NOTE: To view the article with Web enhancements, go to:
http://www.medscape.com/viewarticle/487753

Roy Perlis, MD

Medscape Psychiatry & Mental Health 9(2), 2004. © 2004 Medscape

Posted 08/24/2004

The 3 traditional mood stabilizing medications -- lithium, valproate, and carbamazepine -- have sometimes been referred to as the "chocolate, vanilla, and strawberry" of bipolar disorder. With the proliferation of treatments studied in randomized, controlled trials in bipolar disorder, and an equal or greater number suggested by anecdote or open trials, the range of flavors has increased dramatically. Since some of these newer treatments (lamotrigine, atypical antipsychotics) show similar efficacy to older treatments, determining when and how to apply the various treatment options has become correspondingly more complex.

The rise of evidence-based medicine in psychiatry, as in other medical specialties, offers one way of prioritizing treatments.^[1] Few clinicians would argue with the premise that treatments with proven efficacy should be selected over those with only suggestive data (though the widespread use of gabapentin in the face of negative placebo-controlled data in bipolar disorder might suggest otherwise). Of course, efficacy (how well a treatment works in a trial) does not translate perfectly to effectiveness (how well a treatment works in the real world). Still, as a combined measure of efficacy and tolerability, randomized controlled trials provide convincing evidence that a treatment is worth considering. A major reason for the rise of guideline- or algorithm-based care is to ensure that clinicians provide evidence-based treatment.^[1]

On the other hand, the evidence -- particularly that derived from randomized trials -- rarely addresses the questions of greatest concern to clinicians. Ethical issues aside, placebo-controlled trials are not particularly relevant to clinical practice, as few clinicians routinely prescribe placebo. Active-comparator trials are more useful, particularly when the comparator is a "gold-standard" drug such as lithium.

A different sort of problem is posed by timing, sequencing, and matching treatments. In terms of timing, clinicians often need to decide how soon to intervene, how rapidly to titrate, how long a trial should last, and at what dose. Most phase 3 clinical trials, designed to show a difference in efficacy at end point, provide only limited data to address this issue. For example, how soon after starting an atypical antipsychotic for mania should the success or failure be assessed?

With sequencing, clinicians need to decide which treatment to use first, how to determine when additional treatment is needed, and whether to add or switch to the second (and subsequent) treatments. Which antipsychotic should be tried first? If the antipsychotic in the first example fails to yield sufficient response, how soon should the treatment regimen be changed? Should another atypical be tried, or a traditional mood stabilizer added? Here again, some trials provide indirect evidence, but rarely is this clinically important topic studied directly.

Finally, with matching, clinicians need to consider whether a particular patient type should have a particular treatment prioritized or deferred. Is lithium less useful in rapid cycling, or for patients with a greater number of prior mood episodes? The answer might depend on which studies are consulted.^[2,3] As with the other questions, secondary analyses of clinical trials sometimes provide an answer, though rarely a consistent or well-validated one.

These are the sorts of questions that treatment guidelines and algorithms help the individual clinician to answer. Three typical guidelines -- the American Psychiatric Association (APA) bipolar treatment guidelines,^[4] the British Association for Psychopharmacology (BAP) guidelines,^[5] and the Texas Medication Algorithm Project (TMAP)^[6] -- all attempt to provide more detailed treatment recommendations. A fourth, the Expert Consensus Guidelines,^[7] also provides treatment guidelines but is currently being revised and will not be considered in more detail here.

Before looking at specific recommendations, it is worth considering how the documents are developed. All 3 began with a review of evidence for treatment strategies in bipolar disorder, relying heavily on randomized controlled trials wherever possible. Where evidence is insufficient to address a clinically important question, the groups attempt to achieve a consensus of experts. For example, the APA guidelines rely on the development of evidence tables, grading studies in terms of strength of evidence for various treatments, which are then reviewed by bipolar experts who attempt to achieve consensus on treatments.

In general, the more objective the criteria for recommendations, the closer the guidelines will be to evidence-based, but the less well the guidelines can address questions for which little or no evidence exists. Conversely, the more a guideline relies on expert opinion, the greater the chance that evidence will be de-emphasized, but the better the guideline will be at addressing less well-studied questions. In other words, an advantage of the expert consensus approach is that questions of treatment timing or prioritization are easily addressed. A disadvantage is that, where the experts are inconsistent with each other or with other survey questions, no formal mechanism exists for clarifying their decisions. Perhaps a greater disadvantage is that, while the experts are expected to be familiar with the literature, they may nonetheless make recommendations apparently at odds with the evidence base.

The role of "experts" in developing guidelines raises 2 additional issues. First, because experts are often selected based upon their research expertise (ie, their contribution to the literature) rather than their clinical expertise, they may have only a vague sense of the pressing clinical issues. The true "clinical experts" who see 30 patients a day in community clinics are simply too busy to write guidelines!

An additional consideration has received attention in the general medical literature^[8] as well as the lay press: many experts have potential conflicts of interest, financial or otherwise. They may thus be called on to opine about trials they were paid to design or conduct, or papers they authored.^[9] As many of the most influential trials are industry-funded, this concern becomes increasingly relevant.^[8]

One last consideration is that developing guidelines takes time, and they risk being out of date by the time they are published. The bipolar clinical trial literature, despite dramatic growth in the last 10 years, is still sparse enough that a single large trial has the potential to dramatically impact clinical practice.

The approach to treating bipolar depression highlights similarities and subtle differences among the 3 treatment guidelines. The prevalence and morbidity of bipolar depression makes appropriate treatment critical,^[10] despite ongoing controversies about optimum treatment strategies.^[11] In this section, we consider the recommendations of the APA, BAP, and TMAP guidelines for beginning treatment in bipolar depression, and for managing "breakthrough" depression.

The APA guidelines^[4] favor beginning treatment for bipolar depression with lithium monotherapy, with lamotrigine considered an alternative first-line agent. While less evidence-based, the APA also acknowledges that combination therapy (with lithium and an antidepressant) is appropriate in some circumstances or preferred by some clinicians.

The APA also identifies 2 particular groups that may require distinct treatment. For life-threatening depression (eg, with high risk of suicide), electroconvulsive therapy should be considered. Where psychosis is present, addition of an antipsychotic is advised.

Where patients are already treated with a maintenance medication, or fail to respond to initial treatment, the APA guidelines advise optimization of the maintenance medication, followed by addition of an antidepressant, namely lamotrigine, paroxetine, or bupropion. The precise definition of optimization is not provided, though most often it refers to ensuring adequacy of dose or serum levels.

The role of psychotherapy in relation to pharmacotherapy is not specifically addressed in the APA guidelines, though they note that many studies on bipolar and unipolar depression, especially the latter, justify using cognitive-behavioral therapy and interpersonal therapy for depression.

The APA guidelines do not comment on continuation of antidepressant treatment following acute episodes, due in part to the paucity of evidence. They do acknowledge that, while few data exist, in some circumstances longer-term use of an antidepressant may be required for subthreshold or breakthrough depressive symptoms.

In general, the BAP guidelines^[5] are quite similar to those issued by the APA, though they distinguish between degrees of depression severity. The BAP guidelines advise initiation of an antidepressant plus an antimanic agent (specifically, lithium, valproate, or an antipsychotic); notably, in contrast to the APA guidelines, lamotrigine is omitted from these options. For less severe depression, they advise monotherapy, with lamotrigine, lithium, or possibly valproate. Notably, the BAP guidelines omit lamotrigine from treatment of more severe depression but include valproate as a treatment option either in combination or as monotherapy.

As in the APA guidelines, a form of structured psychotherapy -- in particular, interpersonal therapy or cognitive behavioral therapy -- is also supported, when available. Like the former guidelines, there are no details provided about how or when psychosocial interventions should be combined with pharmacotherapy.

For "breakthrough" depression, like the APA guidelines, the BAP advises optimization followed by addition of an antidepressant.

Distinct from the APA guidelines, the BAP guidelines do address appropriate duration of antidepressant treatment. On the important subject of how soon to discontinue an antidepressant, the BAP guidelines suggest that antidepressants may be discontinued after as little as 12 weeks, though the factors that may influence this decision (eg, history of rapid cycling) are not specified.

Finally, TMAP^[6] also represents expert consensus, derived in part from literature review. In contrast to the previous 2 guidelines, the goal of TMAP is explicitly to develop algorithms -- step-by-step approaches to treatment. For this reason, its recommendations are generally more detailed, and more attention is paid to implementing treatment: titrating doses, managing side effects, and sequencing interventions.

Consistent with their recommendations for the treatment of mania, TMAP always advises initiation of a mood stabilizer -- in particular, lithium, valproate, or olanzapine -- as monotherapy prior to utilizing medication combinations. Thus, like the APA guidelines, monotherapy is favored, though with a different array of treatment options. If initial treatment fails, an antidepressant (which may include selective serotonin reuptake inhibitors, bupropion, or lamotrigine) is added; if this next intervention fails, lithium or another antidepressant may be added, or the initial antidepressant may be switched.

An interesting distinction of the TMAP guidelines is that they represent probably the first formal consideration of atypical antipsychotics as mood stabilizers or antidepressants: olanzapine is categorized as a mood stabilizer, and atypical antipsychotics are proposed as alternatives (albeit third line) for treating depression.

As the TMAP algorithms address medication, psychosocial interventions are not discussed in detail, other than to state that they may be considered. However, a strength of TMAP is that it includes explicit instructions for timing treatment interventions (referred to as "Critical Decision Points"), based upon degree of response: patients are evaluated every 2 weeks during an acute episode, at which time the medication doses may be increased, or the next step in the algorithm may be taken. TMAP also places an upper limit (8 weeks) on the duration of a treatment trial before proceeding to the next step.

Following remission of depression, TMAP recommends continuing antidepressant medication for at least 1-3 months without dose decrease. In general, the antidepressant may then be tapered, though in patients who previously relapsed with antidepressant taper, long-term treatment may be required.

In sum, these 3 sets of treatment guidelines generally draw similar conclusions from their review of the literature. They all favor initiation of a mood stabilizing (antimanic) treatment first, with the decision to add an antidepressant influenced by the degree of depressive severity. All treatment guidelines support electroconvulsive therapy for more severely ill patients and antipsychotic treatment where psychosis is present. Areas of difference include how long to continue antidepressant treatment following remission, where and when lamotrigine or atypical antipsychotics are included in treatment, and the extent to which timing of treatment addition or change is specified.

Psychiatry, like other areas of medicine, has been markedly influenced by the emphasis on evidence-based medicine.^[1] Treatment guidelines represent one means of developing and disseminating a standard of care that is based upon evidence rather than anecdote. In general, contemporary guidelines such as those developed by APA, BAP, and TMAP are similar in their recommendations for bipolar disorder, and for bipolar depression in particular. Unfortunately, because many clinical trials are intended for regulatory purposes rather than clarification of clinical issues, the evidence is often insufficient to answer basic but profoundly important treatment questions. Here the role of experts becomes important; understanding the way in which guideline authors attempt to fill in the gaps in the evidence base can help clinicians to interpret and implement their recommendations.

1. Shaneyfelt TM, Mayo-Smith MF, Rothwangl J. Are guidelines following guidelines? The methodological quality of clinical practice guidelines in the peer-reviewed medical literature. JAMA. 1999;281:1900-1905. Abstract
2. Swann A, Bowden CL, Calabrese JR, Dilsaver SC, Morris DD. Differential effect of number of previous episodes of affective disorder on response to lithium or divalproex in acute mania. Am J Psychiatry. 1999;156:1264-1266. Abstract
3. Baldessarini RJ, Tondo L, Floris G, Hennen J. Effects of rapid cycling on response to lithium maintenance treatment in 360 bipolar I and II disorder patients. J Affect Disord. 2000;61:13-22. Abstract
4. American Psychiatric Association. Practice guideline for the treatment of patients with bipolar disorder (revision). Am J Psychiatry. 2002;159(4 suppl):1-50.
5. Goodwin GM, Young AH. The British Association for Psychopharmacology guidelines for treatment of bipolar disorder: a summary. J Psychopharmacol. 2003;17(4 suppl):3-6.
6. Suppes T, Dennehy EB, Swann AC. Report of the Texas Consensus Conference Panel on medication treatment of bipolar disorder. J Clin Psychiatry. 2000;63:288-299.
7. Sachs GS, Printz DJ, Kahn DA, Carpenter D, Docerty JP. The Expert Consensus Guideline Series: Medication Treatment of Bipolar Disorder 2000. Postgrad Med. 2000;Spec No:1-104. Abstract
8. Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA. 2003;289:454-465. Abstract
9. Fava GA. Conflict of interest in psychopharmacology: can Dr. Jekyll still control Mr. Hyde? Psychother Psychosom. 2004;73:1-4
10. Judd LL, Akiskal HS, Schettler PJ, et al. The long-term natural history of the weekly symptomatic status of bipolar I disorder. Arch Gen Psychiatry. 2002;59:530-537. Abstract
11. Ghaemi SN, Hsu DJ, Soldani F, Goodwin FK. Antidepressants in bipolar disorder: the case for caution. Bipolar Disord. 2003;5:421-433. Abstract

Roy Perlis, MD, Director, Psychopharmacology Consultation Service, Massachusetts General Hospital Mood and Anxiety Disorder Institute; Instructor in Psychiatry, Harvard Medical School, Boston, Massachusetts

Disclosure: Roy Perlis, MD, has disclosed that he has received grants for clinical research from Elan/Eisai and has served as an advisor or consultant for AstraZeneca, Bristol-Myer Squibb, Eli Lilly and Company, GlaxoSmithKline, Janssen, and Pfizer.