Sleep Disorders

Spring 1999
Volume 10, Number 1

 Previous Article  

Introduction. Thomas Edison, unknowingly, created one of the largest business segments in the United States. It was not his impact on the electrical industry, but on expenditures for sleep products. His discovery has resulted in more shift workers, a societal phase delay in our sleep patterns, and chronic sleep deprivation in many. In 1990, Americans spent over $78 000 000 on over-the-counter sleep preparations.1 Physicians in 1992 wrote over 17 000 000 prescriptions to promote better sleep.2

The pharmacotherapy for insomnia began in the 1860’s with chloral hydrate. This was followed by bromide in 1870. In 1960, the modern era of therapy began with benzodiazepine sedative hypnotics. The earlier sedatives and barbiturates are now all but abandoned.3 The use of sleeping agents goes up with patient’s age. This parallels the complaint of insomnia in the elderly. Other sleep disorders increase in the elderly and treating the primary sleep disorder, as opposed to the symptom of insomnia, is indicated.

At the recent American Sleep Disorders Association (ASDA) meeting in New Orleans (1998), much debate was centered on the nosology of insomnia. It is felt that the term insomnia is too vague for medical usage. Another lexicon is needed to communicate more effectively the medical condition and its implications. For now, the treating physician is to remember that insomnia is only a symptom. The pharmacotherapy of this complaint can be done at the level of pure symptomatic management or it can be done at the level of specific diagnosis and treatment. Each approach may be correct when knowingly applied in each individual patient.

Melatonin. The properties of an ideal sedative hypnotic are listed in Table 1. Melatonin has had its proponents that it fits this profile. It indeed is a powerful physiological agent whose long-term effects and adverse reactions remain unknown. As the “tryptophan-myositis” experience has taught us, we need to be extremely cautious when using any molecule, even if on the surface, it appears to be safe. 

Melatonin is synthesized in the pineal gland.4 The pineal gland is located in the center of the brain just posterior to the third ventricle. Melatonin is synthesized from tryptophan via a serotonin intermediary. The pineal glad synthesizes melatonin in response to changes in light. It can be thought of as a chemo-photo-transducer in this regard. When light hits the retina, it activates the suprachiasmatic nucleus of the hypothalamus. This then sends a noradrenergic signal, via the sympathetic nervous system, to the pineal gland which responds by the synthesis of melatonin. Melatonin synthesis begins with decrease in the photo-input to the gland. The dim light melatonin onset (DLMO) is a standard marker for circadian patterns in humans. Onset and production of melatonin is sustained throughout the night. It peaks from 2:00 AM to 4:00 AM and decreases slowly in the morning. Newer melatonin preparations for treatment of sleep disorders may focus on creating this physiologic profile of release.

The concentration of the agent reaches a maximum of approximately 60 picogram/ml (pg/ml) in the dark. During the day it is sustained at 10 pg/ml. An office fluorescent light is enough to inhibit melatonin production to daytime levels and this may be one of several mechanisms that creates problems for night shift workers.4 Melatonin does not appear to be synthesized until the third month of life. The timing of its synthesis with myelination of the central nervous system correlates well with the development of sleep/wake patterns in babies. Melatonin is hydroxylated in the liver and excreted in the urine after conjugation as 6-sulfatoxymelatonin.

Oral doses given in clinical studies or in over-the-counter preparations are usually 1 to 5 mg. This results in a Cmax of approximately 50 times the sustained release level. The Tmax for such preparations is one hour. Return to baseline is 4 to 8 hours. This profile does not mimic, at this time, the physiologic sustained response of the gland.4, 5

Physiologic effects of melatonin are mediated through second messengers. Melatonin receptors work primarily through G-proteins. There are also intracytoplasmic and nuclear receptors. It appears that its hypnotic effect is independent of its circadian effect. Its circadian effect leads to lowering of body temperature which promotes sleep. However, its hypnotic effects can occur independent of any thermal effects. Decreased sleep latency, increased sleep efficiency, increased REM percentages, can all occur in a dose dependent fashion.4, 5

The clinical and systemic effects of melatonin are many. It is a free radical scavenger and an immune activator. These two properties have led to its label as an “anti-aging” medication. It has been promoted as an agent to enhance sexual behavior. This is curious since it, paradoxically, is an inhibitor of the hypothalamic-pituitary-gonadol axis.

It has been found to shift the endogenous circadian phase.6 This has led to its use clinically in disorders such as jet lag and the disorder of shift work. Doses used in many of these studies are 5 to 10 mg. One open-label study from Italy tested 14 physicians attending a conference in Hong Kong.7 They were dosed with 5 mg the day of departure and for 2 days upon arrival in Hong Kong. All subjects had a positive response. A study of 52 international cabin crew members compared 5 mg versus a placebo and found less residual sleepiness and faster recovery times to the new time zone in treated subjects.8 Nurses given 0.5 mg on a rotating shift had better mood and performance versus placebo. Other studies have reproduced these results. Doses are usually given 30 minutes prior to or at bedtime.

Melatonin has a role in transient (days-months) circadian disorders, as well as transient insomnia. Its role as a chronic, therapeutic agent is unknown. It may be safer in the elderly where it might be a replacement therapy. Patients with chronic exposure may be running the risk of developing withdrawal states or abnormal suppressive effects from taking this exogenous hormone.

Sedative hypnotics. The modern age of pharmacotherapy began in the 1960’s with the introduction of flurazepam.2 Since that time there has been the development of 4 other benzodiazepine hypnotics and one nonbenzodiazepine hypnotic (Table 2). The benzodiazepine hypnotics work primarily by binding to GabaA receptors. These are diffusely located throughout the nervous system. The effects are felt to be mediated primarily by postsynaptic receptors. Activation of these receptors increase chloride influx resulting in hyperpolarization of neurons. Zolpidem (Ambien) appears to be more specific in terms of its Gaba receptor physiology. Benzodiazepines bind nonspecifically to all 3 subclasses of the receptor, whereas zolpidem binds specifically to one subclass.

The polysomnographic effects of the benzodiazepines include decrease in slow wave sleep.9 There does not appear to be any change in the REM sleep. They clearly can exacerbate pre-existing sleep apnea and it is debated whether they can induce new apnea in at risk patients. Barbiturates and most antidepressants on the other hand, suppress REM sleep. Ethanol suppresses many aspects of normal sleep physiology and can exacerbate sleep apnea.

The key pharmacokinetic parameters of the major hypnotics are summarized in Table 2. Onset of action is fairly consistent across the agents. Patients should be instructed to not delay sleep onset after taking these agents. Patients who remain up and active, especially the elderly, are more prone to develop injury or cognitive dysfunction prior to falling asleep. The clinical duration of action should be kept in mind depending on whether one is trying to promote sleep onset, sleep maintenance, or suppress early awakenings. The elimination of half-life of the agents and their metabolites usually extends for prolonged periods of time after the clinical effect is seen. This may be responsible for some of the problems of these agents. The main side effects of the hypnotics include residual sedation, anterograde amnesia, and rebound insomnia.10

Residual sedation clearly is a function of dose and increasing half-life. High doses of short half-life agents can indeed produce residual sedation. There is less of this with the benzodiazepines than the older barbiturates. Residual sedation can continue despite behavioral tolerance that may develop. Anterograde amnesia can be seen with any sedative hypnotic, both benzodiazepine and nonbenzodiazepine. The amnesia is felt to be a function of sleep onset latency. If wakeful activity continues after pharmacologic onset, then behavioral amnesia will develop. It has been argued that anterograde amnesia from these agents may really represent retrograde amnesia due to acute sleep onset. Rebound insomnia is described as 1 to 2 nights of poor sleep after acute discontinuation. It is characterized by a delay in sleep onset and poor sleep efficiency.10 It has been described in all sedative hypnotics, both short and intermediate acting. It may be also directly a function of dose. Duration of treatment may also contribute to rebound insomnia. Zolpidem appears to have an advantage in that testing of this agent over 6 to 8 weeks did not result in as much rebound insomnia as other agents.11 However, these major side effects have not been properly assessed in dose equivalent, head-to-head comparisons among agents.

Antidepressants. Classic teaching states that depression and REM sleep are intimately tied. There appears to be decreased REM latency in those who are depressed. This may be due to increased REM pressure or a circadian shift. Antidepressants are, for the most part, REM suppressants. Acute REM sleep deprivation in sleep lab environments can decrease depressive symptomatology. Such a therapeutic approach has its limitations. With the advent of new psychopharmacology, this relationship with REM does not appear to be absolute. For example, bupropion (Wellbutrin) and nefazodone (Serzone) may not change, or may actually increase, REM sleep on polysomnography.12

The mechanism of improved sleep is related to the antidepressants’ ability to block adrenergic and serotonergic transmission. Blocking noradrenergic transmission appears to be very critical in regulating polysomnographic findings and in normalizing REM/nonREM interactions.12, 13 Although antidepressants are not approved for use solely for “insomnia,” they are extensively prescribed and have good clinical efficacy for sleep disorder patients. Antidepressant effects on sleep and polysomnography have not been directly compared in an appropriate study.

Nevertheless, sleep complaints, primarily insomnia, are commonly seen in association with other depressive symptoms or as co-morbidities. They can clearly cause a positive feedback loop. There is usually a delay in the direct antidepressant effects of these medications. Sedating effects may be perceived immediately by the patient prior to direct antidepressant effects. Drugs with such sedating effects may be more helpful in the management of patients presenting primarily with sleep symptomatology (Table 3).

The use of antidepressant medications requires appropriate knowledge of their side effects.13, 14 This is particularly true in the elderly. Cardiovascular disease is an important risk factor, especially when using the tricyclics. Anticholinergic side effects can include exacerbation of urinary retention, orthostatic hypotension, and arrhythmias. Rarely, antidepressants can precipitate periodic limb movements which can paradoxically exacerbate insomnia.

Many of the over-the-counter preparations for insomnia contain antihistamines. Major components include hydroxyzine and diphenhydramine. These also can have anticholinergic side effects and result in residual sedation. The use of such over-the-counter medications should be discouraged in patients who are actively undergoing treatment — be it with nonpharmacologic therapies, pharmacologic therapies, or both.

Conclusion. Both acute and short-term use and chronic long-term use, may be appropriate for sleep-promoting agents in any one patient. This holds true for melatonin, antidepressants, and certain sedative-hypnotic agents. It is important that appropriate patient selection be made. Diagnosing underlying medical and sleep disorders contributing to insomnia are crucial. All patients with any sleep complaint need appropriate behavioral interventions and sleep hygiene techniques.

Address comments and questions to:

Ronald E Kramer, MD
CNI Sleep Disorders Center
701 East Hampden Avenue, Suite 530
Englewood, CO 80113

 Previous Article