Objective: The purpose of this study is to determine the accuracy of the cuff-leak test in a cardiovascular ICU.
Methods: Five hundred twenty-four patients were tested immediately before being extubated 531 times. The cuff-leak test was performed with the ventilator set in assist-control mode at a tidal volume (VT) of 10 to 12 mL/kg. The leak was taken to be the difference between the preset inspiratory VT and the average of the three lowest of the subsequent six expiratory VTs. A leak </= 110 mL was considered to be a positive leak test.
Results: None of the 20 patients with a positive leak test developed problems. Three patients had postextubation stridor. Their leaks were 433, 312, and 350 mL.
Conclusion: The cuff-leak test is inaccurate and cannot be recommended for routine use in this population.