The study is planned for 430 patients (2:2:1 for both drugs and placebo) over 12 weeks of medication. All receive manualized counselling to improve compliance (Medical Management) which is extended to 6 months. Subtyping is primarily done using the acoustic startle reflex, functional magnetic resonance imaging, positron emission tomography (in a subset of patients), and the Inventory of Drinking Situations. Relapsers will be re-randomized into a second study where additional psychotherapy (Cognitive Behavioral Intervention) is used in a stepped care approach. Genotyping and additional analyses such as health economy are being done as well. The study follows the assessment methods, treatments, and medications used in the U.S. based COMBINE study, which will allow for a direct comparison between this U.S. study trial and a study performed in Europe.

Owing to the small sample size, the results of this study need to be interpreted with caution. Short-term moderate alcohol consumption (BAL 0.03%) did not significantly alter objective or subjective parameters of sleep. Higher doses of alcohol resulting in a BAL level of 0.10% immediately before going to bed mainly influenced sleep in the first half of the night, resembling the effects of a short-acting hypnotic drug, including a suppression of phasic aspects of REM sleep (REM density). Interestingly, analysis of the latter part of these nights indicated the immediate presence of withdrawal effects (increased light sleep). No statistically significant effects on sleep parameters were observable during the 2 nights of withdrawal from alcohol at the higher BAL. Interpreted carefully, our data indicate that negative effects on sleep occur already with short-term use of alcohol at doses of BAL of 0.10%, despite hypnotic-like effects during the first hours of sleep, especially during the latter part of the night.

Multiple lines of evidence suggest that the endocannabinoid system is implicated in the development of alcohol dependence. In addition, in animal models, the cannabinoid receptor 1 blocker rimonabant was found to decrease alcohol consumption, possibly by indirect modulation of dopaminergic neurotransmission. This was a 12-week double-blind, placebo-controlled, proof-of-concept study to assess the possible efficacy of the cannabinoid receptor 1 antagonist rimonabant 20 mg/d (2 x 10 mg) in the prevention of relapse to alcohol in recently detoxified alcohol-dependent patients. A total of 260 patients were included, 258 were exposed to medication, and 208 (80.6%) were men. Patients had an alcohol history of 15 years on average. More patients in the rimonabant group (94/131 [71.8%]) completed treatment compared with the placebo group (79/127 [62.2%]). Although there was a modest effect of rimonabant with respect to relapse rate, there were no statistically significant differences between treatment groups. Approximately 41.5% of the rimonabant group had relapsed to drinking at the end of the study compared with 47.7% of the placebo group (obtained from Kaplan-Meier-curve). Differences were more marked but not statistically significant in patients who relapsed to heavy drinking: 27.7% versus 35.6%, respectively. Safety and tolerance of the drug were good. Similar rates of adverse events were reported between the 2 groups; less patients experienced serious events or discontinued the treatment with rimonabant compared with placebo. Rates of depression-related events were low (3.8% with rimonabant compared with 1.6% with placebo). Patients on rimonabant lost weight (Mean, -1.7 kg) compared with baseline, whereas there was no such change in the placebo group. Weight loss was more pronounced in patients with a higher body mass index. In addition, there was a significant decrease in leptin levels in the rimonabant group compared with baseline. Lack of efficacy in this study may be explained by a very high response rate in the placebo group and a relatively short treatment duration. Taking the substantial numbers of animal studies suggesting a possible role of CB1 antagonists for the treatment of alcohol dependence into account, it seems worthwhile to further test cannabinoid blockers in the treatment of alcoholism.

The beta2 increase could be interpreted as a sign of dysfunctional arousal during REM sleep "unmasked" by the additional stressor of sleep environment adaptation. Its determinants are likely to be both premorbid and drinking history related.

No abstract