© 2009 American Thoracic Society
Outcomes in Children with Obstructive Sleep ApneaTo the Editor:We read with great interest the study reported by Dr. Amin and colleagues (1) as it covers an important but neglected area of pediatric obstructive sleep apnea, that is, long-term outcomes after surgical intervention. There are several points that warrant further discussion. First, it is difficult to understand Amin and colleagues' definition of sleep-disordered breathing (SDB) recurrence as an apnea–hypopnea index (AHI) >3, using the value of "1 SD ± the mean value of the comparison group at 1 year," when they defined SDB as an AHI >1.0 at the beginning of the study. As stated in our previous article on the statistical issue of using regression to model the AHI, modeling this index by its categorization based on a cutoff point not only reduces the statistical power, but the choice of cutoff point will affect the results (2). As the cutoff point is post priori defined based on the characteristic of the sample, the conclusion from Amin and colleagues' study cannot be generalized to other populations. Second, Amin and colleagues failed to adjust for the fact that normal values of body mass index (BMI) change with age in children, unlike in adults (3). Failure to account for this point may render the analysis of "rate of BMI change" inaccurate. Transforming the raw BMI to BMI z score using the age- and sex-specific normal values would help. Third, Amin and colleagues used multiple comparisons of data from baseline, 6 months, and 12 months by using repeated simple t tests. This is not advisable because it will increase the probability of type 1 error in at least one of the t tests (4). It is also not possible to deduce which of the statistical significant results is, in fact, a false positive. Omnibus tests, such as repeated-measure ANOVA or Friedman tests coupled with an appropriate post hoc test, should be used. Fourth, there are different methods of tonsillectomy and adenoidectomy (TandA) (5), for example, cold versus hot dissection and suturing the tonsillar wound to reduce collapsibility of the pharynx and to prevent scarring and narrowing of the posterior pharynx versus not suturing. It would be important for Amin and colleagues to report the type of TandA practiced in their center.
Kwong Wah Hospital FOOTNOTES Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
