Time from mechanical ventilation initiation to venovenous extracorporeal membrane oxygenation in COVID-19: A prospective, multicentre, observational study

BackgroundThe impact of the duration of invasive mechanical ventilation (IMV) before venovenous extracorporeal membrane oxygenation (VV ECMO) on patient outcomes in COVID-19 remains unclear.Methods and settingData from the COVID-19 Critical were used to investigate whether the duration of IMV prior to VV ECMO initiation was associated with ICU mortality between January 1st 2020 and December 31st, 2022. Multivariable Cox regression models were used to evaluate the role of the duration of IMV before ECMO on patient outcomes after adjusting for calendar date and key patient covariates.Results919 adult patients with median age 50 and severe COVID-19 infection requiring IMV and VV ECMO were included in the analysis. The ICU mortality for ECMO performed on the same day as IMV was 43%, compared to 47% when ECMO was initiated between 1 and 7 days and 54% when ECMO was initiated at 8+ days. The 8+ day group of IMV received more prone positioning (71%), neuromuscular blockade (80%), and vasopressor support (74%). The multivariable analysis showed the mortality risk increased when ECMO was initiated within the first 7 days from commencement of IMV (Hazard ratio, HR: 1.37; 95% CI: 1.08 to 1.73) or later (HR: 1.51; 1.02 to 2.22), compared with same-day initiation.ConclusionAnalysis indicated a positive association between time on IMV before VV ECMO initiation and ICU mortality, but effects sizes exhibited high uncertainty after adjusting for other patient characteristics. Further studies are needed to confirm our findings.