The purpose of this article is to describe the problem of predicting the lung function recovery in patients with extracorporeal membrane oxygenation for acute respiratory distress syndrome. Data from CESAR and EOLIA clinical trials on the efficacy of extracorporeal membrane oxygenation in patients with acute respiratory distress syndrome have been reviewed and some controversial results discussed. The prognostic PRESERVE and RESP scores developed as prognostic tools on the basis of the results of these studies, are presented, the limitations of their applicability in various forms of acute respiratory distress syndrome are discussed. We propose to subdivide the predictors of the extracorporeal membrane oxygenation outcome in patients with acute respiratory distress syndrome into 4 following groups: 1. Lung injury severity criteria, including parameters of their lung mechanical and functional properties. 2. Time from acute respiratory failure onset to extracorporeal membrane oxygenation initiation, which reflects the rate of pathological processes in lungs and timing of decision to initiate extracorporeal membrane oxygenation. 3. The etiology of pulmonary disorders, directly affecting the reversibility of pathological processes in the lungs. 4. The severity of the patient's general condition, including the severity of manifestations of multiple organ failure, the degree of decompensation of concomitant chronic diseases, including oncological and associated with immunosuppression. Several diseases are associated with a higher risk of specific complications, particularly hemorrhagic, during extracorporeal membrane oxygenation.