Association of Tracheostomy With Outcomes in Patients With COVID-19 and SARS-CoV-2 Transmission Among Health Care Professionals: A Systematic Review and Meta-analysis.

Importance: Approximately 5% to 15% of patients with COVID-19 require invasive mechanical ventilation (IMV) and, at times, tracheostomy. Details regarding the safety and use of tracheostomy in treating COVID-19 continue to evolve. Objective: To evaluate the association of tracheostomy with COVID-19 patient outcomes and the risk of SARS-CoV-2 transmission among health care professionals (HCPs). Data Sources: EMBASE (Ovid), Medline (Ovid), and Web of Science from January 1, 2020, to March 4, 2021. Study Selection: English-language studies investigating patients with COVID-19 who were receiving IMV and undergoing tracheostomy. Observational and randomized clinical trials were eligible (no randomized clinical trials were found in the search). All screening was performed by 2 reviewers (P.S. and M.L.). Overall, 156 studies underwent full-text review. Data Extraction and Synthesis: We performed data extraction in accordance with Meta-analysis of Observational Studies in Epidemiology guidelines. We used a random-effects model, and ROBINS-I was used for the risk-of-bias analysis. Main Outcomes and Measures: SARS-CoV-2 transmission between HCPs and levels of personal protective equipment, in addition to complications, time to decannulation, ventilation weaning, and intensive care unit (ICU) discharge in patients with COVID-19 who underwent tracheostomy. Results: Of the 156 studies that underwent full-text review, only 69 were included in the qualitative synthesis, and 14 of these 69 studies (20.3%) were included in the meta-analysis. A total of 4669 patients were included in the 69 studies, and the mean (range) patient age across studies was 60.7 (49.1-68.8) years (43 studies [62.3%] with 1856 patients). We found that in all studies, 1854 patients (73.8%) were men and 658 (26.2%) were women. We found that 28 studies (40.6%) investigated either surgical tracheostomy or percutaneous dilatational tracheostomy. Overall, 3 of 58 studies (5.17%) identified a small subset of HCPs who developed COVID-19 that was associated with tracheostomy. Studies did not consistently report the number of HCPs involved in tracheostomy. Among the patients, early tracheostomy was associated with faster ICU discharge (mean difference, 6.17 days; 95% CI, -11.30 to -1.30), but no change in IMV weaning (mean difference, -2.99 days; 95% CI, -8.32 to 2.33) or decannulation (mean difference, -3.12 days; 95% CI, -7.35 to 1.12). There was no association between mortality or perioperative complications and type of tracheostomy. A risk-of-bias evaluation that used ROBINS-I demonstrated notable bias in the confounder and patient selection domains because of a lack of randomization and cohort matching. There was notable heterogeneity in study reporting. Conclusions and Relevance: The findings of this systematic review and meta-analysis indicate that enhanced personal protective equipment is associated with low rates of SARS-CoV-2 transmission during tracheostomy. Early tracheostomy in patients with COVID-19 may reduce ICU stay, but this finding is limited by the observational nature of the included studies.