Efficacy and Safety of Thermal Ablation for Indeterminate Thyroid Nodules: A Systematic Review of the Literature and Meta-Analysis

Abstract

Background:

The management of indeterminate thyroid nodules (ITNs), classified as Bethesda III and IV, is challenging due to biopsy limitations in distinguishing benign from malignant nodules. While diagnostic lobectomy is the standard, thermal ablation (TA) is increasingly considered for patients ineligible or unwilling to undergo surgery. This systematic review and meta-analysis therefore evaluate the efficacy and safety of TA for ITNs.

Methods:

A comprehensive search of MEDLINE, EMBASE, and COCHRANE databases was conducted through May 11, 2025, for studies on ITNs treated with TA, with ≥12 months of follow-up and reported clinical or safety outcomes. Case reports, abstracts, and reviews were excluded. Two radiologists independently performed data extraction and quality assessment. Outcomes included volume reduction rate (VRR), regrowth, delayed surgeries, malignancy detection, and complications. The Risk of Bias for Nonrandomized Studies (RoBANS) tool was used for quality assessment. A random-effects model synthesized pooled estimates, with heterogeneity quantified by Higgins’ I ².

Results:

A total of 15 studies with 1149 nodules were analyzed, showing progressive VRR increase, plateauing at 48 months. The pooled 12-month VRR was 81.0% (confidence interval: 76.0–85.9%). Hydrodissection significantly improved VRR at 6 months (p = 0.03), while larger nodules were more prone to regrowth. Major complications occurred in 1.8% (21/1149), with no reported metastasis. Regrowth and delayed surgery occurred in 2.3% (26/1149) and 0.3% (4/1149), respectively, with three malignancies upon delayed surgery.

Conclusions:

TA may be considered a minimally invasive alternative for ITNs who are not candidates for or decline surgery, demonstrating favorable efficacy and safety. However, study limitations, short follow-up, and residual malignancy risk necessitate careful follow-up, particularly for larger nodules. Advanced TA techniques such as hydrodissection may enhance outcomes by increasing the likelihood of complete ablation. Long-term prospective studies and randomized trials are needed to confirm TA’s role in clinical practice.

Introduction

Thyroid nodules are increasingly common in clinical practice, with reported incidence rates in adults reaching up to 67%.¹ These nodules can be classified as benign, malignant, or indeterminate, with indeterminate cases accounting for 19.7%.² The pathological findings for these indeterminate nodules range from benign to malignant and include low-risk neoplasms.³ The Bethesda system estimates the risk of malignancy for indeterminate nodules as 13–30% for Bethesda III and 23–34% for Bethesda IV.⁴

The management of indeterminate thyroid nodules (ITNs) remains a challenge because biopsy methods such as fine-needle aspiration (FNA) and core needle biopsy (CNB) cannot reliably differentiate between benign and malignant nodules. FNA has low sensitivity, whereas CNB offers greater diagnostic accuracy but still has a high rate of false-positive results.^5

–8 This limitation leads to the recommendation for diagnostic lobectomy. Surgery provides a definite diagnosis because it allows the detection of vascular, capsular, and extrathyroidal tissue invasion, but comes with potential complications such as vocal cord palsy, hypothyroidism, and hypocalcemia.⁹ Furthermore, the majority of surgically removed nodules turn out to be benign.¹⁰

Thermal ablation (TA), such as radiofrequency ablation (RFA) and microwave ablation (MWA), is increasingly being used for minimally invasive treatment of thyroid lesions.^{11

–18} Current international guidelines recommend TA for benign nodules, recurrent thyroid cancers, and papillary thyroid microcarcinomas (PTMCs).^18

–23 However, the guidelines do not recommend TA for ITNs because of insufficient evidence.^24,25 Although diagnostic hemithyroidectomy is the standard management for ITNs, recent studies suggest that TA may be an option for patients who cannot undergo or refuse surgery. To further investigate this possibility, we conducted a systematic review and meta-analysis to evaluate the feasibility and safety of TA for ITNs.

Materials and Methods

This systematic review and meta-analysis was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).²⁶

Literature search

A search was conducted of the MEDLINE, EMBASE, and COCHRANE databases to identify original literature on patients with ITNs treated with TA. The following search terms were used: ((Indeterminate OR Bethesda III OR Bethesda IV OR Atypia of undetermined significance OR AUS OR Follicular neoplasm) AND (Thyroid) AND (Thermal ablation OR Radiofrequency ablation OR Radio-frequency ablation OR RFA OR Microwave ablation OR Microwave coagulation OR MWA)). The literature search had no set start date and was updated on May 11, 2025. Only English-language publications were included. In addition, the bibliographies of relevant articles were reviewed to find any further appropriate articles.

Inclusion criteria

Studies satisfying the following criteria were included: (1) patients with ITN(s) (Bethesda III, Bethesda IV, AUS, or follicular neoplasm) on FNA or CNB treated with any kind of TA, such as RFA or MWA; and (2) follow-up data for at least 12 months after ablation.

Exclusion criteria

Studies or subsets of studies were excluded on the basis of the following criteria: (1) letters, editorials, conference abstracts, case reports or series, systematic reviews or meta-analyses, consensus statements, guidelines, and review articles; (2) articles not directly focused on the topic of ITNs treated with TA; and (3) articles with a suspected overlap of populations.

Two radiologists, H.L. and J.H.B., with 7 and 30 years of experience in thyroid imaging, respectively, independently conducted the literature search and selection process, which included two stages: First screening (abstract review) and second screening (full-text review). In the first stage, each reviewer independently evaluated abstracts based on predefined criteria. In the second stage, full-text articles were reviewed, and the results were compared. Any discrepancies were resolved through discussion between the reviewers.

Data extraction

Data extraction was performed using standardized forms developed specifically for this study, and included the following: (1) characteristics of the articles, including author, year of publication, country of origin, study design (prospective or retrospective), duration of patient recruitment, ultrasonography (US) follow-up protocol, mean follow-up, follow-up loss, ablation method, number of patients, number of nodules, mean patient age, male-to-female ratio, diagnosis method, number of nodules categorized by Bethesda categories, detailed inclusion criteria, and baseline nodule volume; (2) details of the TA, including ablation methods, generator, electrode, applied power, mean ablation time, number of ablation sessions, type of anesthesia, and use of sedation; (3) results of ablation and follow-up details, including serial changes in the posttreatment volume reduction rate (VRR) of the index nodule (1-, 3-, 6-, 12-, 24-, 36-, 48-, and 60-month data), with VRR during follow-up calculated as ([initial nodular volume − final nodular volume] × 100)/initial nodular volume)²⁷; and (4) clinical and safety outcomes, including the number of biopsies performed post-TA, reported number of regrowths, delayed surgery, malignancy at delayed surgery, metastasis, and complications related to the ablation procedure.

Complications were assessed according to the reporting standards of the Society of Interventional Radiology.²⁸ Major complications were defined as adverse events that resulted in significant morbidity or disability, required intensive medical care or hospital admission, or significantly extended hospital stays. Any other complications, including unintended or undesired outcomes of the procedure that did not require therapy or medical treatment, were classified as minor complications, as described in previous studies.^29,30 Voice changes were classified as a major complication because in all studies the duration was reported as “within 6 months.” Immediate postablation events, such as parenchymal edema and tolerable mild pain or discomfort that did not require medication, were not considered to be complications.

Quality assessment

Two authors (H.L. and J.H.B.) independently conducted data extraction and quality assessment using the Risk of Bias for Nonrandomized Studies (RoBANS) tool for nonrandomized controlled trials.^31,32

Data synthesis and analyses

The primary outcome of the current systematic review and meta-analysis was the serial VRRs of ablated nodules. The secondary outcomes focused on detailing the clinical and safety outcomes, including regrowth, delayed surgery, pathologically confirmed malignancies after ablation, and complications during postablation follow-up. We estimated pooled proportions using an inverse-variance weighted approach,^33
–35 applying a random-effects model to calculate the overall proportions. The degree of heterogeneity was quantified using Higgins inconsistency index (I ²), where values above 50% were considered indicative of substantial heterogeneity.³⁶ Given the substantial heterogeneity observed, we conducted sensitivity analyses to identify its potential sources. If significant heterogeneity persisted after excluding outlier studies, we applied a random-effects model to account for the variability. Post hoc subgroup analysis was also performed according to the baseline nodule volume, the presence or absence of hydrodissection, the ablation method (RFA vs. MWA), and genetic alteration testing. Subgroup analysis based on baseline nodule volume was performed to assess its impact on procedural efficacy, as larger nodules are less likely to achieve complete ablation and effective volume reduction compared with smaller nodules. Subgroup analysis by hydrodissection was conducted to evaluate its potential to enhance safety and minimize complications by creating a protective margin between critical structures and the target lesion, thereby reducing unablated margins and improving ablation completeness. Subgroup analysis by ablation method was included to compare the clinical outcomes of RFA and MWA, as the two modalities differ in mechanisms of energy delivery, heat distribution, and tissue interactions, which may influence treatment efficacy and complication rates. Subgroup analysis based on molecular testing was conducted, considering its potential impact on outcomes such as regrowth and the risk of malignancy. Metaregression was used for these subgroup analyses. All statistical analyses were performed by one author (S.J.C., with 9 years of experience in conducting systematic reviews and meta-analysis), using the “meta” package in R version 4.4.1 (http://www.r-project.org/).

Results

Literature search

The article selection process is described in Figure 1. The initial search for the systematic literature review identified 235 articles (Supplementary Data). After removing 42 duplicates, 193 articles remained for screening on the basis of their titles and abstracts, resulting in 29 potentially eligible articles. Upon reviewing the full text of the 29 provisionally eligible articles (Supplementary Table S1), it was found that one was a case report,³⁷ six were conference abstracts,^38

–43 three were review articles,^44
–46 one potentially involved overlapping data,⁴⁷ and three were not relevant to the field of interest.^48
–50 Ultimately, 15 articles were included in this study.

FIG. 1.

Flow diagram of the study selection process.

Characteristics of the included studies

Table 1 details the characteristics of the 15 included studies.^{51

–65} A total of 1015 patients were enrolled, with 1149 nodules undergoing TA. The study populations ranged in size from 6 to 375 patients, with an average patient age between 38 and 63 years. The overall male-to-female ratio was 651–364. The mean baseline nodule volume varied between 0.6 and 17.9 mL. The US follow-up protocol varied from simple checkups at 6 and 12 months to comprehensive follow-up spanning 1, 12, 24, 36, 48, and up to 60 months after TA. The studies predominately originated from China, the United States, and Italy, with additional contributions from Taiwan and Korea. Among the 15 studies, 11 were retrospective in design^{51,52,55,57

–60,62

–65} and four were prospective.^53,54,56,61 All studies confirmed the Bethesda III or IV classification of nodules through pathology, using various biopsy methods, including two FNAs, two CNBs, or at least one FNA or CNB. The total number of nodules reported was 745 for Bethesda Category III and 345 for Bethesda Category IV. In addition, 59 nodules were classified as either Bethesda Category III or IV but were not individually distinguished.

Table 1.

Characteristics of the Included Studies

Author (year)	Country	Study design	Enrollment period	US follow-up protocol (months)	Mean follow-up (months)	Ablation method	Patient No.	Nodule No.	Follow-up loss (%)	Mean age (years)	Male: Female	Bethesda categories		Diagnosis method	TI-RADS	Detailed inclusion criteria	Baseline nodulevolume(mL)
Author (year)	Country	Study design	Enrollment period	US follow-up protocol (months)	Mean follow-up (months)	Ablation method	Patient No.	Nodule No.	Follow-up loss (%)	Mean age (years)	Male: Female	III	IV	Diagnosis method	TI-RADS	Detailed inclusion criteria	Baseline nodulevolume(mL)
Chiang PL et al. (2023)⁵¹	Taiwan	Retro.	Sep. 2019 to Sep. 2021	1, 3, 6, and 12	NA	RFA	28	30	0	47.6	5:23	30	0	FNA or CNB	ACR TI-RADS; 1 (1), 2 (12), 3 (8), 4 (9)^a		7.9 ± 13.4
Dobrinja C et al. (2015)⁵²	Italy	Retro.	NA	1, 3, 6, 12, and 24	NA	RFA	6	6	0	61.2	3:3	6^b	FNA	NA	Molecular test (+)^c	17.5 ± 5.6
Dong Y et al. (2024)⁵³	China	Pro.	Jan. 2017 to Dec. 2020	1, 3, 6, 12, and last f/u	40.0 ± 11.0	RFA	62	62	0	38.0	10:52	0	62	FNA	C-TIRADS; 4 (62)^a	≤ 2 cm,C-TIRADS 4A	1.0 ± 0.7
Fugazzola L et al. (2024)⁵⁴	Italy	Pro.	Oct. 2020 to Oct. 2022	1, 6, 12, and 24	NA	RFA	33	33	0	52.7	9:24	33	0	FNA	EU-TIRADS; 3 (27), 4 (6)^a	Solid portion >70%,EU-TIRADS 2–4, molecular test(+)^d	17.3 ± 10.7
Ha SM et al. (2017)⁵⁵	Korea	Retro.	Jan. 2009 to Dec. 2011	1, 12, 24, 36, 48, and 60	66.4 ± 5.1	RFA	10	10	0	41.8	0:10	0	10	CNB	NA	1–2 cm	0.6 ± 0.4
Issa PP et al. (2022)⁵⁶	United States	Pro.	2018 to 2020	1, 3, 6, and 12	NA	RFA	53	53	0	63.0	23:30	53		FNA	NA	Molecular test (+)^e	1.9 ± 0.5^f
Jackson K et al. (2025)⁵⁷	United States	Retro.	Nov. 2021 to Aug. 2024	1, 6, 12, 18, and 24	8.8 ± 7.9	RFA	32	32	97 (by 24M)	59.8	27:5	32	0	FNA	NA	Molecular test (+)^g	15.8 ± 14.4
Li X et al. (2021)⁵⁸	China	Retro.	Jan. 2017 to Dec. 2019	1, 3, 6, and biannual f/u	16.4 ± 5.2	RFA, MWA	30	30	73 (by 24M)	47.3	4:26	0	30	FNA	NA		6.6 ± 9.1
Lin WC et al. (2021)⁵⁹	Taiwan	Retro.	Jan. 2018 to Jul. 2019	6 and 12	10.1^h	RFA	22	22	0	39.6	1:21	0	22	FNAor CNB	NA	> 1 cm, SUVmax <5	12.6 ± 20.9
Liu YT et al. (2025)⁶⁰	China	Retro.	Jan. 2014 to Oct. 2023	1, 3, 6, and biannual f/u	12.0 ± 8.9	RFA, MWA	161	≥161ⁱ	0	46.0	131:30	0	161	FNAor CNB	NA		2.4 ± 2.6^f
Ryu YJ et al. (2025)⁶¹	United States	Pro.	NA	1, 3, 6, and 12	6.9 ± 6.7	RFA	18	20	70 (by 12M)	45.0	16:2	20	0	FNA	NA	Molecular test (+)^g	17.9 ± 21.9^f
Tang Q et al. (2025)⁶²	China	Retro.	Jul. 2020 to Sep. 2023	1, 3, 6, 9, 12, and biannual f/u	6.83^h	MWA	95	101	84 (by 24M)	47.1	79:16	101	0	FNA	NA	Molecular test (+)^j	2.9 ± 4.7
Yang J et al. (2022)	China	Retro.	Jan. 2013 to Sep. 2021	1, 3, 6, and 12	10.1 ± 3.1	MWA	46	47	30 (by 12M)	46.3	11:35	47	0	FNA	ACR TI-RADS; 1 (11), 2 (6), 3 (20), 4 (10)^a		7.5 ± 6.7
Yu XY et al. (2025)⁶³	China	Retro.	Sep. 2016 to Mar. 2024	1, 3, 6, 9, 12, and biannual f/u	16 ± 14.1	MWA	44	60	59 (by 24M)	48.2	34:10	0	60	FNA	TI-RADS; 3 (44)^a	≤3 cm (no. of nodules ≤3), molecular test (+)^k	9.64 ± 17.9^f
Zhou XY et al. (2025)⁶⁴	China	Retro	Jan. 2014 to Jan. 2024	1, 3, 6, 9, 12, and annual f/u	23.6 ± 20.6	RFA, MWA	375	482	90 (by 60M)	43.2	298:77	482	0	FNA^c	NA		11.4 ± 12.6

The numbers in parentheses represent the number of nodules corresponding to each grade.

This study presents the cytopathologic results based on the British and Italian reporting systems for thyroid cytopathology. All the included nodules in this study are classified as Thy3, corresponding to Bethesda Categories III and IV.

Only BRAF- and NRAS-negative nodules were included.

Using the AllPrep DNA/RNA Micro Kit (Qiagen), genetic alterations, including BRAF^V600E , TERT, and RET/PTC1 (RET/CCDC6), were investigated. Only mutation-negative nodules were included.

Two molecular platforms—Afirma (GEC and GSC) and ThyGenX/ThyraMIR—were used for cytological analysis, but the study did not specify whether only mutation-negative nodules were included.

Standard deviation was estimated using either the interquartile range (SD ≈ IQR/1.35, assuming a normal distribution) or the range rule of thumb (SD ≈ [maximum − minimum]/4).

Molecular tests (Afirma® or ThyroSeq® v3) were used to confirm benignity before ablation. Only mutation-negative nodules were included.

The standard deviation was not reported.

The total number of nodules was not reported; however, 103 single and 58 multiple nodules were documented. For the purpose of analysis, the total number was assumed to be 161.

Molecular testing was conducted using an eight-gene panel, including BRAF, KRAS, NRAS, HRAS, TERT, RET, PAX8, and NTRK3, to identify point mutations and gene fusions such as BRAF^V600E , TERT C228T/C250T, and RET/PTC or PAX8-PPARγ fusions. Only mutation-negative nodules were included.

Preablation BRAF^V600E PCR testing was performed, and 29.5% of patients were found to be mutation-positive.

FNA, fine-needle aspiration; CNB, core needle biopsy; M, month; MWA, microwave ablation; NA, not available; No., number; Pro., prospective; Retro., retrospective; RFA, radiofrequency ablation; TI-RADS, Thyroid Imaging Reporting and Data System; US, ultrasonography.

Characteristics of the ablation methods

Table 2 summarizes the technical specifications of the ablation procedures, including the generator type, electrode characteristics (gauge, tip length, and cooling mechanism), power settings, number of sessions performed, anesthesia type, and sedation use. The electrode gauge used was 18G for RFA and 16G to 17G for MWA. The applied power ranged from 3 to 60 W for RFA (15–60 W for monopolar electrodes and from 3W for bipolar electrodes), and from 25 to 35 W for MWA. The majority of the included studies (9 out of 15) used RFA for TA.^{51

–57,59,61} Three studies used either RFA or MWA,^58,60,64 while the remaining three studies used MWA.^62,63,65 The mean ablation times ranged from 161 to 876 seconds. Most studies conducted a single session, although some reported 1–2 sessions. Most studies used local anesthesia, and only one study reported the use of sedation (midazolam).⁵²

Table 2.

Characteristic of the Ablation Methods

Author (year)	Ablationmethod	Generator	Electrode	Appliedpower (W)	Mean ablationtime (s)	Ablationsession No.	Anesthesia	Sedation
Chiang PL et al. (2023)⁵¹	RFA	VIVA (STARmed, Korea); M2004 (RF Medical, Korea)	18G (C., 0.5, 0.7, or 1.0 cm)	30 ∼	NA	Single	Local	No
Dobrinja C et al. (2015)⁵²	RFA	AGN-H-1.0 (HS Hospital Service SpA, Italy)	18G (C., 0.7, 1.0, or 1.5 cm)	30 ∼ 50	NA	Single	Local	Yes (MDZ)
Dong Y et al. (2024)⁵³	RFA	VIVA (STARmed, Korea)	18G (C., 1.0 cm)	40 ∼ 60	161	Single	Local	No
Fugazzola L et al. (2024)⁵⁴	RFA	VIVA (STARmed, Korea)	18G (C., 1.0 cm)	47.1 ± 10.9	540	Single	Local	No
Ha SM et al. (2017)⁵⁵	RFA	Cool-tip RF system (Radionics, USA)	18G (C., 0.5 cm)	18.8 ± 2.3	223	1 ∼ 2	Local	No
Issa PP et al. (2022)⁵⁶	RFA	NA	18G (C., 0.5 or 0.7 cm)	15 ∼	NA	1 ∼ 2	Local	No
Jackson K et al. (2025)⁵⁷	RFA	NA	18G (C., 0.7, 1.0, or 1.5 cm)	NA	NA	1 ∼ 2	Local	No
Li X et al. (2021)⁵⁸	RFA	CelonLabPOWER (Olympus, Germany)	18G (UC., 1.5 cm)	3 ∼	266	Single	Local	No
	MWA	MTI-5DT (Changcheng Microwave System Engineering, China)	17G (0.3 cm)	25 ∼ 35	266	Single	Local	No
Lin WC et al. (2021)⁵⁹	RFA	NA	0.5–1.0 cm	15 ∼ 60	876	1 ∼ 2	Local	No
Liu YT et al. (2025)⁶⁰	RFA	Cooltip Radiofrequency Ablation System (Covidien, Ireland)	C., 0.7 cm	40 ∼ 50	NA	Single	Local	No
	MWA	KY-2000 (Kangyou Applied Research Institute, China); Intelligent Basic Type Microwave Tumor Ablation System (Nanjing ECO Microwave System, China)	0.3 cm	30	NA	Single	Local	No
Ryu YJ et al. (2025)⁶¹	RFA	NA	NA	NA	NA	≧1	Local	No
Tang Q et al. (2025)⁶²	MWA	XR-A and XR-B systems (Nanjing Great Wall Information System, China)	NA	20.4 ± 3.2	NA	Single	Local	No
Yang J et al. (2022)	MWA	KY-2000 (Kangyou Applied Research Institute, China)	16G (C.)	25 ∼ 30	242	Single	Local	No
Yu XY et al. (2025)⁶³	MWA	KY-2000 (Kangyou Applied Research Institute, China); Intelligent Basic Type Microwave Tumor Ablation System (Nanjing ECO Microwave System, China)	17G (C., 0.35 cm)	30 ∼ 35	NA	Single	Local	No
Zhou XY et al. (2025)⁶⁴	RFA	Cooltip Radiofrequency Ablation System (Covidien, Ireland)	17G (C., 0.7–1.0 cm)	40 ∼ 60	NA	1 ∼ 2	Local	No
	MWA	Intelligent Basic Type Microwave Tumor Ablation System (Nanjing ECO Microwave System, China)	16G or 17G (C., 0.3 or 0.5 cm)	30	NA	1 ∼ 2	Local	No

C., cooled; MDZ, midazolam; s, second; UC., uncooled; W, watt.

Volume reduction rate after ablation

Table 3 displays the mean VRRs for the ablated nodules during the follow-up period, as well as the pooled proportions derived from the meta-analysis. The pooled VRRs at 1, 3, 6, 12, 24, 36, 48, and 60 months were 15.9% (confidence interval [CI]: −10.3–42.1%), 50.0% (CI: 40.1–59.8%), 68.0% (CI: 63.6–72.3%), 81.0% (CI: 76.0–85.9%), 85.1% (CI: 73.5–96.7%), 90.6% (CI: 73.2–108.0%), 93.8% (CI: 81.3–106.2%), and 93.0% (CI: 78.9–107.1%), respectively, based on 701, 653, 785, 769, 256, 77, 43, and 47 nodules (Fig. 2). Supplementary Figure S1 shows the trends of the pooled VRRs for nodules treated with TA. The pooled VRR progressively increased, showing a significant reduction in volume within 12 months, with a gradual increase thereafter, plateauing around 48 months. Substantial heterogeneity persisted across all time periods: 1 month (I ² = 99.8%, p = 0), 3 months (I ² = 99.1%, p < 0.01), 6 months (I ² = 93.5%, p < 0.01), 12 months (I ² = 93.9%, p < 0.01), 24 months (I ² = 99.5%, p < 0.01), 36 months (I ² = 96.3%, p < 0.01), 48 months (I ² = 89.6%, p < 0.01), and 60 months (I ² = 89.7%, p < 0.01).

FIG. 2.

Forest plots of pooled volume reduction rates at 1, 3, 6, 12, 24, 36, 48, and 60 months of follow-up.

Table 3.

Ablation Results (Volume Reduction Rate)

Author (year)	Ablationmethod	Baseline nodulevolume (mL)	VRR (%)
Author (year)	Ablationmethod	Baseline nodulevolume (mL)	1 Month	3 Months	6 Months	12 Months	24 Months	36 Months	48 Months	60 Months
Chiang PL et al. (2023)⁵¹	RFA	7.9 ± 13.4	NA	NA	77.9 ± 13.8	87.4 ± 10.8	NA	NA	NA	NA
Dobrinja C et al. (2015)⁵²	RFA	17.5 ± 5.6	33.4 ± 17.8^a	47.5 ± 14.4^a	44.2 ± 41.6^a	63.1 ± 8.61^a	58.4 ± 14.9^a	NA	NA	NA
Dong Y et al. (2024)⁵³	RFA	1.0 ± 0.7	−88.6 ± 2.12	16.0 ± 0.9	59.7 ± 0.59	82.0 ± 0.2	NA	NA	NA	NA
Fugazzola L et al. (2024)⁵⁴	RFA	17.3 ± 10.7	36.8 ± 16.5	NA	59.9 ± 15.5	62.0 ± 15.7	66.9 ± 4.0	NA	NA	NA
Ha SM et al. (2017)⁵⁵	RFA	0.6 ± 0.4	55.2 ± 22.0	NA	NA	94.1 ± 6.1	98.6 ± 2.8	99.2 ± 1.8	99.5 ± 1.0	99.5 ± 1.0
Issa PP et al. (2022)⁵⁶	RFA	1.9 ± 0.5^b	NA	NA	NA	69.2 ± 12.0^b	NA	NA	NA	NA
Jackson K et al. (2025)⁵⁷	RFA	15.8 ± 14.4	46.4 ± 21.4	NA	69.0 ± 9.6	81.6 ± 12.8	77.0	NA	NA	NA
Li X et al. (2021)⁵⁸	RFA, MWA	6.6 ± 9.1	47.4 ± 19.5	66.3 ± 15.5	77.1 ± 15.3	87.2 ± 13.1	94.6 ± 9.0	NA	NA	NA
Lin WC et al. (2021)⁵⁹	RFA	12.6 ± 20.9	NA	49.5 ± 21.0	65.4 ± 24.2	73.7 ± 17.7	NA	NA	NA	NA
Liu YT et al. (2025)⁶⁰	RFA, MWA	2.4 ± 2.6^b	−81.5^c	7.3^c	51.47^c	78.1^c	NA	NA	NA	NA
Ryu YJ et al. (2025)⁶¹	RFA	17.9 ± 21.9	45.1 ± 18.5^a,b	66.6 ± 14.8^a,b	75.5 ± 11.9^a,b	83.6 ± 8.1^a,b	NA	NA	NA	NA
Tang Q et al. (2025)⁶²	MWA	2.9 ± 4.7	NA	43.3 ± 26.9	57.8 ± 18.8	83.5 ± 15.5	90.1 ± 13.6	NA	NA	NA
Yang J et al. (2022)	MWA	7.5 ± 6.7	37.9 ± 22.5	58.6 ± 20.3	73.8 ± 17.2	85.0 ± 10.9	NA	NA	NA	NA
Yu XY et al. (2025)⁶³	MWA	9.64 ± 17.9^b	−48.4 ± 179.7	46.0 ± 113.0	70.9 ± 47.1	94.5 ± 25.5	100.0 ± 3.6	NA	NA	NA
Zhou XY et al. (2025)⁶⁴	RFA, MWA	11.4 ± 12.6	−21.1 ± 337.6	31.0 ± 202.5	58.8 ± 51.7	72.2 ± 39.1	83.9 ± 18.2	81.4 ± 27.5	86.7 ± 23.7	85.0 ± 28.2
No. of studies (nodules)		15 (1149)	11 (701)	10 (653)	13 (785)	15 (769)	8 (256)	2 (77)	2 (43)	2 (47)
Heterogeneity (I², p value)		98.0%, p < 0.01	99.8%, p = 0	99.1%, p < 0.01	93.5%, p < 0.01	93.9%, p < 0.01	99.5%, p < 0.01	96.3%, p < 0.01	89.6%, p < 0.01	89.7%, p < 0.01
Pooled proportions (CI)		8.29 (5.12 − 11.46)	15.9 (−10.3 − 42.1)	50.0 (40.1 − 59.8)	68.0 (63.6 − 72.3)	81.0 (76.0 − 85.9)	85.1 (73.5 − 96.7)	90.6 (73.2 − 108.0)	93.8 (81.3 − 106.2)	93.0 (78.9 − 107.1)

The data were estimated using figure.

Standard deviation was estimated using either the interquartile range (SD ≈ IQR/1.35, assuming a normal distribution) or the range rule of thumb (SD ≈ [maximum − minimum]/4).

While only the overall mean was reported for the entire cohort, serial mean VRRs and standard deviations were separately provided for two subgroups based on baseline nodule volume (≤2 cm and 2–3 cm), and these were used for data pooling.

CI, confidence interval; VRR, volume reduction rate.

Clinical and safety outcomes

Table 4 presents the details of clinical outcomes and complications following the procedures. Of the 15 included studies, three performed post-TA biopsies according to their own protocols. Of these three studies, one conducted biopsies on nodules larger than 1 cm at the 6-month post-RFA follow-up, one performed biopsies on all nodules, and the protocol for the third study was unspecified. A total of 35 biopsies were performed after TA, with no malignancy being detected in the postablation biopsy groups. There were 26 cases of regrowth, among which four patients underwent delayed surgery. Malignancy was confirmed in three of them: one had a follicular neoplasm of indeterminate malignant behavior (baseline volume: 24.5 mL) with regrowth at 24 months; another had a minimally invasive follicular carcinoma (baseline volume: 39.5 mL) with regrowth at 6 months; and the third had metastatic disease from a rare urethral cell carcinoma with regrowth at 36 months accompanied by PET avidity. The remaining patient underwent thyroidectomy due to anxiety, with no malignancy confirmed. No cases of cervical or distant metastasis were reported.

Table 4.

Clinical and Safety Outcomes

Author (year)	Nodule No.	Postthermal ablation					Major complications			Minorcomplications
Author (year)	Nodule No.	Biopsy No.	Regrowth No.	Delayedsurgery No.	Malignancy No.	Cervical or distantmetastasis No.	Total	Transient vocalcord palsy	Nodule rupture	Minorcomplications
Chiang PL et al. (2023)⁵¹	30	8^a	NA	NA	0	NA	2	1^b	1^c	0
Dobrinja C et al. (2015)⁵²	6	NA	2^d	2	2^e (MIFTC, FNwIMB)	NA	NA	NA	NA	NA
Dong Y et al. (2024)⁵³	62	NA	0	NA	NA	0	0	0	0	NA
Fugazzola L et al. (2024)⁵⁴	33	5^f	0	NA	0	NA	0	0	0	NA
Ha SM et al. (2017)⁵⁵	10	NA	0	NA	NA	0	0	0	0	0
Issa PP et al. (2022)⁵⁶	53	NA	5^g	NA	NA	NA	1	1^b	0	0
Jackson K et al. (2025)⁵⁷	32	NA	0	NA	NA	NA	4	4^b	0	4^h
Li X et al. (2021)⁵⁸	30	NA	0	NA	NA	0	0	0	0	0
Lin WC et al. (2021)⁵⁹	22	22ⁱ	0	NA	0	0	1	1^b	0	0
Liu YT et al. (2025)⁶⁰	161	NA	2^j	NA	NA	0	2	2^k	0	0
Ryu YJ et al. (2025)⁶¹	20	NA	1^l	1^l	1^l (Metastasis)	NA	0	0	0	0
Tang Q et al. (2025)⁶²	101	NA	0	NA	NA	0	0	0	0	0
Yang J et al. (2022)	47	NA	0	NA	NA	0	0	0	0	0
Yu XY et al. (2025)⁶³	60	NA	2^m	1ⁿ	NA	0	3	3^b	0	1^h
Zhou XY et al. (2025)⁶⁴	482	NA	14^m	0	NA	0	8	8^b	0	3^h
Total	1149	35	26	4	3	0	21	20	1	8

For eight nodules larger than 1 cm at the 6-month post-RFA follow-up, post-RFA FNA was performed to rule out malignancy.

All cases resolved within 3 months without any treatment.

A minimal isthmic rupture occurred 1-month post-RFA. The patient was closely monitored at the outpatient department and treated with only oral medication, ultimately recovering.

During postthermal ablation follow-up, an increase in VRR compared with prior VRR is defined as regrowth. In one case, regrowth was observed at the 6-month follow-up, while in the other case, it was noted at the 24-month follow-up.

After surgery, one (regrowth at 6-month follow-up) was a minimally invasive follicular carcinoma (baseline volume: 39.5 mL), and the other (regrowth at 24-month follow-up) was a follicular neoplasm of indeterminate malignant behavior (baseline volume: 24.5 mL).

For five patients, 6–12 months after RFA, the FNA results were Bethesda III.

Regrowth is defined as an increase in volume assessed by ultrasound, where the postoperative volume is greater than the preoperative volume. Of the nodules that regrew, the majority (42.1%) did so within the first month.

In Jackson K et al. (2025),⁵⁷ reported complications included skin burn, fever, and transient hyperthyroidism. Yu XY et al. (2025)⁶³ reported dysphagia. Zhou XY et al. (2025)⁶⁴ noted cough, back pain, and hypothyroidism.

All 22 patients underwent postablation fine needle aspiration within 6–12 months, with each cytology result consistent with Bethesda category I.

In this study, two residual tumors were identified and considered to result from regrowth of residual thyroid tissue within the ablation zone.

All cases resolved within 6 months without any treatment.

One case of regrowth at 36 months was later diagnosed as metastatic urethral cell carcinoma following thyroidectomy.

Number of cases of local recurrence.

Among the two patients with local recurrence, one underwent thyroidectomy due to anxiety, and the other continued regular follow-up.

FNwIMB, follicular neoplasm of indeterminate malignant behavior; MIFTC, minimally invasive follicular carcinoma.

Major complications were reported in 21 out of 1149 cases (1.8%).^{51,56,57,59,60,63,64} Voice changes were the most common complication, with 20 cases reported.^{51,56,57,59,60,63,64} All of them resolved within 6 months without any treatment. Nodule rupture occurred in one case and was treated with only oral medication at the outpatient department, with the patient ultimately recovering.⁵¹ Eight cases (0.7%) of minor complications were reported, including skin burn, fever, transient hyperthyroidism, hypoparathyroidism, dysphagia, and back pain.

Subgroup analysis

Subgroup analysis was conducted post hoc to evaluate the impact of hydrodissection, the baseline nodule volume, ablation method, and genetic alteration testing (Supplementary Figs. S2–S7). A significant difference was noted in the pooled VRR according to the presence^{51,53,58,60,61,63
–65} or absence^{52,54

–57,59,62} of hydrodissection at 6 months (p = 0.03), and the gap progressively increased throughout the 24-month follow-up period, as shown in Figure 3. Table 5 shows the subgroup analysis based on baseline nodule volume. Using a cutoff volume of 10 mL, VRR showed a significant difference at 12 and 24 months (p < 0.05). However, no significant difference in pooled VRR was observed between ablation methods (RFA vs. MWA) over the 24-month follow-up. Similarly, there was no significant difference between the molecularly tested and untested groups.

FIG. 3.

Trends in pooled volume reduction rates based on the presence or absence of hydrodissection. A significant difference in pooled VRR was observed at 6 months (70.9% with hydrodissection vs. 62.5% without hydrodissection, p = 0.03). VRR, volume reduction rate.

Table 5.

Subgroup Analysis Based on Baseline Nodule Volume

Baseline nodulevolume (mL)	Follow-upduration(month)	Smaller volumegroup (<X mL)		Larger volumegroup (>X mL)		p
Baseline nodulevolume (mL)	Follow-upduration(month)	No. of studies(nodules)	Pooled VRR (CI)	No. of studies(nodules)	Pooled VRR (CI)	p
<5 mL vs. >5 mL	1	3 (233)	−2.04 (−65.32 to 61.25)	8 (468)	32.14 (16.69–47.59)	0.30
	6	3 (274)	65.20 (56.95–73.45)	10 (511)	69.29 (64.27–74.31)	0.41
	12	5 (322)	84.13 (76.79–91.47)	10 (447)	78.92 (72.29–85.55)	0.30
	24	2 (25)	95.03 (86.81–103.25)	6 (231)	81.24 (65.97–96.50)	0.12
<7 mL vs. >7 mL	1	4 (263)	8.04 (−44.71 to 60.78)	7 (438)	26.59 (5.86–47.32)	0.52
	6	4 (307)	67.77 (59.70–75.83)	9 (478)	68.23 (62.94–73.51)	0.93
	12	6 (352)	84.54 (78.25–90.83)	9 (417)	77.96 (70.82–85.10)	0.18
	24	3 (55)	95.32 (90.93–99.72)	5 (201)	77.77 (59.83–95.72)	0.06
<10 mL vs. >10 mL	1	6 (348)	5.95 (−35.20 to 47. 09)	5 (353)	33.68 (15.21–52.15)	0.23
	6	7 (417)	70.16 (64.24–76.08)	6 (368)	64.76 (59.05–70.48)	0.20
	12	9 (453)	85.58 (80.89–90.26)	6 (316)	72.51 (65.31–79.71)	<0.05^a
	24	4 (73)	96.73 (93.09–100.37)	4 (183)	70.48 (55.95–84.95)	<0.05^a

Indicates a statistically significant difference.

CI, confidence interval; No, number.

Sensitivity analysis

Sensitivity analysis was performed by sequentially excluding each study and reanalyzing the remaining data (Supplementary Fig. S8). No single study had an undue influence on the overall meta-analysis results.

Publication bias

The funnel plot (Supplementary Fig. S9) shows a generally symmetrical distribution, with a few studies outside the pseudo 95% confidence limits. Egger’s test showed no significant asymmetry (p = 0.76), suggesting no evidence of publication bias.

Quality assessment of the studies

The quality of the 15 included studies was assessed using the RoBANS tool. As detailed in Figure 4, all studies showed a low risk of bias in areas such as selective reporting, outcome evaluation, and measurement exposure. However, there were concerns in five areas. Eight studies had a high risk of bias in participant comparability because of inadequate baseline characteristics.^{51,53
–55,59,62,64} All studies showed a high risk in the selection of participants, either because the outcome was unconfirmed at the time of enrollment or because recruitment strategies did not reflect the intended target population. Eleven studies demonstrated a high risk related to confounding variables, as they lacked appropriate identification or control of confounders.^{51
–53,55,56,58
–60,63
–65} In terms of blinding, all studies showed high risk due to the absence of outcome assessor blinding or uncertainty regarding whether blinding was maintained. In addition, seven studies were found to have follow-up losses, leading to a high risk of bias in the incomplete outcome data.^{57,58,61

–65}

FIG. 4.

Quality assessment of the included studies according to the Risk of Bias for Nonrandomized Studies (RoBANS) assessment tool.

Discussion

Our study suggests that TA may be an effective and safe treatment for ITNs, with a sustained decrease in nodule volume observed during the 60-month follow-up. The 6-month pooled VRR with hydrodissection is significantly higher than that without hydrodissection. Although the incidence of regrowth (2.3%) and delayed surgery (0.3%) was very low, three cases that underwent delayed surgery were confirmed to be malignant, raising concerns, although no metastasis occurred. In addition, the overall major complication rate was 1.8%, ranging from 1.2% to 12.5%.

According to the Bethesda system, the indeterminate categories of Bethesda III and IV carry a malignancy risk of 13–30% and 23–34%, respectively.⁴ In this study, the postablation pathologically confirmed malignancy rate was 0.3%. Although this represents a significant reduction compared with the preablation risk, there remains a residual malignancy risk, which poses a concern. This is why current guidelines recommend surgery as the first-line treatment for ITNs (AACE/ACE/AME, ETA), with RFA not being recommended (ATA, KSThR).^24,25,66,67 Nevertheless, TA offers several advantages over surgery. First, the rate of malignancy after TA treatment of ITNs is low. However, given that most included studies, except for two, had relatively short follow-up periods (12–24 months), there remains a possibility of additional cases of regrowth, malignancy, and even distant metastasis over time. Although the malignancy risk can be reduced with complete ablation, the residual risk still necessitates careful follow-up, similar to the approach used for active surveillance or RFA of PTMC.^13,68,69 In this study, three cases confirmed as malignancy after surgery were associated with regrowth, indicating that regrowth during follow-up should raise suspicion for malignancy and warrants consideration of surgery. In addition, compared with surgery, TA has a lower complication rate, shorter procedure time, and lower invasiveness.^65,70 For patients who are poor surgical candidates, elderly, or who refuse surgery, TA offers a minimally invasive alternative.

TA is highly effective for reducing nodule volume, with the pooled VRR continuing to increase up to 48 months, before reaching a plateau. However, considering the malignancy risk associated with Bethesda III and IV nodules, the treatment goal should not be merely volume reduction, but complete ablation to eliminate the risk of malignancy. In this study, the group that underwent hydrodissection showed a significantly higher pooled VRR at 6 months than the group without hydrodissection, suggesting that hydrodissection may increase the likelihood of complete ablation. Moreover, to minimize marginal regrowth, the guidelines propose vascular ablation techniques.^71
–73 These techniques lead to more efficient and complete ablation.⁷² In the subgroup analysis based on baseline nodule volume, setting higher thresholds for dividing the groups led to more significant differences in pooled VRR. Specifically, a statistically significant difference was observed at 12 and 24 months for nodules dichotomized with a 10 mL threshold, indicating that larger nodules have a higher likelihood of regrowth. Of the three cases confirmed as malignancy following surgery for regrowth after TA, two involved large nodules (>24 mL), while the remaining case did not report the nodule volume. A recent study by Ahn et al. found that large nodule size (≥3 cm) independently increased the malignancy risk in Bethesda Category III nodules categorized as low or intermediate suspicion on ultrasound.⁷⁴ In light of this, selecting only nodules of a manageable size for which complete ablation is achievable, and applying advanced techniques such as hydrodissection and vascular ablation, which are commonly used in RFA of benign thyroid nodules, could potentially improve treatment success rates for ITNs.

Although this study found no significant difference in pooled VRR between molecularly tested and untested groups, the high negative predictive value (>94%) of molecular tests suggests that genetic analysis may become a crucial step in deciding between TA and surgery for the treatment of ITNs, provided the cost is acceptable.^75
–77 In addition, a study focusing on ITNs with low SUV (SUV max <5) on PET/CT reported no malignancy among all the ablated nodules confirmed by biopsy.⁵⁹ In light of this, incorporating advanced diagnostic tools, such as molecular analysis and radioisotope imaging, into pretreatment evaluations could help identify nodules with a low likelihood of malignancy, thereby optimizing treatment decisions and improving patient outcomes. Ideally, ITNs with negative molecular profiles and low FDG uptake represent optimal candidates for TA.

The occurrence rate of major complications was low at 1.8%. All cases of transient vocal cord palsy fully recovered within 6 months without any specific treatment, and the one case of nodule rupture was resolved with oral medication. These findings confirm that TA is a safe method for treating ITNs.

This study has several limitations. First, 13 of the 15 studies had a short follow-up period of 12 to 24 months. This limited follow-up period may lead to underestimation, as malignancies that have not yet been detected could still emerge. To derive more definite conclusions about the safety and efficacy of TA, studies with long-term follow-up are crucial. Such studies could provide more reliable insights into delayed outcomes and better inform clinical decision-making. Second, post-TA biopsies were performed on only 3.0% of the nodules, and distant metastasis after TA was evaluated exclusively through US. This raises the possibility of undetected malignancies, potentially leading to an underestimation of the true malignancy rate. Third, since most of the included studies were retrospective, there is a risk of bias. Fourth, heterogeneity was present among the included studies. Fifth, in 10 of the 15 studies, the sample size was relatively small, with fewer than 50 patients. Last, our study lacks prior registration in a repository such as PROSPERO. Although a review of the PROSPERO database confirmed that no similar studies were registered at the time of our analysis, the absence of formal protocol registration limits the ability to fully ensure reproducibility and accountability.

In conclusion, TA may be considered a safe and effective treatment option for patients with ITNs who are not candidates for or decline surgery. However, caution is advised, and a thorough follow-up is essential due to the relatively short follow-up periods in most of the included studies. Larger nodules have a risk of incomplete ablation, and it is therefore important to selectively perform TA on nodules with a high likelihood of achieving complete ablation, using advanced techniques such as hydrodissection and vascular ablation, to ensure complete ablation. Limitations, including short follow-up periods, small study sizes, and study heterogeneity, emphasize the need for further research. Long-term prospective studies and randomized clinical trials are essential to better establish the reliability and clinical utility of TA.

Footnotes

Authors’ Contributions

H.L.: Conceptualization, formal analysis, investigation (lead), methodology, writing—original draft (lead), and writing—review and editing. S.J.C.: Conceptualization, formal analysis (lead), investigation, methodology (lead), supervision, and writing—review and editing (lead). Y.J.: Formal analysis, investigation, and methodology. S.Y.J.: Formal analysis, investigation, and methodology. J.H.B.: Conceptualization (lead), formal analysis, investigation, methodology, supervision (lead), and writing—review and editing.

Author Disclosure Statement

J.H.B. has been a consultant for two radiofrequency companies since 2017, STARmed and RF Medical.

Funding Information

This research received no external funding.

Supplementary Material

Supplementary Data

References

Fisher

, Perrier

. The incidental thyroid nodule. CA Cancer J Clin, 2018; 68(2):97–105; doi: 10.3322/caac.21447

Bongiovanni

, Spitale

, Faquin

, et al. The Bethesda system for reporting thyroid cytopathology: A meta-analysis. Acta Cytol, 2012; 56(4):333–339; doi: 10.1159/000339959

Baloch

, Asa

, Barletta

, et al. Overview of the 2022 who classification of thyroid neoplasms. Endocr Pathol, 2022; 33(1):27–63; doi: 10.1007/s12022-022-09707-3

Ali

, Baloch

, Cochand-Priollet

, et al. The 2023 Bethesda system for reporting thyroid cytopathology. Thyroid, 2023; 33(9):1039–1044; doi: 10.1089/thy.2023.0141

Yoon

, Kim

, Hong

, et al. Sonographic features of the follicular variant of papillary thyroid carcinoma. J Ultrasound Med, 2008; 27(10):1431–1437; doi: 10.7863/jum.2008.27.10.1431

Shih

, Shun

, Su

, et al. Follicular variant of papillary thyroid carcinoma: Diagnostic limitations of fine needle aspiration cytology. Acta Cytol, 2005; 49(4):383–386; doi: 10.1159/000326170

Yoon

, Baek

, Lee

, et al. Diagnosis of thyroid follicular neoplasm: Fine-needle aspiration versus core-needle biopsy. Thyroid, 2014; 24(11):1612–1617; doi: 10.1089/thy.2014.0140

Novoa

, Gürtler

, Arnoux

, et al. Role of ultrasound-guided core-needle biopsy in the assessment of head and neck lesions: A meta-analysis and systematic review of the literature. Head Neck, 2012; 34(10):1497–1503; doi: 10.1002/hed.21821

Roth

, Witt

, Steward

. Molecular testing for thyroid nodules: Review and current state. Cancer, 2018; 124(5):888–898; doi: 10.1002/cncr.30708

10.

Baloch

, Fleisher

, LiVolsi

, et al. Diagnosis of “follicular neoplasm”: A gray zone in thyroid fine-needle aspiration cytology. Diagn Cytopathol, 2002; 26(1):41–44; doi: 10.1002/dc.10043

11.

Lim

, Cho

, Baek

. Comparative efficacy and safety of radiofrequency ablation and microwave ablation in benign thyroid nodule treatment: A systematic review and meta-analysis. Eur Radiol, 2025; 35(2):612–623; doi: 10.1007/s00330-024-10881-7

12.

Lee

, Baek

, Chung

, et al. Radiofrequency ablation of recurrent thyroid cancers: Anatomy-based management. Ultrasonography, 2022; 41(3):434–443; doi: 10.1436/6/usg.21221

13.

Lim

, Cho

, Baek

, et al. US-Guided Radiofrequency Ablation for Low-Risk Papillary Thyroid Microcarcinoma: Efficacy and Safety in a Large Population. Korean J Radiol, 2019; 20(12):1653–1661.

14.

Chung

, Baek

, Choi

, et al. Ten-Year outcomes of radiofrequency ablation for locally recurrent papillary thyroid cancer. Korean J Radiol, 2024; 25(9):851–858.

15.

Fei

Y-L

, Wei

, Zhao

Z-L

, et al. Efficacy and safety of thermal ablation for solitary low-risk T2N0M0 papillary thyroid carcinoma. Korean J Radiol, 2024; 25(8):756–766.

16.

Shin

, Seo

, Lee

, et al. Comparison of the therapeutic efficacy and technical outcomes between conventional fixed electrodes and adjustable electrodes in the radiofrequency ablation of Benign thyroid nodules. Korean J Radiol, 2024; 25(2):199–209.

17.

Xue

, Teng

, Wang

. Efficacy and safety of ultrasound-guided radiofrequency ablation for papillary thyroid microcarcinoma: A systematic review and meta-analysis. Int J Hyperthermia, 2022; 39(1):1300–1309; doi: 10.1080/02656736.2022.2129101

18.

Cho

, Baek

, Lim

, et al. Long-Term follow-up results of ultrasound-guided radiofrequency ablation for low-risk papillary thyroid Microcarcinoma: More than 5-year follow-up for 84 Tumors. Thyroid, 2020; 30(12):1745–1751; doi: 10.1089/thy.2020.0106

19.

, Baek

, Che

, et al. Radiofrequency ablation of benign thyroid nodules: Recommendations from the Asian Conference on tumor ablation task force. Ultrasonography, 2021; 40(1):75–82; doi: 10.1436/6/usg.20112

20.

Papini

, Monpeyssen

, Frasoldati

, et al. 2020 European Thyroid Association clinical practice guideline for the use of image-guided ablation in Benign thyroid nodules. Eur Thyroid J, 2020; 9(4):172–185; doi: 10.1159/000508484

21.

, Lee

, Jung

, et al.; Korean Society of Radiology. Radiofrequency ablation of benign thyroid nodules and recurrent thyroid cancers: Consensus statement and recommendations. Korean J Radiol, 2012; 13(2):117–125; doi: 10.3348/kjr.2012.13.2.117

22.

Garberoglio

, Aliberti

, Appetecchia

, et al. Radiofrequency ablation for thyroid nodules: Which indications? The first Italian opinion statement. J Ultrasound, 2015; 18(4):423–430; doi: 10.1007/s40477-015-0169-y

23.

Mauri

, Hegedus

, Bandula

, et al. European Thyroid Association and Cardiovascular and Interventional Radiological Society of Europe 2021 clinical practice guideline for the use of minimally invasive treatments in malignant thyroid lesions. Eur Thyroid J, 2021; 10(3):185–197; doi: 10.1159/000516469

24.

Haugen

, Alexander

, Bible

, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid, 2016; 26(1):1–133; doi: 10.1089/thy.2015.0020

25.

Kim

, Baek

, Lim

, et al.; Guideline Committee for the Korean Society of Thyroid Radiology (KSThR) and Korean Society of Radiology. 2017 Thyroid radiofrequency ablation guideline: Korean society of thyroid radiology. Korean J Radiol, 2018; 19(4):632–655; doi: 10.3348/kjr.2018.19.4.632

26.

Page

, McKenzie

, Bossuyt

, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Bmj, 2021; 372:n71; doi: 10.1136/bmj.n71

27.

Lim

, Lee

, Ha

, et al. Radiofrequency ablation of benign non-functioning thyroid nodules: 4-year follow-up results for 111 patients. Eur Radiol, 2013; 23(4):1044–1049; doi: 10.1007/s00330-012-2671-3

28.

Ahmed

, Solbiati

, Brace

, et al.; Standard of Practice Committee of the Cardiovascular and Interventional Radiological Society of Europe. Image-guided tumor ablation: Standardization of terminology and reporting criteria–a 10-year update. Radiology, 2014; 273(1):241–260; doi: 10.1148/radiol.14132958

29.

Baek

, Lee

, Sung

, et al.; Korean Society of Thyroid Radiology. Complications encountered in the treatment of benign thyroid nodules with US-guided radiofrequency ablation: A multicenter study. Radiology, 2012; 262(1):335–342; doi: 10.1148/radiol.11110416

30.

Kim

, Lee

, Choi

, et al. Complications encountered in ultrasonography-guided radiofrequency ablation of benign thyroid nodules and recurrent thyroid cancers. Eur Radiol, 2017; 27(8):3128–3137; doi: 10.1007/s00330-016-4690-y

31.

Kim

, Park

, Lee

, et al. Testing a tool for assessing the risk of bias for nonrandomized studies showed moderate reliability and promising validity. J Clin Epidemiol, 2013; 66(4):408–414; doi: 10.1016/j.jclinepi.2012.09.016

32.

Kim

, Seo

, Lee

, et al. Study Design Algorithm for Medical literature of Intervention (DAMI) and Risk of Bias for Nonrandomized Studies (robans) ver 2.0 by HIRA. Seoul: Health Insurance Review & Assessment Service, 2013.

33.

Kim

, Lee

, Choi

, et al. Systematic review and meta-analysis of studies evaluating diagnostic test accuracy: A Practical Review for Clinical Researchers-Part I. General guidance and tips. Korean J Radiol, 2015; 16(6):1175–1187; doi: 10.3348/kjr.2015.16.6.1175

34.

Lee

, Kim

, Choi

, et al. Systematic review and meta-analysis of studies evaluating diagnostic test accuracy: A practical review for clinical Researchers-Part II. statistical methods of meta-analysis. Korean J Radiol, 2015; 16(6):1188–1196; doi: 10.3348/kjr.2015.16.6.1188

35.

Suh

, Park

. Successful publication of systematic review and meta-analysis of studies evaluating diagnostic test accuracy. Korean J Radiol, 2016; 17(1):5–6; doi: 10.3348/kjr.2016.17.1.5

36.

Higgins

, Thompson

, Deeks

, et al. Measuring inconsistency in meta-analyses. BMJ, 2003; 327(7414):557–560; doi: 10.1136/bmj.327.7414.557

37.

Belitova

, Ivanov

, Marinov

, et al. Minimally invasive radiofrequency ablation for large thyroid toxic adenoma. Acta Medica Bulgarica, 2019; 46(3):50–52; doi: 10.2478/amb-2019-0031

38.

Baek

. Thyroid tumours. CardioVascular and Interventional Radiology, 2013; 36:S208–S209.

39.

Cho

. Radiofrequency ablation of thyroid nodules. Ultrasound in Medicine and Biology, 2019; 45:S71; doi: 10.1016/j.ultrasmedbio.2019.07.648

40.

Klopper

, Ahmad

, Lilla

, et al. #1702521 Efficacy of radiofrequency ablation for afirma Genomic Sequencing Classifier (GSC) benign indeterminate thyroid nodules. Endocrine Practice, 2024; 30(5):S147–S148; doi: 10.1016/j.eprac.2024.03.058

41.

Abdelhady

. Radiofrequency ablation of indeterminate thyroid nodules: The First North American comparative analysis. Thyroid, 2023; 33:A139; doi: 10.1089/thy.2023.29161.lb.abstracts

42.

Buti

, Issa

, Omar

, et al. Radiofrequency ablation of indeterminate thyroid nodules; The First North American comparative analysis. Thyroid, 2022; 32:A133; doi: 10.3390/ijms231911493

43.

Yang

, Chu

, Xiang

, et al. Efficacy and safety of UG-MWA versus surgical resection for bethesda Category IV thyroid nodules. Thyroid, 2022; 32(A132); doi: 10.3389/fendo.2022.924993

44.

Baldwin

, Natter

, Patel

, et al. Minimally invasive techniques for the management of thyroid nodules. Endocrinol Metab Clin North Am, 2022; 51(2):323–349; doi: 10.1016/j.ecl.2022.01.001

45.

Bandeira‐Echtler

, Bergerhoff

, Richter

. Levothyroxine or minimally invasive therapies for benign thyroid nodules. Cochrane Database Syst Rev, 2014; 2014(6):CD004098; doi: 10.1002/14651858.CD004098.pub2

46.

Radzina

, Cantisani

, Rauda

, et al. Update on the role of ultrasound guided radiofrequency ablation for thyroid nodule treatment. Int J Surg, 2017; 41 (Suppl 1):S82–S93; doi: 10.1016/j.ijsu.2017.02.010

47.

Zhou

, Yu

, Wei

, et al. Efficacy and safety of microwave ablation for treatment of follicular thyroid neoplasms: A preliminary study. Int J Hyperthermia, 2024; 41(1):2398558; doi: 10.1080/02656736.2024.2398558

48.

Joerger

, Knüsel

, Alejandre-Lafont

, et al. Thermal ablation followed by intratumoral injection of a novel immune stimulant IP-001 in patients with advanced solid tumors: Phase IB part of study SAKK 66/17. Immuno-Oncology and Technol, 2022; 16:100306; doi: 10.1016/j.iotech.2022.100306

49.

Kim

, Shin

, Hahn

, et al. Delayed cancer diagnosis in thyroid nodules initially treated as benign with radiofrequency ablation: Ultrasound characteristics and predictors for cancer. Korean J Radiol, 2023; 24(9):903–911; doi: 10.3348/kjr.2023.0386

50.

Metz

, Omar

, Cironi

, et al. Risk of malignancy following radiofrequency ablation of thyroid nodules. Thyroid, 2022; 32:A53–A54; doi: 10.1089/thy.2022.29138.abstracts

51.

Chiang

, Luo

, Chang

, et al. Radiofrequency ablation for thyroid Bethesda III nodules: Preliminary results. Eur Thyroid J, 2023; 12(6); doi: 10.1530/etj-23-0105

52.

Dobrinja

, Bernardi

, Fabris

, et al. Surgical and pathological changes after radiofrequency ablation of thyroid nodules. Int J Endocrinol, 2015; 2015:576576; doi: 10.1155/2015/576576

53.

Dong

, Zhan

, Zhou

, et al. Volume reduction rate of radiofrequency ablation in ≤ 2 cm Bethesda IV thyroid nodules. Eur Radiol, 2024; 34(3):1597–1604; doi: 10.1007/s00330-023-10185-2

54.

Fugazzola

, Deandrea

, Borgato

, et al. Radiofrequency ablation is an effective treatment for Bethesda III thyroid nodules without genetic alterations. Eur Thyroid J, 2024; 13(3); doi: 10.1530/etj-24-0020

55.

, Sung

, Baek

, et al. Radiofrequency ablation of small follicular neoplasms: Initial clinical outcomes. Int J Hyperthermia, 2017; 33(8):931–937; doi: 10.1080/02656736.2017.1331268

56.

Issa

, Omar

, Issa

, et al. Radiofrequency ablation of indeterminate thyroid nodules: The First North American Comparative Analysis. Int J Mol Sci, 2022; 23(19); doi: 10.3390/ijms231911493

57.

Jackson

, McCabe

, Stern

, et al. Early results of radiofrequency ablation for treating molecular negative Bethesda III thyroid nodules. J Am Coll Surg, 2025; doi: 10.1097/XCS.0000000000001388

58.

, Lan

, Li

, et al. Ultrasound-Guided thermal ablation of Bethesda IV Thyroid Nodules: A pilot study. Front Endocrinol (Lausanne), 2021; 12:674970; doi: 10.3389/fendo.2021.674970

59.

Lin

, Tung

, Chang

, et al. Radiofrequency ablation for treatment of thyroid follicular neoplasm with low SUV in PET/CT study. Int J Hyperthermia, 2021; 38(1):963–969; doi: 10.1080/02656736.2021.1912414

60.

Liu

, Fan

, He

, et al. The preliminary multicenter retrospective study on the efficacy and safety of thermal ablation for follicular thyroid neoplasms ≤ 3 cm. Discov Oncol, 2025; 16(1):219; doi: 10.1007/s12672-025-01984-8

61.

Ryu

, Hsu

, Kuo

, et al. Radiofrequency ablation of Bethesda category III thyroid nodules with benign molecular testing: Preliminary findings from a single institution. Am J Surg, 2025; 241:115929; doi: 10.1016/j.amjsurg.2024.115929

62.

Tang

, Chen

, Zhang

, et al. Safety and feasibility of US-guided microwave ablation for the treatment of Bethesda III Thyroid nodules with negative eight-gene panel mutational profile. Radiology: Imaging Cancer, 2025; 7(1); doi: 10.1148/rycan.240058

63.

, Zhou

, Wei

, et al. A preliminary study of microwave ablation for Bethesda IV follicular neoplasms (≤3 cm). Br J Radiol, 2025; 98(1168):578–585; doi: 10.1093/bjr/tqaf015

64.

Zhou

, Fan

, He

, et al. Efficacy, safety, and risk factors of thermal ablation for follicular thyroid neoplasms: A multicentric study. Endocrine, 2025; doi: 10.1007/s12020-025-04213-y

65.

Yang

, Zhang

, Li

, et al. Efficacy and safety of ultrasound-guided microwave ablation versus surgical resection for Bethesda category IV thyroid nodules: A retrospective comparative study. Front Endocrinol (Lausanne), 2022; 13:924993; doi: 10.3389/fendo.2022.924993

66.

Gharib

, Papini

, Garber

, et al. American association of clinical endocrinologists, American College of Endocrinology, and Associazione Medici Endocrinologi Medical Guidelines for clinical practice for the diagnosis and management of thyroid nodules - 2016 update Appendix. Endocrine Practice, 2016; 22:1–60; doi: 10.4158/EP161208.GL

67.

Durante

, Hegedüs

, Czarniecka

, et al. 2023 European Thyroid Association clinical practice guidelines for thyroid nodule management. Eur Thyroid J, 2023; 12(5); doi: 10.1530/etj-23-0067

68.

Lee

, Lee

, Lim

, et al.; Korean Society of Radiology. Standardized ultrasound evaluation for active surveillance of low-risk thyroid Microcarcinoma In Adults: 2024 Korean society of thyroid radiology consensus statement. Korean J Radiol, 2024; 25(11):942–958.

69.

Ghai

, Goldstein

, Sawka

. Ultrasound imaging in active surveillance of small, low-risk papillary thyroid cancer. Korean J Radiol, 2024; 25(8):749–755.

70.

Che

, Jin

, Shi

, et al. Treatment of benign thyroid nodules: Comparison of surgery with radiofrequency ablation. Am J Neuroradiol, 2015; 36(7):1321–1325; doi: 10.3174/ajnr.A4276

71.

Sim

, Baek

. Long-term outcomes of thermal ablation for benign thyroid nodules: The issue of regrowth. Int J Endocrinol, 2021; 2021:9922509; doi: 10.1155/2021/9922509

72.

Park

, Baek

, Park

, et al. Thyroid radiofrequency ablation: Updates on innovative devices and techniques. Korean J Radiol, 2017; 18(4):615–623.

73.

Offi

, Garberoglio

, Antonelli

, et al. The Ablation of thyroid nodule’s afferent arteries before radiofrequency ablation: Preliminary data. Front Endocrinol (Lausanne), 2020; 11:565000; doi: 10.3389/fendo.2020.565000

74.

Ahn

, Na

, Kim

J-H

. Risk stratification of thyroid nodules diagnosed as Bethesda Category III by ultrasound, size, and cytology. Korean J Radiol, 2024; 25(10):924–933.

75.

Labourier

, Shifrin

, Busseniers

, et al. Molecular testing for miRNA, mRNA, and DNA on fine-needle aspiration improves the preoperative diagnosis of thyroid nodules with indeterminate cytology. J Clin Endocrinol Metab, 2015; 100(7):2743–2750; doi: 10.1210/jc.2015-1158

76.

Patel

, Angell

, Babiarz

, et al. Performance of a genomic sequencing classifier for the preoperative diagnosis of cytologically indeterminate thyroid nodules. JAMA Surg, 2018; 153(9):817–824; doi: 10.1001/jamasurg.2018.1153

77.

Steward

, Carty

, Sippel

, et al. Performance of a multigene genomic classifier in thyroid nodules with indeterminate cytology: a prospective blinded multicenter study. JAMA Oncol, 2019; 5(2):204–212; doi: 10.1001/jamaoncol.2018.4616

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

5.82 MB

0.00 MB