Coordination of central and autonomic nervous systems during sleep
Abstract
Dysfunction of coordination between the central nervous system (CNS) and autonomic nervous system (ANS) has been implicated as a pathologic aspect of many disorders, including sleep apnea, idiopathic Parkinson's disease, sudden infant death syndrome, and diabetes. Quantitative methods have been recently proposed for diagnosis of autonomic dysfunction. If a quantitative method could reliably measure CNS-ANS coordination, diagnosis of CNS-ANS dysfunction and understanding of associated disorders would be enhanced. This study hypothesizes that central and autonomic nervous rhythms are synchronized at respiratory frequencies and that CNS-ANS synchronization with respiration is modulated by physiologic state. In particular, this study seeks to verify a consequence of the hypothesis, namely that spectral coherence between EEG and heart rate variability (HRV) time series at respiratory frequencies varies with sleep state in healthy young adults. Eight healthy young adults volunteered for the study and gave their informed consent. Subjects reported to the sleep laboratory two hours prior to habitual bedtime for preparation. A urine specimen was obtained and qualitatively tested for substances which might confound the study. Electrodes were then applied to record EEG, electrocardiograms, electrooculograms, and submental electromyograms. EEG electrode locations were controlled by using a cap with standard 10-20 system recording sites. Subjects next assumed a supine position in bed and elastic straps containing piezoelectric transducers were placed around the chest and abdomen to monitor respiratory movements. Subjects retired and recording began at the habitual bedtime. Data analysis began by identifying artifact-free episodes of slow wave sleep (SWS) and paradoxical sleep (REM). Epochs 200 seconds in duration were subjected to spectral analysis. Coherence between HRV and each EEG channel was then computed and values in frequency ranges of predominant HRV components were retained for analysis. Measures of EEG-HRV coherence distributions over the scalp, including average and hilliness values, were computed. Coherence metrics were found to be significantly different in SWS and REM. The study concludes that EEG-HRV coherence provides a noninvasive measurement of CNS-ANS association which is responsive to changes in physiologic state.
This paper has been withdrawn.