The neurobehavioral and neuropsychological changes associated with thyrotoxicosis are multiple and varied (5,6,7,8). Patients complain of anxiety and dysphoria, emotional lability, insomnia, and at times, intellectual dysfunction. Their ability to concentrate is particularly impaired; indeed, this may be the earliest disturbance, and it is often associated with increasing restlessness and tremulousness. Patients appear irritable, jittery, and easily moved to anger; some express ideas of reference and frank paranoia. Thoughts and words can come rapidly and are disjointed at times, suggesting a thought disorder.

Motor activity is increased, usually associated with agitation. Although this may mimic manic behavior, the fully developed psychiatric syndrome of mania is surprisingly rare in patients with thyrotoxicosis. Sleep disturbances including vivid dreams and nightmares are common, and daytime energy levels are often decreased. These features are an important distinction from mania, in which increased energy, irritability, and decreased sleep are the common presenting problems. When true mania and hypomania occur in patients with thyrotoxicosis (7,8,9,10), the patients typically have a previous diagnosis of bipolar disorder or a strong family history of that disorder. Episodic anxiety, frequently in association with subjective
awareness of tachycardia or arrhythmia, is also a common symptom in thyrotoxic patients. Indeed, some of these diffuse dysphoric feelings have been reported in normal subjects given high doses of levothyroxine (L-T4) (11,12). In rare patients, the behavioral dysfunction may progress to a nonspecific psychotic illness with bizarre delusional thoughts, often of a paranoid nature (7,11). With the onset of thyrotoxic storm (also rare), delirium, restlessness, and agitation can appear acutely (13).

In contrast to this picture is the mental state of patients with so-called apathetic thyrotoxicosis (14), an uncommon form of presentation of thyrotoxicosis, that mimics a depressive disorder and that usually occurs in elderly patients, although not exclusively so (7,15). This syndrome is characterized by apathy, lethargy, pseudodementia, weight loss, and depressed mood; the patients may be initially diagnosed as having melancholia. The true diagnosis can be overlooked, because many of the common manifestations of thyrotoxicosis in young people, such as tachycardia, hyperphagia, increased perspiration, warm skin, and goiter, are lacking (16). Thus, special vigilance for this type of thyrotoxic syndrome should be maintained. Apathetic thyrotoxicosis also has been described in young adults (17,18).

Because of the considerable overlap between mental and physical complaints, such as loss of energy and tremulousness, the true incidence of neuropsychiatric symptoms and mental disorders in patients with thyrotoxicosis is difficult to estimate. Smaller studies revealed higher scores of anxiety and depression in patients with Graves’ thyrotoxicosis compared with healthy controls (19,20). A questionnaire study of neuropsychiatric complaints in 137 patients with thyrotoxicosis caused by Graves’ disease confirmed that psychiatric symptoms, especially anxiety and irritability, were common (21). In two studies of patients with thyrotoxicosis using modern diagnostic criteria for psychiatric disorders, the prevalence rates for depressive disorders were 31% (22) and 69% (23), and the rates for anxiety disorders were 62% (22) and 61% (23). In contrast, in a recent Norwegian study in an unselected population of more than 30,000 individuals, there was no statistical association between thyrotoxicosis and self-reported levels of depression and anxiety (24) although the psychiatric measure utilized was relatively insensitive for subtle symptoms. However, the subgroup of individuals with former known thyroid disease had an increased risk of both anxiety and depression (24).

Several groups of investigators have sought to quantify the cognitive changes in patients with thyrotoxicosis, by comparing patients with matched normal subjects and by studying patients during their illness and after recovery (6,22,23,25,26). A thorough psychological study of Turkish women with thyrotoxicosis revealed multiple abnormalities (25). Although the range of educational background of the study and control groups was broad, the groups were well matched for this variable and for age and socioeconomic status. Of the 23 women with thyrotoxicosis, 10 were retested after treatment. Before treatment, their responses to visual stimuli were slow, but they were normal after treatment. The auditory reaction times were also slow, but did not return to normal after treatment. Their visuomotor coordination was less accurate, less steady, slower, and more readily fatigued, as compared with that of the normal subjects, and they made more mistakes on mirror-drawing tests (25). This impairment of cognition, directly correlated with the thyrotoxic state, also was found in two subsequent studies. In one study, thyrotoxic patients performed poorly on a nonverbal test of intelligence and on a neuropsychological test of visual attention and task switching (6). In the second study, performance on those tasks that require concentration and memory was impaired in proportion to the degree of increase in their serum thyroxine (T₄) concentrations (27).

Subclinical thyrotoxicosis, defined as a subnormal TSH with normal FT4 and T3 levels, occurs in 1% to 2% of the population. It is unclear whether this is associated with affective or cognitive impairment. Some smaller studies suggest increased rates of anxiety and irritability, decreased vitality, depressive symptoms, or mild decrements in cognition (19,28,29). These findings are qualitatively similar but quantitatively less severe than those in overt thyrotoxicosis, consistent with the idea that subclinical thyrotoxicosis represents a mild degree of thyrotoxicosis. However, two large studies have not confirmed these findings in unselected populations (24,30). From a clinical standpoint, major abnormalities in mood or cognitive function in patients with subclinical thyrotoxicosis should be evaluated and treated as separate disorders.

In most patients, the diagnosis of thyrotoxicosis presents little difficulty. Problems may arise, however, in older patients, as noted above, who are apathetic and in whom behavioral and psychological symptoms and signs (i.e., anxiety, agitation, lowered attentiveness, irritability, depressed mood with poorly defined ideas of persecution, insomnia, or depression) are prominent. They may be diagnosed as having an agitated major depression or involutional paranoid psychosis, if other manifestations of thyrotoxicosis are overlooked. The most frequent misdiagnosis is that of anxiety disorder. The differentiation between thyrotoxicosis and an anxiety state may not be easy, especially in the early stages of either disorder (31). Unlike thyrotoxicosis, which is usually progressive, the intensity of anxiety states tends to vary over time. There also may be a history of other adjustment difficulties, and the anxiety state itself may be associated with specific fears of objects or situations.

Panic attacks may present a particular difficulty in differential diagnosis, especially because of the rapid heart rate and palpitations that accompany them. Since a biochemical diagnosis of (overt) thyrotoxicosis is easily obtained by measurement of serum TSH levels (decreased) and of levels of FT4 and FT3, or both (increased), all patients with these psychiatric syndromes should have these measures of thyroid function
checked. Patients who have panic attacks often awake in the middle of the night with these symptoms. Their resting pulse rates, however, usually are not high, unlike in thyrotoxicosis; the hands and feet are usually cold and clammy, whereas they are warm and moist in patients with thyrotoxicosis. Also, although both anxious and thyrotoxic patients may have difficulty sleeping and may eat more than usual, the former usually do not lose weight, nor do they have the progressive disturbances of memory, calculation, and problem-solving ability that characterize thyrotoxicosis.

The relationship of thyrotoxicosis to mania is complex. Motor acceleration, pressure of speech, and disorganization of thought content are found in patients with thyrotoxicosis. However, the constellation of symptoms that is necessary to fulfill the criteria for secondary mania (defined as mania occurring as a result of a concurrent medical condition like thyrotoxicosis, 32) is far less common (33). Secondary mania can be differentiated from primary mania by a later age of onset in patients with no family history of mania or affective disorders. Initiation of high doses of thyroid hormone therapy in hypothyroid patients can precipitate classic (primary) mania (34), especially in patients with a family history of affective illness. This possibility should be considered in a patient who exhibits bizarre or hyperactive behavior while receiving thyroid hormone.

Thyrotoxicosis factitia is the term used to describe thyrotoxicosis caused by surreptitious ingestion of high doses of thyroid hormones (35). Patients attempting suicide by taking L-T₄ or other thyroid hormone preparations usually suffer from psychiatric illness and should be referred for psychiatric consultation. The disorders that may be associated with thyrotoxicosis factitia include Munchausen’s syndrome (a disorder characterized by the intentional production or feigning of physical or psychological symptoms and signs), and neurotic disorders in patients with poor body image and concerns regarding their weight and sexual identity (35,36).

The role of psychosocial strain and trauma in the pathogenesis of Graves’ disease, the most common cause of thyrotoxicosis, has been the subject of considerable debate. Although anecdotal reports and a considerable body of clinical opinion seem to support an association, objective evidence remains elusive. One problem in assessing any temporal association between the two is determining the precise onset of thyrotoxicosis. The speed of onset of symptoms is variable, and the thyrotoxicosis is probably subclinical for weeks or months before symptoms appear, so that it may already have been present at the time of the supposed precipitating event. Similarly, the psychological reaction to the event may be a reflection of already present thyrotoxicosis, rather than causing it. Information distinguishing these points is virtually impossible to obtain by retrospective study.

Prospective studies would undoubtedly offer better information, but are difficult to conduct. However, in one such study, subjects from the general population who had hyperfunctioning regions on thyroid scintiscans were followed carefully with independent psychological and thyroid evaluations. Among 239 women followed for 12 years, the hyperfunctioning regions appeared to wax and wane in a direct relationship with life stress, and some women developed clinical thyrotoxicosis during conditions of severe or prolonged life strain (37). Two studies explored the onset of Graves’ thyrotoxicosis using a controlled retrospective methodology. In a case-control study, patients developing Graves’ thyrotoxicosis reported more negative life events, such as divorce, bereavement, and educational and occupational failure, than did control subjects (38). Similar results were obtained in a consecutive sample of 70 patients matched for age and sex with normal subjects (39). In this study the patients had greater life change, both positive and negative, in the year preceding the diagnosis of Graves’ thyrotoxicosis than did the control subjects; however, raters unaware of the subjects’ study group judged only negative life events to be significantly greater in the patients than in the control subjects. In summary, psychological stress may be associated with the onset of symptoms of thyrotoxicosis and may influence its clinical course. Assessment of psychological stress and psychotherapeutic interventions (e.g., behavioral therapy) should be considered. In a study of patients with Graves’ disease who were followed-up for 2 to 5 years of antithyroid drug therapy the impact of major life events led to aggravation of the autoimmune disease (40). Behavioral interventions that include coping strategies to handle stressful life events may be useful in improving prognosis in Graves’ disease.