Clinical experience and evidence-based research have provided support for the efficacy of electroconvulsive therapy (ECT) for unipolar depressive episodes (
1), but the relative efficacy of ECT in treatment-resistant bipolar depression remains understudied. Bipolar depression, which dominates the longitudinal course of bipolar disorder, is associated with psychosocial and cognitive impairment and with significant total health care costs (
2). Although the prevalence of treatment-resistant bipolar depression appears to be relatively high, few randomized controlled trials have been conducted for this condition.
Unlike unipolar depression, bipolar depression has only four pharmacological treatments approved by the Food and Drug Administration (FDA): a combination of olanzapine and fluoxetine (
3), quetiapine (
4,
5), and, most recently, lurasidone monotherapy or lurasidone in combination with lithium or valproate (
6,
7). Antidepressant agents that are effective in unipolar depression do not appear to provide similar benefits in bipolar depression (
8). In this issue of the
Journal, Schoeyen and colleagues (
9) address an important scientific and clinical question: does a treatment with proven efficacy in treatment-resistant unipolar depression have the same effect in treatment-resistant bipolar depression? Bipolar depression is undoubtedly one of the psychiatric conditions with the most unmet medical need.
The authors conducted a 6-week randomized controlled trial (N=73) comparing ECT and an algorithm-based pharmacological treatment in patients with treatment-resistant bipolar depression in seven sites across Norway. This study appears to be the first randomized controlled trial comparing the efficacy of ECT and algorithm-based pharmacotherapy in treatment-resistant bipolar depression. Patients assigned to ECT received three sessions per week for up to 6 weeks using right unilateral placement of stimulus electrodes and brief pulse stimulation. The ECT group had adequate seizure duration and number of treatments. As cited by the authors, larger studies have supported comparable efficacy of right unilateral and bitemporal electrode placements in depressive episodes (
10). The results show that ECT was significantly more effective than pharmacological treatment. The mean score on the Montgomery-Åsberg Depression Rating Scale (MADRS) was 6.6 points lower in the ECT group than in the group receiving algorithm-based pharmacological treatment. More than twice as many ECT patients had a significant response (73.9% versus 35.0%, p=0.01), but the remission rate did not show significance and was quite modest for both groups (34.8% versus 30.0%, p=0.74). The low remission rate occurred in spite of a nonstringent remission definition, a MADRS score of 12 or less, rather than the more conservative criterion of 8 or less, as recommended by the International Society for Bipolar Disorders task force (
11).
An important question is whether the recruited patients truly represent a treatment-resistant group. The investigators defined treatment resistance as lack of response to two lifetime trials of antidepressants, lithium, lamotrigine, quetiapine, or olanzapine. For unipolar depression, more stringent definitions of treatment resistance have been utilized (
12). FDA-approved treatments have also defined treatment resistance as lack of response to two or more treatments
in the current episode (
13). Furthermore, the pharmacological treatment algorithm used by the authors, although state of the art in 2007, would currently not be considered standard, as newer treatments for bipolar depression, including the combination of olanzapine and fluoxetine (
3) or lurasidone in monotherapy (
6) or combination (
7), were apparently not included as first-line treatments. The latter was understandably excluded because the data had not been published when the present trial was conducted. Another limitation is that patients with a rapid cycling course, who are frequently treatment resistant, were excluded. The generalizability of the findings is also limited because patients with comorbid substance abuse, which are the majority in U.S. clinical settings, were excluded.
It appears that all patients initially received lamotrigine, valproate, or lithium (
14). These medications have shown limited efficacy in the treatment of bipolar depression (
2). Many patients later received either quetiapine or the combination of olanzapine and fluoxetine, but these were not first-line treatments. A challenge with the interpretation of the data is that the duration of each pharmacological treatment is not provided. As the study was not blinded, bias could have been introduced, especially because the treatment procedures differed considerably (
15). It is not clear that having independent investigators rating audiotaped interviews prevented bias. In spite of the above limitations, this report adds major value to the evidence-based data on the use of ECT as a treatment option for bipolar depression. As stated by the authors, the low remission rate with either treatment highlights the importance of developing better approaches to treatment-resistant bipolar depression.
Another major area of unmet medical need is of course maintenance treatment. The role of ECT in maintenance treatment still needs to be determined. With some possible exceptions, all patients who recover from an episode of bipolar depression will need maintenance treatment. Clinical randomized trials have provided evidence that continuation of the same treatment given during acute phases of the illness may be associated with the best outcome for maintenance treatment (
16). However, there are few data on the efficacy of ECT in maintenance treatment, and such data are desperately needed. The acute phase of this investigation is presented in this issue, but the entire study protocol has been described previously (
14). The longitudinal aspect of this investigation will include 21 weeks of follow-up in both treatment arms, permitting further information regarding characteristics of sustained response versus relapse. Identified by Lisanby as the “most pressing issue in the field” (
17), relapse rates after ECT are as high as 50% despite optimal continuation therapies (
17,
18). The majority of patients that relapse do so within 6 to 9 weeks of completing an ECT series (
18,
19). The treatment-resistant patients in this study who experienced syndromal response, defined as having persistent depressive symptoms after meeting response criteria, may have higher relapse rates at follow-up than the patients who achieved remission of their symptoms following the index ECT series. The field looks forward to the findings of the 21-week follow up results.
Although debated, ECT use in the United States may be declining (
20). Further investigations and rigorous clinical trials will solidify and expedite the use of ECT in treatment-resistant bipolar disorders. Patients need better treatment. We applaud the authors for their study, as it provides additional evidence of the value of ECT in the treatment of bipolar depression.