Graves' disease (GD) is an autoimmune thyroid disorder caused by stimulating auto-antibodies to the thyrotrophin (TSH) receptor on thyroid follicular cells. It is the most common cause of hyperthyroidism and approximately 3% of women and 0.5% of men develop GD in their lifetime.

RAI has been shown to be a cost-effective and safe therapy in patients with GD but with some disadvantages. In addition, despite its proven efficacy and safety, many patients do not wish to undergo RAI because of radiation fear and prefer to either continue ATDs or have surgery.

High-intensity focused ultrasound (HIFU) is a non-invasive procedure that involves the application of a high-energy focused beam for thermal tissue ablation within a targeted zone. Similar to the principle of RAI (i.e. using ionizing radiation to ablate thyroid parenchyma and cause GD remission), we postulated that the heat energy generated from HIFU could also be used to ablate the thyroid parenchyma and cause GD remission. The idea of using heat energy to ablate thyroid parenchyma minimally invasively was recently reported using radiofrequency ablation but to our knowledge, we are one of the first (if not the first) group to propose using HIFU energy to ablate thyroid parenchyma as a definitive treatment for relapsed GD.

Having obtained ethical approval, a pilot study was conducted to examine the efficacy and safety of HIFU as a treatment for relapsed GD. In the pilot study, all patients underwent a safe and successful HIFU ablation for relapsed GD. Based on the results of the pilot study, we hypothesize that a single HIFU treatment to the thyroid gland may be as effective as our standard outpatient fixed-dose of RAI (370MBq) in causing remission of GD at 6-month. If our hypothesis turns out to be true, HIFU could become a treatment option for patients who are indicated for RAI but do not wish to have it because of one reason or another. HIFU appears to induce a faster disease remission and lessen the need of deferring pregnancy and radiation precautions because of the absence of radioactivity.

Graves' disease (GD) is an autoimmune thyroid disorder caused by stimulating auto-antibodies to the thyrotrophin (thyroid stimulating hormone [TSH]) receptor on thyroid follicular cells. It is the most common cause of hyperthyroidism and approximately 3% of women and 0.5% of men develop GD in their lifetime1. In our locality, like Europe and Japan, antithyroid drugs (ATDs) such as carbimazole or propylthiouracil have been preferred over radioactive iodine (RAI) and surgery as the initial treatment of GD2-4. This is because ATDs are relatively easy to administer, could induce disease remission (30-70%) and avoid life-long thyroid hormone replacement, operative risks and radioactivity2-4. However, because of possible side-effects, they are only recommended for a period of 12 to 18 months. Taking ATDs for a longer period does not seem to increase the chance of remission5. Therefore, currently, once the disease has relapsed after a 18-month of ATD treatment, more definitive treatment modalities like RAI or surgery are indicated. Regarding to which is more preferable, surgery is usually advised in patients with large compressive goiter (>80g), suspected or documented thyroid malignancy, planning to become pregnant within 6 months and moderate to severe Graves' ophthalmopathy (GO)2-4. As a result, most patients without these conditions are considered for RAI2-4.

RAI has been shown to be a cost-effective and safe therapy in patients with GD4,6,7. In North America, clinical endocrinologists favor RAI as the initial treatment for GD2-4. However, since RAI is usually prescribed on an outpatient basis, it is necessary to consider nearby individuals' exposure doses and formulate radiation precautions carefully8. Other disadvantages include its slow induction of euthyroidism, potential worsening of GO and deferral of pregnancy2-4. Although various dosing techniques have been described to ensure an adequate radiation delivery to the thyroid gland, the simplest and most effective method has been to administer a fixed dose of RAI2. However, due to local regulations and our densely-populated areas, administering higher RAI doses (>400MBq) is not permitted on an outpatient basis9. As a result, the usual RAI dose usually ranges between 185 to 370MBq (i.e. up to 10mci)10. In addition, despite its proven efficacy and safety, many patients do not wish to undergo RAI because of radiation fear and prefer to either continue ATDs or have surgery11.

High-intensity focused ultrasound (HIFU) is a non-invasive procedure that involves the application of a high-energy focused beam for thermal tissue ablation within a targeted zone. It has been applied in a variety of medical conditions including uterine fibroids and prostate, breast, pancreatic, and liver tumors12. Unlike other ablation devices (like radiofrequency or laser ablation), HIFU does not require needle puncture and is considered safer and less operator-dependent13. Single HIFU treatment has been shown to reduce size of benign thyroid nodules13,14. Histologic examination confirmed that HIFU induced complete tissue necrosis within a targeted area13. Similar to the principle of RAI (i.e. using ionizing radiation to ablate thyroid parenchyma and cause GD remission), we postulated that the heat energy generated from HIFU could also be used to ablate the thyroid parenchyma and cause GD remission. The idea of using heat energy to ablate thyroid parenchyma minimally invasively was recently reported using radiofrequency ablation15 but to our knowledge, we are one of the first (if not the first) group to propose using HIFU energy to ablate thyroid parenchyma as a definitive treatment for relapsed GD.

Having obtained ethical approval, a pilot study was conducted to examine the efficacy and safety of HIFU as a treatment for relapsed GD. Over a 2-month period, 20 patients underwent a single HIFU treatment for relapsed GD. The treatment involved ablating the entire right, left and central (isthmic) lobes with HIFU pulses. To avoid inadvertent injury to heat-sensitive structures like the recurrent laryngeal nerve, parathyroid glands, trachea and esophagus, we deliberately left 2-3ml of thyroid parenchyma non-ablated close to the trachea-esophageal groove on each side. The reason for not leaving a larger amount of non-ablated parenchyma is because from the experience in subtotal thyroidectomy, leaving > 6ml of normal parenchyma may diminish the long-term remission rate16.

In the pilot study, all patients underwent a safe and successful HIFU ablation for relapsed GD. By ultrasound (USG) volumetry, the mean pre-ablation total thyroid volume was 18.5 ± 6.4ml. The average treatment time was 72.7 ± 31.0 mins and the total energy delivered to each patient was 21.4 ± 5.9 KJ. No patients suffered any major complications afterwards except for 2 patients (10.0%) who had minor neck redness in the first week. By 4-week, all patients were either biochemically euthyroid (n=15) (without ATDs) or biochemically hypothyroid (requiring thyroxine replacement) (n=5). At 6-month, 15 (75.0%) patients remained biochemically euthyroid or hypothyroid without ATDs (i.e. remission) while 5 patients (25.0%) had biochemical hyperthyroidism. However, of these, only 3 required ATDs and 2 were kept observed.

Based on the results of the pilot study, we hypothesize that a single HIFU treatment to the thyroid gland may be as effective as our standard outpatient fixed-dose of RAI (370MBq) in causing remission of GD at 6-month. If our hypothesis turns out to be true, HIFU could become a treatment option for patients who are indicated for RAI but do not wish to have it because of one reason or another. Apart from the benefit of being an outpatient treatment, HIFU appears to induce a faster disease remission and lessen the need of deferring pregnancy and radiation precautions because of the absence of radioactivity.

Inclusion Criteria:

1. A diagnosis of GD based on standard clinical criteria (elevated thyroid hormone levels, suppressed TSH, positive TSHR auto-antibody, diffuse goiter ± high RAI uptake).
2. Aged between 18 and 70 years old at the time of informed consent
3. Have had at least one GD relapse. A relapse is defined as recurrent or persistent hyperthyroidism despite completing a continued course of ATDs for 18 months or more.
4. Indicated for RAI treatment
5. Valid consent is obtained

Exclusion Criteria:

1. Prefer or have clear indications for surgery (such as those with large compressive goiter, suspected or documented thyroid malignancy, wishing to become pregnant within 6 months or with moderate to severe GO).
2. Have pre-existing vocal cord palsy
3. Are unable to move or extend their neck
4. Have either right, left or central (isthmic) lobe measuring >30ml in volume on pre-treatment USG volumetry (see later)
5. Have concomitant thyroid nodules which are either indeterminate, suspicious of malignancy or proven malignant on fine needle aspiration cytology (FNAC).
6. Have active or severe Graves ophthalmopathy (GO)
7. Are pregnant, lactating or planning for pregnancy within 6 months
8. Have any medical conditions that would make them too ill to undergo intravenous sedation or treatment
9. Any condition that is unstable or can jeopardize the safety of the patients and their compliance to the study