Psychiatry Guest Editor's Column -- International Consensus Group on Bipolar I Depression Treatment Guidelines: Synopsis and Discussion*

Thomas A. M. Kramer, MD

In This Article


This group of international experts in the treatment of bipolar disorder was convened in New York in December 2003. As a departure point for their discussions, it was noted that the treatment of mania is fairly standardized throughout the world, but the treatment of bipolar depression varies substantially regionally. This group took as its mission to look at the available clinical evidence only. They did not take into account expert opinion, although other published treatment guidelines often do.

They first discussed problems in the treatment of bipolar depression. They noted the fact that at least until recently, this diagnosis has been overlooked in favor of mania in the treatment of bipolar disorder. In many areas of the world, it is either misdiagnosed as unipolar depression or underdiagnosed in general. Regarding clinical evidence, there has been very little up until recently, with most clinical investigation focusing on the treatment of mania. As a result, clinicians used what they knew would work symptomatically, namely antidepressants. Even though currently there are treatments that have been studied and approved specifically for bipolar depression, antidepressants remain the most common treatment, despite of the fact that there are consistent data that antidepressants can induce mania or rapid cycling. In addition, there is evidence that antidepressant monotherapy is less effective in preventing depressive symptoms in bipolar disorder than an antidepressant-mood stabilizer combination. As a result, many treatment guidelines, including the American Psychiatric Association's, recommend against antidepressant monotherapy for bipolar depression.

The group noted, however, that antidepressant monotherapy is frequently used anyway and speculated this was due to a number of factors. They cited 4 myths about bipolar depression treatment that have either been disproved or are not supported by current evidence: (1) bipolar disorder is not a lifelong illness and episodes only need to be treated acutely; (2) antidepressants should only be augmented with mood stabilizers if manic symptoms appear; (3) the addition of an antidepressant to a mood stabilizer has a more rapid onset of action; and (4) recent episode frequency has no effect on treatment selection. In addition to this mythology, they also noted that patients tend to prefer antidepressant monotherapy, particularly if they enjoy their periods of hypomania, so they may put pressure on their physicians. Despite this, the group noted hope-generating trends in practice patterns, particularly the use of second-generation antipsychotics as monotherapy for bipolar depression.

Regarding first-line treatments for the management of bipolar depression, the sheer number of studies demonstrating lithium to be effective, despite the fact that some of them are poorly designed, makes lithium the best-established treatment. The group cited numerous studies demonstrating this. Particularly in studies noting the prophylactic qualities of lithium vs antidepressants, lithium was consistently as good or better than antidepressants in preventing depressive symptomatology and considerably better at preventing manic symptomatology. It is important to note also that these studies showed that placebo was considerably better than antidepressants in preventing manic symptomatology. The group then discussed the studies examining the use of lamotrigine for bipolar depression, and also found it convincing. In a head-to-head comparison between lithium and lamotrigine, lamotrigine seemed to be more effective at delaying depressive episodes, and lithium seemed to be more effective at delaying manic episodes. Most recently, evidence has shown that a combination of olanzapine and fluoxetine had substantially higher rates of response and remission than either olanzapine alone or placebo, although the olanzapine group was significantly more effective than the placebo group.

The group then discussed the issues involved with treatment nonresponse. For nonrapid cycling patients, after optimization of the current treatment proves ineffective, they recommended combining 2 first-line treatments, although they admit there is very little, if any, evidence to support this recommendation. The other option would be to add an antidepressant to the first-line treatments, but they recommended strongly against tricyclics or monoamine oxidase inhibitors, as there is evidence that these antidepressants are the most likely to induce mania. They also noted that there are more recent data that show that patients who respond to antidepressant treatment added to a mood stabilizer are considerably more likely to relapse if the antidepressant is stopped, and they recommended that antidepressant treatment be continued with the mood stabilizer for at least a year after remission of the depression.

For rapid cycling patients, the group first noted that this subtype of bipolar disorder is often unrecognized, with many clinicians underplaying the importance of episode frequency in the management of the illness. They noted recent evidence that makes it less clear whether rapid cyclers are a more severe variant of bipolar disorder, but they did point out that it has some treatment implications. All of the first-line treatments mentioned above have been studied in rapid cycling and found to be effective, and, as such, the group recommended combining 2 of them in nonresponding rapid cycling patients. They also noted that valproate can be very beneficial for rapid cycling patients, particularly for patients with mania and mixed states. There is a great deal of evidence that in rapid cycling patients, manic symptoms are more likely to respond to treatment than depressive symptoms, and, as such, depression often becomes the prominent aspect of the illness. For these patients, they recommended considering augmentation with an antidepressant such as a selective serotonin reuptake inhibitor as a last resort.

For those patients who have psychotic symptomatology as a part of their bipolar depression, the recommendations were to add olanzapine, olanzapine-fluoxetine in combination, or electroconvulsive therapy (ECT). Although the group cited numerous studies showing the efficacy of all these treatments, and in particular of ECT for treatment-refractory bipolar depression, they noted there were very few data on effective treatment for patients with mood congruent vs mood incongruent psychosis.

One of the more common manifestations of treatment nonresponse is breakthrough mania. After the first-line treatment being used is optimized, if breakthrough episodes persist, the group discussed a number of options. The agents with the best evidence for long-term treatment of mania, according to this group, are lithium and olanzapine. They also cited good evidence for the efficacy of valproate and risperidone. There is also evidence that other atypical antipsychotics may be very useful in the treatment of breakthrough mania as augmenting agents. The group cited aripiprazole, ziprasidone, quetiapine, and clozapine as showing promise in this regard.

Finally, the group made some general management recommendations for the treatment of bipolar depression. Since this is a lifelong illness, patients who respond to their first-line treatment should continue that treatment long-term. There is good evidence that first-line treatments are both efficacious and safe for periods ranging from 6 months to 2 years. They pointed out that the courses of bipolar disorder in general and bipolar depression in particular vary so greatly from patient to patient; as such, selection of a treatment should be based on the individual patient. They cited as examples that lithium may be preferential for patients with particularly problematic manic episodes, lamotrigine may be preferential for bipolar patients with particularly difficult depressions, and ECT may be most useful for nonresponsive bipolar patients with psychosis. In closing, they emphasized that although they focused their discussion mainly on drug therapy, psychological treatments are a crucial part of the treatment of bipolar disorder and should be a part of every patient's treatment plan.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.