0021-972X/90/7006-1518$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright© 1990 by The Endocrine Society
Vol. 70, No. 6 Printed in U.S.A.
Current Trends in the Management of Graves' Disease* BARBARA SOLOMON, DANIEL GLINOER, RAPHAEL LAGASSE, AND LEONARD WARTOFSKY Departments of Clinical Investigation and Medicine, Walter Reed Army Medical Center (B.S., L. WJ, Washington, D.C. 20307 the Department of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland 20814 and Hospital Saint Pierre Ecole de Sante Publique Faculty of Medicine, University of Brussels (D.G., R.L.), Brussels, Belgium
ABSTRACT. Members of the American Thyroid Association were invited to participate in a survey of the management of Graves' disease. One primary case and several variations were provided, which differed in respect to age, sex, goiter size, severity, etc. The questionnaire was based on the format used in a similar survey of members of the European Thyroid Association. The aim of the survey was to determine 1) how expert thyroidologists employ diagnostic procedures for this disorder, and 2) the choice of therapy of the three treatment options and its manner of implementation. Questionnaires were sent only to clinically active members. The overall response rate was 62%. Data analysis was possible on 52% of members surveyed and was performed using SPSS and a specific Fortran program. In the laboratory evaluation of the primary case a radioiodine uptake, scan, serum total T4, and basal TSH were requested by
G
RAVES' disease (1, 2) is a fairly common endocrine disorder whose diagnosis may be enhanced through the use of new technologies, such as the measurement of free hormone levels and TSH by new ultrasensitive assays (3). In addition, patients have the option of three different treatment modalities, each of which has advantages and disadvantages. Physician recommendations for therapeutic choice may depend upon cost, convenience, medical and dietary exposure to iodine (4, 5), objective data concerning risks and benefits of each treatment modality, availability of surgical expertise, and personal biases. In addition, the patients may influence the selection of therapy on the basis of prior knowledge, fears, and their own biases. After a report by Glinoer and associates (6) on the management of Graves' disease in Europe, we chose to conduct a similar survey within the U.S. We had three
92%, 47%, 83%, and 66%, respectively, with 84% of respondents using an ultrasensitive TSH assay. For management of the primary case, radioiodine treatment was the first choice of 69% of the respondents. Antithyroid drugs were used briefly (3-7 days) before 131I by 28%, whereas 41% said they would employ thioureas after 131I. Of those using 131I, 66% tailored the dose to achieve euthyroidism as the goal of therapy, while 34% aimed for hypothyroidism requiring T4 replacement. Only 30% of respondents chose thioureas as a first line of treatment (72% propylthiouracil; 28% tapazole). The duration of drug therapy was a predetermined fixed interval for 80% of the respondents, with 90% treating for 1-2 yr. Other specific trends in diagnostic approach and therapeutic preferences were identified for the eight variations on the primary case problem. {J Clin Endocrinol Metab 70: 1518-1524, 1990)
objectives in performing this survey: 1) to determine which tests available for the diagnosis of Graves' disease are currently being used; 2) to determine trends for the management of Graves' disease given the three therapeutic options of antithyroid drugs, radioiodine (131I), or surgery; and 3) to determine how the three therapeutic options are being implemented. Materials and Methods
Received November 2, 1989 Address all correspondence and requests for reprints to: Dr. Barbara Solomon, Endocrine-Metabolic Service 7D, Walter Reed Army Medical Center, Washington, D.C. 20307. * The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense. This work was performed under the auspices of Department of Clinical Investigation Protocol 1398-87.
Minor modifications from the original European survey were made, which consisted primarily of changes in wording within a question to comply with current American idioms. The 12page questionnaire was divided into sections. In section I, respondents were presented with a hypothetical patient (Table 1) and asked to select in vivo and laboratory tests for this index patient as they would in their daily practice. Section II dealt with the choice of therapy for the index patient. Section III asked for more specific information on how the chosen therapy would be implemented. The fourth and final section presented eight variations (Table 2) from the index patient. Respondents were asked if therapy would remain the same with changes in various clinical aspects of the index patient and, if there was a change, to provide information similar to section III. Since we were interested in what expert thyroidologists would actually do, we mailed the questionnaires to clinical members of the American Thyroid Association listed in the
1518
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
MANAGEMENT OF GRAVES' DISEASE 1987/88 membership directory who were residing within the U.S. Participants were kept strictly confidential. Identification of responses corresponded to 4 items, diagnostic procedures to 51 items, and therapeutic modalities to 62 items, and there were 62 items for each of the 8 clinical variations. Computer analysis included a theoretical total of 613 items/questionnaire. All responses were coded, entered onto preprogramed floppy disks using an IBM-PC, and transformed to ASCII format. Data analysis was performed using a specifically designed Fortran IV program and SPSS (Statistical Package for Social Sciences) for statistical evaluations (7). All frequencies were adjusted on a 100% basis after omission of missing or unknown values.
Results Survey response Three hundred and eighty questionnaires were mailed, with 235 returned (62%), and 197 (52%) included in the data analyses. The difference is accounted for by questionnaires returned unanswered, with no forwarding address, or too late to be included. The primary reason given by respondents for returning the questionnaire unanswered was insufficient clinical experience due to a change in focus in their activities. Evaluation of the index patient Diagnostic procedures. Initial diagnosis was conducted in an ambulatory care setting (100%) rather than in a hospital setting. Figure 1 presents the results concerning in vivo diagnostic tests. A radioiodine uptake would be obtained by 92.3% of respondents, with 53% requesting the uptake without a thyroid TABLE 1. Clinical impression: uncomplicated Graves' disease Index patient A 43-yr old woman presents with moderate but overt signs of hyperthyroidism of 2- to 3-months duration. She is healthy, takes no medications, and leads an active working life. She has two children, ages 5 and 10 yr, and does not plan on being pregnant again. This is her first episode of hyperthyroidism. She has a diffuse goiter of 40-50 g, pulse rate of 105 beats/min and regular, and typical but minimal eye signs. TABLE
2. Patient variations
No.
Age (yr)
Sex
Severity
Goiter0
First episode
1 2 3 4 5 6 7 8
43 43 43 43 43 43 71 19
F F F F F M F F
Moderate Moderate Severe Moderate Moderate Moderate Moderate Moderate
80 None 40 40 40 40 40 40
Yes Yes Yes No (prior ATD) No (prior surgery) Yes Yes Yes
The difference from index case is underlined. ATD, Antithyroid drugs. 0 Thyroid weight in grams.
1519
scan. Of the 92.3% who would obtain an uptake, 51.5% would use 123iodine, 25.3% technetium pertechnetate (Tcm-99), and 17.2% 131iodine as their primary isotope. Timing of uptake varied with the isotope used. Of those who preferred 131iodine, 97.6% would obtain a radioiodine uptake at 24 h. When 123iodine was preferred, a 4-h uptake would be obtained by 7.5%, a 6-h study by 17%, and a 24-h study by 7.4%. Additional tests that could be requested included 1) serum protein-bound iodine, 2) thyroid (T3) suppression test, 3) thyroid ultrasound, 4) fine needle aspiration, 5) Achilles tendon reflex relaxation time, or 6) other write-in comments. Although chosen by one or two respondents, these tests were not requested by an appreciable number. The greatest write-in citation was for a measurement of the degree of exophthalmos related to the patient's minimal eye signs.
In vitro laboratory tests (Table 3). Respondents were given a choice of 17 individual laboratory tests and were asked which one(s) they would actually use for the index patient. Each test was considered an individual item separate from any other test, with no option for ordering multiple tests. Serum total T4, total T3, T3 resin uptake, and basal TSH represented the most commonly elected basic laboratory analysis for the index patient. Basal TSH was chosen by 66% of respondents, with 84% using an ultrasensitive assay. Therapeutic choices. Once the diagnosis was established, the question asked was: "In this patient, what is the treatment that you would advocate, assuming that the patient leaves the decision of therapy entirely to you?" For the index patient with moderate hyperthyroidism (Fig. 2), radioiodide therapy was recommended most frequently, with antithyroid drug therapy the alternative therapy. Surgery was chosen by only one respondent. In the final section of the survey, a number of aspects of the index case were varied (Table 2), and respondents were asked how their choice of therapy would be altered based upon these differences in patient presentation. For the patient with a large goiter (variation 1) the percentage of responses for each therapeutic option are presented in Fig. 2. Of these, 28.3% of respondents who had previously chosen antithyroid drug therapy now chose radioiodine therapy, while 13.3% changed from antithyroid drug therapy to surgery. Choices for the patient with no goiter (variation 2) were more closely split between radioiodine and antithyroid drug therapy (Fig. 2; total response), with 25.7% changing from radioiodine to antithyroid drug therapy, while only 3.3% changed from antithyroid drugs to radioiodine therapy. When recurrences after earlier treatment were considered (variations 3 and 4) radioiodine therapy was 90% or greater of all selections (Fig. 2). In patients who had prior antithyroid drugs, 71.7% of respondents changed from electing antithyroid drug therapy again to radioiodine. In those patients who had been treated previously by thyroidectomy and then recurred, 78% who would have employed antithyroid drug therapy initially changed to radioiodine. When the severity of the hyperthyroidism in the theoretical patient was varied from moderate to more severe, the responses were very similar to those for the index patient (Fig. 2 depicts total responses); while 11.7% changed from drugs to radioiodine, an almost equal number (15.4%) changed from radioiodine to drug
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
SOLOMON ET AL.
1520
JCE & M • 1990 Vol70«No6
FIG. 1. Percentage of respondents selecting in vivo diagnostic tests. Bars depict percentage who would order a 24-h radioiodine uptake with or without a scintiscan, and which isotope would be employed.
20
131 I
TABLE 3. Frequency of employment of in vitro tests
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.
Test
% Response
Total T4 TSH T3 resin uptake Total T 3 Free T4 index" Complete blood counts Chemistry profile Antimicrosomal antibodies Antithyroglobulin antibodies Free T4 TSH receptor antibodies (TSI, TSAb, TBII)6 Cholesterol Free T 3 Free T4 indexc Thyroglobulin TRH stimulation test Urinary iodide
82.7 66 63.5 55.8 50.8 47.2 40.1 26.9 19.8 16.7 9.1 8.6 5.1 3.6 1.0 0.5 0.5
Listed in decreasing order of frequency. Product of total T4 and T 3 resin uptake. fc TSI: Thyroid stimulating immunoglobuin; TSAb: thyroid stimulating antibody; TBII: TSH receptor binding inhibitory immunoglobulin. c Product of total T4 and thyroid hormone-binding globulin. 0
Therapy selected for the older person was very similar to that for recurrence (variations 3 and 4), with 65% of respondents changing from antithyroid drug therapy to radioiodine therapy, while only 4.5% moved in the other direction (Fig. 2). Therapeutic preferences for the younger patient, however, formed a pattern totally different from that of any of the other variations of the index patient. This was the first time and only instance in which antithyroid drug therapy was the predominant choice (Fig. 2 depicts total responses); 48.5% of respondents who had selected radioiodine therapy for the index patient changed to antithyroid therapy. However, 33% of respondents still chose radioiodine therapy. In summary (Table 4), radioiodine therapy represented 70% of the total options selected. It was recommended as the primary therapeutic option for all patients except the 19-yr-old female, for whom antithyroid drug therapy was recommended. Although antithyroid drug therapy was not chosen as frequently as radioiodine therapy, our respondents felt that it apparently still has a role as an option for selected patients with moderate hyperthyroidism or in the absence of goiter. Surgery was not recommended as a primary therapeutic option and represented only 2% of the total options selected. It appeared primarily as an option, albeit infrequently, for patients with larger goiters. Modalities of treatment
therapy. When the hypothetical sex of the patient was changed to male rather than female, 6.7% of respondents changed their therapy choice from drugs to radioiodine, while no one moved from radioiodine therapy (Fig. 2). The last two variations presented a difference in age, either older, a 71-yr-old female (variation 7), or younger, a 19-yr-old female (variation 8).
Treatment with radioiodine (Fig. 3). Modalities of treatment were analyzed for the index patient and for the variations in which radioiodine therapy was the primary therapeutic option. In 100% of the responses 131iodine was the isotope chosen. The goal of therapy of two thirds
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
MANAGEMENT OF GRAVES' DISEASE
1521
MKXPATEMT
FIG. 2. Frequency of election of radioiodide, antithyroid drug, or surgical therapy for the index patient (upper left) and for patients with eight variations in presentation. PBEVSWWEHT
El
D
RADIOIODIDE
ANTTTHYROID DRUGS
TABLE 4. Synopsis of therapeutic options chosen (percentage) Radioiodide
Antithyroid drugs
Surgery
Index patient Variations 1 2 3 4 5 6 7 8
69
30
1
75 47 90 93 62 71 86 33
18 52 8 7 37 28 14 63
7 1 2 0 1 1 0 4
Mean %
70
28
2
of the respondents was to restore euthyroidism regardless of the patient situation, while one third aimed for an outcome of hypothyroidism (with subsequent replacement therapy) as the goal of therapy. Eighty-five percent of respondents determined the dose of 131iodine to be given based upon the measured radioiodine uptake and a clinical estimate of thyroid gland size. There was a preference to treat the patient with a single dose (98%),
D
SURGERY
but 100% of respondents would give a second dose if needed. This second dose would generally be given 4-6 months later. Medical treatment was used before radioiodine administration by 22-35% of the respondents. Medical treatment consisted mainly of 0-blockers, with antithyroid drugs tending to be used for only 3-7 days before the radioiodine. Medical treatment before radioiodine therapy also was used most frequently with larger goiters, recurrences, or increased severity of disease. Age or sex did not alter the use of medical therapy before radioiodine. In contrast, 32-50% of respondents chose to use medical therapy after radioiodine treatment. The primary reasons given for choosing radioiodine therapy were ease of administration, paucity of side-effects, and low cost relative to the other choices. Antithyroid drug administration (Fig. 4). Choice of drug: Propylthiouracil was chosen by 69-77% of respondents over methimazole. Treatment was initiated with 300-600 mg/day propylthiouracil, whereas those using methimazole started therapy with 20-30 mg/day. Modification of therapy usually occurred after 4-8 weeks of therapy (64%) or when the patient became euthyroid
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
SOLOMON ET AL.
1522
JCE & M • 1990 Vol70»No6 DOSE IMPLEMENTATION
%
100i
FIG. 3. Bar depicts management parameters of respondents employing radioiodine therapy, including 1) whether goal of therapy is achievement of euthyroid or hypothyroid state, 2) percentage who employ both 24-h radioiodine uptake and gland .size to estimate imrL dosage, 3) whether therapeutic goal is attempted with one dose of radioiodine, 4) whether a second dose is given if necessary, and if so at what interval (in months) since the first dose, and finally 5) the percentage of respondents employing antithyroid drugs (ATD) before and/or after the radioiodine dose.
80
THERAPY GOAL
60
ADJUVANT THERAPY
40
20
EUTHYROID
HYPO THYROID
UPTAKE ™
ONE DOSE
SECOND DOSE
4 6 ATD MONTHS BEFORE
ATD AFTER
ADJUNCTIVE THERAPY TREATMENT MODIFICATION
100
THERAPY DURATION
PTU
MMI
WEEKS OF TREATMENT
EUTHYROID LABORATORY CLINICAL
BETA BLOCKER
ATD ALONE
ATD + THYROID HORMONE
YEARS
FIG. 4. Bars depict management parameters for respondents employing antithyroid drug (ATD) therapy including 1) whether propylthiouracil (PTU) or methimazole (MMI) would be used, 2) whether treatment, i.e. dosage, is modified routinely after several weeks of therapy, when patients attain euthyroidism, or on basis of thyroid function tests and clinical signs or symptoms; 3) whether adjunctive therapy with /3-blockers of thyroid hormone is given vs. antithyroid drugs alone; and finally, 4) the usual duration of therapy.
(36%). Modification of therapy was based on both laboratory and clinical parameters (87-91%); however, 711% would use laboratory data alone, while 1-2% would use clinical criteria alone. Combination therapy with adrenergic blockade: Our next series of questions assessed whether antithyroid drugs were used alone or in combination with either /3blockers or thyroid hormone. Adjunctive therapy with @-
blockers was used in the initial phase of therapy by 9099% of respondents; 1-10% did not use /3-blockers, and no one advocated using 0-blockers throughout the total duration of medical treatment. /3-Blockers were used more frequently in moderate to severe hyperthyroidism and less frequently in the absence of goiter. Adjunctive therapy with thyroid hormone was chosen by only 15% of respondents, whereas 85% used antithyroid drugs
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
MANAGEMENT OF GRAVES' DISEASE alone and adjusted the dose as needed. There were not enough positive responses to this question to analyze how the combination of antithyroid drugs and thyroid hormone would be implemented. Duration of therapy: Since Greer et al. (8, 9) popularized the concept of short term therapeutic trials of antithyroid drugs, it was of interest to determine whether thyroidologists using antithyroid drugs today preferred short (1 yr) term therapy. Long term therapy was the overwhelming response, with 88-91% preferring to treat for 1-2 yr. Long term therapy was especially advocated in the young person and the patient without a goiter. The final question asked was "if you do not treat for a fixed period of time, what criteria do you use to stop treatment?" Clinical and biochemical euthyroidism (19%) and normalization of basal T S H in a supersensitive assay (7%) were the most frequent criteria used to stop therapy. While no one felt it was of particular value to follow titers of antithyroid or antimicrosomal antibodies, several respondents examined T S H receptor antibodies (stimulating or blocking types) for normalization. Serum thyroglobulin, the T 3 suppression test, and T R H test also were not favored as criteria for the discontinuation of therapy. In summary, antithyroid drug administration was used in the absence of goiter and for the index patient, male sex, severe hyperthyroidism, and the 19-yr-old female. Most commonly, propylthiouracil would be started at 300-600 mg/day in conjunction with j8-blockers. As the patient progressed toward euthyroidism, antithyroid drug dosage would be adjusted, and /5-blockers stopped. Antithyroid drug therapy would then continue for 1-2 yr. The factors that influenced choice of medical therapy in these patients included the mildness of symptoms, absence of goiter, age, and, for each of these variations, at least a 50% chance for remission. For severe hyperthyroidism, there was more desire for achievement of a drug-induced euthyroid state before treatment with radioiodine. Surgery. The surgical procedure could only be analyzed for variation 1 (80-g goiter), since it was the only instance for which there was an appreciable response. Even then, analysis is necessarily minimal because of the small response. We wanted to assess whether radical or more conservative surgery was preferred. In 92.3% of all responses, the aim of surgery was conservative, i.e. an extent of thyroidectomy expected to keep the patient euthyroid without replacement T 4 medication. Preoperative treatment was always given, either propylthiouracil plus /3-blockers plus Lugol's solution (69%) or methimazole plus j8-blockers plus Lugol's solution
1523
(23%). Medical treatment was administered until euthyroidism had been reached, but it was not necessary that euthyroidism be maintained for any length of time before the surgery. The mean daily dose of proplythiouracil was 450 mg/day for 5 weeks. The mean daily dose of Lugol's solution was seven drops per day given for 1 week.
Discussion The present survey represents an effort to analyze medical strategies for the management of Graves' disease within the U.S. The high response rate probably indicates that the survey gives an accurate pattern of how management is organized and implemented by thyroid specialists. There was consensus that most patients with Graves' disease may be diagnosed and treated in an ambulatory setting. There was some divergence on in vivo testing, particularly between the use of 123iodine and 131 iodine; while many clinicians chose to obtain a scan along with an uptake, the majority indicated that only the uptake was necessary. There was consensus for in vitro testing with selection of assays for total T4, total T3, free T4 index, and basal TSH in a supersensitive assay predominating. Recent availability of highly sensitive TSH assays (employed by 84% of respondents) and their ability to distinguish between normal and suppressed TSH levels probably account for infrequent use of TRH tests (10). Despite recent development of several techniques for measuring antithyroid antibodies and the potential use of TSH receptor antibodies as an index of disease or remission (11), these determinations do not appear as yet to be an integral part of diagnostic testing. The results from both the in vivo and in vitro test sections indicate the importance of regular review of strategies when new tests are added to existing diagnostic procedures. In 1984, Dunn (12) surveyed 54 thyroidologists and asked for treatment options in a hypothetical patient who had 4 variations (19-yr-old female/male or 29-yrold female/male). Each patient had a 70-g goiter. Dunn found a preference for the use of antithyroid drug therapy in both the 19-yr-female and male (67% and 65%, respectively), with a closer 50-50 split between antithyroid drug therapy and radioiodine for the hypothetical 29-yrold man or woman. Surgery was elected for 6-9% over all the variations, particularly for the larger (70-g) goiter, just as in our survey. In contrast to the diversity of opinion and practice reflected in this earlier survey, remarkable consensus was found in our survey for the selection of radioiodide as primary therapy in the index patient and 7 of 8 variations. Antithyroid drug administration was employed as the primary therapy only in the young female patient, in a percentage similar to that found by Dunn (12), but was also employed by a signifi-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
SOLOMON ET AL.
1524
cant number of respondents in other selected circumstances. Where recurrence after antithyroid drugs was concerned, Dunn (12) reported 17-21% would give a second course of drugs, 40-58% would use radioiodine, and 21-43% would use surgery. In contrast, we found that 90% of respondents chose radioiodide therapy. In general, euthyroidism was the aim of therapy when either radioiodine therapy or surgery was chosen. In spite of this desired outcome, hypothyroidism remains the more common result of radioiodine therapy, indicating inaccuracy or confounding factors affecting the precision of calculation of the dose of radioiodine administered. The survey also indicated that no index or association of indices is considered to be useful in predicting the outcome of disease, including several procedures that have been advocated as useful for this purpose, such as measurement of serum thyroglobulin levels and TSH receptor antibodies or TRH testing. Thus, even through hyperthyroidism due to Graves' disease can be treated in three effective ways, each continues to have perceived advantages or disadvantages in specific clinical settings. It will be of interest to repeat such a survey in the future and monitor changes (and their apparent causes) in practice patterns. Acknowledgments The authors thank Mr. Michel Candeur for the programing and design of the computer program employed, Ms. Estelle Coleman for editorial assistance, and the many members of the American Thyroid Association who participated in this survey.
JCE & M • 1990 Vol70^No6
References 1. Burman KD. Hyperthyroidism. In: Becker KL, ed. Principles and practice of endocrinology, 1st ed. Philadelphia: Lippincott; 1990;331-47. 2. Burman KD, Baker Jr JR. Immune mechanisms in Graves' disease. Endocr Rev. 1985;6:183-232. 3. Ridgway EC. Thyrotropin radioimmunoassays: birth, life, and demise. Mayo Clin Proc. 1988;63:1028-34. 4. Wartofsky L. Low remission after therapy for Graves' disease: possible relation of dietary iodine with antithyroid therapy results. JAMA. 1973;226:1083-8. 5. Solomon BL, Evaul JE, Burman KD. Wartofsky L. Remission rates with antithyroid drug therapy: continuing influence of iodine intake? Ann Intern Med. 1987;107:510-2. 6. Glinoer D, Hesch D, Lagasse R, Laurberg P. The management of hyperthyroidism due to Graves' Disease in Europe in 1986. Results of an international survey. Acta Endocrinol (Copenh). 1987; 185(Suppl):9-37. 7. Nie NH, Hull CM, Jenkins JG, Steinbrenner K, Benta DM. Statistical package for the social sciences, 2nd ed. New York: McGrawHill; 1975. 8. Greer MA, Kammer H, Bouma DJ. Short-term antithyroid drug therapy for the thyrotoxicosis of Graves' disease. N Engl J Med. 1977;297:173-6. 9. Bouma DJ, Kammer H, Greer MA. Follow-up comparison of shortterm versus 1-year antithyroid drug therapy for the thyrotoxicosis of Graves' disease. J Clin Endocrinol Metab. 1982;55:1138-42. 10. Toft AD. Use of sensitive immunoradiometric assay for thyrotropin in clinical practice. Mayo Clin Proc. 1988;63:1035-42. 11. Bliddal H, Kirkegaard C, Siersback-Nielsen K, Friis T. Prognostic value of thyrotropin binding inhibiting immunoglobulins (TBII) in long term antithyroid treatment, 131I therapy given in combination with carbimazole and in euthyroid ophthalmopathy. Acta Endocrinol (Copenh) 1981;98:364-9. 12. Dunn JT. Choice of therapy in young adults with hyperthyroidism of Graves' disease—a brief, case-directed poll of fifty-four thyroidologists. Ann Intern Med. 1984;100:891-3.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
Vol. 70, No. 6 Printed in U.S.A.
Current Trends in the Management of Graves' Disease* BARBARA SOLOMON, DANIEL GLINOER, RAPHAEL LAGASSE, AND LEONARD WARTOFSKY Departments of Clinical Investigation and Medicine, Walter Reed Army Medical Center (B.S., L. WJ, Washington, D.C. 20307 the Department of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland 20814 and Hospital Saint Pierre Ecole de Sante Publique Faculty of Medicine, University of Brussels (D.G., R.L.), Brussels, Belgium
ABSTRACT. Members of the American Thyroid Association were invited to participate in a survey of the management of Graves' disease. One primary case and several variations were provided, which differed in respect to age, sex, goiter size, severity, etc. The questionnaire was based on the format used in a similar survey of members of the European Thyroid Association. The aim of the survey was to determine 1) how expert thyroidologists employ diagnostic procedures for this disorder, and 2) the choice of therapy of the three treatment options and its manner of implementation. Questionnaires were sent only to clinically active members. The overall response rate was 62%. Data analysis was possible on 52% of members surveyed and was performed using SPSS and a specific Fortran program. In the laboratory evaluation of the primary case a radioiodine uptake, scan, serum total T4, and basal TSH were requested by
G
RAVES' disease (1, 2) is a fairly common endocrine disorder whose diagnosis may be enhanced through the use of new technologies, such as the measurement of free hormone levels and TSH by new ultrasensitive assays (3). In addition, patients have the option of three different treatment modalities, each of which has advantages and disadvantages. Physician recommendations for therapeutic choice may depend upon cost, convenience, medical and dietary exposure to iodine (4, 5), objective data concerning risks and benefits of each treatment modality, availability of surgical expertise, and personal biases. In addition, the patients may influence the selection of therapy on the basis of prior knowledge, fears, and their own biases. After a report by Glinoer and associates (6) on the management of Graves' disease in Europe, we chose to conduct a similar survey within the U.S. We had three
92%, 47%, 83%, and 66%, respectively, with 84% of respondents using an ultrasensitive TSH assay. For management of the primary case, radioiodine treatment was the first choice of 69% of the respondents. Antithyroid drugs were used briefly (3-7 days) before 131I by 28%, whereas 41% said they would employ thioureas after 131I. Of those using 131I, 66% tailored the dose to achieve euthyroidism as the goal of therapy, while 34% aimed for hypothyroidism requiring T4 replacement. Only 30% of respondents chose thioureas as a first line of treatment (72% propylthiouracil; 28% tapazole). The duration of drug therapy was a predetermined fixed interval for 80% of the respondents, with 90% treating for 1-2 yr. Other specific trends in diagnostic approach and therapeutic preferences were identified for the eight variations on the primary case problem. {J Clin Endocrinol Metab 70: 1518-1524, 1990)
objectives in performing this survey: 1) to determine which tests available for the diagnosis of Graves' disease are currently being used; 2) to determine trends for the management of Graves' disease given the three therapeutic options of antithyroid drugs, radioiodine (131I), or surgery; and 3) to determine how the three therapeutic options are being implemented. Materials and Methods
Received November 2, 1989 Address all correspondence and requests for reprints to: Dr. Barbara Solomon, Endocrine-Metabolic Service 7D, Walter Reed Army Medical Center, Washington, D.C. 20307. * The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense. This work was performed under the auspices of Department of Clinical Investigation Protocol 1398-87.
Minor modifications from the original European survey were made, which consisted primarily of changes in wording within a question to comply with current American idioms. The 12page questionnaire was divided into sections. In section I, respondents were presented with a hypothetical patient (Table 1) and asked to select in vivo and laboratory tests for this index patient as they would in their daily practice. Section II dealt with the choice of therapy for the index patient. Section III asked for more specific information on how the chosen therapy would be implemented. The fourth and final section presented eight variations (Table 2) from the index patient. Respondents were asked if therapy would remain the same with changes in various clinical aspects of the index patient and, if there was a change, to provide information similar to section III. Since we were interested in what expert thyroidologists would actually do, we mailed the questionnaires to clinical members of the American Thyroid Association listed in the
1518
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
MANAGEMENT OF GRAVES' DISEASE 1987/88 membership directory who were residing within the U.S. Participants were kept strictly confidential. Identification of responses corresponded to 4 items, diagnostic procedures to 51 items, and therapeutic modalities to 62 items, and there were 62 items for each of the 8 clinical variations. Computer analysis included a theoretical total of 613 items/questionnaire. All responses were coded, entered onto preprogramed floppy disks using an IBM-PC, and transformed to ASCII format. Data analysis was performed using a specifically designed Fortran IV program and SPSS (Statistical Package for Social Sciences) for statistical evaluations (7). All frequencies were adjusted on a 100% basis after omission of missing or unknown values.
Results Survey response Three hundred and eighty questionnaires were mailed, with 235 returned (62%), and 197 (52%) included in the data analyses. The difference is accounted for by questionnaires returned unanswered, with no forwarding address, or too late to be included. The primary reason given by respondents for returning the questionnaire unanswered was insufficient clinical experience due to a change in focus in their activities. Evaluation of the index patient Diagnostic procedures. Initial diagnosis was conducted in an ambulatory care setting (100%) rather than in a hospital setting. Figure 1 presents the results concerning in vivo diagnostic tests. A radioiodine uptake would be obtained by 92.3% of respondents, with 53% requesting the uptake without a thyroid TABLE 1. Clinical impression: uncomplicated Graves' disease Index patient A 43-yr old woman presents with moderate but overt signs of hyperthyroidism of 2- to 3-months duration. She is healthy, takes no medications, and leads an active working life. She has two children, ages 5 and 10 yr, and does not plan on being pregnant again. This is her first episode of hyperthyroidism. She has a diffuse goiter of 40-50 g, pulse rate of 105 beats/min and regular, and typical but minimal eye signs. TABLE
2. Patient variations
No.
Age (yr)
Sex
Severity
Goiter0
First episode
1 2 3 4 5 6 7 8
43 43 43 43 43 43 71 19
F F F F F M F F
Moderate Moderate Severe Moderate Moderate Moderate Moderate Moderate
80 None 40 40 40 40 40 40
Yes Yes Yes No (prior ATD) No (prior surgery) Yes Yes Yes
The difference from index case is underlined. ATD, Antithyroid drugs. 0 Thyroid weight in grams.
1519
scan. Of the 92.3% who would obtain an uptake, 51.5% would use 123iodine, 25.3% technetium pertechnetate (Tcm-99), and 17.2% 131iodine as their primary isotope. Timing of uptake varied with the isotope used. Of those who preferred 131iodine, 97.6% would obtain a radioiodine uptake at 24 h. When 123iodine was preferred, a 4-h uptake would be obtained by 7.5%, a 6-h study by 17%, and a 24-h study by 7.4%. Additional tests that could be requested included 1) serum protein-bound iodine, 2) thyroid (T3) suppression test, 3) thyroid ultrasound, 4) fine needle aspiration, 5) Achilles tendon reflex relaxation time, or 6) other write-in comments. Although chosen by one or two respondents, these tests were not requested by an appreciable number. The greatest write-in citation was for a measurement of the degree of exophthalmos related to the patient's minimal eye signs.
In vitro laboratory tests (Table 3). Respondents were given a choice of 17 individual laboratory tests and were asked which one(s) they would actually use for the index patient. Each test was considered an individual item separate from any other test, with no option for ordering multiple tests. Serum total T4, total T3, T3 resin uptake, and basal TSH represented the most commonly elected basic laboratory analysis for the index patient. Basal TSH was chosen by 66% of respondents, with 84% using an ultrasensitive assay. Therapeutic choices. Once the diagnosis was established, the question asked was: "In this patient, what is the treatment that you would advocate, assuming that the patient leaves the decision of therapy entirely to you?" For the index patient with moderate hyperthyroidism (Fig. 2), radioiodide therapy was recommended most frequently, with antithyroid drug therapy the alternative therapy. Surgery was chosen by only one respondent. In the final section of the survey, a number of aspects of the index case were varied (Table 2), and respondents were asked how their choice of therapy would be altered based upon these differences in patient presentation. For the patient with a large goiter (variation 1) the percentage of responses for each therapeutic option are presented in Fig. 2. Of these, 28.3% of respondents who had previously chosen antithyroid drug therapy now chose radioiodine therapy, while 13.3% changed from antithyroid drug therapy to surgery. Choices for the patient with no goiter (variation 2) were more closely split between radioiodine and antithyroid drug therapy (Fig. 2; total response), with 25.7% changing from radioiodine to antithyroid drug therapy, while only 3.3% changed from antithyroid drugs to radioiodine therapy. When recurrences after earlier treatment were considered (variations 3 and 4) radioiodine therapy was 90% or greater of all selections (Fig. 2). In patients who had prior antithyroid drugs, 71.7% of respondents changed from electing antithyroid drug therapy again to radioiodine. In those patients who had been treated previously by thyroidectomy and then recurred, 78% who would have employed antithyroid drug therapy initially changed to radioiodine. When the severity of the hyperthyroidism in the theoretical patient was varied from moderate to more severe, the responses were very similar to those for the index patient (Fig. 2 depicts total responses); while 11.7% changed from drugs to radioiodine, an almost equal number (15.4%) changed from radioiodine to drug
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
SOLOMON ET AL.
1520
JCE & M • 1990 Vol70«No6
FIG. 1. Percentage of respondents selecting in vivo diagnostic tests. Bars depict percentage who would order a 24-h radioiodine uptake with or without a scintiscan, and which isotope would be employed.
20
131 I
TABLE 3. Frequency of employment of in vitro tests
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.
Test
% Response
Total T4 TSH T3 resin uptake Total T 3 Free T4 index" Complete blood counts Chemistry profile Antimicrosomal antibodies Antithyroglobulin antibodies Free T4 TSH receptor antibodies (TSI, TSAb, TBII)6 Cholesterol Free T 3 Free T4 indexc Thyroglobulin TRH stimulation test Urinary iodide
82.7 66 63.5 55.8 50.8 47.2 40.1 26.9 19.8 16.7 9.1 8.6 5.1 3.6 1.0 0.5 0.5
Listed in decreasing order of frequency. Product of total T4 and T 3 resin uptake. fc TSI: Thyroid stimulating immunoglobuin; TSAb: thyroid stimulating antibody; TBII: TSH receptor binding inhibitory immunoglobulin. c Product of total T4 and thyroid hormone-binding globulin. 0
Therapy selected for the older person was very similar to that for recurrence (variations 3 and 4), with 65% of respondents changing from antithyroid drug therapy to radioiodine therapy, while only 4.5% moved in the other direction (Fig. 2). Therapeutic preferences for the younger patient, however, formed a pattern totally different from that of any of the other variations of the index patient. This was the first time and only instance in which antithyroid drug therapy was the predominant choice (Fig. 2 depicts total responses); 48.5% of respondents who had selected radioiodine therapy for the index patient changed to antithyroid therapy. However, 33% of respondents still chose radioiodine therapy. In summary (Table 4), radioiodine therapy represented 70% of the total options selected. It was recommended as the primary therapeutic option for all patients except the 19-yr-old female, for whom antithyroid drug therapy was recommended. Although antithyroid drug therapy was not chosen as frequently as radioiodine therapy, our respondents felt that it apparently still has a role as an option for selected patients with moderate hyperthyroidism or in the absence of goiter. Surgery was not recommended as a primary therapeutic option and represented only 2% of the total options selected. It appeared primarily as an option, albeit infrequently, for patients with larger goiters. Modalities of treatment
therapy. When the hypothetical sex of the patient was changed to male rather than female, 6.7% of respondents changed their therapy choice from drugs to radioiodine, while no one moved from radioiodine therapy (Fig. 2). The last two variations presented a difference in age, either older, a 71-yr-old female (variation 7), or younger, a 19-yr-old female (variation 8).
Treatment with radioiodine (Fig. 3). Modalities of treatment were analyzed for the index patient and for the variations in which radioiodine therapy was the primary therapeutic option. In 100% of the responses 131iodine was the isotope chosen. The goal of therapy of two thirds
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
MANAGEMENT OF GRAVES' DISEASE
1521
MKXPATEMT
FIG. 2. Frequency of election of radioiodide, antithyroid drug, or surgical therapy for the index patient (upper left) and for patients with eight variations in presentation. PBEVSWWEHT
El
D
RADIOIODIDE
ANTTTHYROID DRUGS
TABLE 4. Synopsis of therapeutic options chosen (percentage) Radioiodide
Antithyroid drugs
Surgery
Index patient Variations 1 2 3 4 5 6 7 8
69
30
1
75 47 90 93 62 71 86 33
18 52 8 7 37 28 14 63
7 1 2 0 1 1 0 4
Mean %
70
28
2
of the respondents was to restore euthyroidism regardless of the patient situation, while one third aimed for an outcome of hypothyroidism (with subsequent replacement therapy) as the goal of therapy. Eighty-five percent of respondents determined the dose of 131iodine to be given based upon the measured radioiodine uptake and a clinical estimate of thyroid gland size. There was a preference to treat the patient with a single dose (98%),
D
SURGERY
but 100% of respondents would give a second dose if needed. This second dose would generally be given 4-6 months later. Medical treatment was used before radioiodine administration by 22-35% of the respondents. Medical treatment consisted mainly of 0-blockers, with antithyroid drugs tending to be used for only 3-7 days before the radioiodine. Medical treatment before radioiodine therapy also was used most frequently with larger goiters, recurrences, or increased severity of disease. Age or sex did not alter the use of medical therapy before radioiodine. In contrast, 32-50% of respondents chose to use medical therapy after radioiodine treatment. The primary reasons given for choosing radioiodine therapy were ease of administration, paucity of side-effects, and low cost relative to the other choices. Antithyroid drug administration (Fig. 4). Choice of drug: Propylthiouracil was chosen by 69-77% of respondents over methimazole. Treatment was initiated with 300-600 mg/day propylthiouracil, whereas those using methimazole started therapy with 20-30 mg/day. Modification of therapy usually occurred after 4-8 weeks of therapy (64%) or when the patient became euthyroid
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
SOLOMON ET AL.
1522
JCE & M • 1990 Vol70»No6 DOSE IMPLEMENTATION
%
100i
FIG. 3. Bar depicts management parameters of respondents employing radioiodine therapy, including 1) whether goal of therapy is achievement of euthyroid or hypothyroid state, 2) percentage who employ both 24-h radioiodine uptake and gland .size to estimate imrL dosage, 3) whether therapeutic goal is attempted with one dose of radioiodine, 4) whether a second dose is given if necessary, and if so at what interval (in months) since the first dose, and finally 5) the percentage of respondents employing antithyroid drugs (ATD) before and/or after the radioiodine dose.
80
THERAPY GOAL
60
ADJUVANT THERAPY
40
20
EUTHYROID
HYPO THYROID
UPTAKE ™
ONE DOSE
SECOND DOSE
4 6 ATD MONTHS BEFORE
ATD AFTER
ADJUNCTIVE THERAPY TREATMENT MODIFICATION
100
THERAPY DURATION
PTU
MMI
WEEKS OF TREATMENT
EUTHYROID LABORATORY CLINICAL
BETA BLOCKER
ATD ALONE
ATD + THYROID HORMONE
YEARS
FIG. 4. Bars depict management parameters for respondents employing antithyroid drug (ATD) therapy including 1) whether propylthiouracil (PTU) or methimazole (MMI) would be used, 2) whether treatment, i.e. dosage, is modified routinely after several weeks of therapy, when patients attain euthyroidism, or on basis of thyroid function tests and clinical signs or symptoms; 3) whether adjunctive therapy with /3-blockers of thyroid hormone is given vs. antithyroid drugs alone; and finally, 4) the usual duration of therapy.
(36%). Modification of therapy was based on both laboratory and clinical parameters (87-91%); however, 711% would use laboratory data alone, while 1-2% would use clinical criteria alone. Combination therapy with adrenergic blockade: Our next series of questions assessed whether antithyroid drugs were used alone or in combination with either /3blockers or thyroid hormone. Adjunctive therapy with @-
blockers was used in the initial phase of therapy by 9099% of respondents; 1-10% did not use /3-blockers, and no one advocated using 0-blockers throughout the total duration of medical treatment. /3-Blockers were used more frequently in moderate to severe hyperthyroidism and less frequently in the absence of goiter. Adjunctive therapy with thyroid hormone was chosen by only 15% of respondents, whereas 85% used antithyroid drugs
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
MANAGEMENT OF GRAVES' DISEASE alone and adjusted the dose as needed. There were not enough positive responses to this question to analyze how the combination of antithyroid drugs and thyroid hormone would be implemented. Duration of therapy: Since Greer et al. (8, 9) popularized the concept of short term therapeutic trials of antithyroid drugs, it was of interest to determine whether thyroidologists using antithyroid drugs today preferred short (1 yr) term therapy. Long term therapy was the overwhelming response, with 88-91% preferring to treat for 1-2 yr. Long term therapy was especially advocated in the young person and the patient without a goiter. The final question asked was "if you do not treat for a fixed period of time, what criteria do you use to stop treatment?" Clinical and biochemical euthyroidism (19%) and normalization of basal T S H in a supersensitive assay (7%) were the most frequent criteria used to stop therapy. While no one felt it was of particular value to follow titers of antithyroid or antimicrosomal antibodies, several respondents examined T S H receptor antibodies (stimulating or blocking types) for normalization. Serum thyroglobulin, the T 3 suppression test, and T R H test also were not favored as criteria for the discontinuation of therapy. In summary, antithyroid drug administration was used in the absence of goiter and for the index patient, male sex, severe hyperthyroidism, and the 19-yr-old female. Most commonly, propylthiouracil would be started at 300-600 mg/day in conjunction with j8-blockers. As the patient progressed toward euthyroidism, antithyroid drug dosage would be adjusted, and /5-blockers stopped. Antithyroid drug therapy would then continue for 1-2 yr. The factors that influenced choice of medical therapy in these patients included the mildness of symptoms, absence of goiter, age, and, for each of these variations, at least a 50% chance for remission. For severe hyperthyroidism, there was more desire for achievement of a drug-induced euthyroid state before treatment with radioiodine. Surgery. The surgical procedure could only be analyzed for variation 1 (80-g goiter), since it was the only instance for which there was an appreciable response. Even then, analysis is necessarily minimal because of the small response. We wanted to assess whether radical or more conservative surgery was preferred. In 92.3% of all responses, the aim of surgery was conservative, i.e. an extent of thyroidectomy expected to keep the patient euthyroid without replacement T 4 medication. Preoperative treatment was always given, either propylthiouracil plus /3-blockers plus Lugol's solution (69%) or methimazole plus j8-blockers plus Lugol's solution
1523
(23%). Medical treatment was administered until euthyroidism had been reached, but it was not necessary that euthyroidism be maintained for any length of time before the surgery. The mean daily dose of proplythiouracil was 450 mg/day for 5 weeks. The mean daily dose of Lugol's solution was seven drops per day given for 1 week.
Discussion The present survey represents an effort to analyze medical strategies for the management of Graves' disease within the U.S. The high response rate probably indicates that the survey gives an accurate pattern of how management is organized and implemented by thyroid specialists. There was consensus that most patients with Graves' disease may be diagnosed and treated in an ambulatory setting. There was some divergence on in vivo testing, particularly between the use of 123iodine and 131 iodine; while many clinicians chose to obtain a scan along with an uptake, the majority indicated that only the uptake was necessary. There was consensus for in vitro testing with selection of assays for total T4, total T3, free T4 index, and basal TSH in a supersensitive assay predominating. Recent availability of highly sensitive TSH assays (employed by 84% of respondents) and their ability to distinguish between normal and suppressed TSH levels probably account for infrequent use of TRH tests (10). Despite recent development of several techniques for measuring antithyroid antibodies and the potential use of TSH receptor antibodies as an index of disease or remission (11), these determinations do not appear as yet to be an integral part of diagnostic testing. The results from both the in vivo and in vitro test sections indicate the importance of regular review of strategies when new tests are added to existing diagnostic procedures. In 1984, Dunn (12) surveyed 54 thyroidologists and asked for treatment options in a hypothetical patient who had 4 variations (19-yr-old female/male or 29-yrold female/male). Each patient had a 70-g goiter. Dunn found a preference for the use of antithyroid drug therapy in both the 19-yr-female and male (67% and 65%, respectively), with a closer 50-50 split between antithyroid drug therapy and radioiodine for the hypothetical 29-yrold man or woman. Surgery was elected for 6-9% over all the variations, particularly for the larger (70-g) goiter, just as in our survey. In contrast to the diversity of opinion and practice reflected in this earlier survey, remarkable consensus was found in our survey for the selection of radioiodide as primary therapy in the index patient and 7 of 8 variations. Antithyroid drug administration was employed as the primary therapy only in the young female patient, in a percentage similar to that found by Dunn (12), but was also employed by a signifi-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
SOLOMON ET AL.
1524
cant number of respondents in other selected circumstances. Where recurrence after antithyroid drugs was concerned, Dunn (12) reported 17-21% would give a second course of drugs, 40-58% would use radioiodine, and 21-43% would use surgery. In contrast, we found that 90% of respondents chose radioiodide therapy. In general, euthyroidism was the aim of therapy when either radioiodine therapy or surgery was chosen. In spite of this desired outcome, hypothyroidism remains the more common result of radioiodine therapy, indicating inaccuracy or confounding factors affecting the precision of calculation of the dose of radioiodine administered. The survey also indicated that no index or association of indices is considered to be useful in predicting the outcome of disease, including several procedures that have been advocated as useful for this purpose, such as measurement of serum thyroglobulin levels and TSH receptor antibodies or TRH testing. Thus, even through hyperthyroidism due to Graves' disease can be treated in three effective ways, each continues to have perceived advantages or disadvantages in specific clinical settings. It will be of interest to repeat such a survey in the future and monitor changes (and their apparent causes) in practice patterns. Acknowledgments The authors thank Mr. Michel Candeur for the programing and design of the computer program employed, Ms. Estelle Coleman for editorial assistance, and the many members of the American Thyroid Association who participated in this survey.
JCE & M • 1990 Vol70^No6
References 1. Burman KD. Hyperthyroidism. In: Becker KL, ed. Principles and practice of endocrinology, 1st ed. Philadelphia: Lippincott; 1990;331-47. 2. Burman KD, Baker Jr JR. Immune mechanisms in Graves' disease. Endocr Rev. 1985;6:183-232. 3. Ridgway EC. Thyrotropin radioimmunoassays: birth, life, and demise. Mayo Clin Proc. 1988;63:1028-34. 4. Wartofsky L. Low remission after therapy for Graves' disease: possible relation of dietary iodine with antithyroid therapy results. JAMA. 1973;226:1083-8. 5. Solomon BL, Evaul JE, Burman KD. Wartofsky L. Remission rates with antithyroid drug therapy: continuing influence of iodine intake? Ann Intern Med. 1987;107:510-2. 6. Glinoer D, Hesch D, Lagasse R, Laurberg P. The management of hyperthyroidism due to Graves' Disease in Europe in 1986. Results of an international survey. Acta Endocrinol (Copenh). 1987; 185(Suppl):9-37. 7. Nie NH, Hull CM, Jenkins JG, Steinbrenner K, Benta DM. Statistical package for the social sciences, 2nd ed. New York: McGrawHill; 1975. 8. Greer MA, Kammer H, Bouma DJ. Short-term antithyroid drug therapy for the thyrotoxicosis of Graves' disease. N Engl J Med. 1977;297:173-6. 9. Bouma DJ, Kammer H, Greer MA. Follow-up comparison of shortterm versus 1-year antithyroid drug therapy for the thyrotoxicosis of Graves' disease. J Clin Endocrinol Metab. 1982;55:1138-42. 10. Toft AD. Use of sensitive immunoradiometric assay for thyrotropin in clinical practice. Mayo Clin Proc. 1988;63:1035-42. 11. Bliddal H, Kirkegaard C, Siersback-Nielsen K, Friis T. Prognostic value of thyrotropin binding inhibiting immunoglobulins (TBII) in long term antithyroid treatment, 131I therapy given in combination with carbimazole and in euthyroid ophthalmopathy. Acta Endocrinol (Copenh) 1981;98:364-9. 12. Dunn JT. Choice of therapy in young adults with hyperthyroidism of Graves' disease—a brief, case-directed poll of fifty-four thyroidologists. Ann Intern Med. 1984;100:891-3.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
