[Microstructural characteristics of sleep electroencephalogram in obstructive sleep apnea]

Objective: To analyze the characteristics of electroencephalogram (EEG) power spectral density (PSD), slow wave oscillation, and sleep spindles across different sleep stages, brain regions, and levels of severity in patients with obstructive sleep apnea (OSA). Methods: This case-control study collected data from 1 821 suspected OSA patients, comprising 1 422 males and 399 females, with a median age 39 (32, 49) years, who attended our hospital between July 2018 and July 2024. EEG features, including PSD, slow wave oscillations, and sleep spindles, were extracted from polysomnography recordings across different sleep stages and brain regions. The changing trends of these features were analyzed between the OSA and non-OSA groups, as well as across different OSA severity levels. For non-normally distributed variables, the Mann-Whitney U test was used for two-group comparisons, and the Kruskal-Wallis H test with Bonferroni correction was used for multi-group comparisons. For normally distributed variables, the t-test (for two groups) and one-way ANOVA with Bonferroni correction (for multiple groups) were applied. The chi-square test was used for comparisons of categorical variables. Results: Among the 1 821 subjects, 1 706 were diagnosed with OSA including 302 mild, 351 moderate, and 1 053 severe cases. Compared with the control group, the OSA group exhibited the following whole-brain EEG changes: increased PSD in Delta, Theta, Alpha, and Beta bands during N1 stage; increased PSD in Alpha, Beta, and Gamma bands during N2 stage; and increased PSD in all frequency bands during N3 and REM stages (all P<0.05). Moreover, PSD in all frequency bands showed a significant increasing trend with increasing OSA severity (all P _{for trend}<0.001). During N3 stage, the OSA group showed significantly higher density, duration, and amplitude of slow oscillations compared with the control group (P<0.05); during N2 stage, the OSA group showed significantly lower sleep spindle density compared with the control group (P<0.05). Changes in sleep EEG microstructural features were significantly correlated with OSA severity (all P _{for trend}<0.05). Conclusion: Patients with OSA exhibit EEG features during sleep characterized by cortical hyperarousal and thalamocortical rhythm dysregulation, which intensify with increasing disease severity. The observed alterations in EEG microstructural features suggest their potential value as indicators for assessing OSA-related brain functional impairment.