SEATTLE—Successful treatments for insomnia may work by tapping into functional systems or targeting dysfunctional brain systems, according to an overview presented at the 29th Annual Meeting of the Associated Professional Sleep Societies. People with insomnia may have intact homeostatic and circadian function, for example, but abnormal metabolism in the brain’s resting state networks during sleep, said Daniel J. Buysse, MD. These factors may require distinct treatment mechanisms.
Homeostatic and Circadian Functions
To investigate potential circadian and homeostatic differences between people with and without insomnia, Dr. Buysse, Director of the Neuroscience Clinical and Translational Research Center at the University of Pittsburgh, and colleagues subjected a group of older adults (mean age, 68) to a physiologic challenge. Sixty-two participants underwent polysomnography at baseline during a normal night of sleep. During a subsequent 36-hour period of sleep deprivation, participants performed a “constant routine” with constant posture and activity for 24 hours and had a further 12 hours of wakefulness while moving ad lib. The researchers assessed the population’s circadian rhythms during the initial 24 hours of sleep deprivation. Participants also took repeated sleep latency tests and wake EEG tests during the period of wakefulness. Finally, participants underwent a night of polysomnography after the end of the sleep deprivation. Salivary melatonin sampling and core body temperature measurements were performed throughout the study.
Dr. Buysse’s group found that participants with insomnia had a pattern of sleepiness that was nearly identical to that of good sleepers. “You can determine clear effects of time and circadian rhythms in these data, but no difference between the insomnia and control participants,” he said.
The researchers also saw no difference in non-REM delta power, a measure of homeostatic sleep drive, between patients with insomnia and controls at baseline or after sleep deprivation. The results suggest that patients with insomnia responded to the homeostatic challenge in the same way as good sleepers did.
The phase and amplitude of body temperature did not differ between the two groups, either, although participants with insomnia were approximately 0.2 °C hotter than controls. The investigators also found a direct correlation between baseline temperature mesor and insomnia severity index. They did not see any difference, however, in phase or amplitude of melatonin.
Dr. Buysse’s group next administered a course of cognitive behavioral therapy for insomnia (CBT-I) to the participants with insomnia. The treatment reduced sleep latency, especially in the middle of the night, although the effect failed to achieve statistical significance. Treatment had no effect on non-REM delta power or body temperature. Participants’ baseline temperature mesor correlated directly with improvement in insomnia severity, however.
“Older adults with insomnia have intact sleep propensity, homeostatic sleep regulation, and circadian function,” said Dr. Buysse. “Sleep deprivation and CBT-I may take advantage of intact homeostatic and circadian mechanisms to help people with insomnia, rather than correcting something that’s broken or impaired.”
Dysregulation in Resting State Networks
Sleep is not an all-or-nothing phenomenon, said Dr. Buysse. The literature indicates that the intensity of sleep varies between brain regions according to the intensity of waking activity in those regions. To investigate this regional variation in sleep, Dr. Buysse used an FDG–PET methodology devised by his colleague Eric Nofzinger, MD. The researchers performed PET scans during wakefulness and during non-REM sleep on 30 control subjects and 37 participants with insomnia. They used subtraction images to look for changes in the brain between sleep and wakefulness.
Imaging revealed relative deactivation in the posterior cingulate, precuneus, medial prefrontal cortex, and inferior parietal cortex during sleep among all participants. These regions are parts of the default mode network, which is thought to perform functions such as self-awareness, autobiographical thought, and mind wandering. All participants also had deactivation in brain regions involved in executive control, such as the lateral prefrontal cortex.
Participants with insomnia, however, had relative activation in the anterior cingulate. This region is part of the salience network, which identifies stimuli that may be relevant, such as threats. Unlike control subjects, participants with insomnia did not have deactivation in the thalamus.
Next, Dr. Buysse’s group performed an interaction analysis to understand how people with and without insomnia differ in their wake–sleep change. They found a smaller deactivation of the posterior cingulate and precuneus during sleep in people with insomnia, compared with controls. They also saw more activity in the inferior parietal cortex during non-REM sleep in participants with insomnia. Finally, controls had a relative decrease in activity in the dorsolateral prefrontal cortex, while patients with insomnia had no change or a slight increase in activity in this region.