Background We sought to clarify the benefit of cytokine adsorption in patients with COVID-19 supported with venovenous extracorporeal membrane oxygenation (ECMO).Methods We did a single-centre, open-label, randomised, controlled trial to investigate cytokine adsorption in adult patients with severe COVID-19 pneumonia requiring ECMO. Patients with COVID-19 selected for ECMO at the Freiburg University Medical Center (Freiburg, Germany) were randomly assigned (1:1) to receive cytokine adsorption using the CytoSorb device or not. Randomisation was computer-generated, allocation was concealed by opaque, sequentially numbered sealed envelopes. The CytoSorb device was incorporated into the ECMO circuit before connection to the patient circuit, replaced every 24 h, and removed after 72 h. The primary endpoint was serum interleukin-6 (IL-6) concentration 72 h after initiation of ECMO analysed by intention to treat. Secondary endpoints included 30-day survival. The trial is registered with ClinicalTrials.gov (NCT04324528) and the German Clinical Trials Register (DRKS00021300) and is closed. Findings From March 29, 2020, to Dec 29, 2020, of 34 patients assessed for eligibility, 17 (50%) were treated with cytokine adsorption and 17 (50%) without. Median IL-6 decreased from 357•0 pg/mL to 98•6 pg/mL in patients randomly assigned to cytokine adsorption and from 289•0 pg/mL to 112•0 pg/mL in the control group after 72 h. One patient in each group died before 72 h. Adjusted mean log IL-6 concentrations after 72 h were 0•30 higher in the cytokine adsorption group (95% CI −0•70 to 1•30, p=0•54). Survival after 30 days was three (18%) of 17 with cytokine adsorption and 13 (76%) of 17 without cytokine adsorption (p=0•0016).Interpretation Early initiation of cytokine adsorption in patients with severe COVID-19 and venovenous ECMO did not reduce serum IL-6 and had a negative effect on survival. Cytokine adsorption should not be used during the first days of ECMO support in COVID-19.Funding None.

The novel coronavirus severe acute respiratory syndrome coronavirus 2 is infecting hundreds of thousands of humans around the globe. The coronavirus disease 2019 (COVID-19) is known to generate mild as well as critical courses. Complications on the intensive care units include acute respiratory distress syndrome, acute cardiac, and kidney injury as well as shock. Here, we present the first case report of a successful treatment of a COVID-19 patient presenting with adult respiratory distress syndrome plus refractory combined cardiogenic and vasoplegic shock, which could be successfully stabilized after implantation of a percutaneous ventricular assist device plus an extracorporeal membrane oxygenation. Although such intense treatment might not be feasible in case of a health care disaster as described for the hot spots of the COVID-19 pandemic, it might encourage treatment of younger patients on intensive care units not overcrowded by critically ill patients.

Background Prone positioning (PP) reduces mortality of patients with acute respiratory distress syndrome (ARDS). The potential benefit of prone positioning maneuvers during venovenous extracorporeal membrane oxygenation (ECMO) is unknown. The aim of this study was to evaluate the association between the use of prone positioning during extracorporeal support and ICU mortality in a pooled population of patients from previous European cohort studies. Methods We performed a pooled individual patient data analysis of European cohort studies which compared patients treated with prone positioning during ECMO (Prone group) to “conventional” ECMO management (Supine group) in patients with severe ARDS. Results 889 patients from five studies were included. Unadjusted ICU mortality was 52.8% in the Supine Group and 40.8% in the Prone group. At a Cox multiple regression analysis PP during ECMO was not significantly associated with a reduction of ICU mortality (HR 0.67 95% CI: 0.42–1.06). Propensity score matching identified 227 patients in each group. ICU mortality of the matched samples was 48.0% and 39.6% for patients in the Supine and Prone group, respectively (p = 0.072). Conclusions In a large population of ARDS patients receiving venovenous extracorporeal support, the use of prone positioning during ECMO was not significantly associated with reduced ICU mortality. The impact of this procedure will have to be definitively assessed by prospective randomized controlled trials.