Objectives:

To conduct a systematic review and meta-analysis of the efficacy, comparative effectiveness, and harms of medications (both FDA approved and others) for adults with alcohol-use disorders, and to evaluate the evidence from primary care settings.

Data sources:

PubMed^®, Cochrane Library, PsycINFO^®, CINAHL^®, Embase^®, U.S. Food and Drug Administration Web site, ClinicalTrials.gov, and World Health Organization International Clinical Trials Registry Platform (January 1, 1970, to October 11, 2013).

Review methods:

Two investigators independently selected, extracted data from, and rated risk of bias of studies. We conducted meta-analyses using random-effects models. We graded strength of evidence (SOE) based on established guidance.

Results:

We included 135 studies. Most patients met criteria for alcohol dependence; mean ages were in the 40s. Studies typically included psychosocial cointerventions; effect sizes reflect the added benefits of medications.For acamprosate and oral naltrexone (50 mg per day), numbers needed to treat (NNTs) to prevent 1 person from returning to any drinking were 12 and 20, respectively (moderate SOE); NNT to prevent 1 person from returning to heavy drinking was 12 for oral naltrexone (50 mg per day) (moderate SOE). Our meta-analyses of four head-to-head trials found no statistically significant difference between the two medications for consumption outcomes (moderate SOE). For injectable naltrexone, meta-analyses found no significant benefit for return to any or heavy drinking, but found a reduction in heavy drinking days (low SOE). Evidence from well-controlled trials does not support efficacy of disulfiram, except possibly for patients with excellent adherence. Among medications used off label, moderate evidence supports the efficacy of nalmefene and topiramate for improving some consumption outcomes, and limited evidence supports the efficacy of valproic acid. Evidence from primary care settings was scant. We found insufficient direct evidence to conclude whether medications for alcohol-use disorders are effective for improving health outcomes.

Compared with placebo, patients treated with acamprosate had a higher risk of anxiety, diarrhea, and vomiting; those treated with naltrexone had a higher risk of dizziness, nausea, and vomiting. In head-to-head studies, the risks of headache and vomiting were slightly higher for naltrexone than for acamprosate. Individual trials of topiramate reported a significantly increased risk of paresthesias, anorexia, difficulty concentrating, dizziness, psychomotor slowing, and other adverse effects.

Our meta-analyses for variation in naltrexone response related to OPRM1 polymorphisms found no statistically significant difference between A-allele homozygotes and those with at least one G allele, but confidence intervals were wide and additional studies are needed.

Conclusions:

Acamprosate and oral naltrexone have the best evidence for improving alcohol consumption outcomes for patients with alcohol-use disorders. Head-to-head trials have not consistently established the superiority of one medication. Thus, other factors may guide medication choices, such as frequency of administration, potential adverse events, coexisting symptoms, and availability of treatments.