The effect of extubation failure on outcome in a multidisciplinary Australian intensive care unit.

BACKGROUND A reported association between extubation failure (EF) and increased hospital length of stay and mortality led us to assess outcome of EF in an Australian intensive care unit. DESIGN AND SETTING Non-interventional cohort study in the intensive care/high dependency unit of a tertiary referral hospital, 2000-2003. METHODS EF was defined as reintubation within 72 hours of extubation. Causes of EF were determined by review of the clinical notes and prospective record of the EF event. Patients were excluded if they were aged < or = 14 years, self-extubated, were reintubated to replace a defective endotracheal tube, or had been extubated but were returning to the operating theatre. Physiological variables used to calculate severity of illness score were analysed to ascertain correlation with EF. RESULTS 2761 patients were electively extubated, and 52 (1.8%) fulfilled the criteria for EF. Compared with those successfully extubated, EF patients had a higher 24 h APACHE II score (18.0+/-7.0 [mean+/-SD] v 15.3+/-7.4, P=0.009), significant increases in length of stay in ICU (12.8+/-8.3 v 3.0+/-6.0 days, P<0.001) and hospital (33.5 +/-40.8 v 18.0+/-28.6 days, P<0.001) and tracheostomy rate (38.5% v 3.5%, P<0.001). The commonest cause of EF was excess secretions or aspiration (32%). EF was independently associated with hospital mortality (odds ratio [OR], 2.10; 95% CI, 1.00-4.41; P=0.048) and low serum albumin level on admission (OR, 0.75; 95% CI, 0.55-1.00; P=0.05). Neither aetiology of airway failure (OR, 2.21; 95% CI, 0.56- 8.75; P=0.25) nor time to reintubation (OR, 0.99; 95% CI, 0.97-1.01; P=0.76) were associated with mortality. CONCLUSION Our findings confirm an increased risk of adverse outcomes for patients with EF. We observed a comparatively low EF rate. Confirmation in similar patient cohorts is required.