Rates of nosocomial infection associated with interhospital transfer of patients receiving extracorporeal membrane oxygenation.

Abstract

Critically ill patients requiring extracorporeal membrane oxygenation (ECMO) frequently require interhospital transfer to a center that has ECMO capabilities. Patients receiving ECMO were evaluated to determine whether interhospital transfer was a risk factor for subsequent development of a nosocomial infection. Retrospective cohort study. A 425-bed academic tertiary-care hospital. All adult patients who received ECMO for >48 hours between May 2012 and May 2020. The rate of nosocomial infections for patients receiving ECMO was compared between patients who were cannulated at the ECMO center and patients who were cannulated at a hospital without ECMO capabilities and transported to the ECMO center for further care. Additionally, time to infection, organisms responsible for infection, and site of infection were compared. In total, 123 patients were included in analysis. For the primary outcome of nosocomial infection, there was no difference in number of infections per 1,000 ECMO days (25.4 vs 29.4; P = .03) by univariate analysis. By Cox proportional hazard analysis, transport was not significantly associated with increased infections (hazard ratio, 1.7; 95% confidence interval, 0.8-4.2; P = .20). In this study, we did not identify an increased risk of nosocomial infection during subsequent hospitalization. Further studies are needed to identify sources of nosocomial infection in this high-risk population.