Cyanamide or disulfiram in the treatment of adolescent alcohol misusers?

was the purpose of our double-blind, placebo-controlled study to verify the efficacy of cyanamide in the treatment of adolescent alcohol abusers (Niederhofer et al., 2003). Cyanamide is registered for the treatment of alcohol misusers; we therefore thought it would be appropriate to ascertain its efficacy, and in particular its safety, in adolescents. Disulfiram 442 The Randomised Controlled Trial by Niederhofer et al. (2003) describing the successful use of cyanamide in a group of alcohol-misusing adolescents, is further and convincing evidence of the effectiveness of deterrent or antagonist medication in the management of some types of substance misuse. However, despite the obvious similarities in mode of action between cyan-amide and disulfiram, I am puzzled that the above authors made absolutely no mention of the use of disulfiram in treatment. Ironically, their only reference to disulfiram is to a paper which compares its hepatotoxicity with that of cyanamide. Yet the literature strongly suggests that, whereas cyanamide causes histological changes in many patients, disulfiram causes only rare and idiosyncratic hepatotoxicity, which is commoner in women and is probably due to nickel sensitivity from costume jewellery (Brewer and Hardt, 1999). Although Niederhofer et al. (2003) did not specifically mention it, their paper gives the impression that the patients in this study were subjected to unusually high levels of supervision and control and that the treatment programme did not ignore the very important issue of compliance with medication. This necessarily involves a degree of supervision. Perhaps the main reason for the relatively infrequent use of deterrent medication is the failure of many reviewers of the literature to note the crucial distinction between those studies in which the administration of medication was supervised, which are almost universally positive at statistically significant levels, and those studies in which medication was unsupervised, which almost universally show no more effectiveness than placebo medication (Brewer et al., 2000; Brewer and Streel, 2003). Niederhofer et al. (2003) rightly draw attention to the probable superiority of deterrent medications to drugs, such as naltrexone and acamprosate, in alcoholism treatment. They could have strengthened their argument by mentioning studies comparing disulfiram with naltrexone or acamprosate which support that view (see, e.g. Carroll et al., 1993) Finally, it would be interesting to know why Niederhofer et al. (2003) chose a drug which needs to be given three times daily when they could have used disulfiram, which only needs to be given once daily or even …