Natural killer cell activity in patients with Graves' disease and Hashimoto's thyroiditis.

Natural killer (NK) cell activity of peripheral blood lymphocytes (PBL) against k562 human tumor cell targets was studied in patients with Graves' disease and Hashimoto's thyroiditis. NK activity was measured in a standard 4-hour 51chromium (Cr) release assay. Cytotoxicity was expressed as lytic units (LU)/10(6) PBL. Significantly decreased NK cell activity was demonstrated in both groups of patients, with mean (+/- SE) lytic units of 10.3 (+/- 9.1) and 13.3 (+/- 10.3) for patients with Graves' disease and Hashimoto's thyroiditis, respectively, compared with 36.0 (+/- 26.3) for age- and sex-matched normal subjects. When patients with Graves' disease were analyzed according to their thyroid status; NK activity was significantly depressed in (1) hyperthyroid patients before treatment; (2) hyperthyroid patients receiving antithyroid therapy; and (3) euthyroid patients receiving antithyroid therapy, compared with normal subjects. Graves' disease patients who were hypothyroid after radioactive iodine therapy or thyroidectomy had normal NK activity. No significant differences between hyperthyroid and euthyroid patients or between hypothyroid patients and normal subjects were demonstrated. NK activity in patients with Graves' disease did not correlate with serum levels of thyroxine, the presence or severity of ophthalmopathy, or titers of serum thyroid antibodies. In patients with Hashimoto's thyroiditis there was no correlation between NK activity and goiter size, titers of antithyroid antibodies, or thyroid status. These findings suggest that depression of NK activity in both disorders is secondary to abnormalities of thyroid hormone secretion, although an effect of the underlying autoimmune reactions has not been excluded.

Treatment of Graves' hyperthyroidism with radioiodine: results of a prospective randomized study.

In this paper, we report the results of a randomized prospective study on radioiodine treatment in patients with Graves' hyperthyroidism. Complete data were obtained in group 1 from 98 patients 6 months after application of a standard activity of 15 mCi (555 MBq) of 131I and in group 2 from 107 patients who received a target dose of 100 Gy. In group 1, the overall success rate was 71%, but the results in the subgroups clearly were related inversely to the thyroid volume, ranging from 100% in patients with a thyroid volume < 15 mL to about 20% in those with a thyroid size of > 60 mL. In contrast, patients who received a target of 100 Gy showed very similar results, with success rates of about 40-50% in all but one subgroup. Only patients with a thyroid volume < 15 mL had a success rate of about 80%. But due to an incidental increase of uptake and/or effective half-time from the test to the therapy activity, this subgroup received a target dose of about 160 Gy. Additional calculation of the actual target dose in group 1 (standard activity) showed that, with a dose of 200 Gy, a success rate of 80% was obtained. Also, the thyroid volume reduction was related inversely to the target dose. Because the literature is abundant, only a restricted number of references are discussed that are either in agreement with our results or in sharp contrast to them. The reason for these discrepant results might be the difference in the scheme of pretreatment or the different alimentary iodine supply between, for example, Great Britain and the United States on the one hand and Germany on the the other hand.

[Treatment of euthyroid struma. Comparable volume reduction with 400 micrograms iodine, 100 micrograms levothyroxine combined with 100 micrograms iodine or individually dosed levothyroxine]. / Behandlung der euthyreoten Struma. Vergleichbare Volumenreduktion mit 400 micrograms Jodid, 100 micrograms Levothyroxin kombiniert mit 100 micrograms Jodid oder individuell dosiertem Levothyroxin.

AIM: In patients with euthyroid goitre, the efficacy of treatment with 400 micrograms iodine and 100 micrograms levothyroxine combined with 100 micrograms iodine was compared to that of the previous standard of therapy, individually dosed levothyroxine. PATIENTS AND METHODS: A total of 78 patients presenting with euthyroid diffuse goitre (> or = 25 ml) were prospectively enrolled, randomised and treated for 6 months. The course of thyroid volume was followed using thyroid volumetry. RESULTS: Data of 69 patients were included in the final evaluation (57 women, 12 men, age 31 +/- 1 years, thyroid volume 31.5 +/- 1.4 ml, 23 per treatment group). In the patients treated with individually dosed levothyroxine, the thyroid volume decreased by about 39% (95%-confidence limit [CL]-31% to -41%). However, the volume reductions achieved in the patients treated with 400 micrograms iodine or 100 micrograms levothyroxine/100 micrograms iodine were not significantly different (p = 0.35, variance analysis, mono-iodine -34%, 95%-CL -29% to -43%, 100 micrograms levothyroxine/100 micrograms iodine -39%, 95%-CL -32% to -45%). CONCLUSIONS: In patients with euthyroid diffuse goitre, treatment with mono-iodine or combination of levothyroxine with iodine should have principally the same status as the previous standard of therapy, individually dosed levothyroxine. In the view of the authors, its preferential treatment with mono-iodine appears most reasonable.

[The prevention of the recurrence of endemic goiter. The efficacy of a once-a-week dose of 1.53 mg iodide]. / Rezidivprophylaxe der endemischen Struma. Effektivität der einmal wöchentlichen Gabe von 1,53 mg Jodid.

OBJECTIVE: To investigate the efficacy of 1.53 mg iodide administered once weekly in the prophylaxis of goitre recurrence in patients with endemic euthyroid goitre after initial treatment to reduce the goitre size. PATIENTS AND METHODS: 46 consecutive patients who had undergone initial L-thyroxine, iodide or combined treatment were included in the prospective study. Thyroid volume was measured sonographically at the beginning as well as 6 and 18 months later. An increase in thyroid volume of more than 15% was taken as recurrence. RESULTS: The study was concluded in 41 patients. During the prophylaxis mean thyroid volume increased from initially 21.7 +/- 9.9 ml to 22 +/- 10.9 ml after 6 months and to 24.5 +/- 12.1 ml after 18 months (P < 0.01). While thyroid volume remained unchanged in at least two thirds of patients, a recurrence occurred in 32% (n = 13; from initially 22.7 +/- 9.7 ml, to 26.3 +/- 11.3 ml at 6 months and to 29.7 +/- 12.3 ml at 18 months). In all patients with a recurrence a doubling of the iodide dosage to twice weekly 1.53 mg for 6 months reduced thyroid volume again (mean of 25.7 +/- 10.0 ml; P < 0.01). CONCLUSIONS: In at least two third of patients a once-weekly dose of 1.53 mg iodide is a reliable means of preventing a recurrence of endemic euthyroid goitre, but in the remainder the same dose must be increased to twice weekly.

Reduction in thyroid volume after radioiodine therapy of Graves' hyperthyroidism: results of a prospective, randomized, multicentre study.

The reduction in thyroid size in 92 patients treated with radioiodine for Graves' thyrotoxicosis was monitored by ultrasound volumetry. The patients were randomly treated with either a standard 131I activity of 555 MBq or an activity calculated to deliver a thyroid dose of 100 Gy. Within 1 year after radioiodine treatment, a remarkable volume reduction of about 71% (median) (quantile 25% (Q 25) = 49%, Q 75 = 82%, n = 67) was observed. The bulk of this reduction (median 57%, Q 25 = 21%, Q 75 = 74%, n = 92) was found within the first 6 months. Statistical analysis reveals that the effect was clearly related to the thyroid dose actually achieved during therapy. The median reduction obtained 6 months after radioiodine application was 45% for < 100 Gy, 56% for 100-200 Gy and 67% for > 200 Gy (n = 28, 39, 25 respectively). Twelve months after radioiodine application, the effect became less evident: 53%, 68% and 75% respectively (n = 17, 29, 21). The higher median thyroid dose actually achieved by standard than by calculated activity (215 Gy vs. 116 Gy) explains the more pronounced volume reduction in the standard group than in the calculated group; 60% vs. 47% 6 months (n = 47, 45) after radioiodine treatments and 74% vs. 66% 12 months (n = 31,36) after radioiodine application. The relative reduction in thyroid size was just as marked in patients with large thyroids as in those with small glands. The goitre prevalence (thyroid volume > 20 mL in women and > 25 mL in men) was reduced from 73% to only 16% 1 year after radioiodine treatment. In patients with a thyroid volume of more than 60 mL, the median pretherapeutic thyroid volume of 102 mL was reduced to 29 mL. In conclusion, radioiodine treatment in Graves' hyperthyroidism sufficiently reduces thyroid volume in a dose-dependent manner. The findings of this study demonstrate that radioiodine is also an attractive mode of therapy for Graves' patients with substantial thyroid enlargement.

Susceptibility and resistance alleles of human leukocyte antigen (HLA) DQA1 and HLA DQB1 are shared in endocrine autoimmune disease.

Because particular human leukocyte antigen (HLA) DQ alleles are the major predisposing factors for type 1 diabetes mellitus (IDDM), we investigated whether they are shared by other endocrine autoimmune diseases. We, therefore, analyzed the HLA DQ genotypes of 171 patients with IDDM, 271 with Graves' disease (GD), 65 with Hashimoto's thyroiditis, 51 with postpartum thyroiditis, 53 with Addison's disease (AD), and 271 healthy controls. HLA DQA1 and DQB1 alleles were defined by polymerase chain reaction and sequence-specific oligonucleotide hybridization as well as by single strand conformational polymorphism analysis. HLA DQA1*0501 was significantly more frequent in IDDM (60%), GD (65%), and AD (70%) than in controls (43%); DQA1*0301 was significantly more frequent only in IDDM (67% vs. 30% controls). The heterozygous state DQA1*0301/*0501 was found in 9% of controls and 35% of IDDM (relative risk, 5.6). An arginine at position 52 on either DQA1 allele was significantly more frequent in patients with IDDM (94%), GD (80%), and AD (89%) compared with controls (66%). HLA DQB1*0201 and DQB1*0302 were more frequent in IDDM patients (*0201, 62% vs. 36% in controls, *0302, 59% vs. 19% controls), whereas DQB1*0602 was less frequent in IDDM (4%) and GD (18% vs. 31% of controls). In conclusion, endocrine autoimmunity has a common immunogenetic background; susceptibility is conferred by DQA1*0501 as well as an arginine at position 52 of DQA1 alleles, and protection against IDDM and GD is conferred by DQB1*0602.

Radioiodine therapy of Graves' hyperthyroidism: standard vs. calculated 131iodine activity. Results from a prospective, randomized, multicentre study.

The present prospective, randomized, multicentre study was performed to directly compare for the first time the effectiveness of a standard activity of 555 MBq 131iodine vs. an activity calculated to deliver 100 Gy for treatment of Graves' thyrotoxicosis. Therapeutic success was defined as the elimination of hyperthyroidism 6 months after radioiodine application (range 4.5-8 months). A success rate of more than 90% in eliminating hyperthyroidism was reported for both approaches, but only in retrospective investigations. Investigated prospectively, hyperthyroidism was eliminated in only 71% of the patients receiving standard activity (70/98) and 58% of those randomized for calculated activity (62/107). In the patients with standard activity, therapeutic success was inversely related to thyroid size. The rate was 100% for thyroid volumes < or = 15 mL, 95% for 16-30 mL, 68% for 31-45 mL, 44% for 46-50 mL, 20% for 61-75 mL and 25% for > or = 75 mL. In those patients with an activity calculated to deliver 100 Gy (except in those with a volume < or = 15 mL) this size/outcome dependency was almost compensated. The rates were 86%, 65%, 45%, 61%, 41% and 45%, respectively. Furthermore, detailed statistical analysis revealed a strong correlation between the success of therapy and the radiation dose actually absorbed by the thyroid. The rate was 11% for a target dose of 50 Gy, 50% for 100 Gy, 67% for 150 Gy, 80% for 200 Gy, 84% for 250 Gy, 88% for 300 Gy, 90% for 350 Gy and 93% for 400 Gy.(ABSTRACT TRUNCATED AT 250 WORDS)

Thyroid cytotoxic antibodies in atrophic and goitrous autoimmune thyroiditis.

It is unknown whether in chronic lymphocytic thyroiditis the goitrous (Hashimoto's thyroiditis) and atrophic forms (primary myxedema) are variants of the same disease or different pathogenic entities. Conventional thyroid-related autoimmune parameters are unable to separate both diseases serologically. It is assumed that cellular and humoral cytotoxic events induce gland atrophy and thus should be detectable more often in non-goitrous than goitrous autoimmune thyroiditis. We determined antibody-dependent cell-mediated cytotoxicity in 67 patients with autoimmune thyroiditis, using a 51chromium-release assay against human thyroid cells. Thyroid volume had been measured by ultrasonography. Other thyroid-specific antibodies, like TSH binding-inhibiting antibodies, TSH function-blocking antibodies, thyroglobulin antibodies and thyroid peroxidase antibodies, were determined. Cytotoxic antibody activity was 20.5% (median, range 0-54.5%) in patients with autoimmune thyroiditis and 8.3% (median, range 0-18.4%) in controls (p < 0.0001). Analysis of cytotoxicity regarding thyroid size showed a high incidence of cytotoxic antibodies in atrophic disease (median thyroid volume 6 ml), where cytotoxic antibodies were detectable in 80% versus 39% (x2 = 9.6; p < 0.0001) in goitrous disease (median thyroid volume 36 ml). The specific lysis of 30% (median; 95% confidence limit 23.9-32.9) in non-goitrous thyroiditis patients was significantly higher than in goitrous patients (16.9%; 95% confidence limit 13.2-20.4) (p = 0.0006).(ABSTRACT TRUNCATED AT 250 WORDS)

Meta-analysis evaluation of the impact of thyrotropin receptor antibodies on long term remission after medical therapy of Graves' disease.

Patients with the hyperthyroidism of Graves' disease (GDH) have a higher risk of relapse after antithyroid drug therapy (ATD) therapy when TSH receptor antibodies (TRAb) are positive, but the practical clinical implication of TRAb as a predictor for relapse is still much debated. This study was undertaken to investigate by meta-analysis the results from the literature on the use of TRAb as predictor of long term (i.e. at least 1 yr) relapse after ATD. Eighteen publications from 1975-1991 fulfilled the criteria of 1) availability of TRAb at the end of ATD treatment, 2) at least 1 yr of follow-up after ATD, 3) data presentation in a form suitable for meta-analysis, and 4) no other thyroid-related therapy during the follow-up period. The 10 prospective studies, 5 of which measured TSH binding inhibiting immunoglobulins (total n = 597) and 5 of which measured thyroid-stimulating antibodies (n = 340), were computed together because no significant differences were found. In contrast, retrospective and prospective studies differed. In the prospective studies, the odds reduction of relapse showed 65% less risk of relapse when TRAb were absent compared to that in TRAb-positive patients (P < 0.00001). The present meta-analysis has, thus, confirmed in a large number of patients (n = 1524) that absence of TRAb is significantly protective against relapse of GDH after ATD treatment. However, 25% of the patients are "misclassified," and the main questions arising from the study are, therefore, the following. 1) Is it worthwhile to use TRAb as predictor of relapse? 2) Should patients with GDH continue ATD until TRAb becomes negative, rather than for a fixed period? The available methods for TRAb do not allow sufficiently high prediction of relapse or remission after ATD for the individual patient.

Subclinical hypothyroidism and hyperlipoproteinaemia: indiscriminate L-thyroxine treatment not justified.

It is still under discussion whether subclinical hypothyroidism is a biochemical syndrome or a disease associated with an increased risk for development of vascular diseases due to lipid elevation. Therefore, we investigated lipid values in 40 patients with subclinical hypothyroidism, which is defined in terms of normal (N = 26) or slightly increased (N = 14) basal TSH values and/or an exaggerated TSH response (N = 34) to TRH (> 25 mU/l). Patients with increased lipid values were treated with L-thyroxine and reanalysed three months later. Mean levels of total cholesterol. LDL- and HDL-cholesterol and triglycerides in patients with subclinical hypothyroidism were comparable with those in normal subjects. Individual analysis, however, revealed hyperlipoproteinaemia (HL) in 22.5% of the patients investigated (HL type IIa in seven, type IV in two patients). Thyroid function was the same in affected patients as in those with normal lipid values, whereas higher age was significantly more often associated with this syndrome (p < 0.01). Treatment with L-thyroxine resulted in a significant decrease in total and LDL-cholesterol (p < 0.05), although a normalization of their lipid values could be obtained only in half of the patients. None of the subjects with hyperlipoproteinaemia had a history or clinical signs of actual vascular disease. Although the incidence of hyperlipoproteinaemia in our study group of patients with mild subclinical hypothyroidism (22.5%) is comparable to that of the normal population (21.5%), it is more severe in the former group (LDL-cholesterol in patients 5.26 +/- 0.58 vs 4.8 +/- 0.56 mmol/l in controls; p < 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

Determination of bone mineral density by quantitative computed tomography and single photon absorptiometry in subclinical hyperthyroidism: a risk of early osteopaenia in post-menopausal women.

OBJECTIVE: There is evidence that treatment with L-thyroxine increases the risk of early osteopaenia. The aim of our study was to investigate the effect of subclinical hyperthyroidism in patients on TSH-suppressive L-thyroxine in view of the increased risk of decalcification. DESIGN: Measurements of bone mineral density were performed in patients with subclinical hyperthyroidism at different scanning sites of varying trabecular portion. Bone mineral values as well as biochemical data were compared to those of normal controls. PATIENTS: Fifty patients (nine men, 25 premenopausal and 16 post-menopausal women) on TSH-suppressive doses of L-thyroxine were investigated after removal of thyroid cancer. MEASUREMENTS: Dual energy quantitative computed tomography was used for osteodensitometry in the lumbar spine. Single photon absorptiometry from a 125I source was applied to the calcaneus, midshaft radius and distal as well as proximal scanning sites of the distal radius. Normal bone mineral values for each measurement site were taken from healthy reference populations. RESULTS: A significant decrease of bone mineral density in the calcaneus was found in 26 of 50 patients. Bone mass assessment yielded a 9.1% decrease of mean bone mineral content in all patients compared to controls (P less than 0.01). The decrease in post-menopausal women was 22% (P less than 0.001). In premenopausal women bone mineral density changes in the calcaneus were not statistically significant. Cortical measurement sites like the midshaft radius and the proximal scanning site of the distal forearm showed a 14.8% (P less than 0.05) and 10.8% (P = NS) decalcification in post-menopausal women but normal values at the distal scanning site. The lumbar spine was not affected by subclinical hyperthyroidism in either pre or post-menopausal women. In hypoparathyroid patients, bone density did not essentially differ from normals. There was no significant correlation between bone mineral values and duration of treatment or osteocalcin values. CONCLUSIONS: Our data suggest that TSH suppressive L-thyroxine treatment has a detrimental effect on the appendicular skeleton in post-menopausal women. Additional effects of oestrogen deficiency and subclinical hyperthyroidism might lead to accelerated bone loss requiring close supervision to determine the smallest dose needed for suppression of the pituitary-thyroid axis.

Tumor necrosis factor beta gene polymorphisms in Graves' disease.

The physical mapping of tumor necrosis factor alpha (TNF alpha) and lymphotoxin (TNF beta) genes to the short arm of chromosome 6 in man between the loci for histocompatibility leucocyte antigens (HLA)-B and the complement system focused attention to this genetic region that controls immune responses in many ways. It also holds susceptibility genes for a variety of autoimmune disorders that are linked to specific alleles of loci in the HLA D subregion. We have recently identified a TNF restriction fragment length polymorphism with the enzyme NcoI (K. Badenhoop, G. Schwarz, J. Trowsdale, et al. Diabetologia. 1989;32:445-8). The less frequent fragment of 5.5 kilobase (kb) is in strong linkage disequilibrium with the HLA haplotype A1B8DR3. Since Graves' disease is linked to A1B8DR3, we analyzed TNF gene polymorphisms in a large group of Graves' disease patients and normal controls derived from four Centers. We show here a significant association of TNF beta polymorphisms with Graves' disease. The patients have less homozygotes for the 10.5 kb band (60 of 174, 34%) and more heterozygotes 10.5/5.5 kb (96 of 174, 55%), than 173 controls (49% homozygotes 10.5 kb and 42% heterozygotes; chi 2 = 7.45, P less than 0.03). When DR3+ patients and controls were analyzed separately, heterozygotes were still significantly increased in DR3+ Graves' disease patients (54 of 77, 70%) compared to DR3+ controls (21 of 45, 47%; chi 2 = 6.6, P less than 0.04). Furthermore, TNF fragment heterozygotes were found predominantly in patients, who had TSH-receptor antibodies (29/45, 64%, P less than 0.007), implying that these patients might represent an immunogenetic subset of the disease. Although TNF beta polymorphisms are linked to A1B8DR3, these results suggest that they represent an additional susceptibility marker in Graves' disease.

HLA-DR/DQ gene variation in nongoitrous autoimmune thyroiditis at the serological and molecular level.

The etiology of autoimmune diseases is multifactorial with genetic factors being an important prerequisite. There are two clinical manifestations of autoimmune thyroiditis: the goitrous form (Hashimoto's thyroiditis) and the atrophic variant, which is characterized by hypothyroidism (primary myxoedema). Different genetic markers were assumed to be predisposing factors for the distinct clinical presentation. In the present study, we determined HLA A,B,C,DR,DQ alloantigens serologically and HLA-DQ by gene analysis in patients with nongoitrous autoimmune thyroiditis and randomly chosen controls. To verify the exact classifications, thyroid volume (median 5.85 ml) was measured by ultrasonography. HLA-DR5 was found in 16 of 36 (44%) patients with nongoitrous autoimmune thyroiditis and in only 26 of 175 controls (15%) (Pc = 0.0018). There was a tendency towards a lower frequency of HLA-DR7 with 6% positivity in patients vs. 29% in controls (Pc = 0.052). Regarding HLA-DQ, DQ7 was found in 17 of 35 patients (48%) vs. 21 of 98 controls (21%) (Pc = 0.028) (relative risk 3.5). No other association was found with HLA-A,B,C and HLA-DR and -DQ. Our data indicate that the genetic susceptibility to autoimmune nongoitrous thyroiditis is closely associated to HLA-DR5 and DQ7 and not distinct from goitrous disease. We conclude that factors other than genetic ones explain the different immunological and clinical manifestation of chronic lymphocytic thyroiditis.

The relevance of genetic susceptibility in Graves' disease and immune thyroiditis.

It has been known for a long time that there is an increased incidence of Graves' disease and immune thyroiditis in certain families. Genetic research of this disease has shown that it is most probably transmitted in a multifactorial way, i.e. that environmental as well as genetic factors play a role in the genesis of the diseases. This hypothesis is supported by the fact that the rate of concordance is maximally 50% in identical twins. The following model of threshold values is conceivable for the formal genetics of Graves' disease and immune thyroiditis: the diseases break out whenever the sum of environmental (viruses, bacteria, iodine excess, hormones, stress) and genetic (MHC and non MHC-restricted) factors is higher than a given threshold. One of the major genes influencing the genesis of the diseases seems to be HLA-DR3 (or a closely linked gene in strong linkage disequilibrium). If this gene is present, fewer environmental factors are possibly needed.

Thyroid peroxidase/microsomal antibodies are not identical with thyroid cytotoxic antibodies in autoimmune thyroiditis.

Cytotoxic activity in sera of patients with Hashimoto's thyroiditis was measured with an antibody-dependent cell-mediated cytotoxicity assay. Cytotoxicity was determined in a 51chromium release assay using human thyroid cell targets incubated with heat-inactivated serum or IgG from patients with Hashimoto's thyroiditis. Effector cells were obtained from peripheral mononuclear cells of normal subjects. Cytotoxicity was significantly increased in patients with Hashimoto's thyroiditis (median specific lysis 20.2%, range 2.1-58.8) compared with normals (median specific lysis 8.1%, range 0-19.5; p less than 0.00001). The amount of percent specific lysis did not correlate with the titres of microsomal antibodies determined by different methods: passive hemagglutination technique (r = 0.2), enzyme immunoassay with microsomal antigen (r = 0.16), and radioimmunoassay for thyroid peroxidase antibody (r = 0.02). The cytotoxic activity was located in the IgG fraction, both in microsomal antibody positive and negative sera. After pre-incubation of microsomal antibody/thyroid peroxidase antibody positive or negative sera with purified thyroid peroxidase followed by analysis in the antibody-dependent cell-mediated cytotoxicity assay, cytotoxicity decreased in only 2 cases but was unchanged in the remaining sera. Western blot experiments with solubilized thyroid membranes and immunoblotting with cytotoxic-positive/microsomal antibody negative sera showed no binding to thyroid peroxidase. Our data suggest that cytotoxicity in sera from patients with Hashimoto's thyroiditis is not mediated by antibodies against thyroid peroxidase, but by antibodies not yet identified.

[What is the recurrence rate for Basedow's disease treated with thyrostatic agents. Answers from a prospective study]. / Wie häufig rezidiviert ein thyreostatisch behandelter M. Basedow? Antworten aus einer prospektiven Studie.

Graves' disease is an autoimmune disease whose development involves immunogenetic and exogenous factors such as viruses, bacteria and iodine excess. In a number of patients the disease follows a chronic recurrent course. A recurrence rate of 50% can be expected one year after the end of therapy. Based on the preliminary results of the prospective study presented here, this recurrence rate does not differ in groups where patients underwent long-term treatment with 10 or 40 mg thiamazole. Retrospectively obtained data suggest that the time at which euthyroidism occurs under low-dose treatment is dependent on the alimentary iodine supply. A number of groups have attempted to develop clinically applicable methods to identify patients at risk for recurrence at the end of treatment. All the studies yielded controversial results. A prospective multicenter study was undertaken to reinvestigate the importance of measuring TSH receptor antibodies and performing the TRH test and the suppression test at the end of therapy in connection with this problem. In 451 patients the recurrence rate was 50.3% one year after the end of treatment. The patients in the recurrence group had a significantly higher rate of persistent antibody activity, no increase in TSH after TRH (negative TRH test), no normal suppressibility of thyroid 123 iodine uptake (negative suppression test) and larger goiter. The calculation of sensitivities and specificities shows, however, that these differences are not large enough to be of clinical importance for the individual patient.