Bleeding - a Predictor of Mortality

In clinical trials, up to 30% of patients with acute coronary syndromes (ACS) or undergoing PCI experience bleeding complications, and even higher rates have been reported in contemporary practice.¹

A growing body of data suggests a strong correlation between bleeding and both short- and long-term adverse outcomes including mortality, which is independent of baseline characteristics and remains evident in most trials, despite variations in the definition of major bleeding.¹

Bleeding Events Are Predictors of Mortality in PCI

Large-scale independent analysis demonstrates the link between bleeding and mortality

• In a meta-analysis of the OASIS* and CURE* randomized trials (N=34,146), mortality rates at 30 days were 5 times higher in ACS patients with major bleeding compared to those without²

Reprinted from Circulation, Eikelboom JW, Mehta SR, Anand SS, Xie C, Fox KAA, Yusuf S. Adverse impact of bleeding on prognosis in patients with acute coronary syndromes, 114:774-782, copyright 2006, with permission from American Heart Association, Inc.

CURE=Clopidogrel in Unstable angina to prevent Recurrent ischemic Events (study);
OASIS=Organization to Assess Strategies for Ischemic Syndromes (study).
* OASIS-1 and -2 registries enrolled 11,500 patients without persistent STEMI, randomized to IV UFH or hirudin. CURE enrolled 12,562 patients with ACS without persistent STEMI randomized to clopidogrel or placebo in addition to aspirin for 3 to 12 months. Major bleeding was defined the same way in each study as bleeding that was significantly diabling required transfusion of ≥2 units of packed cells or was life-threatening.²

Postprocedural bleeding is associated with 1-year mortality

• In a pooled analysis of 4 ISAR* trials of patients undergoing PCI (N=5,384), patients with bleeding^† within 30 days had more than 4 times greater risk of mortality at 1 year than did patients without bleeding³

Reprinted from J Am Coll Cardiol. Ndrepepa G, Berger PB, Mehilli J, et al. Periprocedural bleeding and 1-year outcome after percutaneous coronary interventions: appropriateness of including bleeding as a component of a quadruple end point, 51:690-697, copyright 2008.

* ISAR-REACT, ISAR-SWEET, ISAR-SMART-2, and ISAR-REACT-2.

ISAR-REACT=Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action for Coronary Treatment;
ISAR-SMART-2=Intracoronary Stenting or Angioplasty for Restenosis Reduction in Small Arteries-2;
ISAR-SWEET=Intracoronary Stenting and Antithrombotic Regimen: Is Abciximab a Superior Way to Eliminate Elevated Thrombotic Risk in Diabetics.
^† The ISAR analysis included 5,384 patients from 4 randomized placebo-controlled trials on the value of abciximab after pretreatment with 600 mg of clopidogrel; ISAR-REACT, -SWEET, -SMART-2, and -REACT-2. Bleeding (defined according to TIMI criteria) included all bleeding events within 30 days after enrollment. The primary end point was 1 year mortality.³

Bleeding was strongly associated with mortality

• In the ACUITY PCI subset analysis (N=7,789), patients with major bleeding* without myocardial infarction (MI) experienced a 39% increase in mortality compared with patients without major bleeding with MI⁴

* Please see important safety information and bleeding definitions.

• There is a step-wise increase in short- and long-term mortality with increasing bleeding severity^5,6
• Even mild bleeding has been associated with a statistically significant increase in the adjusted hazard for death and death or MI⁶
• Access-site hematoma requiring a blood transfusion leads to increased mortality^5,7
• Major bleeding is associated with^8-10

  – Age >75 years

  – Use of heparin

  – Female gender

  – Renal insufficiency

  – Longer procedure GP IIb/IIIa use

  – More provisional

  – Anemia

  – Complicated lesions

  – Longer hospital stay

Patients who experience a major hemorrhage are more likely to die within 30 days

In REPLACE-2, patients who had a major hemorrhage were 30 times more likely to die within 30 days than those who did not have a major bleeding complication (10/195 [5.2%] vs 9/5007 [0.2%], P < .001).^8,9 Multivariate analysis identified major hemorrhage as the third strongest predictor of 1-year mortality (Odds Ratio, 3.53; 95% CI 1.91-6.53,
P < .0001) after renal function and congestive heart failure.⁸

Impact of In-Hospital Major Bleeding on Early and Late Mortality in REPLACE-2 (pooled analysis)

Retrospective analysis: Bleeding-related in-hospital adverse outcomes⁵

Both Major and Minor Bleeding Events Increase the Risk of Mortality in PCI

• In a retrospective analysis of 10,974 patients undergoing PCI from 1991-2000, Kinnaird et al reported TIMI major bleeding as an independent risk factor for in-hospital mortality (OR 3.5; 95% CI 1.9-6.7; P = .0001)⁵
• Patients who experienced major bleeding had significantly (P <.001) higher rates of death (7.5% vs. 0.6%), Q-wave MI (1.2% vs. 0.2%), non–Q-wave MI (30.7% vs 11.8%) and repeat lesion revascularization (1.9% vs 0.3%) compared to patients with no bleeding complications
• Even TIMI minor bleeding significantly increased the risk of death vs patients without bleeding (1.8% vs 0.6%,
  P <.001)
• At 1 year, mortality rates were 17.2%, 9.1%, and 5.5% for patients with major, minor, and no bleeding, respectively

• There is a significant stepwise increase in adverse outcomes by bleeding severity

Restrospective analysis of the incidence, predictors, and prognostic impact of peri procedural bleeding and tranfusion in 10,974 patients who underwent PCI between 1991 and 2000.

Treatment With Heparin Is the Most Important Risk Factor for Bleeding

• Based on data from a multivariate logistic regression analysis, treatment with heparin plus GP IIb/IIIa is the most important risk variable for major hemorrhage after accounting for the independent effects of all other covariables
• These data further support the conclusion that Angiomax^® (bivalirudin) with or without "provisional" GP IIb/IIIa has a better safety profile than heparin or heparin plus GP IIb/IIIa at similar or superior levels of anticoagulant effect
• The multivariate logistic regression model was developed to identify patient and treatment variables that were able to predict the risk of major hemorrhage in association with treatment. A comprehensive list of potential risk variables was considered, which included almost all candidate predictors of major hemorrhage cited in the literature. Among the risk variables, randomization to treatment with heparin plus GP IIb/IIIa was one of the most important independent predictors of risk of major hemorrhage (odds ratio 1.8; P<.0002).^8,11 Other independent covariables predicting risk of major hemorrhage were:
  – Creatinine clearance
  – Age >75 years
  – Female gender
  – LMWH treatment in prior 48 hours

The intensity of anticoagulation as measured by ACT levels was not found to be predictive of risk of major hemorrhage in any univariate or multivariate analysis performed. These data are consistent with published reports that have shown age, female gender, and renal function to be important predictors of risk of hemorrhage in PCI.

Consistent with REPLACE-2, treatment with heparin is one of the most important risk factors for bleeding in BAT. In a multivariate logistic regression analysis to identify patient and treatment variables that were able to predict the risk of major hemorrhage, randomization to treatment with heparin was the most important independent predictor of risk of major hemorrhage (odds ratio 2.9; 95% CI: 2.2–3.8, for risk of major hemorrhage adjusted for covariables).^12,13

Other independent covariables predicting risk of major hemorrhage (though with less certainty) were:

• Age ≥70 years (odds ratio 2.1; 95% CI: 1.6–2.7)
• Female gender (odds ratio 2.3; 95% CI: 1.8–2.9)
• Small body surface area (odds ratio 1.5; 95% CI: 1.2–2.0)
• Prior heparin within 1 hour of angioplasty (odds ratio 1.4; 95% CI: 1.1–1.9)