An investigation into the contributing factors to survival of ARDS patients supported by veno-venous ECMO

Introduction This study aimed to identify characteristics associated with survival during and post Extra Corporeal Membrane Oxygenation (ECMO) therapy, in patients with acute respiratory distress syndrome (ARDS) during the COVID-19 pandemic. Methods A retrospective observational study on 94 consecutive patients with confirmed COVID-19 induced ARDS supported by ECMO was carried out 49/94 (52.7%) patients survived to hospital discharge. Results Non-survivors were found to have significantly ( p < .05) higher: Pre-ECMO International normalized ratios (INR), carbon dioxide partial pressure ( pCO 2 ), Acute Kidney Injury (AKI) scores and blood urea levels. Also, lower pre-ECMO peak inspiratory pressures (PIP), mean arterial pressure, saturation of arterial oxygen (SaO 2 ), blood bicarbonate levels (HCO3), blood Ph and fewer trials off ECMO with shorter combined trial off times. Patients that did not survive were more likely to have renal impairment and have received peri-ECMO haemofiltration. Poor prognosis was significantly associated with: receiving pre-ECMO nitric oxide (HR = 3.047, CI = 1.247-7.447, p = .015), renal impairment (HR = 3.023, CI = 1.586-5.763, p < .001), AKI of 2 (HR = 3.611, CI = 1.382-9.441, p = .009) or 3 (HR = 3.275, CI = 1.235-8.685, p = .017), peri-ECMO haemofiltration (HR = 2.412, CI = 1.310-4.442, p = .005) and the ABO blood group B (HR = 3.103, CI = 1.335-7.212, p = .008). pre-ECMO high CO 2 (HR = 1.134, CI = 1.031-1.248, p = .010), blood lactate (HR = 1.350, CI = 1.156-1.576, p < .001), INR (HR = 2.571, CI = 1.438-4.598, p=<0.001) and lower blood Ph (HR = 0.023, CI = 0.002-0.210, p < .001). Conclusions Commonly used mortality scores may not be of use in a COVID-19 cohort of ECMO patients. The initiation of ECMO needs to be implemented prior to metabolic derangements, renal and fulminant respiratory failure.