REPORTS

Strategies for Treating Chronic

Anna K. Morin, PharmD

Abstract 85%.2,3 Insomnia is often further classified in Insomnia is a prevalent condition that remains clinical practice, based on the duration of underdiagnosed and undertreated. Recognizing and symptoms, as follows: transient insomnia, treating insomnia are important in decreasing mor- lasting 1 to 3 nights; short-term insomnia, bidity and restoring quality of life for those who lasting 3 nights to 1 month; and chronic experience disturbances. Appropriate treatment insomnia, lasting more than 1 month.1,2,4,5 of insomnia should involve multiple interventions The severity and treatment of insomnia take designed to address not only the symptoms of insom- into account the frequency, intensity, dura- nia itself, but also any coexisting factors that may be contributing to the sleep disturbances. A combination tion, and consequences associated with the 1,4 of pharmacologic and nonpharmacologic therapies sleep disturbances. may be particularly efficacious in those with chronic Chronic insomnia may exist in isolation and debilitating insomnia. Pharmacotherapy is the (primary insomnia) or coexist with other most frequently used intervention for insomnia in medical and psychiatric illnesses, medica- cases where the goal of therapy is immediate relief of tion and substance misuse, behavioral or symptoms, insomnia is accompanied by significant environmental factors, or other primary distress or impairment, nonpharmacologic approach- sleep disorders, such as obstructive sleep es alone are ineffective, or the patient prefers med- apnea.6 Although the direct consequences of ication. The ideal has the following insomnia have not been fully identified, characteristics: rapid absorption, rapid sleep induc- chronic insomnia has been associated with tion, optimal duration of action, preservation of sleep architecture, and a favorable safety profile. This a higher risk for the development of depres- review will discuss currently available treatment sion, cardiovascular disorders, chronic options for insomnia, the benefits of each, and appro- obstructive pulmonary disease, back and priate treatment regimens. hip problems, osteoarthritis, rheumatoid (Am J Manag Care. 2006;12:S230-S245) arthritis, and peptic ulcer disease.6-8 Al- though prevalence estimates of insomnia vary depending on the definitions and crite- haracterized by disorders of initiating ria used, epidemiologic studies indicate that or maintaining sleep, or nonrestora- approximately 30% to 35% of the general Ctive sleep, insomnia is a prevalent population experience at least occasional condition that can coexist with psychiatric or intermittent sleep disturbances.6,7,9,10 and medical illness and cause significant Twenty-five percent of these individuals, or impairment of social and occupational func- 10% of the population, report chronic insom- tioning.1 Patients may experience several nia symptoms accompanied by daytime con- symptoms of insomnia at one time, and the sequences of fatigue, irritability, and pattern of symptoms may change over time.1 impaired concentration, which can lead to In studies evaluating the outcomes of treat- negative effects on overall health, mood, and ments of sleep disturbances, insomnia is functioning.6,7,10 Costs associated with in- often defined by a sleep latency (SL) and/or somnia can include related medical expens- wake after sleep onset (WASO) time period greater than 30 min, with a corresponding sleep efficiency (total amount of sleep time Corresponding author: Anna K. Morin, PharmD, Assistant Professor, divided by the total amount of time spent in Department of Pharmacy Practice, Massachusetts College of Pharmacy and Health Sciences, 19 Foster Street, Worcester, MA 01608. E-mail: bed with the intent to sleep) of less than [email protected].

S230 THE AMERICAN JOURNAL OF MANAGED CARE MAY 2006 Strategies for Treating Chronic Insomnia es, more frequent use of healthcare re- focus on modifying factors that precipitate sources, a higher rate of absenteeism, and perpetuate sleep disturbances. Several reduced subjective productivity, and an behavioral techniques (ie, sleep hygiene [SH] increased risk of accidents relative to those education, relaxation therapies, stimulus without insomnia.6,11 As a result, the overall control, sleep restriction) and cognitive ther- economic burden of insomnia is estimated apy have been shown to be effective in the to exceed $100 billion annually.6,10,12 Popu- treatment of insomnia.2-4,14,16-19 Clinical data lations at particular risk for insomnia may demonstrate that behavioral techniques, include women, elderly persons, people with particularly stimulus control and sleep medical and psychiatric comorbidities, and restriction, are superior to and as shift workers.2,6,13-15 effective as pharmacotherapy in the short- However, despite the fact that insomnia is term treatment of sleep initiation problems a prevalent condition that can be associated associated with insomnia.2,3,17,19 Most behav- with negative consequences, it remains ioral and cognitive interventions are com- underdiagnosed and undertreated. Recog- patible with one another and can be nizing and treating insomnia are important combined to optimize outcome. In general, in decreasing morbidity and restoring quali- advantages of these behavioral and cognitive ty of life for those who experience sleep dis- interventions are minimal adverse effects turbances. A comprehensive assessment of and documented improvement in sleep an individual’s medical, psychiatric, and sustained over 6 to 24 months.2,17,19,20 Limi- pharmacologic history, sleep and wakeful- tations to the implementation of these inter- ness patterns, and family history of sleep ventions include a shortage of personnel disorders is necessary before a diagnosis of trained in the provision of these techniques, insomnia can be made and a treatment plan high cost and limited or no insurance reim- implemented. Appropriate treatment of bursement, patient preferences for pharma- insomnia should involve multiple interven- cologic interventions, and the fact that these tions designed to address not only the symp- techniques are time intensive and require toms of insomnia itself but also any motivation on the part of the individual expe- coexisting factors that may be contributing riencing insomnia symptoms.3,4,16-18 Fur- to the sleep disturbances. thermore, some research has suggested that Many people with insomnia complain the efficacy of behavioral and cognitive inter- about difficulties in falling asleep, and, as a ventions decreases with increasing age.14,19 result, sleep onset has long been the focus of Behavioral and cognitive interventions are both pharmacologic and nonpharmacologic typically implemented when pharmacother- treatment interventions. Sleep maintenance apy is contraindicated as augmentation to (staying asleep) can also be a significant pharmacotherapeutic interventions or as a problem, particularly in the elderly and in result of patient preference. individuals with insomnia coexisting with First outlined in 1977 and based on clini- psychiatric and medical disorders.6,14 Al- cal observations of patients with sleep distur- though many individuals may benefit from bances, SH recommendations have evolved pharmacologic intervention, evidence also into a list of behaviors, environmental con- supports the use of nonpharmacologic treat- ditions, and other sleep-related factors that ments in the management of insomnia. A are believed to be instrumental in promoting combination of pharmacologic and non- improved quantity and quality of sleep.16 pharmacologic therapies may be particular- The assumption that accompanies SH edu- ly efficacious in those with chronic and cation in patients with insomnia is that sleep debilitating insomnia. disturbances arise, to some extent, when these patients deviate from SH behav- Nonpharmacologic Interventions for iors.16,18 Although commonly used as an Treating Insomnia approach to the treatment of insomnia, def- Nonpharmacologic interventions in the initions of SH in the scientific literature treatment of insomnia primarily include vary, depending on investigator and study behavioral and cognitive techniques that focus. However, common components of SH

VOL. 12, NO. 8, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S231 REPORTS

recommendations include limiting the use of that regulates sleep and wakefulness.17,18 the bedroom only for sleep and intimacy; The allowable time spent in bed is increased keeping a regular bedtime and wake-up or decreased by 15 to 20 min each week schedule; avoiding , , caf- until a goal of 80% to 90% sleep efficiency is feine, large meals, and vigorous exercise met.2,18 Because gains in total sleep time are near bedtime; eliminating or minimizing not seen immediately, both stimulus control daytime napping; and modifying the sleep and sleep restriction require patient motiva- environment to eliminate or remove sleep- tion and encouragement. Both techniques disturbing elements, such as bright lights, have been shown to be highly efficacious as extremes in temperature and noise levels, single or combined therapies for sleep-onset and bedroom clocks.4,16,18 Evidence support- and sleep-maintenance insomnia.2,19 ing SH education as a stand-alone approach Cognitive approaches used in the treat- to the treatment of insomnia is limited.2,16,19 ment of insomnia involve restructuring tech- Although these factors are rarely the pri- niques that address maladaptive thought mary cause of chronic insomnia, individuals patterns and attitudes that exacerbate and should always be educated about good SH amplify sleep disturbances.2,3,18 The goal is to and proper sleep habits, regardless of the eti- recognize, challenge, and change patient- ology of the sleep disturbance. specific unrealistic sleep expectations, mis- Relaxation therapies for insomnia specifi- conceptions about the causes of insomnia, cally target the arousal system and focus on and apprehension and anxiety about bed- reducing factors, such as tension and stress, time. Cognitive therapy has not been evalu- that may precipitate sleep disturbances.17,18 ated as a stand-alone treatment for chronic Many individuals, however, do not recognize insomnia, but treatment approaches that the significance of stress reduction in the incorporate cognitive restructuring tech- treatment of their insomnia and do not niques have shown positive outcomes.2,3 invest the time required to learn progres- sive muscle relaxation, meditation, simple Pharmacologic Interventions for breathing retraining, guided imagery, or Treating Insomnia other forms of relaxation. There is currently Pharmacotherapy is the most frequently no evidence to support the use of relaxation used intervention for primary and secondary techniques as the sole intervention in the insomnia when immediate symptom relief is management of insomnia.2,18 needed, when the sleep disturbances pro- The basis for the use of stimulus control is duce significant distress or impairment, to associate the bedroom and bedtime with when behavioral and cognitive interventions sleep rather than wakefulness.2-4,18,19 To in- alone are insufficient in treating the insom- crease sleep efficiency and decrease frustra- nia, or when a patient prefers the use of tion that occurs with extended time awake medication.2,18,21 in bed, individuals are instructed to go to Characteristics of a desirable hypnotic sleep only when drowsy, to keep a regular include rapid absorption, rapid sleep induc- sleep-wake cycle, to use the bed only for tion, optimal duration of action, preserva- sleep and sex, to avoid napping during the tion of sleep architecture, and being free day, and to get out of bed if unable to fall from unwanted effects and interac- asleep within 20 min.2,18 Equally important, tions.21 The desired hypnotic agent should individuals should be encouraged to engage not cause residual daytime effects or memo- in relaxing activities until they feel drowsy ry loss, should not interact synergistically or sleepy again, and to repeat this last step with to produce respiratory depres- as often as necessary. Often used in conjunc- sion, and should not produce tolerance, tion with stimulus control, sleep restriction dependence, or rebound insomnia. limits the amount of time spent in bed to Currently available hypnotic agents (both only the time actually spent sleeping.17,18 and ) Maintenance of a consistent wake time, even have effects that are mediated via after a poor night’s sleep, is necessary to syn- activity at the gamma-aminobutyric acid

chronize the endogenous circadian rhythm type A (GABAA) receptors, leading to

S232 THE AMERICAN JOURNAL OF MANAGED CARE MAY 2006 Strategies for Treating Chronic Insomnia enhancement of the activity of GABA. GABA experiencing numerous awakenings through- is the major inhibitory of out the sleep cycle or early awakening with the mammalian central nervous system (40% an inability to return to sleep, provided that of central nervous system [CNS] neurons are the agent does not produce significant next- estimated to be GABAergic) and is thought to day residual effects. play the pivotal role within the sleep-induc- The use of and other sedating ing and maintenance systems.21,22 These agents for chronic insomnia is less straight- agents differ in their pharmacokinetic pro- forward. is approved for the files, safety profiles, and degree of selectivity long-term treatment of sleep-onset and for GABAA receptor subtypes. Most GABA sleep-maintenance insomnia and is the only receptors are composed of α, β‚ and γ sub- hypnotic that has been studied in two 6- units that combine to form pentameric chan- month, double-blind, placebo-controlled, nels, which open secondary to the binding of randomized trials of nightly administra- GABA or an .23,24 The result is an tion.30-32 , a receptor influx of negative chloride ions into the neu- agonist that binds to receptors regulating ronal cell, neuronal cell hyperpolarization, circadian rhythms, has recently received US and subsequent action potential inhibition. Food and Drug Administration (FDA)

The GABAA receptor subtype accounts for approval and is indicated for the long-term greater than half of all GABA receptors and treatment of sleep-onset insomnia, as evi- consist primarily of 2 α1-subunits, 2 β2-sub- denced in two 5-week, placebo-controlled 23-27 33 units, and 1 γ1-subunit. GABAA receptors trials of nightly administration. containing the α2- or α3-subunits are consid- tartrate extended-release formulation was erably less abundant and α4-, α5-, and α6-sub- also recently approved for the treatment of units account for less than 5% of all GABAA insomnia without any limitation in the receptor subtypes.24,27-29 The heterogeneity of length of use, based on two 3-week, placebo-

GABAA receptor subtypes is responsible, in controlled studies of nightly administra- part, for the various pharmacologic effects of tion.34 Other FDA-approved hypnotics in the benzodiazepines and other GABA . marketplace are indicated for the short-term Based on their elimination half-lives (t ), treatment of insomnia. 1/2 hypnotic agents may be divided into 4 cate- gories: (1) ultrashort-acting agents with a t Benzodiazepines. The site 1/2 of 1 hr, such as ; (2) short-acting in the GABAA receptor is located on the in- agents, with a t of 2.5 to 6 hr, including terface of the - and -subunits, and the 1/2 α γ , and the nonbenzodiazepines effects of benzodiazepines are determined 24,28 zolpidem tartrate and eszopiclone; (3) inter- by the GABAA receptor subtype. The α1- mediate-acting agents, with a t of 9 to 24 hr, subunit appears to mediate the sedative and, 1/2 including and ; and in part, the anterograde and anti- (4) long-acting agents, with a t greater convulsant effects of benzodiazepines.25,26,35 1/2 than 24 hr, including , , The α2-subunit mediates anxiolysis, and the 18,22 and . These differences in half- α3- and α5-subunits share life, along with time for onset of action, and ataxic properties.25,27 The muscle which is based, in part, on the time to max- relaxant and ethanol potentiation effects of imal concentration and GABAA receptor benzodiazepines appear to be mediated by 25,27 subtype activity, may be useful in determin- the α2- and α4-subunits, respectively. The ing which hypnotic agent to employ for a α-subunit of the GABAA receptor appears to specific sleep complaint.18 For example, for a be benzodiazepine insensitive.26 Previous re- primary complaint of difficulty falling asleep, ceptor terminology corresponds to the GABAA an agent with a short half-life and rapid receptor α-subunit, with the α1-subunit onset of action may be the most appropriate reflecting benzodiazepine1 (BZ1, or omega1) choice; administration of a longer-acting pharmacology and a heterogeneous set of agent may increase the risk of residual day- GABAA receptor subtypes containing α2-, α3-, time effects. An agent with a longer half-life or α5-subunits reflecting benzodiazepine2 28,36 might be more appropriate for a patient (BZ2, or omega2) pharmacology.

VOL. 12, NO. 8, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S233 REPORTS Supported by extensive studies confirm- Benzodiazepine dependence is consid- ing safety and efficacy at recommended ered more likely with use of high doses, daily doses, benzodiazepines have been the first- dosing of more than 4 months’ duration, line option in the pharmacologic short-term advanced age, current or previous history of management of insomnia.19,23 Benzodia- sedative hypnotic and/or alcohol depen- zepines that are approved by the FDA for use dence, and use of high-potency, short–half- in insomnia include estazolam, flurazepam, life benzodiazepines.41 Furthermore, abuse quazepam, temazepam, and triazolam. A potential of a particular benzodiazepine is meta-analysis of 22 studies reviewing the linked to both greater lipophilicity and benefits and risks associated with the short- shorter elimination half-life. All benzodi- term use ( 14 days) of benzodiazepines or azepines, however, have abuse potential and ≤ zolpidem to treat insomnia in adults report- all potentiate the effects of alcohol.22 Given ed significant improvements in SL onset, the pharmacology of the benzodiazepines, number of awakenings, total sleep duration, none of these adverse effects are unexpect- and sleep quality compared with placebo.37 ed. However, despite their adverse effects, A second meta-analysis that included short-, benzodiazepines are considered relatively intermediate-, and long-acting benzodi- safe in the general population, and even azepines showed improvements in total excessive doses are rarely fatal unless other sleep duration but not SL onset.38 The use of CNS (ie, ethanol) are taken con- benzodiazepines for the long-term treatment comitantly.22,40 Benzodiazepine hypnotics of chronic insomnia, however, remains con- are approved for the short-term (7-14 days) troversial, because no data have been gath- treatment of insomnia and, like all sedative ered to evaluate the long-term use of these hypnotics, should be used with caution in agents.4,38 individuals with a history of Adverse effects associated with the use of and in the elderly, who may metabolize benzodiazepines, such as alteration of sleep these medications more slowly, placing architecture, tolerance, next-day residual them at risk for greater cognitive impair- effects, cognitive and psychomotor impair- ment and sedative effects.18,38,39 ment, , and rebound insomnia, are dose dependent and vary Nonbenzodiazepines. In an attempt to according to the of the minimize the adverse effects associated with individual agent.4,22,38,39 The next-day hang- the use of benzodiazepines, the develop- over effect is the most common complaint ment of agents has been associated with the use of the longer-acting pursued. Not surprisingly, since the intro- benzodiazepines, with many patients experi- duction of the nonbenzodiazepine hypnotic encing , , decreased men- medications, use of benzodiazepines has tal alertness, and diurnal .22,30 In steadily declined.18 The 4 currently available addition to tolerance to the hypnotic and nonbenzodiazepines (zolpidem tartrate, sedative effects, physiologic dependence zolpidem tartrate extended-release, zale- accompanied by withdrawal symptoms on plon, and eszopiclone) promote sedation via discontinuation can occur with long-term the inhibitory neurotransmitter GABA, but use of benzodiazepines, but not to the possess distinct pharmacologic profiles, pro- extent seen with older and recog- files of receptor specificity, and clinical nized of abuse.4,22,38,39 Discomfort activity.32,42 associated with benzodiazepine withdrawal symptoms may be minimized by avoiding Zolpidem tartrate and zolpidem tartrate abrupt discontinuation and employing a extended-release. Zolpidem tartrate, an imid- gradual tapering of the benzodiazepine azopyridine derivative, is indicated for the dose.22,39,40 This physiologic dependence short-term treatment of insomnia and has a does not appear to be associated with psy- rapid onset of action and short elimination chological dependence or accompanying half-life (t ~2.5 hr).43 It appears to possess 1/2 inappropriate drug-seeking and drug-taking a very high affinity for the GABAA receptor 40 behaviors. α1-subunit, with significantly less affinity for

S234 THE AMERICAN JOURNAL OF MANAGED CARE MAY 2006 Strategies for Treating Chronic Insomnia the α2- and α3-subunits than do the benzodi- first night after treatment discontinuation azepines.26 Zolpidem has almost no activity were significantly higher (P .05) in the ≥ at the α5-subunit and, as a result of its pref- zolpidem tartrate 10-mg group versus 47 erential α1-subunit–binding affinity, has placebo. minimal and anticonvulsant Similar findings with regard to the lack of effects.26,44 The recommended doses are 10 ability of zolpidem tartrate to demonstrate a mg in healthy adults and 5 mg in older decrease in the number of awakenings or patients, debilitated patients, and in those improvement in sleep maintenance have with hepatic dysfunction.43 Adverse events been noted in other randomized, double- most commonly associated with discontinu- blind, placebo-controlled trials. Results of ation from US trials were daytime drowsi- some of these studies showed that zolpidem ness (0.5%), dizziness (0.4%), headache tartrate 10 mg improved sleep maintenance (0.5%), nausea (0.6%), and (0.5%).43 initially but not by the end of the study peri- Adverse events reported statistically signifi- od.46,49,50 Other studies did not find an cantly more often than placebo in short- improvement in sleep maintenance.45,49,51 term (up to 10 nights) clinical trials of Results from studies of the effects of zolpi- zolpidem tartrate (up to 10 mg) were drowsi- dem tartrate 10 mg on total sleep time have ness (2%), dizziness (1%), and diarrhea not been consistent.47-50 Discontinuation (1%).43 During longer-term treatment (28-35 treatment effects evaluated 2 to 3 days post- nights) of zolpidem tartrate (up to 10 mg), therapy include cases of tolerance, rebound adverse effects reported statistically signifi- effects, and withdrawal reactions.45,47,48,51 cantly more often than placebo were dizzi- Because of concern about potential toler- ness (5%) and a “drugged feeling” (3%).43 ance and/or abuse with nightly use, several Overall, adverse effects associated with studies of the effects of nonnightly (inter- zolpidem tartrate are dose related, occurring mittent) use of zolpidem tartrate have been more frequently with ≥20 mg per night. When conducted. In 1 study, intermittent dosing used at recommended doses, zolpidem tar- with zolpidem tartrate 10 mg on all treat- trate does not appear to negatively affect ment nights of the study had a statistically next-day psychomotor or cognitive func- significant impact on total sleep time dur- tion. Several trials have evaluated consecu- ing weeks 1 to 4, but no effect for weeks 5 tive nightly use of zolpidem tartrate.45-49 The to 8.52 No difference in SL was seen for longest placebo-controlled trial of nightly zolpidem tartrate 10 mg versus placebo administration is 5 weeks; therefore, the when examining intermittent dosing during potential for dependence, tolerance, or the entire treatment period. Also, although rebound insomnia on discontinuation after there was a statistically significant lower longer use is uncertain.45 number of awakenings during weeks 1 to 2, In a large study of consecutive, nightly weeks 3 to 8 showed no improvement in the treatment, zolpidem tartrate 10 mg was number of awakenings for zolpidem tartrate found to significantly decrease SL and 10 mg. It was recommended that patients increase total sleep time during the active receive education during intermittent dos- treatment time period of 28 nights versus ing. Specifically, they should be informed placebo.47 No improvement in the number of that relatively poor sleep could occur on sleep awakenings was seen for zolpidem tar- the nights that the medication is not trate versus placebo, and WASO was not taken compared with nights that the med- assessed. On the first night after discontinu- ication is taken. Other studies have shown ation of 4 weeks of treatment, a statistically similar improvements in sleep onset and significant higher number of patients who sleep maintenance with intermittent dosing had received zolpidem tartrate 10 mg of zolpidem.53-55 showed an increase in SL (P .05) and an In studies in which quality of life has been ≥ increase in number of awakenings (P .01) measured, the impact of zolpidem tartrate ≥ relative to baseline versus those who had on these measures has been equivocal.52-54 received placebo.47 Withdrawal effects (de- In 2 trials, patients reported improved fined as 3 or more new symptoms) on the sleep quality when treated with zolpidem

VOL. 12, NO. 8, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S235 REPORTS

tartrate.53,54 However, another study using important to note that no studies have the 36-item short-form health survey to looked at middle-of-the-night zaleplon dos- document quality-of-life changes across ing beyond 2 nights’ total duration or in eld- various domains has not demonstrated erly individuals with insomnia. any significant improvement with zolpi- A large, multicenter, placebo-controlled dem tartrate treatment compared with trial in adult patients with insomnia indicat- baseline.55 ed favorable effects of zaleplon on SL when Zolpidem tartrate extended-release is given at the recommended 10-mg dose.47 formulated to provide extended plasma con- Effects were maintained throughout the centrations beyond 3 hr after administra- duration of the 4-week study and, in con- tion.34 It is indicated for treatment of trast to the findings with zolpidem tartrate, sleep-onset and/or sleep-maintenance in- there was no evidence of rebound insomnia somnia based on 2 double-blind, random- or withdrawal symptoms after discontinua- ized, placebo-controlled polysomnogram tion when given at the recommended doses. (PSG) studies of 3 weeks’ duration. For the In addition, a number of studies in elderly 12.5-mg daily dose in adults, adverse events patients with insomnia have also demon- exceeding 5% were headache (19% vs 16% strated significant decreases in SL with with placebo), somnolence (15% vs 2% with short-term ( 2 weeks) use of zaleplon.63-65 ≥ placebo), dizziness (12% vs 5% with place- Given the safety and efficacy data in pro- bo), and nausea (7% vs 4% with placebo). For moting sleep initiation, zaleplon appears to the 6.25-mg daily dose in elderly patients, be particularly useful in those whose only adverse effects with a greater than 5% inci- sleep disturbance is difficulty falling asleep dence were headache (14% vs 11% with or for sleep maintenance when the patient is placebo), dizziness (8% vs 3% with placebo), expected to get at least 4 hr of additional somnolence (6% vs 5% with placebo), and sleep. nasopharyngitis (6% vs 4% with placebo).34 Eszopiclone is a cyclopyrrolone deriva- Zaleplon. Classified as a pyrazolopyrimi- tive and an of , a seda- dine derivative, zaleplon has an ultrashort tive hypnotic not marketed in the United elimination t of approximately 1 hr. It can States. Although not fully elucidated, zopi- 1/2 be administered up to 4 hr before the antic- clone, and presumably eszopiclone, show

ipated wake time, when difficulty falling similar binding profiles to the α1-subunit of 56 asleep is experienced. Similar to zolpidem, the GABAA receptor as benzodiazepines, zaleplon preferentially binds to the GABA but with α-subunit specificity and affinity receptor α1-subunit but possesses less affini- lying somewhere between those of zaleplon 66,67 ty for the α2-, α3-, and α5-subunits than does and zolpidem. Of the 6 known α-sub- 25,26 zolpidem. types, α1 and α3 have been associated with Zaleplon, at recommended doses of 10 mg the brain systems that control sleep. In vitro

in adults and 5 mg in the elderly, has been studies with functional recombinant GABAA shown to decrease SL with little to no effects receptors suggest that (R,S)-zopiclone on total sleep time or number of nocturnal (racemic zopiclone) and its S-isomer awakenings.42,56,57 Headache is the most (eszopiclone) display considerable activity

commonly reported adverse effect with rec- at GABAA receptors containing the α3-sub- ommended doses.56 The few studies that type, as well as activity at receptors of the 29,68 have specifically evaluated the residual α1-subtype. effects of middle-of-the-night use of zaleplon With an elimination t of approximately 1/2 have reported little or no next-morning 6 hr, eszopiclone is approved for the treat- memory impairment.58-61 Whether adminis- ment of sleep-onset and sleep-maintenance tered during the day, at bedtime, or in the insomnia, and has demonstrated efficacy and middle of the night, zaleplon was associated safety in patients with primary insomnia with less psychomotor and memory impair- using both PSG and patient-reported meas- ment and more rapid resolution of adverse ures of sleep.30,31,69,70 Improvement in effects than zolpidem tartrate.6,57-62 It is patient-reported next-day function was also

S236 THE AMERICAN JOURNAL OF MANAGED CARE MAY 2006 Strategies for Treating Chronic Insomnia noted.69 Eszopiclone has been studied in generally well tolerated, with both transient and chronic conditions, and unpleasant taste reported more often in showing improvement in sleep onset, sleep the eszopiclone group. maintenance, and total sleep time.30,31,71 It is This initial 6-month controlled study was the first sedative-hypnotic agent to be followed by a 6-month open-label extension approved without a limitation of short-term phase in which all patients were adminis- use only.32 tered eszopiclone 3 mg nightly.31 Overall, The most commonly reported adverse patients randomized to placebo in the first 6 events of eszopiclone versus placebo, admin- months and switched to eszopiclone for the istered at recommended adult (2- and 3-mg) open-label extension experienced improve- doses were headache (21% and 17%, respec- ments in all measures of sleep studied. The tively, vs 13% with placebo), unpleasant improvement was evident at every time taste (17% and 34%, respectively, vs 3% with point throughout the entire 6-month period placebo), somnolence (10% and 8%, respec- without any evidence of loss of therapeutic tively, vs 3% with placebo), dizziness (5% effect or increase in eszopiclone dose. In this and 7%, respectively, vs 4% with placebo), open-label phase of the study, patients (5% and 10%, respectively, vs 3% switched from placebo to eszopiclone with placebo), dry mouth (5% and 7%, reported significant improvements in sleep respectively, vs 3% with placebo), dyspepsia and daytime functioning relative to the final (4% and 5%, respectively, vs 4% with place- month of double-blind treatment. These bo), nausea (5% and 4%, respectively, vs 4% improvements were comparable with those with placebo), nervousness (5% and 0%, experienced and sustained by patients who respectively, vs 3% with placebo), and rash received eszopiclone for all 12 months. (3% and 4%, respectively, vs 1% with place- Eszopiclone was well tolerated throughout bo).32 Infection appeared to be primarily due the 12-month period, and there was no evi- to the common cold in both eszopiclone dence of withdrawal on discontinuation of groups, and taste disturbances were attrib- treatment or any worsening rates of adverse uted to the liquid formulation used in many effects across time. In summary, these stud- studies. In the elderly patient (1- and 2-mg) ies demonstrate that long-term use of doses, the most common adverse events eszopiclone is safe and leads to sustained were headache (15.3% and 15.2%, respec- efficacy without tolerance in the long-term tively, vs 15.0% with placebo), unpleasant management of insomnia.30,31 taste (8.3% and 11.4%, respectively, vs 1.3% Similar results have been observed in with placebo), somnolence (6.9% and 3.8%, elderly patients with chronic insomnia. In respectively, vs 8.8% with placebo), and dys- a multicenter, placebo-controlled study, pepsia (5.6% and 1.3%, respectively, vs 2.5% elderly patients were treated with nightly with placebo).70 eszopiclone 1 or 2 mg over a 2-week peri- Eszopiclone use in the long-term treat- od. Patients receiving the 2-mg dose expe- ment of insomnia was studied in a double- rienced significant improvements in blind, placebo-controlled trial evaluating patient-reported measures of sleep onset, the long-term safety and efficacy of eszopi- WASO, and total sleep time, and in several clone 3 mg administered nightly over a 6- next-day efficacy measures (eg, improve- month period to adults with chronic ments in daytime alertness, morning primary insomnia.30 Eszopiclone demon- sleepiness, daytime ability to function, and strated sustained (week 1 through 6 overall sense of well-being) compared with months) clinically and statistically signifi- those taking placebo.70 The 1-mg group cant improvements in all self-reported had significantly shorter SL compared with measures of sleep initiation, sleep mainte- placebo. Patients receiving the 2-mg dose nance, sleep duration, sleep quality, and who napped also reported a significant daytime functioning. There was no evidence reduction in the median number of naps of physiologic tolerance or diminished effi- and in the 2-week cumulative total nap cacy over the entire 6 months of the study time. In a second double-blind, placebo- for all endpoints measured. Eszopiclone was controlled study, elderly patients with

VOL. 12, NO. 8, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S237 REPORTS chronic insomnia received eszopiclone 2 somnolence (5% vs 3% with placebo), dizzi- mg nightly for 2 weeks.72 Compared with ness (5% vs 3% with placebo), and fatigue placebo, eszopiclone significantly reduced (4% vs 2% with placebo).33 Unlike the other average objective SL and WASO, and pharmacologic options, ramelteon is not a increased average sleep efficiency over the controlled substance.33,73 2-week, double-blind treatment period. Similar to the previous study, in patients Other Drugs Used to Promote Sleep. who napped, the cumulative number of is a triazolopyridine antidepres- naps was reduced relative to placebo. sant with sedating properties that, although not FDA approved for this indication, is Ramelteon. The hypnotic ramelteon is a widely used for the treatment of insomnia.75 agonist with affinity pre- A review of the clinical trials for trazodone

dominantly for melatonin 1 (MT1) and mela- showed that over half of the studies were 33,73 tonin 2 (MT2) receptors. Ramelteon is conducted over a period of 3 weeks or less, indicated for the treatment of sleep-onset and no trial exceeded 6 weeks of active insomnia.33 Clinical trials supporting efficacy treatment.76 The majority of the studies in and safety of ramelteon included two 35- this review involved depressed patients, not night, randomized, double-blind, placebo- those with primary insomnia, and enrolled controlled PSG trials of patients with chronic fewer than 30 patients in the trazodone insomnia: 1 study in adults aged 18 to 64 treatment groups. The most common ad- years (using single, nightly ramelteon 8 or 16 verse events seen at dosages of 75 to 500 mg), and a 3-period crossover trial in subjects mg/day in these studies included drowsi- aged 65 years or older (using ramelteon 4 or ness, dry mouth, nausea, vomiting, constipa- 8 mg or placebo).33 PSG was performed on the tion, headache, and blurred vision. first 2 nights in each of weeks 1, 3, and 5 of Only 1 randomized, double-blind, paral- treatment. In these studies, all doses reduced lel, placebo-controlled study has evaluated SL to persistent sleep (PSG results) compared the benefit of trazodone for patients with with placebo. The second study in subjects primary insomnia.46 This study evaluated aged 65 years or older reported subjective the efficacy of trazodone in nondepressed efficacy measures with 4 or 8 mg or placebo patients with insomnia and was a random- for 35 nights, and both doses reduced SL ized, double-blind, placebo-controlled com- compared with placebo.33 A similarly de- parison of trazodone 50 mg and zolpidem signed study in adults aged 18 to 64 years tartrate 10 mg administered nightly for 2 using 8 and 16 mg of ramelteon did not repli- weeks.46 During the first week, patients in cate the finding of reduced patient-reported both the trazodone and zolpidem tartrate SL compared with placebo.33 In a published groups reported significant improvements in evaluating the efficacy, safety, subjective SL, sleep duration, WASO, and and dose response of ramelteon, patients with sleep quality relative to placebo, but tra- chronic primary insomnia were randomized zodone was significantly less effective than into a double-blind dosing sequence of 4, 8, zolpidem tartrate with respect to self-report- 16, and 32 mg of ramelteon and placebo.74 ed SL and sleep duration. During week 2, the This 5-period crossover study included a 5- to trazodone group did not significantly differ 12-day washout period between treatments, from the placebo group on any of the sleep and patients served as their own controls. All efficacy endpoints. The zolpidem tartrate doses of ramelteon resulted in statistically sig- group, however, continued to report signifi- nificant decreases in latency to persistent cant improvement in SL and sleep duration sleep and increases in total sleep time. No compared with placebo throughout the next-day residual effects and no differences in entire 2-week period but not in WASO dur- the number or type of adverse events were ing the second week. Estimated total sleep apparent at any dose of ramelteon compared time did not significantly differ across with placebo.74 The most frequent adverse groups at any time during the study period. reactions in phase 1 to 3 studies of ramelteon Compared with tricyclic , 8 mg were headache (7% vs 7% with placebo), the cardiovascular risk profile of trazodone

S238 THE AMERICAN JOURNAL OF MANAGED CARE MAY 2006 Strategies for Treating Chronic Insomnia is relatively benign.76 However, a review of was associated with a high study discontinu- trazodone studies in patients with depres- ation rate, impaired next-day functioning, sion revealed reports of hypotension, syn- and impairments in “ease of awakening” and cope, and exacerbation of ischemic attacks “feelings on/after awakening”—key meas- associated with trazodone administra- ures used in the Leeds Sleep Evaluation tion.76,77 In addition, torsades de pointes, Questionnaire.76 characterized by prolongation of the QTc Studies evaluating tolerance associated interval, and other arrhythmias have been with the use of trazodone in primary insom- reported with trazodone administration.78,79 nia are nonexistent; therefore, no conclu- Trazodone is metabolized almost exclusively sions can be reached at this time. In the by the hepatic cytochrome P-450 3A4 study by Walsh et al described previously, (CYP3A4) isoenzyme system and is there- the efficacy relative to placebo observed dur- fore subject to elevation of serum levels ing week 1 was lost at week 2, but this effect when it is coadministered with potent was primarily related to improvements in CYP3A4 inhibitors such as .77 the placebo group rather than decrements in These adverse effects could be of particular the trazodone group.46 concern in elderly individuals and in those with preexisting cardiovascular disease. Over-the-counter (OTC) Products. Al- Because no randomized controlled trials though some patients with insomnia seek and have evaluated trazodone use in elderly follow medical advice in addressing their patients with insomnia, careful considera- sleep disturbances, many choose to self- tion should be given when prescribing tra- medicate with OTC sleep aids. The major zodone or any other hypnotic. Cases of categories of OTC sleep aids include antihis- priapism, related to the α-adrenergic antag- tamines and the dietary supplements onist properties of trazodone, have also been melatonin and . First-generation reported.76,77 syndrome has been (eg, , brom- reported when trazodone, even at low doses, , or chlorpheniramine) induce has been used in combination with other sedation and via central antihist- antidepressants.80 It is important to note aminergic and mecha- that dosages of trazodone used in insomnia nisms.81 Diphenhydramine is the most are typically lower (25-100 mg before bed- commonly used in OTC sleep time) than dosages used to treat aids, and studies evaluating its sedative (100-600 mg/day).76 Because so few studies effects have produced data to support the have been conducted in nondepressed clinical observation of sedation after adminis- patients with insomnia, it is difficult to tration.82,83 Only 1 double-blind, randomized, assess the risk associated with the use of placebo-controlled study evaluating subjec- trazodone in the treatment of insomnia. In tive reports of sleep parameters with nightly clinical trials evaluating trazodone in nonde- administration of diphenhydramine 50 mg pressed patients with insomnia, patients over a 2-week period indicated some im- experienced significant CNS adverse effects provement in SL, sleep duration, quality of (eg, headache and somnolence) versus sleep, and number of awakenings.84 This placebo.46,76 study, however, demonstrated evidence of According to Mendelson’s review, the tolerance, and efficacy was not sustained. majority of studies evaluating the efficacy of The adverse effects of diphenhydramine trazodone in depressed or dysthymic pa- were notable compared with placebo and tients with secondary insomnia reported included tiredness, drowsiness, dizziness, improvements in subjective sleep assess- and grogginess. ments, including sleep onset, sleep quality, The limitations of antihistamines for and sleep duration.76 However, as pointed treating insomnia include a negative out previously, these studies used a wide adverse-effect profile, including changes in range of doses, were small (involving <20 sleep architecture (ie, a reduction in rapid patients), and were not randomized or eye movement [REM] sleep secondary to placebo controlled; in addition, trazodone their anticholinergic effects); next-day seda-

VOL. 12, NO. 8, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S239 REPORTS tion and cognitive impairment; the develop- constituents may differ, depending on when ment of tolerance after only a few days of and where the plant was harvested.95 use; the potential for abuse; and significant drug interactions.81,85 Studies in elderly Comparison of Nonpharmacologic and patients have shown that they are at par- Pharmacologic Interventions ticular risk for cognitive impairment and Few studies have directly compared the altered consciousness associated with separate and combined effects of behav- diphenhydramine use.86,87 ioral and pharmacologic therapies for the Dietary supplements, such as the hormone treatment of insomnia. One meta-analysis melatonin and the herb valerian, appear to compared the short-term efficacy of benzo- have some sedative properties, and many diazepines and nonbenzodiazepines (zolpi- patients with insomnia perceive these prod- dem tartrate and zaleplon) versus behavioral ucts as being a safer or more natural alterna- interventions (primarily stimulus control tive to drugs. However, melatonin and herbal with or without sleep restriction) in the preparations are marketed as dietary supple- short-term treatment of primary insomnia.17 ments, which can be sold without premarket Both interventions demonstrated similar review or approval by the FDA.88 These improvements in SL, number of awakenings preparations, therefore, are not standardized per night, WASO, total sleep time, and sleep regarding active ingredients, nor are they quality. The authors of the meta-analysis screened for possible unapproved ingredients. concluded that, overall, behavioral therapy Melatonin is an endogenous hormone that for persistent primary insomnia was as effec- plays a major role in the circadian sleep tive in the short term (2-4 weeks) as phar- process. It promotes a behavioral state macotherapy.17 The authors also stated that resembling quiet wakefulness, which allows pharmacotherapy may be a choice when for normal sleep initiation, rather than symptom reduction is an immediate require- sleepiness or profound sedation.89 The role ment, with behavioral treatment indicated of melatonin in the treatment of chronic in cases where drugs are contraindicated. insomnia is not entirely clear, but studies When costs were considered, sedative hyp- suggest that melatonin may be helpful in notics were more cost-effective over a 5- those with sleep disturbances associated week period, but the authors indicated that with reduced endogenous melatonin and cir- it was possible that long-term use of behav- cadian disorders.89-92 ioral treatment might reduce indirect costs, Valerian is used in many cultures for its thus offsetting the short-term cost benefit.17 purported calming and relaxing effects that A randomized, placebo-controlled trial help initiate sleep and decrease anxiety in involving adults with chronic sleep-onset those with insomnia.93 The mechanisms of insomnia compared the clinical efficacy of valerian’s effects on sleep are unclear, but cognitive behavioral therapy (CBT) and GABAergic activity has been suggested. A pharmacologic (zolpidem tartrate) therapy, systematic review of randomized clinical tri- alone and in combination.3 The behavioral als evaluating valerian for insomnia found components of the CBT included a combina- great inconsistency in terms of study design, tion of sleep restriction, stimulus control, and the quality of these studies from a and relaxation training. The primary sleep methodological viewpoint is questionable. outcome measured was SL; secondary sleep Overall, the meta-analysis showed contradic- outcomes measured were sleep efficiency, tory and inconclusive results of valerian with total sleep time, and measures of daytime respect to sleep improvement, indicating a functioning. CBT alone produced the great- need for rigorous trials to determine its effica- est improvements in SL and sleep efficiency, cy.94 Clinical trials have not evaluated toler- and the combined pharmacotherapy-CBT ance, dependence, or withdrawal and treatment provided no advantage over CBT rebound symptoms on discontinuation. Be- alone. The positive effects of CBT alone and cause dietary supplements are not required in combination with pharmacotherapy on to be standardized in the United States, the SL and sleep efficiency were maintained at nature of valerian preparations and their long-term follow-up (up to 12 months).

S240 THE AMERICAN JOURNAL OF MANAGED CARE MAY 2006 Strategies for Treating Chronic Insomnia

In a placebo-controlled study in elderly Treatment of Insomnia With Comorbid patients with sleep-maintenance insomnia, Conditions the efficacy of CBT and pharmacotherapy Patients with depression and chronic pain (temazepam), alone or in combination, was syndromes may be at increased risk for the evaluated using sleep efficiency measure- development of chronic sleep disturbances ments.20 All 3 active-treatment groups were that may further increase the severity of the more effective than placebo in improving comorbid condition.15,99 Few studies, howev- sleep efficiency during the treatment peri- er, have evaluated the treatment of insomnia od. Follow-up data obtained at 12 and 24 in patients with comorbid medical or psychi- months after treatment discontinuation atric illnesses. Before the start of treatment showed that only the CBT group maintained of insomnia associated with depression or a sleep efficiency of ≥85%. Both the pharma- chronic pain, underlying sleep disorders, cotherapy-alone group and the combined- such as sleep apnea and restless legs syn- intervention group reported a significant drome, should be ruled out. Insomnia coex- loss of therapeutic benefit at follow-up after isting with other conditions responds discontinuation. similarly to benzodiazepines and nonbenzo- Findings from other studies suggest that diazepines as does primary insomnia.99,100 although pharmacotherapy may produce a Appropriate pharmacologic and nonpharma- more rapid improvement in sleep distur- cologic therapy targeted at the sleep distur- bances compared with CBT interventions, bances themselves may also improve the use of nonpharmacologic therapies may comorbid condition, and vice versa. A care- improve insomnia in as many as 70% to 80% ful evaluation of the risk-benefit ratio for any of cases; however, treatment response can treatment modality employed should be be highly variable.2 Therapeutic gains are carefully considered in patients with insom- often seen after 4 to 8 weeks of therapy for nia and comorbid conditions. pharmacotherapy alone, after CBT interven- Many antidepressants have been linked to tions alone, or after a combination of phar- changes in sleep architecture, particularly macotherapy and CBT.2,3,17,20 However, these delayed onset of REM sleep, reduced amount findings should be evaluated in view of more of REM sleep, and increased slow-wave recent data on the long-term use of pharma- sleep.101 Although trazodone does not pro- cotherapy.31 Furthermore, the long-term duce any of these changes, it may increase clinical effects of CBT appear to be main- REM sleep.15,99 Antidepressants, such as tained after treatment discontinuation, selective serotonin reuptake inhibitors although these results warrant further inves- (SSRIs), venlafaxine, and , are con- tigation. Despite the important findings of sidered to be activating, and benzodiazepine studies comparing pharmacotherapy and receptor agonists or sedating antidepres- CBT, CBT remains underused in clinical sants (eg, trazodone) are often copre- practice, perhaps owing to real and pre- scribed.99 As previously discussed, studies sumed barriers, such as practitioner time have documented improvements in clinical constraints, patient nonadherence, and a parameters of depression and its accompany- reluctance to reimburse or underreim- ing sleep disturbance with trazodone bursement of psychosocial interven- monotherapy.76 In addition to trazodone, tions.96,97 In addition, practitioners may , , , and mir- view treatments such as CBT as being far tazapine have been used for their sedative less cost-effective than pharmacotherapy; effects associated with histamine or serotonin other limitations of CBT include the fact postsynaptic receptor blockade.99 that most family practitioners are not The effects of zolpidem tartrate nightly trained to administer it, the daily time con- for 4 weeks were evaluated in SSRI-treated straints that most family practitioners face, patients with depression in a 6-week, dou- and insurance reimbursement constraints. ble-blind, placebo-controlled study with However, newer abbreviated forms of CBT first- and last-week single-blind placebo peri- may be more cost-effective and easier to ods.102 Compared with placebo, zolpidem administer.12,98 tartrate significantly improved total sleep

VOL. 12, NO. 8, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S241 REPORTS time, reduced the number of awakenings, In patients experiencing chronic pain, and improved self-rated sleep measures, sedative antidepressants may be useful in such as sleep quality, daytime functioning, alleviating the coexisting depression, anxi- improved concentration, and overall well- ety, and/or insomnia.100 Amitriptyline has being. No evidence of improved depression been well studied in pain syndromes; tra- during the period of zolpidem tartrate–SSRI zodone, nefazodone, and can coadministration was reported. The most also be used.100 common adverse events during the double- Before the use of sedative hypnotic agents blind period included headache (33.7%), or antidepressants, nonpharmacologic strate- dysmenorrhea (11.6%), and dyspepsia, som- gies should be considered. Nonpharmacologic nolence, and (9.5% each). Post- methods that have been found to be useful in treatment adverse events reported included patients with rheumatoid arthritis, fibro- headache (13.3%) and dysmenorrhea and myalgia, and chronic fatigue syndrome (4% each). No tolerance developed include CBT, SH, and exercise.15,100 during the 4-week double-blind period. After substitution of placebo for zolpidem, the Summary zolpidem-treated patients reverted to pre- As the primary disorder or a comorbidity treatment insomnia symptoms, with signifi- associated with other disorders, insomnia is cant subjective rebound sleep disturbances a common yet complex problem that can for 1 night. result in serious consequences.6-9 Despite its Chronic pain appears to worsen sleep dif- high prevalence, insomnia is often misun- ficulties, and, in turn, sleep difficulties wors- derstood by both physicians and patients en the perception of the severity of pain.15,100 and, as a result, is not well managed.10 The Triazolam was evaluated in the treatment of management and treatment of insomnia patients with insomnia and comorbid should be multifactorial and include both rheumatoid arthritis.103 This double-blind, nonpharmacologic and pharmacologic inter- placebo-controlled, crossover study found ventions.3,4,17 Nonpharmacologic strategies that nighttime triazolam doses (titrated from should focus on the various psychological, 0.125 to 0.5 mg) significantly improved SL environmental, behavioral, and lifestyle fac- and total sleep time, and decreased the tors that precipitate and perpetuate sleep number of awakenings. Furthermore, day- disturbances.17-19 Studies indicate that non- time alertness and duration of morning stiff- pharmacologic interventions, alone or in ness improved during treatment with combination with pharmacotherapy, can be triazolam.103 Similarly, zolpidem has been effective for insomnia.17 found to improve sleep quality in patients The most common approach to the treat- with fibromyalgia.104 However, large-scale ment of insomnia is pharmacologic inter- studies to support these findings are lacking. vention; pharmacologic options include In patients with rheumatoid arthritis and benzodiazepines, nonbenzodiazepine seda- coexisting insomnia, a double-blind, place- tive-hypnotic agents, melatonin and bo-controlled study showed that eszopi- melatonin receptor agonists, sedating anti- clone significantly improved all sleep depressants, sedating antihistamines, and efficacy measures (SL, WASO, nocturnal herbal products.4,18,96 No single pharmaco- awakenings, total sleep time, and sleep logic agent currently available meets criteria depth).105 Sleep quality was also significant- for the ideal hypnotic.21 Therefore, the ly improved, as were daytime alertness, choice of pharmacologic agent should be ability to function, and ability to concen- based on the type and duration of the sleep trate. Changes in scores on the Arthritis disturbance and the patient’s current med- Self-Efficacy Scale were clinically and sta- ical conditions, age, and concomitant drug tistically significant for overall score, pain, therapy.4,18,96 Furthermore, the pharmacoki- and pain plus other symptoms. Patient netic and the adverse event profiles of the assessment of pain severity, assessed using hypnotic agent should be considered. a Likert scale ranging from 0 to 10, was sig- Treating insomnia in the presence of comor- nificantly reduced in the eszopiclone group. bid conditions, such as depressive disorders

S242 THE AMERICAN JOURNAL OF MANAGED CARE MAY 2006 Strategies for Treating Chronic Insomnia and chronic pain syndromes, may also 18. Sateia MJ, Pigeon WR. Identification and manage- improve the symptoms of the comorbid ment of insomnia. Med Clin North Am. 2004;88:567-596. condition.99,100,103,104 19. Morin CM, Culbert JP, Schwartz SM. Nonpharma- cological interventions for insomnia: a meta-analysis of treatment efficacy. Am J Psychiatry. 1994;151: 1172-1180. REFERENCES 20. Morin CM, Colecchi C, Stone J, Sood R, Brink D. Behavioral and pharmacological therapies for late-life insomnia: a randomized controlled trial. JAMA. 1. American Psychiatric Association. Diagnostic and 1999;281:991-999. Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. Washington, DC: American Psychiatric 21. Mendelson WB, Roth T, Cassella J, et al. The treat- Association; 2000. ment of chronic insomnia: drug indications, chronic use and abuse liability. Summary of a 2001 New Clinical 2. Morin CM, Hauri PJ, Espie CA, Spielman AJ, Buysse Drug Evaluation Unit meeting symposium. Sleep Med DJ, Bootzin RR. Nonpharmacologic treatment of chronic Rev. 2004;8:7-17. insomnia. An American Academy of Sleep Medicine review. Sleep. 1999;22:1134-1156. 22. Charney DS, Mihic SJ, Harris RA. Hypnotics and sedatives. In: Hardman JG, Limbird LE, eds; Gilman AG, 3. Jacobs GD, Pace-Schott EF, Stickgold R, Otto MW. consulting ed. Goodman and Gilman’s The Pharma- Cognitive behavioral therapy and pharmacotherapy for cologic Basis of Therapeutics. 11th ed. New York, NY: insomnia: a randomized controlled trial and direct com- McGraw-Hill; 2006:chap 16. parison. Arch Intern Med. 2004;164:1888-1896. 23. Gottesmann C. GABA mechanisms and sleep. 4. Holbrook AM, Crowther R, Lotter A, Cheng C, Neuroscience. 2002;111:231-239. King D. The diagnosis and management of insomnia in clinical practice: a practical evidence-based approach. 24. Burt DR. Reducing GABA receptors. Life Sci. 2003; CMAJ. 2000;162:216-220. 73:1742-1758. 5. Hauri PJ. Insomnia. Clin Chest Med. 1998;19:157- 25. Mohler H, Fritschy JM, Rudolph U. A new benzodi- 168. azepine pharmacology. J Pharmacol Exp Ther. 2002; 6. Walsh JK. Clinical and socioeconomic correlates of 300:2-8. insomnia. J Clin Psychiatry. 2004;65(suppl 8):13-19. 26. Sanna E, Busonero F, Talani G, et al. Comparison of 7. Ancoli-Israel S, Roth T. Characteristics of insomnia in the effects of zaleplon, zolpidem, and triazolam at various the United States: results of the 1991 National Sleep GABAA receptor subtypes. Eur J Pharmacol. 2002;451: Foundation Survey. I. Sleep. 1999;22(suppl 2):S347-S353. 103-110. 8. Ford DE, Kamerow DB. Epidemiologic study of sleep 27. Seighart W, Sperk G. Subunit composition, distribu- disturbances and psychiatric disorders: an opportunity tion and function of GABAA receptor subtypes. Curr Top for prevention? JAMA. 1989;262:1479-1484. Med Chem. 2002;2:795-816. 9. Ohayon MM. Epidemiology of insomnia: what we 28. George GFP. . Lancet. 2001; know and what we still need to learn. Sleep Med Rev. 358:1623-1626. 2002;6:97-111. 29. Smith AJ, Alder L, Silk J, et al. Effect of α-subunit on 10. National Institutes of Health: Members of the allosteric modulation of ion channel function in stably expressed human recombinant γ-aminobutyric acid re- National Heart, Lung, and Blood Institute Working 36 Group on Insomnia. Insomnia: assessment and manage- ceptors determined using Cl ion flux. Mol Pharmacol. ment in primary care; September 1998. NIH Publication 2001;59:1108-1118. No. 98-4088. Available at: http://www.nhlbi.nih.gov 30. Krystal AD, Walsh JK, Laska E, et al. Sustained effi- /health/prof/sleep/insom_pc.pdf. Accessed February 9, cacy of eszopiclone over 6 months of nightly treatment: 2005. results of a randomized, double-blind, placebo-con- 11. Ohayon MM, Caulet M, Philip P, et al. New epi- trolled study in adults with chronic insomnia. Sleep. demiologic findings about insomnia and its treatment. 2003;26:793-799. J Clin Psychiatry. 1992;53(suppl 12):34-39. 31. Roth T, Walsh JK, Krystal A, Wessel T, Roehrs TA. An 12. Stoller MK. Economic effects of insomnia. Clin Ther. evaluation of the efficacy and safety of eszopiclone over 1994;16:873-897. 12 months in patients with chronic primary insomnia. Sleep Med. 2005;6:487-495. 13. Dorsey CM, Lee KA, Scharf MB. Effect of zolpidem on sleep in women with premenopausal and post- 32. Lunesta [package insert]. Marlborough, Mass: menopausal insomnia: a 4-week, randomized, multicen- Sepracor Inc; 2004. ter, double-blind, placebo-controlled study. Clin Ther. 33. Rozerem [package insert]. Lincolnshire, Ill: Takeda 2004;26:1578-1586. Pharmaceuticals America, Inc; 2005. 14. Montgomery P, Dennis J. A systematic review of 34. Ambien CR [package insert]. New York, NY: - non-pharmacological therapies for sleep problems later Synthelabo Inc; 2005. in life. Sleep Med Rev. 2004;8:47-62. 35. McKernan RM, Rosahl TW, Reynolds DS, et al. 15. Benca RM, Ancoli-Israel A, Moldofsky H. Special Sedative but not properties of benzodi- considerations in insomnia diagnosis and management: azepines are mediated by the GABAA receptor α1 sub- depressed, elderly, and chronic pain populations. J Clin type. Nat Neurosci. 2000;3:587-592. Psychiatry. 2004;65(suppl 8):26-35. 36. Carlson JN, Haskew R, Wacker J, Maisonneuve IM, 16. Stepanski EJ, Wyatt JK. Use of sleep hygiene in the Glick SD, Jerussi TP. Sedative and anxiolytic effects treatment of insomnia. Sleep Med Rev. 2003;7:215-225. of zopiclone’s and metabolite. Eur J 17. Smith MT, Perlis ML, Park A, et al. Comparative Pharmacol. 2001;415:181-189. meta-analysis of pharmacotherapy and behavior therapy 37. Nowell PD, Mazumdar S, Buysse DJ, Dew MA, for persistent insomnia. Am J Psychiatry. 2002;159:5-11. Reynolds CF III, Kupfer DJ. Benzodiazepines and zolpi-

VOL. 12, NO. 8, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S243 REPORTS dem for chronic insomnia: a meta-analysis of treatment outpatient study. Int Clin Psychopharmacol. 2002; efficacy. JAMA. 1997;278:2170-2177. 17:9-17. 38. Holbrook AM, Crowther R, Lotter A, Cheng C, 56. Sonata [package insert]. Philadelphia, Pa: Wyeth King D. Meta-analysis of benzodiazepine use in the Pharmaceuticals Inc; 2003. treatment of insomnia. CMAJ. 2000;162:225-233. 57. Barbera J, Shapiro C. Benefit-risk assessment of 39. Chouinard G. Issues in the clinical use of benzodi- zaleplon in the treatment of insomnia. Drug Saf. azepines: potency, withdrawal, and rebound. J Clin 2005;28:301-318. Psychiatry. 2004;65(suppl 5):7-12. 58. Vermeeren A, Danjou PE, O’Hanlon JF. Residual 40. Woods JH, Winger G. Current benzodiazepine effects of evening and middle-of-the-night administration issues. Psychopharmacology (Berl). 1995;118:107-115. of zaleplon 10 mg and 20 mg on memory and actual 41. de las Cuevas C, Sanz E, de al Fuente J. driving performance. Hum Psychopharmacol Clin Exp. Benzodiazepines: more “behavioral” addiction than 1998;13:98-107. dependence. Psychopharmacology (Berl). 59. Danjou P, Paty I, Fruncillo R, et al. A comparison of 2003;167:297-303. the residual effects of zaleplon and zolpidem following 42. Sanger DJ. The pharmacology and mechanisms of administration 5 to 2 h before awakening. Br J Clin action of new generation, non-benzodiazepine hypnotic Pharmacol. 1999;48:367-374. agents. CNS Drugs. 2004;18(suppl 1):9-15. 60. Beer B, Ieni JR, Wu WH, et al. A placebo-controlled 43. Ambien [package insert]. New York, NY: Sanofi- evaluation of single, escalating doses of CL 284,846, a Synthelabo Inc; 2004. non-benzodiazepine hypnotic. J Clin Pharmacol. 1994; 44. Terzano MG, Rossi M, Palomba V, Smerieri A, 34:335-344. Parrino L. New drugs for insomnia: comparative tolera- 61. Stone BM, Turner C, Mills SL, et al. Noise-induced bility of zopiclone, zolpidem, and zaleplon. Drug Saf. sleep maintenance insomnia: hypnotic and residual effects 2003;26:261-282. of zaleplon. Br J Clin Pharmacol. 2002;53:196-202. 45. Scharf MB, Roth T, Vogel GW, Walsh JK. A multi- 62. Troy SM, Lucki I, Unruh MA, et al. Comparison of center, placebo-controlled study evaluating zolpidem in the effects of zaleplon, zolpidem, and triazolam on the treatment of chronic insomnia. J Clin Psychiatry. memory, learning, and psychomotor performance. J Clin 1994;55:192-199. Psychopharmacol. 2000;20:328-337. 46. Walsh JK, Erman M, Erwin CW, et al. Subjective 63. Hedner J, Yaeche R, Emilien G, Farr I, Salinas E. hypnotic efficacy of trazodone and zolpidem in Zaleplon shortens subjective sleep latency and improves DSM III-R primary insomnia. Hum Psychopharmacol. subjective sleep quality in elderly patients with insom- 1998;13:191-198. nia. The Zaleplon Clinical Investigator Study Group. Int J 47. Elie R, Ruther E, Farr I, Emilien G, Salinas E. Sleep Geriatr Psychiatry. 2000;15:704-712. latency is shortened during 4 weeks of treatment with 64. Walsh JK, Fry J, Richardson GS, et al. Short-term zaleplon, a novel nonbenzodiazepine hypnotic. efficacy of zaleplon in older patients with chronic Zaleplon Clinical Study Group. J Clin Psychiatry. insomnia. Clin Drug Invest. 2000;20:143-149. 1999;60:536-544. 65. Ancoli-Israel S, Walsh JK, Mangano RM, 48. Fry J, Scharf M, Mangano R, Fujimori M. Zaleplon Fujimori M. Zaleplon, a novel nonbenzodiazepine hyp- improves sleep without producing rebound effects in notic, effectively treats insomnia in elderly patients with- outpatients with insomnia. Zaleplon Clinical Study out causing rebound effects. Prim Care Companion J Group. Int Clin Psychopharmacol. 2000;15:141-152. Clin Psychiatry. 1999;1:114-120. 49. Leppik IE, Roth-Schechter GB, Gray GW, et al. Double-blind, placebo-controlled comparison of zolpi- 66. Damgen K, Luddens H. Zaleplon displays selectivity dem, triazolam, and temazepam in elderly patients with to recombinant GABAA receptors different from zolpi- insomnia. Drug Dev Res. 1997;40:230-237. dem, zopiclone and benzodiazepines. Neurosci Res Comm. 1999;25:139-148. 50. Dockhorn RJ, Dockhorn DW. Zolpidem in the treat- ment of short-term insomnia: a randomized, double- 67. Davies MJ, Newell G, Derry JM, Martin IL, blind, placebo-controlled clinical trial. Clin Dunn SMJ. Characterization of the interaction of zopi- Neuropharmacol. 1996;19:333-340. clone with γ-aminobutyric acid type A receptors. Mol Pharmacol. 2000;58:756-762. 51. Ware JC, Walsh JK, Scharf MB, Roehrs T, Roth T, Vogel GW. Minimal rebound insomnia after treatment 68. Data on file. Sepracor Inc. with 10-mg zolpidem. Clin Neuropharmacol. 1997;20: 69. Zammit GK, McNabb LJ, Caron, Amato DA, Roth T. 116-125. Efficacy and safety of eszopiclone across 6 weeks of 52. Walsh JK, Roth T, Randazzo A, et al. Eight weeks of treatment for primary insomnia. Curr Med Res Opin. non-nightly use of zolpidem for primary insomnia. 2004;20:1979-1991. Sleep. 2000;23:1087-1096. 70. Scharf M, Erman M, Rosenberg R, et al. A 2-week 53. Cluydts R, Peeters K, de Bouyalsky I, Lavoisy J. efficacy and safety study of eszopiclone in elderly pa- Comparison of continuous versus intermittent adminis- tients with primary insomnia. Sleep. 2005;28:720-727. tration of zolpidem in chronic insomniacs: a double- 71. Rosenberg R, Caron J, Roth T, Amato D. An assess- blind, randomized pilot study. J Int Med Res. ment of the efficacy and safety of eszopiclone in the 1998;26:13-24. treatment of transient insomnia in healthy subjects. Sleep 54. Allain H, Arbus L, Schuck S, et al. Efficacy and safe- Med. 2005;6:15-22. ty of zolpidem administered “as needed” in primary 72. McCall WV, Rosenberg R, Caron J. Polysomno- insomnia: results of a double-blind, placebo-controlled graphic evaluation of the efficacy and safety of eszopi- study. Clin Drug Invest. 2001;21:391-400. clone in elderly patients with chronic insomnia. In: 55. Hajak G, Cluydts R, Declerck A, et al. Continuous Program and abstracts of the 157th Annual Meeting of versus non-nightly use of zolpidem in chronic insom- the American Psychiatric Association; May 1-6, 2004; nia: results of a large-scale, double-blind, randomized, New York, NY. Abstract Nos. 318-319.

S244 THE AMERICAN JOURNAL OF MANAGED CARE MAY 2006 Strategies for Treating Chronic Insomnia 73. McGechan A, Wellington K. Ramelteon. CNS 89. Stone BM, Turner C, Mills SL, Nicholson AN. Drugs. 2005;19:1057-1065. Hypnotic activity of melatonin. Sleep. 2000;23:663-669. 74. Erman M, Seiden D, Zammitt G, Sainati S, Zhang J. 90. Zhdanova IV, Wurtman RJ, Regan MM, Taylor JA, An efficacy, safety, and dose response study of ramelteon Shi JP, Leclair OU. Melatonin treatment for age-related in patients with chronic primary insomnia. Sleep Med. insomnia. J Clin Endocrinol Metab. 2001;86:4727-4730. 2006;7:17-24. 91. Andrade C, Srihari BS, Reddy KP, Chandramma L. 75. Walsh JK, Schweitzer PK. Ten-year trends in the Melatonin in medically ill patients with insomnia: a dou- pharmacological treatment of insomnia. Sleep. ble-blind, placebo-controlled study. J Clin Psychiatry. 1999;22:371-375. 2001;62:41-45. 76. Mendelson WB. A review of the evidence for the 92. Almeida Montes LG, Ontiveros Uribe MP, Cortes efficacy and safety of trazodone in insomnia. J Clin Sotres J, Heinze Martin G. Treatment of primary insom- Psychiatry. 2005;66:469-476. nia with melatonin: a double-blind, placebo-controlled, 77. Desyrel [package insert]. Princeton, NJ: Bristol-Myers crossover study. J Psychiatry Neurosci. 2003;28: Squibb Company; 2005. 191-196. 78. Mazur A, Strasberg B, Kusniec J, Sclarovsky S. QT 93. Gyllenhaal C, Merritt SL, Peterson SD, Block KI, prolongation and polymorphous ventricular tachycardia Gochenour T. Efficacy and safety of herbal associated with trazodone- combination. Int and sedatives in sleep disorders. Sleep Med Rev. J Cardiol. 1995;52:27-29. 2000;4:229-251. 79. Pohl R, Bridges M, Rainey JM Jr, Boudoulas H, 94. Stevinson C, Ernst E. Valerian for insomnia: a sys- Yeragani VK. Effects of trazodone and desipramine on tematic review of randomized clinical trials. Sleep Med. cardiac rate and rhythm in a patient with preexisting car- 2000;1:91-99. diovascular disease [letter]. J Clin Psychopharmacol. 95. Hadley S, Petry JJ. Valerian. Am Fam Physician. 1986;6:380-381. 2003;67:1755-1758. 80. Margolese H, Chouinard G. Serotonin syndrome 96. Benca RM. Diagnosis and treatment of chronic from addition of low-dose trazodone to nefazodone. Am insomnia: a review. Psychiatr Serv. 2005;56:332-343. J Psychiatry. 2000;157:1022. 97. Integration of behavioral and relaxation approaches 81. Skidgel RA, Erdos EG. Histamine, bradykinin, and into the treatment of chronic pain and insomnia. NIH their antagonists. In: Hardman JG, Limbird LE, eds; Technol Assess Statement. October 16-18, 1985:1-34. Gilman AG, consulting ed. Goodman and Gilman’s The 98. Edinger JD, Sampson WS. A primary care “friendly” Pharmacologic Basis of Therapeutics. 11th ed. New cognitive behavioral insomnia therapy. Sleep. 2003;26: York, NY: McGraw-Hill; 2006:chap 24. 177-182. 82. Bender BG, Berning S, Dudden R, Milgrom H, 99. Jindal RD, Thase ME. Treatment of insomnia associ- Tran ZV. Sedation and performance impairment of di- ated with clinical depression. Sleep Med Rev. 2004; phenhydramine and second-generation antihistamines: a 8:19-30. meta-analysis. J Allergy Clin Immunol. 2003;111:770-776. 100. Stiefel F, Stagno D. Management of insomnia in pa- 83. Roth T, Roehrs T, Koshorek G, Sicklesteel J, tients with chronic pain conditions. CNS Drugs. 2004; Zorick F. Sedative effects of antihistamines. J Allergy 18:285-296. Clin Immunol. 1987;80:94-98. 101. Tsuno N, Besset A, Ritchie K. Sleep and depres- 84. Rickels K, Morris RJ, Newman H, Rosenfeld H, sion. J Clin Psychiatry. 2005;66:1254-1269. Schiller H, Weinstock R. Diphenhydramine in insomni- 102. Asnis GM, Chakraburtty A, DuBoff EA, et al. ac family practice patients: a double-blind study. J Clin Zolpidem for persistent insomnia in SSRI-treated Pharmacol. 1983;23:234-242. depressed patients. J Clin Psychiatry. 1999;60: 85. Richardson GS, Roehrs TA, Rosenthal L, Koshorek 668-676. G, Roth T. Tolerance to daytime sedative effects of H1 103. Walsh JK, Muehlbach MJ, Lauter SA, Hilliker NA, antihistamines. J Clin Psychopharmacol. 2002;22: Schweitzer PK. Effects of triazolam on sleep, daytime 511-515. sleepiness, and morning stiffness in patients with rheuma- 86. Basu R, Dodge H, Stoehr GP, Ganguli M. Sedative- toid arthritis. J Rheumatol. 1996;23:245-252. hypnotic use of diphenhydramine in a rural, older adult, 104. Moldofsky H, Lue FA, Mously C, Roth-Schechter community-based cohort: effects on cognition. Am J B, Reynolds WJ. The effect of zolpidem in patients with Geriatr Psychiatry. 2003;11:205-213. fibromyalgia: a dose ranging, double-blind, placebo-con- 87. Agostini JV, Leo-Summers LS, Inouye SK. Cognitive trolled, modified crossover study. J Rheumatol. and other adverse effects of diphenhydramine use in 1996;23:529-533. hospitalized older patients. Arch Intern Med. 2001;161: 105. Schnitzer T, Rubens R, Price J, et al. The effect of 2091-2097. eszopiclone 3 mg compared with placebo in patients 88. Hathcock J. Dietary supplements: how they are used with rheumatoid arthritis and co-existing insomnia. and regulated. J Nutr. 2001;131:1114S-1117S. Arthritis Rheum. 2005;52(9 suppl):S352. Abstract.

VOL. 12, NO. 8, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S245