Clinical predictors and outcomes for patients requiring tracheostomy in the intensive care unit

Objective: To identify clinical predictors for tracheostomy among patients requiring mechanical ventilation in the intensive care unit (ICU) setting a nd to describe the outcomes of patients receiving a tracheostomy. Design: Prospective cohort study. Setting: Intensive care units of Barnes-Jewish Hospital, an urban teaching hospital. Patients: 521 patients requiring mechanical ventilation in an ICU for >12 h ours. Interventions: Prospective patient surveillance and data collection. Measurements and Main Results: The main variables studied were hospital mor tality, duration of mechanical ventilation, length of stay in the ICU and t he hospital, and acquired organ-system derangements. Fifty-one (9.8%) patie nts received a tracheostomy. The hospital mortality of patients with a trac heostomy was statistically less than the hospital mortality of patients not receiving a tracheostomy (13.7% vs. 26.4%; p = .048), despite having a sim ilar severity of illness at the time of admission to the ICU (Acute Physiol ogy and Chronic Health Evaluation [APACHE] II scores, 19.2 +/- 6.1 vs. 17.8 +/- 7.2; p = .173). Patients receiving a tracheostomy had significantly lo nger durations of mechanical ventilation (19.5 +/- 15.7 days vs. 4.1 +/- 5. 3 days; p < .001) and hospitalization (30.9 +/- 18.1 days vs. 12.8 +/- 10.1 days; p < .001) compared with patients not receiving a tracheostomy. Simil arly, the average duration of intensive care was significantly longer among the hospital nonsurvivors receiving a tracheostomy (n = 7) compared with t he hospital nonsurvivors without a tracheostomy (n = 124; 30.9 +/- 16.3 day s vs. 7.9 +/- 7.3 days; p < .001). Multiple logistic regression analysis de monstrated that the development of nosocomial pneumonia (adjusted odds rati o [AOR], 4.72; 95% confidence interval [CI], 3.24-6.87; p < .001), the admi nistration of aerosol treatments (AOR, 3.00; 95% CI, 2.18-4.13; p < .001), having a witnessed aspiration event (AOR, 3.79; 95% CI, 2.30-6.24; p = .008 ), and requiring reintubation (AOR, 2.21; 95% CI, 1.54-3.18; p = .028) were variables independently associated with patients undergoing tracheostomy a nd receiving prolonged ventilatory support. Among the 44 survivors receivin g a tracheostomy in the ICU, 38 (86.4%) were alive 30 days after hospital d ischarge and 31 (70.5%) were living at home. Conclusions: Despite having longer lengths of stay in the ICU and hospital, patients with respiratory failure who received a tracheostomy had favorabl e outcomes compared with patients who did not receive a tracheostomy. These data suggest that physicians are capable of selecting critically ill patie nts who most likely will benefit from placement of a tracheostomy. Addition ally, specific clinical variables were identified as risk factors for prolo nged ventilatory assistance and the need for tracheostomy.