Ecmo for Acute Respiratory Failure Due to Covid-19: A Multicenter Retrospective Study

ABSTRACT

INTRODUCTION: Venovenous (VV) extracorporeal membrane oxygenation (ECMO) has been used successfully to treat COVID-19 patients in severe respiratory failure. The objective of our multi-center study is to evaluate mortality, time of ECMO initiation, and demographics in COVID-19 patients treated with VV-ECMO. METHOD(S) Electronic medical records from March 2020 to October 2021 were studied in 49 sites across the United States. Patients treated with ECMO who tested positive for COVID-19 were included in this retrospective data analysis (N=363, age interquartile range 37-55 years). Odds of inhospital mortality were compared using logistic regression models. At thresholds 1-7 days, patients classified as 'early ECMO' were matched to 'delayed ECMO' patients using coarsened exact matching, resulting in 7 independent analyses for early/delayed ECMO cannulation relative to the number of days of pre-ECMO mechanical ventilation (MV). RESULT(S) There were no significant differences in mortality in patients who received early or delayed ECMO. There were also no significant differences in mortality between races, body mass index (BMI), smoking status, hypertension, chronic kidney disease, coronary artery disease, steroid use, or diabetes pre-cannulation. Lastly, pre-cannulation factors associated with mortality in COVID-19 patients treated with ECMO include the use of vasopressors, which was associated with an 87% increase in mortality (p=0.017, confidence interval [CI] 1.12, 3.15);proning, which was associated with a 85% increase in mortality (p=0.015, CI 1.13, 3.06);and the use of baricitinib, which was associated with a four-fold increase in mortality (p=0.041, CI 1.11, 17.6). CONCLUSION(S) We found no evidence that particular demographic characteristics (including race, BMI, or smoking status) contribute to mortality, nor did we find evidence that the length of time on MV prior to ECMO influences mortality. The analysis of large datasets in the ECMO population may better inform clinical decision making.