Bipolar Disorder

BOCA RATON, FLA.  – Proposed changes for the diagnosis of patients with mixed depressive episodes for the next edition of the Diagnostic and Statistical Manual of Mental Disorders will more closely reflect the patients psychiatrists see in their everyday practice, Ellen Frank, Ph.D., said.

The current DSM-IV-TR definition of a mixed episode requires that a patient meet full criteria for a manic episode and full criteria for a major depressive episode (except for duration) nearly every day for at least a 1-week period. "What did I think was wrong with this definition? As far as I’m concerned, these patients don’t exist," Dr. Frank said at the annual meeting of the New Clinical Drug Evaluation Unit, sponsored by the National Institute of Mental Health.

I would argue that if you really carefully distinguish, for example, agitation from increased goal-directed activity, [and] if you really carefully distinguish the potential overlapping symptoms, these patients are about as rare as unicorns," Dr. Frank said.

"We thought the current criteria too restrictive, and they are rarely adhered to. When we say the patient is mixed, we rarely mean that the patient meets full criteria for both poles," said Dr. Frank, professor of psychiatry and psychology at the University of Pittsburgh. "This results in a lot of confusion, a lack of precision, and fails to identify [those] who are at risk of progression from unipolar to bipolar disorder."

Inappropriate treatment selection is another possibility with the DSM-IV-TR definition, said Dr. Frank, director of the Depression and Manic Depression Prevention Program at the Western Psychiatric Institute and Clinic in Pittsburgh, and a member of the DSM-5 Mood Disorders Work Group. In addition, "the consequence of the current definition in common use is that we underestimate suicide risk, because I do believe these patients are at excessive risk of suicide.

"This is a horrible state to be in, and we need to figure out how to do a better job of treating it."

What Dr. Frank and her collaborators on the working group propose for the DSM-5 is a mixed specifier that indicates the presence of symptoms – not the syndrome – of the opposite pole. The specifier would be applied to both episodes of mania and depression, and would be applicable in the context of both unipolar and bipolar lifetime diagnoses.

The psychiatric history of those who experience mixed states tend to include early onset of illness, multiple previous episodes, suicidal behavior, substance abuse, and other comorbid diagnoses that tend to be associated with bipolar disorder (such as panic).

Under the proposed construct, if a "mixed" patient is predominantly depressed, they meet full criteria for a major depressive episode and have at least three of the following nearly every day during the episode:

• Elevated mood.

• Decreased need for sleep (not insomnia).

• Goal-directed activity.

• Increased energy or visible hyperactivity (again, goal directed).

• Grandiosity.

• Accelerated speech.

• Racing thoughts.

If the patient is predominantly manic or hypomanic, they meet full criteria for the manic or hypomanic episode and feature at least three of the following:

• Subjective depression.

• Worry.

• Self-reproach or guilt.

• Negative evaluation of self.

• Hopelessness.

• Suicidal ideation or behavior.

• Anhedonia.

• Fatigue.

• Psychomotor retardation.

Interestingly, the proposed criteria do not include what the working group considered to be characteristics of both poles: irritability, indecisiveness, and psychomotor agitation (which is distinguished from goal-directed activity). In addition, the proposal does not include distractibility, "which I’ve come to believe is actually a feature of almost every patient anywhere on the bipolar spectrum," Dr. Frank said.

Insomnia or hypersomnia is another shared symptom not included in the specifier. Insomnia "is difficult to distinguish as a manic or depressive symptom," said Dr. Roy H. Perlis, director of the bipolar research program at Massachusetts General Hospital and instructor in psychiatry at Harvard Medical School, both in Boston.

Family aggregation, psychiatric history, concurrent psychiatric symptoms, and, particularly, treatment outcome were used as validators for the proposed specifier, Dr. Frank said. For example, in terms of treatment outcome, she added that a poor response to lithium and hypomanic symptoms during predominantly depressive states are associated with destabilization on antidepressants.

DSM-5 field trials are underway to assess the proposed criteria. The goal is to test the clinical acceptability and utility of the new constructs in both academic and practice settings. In addition, the reliability of diagnosis will be assessed in the academic component of the field trials. The clinics testing the mixed depression specifier include the Mayo Clinic, Rochester, Minn.; University of California,Los Angeles; the University of Texas Health Science Center at San Antonio; and the Veterans Affairs Medical Centers in Dallas and Houston.

Dr. Frank’s disclosures include Guilford Press and Servier International. Dr. Perlis’s disclosures include Concordant Rater Systems, Proteus Biomedical, and RIDVentures.

BOCA RATON, FLA.  – Proposed changes for the diagnosis of patients with mixed depressive episodes for the next edition of the Diagnostic and Statistical Manual of Mental Disorders will more closely reflect the patients psychiatrists see in their everyday practice, Ellen Frank, Ph.D., said.

The current DSM-IV-TR definition of a mixed episode requires that a patient meet full criteria for a manic episode and full criteria for a major depressive episode (except for duration) nearly every day for at least a 1-week period. "What did I think was wrong with this definition? As far as I’m concerned, these patients don’t exist," Dr. Frank said at the annual meeting of the New Clinical Drug Evaluation Unit, sponsored by the National Institute of Mental Health.

I would argue that if you really carefully distinguish, for example, agitation from increased goal-directed activity, [and] if you really carefully distinguish the potential overlapping symptoms, these patients are about as rare as unicorns," Dr. Frank said.

"We thought the current criteria too restrictive, and they are rarely adhered to. When we say the patient is mixed, we rarely mean that the patient meets full criteria for both poles," said Dr. Frank, professor of psychiatry and psychology at the University of Pittsburgh. "This results in a lot of confusion, a lack of precision, and fails to identify [those] who are at risk of progression from unipolar to bipolar disorder."

Inappropriate treatment selection is another possibility with the DSM-IV-TR definition, said Dr. Frank, director of the Depression and Manic Depression Prevention Program at the Western Psychiatric Institute and Clinic in Pittsburgh, and a member of the DSM-5 Mood Disorders Work Group. In addition, "the consequence of the current definition in common use is that we underestimate suicide risk, because I do believe these patients are at excessive risk of suicide.

"This is a horrible state to be in, and we need to figure out how to do a better job of treating it."

What Dr. Frank and her collaborators on the working group propose for the DSM-5 is a mixed specifier that indicates the presence of symptoms – not the syndrome – of the opposite pole. The specifier would be applied to both episodes of mania and depression, and would be applicable in the context of both unipolar and bipolar lifetime diagnoses.

The psychiatric history of those who experience mixed states tend to include early onset of illness, multiple previous episodes, suicidal behavior, substance abuse, and other comorbid diagnoses that tend to be associated with bipolar disorder (such as panic).

Under the proposed construct, if a "mixed" patient is predominantly depressed, they meet full criteria for a major depressive episode and have at least three of the following nearly every day during the episode:

• Elevated mood.

• Decreased need for sleep (not insomnia).

• Goal-directed activity.

• Increased energy or visible hyperactivity (again, goal directed).

• Grandiosity.

• Accelerated speech.

• Racing thoughts.

If the patient is predominantly manic or hypomanic, they meet full criteria for the manic or hypomanic episode and feature at least three of the following:

• Subjective depression.

• Worry.

• Self-reproach or guilt.

• Negative evaluation of self.

• Hopelessness.

• Suicidal ideation or behavior.

• Anhedonia.

• Fatigue.

• Psychomotor retardation.

Interestingly, the proposed criteria do not include what the working group considered to be characteristics of both poles: irritability, indecisiveness, and psychomotor agitation (which is distinguished from goal-directed activity). In addition, the proposal does not include distractibility, "which I’ve come to believe is actually a feature of almost every patient anywhere on the bipolar spectrum," Dr. Frank said.

Insomnia or hypersomnia is another shared symptom not included in the specifier. Insomnia "is difficult to distinguish as a manic or depressive symptom," said Dr. Roy H. Perlis, director of the bipolar research program at Massachusetts General Hospital and instructor in psychiatry at Harvard Medical School, both in Boston.

Family aggregation, psychiatric history, concurrent psychiatric symptoms, and, particularly, treatment outcome were used as validators for the proposed specifier, Dr. Frank said. For example, in terms of treatment outcome, she added that a poor response to lithium and hypomanic symptoms during predominantly depressive states are associated with destabilization on antidepressants.

DSM-5 field trials are underway to assess the proposed criteria. The goal is to test the clinical acceptability and utility of the new constructs in both academic and practice settings. In addition, the reliability of diagnosis will be assessed in the academic component of the field trials. The clinics testing the mixed depression specifier include the Mayo Clinic, Rochester, Minn.; University of California,Los Angeles; the University of Texas Health Science Center at San Antonio; and the Veterans Affairs Medical Centers in Dallas and Houston.

Dr. Frank’s disclosures include Guilford Press and Servier International. Dr. Perlis’s disclosures include Concordant Rater Systems, Proteus Biomedical, and RIDVentures.

BOCA RATON, FLA.  – Proposed changes for the diagnosis of patients with mixed depressive episodes for the next edition of the Diagnostic and Statistical Manual of Mental Disorders will more closely reflect the patients psychiatrists see in their everyday practice, Ellen Frank, Ph.D., said.

The current DSM-IV-TR definition of a mixed episode requires that a patient meet full criteria for a manic episode and full criteria for a major depressive episode (except for duration) nearly every day for at least a 1-week period. "What did I think was wrong with this definition? As far as I’m concerned, these patients don’t exist," Dr. Frank said at the annual meeting of the New Clinical Drug Evaluation Unit, sponsored by the National Institute of Mental Health.

I would argue that if you really carefully distinguish, for example, agitation from increased goal-directed activity, [and] if you really carefully distinguish the potential overlapping symptoms, these patients are about as rare as unicorns," Dr. Frank said.

"We thought the current criteria too restrictive, and they are rarely adhered to. When we say the patient is mixed, we rarely mean that the patient meets full criteria for both poles," said Dr. Frank, professor of psychiatry and psychology at the University of Pittsburgh. "This results in a lot of confusion, a lack of precision, and fails to identify [those] who are at risk of progression from unipolar to bipolar disorder."

Inappropriate treatment selection is another possibility with the DSM-IV-TR definition, said Dr. Frank, director of the Depression and Manic Depression Prevention Program at the Western Psychiatric Institute and Clinic in Pittsburgh, and a member of the DSM-5 Mood Disorders Work Group. In addition, "the consequence of the current definition in common use is that we underestimate suicide risk, because I do believe these patients are at excessive risk of suicide.

"This is a horrible state to be in, and we need to figure out how to do a better job of treating it."

What Dr. Frank and her collaborators on the working group propose for the DSM-5 is a mixed specifier that indicates the presence of symptoms – not the syndrome – of the opposite pole. The specifier would be applied to both episodes of mania and depression, and would be applicable in the context of both unipolar and bipolar lifetime diagnoses.

The psychiatric history of those who experience mixed states tend to include early onset of illness, multiple previous episodes, suicidal behavior, substance abuse, and other comorbid diagnoses that tend to be associated with bipolar disorder (such as panic).

Under the proposed construct, if a "mixed" patient is predominantly depressed, they meet full criteria for a major depressive episode and have at least three of the following nearly every day during the episode:

• Elevated mood.

• Decreased need for sleep (not insomnia).

• Goal-directed activity.

• Increased energy or visible hyperactivity (again, goal directed).

• Grandiosity.

• Accelerated speech.

• Racing thoughts.

If the patient is predominantly manic or hypomanic, they meet full criteria for the manic or hypomanic episode and feature at least three of the following:

• Subjective depression.

• Worry.

• Self-reproach or guilt.

• Negative evaluation of self.

• Hopelessness.

• Suicidal ideation or behavior.

• Anhedonia.

• Fatigue.

• Psychomotor retardation.

Interestingly, the proposed criteria do not include what the working group considered to be characteristics of both poles: irritability, indecisiveness, and psychomotor agitation (which is distinguished from goal-directed activity). In addition, the proposal does not include distractibility, "which I’ve come to believe is actually a feature of almost every patient anywhere on the bipolar spectrum," Dr. Frank said.

Insomnia or hypersomnia is another shared symptom not included in the specifier. Insomnia "is difficult to distinguish as a manic or depressive symptom," said Dr. Roy H. Perlis, director of the bipolar research program at Massachusetts General Hospital and instructor in psychiatry at Harvard Medical School, both in Boston.

Family aggregation, psychiatric history, concurrent psychiatric symptoms, and, particularly, treatment outcome were used as validators for the proposed specifier, Dr. Frank said. For example, in terms of treatment outcome, she added that a poor response to lithium and hypomanic symptoms during predominantly depressive states are associated with destabilization on antidepressants.

DSM-5 field trials are underway to assess the proposed criteria. The goal is to test the clinical acceptability and utility of the new constructs in both academic and practice settings. In addition, the reliability of diagnosis will be assessed in the academic component of the field trials. The clinics testing the mixed depression specifier include the Mayo Clinic, Rochester, Minn.; University of California,Los Angeles; the University of Texas Health Science Center at San Antonio; and the Veterans Affairs Medical Centers in Dallas and Houston.

Dr. Frank’s disclosures include Guilford Press and Servier International. Dr. Perlis’s disclosures include Concordant Rater Systems, Proteus Biomedical, and RIDVentures.

Discuss this article at www.facebook.com/CurrentPsychiatry

Treatment nonadherence among patients with chronic illness is high, and bipolar disorder (BD) is no exception. Approximately 21% to 50% of patients with BD do not adhere to their recommended treatment regimen,¹ which adds to the burden of illness and worsens prognosis.

Although treatment nonadherence is a concern with any psychiatric disorder, we focus on BD because of the high prevalence of the disorder, the lifelong nature of the illness, and its resulting disability. BD is challenging to treat even with motivated patients, and psychiatrists cannot count on individuals to follow their prescribed regimen just because they were told to do so. Choosing the best treatment for each patient is complicated, and as physicians, we need to learn how to connect with our patients, increase our insight into their concerns, and work collaboratively to find a treatment they can follow.

This article describes methods of assessing adherence, factors that affect adherence, and pharmacologic and psychosocial interventions to enhance adherence and improve outcomes.

What is adherence?

As the doctor-patient relationship and medical treatment evolved to become more patient-centered, so have the terms used to describe individuals’ treatment-related behavior. Compliance, a physician-centered term that mandates following instructions to achieve treatment goals, evolved to adherence, the extent to which a person fulfills their part of an agreed-upon treatment plan, followed by concordance, which describes a decision-making alliance between patient and provider that strongly considers patients’ input.

Adherence is considered adequate when it occurs at the minimum level necessary for the patient to respond to treatment and avoid relapse.² Research on adherence in BD can be difficult to interpret because results may be influenced by:

• selection bias (patients who are adherent and insightful are more likely to consent to research)
• complications caused by polypharmacy and comorbidity
• investigators’ ability to choose the proper measure to delineate medication adherence attitudes and behaviors
• patients’ compliance with the adherence-enhancing interventions.²

Assessment methods. Several tools can be used to measure adherence to mental illness treatment. Attitudinal scales capture a person’s subjective feelings (such as being on a medication, insight, perceived strength of the therapeutic alliance, and level of stigma faced) and can reflect attitude change that may result from adherence-enhancing interventions. Adherence behavior scales may be convenient to administer in the office but tend to overestimate patients’ adherence (Table 1).^3-7

Pill counts are inexpensive but patients can manipulate unused medication. Prescription refill counts are easy to obtain but do not confirm that the patient took the medication. Electronic medication monitors capture the time of specific doses and can calculate the adherence rate, but they are expensive and do not ensure that the medication was ingested. Measuring the drug in urine or blood is an objective measure of adherence and can serve as clinical guide to pharmacotherapy, but offers limited correlation with the amount of medication taken and is expensive. A combination of measures to estimate adherence may be best.²

Table 1

Tools for measuring adherence to medications

BD adherence studies

Treatment adherence in BD is challenged by the chronic remission-relapse pattern of the disorder. Manic episodes carry the highest risk of nonadherence.² Scott and Pope⁸ evaluated self-reported adherence to mood stabilizers (lithium, carbamazepine, or valproate) among 98 patients with major depressive disorder and 78 with BD. They found that 32% of patients were partially adherent (defined as having missed >30% of doses in the past month) and >60% of these patients had sub-therapeutic plasma levels of mood stabilizers.

In a study of 106 BD outpatients treated with lithium who completed scales regarding their attitudes toward and knowledge of lithium and the Medication Adherence Rating Scale (MARS), 86% of patients had a therapeutic serum lithium level (.6 to 1.2 mEq/L), and knowledge of lithium was correlated with adherence.⁹ Jónsdóttir et al¹⁰ looked at medication adherence among 280 patients with schizophrenia and BD by comparing patient self-reports to provider reports and measuring serum drug concentrations; adherence was defined as having a serum concentration within the reference level for the specific medication. BD patients had an adherence rate of 66%, and self-reported adherence as measured by MARS and provider reports correlated with serum concentrations.

In a study of 71 adolescents with BD followed for 1 year after their first hospitalization for a manic or mixed episode, DelBello et al¹¹ defined nonadherence as taking medication <25% of the time and partial adherence as taking medication 25% to 75% of the time. They found that 42% of patients were partially adherent and 23% were nonadherent.

Strakowski¹² followed 46 adults from Taiwan and 96 from the United States for 1 year after their first manic or mixed episode and found that 79% of the Taiwanese patients and 50% of U.S. patients were adherent. Using the medication possession ratio (MPR)—which is calculated based the number of days between expected and actual prescription refills—to determine adherence, Sajatovic¹³ found that 54% of 44,637 veterans being treated for BD with lithium or anticonvulsants were fully adherent (MPR >.80), 25% were partially adherent (MPR >.50 to .80), and 21% were nonadherent (MPR ≤.50). In a survey of 131 randomly selected psychiatrists and 429 of their adult BD patients, Baldessarini¹⁴ found that 34% of patients reported missing ≥1 medication dose in past 10 days, but psychiatrists recognized only 18% of patients as nonadherent.

What affects adherence?

Although all BD patients share the same diagnosis, the factors that ultimately result in their medication adherence are as variable as the individuals themselves. Patients’ age, sex, culture, symptom severity, worldview, socioeconomic status, opinion of mental illness, and self-image influence their individual decisions on adhering to a prescribed medication regimen.^1,15

Perception of medication efficacy. Not surprisingly, if a medication does not seem to decrease debilitating symptoms, a patient is unlikely to continue taking it. Patients with BD feel more affected by depressive symptoms than by manic symptoms, and have indicated that they are more likely to adhere to and view as successful treatments that reduce depressive symptoms.^16,17

Tolerability. In an Internet-based survey, 469 patients with BD indicated that medication-related weight gain and cognitive impairment were the most important factors that affected adherence.¹⁶ Individuals’ concerns about possible side effects may contribute more to nonadherence than actually experiencing side effects.¹⁷ Concerns about long-term metabolic side effects from atypical antipsychotics also may limit adherence.¹⁷

Neurocognitive impairment. Whether caused by BD, aging, or a combination of these factors, deficits in memory, attention, and executive functioning can lead to unintentional nonadherence. In a study that assessed medication management ability among middle-aged and older adults, patients with BD were found to make 2.8 times more errors than healthy controls.¹⁸

Therapeutic alliance and psychoeducation. Patients’ expectations for pharmacotherapy vary from specific symptom relief to hopes for a complete cure, and their fears may be influenced by media and advertisements.¹⁷ Nonetheless a positive therapeutic alliance with the treating provider improves illness outcomes.¹⁹

A clinician’s ability to help patients build insight is invaluable for their current and future treatment. In a survey of 435 veterans with BD, nonadherence was greater among patients with limited insight about the role of medication in their illness.²⁰ A study of 65 BD patients that evaluated insight into medication adherence at initial interview and 1 year later found that difficulty with adherence at the initial interview predicted future nonadherence and was correlated with lack of insight.²¹ Rosa et al⁹ found that BD patients in denial of their illness and those who had little psychoeducation were more frequently nonadherent with lithium treatment.

Other factors that may contribute to medication nonadherence in BD patients include comorbid substance abuse or personality disorders, both of which are associated with more frequent relapse.¹⁵ Marriage has a beneficial affect on adherence.¹⁵ A good support system may contribute to treatment adherence; in a study of 107 children and adolescents with BD, nonadherent patients were more likely to experience family dysfunction and have a parental history of psychiatric hospitalization.²²

Adherence and BD course

Treatment adherence decreases the suicide rate among BD patients. Angst et al²³ evaluated the rate of suicide among 406 patients with BD and unipolar depression who were followed for 40 years. They found that 11% committed suicide; untreated patients had significantly higher standardized mortality rates than of those who were treated with lithium, antipsychotics, or antidepressants. Other studies confirm this finding.¹⁵

Repeated relapse may predict poorer cognitive performance. Lopez-Jaramillo et al²⁴ showed that patients with BD who had more manic episodes performed poorer on cognitive tests assessing attention, memory, and executive functioning compared with patients with less episodes and with normal subjects.

Medication adherence in BD is a priority because of potential neurodegeneration in BD and the neuroprotective effects of mood stabilizers and some atypical antipsychotics (Box).

Box

Increasing adherence

Pharmacologic strategies. Adherence in BD often is difficult when patients require a complex medication regimen to control their illness. Patients and clinicians may prefer to use once-daily dosing drug formulations, which can provide consistent serum levels and fewer adverse effects. Divalproex extended-release (ER) allows once-daily dosing and improved tolerability by reducing fluctuations in valproic acid serum concentrations compared with the delayed-release formulation. In a retrospective chart review,²⁵ most patients (62%) who switched to divalproex ER from divalproex delayed-release preferred the ER formulation; 52% showed clinical improvement, 81% did not experience side effects, and 8% demonstrated higher adherence after switching.²⁵ Similarly, an extended-release formulation of carbamazepine is approved for treating acute mania.

Many atypical antipsychotics are FDA-approved for acute mania, acute bipolar depression, and/or maintenance (Table 2). Long-acting injectable formulations (LAIs) may be used as maintenance treatment if nonadherence is an issue. LAI risperidone, which is FDA-approved for maintenance treatment of bipolar I disorder (BDI), was found to be safe and effective in stable BD patients who were switched from an oral antipsychotic.²⁶ Asenapine is provided in a rapidly absorbed, sublingual form and is FDA-approved for treating acute mania or mixed episodes associated with BDI.²⁷ Overall, however, only slightly more than one-half of BD patients are adherent to atypical antipsychotics.¹⁵

Although antidepressant use in BD is controversial, Sajatovic¹⁷ found 44% of depressed BDI patients were treated with antidepressants. Novel extended-release antidepressant formulations—including controlled-release fluvoxamine, paroxetine, extended-release bupropion and venlafaxine, once-weekly fluoxetine, rapidly dissolving mirtazapine, and transdermal selegiline—can optimize drug delivery, minimize side effects, and delay onset of action.¹

Psychosocial strategies used in BD include psychoeducation, cognitive-behavioral therapy (CBT), family-focused interventions, and interpersonal and social rhythm therapy (IPSRT) (Table 3).^28-30 Psychoeducation alone or combined with other interventions can decrease the risk of relapse and hospitalization and improve adherence.²⁸ In a 2-year study of 50 euthymic BD patients treated with lithium who participated in a brief hospital-based psychoeducation program, Even et al³¹ found patients’ knowledge about lithium but not their attitudes changed significantly after the program. The changes persisted 2 years after the intervention, with a trend toward a decreased hospitalization rate.

Miklowitz³² reported on 293 BD patients randomized to receive collaborative care (3 psychoeducational sessions delivered over 6 weeks) or 1 of 3 types of intensive psychotherapy: CBT, IPSRT, or family-focused therapy. Attrition was similar for both groups. Compared with those receiving collaborative care, significantly more patients receiving intensive psychotherapy recovered after 1 year, and did so in shorter time.

In a 3-year, multi-site Veterans Administration (VA) study, 306 BD patients received psychoeducation and support from nurse care coordinators who were responsible for access, continuity of care, and information flow to psychiatrists or usual care according to VA guidelines.³³ Compared with the usual care group, patients who received psychoeducation and support from nurse care coordinators had shorter duration of manic episodes and improved function and quality of life. A meta-analysis³⁰ of 12 randomized controlled trials of CBT in BD showed a medium effect size of CBT on adherence at 6 months post-treatment.

Table 2

FDA-approved medications for adult bipolar disorder

Table 3

Psychosocial interventions for bipolar disorder

Related Resource

• Deegan PE. The importance of personal medicine: a qualitative study of resilience in people with psychiatric disabilities. Scand J Public Health Suppl. 2005;66:29-35.

Drug Brand Names

• Aripiprazole • Abilify
• Asenapine • Saphris
• Bupropion • Wellbutrin
• Carbamazepine • Carbatrol, Tegretol
• Carbamazepine extended- release • Equetro
• Clozapine • Clozaril
• Divalproex • Depakote, Depakote ER
• Fluoxetine • Prozac
• Fluvoxamine • Luvox
• Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Mirtazapine • Remeron
• Olanzapine • Zyprexa
• Olanzapine/fluoxetine • Symbyax
• Paroxetine • Paxil
• Quetiapine • Seroquel, Seroquel XR
• Risperidone • Risperdal
• Risperidone long-acting injectable • Risperdal Consta
• Selegiline • Eldepryl, Emsam
• Valproate • Depacon
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosures

Dr. Foster receives research/grant support from the American Psychiatric Foundation, the National Institute of Mental Health, and Sunovion Pharmaceuticals.

Dr. Sheehan and Ms. Johns report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Discuss this article at www.facebook.com/CurrentPsychiatry

Treatment nonadherence among patients with chronic illness is high, and bipolar disorder (BD) is no exception. Approximately 21% to 50% of patients with BD do not adhere to their recommended treatment regimen,¹ which adds to the burden of illness and worsens prognosis.

Although treatment nonadherence is a concern with any psychiatric disorder, we focus on BD because of the high prevalence of the disorder, the lifelong nature of the illness, and its resulting disability. BD is challenging to treat even with motivated patients, and psychiatrists cannot count on individuals to follow their prescribed regimen just because they were told to do so. Choosing the best treatment for each patient is complicated, and as physicians, we need to learn how to connect with our patients, increase our insight into their concerns, and work collaboratively to find a treatment they can follow.

This article describes methods of assessing adherence, factors that affect adherence, and pharmacologic and psychosocial interventions to enhance adherence and improve outcomes.

What is adherence?

As the doctor-patient relationship and medical treatment evolved to become more patient-centered, so have the terms used to describe individuals’ treatment-related behavior. Compliance, a physician-centered term that mandates following instructions to achieve treatment goals, evolved to adherence, the extent to which a person fulfills their part of an agreed-upon treatment plan, followed by concordance, which describes a decision-making alliance between patient and provider that strongly considers patients’ input.

Adherence is considered adequate when it occurs at the minimum level necessary for the patient to respond to treatment and avoid relapse.² Research on adherence in BD can be difficult to interpret because results may be influenced by:

• selection bias (patients who are adherent and insightful are more likely to consent to research)
• complications caused by polypharmacy and comorbidity
• investigators’ ability to choose the proper measure to delineate medication adherence attitudes and behaviors
• patients’ compliance with the adherence-enhancing interventions.²

Assessment methods. Several tools can be used to measure adherence to mental illness treatment. Attitudinal scales capture a person’s subjective feelings (such as being on a medication, insight, perceived strength of the therapeutic alliance, and level of stigma faced) and can reflect attitude change that may result from adherence-enhancing interventions. Adherence behavior scales may be convenient to administer in the office but tend to overestimate patients’ adherence (Table 1).^3-7

Pill counts are inexpensive but patients can manipulate unused medication. Prescription refill counts are easy to obtain but do not confirm that the patient took the medication. Electronic medication monitors capture the time of specific doses and can calculate the adherence rate, but they are expensive and do not ensure that the medication was ingested. Measuring the drug in urine or blood is an objective measure of adherence and can serve as clinical guide to pharmacotherapy, but offers limited correlation with the amount of medication taken and is expensive. A combination of measures to estimate adherence may be best.²

Table 1

Tools for measuring adherence to medications

BD adherence studies

Treatment adherence in BD is challenged by the chronic remission-relapse pattern of the disorder. Manic episodes carry the highest risk of nonadherence.² Scott and Pope⁸ evaluated self-reported adherence to mood stabilizers (lithium, carbamazepine, or valproate) among 98 patients with major depressive disorder and 78 with BD. They found that 32% of patients were partially adherent (defined as having missed >30% of doses in the past month) and >60% of these patients had sub-therapeutic plasma levels of mood stabilizers.

In a study of 106 BD outpatients treated with lithium who completed scales regarding their attitudes toward and knowledge of lithium and the Medication Adherence Rating Scale (MARS), 86% of patients had a therapeutic serum lithium level (.6 to 1.2 mEq/L), and knowledge of lithium was correlated with adherence.⁹ Jónsdóttir et al¹⁰ looked at medication adherence among 280 patients with schizophrenia and BD by comparing patient self-reports to provider reports and measuring serum drug concentrations; adherence was defined as having a serum concentration within the reference level for the specific medication. BD patients had an adherence rate of 66%, and self-reported adherence as measured by MARS and provider reports correlated with serum concentrations.

In a study of 71 adolescents with BD followed for 1 year after their first hospitalization for a manic or mixed episode, DelBello et al¹¹ defined nonadherence as taking medication <25% of the time and partial adherence as taking medication 25% to 75% of the time. They found that 42% of patients were partially adherent and 23% were nonadherent.

Strakowski¹² followed 46 adults from Taiwan and 96 from the United States for 1 year after their first manic or mixed episode and found that 79% of the Taiwanese patients and 50% of U.S. patients were adherent. Using the medication possession ratio (MPR)—which is calculated based the number of days between expected and actual prescription refills—to determine adherence, Sajatovic¹³ found that 54% of 44,637 veterans being treated for BD with lithium or anticonvulsants were fully adherent (MPR >.80), 25% were partially adherent (MPR >.50 to .80), and 21% were nonadherent (MPR ≤.50). In a survey of 131 randomly selected psychiatrists and 429 of their adult BD patients, Baldessarini¹⁴ found that 34% of patients reported missing ≥1 medication dose in past 10 days, but psychiatrists recognized only 18% of patients as nonadherent.

What affects adherence?

Although all BD patients share the same diagnosis, the factors that ultimately result in their medication adherence are as variable as the individuals themselves. Patients’ age, sex, culture, symptom severity, worldview, socioeconomic status, opinion of mental illness, and self-image influence their individual decisions on adhering to a prescribed medication regimen.^1,15

Perception of medication efficacy. Not surprisingly, if a medication does not seem to decrease debilitating symptoms, a patient is unlikely to continue taking it. Patients with BD feel more affected by depressive symptoms than by manic symptoms, and have indicated that they are more likely to adhere to and view as successful treatments that reduce depressive symptoms.^16,17

Tolerability. In an Internet-based survey, 469 patients with BD indicated that medication-related weight gain and cognitive impairment were the most important factors that affected adherence.¹⁶ Individuals’ concerns about possible side effects may contribute more to nonadherence than actually experiencing side effects.¹⁷ Concerns about long-term metabolic side effects from atypical antipsychotics also may limit adherence.¹⁷

Neurocognitive impairment. Whether caused by BD, aging, or a combination of these factors, deficits in memory, attention, and executive functioning can lead to unintentional nonadherence. In a study that assessed medication management ability among middle-aged and older adults, patients with BD were found to make 2.8 times more errors than healthy controls.¹⁸

Therapeutic alliance and psychoeducation. Patients’ expectations for pharmacotherapy vary from specific symptom relief to hopes for a complete cure, and their fears may be influenced by media and advertisements.¹⁷ Nonetheless a positive therapeutic alliance with the treating provider improves illness outcomes.¹⁹

A clinician’s ability to help patients build insight is invaluable for their current and future treatment. In a survey of 435 veterans with BD, nonadherence was greater among patients with limited insight about the role of medication in their illness.²⁰ A study of 65 BD patients that evaluated insight into medication adherence at initial interview and 1 year later found that difficulty with adherence at the initial interview predicted future nonadherence and was correlated with lack of insight.²¹ Rosa et al⁹ found that BD patients in denial of their illness and those who had little psychoeducation were more frequently nonadherent with lithium treatment.

Other factors that may contribute to medication nonadherence in BD patients include comorbid substance abuse or personality disorders, both of which are associated with more frequent relapse.¹⁵ Marriage has a beneficial affect on adherence.¹⁵ A good support system may contribute to treatment adherence; in a study of 107 children and adolescents with BD, nonadherent patients were more likely to experience family dysfunction and have a parental history of psychiatric hospitalization.²²

Adherence and BD course

Treatment adherence decreases the suicide rate among BD patients. Angst et al²³ evaluated the rate of suicide among 406 patients with BD and unipolar depression who were followed for 40 years. They found that 11% committed suicide; untreated patients had significantly higher standardized mortality rates than of those who were treated with lithium, antipsychotics, or antidepressants. Other studies confirm this finding.¹⁵

Repeated relapse may predict poorer cognitive performance. Lopez-Jaramillo et al²⁴ showed that patients with BD who had more manic episodes performed poorer on cognitive tests assessing attention, memory, and executive functioning compared with patients with less episodes and with normal subjects.

Medication adherence in BD is a priority because of potential neurodegeneration in BD and the neuroprotective effects of mood stabilizers and some atypical antipsychotics (Box).

Box

Increasing adherence

Pharmacologic strategies. Adherence in BD often is difficult when patients require a complex medication regimen to control their illness. Patients and clinicians may prefer to use once-daily dosing drug formulations, which can provide consistent serum levels and fewer adverse effects. Divalproex extended-release (ER) allows once-daily dosing and improved tolerability by reducing fluctuations in valproic acid serum concentrations compared with the delayed-release formulation. In a retrospective chart review,²⁵ most patients (62%) who switched to divalproex ER from divalproex delayed-release preferred the ER formulation; 52% showed clinical improvement, 81% did not experience side effects, and 8% demonstrated higher adherence after switching.²⁵ Similarly, an extended-release formulation of carbamazepine is approved for treating acute mania.

Many atypical antipsychotics are FDA-approved for acute mania, acute bipolar depression, and/or maintenance (Table 2). Long-acting injectable formulations (LAIs) may be used as maintenance treatment if nonadherence is an issue. LAI risperidone, which is FDA-approved for maintenance treatment of bipolar I disorder (BDI), was found to be safe and effective in stable BD patients who were switched from an oral antipsychotic.²⁶ Asenapine is provided in a rapidly absorbed, sublingual form and is FDA-approved for treating acute mania or mixed episodes associated with BDI.²⁷ Overall, however, only slightly more than one-half of BD patients are adherent to atypical antipsychotics.¹⁵

Although antidepressant use in BD is controversial, Sajatovic¹⁷ found 44% of depressed BDI patients were treated with antidepressants. Novel extended-release antidepressant formulations—including controlled-release fluvoxamine, paroxetine, extended-release bupropion and venlafaxine, once-weekly fluoxetine, rapidly dissolving mirtazapine, and transdermal selegiline—can optimize drug delivery, minimize side effects, and delay onset of action.¹

Psychosocial strategies used in BD include psychoeducation, cognitive-behavioral therapy (CBT), family-focused interventions, and interpersonal and social rhythm therapy (IPSRT) (Table 3).^28-30 Psychoeducation alone or combined with other interventions can decrease the risk of relapse and hospitalization and improve adherence.²⁸ In a 2-year study of 50 euthymic BD patients treated with lithium who participated in a brief hospital-based psychoeducation program, Even et al³¹ found patients’ knowledge about lithium but not their attitudes changed significantly after the program. The changes persisted 2 years after the intervention, with a trend toward a decreased hospitalization rate.

Miklowitz³² reported on 293 BD patients randomized to receive collaborative care (3 psychoeducational sessions delivered over 6 weeks) or 1 of 3 types of intensive psychotherapy: CBT, IPSRT, or family-focused therapy. Attrition was similar for both groups. Compared with those receiving collaborative care, significantly more patients receiving intensive psychotherapy recovered after 1 year, and did so in shorter time.

In a 3-year, multi-site Veterans Administration (VA) study, 306 BD patients received psychoeducation and support from nurse care coordinators who were responsible for access, continuity of care, and information flow to psychiatrists or usual care according to VA guidelines.³³ Compared with the usual care group, patients who received psychoeducation and support from nurse care coordinators had shorter duration of manic episodes and improved function and quality of life. A meta-analysis³⁰ of 12 randomized controlled trials of CBT in BD showed a medium effect size of CBT on adherence at 6 months post-treatment.

Table 2

FDA-approved medications for adult bipolar disorder

Table 3

Psychosocial interventions for bipolar disorder

Related Resource

• Deegan PE. The importance of personal medicine: a qualitative study of resilience in people with psychiatric disabilities. Scand J Public Health Suppl. 2005;66:29-35.

Drug Brand Names

• Aripiprazole • Abilify
• Asenapine • Saphris
• Bupropion • Wellbutrin
• Carbamazepine • Carbatrol, Tegretol
• Carbamazepine extended- release • Equetro
• Clozapine • Clozaril
• Divalproex • Depakote, Depakote ER
• Fluoxetine • Prozac
• Fluvoxamine • Luvox
• Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Mirtazapine • Remeron
• Olanzapine • Zyprexa
• Olanzapine/fluoxetine • Symbyax
• Paroxetine • Paxil
• Quetiapine • Seroquel, Seroquel XR
• Risperidone • Risperdal
• Risperidone long-acting injectable • Risperdal Consta
• Selegiline • Eldepryl, Emsam
• Valproate • Depacon
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosures

Dr. Foster receives research/grant support from the American Psychiatric Foundation, the National Institute of Mental Health, and Sunovion Pharmaceuticals.

Dr. Sheehan and Ms. Johns report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Discuss this article at www.facebook.com/CurrentPsychiatry

Treatment nonadherence among patients with chronic illness is high, and bipolar disorder (BD) is no exception. Approximately 21% to 50% of patients with BD do not adhere to their recommended treatment regimen,¹ which adds to the burden of illness and worsens prognosis.

Although treatment nonadherence is a concern with any psychiatric disorder, we focus on BD because of the high prevalence of the disorder, the lifelong nature of the illness, and its resulting disability. BD is challenging to treat even with motivated patients, and psychiatrists cannot count on individuals to follow their prescribed regimen just because they were told to do so. Choosing the best treatment for each patient is complicated, and as physicians, we need to learn how to connect with our patients, increase our insight into their concerns, and work collaboratively to find a treatment they can follow.

This article describes methods of assessing adherence, factors that affect adherence, and pharmacologic and psychosocial interventions to enhance adherence and improve outcomes.

What is adherence?

As the doctor-patient relationship and medical treatment evolved to become more patient-centered, so have the terms used to describe individuals’ treatment-related behavior. Compliance, a physician-centered term that mandates following instructions to achieve treatment goals, evolved to adherence, the extent to which a person fulfills their part of an agreed-upon treatment plan, followed by concordance, which describes a decision-making alliance between patient and provider that strongly considers patients’ input.

Adherence is considered adequate when it occurs at the minimum level necessary for the patient to respond to treatment and avoid relapse.² Research on adherence in BD can be difficult to interpret because results may be influenced by:

• selection bias (patients who are adherent and insightful are more likely to consent to research)
• complications caused by polypharmacy and comorbidity
• investigators’ ability to choose the proper measure to delineate medication adherence attitudes and behaviors
• patients’ compliance with the adherence-enhancing interventions.²

Assessment methods. Several tools can be used to measure adherence to mental illness treatment. Attitudinal scales capture a person’s subjective feelings (such as being on a medication, insight, perceived strength of the therapeutic alliance, and level of stigma faced) and can reflect attitude change that may result from adherence-enhancing interventions. Adherence behavior scales may be convenient to administer in the office but tend to overestimate patients’ adherence (Table 1).^3-7

Pill counts are inexpensive but patients can manipulate unused medication. Prescription refill counts are easy to obtain but do not confirm that the patient took the medication. Electronic medication monitors capture the time of specific doses and can calculate the adherence rate, but they are expensive and do not ensure that the medication was ingested. Measuring the drug in urine or blood is an objective measure of adherence and can serve as clinical guide to pharmacotherapy, but offers limited correlation with the amount of medication taken and is expensive. A combination of measures to estimate adherence may be best.²

Table 1

Tools for measuring adherence to medications

BD adherence studies

Treatment adherence in BD is challenged by the chronic remission-relapse pattern of the disorder. Manic episodes carry the highest risk of nonadherence.² Scott and Pope⁸ evaluated self-reported adherence to mood stabilizers (lithium, carbamazepine, or valproate) among 98 patients with major depressive disorder and 78 with BD. They found that 32% of patients were partially adherent (defined as having missed >30% of doses in the past month) and >60% of these patients had sub-therapeutic plasma levels of mood stabilizers.

In a study of 106 BD outpatients treated with lithium who completed scales regarding their attitudes toward and knowledge of lithium and the Medication Adherence Rating Scale (MARS), 86% of patients had a therapeutic serum lithium level (.6 to 1.2 mEq/L), and knowledge of lithium was correlated with adherence.⁹ Jónsdóttir et al¹⁰ looked at medication adherence among 280 patients with schizophrenia and BD by comparing patient self-reports to provider reports and measuring serum drug concentrations; adherence was defined as having a serum concentration within the reference level for the specific medication. BD patients had an adherence rate of 66%, and self-reported adherence as measured by MARS and provider reports correlated with serum concentrations.

In a study of 71 adolescents with BD followed for 1 year after their first hospitalization for a manic or mixed episode, DelBello et al¹¹ defined nonadherence as taking medication <25% of the time and partial adherence as taking medication 25% to 75% of the time. They found that 42% of patients were partially adherent and 23% were nonadherent.

Strakowski¹² followed 46 adults from Taiwan and 96 from the United States for 1 year after their first manic or mixed episode and found that 79% of the Taiwanese patients and 50% of U.S. patients were adherent. Using the medication possession ratio (MPR)—which is calculated based the number of days between expected and actual prescription refills—to determine adherence, Sajatovic¹³ found that 54% of 44,637 veterans being treated for BD with lithium or anticonvulsants were fully adherent (MPR >.80), 25% were partially adherent (MPR >.50 to .80), and 21% were nonadherent (MPR ≤.50). In a survey of 131 randomly selected psychiatrists and 429 of their adult BD patients, Baldessarini¹⁴ found that 34% of patients reported missing ≥1 medication dose in past 10 days, but psychiatrists recognized only 18% of patients as nonadherent.

What affects adherence?

Although all BD patients share the same diagnosis, the factors that ultimately result in their medication adherence are as variable as the individuals themselves. Patients’ age, sex, culture, symptom severity, worldview, socioeconomic status, opinion of mental illness, and self-image influence their individual decisions on adhering to a prescribed medication regimen.^1,15

Perception of medication efficacy. Not surprisingly, if a medication does not seem to decrease debilitating symptoms, a patient is unlikely to continue taking it. Patients with BD feel more affected by depressive symptoms than by manic symptoms, and have indicated that they are more likely to adhere to and view as successful treatments that reduce depressive symptoms.^16,17

Tolerability. In an Internet-based survey, 469 patients with BD indicated that medication-related weight gain and cognitive impairment were the most important factors that affected adherence.¹⁶ Individuals’ concerns about possible side effects may contribute more to nonadherence than actually experiencing side effects.¹⁷ Concerns about long-term metabolic side effects from atypical antipsychotics also may limit adherence.¹⁷

Neurocognitive impairment. Whether caused by BD, aging, or a combination of these factors, deficits in memory, attention, and executive functioning can lead to unintentional nonadherence. In a study that assessed medication management ability among middle-aged and older adults, patients with BD were found to make 2.8 times more errors than healthy controls.¹⁸

Therapeutic alliance and psychoeducation. Patients’ expectations for pharmacotherapy vary from specific symptom relief to hopes for a complete cure, and their fears may be influenced by media and advertisements.¹⁷ Nonetheless a positive therapeutic alliance with the treating provider improves illness outcomes.¹⁹

A clinician’s ability to help patients build insight is invaluable for their current and future treatment. In a survey of 435 veterans with BD, nonadherence was greater among patients with limited insight about the role of medication in their illness.²⁰ A study of 65 BD patients that evaluated insight into medication adherence at initial interview and 1 year later found that difficulty with adherence at the initial interview predicted future nonadherence and was correlated with lack of insight.²¹ Rosa et al⁹ found that BD patients in denial of their illness and those who had little psychoeducation were more frequently nonadherent with lithium treatment.

Other factors that may contribute to medication nonadherence in BD patients include comorbid substance abuse or personality disorders, both of which are associated with more frequent relapse.¹⁵ Marriage has a beneficial affect on adherence.¹⁵ A good support system may contribute to treatment adherence; in a study of 107 children and adolescents with BD, nonadherent patients were more likely to experience family dysfunction and have a parental history of psychiatric hospitalization.²²

Adherence and BD course

Treatment adherence decreases the suicide rate among BD patients. Angst et al²³ evaluated the rate of suicide among 406 patients with BD and unipolar depression who were followed for 40 years. They found that 11% committed suicide; untreated patients had significantly higher standardized mortality rates than of those who were treated with lithium, antipsychotics, or antidepressants. Other studies confirm this finding.¹⁵

Repeated relapse may predict poorer cognitive performance. Lopez-Jaramillo et al²⁴ showed that patients with BD who had more manic episodes performed poorer on cognitive tests assessing attention, memory, and executive functioning compared with patients with less episodes and with normal subjects.

Medication adherence in BD is a priority because of potential neurodegeneration in BD and the neuroprotective effects of mood stabilizers and some atypical antipsychotics (Box).

Box

Increasing adherence

Pharmacologic strategies. Adherence in BD often is difficult when patients require a complex medication regimen to control their illness. Patients and clinicians may prefer to use once-daily dosing drug formulations, which can provide consistent serum levels and fewer adverse effects. Divalproex extended-release (ER) allows once-daily dosing and improved tolerability by reducing fluctuations in valproic acid serum concentrations compared with the delayed-release formulation. In a retrospective chart review,²⁵ most patients (62%) who switched to divalproex ER from divalproex delayed-release preferred the ER formulation; 52% showed clinical improvement, 81% did not experience side effects, and 8% demonstrated higher adherence after switching.²⁵ Similarly, an extended-release formulation of carbamazepine is approved for treating acute mania.

Many atypical antipsychotics are FDA-approved for acute mania, acute bipolar depression, and/or maintenance (Table 2). Long-acting injectable formulations (LAIs) may be used as maintenance treatment if nonadherence is an issue. LAI risperidone, which is FDA-approved for maintenance treatment of bipolar I disorder (BDI), was found to be safe and effective in stable BD patients who were switched from an oral antipsychotic.²⁶ Asenapine is provided in a rapidly absorbed, sublingual form and is FDA-approved for treating acute mania or mixed episodes associated with BDI.²⁷ Overall, however, only slightly more than one-half of BD patients are adherent to atypical antipsychotics.¹⁵

Although antidepressant use in BD is controversial, Sajatovic¹⁷ found 44% of depressed BDI patients were treated with antidepressants. Novel extended-release antidepressant formulations—including controlled-release fluvoxamine, paroxetine, extended-release bupropion and venlafaxine, once-weekly fluoxetine, rapidly dissolving mirtazapine, and transdermal selegiline—can optimize drug delivery, minimize side effects, and delay onset of action.¹

Psychosocial strategies used in BD include psychoeducation, cognitive-behavioral therapy (CBT), family-focused interventions, and interpersonal and social rhythm therapy (IPSRT) (Table 3).^28-30 Psychoeducation alone or combined with other interventions can decrease the risk of relapse and hospitalization and improve adherence.²⁸ In a 2-year study of 50 euthymic BD patients treated with lithium who participated in a brief hospital-based psychoeducation program, Even et al³¹ found patients’ knowledge about lithium but not their attitudes changed significantly after the program. The changes persisted 2 years after the intervention, with a trend toward a decreased hospitalization rate.

Miklowitz³² reported on 293 BD patients randomized to receive collaborative care (3 psychoeducational sessions delivered over 6 weeks) or 1 of 3 types of intensive psychotherapy: CBT, IPSRT, or family-focused therapy. Attrition was similar for both groups. Compared with those receiving collaborative care, significantly more patients receiving intensive psychotherapy recovered after 1 year, and did so in shorter time.

In a 3-year, multi-site Veterans Administration (VA) study, 306 BD patients received psychoeducation and support from nurse care coordinators who were responsible for access, continuity of care, and information flow to psychiatrists or usual care according to VA guidelines.³³ Compared with the usual care group, patients who received psychoeducation and support from nurse care coordinators had shorter duration of manic episodes and improved function and quality of life. A meta-analysis³⁰ of 12 randomized controlled trials of CBT in BD showed a medium effect size of CBT on adherence at 6 months post-treatment.

Table 2

FDA-approved medications for adult bipolar disorder

Table 3

Psychosocial interventions for bipolar disorder

Related Resource

• Deegan PE. The importance of personal medicine: a qualitative study of resilience in people with psychiatric disabilities. Scand J Public Health Suppl. 2005;66:29-35.

Drug Brand Names

• Aripiprazole • Abilify
• Asenapine • Saphris
• Bupropion • Wellbutrin
• Carbamazepine • Carbatrol, Tegretol
• Carbamazepine extended- release • Equetro
• Clozapine • Clozaril
• Divalproex • Depakote, Depakote ER
• Fluoxetine • Prozac
• Fluvoxamine • Luvox
• Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Mirtazapine • Remeron
• Olanzapine • Zyprexa
• Olanzapine/fluoxetine • Symbyax
• Paroxetine • Paxil
• Quetiapine • Seroquel, Seroquel XR
• Risperidone • Risperdal
• Risperidone long-acting injectable • Risperdal Consta
• Selegiline • Eldepryl, Emsam
• Valproate • Depacon
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosures

Dr. Foster receives research/grant support from the American Psychiatric Foundation, the National Institute of Mental Health, and Sunovion Pharmaceuticals.

Dr. Sheehan and Ms. Johns report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

HONOLULU – Regardless of the type of psychotropic medication used to treat patients with bipolar disorder, outcomes are about the same at 1 year, a retrospective chart review from the University of Toledo suggests.

All 121 patients in the study were treated with a mood stabilizer. Some of them had an added antipsychotic, and a significant number also had an antidepressant, lead author Dr. Ronald A. McGinnis said at the annual meeting of the American Psychiatric Association. Across the board, the findings proved consistent. "We found no medication regimen superior to any other," he said.

Those who did not respond to treatment were "as likely to be on an antidepressant, mood stabilizer, or atypical antipsychotic" as were those who did respond. "We looked at individual medications and combinations of medications. No matter how you sliced the data, there were really no statistical differences in medications," despite "a lot of algorithms to follow and expert advice," said Dr. McGinnis, medical director of the University of Toledo (Ohio) Medical Center.

Despite these findings, Dr. McGinnis said he thinks that people do need to be on medications to recover. The point is that the type of medication the patient is taking does not seem to matter. "There are probably some other factors that we are not aware of that have to do with response," he said.

In the study, those factors seemed to include psychological stressors, pain, and missed appointments, all of which are more common in nonresponders. Poverty was more common, too, but the finding was not statistically significant.

For coauthor Dr. Daniel J. Rapport of the department of psychiatry at the university, the take-home message is not that medication choice is irrelevant. On the contrary, he thinks the study suggests that clinicians have to work extra hard to find drugs that work in a given patient.

The assertion is based on the finding that sustained recovery was associated with longer lag times to recovery. It is likely that during that lag time, clinicians were working closely with patients to find the combination of drugs that ultimately worked for them, he said.

Although "it didn’t seem to matter what medications or combinations of medications these people received," specific drugs appear to be more effective and better tolerated in given individuals. "So don’t give up" looking for the right combination. "It’s not the class, it’s the [specific] drug" selected, Dr. Rapport said.

Overall, 43.8% [53] patients achieved at least 12 consecutive months of either euthymia, a much-improved mood, or improvement followed by relapse; 56.1% patients [68] did not respond to treatment, which also included psychotherapy.

"It’s the minority of people who get a clinically meaningful recovery," Dr. McGinnis noted.

In all, 84 patients were men, and most were white. Their average age was 43 years, and average age at diagnosis 25. They were treated for at least a year.

The 121 patients were narrowed down from an original list of 271 who were billed for bipolar disorder over 18 months. The researchers found that 150 of those patients did not actually meet DSM-IV criteria for the disorder.

"We want to go back and see why all these people were billed with bipolar disorder when they really didn’t meet DSM-IV criteria for it," Dr. Rapport said.

In general, when it comes to treating bipolar disorder, he said, "the first thing that you have to do is to stop the cycling, so you use anticycling agents first, like lamotrigine, lithium, Tegretol, or Depakote." Residual symptoms are tackled with "a touch of antidepressants."

Dr. McGinnis said he has no relevant disclosures. Dr. Rapport said he previously spoke on behalf of Lamictal for its maker, GlaxoSmithKline, and worked with other companies. The study received no outside funding.

HONOLULU – Regardless of the type of psychotropic medication used to treat patients with bipolar disorder, outcomes are about the same at 1 year, a retrospective chart review from the University of Toledo suggests.

All 121 patients in the study were treated with a mood stabilizer. Some of them had an added antipsychotic, and a significant number also had an antidepressant, lead author Dr. Ronald A. McGinnis said at the annual meeting of the American Psychiatric Association. Across the board, the findings proved consistent. "We found no medication regimen superior to any other," he said.

Those who did not respond to treatment were "as likely to be on an antidepressant, mood stabilizer, or atypical antipsychotic" as were those who did respond. "We looked at individual medications and combinations of medications. No matter how you sliced the data, there were really no statistical differences in medications," despite "a lot of algorithms to follow and expert advice," said Dr. McGinnis, medical director of the University of Toledo (Ohio) Medical Center.

Despite these findings, Dr. McGinnis said he thinks that people do need to be on medications to recover. The point is that the type of medication the patient is taking does not seem to matter. "There are probably some other factors that we are not aware of that have to do with response," he said.

In the study, those factors seemed to include psychological stressors, pain, and missed appointments, all of which are more common in nonresponders. Poverty was more common, too, but the finding was not statistically significant.

For coauthor Dr. Daniel J. Rapport of the department of psychiatry at the university, the take-home message is not that medication choice is irrelevant. On the contrary, he thinks the study suggests that clinicians have to work extra hard to find drugs that work in a given patient.

The assertion is based on the finding that sustained recovery was associated with longer lag times to recovery. It is likely that during that lag time, clinicians were working closely with patients to find the combination of drugs that ultimately worked for them, he said.

Although "it didn’t seem to matter what medications or combinations of medications these people received," specific drugs appear to be more effective and better tolerated in given individuals. "So don’t give up" looking for the right combination. "It’s not the class, it’s the [specific] drug" selected, Dr. Rapport said.

Overall, 43.8% [53] patients achieved at least 12 consecutive months of either euthymia, a much-improved mood, or improvement followed by relapse; 56.1% patients [68] did not respond to treatment, which also included psychotherapy.

"It’s the minority of people who get a clinically meaningful recovery," Dr. McGinnis noted.

In all, 84 patients were men, and most were white. Their average age was 43 years, and average age at diagnosis 25. They were treated for at least a year.

The 121 patients were narrowed down from an original list of 271 who were billed for bipolar disorder over 18 months. The researchers found that 150 of those patients did not actually meet DSM-IV criteria for the disorder.

"We want to go back and see why all these people were billed with bipolar disorder when they really didn’t meet DSM-IV criteria for it," Dr. Rapport said.

In general, when it comes to treating bipolar disorder, he said, "the first thing that you have to do is to stop the cycling, so you use anticycling agents first, like lamotrigine, lithium, Tegretol, or Depakote." Residual symptoms are tackled with "a touch of antidepressants."

Dr. McGinnis said he has no relevant disclosures. Dr. Rapport said he previously spoke on behalf of Lamictal for its maker, GlaxoSmithKline, and worked with other companies. The study received no outside funding.

HONOLULU – Regardless of the type of psychotropic medication used to treat patients with bipolar disorder, outcomes are about the same at 1 year, a retrospective chart review from the University of Toledo suggests.

All 121 patients in the study were treated with a mood stabilizer. Some of them had an added antipsychotic, and a significant number also had an antidepressant, lead author Dr. Ronald A. McGinnis said at the annual meeting of the American Psychiatric Association. Across the board, the findings proved consistent. "We found no medication regimen superior to any other," he said.

Those who did not respond to treatment were "as likely to be on an antidepressant, mood stabilizer, or atypical antipsychotic" as were those who did respond. "We looked at individual medications and combinations of medications. No matter how you sliced the data, there were really no statistical differences in medications," despite "a lot of algorithms to follow and expert advice," said Dr. McGinnis, medical director of the University of Toledo (Ohio) Medical Center.

Despite these findings, Dr. McGinnis said he thinks that people do need to be on medications to recover. The point is that the type of medication the patient is taking does not seem to matter. "There are probably some other factors that we are not aware of that have to do with response," he said.

In the study, those factors seemed to include psychological stressors, pain, and missed appointments, all of which are more common in nonresponders. Poverty was more common, too, but the finding was not statistically significant.

For coauthor Dr. Daniel J. Rapport of the department of psychiatry at the university, the take-home message is not that medication choice is irrelevant. On the contrary, he thinks the study suggests that clinicians have to work extra hard to find drugs that work in a given patient.

The assertion is based on the finding that sustained recovery was associated with longer lag times to recovery. It is likely that during that lag time, clinicians were working closely with patients to find the combination of drugs that ultimately worked for them, he said.

Although "it didn’t seem to matter what medications or combinations of medications these people received," specific drugs appear to be more effective and better tolerated in given individuals. "So don’t give up" looking for the right combination. "It’s not the class, it’s the [specific] drug" selected, Dr. Rapport said.

Overall, 43.8% [53] patients achieved at least 12 consecutive months of either euthymia, a much-improved mood, or improvement followed by relapse; 56.1% patients [68] did not respond to treatment, which also included psychotherapy.

"It’s the minority of people who get a clinically meaningful recovery," Dr. McGinnis noted.

In all, 84 patients were men, and most were white. Their average age was 43 years, and average age at diagnosis 25. They were treated for at least a year.

The 121 patients were narrowed down from an original list of 271 who were billed for bipolar disorder over 18 months. The researchers found that 150 of those patients did not actually meet DSM-IV criteria for the disorder.

"We want to go back and see why all these people were billed with bipolar disorder when they really didn’t meet DSM-IV criteria for it," Dr. Rapport said.

In general, when it comes to treating bipolar disorder, he said, "the first thing that you have to do is to stop the cycling, so you use anticycling agents first, like lamotrigine, lithium, Tegretol, or Depakote." Residual symptoms are tackled with "a touch of antidepressants."

Dr. McGinnis said he has no relevant disclosures. Dr. Rapport said he previously spoke on behalf of Lamictal for its maker, GlaxoSmithKline, and worked with other companies. The study received no outside funding.

Many individuals think of postpartum depression as an episode of major depressive disorder within 4 weeks of delivery; however, postpartum depressive symptoms also can occur in the context of bipolar I disorder (BD I) or bipolar II disorder (BD II). Despite the high prevalence of postpartum hypomania (9% to 20%), clinicians often fail to screen for symptoms of mania or hypomania.¹ Determining if your patient’s postpartum depressive episode is caused by BD is essential when formulating an appropriate treatment plan that protects the mother and child.

Postpartum depression literature offers little guidance on the recognition and management of postpartum bipolar disorder.² For example, there is scant evidence on pharmacologic or psychotherapeutic treatment of postpartum bipolar depression, and no studies have evaluated the use of screening instruments such as the Edinburgh Postnatal Depression Scale to detect bipolar disorder.

Misdiagnosis of postpartum depression is more likely in cases of subtle bipolarity—BD II and BD not otherwise specified—than in BD I because:³

• physicians often fail to ask postpartum patients about hypomanic symptoms such as feelings of elation, being overly talkative, racing thoughts, and decreased need for sleep
• DSM-IV-TR does not recognize hypomania with a postpartum onset specifier
• hypomania symptoms overlap normal feelings of elation and sleep disturbance following childbirth.

Clues to postpartum bipolar disorder include:

• hypomania: persistently elevated, expansive, or irritable mood
• depression onset immediately after delivery
• atypical features such as hypersomnia, leaden paralysis, and increased appetite
• racing thoughts
• concomitant psychotic symptoms
• history of BD in a first-degree relative
• antidepressants “misadventures” (rapid response; loss of response; induction of mania, hypomania, or depressive mixed episodes; and poor response).

Treatment strategies

Avoid antidepressants. Bipolar depression does not respond to antidepressants as well as unipolar depression. Moreover, antidepressants can induce mania, hypomania, or mixed states, and can increase mood cycle frequency.

Administer mood stabilizers such as lithium or lamotrigine, or atypical antipsychotics such as quetiapine.

In breast-feeding women the benefits of treatment should be balanced carefully against the risk of infant exposure to medications. Lamotrigine should be used cautiously because of concerns of skin rash and higher-than-expected drug levels in the infant. In light of recent data showing no significant adverse clinical or behavioral effects in infants, breast-feeding while taking lithium should be considered in carefully selected women. The preliminary evidence supporting the use of quetiapine during breast-feeding appears promising; however, data on the safety of atypical antipsychotics in lactating women are limited.

Promote sleep. Sleep disruption can be a symptom of as well as a trigger for postpartum bipolar depression. In women with BD, the benefits of breast-feeding should be balanced carefully against the potential for the deleterious effects of sleep deprivation in triggering mood episodes. Women should consider using a breast pump allowing others to assist with feeding or supplementing breast milk with formula to help get uninterrupted sleep.³

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Many individuals think of postpartum depression as an episode of major depressive disorder within 4 weeks of delivery; however, postpartum depressive symptoms also can occur in the context of bipolar I disorder (BD I) or bipolar II disorder (BD II). Despite the high prevalence of postpartum hypomania (9% to 20%), clinicians often fail to screen for symptoms of mania or hypomania.¹ Determining if your patient’s postpartum depressive episode is caused by BD is essential when formulating an appropriate treatment plan that protects the mother and child.

Postpartum depression literature offers little guidance on the recognition and management of postpartum bipolar disorder.² For example, there is scant evidence on pharmacologic or psychotherapeutic treatment of postpartum bipolar depression, and no studies have evaluated the use of screening instruments such as the Edinburgh Postnatal Depression Scale to detect bipolar disorder.

Misdiagnosis of postpartum depression is more likely in cases of subtle bipolarity—BD II and BD not otherwise specified—than in BD I because:³

• physicians often fail to ask postpartum patients about hypomanic symptoms such as feelings of elation, being overly talkative, racing thoughts, and decreased need for sleep
• DSM-IV-TR does not recognize hypomania with a postpartum onset specifier
• hypomania symptoms overlap normal feelings of elation and sleep disturbance following childbirth.

Clues to postpartum bipolar disorder include:

• hypomania: persistently elevated, expansive, or irritable mood
• depression onset immediately after delivery
• atypical features such as hypersomnia, leaden paralysis, and increased appetite
• racing thoughts
• concomitant psychotic symptoms
• history of BD in a first-degree relative
• antidepressants “misadventures” (rapid response; loss of response; induction of mania, hypomania, or depressive mixed episodes; and poor response).

Treatment strategies

Avoid antidepressants. Bipolar depression does not respond to antidepressants as well as unipolar depression. Moreover, antidepressants can induce mania, hypomania, or mixed states, and can increase mood cycle frequency.

Administer mood stabilizers such as lithium or lamotrigine, or atypical antipsychotics such as quetiapine.

In breast-feeding women the benefits of treatment should be balanced carefully against the risk of infant exposure to medications. Lamotrigine should be used cautiously because of concerns of skin rash and higher-than-expected drug levels in the infant. In light of recent data showing no significant adverse clinical or behavioral effects in infants, breast-feeding while taking lithium should be considered in carefully selected women. The preliminary evidence supporting the use of quetiapine during breast-feeding appears promising; however, data on the safety of atypical antipsychotics in lactating women are limited.

Promote sleep. Sleep disruption can be a symptom of as well as a trigger for postpartum bipolar depression. In women with BD, the benefits of breast-feeding should be balanced carefully against the potential for the deleterious effects of sleep deprivation in triggering mood episodes. Women should consider using a breast pump allowing others to assist with feeding or supplementing breast milk with formula to help get uninterrupted sleep.³

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Many individuals think of postpartum depression as an episode of major depressive disorder within 4 weeks of delivery; however, postpartum depressive symptoms also can occur in the context of bipolar I disorder (BD I) or bipolar II disorder (BD II). Despite the high prevalence of postpartum hypomania (9% to 20%), clinicians often fail to screen for symptoms of mania or hypomania.¹ Determining if your patient’s postpartum depressive episode is caused by BD is essential when formulating an appropriate treatment plan that protects the mother and child.

Postpartum depression literature offers little guidance on the recognition and management of postpartum bipolar disorder.² For example, there is scant evidence on pharmacologic or psychotherapeutic treatment of postpartum bipolar depression, and no studies have evaluated the use of screening instruments such as the Edinburgh Postnatal Depression Scale to detect bipolar disorder.

Misdiagnosis of postpartum depression is more likely in cases of subtle bipolarity—BD II and BD not otherwise specified—than in BD I because:³

• physicians often fail to ask postpartum patients about hypomanic symptoms such as feelings of elation, being overly talkative, racing thoughts, and decreased need for sleep
• DSM-IV-TR does not recognize hypomania with a postpartum onset specifier
• hypomania symptoms overlap normal feelings of elation and sleep disturbance following childbirth.

Clues to postpartum bipolar disorder include:

• hypomania: persistently elevated, expansive, or irritable mood
• depression onset immediately after delivery
• atypical features such as hypersomnia, leaden paralysis, and increased appetite
• racing thoughts
• concomitant psychotic symptoms
• history of BD in a first-degree relative
• antidepressants “misadventures” (rapid response; loss of response; induction of mania, hypomania, or depressive mixed episodes; and poor response).

Treatment strategies

Avoid antidepressants. Bipolar depression does not respond to antidepressants as well as unipolar depression. Moreover, antidepressants can induce mania, hypomania, or mixed states, and can increase mood cycle frequency.

Administer mood stabilizers such as lithium or lamotrigine, or atypical antipsychotics such as quetiapine.

In breast-feeding women the benefits of treatment should be balanced carefully against the risk of infant exposure to medications. Lamotrigine should be used cautiously because of concerns of skin rash and higher-than-expected drug levels in the infant. In light of recent data showing no significant adverse clinical or behavioral effects in infants, breast-feeding while taking lithium should be considered in carefully selected women. The preliminary evidence supporting the use of quetiapine during breast-feeding appears promising; however, data on the safety of atypical antipsychotics in lactating women are limited.

Promote sleep. Sleep disruption can be a symptom of as well as a trigger for postpartum bipolar depression. In women with BD, the benefits of breast-feeding should be balanced carefully against the potential for the deleterious effects of sleep deprivation in triggering mood episodes. Women should consider using a breast pump allowing others to assist with feeding or supplementing breast milk with formula to help get uninterrupted sleep.³

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

HONOLULU – The antipsychotic ziprasidone does not appear to work as well in patients with bipolar disorder who are either obese or hyperglycemic, according to a study funded by the drug’s maker, Pfizer.

Among 267 acutely manic patients on ziprasidone (Geodon) monotherapy for 2-3 weeks, those with body mass indexes below 28.8 kg/m² were about twice as likely to respond to ziprasidone or go into remission during treatment than were those with BMIs above 28.8 kg/m², which roughly defines the border between being overweight and obese.

Among other findings, 52% of patients below that cut-off responded to treatment; for those above it, the response rate was 37%.

Meanwhile, patients with randomly tested blood glucose levels below 140 mg/dL were more than three times more likely to go into remission and more than five times more likely to respond to treatment than were those with blood glucose levels at or above 140 mg/dL, a level rarely reached in people with normal glucose metabolism.

More than half of patients with randomly tested glucose levels below 140 mg/dL – but only 16% of patients who tested at or above that level – responded to treatment.

Obese and hyperglycemic patients also showed less improvement on Global Assessment of Functioning scores. The findings all were statistically significant.

"Patients with bipolar disorder who have elevated blood glucose and/or elevated BMI do not respond as well to ziprasidone treatment of acute mania" and "may have a lower probability of responding" to antipsychotics in general, said lead author Dr. Roger S. McIntyre, associate professor of psychiatry and pharmacology at the University of Toronto.

Obese patients might need higher-than-typical doses to overcome greater body mass, but that’s "not clear at this point. You can increase the drug dose all you want; it may not make any difference," said Dr. McIntyre, who also is head of the mood disorders psychopharmacology unit at University Health Network in Toronto.

In any case, he said the findings offer another good reason to encourage patients to lose weight, and also argue for using antipsychotics such as ziprasidone that are less likely than others to cause weight gain, since excess weight now appears to diminish the effects of antipsychotics.

The problem with hyperglycemia might be related to insulin dysregulation; there’s emerging consensus "that insulin dysregulation manifesting as hyperglycemia might be neurotoxic," Dr. McIntyre noted.

He and his colleagues pooled data from previous Pfizer studies to gauge the effects of BMI on response. "It’s intuitive if you have an [elevated] BMI, that the psychopharmacotherapies you are taking would have different distributions and different concentrations, [but] it’s almost never been studied," Dr. McIntyre said.

Based on the results, the research community needs to rethink the effect of BMI on response, he said. "It’s an important way to stratify data."

The patients in the study were at least moderately manic, with baseline Mania Rating Scale scores of 14 or greater. Remission was defined by a score dropped below 10 by the study’s end; response was defined by a greater than 50% score reduction.

HONOLULU – The antipsychotic ziprasidone does not appear to work as well in patients with bipolar disorder who are either obese or hyperglycemic, according to a study funded by the drug’s maker, Pfizer.

Among 267 acutely manic patients on ziprasidone (Geodon) monotherapy for 2-3 weeks, those with body mass indexes below 28.8 kg/m² were about twice as likely to respond to ziprasidone or go into remission during treatment than were those with BMIs above 28.8 kg/m², which roughly defines the border between being overweight and obese.

Among other findings, 52% of patients below that cut-off responded to treatment; for those above it, the response rate was 37%.

Meanwhile, patients with randomly tested blood glucose levels below 140 mg/dL were more than three times more likely to go into remission and more than five times more likely to respond to treatment than were those with blood glucose levels at or above 140 mg/dL, a level rarely reached in people with normal glucose metabolism.

More than half of patients with randomly tested glucose levels below 140 mg/dL – but only 16% of patients who tested at or above that level – responded to treatment.

Obese and hyperglycemic patients also showed less improvement on Global Assessment of Functioning scores. The findings all were statistically significant.

"Patients with bipolar disorder who have elevated blood glucose and/or elevated BMI do not respond as well to ziprasidone treatment of acute mania" and "may have a lower probability of responding" to antipsychotics in general, said lead author Dr. Roger S. McIntyre, associate professor of psychiatry and pharmacology at the University of Toronto.

Obese patients might need higher-than-typical doses to overcome greater body mass, but that’s "not clear at this point. You can increase the drug dose all you want; it may not make any difference," said Dr. McIntyre, who also is head of the mood disorders psychopharmacology unit at University Health Network in Toronto.

In any case, he said the findings offer another good reason to encourage patients to lose weight, and also argue for using antipsychotics such as ziprasidone that are less likely than others to cause weight gain, since excess weight now appears to diminish the effects of antipsychotics.

The problem with hyperglycemia might be related to insulin dysregulation; there’s emerging consensus "that insulin dysregulation manifesting as hyperglycemia might be neurotoxic," Dr. McIntyre noted.

He and his colleagues pooled data from previous Pfizer studies to gauge the effects of BMI on response. "It’s intuitive if you have an [elevated] BMI, that the psychopharmacotherapies you are taking would have different distributions and different concentrations, [but] it’s almost never been studied," Dr. McIntyre said.

Based on the results, the research community needs to rethink the effect of BMI on response, he said. "It’s an important way to stratify data."

The patients in the study were at least moderately manic, with baseline Mania Rating Scale scores of 14 or greater. Remission was defined by a score dropped below 10 by the study’s end; response was defined by a greater than 50% score reduction.

HONOLULU – The antipsychotic ziprasidone does not appear to work as well in patients with bipolar disorder who are either obese or hyperglycemic, according to a study funded by the drug’s maker, Pfizer.

Among 267 acutely manic patients on ziprasidone (Geodon) monotherapy for 2-3 weeks, those with body mass indexes below 28.8 kg/m² were about twice as likely to respond to ziprasidone or go into remission during treatment than were those with BMIs above 28.8 kg/m², which roughly defines the border between being overweight and obese.

Among other findings, 52% of patients below that cut-off responded to treatment; for those above it, the response rate was 37%.

Meanwhile, patients with randomly tested blood glucose levels below 140 mg/dL were more than three times more likely to go into remission and more than five times more likely to respond to treatment than were those with blood glucose levels at or above 140 mg/dL, a level rarely reached in people with normal glucose metabolism.

More than half of patients with randomly tested glucose levels below 140 mg/dL – but only 16% of patients who tested at or above that level – responded to treatment.

Obese and hyperglycemic patients also showed less improvement on Global Assessment of Functioning scores. The findings all were statistically significant.

"Patients with bipolar disorder who have elevated blood glucose and/or elevated BMI do not respond as well to ziprasidone treatment of acute mania" and "may have a lower probability of responding" to antipsychotics in general, said lead author Dr. Roger S. McIntyre, associate professor of psychiatry and pharmacology at the University of Toronto.

Obese patients might need higher-than-typical doses to overcome greater body mass, but that’s "not clear at this point. You can increase the drug dose all you want; it may not make any difference," said Dr. McIntyre, who also is head of the mood disorders psychopharmacology unit at University Health Network in Toronto.

In any case, he said the findings offer another good reason to encourage patients to lose weight, and also argue for using antipsychotics such as ziprasidone that are less likely than others to cause weight gain, since excess weight now appears to diminish the effects of antipsychotics.

The problem with hyperglycemia might be related to insulin dysregulation; there’s emerging consensus "that insulin dysregulation manifesting as hyperglycemia might be neurotoxic," Dr. McIntyre noted.

He and his colleagues pooled data from previous Pfizer studies to gauge the effects of BMI on response. "It’s intuitive if you have an [elevated] BMI, that the psychopharmacotherapies you are taking would have different distributions and different concentrations, [but] it’s almost never been studied," Dr. McIntyre said.

Based on the results, the research community needs to rethink the effect of BMI on response, he said. "It’s an important way to stratify data."

The patients in the study were at least moderately manic, with baseline Mania Rating Scale scores of 14 or greater. Remission was defined by a score dropped below 10 by the study’s end; response was defined by a greater than 50% score reduction.

I needed an intellectual oasis to deal with the anguish and frustration triggered by the monumental amount of misleading information that was included in the well-written article “Not all mood swings are bipolar disorder” (Current Psychiatry, February 2011, p. 38-52). Fortunately, a commentary by Dr. Irene Abramovich (“Breaking the box,” Comments & Controversies, Current Psychiatry, February 2011, p. 59) appeared as a therapeutic elixir. I believe that the “mood swings” article is filled with examples of how dangerous “cookbook” medicine can be.

Dr. Kowatch and colleagues use an expression that can be applied to the so-called diagnoses oppositional defiant disorder (ODD) and conduct disorder (CD): “Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but they are not diagnostic as an isolated symptom. “ A similar concept was my position in a debate titled “Childhood conduct disorder and oppositional-defiant disorders are common manifestations of bipolar disorder” in which I argued that ODD and CD are behavioral expressions of genuine diagnoses.¹ Besides bipolar disorder, I also have seen obsessive-compulsive disorder, social anxiety disorder, and even sexual abuse labeled as “ODD” because the child refuses to be around people (such as a classroom) or is distracted by intrusive thoughts or flashbacks and turns hostile when reproached in front of the class.

In my view, Dr. Kowatch and colleagues give undeserved credit to the behavioral scales (the “cookbooks” of psychiatry) to make diagnoses and seem to miss warning signs in patients’ family history, ie, “history of depression and anxiety” (many times this translates as agitated/dysphoric mania) and “drinking problems, “ which frequently is found in undiagnosed bipolar spectrum patients who use alcohol to “shoot down” racing thoughts that interfere with normal sleep.

From January 2010 to February 2011, I reviewed charts and interviewed patients and families of 1, 654 patients with diagnoses of attention-deficit/hyperactivity disorder co-morbid with ODD, bipolar disorder, generalized anxiety disorder, and even 2 diagnoses that are not allowed by DSM rules: autism and mental retardation. The data from this study, which covers 12 counties that represent the 5 geographical areas of Florida, are being analyzed. In the meantime, I refer readers to my poster presentation from the 2010 U. S. Psychiatric and Mental Health Congress “Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. “² In this study 44 patients were followed for at least 5 years (10 patients were observed for 7 years and a similar number for 6) and none had “oppositional” behavior after the diagnoses were treated. One caveat is that I placed antipsychotics in the same category as conventional mood stabilizers because 5 patients considered to be “inattentive” and “oppositional” actually had schizophrenia.

I oppose the authors’ assertion that “it can be difficult to differentiate the mood swings and symptoms of ODD from those of pediatric BD. “ My experience is that it is simple if we consider all diagnostic possibilities and obtain a thorough family history, which usually includes alcoholism, cannabis abuse, moodiness, suicide completion, unstable lifestyle, etc.

Manuel Mota-Castillo, MD
Assistant Clinical Professor
St. Matthews University
Voluntary Faculty
University of Central Florida
Lake Mary, FL

References

1. Mota-Castillo M, Steiner H. Childhood conduct disorder and oppositional-defiant disorder are common manifestations of bipolar disorder pro and con. Journal of Bipolar Disorders: Reviews and Commentaries. 2005;3:3,15-17.

2. Mota-Castillo M. Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. Poster presented at: 23rd Annual U.S. Psychiatric and Mental Health Congress; November 20 2010; Orlando, FL.

The authors respond

We never suggested that clinicians use “cookbook medicine. “ The “behavioral scales” we recommended in our article are well-validated and reliable tools that allow a clinician to effectively elicit a great deal of useful information from patients and their parents about presenting problems and symptoms. This information can be used with other clinical information to make an accurate diagnosis and subsequent treatment plan.

The purpose of our article was to share our experiences in the differential diagnosis of mood swings in children and adolescents and to suggest that there are other diagnoses that cause mood swings besides bipolar disorder. Although a family history of mood disorders is important, it is also important to recognize that a recent, state-of-the-art study by Birmaher et al¹ reported that 10% of children of parents with bipolar disorder had a bipolar spectrum disorder. That means that 90% did not have bipolar disorder. It is important to remember this when evaluating children of parents with bipolar disorder. Although these children’s risk for developing bipolar disorder is increased compared with the general population, it is more likely that they will not develop bipolar disorder.

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I needed an intellectual oasis to deal with the anguish and frustration triggered by the monumental amount of misleading information that was included in the well-written article “Not all mood swings are bipolar disorder” (Current Psychiatry, February 2011, p. 38-52). Fortunately, a commentary by Dr. Irene Abramovich (“Breaking the box,” Comments & Controversies, Current Psychiatry, February 2011, p. 59) appeared as a therapeutic elixir. I believe that the “mood swings” article is filled with examples of how dangerous “cookbook” medicine can be.

Dr. Kowatch and colleagues use an expression that can be applied to the so-called diagnoses oppositional defiant disorder (ODD) and conduct disorder (CD): “Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but they are not diagnostic as an isolated symptom. “ A similar concept was my position in a debate titled “Childhood conduct disorder and oppositional-defiant disorders are common manifestations of bipolar disorder” in which I argued that ODD and CD are behavioral expressions of genuine diagnoses.¹ Besides bipolar disorder, I also have seen obsessive-compulsive disorder, social anxiety disorder, and even sexual abuse labeled as “ODD” because the child refuses to be around people (such as a classroom) or is distracted by intrusive thoughts or flashbacks and turns hostile when reproached in front of the class.

In my view, Dr. Kowatch and colleagues give undeserved credit to the behavioral scales (the “cookbooks” of psychiatry) to make diagnoses and seem to miss warning signs in patients’ family history, ie, “history of depression and anxiety” (many times this translates as agitated/dysphoric mania) and “drinking problems, “ which frequently is found in undiagnosed bipolar spectrum patients who use alcohol to “shoot down” racing thoughts that interfere with normal sleep.

From January 2010 to February 2011, I reviewed charts and interviewed patients and families of 1, 654 patients with diagnoses of attention-deficit/hyperactivity disorder co-morbid with ODD, bipolar disorder, generalized anxiety disorder, and even 2 diagnoses that are not allowed by DSM rules: autism and mental retardation. The data from this study, which covers 12 counties that represent the 5 geographical areas of Florida, are being analyzed. In the meantime, I refer readers to my poster presentation from the 2010 U. S. Psychiatric and Mental Health Congress “Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. “² In this study 44 patients were followed for at least 5 years (10 patients were observed for 7 years and a similar number for 6) and none had “oppositional” behavior after the diagnoses were treated. One caveat is that I placed antipsychotics in the same category as conventional mood stabilizers because 5 patients considered to be “inattentive” and “oppositional” actually had schizophrenia.

I oppose the authors’ assertion that “it can be difficult to differentiate the mood swings and symptoms of ODD from those of pediatric BD. “ My experience is that it is simple if we consider all diagnostic possibilities and obtain a thorough family history, which usually includes alcoholism, cannabis abuse, moodiness, suicide completion, unstable lifestyle, etc.

Manuel Mota-Castillo, MD
Assistant Clinical Professor
St. Matthews University
Voluntary Faculty
University of Central Florida
Lake Mary, FL

References

1. Mota-Castillo M, Steiner H. Childhood conduct disorder and oppositional-defiant disorder are common manifestations of bipolar disorder pro and con. Journal of Bipolar Disorders: Reviews and Commentaries. 2005;3:3,15-17.

2. Mota-Castillo M. Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. Poster presented at: 23rd Annual U.S. Psychiatric and Mental Health Congress; November 20 2010; Orlando, FL.

The authors respond

We never suggested that clinicians use “cookbook medicine. “ The “behavioral scales” we recommended in our article are well-validated and reliable tools that allow a clinician to effectively elicit a great deal of useful information from patients and their parents about presenting problems and symptoms. This information can be used with other clinical information to make an accurate diagnosis and subsequent treatment plan.

The purpose of our article was to share our experiences in the differential diagnosis of mood swings in children and adolescents and to suggest that there are other diagnoses that cause mood swings besides bipolar disorder. Although a family history of mood disorders is important, it is also important to recognize that a recent, state-of-the-art study by Birmaher et al¹ reported that 10% of children of parents with bipolar disorder had a bipolar spectrum disorder. That means that 90% did not have bipolar disorder. It is important to remember this when evaluating children of parents with bipolar disorder. Although these children’s risk for developing bipolar disorder is increased compared with the general population, it is more likely that they will not develop bipolar disorder.

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I needed an intellectual oasis to deal with the anguish and frustration triggered by the monumental amount of misleading information that was included in the well-written article “Not all mood swings are bipolar disorder” (Current Psychiatry, February 2011, p. 38-52). Fortunately, a commentary by Dr. Irene Abramovich (“Breaking the box,” Comments & Controversies, Current Psychiatry, February 2011, p. 59) appeared as a therapeutic elixir. I believe that the “mood swings” article is filled with examples of how dangerous “cookbook” medicine can be.

Dr. Kowatch and colleagues use an expression that can be applied to the so-called diagnoses oppositional defiant disorder (ODD) and conduct disorder (CD): “Mood swings are analogous to a fever in pediatrics—they indicate something potentially is wrong with the patient, but they are not diagnostic as an isolated symptom. “ A similar concept was my position in a debate titled “Childhood conduct disorder and oppositional-defiant disorders are common manifestations of bipolar disorder” in which I argued that ODD and CD are behavioral expressions of genuine diagnoses.¹ Besides bipolar disorder, I also have seen obsessive-compulsive disorder, social anxiety disorder, and even sexual abuse labeled as “ODD” because the child refuses to be around people (such as a classroom) or is distracted by intrusive thoughts or flashbacks and turns hostile when reproached in front of the class.

In my view, Dr. Kowatch and colleagues give undeserved credit to the behavioral scales (the “cookbooks” of psychiatry) to make diagnoses and seem to miss warning signs in patients’ family history, ie, “history of depression and anxiety” (many times this translates as agitated/dysphoric mania) and “drinking problems, “ which frequently is found in undiagnosed bipolar spectrum patients who use alcohol to “shoot down” racing thoughts that interfere with normal sleep.

From January 2010 to February 2011, I reviewed charts and interviewed patients and families of 1, 654 patients with diagnoses of attention-deficit/hyperactivity disorder co-morbid with ODD, bipolar disorder, generalized anxiety disorder, and even 2 diagnoses that are not allowed by DSM rules: autism and mental retardation. The data from this study, which covers 12 counties that represent the 5 geographical areas of Florida, are being analyzed. In the meantime, I refer readers to my poster presentation from the 2010 U. S. Psychiatric and Mental Health Congress “Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. “² In this study 44 patients were followed for at least 5 years (10 patients were observed for 7 years and a similar number for 6) and none had “oppositional” behavior after the diagnoses were treated. One caveat is that I placed antipsychotics in the same category as conventional mood stabilizers because 5 patients considered to be “inattentive” and “oppositional” actually had schizophrenia.

I oppose the authors’ assertion that “it can be difficult to differentiate the mood swings and symptoms of ODD from those of pediatric BD. “ My experience is that it is simple if we consider all diagnostic possibilities and obtain a thorough family history, which usually includes alcoholism, cannabis abuse, moodiness, suicide completion, unstable lifestyle, etc.

Manuel Mota-Castillo, MD
Assistant Clinical Professor
St. Matthews University
Voluntary Faculty
University of Central Florida
Lake Mary, FL

References

1. Mota-Castillo M, Steiner H. Childhood conduct disorder and oppositional-defiant disorder are common manifestations of bipolar disorder pro and con. Journal of Bipolar Disorders: Reviews and Commentaries. 2005;3:3,15-17.

2. Mota-Castillo M. Extinction of oppositional-defiant symptoms following treatment with mood stabilizers. Poster presented at: 23rd Annual U.S. Psychiatric and Mental Health Congress; November 20 2010; Orlando, FL.

The authors respond

We never suggested that clinicians use “cookbook medicine. “ The “behavioral scales” we recommended in our article are well-validated and reliable tools that allow a clinician to effectively elicit a great deal of useful information from patients and their parents about presenting problems and symptoms. This information can be used with other clinical information to make an accurate diagnosis and subsequent treatment plan.

The purpose of our article was to share our experiences in the differential diagnosis of mood swings in children and adolescents and to suggest that there are other diagnoses that cause mood swings besides bipolar disorder. Although a family history of mood disorders is important, it is also important to recognize that a recent, state-of-the-art study by Birmaher et al¹ reported that 10% of children of parents with bipolar disorder had a bipolar spectrum disorder. That means that 90% did not have bipolar disorder. It is important to remember this when evaluating children of parents with bipolar disorder. Although these children’s risk for developing bipolar disorder is increased compared with the general population, it is more likely that they will not develop bipolar disorder.

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I thought Dr. Kowatch and colleagues took an important first step in pointing out that not all mood swings in children and adolescents are symptoms of bipolar disorder (“Not all mood swings are bipolar disorder,“ Current Psychiatry, February 2011, p. 38-52). They reviewed some of the other psychiatric conditions known to cause labile moods. One glaring omission is borderline personality disorder (BPD).

I am the medical director of a specialized unit that uses dialectical behavioral therapy (DBT) to treat children and adolescents with BPD. We have treated approximately 300 young women on the residential unit and many present similarly: multiple hospitalizations, multiple robust yet failed medication trials, severe and recurrent self-injury, suicide attempts, and a large degree of hopelessness. Most arrive with previous diagnoses of mood disorder not otherwise specified, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, and others. It is when their outpatient psychiatrists and mental health teams have grown frustrated at the lack of enduring progress and are faced with the treatment demands of the borderline patient that a BPD diagnosis is considered. Even though research suggests that BPD—or at least some of its symptoms—begins in the late latency period of childhood,¹ treatment typically is not sought until late adolescence. This is the case despite the fact that BPD has a better prognosis than other serious mental illnesses, such as bipolar disorder.^2,3 Adult BPD patients almost universally recognize that their inability to regulate their mood started in late childhood and early adolescence. Structural and functional neuroimaging has revealed a dysfunctional network of brain regions that seem to mediate important aspects of BPD symptomatology.^4-6

These children have marked mood swings and great difficulty regulating their moods. The mood swings of BPD are not responsive to current medication unless there is comorbid bipolar disorder, in which case treatment with mood stabilizers helps improve vegetative symptoms such as sleep and energy, and reduce racing thoughts, pressured speech, and irritability. What these medications do not treat is the “reactive” mood swings that are characteristic of BPD. The mood reactivity often is triggered by interpersonal or intrapersonal conflict and rarely is long-lived.

Many children and adolescents are moody and most do not have a major psychiatric disorder. Of those who do, it is a great risk to patients’ health to not consider BPD, especially given new and empirically validated treatments, such as DBT. Astute clinicians should keep this diagnosis in mind when treating adolescents with moodiness, particularly when the mood is predominantly reactive to life’s stressors, when other features of the presentation do not fit neatly into a bipolar picture, and when multiple medications fail. On our unit, we have seen that the cognitive-behavioral strategies of DBT help patients even when BPD is not the diagnosis.

I would like to thank Dr. Kowatch and colleagues for expanding our thinking on mood swings and encourage readers to go one step further.

Blaise Aguirre, MD
Medical Director
Adolescent DBT Residential Program
McLean Hospital
Belmont, MA
Instructor in Psychiatry
Harvard Medical School
Boston, MA

References

1. Zanarini MC, Frankenburg FR, Khera GS, et al. Treatment histories of borderline inpatients. Compr Psychiatry. 2001;42:144-150.

2. Tohen M, Hennen J, Zarate CM Jr, et al. Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am J Psychiatry. 2000;157:220-228.

3. Coryell W, Endicott J, Maser JD, et al. The likelihood of recurrence in bipolar affective disorder: the importance of episode recency. J Affect Disord. 1995;33:201-206.

4. De La Fuente JM, Goldman S, Stanus E, et al. Brain glucose metabolism in borderline personality disorder. J Psychiatr Res. 1997;31:531-541.

5. Soloff PH, Meltzer CC, Becker C, et al. Impulsivity and prefrontal hypometabolism in borderline personality disorder. Psychiatry Res. 2003;123:153-163.

6. Juengling FD, Schmahl C, Hesslinger B, et al. Positron emission tomography in female patients with borderline personality disorder. J Psychiatr Res. 2003;37:109-115.

The authors respond

We welcome comments about the importance of a thorough diagnostic evaluation to tease out possible etiologies of “mood swings, “ including psychosocial factors, as in personality disorders. Nevertheless, the debate about diagnosing personality disorders in children and adolescents is not settled. Developmentally, children and adolescents have continuous changes in biology and brain function. There is significantly more empirical evidence of reactive attachment disorders in childhood and adolescence that integrate the affective changes seen in children who live in chaotic environments. DSM defines BPD as a pervasive pattern of instability of interpersonal relationships that begins by early adulthood.¹ Many children with diagnoses of posttraumatic stress disorder, mood disorder, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, etc. also can have difficulties in relating to others caused by their neurobiologic deficits, which also may limit response to medication. Furthermore, children with learning disorders also can misperceive motives of others and thus have pervasive patterns of relational instability.

The research Dr. Aguirre suggested is based on treatment histories and not rigorous study methodology. Most empirical evidence of personality disorders is strongly influenced by psychoanalytic literature regarding object relations, which is in flux because of emerging attention to attachment theory and progress in neurologic studies in the evaluation of temperamental variations related to the influence of mirror neurons.²

We also take issue with the comment that “BPD has a better prognosis than other serious men tal illn esses, such as bipolar disorder” There have been significant efforts in studying the role family can have in the outcomes of mood disorder treatment.³

Finally, there is evidence that in adults medication can be beneficial in treating the affective deregulation of patients with BPD who do not have comorbid disorders.⁴

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I thought Dr. Kowatch and colleagues took an important first step in pointing out that not all mood swings in children and adolescents are symptoms of bipolar disorder (“Not all mood swings are bipolar disorder,“ Current Psychiatry, February 2011, p. 38-52). They reviewed some of the other psychiatric conditions known to cause labile moods. One glaring omission is borderline personality disorder (BPD).

I am the medical director of a specialized unit that uses dialectical behavioral therapy (DBT) to treat children and adolescents with BPD. We have treated approximately 300 young women on the residential unit and many present similarly: multiple hospitalizations, multiple robust yet failed medication trials, severe and recurrent self-injury, suicide attempts, and a large degree of hopelessness. Most arrive with previous diagnoses of mood disorder not otherwise specified, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, and others. It is when their outpatient psychiatrists and mental health teams have grown frustrated at the lack of enduring progress and are faced with the treatment demands of the borderline patient that a BPD diagnosis is considered. Even though research suggests that BPD—or at least some of its symptoms—begins in the late latency period of childhood,¹ treatment typically is not sought until late adolescence. This is the case despite the fact that BPD has a better prognosis than other serious mental illnesses, such as bipolar disorder.^2,3 Adult BPD patients almost universally recognize that their inability to regulate their mood started in late childhood and early adolescence. Structural and functional neuroimaging has revealed a dysfunctional network of brain regions that seem to mediate important aspects of BPD symptomatology.^4-6

These children have marked mood swings and great difficulty regulating their moods. The mood swings of BPD are not responsive to current medication unless there is comorbid bipolar disorder, in which case treatment with mood stabilizers helps improve vegetative symptoms such as sleep and energy, and reduce racing thoughts, pressured speech, and irritability. What these medications do not treat is the “reactive” mood swings that are characteristic of BPD. The mood reactivity often is triggered by interpersonal or intrapersonal conflict and rarely is long-lived.

Many children and adolescents are moody and most do not have a major psychiatric disorder. Of those who do, it is a great risk to patients’ health to not consider BPD, especially given new and empirically validated treatments, such as DBT. Astute clinicians should keep this diagnosis in mind when treating adolescents with moodiness, particularly when the mood is predominantly reactive to life’s stressors, when other features of the presentation do not fit neatly into a bipolar picture, and when multiple medications fail. On our unit, we have seen that the cognitive-behavioral strategies of DBT help patients even when BPD is not the diagnosis.

I would like to thank Dr. Kowatch and colleagues for expanding our thinking on mood swings and encourage readers to go one step further.

Blaise Aguirre, MD
Medical Director
Adolescent DBT Residential Program
McLean Hospital
Belmont, MA
Instructor in Psychiatry
Harvard Medical School
Boston, MA

References

1. Zanarini MC, Frankenburg FR, Khera GS, et al. Treatment histories of borderline inpatients. Compr Psychiatry. 2001;42:144-150.

2. Tohen M, Hennen J, Zarate CM Jr, et al. Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am J Psychiatry. 2000;157:220-228.

3. Coryell W, Endicott J, Maser JD, et al. The likelihood of recurrence in bipolar affective disorder: the importance of episode recency. J Affect Disord. 1995;33:201-206.

4. De La Fuente JM, Goldman S, Stanus E, et al. Brain glucose metabolism in borderline personality disorder. J Psychiatr Res. 1997;31:531-541.

5. Soloff PH, Meltzer CC, Becker C, et al. Impulsivity and prefrontal hypometabolism in borderline personality disorder. Psychiatry Res. 2003;123:153-163.

6. Juengling FD, Schmahl C, Hesslinger B, et al. Positron emission tomography in female patients with borderline personality disorder. J Psychiatr Res. 2003;37:109-115.

The authors respond

We welcome comments about the importance of a thorough diagnostic evaluation to tease out possible etiologies of “mood swings, “ including psychosocial factors, as in personality disorders. Nevertheless, the debate about diagnosing personality disorders in children and adolescents is not settled. Developmentally, children and adolescents have continuous changes in biology and brain function. There is significantly more empirical evidence of reactive attachment disorders in childhood and adolescence that integrate the affective changes seen in children who live in chaotic environments. DSM defines BPD as a pervasive pattern of instability of interpersonal relationships that begins by early adulthood.¹ Many children with diagnoses of posttraumatic stress disorder, mood disorder, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, etc. also can have difficulties in relating to others caused by their neurobiologic deficits, which also may limit response to medication. Furthermore, children with learning disorders also can misperceive motives of others and thus have pervasive patterns of relational instability.

The research Dr. Aguirre suggested is based on treatment histories and not rigorous study methodology. Most empirical evidence of personality disorders is strongly influenced by psychoanalytic literature regarding object relations, which is in flux because of emerging attention to attachment theory and progress in neurologic studies in the evaluation of temperamental variations related to the influence of mirror neurons.²

We also take issue with the comment that “BPD has a better prognosis than other serious men tal illn esses, such as bipolar disorder” There have been significant efforts in studying the role family can have in the outcomes of mood disorder treatment.³

Finally, there is evidence that in adults medication can be beneficial in treating the affective deregulation of patients with BPD who do not have comorbid disorders.⁴

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

I thought Dr. Kowatch and colleagues took an important first step in pointing out that not all mood swings in children and adolescents are symptoms of bipolar disorder (“Not all mood swings are bipolar disorder,“ Current Psychiatry, February 2011, p. 38-52). They reviewed some of the other psychiatric conditions known to cause labile moods. One glaring omission is borderline personality disorder (BPD).

I am the medical director of a specialized unit that uses dialectical behavioral therapy (DBT) to treat children and adolescents with BPD. We have treated approximately 300 young women on the residential unit and many present similarly: multiple hospitalizations, multiple robust yet failed medication trials, severe and recurrent self-injury, suicide attempts, and a large degree of hopelessness. Most arrive with previous diagnoses of mood disorder not otherwise specified, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, and others. It is when their outpatient psychiatrists and mental health teams have grown frustrated at the lack of enduring progress and are faced with the treatment demands of the borderline patient that a BPD diagnosis is considered. Even though research suggests that BPD—or at least some of its symptoms—begins in the late latency period of childhood,¹ treatment typically is not sought until late adolescence. This is the case despite the fact that BPD has a better prognosis than other serious mental illnesses, such as bipolar disorder.^2,3 Adult BPD patients almost universally recognize that their inability to regulate their mood started in late childhood and early adolescence. Structural and functional neuroimaging has revealed a dysfunctional network of brain regions that seem to mediate important aspects of BPD symptomatology.^4-6

These children have marked mood swings and great difficulty regulating their moods. The mood swings of BPD are not responsive to current medication unless there is comorbid bipolar disorder, in which case treatment with mood stabilizers helps improve vegetative symptoms such as sleep and energy, and reduce racing thoughts, pressured speech, and irritability. What these medications do not treat is the “reactive” mood swings that are characteristic of BPD. The mood reactivity often is triggered by interpersonal or intrapersonal conflict and rarely is long-lived.

Many children and adolescents are moody and most do not have a major psychiatric disorder. Of those who do, it is a great risk to patients’ health to not consider BPD, especially given new and empirically validated treatments, such as DBT. Astute clinicians should keep this diagnosis in mind when treating adolescents with moodiness, particularly when the mood is predominantly reactive to life’s stressors, when other features of the presentation do not fit neatly into a bipolar picture, and when multiple medications fail. On our unit, we have seen that the cognitive-behavioral strategies of DBT help patients even when BPD is not the diagnosis.

I would like to thank Dr. Kowatch and colleagues for expanding our thinking on mood swings and encourage readers to go one step further.

Blaise Aguirre, MD
Medical Director
Adolescent DBT Residential Program
McLean Hospital
Belmont, MA
Instructor in Psychiatry
Harvard Medical School
Boston, MA

References

1. Zanarini MC, Frankenburg FR, Khera GS, et al. Treatment histories of borderline inpatients. Compr Psychiatry. 2001;42:144-150.

2. Tohen M, Hennen J, Zarate CM Jr, et al. Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am J Psychiatry. 2000;157:220-228.

3. Coryell W, Endicott J, Maser JD, et al. The likelihood of recurrence in bipolar affective disorder: the importance of episode recency. J Affect Disord. 1995;33:201-206.

4. De La Fuente JM, Goldman S, Stanus E, et al. Brain glucose metabolism in borderline personality disorder. J Psychiatr Res. 1997;31:531-541.

5. Soloff PH, Meltzer CC, Becker C, et al. Impulsivity and prefrontal hypometabolism in borderline personality disorder. Psychiatry Res. 2003;123:153-163.

6. Juengling FD, Schmahl C, Hesslinger B, et al. Positron emission tomography in female patients with borderline personality disorder. J Psychiatr Res. 2003;37:109-115.

The authors respond

We welcome comments about the importance of a thorough diagnostic evaluation to tease out possible etiologies of “mood swings, “ including psychosocial factors, as in personality disorders. Nevertheless, the debate about diagnosing personality disorders in children and adolescents is not settled. Developmentally, children and adolescents have continuous changes in biology and brain function. There is significantly more empirical evidence of reactive attachment disorders in childhood and adolescence that integrate the affective changes seen in children who live in chaotic environments. DSM defines BPD as a pervasive pattern of instability of interpersonal relationships that begins by early adulthood.¹ Many children with diagnoses of posttraumatic stress disorder, mood disorder, bipolar disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, etc. also can have difficulties in relating to others caused by their neurobiologic deficits, which also may limit response to medication. Furthermore, children with learning disorders also can misperceive motives of others and thus have pervasive patterns of relational instability.

The research Dr. Aguirre suggested is based on treatment histories and not rigorous study methodology. Most empirical evidence of personality disorders is strongly influenced by psychoanalytic literature regarding object relations, which is in flux because of emerging attention to attachment theory and progress in neurologic studies in the evaluation of temperamental variations related to the influence of mirror neurons.²

We also take issue with the comment that “BPD has a better prognosis than other serious men tal illn esses, such as bipolar disorder” There have been significant efforts in studying the role family can have in the outcomes of mood disorder treatment.³

Finally, there is evidence that in adults medication can be beneficial in treating the affective deregulation of patients with BPD who do not have comorbid disorders.⁴

Robert A. Kowatch, MD, PhD
Professor of Psychiatry and Pediatrics

Erin Monroe, CNS
Clinical Nurse Specialist
Division of Psychiatry

Sergio V. Delgado, MD
Associate Professor of Psychiatry
and Pediatrics
Cincinnati Children’s Hospital Medical Center
Cincinnati, OH

You’ve heard about the 2 certainties in life: death and taxes. In psychiatric practice with complex and chronic patients, there is a third certainty: polypharmacy. It ranges from thoughtful to indiscriminate and seems to be entrenched in clinical practice, possibly reflecting practitioners’ desperation in trying to manage severely ill, treatment-resistant patients, usually in the absence of evidence-based guidelines.

I never fail to encounter polypharmacy in hospitals or clinics where I consult. I always wondered how the patient’s doctor knew which drug was exerting a therapeutic effect or which drug was causing side effects (parkinsonism, akathisia, sedation, orthostasis, dizziness, headache, blurry vision, etc. ). Over time, I came to categorize polypharmacy into 4 subtypes that span the spectrum from sensible to absurd. Here is my personal classification, which I trust that you, my readers, have witnessed as well.

Necessary polypharmacy. This variant of polypharmacy is evidence-based and proven in double-blind studies to be more effective than monotherapy. The most prominent example is adding an atypical antipsychotic to a mood stabilizer in bipolar mania. In fact, the superior efficacy of combination therapy in bipolar disorder is one of the oldest forms of rational polypharmacy, is supported by FDA trials, and is indicated whenever mood stabilizer monotherapy is not sufficient. For example, combining lithium and valproate is superior to either drug alone. Another example of FDA-approved combinations is combining small doses of an atypical antipsychotic to an antidepressant for treatment-resistant depression.

Reasonable polypharmacy. Although many of the combinations in this category are not FDA-approved, controlled studies support their use for suffering patients. Examples include:

• An atypical antipsychotic added to a selective serotonin reuptake inhibitor (SSRI) for obsessive-compulsive disorder (OCD) patients who do not improve on SSRI monotherapy.
• Modafinil added to clozapine in patients who suffer substantial and persistent daytime sedation or somnolence.
• Combining 2 antidepressants for major depressive disorder patients who partially respond to 1 antidepressant.
• Combining a mood stabilizer with an antidepressant for bipolar depression to prevent mood switching.

Ridiculous polypharmacy. The sky is the limit to the variations and degrees of ridiculous polypharmacy, but the theme is the same: an absurd concoction of psychotropic drugs across several classes, often including multiple agents from 1 or several classes. Here are examples I have seen in patient records:

• Two atypicals, an anticholinergic, a mood stabilizer, an antidepressant, and 2 benzodiazepines.
• Three antipsychotics (2 atypicals and 1 typical), 2 antidepressants, 3 sedative/hypnotics, and an anticonvulsant for weight control.
• This one takes the cake: 6 antipsychotics (2 typicals and 4 atypicals), plus an anticholinergic, 3 mood stabilizers, 2 antidepressants, 2 sleeping pills, a hypoglycemic agent, 2 antihypertensives, and a statin.

Hazardous polypharmacy. In this category, serious medical complications, toxic effects, or death may occur because of careless combinations of drugs that may interact to produce dangerous kinetic interactions or exacerbate a pre-existing medical condition. Examples include:

• Combining 1 psychotropic with another that may inhibit its metabolism (eg, prescribing fluvoxamine to a severely psychotic patient who developed OCD while receiving clozapine). There have been several toxic reactions and even death because fluvoxamine inhibits cytochrome 1A2, which metabolizes clozapine, thus increasing clozapine blood level by 400% to 500%.
• Combining 2 injectable drugs for agitation that may cause a serious medical complication. An example would be injecting a benzodiazepine such as lorazepam in a patient receiving olanzapine IM, which can cause severe respiratory depression or death.
• Combining several drugs, each of which may prolong the QTc interval, resulting in syncope or torsade de pointes.

Psychopharmacology can relieve the terrible anguish of psychosis, depression, or anxiety, but it also can carry iatrogenic risks if it is not based on scientific evidence. The practice of psychopharmacology requires the fully integrated skills of medical and psychiatric training to maximize benefit while avoiding harm. It also requires basic arithmetic skills: to consider subtracting drugs, not only adding them!

You’ve heard about the 2 certainties in life: death and taxes. In psychiatric practice with complex and chronic patients, there is a third certainty: polypharmacy. It ranges from thoughtful to indiscriminate and seems to be entrenched in clinical practice, possibly reflecting practitioners’ desperation in trying to manage severely ill, treatment-resistant patients, usually in the absence of evidence-based guidelines.

I never fail to encounter polypharmacy in hospitals or clinics where I consult. I always wondered how the patient’s doctor knew which drug was exerting a therapeutic effect or which drug was causing side effects (parkinsonism, akathisia, sedation, orthostasis, dizziness, headache, blurry vision, etc. ). Over time, I came to categorize polypharmacy into 4 subtypes that span the spectrum from sensible to absurd. Here is my personal classification, which I trust that you, my readers, have witnessed as well.

Necessary polypharmacy. This variant of polypharmacy is evidence-based and proven in double-blind studies to be more effective than monotherapy. The most prominent example is adding an atypical antipsychotic to a mood stabilizer in bipolar mania. In fact, the superior efficacy of combination therapy in bipolar disorder is one of the oldest forms of rational polypharmacy, is supported by FDA trials, and is indicated whenever mood stabilizer monotherapy is not sufficient. For example, combining lithium and valproate is superior to either drug alone. Another example of FDA-approved combinations is combining small doses of an atypical antipsychotic to an antidepressant for treatment-resistant depression.

Reasonable polypharmacy. Although many of the combinations in this category are not FDA-approved, controlled studies support their use for suffering patients. Examples include:

• An atypical antipsychotic added to a selective serotonin reuptake inhibitor (SSRI) for obsessive-compulsive disorder (OCD) patients who do not improve on SSRI monotherapy.
• Modafinil added to clozapine in patients who suffer substantial and persistent daytime sedation or somnolence.
• Combining 2 antidepressants for major depressive disorder patients who partially respond to 1 antidepressant.
• Combining a mood stabilizer with an antidepressant for bipolar depression to prevent mood switching.

Ridiculous polypharmacy. The sky is the limit to the variations and degrees of ridiculous polypharmacy, but the theme is the same: an absurd concoction of psychotropic drugs across several classes, often including multiple agents from 1 or several classes. Here are examples I have seen in patient records:

• Two atypicals, an anticholinergic, a mood stabilizer, an antidepressant, and 2 benzodiazepines.
• Three antipsychotics (2 atypicals and 1 typical), 2 antidepressants, 3 sedative/hypnotics, and an anticonvulsant for weight control.
• This one takes the cake: 6 antipsychotics (2 typicals and 4 atypicals), plus an anticholinergic, 3 mood stabilizers, 2 antidepressants, 2 sleeping pills, a hypoglycemic agent, 2 antihypertensives, and a statin.

Hazardous polypharmacy. In this category, serious medical complications, toxic effects, or death may occur because of careless combinations of drugs that may interact to produce dangerous kinetic interactions or exacerbate a pre-existing medical condition. Examples include:

• Combining 1 psychotropic with another that may inhibit its metabolism (eg, prescribing fluvoxamine to a severely psychotic patient who developed OCD while receiving clozapine). There have been several toxic reactions and even death because fluvoxamine inhibits cytochrome 1A2, which metabolizes clozapine, thus increasing clozapine blood level by 400% to 500%.
• Combining 2 injectable drugs for agitation that may cause a serious medical complication. An example would be injecting a benzodiazepine such as lorazepam in a patient receiving olanzapine IM, which can cause severe respiratory depression or death.
• Combining several drugs, each of which may prolong the QTc interval, resulting in syncope or torsade de pointes.

Psychopharmacology can relieve the terrible anguish of psychosis, depression, or anxiety, but it also can carry iatrogenic risks if it is not based on scientific evidence. The practice of psychopharmacology requires the fully integrated skills of medical and psychiatric training to maximize benefit while avoiding harm. It also requires basic arithmetic skills: to consider subtracting drugs, not only adding them!

You’ve heard about the 2 certainties in life: death and taxes. In psychiatric practice with complex and chronic patients, there is a third certainty: polypharmacy. It ranges from thoughtful to indiscriminate and seems to be entrenched in clinical practice, possibly reflecting practitioners’ desperation in trying to manage severely ill, treatment-resistant patients, usually in the absence of evidence-based guidelines.

I never fail to encounter polypharmacy in hospitals or clinics where I consult. I always wondered how the patient’s doctor knew which drug was exerting a therapeutic effect or which drug was causing side effects (parkinsonism, akathisia, sedation, orthostasis, dizziness, headache, blurry vision, etc. ). Over time, I came to categorize polypharmacy into 4 subtypes that span the spectrum from sensible to absurd. Here is my personal classification, which I trust that you, my readers, have witnessed as well.

Necessary polypharmacy. This variant of polypharmacy is evidence-based and proven in double-blind studies to be more effective than monotherapy. The most prominent example is adding an atypical antipsychotic to a mood stabilizer in bipolar mania. In fact, the superior efficacy of combination therapy in bipolar disorder is one of the oldest forms of rational polypharmacy, is supported by FDA trials, and is indicated whenever mood stabilizer monotherapy is not sufficient. For example, combining lithium and valproate is superior to either drug alone. Another example of FDA-approved combinations is combining small doses of an atypical antipsychotic to an antidepressant for treatment-resistant depression.

Reasonable polypharmacy. Although many of the combinations in this category are not FDA-approved, controlled studies support their use for suffering patients. Examples include:

• An atypical antipsychotic added to a selective serotonin reuptake inhibitor (SSRI) for obsessive-compulsive disorder (OCD) patients who do not improve on SSRI monotherapy.
• Modafinil added to clozapine in patients who suffer substantial and persistent daytime sedation or somnolence.
• Combining 2 antidepressants for major depressive disorder patients who partially respond to 1 antidepressant.
• Combining a mood stabilizer with an antidepressant for bipolar depression to prevent mood switching.

Ridiculous polypharmacy. The sky is the limit to the variations and degrees of ridiculous polypharmacy, but the theme is the same: an absurd concoction of psychotropic drugs across several classes, often including multiple agents from 1 or several classes. Here are examples I have seen in patient records:

• Two atypicals, an anticholinergic, a mood stabilizer, an antidepressant, and 2 benzodiazepines.
• Three antipsychotics (2 atypicals and 1 typical), 2 antidepressants, 3 sedative/hypnotics, and an anticonvulsant for weight control.
• This one takes the cake: 6 antipsychotics (2 typicals and 4 atypicals), plus an anticholinergic, 3 mood stabilizers, 2 antidepressants, 2 sleeping pills, a hypoglycemic agent, 2 antihypertensives, and a statin.

Hazardous polypharmacy. In this category, serious medical complications, toxic effects, or death may occur because of careless combinations of drugs that may interact to produce dangerous kinetic interactions or exacerbate a pre-existing medical condition. Examples include:

• Combining 1 psychotropic with another that may inhibit its metabolism (eg, prescribing fluvoxamine to a severely psychotic patient who developed OCD while receiving clozapine). There have been several toxic reactions and even death because fluvoxamine inhibits cytochrome 1A2, which metabolizes clozapine, thus increasing clozapine blood level by 400% to 500%.
• Combining 2 injectable drugs for agitation that may cause a serious medical complication. An example would be injecting a benzodiazepine such as lorazepam in a patient receiving olanzapine IM, which can cause severe respiratory depression or death.
• Combining several drugs, each of which may prolong the QTc interval, resulting in syncope or torsade de pointes.

Psychopharmacology can relieve the terrible anguish of psychosis, depression, or anxiety, but it also can carry iatrogenic risks if it is not based on scientific evidence. The practice of psychopharmacology requires the fully integrated skills of medical and psychiatric training to maximize benefit while avoiding harm. It also requires basic arithmetic skills: to consider subtracting drugs, not only adding them!

Discuss this article at www.facebook.com/CurrentPsychiatry

Among DSM axis I diagnoses, bipolar disorder (BD) has the highest rates of comorbid substance use disorders (SUDs).^1-3 Approximately 60% of patients with bipolar I disorder have a lifetime diagnosis of an SUD.¹ Excluding tobacco, alcohol is the substance most often abused by BD patients, followed by cannabis, amphetamines, and cocaine.^1-3

BD patients with comorbid SUD usually exhibit more severe clinical presentations and poorer outcomes than their counterparts without SUDs. Compared with patients with BD alone, those with BD and SUD comorbidity (BD-SUD) experience earlier onset of mood symptoms; higher rates of anxiety disorders, suicide attempts, accidents, hospitalizations, and rapid cycling; more depressive episodes; and lower treatment compliance.^4-9

Several treatment options are available for patients with BD-SUD, including psychotherapy modalities, medications primarily used to treat BD, and medications primarily used to treat SUDs. Evidence-based support for these treatments remains limited, and no treatment of choice has emerged. This article reviews evidence on the longer-term treatment of BD-SUD, including general strategies and specific psychosocial and pharmacologic interventions. Short-term treatment strategies, such as pharmacotherapy for detoxification, are outside the scope of this review.

General strategies

The causes of BD-SUD are complex. Evidence suggests that the presence of affective symptoms is associated with an increased risk for substance misuse. This should be kept in mind when treating a patient with BD-SUD because controlling mood symptoms probably will help control substance abuse. However, evidence also shows that SUDs may be independent of mood episodes. Therefore, treating only mood symptoms in the hope that doing so will control substance abuse may not be enough.

Because the negative impact of SUDs on BD outcome is well documented, inform patients that limiting their use of alcohol and/or drugs is vital to control their mood disorder. Efforts to educate, stimulate, and support patients to moderate or stop their alcohol and/or drug use are likely to result in positive changes.¹⁰ Therefore, treatment for BD-SUD should follow, in part, the same recommendations for treatment of SUDs in patients with no comorbid axis I disorders:

• identify the problem (ie, the existence of a comorbid SUD)
• share your concerns with your patient
• offer appropriate and specific treatments, such as detoxification and/or self-help and counseling programs.¹⁰

Because SUDs usually are chronic and relapsing conditions, periods of drug and/ or alcohol use should be expected and not considered a sign of treatment failure. In addition, integrating treatment for both conditions probably is better than managing each separately. Therefore, targeting BD symptoms with mood-stabilizing medications and substance abuse with nonpharmacologic modalities such as drug counseling likely will bring about the best results.

Compared with BD patients without comorbid SUD, BD-SUD patients have a 7-fold increased risk of antidepressantinduced mania.¹¹ Therefore, antidepressants should be prescribed cautiously for patients with BD-SUD.

Integrated psychosocial therapy

BD-SUD patients may benefit from attending self-help programs such as Alcoholics Anonymous and Narcotics Anonymous, provided their mood is stable enough to allow them to participate. Other forms of psychotherapy for BD-SUD patients include standard group drug counseling and integrated group therapy that simultaneously addresses both conditions.

Integrated group therapy is based on the premise that changing maladaptive mood cognitions and behaviors will facilitate recovery from SUDs, and changing maladaptive substance use cognitions and behaviors will facilitate recovery from mood disorders.¹² In a recent randomized controlled trial, 62 BD-SUD patients were blindly assigned to integrated group therapy or standard group drug counseling and followed for 3 months.¹² Pharmacotherapy was prescribed as usual. Substance use decreased for both groups. However, compared with patients in the drug counseling group, those who participated in integrated group therapy spent fewer days using substances in general and alcohol in particular, fewer days using alcohol to intoxication, and had a shorter time from treatment initiation to the first abstinent month. There were no differences between groups in number of weeks in a mood episode.

Pharmacotherapy options

For a table that summarizes the dosages and indications of the medications used to treat BD-SUD that are described below, visit this article at CurrentPsychiatry.com.

Table

Medications used to treat substance use disorders in bipolar disorder patients*

Lithium. Given its well-documented mood stabilizing effect, lithium would seem to be a reasonable option to treat BD-SUD patients, but scant evidence supports its role as an anti-alcohol or anti-drug medication (Table 1).^13,14 Lithium’s efficacy was evaluated in a study of 25 adolescents suffering from mood disorders (mostly BD) and comorbid SUDs (mostly alcohol and cannabis) randomized to receive lithium or placebo for 6 weeks.¹³ Lithium was well tolerated and improved psychiatric symptoms. At week 3, patients receiving lithium produced fewer positive results on randomly administered urine drug screens than those receiving placebo.

However, lithium seems to have little efficacy in reducing cocaine use in cocaine-dependent patients with bipolar spectrum disorders.¹⁴ In an open-label study, 10 patients with a history of hypomania or cyclothymia received lithium monotherapy for 12 weeks. Although patients experienced improved mood symptoms and decreased cocaine use, the mean decrease was transitory and not statistically significant. Another factor that may limit lithium’s use for BD-SUD patients is that these patients are more likely to comply with valproate treatment than with lithium.¹⁵

Table 1

Lithium for BD patients with substance use disorders

Anticonvulsants. In a double-blind, placebo-controlled study of 59 alcohol-dependent bipolar I disorder patients, lithium plus divalproex sodium was superior to lithium plus placebo in decreasing number of drinking days and number of drinks per day and in increasing periods of abstinence (Table 2).^16-19 Divalproex sodium was well tolerated and liver function improved in the divalproex sodium group compared with the placebo group, which probably was a benefit of decreased alcohol consumption. In addition, there was a strong association between mood symptoms and alcohol use, which suggests that maximizing mood symptom treatment may decrease alcohol use. However, the divalproex sodium and placebo groups did not differ in measures of mood symptoms, which implies that divalproex sodium might exhibit a positive effect on drinking regardless of its mood-stabilizing properties.

Divalproex sodium also has been used to treat BD comorbid with cocaine dependence. In a small open-label study, 15 patients receiving divalproex sodium plus counseling for mood and substance use disorders were followed for 6 weeks.¹⁷ The 7 patients who completed the trial had significantly more cocaine-abstinent days, spent less money on cocaine, and experienced fewer manic and depressive symptoms. However, divalproex sodium’s effect on cocaine use cannot be determined solely from this study because there was no placebo control group.

Despite its widespread use as a mood stabilizer and potential use in alcohol detoxification, carbamazepine scarcely has been studied in BD-SUD patients. A double-blind, placebo-controlled study of 139 cocaine-dependent patients with BD or other affective disorders found that patients taking carbamazepine for 12 weeks experienced modest reductions in positive urine drug screens and increased time to cocaine use.¹⁸ They also reported less cocaine craving than patients taking placebo, and mood symptoms (mostly depressive) improved.

An open-label study used lamotrigine as adjunctive therapy or monotherapy for 62 cocaine-dependent BD patients followed for 36 weeks.¹⁹ There was some decrease in cocaine craving, money spent on cocaine, and rate of depressive and manic symptoms, but no effect on cocaine use. A placebo-controlled trial is necessary to confirm these modest effects.

No studies have evaluated the potential role of topiramate in treating BD-SUD, despite its FDA-approved indication for alcoholism treatment. Topiramate’s well-known safety and tolerability profile in BD patients make it an interesting option for those with co-occurring alcohol dependence.

Table 2

Studies suggest anticonvulsants may reduce alcohol, cocaine use in BD patients

Atypical antipsychotics. In an open-label study, 16 weeks of quetiapine monotherapy effectively decreased alcohol consumption, alcohol craving, and psychotic and affective symptoms in 28 alcoholics with a variety of psychiatric diagnoses, including BD, schizoaffective disorder, and borderline personality disorder (Table 3).^20-24 However, in a double-blind study of augmentation with quetiapine or placebo for 102 alcohol-dependent BD patients, no significant differences in alcohol use were found between groups.²¹

Quetiapine may be effective for treating BD patients with comorbid cocaine dependence. In an open-label study, 12 weeks of quetiapine augmentation in 17 cocaine-dependent BD patients was associated with decreased cocaine craving and improvement in depressive symptoms.²² In another open-label study, 80 BD patients with comorbid cocaine or amphetamine dependence were randomly assigned to receive quetiapine or risperidone as adjunctive therapy or monotherapy for 20 weeks.²³ Both groups showed significantly decreased drug use and drug craving and improved mood. This study suggests that risperidone also may be an option for BD patients with comorbid cocaine or stimulant dependence.

A 20-week, open-label study of 20 BD-SUD patients found that switching patients from their previous antipsychotic to aripiprazole resulted in less cocaine craving, less alcohol craving, and less money spent on alcohol.²⁴

Olanzapine has not been systematically studied in BD-SUD patients. Some case reports suggest that olanzapine may decrease cocaine craving and use in patients with schizoaffective disorder (bipolar type) and alcohol craving and use in BD patients with comorbid alcohol dependence.²⁵

Table 3

Evidence of efficacy for antipsychotics for BD patients with SUDs

SUD medications. Little evidence guides using medications indicated for treating SUDs—such as naltrexone, acamprosate, and disulfiram—as treatment for BD patients (Table 4).^26-29 In an open-label trial of 34 BD patients with alcohol dependence, naltrexone was well tolerated and associated with decreased alcohol craving and use and modest improvement in manic and depressive symptoms.²⁶

In a double-blind, placebo-controlled study, 50 alcohol-dependent BD patients treated with standard mood-stabilizing therapy and cognitive-behavioral therapy were randomized to receive add-on naltrexone, 50 mg/d, or placebo.²⁷ Patients receiving naltrexone showed decreased alcohol consumption, although no measures were statistically significant. Effect sizes of alcohol use decrease and alcohol craving were moderate to large compared with placebo, which suggests that naltrexone may be effective for treating alcoholism in these patients.

Two other studies evaluated naltrexone and disulfiram in patients with BD or other mood disorders.^28,29 Naltrexone was well tolerated, caused no serious adverse side effects, and was significantly more effective than placebo in decreasing drinking rates and increasing the number of abstinent days.^28,29 Disulfiram was as effective as naltrexone, but the combination of both offered no advantage over use of either drug separately.

There are reports of a new-onset manic episode associated with naltrexone use in a patient with opioid dependence, and a manic episode triggered by naltrexone in a patient with BD with comorbid alcohol dependence.^30,31 At both low and high doses, disulfiram is associated with induction of psychotic mania in alcoholic patients without a personal or family history of BD.^32,33

We found no studies that evaluated treating BD patients who abused other substances, such as cannabis or opiates. We recommend that BD patients with these substance use disorders should be referred to treatment modalities that are condition-specific, such as psychotherapy for cannabis use disorders or methadone or naltrexone treatment for opiate dependence. More severe cases of comorbid SUD probably would benefit from a referral to or consultation with a SUD specialist.

Table 4

Naltrexone and disulfiram for BD patients with alcohol dependence

Related Resource

• Tolliver BK. Bipolar disorder and substance abuse: Overcome the challenges of ‘dual diagnosis’ patients. Current Psychiatry. 2010; 9(8): 32-38.

Drug Brand Names

• Acamprosate • Campral
• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, Equetro, others
• Disulfiram • Antabuse
• Divalproex sodium • Depakote,
• Depakote ER Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Methadone • Dolophine
• Naltrexone • ReVia, Vivitrol
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Valproate • Depacon

Disclosures

Dr. Nery held a temporary work contract as a clinical research physician with Eli Lilly and Company Brazil from May 2009 to November 2009.

Dr. Soares was partly supported by National Institute of Health grants MH 68766, MH 69774, and RR 20571. He receives grant/research support from Bristol-Myers Squibb, Cephalon, GlaxoSmithKline, and Sunovion.

Discuss this article at www.facebook.com/CurrentPsychiatry

Among DSM axis I diagnoses, bipolar disorder (BD) has the highest rates of comorbid substance use disorders (SUDs).^1-3 Approximately 60% of patients with bipolar I disorder have a lifetime diagnosis of an SUD.¹ Excluding tobacco, alcohol is the substance most often abused by BD patients, followed by cannabis, amphetamines, and cocaine.^1-3

BD patients with comorbid SUD usually exhibit more severe clinical presentations and poorer outcomes than their counterparts without SUDs. Compared with patients with BD alone, those with BD and SUD comorbidity (BD-SUD) experience earlier onset of mood symptoms; higher rates of anxiety disorders, suicide attempts, accidents, hospitalizations, and rapid cycling; more depressive episodes; and lower treatment compliance.^4-9

Several treatment options are available for patients with BD-SUD, including psychotherapy modalities, medications primarily used to treat BD, and medications primarily used to treat SUDs. Evidence-based support for these treatments remains limited, and no treatment of choice has emerged. This article reviews evidence on the longer-term treatment of BD-SUD, including general strategies and specific psychosocial and pharmacologic interventions. Short-term treatment strategies, such as pharmacotherapy for detoxification, are outside the scope of this review.

General strategies

The causes of BD-SUD are complex. Evidence suggests that the presence of affective symptoms is associated with an increased risk for substance misuse. This should be kept in mind when treating a patient with BD-SUD because controlling mood symptoms probably will help control substance abuse. However, evidence also shows that SUDs may be independent of mood episodes. Therefore, treating only mood symptoms in the hope that doing so will control substance abuse may not be enough.

Because the negative impact of SUDs on BD outcome is well documented, inform patients that limiting their use of alcohol and/or drugs is vital to control their mood disorder. Efforts to educate, stimulate, and support patients to moderate or stop their alcohol and/or drug use are likely to result in positive changes.¹⁰ Therefore, treatment for BD-SUD should follow, in part, the same recommendations for treatment of SUDs in patients with no comorbid axis I disorders:

• identify the problem (ie, the existence of a comorbid SUD)
• share your concerns with your patient
• offer appropriate and specific treatments, such as detoxification and/or self-help and counseling programs.¹⁰

Because SUDs usually are chronic and relapsing conditions, periods of drug and/ or alcohol use should be expected and not considered a sign of treatment failure. In addition, integrating treatment for both conditions probably is better than managing each separately. Therefore, targeting BD symptoms with mood-stabilizing medications and substance abuse with nonpharmacologic modalities such as drug counseling likely will bring about the best results.

Compared with BD patients without comorbid SUD, BD-SUD patients have a 7-fold increased risk of antidepressantinduced mania.¹¹ Therefore, antidepressants should be prescribed cautiously for patients with BD-SUD.

Integrated psychosocial therapy

BD-SUD patients may benefit from attending self-help programs such as Alcoholics Anonymous and Narcotics Anonymous, provided their mood is stable enough to allow them to participate. Other forms of psychotherapy for BD-SUD patients include standard group drug counseling and integrated group therapy that simultaneously addresses both conditions.

Integrated group therapy is based on the premise that changing maladaptive mood cognitions and behaviors will facilitate recovery from SUDs, and changing maladaptive substance use cognitions and behaviors will facilitate recovery from mood disorders.¹² In a recent randomized controlled trial, 62 BD-SUD patients were blindly assigned to integrated group therapy or standard group drug counseling and followed for 3 months.¹² Pharmacotherapy was prescribed as usual. Substance use decreased for both groups. However, compared with patients in the drug counseling group, those who participated in integrated group therapy spent fewer days using substances in general and alcohol in particular, fewer days using alcohol to intoxication, and had a shorter time from treatment initiation to the first abstinent month. There were no differences between groups in number of weeks in a mood episode.

Pharmacotherapy options

For a table that summarizes the dosages and indications of the medications used to treat BD-SUD that are described below, visit this article at CurrentPsychiatry.com.

Table

Medications used to treat substance use disorders in bipolar disorder patients*

Lithium. Given its well-documented mood stabilizing effect, lithium would seem to be a reasonable option to treat BD-SUD patients, but scant evidence supports its role as an anti-alcohol or anti-drug medication (Table 1).^13,14 Lithium’s efficacy was evaluated in a study of 25 adolescents suffering from mood disorders (mostly BD) and comorbid SUDs (mostly alcohol and cannabis) randomized to receive lithium or placebo for 6 weeks.¹³ Lithium was well tolerated and improved psychiatric symptoms. At week 3, patients receiving lithium produced fewer positive results on randomly administered urine drug screens than those receiving placebo.

However, lithium seems to have little efficacy in reducing cocaine use in cocaine-dependent patients with bipolar spectrum disorders.¹⁴ In an open-label study, 10 patients with a history of hypomania or cyclothymia received lithium monotherapy for 12 weeks. Although patients experienced improved mood symptoms and decreased cocaine use, the mean decrease was transitory and not statistically significant. Another factor that may limit lithium’s use for BD-SUD patients is that these patients are more likely to comply with valproate treatment than with lithium.¹⁵

Table 1

Lithium for BD patients with substance use disorders

Anticonvulsants. In a double-blind, placebo-controlled study of 59 alcohol-dependent bipolar I disorder patients, lithium plus divalproex sodium was superior to lithium plus placebo in decreasing number of drinking days and number of drinks per day and in increasing periods of abstinence (Table 2).^16-19 Divalproex sodium was well tolerated and liver function improved in the divalproex sodium group compared with the placebo group, which probably was a benefit of decreased alcohol consumption. In addition, there was a strong association between mood symptoms and alcohol use, which suggests that maximizing mood symptom treatment may decrease alcohol use. However, the divalproex sodium and placebo groups did not differ in measures of mood symptoms, which implies that divalproex sodium might exhibit a positive effect on drinking regardless of its mood-stabilizing properties.

Divalproex sodium also has been used to treat BD comorbid with cocaine dependence. In a small open-label study, 15 patients receiving divalproex sodium plus counseling for mood and substance use disorders were followed for 6 weeks.¹⁷ The 7 patients who completed the trial had significantly more cocaine-abstinent days, spent less money on cocaine, and experienced fewer manic and depressive symptoms. However, divalproex sodium’s effect on cocaine use cannot be determined solely from this study because there was no placebo control group.

Despite its widespread use as a mood stabilizer and potential use in alcohol detoxification, carbamazepine scarcely has been studied in BD-SUD patients. A double-blind, placebo-controlled study of 139 cocaine-dependent patients with BD or other affective disorders found that patients taking carbamazepine for 12 weeks experienced modest reductions in positive urine drug screens and increased time to cocaine use.¹⁸ They also reported less cocaine craving than patients taking placebo, and mood symptoms (mostly depressive) improved.

An open-label study used lamotrigine as adjunctive therapy or monotherapy for 62 cocaine-dependent BD patients followed for 36 weeks.¹⁹ There was some decrease in cocaine craving, money spent on cocaine, and rate of depressive and manic symptoms, but no effect on cocaine use. A placebo-controlled trial is necessary to confirm these modest effects.

No studies have evaluated the potential role of topiramate in treating BD-SUD, despite its FDA-approved indication for alcoholism treatment. Topiramate’s well-known safety and tolerability profile in BD patients make it an interesting option for those with co-occurring alcohol dependence.

Table 2

Studies suggest anticonvulsants may reduce alcohol, cocaine use in BD patients

Atypical antipsychotics. In an open-label study, 16 weeks of quetiapine monotherapy effectively decreased alcohol consumption, alcohol craving, and psychotic and affective symptoms in 28 alcoholics with a variety of psychiatric diagnoses, including BD, schizoaffective disorder, and borderline personality disorder (Table 3).^20-24 However, in a double-blind study of augmentation with quetiapine or placebo for 102 alcohol-dependent BD patients, no significant differences in alcohol use were found between groups.²¹

Quetiapine may be effective for treating BD patients with comorbid cocaine dependence. In an open-label study, 12 weeks of quetiapine augmentation in 17 cocaine-dependent BD patients was associated with decreased cocaine craving and improvement in depressive symptoms.²² In another open-label study, 80 BD patients with comorbid cocaine or amphetamine dependence were randomly assigned to receive quetiapine or risperidone as adjunctive therapy or monotherapy for 20 weeks.²³ Both groups showed significantly decreased drug use and drug craving and improved mood. This study suggests that risperidone also may be an option for BD patients with comorbid cocaine or stimulant dependence.

A 20-week, open-label study of 20 BD-SUD patients found that switching patients from their previous antipsychotic to aripiprazole resulted in less cocaine craving, less alcohol craving, and less money spent on alcohol.²⁴

Olanzapine has not been systematically studied in BD-SUD patients. Some case reports suggest that olanzapine may decrease cocaine craving and use in patients with schizoaffective disorder (bipolar type) and alcohol craving and use in BD patients with comorbid alcohol dependence.²⁵

Table 3

Evidence of efficacy for antipsychotics for BD patients with SUDs

SUD medications. Little evidence guides using medications indicated for treating SUDs—such as naltrexone, acamprosate, and disulfiram—as treatment for BD patients (Table 4).^26-29 In an open-label trial of 34 BD patients with alcohol dependence, naltrexone was well tolerated and associated with decreased alcohol craving and use and modest improvement in manic and depressive symptoms.²⁶

In a double-blind, placebo-controlled study, 50 alcohol-dependent BD patients treated with standard mood-stabilizing therapy and cognitive-behavioral therapy were randomized to receive add-on naltrexone, 50 mg/d, or placebo.²⁷ Patients receiving naltrexone showed decreased alcohol consumption, although no measures were statistically significant. Effect sizes of alcohol use decrease and alcohol craving were moderate to large compared with placebo, which suggests that naltrexone may be effective for treating alcoholism in these patients.

Two other studies evaluated naltrexone and disulfiram in patients with BD or other mood disorders.^28,29 Naltrexone was well tolerated, caused no serious adverse side effects, and was significantly more effective than placebo in decreasing drinking rates and increasing the number of abstinent days.^28,29 Disulfiram was as effective as naltrexone, but the combination of both offered no advantage over use of either drug separately.

There are reports of a new-onset manic episode associated with naltrexone use in a patient with opioid dependence, and a manic episode triggered by naltrexone in a patient with BD with comorbid alcohol dependence.^30,31 At both low and high doses, disulfiram is associated with induction of psychotic mania in alcoholic patients without a personal or family history of BD.^32,33

We found no studies that evaluated treating BD patients who abused other substances, such as cannabis or opiates. We recommend that BD patients with these substance use disorders should be referred to treatment modalities that are condition-specific, such as psychotherapy for cannabis use disorders or methadone or naltrexone treatment for opiate dependence. More severe cases of comorbid SUD probably would benefit from a referral to or consultation with a SUD specialist.

Table 4

Naltrexone and disulfiram for BD patients with alcohol dependence

Related Resource

• Tolliver BK. Bipolar disorder and substance abuse: Overcome the challenges of ‘dual diagnosis’ patients. Current Psychiatry. 2010; 9(8): 32-38.

Drug Brand Names

• Acamprosate • Campral
• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, Equetro, others
• Disulfiram • Antabuse
• Divalproex sodium • Depakote,
• Depakote ER Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Methadone • Dolophine
• Naltrexone • ReVia, Vivitrol
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Valproate • Depacon

Disclosures

Dr. Nery held a temporary work contract as a clinical research physician with Eli Lilly and Company Brazil from May 2009 to November 2009.

Dr. Soares was partly supported by National Institute of Health grants MH 68766, MH 69774, and RR 20571. He receives grant/research support from Bristol-Myers Squibb, Cephalon, GlaxoSmithKline, and Sunovion.

Discuss this article at www.facebook.com/CurrentPsychiatry

Among DSM axis I diagnoses, bipolar disorder (BD) has the highest rates of comorbid substance use disorders (SUDs).^1-3 Approximately 60% of patients with bipolar I disorder have a lifetime diagnosis of an SUD.¹ Excluding tobacco, alcohol is the substance most often abused by BD patients, followed by cannabis, amphetamines, and cocaine.^1-3

BD patients with comorbid SUD usually exhibit more severe clinical presentations and poorer outcomes than their counterparts without SUDs. Compared with patients with BD alone, those with BD and SUD comorbidity (BD-SUD) experience earlier onset of mood symptoms; higher rates of anxiety disorders, suicide attempts, accidents, hospitalizations, and rapid cycling; more depressive episodes; and lower treatment compliance.^4-9

Several treatment options are available for patients with BD-SUD, including psychotherapy modalities, medications primarily used to treat BD, and medications primarily used to treat SUDs. Evidence-based support for these treatments remains limited, and no treatment of choice has emerged. This article reviews evidence on the longer-term treatment of BD-SUD, including general strategies and specific psychosocial and pharmacologic interventions. Short-term treatment strategies, such as pharmacotherapy for detoxification, are outside the scope of this review.

General strategies

The causes of BD-SUD are complex. Evidence suggests that the presence of affective symptoms is associated with an increased risk for substance misuse. This should be kept in mind when treating a patient with BD-SUD because controlling mood symptoms probably will help control substance abuse. However, evidence also shows that SUDs may be independent of mood episodes. Therefore, treating only mood symptoms in the hope that doing so will control substance abuse may not be enough.

Because the negative impact of SUDs on BD outcome is well documented, inform patients that limiting their use of alcohol and/or drugs is vital to control their mood disorder. Efforts to educate, stimulate, and support patients to moderate or stop their alcohol and/or drug use are likely to result in positive changes.¹⁰ Therefore, treatment for BD-SUD should follow, in part, the same recommendations for treatment of SUDs in patients with no comorbid axis I disorders:

• identify the problem (ie, the existence of a comorbid SUD)
• share your concerns with your patient
• offer appropriate and specific treatments, such as detoxification and/or self-help and counseling programs.¹⁰

Because SUDs usually are chronic and relapsing conditions, periods of drug and/ or alcohol use should be expected and not considered a sign of treatment failure. In addition, integrating treatment for both conditions probably is better than managing each separately. Therefore, targeting BD symptoms with mood-stabilizing medications and substance abuse with nonpharmacologic modalities such as drug counseling likely will bring about the best results.

Compared with BD patients without comorbid SUD, BD-SUD patients have a 7-fold increased risk of antidepressantinduced mania.¹¹ Therefore, antidepressants should be prescribed cautiously for patients with BD-SUD.

Integrated psychosocial therapy

BD-SUD patients may benefit from attending self-help programs such as Alcoholics Anonymous and Narcotics Anonymous, provided their mood is stable enough to allow them to participate. Other forms of psychotherapy for BD-SUD patients include standard group drug counseling and integrated group therapy that simultaneously addresses both conditions.

Integrated group therapy is based on the premise that changing maladaptive mood cognitions and behaviors will facilitate recovery from SUDs, and changing maladaptive substance use cognitions and behaviors will facilitate recovery from mood disorders.¹² In a recent randomized controlled trial, 62 BD-SUD patients were blindly assigned to integrated group therapy or standard group drug counseling and followed for 3 months.¹² Pharmacotherapy was prescribed as usual. Substance use decreased for both groups. However, compared with patients in the drug counseling group, those who participated in integrated group therapy spent fewer days using substances in general and alcohol in particular, fewer days using alcohol to intoxication, and had a shorter time from treatment initiation to the first abstinent month. There were no differences between groups in number of weeks in a mood episode.

Pharmacotherapy options

For a table that summarizes the dosages and indications of the medications used to treat BD-SUD that are described below, visit this article at CurrentPsychiatry.com.

Table

Medications used to treat substance use disorders in bipolar disorder patients*

Lithium. Given its well-documented mood stabilizing effect, lithium would seem to be a reasonable option to treat BD-SUD patients, but scant evidence supports its role as an anti-alcohol or anti-drug medication (Table 1).^13,14 Lithium’s efficacy was evaluated in a study of 25 adolescents suffering from mood disorders (mostly BD) and comorbid SUDs (mostly alcohol and cannabis) randomized to receive lithium or placebo for 6 weeks.¹³ Lithium was well tolerated and improved psychiatric symptoms. At week 3, patients receiving lithium produced fewer positive results on randomly administered urine drug screens than those receiving placebo.

However, lithium seems to have little efficacy in reducing cocaine use in cocaine-dependent patients with bipolar spectrum disorders.¹⁴ In an open-label study, 10 patients with a history of hypomania or cyclothymia received lithium monotherapy for 12 weeks. Although patients experienced improved mood symptoms and decreased cocaine use, the mean decrease was transitory and not statistically significant. Another factor that may limit lithium’s use for BD-SUD patients is that these patients are more likely to comply with valproate treatment than with lithium.¹⁵

Table 1

Lithium for BD patients with substance use disorders

Anticonvulsants. In a double-blind, placebo-controlled study of 59 alcohol-dependent bipolar I disorder patients, lithium plus divalproex sodium was superior to lithium plus placebo in decreasing number of drinking days and number of drinks per day and in increasing periods of abstinence (Table 2).^16-19 Divalproex sodium was well tolerated and liver function improved in the divalproex sodium group compared with the placebo group, which probably was a benefit of decreased alcohol consumption. In addition, there was a strong association between mood symptoms and alcohol use, which suggests that maximizing mood symptom treatment may decrease alcohol use. However, the divalproex sodium and placebo groups did not differ in measures of mood symptoms, which implies that divalproex sodium might exhibit a positive effect on drinking regardless of its mood-stabilizing properties.

Divalproex sodium also has been used to treat BD comorbid with cocaine dependence. In a small open-label study, 15 patients receiving divalproex sodium plus counseling for mood and substance use disorders were followed for 6 weeks.¹⁷ The 7 patients who completed the trial had significantly more cocaine-abstinent days, spent less money on cocaine, and experienced fewer manic and depressive symptoms. However, divalproex sodium’s effect on cocaine use cannot be determined solely from this study because there was no placebo control group.

Despite its widespread use as a mood stabilizer and potential use in alcohol detoxification, carbamazepine scarcely has been studied in BD-SUD patients. A double-blind, placebo-controlled study of 139 cocaine-dependent patients with BD or other affective disorders found that patients taking carbamazepine for 12 weeks experienced modest reductions in positive urine drug screens and increased time to cocaine use.¹⁸ They also reported less cocaine craving than patients taking placebo, and mood symptoms (mostly depressive) improved.

An open-label study used lamotrigine as adjunctive therapy or monotherapy for 62 cocaine-dependent BD patients followed for 36 weeks.¹⁹ There was some decrease in cocaine craving, money spent on cocaine, and rate of depressive and manic symptoms, but no effect on cocaine use. A placebo-controlled trial is necessary to confirm these modest effects.

No studies have evaluated the potential role of topiramate in treating BD-SUD, despite its FDA-approved indication for alcoholism treatment. Topiramate’s well-known safety and tolerability profile in BD patients make it an interesting option for those with co-occurring alcohol dependence.

Table 2

Studies suggest anticonvulsants may reduce alcohol, cocaine use in BD patients

Atypical antipsychotics. In an open-label study, 16 weeks of quetiapine monotherapy effectively decreased alcohol consumption, alcohol craving, and psychotic and affective symptoms in 28 alcoholics with a variety of psychiatric diagnoses, including BD, schizoaffective disorder, and borderline personality disorder (Table 3).^20-24 However, in a double-blind study of augmentation with quetiapine or placebo for 102 alcohol-dependent BD patients, no significant differences in alcohol use were found between groups.²¹

Quetiapine may be effective for treating BD patients with comorbid cocaine dependence. In an open-label study, 12 weeks of quetiapine augmentation in 17 cocaine-dependent BD patients was associated with decreased cocaine craving and improvement in depressive symptoms.²² In another open-label study, 80 BD patients with comorbid cocaine or amphetamine dependence were randomly assigned to receive quetiapine or risperidone as adjunctive therapy or monotherapy for 20 weeks.²³ Both groups showed significantly decreased drug use and drug craving and improved mood. This study suggests that risperidone also may be an option for BD patients with comorbid cocaine or stimulant dependence.

A 20-week, open-label study of 20 BD-SUD patients found that switching patients from their previous antipsychotic to aripiprazole resulted in less cocaine craving, less alcohol craving, and less money spent on alcohol.²⁴

Olanzapine has not been systematically studied in BD-SUD patients. Some case reports suggest that olanzapine may decrease cocaine craving and use in patients with schizoaffective disorder (bipolar type) and alcohol craving and use in BD patients with comorbid alcohol dependence.²⁵

Table 3

Evidence of efficacy for antipsychotics for BD patients with SUDs

SUD medications. Little evidence guides using medications indicated for treating SUDs—such as naltrexone, acamprosate, and disulfiram—as treatment for BD patients (Table 4).^26-29 In an open-label trial of 34 BD patients with alcohol dependence, naltrexone was well tolerated and associated with decreased alcohol craving and use and modest improvement in manic and depressive symptoms.²⁶

In a double-blind, placebo-controlled study, 50 alcohol-dependent BD patients treated with standard mood-stabilizing therapy and cognitive-behavioral therapy were randomized to receive add-on naltrexone, 50 mg/d, or placebo.²⁷ Patients receiving naltrexone showed decreased alcohol consumption, although no measures were statistically significant. Effect sizes of alcohol use decrease and alcohol craving were moderate to large compared with placebo, which suggests that naltrexone may be effective for treating alcoholism in these patients.

Two other studies evaluated naltrexone and disulfiram in patients with BD or other mood disorders.^28,29 Naltrexone was well tolerated, caused no serious adverse side effects, and was significantly more effective than placebo in decreasing drinking rates and increasing the number of abstinent days.^28,29 Disulfiram was as effective as naltrexone, but the combination of both offered no advantage over use of either drug separately.

There are reports of a new-onset manic episode associated with naltrexone use in a patient with opioid dependence, and a manic episode triggered by naltrexone in a patient with BD with comorbid alcohol dependence.^30,31 At both low and high doses, disulfiram is associated with induction of psychotic mania in alcoholic patients without a personal or family history of BD.^32,33

We found no studies that evaluated treating BD patients who abused other substances, such as cannabis or opiates. We recommend that BD patients with these substance use disorders should be referred to treatment modalities that are condition-specific, such as psychotherapy for cannabis use disorders or methadone or naltrexone treatment for opiate dependence. More severe cases of comorbid SUD probably would benefit from a referral to or consultation with a SUD specialist.

Table 4

Naltrexone and disulfiram for BD patients with alcohol dependence

Related Resource

• Tolliver BK. Bipolar disorder and substance abuse: Overcome the challenges of ‘dual diagnosis’ patients. Current Psychiatry. 2010; 9(8): 32-38.

Drug Brand Names

• Acamprosate • Campral
• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, Equetro, others
• Disulfiram • Antabuse
• Divalproex sodium • Depakote,
• Depakote ER Lamotrigine • Lamictal
• Lithium • Eskalith, Lithobid
• Methadone • Dolophine
• Naltrexone • ReVia, Vivitrol
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Valproate • Depacon

Disclosures

Dr. Nery held a temporary work contract as a clinical research physician with Eli Lilly and Company Brazil from May 2009 to November 2009.

Dr. Soares was partly supported by National Institute of Health grants MH 68766, MH 69774, and RR 20571. He receives grant/research support from Bristol-Myers Squibb, Cephalon, GlaxoSmithKline, and Sunovion.