LAUSR

Real-world clinical studies provide unique insights into how randomised controlled study data are translated into routine clinical practice. Real-world data can highlight areas for improvement and generate new hypotheses. When contemporaneously collected and well characterised, like the UK national database of all-comers undergoing percutaneous coronary intervention (PCI), the data tantalise with insights unlikely to come from randomised data.1 2 The authors of the study ‘Prasugrel, Ticagrelor and Clopidogrel in primary percutaneous coronary intervention: a comparative analysis’ address a topic many considered a solved problem.2 The choice of antiplatelet agents in acute coronary syndromes (ACS), specifically acute ST elevation myocardial infarction (STEMI) treated by primary PCI, was initially aspirin and clopidogrel for all patients, until the availability of more potent antiplatelets, prasugrel and ticagrelor, created a crisis of choice. Financial limitations constrained use and the initial roll-out was reserved for STEMI, where a more potent antiplatelet may be advantageous. However, data specific to the STEMI cohort are smaller than imagined. The pivotal studies recruited a spectrum of patients with ACS with STEMI constituting 26% of patients in ‘TRial to assess Improvement in Therapeutic Outcomes by optimizing platelet InhibitioN with prasugrel Thrombolysis In Myocardial Infarction 38’ (TRITON-TIMI-38) for prasugrel, and 38% of patients in ‘PLATelet inhibition and patient Outcomes study’ (PLATO) for ticagrelor.3 4 Although subgroup data were supportive,5 6 the smaller size means statistical power is limited and definitive conclusions become harder without assessing larger data sets. Real-world data provide a solution. Another issue has been the lack of head-to-head studies between prasugrel and ticagrelor. The PRAGUE study group (‘PRAGUE-18’) showed no difference in outcomes between prasugrel and ticagrelor at 1 year.7 However, the study was blighted by high discontinuation rates of the study drug for economic reasons (patients self-paid while clopidogrel was reimbursed). Importantly, it was underpowered, requiring at least 10 times more …