By Stephen V. Sobel, MD
Bipolar disorder causes havoc in patients’ lives. Even in the best of circumstances, successful treatment is challenging. Treatment targets constantly shift; patients are frequently nonadherent; and comorbidity is the rule, not the exception. Diagnosis of bipolar disorder is often difficult. Comorbidities need to be identified and addressed if treatment is to be effective.
The importance of an accurate diagnosis
With apologies to Charles Dickens, bipolar disorder is often experienced as the “best of times and the worst of times.” This polarity often causes bipolar disorder to be undiagnosed, overdiagnosed, or misdiagnosed. Bipolar disorder is associated with a significantly elevated risk of suicide. Moreover, bipolar patients often use highly lethal means for suicide.1 Contributing factors include early age at disease onset, the high number of depressive episodes, comorbid alcohol abuse, a history of antidepressant-induced mania, and traits of hostility and impulsivity.
Bipolar I disorder, with episodes of full-blown mania, is usually easier to diagnose than bipolar II disorder, with episodes of subtler hypomania. Recognizing that the primary mood state may be irritability rather than euphoria increases the likelihood of diagnosis as does the recognition that symptoms often last fewer than the 4 days required for diagnosis by DSM-IV.2 Focusing more on overactivity than mood change further improves diagnostic accuracy, and the use of structured questionnaires is helpful.
Given the greater frequency of depression than manic episodes in bipolar disorder, what clues indicate bipolar disorder rather than unipolar depression? The Table lists factors that may help identify unipolar depression.
A moving target needs moving treatment
Effective personalized treatment recognizes bipolar disorder as a biopsychosocial disorder, but mood-stabilizing medications are the backbone of treatment. These medications fall into 3 categories: lithium, antikindling/antiepileptic agents, and second-generation antipsychotics. The mechanisms of actions by which these medications work are numer-ous and include increasing levels of serotonin, γ-aminobutyric acid, and brain-derived neurotrophic factor (BDNF) and decreasing glutamate levels; modifying dopamine pathways; stabilizing neuronal membranes; decreasing sodium channels; decreasing depolarization; decreasing apoptosis; and increasing neural cell growth/arborization.
Double-blind placebo-controlled studies of the medications—lithium, divalproex, carbamazepine, and atypical antipsychotics—used to treat symptoms of acute mania have demonstrated a response rate of approximately 50% to these drugs. Response was defined as a 50% decrease in symptoms using the Young Mania Rating Scale (YMRS) with onset of response within a few days.
An increasingly intriguing aspect of treatment with lithium and atypical antipsychotics involves their effect on BDNF. In a study of 10 manic patients treated with lithium for 28 days, most (87%) showed an increase in BDNF level (ie, from 406 pg/mL to 511 pg/mL).
Factors that suggest bipolar depression rather than unipolar depression
In a typical 3-week study of acute mania, approximately half of the benefit was seen by day 4. A 3-week, double-blind, inpatient study of olanzapine and risperidone in 274 patients with acute mania found that of 117 patients who had a less than 50% decrease in the YMRS score at 1 week, only 39% responded and 19% had symptom remission at end point. Of 40 patients with a less than 25% decrease in the YMRS score at 1 week, only 25% responded and only 5% had symptom remission at 3 weeks. Of 157 patients who had at least a 50% decrease in the YMRS score at week 1, 84% responded and 64% had symptom remission at 3 weeks.4 Clinically, a medication change should be considered for patients who do not demonstrate substantial benefit by week 1.
A meta-analysis comprising 16,000 patients who had acute mania found that the most effective agents were haloperidol, risperidone, and olanzapine. The least effective were gabapentin, lamotrigine, and topiramate.5
A combination of medications—typically lithium or an antiepileptic with an atypical antipsychotic—is often necessary to successfully treat acute mania. A meta-analysis found the response rate increased from 42% to 62% when an antipsychotic was added.6
Bipolar depression has proved to be more resistant to medication treatment than mania. The same medications are used, with lamotrigine for maintenance treatment. The FDA has approved Seroquel, Seroquel XR, and Symbyax (the combination of olanzapine and fluoxetine), for the acute treatment of bipolar depression. Studies of acute bipolar depression have typically lasted 8 weeks. Approximately half of the benefit oc-curs by week 2, with statistical separation from placebo between weeks 1 and 3.7-9
The best treatment is prevention
Patients who have bipolar disorder almost always require lifelong maintenance treatment, frequently with 2 medications: one to prevent the upside (ie, hypomania/mania), and another to prevent the downside (ie, depression).
Findings from a registration trial showed that lamotrigine more effectively prevented depressions than lithium but lithium prevented mania/hypomania more effectively than lamotrigine.10
Another study added placebo or lamotrigine to lithium treatment for 124 patients. The median time to relapse/recurrence was 3.5 months for those taking lithium monotherapy but 10 months for those who received combination treatment.11
The effectiveness of a combination maintenance regimen was also seen in a study of 628 patients with bipolar I disorder treated for 2 years: 65% of those taking lithium or divalproex alone experienced a recurrence compared with 21% who received quetiapine added to lithium or divalproex.12 However, combination treatment may result in more adverse effects and increased risk of drug-drug interactions.
The best mood stabilizer
The best mood stabilizer for a patient is the one he or she will take. No matter how effective a medication is, it will not relieve symptoms if it is not being taken. The key to effective personalized treatment of bipolar disorder is a good patient-physician connection in which the patient is part of the treatment decision-making process.
Psychotherapy is an integral part of the effective treatment of bipolar disorder, not just an augmentation strategy. Psychotherapies that are helpful include cognitive-behavioral therapy and social rhythm therapy.13 Psychotherapy can focus on several areas, such as education, comorbidities, medication adherence, and interpersonal relationships. In addition, therapy can challenge the automatic, distorted, and dysfunctional thoughts and help the patient maintain social rhythms (eg, consistent sleep). The involvement of family members in treatment enhances success.
Patients may stop taking their medications because the adverse effects become intolerable; they may miss what they perceive as their more satisfying and productive hypomania; and they might believe that a period without symptoms means that they are cured and no longer need medications. One study of 3640 patients with bipolar disorder who made 48,000 physician visits found that 24% of patients were nonadherent (defined as missing at least 25% of doses) 20% of the time. Factors associated with nonadherence included rapid cycling, suicide attempts, earlier onset of illness, anxiety, and alcohol abuse.14
Patients who have bipolar II disorder spend far more time depressed than hypomanic. Lithium appears to be less effective than antikindling agents for rapid cycling as well as for mixed bipolar disorder states.15
Maintenance treatment is necessary for patients with acute mania or acute depression; therefore, choose medications that are more tolerable to the patient to facilitate long-term adherence. Recognize that medications may need to be adjusted or changed—in the acute phase of illness, rapid efficacy is often the priority, while medication adherence is the priority during the maintenance phase.
Other factors to consider when choosing the best medication for a particular patient include:
• A history of treatment response
• A family history of response
• Adverse effects of a particular drug
• Drug interactions
The use of antidepressants in bipolar disorder is controversial because they may induce rapid cycling, especially in patients with episodes of rapid cycling.16 In a study by Altshuler and colleagues,17 patients who had breakthrough depression despite treatment with a mood stabilizer were treated with antidepressants for at least 60 days. Patients who had symptom remission for 6 weeks were followed up for 1 year: 36% of patients who continued antidepressants for longer than 6 months relapsed versus 70% who discontinued antidepressants before 6 months.
A randomized discontinuation study with antidepressants found no statistically significant symptomatic benefit in the long-term treatment of bipolar disorder.18 Trends toward mild benefits, however, were found in patients who continued antidepressants. This study also found, similar to studies of tricyclic antidepressants, that rapid-cycling patients had worsened outcomes with continuation of modern antidepressants, including SSRIs and SNRIs.
An NIMH study of 159 patients who had breakthrough depression despite receiving a mood stabilizer were treated with sertraline (mean dosage, 192 mg/d), bupropion (mean dosage, 286 mg/d), or venlafaxine (mean dosage, 195 mg/d) for 10 weeks with a 1-year follow-up.19 At the end of 1 year, only 16% of the patients had continued remission while more than 55% had switched to mania/hypomania. The worst results were seen with venlafaxine and the best with bupropion.
In a study by Sachs and colleagues,20 patients who had breakthrough depression despite being treated with mood stabilizers were randomized to paroxetine (mean dosage, 30 mg/d), bupropion (mean dosage, 300 mg/d), or placebo. No significant differences on any effectiveness or safety outcome, including remission rates or affective switch frequency, were found.
Overall, these studies indicate that the role of antidepressants is limited and that, in fact, a trial of a mood stabilizer cannot be considered to have failed unless the failure occurs in the absence of an antidepressant. A meta-analysis of 18 studies with 4105 patients found that combination treatment including a mood stabilizer and an antidepressant was not statistically superior to monotherapy.21
When symptoms persist
Establish the context of each appointment by focusing on changes in occupational, social, family, and health status. Evaluate medication regimens, with a focus on effectiveness for carefully chosen target symptoms and adherence to treatment, as well as medication tolerability and patient attitudes. Be alert to the emergence of early symptoms of mood change, and adjust medications if necessary. Remember that treatment modalities often need to change over time.
Mood stabilizers should be optimized with combination therapy for sustained remission. Antidepressants may worsen the disease course, and a true trial of a mood stabilizer can-not occur within the setting of antidepressants. If symptoms persist, ask: Is the patient taking anything that is making symptoms worse, eg, drugs, alcohol, or antidepressants? Is the patient taking the medications? Is treatment adequate? Is another condition (including subclinical hypothyroidism) interfering with treatment? Is psychotherapy being ignored?