Background: Whereas studies have suggested an association between abnormal TSH serum levels on antidepressant efficacy in major depressive disorder (MDD), the impact of normal serum TSH levels on antidepressant efficacy is unknown. This study aimed to investigate whether TSH serum levels within the normal range predict short-term antidepressant efficacy.
Methods: Pooled data from 7 randomized, double-blind, placebo-controlled, MDD clinical trials were analyzed to compare the efficacy of duloxetine depending on baseline serum TSH levels. Adult outpatients with MDD (DSM-IV criteria) received duloxetine (60-120 mg/day, n=1242) or placebo (n=827) for up to 9 weeks. Efficacy was measured based on the 17-item Hamilton Rating Scale for Depression total score change, response rate, remission rate and time to response. TSH serum levels were measured at baseline using Microparticle Enzyme Immunoassay technology. Only patients with a TSH serum level within the normal range were considered for the analysis.
Results: The mean (SD) of baseline TSH serum levels was 1.55 (0.86) mIU/L (median: 1.35; interquartile range: [0.92-1.94]). No significant treatment-by-TSH quartile interaction was evidenced in change from baseline, response, nor remission, indicating that the magnitude of duloxetine's treatment effects did not differ significantly between TSH quartiles. No significant difference in time to response was evidenced between any of the quartiles.
Limitations: This analysis is a post-hoc analysis of pooled data.
Conclusion: In this analysis of pooled data, the overall response to duloxetine in MDD did not differ regarding baseline serum TSH levels when considering TSH within the normal range.
Copyright 2009 Elsevier B.V. All rights reserved.