J Clin Psychiatry 2014 Feb;75(2):162-8
Department of Psychiatry and Psychotherapy, Charité Universitätsmedizin Berlin, Campus Mitte, Berlin.
Objective: Suboptimal availability of circulating thyroid hormones may contribute to the high rate of treatment failures in bipolar disorder. This study tested the efficacy of adjunctive treatment with supraphysiologic doses of levothyroxine in patients with bipolar depression and the hypothesis that women would display a better outcome compared to men.
Method: The aims of this multicenter, 6-week, double-blind, randomized, placebo-controlled fixed-dose (300 μg/d) trial conducted from 2004 to 2009 were to assess efficacy and tolerability of levothyroxine adjunctive to continuing treatment with mood stabilizer and/or antidepressant medication for patients with bipolar I or II disorder, currently depressed (DSM-IV), and to investigate gender differences in treatment response. The primary efficacy variable was mean change in Hamilton Depression Rating Scale (HDRS) score.
Results: Of 74 patients enrolled in the study, 62 (35 with bipolar I; mean age = 44.9 years) were randomized. Mean change in HDRS score from randomization to week 6 was larger in the levothyroxine group compared to the placebo group, with a 2.7-point difference (decline of -7.8 [38.3%] vs -5.1 [25.5%]; last-observation-carried-forward analysis). The course of HDRS scores over time from randomization to week 6 was significantly different between groups at week 4 (P = .046) but not at the end of the placebo-controlled phase (P = .198). The secondary analysis of women (n = 32) revealed a significant difference between groups in mean change in HDRS score (-16.6% placebo vs -42.4% levothyroxine, P = .018). A mixed-effects model for repeated-measures analysis showed a significant between-group difference in HDRS score (6.8, P = .012) for women. High thyroid-stimulating hormone levels, indicating suboptimal levels of circulating thyroid hormones, were predictive for positive treatment outcome in women treated with levothyroxine in a linear regression model (F3 = 3.47; P = .05).
Discussion: This trial demonstrated that patients treated with levothyroxine did numerically better than those treated with placebo; however, the study failed to detect a statistically significant difference between the 2 groups in the primary outcome measure due to a high placebo response rate. Previous findings that women show better improvement in depression scores with levothyroxine compared to men were confirmed.
Trial Registration: ClinicalTrials.gov identifier: NCT01528839.