Abstract
Patients receiving venovenous extracorporeal membrane oxygenation (VV-ECMO) often require extended periods of ventilation. We examined the role of tracheostomy on outcomes of patients supported with VV-ECMO. We reviewed all patients at our institution who received VV-ECMO between 2013 and 2019. Patients who received a tracheostomy were compared with VV-ECMO-supported patients without tracheostomy. The primary outcome measure was survival to hospital discharge. Secondary outcome measures included length of intensive care unit (ICU) and hospital stay and adverse events related to the tracheostomy procedure. Multivariable analysis was performed to identify predictors of in-hospital mortality. We dichotomized patients receiving tracheostomy into an "early" and "late" group based on median days to tracheostomy following ECMO cannulation and separate analysis was performed. One hundred and fifty patients met inclusion criteria, 32 received a tracheostomy. Survival to discharge was comparable between the groups (53.1% vs. 57.5%, p = 0.658). Predictors of mortality on multivariable analysis included Respiratory ECMO Survival Prediction (RESP) score (odds ratio [OR] = 0.831, p = .015) and blood urea nitrogen (BUN) (OR = 1.026, p = 0.011). Tracheostomy performance was not predictive of mortality (OR = 0.837, p = 0.658). Bleeding requiring intervention occurred in 18.7% of patients following tracheostomy. Early tracheostomy (<7 days from the initiation of VV-ECMO) was associated with shorter ICU (25 vs. 36 days, p = 0.04) and hospital (33 vs. 47, p = 0.017) length of stay compared with late tracheostomy. We conclude that tracheostomy can be performed safely in patients receiving VV-ECMO. Mortality in these patients is predicted by severity of the underlying disease. Performance of tracheostomy does not impact survival. Early tracheostomy may decrease length of stay.