Association between obstructive sleep apnea and metabolic profiles across sleep stages in pediatrics.

OBJECTIVE

To evaluate the association between metabolic profiles and obstructive sleep apnea (OSA) during rapid eye movement (REM) and non-REM (NREM) sleep in children.

METHODS

A total of 514 children underwent overnight polysomnography (PSG) and biochemical testing. Obstructive apnea hypopnea indices (OAHI) and oxygen desaturation indices (ODI) were recorded for total sleep time (TST), REM, and NREM. A multivariate analysis of covariance model assessed the associations of OAHITST, OAHINREM, and OAHIREM with overall metabolic parameters, adjusting for body mass index (BMI) z-score and other covariates. Then, multivariable linear regression models were used to quantify the independent effects of sleep variables on metabolic profiles during REM and NREM sleep.

RESULTS

When stratified by OAHI-defined severity of OSA, TP, TCH, TG, HDL-C, LDL-C, LDL-C/HDL-C, TyG, and ALT levels were significantly different during TST and NREM (all P < 0.05). In fully adjusted multivariable linear regression models, both OAHITST, OAHINREM and OAHIREM were independently associated with these metabolic parameters. When stratified by ODI-defined severity of OSA, HDL-C, LDL-C, LDL-C/HDL-C, and TyG levels differed significantly during NREM (all P < 0.05) and were independently related to ODINREM. ODIREM was independently associated with TP, TCH, TG, HDL-C, LDL-C, LDL-C/HDL-C, and TyG levels. And in fully adjusted multivariable linear regression models, associations between ODI and TG, LDL-C, and TCH were evident during REM sleep but not during NREM sleep.

CONCLUSIONS

In pediatric OSA, OAHI was associated with metabolic alterations throughout sleep stages. ODI showed more pronounced metabolic effects during REM sleep, where the associations of TG, LDL-C, and TCH varied across ODI severity categories.