Perspectives on Psychopharmacology: Bipolar Disorder and Evidence-Based Psychiatry

By Neil Sandson, MD

[Spring 2006; Vol. 32, No. 3; Pg 6, 10]

Bipolar disorder is one of the most challenging illnesses that psychiatrists encounter in patients.  Prior to 2000, only lithium, valproate, and chlorpromazine were  FDA approved for the treatment of bipolar disorder.  Since then, eight more agents (aripiprazole, carbamazepine, lamotrigine, olanzapine, olanzpine/fluoxetine combination, quetiapine, risperidone, and ziprasidone) were approved.  While there are patients who are responding better to pharmacotherapy than in the past, our more expanded armamentarium makes adherence to evidence-based practice more important than  before, to prevent us from straying into uncharted and potentially dangerous territory.

One trend I have observed over 10 years, is a tendency for pharmacotherapy to be guided by (seeming) avoidance of risk and a desire for “newer and better” drugs rather than using a drug with superior efficacy.  This accounted for the popularity of gabapentin in the treatment of bipolar patients.  Unfortunately, variations of this theme persist.  One example is the underuse of lithium, despite research showing that it has the best outcome in the treatment of all phases of bipolar illness, not to mention an unmatched antisuicidal effect in bipolar patients.  Another example is the use of drugs like oxcarbazepine and topiramate as first and second-line drugs in bipolar pharmacotherapy.  There is no evidence to support this practice, and yet these agents are used far more frequently than either their demonstrated efficacy or their ancillary benefits (avoidance of bone marrow suppression and drug interactions with oxcarbamazepine and weight loss with topiramate) would justify.  This is not to suggest that these agents can never be used for a bipolar patient.  Rather, they should be used with a reluctance that reflects their highly conjectural status as helpful drugs.

Another difficult issue in the treatment of bipolar disorder is the problem of bipolar depression and the appropriate role of antidepressant medications in the depressed and maintenance phases of the illness.  There are many, like myself, who are particularly averse to using antidepressants as a standard intervention.  Of concern is the propensity of these drugs to produce switches into the manic phase, which is disruptive to patients’ lives and may lead to further deterioration in the course of their illness.  It is also true that  bipolar depression is responsible for more morbidity and mortality.  Thus, many reasonable clinicians maintain that effective treatment and prevention of bipolar depression is a higher clinical priority than the prevention of manic episodes.  My perspective on this issue is that combination “mood stabilizer” therapy (liberally including lithium, various anticonvulsants, and antipsychotic agents) addresses both clinical concerns.  No study has definitively demonstrated that antidepressants are superior in treating bipolar depression than combination mood stabilizer therapies, and some studies suggest that they are comparable in efficacy.  Given that both approaches seem equally effective for bipolar depression, I have pursued combination mood stabilizer therapies as my first treatment option for bipolar depression; this avoids the risk of using an antidepressant that may cause a switch to mania. 

Some point out that the newer antidepressants, such as SSRIs and bupropion, have manic switch rates comparable to placebo in double-blind, placebo-controlled studies.  However, I have observed that switch rates seem higher in clinical populations as compared to patients able to meet the stringent exclusion criteria characteristic in rigorous studies.  Thus, the dangers posed by antidepressants may well be greater than more reliable studies would suggest.

A caveat to the above is to respect prior treatment responses for specific patients.  Altshuler et al. recently published a study that suggests that if a patient maintained good stability for at least two months while on an antidepressant, then removing that antidepressant is more likely to produce de-compensation.  In view of this finding, and the broader philosophical principle of “never change a winning game, always change a losing game”, I don’t remove the antidepressants from stable patients’ regimens in a dogmatic or cavalier manner.  I’ve even been known to voluntarily prescribe them for bipolar depression from time to time.

One final issue with regard to evidence-based treatment of bipolar disorder is the neglect of non-pharmacologic interventions.  Over the past five years, research has proven that cognitive-behavioral therapy and individual and family psycho-education are efficacious in preventing relapse across populations.  While most clinicians are generally aware that combined approaches are “better” than either medication management or psychotherapies alone, there is also a pervasive sense that medication management by itself is “good enough” for many or even most bipolar patients, and that only our most treatment resistant, disruptive, and/or noncompliant patients really require both forms of treatment. Regrettably, this devaluing of psychotherapy is a generalized phenomenon that is not restricted to the treatment of bipolar disorder.  It has its roots in the economic and philosophical changes that have altered our professional mode of practice over the last 15+ years, and systemic problems continue to hinder even the most well-intentioned clinicians from implementing psychotherapy.  However, help may be on the way. 

The STEP-BD (Systematic Treatment Enhancement Program for Bipolar Disorder) study is a large, multicenter, NIMH study designed to evaluate the efficacy of various medications and psychotherapeutic/psychosocial interventions in the treatment of bipolar disorder.  While the publication of the phase one results of the CATIE (Clinical Antipsychotic Trials in Intervention Effectiveness) study has not instantly transformed the treatment of schizophrenia, it has represented a decisive step toward evidence-based practice.  In the February 2006 issue of the American Journal of Psychiatry, some initial analyses from the STEP-BD study were published.  These dealt mainly with specific factors that were associated with illness recurrences, but in the near-future, more specific comparisons of therapeutic interventions will be published.  These results will hopefully exert a significant influence on current practice.