Sleep and Sleep Disorders: Comorbid Conditions and Pathophysiology

Thomas Roth, PhD, is the director of the Sleep Disorders and Research Center at Henry Ford Hospital, a professor at Wayne State University, and a professor at the University of Michigan College of Medicine. Roth has published extensively on the pathophysiology of sleep, fragmented sleep, and sleep disorders; during the presentation, he focused on the evolution of our understanding of insomnia.

Roth began by discussing the nature of insomnia and current treatments with respect to the National Institutes of Health (NIH) conference. He noted that in 1983, insomnia was a symptom, not a primary disorder. By 2005, however, the NIH had defined insomnia as a disorder that is comorbid with many other disorders.

To define a disorder, Roth explained that the condition must have distinct symptoms, must produce clinically meaningful distress, and must involve a pathophysiologic mechanism. Insomnia is a disorder that occurs despite the adequate opportunity to sleep. The presence of a crying child or uncomfortable situation that makes sleep difficult does not necessarily constitute an inability to sleep, Roth stated.

Using a 2-hour diagnostic interview of participants drawn from the general population, Roth demonstrated how the prevalence of insomnia differs depending on the diagnostic criteria employed (eg, The Diagnostic and Statistical Manual of Mental Disorders (DSM), International Classification of Diseases Criteria (ICD), or Research Diagnostic Criteria (RDC). Use of the ICD definition resulted in the lowest prevalence of insomnia (5%), followed by the RDC (around 15%), then DSM (20% to 25%). The reason for this, Roth mentioned, was that according to the ICD definition, a diagnosis of insomnia requires a high degree of daytime morbidity. A surprising finding, Roth pointed out, was that people over the age of 58 years had a lower prevalence of insomnia than individuals from younger age groups.

In discussing the morbidity of insomnia, Roth stated that insomnia involves more than the inability to sleep, but also the inefficiency of sleep. He presented results that demonstrated an increase in pain sensitivity and mortality with low sleep efficiency. These results have also been found in patients with sleep apnea; however, Roth contended that sleep apnea is a separate disorder from insomnia. Insomnia also increases the incidences of depression, anxiety, alcohol abuse, and drug abuse. Difficulty sleeping is also a significant risk factor for falls.

Insomnia has been associated with hypertension, as demonstrated in a study by Vgontazas et al which showed that insomnia increases the risk of hypertension by 82%. The danger, Roth noted, is a convergence of both loss of sleep and insomnia, which increases the risk of developing hypertension even more than either condition alone.

Impairment of homeostasis, the circadian process, and hyper-arousal (the wake system) have been proposed as pathophysiological components of insomnia. Because patients with insomnia also show normal homeostatic response to sleep deprivation, Roth stated that homeostatic imbalance does not explain the pathophysiology of insomnia. Although delayed circadian rhythms also increase the risk of insomnia, Roth does not believe this to be the main etiology either. He spoke of the increased early-morning cortisol levels, increases in the multiple sleep latency test (MSLT) in patients with insomnia, and the relationship between MSLT and sleep time. Insomniacs produced the lowest MSLTs, an exact inverse relationship when compared with normal, healthy individuals. People who sleep 8 hours have higher MSLTs (more alert), and people who sleep 5 hours have lower MSLTs (less alert). For those with insomnia, opposite outcomes are produced. Levels of urinary norepinephrine are also higher among patients with insomnia. Catecholamines in the urine indicate autonomic activity, which suggest a wakefulness- and hyperarousal-related pathophysiology of insomnia.

Seen with a brain scan, patients with insomnia have increased activity in arousal centers of the brain, but what is important to note, Roth said, is that the individuals are currently in phase 2 sleep and are sound asleep during this time. In this study, there were no differences in total sleep times between insomnia-affected individuals and normal individuals. Insomnia is not simply time awake during the night, but is also hyper-arousal during sleep.

The reason behind the hyperarousal may be a differential response to stress, as found by Morin et al in 2003. A hyperaroused response to cold temperatures, as measured by heart rate, revealed a higher heart rate in people with insomnia. Stepanski et al demonstrated similar results in subjects with insomnia, with a higher arousal level following exposure to a loud noise. Cortisol release also occurred at higher levels in patients with insomnia. Even public speaking, investigated by McClure et al in 2004, produced different responses in people with insomnia. When challenged with the task of speaking in public, patients with insomnia slept less and had higher degrees of alertness during the day compared with individuals without insomnia. People with insomnia may respond to the sleep environment itself as a stressor.

Roth continued, “The question becomes, which is first? Insomnia or the hyperarousal?” A study by Jefferson et al found that individuals with insomnia experience poorer sleep in response to stress. Twin studies showed genetic heritability of sleep reactivity, with 29% heritability among females and 43% heritability among males. Investigators have studied people with high FIRST (Ford Insomnia Response to Stress Test) scores in the laboratory setting; a high FIRST score indicates a vulnerability to sleep disturbance in response to stress. The researchers found that subjects with high FIRST scores experienced higher sleep latency and lower sleep efficiency compared with normal controls.

Individuals with high FIRST scores also respond differently to caffeine, demonstrating a greater sensitivity to the effects of caffeine than people with low FIRST scores. However, Roth questioned the relationship between FIRST scores and insomnia. He described a study in which he and collaborators found that subjects with high FIRST scores had an 11.2% risk of developing insomnia over 13 months versus 3.4% in people with low FIRST scores. According to findings by Drake et al (2013), for every point on the FIRST scale, a patient develops a 15% increase in the risk of developing insomnia over a 2-year period.

Insomnia has also been linked with depression. A study of 1396 patients showed that those who developed insomnia were 38% more likely to develop depression afterward.

Comorbidity, Roth emphasized, is an important aspect of insomnia. In a study of eszopiclone plus fluoxetine versus eszopiclone with placebo, a higher response occurred with the combination of eszopiclone and fluoxetine than with eszopiclone with placebo. The sleep medication also augmented the antidepressant response on the Hamilton Depression Rating scale. Roth asked the audience, “Is this a drug effect or a sleep effect?” In a study by Manburg et al (2005), a similar response occurred with a nondrug treatment of cognitive behavioral therapy intervention (CBTI) with sleep. In other words, the improvement in depression was caused not by the effects of sleeping medications, but by the improvement in sleep associated with use of a sleeping medication.

Pain sensitivity also improved following CBTI treatment and an extended period of sleep. In patients about to undergo knee and hip replacements, those who had a sleep extension of 2 hours prior to the surgery required less pain medication and reported less pain. Conversely, individuals who are taking gabapentin also have a lower use of opioids. Another study also demonstrated a decreased use of the pain medication zolpidem, and decreased pain sensitivity after knee-replacement surgery.

Roth asked, “How much of the effect is directly related to CBTI, and how much is an indirect effect related to sleep?” He noted that we do not have the answer to that question, but it is an important inquiry for future of understanding insomnia.

In a study involving patients with 11 different diseases including sleep apnea, depression, neuropathic pain, and rheumatoid arthritis, Roth and collaborators found that 52% of the diseases have some morbidity that is attributable to insomnia. Furthermore, insomnia is responsible for 9.8% of all accidents in the United States. Roth concluded his presentation by stating that if insomnia were eliminated, the productivity of the United States could be increased by 29%.