Rescue pre-operative treatment with Lugol’s solution in uncontrolled Graves’ disease

Introduction

Graves’ disease (GD) is common and can be treated by medication blocking thyroid hormone synthesis with drugs such as methimazole, carbimazole or propylthiouracil. With medication for 12–18 months, cure can be achieved, which is sustained in around 50–55% (1). Another option is surgery, which is often considered in recurrence of GD after medical therapy, in patients with large goitres, and when pregnancy is planned in the near future. The third possibility is treatment with radioactive iodine, where thyrocytes are exposed to local radiation, with a subsequent decline of hormone secretion. In these two latter alternatives hypothyroidism is a deliberate goal, treated by life-long levothyroxine.

Pharmacologic therapy is often chosen in Europe as it is mostly well-tolerated. However, there are well-known side effects. Most feared is agranulocytosis, but other potential side effects include hepatotoxicity, loss of taste, arthralgia and most commonly rash, which can be severe. As medical treatment is planned for a considerable time, even mild side effects can make it necessary to change therapy. Surgery is much safer if the patient is euthyroid, as the thyroid gland is highly vascularised, and anaesthetic as well as bleeding complications are more frequent in untreated thyrotoxicosis (2).

Lugol’s solution (LS) was invented in 1829 by the French physician Jean Guillaume August Lugol as a cure for tuberculosis. It is a solution of elemental iodine (5%), potassium iodide (10%) and distilled water. It has been used as a disinfectant, as a reagent for starch detection in organic compounds, in histologic preparations, in dental procedures and in diagnosis of cervical cell alterations, also known as the Schiller’s test. Already in the 1920s LS was given as a pre-treatment to thyroid surgery (3). It became the standard pre-operative treatment to control hyperthyroidism. However, with the development of pharmacologic agents blocking the thyroid hormone synthesis and radioactive iodine, the need for pre-treatment with LS diminished.

LS causes an increased thyroidal iodide uptake, inhibits thyroid peroxidase which attenuates iodide oxidation and organification (4), as well as blocking the release of thyroid hormones (5). This acute Wolff–Chaikoff effect results in an increase in intrathyroid iodine concentration within 24–48 h and a decrease in thyroid hormone synthesis in normal rats (6). Escape from the acute Wolff–Chaikoff effect is associated with a decrease in thyroid sodium/iodide symporter activity, causing a decrease in intrathyroid iodide concentration (7). Thus, there is a therapeutic window between achieved euthyroidism and an escape phenomenon, illustrated by a patient treated with iodine for more than 30 days who developed serious hyperthyroidism (8).

In untreated hyperthyroidism 0.5 mL (375 mg) daily treatment with LS significantly reduces both free thyroxine (FT4) and free triiodothyronine (FT3) within 5–10 days, while TSH remains below the limit of detection (9). Aside from the effects on FT4 and FT3 levels, iodine also decreases the rate of blood flow, thyroid vascularity and intraoperative blood loss during thyroidectomy (10, 11). When medical therapy with anti-thyroid drugs fails due to side effects, such as agranulocytosis, rescue treatment is warranted to avoid atrial flutter and thyroid storm. One option is treatment with beta-blockers (12, 13), which could be used alone in high doses or preferably together with iodine (14, 15). With the advancement of therapeutic options LS before thyroidectomy is no longer a routine in many countries (16, 17), although it is still advocated in ATA guidelines with 5–7 drops thrice daily for 10 days (18). The use of LS currently in countries where it is not routinely recommended is unclear.

The aim of this study was to investigate which factors lead to iodine treatment in patients with GD, outcome effects of LS on thyroid hormone levels and clinical hyperthyroidism by measure of heart rate changes at a major centre where LS is not the mainstay of therapy before thyroidectomy.

Materials and methods

This retrospective study was conducted at the Department of Endocrinology, Metabolism and Diabetes, Karolinska University Hospital, Stockholm, Sweden. All patients with an International Classification of Diseases version 10 (ICD-10), with a code of E050 (Graves’ thyrotoxicosis) and treated with iodine during 2005–2015 were included and all medical files were reviewed manually. All specialist out-patient visits and admissions in Sweden are coded with ICD-10 codes by the attending physician and are thereafter stored in both local and national databases (19). Treatment with oral drops using the dental solution (iodine and iodide potassium) with the same concentration as LS (APL Pharma Specials, Stockholm, Sweden) was used to control hyperthyroidism. Iodine concentration per drop was also measured.

The study was in accordance with the ethical standards of the institutional, national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. For this type of retrospective study formal consent is not required.

Results

Twenty-seven patients (25 females) were treated with LS. Median age was 39 (23–64) years. All had GD with an elevation of FT4 and FT3, suppressed TSH, and measurable thyroid stimulating hormone receptor auto-antibodies (TRAb). They had all been treated with medical therapy previously, with the exception of one patient with primary neutropenia; eight (30%) had recurrent disease after finishing 18 months of previous medical therapy. One (4%) had also had a prior treatment with radioactive iodine. Four (15%) had thyroid associated ophthalmopathy with CAS ≥3. Two (7%) had moderate thyroid associated ophthalmopathy. Twelve subjects were smokers (44%), five (19%) were ex-smokers and 10 were (37%) non-smokers.

Twenty-six patients (96%) had a prior treatment with methimazole, and 16 (59%) had also tried to switch to propylthiouracil due to side effects. LS treatment was initiated in an out-patient setting in twelve patients (44%) with a dose of drops containing 6.7 mg/drop, most often 5 drops thrice, i.e., 100.5 mg iodine daily (timepoint 0). In all cases a set date for surgery was decided before iodine treatment was started, 7–10 days before thyroidectomy. Therapy was initiated at the ward for remaining fifteen patients (56%). In patients with unsufficient response, measured as elevated heart rate >80 and/or still evidence of biochemical hyperthyroidism, the dose was doubled after 2–5 days (n = 15, 58%) (timepoint 1). A further re-evaluation was made at day 5 (5–9) (timepoint 2). All patients were admitted for the last part of the iodine treatment (6 days (0–12)), and recovery after surgery (1 day (1–3)). In all, patients received 1005 mg iodine (168–2110 mg) (Fig. 1). Initial dose in three patients was 1 drop thrice daily, and the one patient later treated with radioiodine was only taking LS for three days, explaining the wide range. Admission was planned in 22 patients (81%), while five patients (19%) were admitted acutely with a combination of atrial flutter and heart failure in one, rash in two and agranulocytosis and sepsis in two patients. A heart rate <80 was required before surgery. All but one patient (96%) were also treated with the beta-blocker propranolol as a part of the pre-operative treatment. The dose was initially 120 mg daily (40–240), and was increased during LS therapy to 130 mg (80–320) (P = 0.005). Treatment with adjunctive therapy as cholestyramine or corticosteroids was not used.

View Full Size

Accumulative doses of iodine solution in 27 patients with uncontrolled Graves’ disease.

Citation: Endocrine Connections 6, 4; 10.1530/EC-17-0025

• Download Figure
• Download figure as PowerPoint slide

Background information and side effects necessitating treatment with LS are shown in Table 1. The biochemical results at the commencement of LS and at follow-up are presented in Fig. 2. TSH levels did not change during treatment. FT4 levels were initially 53 pmol/L (20–100), after 2–3 days, 42 (18–100) (P = 0.005) and after 6–9 days, 20 (8–52) (P = 0.0002). During these time intervals FT3 levels were 20 pmol/L (6–50), 10 (5.7–36) (P = 0.08) and 6.5 (4.3–8.3) (P = 0.04), respectively. Heart rate changed from 87 beats/min (72–116) to 80 (54–120) (P = 0.08) and 76 (60–96) (P = 0.0007) following treatment (Fig. 2). Twenty-six (96%) subjects had total thyroidectomy and one (4%) was treated with radioactive treatment two months following LS (Table 1). The duration of treatment was 8.5 days (4–13). Four patients (15%) had adverse effects of iodine treatment (rash n = 2, rash and vomiting n = 1 and swelling of fingers n = 1). This was managed with dose reduction in two and stopping LS prematurely in two. These four women were aged 30, 35, 43 and 61 years, respectively, and the LS dose ranged from 225 to 1708 mg. Thyroidectomy was uneventful in all 26 patients who underwent operation. However, 7 patients (27%) developed temporary post-operative hypocalcaemia, treated with alfacalcidol and calcium carbonate in 5 (19%) and with only the latter in two (8%).

View Full Size

Initial TSH, free T4, free T3 and pulse, and changes during Lugol’s solution. *Indicates P < 0.05 compared to timepoint 0.

Citation: Endocrine Connections 6, 4; 10.1530/EC-17-0025

• Download Figure
• Download figure as PowerPoint slide

Discussion

Graves’ disease is a common condition, however, it can have serious consequences if not treated promptly (21). If anti-thyroid drugs fail or cannot be used, surgery or radioiodine therapy are the treatment options. However, pre-treatment is often necessary, especially before surgery to improve outcome. LS has been used in this setting but very little has been published about its use. To the best of our knowledge, this is the largest study to report on LS in patients with Graves’ disease who have failed medical therapy. The most common reason for failure was side effects, but all were successfully treated with LS pre-operatively or before radioiodine treatment.

LS significantly decreased plasma thyroid hormone levels and heart rate. FT4 decreased relatively more than FT3 levels. As the half time for T4 is considerably longer (6–7 days) than that for T3 (24 h) (22), and T4 is more tightly bound to thyroxin-binding globulin, a sharper decrease in FT3 than in FT4 could be expected. However, our findings may be explained by the observation in smaller studies indicating that thyroid hormone release is also affected by LS (5, 23).

Median length of stay at the Karolinska University Hospital for thyroidectomy is 25 h, compared to six days in the present study which includes only patients with more complicated disease. Thus, the cost for these patients is considerable. The final days of pre-operative optimization with LS was performed in the in-patient setting, which may not be necessary. However, in those with adherence issues, or agranulocytosis and sepsis, the in-patient optimization seems reasonable. In cases with contraindication to medical therapy radioiodine treatment could be an option. However, this is rarely performed in uncontrolled GD, since there is a risk of aggravation of the thyrotoxicosis (24), and treatment failure (25, 26) may be due to high I¹³¹ turnover (27). Alternative pre-operative treatment in this cohort could be treatment with beta-blockers alone (12, 13), or radioiodine therapy, which was avoided due to a combination of uncontrolled GD, thyroid size, high TRAb levels and thyroid associated ophthalmopathy. Further potential treatments could have been the addition of cholestyramine (28), and corticosteroids in cases of thyroid storm (29).

During surgery all patients were well, indicating a clinically satisfactory effect of LS. It is also worth noting the non-hormone positive effects of LS on risks of bleeding and thyroid vascularity (10, 11), however this was not evaluated in the current study. The mechanism has been suggested to be mediated by decreased angiogenesis and blood flow since the thyroid arterial blood flow, VEGF and IL-16 are all significantly reduced by LS (30). As surgery in our patients was uneventful, and peripheral thyroid hormones did not become completely normal, a contributing effect on the vasculature of iodine solution is plausible. In this cohort, temporary hypocalcaemia was seen in 27% post-operatively. This was lower than that previously reported in more stable patients with 44–65% experiencing transient hypocalcaemia (31, 32).

Doses of LS differ in the literature. Individualized iodine dose in euthyroid GD patients also treated with anti-thyroid drugs was 0.8 mg/kg iodine daily in the study by Yilmaz et al. (11). Erbil et al. used 10 drops thrice daily for ten days (10), a duplication of the dose we initially administered, whereas the ATA guidelines recommend 5–7 drops thrice daily (18). In another study 150 mg iodine daily was used (33). Thus, different doses have been applied which makes comparisons difficult. It can be argued that the dose in our cohort should be higher as our subjects generally had untreated disease due to side effects of anti-thyroid therapy. Optimal doses and duration of therapy to achieve the positive effects on vascularity and bleeding complications are also unclear. Hypersensitivity to LS is a known side effect and this was developed by 15% of our cohort. We could not find any relationship between LS dose and side effects. Other adverse effects of LS besides allergic reactions include sialadenitis and gastrointestinal disturbances, which none of our patients experienced.

An intriguing aspect is the risk of exacerbating hyperthyroidism by time with prolonged iodine treatment. The therapeutic window has been claimed to be 10–14 days after iodine administration before the blocking effect vanishes. It has been reported that T4 and T3 start to increase again after 21 days of iodine treatment in hyperthyroid patients (33). On the other hand, in Japan long-term treatment with iodine has been used, alone or together with anti-thyroid drugs, to achieve euthyroidism with good results (34). Whether this also works in patients with other ethnic backgrounds and iodine sufficiency is not known and should be explored in future trials.

The inherent limitations of all retrospective studies, particularly in that of ascertainment bias, are also present in this study. Moreover, the number of patients included was limited. However, we did not exclude any patients commenced on LS, we had comprehensive medical files and the study is the largest so far in this setting.

In conclusion, the treatment of uncontrolled hyperthyroidism with LS is safe and it decreases thyroid hormones and pulse frequency. Side effects were limited. LS could be recommended pre-operatively in GD with failed medical treatment, especially if side effects to anti-thyroid drugs have occurred.

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 was an academic investigation with no external funding.