Effect of early versus late or no tracheostomy on mortality and pneumonia of critically ill patients receiving mechanical ventilation: a systematic review and meta-analysis

Abstract

Delay of tracheostomy for roughly 2 weeks after translaryngeal intubation of critically ill patients is the presently recommended practice and is supported by findings from large trials. However, these trials were suboptimally powered to detect small but clinically important effects on mortality. We aimed to assess the benefit of early versus late or no tracheostomy on mortality and pneumonia in critically ill patients who need mechanical ventilation. We systematically searched PubMed, CINAHL, Embase, Web of Science, DOAJ, the Cochrane Library, references of relevant articles, scientific conference proceedings, and grey literature up to Aug 31, 2013, to identify randomised controlled trials comparing early tracheostomy (done within 1 week after translaryngeal intubation) with late (done any time after the first week of mechanical ventilation) or no tracheostomy and reporting on mortality or incidence of pneumonia in critically ill patients under mechanical ventilation. Our primary outcomes were all-cause mortality during the stay in the intensive-care unit and incidence of ventilator-associated pneumonia. Mortality during the stay in the intensive-care unit was a composite endpoint of definite intensive-care-unit mortality, presumed intensive-care-unit mortality, and 28-day mortality. We calculated pooled odds ratios (OR), pooled risk ratios (RR), and 95% CIs with a random-effects model. All but complications analyses were done on an intention-to-treat basis. Analyses of 13 trials (2434 patients, 648 deaths) showed that all-cause mortality in the intensive-care unit was not significantly lower in patients assigned to the early versus the late or no tracheostomy group (OR 0·80, 95% CI 0·59-1·09; p=0·16). This result persisted when we considered only trials with a low risk of bias (511 deaths; OR 0·80, 95% CI 0·59-1·09; p=0·16; eight trials with 1934 patients). Incidence of ventilator-associated pneumonia was lower in mechanically ventilated patients assigned to the early versus the late or no tracheostomy group (691 cases; OR 0·60, 95% CI 0·41-0·90; p=0·01; 13 trials with 1599 patients). There was no evidence of a difference between the compared groups for 1-year mortality (788 deaths; RR 0·93, 95% CI 0·85-1·02; p=0·14; three trials with 1529 patients). The synthesised evidence suggests that early tracheostomy is not associated with lower mortality in the intensive-care unit than late or no tracheostomy. However, early, compared with late or no, tracheostomy might be associated with a lower incidence of pneumonia; a finding that could question the present practice of delaying tracheostomy beyond the first week after translaryngeal intubation in mechanically ventilated patients. Nevertheless, the scarcity of a beneficial effect on long-term mortality and the potential complications associated with tracheostomy need careful consideration; thus, further studies focusing on long-term outcomes are warranted. None.