Myocardial Infarction Risk after Discontinuation of Thienopyridine Therapy in the Randomized DAPT Study (Dual Antiplatelet Therapy)

Abstract

Background: Thienopyridine plus aspirin beyond 1 year after coronary stenting reduces myocardial infarction (MI) risk and increases bleeding risk in comparison with aspirin alone. The hazard associated with late thienopyridine discontinuation and risk factors for MI after discontinuation are poorly defined. Methods: In the DAPT Study (Dual Antiplatelet Therapy), after percutaneous coronary intervention and 12 months of thienopyridine (clopidogrel or prasugrel) plus aspirin, eligible patients remained on aspirin and were randomly assigned to continued thienopyridine versus placebo for 18 months. At 30 months, patients stopped the study drug and were observed for 3 months. Cumulative incidence of MI was assessed over 3 months after randomization (months 12-15) and 3 months after study drug discontinuation (months 30-33). The MI hazard for each of these periods was assessed across randomized treatment arms and by DAPT score values <2 or ≥2. Results: Among the 11 648 randomly assigned patients, the monthly cumulative incidence of MI was lower with continued thienopyridine versus placebo at 12 to 15 months (0.12% versus 0.37%, P<0.001, in all patients; 0.13% versus 0.27%, P=0.02, in patients not treated with paclitaxel-eluting stents), and higher at 30 to 33 months (0.30% versus 0.15%, P=0.013, in all patients; in patients without paclitaxel-eluting stents, 0.18% versus 0.17%, P=0.91). The majority of MIs in both time periods (74% and 76%) were not related to stent thrombosis. After multivariable adjustment, treatment arm independently predicted MI at months 12 to 15 (P<0.001) and 30 to 33 (P=0.011). During months 12 to 15, patients with DAPT scores <2 or ≥2 both had lower rates of MI with continued thienopyridine (MI monthly incidence 0.16% versus 0.51%, P<0.001, for scores ≥2; 0.08% versus 0.24%, P=0.012, for scores<2, interaction P=0.064). Conclusions: Discontinuing thienopyridine after either 12 or 30 months is associated with an early increase in MI risk, mainly unrelated to stent thrombosis; the magnitude of risk is highest in the earlier time frame, and lower in patients not treated with paclitaxel-eluting stents. Although higher DAPT scores identify patients with greater absolute ischemic benefit (relative to bleeding harm) with continued thienopyridine therapy, discontinuation at 12 months increases MI hazard regardless of DAPT score group.