To evaluate the effect of the threat of an aversive reaction on the response during alcohol cue exposure in alcohol dependent patients : (1) the subjective response (craving) and (2) the physiological response (heart rate and blood pressure).

The efficacy of disulfiram in relapse prevention is controversial. Not only are most of the studies dated but their methodological rigor is generally poor. The major obstacle to disulfiram's effectiveness is non-compliance. No study to date has directly explored whether the threat of a disulfiram ethanol reaction (DER), provoked by the ingestion of disulfiram, has an effect on craving. Alcohol dependent patients have difficulty tolerating craving, a phenomenon that is believed to increase the probability of relapse. We propose in this study an evaluation of alcohol craving in relation to the threat of a DER compared to no threat. In both of these experimental conditions, we will use a placebo in order to avoid confounding the pharmacological effect of disulfiram with the psychological effect of the threat. Craving will be evaluated in the context of the multidimensional model of ambivalence (BREINER, STRITZKE and Lang, 1999) which provides two independent dimensions, craving and aversion.

To evaluate the effect of the threat of an aversive reaction on the response during alcohol cue exposure in alcohol dependent patients : (1) the subjective response (craving) and (2) the physiological response (heart rate and blood pressure).

• To evaluate the correlation between the subjective and physiological responses to alcohol cue exposure in relation to the threat of an aversive reaction.
• To evaluate the moderating effects of mood and personality on alcohol cue exposure in relation to the threat of an aversive reaction.

The design of this study is a within-subject, single-blind, randomized, and monocentric. The participants will be exposed to their habitual alcoholic drink. They will receive a placebo with two types of randomized inductions : (1) the threat of an aversive reaction and (2) no threat. The initial inclusion visit will take place a minimum of six days after the patients consumed their last alcohol beverage, the first cue exposure will take place one to seven days after the inclusion visit, and the second cue exposure will take place four to eight days after the first. This study directly benefits the patient because the experience of cue exposure provokes habituation.

The demonstration of an effect of the threat of an aversive reaction on craving may help alcohol dependent patients to better accept treatment using disulfiram as they would view it as alleviating craving instead of strictly as a punitive measure in the event of alcohol intake.

Inclusion Criteria:

• Inpatients
• Alcohol dependence
• Detoxified since at least one week
• Willing to abstain from alcohol for at least 6 months
• Never treated with disulfiram

Exclusion Criteria:

• Any contra-indication to disulfiram
• Treated with and antidepressant or a neuroleptic medication within the 30 previous days
• Treated with acamprosate, naltrexone, betablockers or clonidine within the 7 previous days
• Treated with benzodiazepines within the 3 previous days (except diazepam, maximum 30 mg/d)
• Anosmia