The etiology of substance abuse comorbidity in bipolar disorder is complex and not completely understood.33,34 A view commonly held by patients and clinicians alike is that substance abuse occurs as an attempt to self-medicate mood. While this may hold true for individual patients, people with bipolar disorder also exhibit high rates of substance use even during euthymic periods,34 and in many cases substances exacerbate rather than reduce symptoms (eg, alcohol during depression, cocaine during mania).34–37
Furthermore, the onset of bipolar disorder symptoms may emerge either before or after the onset of substance abuse.38–41 Dually diagnosed patients with early (or antecedent) onset of bipolar symptoms tend to suffer a more severe course of affective illness39–41 and have poorer syndromic and functional recovery whether substance abuse is current or in the past,14,18 suggesting that co-occurring SUDs may be a marker, rather than a determinant, of illness severity in these patients. Unfortunately, no shared genetic and/or neurobiological mechanisms that confer susceptibility to both bipolar disorder and addiction have been identified.
Despite the high prevalence and impact of SUD comorbidity in bipolar disorder, the evidence base to guide clinical decision-making is remarkably limited. Traditionally, efficacy studies of medications for bipolar disorder have excluded subjects with substance abuse, and serious mental illness generally has been an exclusion criterion from treatment trials for SUDs.42 As a result, few randomized placebo-controlled trials have been completed in this clinical population, and most of those conducted to date have been limited to relatively small samples with high attrition rates43–48 (see Table 1, page 192).
Table 1. Published Double Blind, Randomized Controlled Medication Trials Conducted in Adults with Co-occurring Bipolar Disorder and Substance Use Disorders
Diagnostic and Management Implications
Substance dependence and bipolar disorders share a number of characteristics that present challenges for diagnosis and clinical management. Both are heterogeneous remitting/relapsing disorders with fluctuating instability of mood, high impulsivity, altered responsiveness to reward, and impairment of executive function.49 Similar to many SUDs, bipolar disorder may evade diagnosis and treatment for years, until progression of severity and psychosocial consequences prompt clinical attention.50 At the time of clinical presentation, substance intoxication or withdrawal can confuse diagnosis further, leading to either underdiagnosis51 or overdiagnosis52 of mood disorders.
The fluctuating course of illness of bipolar disorder may obscure or parallel the progression of SUDs, with intermittent symptomatic periods interrupted by spontaneous remissions that can reinforce denial of illness. As a result, individuals may be least likely to recognize that they are ill during periods when symptoms are most severe. Accordingly, co-occurring SUDs strongly predict poor treatment adherence in patients with bipolar disorder, especially during periods of active use.15
Medical comorbidities such as metabolic syndrome and cardiovascular disease are overrepresented in bipolar patients in general64 and occur at significantly higher rates in those with co-occurring SUDs, especially nicotine dependence.65 Bipolar adolescents initiate smoking at high rates55 and perhaps due to perceived beneficial effects on impaired cognition,56 patients with bipolar disorder have particularly low quit rates from smoking.57 Yet research suggests that a minority (12%) of psychiatrists advise their patients to quit smoking,58,59 and even when counseling occurs, medications for smoking cessation are rarely prescribed.59 Given that cigarette smoking can reduce plasma levels of many medications used for mood stabilization in bipolar disorder,60 improved treatment of nicotine dependence in these patients has the potential to improve management of mood symptoms as well as reduce long-term health risks, and thus represents one of the most urgent unmet needs in the treatment of bipolar disorder.
Hepatitis C Virus
Another important medical comorbidity in people with bipolar disorder and comorbid SUDs is chronic infection with hepatitis C virus (HCV), which occurs at rates five to eight times higher in bipolar patients with SUDs than in patients without either illness.30–32 Excessive alcohol use in those with HCV is associated with severe hepatic fibrosis, rapid disease progression, and elevated rates of hepatocellular carcinoma and cirrhosis.61 Alcohol dependence is found in up to one-third of people with bipolar disorder.4,8
Furthermore, medications such as valproate that are commonly used for maintenance treatment of bipolar disorder can adversely affect liver function even in non-infected patients, and thus require more careful monitoring in HCV-infected individuals. Although not absolutely contraindicated in these patients, marked elevation of hepatic transaminases upon introduction of valproate occurs more often in those with HCV.62 Since bipolar patients with comorbid SUDs may respond better to anticonvulsants than to lithium,18 valproate maintenance treatment is common. However, assessment of HCV infection status in these patients is advisable because the effects of valproate in actively drinking bipolar patients with HCV are unknown.
Clinicians should recognize that the presence of either bipolar disorder or SUD also complicates the treatment of HCV. Ongoing alcohol use predicts poor treatment response and is generally considered a contraindication for pharmacotherapy of HCV.61 Even in those without active alcohol use, HCV-infected patients with bipolar disorder are significantly less likely to receive treatment for HCV than those without bipolar disorder,63 perhaps reflecting concerns about severe psychiatric side effects such as depression, mania, psychosis, and suicidal ideation that can occur upon treatment with interferon-alpha, the most effective intervention for chronic HCV.64
Integrated Approach to Clinical Management
Patients with co-occurring bipolar disorder and SUDs may be encountered in general psychiatric or specialty (addiction or mood disorder) outpatient clinics. In any of these settings, a thorough substance use history should be obtained in all patients with known or suspected bipolar disorder, and detailed screening for any history of (hypo-) mania and depression is recommended for all intake assessments of patients with addictive disorders. Cautious interpretation of diagnostic history in new patients with SUDs is essential, as the absence of previous bipolar diagnosis or treatment does not rule out the illness.50,51 Conversely, a previous diagnosis of bipolar disorder in patients with SUDs is best considered provisional until supported by collateral history as overdiagnosis of bipolar disorder may occur in the context of ongoing substance use or withdrawal.52
Alhough many patients report virtually uninterrupted substance abuse since late adolescence or early adulthood, focused assessment of mood symptoms prior to regular substance use, or during any extended periods of abstinence, is optimal to make the diagnosis of bipolar disorder with confidence. Careful evaluation of family history of bipolar disorder and SUDs should be prioritized in every initial intake assessment given the high heritability of both conditions.65,66 This can be especially helpful in gauging risks of future comorbidity in patients who present primarily with either addiction or mood disorders in isolation.
When clinical and family history support the diagnosis of bipolar disorder and co-occurring SUDs, more detailed inquiry is warranted (see Table 2, page 194). The age of onset of each disorder should be assessed, as early onset of mania prior to the development of SUDs may predict a more guarded prognosis.38,39 A thorough assessment of substance use history should include inquiry about past withdrawal severity, as a history of withdrawal seizures in an actively drinking bipolar patient may support the addition of an anticonvulsant medication for mood stabilization.
Table 2: Recommended Routine Assessments for Patients with Potential Co-occurring SUDs and Bipolar Disorder
Evaluation of SUD history should specifically inquire about polysubstance abuse or dependence, include screening for any history of injection drug use and other HCV/HIV risk factors such as hypersexuality during (hypo-) mania, and the date of last screening for HCV/HIV in bipolar patients at high risk of infection should be documented. Consistent with the recommendations of the US Centers for Disease Control and Prevention, voluntary HIV (re-) screening should be considered at least annually in all patients at high risk for HIV.67 Assessment of historical and ongoing alcohol and other drug abuse at the initial visit should be followed up by continued monitoring at subsequent visits, especially in all patients with HCV.
Accordingly, screening for nicotine dependence should be a routine part of intake and longitudinal assessment. When feasible, urine drug screening and confirmatory testing of alcohol use biomarkers such as carbohydrate-deficient transferrin and gamma-glutamyltransferase can aid clinical risk assessment, especially in patients with poor insight or low motivation. Finally, it is imperative that treatment planning and management for all bipolar patients with comorbid SUDs include assessment of suicidal ideation at every visit.
In the absence of optimal treatment guidelines for bipolar disorder when complicated by comorbid SUDs, many researchers have begun to advocate an integrated approach to clinical management. As an exclusive focus on either affective illness or substance abuse is inadequate to address the other accompanying underlying condition(s), integrated psychosocial treatments such as that developed by Weiss and colleagues stress the numerous similarities and interplay between addictive and bipolar disorders.68
In this framework, patients and clinicians are encouraged to consider that they are not simultaneously experiencing/treating two distinct disorders; rather the illness being treated is in fact one multifaceted disorder, or “bipolar substance abuse.” This approach emphasizes patient responsibility for participation in treatment and recovery across the various dimensions of the disease, (eg, adherence with mood-stabilizing medications, abstinence from substance use, etc.). In well-controlled randomized comparison trials, this approach has been reported to reduce alcohol and other drug use to roughly half the levels observed in control subjects who receive only group drug counseling.69
Few prospective, controlled randomized clinical trials exist to guide pharmacologic treatment decisions (see Table 1). It is generally accepted that comorbid SUDs, as well as rapid cycling and mixed mood episodes (both of which are more common in patients with comorbid SUDs), predict poor response to lithium.18,49,70 This has promoted interest in the study of anticonvulsant medications such as carbamazepine,71 valproate,43 and lamotrigine72,73 for bipolar patients with comorbid alcohol and/or cocaine dependence. Salloum and colleagues reported fewer heavy drinking days and fewer drinks per heavy drinking day in alcohol-dependent bipolar I patients when maintained on divalproex sodium plus lithium relative to those treated with lithium plus placebo.43
However, in this study the combination treatment conferred no advantage for either manic or depressive symptoms, and rates of remission from depression remained relatively low in both groups. A subsequent 6-month study comparing lithium vs. the combination of lithium and divalproex in individuals with SUDs and rapid-cycling bipolar disorder found no additional benefit of combination treatment over lithium monotherapy in retention, mood outcomes, or substance use outcomes.45 Nonetheless, evidence that anticonvulsants can be beneficial for the treatment in acute alcohol withdrawal74 may support their use as mood stabilizers instead of or in addition to lithium in actively drinking bipolar patients.
Evidence for the use of atypical antipsychotics in comorbid bipolar disorder and SUDs has been mixed.44,46,75–77 Beneficial effects of agents such as aripiprazole76 and quetiapine77 have been reported in bipolar patients with cocaine or methamphetamine dependence, but a controlled trial of quetiapine in bipolar alcoholics found no evidence of efficacy for drinking outcomes,44 a result that was replicated in the only large multisite medication trial conducted in this population.46 The latter study, in which 362 alcohol-dependent subjects with bipolar I disorder were randomized to placebo or flexible-dose quetiapine (300–800 mg/day) for 12 weeks, found no difference in proportion of heavy drinking days or number of drinks per day in the two treatment groups, and somewhat surprisingly, also found no advantage of quetiapine over placebo in improvement of depressive or manic symptoms.46
At present, only two of the three medications approved by the US Food and Drug Administration (FDA) for the treatment of alcohol dependence have been evaluated in bipolar alcoholics.47,48 In a 12-week study of 50 actively drinking alcohol-dependent subjects with bipolar I or II disorder, Brown and colleagues found no statistically significant difference in drinking outcomes in subjects randomized to naltrexone (50 mg/day) or placebo, though a trend toward a greater decrease in drinking days and alcohol craving was reported.47
Similarly, a randomized controlled 8-week trial of acamprosate (1,998 mg/day) in alcohol-dependent subjects with bipolar disorder found no advantage of acamprosate over placebo in time to first drink, time to first heavy drinking day, or percent days abstinent, though post hoc analyses of clinician-rated global impression scale scores favored acamprosate in the final 2 weeks of the study.8
Neither naltrexone nor acamprosate appeared to exert adverse effects on either depressive or manic symptoms, though future studies of larger samples are needed to definitively establish the safety and efficacy of these medications in bipolar alcoholics.
Nicotine Addiction Treatment
Much room for improvement exists in the treatment of nicotine dependence in people with bipolar disorder.10 The 2008 US Public Health Service Clinical Practice Guidelines recommend that smokers with both psychiatric and substance abuse disorders should be offered treatment for tobacco dependence and that clinicians, including psychiatrists, must overcome reluctance to treat these populations.78 As tobacco dependence is a chronic relapsing condition, repeated intervention and motivational enhancement are necessary components to successful treatment.
The optimal timing of a smoking quit attempt is also crucial for many smokers, and no research to date has investigated the optimal timing of a cessation attempt for smokers with bipolar disorder, though clinical judgment would suggest that relative periods of mood stability would be optimal times for prompting a smoking cessation attempt.79
Treatment research on the use of approved pharmacotherapies for smoking cessation, including nicotine replacement therapy, bupropion, and varenicline, in patients with bipolar disorder is virtually nonexistent and desperately needed. Given the risks of mood destabilization and/or manic switch in some bipolar patients upon treatment with antidepressants, the safety and efficacy of bupropion for smoking cessation in this population is not established, though research suggests that bupropion has lower switch rates when compared to other antidepressants such as sertraline or venlafaxine.80
Similarly, in smokers with bipolar disorder, the safety and efficacy of varenicline, an alpha-4, beta-2 nicotinic receptor partial agonist that appears in clinical trials to be the most effective medication for smoking cessation developed to date81,82 is unknown at this time.10 Case reports of neuropsychiatric symptoms associated with the use of varenicline has prompted the FDA to issue a warning regarding suicidality, and recent published cases of manic switches in both bipolar I and II individuals with varenicline suggest that careful psychoeducation, titration, and monitoring are called for if using this medication in patients with bipolar disorder.83,84