Sumeet Subherwal, Richard G. Bach, Anita Y. Chen, Brian F. Gage, Sunil V. Rao, Sumeet Subherwal, Richard G. Bach, Anita Y. Chen, Brian F. Gage, Sunil V. Rao, Tracy Y. Wang, W. Brian Gibler, E. Magnus Ohman, Matthew T. Roe, Eric D. Tracy Y. Wang, W. Brian Gibler, E. Magnus Ohman, Matthew T. Roe, Eric D.

Peterson, Karen P. Alexander Peterson, Karen P. Alexander Duke Clinical Research Institute and Washington University St. LouisDuke Clinical Research Institute and Washington University St. Louis

Sumeet Subherwal, Richard G. Bach, Anita Y. Chen, Brian F. Gage, Sunil V. Rao, Sumeet Subherwal, Richard G. Bach, Anita Y. Chen, Brian F. Gage, Sunil V. Rao, Tracy Y. Wang, W. Brian Gibler, E. Magnus Ohman, Matthew T. Roe, Eric D. Tracy Y. Wang, W. Brian Gibler, E. Magnus Ohman, Matthew T. Roe, Eric D.

Peterson, Karen P. Alexander Peterson, Karen P. Alexander Duke Clinical Research Institute and Washington University St. LouisDuke Clinical Research Institute and Washington University St. Louis

The CRUSADE Bleeding ScoreThe CRUSADE Bleeding Score

A validated risk prediction tool for estimation of baseline risk of in-

hospital major bleeding in patients with NSTEMI

Validated risk stratification tools exist for baseline Validated risk stratification tools exist for baseline ischemic risk (ischemic risk (TIMI, PURSUIT, GRACE ACS), however estimation of baseline bleeding risk in patients with estimation of baseline bleeding risk in patients with NSTEMI is difficult because existing tools: NSTEMI is difficult because existing tools:

include treatment variables (i.e. antithrombotics or include treatment variables (i.e. antithrombotics or invasive procedures) invasive procedures)

derived from highly selected patient populationsderived from highly selected patient populations

Validated risk stratification tools exist for baseline Validated risk stratification tools exist for baseline ischemic risk (ischemic risk (TIMI, PURSUIT, GRACE ACS), however estimation of baseline bleeding risk in patients with estimation of baseline bleeding risk in patients with NSTEMI is difficult because existing tools: NSTEMI is difficult because existing tools:

include treatment variables (i.e. antithrombotics or include treatment variables (i.e. antithrombotics or invasive procedures) invasive procedures)

derived from highly selected patient populationsderived from highly selected patient populations

BackgroundBackground

CRUSADE Quality Improvement Initiative February 15, 2003 to December 31, 2006 n=89,134 NSTEMI patients at 485 US hospitals

•Excluded unstable angina, home warfarin, transfer out, deaths within 48 hours

In-hospital Major Bleeding (Censored at CABG) Absolute HCT drop ≥12% (Baseline – Nadir) Intracranial hemorrhage or retroperitoneal bleed Transfusion if baseline HCT ≥28% Transfusion if baseline HCT <28% AND witnessed bleed

Divided into derivation (80% of N) and validation cohorts (20% of N)

Incorporated clinically and statistically significant univariate associations into a multivariable model using generalized estimating equations (GEE)

CRUSADE Quality Improvement Initiative February 15, 2003 to December 31, 2006 n=89,134 NSTEMI patients at 485 US hospitals

•Excluded unstable angina, home warfarin, transfer out, deaths within 48 hours

In-hospital Major Bleeding (Censored at CABG) Absolute HCT drop ≥12% (Baseline – Nadir) Intracranial hemorrhage or retroperitoneal bleed Transfusion if baseline HCT ≥28% Transfusion if baseline HCT <28% AND witnessed bleed

Divided into derivation (80% of N) and validation cohorts (20% of N)

Incorporated clinically and statistically significant univariate associations into a multivariable model using generalized estimating equations (GEE)

CRUSADE Major Bleeding ModelCRUSADE Major Bleeding Model

VariableDerivation Cohort

(N = 71,277)Validation Cohort

(N = 17,857) Age (years)* 67.0 (56.0, 79.0) 67.0 (56.0, 79.0)

Male 60.2% 60.3%

Family history of CAD 33.9% 33.9%

History of hypertension 70.5% 70.6%

Diabetes mellitus 32.7% 32.5%

Prior vascular disease† 18.4% 18.1%

Current/recent smoker 28.4% 27.8%

Hyperlipidemia 52.0% 51.7%

Prior myocardial infarction 28.1% 27.9%

Prior CABG 18.2% 18.5%

Baseline HCT (%)* 40.7 (36.5, 44.2) 40.7 (36.6, 44.1)

CrCl (mL/min)*‡ 70.3 (43,8, 101.9) 70.8 (44.0, 102.0)

In-Hospital Events

Death > 48 hrs  2.7% 2.6%

Major bleeding 9.4% 9.6%

*Median (25th, 75th percentile)†Prior vascular disease defined as h/o stroke or peripheral arterial disease‡ Creatinine clearance as estimated by Cockcroft-Gault Formula

Baseline CharacteristicsBaseline Characteristics

Prior vascular disease defined as h/o stroke or peripheral arterial diseaseNote: Heart rate is truncated @ <70 bpm; CrCl: Cockcroft-Gault is truncated @ >90 mL/min

Variable 2 Derivation CohortOR 95% CI

Validation Cohort OR 95% CI

Baseline HCT <36% (vs. ≥ 36%) 434.6 2.28 2.11-2.46 2.17 1.92-2.44

CrCl (per 10 mL/min decrease) 433.2 1.12 1.10-1.13 1.11 1.09-1.13

Heart rate (per 10 bpm increase) 159.2 1.08 1.07-1.10 1.09 1.07-1.12

Female 77.8 1.31 1.23-1.39 1.33 1.19-1.50

Signs of heart failure 37.7 1.23 1.15-1.31 1.13 1.01-1.28

Prior vascular disease 30.4 1.19 1.12-1.27 1.10 0.98-1.24

Diabetes mellitus 26.6 1.16 1.10-1.23 1.25 1.12-1.40

SBP ≤ 110 mm Hg (vs. 110-180) 12.6 1.26 1.16-1.36 1.27 1.10-1.47

SBP ≥ 180 mm Hg (vs. 110-180) 1.24 1.14-1.35 1.18 1.02-1.37

c-Statistic 0.72 0.71

Multivariable Predictors of BleedingMultivariable Predictors of Bleeding

CRUSADE Bleeding ScoreCRUSADE Bleeding ScoreCRUSADE Bleeding ScoreCRUSADE Bleeding Score

The CRUSADE Bleeding Score was developed by assigning a weighted integer to each independent predictor based on the predictor’s coefficient in the reduced regression model

The CRUSADE Bleeding Score (range 1-100 points) equals the sum of weighted integers for independent predictors

The CRUSADE Bleeding Score was developed by assigning a weighted integer to each independent predictor based on the predictor’s coefficient in the reduced regression model

The CRUSADE Bleeding Score (range 1-100 points) equals the sum of weighted integers for independent predictors

CRUSADE Bleeding Score Nomogram NomogramCRUSADE Bleeding Score Nomogram Nomogram

Note: Heart rate is truncated @ <70 bpm; CrCl: Cockcroft-Gault is truncated @ >90 mL/min; Prior Vascular disease is defined as prior PAD or stroke

PredictorPredictor RangeRange ScoreScoreBaseline Hematocrit (%) < 31

31-33.934-36.937-39.9

≥ 40

97320

Creatinine Clearance (mL/min) ≤ 15>15-30>30-60>60-90

>90-120>120

3935281770

Heart rate (bpm) ≤ 7071-8081-90

91-100101-110111-120≥ 121

01368

1011

Sex MaleFemale

08

Signs of CHF at presentation NoYes

07

Prior Vascular Disease NoYes

06

Diabetes Mellitus NoYes

06

Systolic blood pressure (mm Hg)

≤ 9091-100

101-120121-180181-200≥ 201

1085135

CRUSADE Bleeding ScoreCRUSADE Bleeding Score

CRUSADE Bleeding Score Risk QuintilesCRUSADE Bleeding Score Risk Quintiles

Patients were categorized into risk quintiles based on Patients were categorized into risk quintiles based on CRUSADE Bleeding ScoreCRUSADE Bleeding Score

Patients were categorized into risk quintiles based on Patients were categorized into risk quintiles based on CRUSADE Bleeding ScoreCRUSADE Bleeding Score

Risk N Min Score Max Score Bleeding

Very low 19,486 1 20 3.1%

Low 12,545 21 30 5.5%

Moderate 11,530 31 40 8.6%

High 10,961 41 50 11.9%

Very High 15,210 51 91 19.5%

Major Bleeding by Risk Quintiles in Derivation and Validation Cohorts

Major Bleeding by Risk Quintiles in Derivation and Validation Cohorts

Major Bleeding by Antithrombotic therapyMajor Bleeding by Antithrombotic therapy

≥2 Antithrombotics (anti-platelet [aspirin or clopidogrel], anti-coagulant, or GP IIb/IIIa; n=50,969; c-index 0.72)<2 Antithrombotics (anti-platelet, anti-coagulant, or GP IIb/IIIa; n=5,931; c-index 0.73)

Major Bleeding by Invasive or Conservative Approach

Major Bleeding by Invasive or Conservative Approach

Conservative Approach (≥2 antithrombotics and no catheterization; n=3,200; c-index 0.68)Invasive Approach (≥2 antithrombotics and no catheterization; n=43,492; c-index 0.73)

Mortality in those who did or did not have a Major Bleed across Risk Quintiles

Mortality in those who did or did not have a Major Bleed across Risk Quintiles

ConclusionConclusion

The CRUSADE Bleeding Score combines 8 predictors The CRUSADE Bleeding Score combines 8 predictors of major bleeding into a simple validated prediction tool of major bleeding into a simple validated prediction tool that estimates baseline risk of in-hospital major bleeding that estimates baseline risk of in-hospital major bleeding in patients with NSTEMI in patients with NSTEMI

Preserved discrimination across treatment subgroupsPreserved discrimination across treatment subgroups

Complements ischemic risk prediction tools to Complements ischemic risk prediction tools to better better enable clinicians to consider the potential adverse enable clinicians to consider the potential adverse outcomes in patients with NSTEMI prior to initiation of outcomes in patients with NSTEMI prior to initiation of therapytherapy

The CRUSADE Bleeding Score combines 8 predictors The CRUSADE Bleeding Score combines 8 predictors of major bleeding into a simple validated prediction tool of major bleeding into a simple validated prediction tool that estimates baseline risk of in-hospital major bleeding that estimates baseline risk of in-hospital major bleeding in patients with NSTEMI in patients with NSTEMI

Preserved discrimination across treatment subgroupsPreserved discrimination across treatment subgroups

Complements ischemic risk prediction tools to Complements ischemic risk prediction tools to better better enable clinicians to consider the potential adverse enable clinicians to consider the potential adverse outcomes in patients with NSTEMI prior to initiation of outcomes in patients with NSTEMI prior to initiation of therapytherapy