Predictors of Mortality in Acute Respiratory Failure Patients Treated with Extracorporeal CO2 Removal

ABSTRACT

Extracorporeal CO2 removal (ECCO2R) is an effective therapy for correcting hypercapnia and respiratory acidosis. However, appropriate patient selection for optimal ECCO2R outcomes is not well defined. The goal of this retrospective, multicenter study was to determine patient and ECCO2R therapy characteristics which predict mortality in acute respiratory failure patients treated with the Hemolung ECCO2R system (ALung Technologies). The Hemolung Registry was queried for patients with acute respiratory failure. The Registry contains patient demographics, baseline and on- ECCO2R physiologic parameters, ICU survival and Hemolung performance data. Predictors of ICU survival were identified using a multivariable logistical regression analysis. 159 Hemolung patients were included in the analysis. Patients primarily had COVID-19 (55%) or non-COVID-19 ARDS (36%). Survival to ICU discharge was 41%. The median age of survivors was lower (49 vs 58 years), the use of adjunct therapies was lower in survivors (35.4% vs 64.9%), and a greater proportion of survivors received the recommended level of anticoagulation (43.1% vs 23.4%). ECCO2R complications were not significantly different between ICU survivors and non-survivors. COVID-19 diagnosis and a P/F < 100 at the start of Hemolung therapy were each independently associated with ICU mortality. This is the first study specifically evaluating patient and ECCO2R therapy characteristics that independently predict mortality in patients presenting with acute respiratory failure. The results of this study can provide insight in patient selection for future clinical trials and real-world use. Due to the retrospective nature of this study survival to hospital discharge data was not available.