Anticoagulation practices and the prevalence of major bleeding, thromboembolic events, and mortality in venoarterial extracorporeal membrane oxygenation: A systematic review and meta-analysis

Purpose: The purpose was to evaluate the safety of anticoagulation in venoarterial extracorporeal membrane oxygenation (VA-ECMO).

Design: We performed a systematic review and meta-analysis using multiple electronic databases. Studies were from 1977 to September 27, 2016. We evaluated the effect of anticoagulation in VA-ECMO on outcomes including major bleeding, thromboembolic events, and in-hospital mortality using a random effects model meta-analysis.

Results: Twenty-six studies (1496 patients) were included. Ten studies only had patients with postcardiotomy shock, 4 studies only included extracorporeal cardiopulmonary resuscitation patients, and 10 studies had a mixture of patients. Most studies (n=17) were low quality with a Newcastle-Ottawa Scale score ≤5. The summary prevalence of major bleeding was 27% (95% confidence interval [CI], 18%-35%), with considerable between-study heterogeneity (I²=91%). Major bleeding requiring reoperation was the most common bleeding event. The summary prevalence of thromboembolic events was 8% (95% CI, 4%-13%; I²=83%). Limb ischemia, circuit-related clotting, and stroke were the most commonly reported events. The summary prevalence for in-hospital mortality was 59% (95% CI, 52%-67%; I²=78%).

Conclusions: The optimal targets and strategies for anticoagulation in VA-ECMO are unclear. Evaluation of major bleeding and thromboembolic events is limited by study quality and between-study heterogeneity. Clinical trials are needed to investigate the optimal anticoagulation strategy.