Fatal intoxications are by far the most common reason for the excess mortality in drug users, followed by various somatic disorders, suicide, trauma and homicide . Research in this area may help to reduce mortality in drug dependence. One of the apparent risk factors in fatal overdose are polyintoxications, and especially the role of alcohol. We read with interest the article from Hickman et al. , which suggests that animal models could provide evidence for pharmacodynamic or pharmakokinetic interactions between alcohol and opioids which should be corroborated within clinical challenge and epidemiological studies.
While some issues on this subject are clear, many are open for debate. In their recent publication, Darke et al.  have discussed most of the issues concerning mortality among illicit drug users and this paper is an an outstanding and highly informative research monograph. Hickman et al.  state with reason that many studies have reported that alcohol is present in half or more cases and that there is an inverse relationship between blood alcohol and blood morphine concentrations . The fact that central nervous system depressants such as alcohol increase the risk for fatal overdose by respiratory depression is not surprising. One of the more interesting questions is the chronological relationship between alcohol and heroin intake. There is some evidence [3,4] that the risk is highest when using heroin following the consumption of alcohol, and not the other way around, but this has to be studied in more detail. There is also evidence for a much faster than usual respiratory depression after heroin consumption in alcohol-intoxicated individuals . A possible explanation might be a decreased metabolization of monoacetylmorphin to morphin in the presence of high blood alcohol levels . This question can be answered only through a translational research approach.
Both from epidemiological and clinical perspectives, it seems noteworthy to point out that the number of individuals with alcohol intoxications in methadone maintenance is lower than in heroin users. Two studies published in Addiction emphasize this: Bryant et al.  examined 7451 overdose deaths in New York City between 1990 and 1998; 1024 were methadone-induced overdose deaths and 4267 heroin-induced deaths. Alcohol was detected in 10.9% of methadone-induced overdose deaths but in 72% of heroin-induced deaths. Similarly, Shah et al.  studied 1120 drug overdose deaths in New Mexico. Alcohol was present in 20.3% of the methadone-related and 30.6% of the non-methadone-related deaths. These studies do not provide information about level of blood alcohol and heroin concentration or the chronological inter-relationship. These issues can be studied only in a translational research approach, not by only analysing toxicological data.