Abstract
BACKGROUND: Bleeding is the principal safety concern of antithrombotic therapy and occurs frequently among patients with coronary artery disease (CAD). AIMS: We aimed to evaluate the prognostic impact of bleeding on mortality compared with that of myocardial infarction (MI) in patients with CAD. METHODS: We searched Medline and Embase for studies that included patients with CAD and that reported both the association between the occurrence of bleeding and mortality, and between the occurrence of MI and mortality within the same population. Adjusted hazard ratios (HRs) for mortality associated with bleeding and MI were extracted and ratios of hazard ratios (rHRs) were pooled by using inverse variance weighted random effects meta-analyses. Early events included periprocedural or within 30-day events after revascularisation or acute coronary syndrome (ACS). Late events included spontaneous or beyond 30-day events after revascularisation or ACS. RESULTS: A total of 141,059 patients were included across 16 studies; 128,660 (91%) underwent percutaneous coronary intervention. Major bleeding increased the risk of mortality to the same extent as MI (rHRsbleedingvsMI 1.10, 95% CI: 0.71-1.71, p=0.668). Early bleeding was associated with a higher risk of mortality than early MI (rHRsbleedingvsMI 1.46, 95% CI: 1.13-1.89, p=0.004), although this finding was not present when only randomised trials were included. Late bleeding was prognostically comparable to late MI (rHRsbleedingvsMI 1.14, 95% CI: 0.87-1.49, p=0.358). CONCLUSIONS: Compared with MI, major and late bleeding is associated with a similar increase in mortality, whereas early bleeding might have a stronger association with mortality.