1. Safety and effectiveness of bivalirudin in routine care of patients undergoing percutaneous coronary intervention JA Rassen, MA Mittleman, RJ Glynn, A Brookhart, S Schneeweiss Dept of Pharmacoepidemiology and Pharmacoeconomics, Brigham & Women’s Hospital; Depts of Epidemiology and Biostatistics, Harvard School of Public Health; Beth Israel Deaconess Eur Heart J, Nov 25, 2009

2. 2 ●Bivalirudin has been studied as an alternative to heparin plus GP IIb/IIIa inhibitor during PCI; trials have indicated that bivalirudin is non-inferior to heparin with respect to death and repeat revascularization and may decrease the risk of major bleeds. ●To evaluate the effectiveness and safety of bivalirudin as used in routine care (transfusion, repeat PCI and death) Authors supported by Agency for Healthcare Research and Quality, National Institute of Aging, and National Institute of Mental Health Objectives Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437

3. Premier Database Background The Premier Perspective Database: ● A comprehensive repository of clinical, financial, and outcomes information that undergo routine quality and completeness checks including data verification, reconciliation, and validation ● Used by the FDA for drug surveillance and by CMS to evaluate next-generation payment models ● Over 5 million inpatient discharges and over 30 million hospital outpatient visits are recorded annually; approximately 1/6 of all US hospitalizations ● Potential to allow greater insight on comparative effectiveness issues Observational database background

4. ● A total of 127,185 PCI procedures were identified from the database between June 2003 through December 2006 ● Patient groups were defined as having received either bivalirudin plus provisional GPIIb/IIIa or the comparator, heparin plus GPIIb/IIIa ● Primary outcome: blood transfusion (whole blood, RBC, FFP, platelets, cryoprecipitate) ● Secondary outcomes: in hospital mortality and repeat PCI within the same hospital admission Methodology: Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437

5. Patient Population Patients with their first inpatient admission between June 2003 and December 2006 in which they underwent PCI (N= 326 556) Recorded charge for bivalirudin (+/- GPI) with no heparin OR Heparin (at least 1000 U) + GPI Overall Analysis Population N=127,185 Exclusions Outpatient procedures Rural hospitals, or those with < 2 PCI/day average No record of antithrombotic treatment, other antithrombotic regimen or GPI only History of hemostatic disorders Data on file. Premier Perspective™ Database. Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437 Treated with bivalirudin N=32, 541 (26%) Treated with heparin+ GPI N=94,644 (74%)

6. Methodology: statistical analysis Adjusted for: ● Patients’ socio-demographic factors: age, sex, race, income, and married or living with partner vs. living alone, year of admission, urgent admission basis. ● Patient co-morbidities: diabetes, hypertension, liver disease, COPD/asthma, cancer, smoking, old MI, old stroke, endocarditis, ischaemic heart disease, peripheral artery disease, and chronic kidney disease. ● Hospital factors: teaching vs. non-teaching status, location (midwest, northeast, south, or west of the USA, and urban/rural), hospital size (number of beds), and high volume hospitals (an average of 10 or more PCIs performed per day). Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437

7. ● Primary analysis: estimated unadjusted and adjusted hazard ratios (Cox proportional hazards models); ● Secondary analysis: examined cumulative incidence on the risk difference (least squares) and odds ratio (logistic regression) scales. ● Sensitivity analyses for urgency of PCI/elective PCI, and for heparin alone group ● Instrumental variable analysis conducted using hospital preference for treatment of PCI patients with bivalirudin Methodology: statistical analysis Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437

8. Results Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437 ●Percentage of hospitals reaching threshold of administering bivalirudin to 80%of patients receiving PCI, and hospitals administering bivalirudin to >1% of those patients

9. Results Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437 ●By quarter of calendar time, % of PCI patients exposed to bivalirudin in hospitals in the Premier Perspective Database.

10. Hazards Ratio ± 95% CI Bivalirudin Better Heparin + GP IIb/IIIa Inhibitor Better Unadjusted IV subpopulation Unadjusted Fully adjusted 0.93 (0.85–1.00) OR (95% CI) 0.77 (0.69–0.86) 0.67 (0.61–0.73) 10 Results: Transfusion* Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437 * At least one unit of any blood product Instrumental variable analysis 0.72 (0.12–4.47)

11. Hazard Ratio ± 95% CI Bivalirudin Better Heparin + GP IIb/IIIa Inhibitor Better Unadjusted IV subpopulation Unadjusted Fully adjusted 0.83 (0.78–0.89) OR (95% CI) 0.82 (0.76–0.89) 0.96 (0.90–1.03) 11 Results: Repeat PCI Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437 Instrumental variable analysis 0.83 (0.49–1.40)

12. Hazard Ratio ± 95% CI Bivalirudin Better Heparin + GP IIb/IIIa Inhibitor Better Unadjusted IV subpopulation Unadjusted Fully adjusted 0.57 (0.49–0.65) OR (95% CI) 0.56 (0.46–0.66) 0.51 (0.44–0.60) 12 Results: Death Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437 Instrumental variable analysis 0.51 (0.34–0.78) ●In-hospital mortality rate: bivalirudin 0.8%, 2.1% heparin + GPI

13. Results ● Hazard ratio point estimates and standard errors for models of mortality in the Premier and clinical trial databases Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437

14. Conclusions ● Bivalirudin is protective compared to heparin plus GPI with regard to the risk of blood transfusions and death ● Mortality benefit may even exceed trial estimates ● Because of conventional analyses’ potential for bias as a result of residual confounding, IVA-based methods, with their known limitations, may help in studying the safety and effectiveness of medications outside the constrained setting of clinical trials. Rassen JA et al Eur Heart J 2009 On line ahead of print Nov 25 doi:10.1093/eurheartj/ehp437