BACKGROUND

Characterized by intermittent hypoxia (IH), sleep fragmentation, and enhanced sympathetic nervous system activity, obstructive sleep apnea (OSA) precipitates oxidative stress, systemic inflammation, and metabolic perturbations. These disturbances manifest as alterations in serum uric acid (SUA) and high-density lipoprotein cholesterol (HDL-C) levels. Recently, the ratio of SUA to HDL-C (UHR) has emerged as a potential biomarker reflecting both inflammatory and metabolic status. This study investigates the association between UHR and OSA.

METHODS

Using a cross-sectional design, data were extracted from adults aged 20 years and older in the National Health and Nutrition Examination Survey (NHANES) database, covering the period from 2015 to March 2020. OSA was determined via the NHANES Sleep Disorders Questionnaire. The investigation employed weighted logistic regression alongside trend tests to evaluate the relationship between UHR and OSA. Nonlinear relationships were examined with restricted cubic spline analysis and threshold effect analysis. Receiver operating characteristic (ROC) curves were utilized to compare the predictive capacities of UHR, SUA, and HDL-C for OSA, with the area under the curve (AUC) calculated to assess the models' predictive accuracy. In addition, mediation analyses were conducted to explore the role of body mass index (BMI) in this association, and sensitivity analyses confirmed the robustness of the findings. Subgroup analyses further assessed the impact of various covariates.

RESULTS

Among the 9985 adults included, 4906 were identified as individuals with OSA. A positive association between UHR and the risk of OSA was observed (OR = 1.02; 95% CI: 1.01, 1.04; P = 0.014). Moreover, a nonlinear relationship was confirmed (P for nonlinearity = 0.024), with an inflection point at a UHR level of 10.23. UHR demonstrated greater predictive accuracy for OSA (AUC = 0.591) compared to SUA (AUC = 0.568) and HDL-C (AUC = 0.580). Additionally, BMI was found to partially mediate the relationship between UHR and OSA, with a mediation proportion of 61.99%. This association remained significant within specific subpopulations (P < 0.05) and was further modulated by factors such as age, alcohol consumption, and diabetes status (P for interaction < 0.05). Sensitivity analyses underscored the stability of these results.

CONCLUSION

UHR is positively correlated with the risk of OSA in adults, with BMI serving as a partial mediator. The findings support UHR as a viable biomarker for early detection and risk assessment in patients with OSA. Strategies focusing on weight management may reduce the risk of OSA among individuals with elevated UHR levels.