Long-Term Naturalistic Follow-Up of Patients With Bipolar Depression and Mixed State Treated With Electroconvulsive Therapy Pierpaolo Medda, MD,* Mauro Mauri, MD,* Sara Fratta, MD,* Benedetta Ciaponi, MD,* Mario Miniati, MD,* Cristina Toni, MD,*Þ Liliana Dell’Osso, MD,* and Giulio Perugi, MD*Þ Objective: The aim of the present study was to evaluate the long-term outcome in a sample of patients with bipolar disorder with severe de- pression or mixed-state resistant to pharmacological treatment who have responded to electroconvulsive therapy (ECT). Method: The study involved 36 patients with major depression (5 pa- tients with bipolar I depression, and 14 patients with bipolar II depres- sion) or mixed state (17 patients) treated with bilateral ECT delivered using a brief pulse stimulator Mecta 5000 Q on a twice-a-week sched- ule. The patients were evaluated before ECT (baseline) and 1 week after the ECT course (final score) using the Hamilton Depression Rating Scale, Mania Rating Scale, Brief Psychiatric Rating Scale, and Clinical Global Improvement. The Longitudinal Interval Follow-up Evaluation was administered every 16 weeks to assess time to relapse (defined as LIFE scores Q5 for at least 2 consecutive weeks or as the need for hospitalization) and periods of response and remission. Results: The mean duration of follow-up was 55.3 T 30.4 weeks (range, 24Y160 weeks). Thirteen patients (36.1%) showed a depressive relapse during the follow-up; the mean time (length) of depressive relapse was 20.4 T 21.8 weeks (range, 2Y60 weeks). Twenty-nine patients (80.5%) fulfilled the criteria for a full remission from depressive symptoms after 6.7 T 7.9 weeks from the last ECT. Seventeen patients (47.2%) were in remission for more than 70% of the time. No manic episodes occurred during the follow-up, only 1 patient had a mixed episode, and 11 patients had a hypomanic episode. Conclusion: Electroconvulsive therapy showed a positive impact on the clinical course of severe and treatment-resistant patients with bipolar disorders, as suggested by the high number of weeks spent in remission during the follow-up period. In our study, the duration of depressive episode was related to early relapse during follow-up period. Key Words: bipolar depression, mixed state, electroconvulsive therapy, follow-up, relapse (J ECT 2013;29:179Y188) T he treatment of bipolar depression is still challenging. A recent study indicates that antidepressant adjunctive treat- ment in a large cohort of patients with bipolar depression is no more efficacious than optimized monotherapy with mood sta- bilizers. 1 Moreover, treatment with antidepressants has been associated with manic switches, mixed state induction, and cy- cle acceleration. 2Y5 Even more complex is the treatment of mixed states, characterized by longer and more severe episodes in comparison with both depression and mania, high frequency of psychotic features, great risk of suicide, poor outcome, 6 and low response rate to pharmacological treatments. 7Y9 Electroconvulsive therapy (ECT) is a highly effective and safe treatment of acute episodes of both depressive and mixed phases of bipolar disorder. 10Y12 However, relapse rates are high once treatment stops and often occur in the period immediately after ECT. 13Y15 Early follow-up studies, conducted in the 1960s showed a relapse rate of approximately 20% in patients who received an active medication (tricyclic antidepressant [TCA] or monoamine oxidase inhibitor) during and after ECT compared to a relapse rate of approximately 50% in the placebo-controlled group. 16Y18 In most recent studies, despite pharmacological continuation treatment, a high rate of relapse/recurrence has been observed (38%Y64%). 13,19Y23 However, conflicting results have been attained in different study populations. 24,25 The different outcomes can be attributed not only to the changes over time in assessment methods or ECT administration but also to the shift in patients’ characteristics. At present, ECT is no longer a first-line treatment for severe depression, but medi- cation resistance has become its primary indication, and patients referred to ECT are frequently chronic and more severely ill. 26,27 Sackeim et al 20 reported relapse rates of 84% for patients treated with placebo, 60% for those treated with nortriptyline, and 39% for nortriptyline-lithium in a 6-month trial of patients with major depression considered responders to ECT. In a 1-year nat- uralistic follow-up study, Birkenhager et al 23 reported a relapse rate of 41% in a sample of 40 patients with major depression. Prudic et al, 13 in a prospective naturalistic 24-week follow-up study involving 154 patients at 7 hospitals, observed a 64.3% of relapse after ECTwith a median time to relapse of 8.6 weeks. Patients with affective episodes resistant to medication have an increased risk of relapse after ECT. 14,22,28 In other studies, re- sidual symptoms, female sex, psychotic features, comorbid con- ditions, number of ECT treatments, have been associated with an increased risk of relapse. 13,20,29Y31 Most of the follow-up studies have been limited to patients with unipolar disorder, and the effect of ECT on long-term out- come in patients with bipolar disorder is not well studied. Con- tradictory conclusions have been attained for both the relapse rate 24,32 and the clinical and symptomatological characteristics related to early relapse, as medication resistance, 22 and psychotic features 22,33 The aim of the present study was to evaluate the long-term outcome in a sample of patients with bipolar disorder with de- pression or mixed state responsive to ECT. Clinical and demo- graphic factors were evaluated to predict outcome. Finally, we used a survival analysis to identify if length of current episode could affect the time of relapse. MATERIALS AND METHODS The study involved 36 patients with treatment-resistant major depression or mixed state who had received ECT between ORIGINAL STUDY Journal of ECT & Volume 29, Number 3, September 2013 www.ectjournal.com 179 From the *Department of Neuroscience, University of Pisa, Pisa, Italy and †Institute of Behavioural Science ‘‘G. De Lisio’’, Carrara-Pisa, Italy. Received for publication December 10, 2012; accepted January 17, 2013. Reprints: Giulio Perugi, MD, Department of Neuroscience, University of Pisa, Via Roma 67, 56100, Pisa, Italy (e-mail: g.perugi@psico.med.unipi.it). The authors have no conflicts of interest or financial disclosures to report. Copyright * 2013 by Lippincott Williams & Wilkins DOI: 10.1097/YCT.0b013e3182887b7a Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.