Desipramine and contingency management for cocaine and opiate dependence in buprenorphine maintained patients

Abstract

Co-dependence on opiates and cocaine occurs in about 60% of patients entering methadone treatment and has a poor prognosis. However, we recently found that desipramine (DMI) could be combined with buprenorphine to significantly reduce combined opiate and cocaine use among these dually dependent patients. Furthermore, contingency management (CM) has been quite potent in reducing cocaine abuse during methadone maintenance. To test the efficacy of combining CM with these medications we designed a 12-week, randomized, double blind, four cell trial evaluating DMI (150 mg/day) or placebo plus CM or a non-contingent voucher control in 160 cocaine abusers maintained on buprenorphine (median 16 mg daily). Cocaine-free and combined opiate and cocaine-free urines increased more rapidly over time in those treated with either DMI or CM, and those receiving both interventions had more drug-free urines (50%) than the other three treatment groups (25–29%). Self reported opiate and cocaine use and depressive and opioid withdrawal symptoms showed no differences among the groups and symptom levels did not correlate with urine toxicology results. Lower DMI plasma levels (average 125 ng/ml) were associated with greater cocaine-free urines. DMI and CM had independent and additive effects in facilitating cocaine-free urines in buprenorphine maintained patients. The antidepressant appeared to enhance responsiveness to CM reinforcement.

Introduction

Opiate and cocaine dependence are major health problems in the United States and associated with medical complications such as human immunodeficiency virus (HIV) infection, hepatitis infection and endocarditis, and social issues such as unemployment and illegal activity. Methadone or buprenorphine maintenance can improve opiate problems, but neither medication targets cocaine abuse and, indeed, cocaine use may increase in some methadone-maintained patients (Kosten et al., 1987, Zweben and Payte, 1990, Sorensen and Copeland, 2000, Ball and Ross, 1991). Combined opiate and cocaine dependence is common with rates of 58% among those entering methadone treatment and has a poor prognosis (Ball et al., 1988, Ball et al., 1989). For example, in a recent study those opiate dependent patients who presented with both opiate and cocaine positive urines were three times more likely than those with opiates alone to drop out of methadone maintenance (30 vs. 10%) and have substantially fewer opiate and cocaine free urines (3 vs. 43%) in the first month (DeMaria et al., 2000). Thus, more effective maintenance strategies need to be developed for combined opiate and cocaine dependence such as the addition of other medications and/or behavioral therapies to opiate maintenance.

In studies examining pharmacotherapies for dual dependence on cocaine and opiates, we have found that desipramine (DMI) may be more effective when combined with buprenorphine than methadone. Previous studies had found that in methadone maintained patients DMI was not effective in reducing cocaine abuse (Arndt et al., 1992, Kosten et al., 1992a), and that buprenorphine alone, which had shown some initial promise for cocaine abuse, was not more effective than methadone alone in reducing cocaine abuse among these dually dependent patients (Kosten et al., 1989, Kosten et al., 1992b, Kosten et al., 1993, Schottenfeld et al., 1993, Schottenfeld et al., 1997). However, we recently found that DMI could be combined with buprenorphine to significantly reduce combined opiate and cocaine use among these dually dependent patients (Oliveto et al., 1999). Combining DMI with buprenorphine was more effective than placebo DMI, or DMI combined with methadone (Oliveto et al., 1999). We did not find that improvement in depressive symptoms was associated with reductions in opiate and cocaine use, although recent work has suggested that an additive effect of the antidepressant-fluoxetine and contingency management (CM) might be due to its antidepressant effect (Schmitz et al., 1998). Thus, combining buprenorphine and DMI appeared promising for combined cocaine and opiate use.

Nevertheless, DMI did not eliminate illicit drug use and to improve its efficacy, we chose a potent behavioral treatment—CM, that might enhance the attainment of initial abstinence, as well as improve both short and long term outcomes of maintenance opiate treatment (Carroll, 1997, Woody et al., 1983, Woody et al., 1995, McLellan et al., 1993, Carroll et al., 1995, Bigelow et al., 1984, Stitzer et al., 1979, Stitzer et al., 1982, Stitzer et al., 1992, Stitzer and Bigelow, 1978, Iguchi et al., 1988, Magura et al., 1988, Milby et al., 1978, McCaul et al., 1984). Several notable studies have indicated that CM can be quite potent in an opiate maintenance setting such as methadone, although review of all the CM studies done in methadone treatment settings has indicated that CM would benefit from added interventions such as the DMI we used in the current study (Chutuape et al., 1999, Griffith et al., 2000, Jones et al., 2001, Silverman et al., 1996, Silverman et al., 1998, Silverman et al., 1999, Silverman et al., 2001).

In order to test these issues we designed a study using DMI and CM alone and in combination among buprenorphine maintained patients who had dual dependence on opiates and cocaine. We had three hypotheses. First, both the DMI and CM groups would show a greater reduction in cocaine and opiate use than the groups not getting these treatments. Furthermore, the combined medication and CM group would have an additive effect and a greater reduction in cocaine and opiate use than either CM or DMI alone. Second, like in our previous study, those getting DMI would show no greater reduction in depressive or withdrawal symptoms than those getting placebo (Oliveto et al., 1999). Third, since we had previously found that higher DMI levels were associated with greater opiate free urines, we hypothesized that higher DMI levels would be associated with more opiate and cocaine-free urines (Oliveto et al., 1999).