Generic Name carbamazepine
Manufacturer Ciba Geigy
Treatment Class Psychopharmacologic and Neurologic
Indication Epilepsy, Trigeminal Neuralgia, Drug Dependence
MSB Review Drug Dependence


The Frustrations of Treating Cocaine Dependence


The success of methadone maintenance for the treatment for heroine addiction has prompted the search for a similar agent for cocaine addiction, one that would not produce euphoria but would still prevent withdrawal. Unfortunately, cocaine dependence has proved to be highly resistant to therapy, primarily because of the powerful reinforcing potency of the drug. Acute cocaine administration increases the sensitivity to stimulation-reward in dopamine pathways, while long-term administration decreases this sensitivity to brain stimulation-reward. Animal studies have demonstrated that tricyclic antidepressants can reverse the long-term effect, but early open trials of tricyclics in cocaine addicts were negative.

The dopamine agonist amantadine (Symmetrel/DuPont) has been shown to reduce cocaine craving during withdrawal, and a single dose of, bromocriptine (Parlodel/Sandoz), was able to reverse cocaine craving. Other drugs that act on dopamine pathways have been tried for cocaine dependence -- including methylphenidate (Ritalin/Ciba Geigy), mazindol (Mazanor/Wyeth-Ayerst), and carbamazepine (Tegretol/Ciba Geigy) -- but with dismal long-term results. Bupropion (Wellbutrin/Burroughs Wellcome) was regarded as a promising candidate for cocaine abuse. Like cocaine, bupropion inhibits dopamine reuptake, but unlike promising candidate for cocaine abuse. Like cocaine, bupropion inhibits dopamine reuptake, but unlike cocaine, it does not produce euphoria or dependence. It also inhibits serotonin reuptake. Recently Hollister and Krajewski reported on a 21- day trial of bupropion in 46 young cocaine addicts. Doses of 100 mg three times daily reduced craving in some addicts, but the drop-out rate was high; only eight subjects in the bupropion group and six in the placebo group completed the 21- day trial. Side effects included dizziness, dry mouth, restlessness and lethargy. (Hollister LE, Krajewski K. Arch Gen Psych 1992; 49: 905. Meyer RE. Arch Gen Psych 1992; 49: 900-905.)

Most trials of drugs for cocaine dependence are open trials, and the results have to be interpreted with caution. Subjects often report what the researchers want to hear, while continuing with whatever behavior suits them. Frequently an addict will claim that the medication reduced cocaine use, while urinalysis will reveal substantial cocaine metabolites, indicating continued high use of cocaine. In light of compliance problems and deceptive behavior on the part of the user, two groups of researchers decided to try a controlled trial of desipramine in methadone-maintained cocaine users who were already attending a clinic every day for a concomitant heroine addiction. Both groups checked urine specimens twice weekly for cocaine and metabolites. The trials were carefully conducted and double blind -- and both reported negative results.

The study by Arndt et al. compared desipramine with placebo and lasted 12 weeks, with 1-, 3-, and 6-month follow-ups. Blood levels of desipramine, maintained at a target 150-300 ng/mL, were found to be unrelated to outcome, although psychiatric patients appeared to have improved on desipramine. Overall, 25% of subjects outcome, although psychiatric patients appeared to have improved on desipramine. Overall, 25% of subjects dropped out of the trial. Approximately 70% of urine specimens were positive for cocaine in the final week of treatment. Paradoxically, the placebo group sho    wed significantly less cocaine abuse during 3- and 6-month follow-up, compared with treated patients. The investigators speculated that the relief of psychiatric symptoms "may have actually allowed them to be able to continue to abuse cocaine." (Arndt IO et al. Arch Gen Psych 1992; 49: 888-893.)

The Kosten study also lasted 12 weeks, and compared desipramine with amantadine and placebo. Blood levels of desipramine were within the target range (170 ng/mL). Again, favorable outcome was unrelated to blood levels of desipramine and depressed addicts appeared to respond more favorably to desipramine. Amantadine was no better than placebo. While addicts reported they had reduced cocaine usage, urinalysis showed high cocaine in the urine. (Kosten TR et al. Arch Gen Psych 1992; 49: 888-893.) These two reports reiterate the importance of randomized controlled trials in the treatment of addiction. As Roger Meyer noted in his review of pharmacotherapies for cocaine dependence, "Self-reporting is not equivalent to biological testing." (Meyer RE. Arch Gen Psych 1992; 49: 900-905.)