EMPROVING outcomes: Evaluating the effect of an ultralung protective strategy for patients with ARDS treated with ECMO.

Abstract

Since the initiation of an adult extracorporeal membrane oxygenation (ECMO) program at our institution, the program has managed well over 200 patients with ECMO in a 3-year time frame. While there is a plethora of research evaluating ECMO for acute respiratory distress syndrome (ARDS), few studies have evaluated the impact that ventilator management after cannulation might have on outcomes. We hypothesized that failure to properly protect the lungs after cannulation would lead to higher mortality. This was a retrospective observational study performed from 1 January 2014 to 8 July 2018. A total of 196 patients were treated with ECMO, 57 of whom were diagnosed with ARDS and treated with venovenous ECMO. The univariable analysis revealed a statistically higher total serum bilirubin and lower total days on ECMO in those who died vs those who lived. During ECMO, higher mean peak inspiratory pressures (PIP) and higher FiO2 were found in those who died vs those who lived. In multivariable analysis, increasing age (odds ratio [OR] = 1.2; confidence interval [CI] = 1.04-1.39, P = .02), increasing mean PIP, and increasing mean FiO2 concentration during ECMO (PIP: OR = 1.40, CI = 1.03-1.89, P = .03; FiO2 : OR = 1.16, CI = 1.02-1.32, P = .02) were all associated with increased mortality. Failing to protect the lungs with a lung protective strategy such as the EMPROVE protocol after ECMO cannulation was associated with mortality. For every 1 mm Hg increase in the mean PIP, the odds of dying increased 1.4 times, and for every 1% increase in the mean FiO2 , the odds of dying increased 1.16 times. For lung rest to truly be effective, the lungs must be relieved of the burden of gas exchange.