Heightened Awareness in Insomnia

Abstract

The often asked question “how did you sleep?” is more metaphysically complex than it appears. In one sense, the answer after a good night’s sleep should be, “I have no idea, I was unconscious at the time.” Certainly many insomnia patients who respond negatively to the question are not really commenting on the time they were asleep, but rather on the time that they remember being awake. Memories of looking at an alarm clock at 1:00, 2:00, 3:00, and 4:00 AM reinforce the belief that they hardly slept, despite the fact that the periods between 1:01 and 1:59; 2:01 and 2:59, etc., may have been filled with sleep spindles, K-complexes, theta EEG activity, or rapid eye movements. The salient data in determining subjective quality in this case are the brief periods of conscious awareness, not the lengthier periods of “unconscious” sleep. Anyone who has undergone general anesthesia, has been asked to count backwards from 100, and then several hours later has no memory of anything that happened after “97,” can attest to the fact that we are only aware of being conscious, and have no frame of reference with which to comprehend the passing of time when unconscious. Sleep, unlike anesthesia or coma, can be thought of as set of rapidly reversible states of unconsciousness, and as such raises interesting questions as to the extent to which conscious awareness occurs throughout the sleep period and how it influences subjective sleep quality.

The paper by Corsi-Cabrera et al. 1 in this issue of SLEEP seeks to investigate the possibility of hyperarousal in insomnia leading to the possibility of conscious awareness occurring in the context of what we have traditionally viewed to be EEG evidence of NREM sleep. This is an important and intriguing idea, investigated in a carefully conducted experiment. Their most interesting finding is the maintenance of frontoparietal coupling of high frequency EEG activity into stage 1 sleep in insomnia patients, indicating continued activation of a frontoparietal network thought to relate to self-awareness. The emergence of such long-range connectivity in the brain has very recently been shown to relate to the recovery of consciousness in patients previously in vegetative or minimally conscious coma states. 2 The normal control EEG data collected by Corsi-Cabrera et al. 1 complement the fMRI findings of Czisch’s laboratory 3 and others in showing evidence of diminished default network coupling during all NREM sleep stages, but with the capability of a dramatically increased temporal resolution relative to fMRI, limited as it is by the biophysics of the hemodynamic response function, inherent to the BOLD response.

The finding of altered connectivity during stage 1 raises some interesting possibilities for future investigations in insomnia sufferers. Stage 1 is itself a heterogeneous state, consisting of rapid oscillations between wake (alpha and beta) and sleep (theta) EEG periods thought to map onto conscious and unconscious microstates. 4 Work pioneered by John Trinder has shown that when these alpha and theta microstates are separated, several aspects of physiology show dramatic and rapidly reversible changes. Thus tidal volume, 5 – 8 upper airway resistance, 9, 10 and upper airway muscle activity 9 – 13 show immediate changes in the first theta breath of a sleep onset when compared to wakefulness or the first alpha breath of an arousal from sleep. Using the techniques developed by Trinder, we were able to show rapid alterations in EEG responses evoked by simple auditory and respiratory stimuli, depending on whether stimuli were presented during alpha or theta microstates within N1. Specifically the P300 response, thought to reflect high-level cognitive processing, was maintained from wakefulness into N1 alpha but replaced by the inhibitory N350 component as the dominant response in N1 theta. 14, 15 In the context of the Corsi-Cabrera findings it would be intriguing to know whether the frontoparietal coupling in insomnia in stage 1 was specific to alpha microstates or persisted into theta as well.

There are however two cautionary notes that need to be considered in the context of evaluating stage 1 sleep and consciousness awareness driving subjective experience of wakefulness or poor sleep. The first is that just having more conscious awareness during sleep may not be sufficient to produce the subjective impression of poor sleep quality. For example, women with premenstrual syndrome have poorer subjective sleep quality in the (symptomatic) luteal phase than the follicular phase of the menstrual cycle. 16 They also show increased WASO in the luteal phase. However, asymptomatic control women showed an almost identical effect of menstrual phase on WASO but with no associated change in subjective ratings of sleep quality. 16 In this case there is clearly an interaction of stage 1 sleep and other psychological factors that produces the subjective experience of poor sleep. The second factor to consider is that not all stage 1 epochs are created equal. For example, in one case a 3-minute period of stage 1 following an arousal might consist of 10 seconds of alpha in the first epoch, followed by 5 epochs of uninterrupted theta activity prior to the first K-complex or spindle. In a second case, a 3-minute stage 1 period might have 6 epochs, each with 10 seconds of alpha. Thus despite the same number of minutes of stage 1, the amount of potential conscious awareness might vary substantially (10 vs. 60 seconds). This argues for microstate evaluation within stage 1 when investigating these issues.

Insomnia is a highly prevalent disease, with far-reaching health, safety, and even economic implications for millions of sufferers. Thus far there is a scarcity of treatments that have long-term effectiveness. The insights provided by the type of novel data presented in Cabrera et al. 1 are necessary if we are to increase our understanding of the disease and develop more effective treatments.