Weaning injured patients with prolonged pulmonary failure from mechanical ventilation in a non-intensive care unit setting

Background: Injured patients with pulmonary failure often require prolonged length of stay in an intensive care unit (ICU), which includes weaning from ventilatory support. In the last decade, noninvasive ventilation modes have been established as safe and effective. One method for accomplishing this mode of ventilation uses a simple bilevel ventilator. Because this ventilator has been successfully used in hospital wards, we postulated that bilevel ventilators could provide sufficient support during weaning from mechanical ventilation of injured patients in a non-ICU setting. Methods: A retrospective review of trauma patients (August 1996-January 1999) undergoing bilevel positive pressure ventilation as the final phase of weaning was conducted. Before ward transfer with bilevel ventilation, conventionally ventilated ICU patients were changed to bilevel ventilation and were required to tolerate this mode for at least 24 hours. All patients had a tracheostomy as a secure airway. Outcomes analyzed included ICU length of stay, hospital length of stay, duration of mechanical ventilation, weaning success, complications, and survival. Results: Fifty-one patients (39 men, 12 women) with a mean age of 53 received more than 24 hours of bilevel positive pressure ventilation. Mean Injury Severity Score was 29, with blunt mechanisms of injury occurring in 90%. Chest or spinal cord injuries that affected pulmonary mechanics were present in 75% of patients. Ventilator-associated pneumonia was treated in 43% of patients. Mean ICU length of stay and hospital length of stay were 21 and 34 days, respectively. Weaning was successful in 89% of patients, whereas 11% were discharged to skilled nursing facilities still receiving bilevel positive pressure ventilation. Two patients died, neither from a pulmonary nor airway complication. Of the remaining 49 patients, 12 were weaned in the ICU and 37 were transferred to the ward with bilevel ventilatory support. The average length of ward ventilation was 6.5 ± 5.4 days (n = 37). Conclusions: Implementation of a program using bilevel ventilation to support the terminal phase of weaning seriously injured patients from mechanical ventilation was successful. After initiating this mode in the ICU, it was satisfactorily continued in standard surgical wards. Because this method enabled the withdrawal of ventilatory support in a non-ICU setting, its major advantage was reducing ICU length of stay.