Association between ultrasonography echogenicity and heterogeneity and thyroid function in autoimmune diffuse thyroid diseases in children and adolescents

Purpose: To evaluate the association between thyroid echogenicity and heterogeneity seen on ultrasonography (US) and thyroid function in pediatric and adolescent populations with autoimmune diffuse thyroid diseases (AITD). Methods: From 2000 to 2020, we reviewed thyroid ultrasound (US) images and thyroid function statuses in 133 children and adolescent AITD patients. Our review of the images focused on decreased echogenicity and heterogeneity, which were classied into four grades. Results: Among patients with overt hypothyroidism or overt hyperthyroidism, 94.2% (65/69) showed a US grade of 3 or 4. In patients with subclinical hyper/hypothyroidism or euthyroidism, 45.3% (29/64) showed grades 1 or 2. There were no overt hyper/hypothyroidism patients with US grade 1. When we compared US grades according to thyroid status, more severe thyroid dysfunction was signicantly associated with higher US grade (p=0.047). Thyroid stimulating hormone (TSH) level differed signicantly according to US grades when we evaluated hyperthyroid (p=0.035) and hypothyroid (p=0.027) states independently. 11 patients showed both US grade and thyroid function status changes on follow-up US. Conclusions: In children and adolescent AITD patients, there was an association between decreased echogenicity and heterogeneity on US and thyroid dysfunction.


Introduction
A wide spectrum of conditions such as de ciency of the iodine that is required to produce thyroid hormones, diffuse thyroid diseases, and thyroid nodules can lead to morphologic changes in the thyroid gland 1,2 . The most common category of diffuse thyroid disease comprises autoimmune thyroid diseases (AITD), including Hashimoto's thyroiditis (HT) and Graves' disease (GD) 2 . These chronic disorders are characterized by an intolerance to self-thyroid antigens and have a prevalence of 1.3% and 9.6% in children and adolescents, respectively 3 .
In managing AITD, biochemical parameters including thyroid function and antibody levels are checked, and US is usually performed to check for parenchymal abnormalities or emergence of focal lesions 1 .
Several previous studies reported US ndings of thyroiditis and an association between thyroid gland echogenicity and thyroid function 2,4−10 . Decreased echogenicity of the thyroid gland on US is associated with overt hypothyroidism 4,10 , and changes in echogenicity have been reported in subclinical hypothyroidism as well [5][6][7][8][9] . Another general population study found that thyroid gland heterogeneity on US is related to thyroid dysfunction regardless of whether it is accompanied by hypoechogenicity 5 .
To the best of our knowledge, no studies have evaluated the relationship between US thyroid features, such as decreased echogenicity and heterogeneity, and thyroid function in child and adolescent patients with AITD, including HD and GD. The aim of this study was to evaluate the association between thyroid echogenicity and heterogeneity seen on US and thyroid function in children and adolescents with AITD.

Patients and thyroid function analysis
This retrospective study was approved by the Institutional Review Board of Kangnam Sacred Heart Hospital, all methods were performed in accordance with the relevant guidelines and regulations and the requirement for informed patient consent was waived. At our institution, 372 child and adolescent patients underwent thyroid US between January 2000 and March 2020.
Among the 372 patients, 239 were excluded for the following reasons: 95 did not meet the AITD diagnostic criteria; 85 showed no parenchymal abnormality on US scan; 29 patients were diagnosed with congenital hypothyroidism; 18 showed focal lesions on US with benign cytology results; nine had focal lesions with malignant pathology results; one patient had thyroid gland aplasia; one patient had an ectopic thyroid gland; and one patient developed thyroid gland hypoplasia after abscess formation due to a fourth branchial cleft cyst (BCC). Finally, 133 patients were included in this study. AITD diagnosis was made according to clinical history, symptoms, thyroid function test (TFT) results, and con rmation of antithyroid peroxidase autoantibodies (TPOAb) or serum TSH receptor antibodies (TRAb).
Laboratory data obtained from patient medical records were retrospectively evaluated. Serum Thyroid function was subdivided into ve statuses: overt hypothyroidism (1), subclinical hypothyroidism (2), euthyroidism (3), subclinical hyperthyroidism (4), and hyperthyroidism (5). Overt hypothyroidism (1) was de ned by low concentrations of fT4 and elevated serum concentrations of TSH. Subclinical hypothyroidism (2) was de ned as elevated serum TSH and normal fT4. Normal TSH, fT4, and T3 levels were de ned as euthyroidism (3). Subclinical hyperthyroidism (4) was de ned as low serum TSH level and normal fT4 level. Overt hyperthyroidism (5) was de ned as elevated concentrations of fT4 and low serum concentrations of TSH. US evaluation US evaluations were performed using a 5-12 MHz linear array transducer (IU22 US or HDI 5000; Philips Healthcare or LogiQ E9; GE Healthcare). One of four radiologists whose experiences with thyroid imaging ranged from 10-18 years performed US examinations. Two radiologists with 10 and 11 years of experience in thyroid imaging reviewed the US ndings and reached a consensus for each case. The reviewers were blinded to patient hormonal status. The US patterns were classi ed into four grades according to hypoechogenicity and heterogeneity degrees. Hypoechogenicity was estimated by comparing with the echogenicity of the anterior strap muscle, and heterogeneity was de ned as any region with an unclear boundary showing a different echogenicity from other parts of the gland. Grades were as follows: Grade 1 (G1) = diffusely enlarged thyroid gland with normal echogenicity (similar to a normal thyroid gland and hyperechoic to the anterior strap muscle) without heterogeneity (Fig. 1); G2 = diffusely enlarged thyroid gland with heterogeneity involving less than one third of thyroid gland, while the rest of the gland shows normal echogenicity (Fig. 2); G3 = diffusely enlarged thyroid gland with heterogeneity involving more than one third of the gland, while the rest of gland shows isoechogenicity compared with the anterior strap muscle (Fig. 3); and G4 = diffusely enlarged thyroid gland with diffuse heterogeneity involving more than one third of the gland with marked hypoechogenicity that is more hypoechoic than the anterior strap muscle (Fig. 4).
Associations were considered signi cant if the P-value was < 0.05. Descriptive data are summarized as means ± standard deviations (SD) or numbers of individuals (percentage) with a condition. Statistical differences were compared using Fisher's exact test for categorical variables. For continuous variables, the Kruskal-Wallis test was used to compare across US grade levels for thyroid function status and antibody titer after testing for normality and equivalent variance.

Discussion
Our study revealed that greater thyroid dysfunction severity was associated with higher US grade in child and adolescent AITD patients. TSH level also differed signi cantly according to US grade.
AITD is the most common cause of thyroid dysfunction in children and adolescents in iodine-su cient populations, and its representative diseases are HT and GD 13 . HT and GD are the most common causes of hypothyroidism and hyperthyroidism, respectively 4,10,13,15−18 . HT, i.e., lymphocytic thyroiditis, is a goitrous form of AITD. In HT, autoantibodies break down thyroid gland cells during immune processes [15][16][17] . GD is caused by thyroid gland stimulation by autoantibodies against the TSH receptor on follicular epithelial cells 18 . These antibodies mimic the effects of TSH, causing overproduction and release of Page 7/14 thyroid hormone. The typical HT histological features include lymphoplasmacytic in ltration, germinal center formation, follicular destruction, Hurthle cell change, and varying degrees of brosis 9,10,16 , while GD is characterized by histopathological hypercellularity, patchy lymphocyte in ltration, little colloid, and scalloping colloid 19 .
US is a diagnostic tool that is widely used as an adjunct to clinical exams for evaluating thyroid size, anatomy, and parenchymal abnormalities 2,12 . On US, healthy thyroid gland echogenicity is homogeneous and higher than that of the surrounding muscle 20 . AITD patients' thyroids present differently, with characteristic US ndings such as variable degrees of thyroid gland enlargement, decreased parenchymal echogenicity, and heterogeneous parenchymal echo pattern. Thyroid gland enlargement is usually diffuse and symmetric. HT may show as poorly de ned, patchy hypoechoic areas and micronodular patterns consisting of multiple small (~ 2-6 mm) hypoechoic nodules. In GD, vascularity tends to be increased, while in HT, it is variable 21 .
Tissue echogenicity of the thyroid gland on US depends on the organ's cellularity and vascularization.
Decreased colloid content, lymphocytic in ltration, and increases in intrathyroidal ow result in hypoechoic tissue patterns 20,22,23 . As mentioned above, in ammatory cells in ltrate and destroy the thyroid gland in HT and GD, which can appear as hypoechogenicity on US.
Heterogeneous echogenicity of the thyroid gland is another well-known nding in AITD [24][25][26][27] . Heterogeneous echogenicity is caused by multiple structures of different acoustic impedances creating variable echogenicity degrees on US, and the normal representative tissue is muscle 28 . When healthy, organs such as the thyroid gland or liver consist of characteristic cells with little variation, thus showing homogenous echogenicity on US 28 . Considering the pathological ndings of AITD, heterogeneity may also appear on US due to in ltration of other cells in the thyroid gland and brosis. Although heterogeneity is a well-known nding in AITD, no studies have evaluated the relationship between thyroid dysfunction severity and heterogenicity degree on US.
Thyroid dysfunction can affect child and adolescent growth and development in various ways. Overt hypothyroidism can cause a potentially fatal medical condition with adverse effects on lipid metabolism and cardiovascular function that occurs in about 10% of HT patients. Onset of this condition is insidious and it may not become clinically apparent until symptoms are abundantly developed 35 . Overt hyperthyroidism in children and adolescents is both less common and more severe than in adults 29,30 . Symptoms of hyperthyroidism include impaired neurodevelopmental outcomes and altered skeletal maturation 28 , such as craniosynostosis and advanced bone age in younger children. Additionally, for school-aged children, poor school performance is common, and may cause severe anxiety in patients and their parents 31 .
AITD, a single disease entity, can manifest in various thyroid function statuses. While subclinical and overt hypo-and hyperthyroidism share similar etiologies, the symptoms of the former are nonspeci c and signs are typically absent 32 . Therefore, diagnosis and treatment decisions for subclinical hypo-and hyperthyroidism mainly depend on TFT results [32][33][34] . Subclinical thyroid dysfunction tends to develop into overt thyroid dysfunction 32,33 , and the risk of progression to overt hypothyroidism in subclinical hypothyroid patients is higher in patients with underlying thyroid disease 12 . It is di cult to predict the risk of progression to a more severe state of thyroid dysfunction. There is no agreement on clinical features, numerical values that indicate mild thyroid dysfunction, or prognosis 10,12,32−24 . In our study, there was an association between US grade and TSH level, which was consistent with results from a previous study 35 , where younger patients showed a stronger association between decreased echogenicity and higher TSH, and the relationship was stronger when the changes were recent. Hypoechogenicity of the thyroid gland is a common nding in HT. Jeong et al 10 , reported that the severity of HT varies depending on hypoechogenicity degree. Other studies have reported that changes in US may be an early sign of more severe thyroid dysfunction and initial hypoechogenicity indicates later development of hypothyroidism 4.9 .
Those studies focused on adults and mainly dealt with HT. Our study ndings indicate that hypoechogenicity and heterogeneity degree on US are correlated with thyroid dysfunction in child and adolescent AITD patients, including those with GD and HT.
In 11 patients in our study, thyroid function status changed in in a positive relationship with US grade, regardless of whether US grade worsened or improved. This is the rst reported nding of not only exacerbation but also remission on US, and it suggests that US changes can re ect thyroid dysfunction status with high sensitivity.
We found no signi cant associations between echogenicity and TPOAb and TRAb levels. A few studies have assessed the correlation between autoantibody level and hypoechogenicity of the thyroid gland, but none found a signi cant relationship 9,36 . However, in mild thyroid dysfunction cases, checking the initial TPOAb level early with US can help predict the course of disease and set the treatment direction 34 .

Conclusion
We found an association between hypoechogenicity and heterogeneity degree and severity of thyroid dysfunction in child and adolescent AITD patients, including those with normal thyroid function, subclinical thyroid dysfunction, or overt thyroid dysfunction. These results suggest that US ndings can be used as another parameter to supplement biochemistry results for thyroid status evaluation. Figure 1 Transverse (A) and longitudinal (B) US images of G1: diffusely enlarged thyroid gland with normal echogenicity, similar to a normal thyroid gland and hyperechoic compared with the anterior strap muscle (asterisk) without heterogeneity.   Transverse (A) and longitudinal (B) US images of G4: diffusely enlarged thyroid gland with diffuse heterogeneity involving more than one third of the gland (area with dashed lines) with marked hypoechogenicity that is more hypoechoic than the anterior strap muscle (asterisk).

Figure 5
Flowchart of overall study population enrollment.

Figure 6
A 16-year-old female patient: initial US (A and B) was classi ed as G2. On follow-up US after two years, hypoechogenicity and heterogeneity were aggravated and classi ed as G4 (C and D). Simultaneously, TFT results were also aggravated (euthyroidism to overt hyperthyroidism).

Figure 7
A 14-year-old female patient: initial US (A and B) was classi ed as G4. On follow-up US after one year, hypoechogenicity and heterogeneity were improved and classi ed as G3 (C and D). Simultaneously, TFT results were also improved (overt hypothyroidism to euthyroidism).