Related Experiment Video
Updated: Oct 4, 2026

Drug-Induced Sleep Endoscopy (DISE) with Target Controlled Infusion (TCI) and Bispectral Analysis in Obstructive Sleep Apnea
Published on: December 6, 2016
Obstructive sleep apnea severity associates with reduced phase-lag index functional connectivity during N2 sleep
1Department of Neurology, Ulsan University Hospital, Ulsan, Korea.
Purpose:
Obstructive sleep apnea (OSA) is increasingly recognized as a risk factor for cognitive impairment and dementia. We investigated stage- and frequency-specific associations between OSA severity and both conventional spectral measures and phase-lag index (PLI)-based functional connectivity.
Methods:
We retrospectively analyzed polysomnography data from 50 patients (mean age 51.7 ± 14.3 years, 70% male) diagnosed with OSA or primary snoring. Six-channel EEG data were analyzed using conventional spectral power ratios (alpha+beta/theta+delta) and PLI-based functional connectivity across delta, theta, alpha, and beta bands. Pearson correlations assessed associations between apnea-hypopnea index (AHI) and EEG parameters across wakefulness, N2 sleep, and REM sleep.
Results:
Median AHI was 25.7 [IQR: 7.2-48.3]. Conventional spectral measures showed no significant AHI correlations in any sleep stage. PLI analysis revealed stage- and frequency-specific associations: during N2 sleep, AHI negatively correlated with delta-band PLI (r = -0.335, 95% CI -0.588 to -0.021, p = 0.037) and theta-band PLI (r = -0.326, 95% CI -0.582 to -0.012, p = 0.043). No significant associations were observed during wakefulness or REM sleep.
Conclusion:
Functional connectivity analysis identified stage-specific associations with OSA severity that were not observed with conventional spectral measures, suggesting that PLI-based connectivity may provide complementary information on OSA-related neural network alterations.
Related Concept Videos
Sleep Apnea
The condition is more prevalent among...
Sleep-Wake Cycles
NREM Sleep
NREM sleep comprises four progressive stages that seamlessly merge: