Restraint Urged in Adding Antidepressant to Mood Stabilizer

Results from the STEP-BD study (Systematic Treatment Enhancement Program for Bipolar Disorder), the largest randomized, double-blind, controlled study on bipolar disorder treatment, suggest that most patients with bipolar depression do not gain additional benefits from adding an antidepressant to mood stabilizers, Gary Sachs, M.D., explained. Sachs, the principal investigator of the National Institute of Mental Health (NIMH)-funded study, is an associate professor of psychiatry at Harvard Medical School and director of the Bipolar Clinic and Research Program at Massachusetts General Hospital.

In STEP-BD 366 patients with acute bipolar depression were randomized to receive a mood stabilizer plus either an antidepressant (bupropion or paroxetine) or a placebo. The rates of response and recovery were slightly higher in the mood-stabilizer-only group than in the mood stabilizer-plus-antidepressant group, although the differences were not statistically significant.

Adjunct intensive psychotherapy was validated in the STEP-BD study.“ Not only did we have a higher response rate and a faster response, but patients stayed well longer,” Sachs said. However, he cited other research suggesting that cognitive-behavioral therapy is effective for patients in the early phase of bipolar disorder but not for those who have been ill for a long time.

A study conducted by the Stanley Foundation Bipolar Collaborative Network (SFBCN) provided somewhat different insights, according to Mark Frye, M.D., a professor of psychiatry and director of the Mood Clinic and Research Program at the Mayo Clinic. This randomized, double-blind study compared three antidepressants with different mechanisms—sertraline, bupropion, and venlafaxine—as an adjunct to mood stabilizers in patients with bipolar I and bipolar II depression. After 10 weeks of treatment, about half of the 174 patients achieved response, and there was no significant difference among the three drugs. About one-third of the patients achieved remission. Unlike in STEP-BD, all patients in this study had had incomplete response to mood stabilizers before enrollment. There was no placebo group.

Two other researchers reviewed evidence on the effectiveness of antidepressants for bipolar depression from different perspectives.

In patients who respond to and tolerate antidepressants, continued antidepressant treatment after the resolution of their acute symptoms seems to be beneficial in delaying or preventing relapse within a year, according to Lori Altshuler, M.D., a professor of psychiatry and director of the Mood Disorders Research Program at UCLA. She cited findings from two observational, naturalistic trials of patients who were treated in the SFBCN.“ Continuing antidepressants for at least a year after remission may protect against depressive relapse,” she noted.

Does this conclusion contradict the STEP-BD results? Not necessarily. The open studies at SFBCN included only patients who had responded to and tolerated antidepressants. Only 14 percent of all 589 bipolar patients who were exposed to antidepressants did well enough on antidepressants for at least two months to enter the studies.

Nassir Ghaemi, M.D., M.P.H., director of the Bipolar Disorder Research Program and an associate professor of psychiatry and public health at Emory University, presented data from a randomized, open-label study on the long-term outcomes of bipolar patients on antidepressants. Patients who had responded to the combination of a mood stabilizer and an antidepressant were randomized to either continue with the drug combination or stay on only the mood stabilizer for one year of follow-up.

The combination of continued antidepressant and a mood stabilizer was effective in reducing subsyndromal depressive morbidity and delayed relapses somewhat, Ghaemi said. However, the benefit of adjunct antidepressants observed was smaller than in the nonrandomized, observational studies above. Long-term adjunct antidepressant use did not increase the time in remission or reduce the frequency of depressive episodes.

“We selected bupropion and paroxetine [for the STEP-BD study] because they are less associated with mood switching than other antidepressants,” Sachs said. The results confirmed that these drugs did not increase the risk of switching these patients to mania or hypomania.

The three-drug comparison study presented by Frye pointed to a difference in the tendency to cause a mood switch among antidepressants. Venlafaxine was associated with a significantly higher rate of switching to mania than were bupropion and sertraline, especially in patients with rapid cycling. Bupropion and sertraline had a similar effect on switching.

Frye listed additional major risk factors that may contribute to mood switching, including use of tricyclic antidepressants, a history of antidepressant-induced mania, a low level of thyroid-stimulating hormone, the presence of certain polymorphisms in the 5HTTLPR gene, an earlier onset of the disorder, and comorbid alcoholism. In addition, patients with bipolar I disorder may have a higher switch rate than patients with bipolar II disorder on antidepressants, and may have minimal manic symptoms during the depressive phase.

The rapid-cycling type poses additional risks. In the randomized, open-label study mentioned by Ghaemi, he and his colleagues observed “a paradoxical phenomenon that antidepressants actually worsened the depression of patients with rapid cycling.”

“Start with mood stabilizers” to treat patients with bipolar depression was the advice from all the presenters to a full house of attendees that spilled into the hallway. An adequate trial of mood stabilizers is critical for a majority of patients. “If they're not already on lithium or lamotrigine, put them on,” said Robert Post, M.D., chief of the Biological Psychiatry Branch at NIMH.

If patients do not fully respond to a mood stabilizer, the panel advised trying newer antipsychotics such as quetiapine or olanzapine. Sachs stressed the proven effectiveness of psychotherapy.

Antidepressants should be used with far more restraint than is exercised in current practice, most presenters agreed. A specific group of patients may respond well to antidepressants, but identifying them can be difficult. Ultimately, psychiatrists must base their clinical decisions on careful assessment of individual patients.

The complexity of bipolar disorder, frequent misdiagnosis, a lack of relevant research, and inappropriate but common prescribing of antidepressants create many challenges to psychiatrists. Although some consensus emerged from the discussion, many important debates remain.

“Most patients we encounter in office practice are bipolar II patients, and you shouldn't extrapolate bipolar I data to bipolar II. Unfortunately, there is a paucity of data,” said Frederick Goodwin, M.D., a professor of psychiatry and director of the Psychopharmacology Research Center at George Washington University Medical Center.

He blamed the short duration of most bipolar studies for much of the confusion in the literature. “The timeframe is critical to understand the distinction between relapse prevention of one continuous episode and prevention of the next episode,” Goodwin said. “The average time between two episodes is 13 to 18 months.”

Although studies suggest that continuing antidepressant treatment for up to a year can prevent relapse into the same depressive episode, there are almost no longitudinal data on how these drugs may affect the next episode, which will probably be manic and may be exacerbated by continued antidepressant use, he pointed out.

Further complicating the picture, “50 percent of bipolar patients are misdiagnosed by psychiatrists as unipolar depression,” Goodwin continued. “What do you think happens in a busy internist's office or a family doctor's office?” He cited recent data indicating that antidepressants are currently the most prescribed type of drugs, and more than 80 percent of all antidepressants are prescribed by nonpsychiatrists.

Patients usually visit physicians when they are in a depressive episode, and past manic symptoms are often missed. Mixed episodes, rapid cycling, and comorbidities pose additional diagnostic difficulties.

“Most adult patients we see have already taken antidepressants before they ever get exposed to a mood stabilizer.”

“Fifty years after antidepressants became available, we still don't know how to best treat bipolar depression,” Post commented. “We need studies from NIMH, comparative data, crossover data.... We need better algorithms that are not made up.” ▪