Recovery from Atrophic Autoimmune Thyroiditis in a Child: Thyroid Stimulation-Blocking Antibody as a Prognostic Marker

Atrophic autoimmune thyroiditis (AAT) is an autoimmune hypothyroidism without goiter, resulting in a failure to produce sufficient quantities of thyroid hormones. AAT is distinguished from Hashimoto’s disease by the absence of a goiter. In contrast to Hashimoto’s disease, which is often suspected due to goiter development, AAT is generally suspected based on the development of hypothyroidism-related symptoms. Therefore, its diagnosis is likely to be delayed, and the severity of hypothyroidism at diagnosis is reported to be higher in pediatric cases [1]. In two studies of pediatric AAT, 19/21 (90%) and 12/18 (67%) enrolled patients had short stature [1, 2]. The mean age of onset of AAT in these studies was 7.7 and 6.5 years, respectively [1, 2]. In one of the studies [1], anti-peroxidase antibody (TPOAb), a hallmark of Hashimoto’s disease, was positive in >90% of patients [1]. Positivity for thyroid stimulation-blocking antibody (TSBAb), a characteristic autoimmune reaction of AAT, was also observed in approximately 40% of pediatric AAT [2], indicating that TSBAb is a specific but not sensitive test for predicting AAT. Generally, Hashimoto’s disease and Graves’ disease are diagnosed using anti-thyroglobulin antibody (TgAb), TPOAb, and TSH receptor antibody (TRAb) measured by electrochemiluminescence immunoassay. In cases of atypical hypothyroidism, such as those lacking goiter, TSBAb measured by cell-based bioassay is used to clarify the mechanism(s) of hypothyroidism.

In the present report, we report a case of AAT who was negative for TgAb and TPOAb and positive for TSBAb at diagnosis. During the 8-year follow-up observation, the patient experienced TSBAb seroconversion, and we successfully discontinued levothyroxine replacement therapy.

The patient was a Japanese boy, aged 10 years at the final observation. He was the first child born to parents with no history of thyroid diseases. The pregnancy was uneventful, and he was born at 39 weeks gestation via a normal vaginal delivery. Because the mother had no history of thyroid disease, no data on thyroid function or thyroid autoantibodies were available. At birth, the infant’s length was 51.2 cm (+1.5 SD), weight 3,580 g (+1.0 SD), and head circumference 34.0 cm (0.0 SD). The result of newborn screening for congenital hypothyroidism, performed on the fourth day of life was negative (blood-spot TSH level 2.7 m IU/L; cutoff <15 m IU/L). At a health checkup performed at age 1 year 6 months, he was 78.7 cm tall (−0.7 SD), weighed 11.2 kg (+1.0 SD), walked independently, and could speak 4 specific words (mom, dad, etc.). However, at the 2-year-old health checkup, he showed significant growth restriction, with a height of 80.5 cm (−1.7 SD). Because no height growth was observed at the subsequent 2-month follow-up, he was referred to us for further investigation. When we first evaluated the boy (age 2 years 2 months), the height was 80.6 cm (−2.0 SD), weight 11.6 kg (−0.3 SD), and head circumference was 48.8 cm (0 SD). The growth charts revealed stunted growth after age 1 year 10 months (Fig. 1a). He was able to run but could not yet speak two-word sentences. On physical examination, he was inactive and showed a large amount of hair loss on his shoulders. No goiter was noted on palpation. The skin was dry. Biochemical tests revealed elevated low-density lipoprotein cholesterol (LDL-C; 259 mg/dL; reference 70–139) and creatine kinase (CK; 795 U/L; reference 62–287). Thyroid function tests showed high TSH (818 mU/L; reference 0.35–4.93), low free T₃ (1.77 pg/mL; reference 1.88–3.18), and low free T₄ (<0.40 ng/dL; reference 0.70–1.48). Levels of TgAb and TPOAb were <10 IU/mL (reference <28) and 14 IU/mL (reference <16), respectively. TRAb was markedly high at >30 IU/L (reference 0–1.9). TSBAb level was elevated at 98.3% (reference <31.7). Ultrasonography revealed a hypoplastic thyroid (thyroid width 1.65 cm, −2.5 SD [3]). Internal echogenicity was low throughout the gland (Fig. 1b). We diagnosed the patient with AAT, and levothyroxine therapy was initiated at 10 μg/day (0.9 μg/kg/day). The dose of levothyroxine was subsequently increased to 30 μg/day (2.6 μg/kg/day). By age 2 years 8 months, serum levels of TSH, free T₃, free T₄, LDL-C, and CK were all normalized. TgAb and TPOAb remained negative, while TRAb and TSBAb remained positive. He showed catch-up growth following the initiation of the levothyroxine therapy (Fig. 1a).

At age 8 years, TRAb decreased (4.5 IU/L) and TSBAb became negative (0.0%). Similar results were obtained at age 10 years and 1 month (Fig. 1a). We therefore decided to discontinue levothyroxine treatment. At 7 months following discontinuation of treatment, thyroid function remained normal (TSH 1.23 mU/L, free T₃ 3.42 pg/mL, free T₄ 0.97 ng/dL), while ultrasound examination revealed a normal-sized thyroid gland (thyroid width 2.42 cm, −1.3 SD [3]) (Fig. 1b).

To investigate autoantibody production in previously reported cases of pediatric AAT, we searched the PubMed database using the key terms “hypothyroidism,” “child,” “atrophic,” and “autoimmune.” Of the 30 English-language papers identified in this search, reports were selected based on the following criteria: (i) overt hypothyroidism with a high serum TSH level and a low free T4 level; (ii) no goiter on ultrasonography and/or palpation; (iii) age at diagnosis <15 years. Six reports met these three criteria [4‒9]. Cases in these reports were diagnosed after newborn screening for hypothyroidism had been implemented in each country. The autoantibody production status of all relevant patients is summarized in Table 1. Overall, 16/17 AAT cases were positive for TgAb and/or TPOAb, of whom two were negative for TSBAb; one was positive for TSBAb; and one was negative for TgAb, TPOAb, and TSBAb.

In this paper, we report the case of a boy with AAT diagnosed at 2 years with the characteristic autoantibody production pattern: negative for TgAb and TPOAb and positive for TRAb and TSBAb. In a literature review of pediatric AAT, 80% of cases were found to be positive for either TgAb or TPOAb, while TSBAb was positive in only 1 patient. These observations suggest that destruction of the thyroid tissue occurs in the majority of pediatric AAT cases, which is followed by secondary antibody production to exposed autoantigens, as in Hashimoto’s disease. The unique feature of our case is that AAT was developed without the production of TgAb and TPOAb. Traditionally, Hashimoto’s disease was thought to be caused by abnormal cell-mediated immunity, while Graves’ disease was related to the production of thyroid-stimulating TRAb. However, recent clinical studies have revealed that this view is an oversimplification: 50–90% of Graves’ disease patients are positive for TPOAb, while 10–20% of patients with Hashimoto’s disease patients are positive for TRAb [10]. In one study of adult Hashimoto's disease, 10% of patients were positive for TSBAb [11]. Further, in one study of adult AAT, 5 of 28 patients were negative for both TgAb and TPOAb and positive for TSBAb [12]. At present, the number of pediatric AAT cases with detailed clinical information is very limited. More cases are required to clarify the phenotypic diversity of pediatric AAT.

This is the first report of pediatric AAT which resolved over several years of follow-up. In one 10-year follow-up study involving 24 adult TSBAb-positive AAT cases, TSBAb disappeared in seven cases, with five of these individuals maintaining normal thyroid function after discontinuing levothyroxine therapy [13]. Further studies are required to determine if similar cases exist in children, what their autoantibody status is, and which are most likely to recover thyroid function. The short observation period following levothyroxine discontinuation is a limitation of this study. Because of the possibility of AAT recurrence during adolescence, when the body’s demand for thyroid hormone increases, longer follow-up is required to conclude that the disease is indeed in remission.

Unlike Hashimoto’s thyroiditis, which typically presents with a characteristic goiter, AAT is generally only diagnosed following growth failure [1, 2]. This delay in diagnosis can lead to persistent hypothyroidism, which disrupts the hypothalamic-pituitary-thyroid axis, resulting in pituitary hyperplasia. One previous report on AAT proposed that pituitary hyperplasia resulting from elevated TRH levels may cause deficient growth hormone secretion and growth failure [9]. This may explain why many AAT cases, including ours, are complicated by severe growth retardation.

Herein, we reported the case of a boy with AAT diagnosed at age 2 years who successfully discontinued treatment following seroconversion of TSBAb during an 8-year follow-up. As in adults, seroconversion of TSBAb may be used as a marker in predicting a favorable prognosis in children with AAT. In cases of hypothyroidism where TgAb and TPOAb are negative, as in our case, measuring TSBAb together with TRAb would provide better understanding on the condition and predict the prognosis.

This study protocol was reviewed, and the need for approval was waived by the Ota Memorial Hospital Ethics Committee. Written informed consent was obtained from the parent of the patient for publication of the details of their medical case and any accompanying images.

The authors have no conflicts of interest to declare.

This study was funded solely by the authors’ institutional resources.

Chieko Kusano identified the patient and wrote the manuscript. Naoaki Hori provided intellectual input for the analyses. Tomonobu Hasegawa and Satoshi Narumi were the overall supervisor of the data collection and wrote the manuscript that was approved by all authors.

All data generated and/or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.