Abstract:

Background
Chronic excessive alcohol consumption may lead to dependence, and to alcohol withdrawal syndrome (AWS) in case of abrupt drinking cessation. Gamma-hydroxybutyric acid (GHB) can prevent and suppress withdrawal symptoms, and improve the medium-term abstinence rate. A clear balance between effectiveness and harmfulness has not been yet established.

Objectives
To evaluate the efficacy and safety of GHB for treatment of AWS and prevention of relapse

Search strategy
We searched Cochrane Drugs and Alcohol Group' Register of Trials (October 2008), PubMed,  EMBASE,  CINAHL (January 2005 - October 2008), EconLIT (1969 to February 2008), reference list of retrieved articles

Selection criteria
Randomized controlled trials (RCTs) and Controlled Prospective Studies (CPS) evaluating the efficacy and the safety of GHB vs placebo or other pharmacological treatments.

Data collection and analysis
Three authors independently extracted data and assessed the methodological quality of studies.

Main results
Thirteen RCTs were included. Eleven studies were conducted in Italy.

For withdrawal syndrome, comparing GHB 50mg with placebo, results from 1 study, 23 participants favour GHB for withdrawal symptoms: WMD -12.1 (95% CI, -15.9 to -8.29) and side effects were more frequent in the placebo group: RR 16.2 (95% CI, 1.04 to 254.9).In the comparison with Chlormetiazole, for GHB 50mg, results from 1 study, 21 participants favour GHB for withdrawal symptoms: MD -3.40 (95% CI -5.09 to -1.71), for GHB 100mg, results from 1 study, 98 participants favour anticonvulsants for side effects: RR 1.84 (95% CI 1.19 to 2.85).

At mid-term, comparing GHB with placebo, results favour GHB for abstinence rate (RR 5.35; 1.28-22.4), controlled drinking (RR 2.13; 1.07-5.54), relapses (RR 0.36; 0.21-0.63), and number of daily drinks (WMD -4.60; -6.18 to -3.02). GHB performed better than NTX and Disulfiram on abstinence (RR 2.59; 1.35-4.98, RR 1.66; 0.99-2.80 respectively). The association of GHB and NTX was better than NTX on abstinence (RR 12.2; 1.79-83.9), as well was the association of NTX, GHB and Escitalopram versus Escitalopram alone (RR 4.58; 1.28-16.5). For Alcohol Craving Scale results favour GHB versus placebo (WMD -1.90; -2.45 to 1.35) and Disulfiram (WMD -1.40; -1.86 to-0.94).

Authors' conclusions
GHB 50mg is effective compared to placebo in the treatment of AWS, and in preventing relapses in previously detoxified alcoholics at 3 months follow-up, but the results of this review do not provide sufficient evidence in favour of GHB compared to benzodiazepines and Chlormethiazole for AWS prevention. GHB is better than NTX and Disulfiram in maintaining abstinence and it has a better effect on craving than placebo and Disulfiram. Side effects of GHB are not statistically different from those with BZD, NTX or Disulfiram. However, concern has been raised regarding the risk of developing addiction, misuse or abuse, especially in polydrug abusers.