Ultrasound-guided percutaneous thermal ablation of parotid tumors: experience from two-centers

Abstract

Objective: To evaluate the efficacy and feasibility of ultrasound-guided percutaneous thermal ablation (TA) for treating benign parotid tumors.

Methods: Patients with benign parotid tumors who underwent ultrasound-guided microwave ablation (MWA) or radiofrequency ablation (RFA) between January 2020 and March 2023 were included in this retrospective study. Change in tumor size (maximum diameter, tumor volume(V), volume reduction rate (VRR)) and cosmetic score (CS) were evaluated during a one-year follow-up period. We also recorded the incidence of any complications associated with TA.

Results: A total of 23 patients (13 males and 10 females; median age 65 years, range 5–91 years) were included. The mean VRR at 1, 3, 6, and 12 months after TA was 37.03%±10.23%, 56.52%±8.76%, 82.28%±7.89%, and 89.39%±6.45%, respectively. Mean CS also changed from 3.39 ± 0.66 to 1.75 ± 0.93 (p < 0.001) by the end of follow-up time. Subgroup analysis showed that tumors with smaller initial maximum diameter had a faster CS reduction rate than those with larger initial diameter. The incidence of facial nerve dysfunction was 8.70%.

Conclusion: Ultrasound-guided percutaneous TA is an effective and safe treatment option for patients with benign parotid tumors.

Keywords:

• Parotid neoplasms
• ultrasonography
• interventional
• ablation techniques
• microwave ablation
• radiofrequency ablation

Introduction

Parotid gland tumors mainly include pleomorphic adenoma, Warthin tumor, basal cell adenoma, and ductal papilloma, etc [1]. Although these tumors are pathologically benign, some have the potential for recurrence or malignant transformation [2]. Therefore, timely treatment is essential.

Currently, surgery is the mainstay for management of the tumors. To avoid damage to important structures such as the facial nerve, surgeons may opt for one of several surgical methods according to the location, size, depth, and relationship of the tumor with the facial nerve [3–5]. However, there is still concern regarding postoperative complications, including facial nerve injury, Frey’s syndrome, sialocele, salivary fistula, and bleeding [6]. The incidence of those complications increases with the extent of the surgical scope [3], which makes large tumors more difficult to treat. In addition, recurrence caused by enucleation with rupture of the tumor capsule and incomplete tumor excision during surgery is a risk [5]. Postoperative scarring can also affect the patients’ quality of life.

Ultrasound (US)-guided percutaneous thermal ablation (TA) includes microwave ablation (MWA), radiofrequency ablation (RFA), and laser ablation (LA). It is a minimally invasive local therapy that can induce irreversible injury by using the energy-tissue interaction generated around an inserted applicator to heat the target cells adequately [7]. Compared with traditional surgery, TA can eradicate tumors with minimal effect on normal tissues. To date, it has mainly been used to treat liver, thyroid, breast, and uterine leiomyoma [8–13]. With the continuous expansion of the scope of application, we explored the use of palliative ablation for head and neck tumors [14,15], which can alleviate symptoms and improve the quality of life of patients with advanced head and neck malignancies and early-stage tongue cancer.

Several studies have been conducted on the application of RFA for Warthin’s tumors, which showed promising volume reduction rates and cosmetic-scale improvements without facial nerve injury. However, the sample size was small and other tumor types were not assessed [16,17]. The objective of the present study was to assess use of US-guided percutaneous parotid gland TA for the treatment of parotid tumors.

Methods

Patients

This retrospective study was approved by the Ethics Committee of Sichuan Cancer Hospital (Ethical batch No. Js-2016-011). After the procedure and risks were described in full, each patient voluntarily chose the treatment plan by signing an informed consent form.

We retrospectively included patients based on the following criteria: (1) Preoperative confirmation that all tumors were diagnosed as benign by core needle biopsy. (2) No suspicious malignant signs on US images. (3) The patient refused surgery or could not tolerate surgery but was willing to undergo TA. (4) Follow-up ≥ 6 months. The exclusion criteria were as follows: (1) Severe cardiopulmonary disease and/or a severe blood coagulation disorder. (2) Pathology confirmed malignancy. (3) Follow-up time of < 6 months.

For subgroup analysis, we divided the patients into three groups based on tumor size. Group 1, group 2, and group 3 included tumors with an initial maximum diameter ≤2cm, 2-4 cm, and >4 cm, respectively.

Pretreatment assessments

All procedures were performed by two interventional radiologists (L. M. and, S. S. W.) with at least 8 years of experience in TA.

Before the operation, all patients underwent US examination, computed tomography (CT), or magnetic resonance imaging (MRI). Two-dimensional US, color Doppler, and contrast-enhanced US(CEUS) were used to evaluate tumor’ size, location, and internal blood supply to guide the next steps of treatment. CT or MRI was used to evaluate the overall condition of the tumor and parotid glands.

Ablation procedure

A Philips EPIQ7 (Philips Healthcare) equipped with L12-5 and eL18-4 probes was used as the ultrasonic diagnostic instrument. The patient was placed in the supine position, with the head tilted toward the healthy side, fully exposing the preauricular region. After determining the puncture point, the operator injected a local anesthetic layer by layer. A suction biopsy needle (22-gauge) was used for hydrodissection with saline injected around the parotid tumor under US guidance to form a separation zone of at least 2 mm thickness to protect the surrounding normal parotid gland tissues and nerves (Figure 2).

MWA was performed using an MWA system (KY2000; Kangyou Medical Instruments) equipped with an MWA generator and a 16-gauge microwave antenna with a 3-mm active tip and 10-cm shaft length. The ablation, with a power of 30 W, began after implanting a microwave antenna into the target location under US guidance. Normal saline was continuously injected during ablation to maintain the separation zone and protect the surrounding tissues. For patients with cystic tumors, the surgeon aspirated the cystic fluid using a 5 ml syringe and then carried out routine ablation (Figure 3). When the hyperechoic area on US completely covered the target area, ablation was paused, and CEUS was performed to evaluate whether the ablation was complete. The lesions that were completely ablated showed no contrast agent perfusion. If there was evidence of incomplete ablation, the surgeons conducted supplementary ablation until there was no more contrast agent in the lesion on CEUS.

RFA was performed using an RFA system (VIVA; STARmed) equipped with an 18-gauge RF generator with a 7-mm active tip and 7-cm shaft length monopolar electrode. As the high electrical conductivity of saline solution would affect the RFA [18], we used 5% glucose solution for hydrodissection. All other steps of the RFA procedure were consistent with those described for MWA.

Technical feasibility was defined as the ability to target the tumor and carry out the ablation as preoperatively planned. Technical success was defined as complete ablation at the end of every procedure.

Post-treatment assessments

Patients were followed up at the first, third, and sixth months after ablation and then every six months. At each follow-up, we used conventional two-dimensional US to evaluate the size of the ablated area and compared it with the preoperative images to evaluate the reduction. Color Doppler imaging was also used to assist in the diagnosis of neovascularization in the ablated area. We performed CEUS in some patients in whom there was suspected residual lesion based on comparison to preoperative imaging and color Doppler results. Changes in CS [19] were also recorded at follow-up. CS was scored as: 1 = no palpable mass; 2 = no visible but palpable mass; 3= the mass is only visible when gritting teeth; and 4= the mass is easily visible.

Image interpretation

All US images and medical records were reviewed by a single, independent radiologist at each center (M. Z., Y. C. L.) with over 6 years of experience in parotid radiology.

For each tumor, the maximal diameter (a) and two orthogonal diameter lines perpendicular to the maximum diameter of each tumor were drawn (b and c) before the procedure and at each follow-up visit. Tumor volume (V) was calculated according to the formula: V = πabc/6. VRR was calculated by the formula: VRR = (initial volume-final volume)/initial volume× 100%.

Statistical analysis

Statistical analysis was performed using SPSS software (SPSS 26.0, IBM). Measurement data are expressed as mean ± standard deviation and were analyzed using Student’s t-test or one-way analysis of variance (ANOVA). Enumeration data are expressed in frequencies and were analyzed using a non-parametric test. Statistical significance was set p < 0.05.

Results

Patient information

From January 2020 to March 2023, a total of 26 patients underwent US-guided percutaneous TA of parotid tumors in Sichuan Cancer Hospital and Fujian Provincial Hospital. After excluding 3 patients with malignant histopathology, 23 patients were finally included in this study. A flowchart of this process is shown in Figure 1. The baseline characteristics of the 23 patients are summarized in Table 1. The median age was 65 years (range 5–91years). There were two(8.70%) patients younger than 10 years and twelve (52.17%) patients older than 65 years. Six (26.09%) patients underwent RFA, and seventeen (73.91%) patients underwent MWA. Sixteen (69.60%) patients had Warthin tumor, five (21.74%) patients had pleomorphic adenoma, one (4.35%) patient had basal cell adenoma, and one (4.35%) patient had hemolymphangioma. The mean follow-up time after TA was 10.17 ± 2.82 months.

Figure 1. Study flowchart.

Figure 2. Hydrodissection of parotid tumor under ultrasound guidance. A, B Before TA, injected normal saline or 5% glucose solution around the parotid tumor under the guidance of ultrasound to form a separation zone about 2 mm thick. C, D during TA, continuously injected normal saline or 5% glucose solution to maintain the separation zone. E schematic diagram of hydrodissection. The yellow arrows showing the 22-gauge biopsy needle, the white arrows showing the ablation antenna, and the red dotted lines showing the separation zone.

Figure 3. A 75-year-old woman who had Warthin tumor for over 2 years suddenly increased in 1 week. A conventional ultrasound showed a cystic and solid tumor in the left lobe of the parotid gland. B CEUS showed blood perfusion in solid components and septations. C use a 5 ml syringe needle to draw out the fluid in the cystic before TA. D place the active tip of the ablation needle at the distal end of the tumor and then perform sectorial ablation from the bottom to the top until the hyperechoic gas completely covers the tumor. E after ablation, performed CEUS again, showing no obvious contrast agent perfusion in the tumor. F the aspirated fluid was purulent, and the pathology suggested that acute and chronic inflammatory cells were involved. The yellow arrow indicated the edge of the tumor; the white arrow in figures C and D indicated the 5 ml syringe needle and ablation antenna, respectively.

Table 1. Baseline characteristics of patients included in our study.

Characteristics	n (%)	Characteristics	n (%)
age (year)		location
	65 (35 ∼ 72)	S + D	2 (8.70%)
sex		S	21 (91.30%)
male	13 (56.52%)	side
female	10 (43.48%)	left	12 (52.17%)
maximum diameter (cm)		right	11 (47.82%)
	2.76 ± 1.61	facial nerve dysfunction
volume (cm^3)		+	2 (8.70%)
	3.47 (0.62-9.02)	–	21 (91.30%)
pathology		ablation method
Warthin tumor	16 (69.57%)	MWA	17 (73.91%)
pleomorphic adenoma	5 (21.74%)	RFA	6 (26.09%)
basal cell adenoma	1 (4.35%)	follow-up time (months)
hemolymphangioma	1 (4.35%)		10.17 ± 2.82

S: superficial lobe; D: deep lobe; VRR: volume reduction rate; MWA: microwave ablation; RFA: radiofrequency ablation.

Changes in tumor size and CS

All tumors were successfully ablated without local recurrence, and no patients required additional ablation. Before TA, the mean maximum tumor diameter was 2.76 ± 1.61 cm, and the mean V was 3.47(0.62-9.02) cm³. During the follow-up period, tumor maximum diameter and V gradually decreased, and the VRR at 1, 3, 6, and 12 months after TA was 37.03%±10.23%, 56.52%±8.76%, 82.28%±7.89%, and 89.39%±6.45% respectively. At the end of the follow-up period, two(8.70%) tumors had completely disappeared, and the VRR of nine(39.13%) tumors reached more than 90%. The change in tumor size (including maximum diameter and V) and VRR are presented in Table 2. Figure 4 shows US images of two representative patients before, and 6 months after ablation.

Figure 4. Ultrasound images of patients before and after TA. A, B a 91-year-old man with a cystic Warthin tumor in the right parotid gland. Figure A showed the ultrasound image before ablation, with a tumor size of 6.3 × 5.2 cm, Figure B showed the image 6 months after ablation, with the tumor size reduced to 2.7 × 1.5 cm. C, D a 51-year-old man with a Warthin tumor in the left parotid gland. Figure C showed the ultrasound image before ablation with a tumor size of 2.6 × 1.7 cm, Figure D showed the image 6 months after ablation, with the tumor size reduced to 1.9 × 1.1 cm. Yellow and white arrows indicated the edge of tumors.

Table 2. Tumor V and VRR before TA and at each follow-up time-point after TA.

Follow-up time	The maximum diameter (cm)	p value	V (ml)	p value	VRR (%)	p value
Before TA (n = 23)	2.76 ± 1.61	NA	3.47 (0.62-9.00)	NA	NA	NA
After TA
1-month (n = 23)	2.40 ± 1.45	0.432	2.17 (0.47-5.87)	0.394	37.03 ± 10.23	NA
3-month (n = 23)	2.13 ± 1.25	0.144	1.61 (0.29-3.96)	0.128	56.52 ± 8.76	0.000
6-month (n = 23)	1.51 ± 1.01	0.003	0.98 (0.09-1.62)	0.002	82.28 ± 7.89	0.000
12-month (n = 16)	1.47 ± 1.01	0.007	0.74 (0.09-1.28)	0.002	89.39 ± 6.45	0.000

p Value of the maximum diameter and V: after TA (1, 3, 6, or 12 months) vs. before TA.

p Value of VRR: after TA (3, 6, or 12 months) vs. 1 month after TA.

NA: not applicable; V: volume; VRR: volume reduction rate; TA: thermal ablation.

Before ablation, the mean CS was 3.39 ± 0.66. It gradually decreased with longer follow-up time, and at 12 months, the mean CS was 1.75 ± 0.93 (Figure 5). There were three (13.04%) patients whose CS did not show a significant decrease. They all had large primary tumors with a maximum diameter of more than 5.0 cm.

Figure 5. Changes in mean CS at each follow-up time point. A CS changes in all patients. B CS changes in subgroup analysis.

To further determine the relationship between tumor size and treatment effect, we divided the tumors into 3 subgroups according to the initial maximum diameter as shown in Table 3. Twelve months after ablation, the improvement in CS was highest in tumors with initial maximum diameter < 2 cm, followed by those with 2 and 4 cm and > 4 cm (p < 0.05) (Table 3 and Figure 5). The change in VRR showed no statistically significant difference between subgroups.

Table 3. Comparison of CS and VRR of different tumor diameters.

Tumor diameters
	≤2cm	2-4cm (n = 9)	>4cm (n = 4)	p value
CS
Initial	3.10 ± 0.74 (n = 10)	3.44 ± 0.53 (n = 9)	4.00 ± 0.00 (n = 4)	0.050
1months later	2.70 ± 0.95 (n = 10)	3.22 ± 0.67 (n = 9)	3.75 ± 0.50 (n = 4)	0.088
3months later	1.80 ± 0.63* (n = 10)	2.67 ± 0.50* (n = 9)	3.25 ± 0.96 (n = 4)	0.002
6months later	1.10 ± 0.32* (n = 10)	1.89 ± 0.33* (n = 9)	2.75 ± 1.25 (n = 4)	0.000
12 months later	1.00 ± 0.00* (n = 6)	1.83 ± 0.41* (n = 6)	2.75 ± 1.25 (n = 4)	0.004
VRR (%)
Initial	–	–	–
1months later	36.68 ± 8.72 (n = 10)	35.33 ± 12.41 (n = 9)	41.75 ± 9.47 (n = 4)	0.600
3months later	56.16 ± 6.22^# (n = 10)	54.97 ± 11.76^# (n = 9)	60.93 ± 6.62^# (n = 4)	0.510
6months later	84.84 ± 8.41^# (n = 10)	80.43 ± 9.14^# (n = 9)	83.70 ± 9.08^# (n = 4)	0.640
12 months later	91.58 ± 6.41^# (n = 6)	86.15 ± 6.00^# (n = 6)	90.95 ± 6.70^# (n = 4)	0.320

*p < 0.05 compared with initial cosmetic scores.

^#p < 0.05 compared with VRR 1 months later.

CS: cosmetic score; VRR: volume reduction rate.

Safety

No serious complications occurred after ablation. All patients developed edema at the ablation site after TA recovered within 3–5 days. Two (8.70%) patients experienced facial nerve dysfunction after TA, which presented as the nasolabial fold of the affected side becoming shallow, failure to close the eyelids, and askew mouth. Both patients had recovered by 3 months (Figure 6).

Figure 6. An 8-year-old girl with pleomorphic adenoma of the right parotid gland. A, B she had an askew mouth and couldn’t close the right eyelids one week after TA. C two months after TA, the patient recovered as usual. D Preoperative ultrasound images indicated a tumor at the right parotid gland. E Performed hydrodissection before TA. F Place the active tip of ablation needle at the distal end of the tumor, and then perform sectorial ablation from the bottom to the top. G 3 years after TA, the tumor had shrunk by more than 95%. H No obvious abnormal blood flow signal found in the right parotid gland. J No obvious blood flow signal found in the tumor. Yellow arrows indicated the edge of tumor; red arrows indicated separation zone; white arrows in figure E and F indicated the biopsy needle and the ablation needle respectively.

Discussion

This retrospective study found that TA is an effective method for treating benign parotid tumors, with no serious complications. By 12 months post-treatment, the tumor had reduced by almost 90% and in two patients, it was completely gone. Additionally, there were significant improvements in CS in most patients. These results suggest that TA is a highly effective treatment that results in patient satisfaction. Subgroup analysis showed that TA provides better cosmetic improvement for patients with smaller initial tumors but there was no difference in VRR, indicating that efficacy of the treatment is not affected by tumor size.

According to the European Salivary Gland Society [4], surgery is the mainstay for treating parotid tumors. In order to reduce the occurrence of complications, surgeons can choose different surgical methods according to the specific conditions of patients [3,5], however, there is controversy surrounding how to balance facial nerve integrity while providing efficacious resection [20]. The literature reports that 8% to 74% of patients who undergo surgery have transient facial nerve dysfunction [2,6,21,22]. In this study, two patients (8.70%) with 2 cm tumors had facial nerve dysfunction. This is higher than the incidence reported by others [16,17]. A possible cause is that the tumor in these patients grew too close to the facial nerve and local heat damage from the ablation may have temporarily damaged the nerve, despite the use of hydrodessection. Hydrodissection to establish a separation zone can effectively protect the recurrent laryngeal nerve from heat injury; however, if the thickness of the separation zone is insufficient, the nerve may be injured [18,23]. Importantly, the dysfunction resolved within three months.

TA has brought new possibilities for tumor treatment as it can accurately kill tumor tissues while preserving normal tissues and limiting the risk of vessel rupture. Compared to surgery, TA can effectively reduce complications in the treatment of various tumors [12,13,24]. According to the Korean Society of Thyroid Radiology, TA can be considered in patients with benign or recurrent thyroid cancers [9], and for malignant head and neck tumors located near major vessels [14].

However, few studies have assessed its efficacy in treating parotid tumors [16,17]. Cha et al. found that the mean VRR of Warthin tumors treated with RFA was 86.7% at 6 months [17], and good local control was achieved during long-term follow-up [16]. The present study yielded similar results with a 12-month VRR of 89.39%±6.45%. In our study, we included patients from two centers: one center used RFA, and the other center used MWA. The ablation process was successful in all patients.

To achieve complete ablation, reduce the local recurrence rate, and preserve the facial nerve, we used additional measures, including CEUS, to ensure the safety and effectiveness of the TA procedure. CEUS has been widely used in TA with great value before, during, and after the TA procedure [14,25–28]. In our study, CEUS allowed visualization of the blood supply to and surrounding the tumor, which helped us estimate the ablation range and the appropriate approach. CEUS also allowed us to determine whether the lesion was completely ablated. We also used hydrodissection to protect healthy tissues around lesions by injecting either saline or glucose between important structures close to the target tumor [18]. Hydrodissection protects delicate structures from heat damage and has been shown to provide a greater safety margin during thyroid nodule ablation [29,30].

The youngest patients included in our study were 5 years old, and the oldest was 91 years old. No serious complications occurred after TA, and no recurrence was found during the follow-up period. This shows that US-guided TA is safe and has great potential for the treatment of parotid tumors. For children, minimally invasive TA avoids leaving facial scars, and for the elderly, especially those with comorbidities, TA provides an effective and low-risk treatment option compared to surgery.

Our subgroup analysis showed that the improvement in CS and reduction in VRR was larger for smaller tumors, which has been reported by others for thyroid nodules [31,32]. For tumors with an initial maximum diameter greater than 4 cm, the decrease in CS after ablation showed no statistically significant difference from that before ablation. This may be because the tumor was too large to completely absorb over a short time period. Although these patients still showed facial asymmetry, their cosmetic condition had obviously improved, and would likely continue to improve as the ablation area continued to be absorbed. For these patients, using two or multiple ablations may reduce tumor burden more safely. For example, three months after the first TA, when the ablation area has significantly shrunk, TA can be performed again to further reduce tumor volume while maintaining patient safety by not ablating more tissue at the initial procedure. We will continue to explore the clinical application of this approach in future studies.

This study has several limitations. First, although our study was performed in two centers, the number of patients included is limited, and there is a lack of long-term follow-up data. Second, as a retrospective study, selection and data collection biases could not be avoided. Third, we did not directly compare outcomes with patients undergoing traditional surgery, which limits our ability to demonstrate the superiority of TA. In the future, multicenter, prospective, and large-sample studies are needed to verify the effectiveness of TA in the treatment of parotid tumors.

In conclusion, this study showed that during short-time follow-up, US-guided TA is safe and effective for the treatment of benign parotid tumors. CEUS and hydrodissection can effectively assist in the ablation procedure, and earlier interventions for patients can achieve a better cosmetic effect.

Acknowledgments

We thank all participants recruited for this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data presented in this study are available on request from the corresponding authors.

Additional information

Funding

This work was supported by National Key R&D Program of China (Grant No. 2019YFE0196700) and National Natural Science Foundation of China (Grant No. 82272015).