An Overview of Sleep Medications

Insomnia is an extremely prevalent disorder affecting up to a third of the adult population in the United States.¹⁻³ While numerous definitions exist, the DSM-IV defines insomnia as the presence of either difficulty initiating or maintaining sleep or nonrestorative sleep.⁴ Also accompanying this disruption in sleep is clinically significant distress or impairment of functioning. Insomnia can be either acute or chronic. It can also be classified as primary, or secondary to a wide variety of psychological or medical conditions.

Insomnia is frequently comorbid with pain. Patients with chronic painful conditions are more likely than the general population to experience insomnia.⁵˒⁶ The relationship is also reciprocal with patients having insomnia reporting higher pain scores than patients without insomnia.⁶ Treatment of insomnia in the presence of pain is thus multifactorial. Both must be treated simultaneously for optimal outcomes. Unfortunately, just as in the treatment of pain, there are numerous treatment options for insomnia but no universal best choice. Choices must be tailored to the individual patient. This paper will review the pharmacologic options for treatment of insomnia. This is not to imply that nonpharmacologic methods are ineffective. Frequently, behavioral modification strategies are the optimal choice in the chronic pain patient, due to the presence of drug interactions and the development of tolerance with many pharmacologic choices. However, they do require motivated patients and their effects may not be observed for several nights.

There are a number of pharmacologic agents available to treat insomnia. Among drugs approved by the Food and Drug Administration (FDA) for insomnia are the benzodiazepines (e.g. temazepam, triazolam), the non-benzodiazepine GABA agonists (e.g. zolpidem, zaleplon), and the melatonin agonist, ramelteon (see Table 1). Several antidepressants have been used off-label for their hypnotic properties (e.g. amitriptyline, trazodone, mirtazapine; see Table 2). Finally, over-the counter options include the antihistamines diphenhydramine and doxylamine (see Table 3). While anticonvulsants and atypical antipsychotics are occasionally used for insomnia, this is not commonplace, is not supported in the literature, and will not be discussed.

Benzodiazepines

Essentially all benzodiazepines share hypnotic properties, and can be used as sleep aids. However, only flurazepam (Dalmane®), estazolam (Prosom®), quazepam (Doral®), temazepam (Restoril®), and triazolam (Halcion®) are formally indicated for treatment of insomnia. All of these agents are available generically. The primary distinction among the benzodiazepines is their half-life. Short half-life agents, such as triazolam, have a rapid onset of effect and little hangover sedation the next day. However, they may not provide sleep throughout the night. Conversely, long half-life agents (flurazepam, estazolam, quazepam) increase sleep maintenance, but may take some time to induce sleep and have a high incidence of adverse cognitive effects the next day. Overall, total sleep duration appears to increase with the use of benzodiazepines.⁷˒⁸

Benzodiazepines are controlled substances, specifically C-IV. Thus, they do possess the potential for dependence and addiction. However, they are less problematic than older sleep aids such as barbiturates. Nonetheless, they should be used with great caution in patients with a history of substance abuse, including alcohol abuse. Development of tolerance to the benzodiazepines— when used as directed for sleep—is controversial. While it has been reported, there are no long term trials that would clarify the incidence.⁷ Regardless of duration of use, rebound insomnia on discontinuation has been reported, with greater effects at greater doses. Withdrawal symptoms have also been noted on discontinuation. Tapering is recommended, especially in patients receiving high doses and/or long duration of therapy.

Benzodiazepines have demonstrated efficacy in patients with chronic pain.⁹ Measures of both sleep and pain improved in rheumatoid arthritis patients when treated with triazolam. However, caution must be exercised when used alongside certain other pain medications. In particular, concurrent use of opiates and benzodiazepines can potentiate not only sedation, but also respiratory depression. Respiratory depression with benzodiazepines may also be problematic in patients with obstructive sleep apnea.

Non-benzodiazepine GABA agonists

Compared to the benzodiazepines, the non-benzodiazepines have a similar affinity to the a1 subunit of the GABAA receptor, but much lower affinities for the other a subunits. Thus, they retain similar hypnotic effects, but lack the muscle relaxant and anticonvulsant properties of the benzodiazepines. Unfortunately, they also appear to retain similar potential for abuse and dependence.¹⁰ All carry a C-IV designation. Like the benzodiazepines, they can be lethal in overdose, but that is very rare unless used in conjunction with ethanol. The agents having FDA approval are zolpidem (Ambien®, Ambien CR®), zaleplon (Sonata®), and eszopiclone (Lunesta®). A fourth agent, indiplon, is under consideration by the FDA. As a group, these are the most prescribed sleep aids in the United States.

Zolpidem, in its immediate release form is the only non-benzodiazepine hypnotic available generically. With its relatively short half-life (~2.5 hours), it has clearly demonstrated efficacy in reducing sleep latency. It is less clear whether the immediate release form affects sleep maintenance. Short term studies are generally positive, but longer duration trials have often found a waning of effect on sleep maintenance. The extended release form is effective for sleep maintenance, but may have a higher incidence of somnolence the next day. Studies of long-term dosing of extended release zolpidem have not been published. Both forms have the potential for rebound insomnia.

Zaleplon has the shortest half-life of the non-benzodiazepine hypnotics: approximately 1 hour. Thus, it has an effect on sleep latency, but unfortunately little, if any, effect on total sleep duration. Its effects are so short acting that it can be administered with as few as four hours available sleep time, and still have minimal cognitive effects the next day. At recommended doses, there is no evidence of either rebound insomnia or withdrawal.

Eszopiclone has a longer half-life than other non-benzodiazepine hypnotics (~6 hours) and is thus beneficial for improving both sleep latency and sleep maintenance. It was also the first to report data from long-term trials, and so is approved by the FDA without limitations on duration of use. The long-term trials also suggested that tolerance did not develop even over a year of treatment.¹¹ As one would expect, the longer duration of effect also correlates with an increase in the risk of next day sedation.

Both zolpidem and eszopiclone have demonstrated efficacy in patients with chronic painful conditions. Zolpidem improved sleep measures in fibromyalgia patients, though it did not alter pain scores.¹² Eszopiclone improved both sleep measures and pain scores in rheumatoid arthritis patients.¹³ As with all benzodiazepines, caution must be observed when using opiates concomitantly.

Melatonin Agonists

Melatonin, which is available over the counter, has been used for many years as a sleep aid. However, controlled clinical trials have generally failed to demonstrate clinically significant effects on objective sleep parameters. The primary exception to this is in patients with disrupted sleep-wake cycles (e.g. shift workers, jet lag). Ramelteon (Rozerem®) on the other hand, has demonstrated decreased sleep latency and increased total sleep time.¹⁴ It is not a controlled substance, and has no stated limitations on duration of treatment. Rebound insomnia and withdrawal have not been observed. However, there are no head-to-head trials comparing the efficacy of ramelteon to either benzodiazepine or non-benzodiazepine hypnotics. Likewise, there is no published evidence in patients with comorbidities such as depression or chronic pain. Thus, its overall place in therapy remains to be determined.

Antidepressants

Although no antidepressant carries a specific indication for insomnia, several have been utilized as sleep aids. In particular, amitriptyline (Elavil®) and trazodone (Desyrel®) have seen wide use, with mirtazapine (Remeron®) gaining in popularity more recently. All three are not controlled substances and are available generically. Despite their frequent use—predominantly in patients with comorbid depression—clinical evidence supporting the benefits of antidepressants as hypnotic agents is scant.¹⁵˒¹⁶ Thus, it is difficult to determine if long-term benefits in sleep are a direct effect of the drug,or secondary to improvements in the underlying depression that disrupts sleep.

In general, antidepressants are associated with shortened sleep latency and increased subjective sleep duration. However, they tend to have relatively high incidences of next-day sedation. Also, while conflicting, there are data suggesting a fairly rapid development of tolerance to the sedating effects of the antidepressants.¹⁷ Side effects are common with these medications. Notable adverse reactions include anticholinergic effects and cardiovascular effects (orthostasis and tachycardia in particular) with amitriptyline; orthostasis, blurred vision, significant weight gain with mirtazapine; and, rarely, priapism with trazodone.

Antidepressants can be very useful in patients with comorbid pain. Amitriptyline is one of the most recommended agents in the treatment of neuropathic pain, usually at doses similar to those used for insomnia. Mirtazapine also has demonstrated some benefits in pain reduction.¹⁸ However, as in the insomnia trials, most patients had comorbid depression. In contrast to the others, trazodone has largely failed to provide benefit in patients with chronic pain.¹⁹

Antihistamines

Essentially all first generation antihistamines can be described as sedating. Of these, diphenhydramine (Benadryl) is the one most commonly utilized as a sleep aid. It can reduce sleep latency and lengthen sleep duration. However, tolerance may develop quite rapidly.²⁰ Sedation and cognitive impairment the following day appears common, especially in the elderly. Finally, anticholinergic side effects can be pronounced. Constipation can be quite problematic in patients receiving concomitant opiates and diphenhydramine.

Conclusion

The ideal hypnotic does not exist, especially for patients with chronic pain. In a given patient, one must attempt to maximize efficacy with regard to onset and duration of sleep, as well as potential benefits for comorbidities, while minimizing tolerance, side effects, potential for abuse, cost, and drug interactions. That being stated, judicious use of hypnotic agents can significantly improve patients’ quality of life and should not inherently be avoided.