Tracheal airleak as a predictor of post-extubation stridor in the paediatric intensive care unit

Objective: To quantify the pressure needed to generate a tracheal airleak around the endotracheal tube that might distinguish successful extubations from those with postextubation stridor. Design: Prospective, observational clinical study. Setting: The Pediatric Intensive Care Unit of a tertiary care, university hospital. Subjects: Fifty-six paediatric patients mechanically ventilated for over 24 hours without parenchymal lung disease or upper airway obstruction. Interventions: None. Measurements: Tracheal airleaks were measured prior to extubation. Main results: A tracheal airleak at 20 cm H₂O of pressure identified post-extubation stridor with 100% sensitivity, 33% specificity and a positive predictive value of 18%. The absence of an airleak at 40 cm H₂O identified postextubation stridor with 43% sensitivity, 80% specificity, and a positive predictive value of 23%. There was no airleak pressure that could successfully distinguish patients at risk for post-extubation stridor. Conclusions: The pressure required to generate a tracheal airleak is an unreliable predictor of post-extubation stridor in paediatric patients mechanically ventilated for over 24 hours for reasons other than parenchymal lung disease or upper airway obstruction.