Clinical aspects

Description

The treatment of chronic insomnia is a combination of behavioral and pharmacological therapies; in fact, adding medication to behavioral therapy improves outcomes during the acute therapy phase. We will focus our discussion on pharmacological treatment of the chronic insomnia disorder.

Medications for the treatment of insomnia include FDA-approved medications indicated for insomnia, FDA-approved medications prescribed off-label for insomnia, and over-the-counter and herbal medications frequently used by patients with insomnia.

Treatment of insomnia should include both pharmacologic and nonpharmacologic strategies to address sleep onset, sleep duration, and early awakening while improving daytime function (84). According to the clinical practice guidelines published by the American Academy of Sleep Medicine (AASM) in 2017 (87) and the American College of Physicians published in May 2016 (76), all patients with chronic insomnia should receive cognitive behavioral therapy. Nonetheless, it is not helpful in patients with dementia who are unable to actively or meaningfully participate in treatment sessions (48). These guidelines recommend a stepwise approach that starts with evaluation of symptoms and comorbid conditions, followed by the initial therapy and a posterior assessment of the patient response and monitoring of long-term treatment. If there is no improvement in insomnia, factors such as life events and psychiatric disorders must be considered. Finally, the clinician should think about other comorbid diagnoses before designating the patients as having treatment-resistant insomnia and instituting pharmacotherapy (76; 10). The consensus statement of the AASM recommends that first-line medications for the treatment of chronic insomnia are the short or intermediate-acting benzodiazepines or the imidazopyridines or ramelteon (statement released before the availability of suvorexant) (87). These should be followed by sedating antidepressants, particularly if there is comorbid anxiety or depression. The next option is a combination of either benzodiazepine or imidazole or ramelteon and sedating antidepressant (87). Lastly, other sedating agents can be considered, such as anticonvulsants or antipsychotics, but only in those who stand to benefit from the primary effect of these drugs (87). Pharmacological therapy, including over-the-counter sleep aids and alcohol, is the most widely used treatment for insomnia, yet no evidence-based clinical practice guidelines have been published to date and neither of the most recent guidelines found strong evidence about the overall efficacy of pharmacotherapy in chronic insomnia (87).

We will first review the FDA-approved medications for the treatment of insomnia. These include benzodiazepines, imidazopyridines, the H1-receptor antagonist doxepin, the melatonin receptor agonist ramelteon, and the dual orexin receptor antagonist suvorexant.

Until the advent of imidazopyridines, benzodiazepines were the only hypnotics of choice for insomnia. They were clinically proven to be efficacious and had rare adverse reactions, but there were concerns regarding abuse potential and withdrawal/dependence. They exert their effects by binding to GABAA receptors throughout the brain and increasing the amount of time chloride channels are open. Benzodiazepines differ from each other mainly by speed of absorption and elimination half-life.

Table 1. Benzodiazepines Approved in the United States for the Treatment of Insomnia

Drug	Elimination half-life	Peak plasma level	Active metabolites	Indication
quazepam	39 hours	2 hours	Yes	Sleep maintenance
flurazepam	2.3 hours	20 minutes	Yes	Sleep maintenance
estazolam	10 to 15 hours	0.5 to 2 hours	No	Sleep maintenance
temazepam	8 to 15 hours	30 to 60 minutes	No	Sleep maintenance
triazolam	1.5 to 5.5 hours	15 to 30 minutes	No	Sleep initiation

All of the above benzodiazepines have been found in numerous trials to be beneficial for treating insomnia in patients 18 to 65 years old. Doses of 0.25 mg of triazolam are recommended for the treatment for sleep-onset insomnia in adults. They are not recommended in the elderly due to risk of falls, cognitive impairment, and motor vehicle accidents (87).

Zolpidem, zaleplon, and eszopiclone are hypnotics belonging to a class of medications called imidazopyridines. Like the benzodiazepines, the imidazopyridines bind to GABAA receptors, prolonging opening of the chloride channel. The imidazopyridines, however, bind with a much higher affinity to the subunits of the GABAA receptor, which are associated with somnolence rather than bind to the subunits associated with anxiolysis and enhanced mood (07).

Zolpidem is approved in its immediate-release formulation (5 and 10 mg) for the treatment of sleep onset insomnia, in its controlled-release preparation (6.25 and 12.5 mg) for the treatment of sleep onset and sleep maintenance insomnia, and in its sublingual middle of the night dose form (1.75 mg for women and 3.5 mg for men) for insomnia characterized by difficulty returning to sleep after middle-of-the-night awakening. The FDA recommends a starting dose of 5 mg for immediate-release zolpidem and 6.25 mg for extended-release forms of zolpidem (87). In its immediate-release formulation, zolpidem has a short half-life of 2.2 hours, and peak plasma levels are reached in 90 minutes. The dose needs to be adjusted in the setting of hepatic impairment but not with altered renal function. There has been a change in FDA dosing recommendations based on sex differences in daytime impairment with nighttime zolpidem (27). The recommended dose for adult women is 5 mg because women seem to metabolize the drug more slowly. In its controlled-release formulation, pharmacokinetic analysis has demonstrated that the time to maximum concentration and terminal elimination half-life of 12.5 mg zolpidem extended-release are similar to those of 10 mg zolpidem. The initial dose of sublingual zolpidem should always be adjusted in women, elderly, and debilitated patients, or in patients with a hepatic disorder. Zolpidem should be reduced or avoided in elderly patients due to case reports of hallucinations in septuagenarian and octogenarians (62). Zolpidem at the recommended doses is associated with minimal rebound insomnia and hangover symptoms of short duration (66).

Zaleplon is the shortest-acting approved hypnotic in the United States and, thus, it has fewer residual daytime effects than zolpidem (07). Zaleplon is approved for sleep initiation insomnia at 5, 10, and 20 mg doses and for sleep maintenance at 5 and 20 mg (both in immediate release and modified release), and for middle-of-the-night awakenings (if there are 4 or more hours of sleep left) at doses of 5 and 10 mg. Zaleplon is recommended for the treatment of sleep onset insomnia in adults based on trials of 10 mg doses (87). The onset of action is approximately 30 minutes, and the duration of action is about 4 hours. Peak zaleplon serum concentrations occur in about 1 hour, and its elimination half-life is also about 1 hour. Similar to zolpidem, its dose needs to be adjusted with altered hepatic function but not with renal impairment. Because of its short half-life, there is no residual sedation when zaleplon is administered in the middle of the night; hence, it is the ideal medication for sleep maintenance insomnia.

Eszopiclone is approved for long-term therapy for the treatment of sleep onset insomnia and sleep maintenance insomnia at doses of 1 to 3 mg. Eszopiclone is the longest acting of the three imidazopyridines; however, there is an FDA warning regarding next-day impairment with eszopiclone. The 3 mg dose can cause impairment to driving skills, memory, and coordination lasting greater than 11 hours after the bedtime dose. As such, the recommended starting dose is now 1 mg (04). Clearance is reduced in the elderly, and extra caution should be employed; however, eszopiclone 1 mg is likely to be safe for the healthy elderly (92). Additionally, it has been shown to be effective for treatment of insomnia for at least 12 months with no evidence of tolerance, dependence, or abuse. It has caused mild, transient memory impairment in some patients. In addition to primary insomnia, eszopiclone has also been studied in insomnia associated with depression and anxiety. When coadministered with fluoxetine, it is relatively well tolerated and associated with rapid, substantial, and sustained sleep improvement, a faster onset of antidepressant response on the basis of Clinical Global Impression (CGI). There is a new clinical trial focused on studying the effectiveness of zopiclone for treatment of insomnia in palliative care (37).

There are very few comparative efficacy studies of imidazopyridines. Eszopiclone 3 mg is as effective as zopiclone 7.5 mg in treating both subjective and polysomnographic measures of insomnia over 4 weeks of nightly usage (74). The two had similar adverse event profiles as well. No studies are available comparing eszopiclone or zolpidem to zaleplon or other older benzodiazepine-type sedative hypnotics.

The standard of practice is to use imidazopyridines intermittently for the treatment of acute periods of insomnia, and they should be reserved for use if the first-line agents are ineffective (60). There is no real consensus on the pharmacological management of chronic insomnia. One study aimed to explore the use of zolpidem as maintenance therapy (73). Seventy-four patients with chronic insomnia were treated with 10 mg zolpidem for 4 weeks and were then randomized for the next 12 weeks to dosing of 10 or 5 mg of zolpidem nightly, intermittent dosing with 10 mg 3 to 5 days weekly, or “partial reinforcement dosing” with nightly pill use where 50% were 10 mg pills of zolpidem, and 50% were placebo.

All groups maintained treatment response over the study period, suggesting a partial reinforcement strategy may be an effective maintenance treatment strategy that can limit the amount of active drug used. Imidazopyridines do not worsen sleep apnea, regardless of baseline apnea-hypopnea index (68).

In 2019, according to the clinical practice guidelines published by The American Academy of Family Physicians, deprescribing is recommended for those 65 years and older with insomnia taking a benzodiazepine receptor agonist regardless of duration. A very slow taper with a 25% dose reduction every 2 weeks should occur, and if possible, 12.5% reductions and medication-free days near the end. Options include 50% reduction and switching to lorazepam (Ativan) or oxazepam if tapering is not possible in 25% increments. On the other hand, benzodiazepine receptor agonists should be continued in patients with other sleeping disorders that may be aggravating insomnia or in cases where benzodiazepines are effective. In this setting, multidisciplinary management should be considered, the underlying condition treated, and substances that worsen insomnia such as caffeine and alcohol should be minimized (15).

Doxepin, an H1-receptor antagonist, is approved for the treatment of insomnia. It is a tricyclic antidepressant that is FDA-approved for depression in doses of 75 to 150 mg and for insomnia in doses of 3 to 6 mg. It is a much more potent and selective H1 antagonist than diphenhydramine and doxylamine, which are both H1 antagonists available over the counter and frequently used for insomnia (45). A systematic review of randomized placebo-controlled trials of doxepin found that although it has a small to medium effect on sleep maintenance and sleep duration, it is not effective for sleep initiation (104). The AASM suggests its use as possible treatment for sleep maintenance insomnia based on trials of 3 mg and 6 mg of doxepin (87).

Adverse effects of the H1 antagonists are anticholinergic effects along with weight gain; however, in the approved 3 to 6 mg dose of doxepin, anticholinergic effects and weight gain are not seen, even in older adults who tend to be more susceptible to anticholinergic effects (82). Headache and somnolence are the most common side effects (104).

Ramelteon is a melatonin receptor agonist that is approved by the FDA for the treatment of insomnia. Melatonin is an endogenous hormone produced by the pineal gland. Melatonin in varying doses is available as an over-the-counter supplement but is not FDA regulated. Ramelteon, along with doxepin, has the distinction of being one of the only FDA-approved pharmacotherapies for insomnia that is not designated by the U.S. Drug Enforcement Administration as having significant potential for abuse (64). Ramelteon is a melatonin ML1/MT1 receptor agonist that demonstrates a statistically significant reduction in latency to persistent sleep and a statistically significant increase in total sleep time with no apparent next-day residual effects in patients with chronic primary insomnia. Ramelteon is recommended treatment of sleep onset insomnia based on trials of 8 mg (87). It seems to be most beneficial for sleep onset insomnia, and its efficacy seems to be relatively modest compared with the previously reviewed hypnotics (88).

In a nationally representative cross-sectional survey of older adults in the United States, more than 20% of respondents reported over-the-counter sleep aid use (61). Several studies have shown efficacy of the over-the-counter form of melatonin in the treatment of circadian rhythm problems, particularly among shift workers (86). Also, there is some evidence that it may subjectively and objectively improve sleep in older adult patients who suffer from insomnia, especially when mixed with with magnesium, zinc, and vitamin b complex (19; 12); sleep disturbances in autism and intellectual disability (31); insomnia associated with childhood ADHD (26); and childhood-onset insomnia. In a metaanalysis of randomized controlled trials, melatonin was found to be both efficacious and tolerable in treating sleep onset insomnia in children and adolescents (100). The study suggested starting with a 1 mg per day dose and increasing as needed up to 12 mg per day. Melatonin 3 mg, given 2 hours before bedtime, has been shown effective in improving subjective sleep in cancer patients with insomnia (46). The data available, however, do not support its use as a wide-spectrum hypnotic for primary insomnia. Guidelines do not recommend melatonin as a treatment for sleep onset or sleep maintenance insomnia based on trials of 2 mg (87). In addition, no randomized placebo-controlled study has been able to demonstrate that melatonin offers a significant benefit to treat insomnia in patients with Alzheimer disease. However, a few small studies have shown benefit despite having some limitations. For instance, in an uncontrolled retrospective study, administration of melatonin 9 mg for 22 to 35 months was associated with decreasing the sundowning effect and improving subjective sleep (83).

Suvorexant, a medication approved for the treatment of insomnia, is part of a new class of medications called orexin antagonists. Suvorexant has been examined in two 3-month, randomized, double-blind, placebo-controlled, parallel-group studies of 40 and 20 mg in nonelderly adults (age < 65 years) and 30 and 15 mg in elderly patients (age ≥ 65 years). It was found to be superior to placebo in improving sleep latency (objective and subjective) and in improving sleep maintenance (objective and subjective) (14). Results from these trials did not suggest rebound or withdrawal with discontinuation (33). A study found the power spectral density of EEG signals during polysomnography of patients on suvorexant to be no different than those of patients on placebo. This suggests that unlike the benzodiazepines and imidazopyridines, suvorexant does not alter sleep architecture (56). Daytime somnolence is a side effect of suvorexant, and impaired next-day driving is seen in doses as low as 20 mg. As such, the FDA has approved doses of 20 mg or less. The AASM suggests its use for sleep maintenance insomnia based on trials of 10 mg, 15/20 mg, and 20 mg doses of suvorexant. Pooled analyses of data from phase 3 randomized controlled clinical trials showed that suvorexant is well-tolerated in elderly patients with insomnia (34). In a large cohort where 232 acute stroke patients received ramelteon combined with suvorexant or a GABA receptor agonist, the addition of suvorexant to ramelteon therapy improved subjective sleep quality and presented a lower proportion of patients showing delirium compared to the addition of a GABA receptor agonist (42).

Medications not approved by the FDA for the indication of insomnia but often used off-label for insomnia therapy are reviewed below. These include antidepressants, antipsychotics, and anticonvulsants.

Due to concerns of tolerance, addiction, and dependence on hypnotics, antidepressants (especially amitriptyline and trazodone) have been prescribed more and more for the treatment of insomnia. There has been an upward trend in the number of antidepressants that were prescribed for insomnia in the last decade and a half. We have already reviewed doxepin, a tricyclic antidepressant approved for insomnia, and reviewed the evidence for its use. Unfortunately, there are limited data regarding the efficacy of most other antidepressants as hypnotics.

A meta-analysis of polysomnographic randomized, controlled trials of pharmacotherapy for insomnia showed that benzodiazepines and imidazopyridines are significantly more effective than antidepressants in reducing subjective and objective sleep onset latency in patients with primary insomnia (102). The authors conclude that physicians should favor benzodiazepines and imidazopyridines over antidepressants for the treatment of primary insomnia but acknowledge that side effect profiles and comorbidities should play a role in choosing pharmacotherapy for insomnia. Benzodiazepines/imidazopyridines have the disadvantage of abuse, dependency, tolerance, and rebound. However, the antidepressants also have the disadvantage of lingering daytime sedation after nighttime administration and other adverse effects (with tricyclics aggravation of restless legs syndrome, anticholinergic effects, etc.) that make them poor choices for the treatment of insomnia except in a select group of patients.

Besides doxepin, other tricyclics used for insomnia (off-label) include trimipramine and amitriptyline. There is a lack of evidence regarding whether these medications are useful in patients with both depression and insomnia. In addition to a lack of data showing efficacy, significant side effects should be considered before prescribing these medications for insomnia, including anticholinergic side effects, weight gain, and cardiac conduction effects—the reason being that despite that its short-term use could be beneficial, long-term treatment with sedative antidepressants is not generally recommended (77).

Trazodone is one of the most frequently used antidepressants for insomnia. Though trazodone does not tend to aggravate restless legs symptoms and periodic limb movements as do the tricyclics, it does have residual daytime sedation and can cause rebound insomnia. The AASM does not recommend treatment with trazodone for sleep onset nor sleep maintenance insomnia because, in previous research, none of the sleep outcome variables improved to a clinically significant degree (87). Findings, however, on preserving cognition via sleep-wake rhythm consolidation encourage further investigation into trazodone effectiveness (47).

Selective serotonin reuptake inhibitors are another class of antidepressants not found to be particularly effective for primary insomnia. However, a case-control study of 90 patients with primary insomnia showed that after 8 weeks, combined treatment with zolpidem and paroxetine was more effective than treatment with zolpidem alone in sleep maintenance and early morning awakenings (105).

Antipsychotics, particularly quetiapine and olanzapine, are increasingly being used to treat insomnia even in patients without psychosis. Despite quetiapine being regularly prescribed for the treatment of insomnia, especially with comorbid depression, there is a lack of clear evidence for its efficacy and place in the management of insomnia. In fact, a review of the literature concluded that quetiapine should not be used in the treatment of insomnia, even when there are comorbid indications for quetiapine, given the lack of studies proving the benefits may outweigh the risks of the drug (03). A systematic review of 13 trials of atypical antipsychotics in primary insomnia concluded that these drugs should be avoided in the first-line treatment of primary insomnia. This recommendation was based on a lack of efficacy and their side effect profile (95). This includes anticholinergic effects, extrapyramidal effects (though less frequent in atypical antipsychotics), cardiac effects, and increased mortality in dementia patients; antipsychotics should not at this time be used for insomnia in patients without comorbid depression or psychosis (17).

Another class of medication used off-label for insomnia is the anticonvulsants, such as gabapentin, pregabalin, and tiagabine. At relatively low doses of 300 and 600 mg, gabapentin has been shown to modestly improve sleep efficiency in primary insomniacs (05). A randomized, blinded, placebo-controlled polysomnography trial of gabapentin at 250 mg for insomnia showed less wake after sleep onset on polysomnogram and higher ratings of sleep quality without next-day impairment (28). In addition, a study of gabapentin (100 to 600 mg per day) or clonazepam (0.5 to 2 mg per day) showed the drugs to be similarly effective in treating residual sleep disturbances in patients recovering from major depression (67). Studies evaluating the efficacy of tiagabine as a hypnotic for primary insomnia have produced contradictory results. When different tiagabine doses (4, 8, 12, and 16 mg) were compared to placebo, there was a significant dose-dependent increase in slow-wave sleep percentage with all tiagabine doses. A trend toward a dose-dependent increase in total sleep time was present without any effect on latency to persistent sleep. Tiagabine is not recommended as a treatment for sleep onset or maintenance insomnia based on trials of 4 mg of the drug (87).

Medications available over the counter often used in patients self-treating insomnia include the antihistamines such as diphenhydramine and doxylamine and over-the-counter supplements such as valerian and melatonin. We have already briefly mentioned the antihistamines and melatonin. One of the antihistamines, diphenhydramine, is very frequently used to treat insomnia despite little evidence that it is beneficial for this indication. Although diphenhydramine may have some effect on sleep maintenance, it has little effect on sleep onset insomnia and has significant drawbacks including anticholinergic side effects, tolerance, and significant next-day impairment (64). Diphenhydramine is not recommended as a treatment for sleep onset or maintenance insomnia based on trials of 50 mg of diphenhydramine (87). A prospective study about self-medication for insomnia in college students has shown that 25% of students approved of sleep aid use, which implies a high prevalence of self-medication among college students (30). Sleep aids were used 1 to 2 nights per week with over-the-counter medications being the most frequently used, followed by marijuana and alcohol. The study has also shown that starting self-medications for insomnia was not associated with improvement in insomnia after 2 months but was associated with an increase in negative consequences. In lieu of the legalization of cannabis products, cannabinoids have emerged as another class of drugs used for self-medication against insomnia. Despite its rising use, there are not enough studies to prove the efficacy of cannabinoids in the treatment of insomnia; however, many randomized controlled trials are currently underway to examine them. Some studies have shown reduced dream recall, reduced sleep onset latency, and improved subjective sleep quality but with hangovers and other side effects. Cannabinoids are not yet recommended in the treatment of insomnia pending further justifications (52).

Many over-the-counter herbal supplements are used for insomnia. Magnesium is one of the supplements notoriously claimed to improve sleep; however, there is a lack of evidence to support this claim. A systematic review and meta-analysis of three randomized controlled studies found that magnesium may improve sleep in insomnia patients, but the results were statistically insignificant. Nevertheless, the authors recommend oral supplementation of magnesium (less than 1 g given up to three times a day) for improving sleep because magnesium is cheap, widely available, and safe to use (57). A systematic review and meta-analysis of 14 randomized controlled trials of herbal remedies for insomnia (ie, valerian, chamomile, kava, and Wuling) did not find any of these drugs to be more effective than placebo or any active control. Kava, chamomile, and Wuling showed similar or less adverse events per person than placebo, but a greater number of adverse events were seen with valerian (53). One promising clinical trial showed that the Ashwagandha root extract significantly improved sleep quality both objectively and subjectively in healthy subjects and patients suffering from insomnia when compared to placebo (50).

Another area of treatment that has been gaining popularity is aroma inhalation therapy. Through stimulating the olfactory sense, aromatic oil particles are able to regulate and impact the endocrine, immune, and nervous system. A systematic literature review and meta-analysis showed that aroma inhalation therapy when using a single aroma (most commonly lavender) may significantly resolve sleep problems, especially in patients suffering from insomnia. The effects of the treatment increase as the number of sessions increases. However, further studies and trials are required to highlight further evidence for the usage of aroma inhalation therapy in the management of insomnia (13).

Consensus guidelines do not currently recommend the use of over-the-counter and herbal agents in the treatment of chronic insomnia. European guidelines describe light therapy and exercise regimes as possibilities for adjunct therapies (77). A review study published in 2022 showed promising results for the use of lavender in the treatment of insomnia in different patient populations (including patients suffering from end-stage renal disease, neuropsychiatric disorders, and respiratory, cardiac, and metabolic diseases) concluding with the need for further studies as mentioned before (55).

The goal of treatment is resolution of the insomnia or that of the underlying cause. If left untreated, insomnia can be the cause of significant economic problems, mortality, and morbidity. It costs the American public about 100 billion dollars annually in medical expenses, ramifications of accidents, and reduced productivity due to absenteeism and decreased work efficiency.

A study empirically analyzed the financial impact of insomnia treatment on health care (101). Surprisingly, it found that insomnia treatment did not lead to a substantial decrease in healthcare utilization. Moreover, the study found that insomnia treatment was correlated with an increase in emergency visits within 1 year. Future research is required for the effects of insomnia on healthcare costs to further understand its impact.

Longstanding untreated insomnia can be a significant risk factor for the development of depression and anxiety disorders (36). Thus, it is of utmost importance to diagnose and treat insomnia early, especially during times like the COVID-19 epidemic. A study has shown a significant increase in the prevalence of insomnia among individuals affected by COVID-19 compared to the general population (11). Not only has there been an increase in the prevalence of insomnia during the epidemic, but there has also been a positive correlation between insomnia severity during the epidemic and the severity of anxiety and suicidal ideation (43). Therefore, it is essential to act swiftly and treat insomnia, which may aid in reducing the risk of suicide among those at risk.

Indications

Insomnia is the most common sleep-related complaint and the second most common overall complaint (after pain) reported in primary care settings. An estimated 4% of adults who have symptoms of insomnia use hypnotic medications (70). Chronic insomnia is defined by the International Classification of Sleep Disorders, 3rd edition (ICSD-3) as difficulty maintaining or initiating sleep despite adequate opportunity and environment for sleep that occurs at least three times per week for 3 or more months and is associated with some form of daytime impairment. In chronic insomnia, difficulties with sleep cannot be explained by another sleep disorder (01).

Insomnia is a frequent symptom associated with many medical or psychological conditions or substance use. Insomnia in these situations often responds to the treatment of the underlying disorder.

Contraindications

Alcoholics and drug abusers have a higher tendency to abuse benzodiazepines, so benzodiazepines should be prescribed with caution in this population. Benzodiazepines are also contraindicated in pregnancy.

Outcomes

No reliable studies have looked at the percentage of responders to sedative hypnotics. A few studies show that most people will respond, at least initially, to nonbenzodiazepine GABAA agonists and benzodiazepines with improved subjective and objective sleep. Some patients may even experience long-term benefits. However, to maintain long-term benefits, a combination of pharmacological and behavioral methods of treatment is usually needed. Among chronic insomnia patients, about 40% do not achieve sustained remission with any of the currently available pharmacological or behavioral treatments (78).

A study in Japan examined whether baseline demographic factors or subjective sleep variables were predictive of outcomes, following treatment with eszopiclone 2 mg. This study found no relationship with demographic factors and treatment outcome, but it did find that patients with shorter sleep latency and lower wake after sleep onset values at baseline showed better outcomes after treatment as measured in changes in sleep latency and wake after sleep onset (35).

Another meta-analysis sought to compare treatment efficacy of medications used for insomnia. Using only polysomnographic, parallel-group, randomized controlled drug trials, this study found higher effect sizes for objective and subjective outcomes for benzodiazepines and imidazopyridines compared with antidepressants, including doxepin (102).

Adverse effects

Traditionally, concerns regarding the use of benzodiazepines and imidazopyridines centered around daytime somnolence and risk of addiction, dependence, and tolerance, although the risk of abuse is lower than previously thought in those properly diagnosed and without a history of abuse (81). In 2013, the FDA issued a drug safety communication recommending label changes to products containing zolpidem to reduce the recommended dose and advised against driving after using extended-release preparations (U.S. Food and Drug Administration 2013). In 2014, the FDA changed the recommended starting dose of eszopiclone to 1 mg because of next-day impairment (04).

However, there have been growing concerns about the relationship between benzodiazepines and mortality. A large-scale study followed more than 10,000 subjects on hypnotics (benzodiazepine receptor agonists, benzodiazepines, and antihistamines) for an average of 2.5 years and compared them with more than 20,000 controls. The results, although not causative, demonstrated a larger than 3-fold increase in mortality associated with taking as little as 18 pills per month. Cancer risk was higher in those taking nightly sedative hypnotics than controls. The study was controlled for prestudy health status (44). Similarly, the published results from two large cohort studies in France and the United Kingdom suggest there is a moderate increase in all-cause mortality at 12 months in incident (and mostly occasional) users of benzodiazepines when compared with controls and nonbenzodiazepine hypnotic/antidepressant users (71).

In addition to concerns about mortality, there have been growing concerns regarding long-term cognitive function and its association with benzodiazepine and imidazopyridine use. A case-control study using data from the Quebec health insurance program database found lifetime benzodiazepine use was associated with an increased risk of Alzheimer disease (adjusted for insomnia, anxiety, and depression) with the strength of the association related to the length of exposure (09). The mechanism is not clear, but studies in mice have suggested altered neuronal activity in the hippocampus (suppressed CA1 calcium signaling) (08). A case-control study based on 20 patients followed at a sleep center for drug discontinuation and 13 healthy matched controls demonstrated that long-term use of high doses of benzodiazepines leads to disruption of sleep microstructure (58). Also, long-term use of short or intermediate half-life benzodiazepine receptor agonists was associated with a higher risk of high blood pressure in a large sample of 1272 individuals with insomnia (32).

Zolpidem has gotten a significant amount of press in the lay media because of automatic behaviors experienced by people taking this medication, including sleep driving, sleep eating, and sleepwalking with no recollection of the events afterward. Few of the high-profile cases occurred when zolpidem dose exceeded 10 mg or when it was taken in conjunction with alcohol or other sedative medications. In all cases, the automatic behaviors stopped after stopping zolpidem. Caution should be used, especially when zolpidem is prescribed with other sedatives, at a dose higher than 10 mg (5 mg in women and men over 65 years of age), or in the setting of head injury, psychiatric illness, untreated sleep disorders, and childhood history of sleepwalking. In fact, the single key risk predictor for complex sleep behaviors with zolpidem is doses higher than 10 mg per day.

As with most sedative drugs, a meta-analysis has shown that Z-drugs are associated with falls and fractures in the elderly (96). A study of 89 chronic users of temazepam, zopiclone, or zolpidem aged 55 or older showed significant and rapid improvements in hand-grip strength and balance with the withdrawal of the medications (69). In 2012, the American Geriatrics Society advised physicians and patients to avoid using benzodiazepines as first-line therapy for insomnia in the elderly because of the risk of falls and fractures (93). However, a study has shown that in nursing homes, insomnia, and not hypnotic use, is associated with a greater risk of subsequent falls. Benzodiazepines and Z-drugs are associated with increased risk of hip fracture, especially in patients new to the medication (20). Also, in a drug event monitoring program in Japan, a safety analysis of zolpidem use suggested that low-dose zolpidem can be safely prescribed to patients 80 years of age or older (40).

Additionally, transient psychosis has been associated with zolpidem use, and persistent psychosis after abuse of high-dose zolpidem was reported in a young male using up to 500 mg of zolpidem for 3 months. His isolated, suspicious, and aggressive behavior continued after weaning off zolpidem, and he has now been diagnosed with schizophrenia (25). A more alarming finding has been the association of zolpidem use and increased risk of suicide in people with and without psychiatric illness. In a case-cohort study from Taiwan, zolpidem exposure was found to be significantly associated with the risk of suicide/suicide attempt with an odds ratio of 2.08 (95% CIs, 1.83 to 2.36) (91). A case report showed that high doses of zolpidem could contribute to interruption of the sneezing pathway; nonetheless, because it was based on only one patient, further investigations are required to confirm or refute this rare side effect (02).

Eszopiclone is similar in its safety profile to zolpidem and zaleplon but has an unpleasant taste (34% vs. 3%) as a commonly reported side effect in patients taking it versus those taking a placebo.

The American Family Physician released new guidelines in 2019 that recommended against using “Z drugs,” including zolpidem, for more than a month in adults aged 18 to 64 years old and recommended against using them for any duration in adults older than 65 years old (23).

Suvorexant, the orexin receptor antagonist, has also been shown to have no clinically relevant respiratory effects in healthy individuals (97), and it has been found to be safe in patients with mild to moderate obstructive sleep apnea at twice the recommended dose (90). It is generally well tolerated with the most common adverse effect reported to be next-day somnolence. At doses of 10 mg in healthy volunteers, there were no residual effects, but higher doses have shown prolonged reaction time and deficits on driving tests (90a). The frequency of somnolence increases in a dose-dependent fashion at doses exceeding FDA-recommended levels (87). Doses greater than 20 mg have been associated with symptoms similar to cataplexy, suicidal ideation, hypnagogic hallucinations, abnormal dreams, and sleepwalking (22).

Diphenhydramine, the active ingredient in most over-the-counter hypnotics, impairs driving performance more than alcohol. In fact, the residual daytime sedation of a single nighttime 50 mg diphenhydramine dose was reliably shown on PET even in the absence of subjective sedation. It also has been associated with sleepiness and psychomotor performance decline lasting into the next day. Cognitive side effects in those taking diphenhydramine include learning impairment in teenagers and increased risk of dementia in adults. Additional side effects include dizziness, xerostomia, and orthostatic hypotension (41; 103).

Antidepressants have the adverse effects of lingering daytime sedation after nighttime administration and other adverse effects (with tricyclics, aggravation of restless legs syndrome, and anticholinergic effects).

Ramelteon and melatonin are generally safe medications. The most common side effects of melatonin are headache and somnolence, and its use is not recommended in men and women trying to conceive as it may affect fertility (64). There have been no reports of fertility issues related to ramelteon use (64). Ramelteon may cause somnolence, dizziness, and nausea (64). One study showed that with acute treatment (within 1 hour of dosing), ramelteon does negatively affect road-tracking performance, visual attention, psychomotor speed, and body balance compared with placebo, but not as significantly as low-dose triazolam (65).

There are no good safety data or FDA-regulated standard formulations of the herbal medications, and as mentioned previously, in a meta-analysis valerian was associated with more adverse events than placebo.

Though none of the medications for treating insomnia are without risk, longstanding untreated insomnia itself comes with significant risks. Chronic insomnia has been associated with the development of depression and anxiety disorders as well as hypertension (36). It has been established that short sleep duration could increase the risk of metabolic syndrome in healthy individuals (18). A clinical trial of 404 patients diagnosed with hypertension and insomnia for 28 days (where 202 cases were treated with estazolam and antihypertensive treatment and 200 matched controls received antihypertensives and placebo) showed lower blood pressure in the treatment group by the end of the experiment (54). On the other hand, a population-based study did not find a correlation between insomnia severity and hypertension (99).

Future developments. Though there are several effective drugs for the treatment of insomnia, there is still a need for more drugs that are effective and do not have the risk of next-day impairment that so many of our current insomnia drugs have.

Medications showing promise include two dual orexin receptor antagonists. In a phase 3 clinical trial, lemborexant, another orexin antagonist that was FDA-approved for the treatment of insomnia, was shown to provide better sleep promotion than zolpidem and placebo in patients older than 55 years of age. Lemborexant was found to be tolerable and efficacious, and its effects were maintained after 1 month (85). Lemborexant, when administered at 25 mg/day, has a safe and linear pharmacokinetic profile and safe pharmacodynamic profile independent of age, sex, and race. At that dose, lemborexant is efficacious for the treatment of insomnia with minimal residual effects during wake time (49). A systematic review and network meta-analysis published in 2022 in The Lancet has shown that lemberoxent and eszopiclone were the best drugs when assessing efficacy, tolerability, and acceptability, with the need for further safety data on lemborexant (16). Another orexin receptor antagonist, daridorexant, has been FDA-approved in the United States and comes in 25 mg and 50 mg doses. In a phase 3 study, daridorexant significantly improved sleep induction and maintenance, with a significant dose-response relationship when compared to placebo in subjects with insomnia disorder. Furthermore, latency to persistent sleep and wake after sleep onset were significantly decreased from baseline to months 1 and 3 with daridorexant 25 mg and 50 mg treatment (72). Daridorexant has also been shown to be safe and tolerable, with no residual next-morning effects or withdrawal symptoms (21; 59; 63). However, a systematic review and network meta-analysis failed to show an overall benefit in the treatment of insomnia with daridorexant (16).

Special considerations

Pregnancy. Sleep maintenance insomnia is more common in women in their third trimester compared to age-matched nonpregnant women. Multiple gestations tend to have higher incidence of insomnia. Restless legs syndrome and resulting insomnia are common complaints in pregnancy affecting up to one in every three women, especially in the third trimester. The pharmacological treatment of insomnia should be undertaken carefully during pregnancy due to potential teratogenicity. Among the imidazopyridines, zolpidem is pregnancy category B, whereas zaleplon is category C. Benzodiazepines are contraindicated during pregnancy as they may cause fetal damage (FDA pregnancy category X). The imidazopyridines are all classified as pregnancy category C. In animal studies using zolpidem, reports have shown decreased skull ossification at 24 to 120 times the human dose, and there are reports of low birth weight and small for gestational age babies whose mothers used zolpidem (07). Most dopaminergic agents, opiates, gabapentin, clonidine, trazodone, and over-the-counter sleep aids are all classified as FDA pregnancy category C, which means there are no studies of these agents in human pregnancy, but their use was associated with some pregnancy-related complications in animal studies. Among examined studies, a review of hypnotics (including benzodiazepines, imidazopyridines, antidepressants, and antihistamines) in pregnancy found there was no correlation between sleep aids and an increased risk of congenital malformations, though the benzodiazepines and imidazopyridines may increase the risk of small gestational age infants and preterm birth (70).

Menopause. In menopause-related insomnia, estrogen replacement therapy may control insomnia along with other menopause-related symptoms, though a randomized, blinded trial of 40 menopausal women with insomnia (but without significant hot flashes or other vasomotor symptoms) did not show estrogen therapy to be any more effective than placebo in improving sleep quality as determined by actigraphy (94). Additionally, medical problems reported with estrogen replacement have led to a search for other alternatives. Escitalopram at 10 to 20 mg per day was found to reduce insomnia symptoms and improve subjective sleep quality (75). Herbal alternatives for the treatment of menopause-related insomnia have also been studied. In one study, daily administration of black cohosh for 6 months was shown to improve sleep quality objectively and subjectively (39). Furthermore, isoflavone treatment has been shown to reduce insomnia symptoms, significantly increase sleep efficiency, and decrease the intensity and frequency of insomnia (51). Eszopiclone 3 mg at bedtime tends to improve both the insomnia and the other ancillary symptoms (hot flashes, etc.) of menopause as do gabapentin at 600 to 1800 mg dose, mirtazapine, and controlled release melatonin (06).

Schizophrenia. Articles reviewed between January 2000 and March 2018 showed that switching antipsychotics to paliperidone or adding eszopiclone or melatonin were solid approaches for the treatment of insomnia in schizophrenic patients. Adding sodium oxybate was also recommended in case the aforementioned methods were ineffective (89).

Alcoholism. Insomnia is considered a common comorbidity of alcoholism. If untreated, there is an increased risk of relapse for recovering alcoholics, with a lower quality of life. Sedative hypnotic drugs and other psychotropic drugs are of concern in this population due to an increased risk of relapse, overdose when combined with alcohol, and an increased risk for abuse of the drugs. There are limited studies investigating the proper treatment of insomnia in alcoholism. Cognitive behavior therapy, muscle relaxation therapy, and bright light therapy have demonstrated a reduction in self-reported subjective symptoms of insomnia; however, objective measurements using polysomnography have not been used. Trazodone 50 to 200 mg/day is frequently prescribed to treat insomnia associated with alcohol cessation despite not being approved by the FDA. Trazodone has been shown to increase sleep in two randomized, double-blind, placebo-controlled studies; however, one of these studies has established an increased risk for alcohol use relapse. Further studies are required (29; 80).

Pediatrics. The first line of treatment should always be behavior therapy. If unsuccessful, pharmacologic management in combination with behavior therapy has been shown to be beneficial. Further research is required for specific insomnia medication in the pediatric population. Therefore, the choice of medication should be based on the individual patient profile and the specific characteristics of the drugs, such as the presence of comorbid psychiatric or medical conditions, drug-drug interactions with other prescribed drugs, etc. (24).

Clinical vignette

A 54-year-old woman presented to a sleep center outpatient clinic for evaluation for occasional insomnia. She was a 6 to 7 hours per night sleeper until her father passed away, and during the acute grief period she had trouble falling asleep and staying asleep. Her sleep, however, normalized and stayed stable for several months, when for no apparent reason, she started having insomnia for several days in a row, mainly sleep onset insomnia. This, again, resolved on its own, and the patient did well until she again had a period of 1 week where she had significant sleep onset insomnia. Her bad nights were described as going to bed around 11 o'clock, tossing and turning for half an hour or so unable to sleep, and then leaving the bedroom, turning on the TV, or reading something until she felt sleepy. She tended to fall asleep in the early hours of the morning, only to wake up with the alarm around 4:40. She usually got 3 to 4 hours of sleep on a bad night, and about 7 to 7.5 hours of sleep on a good night.

The patient was diagnosed with occasional situational-induced insomnia. She was advised to follow strict sleep hygiene regulations; zolpidem 5 mg was prescribed to be taken on an as-needed basis whenever she had a problem falling asleep. At her 6-month follow-up she had no more complaints of insomnia and had used the zolpidem only a few days each month.

Scientific basis

The benzodiazepines are GABA agonists. They potentiate the action of GABA by displacing an endogenous inhibitor of GABA receptor binding. They have potent hypnotic, muscle relaxant, anticonvulsant, and antianxiety properties.

Imidazopyridines, also referred to as the nonbenzodiazepines, benzodiazepine receptor agonists, or “Z” drugs, are structurally unrelated to benzodiazepines. They are specific GABAA agonists, and because of their higher affinity for certain subunits of the GABAA receptor, they have potent hypnotic effects but weak anticonvulsant, anxiolytic, and muscle relaxant properties at recommended doses.

Melatonin, an endogenous hormone released by the pineal gland, regulates circadian rhythms by direct action on the suprachiasmatic nucleus (the circadian pacemaker). The soporific effect of melatonin seems to be only during the day several hours after the administration of the dose, most likely due to a reduction of core body temperature.

The H1 antagonist doxepin is believed to exert its hypnotic effect by blocking the wake-promoting effects of histamine. Doxepin is a tricyclic used for the treatment of depression at doses of 75 to 150 mg but is approved for insomnia treatment at doses of 3 to 6 mg (64).

Orexin receptor antagonists are a new class of drugs for the treatment of insomnia. Suvorexant was just approved by the FDA on August 14, 2014. Orexin is a neuropeptide synthesized by the hypothalamus that promotes arousal and wakefulness, and blocking of its receptors by suvorexant has been found to promote sleep (79). Identification of a novel orexin type 2 antagonist encourages research of compounds targeting orexin type 2 receptors in the treatment of insomnia (38).