Hyperthyroidism and depression: a clinical case of atypical thyrotoxicosis manifestation

The relationship between psychiatric symptoms and thyroid function has been well known and studied since antiquity. The common view is that clinical hypothyroidism is associated with depressive symptoms, whereas the psychiatric manifestations of hyperthyroidism are agitation, emotional lability, hyperexcitability, occasionally accompanied by angry outbursts, and euphoria. The case here reported overturns this conventional medical knowledge. A 73-year-old Italian woman experienced a severe major depressive episode with psychotic and melancholic features during laboratory thyrotoxicosis. No classical clinical signs and symptoms of thyrotoxicosis were present. Psychiatric symptoms improved together with the resolution of the hyperthyroid state. Historically, different cases of so-called ‘apathetic hyperthyroidism’ have been described. Recent neuroimaging and animal studies provided possible neurobiological explanations, showing how the excess thyroid hormones could affect brain structures involved in the regulation of mood, leading to depression. A direct link between hyperthyroidism and depression seems to be likely. This insight may be relevant in facilitating early diagnosis of thyroid disease and the planning of therapeutic strategies.


Introduction
Psychiatric symptoms are common in the context of thyroid diseases. Indeed, hypothyroidism is frequently associated with depression (Williams et al., 2009), whereas hyperthyroidism usually manifests with agitation, delirium, irritability, hyperexcitability, and euphoria (Akamizu et al., 2012). However, the opposite has also been described.
Although rare, cases of so-called 'apathetic hyperthyroidism' have been reported in the literature. They are characterized by symptoms of depression, apathy, and 'intellectual stupor' (Lahey, 1931). These symptoms could mask the classical signs and symptoms of thyrotoxicosis and may delay this diagnosis. In recent years, different studies have investigated the neurobiological mechanisms underlying the relationship between psychiatric symptoms and thyroid function and how thyroid hormones could affect brain structure. Neuroimaging (Schreckenberger et al., 2006;Zhang et al., 2014aZhang et al., , 2014b and animal studies (Ruel et al., 1985;Desouza et al., 2005;Sala-Roca et al., 2008;Martí-Carbonell et al., 2012) seem to suggest that the excess of thyroid hormones and, therefore, hyperthyroidism rather than hypothyroidism could affect brain structures involved in the regulation of mood and emotions, leading to depression.
We herein report the case of a patient with laboratory findings of thyrotoxicosis in the absence of typical physical and psychiatric manifestations of the disease. She showed a clinical picture of severe major depression, which may appear inconsistent with laboratory findings of hyperthyroidism. She indeed showed a depressed mood with psychotic and melancholic features that eventually improved following the improvement of the laboratory hyperthyroid state.

Case
A 73-year-old Italian woman was admitted to our psychiatric ward because of the onset of severe depressive symptoms. She showed depressed mood, diminished interest and pleasure in all activities, anxiety, fatigue, decreased appetite, and sleep disturbances. Delusions with themes of somatic disease, death, and nihilism were also present. She was overly concerned about perceived defects in her physical appearance, and her worries reached a delusional intensity. She was completely convinced; she looked ugly, physically abnormal, and deformed 'like a monster'; her preoccupation was focused on her skin, hair, and also internal organs. Moreover, she presented ideas of reference, olfactory hallucinations (she perceived her skin as smelly and wormy), and cenesthetic hallucinations (feeling a pushing sensation in her visceral organs and pinching the skin). She had no insight and no compliance with treatment, so compulsory treatment was necessary.
One month before the hospital admission, she attended for the first time a psychiatric outpatient clinic because of depressed mood, somatic concerns, and anxiety. Psychopharmacological treatment with olanzapine and sertraline was prescribed, but she never took these medications. Her family psychiatric history was negative. Her medical history included hypertension, a recent diagnosis of endometrial cancer, and a subsequent hysterectomy three months before her admission to our psychiatric inpatient unit. Moreover, five months before her admission, she received the diagnosis of toxic multinodular goiter, and she was treated with antithyroid therapies (first with methimazole and then with propylthiouracil because of dermatological side effects) for two months; then, the patient decided to stop taking medications.
At hospital admission, her Montgomery Åsberg Depression Rating Scale (MADRS) (Montgomery and Asberg, 1979) score was 49, in the context of a clinical diagnosis of a severe major depressive episode; her vital signs showed tachycardia (heart rate 124 beats per min) and hypertension (blood pressure, 155/95 mmHg); her laboratory findings showed thyrotoxicosis [thyroid-stimulating hormone (TSH) = 0.01 µIU/ml, free thyroxine (FT4) = 23.5 pg/ml]. We started a psychopharmacological therapy with sertraline titrated up to 100 mg/day and olanzapine up to 20 mg/day, as well as an antithyroid therapy with propylthiouracil up to 150 mg/day.
In the following weeks, her depressive and psychotic symptoms remained stable despite pharmacological treatments. Moreover, she reported low tolerability to psychotropic medications and side effects such as nausea and psychomotor retardation. We hypothesized she was poorly compliant with olanzapine since we found olanzapine blood levels (25 µg/l) at the lower limit of the therapeutic range (20-80 µg/l); therefore, we switched to aripiprazole oral solution up to 30 mg/day. On day 15, laboratory findings showed an improvement in thyroid function (TSH = 0.01 µIU/ml, FT4 = 15.4 pg/ml). In the following week, depressive and psychotic symptoms gradually improved, and the patient presented much better insight. On day 24, MADRS score was 13, and the patient was discharged.

Discussion
In the presented clinical case, the severity of depressive symptoms closely resembled the levels of thyroid hormones. The patient, even with laboratory findings of thyrotoxicosis, did not show the typical psychiatric manifestations of this condition, which is usually characterized by hyperexcitability, but rather a severe major depressive episode with psychotic and melancholic features. This challenged the common clinical knowledge that depression is associated with clinical hypothyroidism (Williams et al., 2009). Indeed, medical guidelines by the American Association of Clinical Endocrinologists recommend that "the diagnosis of subclinical or clinical hypothyroidism must be considered in every patient with depression" (Goodman et al., 2011). However, we cannot make inferences about the improvement of symptoms in the reported patient as a direct consequence of the normalization of thyroid hormone levels, due to the concomitant coadministration of antithyroid and psychotropic drugs, including a serotonin reuptake inhibitor and second-generation antipsychotics, which may also have influenced the course of the disease. Moreover, the poor compliance with olanzapine made evaluations more difficult. It is also worth noting that the side effects reported by the patient, which in turn may have hampered adherence to treatment, may have been due to hypersensitivity induced by the excess thyroid hormones, according to the hypothesis of a possible interaction between thyroid and catecholamine receptors (Whybrow and Prange, 1981).
The relationship between thyroid hormones and emotional states has been known since ancient times, as alterations of thyroid function were recognized to result in disorders of mental function (Esposito et al., 1997). Previous studies investigated this relationship, but they yielded conflicting and unconclusive results. There is scarce evidence supporting the correlation between thyroid hormones concentration, severity, and duration of depressive symptomatology (Berent et al., 2014). In a multicentric cross-sectional study, Fugger et al. (2018) evaluated the co-occurrence of thyroid disease in patients with major depressive disorder (MDD), finding a prevalence of 13.3% for hypothyroidism and a prevalence of 1.6% for hyperthyroidism. Moreover, this study found that patients with MDD and comorbid hyperthyroidism were significantly older and showed more frequently any somatic comorbidity and heart disease than those without comorbid hyperthyroidism. A significant difference in terms of ethnic origin was reported, as MDD with comorbid hyperthyroidism was more frequent in patients of non-Caucasian origin. The co-occurrence of psychotic and melancholic features was more frequent in MDD comorbid with hypothyroidism. Patients having MDD with psychotic and melancholic features seem to have a hyperactivation of the hypothalamic-pituitary-adrenal axis, leading to cortisol hypersecretion. In turn, hypercortisolemia may induce dysregulation of the hypothalamic-pituitary-thyroid (HPT) axis by altering TSH secretion (Duval et al., 2006).
Patients with hyperthyroidism usually show physical manifestations of a hypermetabolic state and autonomic nervous system hyperactivation (Weinstein et al., 1963;Trzepacz et al., 1989). Common psychological symptoms are agitation, irritability, emotional lability, euphoria, or even hypomania (Gregory, 1956;Weinstein et al., 1963). In terms of physical symptoms, thyrotoxicosis is typically associated with weight loss, osteoporosis, atrial fibrillation, embolic events, muscle weakness, and tremor (Akamizu et al., 2012;Ross et al., 2016). However, back in 1931, in the New England Journal of Medicine, Lahey (1931) stated that there are two different types of hyperthyroidism: the so-called 'activated type' and the so-called 'nonactivated or apathetic type'. To our knowledge, in 1930, Hamburger and Lev (1930) reported the first cases of hyperthyroidism masked by depressive symptoms. Although rare, other similar cases were described in the literature, in which depressive symptoms, sometimes with psychotic features, obscured the usual signs of hyperthyroidism, and the psychiatric manifestations usually disappeared after thyroidectomy (McGee et al., 1959). These 'thyroid melancholics' are usually elderly, show weight loss, and do not show the usual signs and symptoms of hyperthyroidism (Thomas et al., 1970;Taylor, 1975). In order to provide a possible explanation, we hypothesized that the behavior of a hyperthyroid person might depend on an emphasization of his/her underlying habitual personality (Gregory, 1956).
Since the earliest reports, the strongest association between thyroid dysfunction and psychopathology has been with mood disorders. Several neuroendocrine mechanisms underlying this correlation have been proposed: thyrotropin-releasing hormone (TRH) acting as a neurotransmitter with antidepressant properties, interactions between TSH and 5-hydroxy tyramine activity, and alterations of the HPT axis (Williams et al., 2009). Some findings in patients with depression include a blunted TSH response to TRH stimulation and an increased prevalence of antithyroid antibodies compared with the general population. Blunting could be the consequence of the 'exhaustion' of a chronically activated HPT axis or a secondary effect of hypercortisolemia (Esposito et al., 1997;Wu et al., 2013). Transient hyperthyroxinemia could also account for acute blunting of the TRH-induced TSH response (Esposito et al., 1997). However, this relationship remains unclear, and recent studies attempted to provide more insights by using functional brain imaging.
Indeed, Schreckenberger et al. (2006) investigated the changes in regional cerebral metabolic activity measured by fluorodeoxyglucose positron emission tomography, a surrogate marker of neuronal activity, in patients with hyperthyroidism. This study also considered the correlations between thyroid hormones, emotional distress, and regional brain metabolism in the same sample. They evaluated 12 patients with untreated Graves' hyperthyroidism and compared them with euthyroid controls. Patients were assessed for thyroid laboratory values and psychometric tests evaluating anxiety and depression, such as the Hospital Anxiety and Depression Scale. Compared with euthyroid controls, patients with hyperthyroidism showed reduced cerebral glucose metabolism in the limbic system (hippocampus and functionally associated medial cingulate). The study also demonstrated activation in the posterior cingulate and in the inferior parietal lobe, which was correlated with both anxiety and depression scales. The finding of a close relationship between posterior cingulate metabolism and the levels of anxiety and depression in hyperthyroidism was in accordance with functional imaging studies, emphasizing the role of the posterior cingulate cortex for encoding emotional stimuli and memory-related functions. Metabolic alterations in these regions involved in affective regulation could also be related to higher concentrations of triiodothyronine receptors in limbic structures, as it has been shown by animal studies for the amygdala and the hippocampus (Ruel et al., 1985). For this reason, the hippocampus, a core structure of the limbic system critical to emotional processing and regulation of mood, seems to be highly sensitive to thyroid hormones concentration. Desouza et al. (2005) provided the first evidence that thyroid hormones regulate neurogenesis in the adult hippocampus, and subsequent studies demonstrated morphological changes in hippocampal neurons in adult hyperthyroid rats, including reduced dendritic branching and decreased spine density (Sala-Roca et al., 2008;Martí-Carbonell et al., 2012). A more recent neuroimaging study using voxel-based morphometry reported bilateral hippocampal atrophy in patients with hyperthyroidism and found that hippocampal gray matter volume was negatively correlated with both serum thyroid hormone levels and disease duration. This study speculates that excess thyroid hormones not only influences brain function but may lead to morphological abnormalities in mature brain structures critical to cognitive and emotional regulations (Zhang et al., 2014b). Yu et al. (2015) provided further evidence to support the hypothesis that thyroid hormones dysregulations can lead to symptoms of anxiety and depression. Their results suggested that thyroid hormones concentrations may have bidirectional effects on anxiety-and depression-like behaviors in rats, possibly by modulating hippocampal brain-derived neurotrophic factor (BDNF) levels: hypothyroidism led to decreased anxiety-and depression-like behaviors and brain serotonin (5-HT) levels, but to increased hippocampal BDNF levels, whereas the opposite was observed in the hyperthyroid rats. Hyperthyroid rats exhibited higher anxiety-and depression-like behaviors, higher brain 5-HT level, and lower hippocampal BDNF levels than controls.

Conclusion
This case report challenges the widely held belief that depression can only be associated with hypothyroid but not hyperthyroid states. In addition, we offer some insights into possible neurobiological and neuroimaging correlates that may explain what we clinically observed in the patient under consideration. The case here reported further draws attention to the need for careful thyroid screening in patients experiencing depression, which could allow early diagnosis of thyrotoxicosis. The presence of an excess of thyroid hormones in patients with depression could increase sensitivity to anticholinergic and noradrenergic effects of some psychotropic medications, reducing drug tolerability (Whybrow and Prange, 1981). We suggest that depression related to a hyperthyroid state may be responsive to antithyroid treatment, alone or in augmentation with antidepressants and/or second-generation antipsychotics, but further studies will be needed to clarify the efficacy and tolerability of such treatment strategies.