Thyrotropin Receptor Antibodies and Graves' Disease, a Side-Effect of ...

0021-972X/97/$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1997 by The Endocrine Society

Vol. 82, No. 9 Printed in U.S.A.

Thyrotropin Receptor Antibodies and Graves’ Disease, a Side-Effect of 131I Treatment in Patients with Nontoxic Goiter* ¨ S, BIRTE NYGAARD, JENS HELMER KNUDSEN, LASZLO HEGEDU ANNEGRETE VEJE CAND SCIENT, AND JENS ERIK MØLHOLM HANSEN Department of Internal Medicine and Endocrinology, Herlev Hospital (B.N., J.H.K., J.E.M.H.), Herlev; the Department of Internal Medicine and Endocrinology, Odense University Hospital (L.H.), Odense; and the Department of Clinical Physiology and Nuclear Medicine, Holbæk Hospital (A.V.C.S.), Holbæk, Denmark ABSTRACT The use of 131I treatment in patients with benign nontoxic goiter is increasing, and the described side-effects are few. In this paper we describe appearance of TSH receptor antibodies (TRAb) and concomitant development of hyperthyroidism as a side-effect of 131I treatment in patients with nontoxic goiter. In this retrospective study, 191 consecutive patients with 131I-treated nontoxic goiter are described. Nine patients (5%) developed hyperthyroidism 3 months after 131I treatment, and 5 patients (3%) developed radiation thyroiditis within the first month. Frozen sera were analyzed for thyroid peroxidase antibodies (anti-TPO) in 130 patients before 131I treatment. In 21% of these, serum levels of anti-TPO were over 200 U/mL. The complication frequency of Graves’-like hyperthyroidism and hypothyroidism was 51% in patients with elevated anti-TPO (n 5 27) and 15% in patients with normal serum anti-TPO levels (P , 0.00005). TRAb, anti-TPO, and thyroglobulin were followed in patients developing

R

ADIATION thyroiditis is a well known complication when thyroid diseases are treated with radioiodine (131I). Manifestations usually commence a few days after 131I treatment and consist of neck pain, dysphagia, thyroid tenderness, and transient hyperthyroidism (1, 2). In a few patients with toxic multinodular goiter, the appearance of TSH receptor antibodies (TRAb) and the onset of hyperthyroidism after 131I treatment, have been described (3– 6). TRAb are seen in about 85% of patients with Graves’ disease and are thought to be involved in the initiation of hyperthyroidism in Graves’ disease (7). TRAb have also been reported in patients with other thyroid diseases, but are uncommon in nontoxic goiter (7). 131 I therapy is increasingly used in patients with benign nontoxic goiter (8 –14). Therefore, we found it pertinent to describe the occurrence of hyperthyroidism in patients with nontoxic goiter, not as a result of radiation thyroiditis, but most likely caused by TRAb production triggered by 131I treatment. Received March 17, 1997. Revision received May 14, 1997. Accepted June 3, 1997. Address all correspondence and requests for reprints to: Birte Nygaard, M.D., Department of Internal Medicine and Endocrinology F 112, Herlev Hospital, Herlev Ringvej, DK-2730 Herlev, Denmark. * This work was supported by grants from Agnes and Knut Mørks Foundation. The data were published in part as an abstract in Eur J Endocrinol 130(Suppl 1):7, 1994.

hyperthyroidism or radiation thyroiditis and in 10 control patients remaining euthyroid. At the time of 131I treatment, all patients had serum TRAb values within the normal range. Three months after administration of 131I, the patients developing hyperthyroidism had a transient extensive rise in serum TRAb and anti-TPO levels parallel to a rise in the serum free T4 index. In patients developing radiation thyroiditis, serum TRAb values were normal. In control patients, serum TRAb and anti-TPO values were both within the normal range throughout the observation period. In conclusion, hyperthyroidism can be triggered by 131I in patients with nontoxic goiter, not only related to radiation thyroiditis but also as a Graves’-like hyperthyroidism induced by TRAb. Elevated antiTPO pretreatment is a marker of an increased risk of side-effects to 131 I treatment in nontoxic goiter. (J Clin Endocrinol Metab 82: 2926 – 2930, 1997)

Subjects and Methods Patients The study was retrospective, comprising 191 consecutive 131I-treated patients with nontoxic goiter (21 men and 170 women; median age, 52 yr; range, 30 – 83 yr). We have previously described the first 69 patients (8). Nontoxic goiter was defined as the presence of a goiter combined with free T4 index (FT4I) and free T3 index (FT3I) within the normal range; no discrimination was made between patients with or without thyroid autoantibodies. The patients were evaluated before and 1, 3, 6, and 12 months after treatment, and TSH, T3, T4, and T3 uptake ratio were determined. In patients developing hyperthyroidism (patients 1–9); in patients developing radiation thyroiditis defined as neck pain, dysphagia, thyroid tenderness, and transient hyperthyroidism (patients 10 –14), and in 10 randomly selected control patients with normal FT4I and FT3I, we measured thyroid peroxidase antibodies (anti-TPO), TRAb, and thyroglobulin (Tg) for 12–24 months after 131I treatment. Patient data are given in Table 1. The results of thyroid scintiscans before 131I treatment are shown in Table 1. [Scintiscan at the time of hyperthyroidism was performed in only one patient (no. 5), and this demonstrated unaltered heterogeneous uptake.] None of the patients had eye symptoms related to the hyperthyroid phase.

Methods Frozen (220 C) sera from consecutive patients treated with 131I for nontoxic goiter were analyzed. Thyroid function was evaluated by measuring serum levels of T4, T3, and TSH and T3 uptake ratio. Free T4 and T3 indexes were calculated by multiplying the T4 level by the result of the T3 uptake test. Details of the methods have been described previously (8). TRAb were determined by a receptor assay (TRAk-assay,

2926

I

2927

Results

Patients 1–9 developed hyperthyroidism 3 months after treatment; patients 10 –14 developed radiation thyroiditis. Treated with L-T4.

FT4 index

a

370,259 –740 50, 39 – 69

131

Henning Berlin, Berlin, Germany; normal range, ,9 U/L; interassay variation, 7%). Anti-TPO were measured by the LUMI-test from Henning Berlin (normal range, ,200 U/mL; interassay variation, 10%). Tg was measured by the Dynotest Tg from Henning Berlin (normal range, 2–70 ng/mL; interassay variation, 3%). 131 I was given at a dose of 3.7 megabecquerels/g total thyroid mass [evaluated by ultrasound (15)], corrected to a 100% 24-h 131I uptake. The maximum dose given was 740 megabecquerels. Before treatment, a 99m Tc thyroid scan was performed in all patients with a g-camera. For statistical evaluation, Mann-Whitney’s test, Fisher’s exact test, and Wilcoxon’s test were used. P , 0.05 was considered statistically significant.

3, ,9 1, ,9 ,50, ,50 – 61 40, 19 –111 55, 37–140

(n 5 10) (n 5 10)

0.20,0.02–1.1

2 4 1 1 1 1 111

740 222 444

12 13 14 Ten patients who remained euthyroid Median, range

71, F 45, F 36, F 8F/2M

Multinodular Multinodular Multinodular 7 multinodular, 3 diffuse

140

0.01 0.09 0.18

12 3 3 6 2 3 60 22 108 100 37 116 Multinodular Multinodular Diffuse Multinodular F F F F 31, 56, 46, 50, 8 9 10 11

370 740 259 740

40 53

F F F F 33, 50, 48, 48, 4 5 6 7

370 370 555 370

0.39 0.27 1.50 1.20

988 483 6347 (1 month) 30 2605 50

120 368 57 ? (38 after 1 yr) 3 7 6 5 42 19 58 38

0.73 0.01a 0.18 0.22

330 4081 554 2000

35 324 30 6 4 3 646 50 48 53, F 56, F 35, F 1 2 3

555 481 444

Multinodular Multinodular Diffuse with a small solitary adenoma Multinodular Multinodular Multinodular Multinodular

52 49 83

39 23 26

0.48 0.97 0.02

TRAb at the time of treatment (U/L) Scintiscan I dose (MBq)

131

Age (yr), sex Patient no.

TABLE 1. Patient data of

131

I-treated patients with nontoxic goiter

Thyroid vol at the time of treatment (mL)

Thyroid vol 1 yr after treatment (mL)

TSH at the time of treatment (U/mL)

Anti-TPO at the time of treatment (U/mL)

Max. TRAb value (U/L)

GRAVES’ DISEASE TRIGGERED BY

All patients had normal FT4I and FT3I at the time of treatment. In five patients, classical symptoms of radiation thyroiditis were seen within 1 month (patients 10 –14). The levels of FT4 ranged from 150 –233 arbitrary units after 1 month in four patients. In patient 11, no blood samples were obtained after 1 month, but the described symptoms were classic. FT4 levels were normalized after 3 months in all five patients. Nine patients (no. 1–9) had normal FT4 concentrations 1 month after 131I treatment, but developed hyperthyroidism after 3 months. Symptoms of hyperthyroidism were minor in seven patients (no. 2, 3, and 5–9) and severe in two patients (no. 1 and 4). Two patients became hyperthyroid and resolved spontaneously without therapy in 15–18 months (no. 2 and 5), five patients were given antithyroid drugs and remained euthyroid off medication 6 –18 months after 131I treatment (no. 3 and 6 –9), one patient’s hyperthyroidism was treated with a second dose of 131I (no. 1), and one patient had a subtotal thyroidectomy (no. 4). Figure 1 shows changes in serum FT4I related to 131I treat-

FIG. 1. Changes in serum FT4I related to time after 131I treatment of nontoxic goiter in 6 patients (no. 1– 6) developing hyperthyroidism (median, black line; range, CCC) and range of FT4I in 10 control patients (...........).

2928

JCE & M • 1997 Vol 82 • No 9

NYGAARD ET AL.

ment in patients 1– 6 and in 10 control patients. Data from patients 1, 3, 4, and 5 are shown at 18 and 24 months. Patients 7–9 are not illustrated in Figs. 1– 4 because of some missing data. TRAb

Changes in serum TRAb levels in patients 1– 6 and the 10 control patients are shown in Fig. 2. TRAb values in patients developing hyperthyroidism were, although within the normal range, significantly higher (median, 5 U/L; range, 3–7) than those in the control patients (median, 1 U/L; range, 0 – 4; P , 0.001). Three months after 131I treatment there was a considerable increase in serum TRAb values. Serum TRAb values decreased during a period of 18 months, with a slightly slower decrease in the 2 patients who did not receive antithyroid drugs (no. 2 and 5). Patient 1, who received a second 131I treatment, showed a secondary rise in serum TRAb levels, but this was not followed by recurrence of hyperthyroidism. In patient 7, the serum TRAb value measured before and 1 month after treatment was 4 U/L; again after 12 months of treatement, it was 38 U/L. In patients 10 –14, serum TRAb concentrations were within the normal range. Anti-TPO

Serum anti-TPO levels were measured to evaluate changes in another thyroid autoantibody. Changes in serum anti-TPO values were parallel those in serum TRAb, i.e. no changes after 1 month, a considerable increase after 3 months, and no changes in the control group (Fig. 3). Of the patients developing radiation thyroiditis, two had normal anti-TPO values

FIG. 2. Changes in serum TRAb concentration related to time after 131 I treatment of nontoxic goiter in 6 patients (no. 1– 6) developing hyperthyroidism (median, black line; range, CCC) and range of serum TRAb in 10 control patients (...........). A logarithmic scale was used.

without any changes after 131I treatment, and three had elevated values with an increase after 3– 6 months of treatment. To evaluate whether there was a correlation between positive anti-TPO and development of side-effects to 131I treatment, anti-TPO were measured before treatment. Blood samples were available in 130 patients. Anti-TPO levels less than 200 U/mL were found in 103 patients (79%; only 2 patients had values between 100 –200 U/mL). In patients with positive anti-TPO levels, 6 of 27 (22%) developed Graves’-like hyperthyroidism compared to 2 of 103 patients (2%) with normal anti-TPO levels (P 5 0.002). Eight of 27 with positive anti-TPO levels developed hypothyroidism requiring l-T4 (29%) compared to 13 of 103 (13%) without anti-TPO (P 5 0.08). Three of 27 with positive anti-TPO developed radiation thyroiditis (11%) compared to 2 of 103 (2%) without these antibodies (P 5 0.12). Tg

Figure 4 shows changes in serum Tg concentrations. There was a significant increase (P 5 0.04) in serum Tg levels 1 month after treatment in the patients who developed hyperthyroidism, but due to the wide range in the control group (8- 500 ng/mL), no difference between these patients and the control group was seen. Median Tg levels in the control group increased during the first 3 months, but the changes were not statistically significant. Discussion 131

I therapy has recently been described as an attractive alternative to surgery in the treatment of benign nontoxic goiter leading to an approximately 50% reduction in thyroid volume within 1–2 yr (8 –14). In view of this, we found it pertinent to focus attention on possible side-effects of this

FIG. 3. Changes in serum anti-TPO values related to time after 131I treatment of nontoxic goiter in 6 patients (no. 1– 6) developing hyperthyroidism (median, black line; range, CCC) and range of serum anti-TPO in 10 control patients (.. .. ...). A logarithmic scale was used.

GRAVES’ DISEASE TRIGGERED BY

FIG. 4. Changes in serum Tg level related to time after 131I treatment of nontoxic goiter in 6 patients (no. 1– 6) developing hyperthyroidism (median, black line; range, CCC) and serum Tg in 10 control patients (median, gray line; range, .. .. ..).

treatment. In the present paper we describe 9 patients from a consecutive group of 191 patients (5%) who had normal serum TRAb values at the time of 131I treatment as well as 1 month later, but increased serum TRAb values and hyperthyroidism after 3 months. Similar findings were reported by Kay et al. in a study of 14 patients (16). Furthermore, a Graves’-like hyperthyroidism commencing several months after radioiodine therapy for toxic nodular goiter associated with appearance of TRAb has recently been described in a few patients (3– 6). When Graves’ disease is treated with 131I, a transient rise in TRAb levels is seen after approximately 3 months. Also, an increase in serum levels of TgAb and microsomal antibodies has been described (7, 17, 18). These observations have been interpreted as an immunological response caused by the release of thyroid antigens from destroyed follicular cells (7). The increase in circulating antigens cannot be measured, but is believed to correspond to changes in serum Tg levels after 131I therapy. Therefore, we measured serum Tg as a marker of antigen release. We have previously seen a significant increase in serum Tg in the first days after 131I treatment of nontoxic goiter (unpublished data). In the present study we did not measure serum Tg until 1 month after treatment. Therefore, we may have overlooked a larger early increase in serum Tg. No obvious differences between serum Tg in the control group and the patients developing hyperthyroidism could be demonstrated. In this retrospective study we did not measure cAMP production from thyroid tissue, but only TRAb, and we only had the possibility to obtain a thyroid scan in one patient to be completely sure that we were dealing with a Graves’-like disease. However, the concomitant increase in FT4I and TRAb first seen 3 months after 131I treatment, the fact that hyperthyroidism in some patients persisted for over a year,

131

I

2929

and the different antibody profile seen in patients with a classical radiation thyroiditis make this explanation plausible. The fact that TRAb values were significantly higher, although within the normal range, in the patients developing hyperthyroidism could indicate that these patients had minimal preexisting Graves’ disease that was aggravated by 131I. Other environmental factors that cause destruction of the follicular cells have been described as leading to development of Graves’ disease, i.e. external radiation for nonthyroidal disease (19), surgical manipulation of the thyroid during parathyroidectomy (20), or subacute thyroiditis (21). Such cases are rare and may be coincidental, or it could be the environmental influence that triggers a Graves’-like autoimmune response to the TSH receptor. Thyroid autoantibodies have previously been described as a risk factor for the development of hypothyroidism after 131I therapy for single hot thyroid nodules (22). Correspondingly, we found that Graves’-like hyperthyroidism and hypothyroidism after 131I treatment were more likely to appear in patients with an elevated serum anti-TPO level (51%) than in those without (15%). Although we do not have a histological diagnosis, these patients with elevated anti-TPO could be suffering from Hashimoto’s thyroiditis. We, therefore, suggest that 131I treatment should only be used in highly selected patients with elevated anti-TPO. Although Graves’ disease has an autoimmune origin, the pathogenesis of toxic nodular goiter is believed to be of nonautoimmune origin. Progression from nodular nontoxic goiter to hyperthyroidism is thought to be caused by an increasing autonomy in the multinodular goiter (23). However, in the present study we have described a Graves’-like hyperthyroidism based on autoimmune phenomena in patients with typical nodular nontoxic goiter. In conclusion, treatment with 131I in patients with nontoxic goiter does not usually lead to increased serum TRAb levels. In some predisposed individuals, an immunological response, probably triggered by the release of Tg or other antigens from the thyroid, causes a transient (months) rise in TRAb and a concomitant hyperthyroidism starting 3 months after 131I treatment. This condition is different from the radiation thyroiditis, which can be observed during the first month after 131I without detectable TRAb in serum. When treating nontoxic goiter with 131I, one should be aware of this possible side-effect. Patients with high serum anti-TPO levels seem to be at an increased risk and require frequent evaluation. References 1. Cooper DS. 1991 Treatment of thyrotoxicosis. In: Braverman LE, Utiger RD, eds. Werner and Ingbar’s the thyroid, a fundamental and clinical text, 6th ed. Philadelphia, New York, London, Hagerstown: Lippincott; 887–916. 2. Mazzaferri EL. 1991 Carcinoma of follicular epithelium. Radioiodine and other treatment and outcomes. In: Braverman LE, Utiger RD, eds. Werner and Ingbar’s the thyroid, a fundamental and clinical text, 6th ed. Philadelphia, New York, London, Hagerstown: Lippincott; 1138 –1165. 3. Smyth PPA, Neyan D, McMullan NM, Smith DF, McKenna TJ. 1988 Sequential presentation of a case of hyperthyroidism with autonomously functioning nodules and Graves’ disease in the presence of IgG thyroid stimulators. Acta Endocrinol (Copenh). 118:474 – 478. 4. Boddenberg B, Voth E, Schicha H. 1993 Immunogene hyperthyreose nach radio-jod-ablation einer fokalen autonomie. Nucl Med. 32:18 –22. 5. Chiovato L, Ferruccio S, Vitti P, Bendinelli G, Pinchera A. 1994 Appearance of thyroid stimulating antibody and Graves’ disease after radioiodine therapy for toxic nodular goitre. Clin Endocrinol (Oxf). 40:803– 806.

2930

NYGAARD ET AL.

6. Nygaard B, Faber J, Veje A, Hegedu¨s L, Hansen JM. 1995 Appearance of Graves’like disease after radioiodine therapy for toxic as well as non-toxic multinodular goitre [Letter]. Clin Endocrinol (Oxf). 43:129 –130. 7. Bech K. 1983 Immunological aspects of Graves’ disease and importance of thyroid stimulating immunoglobulins [Thesis]. Acta Endocrinol (Copenh). 254(Suppl):1–35. 8. Nygaard B, Hegedu¨s L, Gervil M, Hjalgrim H, Søe-Jensen P, Hansen JM. 1993 Radioiodine treatment of multinodular non-toxic goitre. Br Med J. 307:828 – 832. 9. Huysmans DAKC, Hermus ARMM, Corstens FHM, Barentsz JO, Kloppenborg PW. 1994 Large, compressive goiters treated with radioiodine. Ann Intern Med. 121:757–762. 10. Kay TWH, d’Emden MC, Andrews JT, Martin FIR. 1988 Treatment of nontoxic multinodular goiter with radioactive iodine. Am J Med. 4:19 –22. 11. Klein B, Klein E, Horster FA. 1989 Ergebnisse der fraktionierten radiojodtherapie bei 696 hyperthyreosen und 690 blanden strumen. Nucl Med. 28:129 –136. 12. Hegedu¨s L, Hansen BM, Knudsen N, Hansen JM. 1988 Reduction of size of thyroid with radioactive iodine in multinodular non-toxic goitre. Br Med J. 297:661– 662. 13. Wesche MF, Tiel-v-Buul MM, Smits NJ, Wiersinga WM. 1995 Reduction in goiter size by 131I therapy in patients with non-toxic multinodular goitre. Eur J Endocrinol. 132:86 – 87. 14. Verelst J, Bonnyns M, Glinoer D. 1990 Radioiodine therapy in voluminous multinodular non-toxic goitre. Acta Endocrinol (Copenh). 122:417– 421. 15. Hegedu¨s L, Perrild H, Poulsen LR, et al. 1983 The determination of thyroid

16. 17. 18.

19.

20. 21. 22.

23.

JCE & M • 1997 Vol 82 • No 9

volume by ultrasound and its relationship to body weight, age and sex in normal subjects. J Clin Endocrinol Metab. 6:260 –263. Kay TWH, Heyma P, Harrison LC, Martin FIR. 1987 Graves’ disease induced by radioactive iodine. Ann Intern Med. 107:857– 858. McGregor AM, McLachlan SM, Smith BR, Hall R. 1979 Effect of irradiation on thyroid-autoantibody production. Lancet. 2:442– 444. Feldt-Rasmussen U, Bech K, Date J, Petersen PH, Johansen K, Madsen SN. 1982 Thyroid stimulating antibodies, thyroglobulin antibodies and serum proteins during antithyroid treatment with radioiodine or propylthiouracil of Graves’ disease. Allergy. 32:161–172. Williams ED. 1991 Biologic effects of radiation on the thyroid. In: Braverman LE, Utiger RD, eds. Werner and Ingbar’s the thyroid, a fundamental and clinical text, 6th ed. Philadelphia, New York, London, Hagerstown: Lippincott; 421– 436. Walfish PG, Caplan D, Rosen IB. 1992 Postparathyroidectomy transient thyrotoxicosis. J Clin Endocrinol Metab. 75:224 –227. Bennedbæk FN, Gram J, Hegedu¨s L. 1996 The transition of subacute thyroiditis to Graves’ disease as evidenced by diagnostic imaging. Thyroid. 6:457– 459. Mariotti S, Martino E, Francesconi M, et al. 1986 Serum thyroid autoantibodies as a risk factor for development of hypothyroidism after radioactive iodine therapy for single thyroid “hot” nodule. Acta Endocrionol (Copenh). 113:500 –507. Studer H, Peter HJ, Gerber H. 1989 Natural heterogeneity of thyroid cells: the basis for understanding thyroid function and nodular goiter growth. Endocr Rev. 10:125–135.