Abstract
Background
Mood disorders are common in Graves’ disease despite treatment. The pathogenic mechanisms involved are unknown and so is whether previous psychiatric disease influences these symptoms.
Methods
This is a longitudinal study conducted in Sweden on 65 women with newly diagnosed Graves’ disease and 65 matched controls. Participants were examined during hyperthyroidism and after 15 months of treatment. Examinations included blood sampling, and psychiatric testing with the Comprehensive Psychopathological Rating Scale for Affective Syndromes and the Structured Clinical Interview for DSM-IV – Axis I Disorders. We also performed two analyses of a national population-based registry to determine previous psychiatric diagnoses and previous prescriptions of psychoactive drugs in (i) all patients we asked to participate and (ii) all Swedish women given a diagnosis of hyperthyroidism during 2013–2018, comparing them to matched controls.
Results
There was no increased previous psychiatric comorbidity in Graves’ patients compared to controls. There was no higher prevalence of psychiatric diagnoses and prescriptions of psychoactive drugs between (i) included GD patients compared to those who declined participation and (ii) women with a hyperthyroidism diagnosis in 5 years prior to their diagnosis, compared to matched controls. Depression scores and anxiety scores were higher in patients compared to controls both during hyperthyroidism (depression (median (IQR): 7.5 (5.0–9.5) vs 1.0 (0.5–2.5) P < 0.001), anxiety: 7.7 (5.0–11) vs 2.5 (1.0–4.0) P < 0.001) and after treatment (depression: 2.5 (1.5–5.0) vs 1.5 (0.5-3.5) P < 0.05), anxiety: 4.0 (2.5–7.5) vs 3.0 (1.5-5.0) P < 0.05). Patients with a previous psychiatric condition, mild eye symptoms, and a younger age had more anxiety at 15 months compared to patients without these symptoms and a higher age (all p<0.05).
Conclusion
Graves’ disease affects patients’ mood despite treatment. A previous psychiatric condition, mild eye symptoms, and a younger age increase the vulnerability for long-lasting symptoms and require specific attention.
Introduction
Mental symptoms and reduced quality of life (QoL) are common brain-derived problems in untreated Graves’ hyperthyroidism (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14). In a 2006 review (meta-analysis) the most prevalent mental symptoms included anxiety, fatigue, emotional lability, and reduced emotional well-being (1). These symptoms are often ameliorated by the restoration of thyroid hormone levels to normal by treatment, but a considerable number of patients suffer from long-standing impaired health (3, 7, 15, 16, 17, 18, 19). The mechanism behind these complications remains to be elucidated, as is the pathogenesis underlying them. Hypothetically, the increased prevalence of mental symptoms may be a consequence of the elevated thyroid hormones, the autoimmune reaction, the circumstance that persons with psychiatric conditions may be more prone to develop Graves’ disease (GD), the development of complications like thyroid-associated ophthalmopathy (TAO), or a general stress-induced psychological reaction to disease occurrence. All the abovementioned explanations have been suggested in previous studies (18, 20, 21, 22). Mechanistically, a reduction in medial temporal lobe brain areas in hyperthyroid GD patients has also been observed (18, 23), but its recovery in euthyroidism was not associated with alleviated symptomatology (18).
A few studies in GD subjects have found associations between depression, anxiety, and hormonal levels (9, 23), while others have not (6, 10). An epidemiological study investigating the relationship between thyroid hormone levels and the prevalence of depression and anxiety in the general population found no increased risk of depression or anxiety in hyperthyroid individuals (24). Thus, the evidence for an association between thyroid hormone levels and mental symptoms in GD is inconclusive.
Supporting thyroid-related antibodies or autoimmunity per se as a causative factor for affective symptoms is the fact that depression and anxiety appear more prevalent in autoimmune GD than in toxic nodular goiter (TNG) (4). The presence of TAO has also been described to exacerbate psychiatric symptoms in GD (20, 25).
The influence of previous psychiatric conditions in GD was investigated in a Danish registry study, where the risk of having been hospitalized with a psychiatric disease and treated with antidepressants or anxiolytics before the onset of GD was observed to be increased in persons with GD compared to the general population (21). A common cause therefore cannot be excluded. More specifically, bipolar disease has been reported to be more prevalent in women who later develop GD compared to those who do not (26). In addition, existing psychiatric symptoms often worsen during the phase of hyperthyroidism (27). It is therefore important to assess the relevance of preexisting psychiatric morbidity for prognosis. Still, it is unknown whether latent psychiatric diseases become apparent during the hyperthyroidism and are responsible for the remaining mental symptoms.
In the CogThy study, we hypothesized that it is GD per sethat increases the prevalence of psychiatric symptoms in hyperthyroidism and long after euthyroidism has been attained. The aim was to evaluate the influence of previous psychiatric conditions, thyroid hormone levels, antibody levels, and eye symptoms on the prevalence of depression and anxiety in the hyperthyroid phase and after 15 months of treatment.
Methods and subjects
Design of the CogThy study
The study design of the CogThy study has been described previously (28). Briefly the study was designed as a case-controlled prospective longitudinal observational study with an open treatment period for hyperthyroidism of at least 15 months. The study included 65 women with newly diagnosed GD and 65 matched controls. Both groups underwent assessment at the time of inclusion and after 15 months. Treatment of the hyperthyroidism was in accordance with the local treatment scheme for premenopausal female adult patients with GD where all patients received either antithyroid drugs (ATD) or surgery with previous ATD treatment. As an exception, one patient was treated with radioactive iodine. Hypothyroidism was prevented by early commencement of thyroxine treatment.
The study visits at inclusion included somatic evaluations, laboratory hormonal and antibody assessments, neuropsychological and psychiatric evaluation, and self-administered QoL questionnaires. This was repeated after 15 months. For full details see Holmberg et al. (28).
Patients
From September 2011 to March 2018, 65 patients with newly diagnosed GD prospectively were recruited from the Thyroid Unit at Sahlgrenska University Hospital, Göteborg (n = 64) and from the Department of Medicine at Kungälv’s Hospital, Kungälv (n = 1) in Sweden. They were eligible for the study if they were premenopausal, with fT4 levels ≥50 pmol/L (reference range: 12–22) and/or total T3 ≥6.0 nmol/L (reference range: 1.3–3.1) in combination with positive thyroid-stimulating hormone receptor antibodies (TRAbs) and/or a technetium scintigraphy with a diffuse uptake. Exclusion criteria were pregnancy, serious somatic disease such as other endocrine disease, heart failure, respiratory failure, active malignancy that might interfere with QoL, psychosis, or inability to follow the study protocol for other reasons. Neither were the patients eligible if they had TAO with steroid treatment, nor if glucocorticoid treatment was expected within the next 15 months. Moreover, previous or present glucocorticoid treatment because of other illness, contraindications for magnetic resonance imaging, and GD caused by amiodarone treatment were exclusion criteria.
Control subjects with matching age and sex were selected from a random sample from the general population in Göteborg provided by the Swedish Tax Registry. When a control subject consented for participation, she was also matched regarding smoking status and educational level. In addition to the exclusion criteria stated for patients, controls were excluded if they had thyroid hormone levels outside the normal reference range, previous or ongoing thyroid disease, or TAO.
We approached 116 patients for study participation and 56% were included. Reasons for not participating were exclusion criteria not previously known (n= 5), not wanting to participate (n= 14), not having the time (n= 8), difficulties to plan first visit within 2 weeks from start of ATD treatment (n= 3), no reason given or not possible to get in contact with after first information (n= 21). Of the 65 patients in the CogThy study, only 64 were included in the present analysis due to lack of data from one patient who was excluded immediately following inclusion. Patient and control characteristics are presented in Table 1.
Demographic and clinical characteristics of 64 patients with Graves’ disease and 64 controls matched for age, smoking status, and educational level at inclusion. Data are presented as median (IQR) or as n (%).
| Patients | Controls | P | |
|---|---|---|---|
| n | 64 | 64 | |
| Age (years) | 30.5 (24.5–38.0) | 31.5 (26.5–39.0) | 0.68a |
| BMI (kg/m2) | 22.2 (19.3–25.0) | 23.5 (21.3–25.8) | 0.02a |
| Previous smoker | 13 (20) | 14 (22) | 0.83b |
| Current smoker | 12 (19) | 8 (12) | 0.33b |
| Previous affective psychiatric condition identified by SCID-I | 34 (53) | 35 (55) | 0.86b |
aMann–Whitney U-test; bχ2 test.
BMI, body mass index; IQR, interquartile range; SCID-I, Structured Clinical Interview for DSM-IV – Axis I Disorders.
Clinical assessment of background variables
Information was collected about ethnicity, socioeconomic status, smoking, snuff, alcohol, stress, pregnancy, heredity, previous low-dose irradiation to the head and neck region, other previous and current diseases, and ongoing medications.
Somatic and eye assessments
Clinical visit included assessments of physical status, body weight, and height and electrocardiography. Eyes were assessed with Clinical Activity Score (CAS) (29).
Biochemical methods
Blood specimens were analyzed at the Department of Clinical Chemistry at Sahlgrenska University Hospital, Göteborg, Sweden, for serum thyroxine (T4), free thyroxine (fT4), triiodothyronine (T3), free T3 (fT3), thyroid-stimulating hormone (TSH), and thyroid peroxidase antibodies (TPOAb) for clinical purposes by electrochemiluminescence immunoassay (Roche Elecsys®ECL, Roche Diagnostics International AG). Total TRAbs were analyzed using radioreceptor assay analysis with Brahms Kryptor (Thermo Fisher Scientific). Thyroid Stimulating Immunoglobulins (TSI) were analyzed with Siemens IMMULITE 2000/2000 XPi TSI assay (Siemens Healthcare GmbH) (30). For reference ranges see Table 1.
Psychiatric assessments
The Swedish version of the Structured Clinical Interview for DSM-IV – Axis I Disorders (SCID-I) was used to diagnose previous and present psychiatric disorders (31). SCID-I is a semistructured interview based on open questions. Regardless of whether the interviewee has ever been in contact with health care or has previously been diagnosed with a psychiatric disease, the interview gathers the participant’s symptoms that may or may not fulfill the diagnostic criteria of a DSM-IV axis disorder. The SCID-I interview is extensively used to capture previous and present psychiatric disorders (31). All interviews were performed by either of two psychiatrists experienced in both psychiatric diagnostics and in SCID. The Swedish version translated in 1998 was used.
Depressive and anxiety symptoms the previous week were assessed with the Comprehensive Psychopathological Rating Scale for Affective Syndromes (CPRS-S-A) (32). CPRS-S-A is a self-rating scale developed in Sweden but widely used internationally. The scale was constructed in 1994 as a tool to evaluate symptomatic change in psychiatric disorders. It consists of 19 questions covering depression, anxiety, and obsessive-compulsive disorder. We used the subscales depression and anxiety. A higher score refers to more severe symptoms.
Design of registry search
As all individuals in Sweden have a unique personal identification number and all visits in health care are coded by a diagnosis, the outpatient registry at the Swedish National Board of Health was investigated for the prevalence of psychiatric diagnoses and prescriptions of psychoactive drugs. The following diagnoses were used: harmful use of alcohol or drugs (F100-199), affective disorders (F300-399), and phobias and anxiety (F400-489), and the following drugs were considered: psycholeptics (ACT-code N05), psychoanaleptics (N06), or other nervous system drugs (N07). The search was based on the presence or absence of a diagnosis and/or a prescription during the 5 years preceding the date of inclusion, the date of request for participation, or the date of diagnosis dependent on the group investigated. Thus, several diagnoses or prescriptions during the 5 investigated years did not render higher score than one diagnosis or prescription.
The following groups were investigated: (i) participants compared to those who were asked for participation but denied inclusion and (ii) all females with a first diagnosis of thyrotoxicosis (E05) in Sweden in the years 2013–2018 compared to an age-matched female cohort without any E05 diagnosis, in the same years. For every patient there were ten controls. The rationale for the first search was to explore any selection bias and for the second to investigate if the prevalence of psychiatric diagnoses and use of psychoactive medications in hyperthyroid patients prior to the diagnosis differed from that of matched thyroid healthy controls.
The search was performed by a statistician at the National Board of Health and all results were delivered in an unidentifiable form separated by groups.
Ethical approval
Ethical approval was granted by the Regional Ethical Review Board in Göteborg, Sweden (Ref. no. 190-10 approved May 21, 2010, with additions T912-11 approved November 15, 2011, T854-12 approved November 5, 2012, T955-14 approved February 26, 2015, and T249-16 approved March 15, 2016). The study was conducted according to the Declaration of Helsinki. All participants provided written informed consent.
Statistical analysis
For the statistical analysis, fT3, fT4, TRAb, and TSI levels exceeding the upper detection limit were set as being at this limit, while TRAb, TSI, TPOAb and TSH levels below the lower detection limit were set as being at that limit. When fewer patients than 64 were included in a subanalysis, the included number is noted. Groups were compared with the Mann–Whitney U-test. Prevalence between groups were compared with the chi-square test. Longitudinal comparisons were made with Wilcoxon matched-pairs test. Significance tests were two tailed, and the significance level was set at P = 0.05. No corrections for multiple comparisons were made. All statistical calculations were made using TIBCO Statistica 13.2 (TIBCO Software Inc, Palo Alto, CA, USA).
Results
The patients were assessed at median 8 days after starting ATD treatment. The median age at inclusion was 31 years and the median duration of symptoms was 4 months. Mild eye symptoms were noted in 53% of the patients. The only observed difference between the patients and the controls was that the patients had a lower median weight, as expected in hyperthyroidism. None of the patients had moderately severe/severe TAO. Baseline data characteristics of patients and controls and disease characteristics of the 64 included patients are presented in Tables 1 and 2.
Disease characteristics of 64 patients with newly diagnosed Graves’ disease at inclusion in study within 1–2 weeks from diagnosis when still hyperthyroid. Data are presented as median (IQR) or as n (%).
| Characteristics | Values |
|---|---|
| Age (years) | 31 (25.0–38.0) |
| BMI (kg/m2) | 22.4 (19.4–24.8) |
| Time from blood test before diagnosis to inclusion (days) | 13 (11–20) |
| Time from start of ATDs to blood test at inclusion (days) | 8 (6–12) |
| Duration of symptoms before diagnosis (months) | 4 (3.0–6.5)b |
| Previous smokers | 41 (64) |
| Smokers | 8 (12) |
| Patients treated with beta blockersa | 48 (77)b |
| Patients with ATDs | 64 (100) |
| Propylthiouracil | 2 (3) |
| Thiamazole | 62 (97) |
| Clinical Activity Score | |
| ≥1 point | 34 (53) |
| >3 points | 64 |
aPropranolol or metoprolol; bData available for 62 patients.
ATD, antithyroid drugs; BMI, body mass index.
During hyperthyroidism
At the time of diagnosis, 90.0% (58/63) of the GD patients presented with fT4 ≥50 pmol/L and total T3 was ≥6.0 nmol/L in 56.4% (31/55). In 98.4% (63/64) of the patients elevated TRAbs were observed. The patient with normal TRAbs had an increased, diffuse uptake on technetium scintigraphy.
Thyroid treatment
All patients were treated with ATD, given according to clinical routine in block-and-replacement fashion or surgery following ATD pretreatment. After 15 months, 20/55 (36.4%) had undergone thyroidectomy, one (1.8%) patient was treated with radioactive iodine, 26/55 (47.3%) were on ATD, and 8/55 (16.6%) had no treatment and were in remission.
Biochemistry
The levels of fT3, T3, fT4, T4, TRAbs, TSIs, TPOAbs, and TSH, for patients at the time of inclusion and at follow-ups are presented in Table 3. At follow-up, thyroid hormones had returned to normal levels for most patients, but median fT4 remained higher and fT3 lower in patients compared to controls. At 15 months, TSH levels were below normal in 9/57 (15.8%), above normal in 5/57 (8.8%), and within the normal range in 43/57 (79.2%). TRAb levels improved with treatment but did not yet return to normal in 21/54 (38.9%) of patients.
Disease characteristics of 64 patients with newly diagnosed Graves’ disease at inclusion in study within 1–2 weeks from diagnosis when still hyperthyroid.
| RR | At inclusiona | At 15 months | P | |||
|---|---|---|---|---|---|---|
| n | Median (IQR) | n | Median (IQR) | |||
| S-fT3 (pmol/L) | 3.1–6.8 | 63 | 13 (9.5–19.0) | 55 | 4.3 (3.8–4.8) | 1.1E-10 |
| S-T3 (nmol/L) | 1.3–3.1 | 64 | 3.8 (3.1–5.0) | 56 | 1.6 (1.4–1.8) | 8.0E-11 |
| S-fT4 (pmol/L) | 12–22 | 64 | 38.5 (29.0–49.0) | 57 | 18 (15.0–20.0) | 1.3E-10 |
| S-T4 (nmol/L) | 66–181 | 63 | 170 (140.0–200.0) | 55 | 110 (93–120) | 6.7E-09 |
| S-TRAb (IU/L) | <1.8 | 64 | 10.1 (5.5–23.6) | 54 | 0.8 (0.8-2.17) | 3.6E-09 |
| S-TSI (IU/L) | <0.10 | 62 | 7.6 (4.2–15.9) | 53 | 0.6 (0.3–1) | 1.6E-09 |
| S-TPOAb (kU/L)a | <34 | 51 | 123 (24.0–480.0) | 53 | 12 (6.0–38.0) | 1.4E-07 |
| S-TSH (mlU/L) | 0.30–4.2 | 64 | 0.01 (0.01–0.01)b | 57 | 1.3 (0.6–2.5) | 1.6E-10 |
fT3, free triiodothyronine; fT4, free thyroxine; T3, triiodothyronine; T4, thyroxine; TRAb, TSH receptor antibodies; TPOAb, thyroid peroxidase antibodies; TSI, thyroid-stimulating immunoglobulins; TSH, thyroid-stimulating hormone; RR, reference range.
Psychiatry
Structured Clinical Interview for DSM-IV
At the time of inclusion, 64 patients and 64 matched controls were interviewed according to the SCID protocol. Of these participants, 33/64 (52%) patients and 35/64 (55%) of the controls were estimated to having fulfilled a diagnostic criterion for depression, anxiety, and/or bipolar disorder at least once preceding the GD diagnosis or the control visit (Table 1). In the patients, 15/64 (23%) had previous depression and 28/64 (44%) previous anxiety. Of these, 10/33 (30%) had both depression and anxiety. In the controls 14/64 (22%) had previous depression and 30/64 (47%) previous anxiety. Of these, 9/35 (26%) had both depression and anxiety. There were eight (12%) patients and one (2%) control with bipolar disorder. There were no significant differences in previous depression or anxiety between the patients and controls. Previous bipolar disorder was more common among patients than controls (χ 2 test, P < 0.05).
Results from the registry analyses
In the 5-year period preceding the request for participation in the study, a psychiatric diagnosis was given to 2/64 (3.1%) of the included GD patients and to 1/51 (2.0%) of the invited but not included GD patients (Fig. 1A). During the same time period psychoactive medications were prescribed to 7/64 (10.9%) of the included GD patients and to 3/51 (5.9%) of the invited but not included GD patients (Fig. 1B). None of these differences were significant.
(A) Proportion of patients with Graves’ disease (GD) and a previous psychiatric diagnosis, according to the Swedish diagnosis registry, in the 5 years prior to the GD diagnosis presented for the included GD patients and in the invited but not included GD patients. (B) Proportion of Graves’ disease (GD) patients with at least one prescription of a psychoactive substance, according to the Swedish diagnosis registry, in the 5 years prior to the GD diagnosis in the included GD patients and in the invited but not included GD patients.
Citation: European Thyroid Journal 13, 1; 10.1530/ETJ-23-0247
During the years 2013–2018 a total of 14435 women were given the diagnosis thyrotoxicosis (E05) at least once. A majority of these were likely GD cases due to the higher incidence of GD in Sweden compared to other causes of hyperthyroidism like toxic nodular goiter (33, 34). In this group a total of 1204 women were given a psychiatric diagnosis at any time during the 5 years preceding the E05 diagnosis. The proportion of patients with at least one psychiatric diagnosis was 1204/14,435 (8.3%) in the thyrotoxicosis patient group and 12,953/144,350 (9.0%) in the matched controls (Fig. 2A).
(A) Proportion of patients and controls with at least one psychiatric diagnosis in the 5 years preceding the diagnosis of thyrotoxicosis (E05). Results presented in all Swedish female patients with a thyrotoxicosis diagnosis (E05) in the years 2013–2018, and age- and sex-matched controls (1:10) without thyrotoxicosis during the same years. (B) Proportion of patients with at least one prescription of a psychoactive substance in the 5 years preceding the diagnosis of thyrotoxicosis (E05). Results presented in all Swedish female patients with a thyrotoxicosis diagnosis in the years 2013–2018, and age- and sex-matched controls (1:10) without thyrotoxicosis.
Citation: European Thyroid Journal 13, 1; 10.1530/ETJ-23-0247
In the 5 years prior to the E05 diagnosis, psychoactive medications were prescribed to 5448/14,435 (37.7%) thyrotoxicosis patients and to 58,119/144,350 (40.3%) of the controls (Fig. 2B). It was significantly less common to have been diagnosed with a psychiatric diagnosis or prescribed a psychoactive medication in the thyrotoxicosis patients compared to the controls (Fig. 2A and B).
Comprehensive Psychopathological Rating Scale
CPRS was completed by 64 patients and controls at inclusion in hyperthyroidism and by 53 patients and 39 controls at 15 months in euthyroidism. At inclusion, hyperthyroid patients reported more depression and anxiety than controls (Fig. 3A and B). After 15 months, patients had improved but still reported more symptoms of depression and anxiety compared to controls (Fig. 3A and B). There was no significant difference in the controls between inclusion and after 15 months.
(A and B) Results in CPRS depression (A) and anxiety (B) scoring in 64 premenopausal women with newly developed Graves’ disease at inclusion when still hyperthyroid and after 15 months of treatment in euthyroidism. Controls, matched for age, sex, smoking, and education, were investigated at the same time interval. Results presented as median (IQR) and comparisons done with Mann–Whitney U-test. The line in the box plot represents the median and the X the mean.
Citation: European Thyroid Journal 13, 1; 10.1530/ETJ-23-0247
With hyperthyroidism, patients with previous psychiatric conditions (according to SCID) reported more depression (median (IQR): 8.5 (6.5–11.5) vs 6.0 (4.0–8.5) P < 0.01) but not anxiety 8.0 (6.0–12.5) vs 6.5 (4.5–9.5) P = 0.056 than patients without such history. There was no age difference between those reporting a previous psychiatric condition and those that did not (data not shown). At 15 months, anxiety score but not depression score was higher in patients with a previous psychiatric condition (anxiety scores (median (IQR): 4.3 (3.0–9.0) vs 3.0 (2.5–4.5) P < 0.05, depression scores: 2.8 (1.5–6.5) vs 2.0 (1.5–3.0) P = 0.073) compared to patients without a previous psychiatric condition.
There was also a significant difference in both CPRS depression and anxiety scores between patients who reported eye symptoms (CAS 1–3) at baseline and those who did not: CPRS depression (median: 8.5, IQR: 6.0–10.5 vs median: 6.0, IQR: 4.0–8.0; P < 0.01) and CPRS anxiety (median: 9.0, IQR: 7.0–2.5 vs median: 5.8, IQR: 3.5–8.0; P < 0.001). At 15 months patients who reported eye symptoms at baseline scored higher in CPRS anxiety compared to those who did not report any eye symptoms (median: 4.3, IQR: 3.0–7.5 vsmedian: 2.8, IQR: 1.5–4.5; P < 0.05).
At inclusion, the 18 patients who had neither a previous psychiatric condition nor eye symptoms had higher CPRS depression (median: 5.0, IQR: 3.0–6.5 vs median: 1.0, IQR: 0.0–2.5; P < 0.001) and CPRS anxiety score (median: 5.0, IQR: 4.0–6.5 vs median: 2.0, IQR: 1.0–3.5; P < 0.001) than the controls. This difference did not remain at 15 months.
There was an inverse correlation at inclusion between the patients’ age and CPRS anxiety score (ρ = −0.26, P < 0.05) but not with the CPRS depression score. No significant correlations between CRPS scores and patients’ BMI or smoking status were noted (data not shown).
No correlations were found between the patients’ levels of fT3, fT4, TSH, TPOAbs, TRAbs, or TSIs and their CPRS scores, either at inclusion or at follow-up. There was no difference in any of the thyroid hormone or antibody levels between the patients with a previous psychiatric condition and those without. There was no difference in CPRS scoring at 15 months between those with elevated TRAb and those with normalized TRAb (data not shown).
Discussion
In this prospective study in 64 women with newly developed Graves’ disease, psychiatric symptoms were more common in patients than controls both in the hyperthyroid phase and after 15 months of treatment. We found no signs of a higher prevalence of a previous psychiatric diagnosis in included patients compared to controls or to patients eligible for the study, but not included. Nevertheless, having mild eye symptoms, a history of psychiatric problems, and a younger age aggravate psychiatric symptoms both early in the course and after treatment.
In GD patients, a higher risk of being hospitalized with a psychiatric condition compared to the general population has been reported in a Danish study, also before the onset of GD (21). Although hospitalization, and thereby more severe psychiatric illness, was not investigated in our study, we could not confirm a higher prevalence of psychiatric diagnoses in hyperthyroid female patients compared to matched thyroid healthy controls, neither in our cohort of women newly diagnosed with GD nor in our registry search on females 5 years prior to a hyperthyroidism diagnosis. On the contrary, in the registry search, we found a lower prevalence of psychiatric diagnoses and use of psychoactive substances in the patients with a subsequent hyperthyroidism diagnosis compared to a control group. The difference was small but significant and we find it unlikely that the inclusion of all causes of hyperthyroidism would have such a great impact on our result since hyperthyroidism in Sweden is dominated by GD. There is no obvious explanation for this interesting finding. A woman that contacts the health care with symptoms of depression or anxiety will most likely be tested for thyroid disease. So missed thyroid conditions in the control group are not a plausible cause to the finding.
The conflicting results compared to the previously mentioned Danish study may depend on several factors. The Danish study included severe psychiatric conditions that often result in hospitalization, such as psychosis (F20–25, F28–29), while the Swedish registry search included substance use (F100–199), obsessive-compulsive disorders, stress-related disorders, adjustment disorders, dissociative disorders, and more (F42–49). The Danish study included men, and it is well known that men have a higher prevalence of psychotic conditions compared to women but since the study used matched controls this is still an unlikely explanation. The Danish cohort was more than 30 years older and was followed a longer time. In general, Denmark also has a higher thyroid multinodular goiter (TMNG)/GD ratio than Sweden, due to a later start of an iodinization program.
The SCID interview indicated a previous psychiatric condition in about half of the patients as well as the controls. This was a high number, compared to the number of patients that were given a psychiatric diagnosis prior to the E05 thyrotoxicosis diagnosis in our registry search and compared to the 2013 point prevalence of depression and anxiety in the Swedish general female population (35) as well as to a previous publication on thyrotoxicosis where SCID was used to evaluate previous psychiatric conditions (36). These disparities are unlikely to be caused by selection bias since there was no difference in our registry search between included patients and patients not included. SCID is an interviewer-dependent instrument and our interviewers had vast experience in using this instrument. Nevertheless, there may be interrater differences. Despite the high frequency of previous psychiatric conditions in the SCID interview, these results, together with the result of our registry search, indicate a low probability that the root of the common symptoms of depression and anxiety in GD originate in an increased psychiatric vulnerability in patients that respond to GD with psychiatric symptoms.
Concerning bipolar disorder, the higher prevalence in the patients compared to the controls in the SCID interview confirms previous findings (26). Most previous studies have investigated the increased prevalence of hyperthyroidism and antibodies in patients with bipolar disorder, but hyperthyroidism has previously also been described as a risk factor for bipolar disorder, especially evident in women (37). None of our included patients were on lithium treatment.
That patients without a previous psychiatric condition and without eye symptoms still scored worse in CPRS supports the hypothesis that psychiatric symptoms in the hyperthyroid phase of GD is a direct consequence of the disease. We could not confirm previous findings of correlations between thyroid hormone levels and/or thyroid antibody levels and psychiatric symptoms. The reason for this is likely our inclusion only of patients with very high thyroid hormone levels.
Still, having had a diagnosis of depression, anxiety, or bipolarity before the onset of GD aggravates the psychiatric symptoms in hyperthyroidism and should be noted in the clinical evaluation to prepare for directed support during the course of the disease.
It is well known that active TAO has a deleterious effect on GD patients’ QoL (20, 25) but that mild eye symptoms are associated with the patients’ psychiatric symptoms is a novel finding. Eye complications is a publicly well-known, and feared, consequence in GD and patients may experience that mild eye symptoms are a first sign of coming problems and react psychologically to this. Another possibility is a common biochemical cause to eye symptoms and psychiatric symptoms. Despite the cause, the association between mild eye symptoms and more severe psychiatric symptoms is important to bear in mind in the initial consultation.
Strengths/weaknesses
The CogThy study only includes patients with very high levels of thyroid hormones. This is a strength in that it narrows the group that is studied and minimizes the risk of misdiagnosis but also a limitation since it reduces the generalizability of the study. Therefore, the registry search was added to broaden the perspective.
Swedish registries are based on unique individual personal numbers, which ensures that no individual is registered twice and that all are included. Another strength is that recruitment bias was controlled through a registry search.
Moderately severe and severe TAO is known to impair QoL considerably (38). The CogThy study excluded patients with TAO with ongoing steroid medication or risk for steroid treatment in the coming year. We consider this as a strength of the study since it excludes an important confounder.
The limitation to female participants can be seen as a weakness but is also a strength that it limits the internal differences in the population.
Conclusion
Psychiatric symptoms are common in GD patients, both in the hyperthyroid phase and after treatment. Although having a previous psychiatric condition aggravates these symptoms, they are likely not a consequence of a premorbidity but rather a consequence of the disease per se. Having a history of previous psychiatric conditions, eye symptoms early in the course of the disease, and a younger age are all circumstances associated with worse treatment outcome and should be addressed early in treatment period to limit remaining problems in vulnerable individuals.
Declaration of interest
HFN has received lecturing fees from Siemens Inc, Astra Zeneca, and Bristol Mayer Squibb, IBSA and is head of an advisory board for Oripharm. All other authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study reported.
Funding
The study was financed by grants from the Swedish State under the agreement between the Swedish government and the county councils, the ALF-agreement (ALFGBG-717311, ALFGBG-790271), the Healthcare Board, Region Västra Götaland (Hälso- och sjukvårdsstyrelsen), Sahlgrenska University Hospital research funds, Gothenburg Medical Society, Swedish Medical Society, Swedish Society for Medical Research, Swedish Endocrine Society, Fredrik and Ingrid Thuring’s Foundation, Iris Grant, Jeansson’s Foundation, Tore Nilsson’s Foundation, Wilhelm and Martina Lundgren’s Foundation, Pharmacist Hedberg’s Foundation, Anna-Lisa and Bror Björnsson’s Foundation, Adlerbert Research Foundation, and Åke Wiberg’s Foundation. Region Västra Götaland, Sweden, is acknowledged for generous support.
Author contribution statement
All authors participated in the design of the study and collection of data. MH did all the analyses and wrote the manuscript with important contributions from all the rest.
Acknowledgements
The study is registered in the public project database for research and development in Västra Götaland County, Sweden (https://www.researchweb.org/is/vgr/project/44321). The authors would especially like to thank Jenny Tiberg and Anders Svensson for their dedicated work on the CogThy study.
References
- 1↑
Watt T, Groenvold M, Rasmussen AK, Bonnema SJ, Hegedus L, Bjorner JB, & Feldt-Rasmussen U. Quality of life in patients with benign thyroid disorders. A review. European Journal of Endocrinology 2006 154 501–510. (https://doi.org/10.1530/eje.1.02124)
- 2↑
Elberling TV, Rasmussen AK, Feldt-Rasmussen U, Hording M, Perrild H, & Waldemar G. Impaired health-related quality of life in Graves’ disease. A prospective study. European Journal of Endocrinology 2004 151 549–555. (https://doi.org/10.1530/eje.0.1510549)
- 3↑
Berg G, Michanek A, Holmberg E, & Nystrom E. Clinical outcome of radioiodine treatment of hyperthyroidism: a follow-up study. Journal of Internal Medicine 1996 239 165–171. (https://doi.org/10.1046/j.1365-2796.1996.441788000.x)
- 4↑
Bové KB, Watt T, Vogel A, Hegedüs L, Bjoerner JB, Groenvold M, Bonnema SJ, Rasmussen ÅK, & Feldt-Rasmussen U. Anxiety and depression are more prevalent in patients with Graves’ disease than in patients with nodular goitre. European Thyroid Journal 2014 3 173–178. (https://doi.org/10.1159/000365211)
- 5↑
Fukui T, Hasegawa Y, & Takenaka H. Hyperthyroid dementia: clinicoradiological findings and response to treatment. Journal of the Neurological Sciences 2001 184 81–88. (https://doi.org/10.1016/s0022-510x(0000487-1)
- 6↑
Vogel A, Elberling TV, Hording M, Dock J, Rasmussen AK, Feldt-Rasmussen U, Perrild H, & Waldemar G. Affective symptoms and cognitive functions in the acute phase of Graves’ thyrotoxicosis. Psychoneuroendocrinology 2007 32 36–43. (https://doi.org/10.1016/j.psyneuen.2006.09.012)
- 7↑
Torring O, Tallstedt L, Wallin G, Lundell G, Ljunggren JG, Taube A, Sääf M, Hamberger B & the Thyroid Study Group. Graves’ hyperthyroidism: treatment with antithyroid drugs, surgery, or radioiodine – a prospective, randomized study. Journal of Clinical Endocrinology and Metabolism 1996 81 2986–2993. (https://doi.org/10.1210/jcem.81.8.8768863)
- 8↑
Miao Q, Zhang S, Guan YH, Ye HY, Zhang ZY, Zhang QY, Xue RD, Zeng MF, Zuo CT, & Li YM. Reversible changes in brain glucose metabolism following thyroid function normalization in hyperthyroidism. AJNR American Journal of Neuroradiology 2011 32 1034–1042. (https://doi.org/10.3174/ajnr.A2449)
- 9↑
Schreckenberger MF, Egle UT, Drecker S, Buchholz HG, Weber MM, Bartenstein P, & Kahaly GJ. Positron emission tomography reveals correlations between brain metabolism and mood changes in hyperthyroidism. Journal of Clinical Endocrinology and Metabolism 2006 91 4786–4791. (https://doi.org/10.1210/jc.2006-0573)
- 10↑
Trzepacz PT, McCue M, Klein I, Levey GS, & Greenhouse J. A psychiatric and neuropsychological study of patients with untreated Graves’ disease. General Hospital Psychiatry 1988 10 49–55. (https://doi.org/10.1016/0163-8343(8890084-9)
- 11↑
Gulseren S, Gulseren L, Hekimsoy Z, Cetinay P, Ozen C, & Tokatlioglu B. Depression, anxiety, health-related quality of life, and disability in patients with overt and subclinical thyroid dysfunction. Archives of Medical Research 2006 37 133–139. (https://doi.org/10.1016/j.arcmed.2005.05.008)
- 12↑
Chattopadhyay C, Chakrabarti N, & Ghosh S. An assessment of psychiatric disturbances in Graves disease in a medical college in eastern India. Nigerian Journal of Clinical Practice 2012 15 276–279. (https://doi.org/10.4103/1119-3077.100620)
- 13↑
Riguetto CM, Neto AM, Tambascia MA, & Zantut-Wittmann DE. The relationship between quality of life, cognition, and thyroid status in Graves’ disease. Endocrine 2019 63 87–93. (https://doi.org/10.1007/s12020-018-1733-y)
- 14↑
Johansson B, Holmberg M, Skau S, Malmgren H, & Filipsson Nystrom H. The relationship between mental fatigue, depression, and cognition in Graves’ disease. European Thyroid Journal 2023 12. (https://doi.org/10.1530/ETJ-23-0040)
- 15↑
Perrild H, Hansen JM, Arnung K, Olsen PZ, & Danielsen U. Intellectual impairment after hyperthyroidism. Acta Endocrinologica 1986 112 185–191. (https://doi.org/10.1530/acta.0.1120185)
- 16↑
Abraham-Nordling M, Torring O, Hamberger B, Lundell G, Tallstedt L, Calissendorff J, & Wallin G. Graves’ disease: a long-term quality-of-life follow up of patients randomized to treatment with antithyroid drugs, radioiodine, or surgery. Thyroid 2005 15 1279–1286. (https://doi.org/10.1089/thy.2005.15.1279)
- 17↑
Torring O, Watt T, Sjolin G, Bystrom K, Abraham-Nordling M, Calissendorff J, Cramon PK, Filipsson Nystrom H, Hallengren B, Holmberg M, et al.Impaired quality of life after radioiodine therapy compared with antithyroid drugs or surgical treatment for Graves’ hyperthyroidism. A long-term follow-up with ThyPRO and SF-36. Thyroid 2019 29 322–331. (https://doi.org/10.1089/thy.2018.0315)
- 18↑
Holmberg M, Malmgren H, Heckemann RA, Johansson B, Klasson N, Olsson E, Skau S, Starck G, & Filipsson Nystrom H. A longitudinal study of medial temporal lobe volumes in Graves’ disease. Journal of Clinical Endocrinology and Metabolism 2022 107 1040–1052. (https://doi.org/10.1210/clinem/dgab808)
- 19↑
Calissendorff J, Cramon PK, Hallengren B, Khamisi S, Lantz M, Planck T, Sjolin G, Wallin G, & Holmberg M. Long-term outcome of Graves’ disease: a gender perspective. Women’s Health Reports 2023 4 487–496. (https://doi.org/10.1089/whr.2023.0073)
- 20↑
Kahaly GJ, Petrak F, Hardt J, Pitz S, & Egle UT. Psychosocial morbidity of Graves’ orbitopathy. Clinical Endocrinology 2005 63 395–402. (https://doi.org/10.1111/j.1365-2265.2005.02352.x)
- 21↑
Brandt F, Thvilum M, Almind D, Christensen K, Green A, Hegedus L, & Heiberg Brix TH. Hyperthyroidism and psychiatric morbidity: evidence from a Danish nationwide register study. European Journal of Endocrinology 2014 170 341–348. (https://doi.org/10.1530/EJE-13-0708)
- 22↑
Fukao A, Takamatsu J, Arishima T, Tanaka M, Kawai T, Okamoto Y, Miyauchi A, & Imagawa A. Graves’ disease and mental disorders. Journal of Clinical and Translational Endocrinology 2020 19 100207. (https://doi.org/10.1016/j.jcte.2019.100207)
- 23↑
Zhang W, Song L, Yin X, Zhang J, Liu C, Wang J, Zhou D, Chen B, & Lii H. Grey matter abnormalities in untreated hyperthyroidism: a voxel-based morphometry study using the DARTEL approach. European Journal of Radiology 2014 83 e43–e48. (https://doi.org/10.1016/j.ejrad.2013.09.019)
- 24↑
Engum A, Bjoro T, Mykletun A, & Dahl AA. An association between depression, anxiety and thyroid function--a clinical fact or an artefact? Acta Psychiatrica Scandinavica 2002 106 27–34. (https://doi.org/10.1034/j.1600-0447.2002.01250.x)
- 25↑
Coulter I, Frewin S, Krassas GE, & Perros P. Psychological implications of Graves’ orbitopathy. European Journal of Endocrinology 2007 157 127–131. (https://doi.org/10.1530/EJE-07-0205)
- 26↑
Bunevicius R, Velickiene D, & Prange AJ Jr. Mood and anxiety disorders in women with treated hyperthyroidism and ophthalmopathy caused by Graves’ disease. General Hospital Psychiatry 2005 27 133–139. (https://doi.org/10.1016/j.genhosppsych.2004.10.002)
- 27↑
Sprah L, Dernovsek MZ, Wahlbeck K, & Haaramo P. Psychiatric readmissions and their association with physical comorbidity: a systematic literature review. BMC Psychiatry 2017 17 2. (https://doi.org/10.1186/s12888-016-1172-3)
- 28↑
Holmberg MO, Malmgren H, Berglund P, Bunketorp-Kall L, Heckemann RA, Johansson B, Klasson N, Olsson E, Skau S, & Filipsson Nystrom H. Structural brain changes in hyperthyroid Graves’ disease: protocol for an ongoing longitudinal, case-controlled study in Göteborg, Sweden—the CogThy project. BMJ Open 2019 9 e031168. (https://doi.org/10.1136/bmjopen-2019-031168)
- 29↑
Mourits MP, Prummel MF, Wiersinga WM, & Koornneef L. Clinical activity score as a guide in the management of patients with Graves’ ophthalmopathy. Clinical Endocrinology 1997 47 9–14. (https://doi.org/10.1046/j.1365-2265.1997.2331047.x)
- 30↑
Tozzoli R, D’Aurizio F, Villalta D, & Giovanella L. Evaluation of the first fully automated immunoassay method for the measurement of stimulating TSH receptor autoantibodies in Graves’ disease. Clinical Chemistry and Laboratory Medicine 2017 55 58–64. (https://doi.org/10.1515/cclm-2016-0197)
- 31↑
First MB, Spitzer RL, Gibbon M & & Williams JBW Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research version, Patient ed n .: New York, USA: Biometrics Research, New York State Psychiatric Institute, 2002.
- 32↑
Svanborg P, & Åsberg M. A new self-rating scale for depression and anxiety states based on the Comprehensive Psychopathological Scale. Acta Psychiatrica Scandinavica 1994 89 21–28. (https://doi.org/10.1111/j.1600-0447.1994.tb01480.x)
- 33↑
Abraham-Nordling M, Bystrom K, Torring O, Lantz M, Berg G, Calissendorff J, Filipsson Nystrom HF, Jansson S, Jorneskog G, Karlsson FA, et al.Incidence of hyperthyroidism in Sweden. European Journal of Endocrinology 2011 165 899–905. (https://doi.org/10.1530/EJE-11-0548)
- 34↑
Filipsson Nyström HF, Jansson S, & Berg G. Incidence rate and clinical features of hyperthyroidism in a long-term iodine sufficient area of Sweden (Gothenburg) 2003–2005. Clinical Endocrinology 2013 78 768–776. (https://doi.org/10.1111/cen.12060)
- 35↑
Johansson R, Carlbring P, Heedman Å, Paxling B, & Andersson G. Depression, anxiety and their comorbidity in the Swedish general population: point prevalence and the effect on health-related quality of life. PeerJ 2013 1 e98. (https://doi.org/10.7717/peerj.98)
- 36↑
Placidi GPA, Boldrini M, Patronelli A, Fiore E, Chiovato L, Perugi G, & Marazziti D. Prevalence of psychiatric disorders in thyroid diseased patients. Neuropsychobiology 1998 38 222–225. (https://doi.org/10.1159/000026545)
- 37↑
Hu LY, Shen CC, Hu YW, Chen MH, Tsai CF, Chiang HL, Yeh CM, Wang WS, Chen PM, Hu TM, et al.Hyperthyroidism and risk for bipolar disorders: a nationwide population-based study. PLoS One 2013 8 e73057. (https://doi.org/10.1371/journal.pone.0073057)
- 38↑
Wiersinga WM. Quality of life in Graves’ ophthalmopathy. Best Practice and Research. Clinical Endocrinology and Metabolism 2012 26 359–370. (https://doi.org/10.1016/j.beem.2011.11.001)
