Today’s Webinar will begin at noon

Today’s Webinar will begin at noon

3/27/12

Welcome from Barb DeBaun, RN, MSN, CIC

Introduction

Introduction

• Please do not put your phone on hold; use the mute function or *6

• Please type questions or comments into text box

• If time permits, we will open up the phone lines at the conclusion of the presentation

Heather Young, MD

Surgical Site Infection after Hysterectomy in Colorado

•Heather Young MD•March 27, 2012

WHYWHY

WHAT

HOW

Objectives

• WHY– Public reporting in CA, CO and nationwide– Our SSI problem

• WHAT– Pilot and statewide study

• HOW– Transfusion and outcomes– Transfusion and immunity– Transfusion at our hospital

Why we started this project

WHYWHY

WHAT

HOW

Public Reporting of HAI

Committee to Reduce Infection Deaths 2010

StateState

CMSCMS

>=<

>=<

Pay for performanceOnline hospital comparisons

Public Reporting Agencies

NHSN

NHSN

All orthopedic, cardiac, and GI surgeries including:

Hip prosthesisKnee prosthesisOpen fracture fixationLaminectomySpinal fusionSpinal refusion

Coronary bypass graftingCardiac surgeryPacemaker surgery

Colon surgeryRectal surgerySmall bowel surgeryGallbladder surgeryAppendectomyGastric surgeryBiliary, liver, pancreatic surgeries

Backer 2011

AUGUST 2007Coronary artery bypass graftingPartial & total knee replacementPartial & total hip replacement

AUGUST 2008Herniorrhaphy

AUGUST 2009Vaginal hysterectomy

Abdominal hysterectomy

Reese 2011

JANUARY 2012Colon surgeryBreast surgery

Why we started this project

WHY

WHAT

HOW

What we did and What we found

Preparation for Reporting

Jul-Dec 2006

Jan-Jun 2007

Jul-Dec 2007

Jan-Jun 2008

Jul-Dec 2008

Jan-Jun 2009

NationalRate

Why we started this project

WHY

WHAT

HOW

What we did and What we found

Pilot Data, Study Design

Single center, retrospective chart review

Inclusion:• All patients who had total abdominal hysterectomy (TAH)

• Dec 30, 2005, to Mar 9, 2010• Age ≥ 18 y.o.

Exclusion:• Emergent surgery• Surgery for known infection (tubo-ovarian abscess)

• No follow-up documented in 30 days after TAH

Young 2011

Hypotheses

• Do our surgeons adhere to evidence-based guidelines?

• Are there other published variables that we should consider for intervention?

Study Population N (%)

Total number of patients

229 (100%)

Excluded from analysisEmergent indication

No TAH performedSurgery for TOA

No follow-up in 30 d after TAH

37 (16.2%)26542

Included in analysis

192 (83.8%)

Developed SSI 21 (10.9%)

Inclusion/Exclusion

Young 2011

Patient & Hospital Variables

PMHx, PSHx

Indication for TAH

Length of time in OR

Estimated blood loss (EBL)

Blood transfusion

Concomitant OR procedures

Type of skin incision

Surgeon

ASA score

Wound class

SCIP measures compliance

Olsen 2009; Ahmed 2001; Taylor 1998; Shapiro 1982; Meltomaa 2000; Molina-Cabrillana 2008; Ghezi 2009; Leung 2007; Persson 1996

Adherence to SSI Bundle

0

20

40

60

80

100

120

Perc

enta

ge

No SSI SSI

Antibiotics within 60 minutes of incision

Appropriate antibiotics

PACU arrival temperature ≥36 degrees

CompositeYoung 2011

Green = SSIBlue = No SSI

VariableSSI N, %

No SSI N, %

P

Diabetes 419.1%

1810.5% 0.247

Smoking 419.1%

5833.9% 0.169

ObesityBMI ≥ 30

1466.7%

7645.1% 0.061

Blood transfusio

n

628.6%

1810.5% 0.018

EBL ≥ 500 mL

1047.6%

4725.7% 0.035

Selected Variables and SSI

Young 2011

VariableSSI N, %

No SSI N, %

P

Diabetes 419.1%

1810.5% 0.247

Smoking 419.1%

5833.9% 0.169

ObesityBMI ≥ 30

1466.7%

7645.1% 0.061

Blood transfusio

n

628.6%

1810.5% 0.018

EBL ≥ 500 mL

1047.6%

4725.7% 0.035

Selected Variables and SSI

Young 2011

Laboratory Variables

Pre-operative creatinine

Pre-operative complete blood count

Post-operative hemoglobin / hematocrit

Microbiology, if applicable

Meltomaa 2000

Lab ValueSSI

Mean (SD)No SSI

Mean (SD)P

WBC 7.75 (2.66)7.65 (2.67)

0.869

Platelet 359 (133) 335 (137) 0.448

Hematocrit37.94 (4.10)

39.26 (4.87)

0.234

Post-operative hematocrit

31.81 (4.13)

31.67 (5.74)

0.913

Delta hematocrit

6.13 (4.06)7.62 (4.94)

0.187

Lab Values and SSI

Young 2011

Model I Model II

N = 192EBL <500

mLN = 138

EBL ≥500 mL

N = 54

Blood transfusio

n

3.581.21, 10.62

7.561.59, 35.98

1.140.25, 5.16

BMI ≥302.55

0.96, 6.74--- ---

Multivariate Analysis

Young 2011

Reasons for Blood Transfusion

Variable Frequency“Anemia” 72.7%

“High amount of EBL”

27.3%

“Dizziness” or “shortness of

breath”24.2%

“Comorbid heart disease”

18.2%

“Use of anticoagulants”

9.1%

Young 2011

Discussion

1.Obesity trends toward significance

2.Blood transfusion associated with SSI after TAH, especially if EBL <500 mL

1.Most indications for blood transfusion are subjective and potentially modifiable

Young 2011

Strengths & Weaknesses

Excellent post-discharge follow-up (99%)

Use of standard definitions

Single centerRelatively small number of infections and transfusions

Young 2011

Next Steps

• Expand to multicenter• State of Colorado• Survey of infection control nurses• IRB approved at 35 / 51 hospitals that perform TAH 8 still pending at publication 8 declined participation

• Colorado Hospitals Association (CHA)

Methods

• Survey designed• Website developed, supported by CHA

• Emails sent to all infection control nurses at participating facilities

• Stipend offered for participation

Young 2012

Participants

• 6 facilities• 567 subjects entered into database

• 20 excluded• 547 subjects analyzed

•10 (1.8%) developed SSI•35 (6.5%) had blood transfusion

Young 2012

Variable SSI No SSI P

BMI ≥30 60.0% 40.4% 0.33

Hct <30 0.0% 5.0% 1.00

EBL ≥500 mL 20.0% 6.7% 0.15

Blood transfusion

20.0% 6.2% 0.13

Antibiotic < 60 min before incision

100.0% 96.4% 1.00

Associations with SSI

Young 2012

Variable SSI No SSI P

BMI ≥30 60.0% 40.4% 0.33

Hct <30 0.0% 5.0% 1.00

EBL ≥500 mL 20.0% 6.7% 0.15

Blood transfusion

20.0% 6.2% 0.13

Antibiotic < 60 min before incision

100.0% 96.4% 1.00

Associations with SSI

Young 2012

OR to develop SSI95% Confidence Interval

EBL <500 mL EBL ≥500 mL

Blood transfusion

7.201.37, 37.72

---

Stratification Analyses

Young 2012

Discussion

• Blood transfusion and EBL ≥500 mL continue to be risk factors for SSI across several diverse institutions

• Biggest contribution of blood transfusion to SSI was in cases of EBL <500 mL

Young 2012

EBL 1200 mL

EBL 300 mL

pRBC pRBC

Who is more likely to have SSI?

How blood transfusions relate to patient outcomes

How blood transfusions relate to immune function

How we changed transfusions in hospital practice (sort of!)

WHY

WHAT

HOW

Transfusion-Related Immunomodulation (TRIM)

Vamvakas 2007

Beneficial effects Detrimental effects

↑ renal transplant survival

↓ Crohn’s disease flares

↓ miscarriage rate

↑ cancer recurrence

↑ post-operative bacterial infection

↑ mortality

TRIM and Hospital Outcomes

Blood transfusions are associated with worse

hospital outcomesEspecially in patients who are not severely ill or profoundly anemic

Hebert et al 1999

ICU Mortality– 838 patients, multicenter, Canada– Randomized controlled trial – Blood transfusion threshold of Hg 7-9

vs 10-12 in community ICU patients– Outcome: 30-day death

Hebert et al 1999

Received less blood

Received more blood

Hebert et al 1999

Received less blood

Received more blood

Hebert et al 1999

Received less blood

Received more blood

Rao et al 2004

Acute Coronary Syndrome– 24,112 patients, multicenter,

international– Observational study– Impact of blood transfusion in patients

with ACS– Outcome: 30-day death

Rao et al 2004

– 24,112 patients, international– Impact of blood transfusion on 30-day death in

patients with ACSNadir Hematocrit

20 25 30 35

Adjusted OR

(95% CI)

1.59 (0.95-2.66)

1.13 (0.70-1.82)

168.64 (7.49-3798)

291.64 (10.28-8274)

OR for 30-day death (by Hct)in patients w/ vs w/o blood transfusion

Rao et al 2004

– 24,112 patients, international– Impact of blood transfusion on 30-day death in

patients with ACSNadir Hematocrit

20 25 30 35

Adjusted OR

(95% CI)

1.59 (0.95-2.66)

1.13 (0.70-1.82)

168.64 (7.49-3798)

291.64 (10.28-8274)

OR for 30-day death (by Hct)in patients w/ vs w/o blood transfusion

Blood Transfusion & Surgical Site Infection

Author Year Surgical ProcedureMultivariate Analysis

Bernard 2009Visceral, vascular,

traumaOR 1.3

Rogers 2009Coronary artery

bypassOR 2.3

Poon 2009Laparoscopic colorectal

OR 2.4

Olsen 2008 Breast OR 3.4

Asensio 2008 Liver transplant OR 2.0

Blumetti 2007 Colorectal OR 2.3

Walz 2006Small bowel, colorectal

OR 1.6

Talbot 2004 Sternotomy OR 3.2

Olsen 2003Saphenous vein

harvestOR 2.8

Olsen 2009 Hysterectomy OR 2.4

Schwarzkopf 2010Thoracic, lumbar

spineOR 8.0

Costello 2010 Cardiac, pediatric OR 7.9

Tang 2001 Colorectal OR 5.3

Innerhofer 2005Hip and knee arthroplasty

OR 23.7

Koch et al 2008

Cardiac surgery– Newer vs older blood (<14 vs ≥14 days)– Outcomes: “serious adverse events” and

long term survival– Older blood associated with more serious

complications and deaths

Koch et al 2008

Death

VentilationRenal failureSepsis

Multiorgan failureLimb ischemiaComposite

Koch et al 2008NEWER BLOOD

OLDER BLOODSURV

IVAL

Offner et al 2002

Trauma surgery– Newer vs older blood (<14 vs ≥14 days)– Outcomes: “major infectious complications”– Older blood associated with more infections

Offner et al 2002

>14 Days

>21 Days

Mean # PRBC transfused

Yellow = no infectionBlue = major infection

Offner et al 2002

How does blood change the immune system?

Theory: Soluble factors in blood products accumulate over time and contribute to diminished immune

function.

Muylle et al 1993

Platelet study

Two phases:• (1) IL6, TNFα, IL1β at storage day 3, 5, 7• (2) Correlate IL6, TNFα, IL1β with febrile

transfusion reactions

Muylle et al 1993

(1) Cytokines at storage day 3, 5, 7• ↑IL6 in 8/12 samples at day 3• ↑IL6 in 10/12 samples at day 5, 7• ↑TNFα and IL1β in samples with ↑IL6• Linear increases with storage time• Increased cytokines about 100-1000x per-

storage levels

Similar findings found by Stack & Snyder 1994

Muylle et al 1993

(2) Correlate IL6, TNFα, IL1β with febrile transfusion reactions

• 45 platelet tx, 6 febrile reactions• ↑IL6 and TNFα in plasma of platelets that caused

reactions (p <0.001)

Muylle 1993

Conclusion: Perhaps some

transfusion reactions are due to transfusion of cytokine-rich plasma

Wadhwa et al 2000

Cytokine levels in WBC-reduced pRBC– Does the timing of WBC-reduction

influence cytokine levels over storage time?

– IL8, TGFβ1, RANTES

Wadhwa et al 2000

Conclusion:– Cytokines tend to increase over time in

stored blood products– Pre-storage plasma filtration prevents

cytokine accumulation compared to post-storage

Offner et al 2002

How does blood change the immune system?

Theory: Soluble factors in blood products accumulate over time and contribute to diminished immune

function

Muylle et al 1993

Platelet study

Two phases:• (1) IL6, TNFα, IL1β at storage day 3, 5, 7• (2) Correlate IL6, TNFα, IL1β with febrile

transfusion reactions

Muylle et al 1993

(1) Cytokines at storage day 3, 5, 7• ↑IL6 in 8/12 samples at day 3• ↑IL6 in 10/12 samples at day 5, 7• ↑TNFα and IL1β in samples with ↑IL6• Linear increases with storage time• Increased cytokines about 100-1000x per-

storage levels

Similar findings found by Stack & Snyder 1994

Muylle et al 1993

2) Correlate IL6, TNFα, IL1β with febrile transfusion reactions

• 45 platelet tx, 6 febrile reactions• ↑IL6 and TNFα in plasma of platelets that caused

reactions (p <0.001)

Muylle 1993

Conclusion: Perhaps some

transfusion reactions are due to transfusion of cytokine-rich plasma

Wadhwa et al 2000

Cytokine levels in WBC-reduced pRBC– Does the timing of WBC-reduction influence

cytokine levels over storage time? – IL8, TGFβ1, RANTES

Wadhwa et al 2000

Conclusion:– Cytokines tend to increase over time in

stored blood products– Pre-storage plasma filtration prevents

cytokine accumulation compared to post-storage

Summary

• Immune mediators are present and accumulate over time in blood products

Summary

• Immune mediators are present and accumulate over time in blood products

• The quantity of innate immune mediators in blood products is sufficient to cause febrile reactions to platelets

Summary

• Immune mediators are present and accumulate over time in blood products

• The quantity of innate immune mediators in blood products is sufficient to cause febrile reactions to platelets

• Levels of innate immune inhibitors in patients after receiving blood products have not been studied

Feedback, Follow-up

Jul-Dec 2006

Jan-Jun 2007

Jul-Dec 2007

Jan-Jun 2008

Jul-Dec 2008

Jul-Dec 2009

Jul-Dec 2010

Jan-Jun 2009

Jan-Jun 2010

Young 2011

Thank You

QUESTIONS?

[email protected]

[email protected]

Notes will be on our website

Cynosure Quality Swap MeetMay 21, 2012

www.cynosurehealth.org

Thanks for joining us today