Psychiatric Disorders That Affect Sleep

Rochelle Zozula

Rita Brooks

There is an intimate relationship between emotional functioning and sleep. Psychiatric disorders are common in our society, and as a result, patients with psychiatric disorders will be seen frequently in sleep disorder centers. Understanding the symptoms associated with these conditions is critical to evaluating the patient properly, creating the optimal sleep environment, and addressing the challenges that can arise during the polysomnographic (PSG) testing procedure. Psychiatric disorders are often treated with medications, and this usually occurs well before an overnight sleep study is performed. Some of the changes in sleep architecture observed during the study may reflect the effects of medications rather than primary illness. We will try to differentiate between the effects of the illness as opposed to medications.

This chapter will review the sleep characteristics of the major psychiatric disorders, including mood disorders, anxiety disorders, schizophrenia, and alcoholism. In this chapter, we will discuss the major disorders and their effects on sleep, including defining the disorder, discussing presenting symptoms, and determining what challenges the sleep technologist may face.

MOOD DISORDERS

It is normal for people to have changes in their mood or affect. However, when these mood changes are extreme, or when they shift from one extreme to the other, or when their mood is not consistent with their life events, they are said to have a mood disorder (1). The mood disorders include major depressive disorder (MDD), bipolar disorder, and seasonal affective disorder.

Major Depressive Disorder

MDD requires a minimum symptom duration of 2 weeks with no manic or hypomanic episodes. Symptoms include depressed mood; reduced level of interest in most or all activities; weight loss or gain; difficulty falling asleep or sleeping too much; feelings of inadequacy; feelings of extreme guilt; a reduced ability to think, concentrate, or make decisions; frequent thoughts of death or suicide; fatigue; and behavior that is agitated or slowed (2). MDD occurs some time in the lives of up to 20% of the North American population, and at any given time about 5% of the population is affected (3). Depression is most likely to begin in adolescence or in the early adult years, but the incidence seems to be increasing in younger populations (4). Between puberty and menopause, women are about two times as likely to develop major depression (5). The lifetime prevalence of MDD is 8% to 12% of men and 20% to 26% of women (6). Throughout a lifetime, an individual may experience several episodes of major depression, so it is a recurrent illness. Sleep disturbance is more pronounced during active illness, but sleep problems may persist during remission as well.

Sleep disturbance is so common in mood disorders that many individuals are given a diagnosis of
depression on the basis of the symptom of insomnia. This may be due to a poor understanding in the general medical community of the variety of sleep disorders, or it may be because of an often erroneous assumption that antidepressant medications are good hypnotics. Although depression is perhaps overdiagnosed in patients with insomnia complaints, sleep disturbance is clearly a common feature of most psychiatric illnesses, including depression. For example, subjective sleep complaints are reported by more than 80% of MDD patients (7).

Although the specific cause of mood disorders has not been determined, there are several theories proposed that may elucidate the connection between mood disorders and sleep-related changes. First, a number of neurotransmitters have been implicated in mood disorders. Specifically, relative deficiencies of noradrenergic (NE) or serotonergic (5-HT) activity or increased cholinergic activity may be involved in depression (8). Notably, cholinergic activity is responsible for initiating rapid eye movement (REM) sleep, and norepinephrine and serotonin are responsible for terminating REM sleep and making the transition back to non-REM (NREM) sleep. Changes in these neurotransmitters would be expected to influence REM sleep, and in fact, REM sleep changes are among the most robust findings in depressive disorders. In addition, antidepressant medications typically enhance NE or 5-HT activity (by blocking the neurotransmitter uptake mechanisms), and patients taking these medications usually show characteristic changes in REM sleep (8, 9).

Another theory proposed for depression is that it may be caused by excessive amounts of REM sleep and a resulting decrease in REM sleep pressure (10). According to this theory, total or partial sleep deprivation can result in significant improvement in depression primarily through the mechanism of REM sleep suppression, and thereby, increasing the pressure to get into REM. Sleep deprivation studies have shown significant improvement in depressive symptoms in about 40% to 60% of patients, but the effect is only temporary and goes away after the REM sleep deprivation is stopped and that patient is allowed to have REM sleep again (11). Specifically, the improvement in mood is short-lived, because 50% to 80% of responders have partial or complete relapse after recovery of REM sleep.

A third theory regarding depression involves disruption of circadian and ultradian rhythms that control sleep-wake activity and REM-NREM cycles. It has been suggested that the circadian timing mechanism that controls REM sleep, core body temperature, and the hormone cortisol is phase-advanced (12). It has been hypothesized that REM sleep is phase-advanced relative to the sleep-wake cycle in depressive patients, and that depression occurs because of awakening at sensitive circadian phases. Other circadian models of depression have focused on the flattening (reduction) of the amplitude of circadian rhythms, including the core body temperature rhythm and rest-activity cycles (derived from actigraphy) (13). It has even been hypothesized that this circadian disruption may be linked to an increased susceptibility to having mood disorders across the life span.

Subjective Findings

Sleep disturbance can be assessed with subjective measures such as questionnaires, objectively with PSG, or inferentially with actigraphy. As previously noted, 80% of MDD patients complain about their sleep. These subjective complaints include difficulty falling asleep, frequent awakenings, early morning awakening with difficulty falling back to sleep, nonrestorative sleep, insufficient total sleep time (TST), and disturbing dreams. Some patients report increased daytime fatigue. The word fatigue is often used to reflect low energy or lethargy, as opposed to daytime sleepiness or drowsiness (difficulty remaining awake).

Subjective measures correlate well with some, but not all, objective measures. For example, patients are fairly good at reporting time in bed, TST, and sleep latency. They are less accurate when reporting the number of awakenings, sleep quality, sleep depth, and how rested they feel upon awakening (14).

Subjective reports on sleep quality may have clinical utility in managing MDD. Survey data have demonstrated that sleep disturbances can occur before the onset of other symptoms of MDD. For example, insomnia occurred as the first symptom in 41% of initial cases of MDD, and it was the first symptom in 56% of relapses. This association is so strong that some investigators can almost predict that a depressive episode will occur sometime in the future, perhaps even months or years, after insomnia first develops. Furthermore, improvement in sleep-related disturbances has been found to be positively correlated with the remission of depressive symptoms (15). This is in contrast to anxiety disorder, where the insomnia complaint typically begins with, or after, the onset of anxiety symptoms (16). The astute clinician should anticipate a relapse of depression in a “controlled” patient if such a patient starts complaining about a recurrence of his or her sleep disturbance (17).

Objective Findings

Many studies have objectively documented electroen-cephalographic (EEG) changes associated with MDD. Most studies show changes in the timing or distribution of sleep stages, particularly slow-wave sleep and REM sleep. A symptomatic but unmedicated adult with MDD typically shows a prolonged sleep-onset latency (SOL), bouts of intermittent wakefulness, increased N1 sleep, and reduced N3 sleep (5). The N3 deficits are greatest in the first NREM sleep period, whereas the amount of
N3 sleep is normal or even increased during the second NREM period. Even more profound and more consistent changes are seen in REM sleep parameters. Nonmedicated patients show a substantially shortened latency to the first REM period (REM latency). REM latency in depressed patients is often on the order of 30 to 50 minutes compared with the typical 90 to 100 minutes seen in nondepressed age- and sex-matched controls. There is also a significant increase in REM density (the number of eye movements per unit of time). Some studies show that the first REM period is longer and may have the highest REM density compared with later REM periods (18). This is just the opposite of normal sleepers, who have longer duration and more intense REM episodes later on in the sleep period. These sleep changes are not as robust in depressed children and adolescents (19, 20) as they are in nonmedicated adults with MDD. In fact, younger groups with depression are very similar to non-depressed individuals on most sleep variables.

The severity of the depression may or may not correlate with the severity of the sleep disturbance (21, 22). There is good evidence that disturbed sleep, particularly a shortened REM latency, may be a trait marker (show a propensity for) rather than a state marker (actively depressed). First-degree relatives of depressed individuals also commonly have short REM latencies (23). The abnormal REM changes may also be observed during remission of depression. These findings suggest that a shortened REM latency is associated with an increased risk of developing depression, as opposed to actively manifesting the disorder.

As previously noted, most antidepressant medication will influence REM sleep by lengthening the REM latency, decreasing REM density, and decreasing the REM percentage, particularly during the first third of the night. As stated previously, the mechanism for the antidepressant action was thought to be through REM suppression and increased REM sleep pressure. However, a number of effective antidepressant medications (e.g., bupropion, nefazodone, and trazodone) do not reduce REM sleep, which suggests another potential mechanism of action.

Cognitive behavioral therapy (CBT) has been shown to decrease REM density, but has no effect on REM latency (24). It is suggested that trait-dependent variables, such as shortened REM latency, decreased N3 sleep ratio, and decreased percentage of N3 sleep, remain stable over time; state-dependent measures, such as REM density and sleep efficiency, may improve with CBT treatment. However, a small percentage of subjects in remission have persistent sleep state-dependent abnormalities after CBT (25).

Bipolar Disorder

Bipolar disorder is characterized by the alternating pattern of two emotional extremes: depression and euphoria. The individual with bipolar illness cycles between periods of potentially severe depression and mania, an agitated and often elated emotional state. In addition to the aforementioned symptoms of depression, manic episodes may be characterized by irritable mood, thoughts of grandiosity, rapid or pressured speech, distractibility, and compulsive behavior (e.g., shopping sprees, drinking, sexual activity) (2).

Bipolar disorder is much less common than unipolar depression and has a lifetime prevalence rate of 2.8% to 6.5% (26). However, if the diagnostic criteria stipulate hypomania and not full mania (bipolar II), the incidence could include about one-half of currently diagnosed unipolar depression patients. There is no gender preference for bipolar disorder I, but bipolar II may affect more women than men (27). There may be a seasonal pattern for the occurrence of depressive and manic/hypomanic episodes. The onset of depression in bipolar disorder tends to be in the fall to winter, whereas mania tends to peak in the spring or fall (28).

Subjective Findings

The literature on sleep and bipolar disorder is not as complete as with unipolar depression. Subjective reports indicate that bipolar patients have concerns about their sleep much like unipolar depressives. In one study, 70% of a bipolar patient population had significant sleep concerns, including impaired sleep efficiency, higher levels of anxiety, fear about poor sleep quality, lowered daytime activity levels, and a tendency to misperceive sleep (29). One major difference between unipolar and bipolar disorder is a stronger tendency for bipolar patients to report symptoms of hypersomnia, with extended nocturnal sleep periods, difficulty awakening, and excessive daytime sleepiness, during the depressed phase (30). However, when the level of hypersomnia was objectively tested using the multiple sleep latency test, there was no increased physiologic sleep propensity demonstrated in the bipolar group. The authors stated that the reported hypersomnia might reflect lack of interest, social withdrawal, decreased energy, or psychomotor retardation, rather than a true increase in sleep propensity (31). Another notable subjective difference between bipolar disorder and MDD is that during the manic phase, although sleep may be severely curtailed, the patients have much less concern over their sleep. They feel that they do not need as much sleep (32). Prodromal episodes of insomnia tend to be more common in mania than depression and the shift to mania may be exacerbated by periods of sleeplessness (33).

Objective Findings

Objective data on sleep during mania indeed find decreased TST, increased time awake in the last 2 hours of the study, a short REM latency, and increased REM
density. These findings were reported to be similar to the sleep changes seen in MDD (34, 35).

Manic episodes are often preceded by periods of sleeplessness, and case reports have suggested that mania can be triggered by sleep deprivation (36). It was speculated that the curtailed sleep time associated with mania may be potentially self-reinforcing for the clinical aspects of the disorder (37). However, it appears that sleep deprivation may have the same beneficial effects on bipolar disorder as it does on MDD patients (38). There is some concern that sleep deprivation can precipitate a manic episode. However, the rate of switching from depression to mania following sleep deprivation is no more pronounced than that observed using antidepressant medication (39).

Bipolar type II disorder is characterized by one or more depressive episodes accompanied by at least one hypomanic episode. In addition, there are no psychotic features such as delusions or hallucinations that can occur with bipolar I disorder. It should be noted that bipolar II is not a milder form of bipolar disorder. The depressive phase can be as severe in either disorder. As such, the increased sleep time seen in the depressive phase of either disorder can be similar. The manic phase is less pronounced in type II, but there is still a reduced need for sleep in over 80% of patients (40).

Seasonal Affective Disorder

Seasonal affective disorder (SAD) was first proposed in 1984 and is currently used as a specifier of either bipolar or recurrent MDD. The symptoms usually present during winter and remit during the spring. Symptoms also improve when the patient is exposed to daylight or bright light therapy. Although patients report depression, the other signs of SAD are quite different from those of MDD. SAD patients typically report increased appetite and increased need for sleep. The disorder is more common in young adults, particularly in women (41).

Disturbances of sleep are a primary component of SAD as they are of other mood disorders. However, SAD patients most often report hypersomnia. Among responses of 293 SAD patients on a symptom questionnaire, 95% acknowledged sleep concerns. The vast majority (80%) complained of winter hypersomnia, with only 10% reporting insomnia and 5% reporting insomnia and hypersomnia. TST increased by about 2.7 hours during the fall/winter months compared with the spring/summer months. Increasing sleep length by up to 2 hours during the winter months occurs in about one-half of a random sample of the general population. Thus, this symptom alone is not necessarily indicative of a disorder (42).

In contrast to the subjective impressions of increased sleep, actual sleep measurement of depressed SAD patients in winter showed reduced sleep efficiency, decreased N3 sleep percentage, and increased REM density, but normal REM latency. These findings were consistent when the patients were compared with themselves in the summer, compared with themselves following 9 days of bright light therapy, and compared with age- and gender-matched healthy controls. Although there may be a natural tendency to increase TST in the winter months, SAD patients show sleep architecture changes that are different from normal controls (42). Bright light therapy, especially morning light administration, appears to be an efficacious and safe treatment for this disorder (43). Bright light therapy is now being studied in more traditional forms of depression. There is some concern that light therapy can precipitate hypomanic or manic episodes, so caution is advised in using bright light therapy with bipolar patients (44).

Effect of Antidepressant Medications

A full review of medication and sleep appears elsewhere in this textbook. However, any discussion of psychiatric disorders and sleep must include some comments about the psychotropic medications. Generally, the antidepressant medications are REM-suppressing drugs. They tend to delay the onset of REM sleep and may decrease both the amount and the percentage of REM sleep. The REM suppression is most pronounced early in the treatment and tends to diminish with the long-term treatment (9). However, even in patients using these medications long term, a short REM latency is typically not seen. Although many of the older tricyclic antidepressants (e.g., amitriptyline, imipramine, clomipramine) are sedating, most of the selective serotonin reuptake inhibitor (SSRI) antidepressants (e.g., fluoxetine, paroxetine, sertraline) and other new agents (e.g., venlafaxine, bupropion) tend to be activating or alerting. As such, they may improve sleep by treating depression, but probably have little primary hypnotic efficacy, and they may actually worsen sleep (45). In addition, most antidepressants increase the incidence of periodic limb movements (PLMs) during sleep and may exacerbate restless legs syndrome (46). However, bupropion seems to be an exception, in that the drug does not exacerbate symptoms of restless legs syndrome (47, 48), nor does it suppress REM sleep like most of the other antidepressants (49). The SSRI fluoxetine can cause REMs in NREM sleep and scoring the record could be challenging if the scorer is unaware of this fact (50). Most antidepressants improve muscle tone in the upper airway and can slightly reduce snoring and the apnea-hypopnea index (AHI) in patients with obstructive sleep apnea syndrome (OSAS) (51, 52). This effect has been most studied with the antidepressants fluoxetine and protriptyline. In addition to increasing
muscle tone, the REM-suppressing aspect of these drugs limits the time during sleep that is most conducive to having obstructive respiratory events.

Only gold members can continue reading. Log In or Register to continue