In reference to is polysomnography required prior to tonsillectomy and adenoidectomy for the diagnosis of obstructive sleep apnea versus mild sleep‐disordered breathing in children?

In a recent best practice guideline, the usefulness of polysomnography (PSG) prior to tonsillectomy and adenoidectomy (TþA) for children with suspected obstructive sleep apnea (OSA) was discussed. We certainly agree with the conclusion of the article that the value of PSG for the diagnosis of childhood OSA compared to history and physical examination needs more study. Indeed, PSG is an expensive, time-consuming and not universally available diagnostic tool. However, it is controversial whether PSG should be completely abandoned in children with suspected OSA, simply because the majority of ENT physicians proceed with TþA without preoperative PSG based only on history and physical examination. Would we ever start a child on chronic anticonvulsants for parental report of nocturnal seizure-like activity without performing an electroencephalogram? In an evidence-based consensus statement, the American Academy of Sleep Medicine has described detailed technical requirements for pediatric PSG equipment and clear PSG scoring rules. Hence, there is no controversy in relation to the value of PSG because ‘‘different sleep centers use different equipment and different criteria for defining positive studies.’’ The major challenge is that neither the pediatric nor the ENT scientific communities have ever agreed on universally acceptable indications for TþA as treatment for OSA. The higher the apnea-hypopnea index (AHI), the higher the risk for OSA-related morbidity, but only a small proportion of children with AHI more than 1 episode/h have ‘‘daytime somnolesence’’ or ‘‘poor cognitive performance.’’ Should we treat children with symptoms of OSA but no OSA-related morbidity, if the severity of nocturnal upper airway obstruction is unknown? Similarly, excessive daytime sleepiness, hyperactivity, or learning problems can be present even in completely healthy children. In the individual child, how do we know that these problems are most likely related to obstructive sleep-disordered breathing if the frequency of apneas and hypopneas is not determined by PSG? Systematic reviews have shown that the overall efficacy of TþA in improving AHI is approximately 85% in uncomplicated patients, and it approaches only 50% in obese children. In addition, several nonsurgical treatment options have proven efficacy for treatment of mild OSA (AHI 1–5 episodes/hr) including nasal steroids, leukotriene receptor antagonists, and weight loss for the obese child. The routine use of TþA as treatment for symptoms of obstructive sleep-disordered breathing is even more controversial in view of a recent publication on the natural history of pediatric OSA. Over a 5-year period, OSA (AHI 1 episode/hr) and excessive daytime sleepiness resolve spontaneously in approximately 70% and 60%, respectively, of preadolescent children. In conclusion, more research is urgently needed to clarify the indications of TþA as a treatment option for pediatric OSA. Given the cost and limited availability of PSG, which remains the gold standard for diagnosing OSA severity, future studies should be focused on the development of alternative low-cost diagnostic instruments or OSA-related biomarkers. In this view, we support the proposal by Kheirandish-Gozal et al. to use a combination of criteria in the diagnosis and management of the child with suspected OSA. Given the implications of OSA in childhood, the possibility of residual sleep apnea after surgery, the rare but still existing complications of TþA and the availability of treatment alternatives, we suggest that correct phenotyping of the child with suspected OSA is mandatory. Further research is necessary to identify those individuals with susceptibility to morbidity with even mild sleep apnea, because it is not ethical and logistically feasible to treat every snoring child, whether it is by TþA or by any other treatment modality.