For more than a century, there have been rich descriptions in the clinical literature of prominent depressive symptoms in patients with thyroid disease, particularly clinical hypothyroidism.1,2 This link between the thyroid and depression1,2 provided optimism that study of the thyroid axis in unipolar depression would unravel the mystery of the etiology of the disorder. An enormous research effort over the last 50 years3 failed to achieve this goal but, nonetheless, produced some important findings. First, the vast majority of patients with major depression are euthyroid.3 Second, basal peripheral thyroid hormone levels are not particularly informative,4 although subjects with depression have higher mean thyroxine (T4) levels compared with when they are remitted and also compared with healthy control subjects.3–7 Moreover, significant decreases in T4 levels occur with response to various treatments, including antidepressant drugs,3–5 electroconvulsive therapy6 and cognitive–behavioural therapy.7 Significant changes in triiodothyronine (T3) levels are inconsistent.3,4 Last, one-quarter to one-third of subjects with depression will have a blunted thyrotropin response to thyrotropin-releasing hormone (TRH).5,6 This finding is consistent with the increased T4 levels observed in depression3,4 and suggests overactivity of the thyroid axis in depressive illness,8 the pathophysiological significance of which remains to be clarified. Despite the consistent observations about the thyroid and depression reported in both the psychiatric and endocrine literature, interest in the thyroid as a major factor in the biology of depression has waned, and the thyroid axis has received little attention in current models of the etiology of depression.9,10 Recent clinical and preclinical studies would suggest that the thyroid should not be overlooked.
Thyroid hormones have been shown to potentiate antidepressant response in treatment-resistant depression. This has been observed particularly for T311 but also for T4.12 Although earlier studies have been criticized for methodological limitations, T3 has been shown to be effective in both open and controlled studies,11 with both tricyclic antidepressant drugs11 and selective serotonin reuptake inhibitors,13–15 and has been shown to be significantly more effective than placebo in a meta-analysis of all available studies.11 Moreover, T3 has been shown to be comparably effective to lithium, one of the standard antidepressant augmentation agents, in a small controlled trial16; this has also been shown in a recent report from the STAR*D sequential treatment trial, where T3 compared favourably to lithium augmentation (remission rates of 25% and 15%, respectively) in 142 patients.15 Although T3 has not gained widespread acceptance as a treatment option, and its clinical effectiveness requires further evaluation, the current literature, notwithstanding its flaws, is persuasive that, at least in selected cases, thyroid hormones have antidepressant effects when used in conjunction with other antidepressant drugs.11–16 Given that most patients with depression are euthyroid,3,4 the thyroid hormone appears to have an antidepressant action that is not related to providing replacement therapy for occult thyroid disease.
Progress in understanding the mechanism of thyroid action in the brain was impeded, until the 1970s, by the belief that these hormones had limited impact on the mature brain, although they were crucial to normal brain development.17 Subsequently, the importance of thyroid hormones to mature brain function has been demonstrated in several ways. First, the nuclear T3 receptor, the site of initiation of thyroid hormone action, is widely distributed in the brain, and the Tr-α subtype is specific to the brain.18 Second, there are specific mechanisms for the active transport of thyroid hormone into the brain19; the type II deiodinase enzyme, responsible for the monodeiodination of T4 to the physiologically active hormone T3, is found only in the central nervous system.20 T3 likely exerts its action in the brain by T3 receptor-mediated changes in gene expression,17 although an effect on neurotransmission directly at the synapse is also plausible.21
Endogenous or exogenously administered thyroid hormones may play a role in the etiology of depression and the biology of treatment response at several points. First, the monoamine hypothesis of depression is still an important construct in understanding the biology of the disorder, in guiding research and in developing new treatments.17,22 The thyroid hormone has complex interactions with the catecholamines,17 whereas its effects on serotonin systems are more reproducible and suggest that thyroid hormone levels may have an inverse relation to brain serotonin function, mediated by changes in serotonin receptor sensitivity.22 One of the current models of the etiology of depression proposes that reduced hippocampal neurogenesis is likely related to excess adrenal steroid and cytokine secretion, with consequent reductions in neurotrophic factors, such as brain-derived neurotrophic factor (BDNF).9,10 Reduced hippocampal volume, adrenal hyperactivity and reduced serum BDNF levels have all been demonstrated in major depression,9,10 and all antidepressant drugs appear to have a common effect in enhancing hippocampal neurogenesis.23 Recent studies in rats demonstrate that thyroid hormones enhance neuronal plasticity,24 increase hippocampal neurogenesis25,26 and enhance the expression of various neurotrophins, including BDNF, either directly or through effects on monoamine receptors.27,28 These data therefore support and may provide a potential explanation for the antidepressant effects observed for the various thyroid hormones.
Small changes in brain thyroid hormone levels may be associated with substantial changes in brain function.17 This notion is consistent with the changes observed in thyroid hormone levels with various antidepressant treatments,3–7 the antidepressant potentiating effect of relatively small doses of thyroid hormone11 and the increasing database on thyroid effects on brain neurochemistry and gene expression.24–28 The thyroid is surely not the primary cause of unipolar depression, advanced by the early endocrine literature, nor may study of the thyroid axis in depression yield robust findings, as observed for the adrenal axis. However, the evidence suggests that thyroid hormones cannot be ignored as important factors in the cascade of biological events leading to the onset of depression and to antidepressant response.
Footnotes
Medical subject headings: thyroid; lithium; depression.
Competing interests: None declared.