C-9 | Outcomes of Extracorporeal Membrane Oxygenation in COVID-19: Updated Meta-Analysis

Abstract

Due to recent nature of the COVID-19 pandemic, the goals and guidelines for extracorporeal membrane oxygenation (ECMO) in COVID-19 are similar to non-COVID-19 patients and mostly small observational studies have been conducted. Our aim is to analyze the mortality and other outcomes of ECMO in COVID-19 patients using the data available to date. We collected 2239 articles from Cochrane, Embase, and PubMed. Among these, 51 retrospective and observational studies describing outcomes in COVID-19 patients treated with ECMO were included. The primary outcome was mortality, and secondary outcomes were ECMO complications. Random effects model was used for pooled proportions with logit transformation with a 95% confidence interval (CI), and Higgins I-squared statistics were used to evaluate heterogeneity. In 51 studies including 13400 COVID-19 patients on ECMO, pooled proportion for mortality was 0.41 (95% CI 0.36-0.46, I2 89%, p<0.01) (Fig 1). While pooled proportions were 0.4 (95% CI 0.26-0.57, I2 94%, p<0.01) for bleeding, 0.12 (95% CI 0.09-0.15, I2 76%, p<0.01) for stroke, 0.25 (95% CI 0.19-0.33, I2 89%, p<0.01) for pulmonary embolism/deep venous thrombosis (PE/DVT) and 0.34 (95% CI 0.26-0.43, I2 92%, p<0.01) for renal replacement therapy (RRT) (Fig 2). Mortality of COVID-19 patients on ECMO is higher than overall mortality of COVID-19 patients as patient requiring ECMO tend to be sicker with higher co-morbidity burden and poor outcomes. ECMO complications of bleeding, stroke, PE/DVT and RRT were observed in COVID-19 patients on ECMO but due to hypercoagulable state of COVID-19, the rates of the complications are increased.