This study was a single-center, retrospective investigation conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Xiamen Medical University (Ethical Approval No. 2,023,052). All participants or their legal guardians provided written informed consent. Our study protocol entailed treating patients with GD using ultrasound-guided TA, with subsequent follow-up visits scheduled at 1, 3, 6, and 12 months after the procedure. The procedure was performed by Dr. Lixin Zhou, a thyroid surgeon with extensive experience in managing thyroid diseases.

GD diagnosis was based on clinical manifestations of hyperthyroidism, diffuse thyroid enlargement, suppressed serum thyroid Stimulating Hormone (TSH) levels, and elevated serum thyroid hormone concentrations [15,16,17,18]. Additional diagnostic criteria included exophthalmos, pretibial myxedema, and elevated thyroid-stimulating hormone receptor antibodies (TRAb) levels. In this study, the measurement of TRAb levels served as a substitute for thyroid stimulating antibodies (TSAb) levels, both for the diagnosis and for the prognostic evaluation of GD.

Patients who met these diagnostic criteria were included in the study. Those who declined ATDs, RIT, or surgical intervention for various reasons—such as ATD treatment failure, reluctance for long-term medication, concerns about radioactivity, cosmetic considerations, or risk of postoperative hypothyroidism—were also eligible. In total, 50 out of 57 patients who met these criteria between October 2017 and December 2021 were included in the study, with seven lost to follow-up.

Clinical characteristics, preoperative symptoms, and postoperative complications were meticulously recorded. Thyroid volume was assessed post-procedure using ultrasound imaging, employing the ellipsoid volume formula [19]. The VRR was calculated using the formula [(V0-V1)/V0] x 100, where V0 and V1 represent baseline and post-ablation thyroid volumes, respectively. Descriptive statistics were used to summarize changes in various clinical parameters, including thyroid hormone levels and BMR, calculated using the Gale method [20].

Ultrasound-guided thermal ablation (TA) can be executed utilizing various modalities, including microwave ablation (MWA), radiofrequency ablation (RFA), or laser techniques. Microwave ablation (MWA) therapy is characterized by rapid inactivation of the target tissue with a larger scope, shorter ablation time, less operator fatigue, and a more manageable goal of near-total thyroid ablation. Radiofrequency ablation (RFA) is also used, but the time required to ablate the same volume of target tissue is longer than that of microwave, and laser ablation requires even more time. At our center, microwave and radiofrequency techniques are primarily used for TA. The portable color ultrasound M-Turb (Sono Sound) is employed for ultrasonic image acquisition and TA guidance. The probe for detecting external organs is a 4–9 MHz linear array probe. RFA uses an S-1500 radiofrequency therapy system (Shanghai Meide Medical Co., Ltd.); a radiofrequency catheter model 10-131181 is used, with an active tip of 5, 7, 10, or 15 mm according to the baseline nodule volume. MWA is performed using an ECO-100A1 microwave treatment instrument (YIGAO Microwave System Engineering Co., Ltd., Nanjing, Jiangsu Province, China) and an ECO-100AI3 superficial organ ablation needle (16 G, total length: 10 cm, microwave transmitter away from the shaft tip: 3 mm). The power is set at 35 W, and the adjustment range is 30–40 W. In this study, the selection between MWA and RFA was determined by the technological advancements and equipment availability at our facility during different time periods, rather than by patient-specific factors. This approach was integral to our methodology and is critical for understanding the context of our results. Specifically, from October 2017 to December 2019, our facility was equipped with and hence utilized MWA for treating patients. As our technological capabilities evolved, we transitioned to using RFA from January 2020 to December 2021.

Before undergoing ultrasound-guided TA for hyperthyroidism, patients were treated with ATD to control symptoms of thyrotoxicosis. For patients with a heart rate above 90 beats/min, add β-adrenergic blockers. For patients with difficulty controlling thyrotoxicosis and taking ATD treatment and patients with second-degree thyroid enlargement and above, potassium iodide oral solution is added (Start taking Luger’s solution 14 days before the procedure: 3 drops/time, three times/day, add one drop daily; continue to take it for seven days after the procedure, reduce one drop daily to 9 drops/time and stop the drug) [16, 17, 21].

To ensure that patients are fully informed and able to choose the treatment plan independently, the procedure consent form must be signed by the patient prior to the procedure. Patients were advised to abstain from food and drink for at least six hours prior to the procedure. Prior to the procedure, a routine ultrasound measurement of thyroid size and Color Doppler Flow Imaging (CDFI) to show thyroid blood perfusion is performed, and the puncture route for TA is planned.

The patient is positioned supine with hyperextended neck. The surgical area is routinely disinfected and draped. Conventional ultrasound is used to determine the location, size, and vascular distribution of the thyroid. Layered infiltration anesthesia under ultrasound guidance using 0.5% lidocaine is performed to separate the thyroid from the pre-thyroid muscles, forming an isolation zone to block heat conduction and prevent damage to surrounding structures. The space behind the thyroid capsule is usually not isolated to avoid tearing and bleeding caused by forced separation due to adhesion of the thyroid to the surrounding large blood vessels. A surgical incision is made in the midline of the neck at the projection of the thyroid isthmus. Under ultrasound guidance, the tip of the puncture needle is inserted into the thyroid, penetrating one side of the gland through the isthmus. The path is from the anterior and inner side next to the trachea, about 0.3 to 0.5 cm, puncturing to the deep part of the thyroid. The needle tip approaches the dorsal capsule of the gland but does not penetrate the thyroid capsule. The gland is ablated from inside to outside in a stepwise manner. During puncture, the needle tip is always kept within the ultrasound-visible range. The preset power range is 30 to 40 W. The thyroid is considered as multiple sections, starting from the middle part of the thyroid and ablating layer by layer upwards and downwards. Each layer uses an ablating sequence from inside to outside, deep to shallow. During ablation, a normal gland thickness of about 0.3 to 0.5 cm near the trachea should be preserved to avoid ablating the danger triangle area and protect the recurrent laryngeal nerve. The Moving-shot ablation technique is used, which can achieve near-total ablation of the hyperthyroid gland. The Moving-shot technique is used to ablate the glands sequentially from the inside to the outside until a gas response with hyperechoic gas is observed [22, 23]. After satisfactory ablation on one side, communication with the patient is made to confirm that there is no change in voice, then the same anesthesia and ablation treatment is applied to the other side of the thyroid. For patients with thick isthmus tissue, ablation of the thyroid isthmus is performed to complete near-total ablation treatment of both lobes and the isthmus of the thyroid in one go.

The ablation range is equivalent to the range of near-total thyroidectomy, only preserving a thin layer of glandular tissue on the inner posterior sides of the trachea, namely the danger triangle area [24], with a thickness of about 0.3 cm, measuring approximately 1.0 cm × 1.5 cm, about 3 g of thyroid tissue. This ensures the safety of important organs such as neck nerves, blood vessels, trachea, and parathyroid glands while achieving near-total thyroid ablation.

Following the TA procedure, the patient’s neck is protected from burns by applying compression ice while pressure is applied to stop bleeding and alleviate discomfort. On the first day after the procedure, the patient is closely monitored for complications such as fever, bleeding, dyspnea, cough, hoarseness, and thyroid storm. Additionally, thyroid function is tested during this time. After successful observation, the patient is discharged on the second day after the procedure.

To maintain thyroid function and manage any residual thyrotoxicosis, patients are treated with ATD, β-adrenergic blockers, and potassium iodide oral solution after the procedure [25]. It is important to note that the potassium iodide solution should be discontinued within ten days after the procedure [26]. Treatment with these drugs is typically maintained for approximately three months, after which they are gradually discontinued once symptoms of thyrotoxicosis have resolved and thyroid function has returned to normal levels. Three months after the TA, patients continued exhibiting elevated thyroid hormone levels, paired with suppressed thyroid-stimulating hormone levels. Even upon cessation of ATDs, these individuals maintained high levels of thyroid hormones and diminished TSH levels. The tapering of ATD dosage presented a challenge in preserving normal thyroid hormone levels. The above patients necessitate a secondary TA of the residual thyroid tissue.

In patients co-diagnosed with GD and active GO, prophylactic glucocorticoids are prescribed to curtail the advancement of GO post-TA and to address any ophthalmic complications that may arise postoperatively.

Enumeration data were presented as frequencies and rates, while measurement data were expressed as medians ± one quartile. The Fisher exact test was employed to evaluate dichotomous variables between populations. The significance of the difference in the means of continuous variables between patients who underwent RFA and MWA was assessed via a two-tailed unpaired Wilcoxon test. A two-tailed paired Wilcoxon test was performed to compare the measurements before and after the procedure. The ggplot package was utilized to generate visualizations of thyroid volume changes. A p-value less than 0.05 was considered statistically significant. Statistical analysis was conducted using R statistics software (version 4.2.1).