Table 2.
Randomized, controlled pharmacotherapy studies in BD comorbid with SUD (other than AUD or tobacco use disorder).
| Substance | Study | Substances | Intervention | Bipolar Diagnosis and N | Design | Outcome/Limitations |
|---|---|---|---|---|---|---|
| Amphetamines | Nejtek et al., 2008 [41] |
Methamphetamine or cocaine | Quetiapine or Risperidone | BD I or II. N = 80 |
20 weeks DB, add on to anticonvulsants or antidepressants. No PLC control. | See heading “Cocaine”. No separate results were reported for methamphetamine. |
| Brown et al., 2012 [42] | Methamphetamine | Citicoline | BD I, II and BD-NOS, MDD currently depressed. N = 48 |
12 weeks DB, PLC controlled add-on to TAU | Depressive symptoms (IDS-score) ↓with citicoline. No significant differences in methamphetamine use between groups. Small and heterogenous sample. | |
| Cannabis | Prisciandaro et al., 2021 [43] | Cannabis | Gabapentin | BDI or II and current (within the past 3 months) moderate-to-severe cannabis use disorder N = 22 |
GBP or PLC, 2-week RCT with cross-over after one week, add on to TAU. MRI study. | Primary outcome: 1H-MRS glutamate and GABA levels. GBP↑ dACC glutamate levels in participants with lower levels of substance use and mood symptoms. GBP↑ rBG glutamate levels and pMCC activation to cannabis cues in cigarette-smoking participants. ↑rBG glutamate and dACC GABA levels in participants while on GBP were associated with ↓cannabis use and mood symptoms in those with more severe SUD and mood symptoms |
| Gao et al., 2017 [44] | Cannabis, Alcohol or both | Quetiapine XR | BD I and II, currently depressed + comorbid anxiety. N = 90, but only 34 with cannabis use disorder |
8 weeks DB, PLC Controlled add-on to MS | No significant difference between PLC and quetiapine XR in mood or substance use outcomes. | |
| Kemp et al., 2009 [45] | Alcohol, cannabis, cocaine | Lithium vs. lithium + valproate | Rapid cycling BD I or II Phase 1: N = 149; Phase 2: N = 31 |
6 months open label stabilization followed by 6 months DB RCT. | Phase 1 (open stabilization): Of the 15 subjects with cannabis use disorders, 53% no longer met the criteria for active cannabis abuse or had entered into early full remission. Of the 9 subjects with cocaine use disorders, 78% no longer met the criteria for active cocaine abuse or had entered into early full remission. Compared to baseline, a significant ↓ in mood symptoms was observed for both interventions. Phase 2 (DB RCT): No differences in mood outcomes between interventions. | |
| Geller et al., 1998 [46] | Alcohol, Cannabis, Inhalants |
Lithium | Adolescents with BD I or II, single manic episode, or MDD with at least one predictor of future BD. N = 25, 2 on cannabis only and 14 on cannabis + alcohol | 6 weeks DB, PLC controlled RCT add on to TAU | The lithium group showed significant↓ across mood and substance use outcome measures, compared to placebo. Small sample size, no separate outcome data reported for cannabis. | |
| Cocaine | Nejtek et al., 2008 [41] |
Cocaine or methamphetamine | Quetiapine or Risperidone | BD I or II. N = 80 |
20 weeks DB, add on to anticonvulsants or antidepressants. No PLC control | Both study medications were associated with a significant ↓ in manic, depressive, or mixed mood states compared to baseline scores. Both medications were also associated with ↓ drug cravings and use. Limited evidence in the absence of a PLC control. No separate results were reported for cocaine. |
| Kemp et al., 2009 [45] | Alcohol, cannabis, cocaine | Lithium vs. lithium + valproate | Rapid cycling Bipolar I or II disorder. Phase 1: N = 149; Phase 2: N = 31 |
6 months open label stabilization followed by 6 months DB RCT. | See heading “Cannabis”. Phase 1 (open stabilization): Of 9 subjects with cocaine use disorders, 78% no longer met criteria for active cocaine abuse or had entered into early full remission. | |
| Brown et al., 2012 [47] |
Cocaine | Lamotrigine | BD I, II, BD-NOS and cyclothymia, depressed or mixed, N = 120 | 10 weeks DB, PLC controlled add-on to TAU | No differences in mood outcomes and craving between lamotrigine and placebo. The lamotrigine group showed a greater ↓ in amount spent on cocaine. | |
| Brown et al., 2007 [48] |
Cocaine and at least on other SUD | Citicoline | BD I or II (history of at least one (hypo)manic episode. N = 44 |
12 weeks DB, PLC controlled add-on to TAU | There were no significant changes in psychiatric symptoms between groups. A significant ↓ in the number of cocaine-positive urine screens was observed. | |
| Brown et al., 2015 [49] | Cocaine | Citicoline | BD I (depressed or mixed mood state) N = 130 |
12 weeks DB, PLC controlled add-on to TAU | Citicoline ↓ active cocaine use and the likelihood of relapse. There was no significant difference in craving symptoms between groups. No significant changes in mood symptoms. |
BD I: Bipolar I disorder; BD II: Bipolar II disorder; BD NOS: Bipolar disorder not otherwise specified; dACC: dorsal anterior cingulate cortex; DB: Double-blind; GBP: Gabapentin; IDS: Inventory of depressive symptoms; MDD: Major depressive disorder; PLC: Placebo; pMCC: posterior midcingulate cortex; rBG: right basal ganglia; RCT: Randomized controlled trial; TAU Treatment as usual; ↓ indicates decrease; ↑ indicates an increase.