Extubation failure is associated with poor intensive care unit and hospital outcomes. Minute ventilation recovery time, an integrative measure of a patient's respiratory reserve, has been shown in a pilot study to predict extubation outcome; however, the methodology is subjective and impractical for routine use. The authors hypothesize that minute ventilation recovery time, measured using an objective and simpler method, would predict extubation outcome. A prospective cohort study was performed in adult medical and surgical intensive care unit patients intubated for >24 hours who were weaning from mechanical ventilation. Minute ventilation recovery time was measured using a new, simplified, and objective method following the final spontaneous breathing trial prior to extubation. The primary outcome was extubation failure, defined as reintubation within 7 days. The study cohort comprised 88 patients, of whom 22 (25%) failed extubation after a median of 3 days. Demographic data, weaning parameters, and the proportion of patients who passed an extubation screen were similar between groups (P > .05). Minute ventilation recovery time was significantly longer in patients who failed extubation (15 [5-15] vs 2 [1-5] minutes, P < .001), consistent in both medical and surgical subgroups. Operating characteristics for a preliminary threshold (minute ventilation recovery time >or=5 minutes) for prediction of extubation failure were sensitivity = 0.78, specificity = 0.71, positive predictive value = 0.47, negative predictive value = 0.90, correctly classified = 0.72. Adjustment for significant covariates did not alter the relationship between minute ventilation recovery time >or=5 minutes and extubation failure (odds ratio = 4.9, 95% confidence interval 1.45-16.2, P < .02). C statistic was 0.79 +/- 0.17. It was concluded that minute ventilation recovery time, measured using a feasible methodology, can predict extubation outcome in medical and surgical intensive care unit patients.