BACKGROUND: Adaptive support ventilation (ASV) facilitates ventilator liberation in postoperative patients in surgical intensive care units (ICU).Whether ASV has similar benefits in patients with acute respiratory failure is unclear. METHODS: We conducted a pilot study in a medical ICU that manages approximately 600 mechanically ventilated patients a year. The ICU has one respiratory therapist who manages ventilators twice during the day shift (8:00 AM to 5:00 PM). No on-site respiratory therapist was present at night. We prospectively enrolled 79 patients mechanically ventilated for > 24 hours on pressure support of > 15 cm H 2 O, with or without synchronized intermittent mandatory ventilation, F IO 2 < 50%, and PEEP < 8 cm H 2 O. We switched the ventilation mode to ASV starting at a "%MinVol" setting of 80 -100%. We defined spontaneous breathing trial (SBT) readiness as a frequency/tidal-volume ratio of < 105 (breaths/min)/L on pressure support of < 8 cm H 2 O and PEEP of < 5 cm H 2 O for at least 2 h, and all spontaneous breaths. The T-piece SBT was considered successful if the frequency/tidal-volume ratio remained below 105 (breaths/min)/L for 30 min, and we extubated after successful SBT. The control group consisted of 70 patients managed with conventional ventilation modes and a ventilator protocol during a 6-month period immediately before the ASV study period. RESULTS: Extubation was attempted in 73% of the patients in the ASV group, and 80% of the patients in the non-ASV group. The re-intubation rates in the ASV and non-ASV groups were 5% and 7%, respectively. In the ASV group, 20% of the patients achieved extubation readiness within 1 day, compared to 4% in the non-ASV group (P ؍ <.001). The median time from the enrollment to extubation readiness was 1 day for the ASV group and 3 days for the non-ASV group (P ؍ .055). Patients switched to ASV were more likely to be liberated from mechanical ventilation at 3 weeks (P ؍ .04). Multiple logistic regression analysis showed that, of the independent factors in the model, only ASV was associated with shorter time to extubation readiness (P ؍ .048 via likelihood ratio test). CONCLUSIONS: Extubation readiness may not be recognized in a timely manner in at least 15% of patients recovering from respiratory failure. ASV helps to identify these patients and may improve their weaning outcomes.