Individual alpha frequency during sleep onset in Insomnia: An exploratory study

Introduction: The Alpha rhythm is considered a stable human electroencephalographic (EEG) trait clearly recognizable during the transition from wake to sleep. In Insomnia Disorder (ID), a subjective sleep disorder characterized by difficulties in initiating and/or maintaining sleep, altered alpha rhythm has been linked to a state of hyperarousal impairing the Sleep Onset (SO) process. Within this context, the present study aims to explore the Individual Alpha Frequency (IAF) during the period preceding the SO (Pre-SO) in IDs and Healthy Controls (HCs). We hypothesize (i) higher IAF Power in IDs compared to HCs during the Pre-SO and (ii) a significant relationship between the IAF power during Pre-SO and insomnia severity. Method: Seventeen ID patients (mean age: 45.2 ± 12.1; 6M/11F) and 17 HCs (mean age: 41.7 ± 11.5; 6M/11F) underwent a single night of Polysomnography (PSG) recording, followed by the completion of sleep diaries and the Insomnia Severity Index (ISI) questionnaire. The relative IAF was computed as the specific frequency associated with the strongest EEG power within the extended alpha range (8– 11.75 Hz) considering an occipital (O1, O2) and a parieto-occipital (P3, P4, Pz, O1, O2) cluster during the 5-min interval before the first K-complex or sleep spindle appearance, considered as the Pre-SO period. Results: The nonparametric statistical test showed significantly higher relative IAF power in IDs compared to HCs in occipital (U = 75.0; p = 0.016) and parieto-occipital (U = 84.0; p = 0.038) electrodes during the 5-min Pre-SO. Spearman's correlation analysis revealed a positive association between the ISI total score and the relative IAF power during the 5-min Pre-SO in the parietooccipital (rho = 0.371; p = 0.040) and occipital (rho = 0.366; p = 0.043) clusters. Considering PSG macrostructure indices, Spearman's correlation analysis showed positive correlations between objective sleep latency and relative IAF in the parietooccipital (rho = 0.390; p = 0.023) and occipital (rho = 0.345; p = 0.045) sites. Moreover, parieto-occipital IAF was positively correlated with latency to slow wave (rho = 0.482; p = 0.004) and REM (rho = 0.383; p = 0.026) sleep. Conclusion: Present findings suggest that IAF is a useful parameter in distinguishing ID patients and HC individuals. Specifically, IAF power is significantly higher in patients suffering from ID compared to HCs and positively correlates with insomnia severity and indices of altered sleep. In conclusion, these findings suggest IAF as a potential electrophysiological biomarker in ID.