Impact of delirium on weaning from mechanical ventilation in medical patients

Background and objective: Delirium is an important predictor of negative clinical outcomes in intensive care unit (ICU), including prolonged mechanical ventilation (MV). However, delirium has not yet proven to be directly linked to weaning difficulties. The objective of this cohort study was to evaluate the association between delirium, as observed on the day of the weaning trial, and subsequent weaning outcomes in medical patients. Methods: This is a retrospective analysis with prospectively collected data on weaning from mechanical ventilation (MV) and delirium, as assessed by bedside ICU nurses using the Confusion Assessment Method for the ICU (CAM-ICU) between October 2011 and September 2013. Results: During the study period, a total of 393 patients with MV support underwent a spontaneous breathing trial (SBT) according to the standardized protocol. Of these patients, 160 (40.7%) were diagnosed with delirium on the day of the first SBT. Patients without delirium were more successfully extubated than those with delirium (81.5% vs 69.4%, P = 0.005). Delirium was found to be associated with final weaning outcomes, including difficult (OR 1.962, 95% CI 1.201-3.205) and prolonged weaning (OR 2.318, 95% CI 1.272-4.226) when simple weaning was used as a reference category. After adjusting for potential confounding factors, delirium was still significantly associated with difficult weaning (adjusted OR 2.073, 95% CI 1.124-3.822), but not with prolonged weaning (adjusted OR 2.001, 95% CI 0.875-4.575). Conclusion: Delirium, as assessed by the CAM-ICU at the time of first weaning trial, was significantly associated with weaning difficulties in medical patients.