Safety and efficacy of selective, clopidogrel-based strategies in acute coronary syndrome: a study-level meta-analysis

Objectives: To investigate outcome with selective, clopidogrel-based therapies vs conventional treatment in patients undergoing percutaneous coronary intervention (PCI), especially for acute coronary syndrome. Background: Safety and efficacy of alternative, selective, clopidogrel-based therapies after PCI are not robustly established. Methods: We performed a study-level meta-analysis on six randomized trials investigating selective clopidogrel-based therapies (three on unguided de-escalation, N=3,473; three on guided clopidogrel therapy, N=7,533). Control groups received ticagrelor or prasugrel treatment. Main endpoints were major bleeding, any bleeding, major adverse cardiovascular events (MACE) and net clinical endpoint. Results: The incidence of major bleeding and MACE was similar in the selective, clopidogrel-based therapy vs conventional treatment arm (OR 0·72, 95% CI 0·51-1·01, p=0·06; OR 0·93, 0·72-1·20, p=0·58; respectively). The rates of any bleeding were lower in the selective, clopidogrel-based therapy vs conventional treatment group (OR 0·57, 0·40-0·80, p=0·001); this greater safety was significant for unguided de-escalation (OR 0·43, 0·32-0·58, p=0·00001) and non-significant for guided clopidogrel therapy (OR 0·72, 0·51-1·02, p=0·07; p for interaction 0·03). The incidence of the net clinical endpoint was fewer in the selective, clopidogrel-based therapy vs conventional treatment arm (OR 0·59, 0·41-0·85, p=0·004); this benefit was significant for unguided de-escalation (OR 0·50, 0·39-0·64, p<0·00001) and non-significant for guided clopidogrel therapy (OR 0·85, 0·62-1·16, p=0·30; p for interaction 0·01). Conclusions: As compared with prasugrel/ticagrelor treatment, alternative, selective, clopidogrel-based approaches provide a similar protection from cardiovascular events, reduce the risk of any bleeding and are associated with a greater net benefit. These beneficial effects were prevalent with unguided de-escalation to clopidogrel.