The disulfiram-ethanol reaction (DER) experience.

Abstract

In their valuable editorial, Fuller & Gordis (2004) draw our attention to the role of disulfiram. The drug is not fashionable, but we should not forget it. We still do not know enough about disulfiram to benefit from its full potential. It works for some, but the question is, for whom and how? The effect is not due to disufiram itself but to its active metabolite S-methyl N,N-diethyltiolcarbamate sulphoxide. The latter is produced if the CYP2E1, CYP3A4 and CYP2A6 of the cytochrome P450 enzyme family are active. Inactivity may be genetic or due to liver disease or certain drugs, e.g. erythromycin or cimetidine. Moreover, long-term disulfiram treatment may accumulate enough diethyltiocarbamate to inhibit the cytochrome P450 enzymes and thus block the formation of the active metabolite (Madan et al. 1998). The above differences in metabolism explain why the disulfiram–ethanol reaction (DER) is sometimes weak or non-existent in disulfiram-treated patients. We need to know how to make a distinction between those patients who would potentially benefit, i.e. be at risk of a significant DER and those who would not be at risk. Could a test be developed to ascertain the risk of DER if the patient takes disulfiram? We also need to know the correct dosing patterns. Elimination of the active drug seems to vary much between individuals, because some patients may experience a DER even 14 days after the last dose. It would be useful to be able to predict the relevant pharmacokinetics in potential patients and tailor the dosing accordingly. For some, a tiny dose would be enough and thus the risk of adverse effects minimal. For the average patient on disulfiram, the risk of any adverse reaction has been estimated at one per 200–2000 treatment years in 1968–91 (Poulsen et al. 1992). This estimate was based on spontaneous reports from physicians and is therefore probably an underestimate. Better estimates derived from prospective monitoring of large patient cohorts would be welcome. Nevertheless, the risk of adverse reactions does not seem to be large compared with the risks of heavy alcohol intake. The most serious of adverse effects, hepatotoxicity that is potentially fatal, may be caused in fact by nickel sensitivity (Poulsen et al. 1992; Brewer & Hardt 1999). This should be confirmed and taken into account when counselling patients. ‘Supervised’ is one part of the answer to the question ‘how?’ There is clear evidence on the efficacy of supervised disulfiram both from the studies cited by Fuller & Gordis and from other research (Azrin et al. 1982; Carroll et al. 2000). However, would supervised administration of disulfiram be even more effective if patients were be exposed to alcohol in a controlled and safe environment early after starting to take disulfiram so that they would actually feel what DER is like? The rationale of disulfiram treatment is that it is a negative reinforcer. Experience speaks louder than words. A randomized controlled trial on supervised disulfiram with and without controlled DER would be useful here.