1. Clopidogrel in ACS and CABG Surgery Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92 Clopidogrel use and bleeding after coronary artery bypass graft surgery John Hyung-Jun Kim, MD, MBA, a,b L. Kristin Newby, MD, MHS, b,c Robert M. Clare, MS, b Linda K. Shaw, MS, b Andrew J. Lodge, MD, d Peter K. Smith, MD, d E. Marc Jolicoeur, MD, MS, b Sunil V. Rao, MD, b,c Richard C. Becker, MD, b,c Daniel B. Mark, MD, MPH, b,c and Christopher B. Granger, MD b,c Palo Alto, CA; and Durham, NC Am Heart J. 2008 Nov;156(5):886-92

2. Changing the Calculations for Assessing Guidelines Adherence Anderson HV, Bach RG, J Am Coll Cardiol 2005;46:1488-9. “We need to invert the current equation to calculate an opportunity score for ACS patients rather than a risk score. Patients with higher baseline risks, such as the elderly, would have higher opportunity scores for benefit, even allowing for some of the greater risks from the treatment.”

3. + + Ischemic Discomfort at Rest No ST-segment Elevation Non-Q-wave MIUnstable Angina Q-wave MI ST-segment Elevation + + + + (  : positive cardiac biomarker) EmergencyDepartment In-hospital6-24hrs Presentation Spectrum of Acute Coronary Syndromes NSTEMI

4. 1990199219941996199820002002 20042007 Evolution of Guidelines for ACS 1990ACC/AHAAMI R. Gunnar 1994AHCPR/NHLBIUAE.Braunwald 19961999 RevUpd RevUpd ACC/AHA AMI T.Ryan 2000 2002 2007 2000 2002 2007 Rev Upd Rev Rev Upd Rev ACC/AHA UA/NSTEMI E. Braunwald J. Anderson 2004 2007 2004 2007 Rev Upd Rev Upd ACC/AHA STEMI E. Antman E. Antman

5. Class I Benefit >>> Risk Procedure/ Treatment SHOULD be performed/ administered Class IIa Benefit >> Risk Additional studies with focused objectives needed IT IS REASONABLE to perform procedure/ administer treatment Class IIb Benefit ≥ Risk Additional studies with broad objectives needed; Additional registry data would be helpful Procedure/Treatment MAY BE CONSIDERED Class III Risk ≥ Benefit No additional studies needed Procedure/Treatment should NOT be performed/admini- stered SINCE IT IS NOT HELPFUL AND MAY BE HARMFUL should is recommended is indicated is useful/effective/ beneficial is reasonable can be useful/effective/ beneficial is probably recommended or indicated may/might be considered may/might be reasonable usefulness/ effectiveness is unknown/unclear/ uncertain or not well established is not recommended is not indicated should not is not useful/effective/ beneficial may be harmful Applying Classification of Recommendations

6. I I IIa IIb III “The Guidelines” Classes of Recommendations Intervention is useful and effective Evidence supportive; awaiting confirming data Evidence conflicts/opinions differ; neutral statement Intervention is not useful/effective and may be harmful Intervention is useful and effective Evidence supportive; awaiting confirming data Evidence conflicts/opinions differ; neutral statement Intervention is not useful/effective and may be harmful

7. Evidence-Based Approach to ACS Weighing the Evidence ► ► Class I: Benefit > > Risk ► ► Class IIa: Benefit > Risk ► ► Class IIb: Benefit > Risk ► ► Class III: Risk > Benefit

8. “The Guidelines” Weighing the Evidence ► ► Weight of evidence grades =Data from many large, randomized trials =Data from fewer, smaller randomized trials, careful analyses of nonrandomized studies, observational registries =Expert consensus

9. Antiplatelet Drug Targets Platelet Thrombin ADP Thromboxane A 2 Epinephrine Serotonin Collagen PAR-1 PAR-4 P2Y 1 P2Y 12 TXA2-R 5HT 2 A Anionicphospholipidsurfaces GP IIb GP IIIa GP VI Platelet GP IIIa GP IIb Fibrinogen GP Ia TRA Clopidogrel Prasugrel Aspirin Gp IIb/IIIa inhibitors PAR - 1 P2Y 12

10. Acute (< 24 hrs) Antiplatelet Therapies for High-Risk NSTE ACS 43% 10% 20% 30% 40% 50% 60% 52% 34% GP IIb/IIIa Clopidogrel GP IIb/IIIa + Neither GP IIb/IIIa Clopidogrel GP IIb/IIIa + Neither Clopidogrel Clopidogrel CRUSADE Q4 2003 data 29%

11. Antiplatelet Tx: 2007 I I IIa IIb III Clopidogrel with full loading dose in ASA- allergic patients EIS: clopidogrel or IIb/IIIa administered upstream SIS: clopidogrel initiated “as soon as possible” and continued for at least one month...... and preferably for one year Clopidogrel with full loading dose in ASA- allergic patients EIS: clopidogrel or IIb/IIIa administered upstream SIS: clopidogrel initiated “as soon as possible” and continued for at least one month...... and preferably for one year

12. Antiplatelet Tx: 2007 I I IIa IIb III ICS with recurrent ischemia on ASA, clopidogrel, and anticoag: add IIb/IIIa upstream EIS: it is reasonable to give both clopidogrel and IIb/IIIa upstream EIS: can omit IIb/IIIa if bivalirudin is anticoagulant + at least 300mg clopidogrel given > 6h prior to cath ICS with recurrent ischemia on ASA, clopidogrel, and anticoag: add IIb/IIIa upstream EIS: it is reasonable to give both clopidogrel and IIb/IIIa upstream EIS: can omit IIb/IIIa if bivalirudin is anticoagulant + at least 300mg clopidogrel given > 6h prior to cath

13. New Guidance on Thienopyridines I I IIa IIb III Clopidogrel 75mg/d should be added to ASA in STEMI patients if lysed or if not reperfused If < 75y/o and lysed or if not reperfused, add oral load of 300mg clopidogrel In PPCI, give 600mg clopidogrel as soon as possible Clopidogrel 75mg/d should be added to ASA in STEMI patients if lysed or if not reperfused If < 75y/o and lysed or if not reperfused, add oral load of 300mg clopidogrel In PPCI, give 600mg clopidogrel as soon as possible Antman et al, 2007 Focused Update to 2004 ACC/AHA STEMI GLs King et al, 2008 Focused Update to 2005 ACC/AHA/SCAI PCI GLs Antman et al, 2007 Focused Update to 2004 ACC/AHA STEMI GLs King et al, 2008 Focused Update to 2005 ACC/AHA/SCAI PCI GLs

14. CLARITY-TIMI 28 Primary Endpoint: Occluded Artery (or D/MI thru Angio/HD) LD 300 mg MD 75 mg P=0.00000036 Odds Ratio 0.64 (95% CI 0.53-0.76) Clopidogrel better Placebo better Sabatine N Eng J Med 2005;352:1179. STEMI, Age 18-75 Occluded Artery or Death/MI (%) 1.00.40.60.81.21.6 36% Odds Reduction

15. Dead (%) Days Since Randomization (up to 28 days) Placebo + ASA: 1,846 deaths (8.1%) Clopidogrel + ASA: 1,728 deaths (7.5%) 0.6% ARD 7% RRR P = 0.03 N = 45,852 No Age limit ; 26% > 70 y Lytic Rx 50% Lytic Rx 50% No LD given No LD given COMMIT: Effect of Clopidogrel on Death In Hospital Chen ZM, et al. Lancet. 2005;366:1607.

16. CREDO: 15 Hrs (Not 6 Hrs) Until Clinical Benefit Seen with 300 mg Load Steinhubl S et al, J Am Coll Cardiol 2006;47:939-943 Steinhubl S et al, J Am Coll Cardiol 2006;47:939-943 Placebo Pretreatment (N=915) Death, MI, UTVR (%) Clopidogrel Pretreatment < 15 Hours (N=645) Clopidogrel Pretreatment > 15 Hours (N=202) 3.5% 7.8% 8.3% Days 0 510 15 2025 1086420

17.   I-A recommendation for upstream advanced anti- platelet therapy in high-risk ACS   Clopidogrel straightforward, well-supported ● ● What about bleeding risk and CABG surgery? ● ● New observational study   I-A recommendation for upstream advanced anti- platelet therapy in high-risk ACS   Clopidogrel straightforward, well-supported ● ● What about bleeding risk and CABG surgery? ● ● New observational study Upstream Antiplatelet Therapy: Bottom Line

18. Clopidogrel in ASC and CABG Surgery Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92 ►Background: Short-term use of clopidogrel plus aspirin among patients with acute coronary syndrome reduces ischemic events, but concerns about coronary artery bypass graft (CABG) surgery– related bleeding limit its early use. What does new data show? ►Methods: Using data from 4,794 consecutive CABG procedures in the Duke Databank for Cardiovascular Disease, investigators developed multivariable models for associations with CABG-related bleeding defined as reoperation for bleeding, red cell transfusion, and a composite of reoperation/transfusion/ hematocrit drop ≥15%. ►Study examined clopidogrel use ≤5 days versus no clopidogrel ≤5 days before CABG in each model. Models were adjusted for propensity for clopidogrel use ≤5 days.

19. Clopidogrel in ACS and CABG Surgery Clopidogrel in ACS and CABG Surgery Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92 ►Results: Among the 4,794 CABG patients: ●332 (6.9%) received clopidogrel ≤5 days before CABG ●127 (2.6%) had reoperation for bleeding ●3,277 (68.4%) received red cell transfusion ●4,387 (91.5%) had the composite outcome. ►After adjustment, clopidogrel use ≤5 days was not significantly associated with reoperation (odds ratio [OR] 1.24, 95% CI 0.63- 2.41) or the composite bleeding end point (OR 1.23, 95% CI 0.72- 2.10). ►Clopidogrel ≤5 days was modestly associated with red cell transfusion (OR 1.40, 95% CI 1.04-1.89) but more weakly than other factors, including which surgeon performed the procedure.

20. Bleeding End Point Rates by Timing of Clopidogrel Use No clopidogrel < 5 days (n=4462)Clopidogrel (n=332) Reoperation for bleeding (%) 2.63.3 Reoperation, transfusion, hematocrit drop > 1.5% 91.394.3 Packed red blood cell transfusion (%) 68.270.2 Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92

21. Multivariable Model of Reoperation for Bleeding (Clopidogrel and Propensity for Clopidogrel Forced In) Parameter OR (95% CI) Wald X 2 P Weight (per kilogram increase) 0.98 (0.97-0.99) 10.63.001 Heparin on day of surgery 1.71 (1.14-2.56) 6.68.01 Myocardial infarction 1.57 (1.09-2.26) 5.97.01 Clopidogrel <5 days before surgery 1.24 (0.63-2.41) 0.39.53 Propensity 0.07 (<0.001-8.84) 1.10.30 Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92

22. Multivariable Model of Composite Bleeding Endpoint (Clopidogrel and Propensity for Clopidogrel Forced In) Parameter OR (95% CI) Wald X 2 P Baseline hematocrit (per unit up to 37) 1.33 (1.25-1.41) 87.13<.0001 Cardiopulmonary bypass used 2.69 (1.75-4.15) 20.07<.0001 Surgeon (10 degrees of freedom) 20.590.242 Female sex 1.83 (1.35-2.47) 15.29<.0001 Creatinine clearance (per unit >67) 0.993 (0.989-0.997) 14.07.0002 Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92

23. Multivariable Model of Composite Bleeding Endpoint (continued) Parameter OR (95% CI) Wald X 2 P Angina 1.86 (1.33-2.60) 13.32.0003 Ever use of glycoprotein IIb/IIIa 2.26 (1.43-3.56) 12.22.0004 No. of diseased vessels 1.44 (1.14-1.83) 9.06.003 Clopidogrel < 5 days before surgery 1.23 (0.72-2.10) 0.57.45 Propensity 0.22 (0.007-7.09) 0.72.38 Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92

24. Multivariable Model of Packed Red Blood Cell Transfusion Parameter OR (95% CI) Wald X 2 P Surgeon (10 degrees of freedom) 95.13<.0001 Baseline hematocrit (per unit > 36) 0.92 (0.90-0.94) 54.52<.0001 Female sex 2.05 (1.70-2.49) 54.47<.0001 Creatinine clearance (per unit 30-130) 0.99 (0.985-0.997) 54.04<.0001 Cardiopulmonary bypass used 2.01 (1.54-2.63) 26.08<.0001 No. of diseased vessels 1.49 (1.28-1.74) 25.41<.0001 Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92

25. Multivariable Model of Packed Red Blood Cell Transfusion (cont.) Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92 Parameter OR (95% CI) Wald X 2 P Age (per 10 years) 1.22 (1.12-1.33) 20.64<.0001 Use of glycoprotein IIb/IIIa 1.51 (1.24-1.84) 16.99<.0001 Procedure status urgent 0.74 (0.62-0.87) 12.79.0003 NYHA class 1.16 (1.02-1.31) 5.38.02 Clopidogrel < 5 days before surgery 1.40 (1.04-1.89) 4.91.03 Propensity 0.31 (0.35-2.71) 1.13.29

26. Multivariable Linear Model of Number of Units of Packed Red Blood Cell Transfusion Parameter Estimate FP Age0.6014245.24<.0001 Cardiopulmonary bypass used 1.8974035.95<.0001 Emergent procedure 2.3353132.85<.0001 Surgeon (10 degrees of freedom) 9.58<.0001 Female sex 0.7261413.41.0003 Angina-0.8744112.18.0005 Hypercholesterolemia-0.6477011.37.0008 Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92

27. Multivariable Linear Model of Number of Units of Packed Red Blood Cell Transfusion (cont.) Parameter Estimate FP Clopidogrel < 5 days 1.1823010.70.0011 Number of diseased vessels 0.618939.30.0023 Smoker-0.551808.22.0042 Peripheral vascular disease 0.672257.12.0077 Heart failure 0.697797.02.0081 Propensity3.408321.66.198 Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92

28. Study Conclusions Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92 ►Conclusion: Clopidogrel administration ≤5 days before CABG was not significantly associated with reoperation for bleeding or a bleeding composite, and only weakly with red cell transfusion after surgery. ►Clinical Implication: The impact of withholding clopidogrel acutely in those for whom clopidogrel has proven benefits and the impact of delaying CABG to prevent bleeding among patients treated with clopidogrel should be viewed in the context of other stronger determinants of bleeding.

29. Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92 ►Lack of a clear pattern of transfusion frequency with timing of CABG since the last clopidogrel dose suggests that factors other than a biological effect of clopidogrel contribute to transfusion use. ►Overall high rates of transfusion observed in our single- center study and the comparable high rates of transfusion in the CRUSADE registry and other reported experiences suggest that rather than focus on a single drug, a concerted, prospective effort should be undertaken to understand the general drivers of transfusion and ascertain what can be done to decrease rates of blood transfusion after heart surgery. Study Conclusions

30. Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92 ►With careful analysis of baseline characteristics, concomitant medication use, type of procedure and surgeon, clopidogrel administration ≤5 days before CABG was not significantly associated with reoperation for bleeding or a composite measure of bleeding. ►Clopidogrel was more weakly associated with perioperative red cell transfusion than other factors. Study Conclusions

31. Kim JH, et al. Am Heart J. 2008 Nov;156(5):886-92 ►Impact of withholding clopidogrel acutely in ACS patients or of delaying CABG to prevent bleeding among or of delaying CABG to prevent bleeding among clopidogrel-treated patients should be viewed in the clopidogrel-treated patients should be viewed in the context of managing other stronger determinants of context of managing other stronger determinants of bleeding. bleeding. ►An aggressive effort to understand and limit high rates of transfusion use may be more important overall than transfusion use may be more important overall than continued focus on the effects of a single drug. continued focus on the effects of a single drug. Study Conclusions