLES. B CohenIx, J Ludmlr2, C Towersl, B Slomovlc~x and JG Alvarezz~ Dept of Ob/Gyn, Long Beach Memorial HospitalI, Long Beach, CA, and Beth Israel Hospital2, Harvard Medical School, Boston, MA OBJECTIVE~ D~palmttoyI phosphattdylchohne (DPPC) Is the major surface-ae"a’~e component ~f the ma~re ~e~al lung surfae’tant and ~s found in trace amounts in other biological fluids We have recently reported a novel test (DDPC test) that measures the concentration of DPPC in ammotlC fluid by enzymatic hydrolysis When a DPPC test cutoff value of 12 #g/mL was used to predict fetal lung maturity, the sensitivity and specificity of the test were 100% and 96%, respectively The objective of this study was to evaluate the rehabthty of the DPPC test in the presence of common contaminants of amnlotlc fluid STUDY DESIGN Forty amnlohc fluid samples were obtained by ammocentests at 30 to 40 weeks gestattonal age Each sample was divided in five-25/~L aliquots (total of 200 ahquots) and diluted 1 1, v/v with PBS (control), mecomum (5 mg/mL), blood, semen (lxl0s spermatozoa/mL), or vaginal fluid The concentra~on of DPPC in the resulting mixture was obtained following enzymatic hydrolysis with phosphohpase C and high-pressure hquid chromatography DPPC values in control and contaminated samples were compared using a paired t test RESULTS: DPPC concentrations in the control samples ranged between 0 4 and 50 2 .ug/mL No stgmficant differences were found between the concentrahon of DPPC in control vs contaminated samples in all 40 cases tested (P> 0 15) The DPPC test had a run-to-run precision of 3%, a turn-around time 10 minutes and a requirement of only 25 ,uL of ammotlc fluid per assay CONCLUSION: The DPPC test is a reliable test for the assessment of fetal lung maturity even in the presence of common contaminants of ammotlc fluid

598 IMPROVED DPPC TEST FOR THE ASSESSMENT OF FETAL LUNG MATURITY BY HIGH-PRESSURE LIQUID CHROMATOGRAPHY JG Alvarezx and J Ludmir Dept of Ob/Gyn,

Beth Israel Hospital, Harvard Medical School, Boston, Ma 02215 OBJECTIVE We have recently reported a novel test (DPPC test) that measures the concentration of dipalmltoyl phosphatidylchohne (DPPC) in amntotlc fluid by enzymatic hydrolysis and high-performance thin- layer chromatography (TLC) When a DPPC test cutoff value of 12 ~ug/mL was used to predict respiratory distress syndrome (RDS), the sensitivity and specificity of the test were 100% and 96%, respectively The run-to-run precision of the test was 13% and the turn-around time 30 minutes The objective of this study was to develop an Improved DPPC test by high-pressure hqu~d chromatography (HPLC) that could provide both faster turn-around time and higher precIsion STUDY DESIGN Fifty frozen specimens of ammotlc fluid were used In this study These samples had been previously used m a prospective study where the concentration of DPPC was determined by enzymatic hydrolysis and TLC Of the 50 cases selected, 15 resulted in RDS and 35 had no RDS In the present study, the concentration of DPPC in 25/~L of the thawed samples was obtained by both TLC and HPLC and the resulting values compared by t test The run-to-run precision is expressed as the relative standard deviation of the mean value obtained in ten allquots of the same sample RESULTS: The mean_+ SD DPPC values in the RDS cases were 5 2 _+ 2 3 ,ug/mL by TLC and 3 2 + 1 1 ,ug/mL by HPLC This difference was statistically significant (P = 0 005) Of the 35 cases with no RDS, 5 had values that ranged between 4 and 10 ,ug/mL by TLC and 2 had DPPC values of 7 and 10 ,ug/mL by HPLC No significant differences were found in DPPC values obtained by TLC and HPLC at concentrations above 12 ,ug/mL (P = 0 3) The run-to-run precision by HPLC was 3% and the turn-around time 10 minutes CONCLUSION: Analysis of DPPC by HPLC provides both faster turn-around t~me and higher precision than TLC

600 POTENTIAL MATERNAL AND NEONATAL COST SAVINGS DUE TO ACCURATE ASSESSMENT OF FETAL LUNG MATURITY. J Ludmlr1, DK. Richardson2x, MW Atkmson1 and JG

Alvarezlx Depts of Ob/Gyn1, Neonatology2, Beth Israel Hospital, Harvard Medical School, Boston MA OBJECTIVE: Current tests for fetal lung maturity (FLM) have high sensitivity but invariably lack specificity resulting In unnecessary pregnancy prolongation. This study evaluates the potential impact of improved accuracy of FLM testing on maternal and neonatal costs STUDY DESIGN: Four hundred and twenty five pregnancies delivered at our hospital between January and December 1994 that had FLM testing were evaluated for maternal and neonatal costs Maternal costs were calculated from the time of tesUng to the t~me of delivery Since delivery is often delayed even in the face of mature tests for very premature fetuses, maternal costs were analyzed only from women in whom FLM testing was performed ~ 35 weeks gestation FLM test results were correlated with neonatal respiratory outcome as determined by a neonatologist blinded to the FLM test results RESULTS: Of the 425 cases tested, 301 delivered within 72h of testing Of these, 237 (78%) were correctly predicted as mature, 9 (3%) were correctly predicted as immature, 52 (17%) were incorrectly predicted as Immature and 1% (3/301) were predicted as mature but developed RDS The remaining 124 cases were predacted as immature and delivery was delayed an average of two weeks The average maternal length-of-stay was five days m the delayed group and less than one day in the non-delayed group (P< 0001) Maternal costs averaged $3,500 in the delayed group and $410 In the non-delayed group (P< 0 001) Applying the 85% false immature rate in the non-delayed group (9/61) to the costs of the delayed group, we compute $324,450 in excess maternal costs attributable to delay from false immature test results Neonatal costs for the three false mature RDS cases averaged $24,954 CONCLUSION: In this populataon, use of a more accurate FLM test that correctly identified all mature and immature cases would have resulted in significant maternal and neonatal cost savings


Am J Obstet Gynecol

601 "PREECLAMPTIC" UTEROPLACENTAL PATHOLOGY IS ASSOCIATED WITH INCREASED MATERNAL MEAN BLOOD PRESSURE IN SPONTANEOUS PREMATURITY. C.M. Salafia*~ J.A. Lopez-Zeno*, V.K. Mmlor*, J.C. Pezzullo*, A Ghldml, C. Spong, L M. Ernst*. Pennatal Research Famhty, Depts of Pathology & OB/GYN, Georgetown Umversltv Medical Center, Washsngton DC, UCONN Me&cal Center, Farmln~ton CT, Ponce School of Medicine, Ponce PR. OBJECTIVE. Both preeclampsm (PE) and preterm birth are associated with similar uteroplacental vascular lessons Such lesions are thought to be related to maternal hypertensaon; we hypothesize that in spontaneous prematurlty maternal blood pressure (BP) also reflects uteroplacental vascular pathology STUDY DESIGN From a data base of consecutive non-anomalous s~ngleton hveblrths dehvered at 22-32 weeks (excluding maternal d~abetes melhtus and chromc hypertensmn) over five years (1989-94), we Identified 71 cases of PE and342 cases of spontaneous prematunty [spontaneous premature membrane rupture (PROM N=174), preterm labor, ~ntact membranes (PTL, N=137), and non-hypertensive abruptlon (N=31)] with mean BP (MBP) at admlssmn, and at [east 1 other value at least 6 hrs after admission. These BP and the hsghest antepartum values were studsed In relationshsp to 4 placental hsstologsc les~on categories (utero-placental vascular and related villous lessons, and lesions of chronic inflammation, coagulation, and placental vaso-occluslon) Lesion scores In each category were summed to reflect total pathologic burden Contingency tables, ,~NOVA, and regression analyses corrected for multiple comparisons (p<0 05 significant). RESUETS. MBP at adlmsslo~ of nonhypertens~ve abruptlon (85+15mmHg), PROM (82_+ llmmHg) and PTE (82_+13mmHg) were slgmficantly less than MBP of PE (ll2_+14mmHg, p<0 0001). In non- PE prematurlty, MBP was sagmflcantly related to presence and extent of lesions of uteroplacental vascular pathology (p=0 005) and lntraplacental vaso-occluslol~ (p=0.01), hut not to lesions of acute Inflammation, chromc ~nflamm~tmn or coagulatmn. In PE maternal BPs were not re ated to individual placental lesmns or to les~on categories (p>0 2). CONCLUSIONS. In non-PE prematunty, MBP at admsssmn was associated with hsgher scores for uteroplacental vascular pathology and Intraplacental vaso-occluslon, suggesting that a common vascular pathology can mamfest esther as PE or non-PE prematunty. We hypothesize that pathologic elevation of maternal blood pressure (as In PE) may develop only after a critical "burden" of vascular pathology develops, and m the absence of tissue pathology resulting esther In a uterine contractile response or membrane rapture

603 AMNIOT1C FLUID IL-6 LEVELS DURING THE EARLY SECOND TRIMESTER: A PREDICTOR OF PRETERM DELIVERY. A Ghldml, G S Eghnton, C B Jenkins*, C.Y. Spong*, J C Pezzullo*, M L Shookhoff*, J F Mill* Perinatal Research Facility, Department of OB/GYN, Georgetown University Medical Center, Washington, DC OBJECTIVE" Placental lschemla and acute inflammation are thought to be underlying mechanisms of approxsmately two thirds of preterm dehveries (PTD) lnterleukin (IL)-6 Is a cytokine released In both processes We tested the hypothesis that ammotlc fluid (AF) IL-6 in asymptomatlc gestations at 15-20 weeks Is a predictor of subsequent PTD STUDY DESIGN. We analyzed singleton gestations with second trimester genetsc amniocentesis between 1/93 and 12/94 for which pregnancy outcome information was available Inclusion criteria were 1) gestatlonal age (GA) at amniocentesls 15-20 weeks; 2) no evidence of fetal or neonatal structural or chromosomal anomalies, 3) no maternal condstions known to be associated with preterm delivery AF samples were stored at -20C Levels of AF IL-6 were measured by immunoassay (R&D Systems) and correlated with demographic and pregnancy outcome information Statistical analysis included correlation, ANOVA after log-transformation, contingency tables, logistic regression, and receiver operator characteristic (ROC) curve analyms RESULTS. The PTD rate at < 34 weeks was 8 7% (17/195) There was a significant inverse correlation-between AF 1L-6 levels at 15-20 weeks and GA at delivery (r=0 19, p=0 008). Women delivering at <34 weeks had significantly higher median AF IL-6 levels (540 5 pg/ml vs 329 6 pg/ml, p=0 03), rate of African American race (p=0.0003) and infants with birth weights < 10th cenhle (p=0.01) than women dehvering at > 37 weeks (n=128) Logistic regression analys~s showed that IL-6 was an sndependent predictor of PTD < 34 weeks after controlling for race and birth weight centiles (OR=I 7, 95% CI 1 02-2 67) ROC analysis indicated that the optimal AF IL-6 cut-off level of 1740 pg/ml [corresponding to 5 6 multiples of the median (MOM)] had a sensitivity=26% and specificity=92% In the prediction of PTD at _< 34 weeks CONCLUSION" AF IL-6 at 15-20 weeks can identify patients at risk for PTD at < 34 weeks This finding suggests that preexisting intrauterine lschemla or inflammation is an important risk factor for PTD. An AF IL-6 level > 5 6 MOM is the optimal cut-off for the prediction of PTD at < 34 weeks.

602 DOES THE PREVALENCE OF HISTOLOGIC INTRAUTERINE INFECTION IN PRETERM PROM INCREASE WITH THE DURATION OF THE INTERVAL TO DELIVERY? A. Ghldmi, C M. Salafia*, V K Minlor*, Departments of Ob/Gyn & Pathology, Perlnatal Research Facility, Georgetown University Medical Center, Washmgton, DC, UConn Medical Center, Farmmgton, CT OBJECTIVE: It Is commonly beheved that the frequency of clinical chorloamnlonitls m preterm premature rupture of membranes (PROM) increases with the duratmn of the PROM-delivery interval We tested the hypothesis that the prevalence of hlstologic intrauterine lnfechon (IUI) increases proportionally to the duration of the latency permd STUDY DESIGN: Consecutive cases of PROM < 32 weeks betweeen 6/89 and 3/94 were recorded Prophylactic antibiotics were administered according to ACOG guidehnes 1UI was defined as mild/moderate/severe based on scoring of acute inflammation in chonodecidua, ammon and umblhcal cord at hlstuloglc examination of the placenta and umbilical cord, as previously described The prevalence and severity of[UI was assessed in 5 groups based on the interval bet~veen PROM and delivery Statistical analysis utlhzed Flsher’s exact test, Chl-square, and 2-way ANOVA after log transformation, where applicable RESULTS % or mean + standard dewatton

Variable Interval PROM-to-dehvery (hours) p

< 4 4-12 12-24 24-48 >48 value

(n=7) (n=29) (n=30) (n~36) (n=60)

C/section rate 57% 75% 70% 65% 57% NS

GAatdel.(hr) 192+25 197£16 202£15 200+15 196+_18 NS

IUI total score 8_+7 9+6 7_+7 6_+5 9_+.7 NS

mod+severe IUI 29% 45% 43% 33% 53% NS

CONCLUSIONS The rate of histologxc evidence of chorloamnmnitis m preterm PROM does not Increase with the duration of the interval PROM- delivery

604 AMNIOTIC FLUID LACTATE DEHYDROGENASE AND

INTERLEUKIN-6 IN PREDICTION OF PRETERM DELIVERY

Garry__Dx, Flgueroa R, Sehgal px, Patel Kx, Aguero-Rosenfeld !vl~x, Martinez Ex, Tejam N

New York Medical College, Valhalla, NY

OBJECTIVE: Our purpose was to compare ammotic fluid (AF) lactate

dehydrogenase (LDH) to lnterleukm-6 0L-6) in predicting preterm delivery

(<37 weeks)

STUDY DESIGN: LDH, !L-6, glucose, WBC count and gram stare were

determined sn AF specimens obtained by transabdomlnal ammocentesls in 107

singleton pregnancies with preterm labor and mtact membranes AF cultures

for aerobes, anaerobes and myeoplasma species were performed Reeesver-

operator characteristic curves, Mann-Whitney U test and the Student t-test

were used Diagnostic indices for sndlvldual tests in prediction ofpreterm

dehvery were calculated

RESULTS The prevalence ofposi/~ve AF culture was 16% (17/107) and 76

patients (71%) delivered preterm, 31 patients (29%) delivered at term Cntacal

values of LDH >_223 mg/dl, IL-6 >1 10 ng/ml, glucose _<17 mg/dl, and

WBC count>50 cells/ram3 were selected for optimal performance for the

prediction ofpreterm delivery The ....... ~ ~ ~e

.~ ~v

sensitivity (SENS) of LDH and IL-6 ........ ~ ..... ~. ~,

did not differ (p= 42) and both had ............ ~ ~ ~ ~

prediction of preterm delivery (p< 001 ) All patients with LDH >_ 310 rag/all or

IL-6 ~ 8 4 ng/ml or glucose _< 17mg/dl delivered preterm The median LDH

(265 0 mg/dl) and IL-6 (2 26 ng/ml) In patients dehvenng preterm dsffered

significantly from those dehvenng at term (LDH 169 0 mg/dl, IL-6 0 62 ng/ml,

p< 001)

CONCLUSION: AF LDH and ~ appear equally predictive of preterm

dehvery LDH is a faster, more economical marker which can be evaluated in

any hospital laboratory

V/ilume 174, Number 1, Part 2 SPO Abstracts 475 Am.| Obstet Gynecol

605 AMNIOTIC FLUID LACTATE DEHYDROGENASE AND

PRETERM DELIVERY PREDICTION

_Garry. D_x, Flgueroa R, Aguero-Rosenfdd ME×, Martmez Ex, Telanl N

New York Me&cal College, Valhalla, NY

OBJECTIVE: Investigators continue to evaluate ammotic fired (AF) lbr

the presence o f mfechon and prediction of delivery. Our purpose was to

evaluate AF lactate dehydrogenase (LDH) as a rapid, ~nexpeus~ve test to

predict pretenn dalr~ery m _< 36 hours

STUDY DESIGN: 93 patients &agnosed w~th preterm labor with *ntact

membranes underwent transabdominal amnlocentesls AF was analyzed

for WBC count, glucose, LDH and gram stare Cultures for aerobes,

anaerobes and mycoplasma species were performed Ammocentesls-to-

dehvery interval was calculated. Receiver-operator characteristic curve

analysis, logisUc regression analysis and t-tests ~vere used Sensitivity,

specificity, positive and negatwe predictwe values were calculated for

LDH, glucose, WBC count and gram siam m tbe prediction ofpreterm

delivery m < 36 hours.

RESULTS: The prevalence of positive AF cultures was 11%(11/93) 34

patients (37%) delivered m _< 36 hours alter ammocentesis. The mean LDH

(746 mg/dl) and the me&an LDH (a, t3 ms/all) m patlenU dehvenng ~ 36

hours was stgmficaMy greater than patwnts dehvenng > 36 hours (mean

LDH 199 mg/dl, medmn LDH 179 mg/dl; p< 002) Cmhcal values of LDH

~> 225 mg/dl, glucose <18 mg/dl and WBC count > 50 cells/mm3 were

selected for optimal performance for each test m predlctmn of

ammoceutes~s-to-dehveD’ <_ 36 hours LDfl had the best sens~tlvfiy (73%)

m pre&ctmn of delwery < 36 hou~ after ammocentes~s m contrast to

glucose (47%), WBC count (38%) and gram stare pos~hve (18%) (p< 05).

AF LDH values _> 225 mg/dl confer a RR of 5 62 (95% CI I 78 - 17 83,

p= 0033) for delivery ~< 36 hours

CONCLUSION: LDH ts a readily available, inexpensive, rap~d AF marker

for the pred~chou ofpmterm debveB:.

607 MATERNAL STEROIDS, CHORIOAMNIONITIS AND THE RISK OF PERIVENTRICULAR LEUCOMALACIA IN THE

PRETERM NEONATE.Cantenno Y’,Venna U,Tejam N, Klein SL

Reale IVP, Jeanty M* New York Me&cal College, Valhalla NY

OBJECTIVE The NIH task force has sho~vn that antenatal steroids

reduce the rnc~dence and seventy of ~ntraventncular hemorrhage (IVH)

IVH often co-exlsts ruth penventncular leucomalac~a (PVL)

Chonoammomt~s (CA) increases the ask of PVL This study examines

the relationship of antenatal steroids, CA and the ask of PVL

STUDY DESIGN 716 consecutive ~nborn neonates w~th Bthwt

< 1750 gins and GA _< 34 wks had cramal sonography on days 3 & 7

PVL was &agnosed as hyper or hypoechom lesions in the

penventncular area Neonates were &vided into groups based on

antenatal exposure to steroids and compared for the ask of PVL The

neonates were also divided into groups based on the diagrms~s of CA on

histology The CA groups were stratuSed by’ steroid usage and

compared for the ask ofPVL Flsher’s exact test was used

RESULTS Stermds s~gmficantly decreased the ask of PVL 6111 l vs

76/605 (p < 01) The ~nctdence of CA was not *ncreased with stenods

STERIODS PVL P VALUE

CA+re YES # 39 5

#297 NO # 258 43 0.4

CA-re YES # 72 l

#419 NO # 347 33 0.01

CONCLUSIONS: Steroids slgmficantly decreased the risk of PVL

seen within the first 7 days of life Tl~s netu-oprotectlve effect was lost

m the presence of CA Stermds do not increase the ask of CA

606 CYTOKINE PANEL EVALUATION IN PRETERM LABOR. R Figueroa. D Garryx, E Marhnez×, P Sehgalx, K Patelx, N Tqlmu New York Medical College, Valhalla, NY

OBJECTIVE: To evaluate the pattern of cytokme elevation in the mmuoUc fired (AF) of women in preterm labor and intact membranes (PTL) with regards to success or failure of tocolys~s

STUDY DESIGN. Forty-seven women in PTL underwent transabdommal

anmmcentes~s AF Imerleukm- 1 ct (17b- 1 ct), Imerleukm-I J3 (IL-I ~3),

Lnterleukm-6 (IL-6), h~terleukm-8 (IL-8), and Tumor Necrosis Factor (TNF) were determined by ELISA and reported in pghnl AF was cultured for

aerobic and anaerobic bacteria, Ureaplasma urealyncum, and Mycoplasma hormms Receiver-operator characteristics curves were used to detenmne

critical levels tbr each cytokme in the piedlctlon of positive AF cultures(AFC) Tocolysls was considered successful if pregnaucy was prolonged more than 7 days

RESULTS: iL-la 1L-113 1L-6 lL-8 TNF del n tocol’/sis AFC >l,429 >376 >28,300 >43,670 >239 <24h 18 success ueg 0 0 0 0 2 0 21 failure neg 2 0 5 1 2 12 8 l:aflure pos 7 7 7 7 6 7

11/12(92%) wllh IL-6 elevations delivered in < 24 hours CONCLUSIONS: Failed tocolysls occurs m a subset of women w~th no evidence o1° mfectl on or mI"!mlmlatl Oll ElevatlouofLL-61rrespectiveof culture results is assocmted w~th inevitability of delivery

608 INTERLEUKIN-10 INHIBITS HUMAN FETAL

MEMBRANE PRODUCTION OF INTERLEUK1N-6 BY

TRANSCRIPTIONAL REGULATION

Stephen J. Fortunatq, Ramkumar Menon, Kenneth F. Swan.

The Matetaml Fetal Group and The Middle Tennessee Perinatal

Research Foundation at The Wmnen’s Hospital, Centennial

Medical Center, Nashville, TN

OBJECTIVE : To study lhe regulatory effects of recombinant IL-IO

on IL-6 mRNA and proteni production m human fetal membranes

STUDY DESIGN : Ammocliormme membranes were collected

frmn women undergoing elective C-section Membranes were

maultamed m au organ explaut system and stanulated w~th media

containing eudotuxm (LPS) (50ng/ml) and varmus amounts of

reemnbuiaot IL-10 (10, 50, 100ng/ml) Expemnents were

conducted ni a dose and tnne dependent mamier Transcrlptmn and

traoslatlon of mterleukm-6 were momtored using quanhtatlVe

PCR and ELISA

RESULTS : Ammoehonoa m culture shows a dose dependent

decrease in the produehon ofIL-6 mRNA and protein in response

to IL-10 stnnulatlon This effect was not evident at 3 hours m

culture, however, by 8 hours in culture a marked effect, which

persisted over a 24 hour penod was seen The IL-10 erect was

only present when tissue was concurrently shmulated with LPS, IL-I O inhlba~oo was not effective m the absence of LPS st~mulation,

CONCLUSION : The Addition of IL-10 to the culture me&a

leads to transcriptional regulatiou of IL-6 resulting m decreased

production of both mRNA and protein by human amnlochonon

The decrease oflL-6 is a dose dependent effect ofIL-lO IL-6 has

been associated w~th uifection uiduced preterm labor and causes

prostaglaodni release from fetal membranes Th~s finding may

have stgnlfieant nnphcahons w~th respect to a possible role for

IL-10 or an 1L-10 stanulatory factor m the treatment of

mflammator-5 eylokme associated preterm labor

476 SPO Abstracts JanumT 1996 Am J Obstet Gynecol

609 FLOW CYTOMETRIC CHARACTERIZATION OF MIDTRIMESTER HUMAN

FETAL LIVER HEMATOPOIETIC STEM CELLS

J M Gfllesx, E Bentohlax, D Gebhard~, M Y Dlvon, W K RashbaumX and W

D Lyman~, Deper~ments of Obstetrics & Gynecology and Pathology, Albert Einstein College of Med=mne, Bronx, NY OBJECTIVE: To characterize hematopolet~c stem cell populations In the

mldtnmester human fetal hver that may be suitable for transplantation therapy

STUDYDESIGN: Inforrnedconsentwasobtalned prlortoelectlvetermmatmn

at 11 to 24 week gestations by dilation and evacuation Gestatlonal age was determined by last menstrual period, uterine size, sonographlc evaluation and

by pest-abortaHoot length Dlssocmtad hver cell suspensions from 13 fetuses

were prepared by passing tissue through stainless steel mesh grids Erythrocytes were separated from mononuclear cells through a FlcolI-Hypaque

d~scontmuous density gradleot The cells were =ncubatad with antibodies to Interleukm-3 receptor (CD123), Thy-1 (CDw90), c-kit (CD117), CD33 or with

~sotype controls These antibodies were conjugated, m various combtnat~ons,

to fluorochromes including phycoerythnn, fluorescem, and Tn-color Using mull=pie color analysis mode, the cell populations were analyzed by flow

cytometry RESULTS: The relative percentages of cells staining posRIvely for CDwg0,

CDl17 and CD33 varied considerably and therefore were not statistically s~gmficant Staining for CD123 demonstrated a positive correlation with

geetat~onal age in hver cell populations from fetuses from 11 to 17 gestatlonal weeks, the percentage of CD123-posrtJve cells vaned between 12 to 16 percent

]n contrast, In cell preparat=ons from hvers of fetuses 18 to 24 weeks, this percentage varied from 21 to 34 (ANOVA p=0 006) CONCLUSIONS: Late second tnmester human fetal hver conta=ns a higher

percentage of CD123-pos=We cells than In earher gestations Th=s marker has

been identical as a phenotyplc component of plunpotentlal hematopoletlc stem cells Because the liver increases in mass with advancing gestatlonal age, the

absolute number of cells with this phenotype ~s also greater Therefore, hver

cells from late mldtnmestar fetuses may provide the most suitable source of transplantable hematopo=et=c cells

611 RISK FACTORS FOR PRETERM BIRTH

SUBTYPES. G Berkowttzx, C Princex, R Lapinskix, D

Savitzx, Dept Ob/Gyn, Mount Sinai Medical Center, NY, NY.

OBJECTIVE: This study was undertaken to assess whether

epidermologlc risk factors for preterm birth (<37 weeks of gestation) differ according to etiologic pathway in a heterogenous

urban populanon. STUDY DESIGN: The study population comprised 31,525

singleton live births that took place at a large, urban hosDtal between 1986 and 1994 Preterm births were subdivided into preterm premature rupture of the membranes (PPROM), preterm

labor, and medically induced births. Information regardmg the preterm subtypes and their eptdemiologic risk factors was obtained

from a computerized pennatal database RESULTS: There were 3176 preterm births during the study

period. Of these, 1335 (42.0%) were due to PPROM, 1215

(38.3%) resulted from preterm labor, and 626 (19.7%) were due to medical intervention. Women who were black or Hispanic, those

who had had a previous preterm birth, those who began prenatal care after the first trimester, women who had been exposed to DES

in utero, patients with preexistmg diabetes, and those with antepartum bleeding were at an increased risk for all three pathways. The three groups did not share other risk factors such as

young or advanced maternal age, payment by third party insurance, low body mass index, low weight gain, nulhparity, hypertensive

disorder, pyelonephntts, cigarette smoking, Illicit drug use, and infant birthweight under the 1Oth percentile.

CONCLUSIONS: While some of the risk factors for preterm

birth apply to all three subtypes, a number of factors differ across

the pathways.

610 THE ROLE OF CORTICOTROPIN-RELEAS1NG HORMONE AS A

PREDICTOR OF PRETERM BIRTH. G Berkowltzx, R Lapinsklx, C

Lockwood, P Floriox, C Princex, F Petraghax, Dept OB/GYN, Mount

Sinai Medical Center, NY, NY.

OBJECTIVE: This study was undertaken to assess whether

cortmotropin-releasmg hormone (CRH) is a predictor of spontaneous

preterm delivery

STUDY DESIGN: Maternal serum levels of CRH and its binding

protein were measured from 20 weeks of gestation in a cross-sectional

study of 396 asymptomatm women at high risk for preterm dehvery.

RESULTS: As shown in the table below, CRH levels increased with

gestational age However, gestational-age-spemfic CRH levels were not

increased for preterm compared to term deliveries, whether preterm

delivery was due to preterm labor or preterm premature rupture of

membranes. The binding protein for CRH did not vary according to

gestatlonal age until term, when It declined substantially

Median values of serum CRH (pg/ml) in term, preterm premature

rupture of membranes (PROM), and preterm labor pregnancies,

Mount Sinai Hoseital. 1990-19~9~,

Gestational Age Term Preterm PROM Preterm Labor P Value at Samulln~ twks) CRH (n) CRIt (n’} ~RH (n)

200-23,9 30’ (87) 30 (4) 30 (7) 067 240-289 30 (119) 166 (4) 21 5 (8) 062 290-329 583 (65) 1734 (10) 399 (7) 033 33.0- 369 361 6 (44) -- (0) 284 8 (5) 0 88

__> ~7,0 11176 (36)

* 3 0 denotes below detection limit

CONCLUSIONS: Contrary to previous studies, these data provide no

evidence that elevated maternal CRH levels are predtctlve of preterm

birth Nevertheless, the drop in the CRH-bmdtng protein level at term

suggests that the bloavailablhty of CRH increases as parturition

approaches

612 THE ROLE OF SHORT INTERPREGNANCY INTERVAL IN LOW BIRTH WEIGHT,

PRETERM BIRTH AMONG WHITE, AFRICAN-AMERICAN, AND HISPANIC

WOMEN. G Berkowazx, R Lapmsk~x, R Berkowltz. Dept OB/GYN, Mount Sinai Me&cal

Center, NY, NY

Objective: Th~s mvesnganon tamed to evaluate a h~othesized association between a

short mteIpregnancy interval and an increased risk of low b~rth weight (LBg0, preterm

births, parnculady anaong African-American women.

Study Design: The study populanon consisted of 6754 women who had had two

consecunve, singleton dehvenes at one hospltaI between lanuary 1986 and lune 1995

The intevpregnancy interval was calculated as the length of nme between consecutive

dehvenes minus the gestational age ot the second chdd at b~rth The risk of a LBW,

preterm birth was evaluated according to the duration of the mterpregnancy Interval

for white, grican-Amencan, and H~spamc women after adjustment for potential

confounders

Results: The rate of LBW, preterm b~rths was 2 1% for whites, 9 6% for Afncan-

Amencans, and 6.6% for H~spanlcs In the umvarlate analysis, the preterm rate was

increased ~n all three racial ethmc groups for those with an lnterpregnancy interval of

less than 90 days. However, no consistent associations were evident for intervals of

91-180 days or 181-270 days Furthermore, as shown m the table below, when

adjustments were made for potential confounders, the only statistically sigmficant finding was observed for an interval of 181-270 days for African-Americans

The Risk of LBW, Preterm Birth According to Length of Interpregnancy

Interval by pocial/Ethnic Groups

Wnae Black Hispanic

Ad! OR (95% CI) ad! OR (95%Cl) A& OR (95% CB

<_90 days 2 3 (0.6-7 9) 1 8 (0 6-5.5) 1.5 (0 7-3 5)

91-180 days 0.6 (0 2-2 1) 1 3 (0.6-2.8) 1 4 (0 4-2.6)

181-270d~y~ 18 (096-34) 21 (11-4.2) 16 (08-31)

Conclusion: These data do not provide any consistent evidence that the nsk of LBW,

pretem~ birth Is related to the lnterpmgnancy interval among whites, Afncan-

Amencalls, or Hispanic women

Volume i74, Number 1, Part 2 SPO Abstracts 477 Am.J Obstet Gynecol

613 EXI~RTION. EMPLO YM~NT, AND RELAXATION DURING PREGNANCY:

ASSOCIATIONS WITH BIRTH OUTCOMES. G. Woox, L. Castro, C.

DunkeI-Schetter~, C. Walla, C. Hobel. Dept.’s Ob/Gyn and Psychology, Cedars-Sinai and UCLA Medical Centers, Los Angeles, California.

OBJECTIVE: To explore the relationship between physical activity during

pregnancy and birth outcomes in a comprehensive manner by evaluating

employment and household related exertion, exercise for fitness, work

hours, end job type while controging for coP.founding modifiers,

METHOD~: Two hundred and fifteen ethnically and socio-economically

diverse women were studied prospectively from 18 weeks of gestation

until delivew. Interview measures assessed amount and type of

physical activity carried out at work or st home, exerci~;e for fitness,

classification, work hours, and relaxation time. Relationships between

different activities end birth outcomes {gestatlonal age at delivery and

birth weight} were evaluated by multivariate analysis controlling for

sociodemogrephlo status, health behaviors and medical risk.

RESULTS: The ma~at study t’md~.gs w~’a: 1}statiOnary axert~on

(standing, lifting, bending) was negativaly associated with birth weight

(_B= -146, _b= -,15, p<.01), Each unit increase in stationary activity

was associated with a 146gm decrease in birth weight. 2)Aerobic

activitias (walking, rapidly moving, exercise for fitness) and job

classification had no apparent influence on birth outcomes. 3)Number

of hours employed was ~negatively associated with gectatlonal age

adjusted birth weight (_B= -8.37, b_= -.23, p<.01). 4)There was a

significant interaction between relaxation time and statlona~f exsrtlon

(p < .05) with the lowsst birth weights occurring in those women with

highest sxertlon scores and lowest relaxation times.

CONCLUSIONS: The current study is the most comprehensive to date

and adds insight regarding the types of physical activities that may

affect birth weight end gestational age at deliveW, Our results suggest

variables such as relaxation time and personal control may significantly

modify the effects of physical activity on birth outcomes. Supported

by NICHD R01 HD29553-01.

615 REDUCING LOW BIRTHWEIGHT B/RTH IN AFRICAN-

AMERICAN WOMEN THROUGH INTENSIVE NURSING

INTERVENTION VIA TELEPHONE. M.L. Moore x p.j

Meis, J.M. Ernest Dept. of Ob/Gyn, Bowman Gray School of

Medicine, Wake Forest University, Winston-Salem, NC.

OBJECTIVE: To examine if nursi~ag support through 2~4

telephov.e calls per v4eek from 24-37 v4eeks gestat~,,v~ll ~educe

rates of low birthweight (LBW) and preterm births in a

population of low income African-American women.

STUDY DESIGN: This report of a randomized prospective trial

of nursing telephone intervention included 1,113 singleton births

[557 study subjects (S), 556 control subjects (C)] All subjects

received prenatal care at a clinic staffed by residents and faculty

of a medical school; all were delivered at the same hospital by

resident staff. Both antenatal and intrapartal caregivers were

blind to group assignment. (S) and (C) did not differ in

demographic, health, and social history or measures of stress and

support. No subjects received home uterine ambulatory

monitoring. RESULTS:

S C RR 95% Conf Inter

Weight <2500 gm 11.3% 15,3% 0.74 0.55-1.00

Gestationalage <36 wks 9.4% 12,8% 0.73 0.52-1.02

CONCLUSIONS: Nursing telephone intervention may offer an

effective method for preventing preterm and LBW births in low

risk African-American parturients. Supported by Grant #ROl-

NR02410-01, NINR, NIH.

614 OBSTETRIC CHARACTERISTICS OF GEORGIA ADOLESCENT

PARTURIENTS ENROLLED IN MEDICAID. M Sprauve. M. Lindsay, R

Pace, B, Skellie, A Jordan, M Smith, Dept of Gyneeology & Obstetrics, Emory

Unwersity School of Medicine, Atlanta, Ga

OBJECTIVE: To assess the characteristics of adolescent pregnancies in

Georgm Medicaid recipients and identify predictors of adverse perinatal outcome in this populatmn

STUDY DESIGN A retrospective cohort study was performed using pregnant Medicaid recipients who delivered in the year 1991 The study population

consisted of 10,782 adolescents ages 12-17 years. The comparison group was

33,441 older parturients ages 21-27 years A descriptive analysis of selected

maternal characteristics was completed, then univariate and stepwise logistic

regression was used to ~dentif’y the relationship between maternal youth and

adverse outcomes

RESULTS: The incidence of adolescent pregnancy in 1991 was 14%

Adolescents were significantly more likely than controls to be black (59.65% vs

55 17%), to have hypertension (3.92% "as 33~,%), unnary tract mfect~ons

(15.92% vs 14 36%), pyelonephntIS (1 70% vs 1 36%), syphilis (0 64% vs

0 34%), and gonorrhea (1.54% vs 1 28%), p < 05 Additionally, they were

significantly less likely to have diabetes (4 32% vs 5.46%), penpartum hemorrhage (5 27% vs 5 81%), and vaginitis (10.74% vs 12 16%), or tu abuse

substances (0 10% vs 1 20%), p< 05.

Adolescents were significantly more likely than controls to experience preterm

delivery (6 71% vs 5 21%), crude odds ratio (COR) 1 31, 95% confidence

interval (CI) 1.17, 1 46) After adjusting for confounders including race,

substance abuse, unnary and gen:tal tract refections, the increased risk of preterm delivery persisted (adjusted odds ratio (AOR) 1 28, 95% CI 1 14, 1.44)

The risk of intrauterine fetal demise (OR 1 0, 95% CI 0 67, 2 17), preterm labor

(OR 1 0, 95% CI 0 67, 2 17) and post term dehvery (OR 1 07, 95% CI 0 64,

1.80) was similar between the groups

CONCLUSIONS: In Georgm Medicaid recipients, pregnant adolescents are at

increased nsk of preterm dehvery These pregnancies are characterized by a

higher prevalence of hypertensive &sorders, genitounnary tract infections, and

sexually transmitted dtseases

616 NEONATAL oUTCOME OF TWINS VERSUS SINGLETONS AFTER PRETERM DELIVERY S.A. Friedman, E, Schiff, L. Kao, B.M. Sibai. Department of Obstetrics and Gynecology, University of Tennessee, Memphis.

OBJECTIVE: To determine whether twin fetuses mature earlier than singleton fetuses and thus have a more favorable neonatal outcome following delivery

SPrior to 35 weeks’ gestation. TUDY DESIGN: A matched cohort design was used to study 224 neonates.

Twins ddivered at 2~-35 weeks (n = 112, 56 sets} were matched for gematlond age (5:1 week), race, gender, and mode of delivery to slngleto~a controls (n = 112). All pregnancies had an ultrasound performed prior to 24 weeks. Pregnancies we’re excluded if there were maternal medical problems, premature rupture of membranes, twin-twin transfusion syndrome, or known fetal anomalies. Maternal and neonatal charts were reviewed for a variety of

outcome varlab es. Results are reported as median or incidence and compared using the Wdcoxon signed-rank test or McNemar’s test as appropriate. P values < 0.05 are considered significant. RESULTS: The median gestational age at delivery was 32.9 weeks for twin

significantly lower than in the singleton group (1670 vs. 2030 g, p< 0.001).

Exposure to a full course of [~lucocorticoids was slightly higher in the twin group (54 vs. 43%, p= 0.14). other outcome variables are shown bdow.

Twin Singleton Odds 95%

(n = l12) (n = l12) Ratio Cl

Spectal care admission (%) * 87 60 4 3 2.1-8.9

Special care (days) * 8 5

RDS (%) 24 17 1.6 0.8-3.2

Meehan~at vet~ti|atiot~ (days) "~ 0 0

IVH Grades lII and IV (%) 4 3 1.4 0.3-7.8

NEC Grades II and III (%) 3 5 0.5 0.1-2.3

5-min Apgar_< 6 (%) 16 17 0.9 0.4-2.0

Neonatal death (%) 4 6 0.7 0.2-2.6

CI, confidence interval; RDS, respiratory distress syndrome; IVH,

intraventricular hemorrhage; NEC, necrotizing enterocolitls. * p< 0.001 f p= 0.05

CONCLUSION: Overall, twins were admitted more frequently to the special cute unit, had (onger admissions there, and spent more days on mechanical ventilation, suggesting that they do not mature earlier than singleton fetuses.


617 LONG TERM FOLLOWU’P OF SURVFVORS OY TWIN

TRANSFI.ISION SYNDROME. M A. Johnsonx, Y. Vaucher*, K. Benirschke, T R. Moore. D~sion of Neonatal and Pennatal Medicine, Univ of Calff San D~ego.

Objective: To study the long term neurodevelopmental (ND) consequences of survivors of twin-transfumon syndrome (TTS)

Stady Design: A retrospechve case-control study was performed utilizing 129 TTS cases identffted from pathology records and verified by large A-V anastomoses on vascular ~njechon. 35 of these cases had clinical features of TTS, yielding 62 survivor twins.

15 twins died in the neonatal period, leaving 47 31/47 had .>1 ND followup visits. Two contemporaneous d~zygohc twin controls, matched by gestatlo~tal age, sex and s~mllar followup, were compared to each case. FolIowup was grouped in 4 month intervals by adjusted gestational age. 74% of TTS and 95% of

controls had followup _>12 months. Results: There were no differences between cases and controls ~n antenatal factors, including maternal age, gestat~onal age (29 7,

30.2 wks) panty, and prenatal care. Intrapartum and neonatal factors were sirmlar except cases had more betamethasone (52% vs

29%) and fetal distress (28% vs 6%), but less preterm labor (68% vs

92%), p<.05. Birthweight percentdes of "ITS cases were lower (36% vs 63%, p=.~(YS) and cord blood base deficits higher (6 9 vs 3.8, p= 01). The irut~al ~nc~dence of IVH (41%, 56%), RDS (56%, 50%) and BPD (17%, 11%) were similar. Followup neurolog~c

assessments (MDI, PDI SCores) were slmdar at 4-6 and 7-10 me. Although TTS cases had lower ND scores and a higher xnc~dence of CP at 11-15 me., by" 21 me. both ND scores and the incidence of

CP were s~milar (12% vs 9%). Conclusions: We conclude that initial mortahty and morbidity of TTS ~s l-ugh but that neurodevelopmental outcome, compared to dizygohc controls, is similar by 21 months

619 PERINATAL OUTCOME IN QUADRUPLET GESTATIONS. ]~R Barton, JM O’Brien, DL Jacquesx, NK Bergauerx, GJ Stanzianox, BM Sthai. Central Baptist Hospital, Lexington, KY, Healthdyae Matermty Management, Marietta, GA, Umvers~ty of Tennessee, Memphis, TN. OBJECTIVE: We previously reported a positive correlataon of both maternal height and weight w~th blade weight (BW) and gestational age (GA) at dehvery m triplet gestataons. The purpose of th~s study was to describe the permatal outcome of intensively momtored quadruplet gestataons and to evaluate maternal factors associated with an increasing

GA at delivery and BW of surv~wng neonates m these multiple gestatmns. STUDY DESIGN: Pregnancy and perinatal outcome data were prospectively collected for quadruplet gestataons managed w~th an intensive

preterm labor identification program from Feb 1991 to Oct 1994...All patients received dady perinatal narsuag contact and home utenne actav~ty monitoring. Individual patient management and timing of dehvery were at the discretion of the pataent’s attending phys~cmn. Statastwal analys~s included a linear regressmn model to assess the influence of body mass index (BMI) and gravithty on BW and GA at dehvery (SPSS V.6.1).

RESULTS: 37 palaents w~th quadruplet pregnancies were followed. There was i st~lbirth and 12 neonatal deaths among 148 baths (pennatal loss rate

of 8.8 %). 4 women (11%) delivered at 24-28 weeks gestatmn. All but one woman (vaginal dehvery following PPROM at 21.6 weeks gestataon) underwent cesarean dehvery. Outcome parameters are summarized in the

Table. M~n + SD (Min, Max)

Maternal Age (years) 31.1 ± 3.3 (24, 40)

GA at Enrollment (wk) 20.7 + 2.8 (17.1, 29.0)

GA at De’a’~ery ~wk) 3~.(~ ± 2.7 (Z~.6, 35A)

Birth Weight (gin) * 1486 + 387 (738, 2468)

Neonatal Stay (days) * 42.8 ± 30.5 (8, 190)

¯ Data are for surwving neonates. The linear regressmn model revealed a slgmficant positive influence of

BMI on BW (p=0.003), but neither BMI nor grav~&ty were s~gmficantly associated wl~ GA at delivery. CONCLUSIONS: S~rndar to triplet gestations, a pos~tave correlataon of

body mass index with birth weight exasts in quadruplet pregnancies. However, despite close antepartum superv~smn, quadruplet pregnancies are associated w~th an increased pennatal morbi&ty and mortahty and a h~gh rate of extreme prematunty. These data should be considered ~n counseling patients with quadruplet pregnancies.

618 OUTCOME OF TWIN GESTATIONS WITH PRETERM LABOR PRIOR TO

24 WEEKS’ GESTATION, O.A. Rust, J.C. Morrison, S.J. Schorr, K.G.

Perry, Jr., W.E. Roberts, R.W. Martin. Dept. Ob/Gyn, Univ. of

Mississippb Jackson, MS.

OBJECTIVE: To assess the outcome of pregnancy in women with twin

gestation (TG) and preterm labor (PTL) before 24 weeks’ gestation who

were prescribed home uterine contraction assessment (HUCA) either

before PTL occurred or after PTL had been successfully arrested.

STUDY DESIGN: The patient population in this retrospective,

descriptive study revolved women with TG who had PTL before 24

weeks’ gestation and also were mallaged with HUCA. They were

stratified by whether HUCA was prescribed before or after diagnosis of

PS"L. The main outcomes studied were gesta~ional age at PTL and

delivery, as well as birth weight and neonatal intensive care unit

admission. RESULTS: Fifty-nine women over an 8-year period (6/86-7/94) were

noted to have TG managed with HUCA and PTL at < 24 weeks. Of

these women, 51% (n = 30) were prescribed HUCA after PTL at < 24

weeks was arrested with toeolytic medication (Group II. The remaining

29 patients were prescribed HUCA before the episode of PTL had

occurred prior to 24 weeks (Group II). Of the 30 women in Group I,

PTL was diagnosed st 22.8 __+ 2°9 (mean __+ SO) weeks and the

gestational age (GA) at delivery was 27.4 __+ 4.9 weeks. The birth

weight in Group I was 942 + 440 gm and of the 60 infants, 51 had

neonatal tnrnnsive ears unit admissions. Of those m Group )1 who

developed PTL ( < 24 weeks) after beginning HUCA, the GA st PTL was

23.5 + 2.6 weeks (p = NS) end their GA at delivery was 34.4 + B. 1

weeks Ip < O.01L The mean birth weight of Group II was 239,$ __+

661 gm with 11 of 58 infants requiring neonatal Intensive care unit

admission (p < 0.05, 0.01, respectively).

CONCLUSIONS: Women with TG and PTL before 24 weeks’ gestation

are more likely to deliver later in gestation, have greater birth weight

and fewer neonatal intensive care unit admissions =f HUCA is prescribed

prior to diagnosis of PTL.

620 cIMVIXT ~ M-J. Xenakis, J. Piper, D. ConwaY~, 0. Langsr, DepL Of 0b/Gyn, UTHSC,

San Antonio TX,

~$JECTIV~ To test the hypothesis that use of an integrative induction approach

combining prosta~landin Ez, anmiotomy and high dose ox~tocth can overcome

the deleterious effect of the unfavorable con/ix on success of induction. STUDY D~I~: 597 pregnancies were enrolled in this prospective study. An

integrative induction protocol using PGEz, amaletomy and high dose oxytocin was

used. Fnilod Induction was defined as th~.hlll~ to achieve the active phase of

labor. Patients were stratified by Bishop score at entry (0-3, 4-6, ~7) and parity

for comparison of success of Induction, maternal and fetal complications, and

duration of labor. ~ Toere were 252 {~2%) nulliparas and 345 {58%) multiparas. The Overall

failed induction rate was 3.7% and overall cesarean section rate was 20%.

Success and failure rates did not differ between the 4-7 and >7 Bishop

categories, thus they were combined for all further analysis.

PRIMI~ARAS MULTIPARAS

Bishop score 0-S ~4 0-3 ~4

Failed Induction 13.5% 0.6% 5.2% 1.3%

Cesarean Section 34% 19% 24% 12%

When stratified by parity, comparison of tile 0-3 and ~4 Bishop categories

revealed significantly higher rates of failed induction (RR 10.2 [2.8-44J) and

cesarean section (RR 2.1 [L4-3.2]), Overall, indications for cesarean section

included lnbor abnorm~ties (46%), tet~ distress (31%) and failed induction (17%). Regardless of Bishop score at entry and parity, maternal and fetal complications

were infreqanut and not statistically sigutflcanL

~ONt’INlll~: Women with a B|shop score of >4, regardless of parity, can

anticipate a successful induction and high likelihood of vaglnai delivery, h~

contrast, failure remains considerable in women with a Bishop score

Volume 174, Number l, Part 2 SPO Abstracts 479 kan J Obstet Gynecol

621 REST & DESCEND VS, PUSHING WITH EPIDURAL ANESTHESIA IN THE

2ND STAGE OF LABOR. SL Hansenx, SL Clark. LDS Hospital, SLC, UT

OBJECTIVE: To compare women with epidural anesthesm who were

encouraged to push from the onset of 2nd stage w~th those who had a

penod of rest before pushing began.

STUDY DESIGN: Prospectively, 252 pts with epldural anesthesia during

the 2nd stage of labor were randomized to a rest period (pnmlgravldas =

2 hrs, mult~grav=das = 1 hr) or ~mmethate pushing at complete ddatatmr~.

Measured vanables were rate of fetal descent, length of pushing, the #

and type of fetal heart rate decelerations, Apgar scores, artenal cord

values, permeal injunes, mode of delivery, length of 2nd stage, and

maternal fatigue. Fatigue was measured by using a visual analogue

scale, verified with the Modified Faugue Symptom Checkhst Vanables

controlled for were bladder status, level of anesthesia, posit=on changes,

uterine contracUon pattern, and pushing style and frequency. Data were

collected using a computerized charting monitor.

RESULTS: The duratmn of the 2nd stage was lengthened when a period

of rest was used before pushing (p=.000) (range 2 to 295 m=n.), actual

pushing time was decreased =n pnm=gravldas (p=.021) and ~n

multlgrawdas (~=.001). Fetal weft being, as reflected by Apgar scores

and cord artenal ph values, was shghtly better in the resting group,

though not significantly so. There were significantly fewer decelerations

~n pnm=gravldas (p= .000), and =n mult=gravidas (j2=.004). Fatigue was

significantly less m pnm=grawdas (p= 017). Penneal miunes and mode

of dehvery were s~mdar m both groups.

CONCLUSION: We could document no benefits of a pohcy revolving

Immediate and continued pushing in the 2nd stage compared to allowing

a vanable penod of rest with spontaneous fetal descent Pennatal

outcome was similar, despite prolongation of the 2nd stage (in resting

pts) for up to 4.9 hours. The decrease in variable decelerations, maternal

fatigue, and in actual pushing time seen In the resting group suggests

that a penod of rest may be benefimal =n pts with slgmficant 2rid stage

variables, or in whom fatigue interferes with pushing efforts.

623 RISK OF CESAREAN DELIVERY IN NULLIPAROUS WOMEN WHO

PRESENT IN LABOR WITH AN UNENGAGED FETAL HEAD K

Murphy, MDx L Shah, MD, W.R- Cohen, MD. Dept of Ob/Gyn Jacobi

Medical Center and Albert Einstmn College of Methcme, Bronx, N. Y..

OBJECTIVE.To assess the relatmushlp of fetal station in early labor to

subsequent patterns of dilatation and descent and to the probabthty of

cesarean dehvery.

STUDY DESIGN’One hundred and thirty-two term nulhparous women m

spontaneous labor were evaluated. Selecnon cntena: 1.regular uterine

contractmus 2.adm~ssmn cervical exam < 6 cm. 3.greater than 37 weeks

gestatmn 4.smgleton vertex presentation 5.no ewdence of thabetes melhtus.

For each participant, the labor charactenstms, the labor curve and mode of

delivery were recorded. At birth, the infant’s weight and head mrcumference

were determmed Statistical analysis was performed using appropriate

parametric and non-parametric tests.

RESULTS. Of the 132 participants, 29(22%) presented with an engaged

fetal head (g saatmn or below ), and 103(78%) vresented with an m~ongaged

fetal head When the unengaged group was further stratified, 15(14.5%)

presented w~th a floanng fetal head (-3 station or above), and 88(85 5%)

presented with a dipping fetal head (-2 or -1 station) Overall, 12 (9%)

underwem primary cesarean sectmn, 2 (6.9%) in the engaged group, 6(6 8 %)

in the dipping group and 4(27%) m the floating group. The rate of cesarean

sectmn for the floating fetal head group was mgnifieanfly higher when

compared to the other two groups (p=0 042). In addition, the floating fetal

head group demonstrated longer second stages of labor(p=0 022) Median

values for the second stages were 98 min. (floating group), 52 mln. (dipping

group) and 53 rain. (engaged group).

CONCLUSION’The majority of nulhpara m th~s study presented h~ labor

with an unengaged fetal head. No significant differences were seen between

those women with an engaged vs. unengaged d~ppmg fetal head with respect

to labor patterns or risk of cesarean delivery. However ,women who

presented with a floating fetal head demonstrated significantly longer second

stages of labor and higher rates of cesarean section.

622 PROLONGED LABOR IN NULLIPARAE - LESSONS FROM THE ACTIVE MANAGEMENT OF LABOR. L Malone,X M Geary,~ J Stronge,x

D Chelmow,x P. Boylan,x M. D’Alton Depts of OB/GYN, New England

Medical Center, Boston, and The National Maternity Hospital, Dublin, Ireland

OBJEISrlVE: To outline predictors of prolonged labor (labor >12 hours) m

nulliparae undergoing active management of labor (AML)

STUDY DESIGN: All nulllparae delivered from 1990 to 1994 at the National

Maternity HosDlal, with labor lasting more than 12 hours were reviewed Patients with induced labor, multiple gestanons, noncephabo presentations or

labor at <37 weeks were excluded Each patient was matched with the next

nulliparous patient who debvered w~th a labor lasting <12 bouts Patients were

also matched on the basis of membrane status on admission A!I nulllparae were

managed according to the prewously desenbed AML protocol from the National

Maternity Hospital

RESULTS: 9,018 nulhparae met inclusion criteria, with 147 having prolonged

labor (1 6%) Prolonged labor was due to metticlent utenne actmn (IUA) in

65%, persistent occlp~topestenor posttton (POP) in 24%, and cephalopelvtc

disproportion m 11% of cases 97 9% of IUA patlents delivered vagmally

compared to 50% of POP patients (p=0.00001) The cesarean section rate was

25% m the prolonged labor group compared with 5% m the control group

(p=00001) Umvanate analys~s showed stat~stmally significant (p<005)

differences m maternal body mass index (mean 24 8kg/m for eases, 23 7kg/m

for controls), cervical ddatmn on admlssmn (1 lcm, 2 lcm), oxytocin use (99%,

53%), ep~dural use (90%, 34%), early placeme~lt of epldural (65%, I2%), and

birthweight (3878g, 3465g) Maternal age, gestatmnal age at debvery and

neonatal outcome were similar tn both groups Afler controlling for confounding

¯ ~anables with multivariate conditional logistic regression, the following factors

remained significantly associated w~th prolonged labor

Variable Odds ratm 95% confidence intervals

Cervix <2cm on admission 3 1 1 3-7 3

Epidural placed at <2cm 42 7 7.5-242.0

Epldural placed at >2em 5.1 1 9-13 7

Blrthweight >4,000g 10 2 3 6-29 4

CONCLUSIONS: Prolonged labor is rare m eulbparae managed w~th AML

Less advanced cervical ddataon on admission and epidural anesthesia, especially

when placed early, are strongly associated w~th prolonged labor in nulbparae

624 MATERNAL SERUM DEHYDROEPIANDROSTERONE-SULFATE

LEVELS AND MYOMETRIAL EFFICIENCY AT TERM IN

NULLIPARAS <25 YEARS OF AGE

L Goolsbv. K Schlechtx, C Harzkex, P Ubarriet,x K Reed. Depts of

OB/Gyn, Arizona Health Sciences Center, Tucson, AZ and Texas Tech

Health Sciences Center, El Paso, TX OBJECTIVE: To evaluate the maternal serum dehydroepiandrosterone

sulfate (DHEAS) level as a factor ~nfluenc~ng myometnal contractile

"efficiency" during labor at term.

STUDY DESIGN: Upon admission to the Labor & Delivery unit, blood

was obtained from 55 term nulliparous women < 25 years of age

presenting in active labor Following delivery, umbilical venous cord

blood was also obtained. Pregnancies complicated by diabetes melhtus,

hypertension, intrauterine growth retardation, tobacco use, or

chodoamnionitis were excluded. Serum DHEAS levels were measured by

radioimmunoassay. DHEAS levels and other obstetric variables were

retrospectively correlated with the clinically-determined requirement for

oxytocin augmentation of labor. Student’s t test was utilized for statistical

analysis. A p-value < 0.05 was considered significant.

RESULTS: Q~ytocin Augmentation

~ (n=30) no(n=25) p

Maternal DHEAS (lag/all) 100.27 135.34 0.0339

Maternal age (yrs) 18,72 20.07 ns

Umbilical vein DHEAS (lag/all) 341,77 330.35 ns

Brtthweight (g) 3266.9 3368.5 ns

Gestational age (wkS) 39.09 39 68 ns

(ns=not significant) CONCLUSION: Among term nulliparous women < 25 years of age,

maternal serum levels of DHEAS are significantly lower in those clinically requiring pharmacologic augmentation than in those progressing spontaneously through labor. DHEAS may be an important factor in efficient labor.


625 MULTIVARIATE ANALYSIS OF RISK FACTORS FOR OPERATIVE

DELIVERY. L. Turcot, W. Fraser, S. Marcoux. Depts. Ob/Gyn and Soc.

Prey. Mad., Laval Umvarsity, Quebec, Canada.

OBJECTIVE: To assess sodo-demograpNc, anthropometric and

obstetncal nsk factors for operative delivery (OD).

METHODS; Data were obtained prospectively on 925 nulliparae In

spontaneous, term labor with a single fetus in cephalic presentation. OD

was defined as either a m~dforceps or a cesarean delivery. Variables

were grouped as those present at admission (Group 1), or those which

appeared during labor (Group 2). Univariate analysis identified variables

which were statistically associated with OD These variables were then

introduced into a multiple logistic regression model. To test if the

inclusion of a variable contributed precision to the model, variables were

withdrawn one-by-one, and the difference in deviance was tested by a

Chi-square test (alpha = 0.10).

RESULTS: Univariate analysis found the following variables were

associated with CO - Group 1: maternal height, pregnancy weight gain,

age, race, gestat~onal age, smoking status, and admission cervical

dilatation; Group 2: dystoc~a, abnormal FHR tracing, oxyto~n, and

epidural analgesia. Variables retained m the final logistic regression

model were: maternal height and age, smoking status, dystocia, epidural,

and abnormal FHR tracing. Smoking reduced the odds of OD (OR = 0.6,

95% CI = 0.3 - 0.9). The odds of OD in the presence of epidural was 3.4

(95%CI = 2.0 - 5.8). This effect was similar in the presence and in the

absence of dystocia.

CONCLUSIONS: Epidural is the obstetrical intervention which is most

closely associated with operative delivery. Chnical trials are required to

determJne if this relationship is causal. Prior to evaluation in a chn~cal

setting, the statistical model requires vahdation in a separate but similar

cohort.

627 CORRELATION BETWEEN MAXIMUM CERVICAL DILATION AT

C/SECTION AND SUBSEQUENT VBAC SUCCESS RATE I.A Hoskins, J.L

Gomez. Dept Ob/Gyn, NYU Medical Center, NY.

OBJECTIVE: To determine whether the extent of cervical dilation at C Section

affects the VBAC success rate in a subsequent pregnancy

STUDY DESIGN: The study period extended from July 1, 1987 to December 30,

1994. Relevant medical records and/or operative reports of the index pregnancy

(group I) were reviewed for (1) maximum cervical dilation attained at time of

C.Section, (2) Pltocin use, (3) indication for a C Section, (4) neonatal birth weight

The records of the subsequent pregnancy (group II) were reviewed (1) for the

success or failure of VBAC, (2) use of Pltocin, (3) neonatal birth weight

RESULTS: Complete informataon on the patient in groups I & II was available in

1917 cases. The indications for C.Section in group I were breech/other

malpresentations 98 (5.1%); fetal distress 286 (14.9%), arrest disorders 1533 (80%)

The maximum cervical dilations achieved in the cases of arrest disorders were _<

5 cms m 11%, 6 to 9 cms in 73%; and full dilation in 16% In group II, the VBAC

success rates were 73% for those with previous CSection for

breech/malpresentation and 68% for those with previous C Section for fetal distress

In cases with previous C Section for arrest disorders, whose cervical dilation was

< 5 cms, the VBAC success rate was 67%; and for cervical dilation 6-9 cms, it was

69% Ho~vever, only 13% of those who achieved full dilation in group I, had

successfulVBACs Thls was statistically significant (p <0 05) The rates oflV

Pitocin use were 76% m group I and 66% in group II Neonatal birth weights were

3788 + 412g in group I and 3736 + 419g in group II (NS).

CONCLUSIONS: (1) Even though approximately 2/3 of the patients who

attempted VBAC were successful, only 13% of those who achieved full dilation,

had a subsequent VBAC in spite of equivalent rates of IV Pltocin use and neonatal

birth weights. Thus patients who attempt a VBAC can be accurately counselled

about the chance of a successful delivery and ~n this way minimize possible

adverse outcomes

626 AMNIOTIC FLUID INDEX (AFI) PREDICTS THE RELIEF OF VARIABLE DECELERATIONS FOLLOWING AMNIOINFUSION BOLUS. CY Spong, F.

Mctqndsey~, MG Ross, Dept OB/GYN, Harbor-UCLA Mad Ctr, Torrance, CA

OBJECTIVE: We hypothesize that the intrapartum pre-amnioinfusion AFI can

predict the response to therapeutic amniomfus~on (AI).

STUDY DESIGN: 51 patients w=th intrapartum repetitive variable

decelerations (VD) requ=ring AI (10 ml/min x 60 ram) underwent AFI detsrminat~on prior to and following AI bolus The fetal heart tracing was

scored (blax~d to AFI values) for number and seventy of VD (mild/moderate, severe, atypical) for 30 rain periods prior to and 1 hr after AI. Success was defined as >50% decrease in the total number of VD or >50% decrease in atypical or severe VD after the AI bolus Spantaneous vaginal deliveries prior

to AI bolus complebon were excluded from analysis "Urgent" operative deliveries md~atad because of the heart rate tracing during the AI bolus were classified AI failures Maternal and neonatal outcomes were analyzed in

relation to the pre-AI AFI Data are reported as mean :l: SD Probability oral

success versus AFI was analyzed by Chi-square test for trend.

RESULTS" The ~nitial AFI was 0-4 cm =n 16, 4-8 cm in 20, 8-12 cm in 12, and >12 cm ~n 3 pahents. The mean AFI prior to AI s~gmficantly increased fo~wing the tx:~s (6 2 + 3.3 to 10 2:1; 4 0 cm), wRh a greater increase noted

m patients with low pre-AI AFI (Table). The probabihty of AI success decreased w~h increas=ng pre-AI AFI, though there was no change ~n the inc=dence of urgent oparat~ve deliveries (Table) There were no s=gnificant

d*fferenoes In maternal (infection, cord prolapse) or neonatal (Apgar _<7, umbilK:al pH <7.2) complications or birthwe=ght (3290:1:484 gin) between the

groups *p<0 05

pre-AI AFI (cm) n a AFI Icm) AI success Urgent operative deiiven/ 0-4 16 58 12(81%) 1 4-8 20 35 13(61%) 3 8-12 12 3 3 5 (42%) 2 >12 3 1.4 1 (33%)* 0 CONCLUSIONS: (1) The AFI value prior to AI can predict the success of AI

for rekef of VD, (2) The majority of patients presenting with intrapartum VD have reduced amnio~c fluid volume (AFI < 8cm), and (3) Equilveient At boluses

effect a greater AFI =ncrsese =n patients with lower pre-AI AFI.

628 USE OF THE FRIEDMAN LABOR CURVE AS A PREDICTOR OF OPERATIVE

DELIVERY IN MACROSOMIC BABIES. I.A Hoskms. J.L. Gomez. Dept.

Ob/Gyn, NYU Medical Center, NY.

OBJECTIVE: To assess the use of the Friedman labor curve as a predictor of

operative delivery in macrosoralc pregnancies.

STUDY DESIGN: The medical records of 1141 patients who had delivered babies

>/-M000 grams from July 1, 1987 through December 30, 1994 were reviewed The

variables studied were’ (1) progress of labor as denoted on the Friedman Labor

Curve, (2) use of IV Pltocm, (3) need for operative delivery.

RESULTS: There were 1348 patients with neonatal birth ~veights >/=4000 grams,

but complete records were only available on 1141 of them. For every chart of a

mother with neonatal birth weight >/=4000 grams (Group I), the chart of the next

mother who delivered a neonate <4000 grams (Group II) was reviewed for

comparison

Group I >/=4000g Group II <4000g P Value

Arrest disorder 41% 39% 0.067

IV Pitocm use 59% 63% 0.073

Prolonged 2nd stage 39% 27% 0 059

(>3hrs)

Operative delivery 29% 20% 0 066

CONCLUSIONS: (1) There were no differences in the progress of labor as

deptcted on the Friedraan Labor Curves, in the 2 groups (2) Even though there was

a trend towards a longer 2nd stage, arrest disorder, and operative delivery in Group

I, this was not statistically significant. (3) Abnormalities in the Friedman Labor

Curve was not useful as the sole indication for an operative delivery in macrosomic

babies


629 OBSTETRICAL PREDICTORS OF SUCCESSFUL DELIVERY OF

LARGE INFANTS. R Joshl~, ,l Gandhi, J Maldman, Dept OB/OYN, The

Long Island College Hospital, SI~xIY-HSCB, New York

OBJECTIVE To evaluate the obstetrical predictors of successful vaginal

dehveU of infants with birth weight BWT > 4000 gins

STUDY DESIGN 344 dehverles w~th BWT -~ 4000 gins were selected

from atotal of 3709 MaternaFFetaJ characteristics, Including maternal

diabetes, gestahonal age, VBAC rate, labor paRem, mtrapartum fetal heart rate (FHR), APGAR scores at 5 ram, shoulder dystocla, neonatal and maternal

trauma, were retrospectively reviewed, and data in all cases of Trial Of Labor

(TOL) was analyzed by incrementally Increasing BWT

RESULTS" 46 patients underwent elective C/Section 298 were allowed

TOL 81% ofbabms with BWT 4500-5000 gms delivered Vaglnally One of

five babies with BWT > 5000 gins, was allowed to labor and delivered

vagmally. 21/33 &abetlcs undetxvant TOL, 66% (14/21) delivered vagmally

All these babws weighed - 4500 gms Gestatlonal age (GA) of > 41 weeks

did not influence labor outcome Successful delivery occurred 90.1%

(1s3/203), 70% (53/72) and 40% (8120) when total length of labor was

<12,12 -20, and >20 hours (p<0.02, p<0.05) A second stage of< 1,

>1,>2,->3 hours was associated w~th a vaginal delivery rote of 100%, 92%

(35138), 88% (16/18) and 11% (1/8) respectively (p<0.05, p>&02) 38%

(8/21) successful XrBAC occurred with BWT > 4000 gms vs 65% in general

populatmn (p<0.O2) Reassunng FHR and APGAR scores > 7 at five minutes

occurred in 92% and 99% respectively among pattents with successful

outcome Shoulder dystocla without neurologle deficit occurred in 2.8%

(7/248) Major perlneal trauma occurred In 11.4 %.

CONCLUSION 82% oflnfant with BWT of 4000-5000 gms could be

delivered vagmally with mxnlmal adverse effeets on FHR, birth lnjunes, or

APGAR scores Risk factors for a dehvery rate of less than 50% Included

previous c/section with a BWT or" > 4000 gins, a labor lasting > 20 hours,

and a second stage of > 2 hours Maternal diabetes w~th BWT of 4000-

4200 gms and GA of 41 weeks were not significant adverse risk factors

Maternal permeal Injury rate of 11% is slgmfieant

631 CLINICAL CORRELATION OF SALIVARY ESTRIOL CONCENTRATION AND CERVICAL RIPENING: A POSSIBLE ROLE FOR FETAL ESTRIOL IN THE ONSET OF PARTURITION. Herman L Hednana. Siobhan Parry~, and Wilham M Gdbert. Division of Maternal-Fetal Medicine, UC Davis, Sacramento, CA OBJECTIVE: To determine whether the recently reported (SPO, 1995) increase m salivary estriol concentrabon (SE3) prior to parturition correlates with the progressive ripening of the cervix. STUDY DESIGN: 128 saliva samples were longitudinally collected from 18 normal pregnant patients at each chmc visit from 30 to 42 weeks gestation. Cervical ripening was assessed by a single operator at each visit and recorded as B~shop score (BS). Using an enzymeimmunoassay (EIA) techmque, SE3 or free estriol was extracted from samples after delivery and correlated with BS. For data companson, the SE3 EIA results were divided into 2 groups based on increasing BS (BS>4 and BS_<4). RESULTS: SE3 concentraUons increased linearly w=th advancing BS (r = 0 92, p <0.0001). The mean SE3 concentrabon (2.553 + 1 140 pg/ml (SD)) in group BS>4 was significantly greater than in BS_<4 (1 012 4- 0678 pg/ml, p <0.0001). Regression of SE3 concentrations wRh BS y~elded a sigmficantly better correlation coefficient in group BS>4 (r = 0.85) than BS_<4 (r = 0.64, p = 0 009). Addfl=onally. the mean rate of SE3 increase w=th advancing BS in group BS>4 (0.318 + 0.100 pg/ml/BS) was significantly greater than in BS_<4 (0.228 4- 0.158 pg/ml/BS, p = 0.0005). CONCLUSION: Increasing salivary estfiol concentration correlates with cervical ripening. Because fetal estriol ~s the predominant component in circulating maternal estriol, the rapid SE3 increase associated w=th cerv=cal ripening suggests a poss=ble active fetal role m preparing the cervix for the m~t~abon of parturition. Sahvary estnol values can be used chmcally m conjunction wRh B~shop score to assess the likelihood of a successful labor induction.

630 FAILURE RATES IN PROLONGED ATTEMPTS TO

INDUCE LABOR

Berkowitz, K. and Gazit, G.x University of Southern California

School of Medicine, Los Angeles, CA

OBJECTIVE: To determine the success rate and safety of

prolonging induction attempts beyond 48 hours as compared to

inductions less than 48 hours in length.

STUDY DESIGN: Population survey of deliveries occurring

between 8/1/83 and 7/31/94. Retrospective chart review with

statistical comparisons using t-test and chi-square where

appropriate. Patients were grouped by the length of induction to

delivery interval. Morbidity was def’med as hemorrhage requiring

transfusion, 3rd/4th degree laceration or neonatal ICU admission.

RESULTS: During the study period there were 9,898 deliveries

with an total cesarean rate of 17.3%. 1185 (12.0%) of pts

required induction, 1089 (92,8%) delivered within 48 hours and

96 (7.2%) received three or more days.

<48 HRS >~8 HRS R.R. (C.L)

(n=1089) (n=96)

Cesarean Rate 18.2% 67.7% 7.4 (4.9-11.0)

(198/1089) (65/96)

Morbidity 16.4% 33.3% 2.3 (1.6-3.5)

(178/1089) (32/96)

CONCLUSIONS: Few panents require more than 48 hours

induction. Prolonged attempts rarely succeed and are associated

with increased morbidity. Prolonged attempts at induction will

not significantly decrease the overall cesarean rate.

632 A COMPARISON OF DIFFERING DOSING REGIMENS OF VAGINALLY ADMINISTERED MISOPROSTOL FOR PRE- INDUCTION CERVICAL RIPENING AND LABOR INDUCTION DA Wing, RH Paul, Dept. of Ob-Gyn, Universtiy of Southern California School of Medicine, Los Angeles, CA.

OBO ECTIVE: To compare two dosing regimens of vaginally administered misoprostol for pre-induction cervical ripening and the induction of labor. METHODS: 522 patients with ~ndications for induction of labor and unfavorable cervices were randomly assigned to one of two dosing regimens of vaginally administered misoprostol. Twenty-five microgram tablets of misoprostol were placed in the posterior vaginal fornix either every three hours to a maximum of eight doses or every six hours to a maximum of four doses. The maximal period of cerdca/ripening was 24 hours regard!ess of the number of doses administered. Medication was not given aRer either spontaneous rupture of membranes or beginning of active labor. RESULTS: Among 522 patients enrolled, 261 randomized to the q3h regimen and 261 to the q6h regimen. The average interval from start of

induction to vaginal delivery was shorter in the q3h group (903.3 + 482.1 minutes) than in the q6h group (1410.9 ± 869.1 minutes) (p<0.001). Oxytecin

augmentation of labor occurred more commonly in the q6h group (51.4%) than in the q3h group (41.8%)(p<0.05). There were no significant differences in mutes of delivery. Overall, 108 patients (20.8%) were delivered by Cesarean. There was a shghtly higher prevalence of tachysystole (6 or more uterine contractions in a ten-minute window for two eonsecutxve ten minute periods) in the q3h group (14.6%) than in the q6h group (11.2%), but this difference was not statistically different. There were no sigmficant differences in frequency of uterine hyperstimulation or hypertonus. There was no significant difference in frequency of abnormal fetal heart rate tracings, meconium passage, one-or five-minute Apgar scores < 7, neonatal resuscitations or admissions to the NICU between the two groups. CONCLUSIONS: Vaginaliy admimstered misoprostot is an effective agent for cervical ripening and the induction of labor. Patients with the q6h dosing schedule experienced longer intervals to delivery, more frequently required

oxy/ocin augmentation, and experienced more failed inductions than patients m the q3h regimen. Further investigation to characterize the safety of mlsoprostol is ne~..ded.

482 SPO Abstracts January t996

Am ~[ Obstet Gynecol

633 CHANGES IN FETAL FZBRONECTIN AFTER SUCCESSFUL TOCOLYSIS WITH ATOSIBAN. MM Ra~irez, M Monga, BT Oshiro, KD Larrabeex, D CasalX* RK Creasy. Dept of ob/Gyn. Univ. of Texas Houston and Adeza Biomedical*. Objective: To evaluate the effects of tocolysis on cervicovaginal fetal fibronectin (FFN). Study Design: 14 patients with preterm labor with 4 or more uterine contractions in 30 minutes and documented cervical change were tocolyzed with Atosiban (Phase II trial). Cervicovaginal FFN was collected before and 12 hours after initiation of Atosiban. FFN was assayed by fetal fibronectfn immunoassay (Adeza Corp.). All samples for FFN were batched and assayed after completion of the study. FFN was considered posltive if > 50 ng/ml. Results: Mean gestational age at entry was 30.8 ± 2.9 weeks with a mean cervical dilatation of 1.6 + 0.7 cm, and ~ 50% effacement. All patients were successfully tocolyzedo 78.5% were positlve (n = ll) for FFN prior to tocolysie. After tocolysis, 63.3% (n : 7) of cervical FFN became negative. The positive predictive value of FEN for preterm delivery after tocolysis was 50.0% (2/4). In contrast, the negative predictlve value of FFN was 85.7% (6/7) for delivery at ~ 36 weeks. The mean gestational age at delivery following a negative FFN was 37.6 ± 2.7 weeks. Conclusion: Conversion to a negative FFN following successful tocolysis remains a strong predlctor for patients to deliver > 36 weeks gestation. This suggest that inhibition of uterine activity prevents the expression of FFN in cervicovaginal secretions of thls proteln.

635 Laminaria Tents Plus Vaginal Prostaglandin Versus Vaginal Prostagtandin Alone for Cervical Ripening. M Lemk~-~, M Turnquest, Dept. Ob/Gyn, indiana University Medical Center; ~ndianapolls, Indiana, Objectives: The purpose of this study was to evaluate the efficacy of add=ng taminana tents to sequential mtravaginat prostaglandm E2 gel for cervical ripening. Study Design: A randomized, prospective study was conducted from October 1994 to May 1995. Pregnant women with maternal or fetal ~nd~cations for induction of labor at 37 weeks’ gestation or greater and a Bishop score less than or equal to 4 were eligible. Nineteen patients received laminana tents in addibon to 4 mg ~ntravaginal prostaglandm E2 gel (study group) while 25 patients received prostaglandin gel alone (control group) /,.fter four hours the lamlnada tents were removed and the gel was continued m both groups at 4 hour ~ntervals, Induction w=th oxytocin was ~mtiated after a Bishop score of 5 or more was acNeved Results: The groups were comparaNe with respect to maternal age, parity, gestat~onal age, reason for induction, and initial Bishop score. The addition of lamlnaria tents to sequential intravag~nal prostaglandin gel did not statistically =mprove the t~me to a favorable cervix (control group 12 7_+8.5 hours and study group 10.9+_7.1 hours, p=0.46) or time to delivery interval (control group 22 4-+11.2 hours and study group 23.4-+13 1 hours, p=0.79). Adding laminaria tents did not have a signihcant impact on the vaginal delivery rate with 28.0% of patients in the control group and 26.3% of patients in the study group undergoing cesarean section (p=0.90). Maternal and neonatal complications were rare in both groups. Conclusion: The addition of laminana tents to intravaginal prostag~andin ge~ dees not ~mprove the t~me to cerwca~ ripemng, the time interval to dehvery, or the cesarean section rate.

634 A RANDOMIZED TRIAL OF LOW-DOSE OXYTOCIN BEGUN

IMMEDIATELY AFTER OR IN A DELAYED MANNER AFTER INTRACERVICAL PROSTAGLANDIN E2 FOR INDUCTION

OF LABOR. F. Coleman~ L. Burksx, W. Rayburn, K. Farmerx, J.

Larsonx, G. Tumbullx, Depts. Ob/Gyn & Pharm Admin., Univ. of

Oklahoma ttealth Sc~. Ctr., Okla City, OK

OBJECTIVE: This study compared effects of low-dose oxytucin begun ~mmediately or an a delayed manner after intracervical

placement of prostaglandin E2 (PGE2) gel for the induction of labor

STUDY DESIGN= A total of 50 nonIaboring women at term with

an unfavorable cervix (Bishop score -< 4) were given a 0.5 mg dose

of PGE2 gel (Prepidil). Each was randomized either to be observed

or to recewe a low-dose of oxytocin (2mU/min, increased by 2mU/min at 30-rain intervals as necessary). After the 6 hour observation, the patient was re-examined and a low-dose of oxytoc~n was either begun or continued. RESULTS: There were no differences between the two groups for

maternal, race, parity, gestational age, predose Bishop score,

predose uterine activity, and indication for induction. Uterine

contractions became more frequent (p < .01) and were more intense

(p < .02) when oxytocin was used immediately. No uterine

hyperstimulation or abnormal fetal heart rate pattern was observed

to require discontmuatmn of the oxytocin. The percentages of cases delivering vaginally within 24, 36, and 48 hours were greater when

oxytocin was begun ~mmediately for nulliparous patients (p< .01)

and for those with predose Bishop scores of 3 or 4 (p < .01).

CONCLUSION: Low-dose oxytocin may be started immediately

after instilling intracerv~cal PGE2 without apparent added risk and

w~th an anticipated shortened time until vaginal delwery.

636 EFFECTS OF ERYTHROMYCIN ON CONTRACTILITY OF ISOLATED MYOMETRIUM FROM PREGNANT RATS Granovsky-Gdsaru S, M.D ," Dan I. M.D.* Gdsaru D, Lav=e O M.D."Aboulafia Y M.D. Diamant Z.Y.M.D." Hanani M. Ph.D.S*Department of Obstetncs and Gynecology, Shaare-Zedek Medical Center, Jerusalem "Serelin Matern~ Hospital, TeI-Aviv ~Laboratory of Experimental Physiology, Hebrew University and The Hadassah Heap, Jerusalem, Israel Objective: Erythromycin is a potent stimulant of motor activity in the stomach, apparently by acting on motilin receptors, However, in the uterus ~ effects on the mechamcal act~v~ have not been studied and only ~ts anbbiotlc prope~es have been considered in the treatment of idiopathic preterm labour The purpose of this study is to characterize m-vrtro the effect of e~/thromycin on the contractJl~ of the pregnant utedne smooth muscle. Study Design: Myometrial stnps from pregnant Sprague-Dawley rats (midgestation) were suspended m tissue baths. Isometric conta~ons were monitored by fome transducers in response to various agents used. Results: Erythromycin exposure caused a sustained decrease in spontaneous muscle activity. The effect was statistically s~gnificant (p< 01)espec=ally in the presence of oxytocin. Relaxation wes observed ~n all the variables including, frequency, amplitude, duration and contrecble force, The actions of e~/thromyc~n were concentra~on dependent. The effect (20%) reached a=gnificance at erythromycin concentrations of 10"~M, plateaued (33%) at 2xl0"~M and 5x104M An additional relaxation effect (add=tlona140%) was reached at 10~M. S=m=lar effects were observed in carbachol treated stnp~ Conclusion: We conclude that ewthromycm produces a decrease in the pregnant rat myometnal actiwty in-wtro. This effect was significant and independent of the stimulant. These data may be useful in interpreting the results of vanous cl=nical trials which use erythromycm in the prevention and treatment of preterm labour. Also, it may prowde new insights into the contrsct~le mechanisms of the uterus,


637 OXYTOCIN LABOR STIMULATION OF TWIN

GESTATIONS: EFFECTIVE AND EFFICIENT. AJ Satin,

MB Fausctt, MC Gordon, WH Barth, Jr. Dept OB/GYN, Wilford

Hall Medical Center, Lacidand AFB, TX

OBJ-~CTIVE: The null hypothesis is that oxytoein rOT) labor

stimulation of twin gestations is similar to that of singletons;

regarding dosage, time, complications, and ability to aclaieve

vaginal delivery.

STUDY DESIGN: This investigation included 110 gravi&m

receiving OT for augmentation or induction of labor. Women

with twin gestations (N=55) were matched by parity, cervical

dilatation at initiation of OT, gestational age, OT dosage regimen,

and indications for OT to singleton gravidas receiving OT.

Outcome variables included maximum dosage of OT, incidence of

hyperstimu|ation and fetal heart ~te abnormalities, time from OT

to delivery, cesareans, maternal and neonatal outcomes.

Statistical analysis included use of contingency tables, paired t

test, Wileoxon rank sum test and multiple logistic regression.

RESULTS: Women with twin pregnancies responded sim~i~ly as

singletons regarding maximum OT dosage (22 v. 18 Mu/min,

P=N.S.), time from OT to delivery (359 v. 389 rain, P=N.S.),

and successful vaginal delivery (89 v. 89%, P=N.S.). OT

stimulation of twins resulted in fewer interruptions of the iafnsion

for fetal heart rate abnormalities and hyperstimulation. (OR 0.18;

95% CI. 06-.54).

CONCLUSIONS: Twin gestation has no adverse impact on the

effectiveness or efficiency of OT labor stimulation. Iudesd, twins

exparienca fewer side effects.

639 EXPRESSION OF OXYTOCIN RECEPTOR IN AMNION, CHORION, AND CHORIONIC PLATES IN PREMATURE LABBOR: EFFECTS OF INFLAMMATION, JF Mdl*, S Keasler*, CM Salafia*, KA Starzyk*, M Verburg*. Department of Ob/Gyn, Georgetown University Medical Center, Washington, DC OBJECTIVE: To test the hypothes~s that oxytocln receptor (OTR) mRNA vanes with preterm labor w~th and w~thout acute inflammation. STUDY DESIGN: Placentas and assooated membranes were collected from 4 women dehvering by cesarean section at 31 or 32 weeks gestation, 2 who had histolog~cal choriodeodmtls (1 with spontaneous preterm labor, and one who had no labor), and two w~th no signs of ~nflammation (both with augmented spontaneous labor) The membranes and chonomc plates were d~ssected and fixed in formahn prior re paraffin embedding and sectaomng Ten m~cron secttons were used for in sltu hybridization (ISH) studies utilizing a non-radioactive, d~gox~gemn-based protocol. A &goxigemn-labeled OTR RNA probe was synthesized from a Barn HI-Apa I fragment of the human OTR cq)NA clone, subcloned into the Bluescript in wtro transcription vector. After hybridization and stnngency washing the non-specific RNA probe was d~gested with Rnase A,’rhe remaining specificalIy hybridized signal was vlsuahzed with an anti-digox~genin antibody labeled with alkaline phosphatase, utilizing NBT and Xphos as chromogens. Samples were also taken for RNA preparatmn using the aod/phenol method and analyzed by reverse transcnption-polymerase chmn reaction (RT-PCR), using primers for the OTR eDNA which spanned intromc mate~lal. RESULTS: In the two non-inflamed, augmented spontaneous preterm labor cases OTR mRNA was detected focally at h~gh levels in the ammon epithelium, chor~onic and dec~dual stroma, and the subchoriomc cytotrophoblasts. In the two cases w~th histological chonodecidmt~s no OTR message was detected in any region, though by RT-PCR OTR message was present. CONCLUSIONS: OTR levels increase during gestation, and reach their highest prior to labor. In acutely ~nflamed t~ssues, other mediators of uterine contractihty, such as cytokmes, may be equally or more ~mportant. Our cases with augmented spontaneons labor but without sagns of inflammation showed OTR mRNA expression, consistent with the concept that oxytocin may be responsible for their labor and that premature elevation m OTR may contribute to their prematunty. In the two cases with acute chonodec~duit~s, OTR mRNA levels were not detectable by ISH. In the case with spontaneous preterm labor this ~s consistent the labor bemg driven by non-oxytocm uterotonlns. In the case with no labor, the nondetectable levels of OTR mRNA ~mplies a state in which neither oxytoc~n nor inflammatory cytoklnes were at suffioent levels to ~nduce labor.

638 TUMOR NECROSIS FACTOR-ALPHA SYNTHESIS AND MMUNOREACTIVITY IN BASAL CYTOTROPHOBLAST MAY BE AN INFLAMMATION-INDEPENDENT PART OF NORMAL PARTURITION. C. M. Salafia*, J. F. M~II*, K.A. Starzyk*, M. Ossandon*, S. Keasler*, D. Splcer*, C Lester*. Permatal Research Facility, Departments of Pathology & OB/GYN, Georgetown Umverslty Medical Center, Washington DC. ’ OBJECTIVE: To examine ~f tu~aor necros~s factor-alpha (TNF-(x) immunoreactiv~ty In placenta is gestatlonal age related and affected by histologic evidence oT acute ascending Infection. STUDY DESIGN: S~xteen cases were selected which ~ncluded 8 cases dehvered 37-40 weeks and 8 dehvered between 26-33 weeks for a prmcloal ~ndlcat~on 6f oreterm labor or remature ru tare of m - P " P embranes. Four of each group had moderate hlstolog~c acute marg~natmg choriodecldmt~s, the others had no h~stologlc acute ~nflammat~on. With antibody to TNF-~ (Genzyme, 1:200), relative lmmunohistochemlcal (IHC) reactawty in frozen sections was graded as weak (.+) moderate (++), and intense (+++). Ceils assessed included syncvtiod’ophoblast (ST), cvtotrophoblast ([CT] of basal .plate~ subchorionic flbnn, and perivirlous fibr n), vascular smooth muscle ana stromal cells in choriomc plate and fetal stem and edematous and non- edematous terminal villi. In s~tu hybridization (.I.SH) and RT-PCR were oerformed on adjacent tissue sections w~th ISH utihz~ng &goxtgenln- labelled nboprolSes from a human T~F-~ eDNA, and RT-PCR using TNF-~ p.rimers. RESUETS: IHC and ISH results were essentially ~dent~cal. In t reterm cases .w~.thput ac.ute lnflammatmn, approximately 25% of basalplate CT were t+). Less than 50% of villous stromal cehs, and scattered stromal and vascular smooth muscle cells In the chorionic plate and fetal vessels were (++). ST and subchonomc CT showed scattered (++). Preterra cases with acute inflammation showed subchor on c CT ST, and chorlonlc plate cells (+++), increased both in number of positive cells and intensity comoared to non-inflamed cases. These findings, were confirmed by R’T-PCR, w~th cases of AI wxth vdlous edema showing the most intense bands, n term non-inflamed cases >75% of basal CT’and ~ntervdlous CT were (+++1; most chormmc pl~ate, wllous stroma, and vascular smooth muscle cells, (but few subchonomc CT) were (÷+). In inflamed term cases no difference in TNF-~ reactiwty was seen. TNF-~ reactivity was not ~i’ifferent in edematous versus non-edematous vdh. CONCLUSIONS: TNF-ot synthesis and immunoreactlwty m different types of placental cells generally increases in the third trimester, and especially xn basal CT Acute inflammation increases TNF-~ reactivity ~n ST, subchormnic CT but not in basal CT. TNF-~, an essentml participant ~n allograft rejection, may contribute to normal parturition, but may not play a role in parturition initiated by preterm acute ascending infection.

640 INTRAAMNIOTIC INFECTION: ITS EFFECT ON DURATION OF LABOR. ~, K.A Aqua*. Department of Ob/Gyn, Ar;zona Health Sciences Center, & Tucson Pefinatal Services, Tucson, AZ OBJECTIVE: We hypothesized that women developing intrapartum =ntraamniot=c infections (IAI) do not have any prolongation of labor when compared to matched controls. STUDY DESIGN: We retrospectively reviewed (1989-1994) all patients diagnosed with IAI who delivered vaginally at University Medice~ Center’. IAI was diagnosed by fever (>38 OC), abdominal tenderness, fetal or matamal tachycardia and leukocytosis. Control patients were matched for age, parity, estimated gestational age at delivery, delivery method, use of pitocin and use of epidurel anesthesia. Results were compared using student t-test and Fishers exact test. RESULTS: Multiple demographic var~b/es were found to be statistically similar ~n both groups including the incidence of tobacco use.

IAI C(mtrol p

n:131 I*/o) n=131

IROM (hrs) 16 -+ 14 7 _+ 5 .0001 Stage l(hr) 10_+9 6_+4 .0001

Stage 2 (min) 55_+ 43 45 +__37 .046

Stage 3 (min) 9_+ 7 7 + 4 .0008 Birth weight (gms) 3397_+ 440 3381 _+ 457 NS

Blood loss (co) 428 _+ 213 349_+ 150 .0006 Apgar 1_<6 25(19) 9(/) .0~5

Apgar 5_<6 6 (5) 0 .03

NICU admission 22161 2112! .0001

IROM: Intrapertum rupture of membranes, blS: blot significant

CONCLUSION: In the presence of IAI all stages of labor are signi~cantiy lengthened.

484 SPO Abstracts January 1996 Am .]" Obstet Gynecol

641 A PROMOTER MUTATION IN THE TUMOR NECROSIS FACTOR ~ GENE IS NOT ASSOCIATED WITH RECURRENT MISCARRIAGE. D. Dizon-Townson1, H. Majorix, D.W. Branch1, K. Ward1,2. Depts. of Ob-Gyn1 and Human Genetics2, Univ. of Utah School of Medicine, Salt Lake City, UT. OBJECTIVE: The tumor necrosis factor ~ (TNF o~) gene is expressed in first trimester decidua and may be necessary for normal implantation and placental development. We hypothesized that mutations in the TNF (x gene might be associated with unexplained recurrent abortion. The T2 mutahon in the TNF c~ gene promoter affects TNF o~ transcnption and is in linkage disequilibrium w~th several human major histocompatibility antigen alleles. STUDY DESIGN: DNA was extracted from whole blood of 47 couples with idiopathic recurrent miscarriage and 41 couples with a history of proven fertility (8 or more live births). Polymerase chain reaction (PCR) was used to amplify the promoter region of the TNF o~ gene. The resulting PCR product was subjected to enzymatic digeshon with Ncol for polymorphism detection. Results were analyzed with a ;(2 contingency table. R E S U LT S: TNF ~ Allele Frequencies

T1 T2

RAB Couples (n=47) 0.86 0.14

Fertile Couples (n=41) 0.82 0. f 8

Published Controls 0.84 0.16

No significant differences found in allele frequency between idiopathic recurrent miscarriage couples and fertile couples. CONCLUSIONS: This polymorphism in the TNF o~ gene is not associated with Idiopathic recurrent miscarriage and would not serve as a useful marker during the evaluation of these patients.

643 THE EFFECT OF LABOUR ON MATERNAL CEREBRAL BLOOO FLOU NENOOY- NANICS. KeJth I~ittioms, Susan Wilson, B.C. U~n’s H~pltat~ Div.of Hater~t-Fetat ~ici~, Van.,B.C.Ca~. V6H ~. OBJECTIVE: To dete~f~ the i~ct of c~tract1~ a~ s~ stage ~shing on cat.rat bto~ ft~ vet~ity essess~ [re.craniaL Do~ter to assess the deveto~nt of cetera[ vasos~ ~rf~ the ta~ur a~ delivery. STaY DESIGN: 15 ~t~s{ve ~t~ents in esrty [a~r had ~ternat ~i~[e cerebra[ bto~ fto~ varsity assess~ c~t1~- ~y ~i~h tr~cron~a~ 0o~ter. Asses~nt ~as ~rfo~ ~r-

~ the ~ak of a c~tracti~, ~r~ ~he tr~h of a c~tract* i~ a~ duri~ ~sh~ in the s~o~ stage. ~e ass~s~ systolic, d~astotlc, man cer~ra[ b[~ fto~ vat.try (cm/s~) ~tsatitity i~ex over f~r c~tracti~s. ~e ~ a~tysis of varia~e to detect a significant differe~e in vet.[ties at these three stages. RE~LTS: Nean ~ter~t cer~ra[ bt~ fto~ vet~ity felt s~gnif~cantty (P<.~) in ~ti~ts ~ring s~ staNe ta~r ~ ~shi~. B[~ F~sure signific~tty i~ress~ (P<.05) at the ~ak of a �~tr~t~ a~ ~r~ s~ stage ~h~. ~CLUSZONS: Ni~te cap.rat vessels vas~itat~ ~ri~ sec~

staNe ~shing as acs~s~ ~ tr~cranfat D~ter. This relative vas~itatati~ ~n the cetera[ vessels c~[d ~ a result of changes in car~ di~i~

~ Neur~[ factors. Pushi~ ~n the sec~ StaNe uith the vatsatva ~er is ~t riskier for the ~tt~t for ~vet~- {~ cerebra[ vasos~ than c~tracti~s in the f~rst stage.

Tr~h Peek of 1st Stage 2~ Stage

Resn RC~V ~sec 56~ 52~11

Pulsatitity l~ex .80~3 .~.5 .~.22

Mean Arterial 7~11"* 97~16" Pressure

*(P<.05) vs TR a~ PC **(P<.015) vs PC

642 644 MATERNAL HEMODYNANIC CHANGES DURING LABOUR. Keith l~illial~.., Susan t~iLson# g.C.~omen’s HospitaL, Div. of MaternaL-FetaL Medicine, Van. B.C., Canada. V6H 3VS. OgJECTIVE: A 31X increase in cardiac output during tabour is reported b~t he~xJynamics of secor~J staNe during pushing or during the peak of contraction are ~. ~e investigated the effects

of uterine contractions and saco~J staNe pushing on Left ve~ttricutar function during ~orma[ l~rturition. STUOY DESIGN: 5 normotensive patients in labour were assessed, in emi-rect~bent position with non-invasive hemo~ic mo~itorin~ usir~l a thoracic electrical impedance technique. Indices assessed included: heart rate, left v~ntricular end diastolic volUm~ (LVEDV)# mean arterial pressure, cardiac output. ALl pormlmtera were assessed l~fore contraction, at the peak of a contraction end

during pushing in the aecor~ stage over the �o~Jrse of f~Jr contractione erd eta average value for each pnrmter d~finad. A~slysi$ of variance was used to assess a significant differer~e. RE~JI.TS: Maternal heart rate erd ~ean arterial pressure role during peak of contraction and during puehing. This was offset by

¯ decrease in stroke voltl~e end sul~equentLy, there was no eignificant change in cardiac output. GO~CLUSIONS: Our results differ from previous reported incresen in cardiac output and my b~ related to the se~i-recuml~t pesition of our petiente, ale fot~d no differences in 2nd stage coq~ared to pesk of contract|on$. Second stage pu~hlng wlth the vetes[ve maneuver doe~ not expese the l~rtur~t to tony further

hemodynamic stress than 1st stage contractions. Trough(TR) Contraction

1st Stage (PC)

Cardiac Output 10.+.2.3 10.6:1:2.6

Stroke Vol ~e 123:L5"

Heart Rate 76+?2* 91±1~,.8

Mean Arterie[ ~ .6-+4,. I’~ 80.2±A. 5

Pushing 2nd Stage

10.5+~.8

113.6~8.9

92±17.7

LVEDV 161 .I~6.2 156+19 I~.8~2.9

*(P(.05) versus - PC and PU


645 OUTCOME OF 494 SINGLETON BREECH DELIVERIES IN A TERTIARY

CENTER. Y. Danielx, G. Failx, M.R, Peyserx, J B Lessingx, AJ. Jaffax, I Gullx, M. ~ M.P. Davidx, D Pausnerx, M.J. Kuptarminc Depts

Ob/Gyn "A" and "B", Serlin Maternity Hospital, Saokler School of Medicine,

Tel Aviv University, Israel. OBJECTIVE: To compare the neonatal outcome of term breech ~nfanta

delivered vaginally with those delivered by cesarean eeotion. STUDY DESIGN: 496 consecutive singleton breech detiveriee between 1992

and 1994 were studied Critana for a vaginal trml of labor included a frank or

complete breech presentation with e~tJmated fetal weight of 2000-3800 g and

no pdor history of cesarean delivery (group A, n=304). Patients who did not meet these cdtada were delivered by cesarean ssct~on without a trial of labor

(group B, n=192) Two obstetricians end a pediatrician were present at all

deliveries.

RESULTS: In group A, 226 (75%) patients del~verad vaginally Ninety-four of

155 (61%) null~parous patients and 132 of 149 (88%) multiparous patients delivered vagmally There were no stillbirths or neonatal deaths, Three

neonates ~n group A and 4 neonates in group B had major anomalies The incidences of 5-m~nuta Apgar score ¯ 7, birth trauma and ICU adm~ssbns

are Ilstad in the Table and were similar in groups A and B. The incidences of

these neonatal morbidities were similar in the nulliparous and the multiparous patienta who delivered vaginally.

A (n=304) A Va,q (n=226) B (n=192) 1-min Apger <7 5.6% 6.2% 0%* 5-rain Apgar <7 O 3% O 4% 0% Birth trauma 1.6%1’ 2.2%1 1%1 ]CU admissions 3.6% 4% 3.1% *p ¯ 0.001 compared to group A and A Vag. i’3 chepalohematoma, 1 clavicular fracture, 1 tongue laceration, tl erb pa~sy, 1 femur fracture

CONCLUSIONS: With prudent selection of candidates for term vaginal

breech delivery, 75% of all patients and 60% of nulliparous patlenta can

safely deliver vaginally.

647 A CASE CONTROL STUDY OF UTERINE RUPTURE DURING PREGNANCY. CW E~emto* JC Smul{=an, SE Gray, WE Sccrza, UMDNJ-Robert Wood Johnson Med=cal SchoolJSt Peter’s Medical Center New Brunswick, NJ OBJECTIVE To evaluate factors associated with utenne rupture duneg pregnancy. STUDY DESIGN’ Records of pahents having utenne rupture (UR) between 6/92 and 6/95 were reviewed A control group was selected by matching laboeng UR patients wdh the next

2 pabents having successful vaginal delivery after cesarean secbon (SVBAC) All pahent records were analyzed for demegraph~cs, Nstoncal and ehmcal vanaNes p~tent~atly affecting the risk for UR Patients > 36 wks gestabonal age (GA) were compared to controls Pahents < 36 wks with preterm rupture were descnbed separately Analyses were made using contingency tables, parametnc and nonparametnc tests where appropriate. Signdlcance was set at p<0 05 0ndlcated w{th *) RESULTS Seventeen patients were =dentlfied with UR Three patients (18%) were <36 wks one of which had preterm labor There were 14 patients >36 wks which were in active labor. There were 30 SVABC controls All 44 analyze~ patients had a prewous low segment

transverse cesarean section (C/S) (results given In median and range) _> 36 wks Control

# of prewous C/S 1(0-2) 1(1-2)

GA at f~rst C/S 40(3%42) 39!33-42) Stahon on Admission * -3(-5 - -2) -3(-3 - 1)

Induction 7 9 Oxytocm use 9 17

Max oxytocm dcee(mU)* 10(0-14) 7(4-34) Prusteglandln gel 5 6 First stage of labor(ram)* 525(170-1620) 395(90-840) Secondstage(m~n) 580(280-1620) 420(179-980) Nonreassurmg FHR* 10 2

Umbilical Art PH* 7 12(6 7-7 43) 7 28(7 02-7 39) in UR patients < 36 wks, one pahent had low segment transverse C/S (wdh previous myomectomy), one had a classical C~S, and one had a previous T incision This group had a median gestatlonal age of 34 wks(range 29-34 wks) F~fty percent of UR cases were suspected and 59% of cases with UR had extension beyond the previous scar Duratton of p~toc~n was longer in patients w~th UR Abdominal pan and abnormal shape of abdomen were common In patients with uterine rupture Birthwelght,mecemum staned ftuld, vegmal bleeding, chonoamn~onltls and bloody urine were not significantly ddferent between groups Two of 14(11 7%) of patients with UR had bladder damage, CONCLUSIONS 1)Risk factors for UR were Ngh stahon on admission and greater doses of oxytocln in the setting of prolongued labor 2)Nonreasaunng FHR tracing, abnormal abdominal pain, abnormal shape of abdomen were all clinical identifiers of UR

646 THE EFFECT OF HISPANIC ETHNICITY ON INTRAPARTUM MANAGEMENT OF LABOR, DELIVERY AND ANALGESIA CW Bemto,, CF Thayer~, MF Lake~, RA Knuppel, AM Vmtz=leos UMDNJ-Robert Wood Johnson Medical School/St. Peter’s Med=cal Center, New Brunswick, N J. BACKGROUND: H=spanic ethnic=ty has been shown to be associated with insufficient analgesia in emergency med=c=ne departments Hispanics w=th =sotated long bone fractures were twice as t=kety not to receive pan medication as nonh=spamc whites. OBJECTIVE: The purpose of th=s study was to determine =f H~spanic ethnicity has an effect on management of tabor, method of delivery, procedures dunng labor and analgesia during labor STUDY DESIGN" A retrospectwe case control study of 527 singleton term pregnanc=es del=vered =n !990 at our hospital. The study group cons=sted of 194 H~spamc patients. The control group consisted of 393 pat=eels chosen using the next sequenbal med=cal record number. The following data were analyzed’ age, gravid~ty, panty, length of stay, method of dehvery, ~nd=cabon for cesarean section, procedures dcnng labor and analgesia during labor Star,shoal analyses were performed developing odds rabos (OR) and 95% conbdence intervals (CI). RESULTS: Mean age, gravtd~ty, panty and length of stay were not s~gndlcantly different between the two groups No d=fference was found between dehvery methods Procedures during labor including pitoc=n augmentabon, inducbon of labor, scalp electrode placement, amniotomy, and placement of intrauterine pressure catheter were stud~ed and no d~fferences were found The following table descnbes the use of analgesla(AG)in the study populabon

Companson of H=span=cs vs Nonhispan~c wNtes OR (95%,CI) Any AG vs none 0 99 (0.79-1 25) Epldural AG vs none* 0 28 (0.17-0 46) Ep~dural AG vs. Intravenous AG* 0 29 (015-0 57)

* md~cates statistical signi%ance CONCLUSION: H~span~c ethmcity s~gnff~cantly associated w~th less use of analgesia in labor when compared to nonhispan~c whites Hispanic ethnicity was not associated w~th any method of delivery or intrapartom procedures

648 INTRAHEPATIC CHOLESTASIS OF PREGNANCY (ICP). A MODEL OF PREMATURE ACTIVATION OF THE OXYTOCIN BIOACTION? Germmn

~_~__,, Kate Sx, Gormaz G*, Vlllarroel Lx and Valenzuela GJ Perinat Med Unit, Dept Ob/Gyn, PU Catohca de Chde School of Medicine and SBCMC, San

Bemardino, Calaforma OI~IECTIVE: ICP ~s a prevalent (3%) disease in Chile. Spontaneous preterm

labor is one of the most frequent comphcations, suggesting that an abnormally high uterine activity could be causally related We tested the hypothesis that a premature antivatmn in the oxytocin bioactmn pathways plays a role in the appearance of preterm labor.

STUDY DESIGN: We performed (1) a longitudinal study of the mean 24 h uterine contraction (uc) frequency ~n patients w~th early onset ICP [<32 weeks; n=l 5] and controls [n=16] Continuous external recording of utenne activity

was made at two weeks’ intervals from 28 weeks’ gestation until delivery; (2) an oxytocin challenge test at 38 weeks’ gestation. We calculated the m~nimal amount of oxytocm reqmred to ehc~t a frequency of 4 uc/10 min (minimum

effectwe dose; MED); patients were pmred for gestational age and Bishop score; and (3) a eumulatwe dose-response cur~e to oxytocin [10-t~ to 11~ M] tn myometrlal strips under "in wire" contht~ons from ICP (n=5) and control (n=7) patients at 38 weeks RESULTS: One third (5/15) of the ICP and none of the controls patients

presented preterm labor. No significant difference m the evolutmn of the uc frequency throughout gestation between control and ICP groups (term or preterm) ~vas observed. However, a s~gmficantly lower MED of oxytoc~n was required in ICP (n~5) than control (n=7) patients [1310±583 vs 3570:~749 mU; Mean ± SEM p <0.03]. In addition a h~gher basal and contractile response to oxytocm was observed in myometrial strtps of ICP patients than controls CONCLUSIONS: Our results confirm the association between 1CP and preterm labor However the mean 24 h utenne activity profiles did not differ from normal pregnancies The lower MED and the higher "m vitro" response

to oxytocin observed in ICP patients support the hypothes~s that a premature activation m the oxytocin b~oaction pathways occurs ~n ICP Supported by DIUC 94/06E


649 1N’IqkAHEPA~IIC CHOLESTAS1S OF PREGNANCY (ICP). 1S THERE A ROLE

FOR PLATELET-ACTIVATING FACTOR (PAF)? Germain AM; Kato S~, Zhan

BX, Johnston JmX Dept Ob/Gyn, P U Catobca de Chile School of Methclne,

Santiago, Chlle and Depts Biochem and Ob/Gyn, U Texas SWMC, Dallas, Texas

OBJECTIVE: Spontaneous preterm labor is one of the most frequent

comphcatlons of ICP. PAF is a potent uterotomn and prmnflammatory agents

described. PAF acetllhydrolase (PAF-AH) the enzyme responsible for PAF

reactivation, is produced by macrophageSo The decidua parietabs Js believed to be

a source of PAF-AH The plasma PAF-AH activity decreases progressively during

the third trimester of pregnancy and increases in the puerpenum in response to

hormonal changes In this study, the hypothesis that a premature activation in the

PAF bioaction pathway plays a role in the ICP related complications was examined

STUDY DES.IGN: We performed (1) a longitudinal study of the plasma PAF-AH

specific actavity (SA) at 2 week intervals in patients with early onset ICP (<32

weeks, n=7) as well as controls (CT; n=6); (2) a cross-sectional study of plasma SA

uf PAF-AH at 3g weeks gestation in 23 ICP and 26 CT patients; and (3)

measurements ofmyometrial content of lmmunoreactive PAF (R1A) from pattents

with 1CP (n=6) and CT (n-10) at term RESULTS: A higher SA of PAF-AH was found from 34 weeks’ gestation:

SA PAF÷AH IN ICP AND CT PATIENTS (nmol x mln-~ x ml-~ plasma)

Gest age (wk) 32-336/7 34-356/7 36.37~7

Control (n-6) 23.4±1.5 20 5±1.3 21 4±1 1

ICP (n=7) 32 4~:4.2# 37.2aA.2’ 36 9±3.9*

Mean ¯ SEM, * = p <0.004, # = p <0.08 vs control group.

A similar findmg was observed in a cross sectional study. The plasma PAF-AH

specific activity in patients with ICP was 42.64-3.2 vs 26.9~.1 4 nmol x rain-l x ml-~

plasma for CT (p<0.0001). The myometrial PAF content was higher in ICP than CT

patients but was not statistical significance (4094208 vs 1834-20 fmol x mg-t

protein; p=0.14)

CONCLUSIONS: Based on these findings, tt is suggested that m ICP the PAF

concentration is increased which in turn stamulates PAF-AH secretion and PAF

bioaction may play a role in the compbcatinns of this disease Supported by DIUC

94/06E.

651 POSTPARTUM CHILLS: A MATERNAL HEAT LOSS PHENOMENON? B. Schumacher,

C. Karmelx, G. Saade, ft. Olson. K. Mmse, Jr. Dept OB/GYN, Baylor College of Medicine,

Houston, TX.

OBJECTIVE: To determine the incidence of postpartum chdls and their relationship to the

degree of maternal temperature change.

STUDY DESION: Using the Diatek® 9000 Insta-tempTM tympanic thermometer,

temperatures were measured on 44 healthy, laboring wome~ who had normal term vaginal

deliveries. Recordings were made ol~ admission, ¯ actwe (~r, at ddivep/and every ~5

minutes thereafter for one hour. The maximal temperature drop (AT) was calculated as the

difference between the h~ghest and the lowest maternal temperatures. AT was positive or

negative if the lowest temperature recorded occurred after or prior to the h~ghest

temperature, respectk’ely. Statistical analysis: Knlmegerov.Smimov, Student’s t,

Mann-Whitney, Pearson’s corre]atien, Fisher-Exact and Chvsquared tests as appropriate.

P < 0.05 used for statistical significance.

RESULTS: Of the 44. women, 19 141%) experienced postpartum chills, onset 1 - 30

,~medien B ram) after delivery, lasting 2 to 60 mM Imadien 15 rain), There was no s~gnfficant

difference in maternal age, weight, parity, hematocrit, ep~dural anesthesia, gestutemal age,

birthweight, placental weight or fetal sex between the shiverers and nun-shiverers. There

was a stgnificaetly larger drop in temperature in patients who shivered as compared to

those who did not (1.4 _+ 0.6 vs 0 3 _+ 0.5 OF; P < 0.03). There was no correlation

between combined

infant and placental

weight and AT.

Figure: Time course of

the mean temperature

in the two groups.

CONCLUSIONS: Large

maximal temperature

drops /end to

postpartum chills which

may serve to restore

body temperature.

650 ISOLATED DISSEMINATED INTRAVASCULAR COAGULATION

AND AMNIOTIC FLUID EMBOLISM. TF Port~ SL Clark, GA Dildy,

GDV Hankins. Department of OB/GYN, University of Utah, Salt Lake City,

UT, UTMD Galveston, Galves~n, TX.

OBJECTIVE: To examine clinical factors surrounding an uncommon

amniotic fiuld embolism variant involving disseminated intravaseular

coagulation (DIC) in the peripartum period.

STUDY DESIGN: A retrospective review of medical records was carried out

a~ part of a national registry of amniotie fluid embolism patients. Cases were

specifically identified where isolaOul acute coagalopathy developed without

antecedent hypotonsion, hypoxia or evidence of any other events or disease

proenss~ including placeatal abruption.

I~.~OLTS: Fight such pati~nt~ were identified. The mean (+/-SD) age of

patients was 33.5 (+/- 5.6) years; the mean (+/-SD) geatational age at

delivery was 37.7 (+/- 2.9) weeks. One patient had triplets, two had twins,

and 5 had singaltons. Five patients were delivered by cesarean so~tlon (l for

previa, 1 for triple.as, 1 as repeat, 1 for twins, and I for fetal distress), 2 by

a~isted vaginal ddivery, and 1 by ~ontancous vaginal delivery. Median

(range) Apgar scores were 8 (2 - 9) and 9 (5 - 10) for one and five minutes,

respectively. Pitoein was u~,.d in 3 patients (all of whom had vaginal

deliveries). Aent¢ hemorrhage occurred in the postpartum period in 7

patients, end in the intrapartum period in 1 patient. Uterine atony was noted

ia 51g pattedats; in no cas~ could the volume of blood lost account for the

development of a coagulopathy. The mean (+/-SD) time from delivery to

bleeding was 42.7 (+/- 32.9) minutes. The mean (+/-SD) time from

delivery to confirmation of coagulopathy was 210.3 (+/- 182.3) minutes. Six

of eight patients (75 %) exsanguine~ved de.spite appropriate medical management

and blood component regla~ement. Fetal elements were identified at autopsy

in 314 patients in whom autopsies were performed.

CONCLUSIONS: Isolated, fatal DIC in pregnant patients appears to

represent aformefiuste of amniotic fluid embolism, with mortality rates

similar to that seen in the more classic syndrome.

652 1S ANTENATAL MAGNESIUM AND CORTICOSTEROID EXPOSURE PROTECTIVE AGAINST THE DEVELOPMENT OF INTRAVENTRICULAR HEMORRHAGE IN INFANTS OF WOMEN WITH PREECLAMPSIA? B. Brost. M. Menard, R. Newman, Dept of Ob/Gyn, Methcal Univ. of S. Carohna, Charleston, SC. OBJECTIVE: Recent reports have noted a decrease in intraventricular hemorrhage (IVH) in very-low birth weight (VLBW < 1500 gin) neonates after maternal administration of magnesium sulfate (MG) or corticosterolds (STER). Enhanced cerebral maturation has also been described in pregnancies complicated by preeclampsia. This study was performed to determine if maternally administered MG or antenatal STER decrease the incidence of IVH in VLBW infants with premature delivery secondary to preeclampsia. STUDY DESIGN: The Methcal Umvers~ty perinatal database was accessed to identtfy all patients wath the thagnosis of preeclampsia (1/89 -12/93) with subsequent delivery of a hveborn VLBW infant. Patients were classified as to antenatal exposure to MG and/or STER. All infants underwent cranial ultrasound for the clinical suspicion of IVH or immediately prior to discharge from the neonatal intensive care unit. RESULTS: Of the 162 VLBW neonates of preeclamptic mothers, 20 (12.3%) had Grade I-II IVH and 16(9.9%) had Grade Ill-IV IVH. The incidence and severity of IVH was stmilar in each therapy group. The relative risk of IVH with therapy was calculated compared to no therapy

Tot IVH RR(95%CD III-IV IVH RR(95%CI/

MG/STER 12/59 0.78(.33-1.831 6/59 1.17(.25-5.38) MG only 16/71 0.86(.38-1.95) 7/71 1.13(.25-5.08) STER only 2/9 1A3(.25-5.081 1/9 1.28(.13-12.41) Neither 6/23 1 2/23 1 Each therapy group was similar with respect to maternal smoking status, PROM, mode of delivery, fetal sex, bmhweight, gestational age, Apgar at 5 min, RDS, ventilator use, pneumothorax, and sepsis. CONCLUSIONS: In this cohort of VLBW neonates born to mothers with preeclampsia, maternal antenatal admimstration of MG and STER was not associated with a decrease in the incidence of GMH-IVH.

Volume 174, Nmnber 1, Part 2 SPO Abstracts 487

Am J Obstet Gynecol

653 A NOVEL TREATMENT OF FETAL LUNG IMMATURITY. EV Cosmi, R LaTorre, R Di Iorio, MM Anceschb 2nd Inst Ob/Gyn, Umv.

Sap~enzs", Rome, Italy.

OBJECTIVE: To study whether intraammotic surfsctant Is more effective

than post-delivery intratracheal surfactant for treatment of RDS.

STUDY DESIGN: Six preterm pregnancies (28 to 32 wks) with lUGR and

ARED and non-reactive CTG were studied. After consent,

emniocentesis was performed c~ose to the mouth of the fetus under

ultrasound (US) guidance end AF samples assayed with e rapid test for

FLM followed by other tests; the needle was left In place. A bolus of

240 mg of A was given over 10 min, followed by 0.8 mg/Kg/h. Five to

16 minutes after A, FBMa began at a rate of lO-12/min. Only when

sustained FBMs’ were obtained, naturat SS 180 mg In 1 ml, Curosurf)

was instilled through the amnlocentesls needle. The fetuses were

delivered within 4 hrs. by CS because of deteriorating clinical

conditions.

RESULTS: Birth weight ranged between 1,O90 and 1,7OO g. Time-to

sustained respiration was within 60 sac. There was no need to repeat

SS administration. Clinical outcome of the neonates was uneventful. CONCLUSIONS: This is the first report of successful IA administration

of SS in the human. Given the fact that 1) SS was seen by US in the

upper airways; 2) no extra doses of SS were required; and 3) the clinical

outcome of the newborns was uneventful, we behave this new

therapeutic approach, which requires only one third of the usual dose

used in the neonate, has great potential for becoming a proper option for

the antenatal prevention of RDS (Supported by CNR, Italy).

655 TREATMENT OF TWIN-T0-TWIN-TRANSFUSION SYNDROME. IS THE THERAPEUTIC

AMNIOCENTESIS STILL A GOOD MANAGEMENT

OPTION?

G. Marl.x A. Abuhamad, B. Verpairojkitx, D. Jones, K. Gomez, R.

Bahado-Slngh, Robert Soper, J.A. Copel. Yale University, New Haven, CT, Eastern Virginta Medical School, Norfolk, VA. OBJECTIVE: The optimal treatment of the pregnancy complicated by twin-to-twia-transfusion syndrome (TTTS) is

controversial. Therapeutic amniocentesis has been proposed with survival rates ranging between 18% and 83%. A survival rate of

53% has been reported with laser-therapy. The aim of this study was to report our experience with aggressive therapeutic amniocentesis in the treatment of twin-to-twin-transfusion syndrome. STUDY DESIGN: Over the past five years, thirty-eight twin gestations were referred to our Ultrasound Units with diagnosis of

TTTS. The diagnos~s was confirmed in 24 sets of twins. Treatment ~ncluded one or more amniocentesis in all pregnancies. RESULTS: Gestational age at diagnosxs ranged from 15 to 29

weeks (mean + SD: 21.2 _+ 3.5 weeks); whereas it was between 25.3 and 37 weeks at delivery (mean + SD: 31.5 + 3.3 weeks). The mean number of amniocentesis per pregnancy was 4.5 + 4.1 with a range

between 1 and 15. Twenty-nine neonates survived (29/46 = 63%)

In thirteen sets, both twins survived without complications (54%).

In 3 sets, one twin survived. There were 8 fetal demises and 7 neonatal deaths. One patient underwent electtve 2rid trimester

abortion. One patient miscarriaged following a fetal demise. CONCLUSION: Our results support previous studies suggesting that therapeutic amniocentesis may be useful in the treatment of TTTS. This information may assist in counseling pregnanczes complicated with twin-to-twin-transfusion syndrome.

654 REPEATED COURSES OF STEROIDS TO ENHANCE FETAL LUNG MATURITY IN PRETERM PROM DO NOT INCREASE THE RISK OF HISTOLOGIC CHORIOAMNIONITIS. C M Salafia* A Gh~dml, V.K Mlmor* Dept of Ob/Gyn & Pathology, Perinatal Research Facility, Georgetown Unlversaty Medical Center, Washington, DC, UCONN Medical Center, Farmmgton, CT OBJECTIVE: Antenatal admmtstratmn of steroids (betamethasone 12 mg I M twice q 24 hours) enhances fetal lung maturation and reduces neonatal morbidity In preterm premature rupture of membranes (PROM) However, the risks of repeated adm~nistratmn of steroads 7 days after the ~mtlal course are unknown We evaluated the prevalence of hlstologlc evidence of chorloamnlonltls in patients receiving single vs multiple courses of steroids for fetal lung maturation. STUDY DESIGN: Consecutive cases of preterm PROM at < 32 weeks gestation delivered between 7/88 and 3/94 were recorded Obstetric and clinical reformation was obtained from women who did not recetve steroids for fetal lung maturity (n=55), those who received a single course (n=47), and those w~th > 2 courses of stermds (n-89) Placental pathology examination was performed after dehvery and h~stolog~c evidence of severe (grades 3-4) acute amnlonltlS was determaned based on semi-quantltatxve score on a scale of 0-4 as previously described. Potential confounding variables considered were presence ofohgohydrammos (vertical pocket of ammotic fluid < 1 cm at ultrasound), labor prior to delivery, gestatmnal age at dehvery, and mode of dehvery Prophylactic antzbmt~cs were administered accord=ng to ACOG guldehnes. Statistical analys~s included ANOVA after log-transformation and Chi-square, where apphcable RESULTS: % (No) or mean + SD

Variables No steroids Single course Multiple courses p value

Ammotlc fluid _<1 cm 37% (t 6/43) 45% (19/42) 41% (34/82) NS

Labor prior to del 61% (33/55) 53% (25/47) 63% (56/89) NS

Gest age at del (days) 193+17 198+17 199+16 NS

Cesarean section 67% (37155) 66% (31147) 64% (57199) NS

Severe ammonltls 31% (17/55) 15% (7/47) 10% (9/89) 0 02

CONCLUSIONS We found no evidence that in preterm PROM at < 32 weeks repeated courses of steroids are associated with a h~gher prevalence of severe acute hlstologlc amnlonltlS.

656 ENDOSCOPIC REPAIR OF FETAL SPINA BIFIDA IN

UTERO J Bruner, N. Tubpan~, W. Richard~ Depts. Ob/Gyn, Neurosurgery, and General S~gety, Vanderbilt University, Nashville, TN. OBJECTIVE: Evaluate the sat’ety and efficacy of minimally mvas~ve surgery for the repair of open spma bifida in utero.

STUDY DESIGN: Women ,~th a pregnancy eomphcated by fetal open myelomenmgocele in the mad-second trimester are offered an experimental procedure designed to prevent ongoing exposure of the

spinal cord to armfiohc fluid. The procedure consists of maternal laparotomy w~th exposure of the gravad uterus. Endoscopic ports are placed for camera and operattng instruments. The ammotie fluid ~s removed and replaced wtth carbon dioxide. The fetus is positioned,

and a maternal spht-thickness skin graft is placed over the exposed neural placode The skin graft and a covering of SurgicelTM are attached with fibrin glue prepared from autologous eryopreeipitate. RESULTS: Two fetuses with open lumbar myelomeningocele

underwent endoscopic palhat~on of a spinal lesion at 22 and 23 weeks’ gestation One infant, dehvered by planned cesarean section at 35 weeks’ gestation after demonstration of fetal lung matunty, ~s approaching 1 year of age The other fetus dehvered one week postoperatively after development of ammomtis, and d~ed m the delivery room from extreme prematurity. No Lmmediate surgical comphcatmns occurred CONCLUSIONS: Minimally ~nvas~ve fetal surgery appears to constitute a feasible approach to nonlethal fetal malformations that result in progresswe organ damage


657 OVINE FETAL BLOOD FLOW RESPONSES TO NITRIC OXIDE DEFICIENCY AND NITRIC OXIDE DONOR THERAPY. W Giles,x M Read,x J Falconer,x and S O’Callaghan x Dept Reprod

Med Umv of Newcastle, NSW, AUSTRALIA OBJECTIVE: Pilot stu&es by our group have demonstrated the

associat~on of abnonnal placental fanctian wtth a decrease m the activity of the enzyme Nitric Oxide Synthase (NOS). We undertook to see the effects on fetal blood flow in a ovme fetal model where NOS acttvtty had bean blocked by N-mtro L-argmine (NOLA) and the eftL, cts

of subsequent treatment vath a NO donating drug Glyceryl Truutate (GTN) STUDY DESIGN: Ten Comedale cross ewes w~th dated pregnancies mlderwent surgery at 125 to 130 days. Maternal, fetal artenal and

venous catheters and an ammohc catheter were inserted and the ewe allowed 7 days to recover Five control fetuses received a 40 inmute mfumon of NOLA (30mg bolus then 2 4mg/k~/mm) followed by a 20

minute mfumea of GTN (10 ~ estmmted fetal weight). Five

fetuses received NOLA the inactive D isomer of NOLA (NODA) followed by GTN Fetal and maternal blood gases, blood pressure, and fetal regmnal blood flow (by nucrospheres) were recorded at 3 occasions m each ammal Stalastlcal analysis was by ANOVA.

RESULTS: There were no changes m fetal and maternal blood gases Slgmficant reducttons m fetal regtonal blood flow were noted m the

fetal placenta (55%, P=-005), heart (21%, P=,017), large gut (63%, P=.002), lung (87%, P<.001), kzdney (55%, P= 001) and spleen (72%, P=- 006) m the 5 NOLA treated fetuses GTN significantly reversed tl~ NOLA effect m these fetal organs In the 5 NODA treated fetuses GTN caused a trend towards increased the regional perfuszca m the fetal

heart (P= 05) CONCLUSIONS: GTN reversed the effects of NO deficiency m the fetal ctrculatmn and also showed a trend towards Unproved regional blood flow m the fetal heart m control animals.

659 MANAGEMENT AND OUTCOME OF ANTI-M

ISOIMMUNIZATION. d,J~.y!ex, R Rosex, A. DeYoungx, M. Kennedyx,

AWaheedx, R O’Shaughnessy Dept of Ob/Gyn, Ohio State University Medical Center, Columbus, OH. OBJECTIVE: Anti-M antibody has been associated with severe

e~Ahrob astos~s fetahs and is bem~ detected .w~th increasin~l frequency Our goal was to evaluate the chnlcal use oT maternal antibody titers and to determine the frequency and severRy of hemolytic disease of the newbom among pa~ents w~th anti-M ~soimmuniza~on

STUDY DESIGN: A retrospecbve chart review was oonducte,d en a/I patients diagnosed w~th an ant~-M antibody and followe<~ ~n our

so mmun zat~on Program between September, 1969 and August 1995 We analyzed ~nd=rect antiglobuhn (Coombs’) titers, amn=ot c fluid optical density, d~rect ant~globuhn, M antigen status, antepartum course, and pennataJ outcomes RESULTS: N=nety patients had 115 pregnanc=es affected by an ant~-M antibody Complete records and outcomes were ava=lable for 90 (78%) of these cases T~tratable antibody was present in 84 (73%) pregnancies as follows: <1:8 (n=79), 1.16 (n=l), 1’64 (n=2), 1.128 (n=2), and 1"256 (n=l) There were no cases of fetal hydrops Eight ammocenteses performed among 3 patients with t~ters > 1:128 were all in zone I of the IJley curve Thirty seven (32%) of the infants were M-antigen pos~We, butonly 9 (7.8"/o) had a pos=tive d=rect Coombs’ at delive~ Two of the 9 had strongly positive Coombs’ results and occurred in newborns with maternal antibody t~ters that were weakly pos=tive 1.1 and <1 1. Conversely there were 21 patients w~th a positive ind~reet Coombs whose newborns were M antigen posi0ve and had a negative direct Coombs’ at dehvery Two of these patients had indirect Coombs’ of >1.64. No infants required phototherapy or exchange transfusions. CONCLUSIONS: Although anti-M antibody has been reported to cause severe hemolytic d~sease of t~e ne.wbom., our data i.ndi~ate that it ~s todd in the majonty of cases we nave Dean unable to detect clinically s~gn~ficant erythroblastos~s fetalis in any case over a 26 year

658 RHESUS MONKEY AS AN ANIMAL MODEL FOR EARLY GESTATIONAL FETOSCOPIC PROCEDURES. G. Webbx, T Feng, WH FlemingX, C Turner~, RB Swensonx, J Huddleston. Depts of Ob/Gyn, Medicine, and Pediatric Hematology, Emery Umvermty School of

Medicine, and the Yerkes Primate Research Center, Atlanta, GA. OBJECTIVE: To determme if fetoscopic visual~atmn in early rhesus gestation would be praetacal and beneficial In facilitatmg fetal venous access. When cordocentesis for transplantation of hematopoletic stem ceils in thirteen early rhesus gestations was difficult to ecineve and confn’m on some animals using ultrasound guidance alone, we utilized fetoscopic visualization to enhance the procedure STUDY DESIGN: After sedation and ultrasound confirmation of gestational age at 45-60 days (fetal raze comparable to 12-13 w~eks m human), a 1.9 nun, senu-ndged endoscope (Imagyn Medical) was inserted transabdominally rote the uterine cavity of four separate Rhesus females

under ultrasound guidance. Intravenous fetal access and transplant were attempted a 22-24 gauge spinal needle placed through a separate site Antibiotac prophylaxas was used RESULTS The fetoscope was successfully introduced into the uterine cavity of all animals Fetal anatomy was well visualized in each case w~th video and photos easily obtained Intravenous hematopoietic stem cell mjecUon via cordocentems was attempted m three animals and was successfully completed m two. There were no fetal losses, wound mfectmns, or other identifiable complicataons, CONCLUSIONS Endoscopic visuahzatlon of the Rhesus monkey fetus is possible, even at an early gestational age, w~th mimmal risk of fetal loss

or injury Fetoscopy is of mgnificant benefit in obtaining access to the fetal e~rculatian Rhesus monkey as an anmaal model for fetoscoplc procedures offers the obvious advantages of fetal and maternal anatomic similarity, while fetoscopy may have multiple applications as a diagnostic and therapeutic tool in humans

660 HEMOLYTIC DISEASE OF THE FETUS: A COMPARISON OF THE QUEENAN AND EXTENDED LILEY METHODS. Joseph ,~,. Spinnato, Ann L Clark, Kathleen K. Ralston,~ Eileen R. Greenwell? Dept. of Ob/Gyn, Univ. of Louisville, Louisville, Kentucky 40292. OBJECTIVE: To compare the performance and utility of the linearly extended Liley chart to the Queenan chart. STUDY METHOD: Amniotie fluid bilirubin (AFB) determinations were performed upon 242 specimens from 75 sensitized patients comparing the techniques described by Liley (chloroform extracted) and Que~nan. The frequency of values in the four Queenan zones (Q) was compared to that in the four "zones" of the Liloy (L) graph (middle zone sub-divided). The aecaracy of the methods was judged as 1) correct, 2) appropriate trend, 3) risk overestimated, 4) risk underestimated, b~sed upon the hematocrit at delivery or transfusion. RESULTS: Values on Q graph were more likely than L to be in a higher zone both before and after 28 weeks (p <0.005). The L method as accurately predicted fetal status as Q (correct 82.8 v 87%, 12.1 v 3.3% app. trend, 1.2 v 8.7% overestimated, 0.8 v 0.4% underestimated), For 23 specimens, Q could not be calculated. Neither hydrops nor fetal death occurred using the L method for management. Among 36 cases where the initial hot was known, L performed as well as Q in 29, better in 4 and poorer in 2 cases. CONCLUSIONS: 1) The utdity of the linearly extended Liley Graph compares favorably to the Queenan Graph. 2) Careful trend analysis is important regardless of method. 3) AFB can be safely used in the second trimest .... I

,

Volume 174, Number l, Part 2 SPO Abstracts 489 Am J Obstet Gynecol

661 RETINOL-BINDING PROTEIN (Rbp) AND ALPHA-1 MICRO-

GLOBULIN (a-lm) IN THE EVALUATION OF FETAL

PROTEINURIA. MP Johnson. S Gaddipatix, A Freedmanx, J Cejkax, K

Kithierx, R Qmntero, !vii Evans. Depts OB/GYN, Molecular Medicine &

Genetics, Pathology, Wayne State University Detroit, MI.

Objective: Differences in specific reabsorption and catabolism of

constituents of fetal urine may be reflective of incremental damage to the

proximal renal tubule We evaluated two components of fetal proteinuria as

potential markers of underlying renal damage in congenital obstructive

uropathy.

Study l~slgn’ Rbp and a-lm levels were determined and compared to total urinary protein (TP) values in last of multiple urine samples obtained by serial vesicocenteses in male fetuses with megacystis, hydronephrosis, and oligohydramnios. TP values were determined spectrophotometdcally using Coomassie blue. Rbp and a-lm values were determined by radial unmunodiffusion. Outcome data was def’med as good if infant survived and had a nadir Creatinin (Cr) <1.0 at one year, or poor if nadir Cr >1.0 or if significant renal damage was confirmed at autopsy. Screening thresholds for Rbp and a-lm were determined to optimize differentlatton for absence of renal damage. TP threshold was based on previously reported work.

Results: Protein Markers for Detection of Mimmal Renal Injury

Rhp <_ 4 mg/L 29 0.82 1.00 1.00 0.90

#-l~t <60 mg/L 28 1.00 0.83 0.77 1 00

!’P <20 mg/dl 44 1.00 0.89 0.85 1.00

Conclusion: Fetal proteinuria can help identify the presence or absence of severe renal injury in fetal obstructive uropathy Although TP appears to remain the best overall marker in fetal proteinuria, a combination of protein components may offer maproved detection as some offer better sensitivity and others specificity. Additional cases are needed to elicit this combination and verify this approach.

663 IN UTERO TREATMENT OF RIGHT-SIDED CONGENITAL DIAPHRAGMATIC HERNIA (R-CDH) BY PRENATAL TRACHEAL OCCLUSION. AW Flak~. MP Johnson, M Treadwell, B Mason, C

Cauldwell~, AI Philh’parP, ML Cullenx, J O’Brien, NS Adzickx, MR ~x, MI Evans. Fetal Therapy Program, Depts. Surgery, Cb/Gyn,

Genetics, Pathology, Hutzel HosFntal and Children’s Hospital of M~chigan, WSU, De~oit, MI. OBJECTIVe: The mz~ority of fetuses with CDH have significant

pulmonary hypoplasia and die despite opttmal postamtal treatment. R-

CDH does worse than left. Prelimanary evidence suggests that fetal txacheal occlusion results in acceleration of lung growth wxth reduction of

herniated viscera.

METHODS: Diagnosis of severe, tsolated, R-CDH was made at 21 weeks

by sonography. Fetal karyotype was normal. There was extensive

herniation of the liver into the right chest with marked mechaslinal shift

and eesentlally no visible lung parenchyma. After extensive counseling,

we performed fetal tracheal occlusion by open hystemtomy at 27 weeks

gestation. Preterm labor was controlled by intravenous nitroglycerin The

fetus was electively delivered by cesarean section at 32 weeks. Tracheal

dips were removed, and fetal bronchoscopy and intubation performed

before uteroplacental clrculalion was disconnected

RESULTS: Growth of the lungs was appreciated by sonography by 5 days

after tracheal occlusion and return of the mediastinum to the midllne

occurred by 3.5 weeks. Pulmonary function was remarkably good after

birth, and the diaphragmatic defect was repaired with a patch 24 hours

after delivery. The trachea appeared ran-real in cahber without malaaa bybronchoscopy, pfio~ to extubation, at 2 weeks of age His course was

complicated by RSV pneumonia and a bowel obstruction related to

assucmted malrotalaon, but he is currently thriving at home at 4 months of

age. CONCLUSIONS: Our results suggest that fetal tracheal occlusion

accelerates lung growth reducing the severity of pulmonary hypoplasia in

fetuses w~th CDH This report represents the fi~st successful prenatal

treatment of a fetus w~th R-CDH.

662 PRENATAL DIFFERENTIATION OF RENAL DAMAGE USING GESTATIONAL AGE-BASED THRESHOLDS FOR FETAL URINARY B-2 MICROGLOBULIN (B2m). MP

Johnson. A Freedmanx, K Kithierx, J Cejkax, R Hume, R Quintero,

C Smithx, MI Evans. Depts OB/GYN, Molecular Medicine & Genetics, Pathology, Pediatric Urology, Wayne State University, Detroit, MI.

Objective: Studies of B2m as a predictor of renal damage in fetal obstructive uropathy have been controversial. We investigated the use of alternative thresholds and gestational age (GA) adjustments.

Study Design: B2m values were evaluated by RIA from last of

sequential urine samples obtained by serial vesicocenteses in 36

male fetuses with megacystis, hydronephrosis, and oligohydramnios.

Outcome data were defined as good if infant survived with nadir

Creatinine (Cr) <1.0 at one year, or poor if nadir Cr >1.0 or if the

presence of severe renal damage was documented at autopsy. Cutoff

thresholds for B2m were determined to optimize differentiation for

absence of significant underlying renal damage.

Results: Detection of Minimal Renal Injury Using B2m

Threshold N Sens S ¢ PPV NPV

~6~ts)I 36 I 0.83 I 0.77 I °861 0.71 I :g6(<2Owks)I 20 I 0.69 I 0.71 [ 0.82 ] 0.55 I <6 (>204t1~) [ 16 I 1.00 I 0.83 I 0.91 I 1.00 I

Conclusion: 1) a B2m threshold of< 6 mg/L is better at identifying

the absence of renal damage than previously reported thresholds, 2) B2m < 6 mg/L is a better indicator of minimal renal damage after 20

wks than prior to 20 wks GA, 3) Additional cases will be needed to further refine these thresholds within these two GA periods.

664 CYTOKINE STIMULATION OF HUMAN FETAL HEMATOPOIETIC CELLS. M West~ren, S Ek~, L Markhagx, A Kjeldgaardx, Q RlngdenL Dept Ob/Gyn and Dept Chn Immunology, Hudd~nge Umversity Hospital, Karohnska Institute, Stockholm, Sweden OBJECTIVE: To study the effects of IL 3, IL 6, stem cell factor (SCF), and granulocyte macrophage - colony stimulating factor (GM-CSF) on human fetal hematopomtlC cells (FHC), bone marrow (BM) and cord blood (CB) cells, by colony forming capacW Single and combined cytokmes and prelncubation vs adding cytokmes in culture were investigated STUDY DESIGN: FHC from 28 elective abortions, three BM samples and three CB samples were incubated with cytok~nes and mvesUgated for the presence of BFU-E (Burst forming unlts-erythroid), CPU-GM (Colony fonmng units granulocyte, macrophage) and CFU-GEMM (Colony forming units granulocyte, erythrocyte, macrophage, megacaryocyte) ANOVA was used for statistical analysis RESULTS: 1L 6 alone had the most pronounced effects on BFU-E formation All four cytokines in combmatmn y~elded the highest scores for CFU-GM (p < 0.05) and CFU-GEMM (p < 0 05) The mode of cytoklne exposure was not a significant determinant of colony formanon CONCLUSIONS: IL 3, IL 6, SCF and GM CSF have significant effects on FHC as evaluated by colony formation and may prove useful for the purpose of enhancing cell proliferation prior to fetal transplantation


665 RISTOLOGIC ACUTE INTRAUTERINE INFLAMMATION VERSUS CLINICAL CHORIOAMNIONITIS: WHICH BEST PREDICTS NEONATAL SEPSIS. C.M~ Salafia*, V.K M~n~or*, J.C Pezzullo’, C Y. Spong*, D.M. Sherer, A. Ghid~nl. Perinatal Research Facd~ty, Departments of Pathology & OB/GYN, Georgetown Univ. Medical Center, Washington DC, UCONN Medical Center, Farmmgton CT. OBJEOTIVE; To compare Nstolog~c d~agnos~s of acute inflammahon and chnical chonoamnlonlbs (CA) as predictors of culture proven neonatal seps~s (CPNS). STUDY DESIGN: From an estabhshed set of 465 infants dehvered <32 weeks were excluded preeclamps~a {n=75) and 9 cases in which a clin~caJ d~agnosts of CA was not supported by chart rewew The remaining 38! cases ~ncluded premature rupture of membranes (PROM) (n=188, 49%), preterm labor intact membranes (n=160, 42%), and non-hypertensive abruption (n=33, 9%). Maternal and neonatal charts were reviewed Chnical CA was diagmosed ~n 122, suspected m 119, and neither d~agnosed nor suspected ~n 140 patients. Acute inflammation ~n amnion (maternal ~nflammabon) and in umNical and chorionlc vessels (fetal inflammahen) was moderate ~f graded 1-2 and severe ~f graded 3-4 on a 0-4 scale Sens]t~wty Sens) and speclfidty (Spec) were calculated from 2X2 tables. RESULTS: 62 (16%) cases of CPNS were identified, Of the 122 clinical CA, 22 (18% had CPNS (Sens 69%, Spec 36%) When CA was d~agnosed or suspected (n=241), 30 (12%) had CPNS (Sens 40%, Spec 19%) In 22/86 (26%) cases severe acute amn~on~t~s had CPNS (Sans 80%, Spec 36%). Combining any acute amn~on~t~s with any fetal inflammahon gave the best sens~t~vity (Sens 90%, Spec 23%). No one critenon required for diagnosis of CA (foul smelhng amnlotic fluid (AF), purulent vagina! discharge, maternal temperature {T) >37 8°O, WBC>20,000/mm3, or utenne tenderness) was related to the ~nc~dence of CPNS (each p>0.20). However severe acute amn~onlhs was predicted by maternal T>37.8°C {p=0 005, Sens 87%, Spec 26%), maternal WBC>20,000/mm3 (p<0.05, Sans 80%, Spec 33%), foul smelhng AF p<0 0001, Sans 91%, Spec 35%), and purulent vaginal discharge (p<0.0001, Sens 91%, Spec 34%) Slmdar relationships were seen {or severe fetal Inflamm~on and WBC>20,000/mm3, foul smelhng AF, and maternal T >37.8°C (each p<0.01) CONCLUSIONS. Cntena for dlagnos~s of CA are individually excellent predictors of h~stologlc severe acute inflammation ot maternal and fetal origin, but not of CPNS H~stologic acute inflammation Is up to 90% sensihve m the prediction of CPNS, and ~s more sensihve than a clinical d~agnosis of CA Frozen section d~agnos~s of amn~on~t~s and umbd~cal and!or chor~onlc vascuht~s at the t~me of h~gh- risk dehvery may enhance management wNle pennatal/neonatal cultures are pending.

667 SEVERE FETAL HEART RATE VARIABLE DECELERATIONS ARE ASSOCIATED WITH UMBILICAL VASCULITIS/HISTOLOGIC AMNIONITIS IN PRETERM INFANTS. CY Spong*, CM Salafia*, DM Sherer, A Ghidini, VK Minior*. Pe~earch .Facility, .D.ep,.art ,m%nts of O.B.,/G,YN a.nd Pathology, GeorKetown University Memcm t, enter, washington DC and’Univ, of Connecticut, Medical Center, Farmington, CT: OBJECTIVE: To evaluate whether severe fetal heart rate (FHR) variable decelerations in patients with preterm delivery are associated with histologic evidence of infection. STUDY DESIGN: Intrapartum FHR tracings from an established database of 465 consecutive patients delivering between 22 and 32 weeks gestation (excluding stillbirth, fetal structural anomalies, materna~l diabetes and hypertension) were assessed. Severe variable decelerations were defined as F~HR _< 70 ,bpm la, sting > 60 seconds. Histopathotogy examination oi amnion tconsiuereurepresentatlve of maternal inflammation) and umbihcal cord (reflecting fetal inflammation) were performed by a single pathologist blin~ted to clinical data. The d~agnosis of acute amnlonitis and umbilica! vasculitis were scorea as absent, mild/moderate, or severe. AmmotIc fluid volume (AFV)~w, as determined sonographic,all~’ a,s part of a fetal blophy.slcal prome. Outcome parameters inmuueu: ~estational age at anal indication for delivery, umbilical at’tery pH, 5irthwel~ht, initial neonatal hematocrit, leu.~ocyte and band count. Statistica"l analyses, includ~ed contingency tables and ANOVA with p< 0.05 considereo slgnincant. RESULTS: Severe FHR variable decelerations were oresent in 75 (16%) and absent in 390 (84%) patients. Severe acute ammonitIs and umbihcal vascuhtis were present in 23 (31%) and 25 (33%) cases respectivelyof theoatients with severe variable decelerations, versus 65 (17%) and 64 (16%) without severe variable decelerations, each p < 0 005. Umbilical artery pH was si~mflcantly lower in cases with severe variable decelerations (7.27 +’0.09 vs 7.31 + 0.08, p = 0.0009). The incidence of severe FHR variable decelerations was indeoendent of AFV. Gestational age at and indication for delivery, birff~weight, initial neonatal hematocrtt, leukocyte and band count were not significantly different between the groups. CONCLUSION: The incidence of severe variable decelerations at < 32 weeks gestatmn IS sigmficantlygreater when either severe maternal or f&al parameters of acute inflammation are present. This observation may reflect an enhanced vasoconstrict~ve response of the umbilical cord vessels in an inflamed fetal environment. These findings are consistent w~th prior studies in patients at term

666 TIlE COMBINATION OF SEVERE FETAL HEART RATE VARIABLE DECELERATIONS AND DECREASED VARIABILITY ARE ASSOCIATED WITH ELEVATED INITIAL nRBC VALUES IN PRETERM NEWBORNS. CM Saiafla*, CY Spong*, VK Mimor*, DM Sherer. PerinataI Research Faclbty, Depts. of OB/GYN & Pathology, Georgetown Umverslty Med. Ctr, Washington, DC, Umvers~ty of Connecticut Medical Center, Farmington, CT. OBJECTIVE: To examine the relationship between antrapartum severe fetal heart rate (FHR) variable deceleratmns, decreased FHR variability, hlstologm acute fetal mflammatmn and mltial neonatal nRBC/dl. STUDY DESIGN: From an estabhshed database of 465 consecutive nonhypertenslve, nondlabetm patients with singleton, nonanomalous fetus+s dehvering <32 weeks with imt~al neonatal bond count obtained -<3 hours and mtrapartum FHR tracings (w~th severe variable decelerations and decreased vanablhty defined according topublished criteria) were Identified nRBC/dl was calculated as WBC.nRBC/100WBC. Acute umblbcal and chortomc (fetal) vascuhtls were scored on a 0-4 scale by a single pathologist bbnded to clinical data Anmlotlc fired voIume (AFV) within 24 hours of delivery, gestational age at birth, and membrane rapture to debvery interval were extracted. StatistlcaI analyses included ANOVA, %2, Kruskal-Wallis test, and multivariate analys~s with p<0 05 significant. RESULTS. 411 patmnts met the mclusion criteria Severe FHR varmble decelerations were seen in 66 (16%), decreased varlablhty in 80 (19%) and both m only 13 patients (3%) Patients with severe FHR variable deceleratmns had s~gmficantly ~ncreased nRBC/dl (1 7 -+ 1 2 vs 1 3 _+ I 1, p=0.02). Increased sum of scores for fetal vascuhns was also slgnlhcantly reIated to increased nRBC/dl (p<0.0{~l). While decreased variability was not associated w~th nRBCs/dl, there was a trend to increasing nRBCs/dl when both FHR changes were present (p=0.068) The presence of neither, either or both FHR changes, and the sum of fetal vasculius were ~ndependent prethctors of nRBCs/dl (p=0.04, p=0.0004, p<0 0001, respectively). AFV, gestatmnal age and rapture of membranes to delivery Interval were not related to nRBCs/dl CONCLUSIONS: These data inthcate the combination of severe FHR variable decelerations and decreased FHR varlablhty, and fetal vascuhtls are s~gmflcant markers of ~ncreased nRBCs/dl The mult~vanate analys~s demonstrates that I~HR chauges and fetal vascuhtls may operate in synergy. We propose that acute Inflammation may affect umblhcal vasomotor stablhty Following cord compression, the sensitized amblhcal vessels respond with vasospasm, causing severe FHR variable decelerations. This leads to abnormal fetal perfusron, changes ~n FHR vanaNhty and an increased nRBCs/dl m the m~tml neonatal blood count.

668 TRANSABDOMINAL AMNIOINFUSION IN OLIGOHYDRAMNIOS AT TERM WITH INTACT MEMBRANES PRIOR TO INDUCTION OF LABOR: A RANDOMIZED CLINICAL TRIAL ~, P Cerut~*, N= Strobelt*, A Loeatelh*, P D’Oria*, S Manam*, A Gh~d~m Department of Ob/Gyn, S Gerardo Hospital Umversity of Milan, and Pennatal Research Facility, Georgetown Umv Medmal Center, M~ an, Italy, and Washington, D C OBJECTIVE: Transcervical ammomfusmn ~s frequently used after rupture of membranes to improve pennatal outcome In the presence of variable decelerations dunng labor or mecomum stained ammot~c fluxd (AF) We tested the hypothesis that transabdommal amnmmfus~on (TA) pnor to induction of labor ~n pnm~paras with ohgohydrammos and intact membranes would decrease the rate of cesarean delwcnes performed for fetal distress STUDY DESIGN: Pnmlparas at term with a sonograph~e diagnosis of oligohydrammos admitted for ~nduetmn of labor were considered for melusmn in th~s randomized trial if they had’ 1 ) s~ngleton gestatmn, 2) vertex presentation, 3) ultrasonographieally estimated fetal weight _> 2500 grams, 4) reactive nonstress test, 5) cervical Bishop’s score _ 6, and 6) intact membranes Excluded were all patients undergoing an eleehve cesarean section Consenhng patients were randomized by computer generated tables to ultrasound-graded TA pnor to lnduetmn (TA group, n=39), or to direct labor reduction (no TA group, n=40) Statasheal analysis utihzed conUngency tables, Student’s t-test or Wdkoxon rank- sum test, where appbcable RESULTS The two groups had similar maternal and obstetric eharactenstms, duration of first and second stage of labor, as well as membrane-rupture to dehvery mterval. TA was successful in all women randomized to the procedure The ~neldence of abnormal fetal heart rate tracing was h~gher in the no TA than TA grnup (42% vs 5%, RR=2 3, 95% CI 1 6-3 3) The rate of spontaneous vaginal dehvery was s~mdar tn the two groups [87% (34/39) vs 70% (28/40), p=0 11] The rate of cesarean section for fetal distress was s~gmficantly higher m the no TA than TA group [25% (10/40) vs 5% (2/39), RR=I 9, 95% CI 1 3- 2 7] No bleethng or infectious complleatmns were noted CONCLUSIONS TA prior to mduchon of labor m a safe and effective optmn

to prevent fetal &stress In pregnancres w~th ohgohydramnlos at tema, intact membranes and unripe cervix

Volume 174, Number l, Part 2 SPO Abstracts 491 AmJ Obstet Gynecol

669 HEART RATE PATTERNS IN FETUSES WITH FETAL ALCOHOL SYNDROME. Y Sorukln, L. Chlkx, C Brown~, S. Martler~, R Sokol.

Dept Ob/Gyn, WSU/Hutzel Hospital, Detroit, MI OBJECTIVE: More non-reassuring FHR patterns in fetal alcohol syndrome

(FAS) pregnancies may explain their reported high Cesarean birth rate. Our objectwe was to compare FHR patterns in FAS and non-FAS pregnancies. STUDY DESIGN: A study group of 2,282 African American gravidas was

followed prospectively through pregnancy and delivery with repeated detailed

alcohol and drug histories Babies were examined at birth by nurses trained in dysmorphohigy examinations 59 FAS cases were identified. Non-FAS

pregnancies were matched for gestational age. There were 100 (33 FAS and

67 controls) with at least 60 minutes of FHR data. The last 60 minutes of available lntrapartum FHR was scan converted for interactive computer display The first 30 minutes and last 30 minutes of each 60 minute strip were wsually interpreted for FHR pauerns by one observer who was blinded to

chmcal data FHR patterns included baseline (BL), short term (STV) and long term (LTV) var~abdity accelerations (A), decelerations (early [El, variable

[V], late ILl), and bradycartha (BR). The strips were coded for overall reassurance (R) non-reassurance (NR) and artifacts (ART) Our hypothesis’

There are no chnlcally significant differences in FHR patterns A stepwise dlscnmmant function analysis was run using L,V,A,BR and NR for FAS versus non-FAS status. Significance was considered at p<0 01. RESULTS: There were no stanstlcally significant differences between FAS

and non-FAS strips when the first 30 mmutes of tracing were compared. Analysis of the last 30 minutes of tracing yielded small statistically significant, but no chmcally important differences for V(~), L(~), and BR(t,).

F[1,981 p

V2 4 47 .035 Increased m FAS group

L2 4.84 028 Decreased

Brady2 4 59 032 Decreased

CONCLUSIONS: We found no clinically important differences of FHR patterns between FAS and non-FAS pregnancies. The reported high Cesarean birth in FAS pregnancies may have explanations other than "fetal distress"

671 VISUAL ONLINE SIGNAL IDENTIFICATION AND THE ACCURACY OF FETAL PULSE OXIMETRY IN SECOND STAGE OF LABOUR. G. Ar~ka~. M.C.H. Haeusler, F. Kaine~ J

Haas.~ Depart. Obs & Gyn, Karl - Franz. Univ., Graz, Austria. OBJECTIVE: The aim of this study was to determine the accuracy and feasibility of pulse oximetry during the second stage of labour with an atraumatic device. PATIENTS AND METHODS: 61 healthy term fetuses in vertex position were enrolled in the study. There were 52 spontaneous vaginal deliveries. 52 women had peridural anaesthesia. In 50 of these fetuses reflection pulse oximetry monitoring could be performed until delivery. Fetal oxygen saturation (SO2 POX) during the last 10 minutes before delivery was compared to oxygen saturation in the umbilical vein (SaO2) after delivery, SO2 POX was measured with an RPOX MK2 (ZLirich) with an atraumatic suction probe, This system allows online identification of (in)valid SO2 POX readings by showing the original red (660 nm) and infrared (920 nm) AC/DC signals. Umbilical vein SaO2 was measured by a spectrophotomater (AVL CO- Oxylite 912) after delivery. RESULTS: The uptake of valid signal during the last 10 minutes before delivery was 5 - 70.9 % (mean 29.9 %). fhere was a significant relationship between SO2 POX measurements during the last 10 minutes before delivery (mean 46.7 %) and the umbilical vein SaO2 (mean 49.4 %) (r - 0.47; p - 0.001). SO2 POX measurements during the last 30 or 60 minutes before delivery did not correlate well with the umbilical vein SaO2. We could not find a significant correlation o~SO2

POX measurements to umbilical vein pH after delivery (7 to 7.4, mean 7.3) (r = 0.45; p- 0.1 ). CONCLUSION: SO2 POX measurements with a new reflection pulse oximetry system with an atraumatic suction device reflect actual oxygen saturation in fetal arterial blood. The system allows visual online data analysis to exclude invalid signals, which is crucial in second stage of labour where artifacts are frequent. Supported by the Austrian Science Foundation (Grant # 9344)

670 INTRAPARTUM FETAL PULSE OXIMETRY: PUERPERAL

MORBIDITY. GA Dildy, CA Loucks~, SL Clark. Department of Ob/Gyn,

University of Utah, Salt Lake City, UT and UVRMC, Provo, UT.

OBJECTIVE: To determine if intmpartum monitoring of fetal arterial oxygen

saturation (SpO2) with a reflectance pulse oxameter is associated with an

increase in maternal pefipaaum complications.

STUDY DESIGN: The Study Group (SG) consisted of 100 women at term (2

37.0 weeks) in labor with singleton vertex fetuses and ruptured membranes,

momtored with a reflectance fetal pulse oximeter (Nelicor NM00 Fetal Pulse

Oximeter and FS-10 Oxisensor); SpO2 data were not used for clinical

management. The Control Group (CG) were 100 women with singleton vertex

fetuses in labor wth roptured membranes, matched for gestational age. 50

demographic and outcome parameters were evaluated. Comparisons were made

by the Mann-Whitaey U, Chi-Square, and Fisher’s Exact test, with significance

considered at P < 0.05.

RESULTS: The mean (_+ SD) duration of SpO2 monitoring in the SG was 2.9

± 1.7 hours. There were no differences between the SG and CG in the duration

of hospitalization postdelivery, intmpartum maximum temperature (Tmax),

Tmax postpartum day (PPD) 2, PPD-1 hematocrit, blood transfusions, or

cesarean sections. There were differences noted in number of vaginal exams

pest rupture of membranes (9.9 vs 6.9, P < 0.000001), duration of ruptured

membranes (6.8 vs 7.6 hours, P < 0.00001), duration of labor (9.3 vs 7 6 hours,

P < 0.001), and Tmax PPD-1 (99.6 vs 99.2°F, P = 0.004). The SG had a

statistically higher Tmax on PPD-I, but there was no statistically significant

difference in the proportion of women with T 2 100.0°F (34% vs 21%, P =

0 06) on PPD-I. The one blood transfusion (SG) occurred following hemorrhage

post tubal ligadon.

CONCLUSIONS: When compared with women receiving standard intrapartum

fetal monitonng, those with SpO2 monitoring had a statistically, but not

clinically, significant increase in Tmax on PPD-I. This may be explained by the

increased duration of labor, duration of ruptured membranes, and pelvic exams

after rupture of membranes in the SG, and may in part be a selection bias, as

women who were not expected to immediately deliver were recruited into the

SG. Fuaher analysis ~s indicated.

672 The Effect of Butorphanol on the Fetal Heart Rate Reactivity During Labor ~x AP Sarno, RS Egerman, BM Sibai. Dept. of Ob/Gyn, University of Tennessee, Memphis Objective: To evaluate the effects of butorphanol (Stadol®)on the fetal heart rate reactivity observed for 60 minutes after Stadol® administration. Methods: Twenty-nine women receiving 1 mg of intravenous Stadol® for pain relief in labor were studied. Prior to the administration of the drug all patients had a fetal heart tracing for at least 20 minutes revealing the presence of two accelerations of 15 beats above the baseline lasting 15 seconds. Fetal heart rate reactivity spontaneously or in response to fetal acoustic stimulation at baseline was compared to that at 15 minutes. Tracings that were deemed nonreactive at 15 minutes were observed for a reslxmse to fetal acoustic stimulation. All tracings were again reviewed at 60 minutes. A power analysis was calculated to estimate the sample size needed to detect a loss of reactivity in 25% of patients. The sample size of 29 women had a power of 80% and an alpha of 0.05. Results: Thirteen fetuses (44%) continued to be reactive at 15 minutes after Stadol® administration. The remaining 16 underwent acoustic stimulation; 13 of them (78%) became reactive. The final 3 nonreactive fetuses became reactive by one hour. Conclusions: The use of Stadoi®in labor may be associated with a transient decrease in fetal heart rate reactivity. A vibroacoustic stimulator will shorten lbe period of fetal nonreactivity following a dose of Stadol®, thereby realtirming fetal well being.


673 Fetal pulse oximetry for monitoring deliveries with

suspected chorionic amnionitis (CA)

Luttkus A.*, Frledmann W.*, Dudenhausen J. W.

Dep. of Obstetrics.; Virchow Klinlkum; Humboldt Universit~it

Berlin; Germany

OBJECTIVES: Does additional pulse oxlmetry momtormg improve

the chnical management of dehverles with suspicion of infection ~

STUDY DESIGN: We enrolled into th~s prospective trial 85

deliveries inomtored by a blinded fetal pulse oximeter (N 400,

Nellcor Inc., Pleasanton, CA) and a fetal oxlsensor (FS14; ~,

=’735+890). The medmn of O2satiirat~on was calculated for all

fetuses without infection spontaneously delivered with l’min

Apgar>6 and pH of umbdical artery >7.19. Add~tmnally we

established the median of O2saturation of all fetuses with a clmtally

defined CA (maternal temperature >38.0°C, CrP>2.0 mg/dl or

leucocytes > 160001~tl).

RESULT: Wh~le the obsterical risk factors (non reassurmg FHR

i.e.) were significantly higher in the group w~th amnlon infection,

the median of O2saturatlon did not differ s~gnificantly (p>0.5). m

periods w~th normal and tachycard FHR. The dnrataon of antibmnc

treatment, mechanical ventdatmn or phototherapy differed as the

number of neonatal infections.

CONCLUSION: In the group of fetuses with CA there was no

increase or decrease of the median o1 O2saturat~on. These data

suggest that during delivery the fetal oxygenation remains lmrmal

for a long t~me before the fetal decompensation starts. Th~s moment

can be detected easily by pulse ox~metry as shown in individual

cases.

675 LACK OF CONSISTENCY IN DEFINITIONS OF FETAL

HEART RATE (FHR) PATTERNS. Parer JT, Quilhgan EJ

Department of Obstetrics, Gynecology and Reproductive Sciences and

CVRI, University of California, San Francisco and Department of

Obstetrics and Gynecology, University ol California, Irvine.

OBJECTIVE: To determine the extent of agreement amongst a group

of academic educator/chnicians in definitions of FHR patterns.

STUDY DESIGN: Fifteen senior academic North American

permatologists were selected on the following bases: a) FHR

publications, b) active in chmcal care and education, c) diversity of

institutional and geographic background. A written survey with 12

categories and 48 subquestions was submitted.

RESULTS: Fourteen responded. The following are selected

definitions, with numbers of respondents in parentheses

Bradycardlas: Baseline FHR bpm. <110 (6); <120 (7); <90 (1)

Duration. 2 min (4); 3 rain (4); 5 man (1); 10 rain (5).

Tachycardia. Baseline FHR bpm: >150 (1); >160 (12), >180 (1).

Duration’ 2 mm (2), 3 rain (I); 10 rain (8); >10 rain (3)

Late Shape’ U, uniform, smooth (12); uniform or irregular

Deceleration: (1); not stated (1). Tmung to UC: Late (13); usually late

(1). Frequency: >50%(3); >80%(1); persistent, repetitive

(6); not used (4).,

Early Shape: U (11), classed as variable (1); classed as reflex

Deceleration. late (1); not stated (1). Frequency: >50% (3); any (2),

most (2), not stated, or no criteria (7)

Variable Shape: variable, abrupt, Lrregular (13); other than late

Deceleration. or early (I). Timing variable (14) Frequency: variable (6); inconsistent (1); persistent or >50% (4); 0 input (3)

FHR Short term: beat to beat or R-R mt (11); other (3) Long

Variability: term: irregular fluctuations, etc. (13), 0 response (1). CONCLUSIONS: Fetal heart rate pattern definitions amongst these

influential educator/chnlcian/mvestigators are remarkably varied We

speculate that this diversity precludes any meaningful research work on

etiology and efficacy of FHR monitoring or recommendations for

management until standardization of definitions is achieved.

674 THE VALIDITY OF 1NTRAPARTUM PULSE OXIMETRY: A

QUANTITATIVE ANALYSIS. H.W Jongsma, P P van den Berg~, J J M.

Menssen~, JG. Nijhuls~ Dept Ob/Gyn, Univ. Hosp Nijmegen, The Netherlands,

G.A Dlldy, Unlv of Utah, Salt Lake City, A. Luttkus~, Humboldt Univ., Berbn

OBJECTIVE: To perform quantitative analysis of fetal pulse oximetry (SpO:)

tracings as a tool to assess the mtrapartum fetal conditlou

STUDY DESIGN: In 5 centres (2 USA, 3 Europe) 308 intrapartum recordings

of cardlotocography (CTG) and SpO~ (with 660/935 nm prototype fetal sensor,

Nellcor, Pleasanton, CA) were made Selected were 119 cases fulfilling two

criteria. I) >45 rain of tracings available and <15 rain tlme interval between

removal of the sensor and birth, 2) and available umbihcal artery (UA) pH.

Computer-stored SpO~ tracings were characterized by two parameters, the mean

SpOt, andthe areaofthe SpO2 tracing below a ’critical’ level of 30% (’SpO~

area’) Receiver Operating Characteristic (ROC) curves for the prediction of UA

pH <7. l 5 were calculated

RESULTS: The median percentage of time with SpO~ signals available was

44% (range 3-85%) Mean SpO~ varied from 25 2-87.5% (median 55.0%). SpO~

area ranged from 0-7 5% (95a’ centile 2 0%) The two SpOz parameters were

correlated (Spearman r-0 67, p<104) The ROC curves for mean SpOz (dashed

In fig ) and SpOz area (continuous hne) are compared with the visual assessment

of referees (open square) of the same tracings (previous study J Soc Gynecol

Invest 1995, 2/2 213). At a specificity of 86% the sensitivity of the opinion of

the four referees was comparable with the sensitivity based on SpO~ area, the

sensitivity based on mean SpO~ was lower (McNemar test. p=0 01) CONCLUSION: Disappointing l ’:"

due to limited accuracy of the 660/935 nm

sensors and the small fraction oftlme with ~ ~ good pu, ..... nIetry signals

ii~"~ "--- J-- ~

Improvements may be obtained by

combining the diagnostic value of

cardlotocography an__d SpOz, and refining

quantitative analysis of the SpO2 tracings , r

based on a better understanding of the ~00 so ~ 40 20 consequences of deficient oxygenation S~ec~

676 INTRAPARTUM FETAL HEART RATE PATTERNS IN 209 BRAIN

DAMAGED INFANTS Ahn MO, Korst L, Phelan JP, Depts of OB/GYN

Cha Women’s Hospital, Seoul, Korea and Pomona Valley Hospital Medical

Center, Pomona, CA

OBJECTIVES To describe the fetal heart rate (FHR) patterns of 209 term

brain damaged infants

STUDY DESIGN. The fetal monitor strips of 209 singleton term

neurologically lmpmred neonates were retrospectively analyzed

RESULTS Of the 209 infants, the admission fetal heart rate (FHR) patterns

were reactive or nonreactive in 113 (54%) and 96 (46%), respectively In the

nonreactive group, the FHR patterns remained nonreactive throughout the

intrapartum period In the reactive group, 14 (12%) patients renamed

reactive Of the renaming 99 (88%) patients, 53 (54%) developed a FHR

tachycardla, repetitive decelerations and absent variability with or without a

prolonged FHR deceleration in 23 (43%) and 30 (57%) patients respectively

In the remaining 46 (46%) patients a sudden, prolonged FHR deceleration

lasnng 34 2 + 13 8 minutes with a range of 15 to 81 minutes, developed.

CONCLUSION Based on our analysis of the FHR patterns of 209 term

neurologically impaired neonates, there is not a single, unique FHR pattern

giving rise to fetal injury Nonetheless, our data do suggest that continuous

electronic fetal monitoring does assist m ~dennfymg the fetus that develops

mtrapartum fetal distress

Subject Index

Volume 174, Number 1, Part 2 Subject index 495 Am J Obstet Gynecol

Abmplio placenta. 109, 196, 294, 352

Acid-Base Status -Fetal: 52, 94, 151, 153,177, 183,216, 257, 259-262, 437, 490, 615,670, 673-674

Alpha-fetoprotein: 22, 379, 415,457-459, 461-463,474-477, 479, 481,498

Amnioinfi~sion: 176, 199, 280, 626, 668

Amniofic Fluid Volume: 24, 82, 84-85, 92, 113, 115, 131,246, 248-249, 269, 383, 385-386, 470, 529, 562, 568, 626

Antepartum Fetal Evaluation: 41, 80, 83, 86-87, 94-95, 109, 130, 143, 196, 250-251,253,301, 379, 499, 554, 659

Acoustic Stimulation: 86-87, 252, 672

Amniotic Fluid Volume: 24, 82, 84-85, 92, 113, 115, 131,246, 248-249, 269, 383, 385-386, 470, 529, 562, 568, 626

Biophysical Profile: 84, 92, 339, 381

Doppler: 45, 188, 227, 235, 237, 420, 539-541,545-547, 549-553, 643

Fetal Movement: 42, 82, 113

Fetal Well Being: 93,238, 384, 544, 551

N~n-Stress Testing: 86, 88-89

Other: 9, 29, 68, 72, 78, 93, 108-109, 131, 202, 253, 255-256, 272, 277, 297, 314, 317, 327, 341,374, 392, 409-410, 413, 416, 425, 427, 429, 439, 443,513, 530, 539, 636

Antiphospholipid Antibodies: 7(I, 187, 296-298, 615

Atrial Natriuretic Hormone: 269

Birth Trauma: 71, 106, 133-135, 137, 148-150, 152, 433,645

Cardiovascular Hemodynamics: 47-49, 54-55, 72, 75, 198, 223-224, 232-236, 239-240, 242, 259, 271, 284, 444, 554, 644

Cervical Examination: 9, 145, 190, 375, 377-378, 440, 557, 592, 623,627

Cervical Ripening: 44, 64, 142, 154, 180, 631-632, 634-635

Cesarean Section: 4, 66, 104, 119-120, 130, 133, 135, 140, 155, 158, 160-169, 171,439-440, 442, 597, 623,625,627-628, 630, 647, 669

Computers: 80, 88, 97, 147, 156, 250, 363,372, 438, 548

Congenital Ano~nalies: 21, 23, 57, 105, 387, 401,408, 411-412, 421-422, 431,450-451,463, 478, 486, 494-495;

497, 656

Cordocentesis: 299, 335

Diagnostic: 241,329, 335, 575

496 Subject Index January 1996 Am | Obslet Gynecol

Diabetes: 1, 19, 26, 129, 187, 220, 302-303,305-308, 310-312, 314-316, 396

Complications: 302, 304-305, 309, 312, 314, 316, 356, 433

Fetal Evalualion: 222, 313,396, 432-433

Ghlcose Metabolism: 221,228, 303~ 541

Management: 26, 71,299, 302, 307-308, 313,315-316

Pathophysiology: 221,303,403,536

Testing: 309-312

Doppler: 45, 188, 227, 235, 237, 420, 539-541,545-547, 549-553,643

Drug Effects On: 198

Findings/OB Complications: 72, 382, 551-552

IUGR: 80, 88, 94, 114, 141,214-215, 239-241,257, 265-266, 279, 295, 317, 420, 496, 519, 524, 529, 540,

547, 549-550, 553,555, 558, 609, 653,657

Preeclampsia: 74, 524

Regional Fetal Blood Flow: 246, 540, 544-545, 547-548, 555

Epidural Anesthesia: 75. 245, 437-440, 442, 621,625, 642, 646

Exercise in Pregnancy: 200, 237-238, 613

Fetal Anomalies: 1, 19, 33, 37, 65, 103, 105-106, 387, 389-391, 393-395, 400-402, 408, 414, 417, 421-423,

426, 431,445, 448, 455, 465, 467, 471-473,475,477-478, 488-489, 492, 494, 498, 543,656

Fetal Growth: 98-99, 124, 230, 23%240, 254, 258, 265, 279, 403-404, 425, 427-428, 449, 593

Fetal Growth Abnormalities: 41, 136, 297, 519, 558, 609

IUGR: 80, 88, 94, 114, 141,214-215, 239-241,257, 265-266, 279, 295, 317, 420, 496, 519, 524, 529, 540,

547, 549-55(I, 553,555,558, 6(19, 653,657

Macrosomia: 66, 71, 134, 136-137, 139, 162, 436, 449

Fetal Lung Maturity: 23, 78, 129, 222, 264, 267, 273,276, 313,432, 566-567, 574, 598-600, 653-654

Fetal Therapy: 23-24, 27, 57-58, 61, 69;270, 279, 480, 546, 653,655-657, 659, 660, 664

Genetics: 22, 25, 27, 33, 91,110, 179, 292, 374, 408, 412, 416, 445, 448, 450-455, 457-464, 466-467, 471-472,

474-476, 479, 481,483-485.487-489, 495, 498, 501-503,641

Genetics-Diagnostic Technique: 21, 25, 91, 111,362, 393,409, 411, 421,447, 449, 452-453,459, 464-465,467,

471,473-474, 476, 478-480, 484-490, 496, 499

Hematology: 27, 69, 132, 179, 204, 244, 300, 306, 329, 454, 512, 664

Volume 174, Number l, Part 2 Subject index, 497 Am J Obstet Gynecol

Hemodynamic Monitoring: 5, 49, 76, 232-233,284, 293,510-511,554, 642-644

Hemolytic Disease of the Fetus: 55, 89, 132, 242, 480, 484, 499, 659-660

Hemorrhage - Poslpartum: 650

Hydrops Fetalis - Nonimmune: 335

Hypertensive Disease in Pregnancy: SPO Foundation Fellow Award Winner for Academic Year 1994-95 and

7, 29, 45, 74, 110, 186, 191,206-207, 228, 265, 275, 281,295, 500-502, 504, 506-509, 512-516, 518, 520 -523, 525-532, 534, 536-538, 601

Complications/Fetal: SPO Foundation Fellow Award Winner for Academic Year 1994-95 and 39, 519

Complications/Maternal: SPO Foundation Fellow Award Winner for Academic Year 1994-95 and 47

107, 278, 504-505, 510, 512, 515, 523, 526-528, 532-533, 535, 538

Evaluation & Diagnosis: 107, 110, 223,226, 500, 505, 507, 513-516, 518, 526, 531,533

HELLP: 168, 195, 275, 300, 503,527-528, 532

Treatment: 53, 76, 230, 506, 508, 511,514, 517, 537-538

Hypoxia & Asphyxia -Fetal: 38-40, 51-53, 172, 177, 212-213,251,257, 261,320, 437, 671,674-676

Incompetent Cervix: 101-102, 376-378, 429, 557

hffections/Maternal: 2, 31, 43, 46, 62, 104, 118, 168, 205,281,287, 291,321-324, 327, 332-334, 337-338, 340- 341,343-345, 347-349, 351,353-355, 357-364, 366, 368, 614, 665

Chorioamnionitis: 2, 25, 50, 67, 115, 211,322-323,325-326, 331,339-340, 349, 453,607-608, 638-639, 666

Chlamydia: 182, 357, 359-362, 364

Endometritis: 333-334, 340

Group B Streptococcus: 197, 344-345, 347, 351,353-354, 356, 565

HIV: 348, 366-369

Preterm L~Ibor/PROM: 6, 10, 13, 17, 46, 68, 112, 192, 203, 213, 217, 318-319, 322, 324-326, 328-330, 563, 565, 570, 57,1-575, 583,591,596, 598-600, 602-605, 608, 636, 654

Intrapartum Fetal Assessment: 38, 41, 58, 77, 108, 114, 208, 252, 320, 438, 621,665,666, 669-673,675-676

Intrauterine Transfilsion: 545-546, 660

Labor: 4, 44, 50, 56, 59, 108, 130, 133-134, 136, 140, 143, 145-146, 148-150, 154, 156, 173-174, 176, 180, 203- 204, 209-210, 229, 244-245, 353,366, 375,442, 558, 620-621,623-625, 627-632, 635,637-639, 642-646, 648-649, 651,668, 671

Meconium: 5, 176-177, 264, 270, 280, 301

498 Subject index January 1996 Am J Obstet Gynecol

Medical Complications: 45, 63, 76, 123, 169, 201,277, 281,283-284, 286, 290-294,296, 298-300, 305, 315, 321,

323,355, 495, 497, 501-502, 510-511, 520, 525, 534, 552, 585, 646, 650

Medications in Pregnancy: 10, 182, 287, 333, 351,363-364, 441,443,506, 586, 672

Antihypertensives: 53, 517, 533

Calciuln Channel Blockers: 517, 579

Cocaine: 286, 441,493

Corticosteroids: 3 l, 35, 70, 83,267, 270, 273,298, 318, 573-574, 576-577, 607

Prostaglandin Inhibitors: 42, 60, 571-572, 624

Prostaglandins: 44, 64, 103,229, 268, 620, 634-635

Tocolytics: 12, 15, 16, 18, 30, 34, 48, 60-61,268, 278, 443, 571-572, 578-582, 604, 633,636

Multiple Gestation: 24, 57, 87, 93, 101,117, 122-128, 231,247, 262, 383, 386, 399, 460-461

535, 553,573, 584, 590, 606, 616-619, 637, 655

OligohydramlfiOS. 85, 247, 249, 280, 385, 559, 562, 626, 668

Other Screening: 107, 307-308-311

Oxytocin: 42, 50, 64, 103, 180, 620, 630, 632, 637, 639, 648

Perinatal Outcome: 2, 8, 11, 32, 34, 36, 38, 40, 43, 58, 63, 70, 78, 98, 106, 116,-118, 121-122, 125, 128, 135,

137, 141-142, 147, 149, 151, 153, 158, 160, 164-165, 175, 181,202, 214-215, 219, 260, 267, 283, 294, 296,

301,317-319, 343-344, 347, 352, 356, 370, 380, 382, 405,423, 431,452, 457, 472-473, 487, 524, 558-560,

569, 580, 582, 586-587, 595, 596-597, 607, 613-614, 616-617, 629, 632, 645,670, 674-676

Perineal Trauma: 36, 62, 148, 151-153, 173-174

Physiology: 29, 51, 56, 77, 178, 185, 188-189, 191, 194, 202, 206-207, 214-215, 218-219, 227, 231,235,

241,255,256, 268, 271-272, 343,444, 500, 525, 537, 539, 641,543-544, 548-549,577, 641,651

Fetal: 52, 54-55, 69, 82-83, 183-184, 216-220, 234, 244, 248-253, 255-256, 258-259, 261-262, 264, 269,

273,276, 320, 352, 493,577, 631,657, 664

Maternal: 5, 47-49, 56, 75, 99, 169, 184, 194, 199-200, 204, 216, 218, 220-221, 223-233,236, 242, 245,

263, 266, 271-272, 277-278, 290-291,304, 306, 324, 444, 504-505, 507, 523,530-531,624, 651

Membranes/Placenta/Cord: 51, 178-179, 181, 183-185, 189, 194-195, 209-210, 248, 258, 266, 331-332,

338, 382, 509, 522, 536, 608, 641,649

Placenta Accreta: 161, 170

Placenta Previa: 8, 119, 121, 161

Placental Pathology: 14, 39, 170, 181, 186-192, 195-196, 205-213,217, 275, 295, 332, 337-338, 368-369, 380,

415, 454, 494, 503,508, 550, 601-603,638, 654, 665

Volmne 174, Number 1, Part 2 Subject index 499 Am J Obstet Gynecol

Polyhydramnios: 246-247

Post-Dates/Prolonged Pregnancy: 85, 139-144

Premalnre Labor: 3, 9, 12-16, 18, 30, 32, 34, 46, 59-60, 67, 102, 190, 203,224, 330-331,561, 563, 569, 571-572, 575, 579-581,583,585, 587-592, 594, 604-606, 610-612, 618-619, 633,648-649

Premature Rupture of Membranes: 4, 6, 10-11, 32, 35, 115-118, 208, 325, 327-328, 381,559-568, 585, 610-612

Prematurity: 3, 6, 11-17, 30, 35, 59, 67-68, 98, 101-102, 113-114, 125, 128, 160, 186, 192, 197, 205, 211-212, 254, 260, 319, 328, 330, 337, 339, 376, 405, 522, 560-561,564, 566-567, 569-570, 576, 578, 582-584, 586- 589, 593-597, 60!-603, 605-606, 609-612, 614, 616-619, 666

Prenatal Care: 8, 43, 91, 95, 97, 99, 112, 116, 121, 132, 144-145, 155, 200, 287, 304, 341,345, 348, 354, 361, 367, 450-451,460, 462, 481,483,534, 576, 589, 592

Psychosocial Aspects of Pregnancy: 63, 95, 97, 112, 254, 283,359, 367, 483, 594, 613

Respirato~2� Distress Syndrome: 129, 222, 276, 293,370, 570, 573,591,598-600

Selective Fetal Reduction: 123-124, 126, 486, 535

Sickle Cell Hemoglobinopathies: 74, 292

Subslance Abuse in Pregnancy: 111, 198-199, 286, 321,355, 441,548, 470, 493,669

Teratology: 1, 19, 447, 455,466, 496-497

Thyroid: 290, 410

Tocolytics: 12, 15, 16, 18, 30, 34, 48, 60-61,268, 278, 443, 571-572, 578-582, 604, 633,636

Ultrasound: 21-22, 61, 89, 92, 111, 131,370, 372-374, 377, 379-381,383-384, 386, 389-391,393-395, 399-401, 411-417, 423,426, 434-436, 445-447, 464-465,470, 490, 593

Estimated Fetal Weight; 66, 139, 404-405, 434-436, 629

Fetal Anomalies: 1, 19, 33, 37, 65, 103, 105-106, 387, 389-391,393-395, 400-402, 408, 414, 417, 421-423, 426, 431,445, 448, 455, 465, 467, 471-473, 475,477-478, 488-489, 492, 494, 498, 543,656

Fetal Echocardiography: 54, 234, 237-238, 394, 396, 399, 402-403,492, 543

Fetal Growth: 98-99, 124, 230, 239-240, 254, 258, 265, 279, 403-404, 425, 427-428, 449, 593

Gestational Age Assessment: 144, 392, 428, 432

Multiple Gestation: 24, 57, 87, 93, 101, 117, 122-128, 231,247, 262, 383,386, 399, 460-461, 535, 553,

573, 584, 590, 606, 616-619, 637, 655

Other: 9, 29, 68, 72, 78, 93, 108-109, 131,202, 253,255-256, 272, 277, 297, 314, 317, 327, 341,374, 392, 409-410, 413,416, 425, 427, 429, 439, 443, 513,530, 539, 636

Technical: 37, 65, 372, 483, 409, 413, 426-429, 434

500 Subject index Janualy 1996 Am J Obstet Gynecol

Vaginal Sonography: 17, 37, 65, 375-378, 392, 557, 584

Uterine Rupture: 120, 163, 166-167, 171-132, 647

Vaginal Birth After Cesarean Section: 120, 154-156, 162-167, 171-172, 628, 647

Author Index

Volmne 174, Number 1, Part 2 Author index 503 Am J Obstet Gynecol

Aal, N" 56 Abbott, OH. 273 Abboud, M 559 Abontafia, Y: 590, 636 AbnLZzo-Fogarassy,M: 457

Abuhamad, A 310, 388-389, 542, 547, 655 Adair, D: 145 Admr, DC: 29 Adamson, SL: 51m 53 Adashek, J: 433 Adashek, JA: 66, 165,291 Addonizio, P: 554 Ader, HJ 12 Adhoot, D: 291 Adler, L: 520 Adra, AM: 423 Adzick, NS: 663 Ager, J" 111,470 Aguero-Rosenfeld, ME: 604-605 Aguila-Mm~silla, N: 493 Alto, MO. 38, 40, 676

Ahokas, RA: 47 Ait-Allah, A: 26,466 AI, M. 263 A1 Anmaar, M: 545-546 A1-Malt, A. 559 Albrecht, J. 56 Ah, Y. 182 Ali-Khan, Z" 458 Allbert, J 376, 566 Allen, J. 203,325 Allen, L: 335 Allen, S. 363 Alsulyman, O 89, 301,387

Alvarez, JG. 525,598-600 Alvarez, M. 461 Amankwah, K: 166

Ambrose, A. 174 Amiet, A: 489

Amlni, S: 559

Amon, E: 559 Anceschi, MM:653 Andersen, HF. 97 Andersen, RN: 228, 531 Anderson, CF: 544 Anderson, T: 247 Andres, RL: 441 Andrew, M: 34, 309 Andrews, WW: 3, 25, 46, 104, 361 Aneed, R: 193 Anfanger, P J: 383

Ansley, D: 433

Antoine, C. 350 ~yaegbanam, A: 292 Apu7~,io, J: 154,459 Aqna, KA’ 640

Ardumi, D. 13

Arefi, L: 545-546 Arias, F: 564

Arikan, G: 571

/~pl, M: 73 Arr~ola, J. 299 Arriola, R: 34 Arsich, J: 195

Artal,R: 59, 315,534 Ascb, DA: 18 Ash, K: 158, 399 Asrat, T: 134,264, 291,345,465 Aston, CE. 33,448 Athanassiou, A: 577 Atk~nson, .MW: 600 Atterbtkry, JL" 98, 255-256

Anbry, R. 534

Audibert, F. 528-529, 532 Avraham,Z: 431 Awaloei, H: 187 Ayala, JA: 180 Baba, D: 579 Bader, TJ: 18 Bahado-Singh, R: 22, 476-477, 655 Baig, S 541 Baja.}, R 16

Baker, DA: 324 Baldinger, S: 486

Batlard, PL 576 Ballard, RA: 576 Bambach, B: 27 Bansal, RK: 173,561 Bar-Hava, I 83

Barad, DH: 457 Barbera, A: 88, 146 Barjot, PH 299 Bamard, J. 259 Bamea, A: 493 Baron, A 221 Barr, Jr, M: 455 Barret, T: 459 Barrett, JFR: 335 Barrett, PIt: 530 Barry, D: 442 Barry, J: 92

Barth, Jr, WH: 637 Barton, JR: 101,619 Barton, L: 387 Bashan, N: 19 Battaglia, FC" 24 ! Batton, D: 595 Banmgarten, A: 476-477 Baxi, L: 526 Bayer, S. 96 Bayhi, DA: 15

Bear, MB. 593 Beasley, D: 570 Beazely, D: 284 Bebbington, MW: 62

Beghard, D: 609

Beguin, F: 463 .Belanger, K: 373

Bel,fort, M: 60, 72, 84, 148, 193,517

504 Author index January 1996 Am J Obstet Gynecol

Belfort, IdA 537-538 Bellotti, M: 88,548 Ben-RafaeI, Z: 83 Beneck, D. 524 Benedetti. T J: 323 Bemrsctkke, K 6 t 7 Bemto, C 80, 130, 181,408,438, 557 Benito, CW: 136,646-647 Bentohla, E: 609 Bentz, LS 255-256 Berbeken, EA: 23 Berg, R: 354 Bergauer, N. 587 Bergauer, NK. 101,619 Berger, H: 539 Berghella, V’ 39%398

Bergmans, M 77 Berkowltz, G: 610-612 Berkowatz, K 222, 432,630 Berkowltz, R’ 612 Bennan, C. 475 Bemasko, JW’ 127, 150 Bemstem, P. 355 Berry, S: 331

Bero,, SM: 257, 575, 596 Besinger, RE 48,580 Beydoun, SN. 423

Beyer, EC: 56 Beyth, Y: 489

Bhandari, V: 219 Bhatia, S: 274 Bialy, G: 226 Bianchi, DW. 424 Bieniarz, A" 381 B~enstock, J: 597 Blllett, H. 292 Bimholz, JC: 404 Bird, lid: 258, 272-273 Birnholz, J. 124 Bischof, P: 463 Bitonti, OM" 332 Blackstone, J: 379,488 Blackwell, S’ 28I ]~lake, PG: 293

Blakemore, K. 27, 112, 149,280,475,558 Blanco, J. 129 Blazer, K 483 Blea, C: 495 Bleker, OP 12 Bloom, S: 58, 177, 416 Bloom, SL" 351 Boardmaa~, C" 576 Bobby, P: 261,437 Bobrowsk~, R" 290, 505 Boden, G: 303 Bofill, JA’ 34, 36, 151-153,309 Bogdan, D. 542 Boge, L’ 504,505 Boggess, KA. 323

Bombard, A: 462

Bombard, AT. 457, 498 Bombardier, C: 70 Bonebrake, J: 122 Bonin, B: 158 Booth, A" 456 Boots, L: 474 Bootstaylor, B 92 Borgida, AF: 131. 170.262, 385,394 Borgida, W’ 393

Bork, M’ 131 Bork, MD. 393-394 Bottoms, DF’ 135 Bottoms, S: 105,569,585-586 Bottoms, SF 35, 160, 260,282, 304, 582 Boucher, M: 61,132,366 Boulvmn, M. 589 Bowen, E’ 227 Bowennan, RA’ 455 Bowling, S. 267 Boylm~, P’ 622 Boyle, J’ 109, 562,659

Brace, RA 183 Bradley, K’ 270 Braget, D. 202

Bragg, EJ 175 Brambat~, B 450 Branch, DW. 161,179~ 296,641 Brandt, D" 486 Braynt, DR: 282 Bremmelgaard, B 73 Briggs, R 168,570 Briggs, RG: 287 Brmkmm~, I11, CR: 314 Brinsden, P: 98 Briones, DL: 29 Britton, K: 140 Brockdnma, D. 229

Broder, M: 483 Brodin, AG: 275 Bronshtein, M. 407, 431,445 Brooks, G: 507-508 Brosens, IA. 23 Brost, B: 81,159,342,652 Brown, C: 669

Brown, CL: 257, 374,469 Brown, E: 507-508 Brown, H: 140,487 Brown, I-RT,: 7, 225 Brown, RC 151, 153 Browne, P; 402

Bnunfield, C: 63 Bmmfield, CG: 471-473 Bruner, J: 247, 656 Bryant, DR: 135 Buchanan, P. 450-451 Budelier, WT: 479 Budorick, NE: 413 Buhur, A: 517 Bullaboy, C’ 542

Bttrks, L: 634

Volume 174, Number l, Part 2 Author index 505 Am J Obstet Gynecol

Burhson, RD 581 Burrus, D 145 Buscagha, M: 241

Bush, J. 249 Buttmo, L 90 Byme, BM" 51, 53

Cabalum, MT" 314, 316 Crone, ME: 278 Calhoun, B: 96, 372 Calhoun, S 385 Callan, FT: 384 Callan, NA: 149 Campbell, I3: 93, !08, 342 Cmnpbell, C: 227 Campbell, D: 107

Campbell, WA: 21, 170,216-219,262,393-394, 411

Camck, JA: 498 Cantermo, J: 318, 377, 524,607 Capeless, E 138, 157 Caplan, M: 520 Caplma, MS 265,275 Capponi, A’ 13

Carb~en, A. 548 Cardomck, E. 285,457 Carelh, MPC 491 Carey, J 31

Canmchael, L 243 Carpenter, !vlW 75, 306

Carpenter, RJ: 482

CalT, M~I: 66,291 Carl SR. 418--419 Carreno, C 57 Carroll,SF. 291 Carter, SM. 457

Casal, D: 180,592 Casal, K 633 Case, KS 161 Casele, H 23I Cassen, EM. 591 Casson, PR: 228 Castro, L: 613

Castro, MA: 301 Cauldwell, C: 663 Cefalo, RC 444 Cejka, J 661-662

Ceklenlak, N. 297 Cemti, P 668

Cetm, I 241 Cetnflo, C’ 128 Chaffm, DG: 259,544 Challis, DE: 551 Chamberlain, C: 391 Chan, DW’ 196 Chan, L 37, 65,425,427-429,555 Chandra, G: 46 Chang, R. 198 Chang, TC: 99, 540,

Chmfl~, TC. 343 Chapman, S: 35, 163-164 Chard, R: 379

Chart, RS: 168 Chasen, S 100 Chatterjee, M: 362

Chelmow, D: 128, 622 Chen, 13. 527 Chen, CL: 370 Chen, X: 303 Chervenak, F: 123-124,404 Chik, L: 371,669 Chiu, TH: 543 Cho, K" 564 Choi, JH: 68 Christensen,H: 3 I Chu, GP: 65,427-428 Chung, H: 249 Chung, YK" 286,514-515

Cifuentes, R: 563 Cmffi, F: 226 Clark, A: 442 Clark, AL 660 Clark, BA: 525 Clark, KE’ 271 Clark, P: 344 Clark, SL: 5, 32, 49, 87, 161,232-233,492, 621,650,670 Clavin, D: 227 Clawn, DK: 15 Chver, SP. 3 Cnaan, A: 576 Coata, G: 322 Coburn, C: 576 Cochen, D: 158 Codipilly, CN 324

Cohen, B’ 433,599 Cohen, BF: 307-308 Cohen, H: 311-312 Cohen, J: 330-331,453 Cohen, RA. 75 Cohen, WR 623 Cohlan, B: 559 Colden, KA: 469

Cole, C: 577 Coleman, E: 634

Collea, J. 102, 245,443 Collins, D: 372 Collins, JC: 283 Colhns, JE: 378 Comfort, M’ 285 Comport, K: 278

Comstock, CH: 395-396,400,490 Conway, D: 42, 71,620

Cook, C: 11, 167, 567 Cook, V’ 11,341,442,567 Cooper, RL: 46 Copel, J: 476-477, 545,547 Copel, JA: 196, 286, 355,380, 514-515,546,555,655

Copen, J’ 22 Copper, R: 2, 583-584 Copper, RL: 17 Cordero, L: 519 Cork, RC: 15

506 Author index JanuaE¢ 1996 Am J Obstet Gynecol

Coronado, C. 148 Corson, V: 27 Cortma, R’ 388 Costal, EV: 653 Cosper, P. 471-473 Cotton, DB" 147,232-233,257, 277, 281,504-506, 537-538, 575 Coulson, C" 444 Coultnp, L: 133,563 Cox, S: 9 Cox, SM" 351,574 Crane, S: 534 Craven, C: 494 Creasy, RK: 633 Cnspens, MA: 163 Cro~toru, D. 388 Crombleholm, T 28 Crowe, S: 93 Cullen, M: 445 Cullen, ML: 663 Cmmnings, B’ 97

Cunningham, FG: 574 Curet, L 169, 560 Cusick, W. 90, 393 Cutt~.ta, F: 184 D’AI~op_, ME: 128,424, 577, 622

D’Oria, P 668 Dado, GM: 185 Daemen, M 178 Dan, I: 636 Daniel, Y: 645 Danilenko-Dixon, DR: 75 Darag~ati, C: 198 Das, A: 2, 6, 17, 584 Davee, M: 487 David, MP. 645 Day,s, N: 376

Davis, RO: 471--473 Davoli, E’ 241 Days, L: 55

Dax, J 9 Day, D: 375 Day, L: 269 Day-Salvatore, D: 408-409,412,449 De Comenico, P: 322 de Veciana, M: 310 Degani, S: 252, 539

De’kker, GA’ 12,254, 516 Del Valle, G: 44 Delke, I: 44 Dell’Atma, D: 536

Demasio, K: 350 Den~nitt, J: 357 Dmme, S: 221 Deprest, JA 23 Deren, O: 22,476-477 Desjardins, C: 399

Devidas, M: 36, 152 DeVnes, KL: 401 DeYoung, A: 659 Di Iorio, R: 653

Di Renzo, G-C: 322,420 Diamant, Y: 590 D~mnant, ZY: 636 Diaz, FG: 119 Dickinson, JE: 105 Dildy, GA: 32, 87, 492, 650, 670, 674 DiLeone, G: 306 Di~ranitt, J’ 359 Diogo, M: 10 Divon, M: 550 Divon, MY: 69, 94, 141,188-189, 235-235, 261,437, 609 Dizon-Tovmson, D. ! 10, 179,503,641 Doany, W. 143 Dobrin, PB: 185 Doi, S: 250-251 Dombrowski, M: 290 Dombrowski, MP: 257, 321,596 Dom~enfeld, A: 453 Dooley, S: 231

Dorman, K: 60 Dowd, T: 415 Dozier, B: 288 Dresden, S: 156 Drgml, A: 407

Droste, S 496 Dnlgm~, A. 469

Druzin, ML. 527 DuBard, M: 25, 46, 164, 435, 471,473,583 Dubey, A: 577 Dudenhausen: JW: 673 Dudley, D: 592 Duff, P: 344 Dugofl; L: 417

Dullien, V: 59 Dunkel-Schetter, C 613 Dunston-Boone, G. 378 Dupra-Wilson, P 573 Dye, T: 534 Dykman, RA¯ 255

Dzmczkowsk~, J: 290 Easton, K: 110 Ebrahim, SAD: 406 Ecker, JL: 173 Edelstone, D: 340 Edwards, C: 458 Edwards, M: 507-508 Edwards, MS: 134,581 Efrat, Z: 83 Egret, J: 80 Egan, J-FX: 131,262, 385,393-394, 411 Egennan, RC: 228 Egennm~, RS: 300,672 Eglinton, GS: 7, 14, 102, 205,214-215,225,603 Elu-emkrmg, RA: 476

Eis, A: 229 E1-Sayed, Y. 579

Elderkin, R: 167 Etdor, A: 512 Eller, D: 81,342 Ely, JW: 95

Volume 174, Number 1, Part 2 Author index 507 Am J Obstet Gynecol

Endicett, B: 446 Eng, i~: 262 Engle, W: 574

Elfrat, Z: 235 Enksen, NL: 129, 367 Eriksson, M: 118 Ernest, J: 145 Ernest, JM: 615 En’st, L’ 170,217

Ernst, LM: 39,208,212, 601 El vin, MG: 269 Eschenbach, DA 323,591 Esplin, SE: 333 Eucker, B: 444 Evans, A: 388-389, 542 Evans, AT: 310 Evans, M: 171,663 Evans, MI: 20, 57, 406-407, 467-469, 482, 661-662 Evrard, VA’ 23 Extennam~, P 463 Eylath, U. 590 Fabbri, EL 394 Facchinetti, F: 533

Fair, G: 645 Faix, R: 197 Falconer, J: 657 Fall, O: I 18 Fanelh, R: 241 Farah, L: 44 Farhat, MY: 7, 225 Farme, D: 4, 70,106, 116-117,298,311-312,335,552 Farmer, K: 634 Faron, G 589 Fausett, MB. 637 Fauvel, M: 366 Fee, S 354 Feeney, LK 394 Fejgm, MD: 489

Fehx, J 484 Feng, T 658

Ferland, J: 306 Fernandez, CO. 58 Ferrazzi, E 88, 146, 548 Fey, J¥" 378

Fmld, N: 42 Field, S J" 257 Fxelding, L. 70 F~gueroa, R: 67, 604-606

Fmk, N: 399 Finlay, T. co-author on research presented is for the SPO

Fmmdation Fellow Award for Academic Year 1994-95 Finley, S: 471

Fisher, S. 580 Flake, A: 422 Flake, AW: 663 Fleisher, J" 144 Fleming, WH: 658

Florio, P: 610 Fong, J 439

Fontenot, MT" 134,581

Fontenot, T: 93, 107-108 Foreman-v Drongelen, M: 263

Forest, JC: 199 Forouzan, I: 553 Fortunato, SJ: 608 Frangich, AY: 528 Frangieh, AY: 386, 532, 568

Fraser, W: 155, 625 Freedman, A: 661-662 Freeman, R: 264,465

Freeman, RK: 66 Fricke, J: 568, 570 Friedman, SA: 47, 386, 528-532, 616 Friedmalm, W: 673 Frye, A: 145

Fuentes, A: 327

Fulton, A: 479, 481 Furman, B: 220 Gabbe, SG: 519, 562 Gaddipati, S: 661 Gakman, R: 220 Galetta, J" 385

Gall, S: 341 Gallagher, M: 184 Galloway, DA: 43

Gandhi, J: 403 Gandhi, J: 629 Garcaa, A: 180 Garcm, P: 565 Garcm, PM: 122, 368 Gardener, M" 142, 363,560 Gare, D: 311-312 Garite, TJ: 10, 180,433 Gannel, SH’ 28 Garry, D: 24, 67, 604-606 Gauthier, D: 381 Gazit, G: 630 Geary, M: 622 Geifinan-Holtmnan, O: 499

Gelertner, N: 436 Gell, JW: 147 Georgiou, S: 356

Gerber, MA: 308 Gennain, AM: 648-649 Ghezzi, F: 319, 327, 331 Ghidini, A: 14, 39, 113, 115, 192,208, 212-215,338-339, 601-

603,654, 665, 667-668 Gim~opoulos, JG: 580

Gibbs, R: 594 Gibson, M: 195 Gielen, A: 112, 558 Gilbert, WM: 183, 631 Giles, W" 657 Gill, RW: 551 Gilles, JM: 69, 609 Gilstrap, L, 177, 35l Ginsberg, N: 460

Girz, B: 94 Gittens, L: 154 Giuliani, A: 244

508 Author index January 1996 Am l Obstet Gynecol

Gmseppe, L: 420

Gleason, D: 140 Gleicher, NA: 90 Glover, D: 195 Godmilow, L: 452 Goepfert, AR: 103-104, 586

Gohr, A: 346 Goldberg, JC: 378 Goldberg, JD: 482 Goldenberg, R: 2, 6, 142, 435, 583-584 Goldenberg, RL: 17, 35, 46, 260,582, 586 Goldman, D: 43 Goldstein, I: 426, 430-431 Golichowski, A: 221 Gomez, FL: 429 Gomez, JL: 627-628 Gomez, K: 655 Gomez, R: 327, 329, 331,453,549 Gomillion, M: 439 Goncalves, L: 57 Gonik, 13: 79, 281,290,332, 337 Gonzales, C: 31 Gonzalez, D: 459 Gonzalez, E: 508 Goltzalez, .rL: 560 Gonzalez, R: 549 Goodwin, TM: 59, 294 301,326 Goolsby, L: 624

Gordon, MC: 637 Gonnaz, G: 648 Goustin, AS: 369 Gouws, F: 45 Grable, I: 565 Graham, E. 553 Granovsky-Grisam, S: 298, 552, 636 Grant, S: 479, 481 Gratton, RJ: 33,448 Graves, C: 76,510-511 Gravson, M’ 202 Gray, K’ 184 Gray, SE: 647

Gr~iano, C: 420 Greb, A: 467 Greenberg, S: 227 Greenberg, SG: 271 Greenhagen, JB: 592 Greenspoon, J’ 353

Greenwell, ER: 660 Greig, P 203,325 Gribble, R: 354 Griffin, C: 27 Griffin, M’ 560 Grisam, D 512 Grisam, D’ 636 Gnsam, S: 106 Groome, LJ: 98, 255-256 Gross, S: 462 Gross, S J: 457 Guerra, F: 359

Guillenno, L: 316

Guinn, D: 585 Guitm, DA: 104 Gull, I: 645 Gurczynski, J: 401

Gurewitsch, E: 439 Gurski, M: 507 Gurski, M: 508 Guz~nan, E: 80, 130, 181,408,412, 438, 557 Guzman, ER: 21,411 Gviazda, I: 294 Haas, MJ: 347 Haeusler, MCH: 671 Haglund, 13:141

Hallahan, T:450-451 Hanadi, A: 545-546 Hmmnerman, C: 590 Hamner, lII, L: 402 Hanani, M: 636 Hawkins, C: 366 Hm~kins, GDV: 5, 49,232-233,650 Hanley, ML. 162, 557

Harem, GV: 149 Itam~ah, M: 4, 70, 571-572 Hansen, SL: 621 Hardardottir, H" 131,170, 198, 385,393-394 Harman, C: 41 Haq~er, CE: 490

Hams, D: 64 Harris, DT: 456 tlarrison, MR: 663 Harvey, C: 5, 49, 232-233 Harvey-Wilkes, KB: 279

Harzke, C: 624 Hasaart, T: 77 Hass, J: 671 Hatangadi, S: 16

Hatten, B: 486 Hanth, JC. 35, 46, 103-104, 164, 260, 582, 585-586 Haydon, B: 75 Haynes De Regt, R’ 378

Hedriana, HL: 183, 631 Iteine, RP: 521 Heinen, AGJJ: 254 HeIlZand, B: 126,485 Helfgott, A: 367

Helmbrecht, GD: 7, 225 He~nstreet, GP. 25 Hcndricks, SK: 273,495-496 Herbert, W: 203,325 Herlicoviez, M: 299 Herrschap, A: 253

Hcrshkovitz, R: 19 Herzog, M: 341 ttewson, S: 4 Hiett, A: 140 Hill, LM: 382 Hillier, SL: 591 Itinton, RY. 200 Hirano, T: 250 Hirshberg, J: 297


Hltti, J: 591 Hobbms, JC: 146, 373,391,417,447 Hobel, CJ: 593~ 613 ttodnett, E: 4, 155 Hoffman, llI, C: 438

Hol"tinmm, M: 45 Hogge, WA: 33,448

Hohimer, A: 55 Holbro~Jl:, JR, RH: 579 Holden, A: 594 Holland, SB: 255-256 Holman, R: 230 Hohnes, KK: 43 Hohnstead, M: 200 Homan, J: 243 Homko, C: 1,302-303 Hong, SA: 319 Hong, SF: 514-515 Hoppe, S: 594

Homstra, G: 263 Horton, C 389 Horton, S: 389 Hoskins, LA: 627-628 Houlihan, C: 438 Houwehngen, AV: 263 How, H: 11,567 Howard, RB: 51, 53 Howard, S 373 Howell, K: 560 Howser, DM: 95 Hsieh, CC. 99 Hs~eh, PC. 543 Hs~ell, TT 99,543 Hsu, CD: 196, 286,514-515 Hsu, J 435 Huddleston, J: 346, 658 Hughes, S: 341

Huhta, JC" 54, 234, 397 Hume, RF. 372, 4-6,468-469, 662 Humphrey, A’ 362

Hunter, C: 592 Hunter, D: 22 Hutson, J: 535 Hutson, M: 123-124 Iams, JD 2, 17, 405, 562, 569, 584 Iammcc~, TA: 580 Ibrahim, M: 194 Ikeda, T 251

hnsels, J: 581 h~glis, SR: 360, 588 hlSUnZa, A: 549 Inon, O 199,589 Iriye, BK’ 66,291

Itskovlcz-Eldor, J 235,407 Iverson, C: 563 Izqnierdo, JC: 180

Jackson, GM: 59, 87, 91

Jacksou, L: 20 Jacques, D: 587 Jacques, DL: 101,619

Jacques, J: 337 Jacques, SM: 79

Jadali, D: 292 Jaekle, RK: 324,370, 384 Jaffa, AJ: 645 Jeanty, M: 318, 524, 607

Jelsema, R: 167 Jeng, G: 161 Jenkins, CB: 14, 189, 214-215,603

Jense, CP: 336 Jensen, IP: 73,336

Jetme, B: 73, 336 Jiminez, G: 180 Jinunerson, C: 203,325 Jiyamapa, D: 368 Johnson A: 446 Johnson DD: 413

Johnson F: 2, 17, 584 Jolmson J: 44 Johnson JM: 478 Jolmson !vIA: 617 Johnson, MP: 20, 57, 406-407, 467-469,482, 661-663 Johnson, SC: 282 Johnson, TRB: 197 Johnston, JM: 649 Johnston, K" 583 Johnston, KE: 25 Johnston, L: 243 Jones, D: 655 Jones, J: 587 Jones, P: 405,569 Jones, R: 27 Jones, TB: 369 Jongsma, HW: 674

Jordan, A: 614 Jonlsay, D: 541 Joseph, M: 81 Joseph, MA" 378

Joshi, R: 629 Judd, VE: 492 Jun, JK: 68, 319, 328-330 Jurcovicova, J: 508 Kadner, S: co-author on research presented is for the SPO

Foundation Fellow Award for Academic Year 1994-95

Kainer, F: 671 Kaltenbach, K: 285 Kambich, M: 485 Kanayama, M: 230 Kang, A: 510

Kao, L: 530-531,616 Karinfi, A: 484 Kannel, C: 651 Karpel, BM: 324, 384

Kato, S: 648-649 Katz, N: 94 Kaulinan, G: 128 Kaufmmm, M: 397 Kay, H: 204

Keasler, S" 50, 638-639 Kellner, LH: 498

510 Author index Janua~Ty 1996 Am I Obstet Gynecol

Kelly, E: 116-117 Kelly, EN: 106 Kelly, MA: 574 Kemp, P: 493 Kelmedy, M: 659

Kelmedy, RC: 343 Kern, I: 244 Kerr, J: 385 Kestin, A: 306 Keunen, H: 77 Khan, R: 545-546 Khandelwal, M: 37, 428, 554 Khoury, JC: 305 Khouza~ni, A: 27, 112, 558 Kidron, D: 489 Kilpatrick, SJ: 173

Kiln, CJ: 319, 328,330 Kiln, IO: 68 Kiln, J: 109 Kim, JL: 276 Kim, JO. 329 Kim, S: 245 Kiln, Y: 245. 443

Kimberlin, DF: 260, 582,585 Kincade, T: 309 King, E: 181 King, IB: 501 King, J. 102 King, R’ 166 Kinneberg, K: 486 K~mley, S’ 179 Kirk, JS. 395-396, 400, 490 Kithier, K: 661-662 Kitzman, DR. 237-238 Kivlat, NB 43 Kjeldgaard, A: 664 Kjos, S: 222,432

Klas, J: 391 Klein, S’ 67, 317-318,624,607

Klennnan, G’ 267 IZdmppel, R. 16,408, 412 Knuppel, RA: 136, 646 Kobo, M: 83 Kolmen, G: 178 Koiphlnan, A: 220 Kolderup, LB: 137

Konstantiniuk, P: 244 Kopczynski, T: 566 Kopehnan, J: 372

Komfield, I" 166 Korst, L: 38, 676 Korst, LM: 40, 172 Kosmas, E. 499 Kosler, J: 446 Koutsky, LA: 43 Kovacs, BW. 387. 484 Kowalik, A: 123-124,535 Koyle, M: 391 Kramer, R’ 169 Kramer, W: 60, 84, 148

Krantz, D: 450-451 Kraus, F: 564 Kreindler, S: 590 Kreitlow, S: 133 Krolm, MA: 591 Knleger, S: 507-508 Kuboshige, J: 314, 316

Ktddman, K: 398 Kullama, LK: 269-270 Ktmg, M: 410

Kno, DM: 543 Kupfenninc, M J: 520, 645 Kupsky, WJ: 538 Kurtzberg, J’ 204

Kuypers, J: 43

LaBella, C: 340 Lachelin, CGL: 59 Ladfors, L: 118 Lage, JM: 186, 191,205-207, 338 Lagrew, DC: 66, 165,421,465

Lai, Y. 80 Lake, M: 80 Lake, MF: 136. 162. 646 Laks, M: 462 Laksluni, V: 509 Lambers, DS: 271 Lameier, L: 107 Lampinen, J: 406, 469 Lampley, EC: 518

Landon, MB. 519 Landweltr, JB: 135,282, 467-468, 575 Lange, FM: 12 Langer, O: 42, 71,313,356, 620

Lam~i, SM: 174 Lalming, J: 142

Lanouette, JM: 257, 401,467, 596 Lapinski, R: 461,610-612 Lapointe, N: 366 Laros, RK: 137, 561 Larrabee, KD: 74, 367,441,633

Larsen, J 450--451 Larson J: 634 Larson, L: 289, 306 Laskin, C: 70, 298 Lasser, D: 526 LaTorre, R: 653

Laurent, S: 566 Lauria, MR: 79 Lauro, R: 536 Lavie, O: 590, 636 Lazebnik, N: 382.449 Leaphart, L: 138

Leavitt, M: 502 Lee, H: 361 Lee, J: 578 Lee, S-K: 43 Lee, W. 395-396, 400, 490 Leeth, E: 485 Legault, C: 240

Leibennan, JR: 19, 220

Volume 174, Number 1, Part 2 Author index 511

Am J Obstet Gynecol

Leiken, E: 156,317

Leiva, MC: 297, 452 Lelnke, M: 635 Lencki, S" 205 Lenke, R_R: 390-391

Leonardi, MR: 135, 160,282, 304, 467, 575 Leporrier, M: 299

Len~t, T: 23 Lesis, R: 568 Leslie, KK: 201 Lessing, JB: 645

Lester, C: 209-211,638 Lev, S. 83 Levaltier, J: 299 Leveno, KJ: 9, 351,574 Levy, G: 299 Levy, J: 220 Lewinsky, RM: 252, 539

Lewis, D 108. 507-508 Lewis, DF: 134, 581 Lewis, R: 30, 386, 529 Liberati, M: 83 Librizzi, R: 297, 452 Lin, WM: 428 Lmdbloom, 13 236 Lindsay, M: 614

Lira, P: 327 Little, BB: 9, 177, 493 Littlefield, R: 372 Lin, Y-L. 268 Locatelli, A: 668 Lochbaum, L: 7, 225 Lockhart, C: 134

Lockwood, C: 610 London, SN: 134 Longo, S: 264,465 Lopez, G: 182 Lopez-Zeno, JA: 522, 601

Lorenz, RP: 396,490 Loucks, CA: 670 Lovelt, S. 10 Loyd, G: 442

Lucas, M’ 415-416 Ludmtr, AG: 525 Ludmir, J. 525,598-600 Ludom~rskL A: 37, 54, 65,234, 427429, 546. 554-555 Lnke, B 125 Lukes, A. 3 Ltmt, C: 202 Lnpo, VR. 133, 198 Luthy, DA. 347, 500-501

Luttkus, A: 673-674 Lyden, TW: 187 Lye, S: 56 Lyman, WD: 609 Lynch, L: 127, 150,461 Lynch, S: 195 Lyons, C: 597 Lysikiewicz, A: 24, 156, 377

Mabie, WC: 284, 287

Macedonia, C: 372 MacGregor, S: 485,578

MacKay, DH: 278

Macones, GA: 18, 364,440 MacPherson, C: 35,260,405,569,582, 586

Macri, CJ: 184 Macri, J: 450451 Magmm, EF: 383 Maghak, B: 475 Magness, RR: 258,272-273

Magnuson, SR: 266 Magriples, U: 196, 355,380 Magtibay, P: 64

Maher, J: 267 Mahoney, J: 22, 476477

Maidman, J: 403,629 Major, CA: 307-308 Major, H: 503,641 Makhail, LN: 497 Malka, E: 226 Malloy, M: 577 Mallozzi-Eberle, A: 425

Malone, FD: 622 Mammen, EF: 454 Mandel, F: 70 MamL SE: 248 Maturing, F: 41 Many, A: 382 Marangi, M: 322

Marbach, M: 220 Marconi, AM: 241 Marcoux, S: 625 Marcovina, SM: 501 Mares, A: 72

Mari, G: 388, 545-547, 549, 555,655 Mariani, S: 668 Markliag, EK: 664

Markson, L: 365 Marley, E: 338 Marquette, GP: 61,132

Martens, M: 349 Martier, SS: 111,321,470, 669 Martin, C: 55 Martin, D’. 560 Martin, GI: 38 Martin, JG: 382 Martin, L: 414-415 Martin, RW: 152-153,293,618 Martin, Jr, JN: 36, 152-153,276, 293 Martinez, A. 184 Martinez, E: 67, 555,604-606 Martinez, MA: 327

Martins, M: 557 Martins, ME: 409 Maslow, AS: 372 Mason, BA: 281,505,537-538, 663 Matsuura, M: 250

Mattsson, LA: 118 Maunsell, E: 155 Mayer, DC: 444

512 Author index lanua,y 1996 Am J Obstet Gynecol

Mayes, M: 60 Mayfield, J: 358 Mays, J’ 24

Mazor, M: 19, 220, 328, 331,453 McCalla, C: 509 McCalla, S: 144,249 McCarton, CC 69 McClean, D: 557 McClurkan, IVIt3:383 McDonald, FD: 147 McEvoy, C: 267 McGregor, JA: 59 McKindsey, F: 626

McLean, DA: 21,162,409 McLean, LK: 482 McNamara, H: 16 McNamara, M: 337 McNeil, S: 239 McNelhs, D: 2, 6, 17, 35,405,569, 582,584-586 McParland, P: 106 Medeans, AL: 593 Meier, P. 354 Meis, PJ: 2, 17, 584,615 Meller, R 73 Menard, M: 652 Menon, R: 608 Menssen, JJM: 674 Mentakis, A: 314 Ment~coglou, S: 41

Mercer, BM. 2, 6, 17, 30, 168,386, 4C5,568-570, 584 Merenstem, G: 6 Metzger, B: 231 Meyer, BA: 370, 384 Meyer, M: 138 Meyer, W: 381 Mezzopane, R: 88

Michaud, J: 488 Mikhail, MS: 292 Mill, JF: 14, 50, 189,209-211,603,638-639 Millar, LK: 8, 121 Miller, D: 86, 119-120, 294 Miller, E: 309 Miller, F" 86, 107-108 M~ller, HS: 456,640 Miller, JM: 15 Miller, M J" 184 Milligan, D 93,107-108 Mdsaps, P: 237~238 Milwltsky, U: 590

Mimon, E: 19 Mmg, P" 466

Mmior, VK: 39, 94, 113-115,188-189, 192, 208, 212,339,522, 550, 601-602,654,665-667

M~kkoI1; H: 144,249 Minton, SP. 32 Minudri, R: 323 Mmdow~ik, M: 6, 175,305 Mirabile, Jr, CP: 370 M~tchell, M: 592 Mittendorf, R 497

Moawad, A: 2, 17, 584 Mohamed, O: 466 Mohapatra, A: 107 Moise, Jr, K: 60, 84, 148, 651 Molina, R: 402 Moline, L: 110 Moll, K: 446 Moller, BR: 336 Molsted, K: 73 Mondion, M: 549 Monga, M: 7% 367, 441,633 Monteagudo., A: 392

Monteiro, H: 89 Montoro, H: 387 Moodley, J: 45

Moody, T: 184 Mooney, DM: 255

Moore, ML: 615 Moore, TR: 82,413,617 Morgan, J: 310 Morgan, M: 576 Morgan, MA: 440

Morrison, JC: 34, 36, 151-153,309, 383,618 Morrow, RJ: 51, 53,335, 478 Mosby, V: 386 Moser, H: 27 Molt, CW: 361 Moutquin, JM: 155, 199 Moy, F: 123,249

Mrkvicka, R 185 Mullen, T. 435,520 Muller, G 299 Muller, JP. 373 Munoz, H: 68 Murata, Y: 250-251 Murphy, K: 623 Murray, HG: 52

Murtha, A: 203,325 Musci, TJ: 137 Myatt, L: 229 Myers, SA: 518 Myhr, T: 4 Myles, T: 381 Nagata, N: 250-251 Nageotte, M: 345, 465

Naidu, S: 45 Nasseri, A: 94 Nathan, L: 288 Nathaniels, P 16 Naylor, D: 311-312

Neerhof, MG: 266,275 Neggars, Y" 583

Nelson, K: 63 Nelson, L: 91,110 Nelson, S: 535 Nelson, TR: 413 Neri, I: 533 Nerrhof, MG: 265 Nesbitt, T: 204 Neuer, M: 498


Newman, A: 259 Newanan, AT: 544 Ne\vmm~, R: 93, 159, 342, 652 Newton, E: 357, 359, 594 Newton, ER: 334 Ney; J: 486

Ne~nan, S: 177 Nguyen, T: 381 Nichols, CR: 105 Nicolaides, K: 263,445

Nicolini, U: 263 Nielsen, H: 577 Nijhuis, JG. 674 Nijland, MJM: 248, 270

Nisell, H. 141,236 Nitowsky, HM: 457, 462 Noga, S: 27

Nolan, T. 227 Nolan, TE’ 15 Nores, Jr., JA: 418-419,424 Normal, G: 569 Norman, G: 405 Nugent, CE: 455 Nuwayhid, B: 381

Nwosu, UC: 268 Nyberg, D. 445

O’Bnen, J: 663 O’Brien, JM" 101,619 O’Ca!laghan, S: 657 O’Campo, P’ 112,558 O’Day, MP 434 O~Donnell, C: 28 O’Reilly-Green, C: 85, 139 O’Shauglmessy, R: 659

O’Sulhvan, E: 129 Odom, MF. 149 Ogasawara, KK. 326

Ogbum, Jr., P 64, 171,230

Olmna, S. 92 Ohlsson, A 4 Olofsson, P: 545-546 Olov Lunell, N: 236 Olsechuk, C 436 Olson, G: 651 Omar, H 195 Oncken, C: 198 Orsino, A: 56 Osathanondh, R: 498

Oshiro, BT: 333,633 Oskmnp, M: 116-117

Ossandon, M: 14, 186, 191,207, 209-211,214-215,638 Otterblad, P: 141 Ouzoun~an, J’ 89,301,387 Ovalle, A. 327 Owen, J: 35, 103,163-164,435,471-474, 585-586 Oyarzun, E: 549 Paarlberg, KM: 254

Padilla,L. 487 Pagotto, L: 397

Palmer, K: 487

Palo~nak], GE: 498 Pm~, Y: 379, 488 Panter, K: 571-572 Pantuck, CB: 223-224 Panyutich, AV: 521

Paolini, C: 241 Papatsonis, DNM: 12

Parchment, A: 150 Pardi, G: 88, 146,241,548 Parer, JT: 576, 675 Parisi, VM: 370

Park, KH: 68, 319, 328-330 Park, S: 251 Parkash, V: 186, 191,205,207, 338 Parker, Jr., CR: 320 Parry, S: 631

Passchier, J: 254 Patel, K: 67, 145,604, 606 Patel, V. 534 Patience, T: 372 Patterson, BK: 368 Patterson, P. 63

Palton, DL: 591 Paul, R: 86, 294, 301,405, 569 Paul, RH: 8, 119-121,632 Pausner, D: 645 Payne, AJ: 45

Peace~nan, AM: 3, 122 Pearhnm~, M: 197, 352 Peeters, LLH: 178 Pemberton, P: 105 Pennacchi, L: 322 Perorate, G: 548 Perusing, S: 433 Pe~uy, M: 246

Penzias, A: 128

Perdue, S: 357, 359, 594 Perle, LV: 506 Perlman, E: 27 Perry, D: 565 Perry, Jr, KG: 36, 293,383,618

Persutte, WH: 390-391,447 Perugino, G: 146

Peters, S: 348 Petraglia, F: 610 Petrikovsky, B: 436, 541

Petterkin, C: 144 Peyser, MR: 512, 645

Pezzullo, JC: 14, 39, 206, 208, 212-215,338, 348, 522, 550, 601, 603,665

Phelan, JP: 38, 40, 172, 676 Phernetton, TM: 259,272-273 Phillipart, AI" 663 Piacquadio, KM: 82 Pijnenborg, R: 186, 191,206-207 Pinckert, T: 102, 348

Pinette, MG: 379, 488

Pinette, SG: 379, 488 Piper, JM: 42, 313,356-357, 359, 620

Pistoia, L: 102

514 Author index January 1996 Am l Obstet G-ynecol

Planer, BC: 576 Podratz, K: 171 Polm~eczky, M: 360, 535

Pollard, JK: 157 Pombar, X: 146 Port, R: 578 Porter, TF: 32, 87, 650 Porter, W: 107 Porto, M: 66,433 Posner, S: 345 Powers, D: 227 Powriex, R: 289 Pratt, L: 273,496 Prentice, E: 496 Press, R: 439 Presser, SC: 588 Pressman, E: 280,475,597 Press~nan, EK. 149 Pretorius, HD: 413 Prieto, JA: 129,521 Prince, C: 610-611 Promecene, P 434 Puder, K: 290, 401,596 Puhjaal, V 462

Qualls, C. 169 Queenan, J: 102 Quilligan, EJ: 251,675

Quintero, R: 661-662

Qumtero, RA: 57 QureshL F: 337 Qureski, F: 79 Rabello, Y. 6, 86 Ragm, A. 126

Ralunan, F: 545-546 Ralston, KK: 660 Ramin, K: 64, 171 Ramin, S: 177 Rmmrcz, MM: 633 Rmnirez, R: 195 Ralnos, J" 202 Ramos, R’ 359 Rmnsey, DK: 493 Ramsey, R 6

Rmnus, R 358, 414416 Ramwell, PW: 7, 225

Rand, WM" 577 Randall, R: 421 Ransom, SB: 147 Ranzim, AC 410,449 Rappaport, V: 483

Rasa-Kublickiene, K: 236 Rasanen, J: 54, 234, 554 Rashbaum, WK: 609 Ray, V 306 Raybon, B: 349 Raybun~, W: 31,634

Read, J: 93, 107-108

Read, M" 657 Reale, M. 67, 318, 607 Reece, EA: 1, 19, 26, 37, 65,220, 302-303, 425429, 466, 554-556

Reed, KL: 259, 544, 624 Reguero, W: 476477

Reich, H: 5? Reimbold, T: 180, 307 Reisner, DP: 347 Reman, O: 299 Resnick, S: 578 Resta, RG: 500 Reyes, C: 222 Rezafikov, L: 201

Riccabona, M: 413 Ricci-Goo&nan, J: 354

Rice, PJ: 268. 614 Richards, W: 656 Richardson, B: 243 Richardson, DK: 600 Richardson, M: 438 Rigano, S: 88 Rigaud-Echols, S: 92 Rigby, F: 227 Rigby, FB: 15 Riggs, TW. 395,400 Rijhs~nghani, A: 274,481

Rinaldo, D: 13 Rindfusz, D: 348 Rilffret, D: 61,132 Ringden, Q: 664 Risley, C: 344

Ritchie, JWK: 116-117, 552 Ritchie, K: 70, 311-312 Rizk, B: 98 Ri~zo, G: 13 Roberson, JR: 454 Roberts, A: 419 Roberts, C: 595 Roberts, J: 2, 6, 17, 144,405,462, 569, 584 Roberts, WE: 151,293,618 Robinson, CA: 440 Robinson, RP: 377 Roden, WJ: 361 Rodis, JF" 21,216-219, 262, 393,411 Rodriguez, C: 412

Rolmer, D: 365 Roitlnan-Jolmson, B: 203,325 Rojas, I: 549 Rokey, R: 72

Romanini, C: 13,536 Romero, R: 57, 68, 319, 327-332,453,575 Ronzoni, S: 241 Rosa, C: 44 Rose, R: 659 Rosenberg, JC: 409

Rosene, K: 306 Rosene Montella, K: 289 Rosenfeld, CR’ 574 Rosem~, BM: 175,305 Rosem~, M: 364 Rosenwaks, Z: 123-124, 535 Rosnes, J: 246 Ross, B: 270

Volume 174, Number 1, Part 2 Author index 515

Am J Obstet Gynecol

Ross, MG: 248,269-270, 626

Rossiter~ P: 475 Rote, NS: 187 Roth, N: 440 Rotmensch, S: 83 Rottem, S: 445 Rouse, DJ: 142, 363 Rowlm~ds, S: 295 Rubio, R: 327 Ruddy, M: 226 Ruggiero, R: 182 Runmey, PJ: 291,345

Russell, E: 504 Russell, T: 126, 485,520 Russo, LR: 48 Rust, OA: 34, 36, 151-153,309, 618 Rutherlbrd, SE: 464 Ryan, G: 106, 116-117, 335,478, 552

Saade, G" 60, 72, 84, 148, 193,517, 651 Sadaat, P: 222 Sadovsky, Y: 56

Sahavi, Z: 83 Salafia, CM: 39, 50, 113-115, 186-192, 2(15-213,338-339,522,

550, 601-603,638-639, 654-667 Salama, M: 30 Salmneh, W: 26 Salari, V. 371 Saleh, AA: 454 Salem, H 466 Saller, DN: 498 Saltzman, D: 92 Salnadi, R: 294,353,432 Samsou, J: 366 Smnuels, P: 109, 519 Sanchez-Rmnos, L: 29, 44 Sm~dberg, P: 340

Sm~ders, MM: 170 Sm]tos, A: 261,437 Santos-Ocampo, C: 577

Santos-Ramos, R: 415

Samo, A. 78 Samo, AP: 672 Satiu, AJ: 637 Savage, K: 163-164

Savitz, D. 611 Schenkel, C: 154

Schiff, E: 530-531,616 Schlecht, KG: 456, 624 Sclmeider, D: 542 Sctmeider, E: 541 Sclmeider, J: 495 Scholl, W: 244 Schorr, SJ: 151, 153,618

Schucker, JL: 529, 672 Schumacher, B: 651 Schwartz, W: 31 Schwarz, K: 230

Schweikert, S: 82 Scorza, W: 130 Scorza, WE: 647

Scott, JR: 296 Sea,yard, G: 552

Seaward, PGR: 116-117 Secbeck, M: 523 Sehgal, P: 67, 453, 604, 606 Sehgman, S: AWARD WINNER, SPO Fotmdation Fellow for

Academic Year 1994-95

Seusini, A: 322 Senner, M: 311-312 Sesti, G: 536 Settlage, R: 294

Settles, D: 172 Seydel, FD: 214-215,348 Shaffer, W: 380 Shah, L: 623 Shahbahrami, B: 484

Shahem, S: 182 Shain, R: 357, 359, 594 Shmmou, S: 63 Shaw, CE: 258,272 Shearer, !viii: 343 Shen, O: 590 Shen-Schwarz, S: 181,408, 412 She~flmv, M: 645 Sheppard, C: 258

Sherer, DM: 39, 113-115, 189, 208, 212,214-215,339, 665-667 Sheraton, LM: 574 Shields, LE: 464

Shflctto, N: 106, 298, 335 Shin, Y: 245,443 Slmeider, B: 355 Sholl, J: 126,485 Shookhoff, ML: 603 Shumway, J: 112,558

S~bai, B: 30, 47, 168, 228, 284, 287, 300, 386, 528-532, 568, 570, 616, 619, 672

Siddiqi, T: 229 Siddiqi, TA: 175, 305 Sigman, R: 176

Silbennau, L: 187, 190

Siler-Khodr, TM: 202 Silva, MEG: 269 Silver, H: 523 Silver, R: 126, 485,520, 578 Silver, RK: 265, 275 Silver, RM: 161,296, 333 Silvennan, N: i00, 285, 364-365 Silvennan, R: 315

Sit, SY: 527 Sivan, E: 1,302-303,425,427 Skmmal, D: 229 Skellie, B: 614 Skupski, D: 124, 535 Slangen, B: 178

Sloau, CS: 490 Slomovic, B: 599 S~neltzer, JS: 198, 262, 394

Smikle, CB: 202 Smiley, RM: 223-224

Smith, C: 662

Author index January 1996 Am J Obstet Gynecol

Smith, M. Smith, P: 200 Smith, RS: 395-396,400, 490

Smith-Leviin, M: 123-124,404, 535 Smnlian, JC: 16, 21,162, 181,216-219, 409-41 i, 449, 647 Snyder, E: 355

Socol, M: 565 Sodovsky, E: 590 Sokol, RJ: 111,321,371,406, 468,470. 669

Solomon, J: 424 Song, J: 295 Songster, G: 89, 387

SoEtag, B: 267 Soper, R: 655

Sorem, KA: 202 Sorensen, TK: 500-501 Sorokin, Y: 79, 669 Spencer, DK: 202 Spicer, D: 209-211,638 Spiegel, D 539 Spmnato, JA: l 1,341,442, 567, 660 Spitzer, K: 70, 298 Spong, CY: 14, 113-115, 189,207, 213-215,316, 339, 601,603,

626,665-667 Sprauve, M: 288, 614 Sprayberry, N: 63 Stm~dley, CA: 194, 277, 504-506,537-538 Stm~dley, PR: 194,504

Stanell, W. 556 Stanislawski, D: 368 Stanley, JR 343 Stanziano, GJ: 101,587, 619 Star, J: 306 Starzyk, KA: 50, 186, 191,206-207, 209- 211,522, 638-639 Steele, L: 239-240 Steiger, R: 421

Stek, A: 353 Stevens, D: 375 Stevens, G 77 Stewart, K’ 237-238 Stewart, M. 267 Stdler, RJ: 378 Stone, J. 461 Stotzer, D 63 Stranun, S. 81,159, 342 Strassberg, S: 154

Stranss, A: 264 Streicher, P’ 290 Strobelt, N. 668 Strom, C: 460 Stronge, J: 622 Stubbs, T: 309 Suda, K: 250 Sujov, P: 431

Stanm~, V: 64 Sury, V: 453

Svendsen, T: 310 Svinarich, DM: 332 Swads,T 595 Swan, KF: 608

Swenson, RB: 658 Swiber, MJ: 256 Symecko, HL: 378 Syn, HC: 319, 328-330

Tamir, A: 430 Tamura, T: 25, 583 Tan, A: 22,476-477 Tm~, B: 571-572 Tan, WM: 173 Tarczy-Homoch, P: 591

Taskin, O: 517 Tassi, C: 322 Tatsugawa, Z: 483

Tatum, K: 237-240 Tedeschi, D: 13 Tefft, L: 75 Teideken, K: 446

Teixera, L: 398 Tejani, N: 24, 67, 156, 317-318, 377, 524,604, 607 Teoh, TG: 106, 298,498, 552 Tepper, R: 489 Teuclmer, B: 244 Tey, A: 129 Thaete, LG: 265-266,275 Thayer, CF: 136, 646 Theriault, D: 61 Thezzi, F: 453 Thorn, E: 2, 6, 142,405,569 Thomas, RA: 369 Thomas, RL: 278 Thomas, Jr, JX: 48 Thompson, K. 274

Thom, E: 17, 35,260, 582, 584, 586 Thorp, Jr., JM: 3,444

Thorsen, P: 73,336 Thunmu, G: 6 Timor-Tritsch, IE: 392 Tolosa, JE: 297, 331,452-453 Tomich, PG: 48, 185,580 Tomlinson, MW: 281,422, 537-538 Tooke-Miller, C: 32 Towers, CV: 264, 291,345,465,599 Tran, T: 366, 563 Treadwell, M: 371,663 Treadwell, MC: 374, 422,469 Treissman, MJ: 62 Trippel, D: 542

Troiano, N: 76, 510-511 Tmdinger, B J: 295

Tsao, FHC: 276 Tnlipan, N: 656

Tnlui, L: 450 Turcot, L: 625 Tumbull, G: 634 Turner, B: 365

Turner, C: 658 Turner, GW: 262 Turner, T: 449 Tnmqnest, M. 635

Turrentme, !vIA: 573


Tutt, E: 107-108

Tuft, L: 93 Tway, V: 446

Twickler, D: 414-416 Tygrett, L: 274

Uckan, E: 49, 232-233 Udom-Rice, I: 588 Uerpairojkit, B: 37, 65,427-429, 547, 555 Uharriet, P: 624 Uhrich, SB: 464 Ulrich, C: 28 Umstot, ES: 228 Ulnstot, US: 531 Usta, I: 528 Valderrama, O: 327 Valenslse, H: 536 Valenzuela, GL: 648 Van Assche, FA: 23 van Ballaer, PP: 23 van de Ven, CJM: 455 van den Berg, PP: 253,674 van Gmjn, HP: 12,254 van Someren, J: 169 vma Straaten, H: 178 van Vngt, J: 445 van den Berghe, K: 23

Vm~ Dyke, DL: 454 Van Wagoner, J: 592

Vm~ Winter, J: 171,230 Varcmysse, L: 207

Varm, J: 132 Vamer, M: 592 Vats, S: 454 Vaucher, Y. 617 Vedernikov, Y: 193 Veille, JC: 145,237-240, 246 Verburg, M: 50 Verburg, M: 639 Vercmysse, L’ 186, 191,206 Vergani, P’ 668

Verkeste, C. 178 Verlinsky, Y: 460

Venna, U: 67. 317-318, 524,607 Verpairojkit, B: 655 Vestergaard, BF: 336 Viano, D: 352 Victoria, A: 564 Villa, M: 180 Villarroel, L: 648 Vlngerhoets, AJJM: 254

Vintzdeos, AM: 21, 80, 90, 130, 136, 162, 181,216-219,408-412, 438, 449, 557, 646

Visintainer, P: 67 Volpe, A: 533 Volpe, L: 439

Vomero, A: 324 yon Oeyen, PT: 490

Vosatka, RJ: 279 Voss, D: 341 Vroon, D: 346

Wagner, E: 84

Waheed, A: 659 Waldsclunidt, T: 274 Walerius, H: 109

Walker, CK: 497 Walker, RP: 498 Wall, S: 122 Wall, SN: 283 Walla, C: 613 Wallace, I: 69 Wallis, T: 337 Walsh, M: 30 Walters, D: 465 Wang, E: 4 Wapner, R: 20, 397-398, 587 Ward, BEW: 491 Ward, K: 91, 110, 179, 494, 502-503,641 Ward, N: 96 Warren, PS: 551 Wass, T: 447 Watson, W: 375 Watts, DH’ 43,323 Webb, G: 346,658 Webb, M. 171 Wehbeb, H: 182 Weiner, S: 398, 452 Weiner, Z: 94 Weiner, Z: 235 Weinhouse, E: 395,400 Weiss, J: 10 Weissinger, M: 511

Welsch, C: 78 Welter, S: 375 Wen, T: 72 Wendel, G: 58, 358 Wenger, D: 27 Wenstrom, K: 25,471-474 Westgren, M: 141,236, 664

Weston, J: 4 Wheeler, T: 76, 510-511 White, DA: 149

White, M: 291 Whiteman, VE’ 65, 556 Whittington, S: 522 Whitty, JE: 281,321,505

Whitworth, NS: 383

Wilkins, IA: 434, 573 Willmi, A: 4 Williams, K: 199, 242,642-644 Willimns, MA: 347, 500-501 Willimns, P: 10 Willimnson, R: 479, 481 Wilson, S: 242, 642-644 Wing, DA: 8, 121,632

Wilm, H: 559 W~nsor, EJT: 478 Winter, TC: 464

Witkin, SS: 360, 588 Witter, F: 112, 558

Wiza, I: 106

518 Author index January 1996 Am J Obstet Gynecol

Wizmtzer, A. 19,220 Wojtowycz, M: 499 Wolfe, I-IM: 111,374, 401,422, 467,470, 575 Wolkoft; L: 577 Wong, S: 129,441 Woo, G. 613 Wood, DC: 54,234 Woods, J: 78 Wright, MTW: 491 Wu, SW: 496 Wu, Y-K: 1, 26 Wu, Z 295

W.¢se, L: 541 Xenakis, EIVlJ: 42, 620 Yadgarova, KT: 7 Yalcinoglu, A: 517

Yah, YL: 540 Yancey, M: 78 Yang, SH: 68, 328-330

Yankowitz, J: 95,479-481 Yaron, Y: 468 Yelian, F: 194 Yeo, GSH: 540 Yeong, M: 70

Yoder, E: 202 Yoon, BH: 68, 319,328-330 Yu Liu, Y: 536 Yun, E 261,437 Yun, H: 588 Zachar, V: 369 Zacharias, R: 464 Zacharova, V 369 Zachary, J: 142

Zador, IE: 371,406 Zarou, D: 182 Zebehnan, AM: 500 Zeng, XM: 423 Zhang, J: 476 Zhang, X: 403,477

Zhao, B: 649 Zin~ner, D" 221 Zmuner, EZ: 235, 431 Zinger, K: 130 Zingheim, RW: 347, 500-501

Zuckennan, G: 460 Zuckennan, R: 27 Zuidmna, L: 167 Zwang, E: 512

Institution Index

Volume 174, Number 1, Part 2 Institution index 521 Am.] Obstet Gynecol

Abbott Laboratories, Abbott Park, IL: 266, 361 Abbott Northwestern Perinatal Center, Minneapolis, MN: 486 Adeza Biomedical, Sunnyvale, CA: 592, 633 Albert Einsteit~ College of Medicine, New York, NY: 69, 85,

94, 139, 141, 188, 189, 236, 261,292, 437, 457, 498, 550, 609. 623

Arizona Health Sciences Center, Tucson, AZ: 259, 544, 624, 640

Arkansas Children’s Hospital, Little Rock, AR: 255 B.C. Women’s Hospital, Vancouver, British Columbia,

CANADA: 242, 642, 643,644 Baylor College of Medicine, Houston, TX, : 60, 72, 84, 148,

193,434, 517, 537, 538, 651 Ben-Gurion University, ISRAEL: 19, 220 Beth Israel Hospital, New York, NY: 92, 525, 598, 599, 600 Biex, Inc, Boulder, CO: 59 Bnai Zion Medical Center, Haifa, ISRAEL: 252, 539 Bourne-Hallam Clinic, Bourn, Cambridge, UK: 98 Bowman Gray School of Medicine, Winston-Salem, NC: 145,

237, 238, 239, 240, 246, 615 Bridgeport Hospital, Bridgeport, CT: 378 Brigham and Women’s Hospital, Boston, MA: 498 Brown University, Providence, RI: 75, 306, 498, 523 Bruce Rappaporl Faculty of Medicine, Haifa, ISRAEL: 430 Butterworth Hospital, Grand Rapids, MI: 167 CAP Study Gronp, Univ. of Laval, Quebec, CANADA: 155

Carolinas Medical Center, Charlotte, NC: 309, 376, 566 Cedars-Sinai Medical Center, Los Angeles, CA, 593, 613

Ccntemfial Medical Center, Nashville, TN: 608 Center for Human Reproduction, Chicago, IL: 90 Central Baptist Hospital, Lexington, KY: 101,619 Centre For AIDS Studies, Montreal, Quebec, CANADA: 366 Centre for Surgical Technologics/K.U. Leuven, BELGIUM: 23 Cha Women’s Hospital, Sconl, KOREA: 676

Chang Gung Memorial Hospital, Taipei, TAIWAN: 99, 543 Clucago Medical School, Chicago, IL: 518 Chief Medical Examiner’s Office, Oklahoma City, OK: 493 Chieti University, ITALY: 83 Children’s Hospital of MI, Detroit, MI: 401,663 Children’s Hospital of Philadelphia, PA: 576 Ciba Coming Diagn., Alameda, CA: 498 College of Physicians and Snrgeons, New York, NY: 526 Columbia University, New York, NY: 223,224, 392, 526 Cornell University Medical College, New York, NY: 16, 123-

124, 360, 404, 439, 535, 588 Dallas County Health Dept., Dallas, TX: 358 Danbury Hospital, Danbury, CT: 187, 190 Duke University, Durham, NC: 203,204, 325 Dynacare/Laboratory of Pathology, Seattle, WA: 500 East Bay Perinatal Medical Associates, Oakland, CA: 10

Easl Tennessee State University, Jolmson City, TN: 268 Eastern Virginia Medical School, Norfolk, VA: 310, 388, 389,

542, 547, 655 Einstein Medical Center., Philadelphia, PA: 425

Emory University, Atlanta, GA: 288, 346, 402, 614, 658 Evanston Hospital, Evanston, IL: 266, 275 Fetal Fibronectin Study Group: 3 Foundation for Blood Research, Scarborough, ME: 498

Fred Hutchinson Cancer Research Center, Seattle, WA: 501 Free University Hospital, Amsterdam, NETHERLANDS: 12,

254, 516 GeneCare Medical Genetics Center, Chapel Hill, NC: 450, 451 General Motors Corporation, Warren, MI: 352 George Washington Univ., Washington, DC: 142, 450-451 Georgetown University, Washington, DC: 7, 14, 39, 50, 102,

113-115, 186-192, 205-215,225, 245, 338-339, 348, 443, 522, 550, 601-603,638-639, 654, 665-668

Gessler Clinic, WinterhavencFL: 373 Golda Medical Center, Tel Aviv, ISRAEL: 83 Harbor-UCLA Medical Center, Torrance, CA: 248, 269-270,

314, 316, 626 Harvard Medical School, Boston, MA: 525,598-600 HCMC, Minneapolis, MN: 133,563 Healthdyne Perinatal Services, Marietta, GA: 101,587, 619 Hebrew University, ISRAEL: 636 Hennepin County Medical Center, Minneapolis, MN: 349 Henry Ford Hospital, Detroit, MI: 454 Hoag Memorial Hospital, Newport Beach, CA: 345 Hormel Instilute, Austin, TX: 230 Hospital Luis Castelazo Ayala, Mexico City, MEXICO: 180 Hospital San Borja Arriaran, Santiago, CHILI: 327 Huddinge University Hospital, Stockholm, SWEDEN: 141,

236, 664 Hmnboldt University, Berlin, GERMANY: 673-674 Illinois Masonic Medical Center, Chicago, IL: 460 Indiana University Medical Center, Indianapolis, IN: 140, 221,

487,635 Inonu University, TURKEY: 517 Integrated Genetics, Framingham, MA: 491 Internat’l Registry of the Onset of Fetal Anomalies, Brooklyn,

NY: 445 ISBM San Paolo University Milan, ITALY: 88, 146, 548

Jacobi Medical Center, Bronx, NY: 457, 623 Jersey City Medical Center, Jersey City, NJ: 360, 588 Johns Hopkins University School of Medicine, Baltimore, MD:

27, 112, 149, 196, 230, 280, 475, 558, 597 Kandang Kerbau Hospital, Singapore, CHINA: 540 Karl-Franz, Univ., Graz, AUSTRIA: 671 Keesler Medical Center, Biloxi, MS; 96, Kennedy Krieger Inst., Baltimore, IVID: 27 King Edward Memorial Hospital for Women, Perth,

AUSTRALIA: 105 King Faisal Specialist Hospital, Riyadh, SAUDI ARABIA:

545-546 Laboratoire de sante unique du Quebec, Montreal, Quebec,

CANADA: 366 LAC/USC Medical Center, Los Angeles, CA: 484 Laval University, Quebec, CANADA: 155, 589, 625

522 Institution ~ndex .Januaxy 1996 Am J Obstet Gynecol

LBJ Hosptial, Houston TX: 129 LDS Hospital, Salt Lake City, UT: 32, 492, 621 Little Company of Mary Hospital, Torrance, CA: 345 Loma Linda University, Loma Linda, CA: 97, 593 Long Beach Memorial Medical Center, Long Beach, CA: 66,

134, 264, 291,345, 465, 599 Louisiana State University Medical Center, Shreveport, LA: 15,

108, 134, 227, 507-508, 581 Loyola University-Chicago, Chicago, IL: 48, 185, 580 Lutheran Medical Center, Brooklyn, NY: 182 M. Negri Institute for Pharmacol Res., Milano, ITALY: 241 Madigan Arn~y Medical Center, Tacoma, WA: 372 Magee-Women’s Hospital, Pittsburgh, PA: 340, 382, 521 Maine Medical Center, Portland, ME: 379, 488 Marshall University, Huntington WV: 11,567 Marshfield Clinic, Marstffield, WI: 72, 354 Mayo Medical Center, Rochester, MN: 64, 171,230 Medical University of South Carolina, Charleston, SC: 81,342 Meharq¢ Medical College, Nashville, TN: 362 Moludals Hospital, SWEDEN: 118 Montefiore Medical Center, Bronx, NY: 457, 462, 498 Montreal General Hospital, Montreal, Quebec, CANADA: 366 Mr. Sinai Hospital Medical Center, Chicago, IL: 518 Mr. Sinai Hospital, Toronto, CANADA: 5 i, 53, 56, 298, 552 Mr. Sinai Medical Center, New York, NY: 127, 150, 461,610-

612 National Cancer Institute, Bethesda, MD: 184 National Institutes of Health, Bethesda, MD: 184 National Maternity Hospital, Dublin, IRELAND: 106

National Naval Medical Center, Bethesda, MD: 184 Naval Hospital, San Diego, CA: 82 New England Medical Center, Boston, MA: 128, 279, 577 New Jersey Medical School, Newark, NJ: 459 New York Hospital, New York, NY: 123-124, 404, 439, 535,

588 New York Medical College, Valhalla, NY: 24, 67, 156, 317

-318, 377, 453,524, 604-607 New York University Medical Center, New York, NY: SPO

Foundation Fellowship Award Winner For Academic Year 1994-95, and 350, 627-628

NICHD MFMU Network, Bcthesda, MD: 2, 6, 17, 35, 57, 68, 142, 260, 297, 319, 327-332, 405, 452-453,569, 575, 582, 584-586

North Shore University Hospital,, Manhasset, NY: 436, 541 Northwestern University Medical School, Chicago, IL: 3, 122,

231,265-266, 275, 283,368, 565, 578 Northwestern University/Evanston Hospital, Evanston, IL:

126, 485, 520 Norwalk Hospital, Norwalk, CT: 22 NTD Laboratories, Inc., Huntington Station, NY: 450-451 O,~wood Hospital, Dearborn, MI: 393-394 Odense University Hospital, DENMARK: 73,336 Ohio State University, Columbus, OH: 109, 519, 562, 659 Olive View - UCLA Medical Center, Sylmar, CA: 143,483

Oregon Health Sciences University, Portland, OR: 55 Ostra Hospital, Gothenburg, SWEDEN: 18 Pacific Northwest Laboratories, Richland, WA: 372 Penn Skate University, Hershey, PA: 174 Pennsylvania Hospital,, Philadelphia, PA: 54, 234, 297, 452

-453,554 Politechnics of Milan, Milan, ITALI: 548 Pomona Valley Hospital, Pomona, CA: 38, 40, 172, 676

Ponce School of Medicine, Ponce, PR: 522, 601 Princess Margaret Hospital, Perth, AUSTRALIA: 105 PU Catolica de Chile School of Medicine, Santiago, CHILI:

549, 648-649 R&D Systems, Minneapolis., MN: 203,325 Rambam Medical Center, ISRAEL: 235,426, 430-431 Reproductive Genetics Institute, Chicago, IL: 460 Robert C. Byrd Health Sciences Center, Charleston, WV: 259 Royal Hospital for Women, Sydney, AUSTRALIA: 551 RWTH, Aachen, GERMANY: 178 S. Gerardo Hopsital, Milan, ITALY: 668 Sackler School of Medicine, Tel Aviv, ISRAEL: 512, 645 Saddleback Memorial Med Ctr, Laguna Hills, CA: 66, 165, 46 San Bernardino County Meal Ctr, San Bernardino, CA: 648 San Marcos University, Lima, PERU: 525 Sapir Medical Center, Kfar Saba, ISRAEL: 489 Seoul National University, Seoul, KOREA: 68, 319, 328-330 Serlin Maternity Hospital, Tel Aviv, ISRAEL: 636, 645 Shaare-Zedek Medical Center, Jerusalem, ISRAEL: 590, 636 Sinai Hospital of Baltimore, Baltimore, MD: 458 Sioux Valley Hospital, Sioux Falls, SD: 375

Soroka Medical Center, ISRAEL: 19, 220 Sotero del Rio Hospital, CHILI: 549

Southwest Foundation for Biomedical Research, San Antonio, TX: 343

St. Francis Hospital, Hartford, CT: 131,385 St. John’s Mercy Medical Center, St. Louis, MO: 564 St. Joseph Hospital, Detroit, MI:454 St. Louis University, St. Louis, MO: 112, 558-559

St. Peter’s Medical Center, New Brunswick, NJ: 16, 21, 80, 90 130, 136, 162, 181,216-219, 408-412, 438, 449, 557, 646- 647

St. Raphael Hospital, New Haven, CT: 477 St. Vincent Hospital, Indianapolis, IN: 390-391 Stanford University Medical Center, Stanford, CA: 527 Stanford University Medical Center, Stanford, CA: 579 Ste-Justine Hospital, Montreal, Quebec, CANADA: 61, 132,

366 Stritch School of Medicine, Maywood, IL: 48 SUNY @ Stony Brook, Stony Brook, NY: 324, 370, 384 SUNY Health Science Center, Brooklyn, NY: 144, 249, 445,

509, 629 SUNY Health Science Center, New York, NY: 403 SUNY Health Science Center, Syracuse, NY: 59, 315, 499, 53z Swedish Medical Center, Seattle, WA: 347, 500-501 Tel Aviv Sourasky Medical Center, Tel-Aviv, ISRAEL: 512

Volume 174, Number 1, Part 9 Institution index 523 Am J Obstet Gynecol

Tel Aviv University, Tcl-Aviv, ISRAEL: 83

Temple University School of Medicine, Philadelphia, PA: 1, 19, 26, 37, 54, 65,220, 234, 302-303,425-429, 466, 546, 554-556

Texas Tech Health Sciences Center, E1 Paso, TX: 624 The Children’s Hospital, Denver, CO: 391 The Long Island College Hospital, Long Island, NY: 403,629 The Toronto Hospital, Toronto, CANADA: 298 The Union Memorial Hospital, Baltimore, MD: 200 The Western Pem~sylvania Hospital, Pittsburgl~, PA: 278 The Women’s Hospital, Nashville, TN: 608 Thomas Jefferson University, Philadelphia, PA: 18, 20, 27,

100, 285,297, 364-365, 397-398, 446,452-453,457, 587, Tor Vergata University, Rome, ITALY: 13, 536 Triplet Army Medical Center, Honolulu, HI: 78 Tnfts University School of Medicine, Boston, MA: 28, 128,

279, 418-419, 424, 577. 622 UMDNJ-Roberl Wood Johnson Medical School, New

Brunswick, NJ: 16, 21, 80, 130, 136, 154, 162, 181,216- 219, 226, 408-412, 438, 449, 557. 646-647

University "La Sapienze", Rome, ITALY’ 653 University College, London, ENGLAND: 59 University Hospilal Center, Caen, FRANCE: 299 University Hospital Nijmegen, NETHERLANDS: 253,674 University Hospital, Lenven, BELGIUM: 23, 191,206-207 University of Alabama at Birmingham, AL: 3, 25, 35, 46, 63,

103-104, 142, 163-164, 260, 320. 361,363,435, 471-474, 582-583,585-586

University of Amsterdam, NETHERLANDS: 12 University of Arizona Health Sciences Center, Tucson, AZ: 456 University of British Columbia, Vancouver, British Columbia,

CANADA: 62, 199 University of California San Francisco, CA: 173,561,576, 675 University of California, lrvine, CA: 66, 165, 180, 250-251,

264, 291, 307-308, 421,433,465, 675 University of California, Davis, Sacramento, CA: 183, 631 University of California, Los Angeles, CA: 121, 143,521,613 University of California, San Diego, CA: 82, 183,413,617 University of California, San Francisco Fetal Treatment Center,

San Francisco, CA 137 University of Chicago, Chicago, IL: 497 University of Chili, Santiago, CHILI: 327 University of Cincinnati, OH: 175, 229, 271,305 University of Colorado School of Medicine, Denver, CO: 59,

146, 201,241,373,390, 417, 447 University of Connecticut Medical Center, Farmington, CT: 21,

39, 80, 113-115, 131, 170, 192, 198, 208, 212, 216-219, 262, 339, 385, 393-394, 411,601-602, 654, 665-667

University of Denver, Denver, CO: 447 University of Florida al Gainesville, FL: 344 University of Florida, Jacksonville, FL: 29, 44 University of Florida, Pensacola, FL: 267 University of Geneva, SWITZERLAND: 199, 463,589 University of Goteburg, SWEDEN: 118

University of Graz, AUSTRIA: 244 University of Illinois at Chicago, IL: 381 University of Iowa, Iowa City, IA: 95, 161,479-481 University of Iowa Hospitals and Clinics, Iowa City, IA: 274 University of Kentucky, Lexington, KY: 86, 93, 107-108 University of Leuven, BELGIU2VI: 186 University ofLimburg, Maastricht, NETHERLANDS: 77, 178,

263 University of Louisville, KY: 11, 341,442, 567, 660 University of Manitoba, Manitoba, CANADA: 41 University of Miami, Miami, FL: 423 University of Michigan, Detroit, MI: 125, 197, 352, 455 University of Milan, ITALY: 88, 241,450, 668 Universily of Mississippi, Jackson, MS: 34, 36, 151-153, 293,

309, 383,618 University of Modena, ITALY: 533 University of Montreal, Montreal, Quebec, CANADA: 61, 132 University of Natal, SOUTH AFRICA: 45 University of New Mexico, Albuquerque, NM: 142, 169, 560 University of Newcastle, NSW, AUSTRALIA: 657 University of North Carolina at Chapel Hill, NC: 3, 444 University of Oklahoma, Oklahoma City, OK: 31,343,634 Universily of Ottawa, Ottawa, CANADA: 158, 399 University of Pennsylvania, Philadelphia, PA: 18, 440, 452,

553, 576 University ofPerugia, Perugia, ITALY: 322, 420 University of Pittsburgh, PA: 33,340, 448, 521 University of Rochester, Rochester, NY: 498 University of Southern Alabama, Mobile, AL: 98, 256 University of South Carolina, Charleston, SC: 159, 652 University of Southern California, Los Angeles, CA: 8, 59 86,

89, 119-120, 222, 294, 301,326, 353,387, 432, 630-632 University of Sydney, NSW, AUSTRALIA: 295 University of Tennessee, Memphis, TN: 30, 47, 168, 228, 284,

287, 300, 386, 528-530, 532, 568, 570, 616, 619, 672 University of Texas HSC, Houston, TX: 74, 129, 334, 367, 434,

441, 573,633

University of Texas HSC, San Antonio, TX: 42, 71, 202, 313, 356-

357, 359, 594, 620 University of Texas MB, Galveston, TX: 5, 49, 232-233, 650 University of Texas Southwestern Med Ctr @ Dallas, TX: 9,

58, 177, 351,358, 414-416, 493, 649 University of Toronto, Toronto, CANADA: 4, 70, 106, 116-

117, 155, 298, 311-312, 335, 478, 571-572 University of Utah, Salt Lake City, UT: 32, 49, 59, 87, 91, 110,

161, 179, 232-233,296, 333,492, 494, 502-503, 592, 641, 650. 670, 674

University of Vermont, Burlington, VT: 157 University of Washington, Seattle, WA: 43,323,464, 504, 591 University of Waterloo, Waterloo, Ontario, CANADA: 70 University of Western Ontario, CANADA: 243

524 Institution index Janua~T 1996 Am J Obstet Gynecol

University of Wisconsin, Madison, WI: 258-259, 272-273,276, 495-496

Uppsala University, SWEDEN: 236 Utah Valley Regional Medical Center, Provo, UT: 492, 670 Vanderbilt University, Nashville, TN: 76, 247, 510-511,656 Wake Forest University, Winston-Salem, NC: 145,615 Washington University, St. Louis, MO: 56, 176 Wayne State University, Detroit, MI: 57, 79, 232-233, 332, 454 Wayne State University/Hutzel Hospital: 68, 111, 135, 147,

160, 194, 257, 277, 281-282, 290, 304, 321,331,337, 369, 371,374, 401,406-407, 422, 467-470, 482, 504-506, 537- 538, 575, 596, 661-663,669

West Virginia University, Morgantown, WV: 195 Westmead Hospital, AUSTRALIA: 52 Wilford Hall Medical Center, San Antonio, TX: 202. 637 William Beaumont Hospital, Royal Oak, MI: 395-396, 400,

490, 595 Winthrop-University Hospital, Mineola, NY: 498 Women and Infant’s Hospital, Providence, RI: 289, 306, 418,

498 Women’s College Hospital, Toronto, Ontario, CANADA: 166 Wright State University, Dayton, OH: 187 Wyoming "Valley OB/GYN Associates, Kingston, PA: 57 Yale University School of Medicine, New Haven, CT: 22, 186,

191, 196, 205, 207, 286, 338, 355, 378, 380, 476-477, 514- 515, 545-547, 549, 555, 655

Yeshiva University, Bronx, NY: 457 Zuiderzee Hospital, Lclystad, NETHERLANDS: 12

ETRICS - American Journal of Obstetrics & Gynecology

Documents