Anxiety in bipolar disorder: data on cotreatment are lacking, but anecdotally, antimanics and antipsychotics have proven useful.

Until recently, discussions of bipolar comorbidity centered on substance abuse. It is now recognized, however, that anxiety is at least as prevalent in bipolar patients and has a substantial negative effect on the course of the mood disorder.

Analysis of data from the first 500 participants in the multicenter Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) found that more than half had a lifetime history of an anxiety disorder, that current prevalence was nearly one-third, and that rates were higher for bipolar I than II. Figures reported elsewhere have been even higher. The presence of comorbid anxiety was associated with lower rates of recovery, reduced functioning, and an increased risk of suicide attempts (Am. J. Psychiatry 2004;161:2222-9).

Various studies have identified panic disorder, generalized anxiety disorder (GAD), and social phobia as the most common comorbidities, although obsessive-compulsive disorder (OCD) and posttraumatic stress disorder (PTSD) are seen as well.

The nature of the bipolar/anxiety association is uncertain, said Naomi Simon, M.D., associate director of the anxiety disorders program at Massachusetts General Hospital, Boston, and lead author of the STEP-BD report. "Whether anxiety comorbidity is a bipolar subtype and a marker of poor outcome or they are two distinct conditions is not clear," she said.

In either case, the reasons for identifying anxiety are compelling. "Bipolar patients should be screened for comorbidity because it's such a powerful risk factor for suicide and poor outcomes. Patients need to be monitored more closely in the presence of anxiety," Dr. Simon pointed out.

Although worthy for its own sake, effective anxiety treatment alone hasn't been shown to ameliorate bipolar disorder, nor is there evidence of efficacy, in this context, for any particular drug. "We are so profoundly behind the curve in understanding and identifying anxiety in bipolar disorder that we don't have data to talk about cotreatment," said Mark S. Bauer, M.D., professor of psychiatry and human behavior at Brown University in Providence, R.I.

Without such data, the most reasonable treatment options would seem to be drugs that are used for anxiety generally. This decision, however, is complicated by the fact that the two medication classes validated for this indication--benzodiazepines and antidepressants--carry substantial risks in the context of bipolar disorder: abuse with the first and manic switching with the second.

Bipolar treatment itself may ameliorate anxiety. But in light of the dangers associated with the primary disorder, this must have a second level of priority in drug selection after primary mood stabilization. "There is probably a difference in the anxiolytic effects of mood stabilizers, but good long-term, well-tolerated mood stabilization is truly the first thing," Dr. Simon noted. Atypical antipsychotics as a class appear to reduce anxiety, and lithium does not, "but there's such good evidence for effective mood stabilization and protection against suicide, I might choose it anyway as the first-line mood-stabilizing agent."

In some cases, anxiety resolves when mood stabilization is optimized. If not, adjunctive medication is needed.

On the use of antidepressants, there is no consensus. "I'd say that people [are becoming] more aggressive in treating depression with antidepressants, and the same would be true of anxiety disorders," Dr. Bauer said.

With "a good antimanic agent on board ... there's less worry about switching and more worry about chronic, undertreated anxiety. I wouldn't shy away from [an SSRI] or a tricyclic for panic, or fluvoxamine or clomipramine for OCD," he said. Its long half-life makes fluoxetine a less attractive choice among SSRIs, whereas an advantage with tricyclics is the ability to monitor serum levels. Dr. Bauer prefers the secondary amines nortriptyline or desipramine, or doxepin or imipramine for patients who need help sleeping.

To Patricia Suppes, M.D., of the department of psychiatry at the University of Texas, Dallas, antidepressants are best avoided when possible. The newly revised Texas Implementation of Medication Algorithms for bipolar disorder, for which Dr. Suppes was the lead author, does not address comorbidity per se, but it places antidepressants below antimanics and antipsychotics in the hierarchy of treatment for depressive episodes (J. Clin. Psychiatry 2005; 66:870-86).

When an antidepressant (preferably an SSRI, a selective norepinephrine-serotonin reuptake inhibitor, or bupropion) is prescribed for anxiety, it is necessary to "optimize" mood stabilization, Dr. Suppes said: "If you're using one antimanic, this may mean going to a higher dose, or adding another. In the most difficult cases, you'll probably need to go to clozapine; it's the strongest antimanic we have."

Benzodiazepines have the most robust evidence of efficacy for anxiety disorders in general, but they require caution for some individuals in light of the heightened risk of substance abuse in bipolar patients, Dr. Simon said. "If the patient had a history of substance abuse or dependence, or a lot of impulsivity, I'd be particularly concerned," she said. "Without such a history, benzodiazepines might be the most effective option."

Sedation in patients who also take other medications can be a limiting factor, she said.

Dr. Simon tends to use clonazepam, which is potent and long acting, and which carries less chance of interdose rebound than do shorter-acting agents like alprazolam.

Atypical antipsychotics have neither mood-destabilizing nor abuse liability. Anecdotally, olanzapine (Zyprexa) and quetiapine (Seroquel) may be used more than others for acute anxiety, but many atypicals appear to be anxiolytic. This is not unlike the anxiolytic effects of the typical antipsychotics, the use of which was limited by long-term side effects, Dr. Simon said. Side effects, particularly metabolic changes, are often the limiting factor with atypical agents.

Some anticonvulsants may be useful as well. "I've seen cases on both sides of the fence: The anticonvulsant gets good results for anxiety, or it doesn't touch it," Dr. Suppes said. Gabapentin does not worsen the primary disorder or abuse, and the side effects are less troublesome than those of atypical antipsychotics, but its anxiolytic effect is modest, she pointed out.

Buspirone has similar advantages and drawbacks, and may deserve a trial for the cognitive symptoms of PTSD or GAD, Dr. Bauer said. Some data also suggest that the antihypertensive drug prazosin may be helpful for PTSD nightmares, as clonidine may be for hypervigilance.

The difficulties surrounding medication make psychotherapy a particularly attractive option, Dr. Simon said. Cognitive-behavioral therapy has been shown to be effective for panic, GAD, and OCD. "It requires a good deal from the patient, and those who are very ill or disorganized may not be able to do it," although the therapy is frequently accessible to moderately depressed or euthymic patients, she said.

BY CARL SHERMAN

Contributing Writer