Search for a command to run...
Patients with multiple traumas associated with cardiopulmonary failure have a high mortality rate; however, such patients can be temporarily stabilized using extracorporeal membrane oxygenation (ECMO), providing a bridge to rescue therapy. Using a retrospective study design, we aimed to clarify the prognostic factors of post-traumatic ECMO support.From March 2006 to July 2016, 43 adult patients (mean age, 37.3 ± 15.2 years; 7 females [16.3%]) underwent ECMO because of post-traumatic cardiopulmonary failure. Pre-ECMO demographics, peri-ECMO events, and post-ECMO recoveries were compared between survivors and nonsurvivors.The most common traumatic insult was traffic collision (n = 30, 69.8%), and involved injury areas included the chest (n = 33, 76.7%), head (n = 14, 32.6%), abdomen (n = 21, 48.8%), and fractures (n = 21, 48.8%). Fifteen patients (34.9%) underwent cardiopulmonary resuscitation and 22 (51.2%) received rescue interventions before ECMO deployment. The mean time interval between trauma and ECMO was 90.6 ± 130.1 hours, and the mode of support was venovenous in 26 patients (60.5%). A total of 26 patients (60.5%) were weaned off of ECMO and 22 (51.6%) survived to discharge, with an overall mean support time of 162.9 ± 182.7 hours. A multivariate regression analysis identified 2 significant predictors for in-hospital mortality: an injury severity score (ISS) >30 (odds ratio [OR], 9.48; 95% confidence interval [CI], 1.04-18.47; P = 0.042), and the requirement of renal replacement therapy (RRT) during ECMO (OR, 8.64; 95% CI, 1.73-26.09; P = 0.020). These two factors were also significant for the 1-year survival (ISS >30: 12.5%; ISS ≤30, 48.1%, P = 0.001) (RRT required, 15.0%; RRT not required, 52.2%, P = 0.006).Using ECMO in selected traumatized patients with cardiopulmonary failure can be a salvage therapy. Prompt intervention before shock-impaired systemic organ perfusion and acute renal failure, especially in high ISS patients, is crucial for both hospital and one-year survival.