The "air leak" test around the endotracheal tube, as a predictor of postextubation stridor, is age dependent in children.

OBJECTIVES: The airleak test is measured with a manometer as the pressure necessary to generate an audible airleak around the endotracheal tube. Our objectives were to determine whether the airleak test predicts postextubation stridor in children and if age affects its sensitivity and specificity.

DESIGN: A retrospective study.

SETTING: Pediatric intensive care unit.

PATIENTS: We studied all intubated patients admitted to our pediatric intensive care unit between July 1998 and December 1999. Patients were excluded if they had acute viral croup, tracheal surgery, hypotonic airway, or vocal cord paralysis or if they died before extubation.

INTERVENTIONS: Medical records were reviewed for patient demographics, presence of an airleak on the day of extubation, airleak values, presence of postextubation stridor, and extubation failure secondary to upper airway obstruction.

MEASUREMENTS AND MAIN RESULTS: One hundred and five patients met our inclusion criteria and had an airleak test at the time of extubation. In children <7 yrs of age, the incidence of postextubation stridor was similar in patients with or without an airleak at >20 mm Hg (50% vs. 67.7%; >.05) with a sensitivity of 65.6% (95% confidence interval, 46.9-80.8). In children >or=7 yrs of age, the incidence of postextubation stridor was greater in patients with an airleak at >20 mm Hg (55.5% vs. 5.8%; p=.01) with a sensitivity of 83.3% (95% confidence interval, 36.8-99) in predicting postextubation stridor.

CONCLUSIONS: Our study suggests that the airleak test has a low sensitivity when used as a screening test to predict postextubation stridor in young children (<7 yrs old), whereas in older children (>or=7 yrs old) the airleak test may predict postextubation stridor.