The prognosis of severe subclinical hyperthyroidism with TSH below 0.1 μU/mL due to Graves’ disease in the Japanese population

Abstract

To determine the prognosis of Graves’ disease initially presenting with severe subclinical hyperthyroidism, we investigated 110 patients with Graves’ disease with normal FT3 and FT4 levels and TSH below 0.1 μU/mL. Graves’ disease was diagnosed based on the diffuse accumulation of radioiodine in the thyroid in 83 patients, while the other 27 patients were diagnosed based on positive anti-TSH receptor antibodies. Seventy patients did not receive immediate medical treatment for the hyperthyroidism. Forty-four patients developed overt hyperthyroidism after 1–131 (median 3) months. In 19 patients, TSH levels returned to normal after 1–43 (median 6) months. One patient developed persistent hypothyroidism after two months, and another six had subclinical hyperthyroidism during the observation period. The positivity of TSH receptor antibodies was significantly higher (p = 0.0445) in patients who developed overt hyperthyroidism (86.0%) than in other patients (65.4%). Seventeen patients were treated immediately after diagnosis. Seven patients remitted after 2–94 (median 9) months of medical treatment. Another 10 patients remained euthyroid under the continuous administration of small amounts of medication. Some patients with severe subclinical hyperthyroidism due to Graves’ disease develop overt hyperthyroidism. If patients are at risk due to cardiovascular diseases, osteoporotic fractures, or an older age, then immediate treatment can be considered. Otherwise, careful monitoring of the thyroid function without treatment for 6 months is considered to be reasonable. TRAb has been suggested to play a role in the progression of subclinical hyperthyroidism due to Graves’ disease.

Introduction

Subclinical hyperthyroidism is defined as suppressed TSH levels with normal free T3 and free T4 levels. Most cases of endogenous subclinical hyperthyroidism are caused by Graves’ disease and autonomously functioning thyroid nodules (AFTNs). Graves’ disease is dominant in iodine-replete areas or in younger patients, whereas AFTNs are dominant in iodine-deficient areas or elderly patients [1]. Although Graves’ disease is the most prevalent cause of overt hyperthyroidism, its diagnosis during subclinical hyperthyroidism is relatively rare, even in iodine-replete areas. Therefore, the prognosis for these conditions is not well understood. We retrospectively investigated patients with severe subclinical hyperthyroidism due to Graves’ disease to determine their prognoses.

Subjects and Methods

Patients

We found 966 patients showing severe subclinical hyperthyroidism with TSH levels below 0.1 μU/mL among patients who visited Noguchi Thyroid Clinic and Hospital for the first time in 2008–2022. Of the 966 patients, 110 diagnosed with Graves’ disease were included in this study (Fig. 1). Graves’ disease was diagnosed when radioactive iodine (RAI) accumulated diffusely in the thyroid with an uptake ratio above 10% (definite group; n = 83). Among patients not evaluated for RAI uptake ratios, those with positive anti-TSH receptor antibodies (TRAbs) were diagnosed with probable Graves’ disease (probable group; n = 27). Patients suspected to have painless thyroiditis due to subsequent transient hypothyroidism were excluded.

Fig. 1  Causes of severe subclinical hyperthyroidism with TSH <0.1 μU/mL in 966 patients diagnosed between 2008 and 2022.

Seventy patients (51 definite and 19 probable) were observed without immediate treatment with methimazole (MMI) or potassium iodide (KI). Seventeen patients (14 definite and 3 probable) were treated with MMI or KI immediately after diagnosis because of their older age or complaints, such as palpitation. None of the patients were instructed to consume excess iodine. The remaining 23 patients could not be followed-up.

This study was approved by the Ethics Committee of our hospital (approval number 046) and conducted in accordance with the Declaration of Helsinki. The patients were given the right to opt out of the study.

Methods

We analyzed age, sex, estimated volume of the thyroid glands, and levels of free T3, free T4, TSH, TRAb, anti-thyroglobulin antibodies (TgAbs), anti-thyroid peroxidase antibodies (TPOAbs), and RAI uptake at the first presentation. In patients who were observed without immediate treatment with MMI or KI, thyroid function was observed for 1–170 (median 6) months. Predictive factors for progression to overt hyperthyroidism were investigated. Changes in TRAb levels were also analyzed in patients who developed overt hyperthyroidism and in those who returned to euthyroid conditions. The prognosis of the patients treated with MMI, propylthiouracil (PTU) or KI was also investigated.

Assays for serum hormones and anti-thyroid autoantibodies

Free T3, free T4, and TSH levels were measured by electrochemiluminescent immunoassay (ECLIA) using Roche Elecsys FT3, FT4, and TSH kits (Roche Diagnostics, Basel, Switzerland). Reference ranges of these were 2.2–4.3 pg/mL for FT3, 0.9–1.7 ng/dL for FT4 and 0.50–5.00 μU/mL for TSH. TRAb levels were measured by enzyme-linked immunosorbent assay (ELISA) using TRAb ELISA Cosmic (Cosmic Corporation Co., Ltd, Tokyo, Japan) (TRAb 2^nd) until March 2009. They were then measured by ECLIA using a Roche Elecsys Anti-TSHR (Roche Diagnostics) (TRAb 3^rd). Until May 2010, TgAb and anti-microsomal antibody values were determined by particle agglutination assays using Serodia ATG and AMC (Fujirebio Co., Hachioji, Japan), respectively. TgAb and TPOAb levels were determined by fluorescent enzyme immunoassay (FEIA) using E Test TOSOH II (Tosoh Corporation, Tokyo, Japan) between June 2010 and April 2018. They were then measured by chemiluminescent enzyme immunoassay (CLEIA) using an AIA-PACK CL (Tosoh Corporation, Tokyo, Japan).

Statistical analyses

The difference in the means of the two groups was analyzed using Wilcoxon’s rank-sum test or Wilcoxon’s signed-rank test, as appropriate. Categorical variables were analyzed using Fisher’s exact test. Statistical analyses were performed using SAS JMP software, ver. 11.0.0 (SAS Institute Inc., Cary, NC, USA).

Results

Patient demographics

Patient demographics are shown in Table 1. There were no significant differences in age, sex, thyroid volume, free T3, free T4 or TSH levels between the definite and probable groups. No significant differences were observed in the positivity of TgAb and TPOAb between the two groups. However, the levels of TRAb 3^rd and positivity for any TRAb in the probable group were significantly higher (p < 0.0001) than those in the definite group. The age of the 17 patients treated immediately after the diagnosis was 30–84 (median 63) years, which was significantly higher (p = 0.0016) than the 14–78 (median 46) years in the 70 patients who were observed without immediate treatment.

Table 1 Demography of the patients.

	Definite	Probable	p value
Number of patients	83	27
Age (yr)	14–83 (47)	26–84 (43)	N.S.
F/M	69/14	24/3	N.S.
Volume (mL)	5.0–48.1 (11.7)	3.3–21.2 (10.6)	N.S.
FT3 (pg/mL)	2.2–4.3 (3.6)	2.9–4.2 (3.7)	N.S.
FT4 (ng/dL)	0.9–1.7 (1.4)	1.1–1.7 (1.5)	N.S.
TSH (μU/mL)	0.01 > –0.09 (0.01)	0.01 > –0.09 (0.01)	N.S.
RAIU (% at 24 hrs)	11.2–49.1 (26.9)	N.D.
TRAb 2nd (%)	0.6–91.7 (54.4)	37.9–44.8 (41.4)
TRAb 3rd (IU/L)	0.3 > –20.1 (2.6)	2.0–38.3 (4.3)	p < 0.0001
TRAb positivity (%)	63.0	100	p < 0.0001
TgAb positivity (%)	39.0	27.3	N.S.
TPOAb positivity (%)	40.8	23.8	N.S.

Data are shown as ranges and median in parentheses. RAIU: radioactive iodine uptake ratio, TRAb: anti-TSH receptor antibodies, TgAb: anti-thyroglobulin antibodies, TPOAb: anti-thyroid peroxidase antibodies or anti-microsomal antibodies, N.S.: not significant, N.D.: not detected.

The prognosis of patients who were observed without immediate treatment

As shown in Table 2 and Fig. 2, among the 70 patients (definite 51 and probable 19) observed without immediate treatment, 44 patients (62.9%) developed overt hyperthyroidism after 1–131 (median 3) months. Twenty-five patients developed overt hyperthyroidism within three months, and the other eight patients developed overt hyperthyroidism between three and six months. No significant difference in the prevalence of progression to overt hyperthyroidism was observed between the definite and probable patients (62.7% vs. 63.2%).

Table 2 Prognosis of the patients observed without immediate treatment after the diagnosis of subclinical hyperthyroidism.

	Total	Definite	Probable
Number of patients	70	51	19
Overt hyperthyroid	44 (62.9%)	32 (62.7%)	12 (63.2%)
Return to euthyroid	19 (27.1%)	14 (27.5%)	5 (26.3%)
Persistent subclinical hyperthyroid	6 (8.6%)	4 (7.8%)	2 (10.5%)
Persistent hypothyroid	1 (1.4%)	1 (2.0%)	0

Fig. 2  Courses of 110 patients with subclinical hyperthyroidism due to Graves’ disease.

The other 26 patients (definite 19 and probable 7) did not develop overt hyperthyroidism during the observation period. TSH levels returned to normal in 19 of the 26 patients within 1–43 (median 6) months. Persistent normal TSH levels were confirmed in 14 patients for 4–68 (median 29.5) months after initial normalization of TSH levels. In contrast, 6 patients had persistent subclinical hyperthyroidism for 3–92 (median 12) months. TSH levels at the final observation were <0.1 μU/mL in 3 and 0.1–0.5 μU/mL in 3. One patient in the definite group developed persistent hypothyroidism two months after diagnosis. No complications, such as cardiovascular events or fractures, were noticed during the observation period without administering any treatment.

Courses of patients treated immediately after the diagnosis

Ten patients were immediately administered with MMI at their first visit. The maximum dose of MMI was 2.5–10 (median 5) mg in 8 patients, but either 30 mg or 15 mg of MMI was required in each patient. Seven patients were administered 10–50 (median 25) mg of KI without MMI.

MMI or KI was withdrawn in 7 patients after 2–94 (median 9) months. None of the patients relapsed after 10–38 (median 29) months. Another 10 patients remained euthyroid under continuous administration of small amounts of medications (Fig. 2).

Course of patients treated after progression to overt hyperthyroidism

Among 44 patients who developed overt hyperthyroidism, MMI or PTU was administered to 35. The maximum dose of MMI was ≤15 mg in 27 patients, 20 mg in 1 patient and 30 mg in three patients. The maximum dose of PTU was 200 mg in 2 patients and 300 mg in 2 patients. Other nine patients were administered KI. The maximum dose of KI was ≤50 mg in 7 patients. MMI, PTU, or KI was withdrawn within 1–65 (median 23.5) months in 24 patients. Five of the 24 patients relapsed. Two patients had allergic skin symptoms and the other patient had liver injury caused by MMI. Thyroidectomy and radioiodine therapy were performed in each patient. In 9 patients, the administration of MMI, PTU or KI was continued. The remaining 9 patients could not be followed-up.

Predictive factors for development of overt hyperthyroidism

Between the 44 patients who developed overt hyperthyroidism and 26 who did not develop overt hyperthyroidism, there were no significant differences in age, sex, estimated thyroid volume, positivity of TgAb or TPOAb, or levels of free T3, free T4, TSH, or TRAb 3^rd. TRAb 2^nd or 3^rd positivity was detected in 86.0% of patients who developed overt hyperthyroidism, which was significantly higher (p = 0.0445) than the rate of 65.4% among patients who did not develop overt hyperthyroidism.

Levels of TRAb 3^rd according to changes in the thyroid functions

Among the patients who developed overt hyperthyroidism, TRAb 3^rd level at the time of deterioration was available for 28. Levels of TRAb 3^rd increased significantly (p = 0.0013) from <0.8–18.3 (median 3.15) IU/L to 1.1–35.0 (median 5.6) IU/L at the time of development of overt hyperthyroidism (Fig. 3).

Fig. 3  TRAb 3^rd titers in patients who developed overt hyperthyroidism. TRAb levels are significantly elevated in patients with overt hyperthyroidism.

Among the patients who returned to a euthyroid condition, TRAb 3^rd level at the time of TSH normalization was available for 16. The levels of TRAb 3^rd decreased from <0.8–20.1 (median 2.8) IU/L to <0.8–6.7 (median 1.1) IU/L according to normalization of TSH levels. However, these changes were not statistically significant (Fig. 4).

Fig. 4  Titers of TRAb 3^rd of the patients who remitted spontaneously. Changes in the titers of TRAb 3^rd were not significant. N.S.: not significant.

Discussion

The prevalence of subclinical hyperthyroidism is reported to be 0.63%–2.26% among the general population when defined as TSH <0.4 or 0.5 μU/mL [2-4] and 0.24%–0.7% when defined as TSH <0.05 or 0.1 μU/mL [4, 5]. It is also reported that the prevalence of subclinical hyperthyroidism in Japan is 0.27% (TSH <0.15 μU/mL) [6] or 0.82% (TSH <0.1 μU/mL) [7]. The major causes of endogenous subclinical hyperthyroidism are Graves’ disease and AFTNs. Graves’ disease is dominant in iodine-replete areas, whereas AFTNs are dominant in iodine-deficient areas [1]. A retrospective analysis in New Zealand revealed that 12.5% of 96 patients with subclinical hyperthyroidism had Graves’ disease, and the other 87.5% had AFTNs [8]. Another study performed in the UK revealed that 5.6% of 323 patients had subclinical hyperthyroidism caused by Graves’ disease, whereas 69.3% had it due to AFTNs [9]. Among the Japanese population, Kasagi et al. reported that 7 of 20 patients with endogenous subclinical hyperthyroidism had Graves’ disease, 9 had euthyroid ophthalmic Graves’ disease, and 4 had AFTNs [7].

One of the clinical implications of subclinical hyperthyroidism is the risk of its progression to overt hyperthyroidism. Several observations of patient groups consisting 5.3%–12.5% with Graves’ disease and 69.3%–89.3% with AFTNs revealed that 2.9%–45.3% of these patients developed overt hyperthyroidism [8-11]. A TSH level <0.1 μU/mL is reported to be the only predictive factor for progression to overt hyperthyroidism [9, 10]. As for subclinical hyperthyroidism due to Graves’ disease specifically, Zhyzhneuskaya et al. [12] reported that 34% out of 44 patients developed overt hyperthyroidism, while 34% returned to a euthyroid status and 30% had persistent subclinical hyperthyroidism. Among 36 patients with TSH ≤0.1 μU/mL, 39% developed overt hyperthyroidism. They also mentioned that older age and positive TPOAb test results were risk factors for progression to overt hyperthyroidism. Kasagi et al. reported that three out of seven patients with subclinical hyperthyroidism due to Graves’ disease developed overt hyperthyroidism [7]. In the present study, 62.9% of patients developed overt hyperthyroidism. Furthermore, 75% of these patients progressed to overt hyperthyroidism six months after diagnosis. It is unknown why progression was more prevalent in our patients than in a previous report [12]. One speculation for the difference in the prognosis between the two studies is that our study was conducted in an iodine-replete area, whereas the study of Zhyzhneskaya et al. [12] was conducted in the UK. The difference in iodine sufficiency between the two areas might contribute to the difference in the progression to overt hyperthyroidism.

TRAb level is a useful parameter for diagnosing Graves’ disease. The sensitivity and specificity for differentiating Graves’ disease from destructive thyrotoxicosis are reported to be 97.0% and 99.1%, respectively [13]. TRAb titers are known to represent the severity of hyperthyroidism [14]. In this study, TRAb positivity was slightly higher in the patients who developed overt hyperthyroidism. In these patients, TRAb levels significantly increased during the development of overt hyperthyroidism. In contrast, TRAb levels did not increase in patients in spontaneous remission of subclinical hyperthyroidism. These findings show that TRAb may contribute to disease progression, even in subclinical hyperthyroidism caused by Graves’ disease.

Another clinical implication of subclinical hyperthyroidism is that it can cause cardiovascular morbidity, such as atrial fibrillation [15, 16], congestive heart failure [17], and coronary heart disease-related mortality [16]. In addition, subclinical hyperthyroidism increases the risk of fractures [18]. These increased risk of cardiovascular morbidity and fractures are higher in the patients TSH <0.1 μU/mL than those with TSH ≥0.1 μU/mL [15-18]. Therefore, management is recommended for patients with persistent subclinical hyperthyroidism when TSH levels are <0.1 μU/mL and patients are ≥65 years old [19]. In our cohort, 17 patients were administered MMI or KI immediately after diagnosis because of their older age or complaints, such as palpitation. Most patients were well-controlled with a small amount of medication. The reasons why these patients had a good response to medication are thought to be due to the fact that the treatment was started at an early stage of the disease while, in addition some patients may have demonstrated a spontaneous remission. Nevertheless, we believe that patients at risk of cardiovascular diseases, osteoporotic fractures, or older age, can be treated immediately to avoid complications. On the other hand, observation without immediate treatment for 6 months seems reasonable for patients at low risk for developing complications. This is because some patients spontaneously return to a euthyroid state and progression to overt hyperthyroidism was observed within 6 months in seventy-five percent of the patients who developed overt hyperthyroidism. Moreover, the prognosis of most patients is good even if they are treated after progression to overt hyperthyroidism. When persistent severe subclinical hyperthyroidism is observed at 6 months, treatment should be considered.

Several limitations of the present study warrant mention. First, RAI uptake ratios were not determined in some patients (probable group). However, no significant differences in prognosis were found between the definite and probable disease groups (Table 2). We believe that patients in the probable disease group may have Graves’ disease. Second, the observation period was insufficient in some patients. A longer observation period is required to establish further evidence. Third, patients with subclinical hyperthyroidism whose TSH levels were between 0.1 μU/mL and the lower limit of reference ranges were not included in our study. However, we believe that the prognosis of patients with TSH <0.1 μU/mL is a more important issue, since the risks of progression to overt hyperthyroidism and the risks of cardiovascular morbidity and fractures are reported to be higher than those of patients with ≥TSH 0.1 μU/mL [19, 20]. Finally, we were unable to analyze TSH receptor mutations. Although non-autoimmune hyperthyroidism is a rare genetic disease [21], it cannot be ruled out in our patients.

In conclusion, some patients with severe subclinical hyperthyroidism due to Graves’ disease develop overt hyperthyroidism. If patients are at risk due to cardiovascular diseases, osteoporotic fractures, or an older age, then immediate treatment can be considered. Otherwise, careful monitoring of the thyroid function without treatment for 6 months is considered to be reasonable. TRAb has been suggested to play a role in the progression of subclinical hyperthyroidism due to Graves’ disease (Graphical Abstract).

Graphical Abstract 

Acknowledgements

We appreciate Ms. Keiko Suga for cooperation in the extraction of clinical data from the electronic medical records.

Author Contributions

All authors helped to edit the original draft and approved the submitted manuscript. Y.N. analyzed the data and drafted the manuscript. T.M. contributed to the discussion and edited the manuscript. N.H., J.T., and H.N. assisted with the data management.

Funding Information

This study did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author Disclosure Statement

None of the authors have any potential conflicts of interest associated with this research. Tsukasa Murakami is a member of the Editorial Board of Endocrine Journal.

References

• 1   Carlé  A,  Andersen  SL,  Boelaert  K,  Laurberg  P (2017) Management of endocrine disease: subclinical thyrotoxicosis: prevalence, causes and choice of therapy. Eur J Endocrinol 176: R325–R337.
• 2   Vejbjerg  P,  Knudsen  N,  Perrild  H,  Laurberg  P,  Carlé  A, et al. (2009) Lower prevalence of mild hyperthyroidism related to a higher iodine intake in the population: prospective study of a mandatory iodization programme. Clin Endocrinol (Oxf) 71: 440–445.
• 3   Vadiveloo  T,  Donnan  PT,  Cochrane  L,  Leese  GP (2011) The Thyroid Epidemiology, Audit, and Research Study (TEARS): the natural history of endogenous subclinical hyperthyroidism. J Clin Endocrinol Metab 96: E1–E8.
• 4   Bjørndal  MM,  Sandmo Wilhelmsen  K,  Lu  T,  Jorde  R (2008) Prevalence and causes of undiagnosed hyperthyroidismin an adult healthy population. The Tromsø study. J Endocrinol Invest 31: 856–860.
• 5   Hollowell  JG,  Staehling  NW,  Flanders  WD,  Hannon  WH,  Gunter  EW, et al. (2002) Serum TSH, T₄, and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 87: 489–499.
• 6   Konno  N,  Yuri  K,  Taguchi  H,  Miura  K,  Taguchi  S, et al. (1993) Screening for thyroid diseases in an iodine sufficient area with sensitive thyrotrophin assays, and serum thyroid autoantibody and urinary iodide determinations. Clin Endocrinol (Oxf) 38: 273–281.
• 7   Kasagi  K,  Takeuchi  R,  Misaki  T,  Kousaka  T,  Miyamoto  S, et al. (1997) Subclinical Graves’ disease as a cause of subnormal TSH levels in euthyroid subjects. J Endocrinol Invest 20: 183–188.
• 8   Schouten  BJ,  Brownlie  BE,  Frampton  CM,  Turner  JG (2011) Subclinical thyrotoxicosis in an outpatient population - predictors of outcome. Clin Endocrinol (Oxf) 74: 257–261.
• 9   Das  G,  Ojewuyi  TA,  Baglioni  P,  Geen  J,  Premawardhana  LD, et al. (2012) Serum thyrotrophin at baseline predicts the natural course of subclinical hyperthyroidism. Clin Endocrinol (Oxf) 77: 146–151.
• 10   Díez  JJ,  Iglesias  P (2009) An analysis of the natural course of subclinical hyperthyroidism. Am J Med Sci 337: 225–232.
• 11   Rosario  PW (2010) Natural history of subclinical hyperthyroidism in elderly patients with TSH between 0.1 and 0.4 mIU/l: a prospective study. Clin Endocrinol (Oxf) 72: 685–688.
• 12   Zhyzhneuskaya  S,  Addison  C,  Tsatlidis  V,  Weaver  JU,  Razvi  S (2016) The natural history of subclinical hyperthyroidism in Graves’ disease: the rule of thirds. Thyroid 26: 765–769.
• 13   Yoshimura Noh  J,  Miyazaki  N,  Ito  K,  Takeda  K,  Hiramatsu  S, et al. (2008) Evaluation of a new rapid and fully automated electrochemiluminescence immunoassay for thyrotropin receptor autoantibodies. Thyroid 18: 1157–1164.
• 14   Vos  XG,  Smit  N,  Endert  E,  Tijssen  JG,  Wiersinga  WM (2008) Frequency and characteristics of TBII-seronegative patients in a population with untreated Graves’ hyperthyroidism: a prospective study. Clin Endocrinol (Oxf) 69: 311–317.
• 15   Selmer  C,  Olesen  JB,  Hansen  ML,  Lindhardsen  J,  Olsen  AM, et al. (2012) The spectrum of thyroid disease and risk of new onset atrial fibrillation: a large population cohort study. BMJ 345: e7895.
• 16   Collet  TH,  Gussekloo  J,  Bauer  DC,  den Elzen  WP,  Cappola  AR, et al. (2012) Subclinical hyperthyroidism and the risk of coronary heart disease and mortality. Arch Intern Med 172: 799–809.
• 17   Gencer  B,  Collet  TH,  Virgini  V,  Bauer  DC,  Gussekloo  J, et al. (2012) Subclinical thyroid dysfunction and the risk of heart failure events: an individual participant data analysis from 6 prospective cohorts. Circulation 126: 1040–1049.
• 18   Blum  MR,  Bauer  DC,  Collet  TH,  Fink  HA,  Cappola  AR, et al. (2015) Subclinical thyroid dysfunction and fracture risk: a meta-analysis. JAMA 313: 2055–2065.
• 19   Biondi  B,  Bartalena  L,  Cooper  DS,  Hegedüs  L,  Laurberg  P, et al. (2015) The 2015 European Thyroid Association guidelines on diagnosis and treatment of endogenous subclinical hyperthyroidism. Eur Thyroid J 4: 149–163.
• 20   Hashimoto  K (2022) Update on subclinical thyroid dysfunction. Endocr J 69: 725–738.
• 21   Nishihara  E,  Chen  CR,  Higashiyama  T,  Mizutori-Sasai  Y,  Ito  M, et al. (2010) Subclinical nonautoimmune hyperthyroidism in a family segregates with a thyrotropin receptor mutation with weakly increased constitutive activity. Thyroid 20: 1307–1314.