Pharmaceutical Treatment of Insomnia
, Volume , Number ,
Insomnia is one of the most common medical complaints, and there are a number of medications to help treat it. The ideal sedative-hypnotic agent would do it all—reduce the time it takes for an individual to get to sleep, improve sleep maintenance, enhance daytime functioning, have no effect on or improve sleep architecture, and be safe in the elderly while having no residual next day effects or abuse potential. (Sleep architecture describes the structure and pattern of sleep, including the cyclical pattern as it shifts between the different sleep stages.) Unfortunately, to date, no sedative-hypnotic meets all of these criteria. In fact, in long-term treatment, the most effective tool for insomnia is not a medication at all; it is cognitive behavioral therapy (Mitchell MD et al, BMC Fam Pract 2012;13–40). Therefore, whenever possible, clinicians should use non-pharmacologic treatments such as relaxation techniques, stimulus control, and cognitive behavioral therapy (Schutte-Rodin S et al, J Clin Sleep Med 2008;4(5):487–504).
However, for those patients requiring a sedative-hypnotic, treatments range from over-the-counter (OTC) sleep aids and “natural” treatments such as melatonin, valerian, and chamomile to prescription medications. While natural treatments tend to be safe, they are often of limited use. OTC and prescription medications can be effective but are often saddled with problematic side effects such as increased falls, next day sedation, and cognitive impairment (Meoli A et al, J Clin Sleep Med 2005;1(2):173–187). (See the table “FDA-Approved Sedative-Hypnotic Medications” for a review of commonly used sleep agents.)
Pharmacologic treatment of insomnia often leads to small improvements in sleep initiation and total sleep time. However, data supporting improvements in other domains, such as cognition, are limited and are reviewed where applicable in this article. In addition, it is important to warn patients not to drive while taking these medications as they can impair coordination and reaction time.
Let’s take a look at some of the medications used to treat insomnia.
Over-the-Counter Sleep Aids
OTC sleep aids consist of sedating antihistamines. A related agent, hydroxyzine (Vistaril), is available by prescription only. Although widely used, OTC sleep aids are not well studied (Schutte-Rodin op cit). In addition they often lead to concerning side effects such as next-day sedation, falls, and anticholinergic side effects that can include dry mouth, constipation, and delirium (Ramakrishnan K and Scheid DC, Am Fam Physician 2007;76(4):517–526).
Given the lack of data regarding these agents, they are not recommended by clinical guidelines for the treatment of insomnia (Schutte-Rodin op cit).
Prescription Sleep Aids
Prescription sleep aids consists primarily of benzodiazepines, the related non-benzodiazepine hypnotics or Z-drugs, the melatonin agonist ramelteon (Rozerem), and non-hypnotic medications with sedating side effects such as antidepressants and antipsychotics.
All benzodiazepines modulate gamma-aminobutyric acid-A (GABA-A) and some are indicated for the short-term treatment of insomnia (Ramakrishnan op cit). Since they all work via the same mechanism, their primary differences are related to their pharmacokinetic properties and FDA-approved indications.
Of the benzodiazepines FDA-approved for the treatment of insomnia, only triazolam (Halcion) has a short half-life, a measure of how long it takes for half of the drug to be eliminated from the bloodstream. Triazolam can be associated with rebound anxiety and is not generally recommended as a first-line hypnotic (Schutte-Rodin op cit).
The remaining agents are all associated with next day sedation and/or cognitive impairment (Roehrs T and Roth T, Neurotherapeutics 2012;9(4):728–738). However, when used in high doses or with other central nervous system (CNS) depressants, even short-acting agents can lead to residual symptoms the next day.
In general, benzodiazepines lead to a dose-related decrease in the time it takes an individual to get to sleep and an increase in total sleep time (Schutte-Rodin op cit). However, they are plagued by side effects including memory impairment, next day sedation, and tolerance to sleep-promoting effects. In addition, they are subject to dependence and abuse (Holbrook AM et al, CMAJ 2000;162(2):225–233).
Therefore, benzodiazepines are no longer the most commonly used medications for insomnia. When they are used, the prescriber should choose an appropriate agent based on the clinical presentation of the patient and the pharmacokinetic profile of the benzodiazepine. For example, an intermediate acting benzodiazepine, such as estazolam (which was previously available under the brand name ProSom), would be preferred for those with early morning awakenings.
Z-drugs are a group of non-benzodiazepine hypnotics, so-called because most of their generic names start with the letter “Z”. They are more selective for receptors in brain regions affecting sleep, wakefulness, and sedation (Morin AK, Am J Manag Care 2006;12(8 Suppl):S230– 245). This accounts for lower abuse potential of Z-drugs compared with traditional benzodiazepines. The drugs include zolpidem (Ambien, Edluar, Zolpimist, Intermezzo) and zaleplon (Sonata). Eszopiclone (Lunesta) appears to have improved efficacy in insomnia for patients with depression and anxiety (Fava M et al, J Clin Psychiatry 2011;72(4):473–479). Zolpidem has been shown effective for insomnia in patients with depression, but has not had beneficial effects for depressive symptoms (Fava M et al, J Clin Psychiatry 2011;72(7):914–928).
In a large meta-analysis of all studies submitted to the FDA, Z-drugs had only limited benefit for insomnia with a mean drug placebo difference for sleep latency of 22 minutes on polysomnography and seven minutes by subjective report. While this result is of dubious clinical benefit, the authors do report a combined drug and placebo effect of 42 minutes decreased sleep latency, suggesting Z-drugs’ effects may be enhanced by the placebo effect to achieve clinically significant benefits (Huedo-Medina TB et al, BMJ 2012;345:e8343). In addition, clinical guidelines support the use of Z-drugs for patients with primary and possibly secondary insomnia (Schutte-Rodin op cit).
Although, DSM-5, scheduled for publication in May 2013, eliminates the distinction between primary and secondary insomnia, those terms are used in this article to reflect findings from research that has, for many years, incorporated a primary/secondary distinction in the study of insomnia.
Ramelteon is one of two FDA-approved prescription sleep medications not “scheduled” by the US Drug Enforcement Administration as a controlled substance. It is indicated for insomnia characterized by difficulty initiating sleep. It is generally well tolerated with somnolence, dizziness, nausea, fatigue, and headache being the most common adverse events. Most studies demonstrated improvement in total sleep time, while data were mixed for subjective sleep latency. It should be taken 30 minutes prior to bedtime. It is recommended to not be taken with or immediately following a high fat meal. Ramelteon is metabolized via CYP450 1A2. Strong inhibitors of 1A2, such as fluvoxamine (Luvox) result in a 70-fold increase in ramelteon’s plasma concentrations. Therefore, ramelteon should not be used in conjunction with strong inhibitors of 1A2. Inhibitors of 3A4, such as ketoconazole (eg, Nizoral) and 2C9, such as fluconazole (Diflucan) increase ramelteon concentrations to a lesser amount. Patients on inhibitors of these enzymes can receive ramelteon with careful observation (Simpson D and Curran MP, Drugs 2008;68(13):1901–1919).
Although there are limited data supporting the use of antidepressants in insomnia, many sedating antidepressants are prescribed for the treatment of insomnia. Some studies indicate that off-label use of antidepressants for insomnia may exceed that of FDA-approved hypnotics, with trazodone being prescribed 32% more often than the most widely prescribed FDA-approved hypnotic (McCall C and McCall WV, Curr Psychiatry Rep 2012;14(5):494–502).
This is of concern, as few studies have assessed antidepressants for primary insomnia and the majority of antidepressant insomnia studies focus on insomnia associated with depression (Mendelson WB, J Clin Psychiatry 2005;66(4):469–476).
There is also little data supporting the use of amitriptyline in primary insomnia. At this time, use of amitriptyline as a hypnotic agent is not supported or refuted by the medical literature. Therefore, clinicians should consider alternate treatments.
Doxepin (Silenor) is the only antidepressant specifically approved for the treatment of insomnia. At doses of 3 mg to 6 mg, doxepin is reported to function as an antihistaminic medication (Markov D and Doghramji K, Curr Psychiatry 2010;9(10):67–77).
At this low dose, doxepin has been shown to decrease early morning awakenings and improve sleep maintenance with a side effect profile similar to placebo. It has been shown to be safe and effective for the treatment of insomnia in adult and geriatric populations (Silenor [package insert]. San Diego, CA: Somaxon Pharmaceuticals, Inc; 2010).
As only one of two (along with ramelteon) non-scheduled FDA-approved hypnotics, doxepin may be particularly useful in those individuals with a history of substance abuse or intolerance to GABA-ergic hypnotics.
Mirtazapine (Remeron) has not been studied for the treatment of primary insomnia. In addition, its delayed peak (two hours) and long half-life (20 to 40 hours) make it undesirable as a hypnotic (Remeron [package insert]. Whitehouse Station, NJ: Merck and Company, Inc; 2012). This is borne out in studies in which mirtazapine has been associated with residual next day impairments in cognition and driving abilities with acute use. As these impairments are no longer present with chronic use, it makes the use of mirtazapine on an as needed basis particularly worrisome (Wingen M et al, J Clin Psychiatry 2005;66(4):436–443). Therefore, use of this agent for primary insomnia should be discouraged. Longer term studies have found improvements in next day functioning in patients with depression, and it may have a role for patients with depression and insomnia (Shen J et al, J Clin Psychiatry 2009;70(3):370–377).
Few studies have examined the use of trazodone for primary insomnia. In one small study, trazodone was reported to decrease awakenings, stage 1 sleep, and patient subjective complaints of insomnia (Mendelson W, J Clin Psychiatry 2005;66(4):469–476). Other studies have found improvements in slow-wave sleep which may indicate improved sleep quality. It has also been shown in mostly small studies to improve subjective reports of sleep for patients with dementia, depression, and fibromyalgia, although objective measures of sleep do not support these reports (Walsh JK et al, Hum Psychopharmacol 1998;13(3):191–198). Therefore, trazodone should be used cautiously for the treatment of primary insomnia, although it may have a role for some forms of secondary insomnia by subjective patient report.
One antipsychotic medication, quetiapine (Seroquel) is prescribed extensively for insomnia, but to date there is only one study of this drug for the treatment of primary insomnia. This study reported nonsignificant improvements in total sleep time and sleep latency (Tassniyom K et al, J Med Assoc Thai 2010;93(6):729–734). Other studies suggest some benefits of quetiapine for patients with comorbid disorders such as anxiety, depression, and autism. Benefits for sleep may be offset by increased side effects such as periodic leg movements and alterations in rapid eye movement (REM) sleep (Wine JN et al,Ann Pharmacother 2009;43(4):707–713). Therefore, treatment guidelines recommend reserving use of quetiapine for those with comorbid insomnia who may benefit from quetiapine for the treatment of another primary psychiatric disorder (Schutte-Rodin op cit).
Adverse Effects of Sleep Medications
There are a number of common adverse effects associated with medications for insomnia.
Amnesia: All benzodiazepines are associated with amnesia due to their GABA-A effects. Studies indicate amnesia can occur with other medications such as sedating antidepressants, antihistamines, and antipsychotics (Canaday BR, Pharmacotherapy 1996;16(4):687–689). Reports of global amnesia of significant periods of time prior to hypnotic use are rare.
Complex Sleep Behaviors: FDA-approved prescription sedative-hypnotics have a boxed warning regarding complex sleep behaviors. The complex behaviors are most commonly reported with zolpidem, zaleplon, and triazolam. In one case series, all patients reported episodes of amnesia of three to five hours after ingesting zolpidem despite seeming to interact normally with the environment (Poceta JS, J Clin Sleep Med 2011;7(6):632–638). The majority of cases have been reported with high dose hypnotic use and when hypnotics are ingested with other CNS depressants. Therefore, it is important to review concomitant medications and substance abuse history before prescribing hypnotics. In general, hypnotics should not be dosed above the FDA-approved dosages.
Next-Day Effects: Use of benzodiazepines are associated with motor vehicle collisions (Vermeeren A and Coenen AM, Prog Brain Res 2011;190:89–103). Zolpidem has been reported to impair driving and cognition at 9 am and 10 am following nighttime and early morning administration (Leufkens TR et al, J Sleep Res 2009;18(4):387–396).
No morning-after residual impairments on driving were noted for patients who received a single nighttime dose of triazolam (0.25 mg) or temazepam (Restoril) (20 mg). Zaleplon was associated with altered driving at 9 am but not 11 am the morning after a 7.5 mg dose. Zolpidem (10 mg) was associated with driving impairment at one, three and a half, and five hours after a dose, but had normalized by the next morning (Rapoport MJ and Banina MC, CNS Drugs 2007;21(6):503–519).
However, recent studies indicate that higher blood levels are associated with impaired next-day driving. Since women have a decreased ability to metabolize zolpidem, the FDA has recently released dosing guidelines that recommend doses of 5 mg immediate-release and 6.25 mg controlled-release tablets nightly. The new guidelines reduce the recommended doses for women by half. Although the recommended doses for men have not been changed, the new guidelines instruct clinicians to consider the lower doses for men as well (FDA Drug Safety Communication: Risk of next-morning impairment after use of insomnia drugs). You can find that guideline issued by the FDA on January 10, 2013, at http://1.usa.gov/XyP4qL.
Not Recommended for Elderly: In general, use of long-acting benzodiazepines, tricyclic antidepressants, and anticholinergics are not recommended for use in geriatric patients as these agents are associated with falls and/or delirium (Vermeeren A and Coenen AM, Prog Brain Res 2011;190:89–103). Zolpidem is associated with an increased risk of falls and hip fractures (Kolia BP et al, J Hosp Med 2013;8(1):1–6).
In conclusion, there are multiple treatment options for insomnia. However, most non-FDA approved treatments lack convincing efficacy and/or safety data for use in primary insomnia. When treating primary insomnia, use of the lowest effective dose of FDA-approved treatments may minimize risks. Clinicians should consider nonpharmacologic treatments.
TCRBH’s Take: While there is evidence that suggests that the most effective treatment for insomnia is CBT, it is clear that pharmacotherapy can play an important role. Medication can be used as an adjunct to CBT, or, in some cases, in lieu of CBT. Although medication does not address the underlying reasons for insomnia, it can, for example, be a very important short-term tool to address or prevent sleep deprivation. It is important to match each patient with the appropriate treatment. And determining the most appropriate treatment requires having a broad base of knowledge of all available treatment options and, in that context, being able to give careful consideration to the patient’s treatment preferences. See “Is CBT Plus Medication the Best Treatment for Insomnia?”