P2Y12 inhibitor adherence trajectories in patients with acute coronary syndrome undergoing percutaneous coronary intervention: prognostic implications

Abstract Aims Post-acute coronary syndrome (ACS) P2Y12 inhibitor non-adherence is common and associated with greater risk of major adverse cardiovascular events (MACEs). Non-adherence can follow different trajectories from an inability to initiate, implement, or continue therapy for the intended duration. We aimed to evaluate P2Y12 inhibitor adherence trajectories among ACS patients treated with percutaneous coronary intervention (PCI), their frequency, and association with MACE. Methods and results We conducted a cohort study of adults discharged alive after PCI for ACS (2012–16) using the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease registry linked with administrative data. The primary outcome was P2Y12 inhibitor adherence trajectory in the year after PCI assessed using group-based trajectory modelling. We used logistic regression and Cox proportional-hazards regression to assess associations of trajectories with risk factors and MACE, respectively. We included 12 844 patients (mean age 62.4 years, 23.6% female). Five trajectories were identified: early consistent non-adherence (11.0%), rapid decline (7.7%), delayed initiation (6.0%), gradual decline (20.5%), and persistent adherence (54.8%). Compared with persistent adherence, rapid decline [hazard ratio (HR) 1.23, 95% confidence interval (CI) 1.01–1.49] and delayed initiation (HR 1.41, 95% CI 1.12–1.78) were associated with higher MACE in the overall cohort, whereas early consistent non-adherence was associated with higher MACE only in the subgroup receiving a drug-eluting stent (HR 2.44, 95% CI 1.60–3.71). Conclusion After PCI for ACS, patients followed one of five distinct P2Y12 inhibitor adherence trajectories. Rapid decline and delayed initiation were associated with a higher risk of MACE, whereas early consistent non-adherence was only associated with higher MACE risk in patients with a drug-eluting stent. Key questions Key findings Take-home message