| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Feature |
Section of Nuclear Medicine, University of Manitoba (W.D.L.); Department of Nuclear Medicine, St. Boniface General Hospital (W.D.L., L.W.); and Section of Endocrinology and Metabolism, University of Manitoba and St. Boniface General Hospital (W.D.L., E.A.S., S.L., R.C.R., E.A.C.), Winnipeg, Canada R2H 2A6
Address all correspondence and requests for reprints to: Dr. William D. Leslie, Department of Medicine (C5121), 409 Tache Avenue, Winnipeg, Canada R2H 2A6. E-mail: bleslie{at}sbgh.mb.ca.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
| Patients and Methods |
|---|
|
|
|---|
All patients being referred to the Nuclear Medicine Department at St. Boniface General Hospital for a first radioiodine treatment of Graves hyperthyroidism were potentially eligible for this study. Recruitment began in January 1992 and ended in May 1996. Graves hyperthyroidism was diagnosed on the basis of compatible symptoms, suppressed TSH levels, elevated serum thyroid hormones (total T4 and/or total T3), radioiodine uptake (RAIU), and diffuse gland enlargement (clinically or on imaging when performed). Hormone assays were performed on two different systems [Delpia System (Wallac, Inc., Turku, Finland) before 1994 then the Immuno-1 (Bayer Corp., Elkhart, IN)]. Comparison between the methods on these two systems showed no appreciable differences in any of the three assays with the following normal ranges: TSH, 0.44.2 mU/liter; total T4, 70150 nmol/liter; and total T3, 1.22.8 nmol/liter. Only patients who had previously been treated with radioactive iodine were specifically excluded.
During the enrollment period the patients underwent a complete physical examination. Data about previous antithyroid drug treatment and ophthalmopathy were extracted from the medical record. The protocol was approved by the institutional review board for St. Boniface General Hospital and the research ethics board of University of Manitobas Faculty of Medicine and was conducted in accordance with the guidelines proposed in the Declaration of Helsinki. Informed written consent was obtained from each participant.
Radioiodine treatment
Most patients received antithyroid medication before radioiodine administration. Antithyroid drugs were discontinued 5 d before radioiodine and were not restarted unless there was evidence of persistent or worsening hyperthyroidism at the initial visit 6 wk posttreatment. ß-Blockers were not systematically stopped and were allowed for symptom control during the peri-radioiodine period. Pregnancy was excluded in females of child-bearing potential.
Each subject underwent RAIU measurements 4 and 24 h after a tracer dose of iodine-131 (0.2 MBq). A single investigator (W.D.L.) examined all patients and estimated thyroid gland size. A subset of the patients (n = 20) underwent simultaneous estimation of thyroid volume with ultrasound, and there was good concordance between clinical and ultrasound-derived estimates of thyroid volume (r = 0.78; P = 0.0001). RAIU at 4 h was used for stratification and randomization, whereas RAIU at 24 h was used for dose calculation. The ratio of early to late uptake was used as an index of iodine-131 retention. The dose adjusted for gland size and RAIU at 24 h was also calculated [dose (MBq) x RAIU (%) x 0.01 x volume (g-1)].
Subjects were stratified according to the 4-h RAIU (<50% or >50%) and thyroid size (<40, 4080, or >80 g) and were then randomized and treated according to one of four iodine-131 dose calculation methods: low-fixed, 235 MBq; high-fixed, 350 MBq;, low-adjusted, 2.96 MBq (80 µCi)/g thyroid adjusted for 24-h RAIU; and high-adjusted, 4.44 MBq (120 µCi)/g thyroid adjusted for 24-h RAIU. Within each stratum, patients were treated in blocks of four in random order using a random number generator. The randomization list was generated and kept at the Nuclear Medicine Department and was unavailable to the referring endocrinologists (allocation concealment). The Nuclear Medicine Department was aware of the treatment dose, but the patient and referring endocrinologist remained blinded until the end of the follow-up period. Radioiodine therapy was accompanied by standardized radiation protection guidelines, information that permanent hypothyroidism was a likely outcome from radioiodine therapy, and verification that the patient had a follow-up assessment with the referring endocrinologist 6 wk after radioiodine treatment.
Outcomes assessment
The same endocrinologists as those in the preradioiodine assessment were involved in clinical follow-up. Follow-up was left to the discretion of the endocrinologist, who monitored for the following clinical outcomes: persistent/recurrent hyperthyroidism requiring additional radioiodine therapy; hypothyroidism, confirmed biochemically with elevated serum TSH on two occasions 4 wk apart and requiring permanent L-T4 treatment; and euthyroidism, a clinically and biochemically euthyroid state to the end of follow-up in the absence of further thyroid treatment. Patients were followed until they were euthyroid and/or stabilized on L-T4 replacement. Secondary outcomes (e.g. hypothyroidism developing after a second radioiodine treatment for recurrent hyperthyroidism) were not considered.
Statistics
All statistical analysis was performed with CSS:Statistica (version 5.1, Statsoft, Inc., Tulsa, OK). P < 0.05 was taken as indicating a statistically significant effect. The baseline characteristics of the study population and nonparticipants were compared using pairwise t tests (quantitative variables) or
2 test (qualitative variables). Comparison of the baseline characteristics for the four treatment arms was performed with ANOVA or nonparametric Kruskal-Wallis test, as appropriate for quantitative and qualitative variables, respectively. Logistic regression was performed to compare rates of clinical outcomes (hyperthyroidism, hypothyroidism, or euthyroidism) according to treatment using coding variables for dose (low vs. high) and calculation method (fixed vs. adjusted). Survival analysis using Kaplan-Meier product-limit curves was used to estimate the time to the clinical outcomes (hyperthyroidism, hypothyroidism, or euthyroidism). Patients were not censored after an initial hyperthyroid or hypothyroid outcome to estimate cumulative rates of euthyroidism. Comparison of time to clinical outcome was assessed with the log-rank statistic. Adjustment for single covariates was performed using the Cox proportional hazards model and treatment coding with variables for dose (low vs. high) and calculation method (fixed vs. adjusted).
| Results |
|---|
|
|
|---|
Of the 264 potentially eligible subjects, 88 (33%) agreed to participate in the study. The characteristics of the participating and nonparticipating subjects are summarized in Table 1
. In general, nonparticipants were slightly older (P < 0.03) and had lower serum total T4 (P < 0.001), but were otherwise similar. The major reasons for nonparticipation (when this was recorded) were patient preference for same-day therapy (42%), patient from a remote location (28%), patient refusal (11%), and physician refusal (4%).
|
|
The mean duration of follow-up was 80 months (range, 10111 months), and no deaths were recorded. During follow-up, 21 subjects (23.8%) experienced persistent/recurrent hyperthyroidism, 61 (69.3%) developed hypothyroidism, and only 6 (6.8%) were euthyroid to the end of follow-up. Final clinical outcome was unrelated to radioiodine treatment assignment (Table 3
), even when the logistic regression model included additional explanatory variables (patient age, gender, thyroid volume, radioiodine uptakes at 4 and 24 h, radioiodine retention index, actual treatment dose, and dose intensity). Mean follow-up was slightly shorter in euthyroid subjects (51 ± 33 months) compared with those with persistent/recurrent hyperthyroidism (75 ± 21 months) or hypothyroidism (85 ± 17 months), and the difference was statistically significant (by ANOVA, P < 0.01). Three patients had less than 4 yr of follow-up (10, 24, and 30 months), and each was euthyroid biochemically up to the time that contact was lost. The diagnosis of recurrent/persistent hyperthyroidism was associated with mean total T4 of 192 ± 71 mmol/liter, total T3 of 9.8 ± 10.3 mmol/liter, and suppressed TSH. Hypothyroidism was diagnosed at a mean total T4 of 40.1 ± 29.3 mmol/liter, total T3 of 1.1 ± 0.7 mmol/liter, and serum TSH of 41.8 ± 34.7 µ/liter. Treatment assignment did not affect the mean biochemical values at which hyperthyroidism and hypothyroidism were diagnosed (data not shown), which is not surprising because outcome assessment was blinded to the treatment dose and dose calculation method. Outcome was unrelated to the administered radioiodine dose (hyperthyroid, 334 ± 109 MBq; euthyroid, 309 ± 86 MBq; hypothyroid, 334 ± 109 MBq; by ANOVA, P = 0.86), but showed a trend to a lower dose intensity in those that developed recurrent/persistent hyperthyroidism (hyperthyroid, 2.81 ± 1.52 MBq g -1 %-1; euthyroid, 3.75 ± 1.01 MBq g-1 %-1; hypothyroid, 3.67 ± 1.53 MBq g-1 %-1; by ANOVA, P = 0.08).
|
The time to each of the prespecified clinical outcomes was assessed using survival analysis techniques and is presented as Kaplan-Meier plots for fixed vs. adjusted dose calculation (Fig. 1
) and low vs. high dose calculation (Fig. 2
). Time to recurrent/persistent hyperthyroidism or permanent hypothyroidism was independent of the dose calculation method (Table 4
). When these outcomes were combined, the time to the composite end point was similar for fixed/adjusted methods, but was paradoxically shorter in the low dose arms compared with the high dose arms (P = 0.02), although the difference (8 ± 12 vs. 10 ± 14 months) is probably not clinically significant.
|
|
|
| Discussion |
|---|
|
|
|---|
Although radioactive iodine is relatively inexpensive, costs of treatment are increased if there is a requirement for complex measurements of radioiodine uptake, turnover, and thyroid volume. For Canada, the combined technical and professional costs for performing two radioiodine uptake measurements and a thyroid ultrasound exceed the cost of the radioiodine treatment itself. Some groups have even proposed a role for positron emission tomography in dose calculation, which would clearly add considerably to costs (14, 15). Therefore, our finding that a fixed dose method works as well as more complicated adjusted methods could lead to greater patient convenience while at the same time reducing medical costs (16). The measurement of radioiodine uptake before radioiodine therapy, even if a fixed dose is used, is still recommended to prevent the inappropriate administration of radioiodine to a patient with lymphocytic (silent) thyroiditis.
Our finding that a low dose method was associated with a shorter duration of euthyroidism (predominantly related to hypothyroidism) than high dose methods is unexpected and contrary to what would be expected. It is possible that this reflects a statistical aberration (type 1 error) or the slight imbalance in patient ages, because the finding was no longer statistically significant after adjustment for age. Although Beierwaltes et al. (17) previously observed a higher incidence of hypothyroidism in patients receiving a higher dose of radioiodine, a selection bias was probably a factor in this nonrandomized study. Most studies have shown the predicted dose-response relationship between administered radioactivity and rates of hypothyroidism, as demonstrated in a recent meta-analysis (18). The ability to predict permanent hypothyroidism remains relatively poor, with an accuracy of only 60% in a multivariable logistic regression model (19).
Several weaknesses in our study must be acknowledged. The possibility of a type 2 error must be considered, because the study population was relatively small. The radioiodine doses used in the low and high dose groups varied by only 50%, and it is possible that a larger difference might have affected the group outcomes. Although it is possible that knowledge of the radioiodine dose and dose calculation method could have affected follow-up and outcome determinations, this was minimized by blinding the patient and referring endocrinologist to these variables. We found similar thyroid biochemical assessments used in the diagnosis of recurrent/persistent hyperthyroidism and hypothyroidism across all treatment arms. Although permanent hypothyroidism was diagnosed when serum TSH was elevated on two occasions, 4 wk apart, it is still possible that some of these cases represented transient hypothyroidism (20, 21, 22). Radioiodine dose adjustment was based upon clinically estimated thyroid gland size and radioiodine uptake at 24 h. Clinical estimate of thyroid gland size is known to be inaccurate, but this is still the most widespread method used for estimating gland size. Therefore, our methods reflect clinical practice. Furthermore, only a single observer was used to estimate thyroid size in an effort to minimize intraobserver variability, and a subgroup of subjects underwent simultaneous thyroid ultrasound volume assessment with a good correlation. Secondly, our use of the radioiodine uptake at 24 h does not make any adjustment for iodine kinetics. Some groups have advocated additional uptake measurements after 24 h to characterize the rate of radioiodine turnover, which is then incorporated into the radioiodine treatment dose (10, 12). Therefore, we cannot state that a more finely adjusted dose based upon thyroid ultrasound volume and/or radioiodine turnover would not give superior results to a fixed dose, as has been implied (23).
There have been few randomized clinical trials that have directly compared different doses of iodine-131 in the treatment of thyrotoxicosis. The earliest and largest trial compared a conventional dose [5.18 MBq (140 µCi)/g adjusted for gland size by palpation, but not for radioiodine uptake] with a half-dose [2.59 MBq (70 µCi)/g] in 546 patients with unspecified causes of hyperthyroidism (24). Initial control of hyperthyroidism was slower in the low dose group, but after 3 yr results were similar. The mean doses of radioiodine used in this study (185 MBq in the conventional group and 104 MBq in the half-dose group) were much less than those currently used, and results may not be applicable to current practice. A more recent trial specifically compared standard and calculated iodine-131 activity for Graves hyperthyroidism in an open randomized trial that directly compared a standard activity of 555 MBq iodine-131 with an activity calculated to deliver 100 Gy (taking into consideration thyroid volume by ultrasound and radioiodine kinetics) (25). A strong correlation was found between the success of therapy and the radiation dose actually absorbed by the thyroid, leading the researchers to recommend individualization of the administered radioiodine activity with a target dose in the range of 200 Gy. The study data actually found higher success rates (defined as elimination of hyperthyroidism 6 months posttreatment) with fixed activity (71%) vs. adjusted activity (58%), a difference that was marginally significant (P = 0.05). The median activity given to the adjusted dose group was 298 MBq, which prevents direct comparison with outcome results from the fixed dose of 555 MBq. In a third open randomized trial, 221 consecutive hyperthyroid patients were randomized to receive a fixed dose of iodine-131 (185, 370, or 555 MBq depending upon gland size) or an individualized dose (3.7 MBq/g total thyroid volume by ultrasound adjusted for 24-h radioiodine uptake measurement, to a maximum dose 740 MBq) (26). The researchers concluded that a semiquantitative approach is probably as good as using more elaborately calculated radioiodine doses and is more cost effective. The rates of permanent hypothyroidism were remarkably low (79%), with an unacceptably high rate of persistent hyperthyroidism (3541% at 12 months). These findings probably relate to the fact that the majority of the subjects had toxic nodular thyroid disease, and results may not be applicable to Graves disease.
In summary, we have completed prolonged clinical follow-up in individuals with Graves hyperthyroidism who were treated with radioactive iodine using a variety of dose calculation methods. We have not found any advantage to dose adjustment over a fixed dose of radioiodine and conclude that any advantage, if it exists, must be small and of little clinical significance. The use of a fixed dose approach simplifies the approach to treatment with potential cost savings, a position also supported in a recent guidelines statement from the United Kingdom (27).
| Footnotes |
|---|
Received May 24, 2002.
Accepted September 17, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. A. Brent Graves' Disease N. Engl. J. Med., June 12, 2008; 358(24): 2594 - 2605. [Full Text] [PDF] |
||||
![]() |
J. C. Sisson, A. M. Avram, S. A. Lawson, P. G. Gauger, and G. M. Doherty The So-Called Stunning of Thyroid Tissue J. Nucl. Med., September 1, 2006; 47(9): 1406 - 1412. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. N. Mamelak, S. Rosenfeld, R. Bucholz, A. Raubitschek, L. B. Nabors, J. B. Fiveash, S. Shen, M.B. Khazaeli, D. Colcher, A. Liu, et al. Phase I Single-Dose Study of Intracavitary-Administered Iodine-131-TM-601 in Adults With Recurrent High-Grade Glioma J. Clin. Oncol., August 1, 2006; 24(22): 3644 - 3650. [Abstract] [Full Text] [PDF] |
||||
![]() |
Radioiodine therapy for hyperthyroidism DTB, June 1, 2006; 44(6): 44 - 48. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Horacek and J. A. Franklyn Radioiodine Treatment of Graves' Hyperthyroidism J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 6113 - 6113. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |