Ethanol and thermal ablation for malignant thyroid tumours

Abstract

Papillary thyroid carcinoma (PTC) is the most common subtype of thyroid malignancy and has a good prognosis and low mortality rate. Surgery is the standard treatment for patients with primary and recurrent thyroid cancer. Although patients with PTC usually exhibit excellent treatment outcome, the incidence of recurrence in the neck ranges from 20% to 59%. When a patient with thyroid cancer is contraindicated for surgery, ultrasound (US)-guided ablation techniques, including ethanol ablation (EA) and thermal ablations, are suggested alternatives. In this review, we evaluated the indications, devices, techniques, clinical outcomes, and complications associated with US-guided EA and thermal ablations based on available scientific evidence and expert opinions regarding the use of ablation for primary and recurrent thyroid cancers.

Keywords:

• Thyroid
• thyroid cancer
• ethanol ablation
• radiofrequency ablation
• ultrasound

Introduction

Papillary thyroid carcinoma (PTC) is the most common subtype of thyroid cancer and has a good prognosis and low mortality rate [1]. Surgery is the standard treatment for patients with primary and recurrent thyroid cancer, followed by radioactive iodine therapy and/or thyroid hormone therapy. Although patients with PTC usually exhibit excellent treatment outcomes, the incidence of recurrence in the neck ranges from 20% to 59%, depending on the risk factors for recurrence. However, when surgery is not suitable for recurrent thyroid cancers, the current guidelines suggest the use of radioactive iodine therapy, external beam radiation therapy, chemotherapy, or image-guided ablation [1]. In primary thyroid cancer, particularly in cases with low-risk papillary microcarcinoma (PMC, size <1 cm), active surveillance has been suggested as first-line management, rather than immediate surgery [2,3].

Although surgery is the standard treatment, the risk of complications associated with surgery is high due to the distortion of the neck anatomy as a result of scar tissue formation, particularly in patients with repeated neck dissections. Meanwhile, active surveillance for patients with low-risk PTC is also associated with certain problems. In these patients, ultrasound (US)-guided ablation techniques, including ethanol ablation (EA) [4–8], radiofrequency ablation (RFA) [9–13], laser ablation (LA) [14–16], and microwave ablation (MWA) [17–19], are suggested as alternatives.

The purpose of this review is to evaluate the possible indications, devices, techniques, clinical outcomes, and complications associated with US-guided EA and thermal ablations based on available scientific evidence and expert opinions regarding the use of ablation for primary and recurrent thyroid cancers.

Materials and methods

Indications

Recurrent thyroid cancer in the neck lymph nodes

The Korean Society of Thyroid Radiology [20] suggests that RFA can be used in patients with high surgical risk and in those who refuse to undergo repeated surgery. A recent Italian report proposed similar indications for recurrent thyroid cancers to those proposed by the Korean Society [21], that is, the presence of recurrent thyroid cancer (operation bed and lymph nodes) and high surgical risk. Before performing RFA, cancer recurrence should be confirmed via US-guided fine-needle aspiration cytology and/or measurement of the washout thyroglobulin concentration [20,22].

The treatment strategies of US-guided ablations for recurrent thyroid cancer are yet to be clearly established. However, two treatment strategies have been commonly suggested thus far, including complete treatment and conservative treatment. Complete treatment strategy is the treatment of any recurrent cancers visible on US and/or computed tomography (CT) [23]. To achieve complete ablation of recurrent cancers, several studies have indicated that treatment should be restricted to patients with ≤3 recurrent thyroid cancers in the neck and without metastatic cancers beyond the neck during treatment [4,9,13,24]. Meanwhile, conservative treatment has been used to treat large cancers that can lead to cosmetic and/or symptomatic problems, such as discomfort, pain, dysphagia, hoarseness, and dyspnoea, given its involvement in critical structures in the neck [23,25].

Primary thyroid cancers

The Korean Society does not recommend the use of RFA for primary thyroid cancers or follicular neoplasms as no evidence of a treatment benefit is available regarding RFA in primary thyroid cancers [20]. However, several recent studies have suggested the application of RFA for low-risk PMC [26–28]. Kim et al. [12] reported their indications as follows: (a) pathologically confirmed PTC without cytological aggressiveness, (b) single PTC without extrathyroidal extension, (c) no metastatic tumours at the time of treatment, and (d) ineligibility for surgery. Recently, papillary thyroid carcinoma variants detected via US have been reported [29]. Among them, aggressive variants should be carefully evaluated before ablation. Furthermore, the operator should carefully evaluate thyroid capsule invasion or lymph node metastasis [30,31].

Devices

Two types of electrodes are used for thyroid RFA, namely, straight internally cooled and multi-tined expandable electrodes [32–38]. A modified straight-type internally cooled electrode has recently been developed specifically for thyroid RFA in Korea [39]. The shaft length of this electrode is short (7 cm) and thin (18–19 G), enabling easy control (as the thyroid gland is a superficial organ) and minimising injury to the normal thyroid [35]. A thin electrode, particularly 19-G, can easily penetrate small metastatic cancers [9,13]. Technicians can also use active tips of various sizes (0.38, 0.5, 0.7, or 1 cm). The electrode tip size is chosen based on the tumour size and the status of the surrounding critical structures. An electrode with a small active tip (i.e. 0.38 or 0.5 cm) is effective and safe for treating small primary and/or recurrent tumours or those that are close to critical structures [40]. Recently, bipolar electrodes have been introduced for patients with heart problems or are pregnant [41–43].

In our institution, we only have experience with EA and RFA; therefore, we briefly reviewed articles regarding LA. In LA, one or two 21-G spinal needles were inserted into the target lymph node, and a 300-lm quartz optical fibre. Real-time US or fusion imaging was used to target the lymph nodes [14–16,44].

Techniques

The Korean Society recommends the use of local anaesthesia rather than general anaesthesia or sedation [20]. Accordingly, patients are placed in a supine position with their neck extended. Many Korean institutions use 1–2% lidocaine without premedication [9,13,24,25]. Local anaesthesia using lidocaine administered to the puncture site and soft tissue around the tumour is a simple and effective pain control method. Meanwhile, Monchik et al. [45] reported the use of intravenous drugs, including a combination of fentanyl citrate (100–400 mcg) and midazolam (1–4 mg), as a possible anaesthetic alternative.

To prevent haemorrhage, the neck vessels along the surgical site should be carefully evaluated [20]. The nerves in the peritumoral area should also be carefully evaluated [46,47]. If a metastatic tumour exists adjacent to these nerves, the hydrodissection technique is useful to prevent thermal injury; in such cases, a 5% dextrose solution is carefully injected between the nerve and the tumour [9,13,24,48]. Furthermore, for primary thyroid cancers, we recommend the “trans-isthmic approach.” Meanwhile, the moving-shot technique has also been proposed as a standard method for the treatment of primary and recurrent thyroid cancers [13,20,28,49]. In this method, thermal damage to the surrounding structures can be minimised. Before starting ablation, the target tumour is divided into multiple conceptual ablation units, and RFA is performed in a unit-by-unit manner by moving the electrode tip [20,23,50]. These conceptual units are smaller at the margin of the tumour and/or the portions adjacent to critical structures (e.g. nerves, trachea, and oesophagus) and are larger in the central, safe portion of the tumour. However, for small tumours, the electrode should be fixed to the centre of the tumour and should not be moved during the procedure. Initially, the electrode tip is positioned in the deepest and most remote conceptual unit of the cancer, which enables easy monitoring of the electrode tip without any disturbance due to microbubbles.

Ablation is initiated with 5–10 W of power for an active tip of 0.38 cm, 10–15 W for an active tip of 0.5 cm, 20–30 W for an active tip of 0.7 cm, and 40–50 W for an active tip of 1 cm. If patients cannot tolerate the pain during ablation, lidocaine injection around the tumour is recommended. Ablation should be terminated when all conceptual units have changed to transient hyperechoic zones. RFA is usually performed on both the tumour and the surrounding normal tissue in order to prevent local recurrence [9]. Dupuy et al. [10] also reported the use of internally cooled electrodes, although they used a different technique. With regard to their technique, internal cooling was only used for masses >1 cm across the greatest dimension; otherwise, the electrode tip temperature was maintained at 90 °C for 2 min without internal cooling. If the temperature within the mass is <50 °C, then additional RFA treatments should be performed at another electrode position. Recurrent cancer with cystic portions can also be effectively treated via RFA [45]. After ablation, the technicians should check for any discomfort or complications, and the patient should be observed for 1–2 h at the hospital.

Meanwhile, in EA, 22–25 G needles that are 3–4 cm long, attached to a 1-ml syringe containing 95–99.9% ethanol, are used [5,7,8,23,45]. Local anaesthesia is necessary for pain control. During a single session, 0.1–1 ml ethanol is injected to the deep and peripheral portion of the recurrent thyroid tumour under US guidance. Immediately after the ethanol injection, the injected area becomes echogenic on US. After a short time (typically <1 min), the echogenic area will decrease, which allows for better visualisation of the needle. The needle is then repositioned at the untreated vascular area using colour Doppler US, and ethanol injections are continued until the entire lymph node is adequately treated. If ethanol leakage to the outside of the lesion is detected on US or if the patient complains of severe pain, then the injections are immediately terminated. When the entire tumour has changed completely to transient hyperechoic zone, the procedure is terminated [8].

During the follow-up period, the indications for repeated RFA or EA include the presence of power Doppler signals or enhancing portion on a CT scan, despite a reduction in the size [5,6,9,13]; volume reduction <50% [8]; or the presence of residual cancer tissue on core-needle biopsy or fine-needle aspiration cytology [5,8,10,51,52]. An additional treatment may be required for large cancers. Dupuy et al. [10] treated a 5-cm recurrent tumour over five sessions using RFA, and Baek et al. [9] treated a 2.8-cm tumour over two sessions also using RFA. Lewis et al. [7] also suggested that the undertreated portion can usually be detected using Doppler US in patients with large recurrent tumours that are located deeply.

Meanwhile, in LA, the introducer needle is receded at a distance of 5 mm, leaving the tip of the fibre in direct contact with the lesion [14–16,44]. Laser illuminations are performed with an output power of 3–4 W [14–16,44]. The total amount of energy is decided on a case-to-case basis according to the technician’s discretion and gas formation on US during ablation [14,15]. At the end of the ablation, contrast enhanced US is performed [14,15,44]. In cases of residual enhancement, the fibre is reinserted, and an additional ablation is performed during the same treatment session [14,15].

Results

Recurrent thyroid cancers

Treatment efficacy is evaluated using several parameters: volume reduction [38], therapeutic success rate (volume reduction >50%) [53], complete disappearance of the treated cancer, serum thyroglobulin concentration, cancer perfusion, and changes in echogenicity [20].

The efficacy of RFA is summarised in Table 1. Following RFA, the mean reduction in tumour volume ranged from 50.9% to 8.4% [9,24,25]; complete disappearance is noted in 25–94% of cancers [9,10,24,45,54]; therapeutic success rates range from 75% to 97% [9,10,13,45]; symptom improvement is observed in 64% of patients [25]; and decreases in serum thyroglobulin concentration are noted in most patients [9,10,25,45]. However, no long-term follow-up data have been published thus far.

Table 1. Publications of radiofrequency ablation for recurrent thyroid cancers.

								Serum thyroglobulin concentration (ng/ml)
	Publication (year)	Patient no.	Nodule no.	Follow-up (month)	Largest diameter (cm)	Volume reduction (%)	Completely disappear (%)	Before	After	Patient numbers^b	Volume reduction >50% (%)	Recurrence
Dupuy et al. [10]	2001	8	11	10.3	2.4	From 24 to 18 mm^a	58.2	34.1	2.6	6/8 (75)	84.8	2/8
Monchik et al. [43]	2006	16	24	40.7	1.7		42	37	1.9		75	3/16
Baek et al. [9]	2011	10	12	23	1.4	93	50	8.1	2.2	7/10 (70)	91.6	1./12
Guenette et al. [11]	2013	14	21	61.3	1.5
Lee et al. [24]	2014	32	35	30	0.81	96.4	94	16.7	10.5			1/35
Wang et al. [52]	2014	8	20	9.4	1.0	77	25	7.7	3.4	6/8 (75)	92	0/20
Lim et al. [13]	2015	39	61	26	0.8	95.1	82	1.2	0.5	13/29 (44.8)	97	4/61
Kim et al. [12]	2015	27	36	36	0.98	98.4	86					3/26
Long et al. [58]	2015	12	12	4-6
Guang et al. [68]	2017	33	54	21	1.2	94.9	61.1	10.2	1.1		100	1

^a Volume reductions in these articles are not suggested, and the mean size of the lymph nodes is suggested.

^b Number of patient with percentage in decreased serum thyroglobulin.

The efficacy of EA is summarised in Table 2. The reduction in the mean cancer volume following EA reportedly ranges from 37.5% to 96%[7,8]; complete disappearance is noted in 31–65% of treated cancers [5–8]; therapeutic success rates range from 70.8% to 98% [6,7]; and reductions in serum thyroglobulin concentration are noted in some patients [5,8,45]. Due to the excellent ablation of local cancer, Heilo et al. [5] suggested that US-guided EA can replace “berry picking” surgery.

Table 2. Efficacy of ethanol ablation for recurrent thyroid cancers.

								Serum thyroglobulin concentration (ng/ml)
	Publication (year)	Patient no.	Nodule no.	Follow-up (month)	Mean size (mm or mm^3)	Volume reduction >50% (%)	Completely disappearance (%)	Before	After	Ethanol amount (cc)	Recurrence
Monchik et al. [43]	2006	6	7	19				6.1	2.0 (3/6)^a
Lim et al. [8]	2007	16	24	24	9.9 mm		42	2.53	0.2	1.1/LN
Lewis et al. [7]	2002	14	29	18	492 mm^3	70.8	31			0.34/LN
Kim et al. [6]	2008	27	47	28.2	11.6mm, 678.8 mm^3	98	44.7			2.4/LN, 1.1/session
Heilo et al. [5]	2011	63	109	38.4						0.1-1/LN	2
Hay et al. [4]	2013	25	37	65	14mm, 223 mm^3		46		≤0.2^b (55%)	0.5	2

^a In this article, serum thyroglobulin level was followed for three of the six patients.

^b Serum thyroglobulin levels ≤0.2 ng/ml were found in 12 of the 22 (55%).

A recent meta-analysis that included 10 studies with a total sample size of 270 patients (415 thyroid nodules) compared the efficacy and safety of RFA and EA [55]. The study indicated that both RFA and EA are acceptable management methods for recurrent thyroid cancers; however, the mean number of RFA sessions was <1.3 in 83.3% of the studies (5/6 RFA studies), whereas that of EA sessions was >2 in 75% of the studies (3/4 EA studies) [55]. Lewis et al. [7] also suggested that RFA has an advantage over EA in terms of the number of treatment sessions required. To evaluate treatment efficacy, US and CT serve as key imaging modalities [20,28]. During the follow-up period, well-treated cancers demonstrate increased echogenicity surrounded by echogenic soft tissue (normal ablated tissue), loss of colour Doppler signals or contrast enhancement, and decreased cancer size on US [9,13]. Contrast-enhanced US has better detection capability for the undertreated cancer portion [16,44]. Moreover, incomplete treatment reportedly occurs due to intractable pain, major vessel encasement, severe calcification [25], and tumour located adjacent to the vagus nerve [9]. Local cancer recurrence reportedly occurs at a rate of 0–25% following RFA [9,10,45] and 3.2–33% following EA [5,7,8,45]. In a 5-year follow-up study of EA, Kim et al. [56] treated 41 cases of recurrent lesions (19 in the central neck and 22 in the lateral neck) in 29 patients. EA resulted in increased lymph node size in seven cases (17.1%), no changes in 10 (24.4%), and decreased size in 24 (58.5%). The incidence of increased lymph nodes was high in elderly patients (65.3 ± 14.4 vs. 48.2 ± 16.3 years; p = 0.02) and in those with small lymph node involvement (9.3 ± 1.0 vs. 12.3 ± 6.4 mm; p = 0.012) [56].

Table 3 shows the patient demographic data and efficacy of LA, which was evaluated using US, PET-CT, and CEUS. The mean reduction in the tumour volume ranged from 88% to 99%, and the mean number of treatment session was 1–1.1. Complete disappearance of lesions was noted in 84% of cases [14–16,44]. When we compared the efficacy of RFA and LA, the two treatment modalities were found to be comparable for treating recurrent thyroid cancers. The advantages of LA during the treatment of recurrent thyroid cancer include a small bore needle, precise control of thermal energy, and use of multiple fibres to achieve large ablation zone [14]. By contrast, RFA has the advantages of using a moving-shot technique and an internally cooled electrode [57].

Table 3. Demographic date, efficacy, and complications following LA.

	Publication (year)	Number of tumours	Mean size (cm)	Volume reduction (%)	Treatment session	Disappearance (%)	PET-CT/CEUS(−) (%)	Local tumour progression	Voice change	Skin burn
Panini et al. [16]	2013	8	0.64 ml	88	1	–	63		1
Mauri et al. [15]	2013	24	1.1		1	–	80	8.3	1
Mauri et al. [14]	2016	46	1.0		1.1		79.1	17.4	1	1
Zhou et al. [43]	2016	21	0.75	99	1.1	84	89	11	1

Only one study reported on the use of MWA to treat 17 patients (23 recurrent thyroid tumours), with a mean treatment session of 2.3 (range 1–4) [18]. The mean reduction in tumour volume was 91% at 18 months follow-up. A total of 30.4% of the nodules had completely disappeared, whereas 52.2% remained as small scar-like lesions [18].

Primary thyroid cancers

The efficacy in treating primary thyroid cancers is evaluated via volume reduction [26,28]. No local tumour recurrence or metastasis to lymph nodes in the neck has been reported during short-term follow-up after RFA [26], LA [58], and MWA [19]. Moreover, a Korean multicenter study [28] reported no major complications and no tumour recurrence in their 4-year follow-up results (Table 4).

Table 4. Complications and demographic data of radiofrequency ablation, laser ablation, microwave ablation for primary thyroid cancers.

	Modality	Patients	Follow up period (months)	Distant metastasis	Voice change	Other complications
Yue et al. [19]	Microwave	21	11	0	4^a	Temporary or local discomfort, 2 chocking and coughing^b
Zhang et al. [26]	Radiofrequency	92	7.8	0	4^c	1 Moderate pain
Kim et al. [28]	Radiofrequency	6	48	0	0	1 Mild headache 1 Mild neck pain
Zhou et al. [55]	Laser	30	13	0	0	Pain and regional discomfort

^a All of patients recovered within 3 months spontaneously.

^b Choking and coughing appeared in two patients, and disappeared in 24 h and 7 d, respectively, without any treatment.

^c All of patients recovered within 3 h.

Furthermore, Valcavi et al. [59] reported the results of three patients with low-risk PTC who underwent surgery after LA. LA was effective for local tumour control of the primary cancers; however, at surgery after LA, multiple minute tumours in the thyroid gland and microscopic metastases in the central neck lymph nodes were detected. This result indicates that thermal ablation is effective for the management of primary thyroid cancer but may have limited efficacy in the control of regional microscopic metastasis or minute multifocal cancers [60]. Comparison of the benefits and limitations of surgery, active surveillance, and image-guided ablations is needed in future studies.

Complications

The treatment of recurrent cancers using RFA and EA is associated with various complications, including pain, neck swelling, haematoma, burns, confusion, shoulder droop, and voice changes (Tables 5 and 6) [9–11,13,24,25,45,54,61]. In particular, voice change is among the most common and serious complication following ablation. A review of previous publications showed that the incidence of voice complications is higher with RFA (6.2%, 19/306) [9–11,13,24,45,62] than with EA (2.7%, 4/151) [4,6,8,45]. In three patients, the voice changes following EA were transient, and their voices recovered after 3 h to 7 d after treatment. However, in a report by Hay et al. [4], one patient exhibited voice recovery only after 9 months. By contrast, six patients with voice changes following RFA did not recover fully. RFA can produce a larger zone of destruction than EA; thus, RFA is more likely to cause permanent damage to the adjacent nerves.

Table 5. Complications following radiofrequency ablation of recurrent thyroid cancer.

	Publication (year)	Country	Number of patients	Voice change	Skin burn	Shoulder drop	Confusion	Hematoma	Swelling	Pain (self-limited)
Dupuy et al. [10]	2001	Ireland	8	1^a	1					+
Monchik et al. [43]	2006	Ireland	16	1^a	1					+
Baek et al. [9]	2011	Korea	10	1^a						+
Park et al. [25]	2011	Korea	11		1					+
Guenette et al. [11]	2013	Ireland	14	1^b						+
Lee et al. [24]	2014	Korea	32	6^a						+
Wang et al. [52]	2015	China	8							+
Lim et al. [13]	2015	Korea	39	3^a						+
Kim et al. [12]	2015	Korea	27	1^a						+
Long et al. [58]	2015	China	12			–			1	5
Kim et al. [59]	2016	Korea	129	5^a		3	1	1	1	2
Guang et al. [68]	2017	China	33							1
Total	–	–	339	19	3	3	1	1	2	–

^a Voice change during the treatment of central neck lesions.

^b Voice change during the treatment of lateral neck lesions.

Table 6. Complications following ethanol ablation of recurrent thyroid cancers.

	Publication (year)	Voice change (period)	Pain	Discomfort
Monchik et al. [43]	2006	1		100% Neck swelling and regional discomfort^a
Kim et al. [6]	2008	1 (1 week)	73% localised pain
Lim et al. [8]	2007	1 (5 d)	Self-limited pain
Lewis et al. [7]	2002	0
Heilo et al. [5]	2011			10%
Hay et al. [4]	2013	1		100%

^a Self-limited neck swelling and regional discomfort were reported in all 16 patients and resolved within 1–2 weeks.

All LA studies reported voice changes after ablation, but all patients recovered spontaneously within 1–3 months [14–16,44]. One case of first-degree burn with complete cosmetic recovery in 1 month was reported [14]. One patient experienced transient dysphonia immediately after MWA [18].

Voice changes commonly occur due to recurrent injury to the laryngeal nerve [62], although voice changes due to vagus nerve injury during treatment of the lateral neck lymph node was suspected in one case [11]. Within the lateral neck, the vagus nerve is a critical structure related to voice production. The vagus nerve can be easily detected on US and is located within the carotid sheath, usually posterolateral to the common carotid artery and posteromedial to the internal jugular vein [46,63]. To avoid thermal injury of the vagus nerve, technicians should be aware of its location and variations of appearance on US. This complication can be overcome by using the hydrodissection technique [13,62] and through verbal communication during the procedures. However, in the hydrodissection technique, continuous fluid infusion is necessary, because the injected fluid spreads gradually along the muscle plane of the neck. The use of a unidirectional ablation electrode is another strategy for preventing thermal injury [64]. This electrode was used to modify the direction of the ablation and produce asymmetrical half-moon-shaped ablation zones in an ex vivo study [64]; however, it has not been reported in any human studies thus far.

The most common symptoms of complications are discomfort and pain in the neck; however, they are usually self-limited [62]. Pain sometimes radiates to the teeth, jaw, head, and chest but is usually resolved after the procedure. Peritumoral lidocaine injection is the most effective method to relieve pain [7,8,62,65]; however, the injected lidocaine can also cause an anaesthetic effect (although transient) in the nerve. During RFA, heat propagation to the surrounding tissue is a main cause of pain; meanwhile, during EA, pain is most likely related to the leakage of the injected ethanol into the surrounding normal tissue. Park et al. [25] reported termination of RFA in five patients due to intolerable pain; however, pain does not normally affect RFA or EA procedures in most patients. Hematoma or neck swelling due to vessel injury has been reported [61,62] and may be associated with pain. Manual compression is recommended for the management of these complications.

Burns at the sites of electrode puncture and pad attachment have been observed following RFA to treat liver cancer [66,67]. Burns have also been reported at the electrode insertion site during the use of RFA to treat recurrent thyroid cancers [10,25,45]. However, burns at the site of pad attachment in RFA for the treatment of recurrent thyroid cancers have not been reported because the energy used in such cases is lower than that used for the treatment of liver cancer [50]. Nevertheless, frequent ice bag application [50] or the injection of cold 5% dextrose solution or saline between the tumour and the skin [68,69] during the procedure may prevent burns.

Kim et al. [62] recently described three patients with injury of the spinal accessory nerve during RFA for recurrent cancers. The spinal accessory nerve is a superficially located motor nerve that supplies the sternocleidomastoid and trapezius muscles. Therefore, injuries to this nerve can lead to limited abduction of the shoulder joints and drooping of the affected shoulders [46,70]. The spinal accessory nerve was identified in 98–100% of segments, and continuous US monitoring of this nerve was possible in most patients, although the mean diameter was only 0.54 mm [70]. In the three patients who developed injury of the spinal accessory nerve in the study by Kim et al. [62], the recurrent lymph nodes were all located at level III or V.

Complications are relatively rare and transient during the treatment of primary thyroid cancers, because they are usually small in size. The incidence of voice changes after RFA and MWA is reportedly 5.4% (8/149) for primary thyroid cancers [19,26], and no other life-threatening complications have been reported.

Fatal complications, such as injury to the oesophagus, trachea, or other nerves (i.e. phrenic nerve and brachial plexus), following the treatment of primary or recurrent thyroid cancers, which are often located adjacent to these structures, have not been reported. To prevent these complications, technicians should be aware of the neck anatomy (determined via US imaging) and always trace the electrode tip during the procedure.

Discussion

Both RFA and EA are possible alternatives for the treatment of recurrent thyroid cancers in patients at high risk of surgical complications or those who refuse repeated surgeries. RFA is superior to EA because of its better efficacy, fewer mean number of treatment sessions required, and wider extent of the ablation zone. However, RFA demonstrates a higher tendency and severity of voice complications than EA in the treatment of central neck lesions. To minimise these complications, technicians should be aware of the various possible complications and the preventive techniques, such as hydrodissection.

In primary thyroid cancers, RFA achieved an excellent effect on local tumour control; however, the use of the ablation method to control regional microscopic metastasis or tiny multifocal cancers is limited.

Conclusions

Both thermal ablations and EA are possible alternatives for the treatment of recurrent thyroid cancers; however, reports on the use of thermal ablations in patients with primary thyroid cancers are limited. These procedures may be applied in patients with high surgical risk or those who refuse repeated surgeries.

Disclosure statement

The authors report no conflicts of interest, but Dr. Jung Hwan Baek is a consultant of two RFA companies, STARmed and RF Medical.