Extracorporeal membrane oxygenation (ECMO) poses unique thrombotic and hemorrhagic risks, and the optimal anticoagulant choice is unknown. We systematically searched Ovid EBM Reviews, Ovid Embase, Ovid Medline, Scopus, and Web of Science Core Collection for randomized-, crossover-, retrospective cohort-, or parallel-designed clinical studies of adult patients receiving ECMO that compared heparin recipients with bivalirudin recipients. Meta-analysis was performed with random-effects models. The ROBINS-I tool was used to assess the risk of bias. Six retrospective observational studies met the inclusion criteria for the qualitative summary. Five studies were suitable for meta-analysis. Those who received heparin were more likely to experience circuit-related thrombosis (odds ratio [OR] 2.05, 95% confidence interval [CI] 1.25–3.37, p = 0.005, I² = 0%) and die (OR 1.62, 95% CI 1.19–2.21, p = 0.002, I² = 0%) compared with those who received bivalirudin. There were no differences in major bleeding events between heparin and bivalirudin recipients (OR 1.83, 95% CI 0.55–6.09, p = 0.33, I² = 82.7%). In retrospective settings compared with heparin anticoagulation, bivalirudin was associated with less circuit-related thrombotic events and greater survival in adults supported on ECMO, without contributing to more bleeding complications. Prospective controlled studies comparing heparin and bivalirudin in adult ECMO patients are warranted to corroborate these findings.