Uncertainty over vascular occlusive events warrants an adequately powered RCT Joint replacement accounts for a large share of the 230 million major operations carried out each year worldwide. Each year in England and Wales alone there are about 180 000 hip and knee replacements.1 Bleeding is an important complication, and many patients require a blood transfusion. One strategy to reduce surgical bleeding is to use the antifibrinolytic drug tranexamic acid. Tranexamic acid inhibits clot breakdown by reducing the binding of plasminogen to fibrin. A recent systematic review and meta-analysis of randomised controlled trials showed that tranexamic acid reduces surgical bleeding and blood transfusion by about one third.2 However, the effect of the antifibrinolytic on the risk of vascular occlusive events remains uncertain. Perioperative myocardial infarction often goes undetected as many patients do not experience ischaemic symptoms.3 Nevertheless, an increased cardiac troponin level, a sensitive marker of myocardial injury, is common after major surgery and is associated with appreciable morbidity and mortality.4 In a linked article (doi:10.1136/bmj.g4829), the effort by Poeran and colleagues to resolve the uncertainty …