Objective: The aim of this study was to investigate a novel standardized protocol in this neurological weaning unit in order to optimize the weaning process for patients subjected to prolonged mechanical ventilation. Of primary interest were the frequency of and reasons for deviating from the protocol as well as risk factors for weaning failure and mortality.
Patients and methods: All patients admitted to the weaning unit were enrolled in this prospective observational study. The weaning protocol consisted of 22 discrete weaning steps. An individual weaning approach was only begun if the standardized protocol failed. Variables for risk analysis included age, body mass index, APACHE II score, reason for initiating mechanical ventilation, total duration of inpatient stay before admission to the weaning unit, extent of mechanical ventilation period before admission, chronic pulmonary diseases and dialysis.
Results: Between October 2007 and December 2008 a total of 644 consecutively admitted patients were enrolled in the study. The mean age was 67.6 ± 12.3 years, the mean inpatient stay before admission was 45.8 ± 34.9 days and the mean duration of mechanical ventilation before admission to the unit was 38.1 ± 28.7 days. The mean APACHE II score was 19.0 ± 7.2, 68 % of the patients were male, 98.6 % had a tracheotomy tube, 30.9 % had a history of chronic pulmonary disease and 13.7 % required dialysis. Reasons for initiating ventilation were cerebral 33.1 %, pulmonary 28.7 %, cardiovascular 31.5 %, neuropathic 2.8 %, myopathic 0.9 %, spinal cord injuries 1.9 % and 1.1 % were unclear. Weaning was successful in 77.3 % (498 patients) of all cases with a mean duration of 22.0 ± 33.9 days. Of those successfully weaned, a total of 85.9 % (n = 428) were weaned according to the standard protocol. The weaning process was also shorter (20.8 ± 35.6 versus 29.0 ± 19.9 days) for those patients weaned according to the protocol compared to those patients where the protocol failed. Protocol failure was normally due to complications during the weaning process. Among the patients where the protocol failed, chronic pulmonary disease (41.4 versus 28.3 %; p < 0.02), a longer duration of ventilation (42.3 ± 22.8 versus 35.9 ± 25.3 days; p < 0.01) and a longer in-hospital stay (52.7 ± 41.4 versus 42.4 ± 30.1 days; p < 0.01) prior to admission were significantly more common. A total of 23.0 % (n = 148) of the patients died and 9.8 % (n = 63) of the patients were discharged into a home care ventilation program. Chronic pulmonary disease and the duration of inpatient stay prior to admission were predictors of weaning failure. The APACHE II score, age and acute renal failure with concomitant need for dialysis were the factors best predicting mortality.
Conclusions: The majority of patients receiving prolonged mechanical ventilation can be successfully weaned using a standardized protocol. Failures of standardized weaning per protocol occurred most often in patients with chronic pulmonary disease and following longer inpatient stay. These patients also had a higher risk of final weaning failure.