Duration of extracorporeal membrane oxygenation is a poor predictor of hospital survival

Abstract

Purpose: Venovenous extracorporeal membrane oxygenation (ECMO) is increasingly used in patients with respiratory failure. In patients without the option of lung transplantation, prognostication is challenging. We hypothesized that duration of ECMO therapy is inversely correlated with the chance of recovery and therefore hospital survival.

Materials and methods: A single-center retrospective register analysis was performed. All bridge-to-recovery venovenous ECMO patients without option for lung transplantation treated between October 2010 and September 2015 were included.

Findings: A total of 175 patients (mean age, 51.61 ± 2.11 years) were detected. Medium time on ECMO was 9.26 ± 1.91 days. Time on ECMO was not significantly shorter in survivors compared to nonsurvivors (8.23 ± 2.04 and 10.15 ± 3.07, respectively; P = .327). Rate of hospital survival and time on ECMO did not correlate (P = .103). The predictive value of ECMO duration on hospital survival was 0.503 in a receiver operating characteristic analysis.

Conclusions: According to our registry data, duration of ECMO therapy by itself could not predict hospital survival. Prospective studies are needed to confirm this finding.

Keywords: ECMO; Extracorporeal membrane oxygenation; Survival.