The role of opioid agonists and antagonists has been explored in psychostimulant users with and without concurrent opioid dependence. Increasingly polydrug use is a common feature both among recreational drug users and dependent opioid users. Cocaine use has been identified as a significant treatment impediment in methadone maintenance patients that undermines the effectiveness of methadone maintenance (Condelli, Fairbank, Dennis & Rachal, 1991) and is a commonly reported reason for treatment failure (Strug, Hunt, Goldsmith, Lipton & Spunt, 1985). There have been reports of subgroups of methadone patients using cocaine to obtain the euphoria not available when receiving methadone (Grabowski, Rhoades, Elk, Schmitz & Creson, 1993; Tutton & Crayton, 1993; van Beek et al., 2001). Buprenorphine was believed by some earlier reviewers (Platt, 1997; Tutton & Crayton, 1993) to hold promise in treating comorbid cocaine-opioid dependence. However, no evidence for cocaine specific efficacy or differential efficacy compared to methadone was found in a recent meta-analysis involving five studies and 779 participants (Mattick, Kimber, Breen & Davoli, 2002).

The opioid antagonist, naltrexone, may have potential in cocaine use via its euphoria blocking effects on opiate pathway reinforcers, similar to its postulated mechanism of action in alcohol dependence. In a placebo-controlled study of relapse prevention treatment, 85 abstinent cocaine-dependent volunteers were randomised into one of four combined conditions; naltrexone (0 vs. 50 mg) with relapse prevention versus drug counselling (Schmitz, Stotts, Rhoades & Grabowski, 2001). The combined naltrexone/relapse prevention group significantly reduced cocaine use over time compared to the other conditions. This study represents an encouraging contribution to integrated behavioural and pharmacological treatment approaches although it is limited by the selection of an abstinent study population and by the difficulties and risks of naltrexone induction and maintenance for opioid dependent patients. Intriguingly, cocaine use (measured by self-report and urinalysis) declined in a cohort of 266 heroin and cocaine users receiving prescribed heroin maintenance from 84% to 48% over 18 months (Blätter, Dobler-Mikola, Steffen & Uchtenhagen, 2002). The study sample was drawn from the participants in the first Swiss trial of prescribed heroin that included older, treatment-refractory heroin users. The authors suggested that high treatment retention in an extensively structured medically prescribed heroin program may have been explanatory. Declines in factors associated with cocaine use including criminality, prostitution and drug scene contact may also have contributed.