Radiofrequency ablation for papillary thyroid microcarcinoma with a trachea-adjacent versus trachea-distant location

Abstract

Objective

To evaluate the outcomes of radiofrequency ablation (RFA) for papillary thyroid microcarcinoma (PTMC) adjacent to the trachea and compare them with those of PTMC distant from the trachea.

Methods

Patients who received RFA for solitary low-risk PTMC between June 2014 and July 2020 were reviewed and classified into adjacent and distant groups. To balance between-group confounders, the propensity score matching approach was employed. Volume, volume reduction ratio (VRR), tumor disappearance, complications, and disease progression were assessed and compared between the groups. Furthermore, factors affecting disease progression were evaluated.

Results

A total of 122 and 470 patients were included in the adjacent and distant groups, respectively. Overall VRR was 99.5% ± 3.1 and cumulative tumor disappearance rate was 99.4% after a mean follow-up time of 40.1 months ± 16.2. Overall disease progression and complications incidence were 3.7% and 1.0%, respectively. No substantial differences were observed between the two groups in the latest volume (0.8 mm³ ± 4.1 vs. 0.9 mm³ ± 4.2, p = .77), VRR (99.7% ± 1.6 vs. 99.5% ± 2.7, p = .75), cumulative tumor disappearance rate (92.6% vs. 94.2%, p = .58), and incidence of disease progression (4.1% vs. 4.5%, p = .70) and complication (1.7% vs. 0.8%, p = .86) after 1:2 matching. Additionally, tracheal adjacency exhibited no association with disease progression in multivariate Cox regression analysis (p = .73).

Conclusion

For eligible patients with PTMC located adjacent to or distant from the trachea, RFA may offer a safe and effective alternative treatment method.

Keywords:

• Ablation techniques
• radiofrequency ablation
• papillary thyroid microcarcinoma
• ultrasonography

Introduction

Thyroid carcinoma, the most common endocrine malignancy worldwide, is responsible for more than 586,000 new cases in 2020 and ranks 9th in terms of incidence among all cancers [1]. Over the last few decades, the global incidence of thyroid carcinoma has significantly increased, whereas mortality due to this cancer has remained steady or even declined [2–4]. This unusual phenomenon has been ascribed to the prevalence of papillary thyroid microcarcinoma (PTMC), which is defined as papillary thyroid carcinoma (PTC) with a maximal diameter of no more than 1.0 cm [5]. PTMC shows favorable progression with a 10-year survival rate of > 99% [5,6].

A significant proportion of thyroid cancer cases are located adjacent to the trachea, with no intervening thyroid tissue. Approximately 38.0% of PTCs and 26.0% of all thyroid cancers are identified as trachea-adjacent tumors [7,8]. A risk of tracheal invasion is posed by trachea-adjacent PTCs, especially when an obtuse angle between the tumor and trachea is observed, which is associated with a higher risk of tracheal invasion [8]. Prompt surgical intervention is essential after identifying tracheal invasion [5]. However, the optimal treatment approach for PTCs that abut the trachea without clinical or imaging evidence of tracheal invasion remains unclear. Thyroidectomy has traditionally been the primary treatment for PTMC [5]. Although the primary tumor can be effectively removed through surgery, it may impact the quality of life by causing scarring, lifelong thyroid hormone replacement, and recurrent laryngeal nerve injury [9]. The American Thyroid Association guidelines now consider active surveillance (AS) as an alternative management strategy for low-risk PTC (T1aN0M0) to avoid overtreatment [5]. AS is recommended by the Japan Association of Endocrine Surgery for low-risk PTMCs that simply touch the trachea [10]. However, psychological anxiety about disease progression throughout the follow-up period may be experienced by patients who opt for AS, and delayed surgical resection is ultimately chosen by many [11]. For benign thyroid neoplasms and primary low-risk PTMC, thermal ablation (TA) has been demonstrated as an effective, safe, and minimally invasive treatment modality [12–14]. Studies have demonstrated favorable findings for TA for treating low-risk PTMC [14–17]. Furthermore, the efficacy of ablation for PTCs at specific locations in the thyroid gland has been investigated in several studies. Song et al. assessed the efficacy of TA for PTMC located in the isthmus and discovered that TA yielded outcomes similar to those of total thyroidectomy [18,19]. Information on the effectiveness of TA for PTMC touching the thyroid capsule was provided by another study by Zheng et al. [20]. To the best of our knowledge, the efficacy and safety of TA in PTMC adjacent to the trachea have not been investigated in any studies.

Therefore, this study aimed to assess the efficacy and safety of TA for solitary PTMC adjacent to the trachea and compare the outcomes with those of PTMC distant from the trachea.

Materials and methods

Patients

The Institutional Review Board of our institution approved this retrospective research. Written informed consent was obtained from all patients who underwent radiofrequency ablation (RFA) before treatment. Ablation was recommended for eligible patients who refuse surgery and AS, and a comprehensive discussion of the advantages and disadvantages was conducted before the ablation procedure.

Patients with PTMC who underwent RFA at our institution between June 2014 and July 2020 were reviewed. The inclusion criteria were as follows: (1) pathologically diagnosed PTC through core-needle biopsy or fine needle aspiration; (2) a unifocal tumor with a maximal diameter of ≤ 10 mm; (3) no clinical or imaging evidence indicating gross extrathyroidal extension (ETE) [6], lymph node metastasis (LNM), or distant metastasis; and (4) a tumor adjacent to the trachea (with no thyroid tissue between the thyroid cancer and the trachea on imaging) or distant from the trachea (with a distance of > 2 mm between tumor and the trachea). The exclusion criteria were as follows: (1) tumor with evidence of aggressive PTC by biopsy; (2) patients with a history of neck irradiation or surgical resection; (3) tumors with an obtuse angle between the tumor and trachea (Supplementary Figure 1); (4) patients with other malignant diseases; and (5) patients with a follow-up time < 24 months. The patients were categorized into two groups: the adjacent group (group A), which included patients with PTMCs located adjacent to the trachea, and the distant group (group D), which comprised patients with PTMCs located distant from the trachea (Figure 1).

Figure 1. The flowchart of patient enrollment.

Pre-RFA assessment

All the patients underwent laboratory tests (complete blood count, coagulation tests, and thyroid function tests) and imaging examinations before treatment. Ultrasound (US) and contrast-enhanced ultrasound (CEUS) (SonoVue, Bracco International) were used to assess the tumor size ($\mathit{\text{volume}}=\pi \mathit{\text{abc}}/6,$ a: transverse diameter; b: vertical diameter; c: anteroposterior diameter) and tumor characteristics. CEUS enhancement modes were categorized into hypo-enhancement, iso-enhancement, and hyper-enhancement compared with the thyroid parenchyma intensity at peak enhancements. To detect tracheal invasion, LNM, and distant metastases, computed tomography (CT) was performed. US-guided core-needle biopsy was performed to determine metastatic status when cervical LNM was suspected.

RFA procedure

Before treatment, communication was performed to reduce nervousness and promote cooperation. Patient’s posture and sterilization procedures were performed in compliance with previously published protocols [15,19]. All procedures were conducted under local anesthesia induced by administrating 1% lidocaine, which was injected into both the subcutaneous puncture site and the anterior capsule of the thyroid. Ablation was performed using an 18-gauge bipolar RF applicator with a 0.9 cm active tip (CelonProSurge micro 100-T09, Olympus Surgical Technologies Europe) and was powered by a bipolar radiofrequency ablation generator (CelonLabPOWER, Olympus Surgical Technologies Europe). The ablation power was set to 4–9 W. Several approaches were employed to prevent thermal injury during the process (Figure 2). First, the moving-shot technique was used by dividing the target tumor into many small ablation units, which were ablated one by one, starting from the deepest portion and moving toward the most superficial portion [21]. Ablation was initiated from the portion closest to the trachea in cases where the tumors were located in close proximity to the trachea. Second, the trans-isthmic approach was employed by inserting the ablation needle into the target tumor in the lateral thyroid lobe via the isthmus, moving from the midline-to-lateral direction [21]. Furthermore, for tumors adjacent to the trachea or those within 5 mm of other critical neck structures, the hydrodissection technique was performed [22]. To achieve this, normal saline was injected using a 23-gauge needle to create a distance ≥ 5 mm from the critical structures. Additionally, the leverage pry-off approach was employed to raise or lower the ablation needle, using either the hands of the operator or the trachea of the patient as a fulcrum [23]. The capsule was ablated during the ablation procedure because of the thinness of the capsule between the tumor and the trachea [18]. To ensure complete ablation, the ablation area was exceeded by at least 3–5 mm beyond the tumor edge. CEUS was conducted following RFA to evaluate the completeness of the ablation. Consistent with previous studies and guidelines [24–26], complete non-enhancement that covered the tumor and a 3–5 mm margin around the tumor on CEUS were considered to indicate complete ablation. An additional ablation was performed if incomplete ablation was detected on CEUS. Patients were instructed to apply pressure to the neck for at least 30 min and to remain in the observation room for a minimum of two hours.

Figure 2. Ultrasound (US) images of the ablation procedure performed on a 31-year-old female with papillary thyroid carcinoma adjacent to the trachea. (a) US image displaying the tumor (arrowheads) before ablation; (b) the hydrodissection technique (arrows) was applied wherein normal saline was injected into the space between the tumor (arrowheads) and the trachea as well as the space between the anterior thyroid capsule and superficial soft tissue; (c) the ablation needle (arrows) was inserted into the target tumor using the trans-isthmic approach; the ablation was initiated from the deepest portion of the tumor and the portion closest to the trachea. The tumor was elevated away from the trachea using the operator’s finger as the fulcrum (utilizing the leverage pry-off method) to further increase the distance (double side arrow) between the tumor and the trachea during ablation; (d) the ablation needle (arrows) was moved toward to the superficial portion of the tumor (arrowheads).

Follow-up

After ablation, follow-ups were scheduled for all patients at one, three, six, and 12 months with subsequent follow-ups every 6–12 months. At each follow-up, US, CEUS, and laboratory tests were performed. Figure 3 illustrates a typical case of RFA treatment for PTMC adjacent to the trachea and the follow-up conducted after RFA. Core-needle biopsy was performed to assess the efficacy of ablation at 3 or 6 months after ablation unless the ablation zone disappeared. Biopsy was also performed if residual cancer, new thyroid cancer, or LNM was suspected. Annual CT scans were performed. PET or bone scans were performed if distant metastases were observed.

Figure 3. The US and contrast-enhanced ultrasound (CEUS) images were obtained in a 64-year-old male with papillary thyroid carcinoma adjacent to the trachea in the right thyroid lobe. (a,b) transverse and longitudinal planes of US images of the tumor (arrows) before RFA; (c,d) CEUS images illustrating non-enhancement in the ablation area (arrows) immediately after RFA; (e,f) CEUS images of the ablation area (arrows) at 1-month follow-up; (g,h) CEUS images of the ablation area (arrows) at 3-months follow-up; (i,j) the ablation area displaying scar-like non-enhancement (arrows) at 12-month follow-up; (k,l) the ablation area completely disappeared at the 18-month follow-up.

Evaluation criteria

Technical success was defined as complete non-enhancement of the target tumor on CEUS after ablation [24]. Volume, volume reduction rate (VRR, calculated as $\mathit{\text{VRR}}=[\text{initial volume}-\text{final volume}]\times 100/\text{initial volume}$), accumulative tumor disappearance rate, and disease progression were included in the assessment of technique efficacy. Tumor disappearance was defined as the complete disappearance of the ablation area of PTC on US or the ablation area remaining scar-like on US but showing no enhancement on CEUS. Disease progression was defined as (1) the presence of a persistently detected lesion in the ablation zone confirmed by biopsy, (2) pathologically confirmed a new malignant thyroid tumor, (3) pathologically confirmed LNM, (4) detection of distant metastasis through imaging, and (5) death result from PTC [27]. Complications were identified based on the standards for image-guided thyroid ablation [28].

Statistical analysis

Statistical analyses were performed using R software (version 4.1.0; R Foundation for Statistical Computing, Vienna, Austria). A 1:2 propensity score matching (PSM) method was used to reduce potential bias between the groups. Standardized mean differences (SMDs) in baseline characteristics were computed before and after matching to indicate balance. Balance was considered achieved when the SMD values were no greater than 0.1. Matching factors included age, sex, volume, mode of CEUS, presence of Hashimoto’s thyroiditis (HT), and follow-up time. The nearest neighbor matching with a caliper of 0.20 without replacement was used to match patients. Continuous variables are presented as mean ± SD and analyzed using the Wilcoxon signed-rank test or Mann–Whitney U test. Categorical variables are presented as numbers with percentages and compared using the Chi-square or Fisher’s exact test. Kaplan–Meier approach was employed to generate progression-free survival curves and cumulative tumor disappearance rate curves, and the log-rank test was employed to compare between-group differences. To identify the variables affecting disease prognosis, the Cox proportional hazards regression model was employed. Statistical significance was set at p < .05.

Results

Demographic and tumor characteristics

A total of 122 patients (94 females, 28 males; mean age: 43.7 years ± 10.4) were enrolled in group A and 470 patients (376 females, 28 males; mean age: 44.0 years ± 10.0) were enrolled in group D. The mean follow-up time was 40.1 months ± 16.2 (range, 24.0–96.4 months). Although no significant differences in baseline characteristics between the two groups were observed, the SMD values in volume, follow-up time, and CEUS mode were all greater than 0.10 before PSM. After 1:2 PSM, 121 and 242 patients were included in groups A and D, respectively. After PSM, all SMD values were less than 0.10, indicating that the baseline characteristics were well-matched between the two groups (Table 1).

Table 1. Baseline characteristics of patients before and after propensity score matching.

	Before matching					After matching
Variables	Adjacent group (n = 122)	Distant group (n = 470)	p value	SMD	Adjacent group (n = 121)	Distant group (n = 242)	p value	SMD
Age (years)	43.7 ± 10.4	44.0 ± 10.0	.68	0.004	43.7 ± 10.5	43.8 ± 10.1	.93	0.004
≥45	57 (46.7)	237 (50.4)	.47	−0.07	57 (47.1)	122 (50.4)	.55	−0.07
<45	65 (53.3)	233 (49.6)			64 (52.9)	120 (49.6)
Sex			.47	0.07			.93	0.01
Female (%)	94 (77.3)	376 (78.9)			93 (76.9)	187 (77.3)
Male (%)	28 (22.7)	94 (21.1)			28 (23.1)	55 (22.7)
Volume (mm^3)	131.1 ± 105.4	101.5 ± 92.6	.06	0.15	130.4 ± 105.6	127.5 ± 106.7	.25	0.06
Follow-up (months)	38.2 ± 13.7	40.6 ± 16.8	.42	−0.18	38.3 ± 13.7	37.3 ± 14.0	.28	0.07
CEUS			.36	−0.11			.86	−0.02
Hypo-enhancement	116 (95.1)	436 (92.8)			115 (95.0)	231 (95.5)
Iso-/hyper-enhancement	6 (4.9)	34 (7.2)			6 (5.0)	11 (4.5)
HT			.34	−0.10			.74	−0.04
Present	33 (27.0)	148 (31.5)			33 (27.3)	62 (25.6)
Absent	89 (73.0)	322 (68.5)			88 (72.7)	180 (74.4)

CEUS: contrast-enhanced ultrasound; HT: Hashimoto’s thyroiditis; MD: maximal diameter; SMD: standardized mean difference.

Data are presented as mean ± SD or number (percentages).

Changes in tumor characteristics

After RFA, all target tumors exhibited complete non-enhancement on CEUS, achieving a technical success rate of 100%. The overall volume was 0.7 mm³ ± 3.6 and VRR was 99.5% ± 3.1 at the latest follow-up. The mean volume and VRR at each follow-up point are presented in Table 2. At any of the follow-up points before and after matching, no significant differences in volume between groups A and D were observed (all p > .05). At 1-month follow-up, the mean VRR in group A was higher than that in group D (−234.9% ± 253.1 vs. −721.3% ± 4132.8, p = .03). However, no differences were observed at other follow-up points (all p > .05). At the most recent follow-up, the mean volume and VRR for group A vs. group D were 0.8 mm³ ± 4.1 vs. 0.9 mm³ ± 4.2 and 99.7% ± 1.6 vs. 99.5% ± 2.7, respectively, indicating no significant between-group differences (p = .77, p = .75, respectively).

Table 2. Changes in volume and VRR of PTMCs in adjacent and distant groups at each follow-up point.

	Volume (before matching)				Volume (after matching)
(a) Time	Total	Adjacent group (n = 122)	Distant group (n = 470)	p value	Adjacent group (n = 121)	Distant group (n = 242)	p value
Baseline	107.8 ± 96.2	115.7 ± 94.4	101.6 ± 84.9	.06	113.1 ± 90.5	107.1 ± 97.6	.25
1 month	475.4 ± 2489.2a	320.6 ± 315.7^a	514.2 ± 2778.5^a	.26	313.8 ± 310.6^a	642.6 ± 3777.8^a	.10
3 months	141.3 ± 215.5	133.4 ± 179.0	143.3 ± 224.1	.65	132.0 ± 179.1	151.4 ± 254.2	.61
6 months	53.9 ± 115.6^a	40.0 ± 74.9^a	57.3 ± 123.5^a	.49	38.6 ± 74.7^a	61.7 ± 145.9^a	.52
12 months	18.2 ± 52.7^a	18.7 ± 63.8^a	18.0 ± 54.9^a	.79	19.0 ± 64.0^a	17.0 ± 44.3^a	.72
24 months	4.7 ± 18.4^a	5.3 ± 18.0^a	4.5 ± 87.7^a	.89	5.4 ± 18.1^a	4.0 ± 15.5^a	.85
36 months	1.0 ± 4.0^a	1.0 ± 4.9^a	0.9 ± 5.0^a	.91	1.0 ± 4.9^a	1.3 ± 6.0^a	.64
48 months	0.7 ± 3.6^a	0.8 ± 4.1^a	0.6 ± 3.5^a	1.00	0.8 ± 4.1^a	0.9 ± 4.2^a	.77
	VRR (before matching)				VRR (after matching)
(b) Time	Total	Adjacent group (n = 122)	Distant group (n = 470)	p value	Adjacent group (n = 121)	Distant group (n = 242)	p value
1 month	−517.3 ± 2719.7	−234.7 ± 251.9	−588.1 ± 3035.4	.002	−234.9 ± 253.1	−721.3 ± 4132.8	.003
3 months	−73.5 ± 327.0	−34.7 ± 187.1	−83.5 ± 353.8	.17	−34.9 ± 187.7	−60.5 ± 231.7	.49
6 months	31.8 ± 191.6	60.8 ± 79.7	24.6 ± 209.9	.05	61.2 ± 80.1	41.8 ± 143.1	.19
12 months	80.3 ± 67.7	81.5 ± 62.7	80.0 ± 69.0	.97	81.2 ± 62.9	84.2 ± 36.2	.69
24 months	94.6 ± 25.5	96.3 ± 12.6	94.2 ± 118.8	.94	96.2 ± 12.6	96.3 ± 15.8	.85
36 months	99.2 ± 4.7	99.4 ± 2.9	99.1 ± 5.0	.89	99.4 ± 2.9	99.2 ± 5.3	.63
48 months	99.5 ± 3.1	99.7 ± 1.6	99.5 ± 3.3	1.00	99.7 ± 1.6	99.5 ± 2.7	.75

^a Statistically significant difference (p value <.05 versus baseline).

Data are presented as mean ± SD.

At the last follow-up, 559 (94.4%) tumors completely disappeared on CEUS before PSM, and 340 (93.7%) tumors completely disappeared after PSM, with 112 tumors (92.6%) and 228 tumors (94.2%) in groups A and D, respectively. The Kaplan–Meier curves in Figure 4 illustrate that the tumor disappearance rate between groups A and D was not significantly different (p = .58).

Figure 4. Graph illustrating Kaplan–Meier curves for tumor disappearance rate between the adjacent and distant groups after propensity score matching.

Disease progression

Disease progression was experienced by 22 patients, with a mean time of 31.5 months ± 17.3 (Supplementary Table 1). The overall incidence of disease progression was 3.7% (22/592), including 3.0% (18/592) new thyroid cancers, 0.3% (2/592) residual cancers, and 0.3% (2/592) LNM. The disease progression rate was comparable between groups A and D (4.1% vs. 3.6%, p = .67), and no significant difference was observed in the incidence of new thyroid cancer (3.3% vs. 3.0%, p = .71), LNM (0 vs. 0.4%, p = .66), and residual cancer (0.8% vs. 0.2%, p = .34) between the two groups (Table 3). the disease progression rate was 4.4% (16/363) with a mean time of 29.2 months ± 17.7 after PSM. No significant differences in disease progression were observed between the two groups (4.1% vs. 4.5%, p = .70), which was consistent with the findings before matching. Furthermore, no significant differences in progression-free survival curves were observed between the two groups after PSM, as illustrated in Figure 5. Additionally, as presented in Table 3, no statistically significant differences in the incidence of new thyroid cancer (3.3% vs. 3.7%; p = .67), residual cancer (0.8% vs. 0.4%; p = .68), or LNM (0% vs. 0.4%; p = .62) were discovered between the two groups. According to the data presented in Table 4, tracheal adjacency was not independently associated with disease progression, even after adjusting for confounding factors, including age, sex, tumor volume, HT, and CEUS mode, as indicated by the findings of the multivariable Cox regression analysis (all p > .05).

Figure 5. Graph displaying Kaplan–Meier curves for progression-free survival between the adjacent and distant groups after propensity score matching.

Table 3. Comparison of outcomes between adjacent and distant groups.

Outcomes	Before matching				After matching
	Adjacent group (n = 122)	Distant group (n = 470)	HR (95%CI)	p value	Adjacent group (n = 121)	Distant group (n = 242)	HR (95%CI)	p value
Disease progression	5 (4.1)	17 (3.6)	1.2 (0.5, 3.4)	.67	5 (4.1)	11 (4.5)	0.8 (0.3, 2.4)	.70
Residual cancer	1 (0.8)	1 (0.2)	3.8 (0.2, 61.5)	.34	1 (0.8)	1 (0.4)	1.0 (0.1, 32.0)	.62
New cancer	4 (3.3)	14 (3.0)	1.2 (0.4, 3.8)	.71	4 (3.3)	9 (3.7)	0.8 (0.2, 2.5)	.67
Ipsilateral lobe	0 (0)	4 (0.9)	0 (0, 1523.5)	.54	0 (0)	3 (1.2)	0 (0, 586.8)	.48
Contralateral lobe	4 (3.3)	10 (2.1)	1.7 (0.5, 5.6)	.35	4 (3.3)	6 (2.5)	1.1 (0.3, 4.0)	.86
LNM	0 (0)	2 (0.4)	0 (0, 88205.2)	.66	0 (0)	1 (0.4)	0 (0, 991395.1)	.68
Ipsilateral neck	0 (0)	2 (0.4)	0 (0, 88205.2)	.66	0 (0)	1 (0.4)	0 (0, 991395.1)	.68
Contralateral neck	0 (0)	0 (0)	…	…	0 (0)	0 (0)	…	…

CI: confidence interval; HR: hazard ratio; LNM: lymph node metastasis.

Data are presented as numbers (percentages).

Table 4. Univariate and multivariate analyses of risk factors affecting disease progression.

	Before matching				After matching
	Univariate analysis		Multivariate analysis		Univariate analysis		Multivariate analysis
Variables	HR (95%CI)	p value	HR (95%CI)	p value	HR (95%CI)	p value	HR (95%CI)	p value
Age (years)	1.0 (0.9, 1.0)	.35	1.0 (0.9, 1.0)	.40	1.0 (1.0, 1.1)	.98	1.0 (1.0, 1.1)	.84
Sex (Male)	0.9 (0.4, 2.2)	.64	0.7 (0.5, 3.9)	.53	0.2 (0, 1.8)	.16	3.7 (0.0, 2.1)	.21
Volume (mm^3)	1.0 (1.0, 1.0)	.91	1.0 (1.0, 1.0)	.97	1.0 (1.0, 1.0)	.95	1.0 (1.0, 1.0)	.99
CEUS (hyper/isoenhancement)	1.3 (0.3, 5.4)	.75	0.8 (0.3, 5.7)	.71	1.9 (0.2, 14.3)	.55	0.5 (0.3, 17.2)	.46
HT	0.8 (0.3, 2.4)	.46	0.7 (0.6, 3.4)	.45	2.0 (0.7, 5.3)	.18	0.6 (0.6, 4.5)	.35
Tracheal adjacency	1.2 (0.5, 3.4)	.67	0.8 (0.5, 3.5)	.63	0.8 (0.3, 2.4)	.70	1.2 (0.3, 2.4)	.73

CEUS: contrast-enhanced ultrasound; CI: confidence interval; HR: hazard ratio; HT: Hashimoto’s thyroiditis.

Data are presented as hazard ratios (95% CIs).

Among the 22 patients who experienced disease progression, one chose to undergo surgical resection, two selected AS, and 19 underwent additional RFA (Supplementary Table 1). The tumor sizes remained stable for the two patients who selected AS, and no signs of local or distant progression were detected during follow-up periods of 18.1 months and 11.9 months. Successful treatment without disease progression was achieved in all patients who opted for additional RFA or surgery. Of the 19 patients who underwent additional RFA, 15 patients (78.9%) exhibited complete disappearance of lesions during the follow-up period, and the remaining four patients (21.1%) had negative biopsy results of the ablation lesions after additional RFA.

Safety

None of the patients experienced life-threatening or permanent complications during or after RFA. Complications were observed in six patients (Table 5), including two cases of hoarseness (recovered within three months), two cases of cough, one case of moderate pain, and one case of hematoma (relieved within a week). In this study, the overall incidence of complications was 1.0% (6/592), including 1.6% (2/122) in group A and 0.9% (4/470) in group D (p = .79). The difference in the incidence of complications in groups A and D was insignificant after matching (1.7% vs. 0.8%, p = .86). Furthermore, there was no significant difference in the incidence of hoarseness, cough, moderate pain, or hematoma between the two groups before and after matching (all p > .05).

Table 5. Comparison of complications between adjacent and distant groups.

Complication	Before matching			After matching
	Adjacent group (n = 122)	Distant group (n = 470)	p value	Adjacent group (n = 121)	Distant group (n = 242)	p value
Total complication	2 (1.6)	4 (0.9)	.79	2 (1.7)	2 (0.8)	.86
Hoarseness	1 (0.8)	0 (0)	.21	1 (0.8)	0 (0)	.33
Cough	1 (0.8)	2 (0.4)	.50	1 (0.8)	1 (0.4)	1.00
Hematoma	0 (0)	1 (0.2)	1.00	0 (0)	1 (0.4)	1.00
Moderate pain	0 (0)	1 (0.2)	1.00	0 (0)	0 (0)	…

Data are presented as numbers (percentages).

Discussion

This study is the first to investigate the effectiveness and safety of RFA as a treatment modality for PTMC situated adjacent to the trachea. A comparison was made between the clinical outcomes of PTMCs in trachea-adjacent and trachea-distant locations. In this study, the technical success was 100%, and the complication rate was 1.0%. Over a mean follow-up period of 40.1 months ± 16.2, no significant differences in volume, VRR, cumulative tumor disappearance rate, or disease progression rate were discovered between PTMCs with trachea-adjacent and trachea-distant locations. Additionally, a trachea-adjacent location was not associated with disease progression based on the results of univariate and multivariate Cox regression analyses. Therefore, the findings of this study reveal the safety and mid-term effectiveness of RFA as an alternative option for PTMC situated adjacent to the trachea.

The optimal management strategy for low-risk PTMC remains controversial. Currently, management strategies available for low-risk PTMC include surgical resection, TA, and AS, and each strategy has pros and cons [29–31]. RFA, as a type of TA, is gaining attention as an increasingly attractive option, particularly for patients unsuitable for or reluctant to undergo surgical resection and those who experience significant anxiety about AS [12,14]. Several studies investigating the use of RFA for treating primary PTC have reported promising outcomes [17,32–34]. In a recent meta-analysis, the VRR and tumor disappearance rate for PTMC at 12 months after RFA were 93.3% and 64%, respectively. Furthermore, the incidences of residual cancer, new thyroid cancer, and LNM after RFA were 0.3%, 2.5%, and 1.0%, respectively [14]. In the current study, 592 patients with solitary PTMC who underwent RFA were investigated, with a follow-up time of 40.1 months ± 16.2. At the latest follow-up, the mean VRR was 99.5% ± 3.1, the cumulative tumor disappearance rate was 94.4%, and the disease progression rate was 3.7%. These results are consistent with those reported in previous studies [14,15,27,35], indicating that RFA could be an alternative treatment approach for low-risk PTMC.

Some researchers have questioned the feasibility of RFA for PTMCs adjacent to the trachea, stating the possibility of minimal ETE and an increased risk of disease progression in tumors touching the medial capsule surrounding the trachea. However, the eighth edition of the AJCC guidelines has excluded minimal ETE from the T-stage classification due to limited evidence of its impact on disease-related mortality and prediction of persistence/recurrence in PTC [6,36]. Additionally, previous studies have investigated the feasibility of ablation for PTMCs with ultrasound-visible minimal ETE or PTMCs situated in the isthmus, revealing favorable efficacy and safety [18,19,37]. Nevertheless, to date, no study has explored the outcomes of TA in PTMCs adjacent to the trachea. In this study, after controlling for confounding factors using PSM, RFA efficacy in PTMCs with trachea-adjacent and trachea-distant locations was compared. For PTMCs adjacent to the trachea, a cumulative tumor disappearance rate of 94.4%, and a VRR of 99.5% ± 3.1 were observed, with no significant difference found compared with PTMCs distant from the trachea. Furthermore, no significant differences in the incidence of new cancer, residual cancer, or LNM were discovered between PTMCs with trachea-adjacent and trachea-distant locations. In addition, Kaplan-Meier analysis revealed no substantial differences in progression-free survival between PTMCs abutting the trachea and PTMCs distant from the trachea. An analysis was conducted to assess whether a trachea-adjacent location could be a risk factor affecting disease progression using the Cox proportional hazards regression model. The results did not provide any evidence to demonstrate that a trachea-adjacent location influences disease progression after adjusting for factors such as age, sex, tumor volume, HT, and CEUS mode. These findings indicate that RFA may be an alternative treatment for patients with low-risk PTMC adjacent to the trachea.

An overall incidence of complications of 1.0% was revealed in this study, with 1.6% for PTMC situated adjacent to the trachea and 0.9% for those situated distant from the trachea. No significant difference in the incidence of complications was observed between PTMC adjacent to and distant from the trachea before and after PSM (all p > .05). Permanent recurrent laryngeal nerve injury was the most common complication observed after surgery, with prior studies reporting an incidence ranging from 2.3% to 3% [38,39]. In contrast to these results, no patient experienced any permanent complications after ablation, and only two patients experienced hoarseness, which subsequently resolved within three months. This might indicate that ablation-induced nerve damage was temporary and relatively moderate. A case of tracheal necrosis after RFA was reported by Morvan et al. in which a patient underwent RFA for a 38-mm thyroid nodule situated at the isthmus under general anesthesia. Tracheal necrosis was subsequently confirmed by MRI after ablation [40]. In our institution, local anesthesia is routinely administered to prevent damage to the surrounding structures and facilitate real-time patient feedback. In this study, the complication rates were slightly lower than those previously reported (ranging from 1.8% to 3.6%) [12,14,35,41]. We speculate that this difference could be attributed to several factors. First, conducting a comprehensive pre-ablation evaluation is crucial. This involves the use of US, CEUS, and CT to evaluate tumor characteristics and determine the appropriate ablation route. Second, it is imperative to employ a combination of various approaches to minimize thermal injury to the surrounding structures, especially when dealing with tumors situated adjacent to the trachea. The hydrodissection technique was implemented to increase the distance between the tumor margin and surrounding structures, with a further extension of this distance achieved by applying the leverage pry-off method. In addition, obtaining real-time feedback from patients can facilitate necessary adjustments and reduce the incidence of complications.

This study had several limitations. First, it was a retrospective study conducted at a single center, which may have introduced selection bias. Second, the retrospective nature of this study could have led to an underestimation of the incidence of complications. Third, the accuracy of identifying aggressive pathological subtypes of PTMC may not have been compromised since PTMC was diagnosed by biopsy. Additionally, the study sample size and follow-up time were insufficient for indolent tumors. Finally, there is a need for further large-cohort prospective studies conducted at multiple centers with extended follow-up periods.

In conclusion, our findings indicate that no significant difference in mid-term clinical outcomes was observed between PTMC located adjacent to and distant from the trachea. Thus, RFA may be an effective and safe alternative approach for managing eligible low-risk PTMC adjacent to the trachea.

Author contributions

Haoyu Jing: Conceptualization, Methodology, Formal Analysis, Writing – Original Draft and Review & Editing; Lin Yan: Methodology, Supervision, Writing – Review & Editing; Jing Xiao: Methodology, Writing – Review & Editing; Xinyang Li: Supervision, Writing – Review & Editing; Bo Jiang: Supervision, Writing – Review & Editing; Zhen Yang: Data Curation, Writing – Review & Editing; Yingying Li, Writing – Review & Editing; Bin Sun, Writing – Review & Editing; Mingbo Zhang: Conceptualization, Writing – Review & Editing, Supervision, Project Administration; Yukun Luo: Conceptualization, Writing – Review & Editing, Supervision, Project Administration.

Ethics statement

This study was approved by the Institutional Review Board of China PLA General Hospital (Number: S2019-211-01).

Patient consent

Written informed consent for study inclusion was waived because of the retrospective nature, and written informed consent for ablation treatment was obtained.

Statement

The article currently submitted to your journal is a revised version based on the previous paper submitted to the repository (DOI:10.21203/rs.3.rs-2937812/v1). We have added the author (B S) to the submission to IJH because the author has contributed to the revisions (including updating the follow-up time of enrolled patients and editing the paper with other authors) for the current paper.

Clinical trial registration

This study has been registered on the Chinese Clinical Trial Registry (ChiCTR2000039563).

Acknowledgments

The authors thank the patients who participated in this study. The paper has been posted in a repository (DOI:10.21203/rs.3.rs-2937812/v1).

Disclosure statement

The authors report there are no competing interests to declare.

Data availability statement

The data supporting the findings of this study are available from the corresponding author upon reasonable request.

Additional information

Funding

This research did not receive any specific grants from funding agencies in public, commercial, or not-for-profit sectors.