To the Editor,
I read with interest the article by Hanlon and colleagues.1 The authors conducted a cross‐sectional study to know the association between obstructive sleep apnea (OSA) and left ventricular hypertrophy (LVH) in obese children with hypertension. Adjusted odds ratios of severe apnea‐hypopnea index (AHI > 10) for LVH significantly increased. I have queries about their study with special reference to statistical validity.
Although significant increases were observed in any criteria of LVH, wide ranges of 95% intervals might be related to insufficient number of children with LVH, especially selecting higher values of left ventricular mass index. The authors used age, sex, race, and BMI z‐score for the adjustment, and the number of events per variable less than 10 has a problem of prediction by logistic regression analysis.2, 3 Although there is an opposite opinion that EPV value less than 10 is also acceptable to know the association instead of prediction,4 I suppose that more stable estimates might be performed by increasing the number of events.
In addition, the authors concluded that dose‐response relationship between OSA and LVH existed. But I think that information is more required to confirm such relationship. Although an interventional study with treatment of OSA and subsequent the change of CVD risk in children is important, there is a gap between a cross‐sectional study and interventional trial.
CONFLICT OF INTEREST
None declared.
REFERENCES
- 1. Hanlon CE, Binka E, Garofano JS, et al. The association of obstructive sleep apnea and left ventricular hypertrophy in obese and overweight children with history of elevated blood pressure. J Clin Hypertens (Greenwich). 2019;21(7):984‐990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Novikov I, Fund N, Freedman LS. A modified approach to estimating sample size for simple logistic regression with one continuous covariate. Stat Med. 2010;29(1):97‐107. [DOI] [PubMed] [Google Scholar]
- 3. Peduzzi P, Concato J, Kemper E, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49(12):1373‐1379. [DOI] [PubMed] [Google Scholar]
- 4. Vittinghoff E, McCulloch CE. Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol. 2007;165(6):710‐718. [DOI] [PubMed] [Google Scholar]