A comparative study of short-term efficacy and safety for thyroid micropapillary carcinoma patients after microwave ablation or surgery.

Background: Although papillary thyroid microcarcinoma (PTMC) has a high incidence and excellent clinical outcome, debate continues as to the therapeutic approach that would be most appropriate after confirming the diagnosis. Methods: We retrospectively analyzed the medical records of 311 patients with T1aN0M0 PTMC between January 2013 and September 2018. In all, 168 underwent microwave ablation (MWA), and 143 underwent surgery. MWA was performed using extensive ablation with hydrodissection. The surgery comprised thyroid lobectomy (TL) with unilateral central lymph node dissection (CND). We examined clinical outcomes during mean follow-up periods of 824 ± 452 days for the TL group and 753 ± 520 days for the MWA group. Results: Postprocedural follow-up revealed that, in the MWA group, the tumors had completely disappeared in 34 patients, and the remainder were reduced to necrotic or carbonized tissue. The incidence of transient hypoparathyroidism was significantly lower in the MWA group than in the TL group (p < .001). In addition, during the follow-up, we found no statistically significant differences between the two groups (TL vs MWA) for PTMC recurrence (1 vs 2 cases), lymph node metastasis (5 vs 5 cases), or disease-free survival [2001 days (5.5 years) vs 1702 days (4.7 years)] (p = .659, p = .795, and p = .974, respectively). Conclusions: If low-risk thyroid carcinoma (i.e., T1N0M0 PTMC) is accurately diagnosed early, MWA could be a minimally invasive alternative to surgery based on our short-term follow-up regarding recurrence and the low rates of complications and disease-free survival.


Introduction
The recorded incidence of papillary thyroid microcarcinoma (PTMC) has recently increased following widespread screening and technical improvements in thyroid ultrasonography (US) and fine-needle aspiration biopsy (FNAC) [1][2][3]. Despite this increased incidence, the mortality rate remains unchanged at 0.5 deaths per 100 000, although the longterm prognosis is usually excellent, with 10-year survival rates of up to 98% [4][5][6]. The American Thyroid Association (ATA) guidelines [7] recommend active surveillance, but many patients refuse this approach because of the possibility of lymph node and distant metastases. Currently, thyroid lobectomy (TL) is commonly used to treat PTMC [8]. A disadvantage of traditional surgery is the high rate of severe complications, with morbidity exceeding that of the disease itself [8]. Hence, the ideal PTMC treatment is still controversial [9].
With advances in medical technology, thermal ablation has emerged as a new strategy. The thyroid radiofrequency ablation guideline [10] reported that radiofrequency ablation, microwave ablation (MWA), and laser ablation effectively ablate local PTMCs over the short term and after a 4-year follow-up period. MWA provides a high intratumoral temperature, allows ablation of large volumes, and has a rapid ablation speed. It may also be far less invasive than surgery [11,12]. Many studies have confirmed that MWA treats solitary T1aN0M0 PTMCs effectively [13,14]. We also confirmed the effectiveness of MWA in an earlier study of thermal ablation versus surgery for PTMC [15].
The present study aimed to compare follow-up outcomes for MWA versus surgery, focusing on lymph node metastasis, recurrence, and postoperative complications.

Patients
The institutional review board of the Beijing Friendship Hospital at Capital Medical University approved the study, and informed consent was obtained from all patients before surgery or MWA.
Altogether, 220 PTMC patients were treated with USguided MWA between January 2013 and September 2018. Among them, 52 patients were excluded (32 were lost to follow-up, and 20 had incomplete information). The remaining 168 patients were enrolled in this study as group A (the MWA group). We also selected 143 patients without clinical or US evidence of neck lymph node metastasis (cN0) in chronological order for the surgical group (group B).
All enrolled patients fulfilled the following criteria: (1) solitary PTMC confirmed by US-guided FNAC; (2) presence of a solitary tumor measuring 10 mm along its greatest dimension; (3) tumor did not contact or distort the thyroid capsule; (4) no extra-thyroidal spread (to the trachea or esophagus); (5) no lymph node involvement identified on imaging or biopsy evaluation; (6) no history of neck irradiation. The MWA patients (group A) were ineligible for, or refused to undergo, surgery because they were either considered at too-high risk for thyroid surgery or for other reasons. The group B patients had requested surgery because of physical or psychological discomfort or neck extension disorders that made MWA intolerable. The exclusion criteria were (1) biopsy confirmation of a different carcinoma or a coexisting thyroid malignancy such as medullary carcinoma; (2) imaging examinations or biopsy that revealed lymph node or distant metastasis; (3) pregnancy; (4) presence of severe heart, respiratory, liver, or renal failure; (5) presence of a coagulation disorder associated with a severe bleeding tendency ( Figure 1).

Preoperative assessment for all patients
All included patients underwent a preoperative physical examination that involved high-resolution neck US examination of suspicious tumors or lymph nodes to confirm the number of lesions (single or multiple) (TI-RADS 4a) [16] in the thyroid gland, followed by FNAC. All patients were forbidden to take antiplatelet or anticoagulant medications for at least 1 week before surgery.
Exclusion and diagnostic standards of lymph node metastasis US findings of lymph node metastasis were as follows: round shape (length/short diameter ratio <2:1); unclear division of cortex and medulla; absent hilus; irregular margins; fusion tendency; peripheral halo; high-echo, low-echo, or cortical calcification [17]. Two doctors (LQ, XH) with more than 10 years of experience assessed the lymph node status. When more than two US images exhibited features of lymph node metastasis, we defined the lymph node as "suspicious," which was then confirmed by FNAC.

MWA instruments and procedures
Before the MWA procedure, each group A patient underwent US and contrast-enhanced US (CEUS) to determine the site, vascularity, and size of the tumor [18]. Two doctors (LQ, YL), each with more than 5 years of experience, performed the MWA using HI VISION Ascendus (Hitachi, Tokyo, Japan) equipped with a 5-to 12-MHz linear-array transducer. The patient was placed supine with the neck fully exposed and with continuous electrocardiographic, respiratory, and blood pressure monitoring. Routine sterile skin preparation was performed with 1% lidocaine as topical anesthesia. A mixture of 1% lidocaine and physiological saline solution was carefully injected into the thyroid capsule as hydrodissection to protect major structures (carotid vessels, trachea, recurrent laryngeal nerve, esophagus) from thermal injury [19].
MWA was performed using the moving-shot technique for multipoint and multidimensional ablation from the tumor's lower pole to its upper pole [15]. A 17-gauge ablation needle was inserted into the tumor's lower pole under US guidance. The tumor and adjacent thyroid gland were included in the ablated zone (5 mm) to prevent marginal recurrence. A power output of 30 W at 2450 MHz for 20-120 s was routinely used, inducing coagulation necrosis in the tumor. The therapy was sustained until the entire tumor was hyperechoic. When withdrawing the antenna, the needle track was coagulated to prevent tumor cell seeding. We spoke with the patients intermittently during the entire procedure to monitor their vocal status. At 1 h after MWA completion, CEUS was repeated to evaluate the ablated area.

Surgical procedures
Surgery for group B patients, performed under general anesthesia, was undertaken by general surgeons with 10 years of clinical experience. The surgical procedure comprised TL (143 patients) with unilateral CND, according to ATA guidelines [16].

Postprocedural follow-up
In the MWA group, the size, volume, vascularity, and characteristics of each tumor were carefully evaluated 1 h after ablation, at each follow-up point (1, 3, 6, 9, and 12 months), and over the years thereafter. Thus, all patients in all groups underwent neck US every 3-6 months postoperatively.
We defined locoregional recurrence as recurrence in thyroid tissue. Lymph node recurrence was defined as either cytological evidence of disease in the central or lateral neck compartment or evidence of disease on US. A diagnosis of recurrent PTMC was based on evidence of disease in the ablated area of the thyroid bed confirmed by FNAC and/or US. Distant metastasis was defined as evidence of disease outside the neck. Complications-transient or persistent hypoparathyroidism confirmed by serum calcium levels <8.5 mg/dl (2.12 mmol/l); symptoms of hypocalcemia; transient or persistent recurrent laryngeal nerve injury (presenting as voice change, hoarseness)-were carefully monitored after the procedures and at each follow-up assessment. If suspected, recurrent laryngeal nerve injury was confirmed by laryngoscopy.

Disease-free survival
Disease-free survival (DFS) following PTMC was defined as the duration of time from treatment until the lack of disease and/or recurrence was confirmed postoperatively [20]. Event occurrence was defined as postoperative lymph node metastasis and/or tumor recurrence [21].

Statistical analysis
Quantitative data measurements are described as means ± standard deviation (SD) and range. We compared data for the two groups using Student's unpaired t test. Categorical data comparisons were performed using the v 2 test or Fisher's exact test. Comparisons of changes in the mean diameter of each tumor before MWA and at each follow-up were evaluated using the analysis of variance repeated-measures test. DFS curves were determined using the Kaplan-Meier estimation and survival curves using the logrank test. All analyses were performed with SPSS statistical software (version 19; IBM, Armonk, NY, USA). p < .05 was considered to indicate statistical significance.

Postprocedural follow-up findings
The patients' characteristics are shown in Table 1. There were no statistically significant differences in age or sex between groups A and B (p ¼ .185 and p ¼ .804, respectively).

Therapeutic effect of MWA
At the time of this writing, 89.3% (150/168) of the MWA patients had undergone more than 1 year of follow-up. During that time, the tumors in 34 patients had completely disappeared. One patient's tumor disappeared as early as 3 months after treatment, with the others having completely disappeared at 6 months (eight patients), 9 months (seven patients), and 1 year (18 patients). The remaining patients' tumors were obviously unchanged 1 year later. CEUS confirmed no enhancement, and FNAC showed necrosis. Changes in the mean diameters after MWA at 1 h and 1, 3, 6, 9, and 12 months after MWA are shown in Table 2.

Complications
Transient hypoparathyroidism was recorded in 10 patients in the surgical group (7%). In contrast, no patients who underwent MWA developed permanent or transient hypoparathyroidism. Transient recurrent laryngeal nerve palsy was reported in 6 (4.2%) patients and permanent laryngeal nerve damage in 1 (0.7%) in the surgical group. The rates of transient and permanent recurrent laryngeal nerve injury in the MWA group were 3.6% and 0.6%, respectively. Hence, the complication rate was significantly lower after MWA than after surgery (p < .001). In addition, all patients in the surgical group required thyroid replacement therapy, whereas none of the patients in the MWA group did so (Table 3).  Figure 2 and Table 4.

Discussion
Thyroid carcinoma exhibits an interesting phenomenon in that, despite an increasing disease incidence, the associated morbidity and mortality rates remain stable. Thyroid cancer also generates much controversy regarding over-diagnosis and over-treatment. The ATA guidelines [16] support active surveillance of patients with low-risk PTMC, especially those who are at high surgical risk and/or are projected to have a short lifespan. Nevertheless, most patients considered to have maintained a stabilized condition prefer to eliminate the cancer. Along with active surveillance, surgery is therefore the most common, traditional treatment, and MWA is an emerging, minimally invasive approach. During the period 2008-2017, laser ablation [22][23][24], radiofrequency ablation [18], and MWA [13] were reported as treatments for low-risk PTMC. MWA is more effective than other ablation techniques because of its good therapeutic effect and rapid rate of temperature increase [11,19,25]. We reported a significant reduction in the mean lesion volume (from 81.60 ± 9.99 to 21.41 ± 4.02 mm 3 ) and complete disappearance of the ablation zones (22.7%, 34/150) during follow-up, indicating that MWA is an effective "local" treatment for PTMC.
In contrast, TL is a relatively "complete" surgical procedure for PTMC with the advantage of low recurrence, but it also carries a risk of injury to recurrent laryngeal nerves or parathyroid glands. Many patients experience postoperative hoarseness and hypoparathyroidism, thus requiring lifelong thyroid hormone replacement therapy. In contrast, MWA minimizes recurrent laryngeal nerve and parathyroid injury rates because of real-time US guidance and the use of the hydrodissection technique [19]. Our experience confirms the reported high rates (7%) of transient hypoparathyroidism after surgery (TL þ unilateral CND). With zero incidence of transient hypoparathyroidism in the MWA group, the difference between groups was statistically significant (p < .001). We found no statistically significant difference in recurrent laryngeal nerve injury between the two groups (p ¼ .527), although the rate was somewhat lower in the MWA group. Compared with RFA and PLA, MWA has the features of thermal conductivity, a faster rate of temperature increase, and powerful thermal injury [11,25]. Zhang et al. [18] reported 5.43% (5/92) as the overall RFA complication rate. In an MWA study, the rates of transient and permanent hoarseness (3.6% and 0.6%) were higher in the present study than the hoarseness rate (2.7%) reported by Teng et al. [26] but lower than the 19% rate found by Yue et al. [13]. Korkusuz et al. [27] reported median pain scores of 3 and 4 points during RFA and MWA, respectively, and their rates of mild hematoma after RFA and MWA treatment were 47.3% (26/ 55) and 44.7% (21/47), respectively. The complications in the present study were similar to those in other studies [27,28].
Recurrence is of greater concern than postoperative complications in patients with thyroid carcinoma. US is the most frequently recommended imaging modality for both preoperative and postoperative evaluations of the thyroid gland and lymph nodes [29]. All patients in our study underwent preoperative and postoperative follow-up US evaluations, with FNAC added to address suspicious lesions (to confirm recurrent disease).
The present study found that 4.2% (7/168) and 4.2% (6/ 143) of patients in the MWA and surgical groups, respectively, developed lymph node metastasis and PTMC recurrence, with similar rates of neck recurrence in the two groups. We speculated that it resulted from an antitumor immune response activated by thermal ablation [30]. Some studies have suggested that thermal ablation can induce a so-called abscopal effect-i.e., spontaneous regression of a remote, non-targeted tumor [31,32]. In our study, recurrent tumors found after MWA or surgery were successfully treated by repeat MWA. Repeated operations are usually challenging for recurrent carcinoma in the thyroid bed and cervical lymph nodes because of distortion of normal tissue planes and fibrosis caused by scar tissue. Thus, the 2017 Thyroid Ablation Guideline [10] recommends ablation to avoid complications. Some studies [33,34] had reported use of MWA to treat recurrent thyroid carcinoma or metastatic lymph nodes.
Given the high disease-specific survival and lack of disease-specific mortality among the treatment groups, the need to perform more radical neck surgery remains controversial [1,8,35]. In our study, there was no disease-specific mortality or distant metastasis in either treatment group. Therefore, when considering the short-term survival status and lack of complications, simply eliminating the tumor may be the better approach. Based on our results and previous reports [11,12,14,15,19], MWA appears to be a good alternative to surgery in low-risk patients (i.e., those with T1N0M0 PTMCs).
Our study has several limitations. First, the sensitivity of US to detect lymph node metastases is relatively low (23-72%), although its specificity is 69-92% [29,36,37]. The main reason is that region VI is a blind area for US [38]. Second, five patients in the surgical group had multifocal disease according to the final pathology, which showed small (1-2 mm) and occult (<2 mm) carcinomas that could easily be missed by US [39]. Thus, the conventional examinations (US, FNAC) could not evaluate a solitary tumor with great validity. Finally, the follow-up time is short (1.0-3.5 years), and the number of cases is small.
In conclusion, MWA is superior to surgery in terms of the appearance of complications and its minimal invasiveness. As a technique that could accurately assess lymph node and lesion status preoperatively, and on the basis of our shortterm follow-up for recurrence and the DFS rates, MWA may be considered an alternative to surgery for low-risk patients with a PTMC. Above all, the most appropriate treatment for PTMC has yet to be determined.  MWA  18  59  34  33  24  0  168  Surgery  25  49  27  30  10  2  143 MWA: microwave ablation.