The highest incidence of relapse in patients with alcohol dependence occurs during the first few months following cessation of drinking. In the United States, currently available drug therapies for alcohol dependence include aversion therapy with disulfiram, and naltrexone, an opioid antagonist. Naltrexone has been moderately successful in some small trials, but dose-dependent liver toxicity and a high incidence of nausea limit its use. Acamprosate, available internationally, is being studied in the United States as a non-aversive pharmacotherapeutic agent. Nalmefene is a newer opioid antagonist that, like naltrexone, has no agonist activity and no abuse potential. Additional advantages include a longer half-life, greater bioavailability and no dose-dependent liver toxicity. Mason and colleagues performed a randomized trial of nalmefene in patients who were alcohol-dependent.

Following recruitment through telephone interviews, patients were given a two-week trial of a placebo medication while eligibility criteria were assessed. The primary criteria required that patients be between 18 and 65 years of age and have a Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R) diagnosis of alcohol dependence. At baseline, all patients underwent electrocardiography, laboratory tests (including determination of gamma-glutamyltransferase level) and a urine toxicology screen for drug abuse. Exclusion criteria included a history of illicit drug dependence, current use of narcotic or psychotropic medications, history of cirrhosis or baseline liver function studies more than twice the normal levels.

Patients meeting eligibility criteria were randomized to receive nalmefene in a dosage of 20 mg per day or 80 mg per day, or placebo. All tablets were identical in appearance. Compliance was monitored by a pill count at each weekly visit and the use of a cap that contained an electronic device to record the date and number of times the bottle was opened. In addition to medical follow-up, all participants received cognitive-behavioral therapy during the 12-week study. Data collected included the quantity and frequency of drinking from 90 days before the study through the 12-week medication/placebo phase. Two different measures of alcohol craving were assessed at each follow-up visit. Outcomes assessed included the rate of relapse to heavy drinking (more than six drinks per day in men and more than four drinks per day in women), percentage of days abstinent and change in the number of drinks consumed per day during the study.

The 105 patients enrolled were randomized into two groups: 70 in the nalmefene group and 35 in the placebo group. In both groups, the average age was approximately 42 years. Approximately two-thirds of participants were men. The rate of medication compliance was 89.9 percent based on return pill counts. During the first week of the study, 34.3 percent of the patients receiving placebo reported heavy drinking compared with 15.7 percent of patients receiving nalmefene. Overall, 58.9 percent of the placebo group relapsed during the study compared with 37.1 percent in the nalmefene group. The chance of relapse was 2.4 times greater in patients receiving placebo than in those receiving nalmefene. Other findings in the patients receiving nalmefene included shorter durations of relapse, a greater number of days until first relapse and fewer drinks consumed on the days that a relapse occurred. None of these differences, however, reached statistical significance. No significant difference was noted in the total number of abstinent days among all patients or in the self-reported measures of alcohol craving. Adverse events most commonly reported included headaches, followed by insomnia, fatigue and nausea, with only nausea occurring much more often in the treatment group.

The authors conclude that nalmefene is effective for the prevention of relapse in alcohol-dependent patients. It should be considered a first-line therapy when it becomes routinely available in the United States. The observed treatment response was prompt, but more than one third of participants receiving nalmefene relapsed, despite weekly follow-up visits and cognitive behavioral therapy.