INTRODUCTION:The extubation period is one of the most challenging aspects for intensive care teams. Timely recognition of the return to spontaneous ventilation is essential for reducing costs, morbidity, and mortality. Several weaning predictors were studied in an attempt to evaluate the outcome of removing ventilatory support. The purpose of this study was to analyze the predictive performance of the modified integrative weaning index (IWI) in the extubation process. METH-ODS: A prospective study was performed in an ICU in a public hospital in Porto Alegre, Brazil, with 59 adult medical-surgical beds. The final population of the study comprised 153 patients receiving mechanical ventilation for over 48 h who were extubated during the period from February to November 2011. Demographic data and clinical parameters were collected in addition to extubation predictors, including static compliance of the respiratory system, ratio of breathing frequency to tidal volume, tracheal airway-occlusion pressure 0.1 s after the start of inspiratory flow, and modified IWI. RESULTS: Extubation failure was observed in 23 of the subjects (15%). Subjects with greater positive fluid balance, lower hemoglobin levels, and lower levels of bicarbonate presented a higher rate of reintubation. The 3 modified IWI values (the first and 30th minute of the spontaneous breathing trial and the difference between them), as well as the other ventilatory parameters and extubation predictors, displayed poor extubation outcome discrimination accuracy. All indexes presented small areas under the receiver operating characteristic curve, and no accurate cutoff point was identified. CONCLUSIONS: We concluded that modified IWI, similar to other extubation predictors, does not accurately predict extubation failure.