Alexander DeLuca, M.D.
Addiction, Pain, & Public Health website  
[Home] [Library]  [Slides]  [Search]  [Medline]  [Links]
Statement of Purpose; Privacy policy; Statement of Ownership; Statement of Sponsorship; Advertising policy 

Treating Insomnia in Patients With Substance Use/Abuse Disorders

by Maher Karam-Hage, M.D.; Psychiatric Times; February 2004; Vol. XXI; Issue 2. Originally posted 7/24/2004.. []. This document also available from: <>

[Print version]

Despite the fact that about 30% of our life is spent sleeping and decades of research have been spent on sleep, we still do not know its real function. What we do know is lack of sleep can have serious implications, such as increased risk of depressive disorders, impaired breathing and heart disease. On the other hand, nighttime sleep disturbance is usually followed by excessive daytime sleepiness that is associated with delayed problems like memory deficits and impaired social and occupational function, and immediate consequences such as car accidents (Kupfer and Reynolds, 1997; Roehrs and Roth, 1995). People who abuse alcohol and other substances are at high risk for sleep disturbances due to the direct effect of the substance or its withdrawal on their sleep architecture and their sleep-wake cycle or its effect on their behavior and daily functioning, which in turn impacts their daily need for sleep.

Two states of sleep alternate throughout the night, characterized in part by different types of brain electrical activity (National Institute on Alcohol Abuse and Alcoholism [NIAAA], 1998). There is slow wave sleep (SWS), during which the brain waves are very slow (commonly referred to as Stage III and IV), and rapid eye movement (REM) sleep, in which the eyes undergo rapid movements although we remain asleep. Restful SWS has the lion's share of sleep time, while REM sleep occurs periodically, resulting in about 25% of sleep time in the young adult. Normally present every 90 minutes, the REM stage lasts five to 30 minutes and is associated with dreaming, but no clear function is known for it. There are about four occurrences of REM in total, the first is shortest and the last is usually longest (NIAAA, 1998).

Effect of Alcohol and Substances of Abuse on Sleep

Little is known about how the different substances of abuse affect sleep in humans, although there are more data on alcohol's effect. When consumed at bedtime, alcohol has an initial stimulating effect among nonalcoholics, followed by a decrease in time to fall asleep (NIAAA, 1998). Many people with insomnia consume alcohol to induce sleep, either by experience or by others' suggestion that it is a sedating agent. Alcohol consumed six hours before bedtime was found to disrupt the second half of the sleep period (Landolt et al., 1996). One review suggested that with continued consumption until bedtime, alcohol's disruptive effects continued or increased and its sleep-inducing effect may decrease (Vitiello, 1997).

In actively drinking alcoholics, specific sleep disturbances are reported, such as increased time required to fall asleep, frequent awakenings and a decrease in subjective sleep quality associated with daytime fatigue (Aldrich, 1998). Further, these individuals undergo a vicious cycle when they attempt to stop drinking since an abrupt reduction or end to drinking usually triggers alcohol-withdrawal syndrome accompanied by pronounced insomnia and sleep fragmentation. Decreased SWS during withdrawal may reduce the amount of restful sleep. Beyond withdrawal, sleep patterns may never return to normal in people with alcoholism (Aldrich, 1998). After years of abstinence, alcoholics tend to sleep poorly, with decreased amounts of SWS and increased nighttime wakefulness contributing to daytime fatigue. When heavy drinking recurs, it leads to increased SWS (restful sleep) and decreased wakefulness. This apparent improvement in sleep continuity may promote continued drinking by associating the return to drinking with improved sleep (NIAAA, 1998). Unfortunately, as drinking continues, sleep patterns get disrupted, closing the cycle (Aldrich, 1998).

[Top of Page]

Clinical Consequences of Insomnia on Abstinence

When compared to patients without insomnia, patients with insomnia were more likely to report frequent alcohol use for sleep (55% versus 28%), had significantly worse polysomnographic measures of sleep continuity, and had more severe alcohol dependence and depression (Brower et al., 2001). In 1994, Gillin et al. measured REM sleep during the admission of patients to a one-month alcoholism treatment program. Higher levels of REM predicted relapse within three months after hospital discharge in 80% of patients. In other studies, those who eventually relapsed exhibited a higher proportion of REM and a lower proportion of SWS at baseline, compared with those who remained abstinent. It is believed that sleep problems in alcoholics increase rates of relapse as evidenced by subjective and polysomnographic sleep predictors (Brower et al., 1998).

Treatment for Insomnia in Substance Users

If sleep problems lead to relapse, then treatment of those problems should improve relapse rates (Brower, 2001). Similar to the general patient, those who have alcohol- and substance use-associated insomnia are presented with three options: behavioral treatments, over-the-counter (OTC) medications or prescription medications (Table).

Behavioral treatments.
No particular behavioral technique has been validated or proven to be superior than others in the substance user. In one study, 10 sessions of progressive relaxation training over a two-week period were found to improve insomnia among institutionalized alcoholics (Greeff and Conradie, 1998). However, in a meta-analysis, no overall difference was found between behavioral approaches and pharmacotherapy for short-term treatment outcome, except that behavioral therapy resulted in greater reduction of sleep latency (time to sleep onset) (Smith et al., 2002).

Over-the-counter medications.
Most OTC remedies for insomnia contain a certain dose (around 25 mg) of the popular antihistaminic-agent diphenhydramine (Benadryl). Due to its deleterious effect on natural sleep's stages and its anticholinergic effects, which cause morning drowsiness and grogginess, diphenhydramine is not a good treatment for insomnia. In lab animals, at a dose slightly above what is effective therapeutically, diphenhydramine markedly affected behavior and all sleep stages (Marzanatti et al., 1989). In particular, it depressed REM and increased SWS (drowsiness). These agents are not recommended in people who use alcohol and other substances, as drowsiness and grogginess might be associated with their use as well.

Melatonin, an OTC food supplement, is a brain chemical implicated in sleep regulation that has been used with mixed results and seems to be more successful among people with low endogenous (internal) levels. In 1996, Schmitz et al. suggested that melatonin is decreased in alcoholics--a theory that deserves further study.

Valerian (Valeriana officinalis) is an herbal product available over the counter that improves subjective experiences of sleep when taken nightly over one- to two-week periods. It appears to be a safe sedative/hypnotic choice in patients with mild-to-moderate insomnia (Hadley and Petry, 2003). Nevertheless, no studies among substance users have been reported to date.

 [Top of Page]

Prescription medications.
Prescription medications for sleep disturbances have traditionally belonged to the benzodiazepine group. Their use has been controversial in the substance user, specifically the alcoholic patient (Ciraulo and Nace, 2000). In part, this is due to benzodiazepine's nonselective action at two central g-aminobutyric acid (GABA)A receptor sites: w(1) and w(2). The sedative action of benzodiazepines is related to w(1) site, whereas w(2) sites are thought to affect memory and cognitive functioning. Since alcohol is a GABA-ergic drug, these medications have a high liability for misuse, abuse and dependence by patients who are alcoholics. The ideal pharmacotherapy for this group of patients would need to be sedating, have a short half-life, possess little or no interaction with other prescriptions, have minimal or no liver metabolism, and should not induce reward, which would cause it to have low abuse potential.

A new group of benzodiazepine-like medications have been promoted as the ideal drugs due to their selective action on the w(1) subunit of the GABAA receptor. These include zolpidem (Ambien), zaleplon (Sonata) and zopiclone. Unfortunately, several case reports and case series have referred to abuse of the w(1) subunit drugs. Especially at high doses, these drugs become undistinguishable from benzodiazepines (Liappas et al., 2003). In a controlled environment, one study examined the acute behavioral effects and abuse potential of three drugs commonly used to treat sleep disorders (trazodone [Desyrel], zolpidem and triazolam [Halcion]) and placebo in 10 male volunteers with histories of alcohol and drug abuse (Rush et al., 1999). The measured effects of trazodone on subject-rated items and therefore abuse potential (e.g., subject ratings of "Willing to Take Again") were less than those observed with triazolam. Zolpidem and triazolam produced comparable effects on these measures. The authors suggested that trazodone has less abuse potential than triazolam and may be a viable alternative to benzodiazepine hypnotics in individuals with histories of alcohol or drug abuse. The main drawbacks of trazodone are that it causes morning drowsiness and has a one in 5,000 risk of inducing priapism, which renders it risky, considering that at least two-thirds of alcoholics in treatment are men.

Since 1998, my colleagues and I found gabapentin (Neurontin) to fit the above description of the ideal pharmacotherapy for insomnia among alcoholics and substance users (Karam-Hage and Brower, 2000). To date, we have yet to observe any abuse or subjective effects reported by patients. In 2000, we reported an open-label study of 15 successfully treated cases. Gabapentin was started at 300 mg hs, the dose was increased to response by 300 mg/day to reach a maximum of 1800 hs (average dose=900 mg) with follow-up at one month.

Antidepressant drugs with 5-HT2 blocking properties, such as mirtazapine (Remeron) or nefazodone (Serzone), are believed to alleviate insomnia and improve sleep architecture. In patients with depression, mirtazapine produces a significant shortening of sleep-onset latency, increases total sleep time and leads to a marked improvement in sleep efficiency (Thase, 1999). No data are available on nefazodone, mirtazapine or other sedating antidepressants among substance users. My colleagues and I reported a comparison of trazodone (25 mg to 100 mg, average dose=75 mg hs) to gabapentin (300 mg to 1800 mg, average dose=1200 mg hs) (Karam-Hage and Brower, 2003). We found the latter to be significantly superior to trazodone on the overall measure and on four of the five sleep questions asked to patients. Most recently, our group conducted a double-blind, placebo-controlled study with gabapentin among alcoholics who continued to have insomnia after one week to one month of abstinence from alcohol (Brower et al., 2003). Gabapentin faired better than placebo not only on sleep improvement compared with baseline but it also had a direct positive effect on preventing relapse to heavy drinking, independent from its effect on sleep.

Sedating atypical antipsychotics (olanzapine [Zyprexa] and quetiapine [Seroquel]) have been suggested as alternative sleep aids for the substance user, they may be useful due to their sedative effect and impact on reducing background anxiety. Other possibilities, like newer mood-stabilizing agents with sedating qualities (e.g., topiramate [Topamax]), need to be explored. Finally, experimental agents like CEE-03-310 (a selective dopamine D1-like receptor antagonist in Phase II development) have demonstrated a dose-dependent enhancement of non-REM sleep at the beginning of the night without any effect on the quantity of REM sleep (Eder, 2002). These effects could be of benefit for substance users/alcoholics.

Dr. Karam-Hage is clinical assistant professor of psychiatry at the University of Michigan. He is also medical director of Chelsea Arbor Addiction Treatment Center, a joint program between the University of Michigan and Chelsea Community Hospital.

 [Top of Page]


Aldrich MS (1998), Effects of alcohol on sleep. In: Alcohol Problems and Aging. NIAAA Research Monograph No. 33. Lisansky Gomberg ES, Hegedus AM, Zucker RA, eds. NIH Pub. No. 98-4163. Bethesda, Md.: NIAAA.

Brower KJ (2001), Alcohol's effects on sleep in alcoholics. Alcohol Res Health 25(2):110-125.

Brower KJ, Aldrich MS, Hall JM (1998), Polysomnographic and subjective sleep predictors of alcoholic relapse. Alcohol Clin Exp Res 22(8):1864-1871.

Brower KJ, Aldrich MS, Robinson EA et al. (2001), Insomnia, self-medication, and relapse to alcoholism. Am J Psychiatry 158(3):399-404.

Brower KJ, Kim HM, Karam-Hage M et al. (2003), A double-blind randomized clinical trial of gabapentin vs. placebo for treating alcohol dependence. Biol Psychiatry 53(8S):84S-85S.

Ciraulo DA, Nace EP (2000), Benzodiazepine treatment of anxiety or insomnia in substance abuse patients. Am J Addict 9(4):276-279; discussion 280-284.

Eder DN (2002), CEE-03-310 CeNeS pharmaceuticals. Curr Opin Investig Drugs 3(2):284-288.

Gillin JC, Smith TL, Irwin M et al. (1994), Increased pressure for rapid eye movement sleep at time of hospital admission predicts relapse in nondepressed patients with primary alcoholism at 3-month follow-up. Arch Gen Psychiatry 51(3):189-197.

Greeff AP, Conradie WS (1998), Use of progressive relaxation training for chronic alcoholics with insomnia. Psychol Rep 82(2):407-412.

Hadley S, Petry JJ (2003), Valerian. Am Fam Physician 67(8):1755-1758.

Karam-Hage M, Brower KJ (2000), Gabapentin treatment for insomnia associated with alcohol dependence. Am J Psychiatry 157(1):151 [letter].

Karam-Hage M, Brower KJ (2003), Open pilot study of gabapentin versus trazodone to treat insomnia in alcoholic outpatients. Psychiatry Clin Neurosci 57(5):542-544.

Kupfer DJ, Reynolds CF 3rd (1997), Management of insomnia. N Engl J Med 336(5):341-346 [see comment].

Landolt HP, Roch C, Dijk DJ, Borbely AA (1996), Late-afternoon ethanol intake affects nocturnal sleep and the sleep EEG in middle-aged men. J Clin Psychopharmacol 16(6):428-436.

Liappas IA, Malitas PN, Dimopoulos NP et al. (2003), Zolpidem dependence case series: possible neurobiological mechanisms and clinical management. J Psychopharmacol 17(1):131-135.

Marzanatti M, Monopoli A, Trampus M, Ongini E (1989), Effects of nonsedating histamine H1-antagonists on EEG activity and behavior in the cat. Pharmacol Biochem Behav 32(4):861-866.

NIAAA (1998), Alcohol Alert: Alcohol and Sleep. Pub 98-41. Available at: AccessedJan. 8, 2004.

Roehrs T, Roth T (1995), Alcohol-induced sleepiness and memory function. Alcohol Health Res World 19(2):130-135.

Rush CR, Baker RW, Wright K (1999), Acute behavioral effects and abuse potential of trazodone, zolpidem and triazolam in humans. Psychopharmacology (Berl) 144(3):220-233.

Schmitz MM, Sepandj A, Pichler PM, Rudas S (1996), Disrupted melatonin-secretion during alcohol withdrawal. Prog Neuropsychopharmacol Biol Psychiatry 20(6):983-995.

Smith MT, Perlis ML, Park A et al. (2002), Comparative meta-analysis of pharmacotherapy and behavior therapy for persistent insomnia. Am J Psychiatry 159(11):5-11.

Thase ME (1999), Antidepressant treatment of the depressed patient with insomnia. J Clin Psychiatry 60(suppl 17):28-31; discussion 46-48.

Vitiello MV (1997), Sleep, alcohol and alcohol abuse. Addict Biol 2(2):151-158.

 [Top of Page]



Dr. DeLuca's Addiction, Pain, and Public Health Website

Alexander DeLuca, M.D., FASAM.

[Top of Page]

Originally posted:  7/24/2004

All website Email to:

Statement of Purpose; Privacy policy; Statements of Confidentiality, Ownership, & Sponsorship; Advertising policy

Most recently revised: 7/24/2004
Copyright 2003 All rights reserved.