Susan had managed to balance a busy career in business with raising a family. She was a high-energy woman who put a lot into her work; when she felt overwhelmed, as she often did, she relied on her executive assistant to help keep her organized. Susan had probably had a first depression in her teens from which she’d recovered without any treatment. The second episode came in her mid-40s, after the death of her father, with whom she’d been very close. She was treated with venlafaxine and appeared to have a good response. She also started psychotherapy to address some life-long problems.

At first, things went well, but over the next 2 years, she found herself on a “roller-coaster” of moods — high energy for a few days, followed by periods of lethargy and despair. Lithium was added and Susan felt a bit less driven, but she settled into a steady state of mixed moods, feeling quite energetic at times, while low and preoccupied with losses at others. When on a high, she was overly sure of herself and made what in retrospect were several mistakes in judgement.

Diagnosis at last

Susan was finally diagnosed with bipolar II mood disorder. One critical question in the treatment of bipolar disorder is whether antidepressants are helpful or whether they destabilize the illness. Now, there’s some new evidence from the largest treatment study of bipolar illness ever performed, the Systematic Treatment Enhancement program for Bipolar Disorder (STEP-BD), which suggests that antidepressants are “mood destabilizers” in this group of patients (Am J Psychiatry 2008; 165:370-7). Previous trials have suggested that patients who are “rapid cyclers,” meaning they have more than 4 mood episodes in a year (depression, mania or hypomania, or mixed mood states — as Susan had), become worse while taking antidepressants, experiencing even more cycles. Most rapid cyclers are women and bipolar II. Episodes are demarcated by either 2 months of normal mood or a sudden shift in polarity.

In this new study, 1,742 bipolar I and II patients were followed for up to a year of treatment. Clinicians were trained in “best treatment available” based on current evidence and then free to treat subjects as they saw fit. The major predictor of worse outcome was antidepressant use. Sixty percent of people in the study received them, usually with a mood stabilizer. In each of 4 outcome groups, the likelihood of developing mood cycling increased linearly with antidepressant use. In an accompanying editorial, Nassir Ghaemi suggests that the mood destabilizing effects of antidepressants are the most common cause of refractory bipolar illness and that stopping antidepressants may be the “sine qua non” of treatment in bipolar disorder.

Susan slowly recovered and began to feel like her old self after her antidepressants were slowly reduced and another mood stabilizer, lamotrigene, was added. The bottom line: check for personal or family history suggestive of bipolar illness before starting antidepressants. If there is some doubt (there often is), follow these patients closely; if they aren’t getting better or get worse, consider a psychiatric consultation.