Effect of Early Versus Late Tracheostomy or Prolonged Intubation in Critically Ill Patients with Acute Brain Injury: A Systematic Review and Meta-Analysis

Background The optimal timing of tracheostomy placement in acutely brain-injured patients, who generally require endotracheal intubation for airway protection rather than respiratory failure, remains uncertain. We systematically reviewed trials comparing early tracheostomy to late tracheostomy or prolonged intubation in these patients. Methods We searched 5 databases (from inception to April 2015) to identify randomized controlled trials comparing early tracheostomy (≤10 days of intubation) with late tracheostomy (>10 days) or prolonged intubation in acutely brain-injured patients. We contacted the principal authors of included trials to obtain subgroup data. Two reviewers extracted data and assessed risk of bias. Outcomes included long-term mortality (primary), short-term mortality, duration of mechanical ventilation, complications, and liberation from ventilation without a tracheostomy. Meta-analyses used random-effects models. Results Ten trials (503 patients) met selection criteria; overall study quality was moderate to good. Early tracheostomy reduced long-term mortality (risk ratio [RR] 0.57. 95 % confidence interval (CI), 0.36–0.90; p = 0.02; n = 135), although in a sensitivity analysis excluding one trial, with an unclear risk of bias, the significant finding was attenuated (RR 0.61, 95 % CI, 0.32–1.16; p = 0.13; n = 95). Early tracheostomy reduced duration of mechanical ventilation (mean difference [MD] −2.72 days, 95 % CI, −1.29 to −4.15; p = 0.0002; n = 412) and ICU length of stay (MD −2.55 days, 95 % CI, −0.50 to −4.59; p = 0.01; n = 326). However, early tracheostomy did not reduce short-term mortality (RR 1.25; 95 % CI, 0.68–2.30; p = 0.47 n = 301) and increased the probability of ever receiving a tracheostomy (RR 1.58, 95 % CI, 1.24–2.02; 0