Spontaneous breathing trials as predictors of extubation outcomes in neurocritical care: insights from the ENIO study.

Abstract

In critically ill patients, extubation readiness is typically assessed using a spontaneous breathing trial (SBT). Among patients with acute brain injury (ABI), the optimal SBT method remains uncertain. We conducted a post-hoc analysis of the ENIO study (NCT03400904), including mechanically ventilated ABI patients with available SBT data, undergoing extubation attempt. SBTs were classified as T-piece, pressure support ventilation (PSV), or continuous positive airway pressure (CPAP). The primary outcome was extubation failure within 5 days. Associations between SBT modality and extubation failure were assessed using multivariable logistic regression and inverse probability of treatment weighting. Of 1,512 patients enrolled in ENIO, 839 met the inclusion criteria, of whom 270 (32.2%) were female and 396 (47.2%) had traumatic brain injury as the cause of admission. SBTs were performed with PSV in 430 (51.3%), T-piece in 329 (39.2%), and CPAP in 80 (9.5%). SBT median duration was 60 min in PSV and T-piece, while 120 min in CPAP. Extubation failure occurred in 177 (21.1%) cases. In multivariable analyses, there was no significant association between SBT modality or duration and extubation outcome. Results were similar in ABI subgroup analyses. After inverse probability weighting, vigorous cough remained the only significant predictor of extubation success. In this large international ABI cohort, neither SBT mode nor duration was associated with extubation failure.