BACKGROUND The air leak test is recommended for assessing the appropriate size of an uncuffed tracheal tube (TT) in children. Our objectives were to determine whether there is a certain threshold air leak value beyond which a higher risk for adverse events after removal of TT can be predicted and to define other risk factors related to extubation. METHODS We enrolled 234 cases ranging from newborn to 9 years of age requiring tracheal intubation for elective or emergency surgery. General anesthesia was induced by a mask or intravenously. The TT size was calculated using the formula: [age (years)/4] + 4.5. After the induction of anesthesia, the patient's trachea was intubated and the correct position was confirmed. The attending anesthetist assessed the leak pressure. Incidences of adverse events (prolonged or barking cough, obstructed or prolonged inspiration or expiration, subcostal and sternal retractions, arterial desaturation, or laryngospasm) were recorded after removal of TT. RESULTS Ten patients were excluded from the study. A total of 218 children underwent 224 operations under general anesthesia. Children who had an absent air leak at 25 cmH(2)O pressure had 2.8 times more adverse events during emergence from anesthesia than those with an audible air leak. Adverse events after the removal of TT were 3.7 times more likely to occur in children whose anesthesia was provided by a less experienced anesthesia trainee. CONCLUSIONS Adverse events after removal of TT were more likely to occur in children with an absent air leak at 25 cmH(2)O pressure and in children whose anesthesia was provided by a less experienced anesthetist.