OBJECTIVE

To determine the relationship between tracheostomy tube in place after intensive-care-unit (ICU) discharge and hospital mortality.

METHODS

We conducted a prospective observational cohort study in a medical-surgical ICU in a tertiary-care hospital that does not have a step-down unit. We recorded clinical and epidemiologic variables, indication and timing of tracheostomy, time to decannulation, characteristics of respiratory secretions, need for suctioning, and Glasgow coma score at ICU discharge. We excluded patients who had do-not-resuscitate orders, tracheostomy for long-term airway control, neuromuscular disease, or neurological damage.

RESULTS

A total of 118 patients were tracheostomized in the ICU, and 73 were discharged to the ward without neurological damage. Of these, 35 had been decannulated. Ward mortality was 19% overall, 11% in decannulated patients, and 26% in patients with the tracheostomy tube in place; that difference was not statistically significant in the univariate analysis (P = .10). However, the multivariate analysis, which adjusted for lack of decannulation, age, sex, body mass index, severity of illness, diagnosis at ICU admission, duration of mechanical ventilation, Glasgow coma score, characteristics of respiratory secretions, and need for suctioning at ICU discharge, found 3 factors associated with ward mortality: lack of decannulation at ICU discharge (odds ratio 6.76, 95% confidence interval 1.21-38.46, P = .03), body mass index > 30 kg/m(2) (odds ratio 5.81, 95% confidence interval 1.24-27.24, P = .03), and tenacious sputum at ICU discharge (odds ratio 7.27, 95% confidence interval 1-55.46, P = .05).

CONCLUSIONS

In our critical-care setting, lack of decannulation of conscious tracheostomized patients before ICU discharge to the general ward was associated with higher mortality.