ECMO in trauma: What are the outcomes?
We read with interest the case series and systematic review of extracorporeal membrane oxygenation (ECMO) for ARDS in trauma by Robba et al.1 They report that of 31 trauma patients who required ECMO (27 from systematic literature review and 4 in their institutional case series) only 4 died, with a survival rate of 87.1%. This survival rate was noted to be high compared to other published institutional, national and international ECMO reports.
The use of ECMO in the trauma population has been reported to have a survival benefit in patients with severe refractory hypoxic respiratory failure. Critical appraisal of available evidence is important in assessing high-cost critical care therapies. But this current review includes only selected case reports and series based on a desire to focus on anticoagulation management in this population. We believe this introduced a significant positive outcome bias in the results since most case reports are used by authors to report interventions that result in favorable outcomes. Consequently, the authors present a serious misrepresentation of the ECMO survival rate which is a significant limitation of this review.
We also note that the limited focus of this review led the authors to exclude some important retrospective cohort studies on post-traumatic ARDS requiring ECMO. Including the following studies would have significantly increased the sample size for this analysis.
Guirand et al reported a two-institution series of 26 trauma patients who required VV-ECMO for ARDS following severe traumatic injuries. Of the 26 ECMO patients, 15 survived to hospital discharge (unadjusted ECMO survival rate 57.7%) vs. conventional management (42 of 76, unadjusted survival rate 55%). Using propensity score matching for age and PaO2/FiO2 ratio with 17 ECMO and 17 conventional patients, survival was significantly greater in ECMO patients (Kaplan Meier survival 64.7% vs. 23.5%, p = 0.01; multivariable logistic regression AOR 0.038, 95% CI 0.004-0.407, p = 0.007).2
Bosarge et al. compared early initiation of ECMO (n = 15) to historical controls (n = 14) for trauma patients with severe ARDS. These authors found a lower mortality in the ECMO cohort (13.3% vs. 64%, p = 0.01), but propensity matching and regression analyses were not performed.3
Ried et al published a retrospective observational analysis of prospectively collected data from the Regensburg ECMO Registry database for all trauma patients with acute respiratory failure requiring extracorporeal lung support during a 10-year interval. This report included 52 patients, 26 supported with VV ECMO and 26 with pumpless extracorporeal lung assist (PECLA). Survival was higher with VV-ECMO (81%) compared to PECLA (77%).4
Fang et al reported the military experience of ECMO in trauma by the Landstuhl Acute Lung Rescue Team. This included 2 patients with severe bilateral pulmonary contusions after motor vehicle collisions and severe hypoxemia and hypercarbia who required VV-ECMO support (initiated at Landstuhl Regional Medical Center and transported to the University of Regensburg for subsequent care) and both patients recovered from their lung injuries.5
There are additional published ECMO case series in trauma that have not been included in this current review, with varying survival rates reported between 35% and 90%. Given these large differences in outcomes, it is imperative to use multi-institutional registry data to avoid selection and positive-outcome bias and to determine actual survival rates associated with ECMO in trauma.
The July 2016 International Summary ECLS Registry Report from the Extracorporeal Life Support Organization (ELSO, https://www.elso.org/) provides an important multi-national point of reference for ARDS patients managed with ECMO. Of the total Adult Respiratory failure patients (n = 10,601), 58% survived to discharge or transfer (n = 6121).