All-night spectral and microstate EEG analysis in patients with recurrent isolated sleep paralysis

Front Neurosci. 2024 Feb 8:18:1321001. doi: 10.3389/fnins.2024.1321001. eCollection 2024.

Abstract

The pathophysiology of recurrent isolated sleep paralysis (RISP) has yet to be fully clarified. Very little research has been performed on electroencephalographic (EEG) signatures outside RISP episodes. This study aimed to investigate whether sleep is disturbed even without the occurrence of a RISP episode and in a stage different than conventional REM sleep. 17 RISP patients and 17 control subjects underwent two consecutive full-night video-polysomnography recordings. Spectral analysis was performed on all sleep stages in the delta, theta, and alpha band. EEG microstate (MS) analysis was performed on the NREM 3 phase due to the overall high correlation of subject template maps with canonical templates. Spectral analysis showed a significantly higher power of theta band activity in REM and NREM 2 sleep stages in RISP patients. The observed rise was also apparent in other sleep stages. Conversely, alpha power showed a downward trend in RISP patients' deep sleep. MS maps similar to canonical topographies were obtained indicating the preservation of prototypical EEG generators in RISP patients. RISP patients showed significant differences in the temporal dynamics of MS, expressed by different transitions between MS C and D and between MS A and B. Both spectral analysis and MS characteristics showed abnormalities in the sleep of non-episodic RISP subjects. Our findings suggest that in order to understand the neurobiological background of RISP, there is a need to extend the analyzes beyond REM-related processes and highlight the value of EEG microstate dynamics as promising functional biomarkers of RISP.

Keywords: EEG; EEG spectral analysis; microstate analysis; polysomnography; recurrent isolated sleep paralysis (RISP).

Grants and funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was supported by the Grant Agency of the Czech Technical University in Prague, Reg. No. SGS22/200/OHK4/3T/17 and Reg. No. SGS21/140/OHK4/2T/17, by the Czech Science Foundation, Reg. no. 22-16874S, and by Programme Cooperatio Neurosciences of Charles University.