Depressed patients with bipolar disorder who receive mood stabilizer therapy do no better with addition of an antidepressant than with addition of placebo, according to an article published online March 28 by the New England Journal of Medicine.

The results are part of the large-scale, multi-site Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD), a $26.8 million clinical trial funded by the National Institutes of Health's National Institute of Mental Health. STEP-BD was designed to identify the best treatment options.

"Treating depression in people with bipolar disorder is notoriously difficult," said National Institute of Mental Health Director Thomas R. Insel. "STEP-BD sought to determine if adding an antidepressant to a mood stabilizer is effective and safe in treating depressive episodes. The results suggest that antidepressants are safe but not more effective than placebo as assessed in a large number of people with bipolar disorder."

Lead author Gary Sachs, MD, of Massachusetts General Hospital and colleagues studied 366 participants at 22 sites in the USA. Unlike most clinical studies, participants were recruited from clinical settings and were included in the study even if they were being treated for coexisting disorders such as substance abuse, anxiety or psychotic symptoms. Such open recruitment criteria allows the study's results to have broader applicability than a tightly controlled trial in which people are excluded from participating if they have coexisting disorders.

Before participants were randomized to bupropion, paroxetine, or placebo, doctors trained in the treatment of bipolar disorder adjusted participants' mood stabilizer doses to optimal levels to ensure they were receiving the most appropriate dose.

After about 26 weeks, 24 percent of patients who had been randomized to antidepressants stayed well for at least eight consecutive weeks -- the study's stringent standard for recovery; 27 percent of those randomized to a placebo stayed well long enough to meet the eight-week recovery standard.

In addition, about 10 percent of each group experienced emerging symptoms of mania, indicating that the antidepressants did not trigger a manic switch any more than placebo. Finally, when comparing the two antidepressants to each other, both showed similar rates of response and manic switch.

"Results of STEP-BD indicate that careful management of mood stabilizer medications is a reasonable alternative to adding an antidepressant medication for treating bipolar depression," said Sachs.

Future STEP-BD results will shed light on other treatment options for bipolar disorder, including psychotherapeutic treatments.