Tracheostomy outcomes in critically ill patients with COVID-19: a systematic review, meta-analysis, and meta-regression.

ABSTRACT

BACKGROUND: We performed a systematic review of mechanically ventilated patients with COVID-19, which analysed the effect of tracheostomy timing and technique (surgical vs percutaneous) on mortality. Secondary outcomes included intensive care unit (ICU) and hospital length of stay (LOS), decannulation from tracheostomy, duration of mechanical ventilation, and complications. METHODS: Four databases were screened between January 1, 2020 and January 10, 2022 (PubMed, Embase, Scopus, and Cochrane). Papers were selected according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and the Population or Problem, Intervention or exposure, Comparison, and Outcome (PICO) guidelines. Meta-analysis and meta-regression for main outcomes were performed. RESULTS: The search yielded 9024 potentially relevant studies, of which 47 (n=5268 patients) were included. High levels of between-study heterogeneity were observed across study outcomes. The pooled mean tracheostomy timing was 16.5 days (95% confidence interval [CI] 14.7-18.4; I2=99.6%). Pooled mortality was 22.1% (95% CI 18.7-25.5; I2=89.0%). Meta-regression did not show significant associations between mortality and tracheostomy timing, mechanical ventilation duration, time to decannulation, and tracheostomy technique. Pooled mean estimates for ICU and hospital LOS were 29.6 (95% CI 24.0-35.2; I2=98.6%) and 38.8 (95% CI 32.1-45.6; I2=95.7%) days, both associated with mechanical ventilation duration (coefficient 0.8 [95% CI 0.2-1.4], P=0.02 and 0.9 [95% CI 0.4-1.4], P=0.01, respectively) but not tracheostomy timing. Data were insufficient to assess tracheostomy technique on LOS. Duration of mechanical ventilation was 23.4 days (95% CI 19.2-27.7; I2=99.3%), not associated with tracheostomy timing. Data were insufficient to assess the effect of tracheostomy technique on mechanical ventilation duration. Time to decannulation was 23.8 days (95% CI 19.7-27.8; I2=98.7%), not influenced by tracheostomy timing or technique. The most common complications were stoma infection, ulcers or necrosis, and bleeding. CONCLUSIONS: In patients with COVID-19 requiring tracheostomy, the timing and technique of tracheostomy did not clearly impact on patient outcomes. SYSTEMATIC REVIEW PROTOCOL PROSPERO CRD42021272220.