Observations on Optimum Time for Operative Intervention ...€¦ · Observations on the Optimum Time for Operative Intervention for Aortic Regurgitation II. Serial Echocardiographic

Observations on the Optimum Time for OperativeIntervention for Aortic Regurgitation

II. Serial Echocardiographic Evaluation of Asymptomatic Patients

WALTER L. HENRY, M.D., ROBERT 0. BONOW, M.D.,DOUGLAS R. ROSING, M.D., AND STEPHEN E. EPSTEIN, M.D.

SUMMARY A recent echocardiographic study of symptomatic patients who had aortic valve replacementfor isolated aortic regurgitation indicated that patients in whom preoperative left ventricular end-systolicdimension (LVD[SYS]) exceeded 55 mm or fractional shortening (%FS) was less than 25% were at high risk ofdeveloping congestive heart failure and dying after an otherwise successful operation. Because indices of leftventricular systolic function might identify asymptomatic patients with aortic regurgitation who might benefitfrom earlier operation, 37 such patients were evaluated with serial echocardiograms (mean follow-up 34months). Fourteen patients (38%) subsequently developed symptoms and were recommended for operation(SUBSQ OP). Twenty-three patients (62%) remain asymptomatic during follow-up (NON OP). LVD(SYS)and %FS were the most sensitive measurements for distinguishing on initial examination the patients who sub-sequently required operation from those who have not (LVD[SYS] 53.0 mm SUBSQ OP vs 44.3 mm NONOP, p = 0.001; %FS 28.8% SUBSQ OP vs 33.9% NON OP,p = 0.002). During serial studies, the maximumrate of change in end-systolic dimension exceeded 7 mm per year in only one patient. Four of five patients(80%) with end-systolic dimension greater than 55 mm developed symptoms and came to operation during amean follow-up of 39 months. Of the 20 patients whose initial end-systolic dimension was 50 mm or less, onlyfour patients (20%) developed symptoms and required operation, and none died during follow-up. Thus, anasymptomatic patient with aortic regurgitation whose end-systolic dimension is less than 50 mm appears to beat low risk and can be safely followed with echocardiograms at yearly intervals. Asymptomatic patients withend-systolic dimension of 50-54 mm are being followed with serial echocardiograms every 4-6 months. Opera-tion is now being recommended to patients with end-systolic dimensions of 55 mm or greater, even in theabsence of symptoms.

IT IS OFTEN DIFFICULT to decide when to inter-vene with operation in patients with long-standingaortic regurgitation.' Although indications vary,many clinicians only recommend operation when cer-tain symptoms are present. This approach is based onthe observation that when angina, syncope or con-gestive heart failure develop, prognosis over the nextseveral years is poor.2-7 A corollary of this concept isthat by not recommending operation until these symp-toms occur, operative risk will be postponed, as willthe long-term morbidity and mortality associated withexisting prosthetic valves. Also, by postponing opera-tion, the patient may benefit from subsequent im-provements in operative technique and prostheticvalve design.The major drawback to this approach is that by the

time angina, syncope or congestive heart failuredevelop, some patients have already suffered irrevers-ible left ventricular dysfunction." 8-18 Many of thesepatients develop severe congestive heart failure anddie after operation, despite the elimination of aorticregurgitation. The present diagnostic challenge, there-fore, is to identify the asymptomatic patient with aor-tic regurgitation just before irreversible left ven-

From the Cardiology Branch, NHLBI, NIH, Bethesda,Maryland.

Address for correspondence: Walter L. Henry, M.D., Universityof California at Irvine Medical Center, Cardiology Divis'on, 101City Drive South, Building 53, Orange, California 92668.

Received September 21, 1978; revision accepted May 22, 1979.Circulation 61, No. 3, 1980.

tricular dysfunction occurs so that earlier operationcan be undertaken." 9, 12, 14, 15, 17

Recently, we analyzed our experience with sympto-matic patients who had aortic valve replacement forlong-standing aortic regurgitation.18 This analysis in-dicated that patients who died postoperatively of con-gestive heart failure had preoperative echocar-diographic evidence of poor left ventricular systolicfunction. These data suggested that echocardiog-raphy might be useful for identifying patients whowould benefit from earlier operative intervention.Recommendations for earlier "prophylactic"

operation that are based on echocardiographic obser-vations in patients who had operation because ofsymptoms must be applied with caution to the asymp-tomatic patient. For example, if asymptomaticpatients in the subgroup who might benefit fromearlier operation do not cross into the high-risk groupor develop severe symptoms for many years, it isreasonable to question the wisdom of earlier opera-tion. However, if such patients develop symptoms andcome to operation within a few years, a significantdecrease in operative mortality or valve-related com-plications would probably not be realized by postpon-ing operation. Hence, it is important to know thechanges in echocardiographic measurements andsymptomatic status that occur during the naturalhistory of initially asymptomatic patients with aorticregurgitation. The present study was designed toprovide this information and to elucidate the role ofechocardiography in managing the patient with long-standing aortic regurgitation.

484

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

SERIAL ECHOES IN ASYMPTOMATIC AR/Henry et al.

MethodsPatient Population

The patient population consisted of all patients withaortic regurgitation who had been followed in the Car-diology Clinic of the National Heart, Lung, and BloodInstitute in whom two or more serial echocardiogramshad been obtained over at least a 1-year period. Pa-tients were considered to have aortic regurgitation ifthey had a blowing diastolic murmur at the base of theheart on physical examination and a left ventricularend-diastolic dimension by echocardiography that wasabove the normal range. No patient had a decreasedcarotid upstroke suggestive of aortic stenosis or anapical holosystolic murmur of mitral regurgitation.Also, no patient had echocardiographic evidence ofmitral or aortic stenosis. All but one patient had high-frequency diastolic fluttering of the anterior leaflet ofthe mitral valve.

Initial echocardiographic evaluation of these pa-tients was performed between January 1972 andDecember 1975. The minimum potential follow-up forany patient in this study was 26 months. When the ini-tial evaluations were performed (and during the courseof the study), operation was recommended for patientswith aortic regurgitation only when patients presentedwith or developed severe dyspnea on exertion, anginapectoris, syncope or congestive heart failure. Thus,none of the patients had these symptoms at the time ofinitial study. In addition, none developed these symp-toms (and, therefore, none came to operation) within 1year after the initial examination.

Patient StudiesHistory, physical examination, 12-lead ECGs and

M-mode echocardiograms were obtained in allpatients at the initial and subsequent evaluations.Romhilt-Estes scores were computed from the ECGsas previously described.1' Cardiac catheterization datawere available in every patient who subsequently cameto operation.

Echocardiographic examination was performedusing either an Ekoline 20A or Hoffrel 201 ultrasoundtransceiver interfaced to a Honeywell 1856 strip-chartrecorder. A 12.5-mm diameter, 2.25-MHz, unfocusedultrasound transducer was used. Echocardiographicmeasurements included left ventricular transversedimensions at end-diastole and end-systole and ven-tricular septal and left ventricular posterior free wallthicknesses.'9' 20 These measurements were obtainedusing the T-scan technique,21 22 with the ultrasoundbeam passing through the left ventricle caudal to thetips of the mitral leaflets. The left ventricular trans-verse dimensions at end-diastole and end-systole weretaken as the maximum and minimum distancebetween the left side of the ventricular septum and en-docardium of the left ventricular posterior free wall.Wall thickness measurements were made in latediastole but before atrial systole. Aortic root and leftatrial dimensions were measured as previouslydescribed.19 A switched-gain circuit was used tosimplify measurement of left ventricular wall

thickness and left atrial dimension.23 From theprimary echocardiographic measurements, left ven-tricular ejection fraction,2' fractional shortening of theleft ventricle25 and estimated left ventricular mass26were calculated.

Patients were subdivided into two groups based onwhether they developed symptoms and requiredoperation or remained asymptomatic. Measurementsin the two groups were compared using an unpaired ttest.

Results

Patient Experience

Thirty-seven patients met the selection criteria andwere included in the present study. There were 20 menand 17 women, ages 16-64 years (mean 35 years).Fourteen of the 37 patients (38%) subsequently devel-oped congestive heart failure, severe dyspnea on exer-tion, angina or syncope and were recommended foroperation. These patients will be referred to as the"subsequent operation" group. In all 14 patients, aor-tic root cineangiography revealed aortic regurgitationsevere enough to produce opacification of the left ven-tricle that failed to clear during the subsequent fewcardiac cycles. No patient had an aortic valve gradientexceeding 20 mm Hg or evidence of significant diseaseof other heart valves. Two of the 14 patients (14%) hadcoexistent coronary artery disease. All 14 patientshave had aortic valve replacement. One of the 14patients required aortic root reconstruction in addi-tion to aortic valve replacement. In these 14 patients,mean follow-up between the initial echocardiographicexamination and the echocardiographic study im-mediately before operation was 28.2 months (range13-57 months). Seven of the 14 patients (50%) came tooperation within 30 months of the initial evaluation.

Twenty-three of the 37 patients (62%) have notdeveloped severe dyspnea on exertion, congestiveheart failure, angina or syncope, and operation hasnot been recommended. These patients will be referredto as the "nonoperative" group. In these 23 patients,the mean time between the initial and most recentechocardiographic examination was 38.7 months(range 26-64 months). All 23 patients are still alive.The time between initial echocardiographic examina-tion and the most recent contact averaged 42.6 months(range 29-70 months).

Initial Evaluation

The electrocardiographic and echocardiographicmeasurements in the two groups are summarized infigures 1 and 2 and table 1. Left ventricular end-diastolic and end-systolic dimensions were larger, leftventricular fractional shortening was less, and theRomhilt-Estes score was greater in the subsequentoperation group than in the nonoperative group. Ofthe four measurements, left ventricular end-systolicdimension and left ventricular fractional shorteningmost strongly distinguished the subsequent operationgroup from the nonoperative group. There were nosignificant race or sex differences between the two

485

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

CIRCULATION VOL 61, No 3, MARCH 1980

INITIAL EXAM

90

80 _

70 _U)wLUW

-

z20

_)

2E

z0U)LSW]>

60 h

50 _

40 _

30 h

20 K

10

LVD (DIA)0

0

iS4to 0

p<.O

NONOP SUBSE-QUENTOP

LVD (SYS)

0

-

L

0

0

p<.+0

p<O.01

NONOP SUBSE-QUENTOP

LV FRACTIONALQWnPTPKIM(Z<

45

40

zw_ 35

U

z

z..

U)< 25I

-J

<=15

LV MASSS

800 K

900 r 0

0

700K *

H

10 _

5K

NONOP

0

U)

4er

z

cn

U)

cnU)

W

0-

4n

LU)0

p<0.01

'SUBSE-QUENTOP

FIGURE 1. Plot of left ventricular dimensions (LVD) atend-diastole (DIA) and end-systole (SYS) and left ven-

tricular (L V) fractional shortening. Patients are subdividedinto those who have remained asymptomatic during follow-up (NONOP) and those who developed symptoms and were

recommended for operation (SUBSEQUENT OP). Themean values are indicated by open circles. The stippled area

indicates the normal range offractional shortening.

600 1

500 H

400 K

pS

300 h

200 F-

100

S

S

e t

0

0

N.S.

16

V)CE

W

LI

H

4

LliZY.:

LJz

,r

LLJ>L

14

12

10

8

6

4

2

INITIAL EXAM

LV PW

- S.

S.

-SO

0

SW

4-

NS.

N.S.- [ L _ L 1 L ____

NONOP SUBSE- NONOP SUBSE-QUENT QUENTOP OP

FIGURE 2. Plot of estimated left ventricular (L V) mass,

LV posterior free wall thickness (LV PW) and Romhilt-Estes point score from the ECG. Open circles representmean values. Shaded areas represent the normal range ofvalues. NONOP = patients who remained asymptomaticduring follow-up; SUBSEQUENT OP = patients whodeveloped symptoms and were recommended for operation.

TABLE 1. Comparisons of Nonoperative and Subsequent Operation Patients

Initial examination Recent examinationSubsequent Subsequent

Echo Nonoperative operation p Nonoperative operation p

LV dimension (DIA)(mm) 66.7 8.8 74.4 - 7.9 0.009 67.1 8.4 76.6 - 7.6 0.002

LV dimension (SYS)(mm) 44.3 - 7.6 53.0 - 6.3 0.001 45.1 7.2 56.4 7.3 0.001

LV fractionalshortening (%) 33.9 - 4.2 28.8 - 4.3 0.002 32.9 4.4 26.4 - 6.3 0.003

LV free wallthickness (mm) 12.6 - 1.6 12.7 - 1.3 NS 12.8 - 1.7 12.9 - 1.7 NS

LV mass (g) 515 - 193 605 - 175 NS 534 - 193 642 - 214 NSAortic root dimension(mm) 36.5 5.4 36.1 - 6.2 NS 37.6 6.0 36.7 - 5.7 NS

LA dimension (mm) 39.9 - 6.0 41.5 - 6.2 NS 41.4 - 6.0 45.4 - 6.5 NS

ECGRomhilt-Estes score 3.5 - 2.5 5.2 - 2.1 0.03 3.7 - 2.8 5.9 - 2.1 0.01Heart rate (beats/min) 66 - 9 69 - 12 NS 68 - 12 74 t= 10 NS

Data are mean =f SD.

Abbreviations: LV= left ventricular; DIA = end-diastole; SYS

486

ESTES

8

7

6

5

4

ir0U)

Uf)

H

ULLLi

-0_

0

3

2 0

NONOP SUBSE-QUENTOP

0~ p<0.05

end-systole; LA = left atrial.

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

SERIAL ECHOES IN ASYMPTOMATIC AR/Henry et al.

groups. However, the mean age ofgroup was significantly less thansequent operation group (35.5 ± 1113.1 years; p < 0.05).The echocardiographic data from

summarized in figure 3 as a percentUvalue obtained using regression eqour laboratory from a large groupolder20 normal subjects.

Recent Evaluation

The mean values obtained atevaluation are also summarized in toperative group, left ventricular ersion did not change significantly dui39 months that elapsed between theevaluations (fig. 4). In the subsgroup, left ventricular end-systol

LVD LVD LV PW(DIA) (S LVS P

0rac

8

000

Ao

0

0

8

r tne nonoperative creased significantly between the initial examinationthat of the sub- and the examination immediately before operation.

1.8 years vs 45.5 + The data were examined to determine the rate ofchange of left ventricular end-systolic dimension dur-

all 37 patients are ing routine serial study. First, the absolute change wasage of the expected computed for each patient by subtracting the value ob-uations derived in tained at initial examination from the value obtainedof younger"9 and at the most recent examination. The average rate of

change was then computed by dividing the absolutechange by the time period between the initial and themost recent examination. The results of the average

the most recent rate of change in end-systolic dimension are also.able 1. In the non- shown in figure 4. The average rate of change ex-nd-systolic dimen- ceeded 4 mm per year in only one of the 37 patientsring the average of (3%). The average rate of change for patients with leftinitial and recent ventricular end-systolic dimension less than 50 mm

sequent operation (2.6 + 1.2 mm per year) was similar to that for pa-lic dimension in- tients whose end-systolic dimension was 50 mm orgreater (2.7 + 1.1 mm per year).

The maximum rate of change of end-systolic dimen-sion also was determined by computing the rate of

LA LV MASS change between sequential studies in each patient. Forexample, most patients had three or more sequential

S

0

z

W

zW

WU

XW

*1.gD

U)en0

x

400 hS

0

350 -

300 f

250F

200

150

r0

_t0

8

75 --0C- H- 0

so

!--

0-

FIGURE 3. Plot of echocardiographic measurements ob-tained at initial examination in 37 initially asymptomaticpatients with isolated aortic regurgitation. Patients who sub-sequently developed symptoms and were recommended foroperation are indicated by the filled circles; those who havenot yet developed symptoms are indicated by the opencircles. The echocardiographic measurements are expressedas a percentage of the expected value computedfrom the pa-tient's age and body surface area. LVD = left ventriculardimension; SYS = end-systole; DIA = end-diastole; LVPW= left ventricular posterior wall thickness; Ao = aortic rootdimension; LV = left ventricular; LA = left atrial dimen-sion.

NON

801

U)ccW

zw

aiF-

U)

(n

z0U)

70 -

60H

50V

40

:;

cr0-

ABSOLUTE CHANGE

IOPERATIVE SUBSEQUENTOP

-0

.

/

-0--

RATE OF CHANGE

AVERAGE MAXIMUM

..:W>- 14

z12X

-o0U)>- 10

0U)z 8llJED5

_-

_-

H 0

W30 6

~~~~~~~W3

204

<

U-

010 W2

N.S. p<0.01

INITIAL RECENT INITIAL RECENT(PREOP)

-2

FIGURE 4. Plot of left ventricular (LV) end-systolicdimension. Data in the four columns on the left show themeasurements for the initial and most recent echo-cardiographic examinations for both the nonoperative andsubsequent operation groups. The two columns on the rightshow the average and maximum rate of change in L V end-systolic dimension. See text for details ofthe method used tocalculate the rate of change.

275 H

250 -

225[-

200 H

175 [

150 F

zW

CLzW

Ul-J

W

w

CLLX

xww

0

125

487

C' Al-

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

VOL 61, No 3, MARCH 1980

echocardiographic studies performed at intervals of6-18 months. The most rapid change noted betweenthese sequential studies was used as the maximum rateof change for that patient (fig. 4). The most rapidchange observed between sequential studies exceeded7 mm per year in only one patient.The most recent data from the nonoperative and

subsequent operation groups were compared with theechocardiographic and electrocardiographic measure-ments of 43 patients in our operative series'8 who hadgood quality echocardiograms and who presented withsymptoms and came to operation within 3 months ofinitial evaluation (table 2). These latter patients will bereferred to as the "immediate operation group." Themean values in the immediate operation patients weresimilar to the mean values in the subsequent operationgroup, except for heart rate and left ventricular freewall thickness, which were less in the subsequentoperation group. In contrast, all measurements in theimmediate operation group were significantly greaterthan those in the nonoperative group, except for frac-tional shortening, which was less, and aortic rootdimension, which was not different.

Preoperative hemodynamic data from the 43patients in the immediate operation group were com-pared with preoperative data in the 14 patients in thesubsequent operation group. None of the hemody-namic measurements, including right- and left-heartpressures, aortic and pulmonary artery pressures, car-

diac output and cardiac index, were significantlydifferent between the two groups.

Relation of Left Ventricular End-systolicDimension to Clinical Course

In figure 5, patients are divided into three subgroupsbased on the left ventricular end-systolic dimensionmeasured at initial echocardiographic evaluation.

End-systolic Dimension < 50 mm

Four of the 20 patients (20%) whose left ventricularend-systolic dimension was less than 50 mm at initial

LEFT VENTRICULAR DIMENSION (END SYSTOLE) IN MILLIMETERS

30 35 40 45 50 55 60 65 70

Initial Exam Patients 12 Patients 5 Patients

(Asymptomatic) (20%-POP) (50%-1POPi (=80%-.OP)

( 1)l (6) 1)

Recent Exam 15 P 1 Nonop No

(Asymptomatic) Symptoms

Symptoms...... .....-........". 0-- .0 0- -ftS . ''''0 0; ijg' 0 0:' 00 A iX; ff0 ..E ..'.-

Recent Exam

(Symptoms-Preop)1 3 Op 5

FIGURE 5. Diagram of the influence of left ventricular(L V) end-systolic dimension on the clinical course of the 37initially asymptomatic patients with aortic regurgitation.Patients are subdivided into three groups: 1) 20 patientswhose initial LV end-systolic dimension was less than 50mm, 2) 12 patients whose initial LV end-systolic dimensionwas 50-54 mm and 3) five patients whose initial LV end-systolic dimension was 55 mm or greater. Fourteen patientsdeveloped symptoms (stippled area) and required operation(Op) while 23 patients have remained asymptomatic duringfollow-up (Nonop).

evaluation developed symptoms and were recom-mended for operation. Left ventricular end-systolicdimension did not exceed 55 mm during follow-up in

these four patients or in the 16 clinically stable pa-

tients who remained asymptomatic (figs. 4 and 5).

End-systolic Dimension of 50-54 mm

Six of 12 patients (50%) who were in this subgroupat initial evaluation developed symptoms requiring

TABLE 2. Comparison of Immediate Operation with Nonoperative and Subsequent Operation Patients

Nonoperative Immediate SubsequentEcho group p operation group p operation group

LV dimension (DIA) (mm) 67.1 - 8.4 0.02 72.1 - 7.5 0.07 76.6 - 7.6

LV dimension (SYS) (mm) 45.1 - 7.2 0.002 52.5 - 10.6 NS 56.4 - 7.3

LV fractional shortening (%) 32.9 - 4.4 0.004 28.2 - 8.5 NS 26.4 - 6.3

LV free wall thickness (mm) 12.8 - 1.7 0.007 14.1 - 1.8 0.03 12.9 1.7

LV mass (g) 534 - 193 0.01 656 - 148 NS 642 +214Aortic root dimension (mm) 37.6 - 6.0 NS 37.6 - 6.3 NS 36.7 5.7

LA dimension (mm) 41.4 - 6.0 NS 44.4 - 7.0 NS 45.4 6.5

ECGRomhilt-Estes score 3.7 - 2.8 0.001 6.6- 2.6 NS 5.9 2.1

Heart rate (beats/min) 68 - 12 0.001 83- 17 0.02 74 10

Data are mean - SD.Abbreviations: LV left ventricular; DIA = end-diastole; SYS = end-systole; LA = left atrial.

488 CIRCULATION

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

SERIAL ECHOES IN ASYMPTOMATIC AR/Henry et al.

operation within 13-52 months (mean 23 months).Four of these six patients developed symptomswithout the left ventricular end-systolic dimension ex-ceeding 55 mm. The other two patients had left ven-tricular end-systolic dimensions greater than 55 mmby the time symptoms had developed (figs. 4 and 5).

End-systolic Dimension > 55 mmOf the five patients in this subgroup at initial

evaluation, four (80%) developed symptoms and wererecommended for operation within 22-57 months(mean 39 months) after initial study. The percentageof patients with an end-systolic dimension of at least55 mm who developed symptoms and came to opera-tion was significantly greater than the percentage ofpatients with an end-systolic dimension less than 50mm who required operation (p < 0.05 by Fisher's ex-act test27). An additional two patients crossed into the> 55 mm subgroup during the study (fig. 4). Both patientsdeveloped symptoms and came to operation, onepatient in 9 months and the other in 16 months aftercrossover (figs. 4 and 5).

DiscussionRecently, we reviewed our experience with symp-

tomatic patients who required operation to correctisolated aortic regurgitation.18 This study showed thatthe left ventricular end-systolic dimension and the leftventricular fractional shortening (measured by echo-cardiography) are excellent measurements to usepreoperatively to identify patients at high risk of hav-ing developed irreversible left ventricular dysfunc-tion.'8 This conclusion is based on the observation thatnine of 17 patients (53%) with a preoperative left ven-tricular end-systolic dimension greater than 55 mmeither died at operation (two patients) or died latepostoperatively of congestive heart failure (seven pa-tients). Eleven of 13 patients (85%) who had both apreoperative left ventricular end-systolic dimensiongreater than 55 mm and a left ventricular fractionalshortening below 25% either died at operation (twopatients) or died late postoperatively of congestiveheart failure (seven patients) or had moderate orsevere residual systolic dysfunction postoperatively(two patients). In contrast, of the 32 patients with leftventricular end-systolic dimension of 55 mm or less,only one (3%) died late postoperatively of congestiveheart failure. That patient suffered an intraoperativemyocardial infarction and manifested markedly im-paired left ventricular systolic function post-operatively.These data strongly suggest that operation should

be performed in patients with aortic regurgitation im-mediately before or shortly after the left ventricularend-systolic dimension exceeds 55 mm, regardless ofsymptomatic status. However, our previous study didnot provide several important pieces of information.First, we did not know how long an asymptomaticpatient with an end-systolic dimension greater than 55mm would remain asymptomatic. If symptoms wereto develop only after many years, the case foroperating before the onset of symptoms would be

weaker. Nor did we know how rapidly the end-systolicdimension changed during serial follow-up studies orwhether this measurement could be used to identifypatients who were likely to develop symptoms andrequire operation.The results of the present study are relevant to these

questions. We evaluated 37 patients who had isolated,chronic aortic regurgitation and had been followed forat least 1 year with two or more serial echocardio-grams. At the initial evaluation, none of the patientshad symptoms serious enough to warrant operation.Fourteen of the 37 patients (38%) subsequently de-veloped either severe dyspnea on exertion, congestiveheart failure, angina or syncope. Based on thedevelopment of symptoms, operation was recom-mended.The echocardiographic and electrocardiographic

measurements in these 14 patients were comparedwith measurements obtained in the 23 patients whohave not developed symptoms and thus have not cometo operation. Left ventricular end-systolic dimensionand left ventricular fractional shortening were themeasurements that on initial study most strongly dis-tinguished patients who eventually required operation(table 1). These measurements are highly correlatedand both reflect systolic function.'8 Left ventricularend-diastolic dimension and the Romhilt-Estes scorealso separated the two groups, but with lowerstatistical significance.When patients were subdivided based on their left

ventricular end-systolic dimension, operation was re-quired in four of the five patients (80%) with a left ven-tricular end-systolic dimension greater than 55 mm.The time between initial study and operation in thesefour patients averaged 39 months. Operation was alsorequired in six of 12 patients (50%) with an end-systolic dimension of 50-54 mm during an averagefollow-up of 23 months. In contrast, only four of the20 patients (20%) whose end-systolic dimension wasless than 50 mm required operation (fig. 5).

Findings in our operative series of symptomatic pa-tients with isolated aortic regurgitation are pertinentto these observations.'8 Approximately two-thirds ofthe operative patients had left ventricular end-systolicdimensions of 50 mm or greater at the time theydeveloped symptoms and came to operation. Of the 17operative patients with an end-systolic dimension lessthan 50 mm, seven (41%) had associated coronaryartery disease (fig. 6). In contrast, coronary diseasewas present in only three of the remaining 32 patients(9%) with end-systolic dimensions greater than 50 mm(p < 0.05 by Fisher's exact test27). Thus, when signifi-cant symptoms develop in patients with aortic regurgi-tation and left ventricular end-systolic dimension lessthan 50 mm, associated coronary disease is often pres-ent. Once this dimension is reached, half of initiallyasymptomatic patients with aortic regurgitation willdevelop symptoms and require operation during anaverage follow-up of 2-3 years.

Left ventricular end-systolic dimension may bemarkedly increased in some patients who do not com-plain of symptoms. Our experience with graded tread-

489

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

VOL 61, No 3, MARCH 1980

mill exercise testing is consistent with this observationin that exercise tolerance may be only mildly impairedin many patients with aortic regurgitation, even whenthe left ventricule is very large and its systolic func-

OPERATIVE GROUP9or

80

U,)

LLW

z

2

i

_n2

z

Q

0

z

C)

_>

701

60h

501-

401-

0

0U.

r

F

3rto

30t-

201h

10-

0

0

0

80

CD

0

0

Q_

0

0

a_______________ _______________

ALL PTS S PTS WITHPATIENTS CAD CAD

FIGURE 6. Plot of left ventricular (L V) end-systolicdimension for the 49 patients reported in our operativeseries.'8 The data shown in the left-hand column have beensubdivided and are shown in the two columns on the right.Subdivision of the patients was based on whether they hadcoronary artery disease (CAD) in addition to aortic valvereplacement. S = without.

tion considerably depressed (Bonow RO, Borer JS,Rosing DR, Henry WL, Pearlman AS, McIntosh CL,Morrow AG, Epstein SE: unpublished observations).The patients who remained asymptomatic during

follow-up were significantly younger than those whodeveloped symptoms and were recommended foroperation.9 However, there was considerable overlapin the ages in the nonoperative and subsequent opera-tion groups. Also, even relatively young patients(including two in their late twenties) presented withlarge, poorly contracting left ventricles.The data in the present study were evaluated to

determine the rate of change in left ventricular end-systolic dimension during routine clinical follow-upstudies. These serial echocardiographic data demon-strated that it is very uncommon for the end-systolicdimension to increase more rapidly than 7 mm peryear.

Before recommending that operation should be per-formed as soon as the end-systolic dimension exceeds55 mm, it is important to examine the alternative, i.e.,delaying operation until symptoms develop. Onepotential benefit of postponing operation is that im-provements in operative techniques may reduceoperative mortality. However, operative mortalitywas only 6% (excluding patients who required aorticroot reconstruction) in our recent series of patientswith isolated aortic regurgitation. This low operativemortality is similar to that noted in several other re-cent series.'2' 13, 17 Because six of seven patients (86%)with end-systolic dimensions greater than 55 mmdeveloped symptoms and came to operation during anaverage follow-up of 30 months (including the twoasymptomatic patients whose end-systolic dimensionincreased to more than 55 mm during serial study), itdoes not appear that operative risk will be significantlydecreased by waiting until asymptomatic patients withend-systolic dimensions greater than 55 mm developsymptoms. Moreover, there is evidence in ouroperative series that the risk of operative damage maybe increased in patients with very large left ventricularend-systolic dimensions.18 By waiting until symptomsdevelop in patients with large end-systolic dimension,one runs the risk that the end-systolic dimension mayincrease further and, as a result, the chances ofoperative damage may be increased and the likeli-hood of left ventricular function improving postop-eratively may be decreased.Another potential benefit of postponing operation is

that improvements in prosthetic valve design willdecrease late morbidity and mortality after valvereplacement. For instance, if the present mortality dueto valve malfunction were greater than the risk ofdeveloping congestive heart failure and dying, earlyintervention would be difficult to justify. However, inour recent operative series of symptomatic patientswith either isolated aortic regurgitation,"8 or isolatedaortic stenosis,28 only three of the combined total of 85patients (4%) who survived operation died of valve-related complications. In contrast, nine of 17 sympto-

CIRCULATION490

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

SERIAL ECHOES IN ASYMPTOMATIC AR/Henry et al.

matic patients (53%) operated upon for aorticregurgitation who had a preoperative left ventricularend-systolic dimension greater than 55 mm either diedat operation or developed late postoperative con-gestive heart failure and died. These deaths occurredwithin 46 months of operation. If patients with both apreoperative end-systolic dimension greater than 55mm and a fractional shortening less than 25% are con-sidered, the likelihood of a poor result is further in-creased: 85% of such patients had a poor operativeresult.'8Even if nonfatal valve complications, including

anemia, cerebrovascular accidents, infection, poppetmalfunction and perivalvular leak, are combined withthe fatal valve complications (unpublished obser-vations), the total incidence of all valve complications(20 of 85 patients or 24%) is still half (or less) of thecongestive heart failure mortality in the high-riskgroups. Therefore, it appears that postponing opera-tion until symptoms develop in an asymptomaticpatient with aortic regurgitation who has an end-systolic dimension greater than 55 mm results in agreater overall late mortality than the combined mor-bidity and mortality associated with prosthetic valve.Future improvements in prosthetic valve design willonly strengthen this argument.As a result of this information, we have changed our

previous policy of recommending operation only topatients with aortic regurgitation who have severedyspnea on exertion, congestive heart failure, anginaor syncope. We still recommend operation to patientswho present with or develop these symptoms duringfollow-up, regardless of the size or function of the leftventricle. Specifically, we do not withhold operation inpatients with a left ventricular end-systolic dimensiongreater than 55 mm or a fractional shortening lessthan 25% for two reasons. First, a few patients with anend-systolic dimension greater than 55 mm had asignificant increase in systolic function after opera-tion. This was particularly true if preoperative frac-tional shortening was greater than 25%, but much lesslikely in patients whose fractional shortening wasbelow 25%.18 Second, a few patients with lowpreoperative systolic function that did not improvepostoperatively are still doing reasonably well afterseveral years of follow-up.

In contrast to our previous policy of only recom-mending operation to symptomatic patients, we nowperform an extensive inpatient evaluation in allasymptomatic or minimally symptomatic patientswho have a left ventricular end-systolic dimension byechocardiography of 55 mm or greater. Unless leftventricular cineangiography and radionuclide angiog-raphy29 are in major conflict with the echocardio-graphic data (a situation that, in our experience, is un-common in patients with aortic regurgitation), werecommend operation, even in the absence of severedyspnea on exertion, overt congestive heart failure,angina or syncope.Asymptomatic patients with aortic regurgitation

who have a left ventricular end-systolic dimension lessthan 50 mm are followed medically. In our previousoperative study, we found that none of the patientswho were in this subgroup died at operation or duringpostoperative follow-up, which averaged 4 years.'8Therefore, there appears to be no reason to proceedwith operation in such patients in the absence of symp-toms, as long as patients are followed every 1-2 years.With this approach, some patients will develop symp-toms requiring operation while maintaining goodsystolic function (fig. 5) and should have an excellentlong-term postoperative prognosis. Alternatively,some patients will develop progressive impairment ofleft ventricular systolic function before the onset ofsymptoms. If these patients are followed closely byserial echocardiographic studies, operation can be per-formed as soon as the patient manifests an end-systolic dimension of 55 mm or fractional shorteningof 25%.Asymptomatic patients with severe aortic regurgi-

tation who fall into the subgroup characterized by aleft ventricular end-systolic dimension of 50-54 mmpresent a difficult therapeutic problem. If operation isperformed in these patients, irreversible left ven-tricular dysfunction can probably be prevented. Onlyone of the 32 patients (3%) in our operative series'8who had preoperative left ventricular end-systolicdimension of 55 mm or less developed congestiveheart failure in the late postoperative period, and thatpatient, as mentioned previously, suffered an in-traoperative myocardial infarction resulting in severeleft ventricular dysfunction.However, operating on patients within this sub-

group before symptoms develop would have beenpotentially advantageous for only two of the 12patients (19%); i.e., those who developed a left ven-tricular end-systolic dimension greater than 55 mmbefore coming to operation (figs. 4 and 5). Moreover,although six of 12 patients (50%) came to operationduring an average follow-up of 23 months after the ini-tial echocardiographic study, the other six (50%) arestill asymptomatic after a follow-up of 2-5 years.Operation in these latter six patients would have beenof questionable benefit. They would have been sub-jected prematurely to the risks of operation (6%operative risk in our series), as well as to the hazardsinherent in a prosthetic valve. Thus, while operatingon asymptomatic patients with aortic regurgitationwhose left ventricular end-systolic dimension is 50-54mm is not an unreasonable approach, the majority ofsuch patients can be followed with only a small risk ofdeveloping irreversible left ventricular dysfunctionbefore symptoms occur. Therefore, we do not rec-ommend operation to asymptomatic patients in thissubgroup.With these considerations in mind, our current ap-

proach to the clinical management of the asymp-tomatic patient with chronic aortic regurgitation is toassess left ventricular size and function by serialechocardiograms. The frequency of examination and

491

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

VOL 61, No 3, MARCH 1980

TABLE 3. Clinical Management and Left Ventricular (LV)Dimension

LV dimension(systole) Clinical management

< 45 mm Medical therapy; echo examevery 2 years.

45-49 mm Medical therapy; echo examevery year.

50-54 mm Medical therapy; echo examevery 4-6 months.

> 55 mm Operate, even withoutsymptoms.

our clinical decisions are based on the left ventricularend-systolic dimension as shown in table 3.

Acknowledgments

The authors appreciate the cooperation and assistance of CoraBurn, Estelle Cohen and Joyce McKay, who performed the echo-cardiographic studies. The statistical analysis of the data was per-formed by Dr. James Ware, who was aided by Erica Britain. Theirhelp was essential to the completion of the study. The authors alsogreatly appreciate the skillful assistance of Exa Murray, who typedthe manuscript.

References1. Selzer A: Cardiac valve replacement: an unanswered question.Am J Cardiol 37: 322, 1976

2. Segal J, Harvey WP, Hufnagel C: A clinical study of one hun-dred cases of severe aortic insufficiency. Am J Med 21: 200,1956

3. Bland EF, Wheeler EO: Severe aortic regurgitation in youngpeople. N Engl J Med 256: 667, 1957

4. Massell BF, Amezcua FJ, Czoniczer G: Prognosis of patientswith pure or predominant aortic regurgitation in the absence ofsurgery. Circulation 34 (suppl III): 111-164, 1966

5. DeGeorges M, Delzant JF: Elements de prognostic d l'in-suffisance aortique isolee recueillis chez 206 malades ages demoins de 50 ans. Sem Hop Paris 42: 1171, 1966

6. Spagnuolo M, Kloth H, Taranta A, Doyle E, Pasternack B:Natural history of rheumatic aortic regurgitation: criteriapredictive of death, congestive heart failure, and angina inyoung patients. Circulation 44: 368, 1971

7. Rapaport E: Natural history of aortic and mitral disease. Am JCardiol 35: 221, 1975

8. Gault JH, Covell JW, Braunwald E, Ross J: Left ventricularperformance following correction of free aortic regurgitation.Circulation 42: 773, 1970

9. Goldschlager N, Pfeifer J, Cohn K, Porrer R, Selzer A: Thenatural history of aortic regurgitation: a clinical andhemodynamic study. Am J Med 54: 577, 1973

10. Cohn PF, Gorlin R, Cohn LH, Collins JJ Jr: Left ventricularejection fraction as a prognostic guide in surgical treatment ofcoronary and valvular heart disease. Am J Cardiol 34: 136,1974

11. Hirshfeld JW, Epstein SE, Roberts AJ, Glancy DL, MorrowAG: Indices predicting long-term survival after valve replace-ment in patients with aortic regurgitation and patients with aor-tic stenosis. Circulation 50: 1190, 1974

12. Isom OW, Dembrow JM, Glassman E, Pasternack BS, SacklerJP, Spencer FC: Factors influencing long-term survival afterisolated aortic valve replacement. Circulation 50 (suppl II): II-154, 1974

13. Barnhorst DA, Oxman HA, Connolly DC, Pluth JR, DanielsonGK, Wallace RB, McGoon DC: Long-term follow-up ofisolated replacement of the aortic or mitral valve with the Starr-Edwards prosthesis. Am J Cardiol 35: 228, 1975

14. Bristow JD, Kremkau EL: Hemodynamic changes after valvereplacement with Starr-Edwards prostheses. Am J Cardiol 35:716, 1975

15. Smith HJ, Neutze JM, Roche AHG, Agnew TM, Barratt-Boyes BG: The natural history of rheumatic aortic regurgita-tion and the indications for surgery. Br Heart J 38: 147, 1976

16. Fischl SJ, Gorlin R, Herman MV: Cardiac shape and functionin aortic valve disease: physiologic and clinical implications.Am J Cardiol 39: 170, 1977

17. Copeland JG, Griepp RB, Stinson EB, Shumway NE: Long-term follow-up after isolated aortic valve replacement. J ThoracCardiovasc Surg 74: 875, 1977

18. Henry WL, Bonow RO, Borer JS, Ware JH, Kent KM,Redwood DR, McIntosh CL, Morrow AG, Epstein SE: Obser-vations on the optimum time for operative intervention for aor-tic regurgitation. I. Evaluation of the results of aortic valvereplacement in symptomatic patients. Circulation 61: 471, 1980

19. Henry WL, Ware J, Gardin JM, Hepner S, McKay J, WeinerM: Echocardiographic measurements in normal subjects:growth-related changes that occur between infancy and earlyadulthood. Circulation 57: 278, 1978

20. Gardin JM, Henry WL, Savage DD, Ware JH, Burn C, BorerJS: Echocardiographic measurements in normal subjects:evaluation of an adult population without clinically apparentheart disease. J Clin Ultrasound 7: 439, 1979

21. Feigenbaum H: Echocardiography, 2nd ed. Philadelphia, Lea& Febiger, 1976, p 464

22. Henry WL, Clark CE, Epstein SE: Asymmetric septal hyper-trophy (ASH): echocardiographic identification of thepathognomonic anatomic abnormality of IHSS. Circulation47: 225, 1973

23. Griffith JM, Henry WL: Switched gain - a technique forsimplifying ultrasonic measurement of cardiac wall thickness.IEEE Trans Biomed Eng 22: 337, 1975

24. Popp RL, Harrison DC: Ultrasound cardiac echography fordetermining stroke volume and valvular regurgitation. Circula-tion 41: 493, 1970

25. Fortuin NJ, Hood WP Jr, Sherman ME: Determination of leftventricular volume by ultrasound. Circulation 44: 575, 1971

26. Troy BL, Pombo J, Rackley CE: Measurement of left ven-tricular wall thickness and mass by echocardiography. Circula-tion 45: 602, 1972

27. Kendall MG, Stuart A: Advanced Theory of Statistics, vol. 3.New York, Hafner Publishing Co, 1961

28. Henry WL, Bonow RO, Borer JS, Kent KM, Ware JH,Redwood DR, Itscoitz SB, McIntosh CL, Morrow AG, EpsteinSE: Evaluation of aortic valve replacement in patients withvalvular aortic stenosis. Circulation 61: 1980 In press

29. Borer JS, Bacharach SL, Green MV, Kent KM, Henry WL,Rosing DR, Siedes SF, Johnston GS, Epstein SE: Exercise-induced left ventricular dysfunction in symptomatic and asymp-tomatic patients with aortic regurgitation: assessment byradionuclide cineangiography. Am J Cardiol 42: 351, 1978

CIRCULATION492

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

W L Henry, R O Bonow, D R Rosing and S E EpsteinSerial echocardiographic evaluation of asymptomatic patients.

Observations on the optimum time for operative intervention for aortic regurgitation. II.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1980 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.61.3.484

1980;61:484-492Circulation. 

http://circ.ahajournals.org/content/61/3/484.citationthe World Wide Web at:

The online version of this article, along with updated information and services, is located on

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from