Spectral analysis in untreated primary insomniacs: evidence for cortical hyperarousal and prefrontal hypometabolism during sleep

Abstract

Introduction Primary insomnia is characterized by complaints of non-restorative and/or insufficient sleep during at least one month in the absence of other pathologies according to the DSM-IV. However, the pathophysiology of insomnia is not well understood. According to a number of studies, primary insomniacs display cortical hyperarousal during wake and sleep/wake transition which could explain both the lack of objective evidence of cognitive dysfunction and the difficulty to initiate or maintain a good sleep in insomnia. In support of the hyperarousal hypothesis, sleep spectral analysis in insomniacs consistently reveal an increase in beta power during NREM sleep in comparison to good sleepers. Recent studies investigated the activity and the metabolism in the prefrontal cortex during wake in insomnia and proposed that insomniacs have a hypometabolism and a hypoactivation of the prefrontal cortex during wake. Previous spectral analyzes in insomnia have been performed only in central electrodes (Krystal et al., 2002; Spiegelhalder et al., 2012). The aim of the present study was to assess delta, sigma, alpha and beta power spectra in prefrontal, occipital, temporal and central cortical areas and to compare activity in prefontal regions to other scalp regions. Materials and methods Overall, 14 insomniacs and 10 good sleepers participated in the study. They completed one night of polysomnography in the sleep lab. Power spectra were calculated using the Fast Fourier Transformation during stage 2 and stages 3 + 4 of sleep and during REM at prefrontal, occipital, temporal and central electrode positions. Results Our results showed that insomniacs exhibited a higher beta power spectrum centrally, temporally and occipitally during stage 3 + 4 in comparison with good sleepers. However, power spectra did not differ between groups at prefrontal electrodes. During the REM stage, beta1 power spectrum significantly decreased at prefrontal electrodes in insomniacs in comparison with good sleepers. Finally, we also found an increase in the delta mean frequency during NREM sleep in insomniacs in comparison with good sleepers. Conclusion Main results are in line with the literature and the concept of hyperarousal. Interestingly, our data suggests hyperarousal in large parts of the brain of insomniacs except for the prefrontal cortex. The latter may be related to selective hypometabolism in this region as previously reported in insomniacs during wake (Altena et al., 2008; Notzfinger et al., 2004). Acknowledgements We want to thanks the participants for their implication in this study.