Conventional weaning parameters do not predict extubation failure in neurocritical care patients.

OBJECTIVE: Predicting extubation failure (EF) is one of the most challenging aspects of critical care medicine. The literature on EF in neurocritical care patients is very scarce. We sought to determine the ability of traditional weaning parameters to predict EF in neurocritical patients.
METHODS: This is a retrospective observational study performed at a tertiary level, academic, Neurological Intensive Care Unit (NCCU). We included all adult patients intubated for neurological reasons in whom an attempt to perform extubation was performed. We compared traditional weaning parameters between patients who failed extubation and those successfully extubated. Fishers exact test was used for categorical variables and t-test for continuous variables. Weaning parameters were analyzed as categorical variables and when appropriate as continuous. We incorporated a coma scale (Four Score) in an attempt to determine if neurologic dysfunction could account for EF.
RESULTS: The study sample compromised 62 patients undergoing extubation trial. None of the individual weaning parameters predicted EF: rapid shallow breathing index (P = 0.62), minute ventilation (P = 0.7479), respiratory rate (P = 1.0), negative inspiratory force (P = 0.62), tidal volume, and PaO2/FIO2 ratio (P = 1.0). There was no significant difference in Four Scale score between EF and successfully extubated patients (0.44, proportions of the mean, t-test). There was no combination of weaning parameters that allowed prediction of EF. All patients had at least three normal weaning parameters, but there was no combination of parameters that accurately predicted EF. Overall, weaning parameters had better specificity than sensitivity for predicting EF.
CONCLUSIONS: In this sample of neurocritical care patients undergoing extubation trial, traditional weaning parameters do not predict extubation failure.