This study explored the extended follow-up findings of participants in a previously reported randomized, parallel-group clinical trial initiated in April 2005 in Tehran, Iran, an area of iodine sufficiency [17]. The clinical trial was registered and approved (IRCT201009224794N1) with an ethics code of 275EC. The research protocol was approved by the ethics committee of the Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences (IR.SBMU.ENDOCRINE.REC.1400.001) and informed consent was obtained.

The details of the research design and results of 8-year follow-up have been previously published [13]. Briefly, patients aged ≤60 years with untreated TMNG were randomized at a ratio of 1:1 to undergo RAI or LT-MMI treatment. Upon conclusion of recruitment, 130 patients met the inclusion criteria. Fig. 1 shows the study flow diagram. The diagnosis was made based on documented hyperthyroidism (serum thyroid-stimulating hormone [TSH] <0.4 mIU/L, free thyroxin [fT4] >23 pmol/L, and/or serum triiodothyronine [T3] >200 ng/dL) and palpation, thyroid sonography/thyroid scintigraphy, and/or negative serum TSH receptor antibodies. Patients with cardiovascular disease, cirrhosis, chronic kidney disease, pregnancy, breastfeeding, use of iodine or amiodarone, and altered mental function were excluded.

At the first visit after randomization, baseline clinical information was obtained, a physical examination was performed, and the thyroid weight was estimated. Patients were followed every 2 to 3 months in the 1st year and every 6 months thereafter, for a total of 108 to 148 months. In patients with overt hypothyroidism or hyperthyroidism, the intervals between follow-up visits were shortened to 1 to 3 months until euthyroidism was achieved. Serum TSH, fT4, and T3 concentrations were determined at each visit.

Of the 65 patients randomized to the RAI group, five were lost to follow-up and six chose other treatment modalities. Therefore, 54 patients received 50 µCi of I¹³¹ per gram of estimated thyroid gland weight divided by 24-hour RAI uptake (as a decimal). Sixty-five other patients received 10 to 20 mg of MMI daily for the 1st month, after which the titration method was applied to maintain serum fT4 and TSH within the normal range. In the MMI group, four patients experienced adverse events related to the medication, three chose to undergo thyroidectomy, and five were lost to follow-up. The remaining 53 patients were treated with LT-MMI. The mean duration of followup was 120 and 132 months for the LT-MMI and RAI groups, respectively.

When subclinical hypothyroidism or hyperthyroidism occurred in the MMI patients, the dose of MMI was adjusted with the target of keeping thyroid function tests within the normal ranges. In the RAI group, the persistence of overt hyperthyroidism for longer than 6 months was treated with additional dose(s) of RAI or LT-MMI treatment (according to the abovementioned protocol). On the contrary, subclinical hypothyroidism or hyperthyroidism was not treated (except for patients aged ≥65 years with subclinical hyperthyroidism). Overt hypothyroidism in both groups was treated with levothyroxine.

Hyperthyroidism was defined as TSH <0.4 mIU/L (normal reference range, 0.4 to 5.06) with fT4 >23 pmol/L (normal reference range, 9 to 23) and/or T3 >200 ng/dL (normal reference range, 75 to 200). Subclinical hyperthyroidism was defined as TSH <0.4 mIU/L with serum fT4 and T3 within the reference range. Overt hypothyroidism was considered as TSH above the upper limit of normal (i.e., 5.06 mIU/L for the Tehranian population) with fT4 <9 pmol/L. Subclinical hypothyroidism was defined as TSH >5.06 mIU/L with normal serum fT4 and T3 concentrations [18].

Prediabetes and diabetes were defined as recommended by American Diabetes Association [19]. Hypertension was diagnosed with a confirmed systolic blood pressure >140 mm Hg and/or diastolic blood pressure >90 mm Hg or the use of antihypertensive medications.

The primary outcome was sustained euthyroidism until the end of the follow-up period. Key secondary outcomes were both overt and subclinical hypothyroidism and hyperthyroidism. For the present study, the time to euthyroidism and percentage of time spent in euthyroid, hypothyroid, and hyperthyroid states were calculated, and information regarding cardiovascular events and death and other comorbidities was obtained from 12 years of follow-up data collected throughout the course of the study. Adverse events such as agranulocytosis, skin reactions, arthralgia, hepatic side effects, and the occurrence of malignancy were also investigated.

The sample size for the trial was calculated to detect non-inferiority if the risk was at least 10%, the lower bound of the confidence interval for risk ratio of MMI versus RAI was <1.1, with a type 1 error of 0.05 and a power of at least 0.8, including an attrition rate of 20%. The baseline characteristics of the participants are shown as mean±standard deviation or frequency (%). Significant differences between the two study groups were assessed by the chi-square and Fisher exact tests for categorical variables and the Mann-Whitney test, analysis of variance, and the Student t test for continuous variables. All analyses were conducted on an intention-to-treat basis.

The time to euthyroidism (i.e., swift amelioration of hyperthyroidism) is considered important because prompt restoration of euthyroidism is critical in the prevention of adverse cardiovascular events [17,18]. Therefore, the velocity of decline and normalization of serum fT4 and T3 and the increase in suppressed TSH to normal levels were compared in both study groups. In addition, the time spent in a euthyroid state after achieving it is another critical outcome for the prevention of adverse events. Therefore, the percentage of time that serum TSH, T4, and T3 stayed in normal range during years of follow-up was calculated and compared between the two groups. The time to euthyroidism was calculated from the start of the intervention until the achievement of normal serum fT4, T3, and TSH. The event date for thyroid status cases was described as the midpoint between the date of the follow-up visit at which a particular thyroid status was detected, and the most recent follow-up visit preceding the diagnosis. All analyses were performed using SPSS for Windows version 19 (IBM Corp., Armonk, NY, USA), with a two-tailed P<0.05 being considered significant.

At baseline, there were 53 patients in the MMI group and 54 patients in the RAI group, with average ages of 49.5±6.0 and 50.0±5.7 years, respectively. The baseline demographics, clinical characteristics, and laboratory variables of the two groups, including sex, smoking, goiter size, and serum concentrations of fT4, T3, and TSH, did not show statistically significant between-group differences (Supplemental Table S1) [13].

In the MMI group, the mean daily dose of MMI declined from 14±5 to 4.0±1.3 mg by the end of 12 years of MMI therapy. At the end of the study, one patient was lost to follow-up; all 50 patients who completed sustained MMI treatment were euthyroid, and two patients on levothyroxine therapy were also euthyroid (Table 1).

In the RAI group, 12 years after receiving 16.7±2.7 mCi (618±100 MBq) of RAI, 25, 25, and four patients were euthyroid, hypothyroid, and subclinically hyperthyroid, respectively (Table 1). Twelve patients experienced recurrence of hyperthyroidism within the 12 years of follow-up: three preferred continuous LT-MMI therapy, three became hypothyroid after receiving a second dose of RAI, two were subclinically hyperthyroid, and four achieved euthyroidism after additional doses of RAI. One patient received a third dose and another patient was treated with third and fourth doses of RAI. Accordingly, in the RAI group, three subgroups could be recognized: euthyroid (n=25), hypothyroid after the first dose of RAI (n=17), and the subgroup who experienced relapse and required additional rounds of RAI therapy for hyperthyroidism (n=12).

The mean time to reach a normal serum TSH concentration was 4.3±1.3 months (range, 2 to 15) in the MMI group and 11.4±6.1 months (range, 3 to 24) in the RAI group (P<0.001) (Fig. 2).

Patients treated with MMI remained in a euthyroid state for 95.8%±5.7% of the 12 years of follow-up (range, 78% to 100%); however, in the RAI group, the mean time spent in euthyroidism was 72.4%±14.8% (range, 50% to 100%) (RAI group; P<0.001). The lowest value belonged to those patients who experienced recurrence of hyperthyroidism after the first dose of RAI treatment, who only spent 59%±10% (range, 50% to 75%) of the 12-year follow-up time in euthyroidism (Table 2).

The time spent in subclinical hyperthyroidism was 3.3%±6.0% (range, 0% to 22%) among patients treated with MMI and 17.4%±12.8% (range, 0% to 50%) in the RAI group (Table 2). This frequency was lowest in those who became hypothyroid and highest in patients who experienced recurrence after RAI treatment. Relapse of overt hyperthyroidism was not observed in patients in the MMI group or patients who attained euthyroidism or hypothyroidism after the first round of RAI therapy. However, patients who experienced relapse of hyperthyroidism after RAI therapy spent 7.5%±6.2% and 40%±23% of the 12-year period in overt and subclinical hyperthyroidism, respectively.

During treatment with MMI, no cases of overt hypothyroidism were seen, and TSH levels above 5.01 mIU/L occurred during 0.9%±0.5% of the treatment time. In the RAI group, the occurrence of subclinical and overt hypothyroidism accounted for 6.0%±9.3% and 2.5%±3.3% of the total duration of 12 years, respectively. Table 2 demonstrates the time spent in various thyroid states in both study groups during 12 years of intervention.

The mean weight gain was 1.1±3.1 and 2.7±4.0 kg in the MMI and RAI groups, respectively (P<0.021). In the RAI group, the mean weight gain was 0.4±1.8, 3.3±4.2, and 3.4±4.6 kg in euthyroid, relapsed, and hypothyroid patients after RAI therapy, respectively (P<0.011).

The prevalence of diabetes, prediabetes, and hypertension was 6%, 15%, and 22% in the MMI group and 8%, 14%, and 25% in the RAI group, respectively. There were no cases of major cardiovascular events, including myocardial infarction, stroke, atrial fibrillation, or death, in either group. Two and four cases of breast cancer occurred in the MMI and RAI groups, respectively.

At the start of the intervention, all patients were less than 61 years of age. At the end of the study, however, 12 and 13 patients were older than 65 years of age in the MMI and RAI groups, respectively. Table 3 shows the characteristics of these patients and compares outcomes between the two study groups. The time to euthyroidism was significantly shorter in MMI-treated patients than in the RAI group (3.3±1.0 months vs. 13.9±8.8 months, P<0.001), and five out of the 12 (42%) spent 100% of the follow-up time in euthyroidism. The other seven patients in this group spent 90% to 95% of the 12-year period in euthyroidism. In the RAI group, only one patient was euthyroid 100% of the time; eight of the 13 patients (61%) spent ≤80% and four (31%) spent ≤58% of the follow-up time in euthyroidism.

In the MMI group, three cases of skin reactions and one case of elevated liver enzymes occurred in the first 3 months of therapy. Mild and transient discomfort occurred in six patients in the RAI group. No other adverse events were reported after fourth month of the study in either group.