Coronary Artery Bypass Graft Versus Percutaneous Coronary Intervention

BACKGROUND: Multiple studies have compared coronary artery bypass graft (CABG) with percutaneous coronary interventions (PCI) for coronary revascularization. There is considerable evidence that adherence to medical therapy can affect the outcomes of therapeutic interventions. However, the long-term influence of compliance with recommended medical therapy on the comparative outcomes of CABG versus PCI remains to be defined. METHODS: All non-ST-segment-elevation myocardial infarction patients undergoing coronary revascularization in an 8-hospital network were followed for up to 8 years to determine medication history and major adverse cardiac events: all-cause mortality, nonfatal myocardial infarction, and reintervention. All mortalities were checked against the Social Security Death Index. Survival curves were derived with Kaplan-Meier methods; hazard ratios were calculated with the Cox proportional hazard model; and propensity score matching was used to account for differences in patient selection. RESULTS: Among the 973 CABG and 2255 PCI patients, Kaplan-Meier major adverse cardiac event-free survival curves demonstrated a significant benefit for antiplatelet, lipid-lowering, and β-blocker therapy in both the CABG and PCI groups (P=0.001 for all 3 medications). Cox regression identified compliance with optimal medical therapy as a more powerful predictor of major adverse cardiac event-free survival than choice of therapy (hazard ratio for noncompliance=2.79; 95% confidence limits, 2.19-3.54; P<0.001; hazard ratio for PCI versus CABG=1.68, 95% confidence limits, 138-2.04; P<0.001). In propensity-matched patients, CABG outcomes were superior to PCI outcomes in patients nonadherent to optimal medical therapy (P=0.001) but were not different in patients adherent to optimal medical therapy (P=0.574). CONCLUSIONS: Regardless of coronary revascularization strategy, medication adherence has a dramatic effect on long-term outcome. Among comparable patients who adhere to optimal medical therapy, outcomes of PCI and CABG may not differ; however, among nonadherent patients, CABG affords better major adverse cardiac event-free survival. Therefore, patient compliance with medical therapy may inform clinical decision making and should be incorporated into all future comparative studies of comparative coronary revascularization strategies.