High prevalence of thyroid hormone autoantibody and low rate of thyroid hormone detection interference

Abstract Objective Thyroid hormone autoantibody (THAb) is a common antibody in autoimmune disease and can interfere with the detection of thyroid hormone (TH). There was no research reporting the prevalence of THAb in Chinese and the rate of THAb interfering with TH detection. Methods We collected 114 patients with autoimmune thyroid disease (AITD) (Hashimoto's thyroiditis, 57 cases; Graves’ disease, 57 cases), 106 patients with nonthyroid autoimmune diseases (NTAID), and 120 healthy subjects. According to the presence or absence of thyroid antibodies, patients with NTAID were divided into two groups: NTAID‐AITD and NTAID groups. Radioimmunoprecipitation technique was used to detect THAb in all subjects. TH was detected on Abbot and Roche platforms in patients with positive THAb. Results The prevalence of THAb was 22.8% in Hashimoto's thyroiditis and 45.6% in Graves’ disease. The prevalence of THAb in AITD group was lower than that in NTAID or NTAID‐AITD groups (34.2% vs. 61.5%, p = 0.014; 34.2% vs. 71.3%, p < 0.01). Among total 98 patients with positive THAb, TH levels of 9 patients were falsely elevated (9.18%). Conclusion The prevalence of THAb in AITD patients was lower than that in NTAID patients. Although THAb had a high frequency in various autoimmune diseases, the prevalence of THAb interfering with TH detection was only 9.18%.

common THAb. The prevalence of THAb was approximately 1% in the general population. 2 A study in Italy revealed that the prevalence of THAb was 20% in Hashimoto's thyroiditis (HT) and 32% in Graves' disease (GD). 3 It was believed that the prevalence of THAb in autoimmune thyroid disease (AITD) was lower than that in nonthyroid autoimmune diseases (NTAID). [3][4][5][6] The mechanism of the presence of THAb in NTAID is still unclear. The above studies were conducted in Italy, and other countries have not carried out any epidemic study on THAb. THAb can bind to thyroid hormone analogs in the detection system, resulting in a falsely elevated level of thyroid hormone. Since THAb was firstly discovered in 1956, 7 more than ten cases of THAb interfering with thyroid hormone detection have been published. [8][9][10][11][12][13][14][15][16][17] THAb is a common thyroid autoantibody in autoimmune diseases, but the rate of thyroid hormone detection interference in THAbpositive patients has not been reported.
A 75-year-old woman who suffered from Hashimoto's thyroiditis for 18 years visited our endocrine clinic. Previous results of thyroid function showed that the thyroid-stimulating hormone (TSH) and free thyroxine (FT4) levels were simultaneously higher than the upper limit of normal value in multiple periods. We suspected that there was interference in FT4 detection, because when we detected FT4 on another platform, the result was significantly different (Roche: 12.7 pmol/L; Abbot: 7.09 pmol/L). THAb is the most common substance interfering with thyroid hormone detection.
So, we suspected that THAb interfered with the detection of FT4.
However, there was no report on THAb in China, and the normal value of THAb in Chinese was unknown.
Collectively, the current study aimed to establish the reference value of THAb in the Chinese population, determine the prevalence of THAb in Chinese patients with AITD, and evaluate the rate of THAb interfering with thyroid hormone detection in patients with positive THAb.

| Patients' selection
We collected 114 patients with AITD who visited in the endocrine clinic of Shandong Provincial Hospital from 2019 to 2020 (57 cases for HT and 57 cases for GD), in which 76 patients were newly diagnosed and 38 patients were under treatment. The exclusion criteria included the following: NTAID, thyroid surgery, and malignancy. We    20 Other collagenases included connective tissue disease (CTD, 7 cases), vasculitis (7 cases), ankylosing spondylitis (4 cases), osteoarthritis (4 cases), and dermatomyositis (3 cases). Patients were defined as having those collagenases when those collagenases were listed in the diagnoses of discharge letters from hospital. The diagnosis of these collagenases was made by two experienced rheumatologists. NTAID patients were divided into NTAID-AITD and NTAID groups according to the absence or presence of thyroid antibodies.
Twenty microliters of this mixture was then incubated with 150 μl anti-human IgM-Agarose or Protein G, both prediluted 1:10 with saline containing bovine serum albumin (BSA) (Sigma) at a final concentration of 0.5%. After 24-h incubation at 4℃, tubes were centrifuged at 2,000× g for 1 min and the supernatant was aspirated. Finally, the radioactivity of immunoprecipitation was detected. The percentage of THAb was equal to total radioactivity divided by immunoprecipitation radioactivity.

| Thyroid function and autoantibody measurements
Thyroid function and autoantibodies were detected by electro-

| Statistical analysis
Continuous variables were expressed as the mean (M) ± standard deviation (SD). The Kolmogorov-Smirnov test was used to examine whether the data of THAb in the healthy obeyed Gaussian distribution. Categorical variables were expressed by percentages.
Categorical variables were compared by the chi-squared test and

Fisher tests. Continuous variables were compared by t tests and
Kruskal-Wallis test. SPSS version 25 was applied in statistical analysis. p < 0.05 indicated that the difference was statistically significant.

| Demographic characteristics and the prevalence of THAb in AITD, NTAID, and NTAID-AITD groups
There was no significant difference in the proportion of females among the three groups. The mean age of the NTAID-AITD group was higher than that of the AITD group (47.77 ± 14.80 years vs. 38.82 ± 12.25 years, p = 0.002). There was no significant difference in mean age between the NTAID-AITD group and the NTAID group or between the AITD group and the NTAID group. The prevalence of THAb in AITD group was lower than that in NTAID or NTAID-AITD groups (34.2% vs. 61.5%, p = 0.014; 34.2% vs. 71.3%, p < 0.01).

| Distribution of different kinds of THAb in HT, GD, SLE, RF, pSS
The overall prevalence of THAb in HT was lower than that in SLE, RF,

TA B L E 1
Demographic characteristic and the prevalence of different kinds of THAb in AITD, NTAID, and NTAID-AITD groups prevalence of T3-T4Ab was higher in SLE than that in HT (p = 0.012).

RA and pSS groups had a higher prevalence of isolated IgM-Ab than
HT group (p < 0.001; p < 0.001). SLE group had a higher prevalence of IgG and IgM-Ab than HT group (p < 0.001). GD group had a lower prevalence of T3-T4Ab and IgG and IgM-Ab than SLE group (p = 0.044; p = 0.010) and had a lower prevalence of overall THAb and isolated IgM-Ab than pSS group ((p = 0.025; p = 0.016) (Figure 1).

| DISCUSS ION
In our research, we calculated the normal value of THAb: 0-4.45% Although THAb can bind to thyroid hormone, thyroid hormone is a hapten and cannot directly elicit the production of antibody. In

1972, Yukio et al. reported that rabbits immunized with mild dena-
tured thyroglobulin (Tg) produced antibodies that could bind to thyroid hormone in the gamma-globulin region. 22 In 1997, Benvenga found that iodinated, heterologous thyroglobulin could lead to the production of THAb in patients with FNAB, and THAb had a positive relationship with TgAb. 1 In this research, Benvenga also demonstrated that the prevalence of THAb in HT was 10-fold higher than that in nonautoimmune thyroid diseases. According to the aftermentioned studies, it was generally believed that THAb was generated due to the leakage of Tg and was the subtype of TgAb. However, in our study, there was no difference in the prevalence of THAb between NTAID and NTAID-AITD groups. Meanwhile, the distribution of THAb in NTAID and AITD groups was significantly different. a higher proportion of single THAb, while NTAID   patients had a higher proportion of composite THAb (T3-T4Ab, IgG, and IgM-Ab). Previous studies have reported that the prevalence of THAb in NTAID was much higher than that in AITD. The prevalence of THAb is 50% in SS, 26% in RA, 3 92.3% 5 in type 1 diabetes (DM1), and 97% in vitiligo. 4 Most of these patients were euthyroid and thyroid autoantibody negative. Researches on patients with FNAB 1 and patients receiving tyrosine kinase inhibitor treatment 23

indicated that
THAb was the earliest thyroid autoantibody occurring in early thyroid damage. Furthermore, a study reported that RA patients with positive THAb would develop to autoimmune hypothyroidism in the following 10 years. 3 The above studies implied that the production of THAb required the damage of the thyroid rather than the presence of TgAb. Collectively, we speculated that the mechanism of the production of THAb in AITD and NTAID was different. In NTAID, thyroid damage comes from systemic immune injury rather than AITD.
In NTAID, immune complexes can deposit in the vessel wall of the thyroid. Complement activation will cause vascular occlusion, which will lead to insufficient oxygen supply in thyroid tissue and eventually result in thyroid injury. Research on THAb and DM1 found that THAb had a relationship with microangiopathy. 5  Although the prevalence of immunoassay interference was less than However, this study also had some limitations. The sample size of patients in this study was small, and there might be selection bias. On the other hand, the use of glucocorticoids and immunosuppressants (about 56.6% of NTAID patients had a history of glucocorticoid or immunosuppressant use in recent 6 months, but the duration and dose of glucocorticoid use were unknown) was not collected.

| CON CLUS ION
The prevalence of THAb in the general population was lower than that in AITD patients, and the prevalence of THAb in AITD patients was lower than that in NTAID patients. Although THAb has a high frequency in various autoimmune diseases, the rate of THAb interfering with thyroid hormone detection was only 9.18%.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.