Extracorporeal Membrane Oxygenation Support in Trauma Versus Nontrauma Patients with Noninfectious Acute Respiratory...

The utility of extracorporeal membrane oxygenation (ECMO) in patients with acute respiratory distress syndrome (ARDS) of noninfectious origin remains unclear. Data on patients with ARDS of noninfectious origin who underwent ECMO were reviewed retrospectively. We compared the pre-ECMO characteristics and hospital outcomes of patients with traumatic and nontraumatic ARDS. In total, 23 patients (trauma, n = 9; nontrauma, n = 14) were included in the study. The mean patient age was 42 years, there were three females, and the mean pre-ECMO Simplified Acute Physiologic Score (SAPS) II was 60.0 (49.0-71.0). The hemoglobin level was lower and the prothrombin time (PT) more prolonged, prior to initiation of ECMO, in traumatic compared with nontraumatic ARDS patients. During the first 48 h of ECMO support, the coagulation parameters did not differ between the two groups, but the platelet counts, PT, and activated partial thromboplastin time indicated that coagulopathy was developing in all patients. The hospital and 28-day mortality rates were 21.7 and 13.0%, respectively, and serious neurological outcomes (cerebral performance category [CPC] of three points or more) developed in 26.1% of all patients; however, the extent of such outcomes did not differ between traumatic and nontraumatic ARDS patients. Upon multivariate analysis, the pre-ECMO SAPS II tended to be associated with composite events (i.e., hospital death and/or a CPC of three points or more) (P = 0.051). Additionally, a history of hypertension and an elevated pre-ECMO SAPS II were significant risk factors for serious neurological outcomes among hospital survivors (n = 18). In conclusion, ECMO support can be associated with favorable outcomes in patients with ARDS of noninfectious origin, irrespective of whether the ARDS is associated with trauma. The pre-ECMO SAPS II and a history of hypertension may be independent risk factors for poor outcomes.