This is an updated version of the original Cochrane review published in Issue 2, 2013.We included two randomised controlled clinical trials (RCTs) with a total of 81 participants. One study focused on the effect of baclofen versus diazepam for people with AWS. The other study aimed to determine the effect of baclofen versus placebo for adult inpatients who met the AWS criteria in Diagnostic and Statistical Manual of Mental Disorders, 4th Edition and had a score of at least 11 on the Clinical Institute Withdrawal Assessment of Alcohol Scale, Revised (CIWA-Ar).
Overall, we found only low-quality evidence for the comparisons reported in this review.
Regarding the efficacy, one study found that both baclofen and diazepam significantly decreased the CIWA-Ar score, without any significant difference between the two interventions. The other study showed no significant difference in CIWA-Ar score between baclofen and placebo, but did show a significantly decreased dependence on high-dose benzodiazepines with baclofen compared to placebo.
Only one study reported safety outcomes, finding no side effects in either the baclofen group or the diazepam group.
The evidence for recommending baclofen for AWS is insufficient. We require more well-designed RCTs to prove its efficacy and safety.