Abstract
BACKGROUND: Tracheostomy is recommended for patients receiving mechanical ventilation (MV) for 14 days or more in the intensive care unit (ICU). Nevertheless, many patients undergoing prolonged MV remain intubated via the translaryngeal route. The aim of this study was to examine the influence of tracheostomy and persistent translaryngeal intubation on short-term outcomes in patients mechanically ventilated for ≥14 days. METHODS: A retrospective study was conducted using the admissions database of a 75-bed ICU from January 1, 2012, to December 31, 2012. Patients who required prolonged MV without tracheostomy at the time of initiation of a ventilator were included. The outcomes were successful weaning, and ICU and in-hospital death. Cox models were constructed to calculate the influence of tracheostomy on the outcome measures while adjusting for other potentially confounding factors. RESULTS: Of the 508 patients requiring prolonged MV, 164 were tracheostomized after a median 18 days of MV. Patients in whom translaryngeal intubation was maintained had significantly higher ICU (42.7% versus 17.1%, p <0.001) and in-hospital (54.1% versus 22.0%, p <0.001) mortality rates, and a significantly lower successful weaning rate (40.4% versus 68.9%, p <0.001). The results were consistent after matching for the propensity score of performing tracheostomy. Furthermore, a time-dependent covariate Cox model showed that a tracheostomy was independently associated with lower in-hospital mortality (adjusted hazard ratio [aHR], 0.26; 95% confidence interval [CI], 0.18-0.39) and higher successful weaning rate (aHR, 2.05; 95% CI, 1.56-2.68). CONCLUSIONS: Tracheostomy is associated with lower in-hospital mortality and higher successful weaning rates in ICU patients receiving prolonged MV. However, the cost-effectiveness and long-term outcomes of tracheostomy for this cohort require further study.