26 Clinical outcomes following primary percutaneous coronary intervention: a comparison of clopidogrel, prasugrel and ticagrelor

Introduction The West Yorkshire Primary Percutaneous Coronary Intervention outcomes study was established to identify factors that are associated with clinical outcomes following primary percutaneous coronary intervention (PPCI) for ST-segment elevation myocardial infarction (STEMI). We assessed the association of procedural oral P2Y12-inhibitor with clinical outcomes in a large consecutive patient-series. Methods Demographic and procedural data for all patients undergoing PPCI between 1-1-2009 and 31-12-2011, and 1-1-2013 and 31-12-2013, in Leeds General Infirmary, were collected prospectively. Minimum 30 day follow-up data were collected for all. Patients with pre-procedural cardiogenic shock and/or cardiac arrest were excluded. Clinical endpoints were 30 day major adverse cardiovascular event (MACE) – defined as all-cause mortality, myocardial infarction (MI) and repeat coronary revascularisation, and 30 day major bleeding (HORIZONS criteria). Multivariable analyses for MACE and major bleeding comparing procedural P2Y12-inhibitors were performed with Cox proportional hazards models, adjusting for major cardiovascular risk factors. Results 4056 patients underwent PPCI during the study period, 464 of whom were excluded due to pre-procedural cardiogenic shock and/or cardiac arrest. Data for 30 day bleeding and MACE were available for 3381 of 3592 (94.2%) patients. Multivariable analysis showed no significant difference in MACE, mortality or major bleeding between patients pre-treated with clopidogrel (n=1492), prasugrel (n=1152), and ticagrelor (n=737) (Table 1). However, there was a significantly lower probability of 30 day MI with ticagrelor compared to clopidogrel (HR 0.38 (0.17–0.84)). The differences in 30 day MI between prasugrel and clopidogrel (HR 0.59 (0.33–1.04)), and prasugrel and ticagrelor (HR 1.55 (0.67–3.61)), were not statistically significant. There were no statistically significant differences in mortality between clopidogrel and ticagrelor, between prasugrel and ticagrelor, and between ticagrelor and prasugrel following multivariable analysis. Conclusion This large consecutive real-world series has shown that pre-treatment with ticagrelor was associated with lower probability of 30 day MI compared to clopidogrel, with no overall difference in bleeding, MACE or mortality. There was no significant difference in bleeding, MACE or mortality between ticagrelor and prasugrel, or between prasugrel and clopidogrel.