Tracheostomy is associated with decreased in‐hospital mortality during severe COVID‐19 infection

Abstract Objective Tracheostomy is often performed in patients with a prolonged course of endotracheal intubation. This study sought to examine the clinical utility of tracheostomy during severe Coronavirus disease 2019 (COVID‐19) infection. Study Design A retrospective single‐system, multicenter observational cohort study was performed on patients intubated for COVID‐19 infection. Patients who received intubation alone were compared with patients who received intubation and subsequent tracheostomy. Patient demographics, comorbidities, and hospital courses were analyzed. Setting The University of Pennsylvania Health System from 2020 to 2021. Methods Logistic regression analysis was performed on patient demographics and comorbidities. Kaplan–Meier survival curves were generated depending on whether patients received a tracheostomy. Results Of 777 intubated patients, 452 were male (58.2%) and 325 were female (41.8%) with a median age of 63 (interquartile range [IQR]: 54–73) years. One‐hundred and eighty‐five (23.8%) patients underwent tracheostomy. The mean time from intubation to tracheostomy was (17.3 ± 9.7) days. Patients who underwent tracheostomy were less likely to expire during their hospitalization than those who did not undergo tracheostomy (odds ratio [OR] = 0.31, P < 0.001), and patient age was positively associated with mortality (OR = 1.04 per year, P < 0.001). Likelihood of receiving tracheostomy was positively associated with being on extra‐corporeal membranous oxygenation (ECMO) (OR = 101.10, P < 0.001), immunocompromised status (OR = 3.61, P = 0.002), and current tobacco smoking (OR = 4.81, P = 0.041). Tracheostomy was also associated with a significantly longer hospital length of stay ([57.5 ± 32.2] days vs. [19.9 ± 18.1] days, P < 0.001). Conclusions Tracheostomy was associated with reduced in‐hospital mortality, despite also being associated with increased comorbidities. Tracheostomy should not be held back from patients with comorbidities for this reason alone and may even improve survival in high‐risk patients.