RSS Feeds
Discontinuing Antithyroid Drug Therapy before Ablation with Radioiodine in Graves Disease
- Henry B. Burch, MD;
- Barbara L. Solomon, DNS;
- Leonard Wartofsky, MD; and
- Kenneth D. Burman, MD
- From the Walter Reed Army Medical Center and Washington Hospital Center, Washington, D.C. Request for Reprints: Henry B. Burch, MD, Endocrine-Metabolic Service, Walter Reed Army Medical Center, Washington, DC 20307-5001. Disclaimer: The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Department of the Army or the Department of Defense. Acknowledgments: The authors thank the nurses of the Kyle Metabolic Unit for assistance in collecting and organizing serum samples; Ms. Robin Howard and Dr. Audrey Chang for assistance with the statistical analysis; and Mr. Mack Burton for administrative support. This work was performed under the auspices of a research protocol approved by the Human Use Committee of the Department of Clinical Investigation, Walter Reed Army Medical Center.
Abstract
Objective: To determine the relative effects on thyroid hormone levels of discontinuing antithyroid drug therapy and subsequent ablation with radioiodine in patients with hyperthyroid Graves disease.
Design: A clinical trial with a prospective analysis of the relative change in thyroid hormone levels over time in response to therapy in two study groups.
Setting: An outpatient endocrine clinic at a tertiary care hospital.
Patients: 21 patients with a clinical diagnosis of hyperthyroid Graves disease scheduled to receive ablation therapy with radioiodine (Iodine-131): 17 patients were pretreated with antithyroid drugs, and 4 were not.
Methods: Antithyroid drugs were stopped 6 days before radioiodine therapy. Patients were monitored clinically and biochemically with measurement of free and total levels of thyroxine (T4) and triiodothyronine (T3) on days −6, −3, −1; the day of radioiodine therapy; and days 1, 2, 3, 4, 5, 7, and 14.
Results: Before radioiodine treatment and compared with baseline measurement, the mean increase in free T4 levels after discontinuation of antithyroid therapy was 86% (95% CI, 16.1% to 156%), with a concurrent mean increase in free T3 levels of 71.6% (CI, 31% to 112%). Radioiodine therapy resulted in a mean decrease in free T3 levels of 28.7% (CI, −44.1% to −13.2%), a mean decrease in total T3 levels of 22.9% (CI, −39.4% to −6.4%), and stability in free and total T4 levels rather than aggravation of thyrotoxicosis. A smaller group of patients not receiving antithyroid drugs experienced a course qualitatively similar to that of pretreated patients after Iodine-131 treatment, with a mean reduction in free T4 levels of 39.8% (CI, −69.9% to −9.7%)and a mean decrease in free T3 levels of 49.4% (CI, −93.7% to −5.1%).
Conclusion: Short-term increases in thyroid hormone levels in patients with Graves disease receiving radioiodine ablation occur primarily as a result of discontinuing antithyroid therapy rather than as a result of treatment with Iodine-131. Stability or decrease in thyroid hormone levels, rather than further elevation, occurs during the 2-week interval after ablation therapy with Iodine-131. Antithyroid drug therapy before radioiodine ablation may have little effect on the short-term biochemical course after Iodine-131 therapy for Graves disease. The homogeneity of our sample regarding age, diagnosis, and general health may prevent application of these findings to other populations without further study.
Radioactive iodine has been used to treat hyperthyroidism associated with Graves disease for nearly 50 years [1]. The efficacy, safety, and low cost of this therapy have made it the preferred definitive treatment in most patients with this disorder [2]. Radioiodine (Iodine-131) produces an intense radiation thyroiditis followed by progressive interstitial fibrosis and glandular atrophy, resulting in the destruction of the synthetic capacity of the gland [3]. Thyroiditis after Iodine-131 therapy is believed to peak between 10 and 14 days after ablation and may be associated with an exacerbation of hyperthyroidism as early as 1 to 7 days after therapy [4-6]. This exacerbation has been presumed to be caused by the release of hormones from the damaged thyroid and may occasionally be of sufficient severity to precipitate a life-threatening thyroid crisis or storm [6, 7].
Concern over the rapid release of glandular hormone stores after Iodine-131 ablation represents a rationale for adjunctive therapy with antithyroid drugs before or after ablation, a practice adhered to by 20% to 40% of thyroidologists according to a recent international survey [2]. However, because these agents inhibit the organification of iodine within the thyroid, they may limit the effectiveness of radioiodine therapy and therefore are generally stopped 4 to 7 days before treatment with Iodine-131 [8]. The discontinuation of antithyroid therapy might be expected to be associated with a rebound in thyroid hormone synthesis, such as to exacerbate thyrotoxicosis; in fact, discontinuation of antithyroid therapy is a frequently cited but probably rare precipitant of thyroid storm [7, 9, 10]. To the best of our knowledge, a systematic study of short-term changes in thyroid hormones after discontinuation of antithyroid therapy has not been previously published.
Little information exists regarding short-term changes in thyroid hormones after administration of Iodine-131. Further, the existing literature in this area provides conflicting information: Some studies show short-term elevation in thyroid hormones after Iodine-131 therapy [11-13], others show no consistent short-term changes in hormone levels [14, 15], and still others show short-term decreases in these levels after radioiodine therapy [16]. None of the above studies compared the relative effects of discontinuing antithyroid therapy and administration of Iodine-131, and no study has measured changes in free thyroid hormone levels, which, as opposed to total thyroid hormone levels, may have predictive value for the development of thyroid storm [17].
We sought to determine prospectively the relative effects of stopping antithyroid therapy and subsequent radioiodine administration on free and total thyroid hormone levels in patients with Graves disease who have received thyroid ablation therapy. We also studied a subgroup of patients not receiving antithyroid therapy to evaluate short-term changes in thyroid hormone levels that occurred in the absence of pretreatment. Finally, we did an analysis of patient characteristics associated with the biochemical and clinical course after antithyroid therapy was discontinued and radioiodine was administered.
Methods
Patient Selection
Patients were diagnosed as having hyperthyroidism caused by Graves disease on the basis of elevated thyroid hormone levels and suppressed thyroid-stimulating hormone levels by sensitive assay in the setting of diffuse goiter, elevated 24-hour radioiodine uptake, and in most cases, detectable levels of antibodies against the thyroid-stimulating hormone receptor. Patients with other causes of hyperthyroidism or cold nodules on pertechnetate scanning were excluded from the study. Unless contraindicated, patients had been treated with the antithyroid medications methimazole or propylthiouracil until they showed biochemical improvement or returned to a euthyroid state before referral for radioiodine ablation.
Baseline Assessment
At baseline, the presence and duration of antithyroid drug therapy were noted, a smoking history was obtained, and thyroid size was assessed by a single examiner. Levels of thyroid-stimulating immunoglobulins and thyroid-stimulating hormone-binding inhibiting immunoglobulins and basal levels of free and total thyroxine (T4) and triiodothyronine (T3) were measured.
Treatment Protocol and Serial Evaluation
Six days before patients were given Iodine-131, antithyroid therapy was discontinued in patients who received these agents. Serial clinical and laboratory assessment was done on days −6, −3, and −1 before therapy; the day of treatment; and days 1, 2, 3, 4, 5, 7, and 14 after treatment. In patients not receiving antithyroid drugs, assessment was initiated from the day before therapy with Iodine-131. On each morning listed, free and total levels of T4 and T3 were measured, as were pulse and estimated thyroid size by palpation. Patients graded their symptoms of heat intolerance, nervousness, confusion, insomnia, dyspnea, palpitations, and hyperdefecation on a scale ranging from 0 to 3 as follows: 0 = none, 1 = mild, 2 = moderate, and 3 = severe. On the day of treatment, radioiodine uptake was measured to allow calculation of the dose of Iodine-131 according to the following formula: 200 to 250 microcuries/g of estimated thyroid weight/radioiodine uptake. Unless contraindicated, β-adrenergic blocking agents (atenolol or metoprolol) were given to all patients throughout the study. No patient received antithyroid drug therapy during the 14 days after Iodine-131 treatment.
Laboratory Testing
Thyroid hormone assays were done at Hazelton Laboratories (Vienna, Virginia) on batched serum samples that had been stored at −20°C pending study completion. Total T4 and T3 levels were measured using radioimmunoassay, and free T4 and T3 levels were measured using Coat-a-Count assay kits (Diagnostic Products, Los Angeles, California). Interassay coefficients of variation for values in the euthyroid range are as follows: total T4, 8.1%; free T4, 7.4%; total T3, 6.3%; and free T3,3.9%. For values in the hyperthyroid range, interassay coefficients of variation are the following: total T4, 5.9%; free T4, 6.8%; total T3, 5.7%; and free T3, 4.8%. Thyroid-stimulating hormone-receptor antibody assays were done at SmithKline Bio-Science Laboratories (Baltimore, Maryland). We measured thyroid-stimulating immunoglobulin levels using bioassay for production of cyclic adenosine monophosphate by cultured thyrocytes and expressed as thyroid-stimulating hormone equivalents. Thyroid-stimulating hormone-binding inhibiting immunoglobulin was measured with a radioreceptor assay and was expressed as the percentage inhibition of thyroid-stimulating hormone binding. Normal ranges for each of these assays are given in Table 1.
Statistical Analysis
We used StatView statistical software (Abacus Concepts, Calabasas, California) for the statistical analyses. Changes in thyroid hormone levels over time were assessed for significance using repeated-measures analysis of variance. Post hoc testing was done using the Fisher protected least significant difference and the Scheffe F-test procedures. Whenever relevant, we calculated confidence intervals to supplement null hypothesis testing. Comparison of the rate of change in thyroid hormone levels after radioiodine therapy in patients who received and those who did not receive pretreatment was done by linear regression analysis of disappearance curves of individual patient hormone, followed by the Mann-Whitney U-test. Correlation between patient clinical measurements (duration of antithyroid drug therapy, current smoking status, goiter size, radioiodine uptake, dosage of Iodine-131, and basal thyroid hormone and thyroid-stimulating-receptor antibody levels) and change in thyroid hormone levels over time was assessed using Pearson's correlation procedure for continuous variables and analysis of variance for calculations involving discrete dependent variables.
Results
Patient Characteristics
Twenty-one patients (12 women and 9 men) gave informed consent to participate in the study, which was approved by the Walter Reed Army Medical Center Human Use Committee. Patients ranged in age from 19 to 71 years (mean, 38.1 ± 11.2 years). Seventeen patients received pretreatment with antithyroid drugs before receiving radioiodine: Nine patients received methimazole, 5 received propylthiouracil, and 3 received both agents sequentially because of rash or pruritus with the first agent used. Four patients could not receive either preparation because of rash (2 patients), a history of hepatitis (1 patient), and social circumstances (1 patient). Estimated baseline thyroid size ranged from 25 to 80 g, with a mean of 44 ± 11 g. Duration of antithyroid therapy ranged from 4 to 52 weeks, with a mean duration of 14 weeks. Baseline clinical and laboratory characteristics are summarized in Table 1. Treatment doses of Iodine-131 ranged from 9.8 to 22.0 mCi, with a mean dose of 16.6 ± 3.7 mCi.
Clinical Course
All patients remained clinically stable throughout the study. No consistent change occurred in symptoms during the study period; nearly equal numbers of patients experienced worsened and improved symptoms, either during preparation for Iodine-131 therapy or after therapy.
Changes in Thyroid Hormone Levels
Figure 1 shows the short-term changes in thyroid hormones after antithyroid therapy was discontinued in the 17 patients receiving this treatment before administration of Iodine-131. Two patients were missing a single time point for which data were derived by regression analysis. Compared with baseline measurements, levels of each of the hormones measured increased significantly during the 5 days after antithyroid therapy was discontinued and before patients received Iodine-131 ablation therapy. As assessed by analysis of variance, changes in each of these levels were highly significant (P < 0.001). After antithyroid drugs were withdrawn and before Iodine-131 ablation was started, mean free T4 levels increased from 19.3 pmol/L to 28.3 pmol/L (1.5 to 2.2 ng/dL) (an increase of 46.7%; P < 0.001). This represented an average free T4 increment of 86.0% (95% CI, 16.1% to 156%). The additional increase in mean free T4 after Iodine-131 therapy was not statistically significant by post hoc testing but was associated with an average increment of 27.2% (CI, 4.8% to 49.5%) for the group as a whole. The subsequent decrease in free T4 levels became significant 7 days after radioiodine ablation (P < 0.05). Mean levels of free T3 increased from 4.4 pg/mL to 7.3 pg/mL (an increase of 65.9%; P < 0.001) during the period between discontinuation of antithyroid therapy and administration of Iodine-131. These changes represented an average increment of 71.6% (CI, 31.0% to 112%) for the group as a whole. The further increase in free T3 levels after Iodine-131 therapy was not statistically significant. Free T3 levels decreased precipitously over the next 2 weeks, with an average decrease of 28.7% (CI, −13.2% to −44.1%)by the final day of testing. Mean levels of total T4 increased from 112.0 nmol/L to 149.3 nmol/L (8.7 µg/dL to 11.6 µg/dL) (33.3%; P < 0.001) before treatment with Iodine-131, representing an average increment of 61.8% (CI, 5.7% to 118%) for the group as a whole. Subsequent changes in total T4 levels were not significant. Mean levels of total T3 increased from 2.4 nmol/L to 3.8 nmol/L (154.5 pg/dL to 246.0 pg/dL) (59.2%; P < 0.001) in the period before Iodine-131 therapy, representing an average increment of 63.7% (CI, 35.8% to 91.7%). The further increase in total T3 levels after Iodine-131 were not significant, with an average increment of 5.5% (CI, −2.6% to 13.7%). Mean levels of total T3 decreased rapidly thereafter, achieving significance compared with the peak levels 3 days after Iodine-131 therapy by post hoc analysis (P < 0.01). By the final day of testing, the average decrease in total T3 levels after Iodine-131 ablation was 22.9% (CI, −39.4% to −6.4%).
![Figure 1. Values shown represent the mean; bars indicate the standard error of the mean. Discontinuing antithyroid therapy resulted in a rapid increase in all thyroid hormone levels measured, peaking on either day 1 (free triiodothyronine [T ] and total T3) or day 2 (free thyroxine [T ] and total T ) after Iodine-131 therapy. Thereafter, free T and total T levels rapidly decreased, followed by a gradual decrease in free T levels and maintenance of total T4 levels at a plateau value over the period of observation. * = significant at < 0.05 compared with baseline values; ** = < 0.001 compared with baseline values; † = < 0.05 compared with values from T ; ‡ = < 0.005 compared with values from T ; § = < 0.05 compared with values from T . T = the day of Iodine-131 treatment. To convert total T values to nmol/L and to convert free T values to pmol/L, multiply by 12.87. To convert total T3 values to nmol/L, multiply by 0.01536.](553/F1.small.gif)
The biochemical course in four patients who received Iodine-131 without pretreatment with antithyroid agents is shown in Figure 2. Compared with values obtained on the morning of Iodine-131 administration, free T4, free T3, and total T3 levels each decreased significantly during the study period; mean levels of free T4 decreased 42.5% (P < 0.001), mean levels of free T3 decreased 60.6% (P < 0.001), and mean levels of total T3 decreased 55.8% (P < 0.001). Mean T4 levels also decreased significantly after Iodine-131 therapy compared with the peak value on the day before Iodine-131 was administered (a decrease of −30.5%; P < 0.05). Despite the relatively small size of the group of patients not receiving antithyroid drugs before Iodine-131 therapy, average decreases in thyroid hormones were significant for free T4 levels (a mean decrease of −45.3%;CI, −12.8% to −77.8%) and total T4 levels (a mean decrease of 29.4%; CI, −3.0% to −55.9%) but not for free or total T3 levels. Comparison of the rate of change in thyroid hormones after Iodine-131 therapy in the two patient groups showed that free T4 levels decreased more rapidly in patients not receiving pretreatment with antithyroid drugs (P = 0.016). No difference was discernible between the two groups for the rates of decrease for total T4, total T3, and free T3 levels. Table 2 provides a summary of thyroid hormone levels before and after treatment with Iodine-131.
Correlational analysis showed that patient age was negatively associated with peak total T3 levels (r = −0.63; P < 0.01) and that radioiodine uptake was positively correlated with an increase in total T4 levels (r = 0.63; P < 0.01). No correlation was found between changes in thyroid hormone levels and patient age, sex, goiter size, duration of antithyroid therapy, thyroid-stimulating hormone-receptor antibody titer, current smoking status, basal levels of thyroid hormones, radioiodine uptake, or dosage of Iodine-131.
Discussion
Although worsening thyrotoxicosis is frequently cited as an early complication of Iodine-131 administration, few data are available about short-term changes in thyroid hormone levels after radioiodine therapy. Interpretation of the existing literature in this area is limited by small study size, heterogeneity of diagnosis within study groups, variability in treatment protocols, and incomplete hormonal analysis after Iodine-131 therapy. In an early study, Maloof and Chapman [15] noted a gradual decrease in protein-bound iodine levels in 7 of 9 hyperthyroid patients treated with Iodine-131 alone. In the remaining two patients, these levels increased, peaking 5 to 10 days after the patients received Iodine-131. However, release of metabolically inert substances such as thyroglobulin and other iodoproteins would also have been measured by the protein-bound iodine assay, which hampers interpretation of these results. Shafer and Nuttall [11] studied changes in serum levels of total T4 and T3 in 11 hyperthyroid patients receiving an average dose of 13 mCi of Iodine-131 and 2 patients receiving large ablative doses of 29 and 50 mCi. No patient received pretreatment with antithyroid drugs. These investigators noted significant increases in mean T3 and T4 levels after patients received Iodine-131, but this finding may have been inappropriately weighted by the inclusion of a patient who received high-dose Iodine-131 and experienced marked increases in thyroid hormone levels because of an apparent radiation thyroiditis. Creutzig and colleagues [13] noted slight increases in total T3 levels after Iodine-131 therapy in four groups of patients with Graves disease who were distinguished on the basis of severity of thyrotoxicosis. Mean T3 levels increased from 1.2% to 16% and returned to baseline no more than 7 days after Iodine-131 therapy. In three patients, symptoms of thyrotoxicosis appeared to worsen at unspecified time intervals after Iodine-131 therapy. Conversely, Wise and colleagues [16] noted an abrupt decrease in total T3 levels in five patients with Graves disease who received 10 to 15 mCi of Iodine-131 and did not receive antithyroid drug pretreatment [16]. Tamagna and associates [14] studied 14 patients with hyperthyroidism of various causes, including 12 patients with diffuse toxic goiter who received 2 to 25 mCi of Iodine-131 and 2 patients with toxic multinodular goiter who each received 100 mCi. In all patients, antithyroid therapy was discontinued at least 1 month before radioiodine ablation. Although the group as a whole experienced no significant short-term changes in total T3 and T4 levels, a subgroup of 3 patients experienced T3 and T4 increases of 91% and 28%, respectively; in 1 of these patients, the symptoms of thyrotoxicosis worsened.
Our study yielded several noteworthy findings. The short-term increase in mean thyroid hormone levels observed after methimazole or propylthiouracil was withdrawn ranged from 46% to 66% greater than baseline values. Unexpectedly, because few to no further increases in hormone levels occurred after Iodine-131 therapy, the primary effector mediating this change appears to be the discontinuation of therapy with these drugs rather than the radioiodine therapy itself. Conversely, hormone levels began to decrease as early as 2 days after ablation therapy. The decrease in free and total T3 levels after Iodine-131 therapy occurred at a rate that surpassed that observed for T4 levels. This may represent direct effects of Iodine-131 on thyroid secretion of T3, the effect of decreased T4 availability for peripheral conversion to T3, the inhibitory effects of β-blockade on the T4-to-T3 conversion, or a combination of these factors. The reason for the disparity between our results and those in the studies by Shafer and colleagues [11] and Creutzig and coworkers [13] is unclear, although several features of our study distinguish it from previous analyses, including a greater uniformity in our patients (all of whom had Graves disease of relatively recent onset), the pretreatment of most of our patients with antithyroid drugs, and a consistent and relatively high dosing regimen for Iodine-131 ablation.
Another unexpected finding of our study is that previous treatment with antithyroid drugs did not produce qualitative differences in the short-term biochemical course after Iodine-131 therapy. Specifically, we observed rapid decreases in free and total thyroid hormone levels in the group of patients not receiving this pretreatment. Interestingly, the rate of decrease in free T4 levels was more rapid in patients not receiving antithyroid drug pretreatment. This finding may support the previously suggested concept that antithyroid therapy renders the thyroid relatively radioresistant [18, 19]. Although the decrease in mean free thyroid hormone levels in these patients was statistically significant, our results do not permit generalization to all patients receiving Iodine-131 who have not received antithyroid drug pretreatment. Further, because patients were not randomly assigned to receive radioiodine without pretreatment, it is not possible to exclude selection bias in favor of patients with milder disease in this subgroup. The finding that patients who received antithyroid drug pretreatment were more likely to sustain a rapid increase in thyroid hormone levels than those treated with Iodine-131 alone is noteworthy in that this is a potential mechanism through which thyroid storm may be precipitated, as has been observed after thyroidectomy, administration of iodinated contrast agents, vigorous thyroid palpation, and ingestion of massive doses of exogenous thyroid hormone [7]. Although little correlation exists between absolute thyroid hormone level and the risk for thyroid storm, several reports have suggested an association between elevation in free hormone levels and precipitation of this medical emergency [10, 19, 20]. Our findings of dramatic increases in free T3 and free T4 levels in preparation for Iodine-131 therapy may be significant in this regard for several related reasons. First, free hormone levels are increased in nonthyroidal illness [21, 22]. Second, it is likely that older patients with underlying cardiorespiratory or infectious disease may be particularly prone to radioiodine-induced thyroid storm [6]. Thus, an additional propensity to increased free hormone levels similar to that seen after discontinuation of antithyroid therapy might enhance the likelihood of this complication in vulnerable patients. Accordingly, clinicians should be especially wary of withdrawing antithyroid drugs and subsequent Iodine-131 ablation therapy in older patients with systemic illness.
In summary, our study shows that patients receiving Iodine-131 therapy for Graves disease after pretreatment with antithyroid drugs experience a short-term elevation in thyroid hormone levels that occurs because of the discontinuation of the antithyroid therapy rather than because of the Iodine-131 treatment itself. Clinical measurements such as goiter size and baseline thyroid hormone levels are poor predictors of the magnitude of the increase in thyroid hormones after antithyroid therapy is discontinued. Relative stability or a short-term decrease in thyroid hormone levels, rather than further elevation, generally follows treatment with Iodine-131 for Graves disease. Although a comparison with a smaller group of patients not receiving antithyroid drugs before Iodine-131 therapy suggests that pretreatment has little effect on the short-term biochemical course after radioiodine therapy, the relatively small size of this subgroup, as evidenced by the broad confidence intervals obtained for hormonal changes after Iodine-131 therapy, prevents a broad recommendation for abandoning this practice at this time. Rather, our results suggest the need to reassess the risk-to-benefit ratio for pretreatment with antithyroid drugs in this setting. It should also be noted that the homogeneity of our study population—relatively young and otherwise healthy persons with Graves disease—precludes a general application of these findings to other patient groups such as older persons with toxic multinodular goiter or concurrent nonthyroidal illness.
Presented in part at the 69th Annual Meeting of the American Thyroid Association, Tampa, Florida, November 1993.
- Copyright ©2004 by the American College of Physicians
Related Clinical Content
