The decision to perform tonsillectomy and adenoidectomy for treatment of pediatric obstructive sleep apnea syndrome is often made on a clinical basis without formal polysomnography. To examine the accuracy of the clinical diagnosis of pediatric obstructive sleep apnea syndrome, we prospectively evaluated 30 children with obstructive symptoms by a standardized history, physical examination, and review of a tape recording of breathing during sleep. On the basis of this clinical evaluation, patients were divided into three predictive groups: (1) definite obstructive sleep apnea syndrome, (2) possible obstructive sleep apnea syndrome, and (3) unlikely to have obstructive sleep apnea syndrome. Nocturnal polysomnography was used to determine the presence or absence of true sleep apnea. Ten of 18 (55.6%) patients predicted clinically to have definite obstructive sleep apnea syndrome had positive nocturnal polysomnographies. Two of six (33.3%) patients predicted to have possible obstructive sleep apnea syndrome had positive nocturnal polysomnographies. One of six (16.7%) patients predicted to be unlikely to have obstructive sleep apnea syndrome had a positive nocturnal polysomnography. Six nocturnal polysomnographies negative by conventional criteria were suspicious for apnea, but considering these positive for obstructive sleep apnea syndrome did not improve the specificity of the clinical prediction. Our results show that clinical assessment of obstructive sleep apnea syndrome in children is sensitive (92.3%) but not specific (29.4%) for making the diagnosis of obstructive sleep apnea syndrome as compared with nocturnal polysomnography and may contribute to the decision to obtain nocturnal polysomnography in specific circumstances.