Abstract
Objective
As thionamide is associated with various adverse effects, we re-evaluated the practical efficacy of potassium iodide (KI) therapy for Graves’ hyperthyroidism (GD).
Methods
We administered KI (mainly 100 mg/day) to 324 untreated GD patients and added methimazole (MMI) only to those remaining thyrotoxic even at 200 mg/day. When the patient became hypothyroid, MMI, if taken was stopped, then levothyroxine (LT4) was added without reducing the KI dose. Radioactive iodine (RI) therapy or thyroidectomy was performed whenever required. We evaluated the early effects of KI at 2–4 weeks and followed patients for 2 years.
Results
At 2 weeks, serum thyroid hormone levels decreased in all 324 patients. At 4 weeks, fT4, fT3, and both fT4 and fT3 levels became normal or low in 74.7%, 50.6%, and 50.6% of patients, respectively. In a cross-sectional survey over 2 years, GD was well-controlled with KI or KI + LT4 (KI-effective) in >50% of patients at all time points. Among 288 patients followed for 2 years, 42.7% remained ‘KI-effective’ throughout the 2 years (KI Group), 30.9% were well-controlled with additional MMI given for 1–24 months, and 26.4% were successfully treated with ablative therapy (mainly RI). Among ‘KI-effective’ patients at 4 weeks, 76.5% were classified into the KI Group. No patients experienced adverse effects from KI.
Conclusion
KI therapy was useful in the treatment of GD. A sufficient dose of KI was effective in >50% of GD patients from 4 weeks to 2 years, and 42.7% (76.5% of ‘KI-effective’ patients at 4 weeks) remained ‘KI-effective’ throughout the 2 years.
Introduction
About 100 years ago, iodide was introduced for the treatment of Graves’ hyperthyroidism (GD) (1, 2). Inorganic iodide administration rapidly reduced serum thyroid hormone levels in GD patients (3, 4, 5, 6), however, a high prevalence of insufficient decline or re-elevation of thyroid hormone levels, the so-called ‘escape phenomenon from the Wolff-Chaikoff effect’ (7, 8), was observed. Since the introduction of thionamide antithyroid drugs (9), the use of iodide has been limited to thyrotoxic storm, preoperative treatment (10, 11, 12), post-treatment with radioactive iodine (RI) therapy (13, 14), and recently for short-term assistive use in combination with thionamide therapy (15, 16).
Among thionamide drugs, methimazole (MMI) has been recommended because of its higher efficacy (17, 18, 19). However, MMI, as well as other thionamides, is associated with various adverse effects, including urticaria, liver dysfunction, polyarthritis, and granulocytopenia. The overall incidence of adverse effects in patients taking MMI at 15 and 30 mg/day was reported to be 13.9–21.9% and 30.0–41.7%, respectively (19, 20).
In 2014, Okamura et al. (21) reported that 65.9% of 44 GD patients exhibiting adverse effects of thionamide were well-controlled with large doses of potassium iodide (KI). Then several studies evaluated the efficacy of KI in GD patients avoiding adverse effects of thionamide (22, 23), or untreated patients with mild to moderate GD (24, 25). Recently, Okamura et al. (26) reported that 33.7% of 504 untreated GD patients showed high sensitivity to KI and a good prognosis for 2–23 years.
We reevaluated the efficacy of KI in 324 newly registered untreated GD patients and elucidated practical strategies of KI treatment that minimize the use of thionamide.
Materials and methods
Participants
Between September 2007 and October 2020, a total of 380 patients with untreated GD visited the Fujikawa-Megumi Clinic (Fig. 1). At their first consultation, they received detailed information about all treatment methods (e.g. thionamides, KI, RI, and thyroidectomy) and prognosis for GD. Forty-one patients started with a combination of KI and MMI, requiring rapid and definitive relief of thyrotoxicosis, most of whom had serious symptoms (e.g. heart failure, atrial fibrillation, severe hyperglycemia, and psychiatric symptoms) and some of whom had requested prompt thyroidectomy. Nine patients chose propylthiouracil (PTU) because they were breastfeeding, three patients chose MMI, and three patients requested immediate RI therapy. The subjects of this study were the remaining 324 patients who chose KI as the first treatment for GD and met all of the following conditions: i) TSH <0.01 mU/L, ii) fT4 >1.9 ng/dL or fT3 >4.5 pg/mL, ii) TSH receptor antibody (TRAb) >2 IU/L or thyroid-stimulating antibody (TSAb) >180%, iv) no imminent serious symptoms, and v) followed for at least 6 weeks. Their clinical data are presented in Table 1. As 3–16 patients dropped out in each interval, the number of subjects at each time point was as follows – 6 weeks: n = 324; 3 months: n = 321; 6 months: n = 314; 1 year: n = 304; and 2 years: n = 288 (Fig. 1).
Flowchart showing the inclusion process for participants. GD, Graves’ hyperthyroidism; KI, potassium iodide; MMI, methimazole; RI, radioactive iodine therapy; PTU, propylthiouracil. *Subjects of the early response study, **Subjects of the cross-sectional follow-up study, ***Subjects of the treatment groupings and the comparative study.
Citation: European Thyroid Journal 13, 6; 10.1530/ETJ-24-0085
Pretreatment clinical data of the patients with Graves' hyperthyroidism (GD) initially treated with potassium iodode (KI). The 288 patients followed for 2 years were classified into three groups: (I) GD was well-controlled with KI throughout 2 years (KI Group), (II) additional meethimazole (MMI) was required even temporarily without ablative therapy (KI+MMI Group), and III) radioactive iodine (RI) therapy or thyroidectomy was performed (Ablation Group). Data are presented as n or as median (IQR).
| Patients followed >6 weeks | Patients followed for 2 years | |||||||
|---|---|---|---|---|---|---|---|---|
| Total (I)+(II)+(III) | (I) | (II) | (III) | P value* | (I) vs (II)+(III)† | |||
| Odd's ratio (95% CI) | P | |||||||
| n | 324 | 288 | 123 | 89 | 76 | |||
| Sex (male / female) | 52 / 272 | 48 / 240 | 19 / 104 | 17 / 72 | 12 / 64 | 0.6316 | 1.0863 (0.4119-2.8646) | 0.8674 |
| % males | 16.0% | 16.7% | 15.4% | 19.1% | 15.8% | |||
| Age (years) | 44.0 (34.0-55.0) | 44.0 (35.0-55.0) | 49.0 (39.5-59.5) | 47.0 (30.0-52.0) | 37.5 (28.0-45.0) | <0.0001 | 1.0370 (1.0080-1.0667) | 0.0095 |
| fT4 (ng/dl) | 3.6 (2.4-5.4) | 3.6 (2.4-5.2) | 2.5 (2.0-3.6) | 3.9 (3.1-5.9) | 4.4 (3.5-5.9) | <0.0001 | 1.0781 (0.7696-1.5104) | 0.6627 |
| fT3 (pg/ml) | 13.1 (8.0-18.7) | 12.9 (7.6-18.6) | 7.8 (5.9-12.0) | 15.5 (11.5-20.2) | 16.4 (12.6-21.0) | <0.0001 | 0.8624 (0.7686-0.9677) | 0.0095 |
| TRAb (IU/l) | 8.2 (4.7-15.7) | 8.3 (4.8-15.7) | 6.3 (4.0-11.6) | 9.8 (5.4-19.7) | 10.4 (6.4-20.7) | 0.0001 | 0.9961 (0.9799-1.0126) | 0.6354 |
| TSAb (%) | 489a (240-1230) | 491c (249-1260) | 404e (188-1015) | 619g (235-1346) | 561i (335-1293) | 0.0608 | 1.0000 (0.9997-1.0004) | 0.9648 |
| Thyroid volume (mL) | 26.9 (20.2-30.6) | 26.5 (20.2-35.9) | 21.6 (20.2-30.6) | 27.8 (22.7-36.6) | 35.3 (25.7-48.6) | <0.0001 | 0.9832 (0.9546-1.0126) | 0.2359 |
| TPOAb (+ve / -ve) | 214 / 95b | 191 / 84d | 65 / 50f | 67 / 18h | 59 / 16j | <0.0001 | 0.4147 (0.1835-0.9374) | 0.0338 |
| % positive | 69.3% | 69.5% | 56.5% | 75.3% | 77.6% | |||
| TgAb (+ve / -ve) | 215 / 94b | 192 / 83d | 75 / 40f | 59 / 26h | 58 / 17j | 0.1588 | 0.9637 (0.4317-2.1511) | 0.9281 |
| % positive | 69.6% | 69.8% | 65.2% | 66.3% | 76.3% | |||
The reference values in this study were as follows – serum fT4: 0.9-1.8 ng/dL; fT3: 2.3-4.4 pg/mL; TRAb: <1.0 IU/L; TSAb: <120%.
TRAb, TSH receptor antibody; TSAb, thyroid stimulating antibody; TPOAb, anti-thyroid peroxidase antibody; TgAb, anti-thyroglobulin antibody.
an = 223; bn = 309; cn =200; dn = 275; en = 67; fn = 115; gn = 64; hn = 85; in = 69; jn = 75; *Wilcoxon's rank-sum test or Pearson's χ2 test; †multiple logistic regression analysis
Protocol of GD treatment
i) All patients received 50-mg KI pills containing 38.25 mg of iodine (Nichi-Iko Pharmaceutical Co., Ltd., Tokyo, Japan). The starting dose of KI was 100 mg/day in 309 patients, 50 mg/day in 12 patients who showed normal fT4 but high fT3 levels, and 200 mg/day in three patients who complained of severe malaise with high levels of fT4 (>7 ng/dL) and fT3 (>20 pg/mL).
ii) If fT4 remained >1.8 ng/dL or fT3 remained >4.4 pg/mL at 2 or 4 weeks after starting KI, the dose of KI was doubled in patients receiving 50 or 100 mg/day (i.e. 50 mg increased to 100 mg/day or 100 mg increased to 200 mg/day). In those taking 200 mg/day, low-dose MMI was added to KI (MMI 5 mg when only fT3 was high, 10 mg when both fT4 and fT3 were high).
iii) When the patient became hypothyroid during combined KI and MMI treatment, MMI was first tapered and stopped. Exceptionally, if a relapse of hyperthyroidism occurred every time MMI was discontinued, the dose of KI was reduced to 100 or 50 mg/day without discontinuing MMI.
iv) When the patient became hypothyroid during KI monotherapy, levothyroxine (LT4) was added without reducing the dose of KI.
v) When TRAb levels decreased markedly and became negative without suppressed serum TSH levels, both KI and LT4 were tapered and stopped. KI was reduced by half (e.g. 200 reduced to 100 mg/day, 100 reduced to 50 mg/day) and LT4 was reduced by 25 μg (e.g. 75 reduced to 50 μg/day, 50 reduced to 25 μg/day).
vi) When thyrotoxicosis relapsed, LT4 was stopped if taken, and the dose of KI was increased (up to 200 mg/day), or MMI was added to KI. If both KI and MMI were already being administered, the dose of MMI was increased.
vii) RI therapy or thyroidectomy was performed whenever the patient wanted, even when GD was well-controlled at that time. When RI therapy was chosen, KI or KI + MMI was stopped 5–7 days before, then 480–500 MBq of 131I was taken after a 5-hour 123I uptake measurement. KI was restarted 3 days later and continued until the patient became eu- or hypothyroid.
viii) Possible remission (PR) was defined as both non-suppressed TSH and TRAb negativity without drugs at that point.
A simplified treatment schema is shown in Fig. 2.
Simplified treatment schema for Graves’ hyperthyroidism in this study. Participants were constantly moving between different treatment groups depending on their conditions. KI, potassium iodide (mg/day); LT4, levothyroxine (μg/day); MMI, methimazole (mg/day); Hyper, hyperthyroid; Eu, euthyroid; Hypo, hypothyroid. The ranges of doses for each drug used were as follows: KI, from 50 mg every other day to 200 mg/day; LT4, from 25 to 100 μg/day; MMI, from 5 mg every other day to 30 mg/day.
Citation: European Thyroid Journal 13, 6; 10.1530/ETJ-24-0085
Methods of clinical evaluation
First, changes in serum fT4 and fT3 levels at 2 and 4 weeks after starting KI were evaluated in 324 patients. Next, a cross-sectional survey was conducted to determine how patients maintained their euthyroid status at each of the following time points: 6 weeks, 3 months, 6 months, 1 year, and 2 years. If GD was well-controlled with KI alone or KI + LT4, the patient was considered to be in a ‘KI-effective’ state. Finally, the 288 patients followed for 2 years were classified into the following three groups: i) those who remained ‘KI-effective’ throughout the 2 years (KI Group), ii) those who required additional MMI even temporarily without ablative therapy (KI + MMI Group), and iii) those who received RI therapy or thyroidectomy (Ablation Group). The clinical characteristics of each group were evaluated.
This study was a retrospective cohort study approved by the Ethics Committees of the Fujikawa-Megumi Clinic and Kyushu University and was conducted according to the Declaration of Helsinki. Informed consent was obtained from all subjects and they were allowed to opt out of the study at any time.
Measurements
Serum levels of fT4, fT3, and TSH were measured using a chemiluminescence immunoassay (Chemi-Lumi, Siemens Diagnostics, Tokyo, Japan) or an electrochemiluminescence immunoassay (ECLusys, Roche Diagnostics). The correlations between the assays were r = 0.908 for fT4, r = 0.947 for fT3, and r = 0.987 for TSH. The reference values in this study were as follows: serum fT4, 0.9–1.8 ng/dL; fT3, 2.3–4.4 pg/mL; and TSH, 0.3–4.8 mU/L. Serum TRAb was measured using a two-step radioreceptor assay (DYNO test TRAb human, Yamasa Corp., Choshi, Japan; normal range <1.0 IU/L) or an electrochemiluminescence immunoassay (ECLusys, Roche Diagnostics; normal range <2.0 IU/L) (r = 0.939). Serum TSAb was measured using a radioimmuno-bioassay (Yamasa Corp; normal range <180%) or an enzyme immuno-bioassay (Yamasa Corp; normal range <120%) (r = 0.841). The thyroid volume was estimated by ultrasonographic measurement of the width, depth, and length of each lobe, and calculated using the following formula: 0.7 × width (cm) × depth (cm) × length (cm). Anti-thyroid peroxidase antibody (TPOAb) and anti-thyroglobulin antibody (TgAb) were measured by a radioimmunoassay (Cosmic II, Cosmic Corp, Tokyo; normal range <0.3 and <0.3 U/mL, respectively) or electrochemiluminescence immunoassay (ECLusys, Roche Diagnostics; normal range <16 and <28 IU/mL, respectively). The values of the former kits for fT4, fT3, TRAb, and TSAb were converted to the values of the latter kits using the corresponding conversion formulas for statistical processing. These converted values were almost the same as the raw values around the normal range.
Statistical analysis
Comparisons between categorical variables were made using Pearson’s χ 2 test, while comparisons between continuous variables were made using Wilcoxon’s rank-sum test. Multiple logistic regression analysis was also performed to compare the KI Group and the non-KI Group. All data collected were analyzed using JMP (version 15.0, SAS Institute, Cary, NC, USA).
Results
Early response to KI treatment
The changes in serum fT4 and fT3 levels in the 324 patients at 2 and 4 weeks after starting KI are shown in Fig. 3. Those who started combination therapy with LT4 (n = 8) or MMI (n = 3) at 2 weeks were not plotted at 4 weeks in these graphs. Both fT4 and fT3 levels decreased rapidly at 2 weeks in all 324 patients. Some patients showed re-elevation at 4 weeks, but not beyond the pretreatment levels, except for one patient. The median and interquartile range of fT4 and fT3 levels were 3.6 (2.4–5.4) ng/dL and 13.1 (8.0–18.7) pg/mL, respectively, before treatment, 1.5 (1.2–1.9) ng/dL and 4.7 (3.4–6.2) pg/mL at 2 weeks, and 1.4 (1.0–1.8) ng/dL and 4.5 (3.2–6.7) pg/mL at 4 weeks.
Changes in serum levels of fT4 (A) and fT3 (B) at 2 and 4 weeks after starting potassium iodide (KI) treatment in 324 patients with untreated Graves’ hyperthyroidism. 2W, 2 weeks; 4W, 4 weeks after starting KI. [1] High sensitivity to KI: fT4 <1.8 ng/dL or fT3 <4.4 pg/mL at both 2 and 4 weeks. [2] Clinical escape from the KI effect: fT4 <1.8 ng/dL or fT3 <4.4 pg/mL at 2 weeks, but fT4 >1.8 ng/dL or fT3 >4.4 pg/mL at 4 weeks. [3] Sensitive but required a higher dose of KI: fT4 >1.8 ng/dL or fT3 >4.4 pg/mL at 2 weeks, but fT4 <1.8 ng/dL or fT3 <4.4 pg/mL at 4 weeks after doubling the dose of KI. [4] Resistance to KI: fT4 >1.8 ng/dL or fT3 >4.4 pg/mL at both 2 and 4 weeks. *Only one patient showed a higher fT3 level at 4 weeks than before treatment.
Citation: European Thyroid Journal 13, 6; 10.1530/ETJ-24-0085
As shown in Fig. 3A-[1], 3B-[1], serum fT4 and fT3 levels became normal or low at 2 and 4 weeks in 202 (62.3%) and 128 (39.5%) patients, respectively. Serum fT4 and fT3 levels normalized at 2 weeks but re-elevated beyond the normal ranges at 4 weeks in 25 (7.7%) and 20 (6.2%) patients, respectively (Fig. 3A-[2], 3B-[2]). Serum fT4 and fT3 levels remained high at 2 weeks but normalized at 4 weeks after doubling the KI dose in 41 (12.7%) and 37 (11.4%) patients, respectively (Fig. 3A-[3], 3B-[3]). Although the KI dose was doubled at 2 weeks, fT4 and fT3 levels remained high at 2 and 4 weeks in 56 (17.3%) and 139 (42.9%) patients, respectively (Fig. 3A-[4], 3B-[4]).
Improvement in thyroid hormone levels in each patient was further evaluated (Table 2). At 2 weeks, fT4, fT3, and both fT4 and fT3 levels became normal or low (sufficient decrease) in 227 (70.1%), 145 (44.8%), and 143 (44.1%) patients, respectively. Among the 227 patients with a sufficient decrease in fT4 at 2 weeks, 84 (37.0%) still showed high fT3 levels. At 4 weeks, sufficient decreases in fT4, fT3, and both fT4 and fT3 were observed in 242 (74.7%), 164 (50.6%), and 164 (50.6%) patients, respectively. Among the 242 patients with a sufficient decrease in fT4 at 4 weeks, 78 (32.2%) still showed high fT3 levels.
Number and percentage of patients with each fT4 and fT3 levels at 2 and 4 weeks after starting potassium iodide (KI).
| Thyroid hormone levels | Number of patients (%) | ||
|---|---|---|---|
| fT4 (ng/dL) | fT3 (pg/mL) | 2 weeks | 4 weeks |
| A) < 0.9 | a) < 2.3 | 6 (1.9) | 8 (2.5) |
| b) 2.3 – 4.4 | 7 (2.2) | 33 (10.2) | |
| c) > 4.4 | 1 (0.3)* | 1 (0.3)* | |
| B) 0.9 - 1.8 | d) < 2.3 | 1 (0.3) | 1 (0.3) |
| e) 2.3 – 4.4 | 129 (39.8) | 114 (35.2) | |
| f) > 4.4 | 83 (25.6)* | 77 (23.8)* | |
| C) > 1.8 | g) 2.3 – 4.4 | 2 (0.6) | 0 |
| h) > 4.4 | 95 (29.3) | 79 (24.4) | |
| D) LT4 started after 2 weeks | 8 (3.5) | ||
| E) MMI was started after 2 weeks | 3 (0.9) | ||
| Total | 324 (100) | 324 (100) | |
| Sufficient fT4 decrease | 227 (70.1) | 242 (74.7) | |
| (A + B + D) | |||
| Sufficient fT3 decrease | 145 (44.8) | 164 (50.6) | |
| (a + b + d + e + g + D) | |||
| Sufficient both fT4 & fT3 decrease | 143 (44.1) | 164 (50.6) | |
| (a + b + d + e + D) | |||
LT4: levothyroxine, MMI: methimazole, Sufficient fT4 decrease: fT4 <1.8 ng/dL, Sufficient fT3 decrease: fT3 <4.4 pg/mL, * T3 toxicosis was observed in about 25% of the patinets at both 2 and 4 weeks. The reference values in this study were as follows: serum fT4, 0.9-1.8 ng/dL; fT3, 2.3-4.4 pg/mL.
Treatments that maintained euthyroid status: cross-sectional survey for 2 years
At 6 weeks (n = 324) (Fig. 4A), a total of 172 (53.1%) patients, including 158 (48.8%) taking KI only and 14 (4.3%) taking KI + LT4, were well-controlled with KI (‘KI-effective’). The remaining 152 (46.9%) patients received both KI and MMI, but only 31 (9.6%) required MMI ≥15 mg/day.
How the patients with Graves’ hyperthyroidism were treated to maintain euthyroid status at each point (A: 6 weeks, B: 3 months, C: 6 months, D: 1 year, and E: 2 years): Cross-sectional survey. [1] P/R (possible remission): showing non-suppressed TSH and negative TRAb without medications [violet area]. [2] KI + LT4: taking potassium iodide (KI) and levothyroxine (LT4) (navy area). [3] KI only: taking KI alone (blue area). [4] KI + MMI < 15: taking KI and methimazole (MMI) <15 mg/day (yellow area)]. [5] KI + MMI ≥ 15: taking KI and MMI ≥15 mg/day (orange area). [6] RI: within 3 months after 131I treatment (RI) (black area). [7] KI + MMI after RI: taking KI and MMI >3 months after RI (pink area). [8] KI after RI: taking KI or KI + LT4 >3 months after RI (green area). [9] Hypo after RI: taking LT4 alone (hypothyroid) after RI (brown area). [10] Op: taking LT4 alone (hypothyroid) after total thyroidectomy (red area). The numbers of patients who dropped out in each interval were as follows: (between A and B) KI only, n = 1; KI + MMI < 15, n = 2, (between B and C) KI only, n = 3, KI + MMI < 15, n = 3; KI + MMI ≥ 15, n = 1, (between C and D) KI + LT4, n = 2; KI only, n = 3; KI + MMI < 15, n = 3; KI after RI, n = 2, (between D and E) P/R, n = 1; KI+LT4, n = 1; KI only, n = 6; KI + MMI < 15, n = 4; KI after RI, n = 2; Hypo after RI, n = 2. (F) Summarized classification of 288 patients followed for 2 years. KI Group: patients who were well-controlled with KI alone or KI + LT4 throughout 2 years. KI + MMI Group: patients who required combination therapy of KI + MMI even for a short period, without ablative therapy, to maintain euthyroid status. Ablation Group: patients who were treated with 131I therapy (RI) or total thyroidectomy (Op) during 2 years.
Citation: European Thyroid Journal 13, 6; 10.1530/ETJ-24-0085
At 3 months (n = 321) (Fig. 4B), 177 (55.1%) patients were ‘KI-effective’, and 36 (11.2%) had already received RI (‘RI’). In this article, patients within 3 months after RI therapy were classified as ‘RI’.
At 6 months (n = 314) (Fig. 4C), 54.7% of patients were ‘KI-effective’. Among 32 patients who had received RI >3 months previously, 29 (9.2%) were well-controlled with KI or KI + LT4 (‘KI after RI’), while MMI was required in three (1.0%) patients (‘KI + MMI after RI’).
At 1 year (n = 304) (Fig. 4D), 52.0% of patients were ‘KI-effective’. KI was withdrawn in four (1.3%) patients, suggesting possible remission (‘P/R’). Eight (2.6%) patients became hypothyroid after RI therapy (‘Hypo after RI’), and four (1.3%) had undergone thyroidectomy (‘Post Op’).
At 2 years (n = 288) (Fig. 4E), 146 (50.6%) patients were ‘KI-effective’, seven (2.4%) achieved ‘P/R’, and 59 (20.4%) were ‘KI + MMI’. The numbers of ‘KI after RI’ and ‘Hypo after RI’ increased to 39 (13.5%) and 27 (9.4%), respectively.
Classification of 288 patients followed for 2 years
The treatments during the 2 years of follow-up are summarized in Fig. 4F. A total of 123 (42.7%) patients, including 79 (27.4%) who had received LT4 with KI, remained ‘KI-effective’ throughout the 2 years and were classified into the KI Group. Eighty-nine (30.9%) patients required additional MMI, even if temporarily without ablative therapy, and were classified into the KI + MMI Group. Seventy-six (26.4%) patients, including 72 treated with RI therapy and four with thyroidectomy, were classified into the Ablation Group.
During the 2-year period, no patients developed adverse effects of KI, whereas adverse effects of MMI were observed in 21 (16.2%) of 130 patients who received MMI (urticaria: n = 14; liver dysfunction: n = 3; polyarthritis: n = 2; granulocytopenia: n = 2).
TRAb negativity rate at 2 years
At 2 years, 35 of 123 patients (28.5%) in the KI Group, 19 of 89 (21.3%) in the KI + MMI Group and eight of 76 (10.5%) in the Ablation Group, a total of 62 of 288 patients (21.5%), had become TRAb negative. Among them, 15 patients had stopped taking KI, and the remaining 47 were in the process of tapering off their medications.
Clinical data before treatment of 288 patients followed for 2 years: Comparisons between the KI Group and other groups
The pretreatment data of 288 patients followed for 2 years (‘Total’), and those classified into KI Group (n = 123), KI + MMI Group (n = 89), and Ablation Group (n = 76) are shown in Table 1. As the multiple logistic regression analysis showed no overall significant differences between KI+MMI Group and Ablation Group (P = 0.2563) except for a slightly larger thyroid volume in Ablation Group (P = 0.0258), these two groups were combined into the non-KI Group, and the clinical difference was evaluated between KI Group and the non-KI Group. In KI Group, the age was significantly older (P < 0.0001), fT4 (P < 0.0001), fT3 (P < 0.0001), and TRAb (P = 0.0001) levels were lower, the thyroid volume was smaller (P < 0.0001), and positive TPOAb was less frequent (P < 0.0001) compared to the non-KI Group in Wilcoxon’s rank-sum test and Pearson’s χ 2 test. In the multiple logistic regression analysis, significant differences were observed in age (P = 0.0095), fT3 (P = 0.0095), and TPOAb (P = 0.0338).
The distribution of age, fT4, fT3, TRAb, and thyroid volume in the KI Group and the total 288 patients are shown in Table 3. Among patients who were older or had lower levels of fT4, fT3, TRAb, or a smaller thyroid volume in comparison to the total median, 52–66% were classified into the KI Group. However, even when the patients were younger, their fT4, fT3, or TRAb levels were higher, or their thyroid volume was larger than the total median, 10–40% of the patients were classified into the KI Group.
Distribution of patients in the KI Group and all patients (n=288) followed for 2 years for each clinical parameter in which a significant difference was observed between KI Group and other groups.
| Clinical data before KI treatment | KI Group, n | All patients, n | n / total n (%) |
|---|---|---|---|
| Age (years), median = 44* | |||
| <30 | 5 | 45 | 11.1 |
| 30 – 39 | 26 | 62 | 41.9 |
| 40 – 44 | 17 | 37 | 45.9 |
| >44# | 75 | 144 | 52.1# |
| fT4 (ng/dL), median = 3.6* | |||
| >7.0 | 3 | 33 | 9.1 |
| 5.1 – 7.0 | 13 | 46 | 28.3 |
| 3.6 – 5.0 | 18 | 64 | 28.1 |
| <3.6# | 89 | 145 | 61.4# |
| fT3 (pg/mL), median = 12.9* | |||
| >20.0 | 7 | 52 | 13.5 |
| 15.1 – 20.0 | 11 | 54 | 20.4 |
| 12.9 – 15.0 | 9 | 37 | 24.3 |
| <12.9# | 96 | 145 | 66.2# |
| TRAb (IU/L), median = 8.3* | |||
| >30.0 | 7 | 29 | 24.1 |
| 20.1 – 30.0 | 9 | 30 | 30.0 |
| 8.3 – 20.0 | 32 | 85 | 37.6 |
| <8.3# | 75 | 144 | 52.1# |
| Thyroid volume (ml), median = 26.5* | |||
| >50.0 | 3 | 30 | 10.0 |
| 35.1 – 50.0 | 7 | 42 | 16.7 |
| 26.5 – 35.0 | 28 | 72 | 38.9 |
| <26.5# | 85 | 144 | 59.0# |
| Total n and % | 123 | 288 | 42.7 |
KI, potassium iodide; TRAb, TSH receptor antibody.
*Indicates median of all 288 patients followed for 2 years; #patients were older or had lower levels of fT4, fT3, TRAb, or a smaller thyroid volume than the total median, high prevalence (52-66%) of them were classified into the KI Group. However, even in other rows, 10-40% of the patients were also classified into the KI Group.
The reference values in this study were as follows – serum fT4: 0.9–1.8 ng/dL; fT3: 2.3–4.4 pg/mL; TRAb: <1.0 IU/L.
Classification of patients who showed sufficient responses to KI at 4 weeks
Of the 164 patients who showed a sufficient decrease in both fT4 and fT3 levels (KI-effective) at 4 weeks, 149 were followed for 2 years. Of them, 114 (76.5%) remained ‘KI-effective’ for 2 years and were classified into the KI Group, 23 (15.4%) were classified into the KI + MMI Group, and 12 (8.1%) into the Ablation Group.
Additional information on KI + MMI Group
Of the 89 patients in the KI + MMI Group, only 30 (33.7%) continuously required MMI for 2 years, while 23 (25.8%) required it for less than 1 year (Fig. 5A). The maximum daily dose of MMI was 12.2 ± 4.4 mg on average, and it was 5–10 mg in 47 (52.8%) patients (Fig. 5B).
Data on combination therapy with potassium iodide (KI) and methimazole (MMI) or radioactive iodine (RI) therapy in patients with Graves’ hyperthyroidism. (A) Duration (months) for which MMI was required during KI treatment in the KI + MMI Group. Thirty (33.7%) patients continuously required MMI for 2 years. In 23 (25.8%) patients, MMI was used for a short period of less than 1 year. (B) Maximum daily dose of MMI required during KI treatment in the KI + MMI Group. The maximum daily dose of MMI was 12.2 ± 4.4 mg on average, and it was only 5–10 mg/day in 47 of 89 (52.8%) patients. In the KI + MMI Group, only one patient took MMI up to 30 mg/day, and most of the patients who required high doses of MMI (>15 mg/day) promptly chose ablative therapy before they actually took such high doses of MMI. One patient who had required MMI 30 mg/day could reduce the dose to 15 mg/day. In six patients who had taken MMI 20 mg/day, two could reduce the dose to 10 mg/day, three to 5 mg every other day, and one could stop MMI, resulting in ‘KI effective’ state. (C) 123I uptake just before RI therapy in KI-treated patients after restriction of iodide for 5–7 days. Most of the patients could achieve sufficiently high thyroid 123I uptake (median 65.0%/5 h). In our district, the urinary iodide excretion was 585 (274–1362) µg/gCr in euthyroid controls (n = 93), and 228 (124–723) µg/gCr in untreated Graves’ hyperthyroidism (n = 100) (unpublished data).
Citation: European Thyroid Journal 13, 6; 10.1530/ETJ-24-0085
Additional information on Ablation Group
Of the 76 patients in the Ablation Group, 56 patients (including 21 who exhibited adverse effects of MMI) were poorly controlled with medication but were reluctant to even start MMI or increase the dose of MMI due to concerns about adverse effects. In the remaining 20 patients who were well-controlled with KI (n = 10) or KI + MMI (n = 10), the reasons for ablative therapy were as follows: ‘large thyroid volume’, ‘planning future conception’, ‘having complications (e.g. heart disease, diabetes mellitus)’ and ‘for work’.
Regarding the 72 patients treated with RI, 45 took KI alone and 27 took KI + MMI until 5–7 days before. The thyroidal 123I uptake just before taking 131I was 65.0 (46.6–74.3) %/5 h (Fig. 5C). The thyroid volume, the oral dosage of 131I, and the estimated amount of 131I taken into the thyroid gland were 42.2 (31.9–58.6) mL, 481.0 (477.3–499.5) MBq, and 6.43 (4.90–8.38) MBq/mL, respectively. All 72 patients had a reduction in their thyroid volume and eventually became well-controlled by KI 50 mg + LT4 or LT4 only.
Changes in thyroid volume in each group
In the KI Group (n = 123), no significant differences in thyroid volume were observed between pretreatment and 2 years: 21.6 (20.2–30.6) mL and 21.0 (16.0–27.0) mL, respectively (P = 0.8257). In contrast, a significant increase was observed in the KI + MMI Group (n = 89): 27.8 (22.7–36.6) mL and 33.2 (26.1–45.1) mL, respectively (P = 0.0009). In the Ablation Group (n = 76), a significant increase was observed between pretreatment and at the time of ablative therapy: 35.3 (25.7–48.6) mL and 42.2 (31.8–58.6) mL, respectively (P < 0.0001).
Discussion
We evaluated the early effects of KI at 2–4 weeks in 324 patients with untreated GD, and the subsequent clinical course for 2 years in 288 patients.
After starting KI, fT4 and fT3 levels decreased in all 324 patients and became normal or low in 74.7% and 50.6% at 4 weeks, respectively. This suggested that thyrotoxicosis could be controlled with KI (KI-effective) within 1 month in approximately 50% of GD patients. From the result that one-third of patients with a sufficient decrease in fT4 still showed high fT3 levels, the measurement of fT3 is important to evaluate the efficacy of KI when TSH is not detectable.
The first concern is the subsequent efficacy of KI. In our cross-sectional survey over 2 years, >50% of the patients were ‘KI-effective’ at any time point, and 42.7% of the 288 patients followed for 2 years remained ‘KI-effective’ throughout the 2 years (KI Group). As our result was similar to that reported by Okamura et al. (26), the percentage of KI-sensitive patients, in whom GD can be well-controlled with KI, may be 34–43% in Japan. This high ratio of KI-sensitive patients might be partly explained by the increase in mild GD patients found by widespread medical checkups. However, the use of high doses of KI and the addition of LT4 without immediately reducing the KI dose, as seen in two-thirds of the KI Group, may have increased the number of KI-sensitive patients in this study.
The threshold of the serum non-hormonal iodine (NHI) level for the onset of the Wolff-Chaikoff (WC) effect was estimated to be 2.5 μM (7). Okamura et al. (26) reported that the serum NHI levels in patients taking KI 100 and 200 mg/day were 19.4 (14.1–25.7) μM and 32.2 (25.6–40.9) μM, respectively, and that there was a significant increase in KI-resistant patients when the serum NHI was <10 μM. We speculate that the effective NHI threshold at which the WC effect can be sustained without clinical escape is much higher than the threshold for its onset, varies between individuals, and depends on the activity of GD and sodium/iodide symporters (8). Even when a strong KI effect is observed, if the KI dose is reduced and the NHI level falls below the effective threshold, thyrotoxicosis will easily recur (26). Therefore, it seems important to maintain a sufficient KI dose by using combination therapy with LT4 until the activity of GD declines.
The second concern is how to differentiate KI-sensitive patients from KI-resistant ones. Previous reports suggested that KI was more effective in older patients with mild GD (21, 22, 24-26), and these results were similar to those of our study. However, as shown in Table 3, the pretreatment data in the KI Group showed a high degree of variation, and even when the patients had shown some worse data in comparison to the median value of the total population, 10–40% of the patients were still classified into the KI Group.
Regarding this problem, it is a remarkable finding that 76.5% of patients in whom KI was effective at 4 weeks remained ‘KI-effective’ for 2 years and were classified into the KI Group. This suggests that the efficacy of KI at 4 weeks may be a good predictor of its long-term efficacy. As the degree of decline in thyroid hormone levels at 2–4 weeks was extremely varied (Fig. 3), it seems impossible to predict patients for whom KI will be effective at 4 weeks according to pretreatment data. Therefore, we propose a sufficient dose of KI (e.g. 100 mg/day) as the first treatment for GD. In half of GD patients, KI may become effective at 4 weeks and may remain effective for 2 years in three-quarters of them. In the other KI-resistant patients, MMI can be added at 4 weeks before the exacerbation of thyrotoxicosis.
The third concern is the clinical course of KI-resistant patients. Among the 89 patients in the KI + MMI Group, only 33.7% continuously required MMI for 2 years, and the average maximum daily dose of MMI was only 12.2 mg. These results suggest that KI resistance may not be permanent and that preceding KI might increase the efficacy of MMI (26). However, since the Ablation Group included patients for whom GD was poorly controlled by KI+MMI, the additive effect of KI + MMI may not be apparent in some refractory cases. Regarding the possible increase in thyroid volume during KI therapy (12, 21, 25), it might depend on the activity of GD, because no significant increase was observed in the KI Group. It seems reasonable to recommend ablative therapy for patients with a marked increase in thyroid volume.
In this study, 72 patients received RI therapy. As shown in Fig. 5C, they could achieve sufficiently high thyroid 123I uptake (median 65.0%/5 h), as previously reported (26, 27). They were well-controlled with KI or KI + LT4 after RI therapy (‘KI after RI’) until they became hypothyroid. If ‘KI after RI’ and ‘KI-effective’ were summed up, >60% of patients were well-controlled with KI from 6 months to 2 years.
The limitations of this study included some dropped out during the study and relatively short follow-up period, which did not allow for the assessment of long-term outcomes. However, Okamura et al. previously reported that the long-term prognosis of KI-treated GD patients over 5-10 years was almost the same as that of thionamide-treated patients (21, 26, 28). KI is inexpensive (JPY 5 per 50-mg pill) and is rarely associated with adverse effects. Although novel antithyroid drugs such as TSH receptor antagonists (29, 30) have recently been introduced, the clinical usefulness of KI should be re-examined, at least in iodine-sufficient areas.
Conclusions
A sufficient dose of KI (e.g. 100–200 mg/day) was effective in >50% of GD patients at any time point from 4 weeks to 2 years, and 42.7% of GD patients (76.5% of ‘KI-effective’ patients at 4 weeks) remained ‘KI-effective’ throughout the 2 years. In the KI-resistant period, which might not be permanent, MMI was added or RI therapy was performed successfully.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agencies in the public, commercial, or not-for-profit sectors.
Acknowledgements
The authors thank Professor Takanari Kitazono of the Department of Medicine and Clinical Science at Kyushu University for his assistance in conducting this study, and Dr Brian Quinn for correcting the English.
References
- 1↑
Plummer HS. Results of administering iodine to patients having exophthalmic goiter. JAMA 1923 80 1955.
- 2↑
Thompson WO, Thompson PK, Brailey AG, & Cohen AC. Prolonged treatment of exophthalmic goiter by iodine alone. Archives of Internal Medicine 1930 45 481–502. (https://doi.org/10.1001/archinte.1930.00140100003001)
- 3↑
Emerson CH, Anderson AJ, Howard WJ, & Utiger RD. Serum thyroxine and triiodothyronine concentrations during iodide treatment of hyperthyroidism. Journal of Clinical Endocrinology and Metabolism 1975 40 33–36. (https://doi.org/10.1210/jcem-40-1-33)
- 4↑
Roti E, Robuschi G, Manfredi A, D’Amato L, Gardini E, Salvi M, Montermini M, Barlli AL, Gnudi A, & Braverman LE. Comparative effects of sodium Ipodate and iodide on serum thyroid hormone concentrations in patients with Graves’ disease. Clinical Endocrinology 1985 22 489–496. (https://doi.org/10.1111/j.1365-2265.1985.tb00148.x)
- 5↑
Tan TT, Morat P, Ng ML, & Khalid BA. Effects of Lugol’s solution on thyroid function in normals and patients with untreated thyrotoxicosis. Clinical Endocrinology 1989 30 645–649. (https://doi.org/10.1111/j.1365-2265.1989.tb00270.x)
- 6↑
Philippou G, Koutras DA, Piperingos G, Souvatzoglou A, & Moulopoulos SD. The effect of iodide on serum thyroid hormone levels in normal persons, in hyperthyroid patients, and in hypothyroid patients on thyroxine replacement. Clinical Endocrinology 1992 36 573–578. (https://doi.org/10.1111/j.1365-2265.1992.tb02267.x)
- 7↑
Wolff J, & Chaikoff IL. Plasma inorganic iodide as a homeostatic regulator of thyroid function. Journal of Biological Chemistry 1948 174 555–564. (https://doi.org/10.1016/S0021-9258(1857335-X)
- 8↑
Eng PH, Cardona GR, Fang SL, Previti M, Alex S, Carrasco N, Chin WW, & Braverman LE. Escape from the acute Wolff-Chaikoff effect is associated with a decrease in thyroid sodium/iodide symporter messenger ribonucleic acid and protein. Endocrinology 1999 140 3404–3410. (https://doi.org/10.1210/endo.140.8.6893)
- 9↑
Astwood EB. Treatment of hyperthyroidism with thiourea and thiouracil. JAMA 1943 122 78–81. (https://doi.org/10.1001/jama.1943.02840190008003)
- 10↑
Feek CM, Sawers JSA, Irvine WJ, Beckett GJ, Ratcliffe WA, & Toft AD. Combination of potassium iodide and propranolol in preparation of patients with Graves’ disease for thyroid surgery. New England Journal of Medicine 1980 302 883–885. (https://doi.org/10.1056/NEJM198004173021602)
- 11↑
Langley RW, & Burch HB. Preoperative management of the thyrotoxic patient. Endocrinology and Metabolism Clinics of North America 2003 32 519–534. (https://doi.org/10.1016/s0889-8529(0300010-0)
- 12↑
Yabuta T, Ito Y, Hirokawa M, Fukushima M, Inoue H, Tomoda C, Higashiyama T, Kihara M, Uruno T, Takamura Y, et al.Preoperative administration of excess iodide increases thyroid volume of patients with Graves’ disease. Endocrine Journal 2009 56 371–375. (https://doi.org/10.1507/endocrj.k08e-240)
- 13↑
Schimmel M, & Utiger RD. Acute effect of inorganic iodide after 131I therapy for hyperthyroidism. Clinical Endocrinology 1977 6 329–332. (https://doi.org/10.1111/j.1365-2265.1977.tb02018.x)
- 14↑
Ross DS, Daniels GH, Stefano PD, Maloof F, & Ridgway EC. Use of adjunctive potassium iodide after radioactive iodine (131I) treatment of Graves’ hyperthyroidism. Journal of Clinical Endocrinology and Metabolism 1983 57 250–253. (https://doi.org/10.1210/jcem-57-2-250)
- 15↑
Sato S, Noh JY, Sato S, Suzuki M, Yasuda S, Matsumoto M, Kunii Y, Mukasa K, Sugino K, Ito K, et al.Comparison of efficacy and adverse effects between methimazole 15 mg+inorganic iodine 38 mg/day and methimazole 30 mg/day as initial therapy for Graves’ disease patients with moderate to severe hyperthyroidism. Thyroid 2015 25 43–50. (https://doi.org/10.1089/thy.2014.0084)
- 16↑
Takata K, Amino N, Kubota S, Sasaki I, Nishihara E, Kudo T, Ito M, Fukata S, & Miyauchi A. Benefit of short-term iodide supplementation to antithyroid drug treatment of thyrotoxicosis due to Graves’ disease. Clinical Endocrinology 2010 72 845–850. (https://doi.org/10.1111/j.1365-2265.2009.03745.x)
- 17↑
Shiroozu A, Okamura K, Ikenoue H, Sato K, Nakashima T, Yoshinari M, Fujishima M, & Yoshizumi T. Treatment of hyperthyroidism with a small single daily dose of methimazole. Journal of Clinical Endocrinology and Metabolism 1986 63 125–128. (https://doi.org/10.1210/jcem-63-1-125)
- 18↑
Okamura K, Ikenoue H, Shiroozu A, Sato K, Yoshinari M, & Fujisima M. Reevaluation of the effects of methylmercaptoimidazole and propylthiouracil in patients with Graves' hyperthyroidism. Journal of Clinical Endocrinology and Metabolism 1987 65 719–723. (https://doi.org/10.1210/jcem-65-4-719)
- 19↑
Nakamura H, Noh JY, Itoh K, Fukata S, Miyauchi A, Hamada N & Working Group of the Japan Thyroid Association for the Guideline of the Treatment of Graves’ Disease. Comparison of methimazole and propylthiouracil in patients with hyperthyroidism caused by Graves’ disease. Journal of Clinical Endocrinology and Metabolism 2007 92 2157–2162. (https://doi.org/10.1210/jc.2006-2135)
- 20↑
Ostuka F, Noh JY, Chino T, Shimizu T, Mukasa K, Ito K, Ito K, & Taniyama M. Hepatotoxicity and cutaneous reactions after antithyroid drug administration. Clinical Endocrinology 2012 77 310–315. (https://doi.org/10.1111/j.1365-2265.2012.04365.x)
- 21↑
Okamura K, Sato K, Fujikawa M, Bandai S, Ikenoue H, & Kitazono T. Remission after potassium iodide therapy in patients with Graves’ hyperthyroidism exhibiting thionamide-assosiated side effects. Journal of Clinical Endocrinology and Metabolism 2014 99 3995–4002. (https://doi.org/10.1210/jc.2013-4466)
- 22↑
Honda A, Uchida T, Komiya K, Goto H, Takeno K, Sato J, Suzuki R, Himuro M, & Watada H. Relationship between the effectiveness of inorganic iodine and the severity of Graves’ thyrotoxicosis: a retrospective study. Endocrine Practice 2017 23 1408–1413. (https://doi.org/10.4158/EP-2017-0044)
- 23↑
Okamura K, Bandai S, Fujikawa M, Sato K, & Kitazono T. Clinical experience of treating Graves’ hyperthyroidism complicated with malignancy - the possible role of potassium iodide for avoiding the risk of thionamide-associated neutropenia. Endocrine Journal 2020 67 751–758. (https://doi.org/10.1507/endocrj.EJ20-0016)
- 24↑
Uchida T, Goto H, Kasai T, Komiya K, Takeno K, Abe H, Shigihara N, Sato J, Honda A, Mita T, et al.Therapeutic effectiveness of potassium iodine in drug-naïve patients with Graves’ disease: a single-center experience. Endocrine 2014 47 506–511. (https://doi.org/10.1007/s12020-014-0171-8)
- 25↑
Suzuki N, Yoshimura Noh J, Sugisawa C, Hoshiyama A, Hiruma M, Kawaguchi A, Morisaki M, Ohye H, Suzuki M, Matsumoto M, et al.Therapeutic efficacy and limitations of potassium iodide for patients newly diagnosed with Graves’ disease. Endocrine Journal 2020 67 631–638. (https://doi.org/10.1507/endocrj.EJ19-0379)
- 26↑
Okamura K, Sato K, Fujikawa M, Bandai S, Ikenoue H, & Kitazono T. Iodine-sensitive Graves’ hyperthyroidism and the strategy for resistant or escaped patients during potassium iodide treatment. Endocrine Journal 2022 69 983–997. (https://doi.org/10.1507/endocrj.EJ21-0436)
- 27↑
Mikura K, Noh JY, Watanabe N, Aida A, Yoshimura R, Kinoshita A, Suzuki A, Suzuki N, Fukushita M, Matsumoto M, et al.Radioiodine uptake after monotherapy with potassium iodide in patients with Graves’ disease. Endocrine Journal 2023 70 541–549. (https://doi.org/10.1507/endocrj.EJ22-0505)
- 28↑
Bandai S, Okamura K, Fujikawa M, Sato K, Ikenoue H, & Kitazono T. The long-term follow-up of patients with thionamide-treated Graves’ hyperthyroidism. Endocrine Journal 2019 66 535–545. (https://doi.org/10.1507/endocrj.EJ18-0418)
- 29↑
Neumann S, Eliseeva E, McCoy JG, Napolitano G, Giuliani C, Monaco F, Huang W, & Gershengorn MC. A new small-molecule antagonist inhibits Graves’ disease antibody activation of the TSH receptor. Journal of Clinical Endocrinology and Metabolism 2011 96 548–554. (https://doi.org/10.1210/jc.2010-1935)
- 30↑
Furmaniak J, Sanders J, Sanders P, Li Y, & Smith BR. TSH receptor specific monoclonal autoantibody K1-70 targeting of the TSH receptor in subjects with Graves’ disease and Graves’ orbitopathy - Results from a phase I clinical trial. Clinical Endocrinology 2022 96 878–887. (https://doi.org/10.1111/cen.14681)
