In intubated patients the presence of a cuff leak (CL) is used as a predictor of successful extubation. CL is proposed to indicate laryngeal edema and predict which patients may develop complications such as postextubation stridor and eventual reintubation. Our objective was to evaluate the reliability of CL in our population of critically ill trauma patients. A retrospective chart review was performed of patients admitted to the trauma service who required mechanical ventilation. All patients undergo the CL test by a single respiratory therapist team before attempted extubation. Data collected included body mass index (BMI), endotracheal tube (ETT) size, length of time of mechanical ventilation, tidal volumes (Vt), and the size of the patient's trachea based on CT scan. The test is performed by the respiratory therapists and involves measuring expired Vt before and after the ETT cuff has been deflated and listening for an audible leak. A positive test result is defined as a CL greater than 10 per cent of Vt or, when volumes are not available, as audible air expired. From October 2005 to May 2006, 150 mechanically ventilated patients were identified and 49 charts were available for review. Forty-one patients had a cuff leak (+CL), whereas eight did not (-CL). The two cohorts were similar in age (+CL = 36.5 years, -CL = 38.1 years, P = 0.82), male gender (+CL = 70%, -CL = 50%, P = 0.25) ETT size (+CL = 7.4, -CL = 7.4, P = 0.57), and BMI (+CL = 28 kg/m2, -CL = 27 kg/m2, P = 0.71). The average tracheal diameter (+CL = 17.4 mm, -CL = 17.5 mm, P = 0.90) as well as the ratio of ETT and tracheal diameter was similar for the two cohorts (+CL = 0.65, -CL = 0.64, P = 0.73). Four patients (10%) in the +CL cohort failed extubation, whereas none of the -CL cohort failed (0%) (P = 0.40). The CL test does not reliably identify those patients who will require reintubation in our trauma population. In addition, the ratio of ETT and tracheal diameter is not predictive of successful extubation.