Risk Factors for Neck Nodal Metastasis in Papillary Thyroid Microcarcinoma: A Study of 1066 Patients

Context:

The surgical management of papillary thyroid microcarcinoma (PTMC), especially regarding the necessity of central lymph node dissection, remains controversial.

Objective:

The objective of the study was to describe the clinicopathological features of PTMC and to identify the risk factors for central lymph node metastasis (CLNM) that can guide surgical strategies for patients with PTMC.

Design:

In this retrospective cross-sectional study, risk factors and outcome variables were assessed at the time of surgery for the primary tumor.

Setting:

The study was conducted at a university-based tertiary care cancer hospital.

Patients:

Data from the medical records of 1066 consecutive patients diagnosed with PTMC over a 5-yr period were analyzed.

Results:

Our multivariate logistic regression analysis found male gender, younger age (≤45 yr of age), multifocal lesions, extrathyroidal extension, and larger size of the primary tumor (>6 mm) to be associated with CLNM; multifocal lesions were associated with the highest risk (odds ratio 4.476, 95% confidence interval 2.975–6.735). Extrathyroidal extension, multifocal lesions, and CLNM were associated with lateral neck lymph node metastasis (LLNM). In patients with a solitary primary tumor, tumor location in the upper third of the thyroid lobe was associated with a lower risk of CLNM and a higher risk of LLNM.

Conclusions:

Prophylactic central lymph node dissection need be considered in PTMC patients presenting with risk factors. In PTMC patients with a solitary primary tumor, tumor location can assist in the evaluation of LLNM. We recommend multicenter research and long-term follow-up to better understand the risk factors and surgical management of PTMC.

Papillary thyroid microcarcinoma (PTMC) is defined as a papillary thyroid carcinoma (PTC) 10 mm or smaller in diameter (1). Due to the recent widespread use of ultrasound (US) and US-guided fine-needle aspiration (FNA), increasing numbers of patients are preoperatively diagnosed with PTMC without palpable thyroid nodules (2–4). Although PTMC has an indolent course, the incidence of central lymph node metastasis (CLNM), which is 24.1–64.1% for PTMC (5–13), is not low compared with the incidence of CLNM in PTC.

Central lymph node dissection (CLND) and total thyroidectomy (TT) currently constitute the common initial surgical management of patients with larger PTC, with or without clinically apparent lymph nodes. This approach is thought to reduce locoregional recurrence and improve disease staging (14, 15). However, the role of CLND in PTMC remains uncertain because no evidence has demonstrated that CLND improves locoregional control or survival in PTMC (16–18). Wada et al. (5) compared the recurrence rate of 235 patients with PTMC who underwent prophylactic neck dissection (185 CLND + lateral neck dissection, 50 CLND only) with that of 155 patients with incidental PTMC who did not undergo neck dissection. After a 60-month follow-up, the recurrence rate was 0.43% (one of 235) for the dissection group and 0.65% (one of 155) for the no-dissection group. No statistical significance was observed. Therefore, the role of prophylactic CLND in PTMC is still and will remain controversial until more evidence emerges. Furthermore, the potential benefits of prophylactic CLND should be weighed against the associated complications, such as hypoparathyroidism and recurrent laryngeal nerve injury. However, better knowledge about the risk factors for CLNM may guide clinical decisions regarding which cases require CLND.

Multiple retrospective studies have described the clinical features of PTMC and identified several predictive factors for CLNM. Size and location of the primary tumor, which can be measured relatively objectively, were two essential characteristics analyzed by Wada et al. (5). However, their results were inconsistent with later studies (6, 9, 12). Therefore, we investigated the frequency, pattern, and predictive factors for CLNM in PTMC using a large group of Chinese patients.

Materials and Methods

Patients

The study subjects were patients all treated at the Department of Head and Neck Surgery, Cancer Hospital of the Fudan University, Shanghai, China, from January 2005 to December 2009. All patients provided written informed consent for their information to be stored in the hospital database and used for research, and this study was approved by the Ethical Committee of Cancer Hospital of Fudan University. The patients included in the study met the following criteria: no previous thyroid surgery and the availability of an adequate medical history. Patients with incidental PTMC found during thyroidectomy for benign conditions (without neck dissection) were excluded.

Surgical treatment

Before surgery, each patient underwent a US examination. US-guided FNA was not performed routinely. Patients with lymphadenopathy in the central or lateral neck, as detected by palpation or US, underwent a computed tomography (CT) scan of the neck as well as US-guided FNA of the suspected lymph node.

The following three criteria were used for selecting patients before surgery: 1) suspicious US images, including microcalcifications, hypoechoic, increased nodular vascularity, infiltrative margins, and length greater than width on the transverse view, and if more than one nodule was found by US, each nodule in the entire thyroid gland was evaluated by a clinical physician; 2) changes in the US images during the 3- to 6-month follow-up; and 3) patient consent.

From January 2005 to December 2009, 5518 patients with primary lesions 1 cm or smaller in diameter underwent US examinations at our hospital. Using our selection criteria, 1440 patients underwent initial surgery in our department. Of these 1440 patients, 1124 were diagnosed with PTMC histologically, and the remaining patients were diagnosed with benign tumors.

Lobectomy plus ipsilateral CLND was typically performed as the initial surgical treatment for PTMC patients with malignant lesions that were limited to a single lobe. The histology of the frozen sections (FS) guided the extent of the surgical procedures in these patients. Initially, the patients underwent a lobectomy according to US results; then if the nodule or nodules were found to be malignant by FS, an ipsilateral CLND was performed. When a benign or undetermined nodule was detected in the contralateral lobe by US, a subtotal lobectomy was performed in our hospital, including nodule enucleation, which was performed on the suspicion of a lesion in the contralateral lobe after preoperative US or partial lobe resection, in which approximately one fourth to two thirds of the contralateral lobe was resected on the suspicion of more than one lesion in the contralateral lobe after a preoperative US. When malignant lesions were found in both lobes of the thyroid by FS, a TT plus a bilateral CLND was performed.

A modified lateral lymph node dissection (LLND), including levels II–V, with the preservation of the sternocleidomastoid muscle, internal jugular vein, and spinal accessory nerve, was performed only in cases with clinically evident lateral neck lymph node metastasis (LLNM). The cervical compartment was defined using the Memorial Sloan-Kettering Cancer Center's classification of cervical lymph node regions (levels II–VI) (21).

Follow-up

All patients received TSH-suppressive hormonal therapy after surgery. Radioactive iodine therapy was limited to patients who had distant metastasis because of its strictly controlled use in China. Radioactive iodine therapy was not routinely prescribed for PTMC patients after surgery. Postsurgical physical examinations were performed every 3–6 months. During the follow-up visits, all patients underwent US examinations of the neck. A CT scan or an FNA was performed to evaluate suspected recurrences in the thyroid bed and lateral neck. Locoregional recurrence was diagnosed by US or CT plus cytological examination when necessary.

Clinicopathological variables assessed

The following variables were used for the analysis of risk factors for lymph node metastasis: gender, age at diagnosis, maximal microscopic tumor size [the largest diameter; for multifocal lesions, including bilateral lesions, the maximal microscopic tumor size of the largest lesion was used, as outlined in the American Joint Committee on Cancer seventh edition (22)], primary tumor extrathyroidal extensions, multifocal/solitary lesions, bilateral lesions(indicating that both lobes of the thyroid were affected by the malignant lesion), and the histological subtype of the PTC. Central lymph node status was analyzed as a binary variable to indicate whether any malignant lymph nodes were detected. Multifocal primary lesions were defined as two or more cancer sites within the thyroid; a solitary intrathyroidal primary lesion represented only a single cancer site within the thyroid. Extrathyroidal extensions of the primary tumor consisted of gross extrathyroidal extensions and microscopic extrathyroidal extensions. The multifocal and extrathyroidal extensions were both assessed in the final pathology.

The location of the primary thyroid tumor was categorized as previously described: upper third, middle third, lower third, isthmus, multifocal in the affected lobe, or multifocal in both lobes (5).

Statistical analysis

The results are expressed as the mean ± sd. The statistical analysis was performed using a Student's t test, χ² test, or Mann-Whitney test, as appropriate. A receiver-operating characteristic (ROC) analysis was used to identify the cutoff point of the size of primary tumor for defining the risk of CLNM and LLNM. The odds ratio (OR) and the 95% confidence interval (CI) for relationships between each variable and lymph node metastasis (yes or no) were calculated using binary logistic regression. A P < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS Software version 12.0 (SPSS, Chicago, IL).

Results

Patient characteristics

There were 2939 patients with PTC initially treated at our hospital between January 2005 and December 2009; of these, 1124 (38.2%) were diagnosed with PTMC histologically, although 58 patients were excluded because their primary tumors were found incidentally during thyroidectomy. Thus, there were a total of 1066 patients who met the inclusion and exclusion criteria of the study.

The age of these 1066 patients ranged from 15 to 79 yr (mean 45.2 ± 11.5 yr). Based on US and FS results, three different types of surgical procedures were performed: 1) lobectomy with ipsilateral CLND (649 patients); 2) lobectomy, ipsilateral CLND, and subtotal lobectomy of the contralateral thyroid lobe (341 patients: 158 patients underwent nodule enucleation and 183 patients underwent partial lobe resection); and 3) TT with a bilateral CLND (76 patients). Of the 105 patients who underwent LLND, the final pathological results revealed that 100 patients had positive malignant lymph node metastasis in the lateral neck. The mean number of harvested central lymph nodes was 3.76 ± 2.82 (range 1–22), and the mean number of harvested lateral lymph nodes was 21.01 ± 8.85 (range 5–48). Skip metastases, which indicate LLNM without CLNM (23), were found in 24 patients (2.2%).

There were no patients with a history of head and neck radiation before surgery, and no patient had distant metastasis. The characteristics of these patients are listed in Table 1.

Table 1.

Characteristics of patients

Parameter	Total (n = 1066)	CLNM− (n = 614)	CLNM+ (n = 452)	P value
Gender
Male	253 (23.7%)	114 (18.6%)	139 (30.8%)	0.001
Female	813 (76.3%)	500 (81.4%)	313 (69.2%)
Size (cm, mean ± sd)	0.66 ± 0.23	0.64 ± 0.24	0.69 ± 0.22	0.001
Age (yr, mean ± sd)	45.2 ± 11.5	46.0 ± 11.4	44.0 ± 11.5	0.003
Location of the primary tumors
Solitary lesion	850 (79.7%)	559 (91.0%)	291 (64.4%)	0.001a
Upper third	278 (26.1%)	242 (39.4%)	36 (8.0%)	0.001b
Middle third	339 (31.8%)	234 (38.1%)	105 (23.2%)
Lower third	212 (19.9%)	70 (11.4%)	142 (31.4%)
Isthmus	21 (2.0%)	13 (2.1%)	8 (1.8%)
Multifocal lesions	216 (20.3%)	55 (9.0%)	161 (35.6%)
Multifocal in both lobes	76 (7.1%)	12 (2.0%)	64 (14.2%)	0.001c
Multifocal in affected lobe	140 (13.1%)	43 (7.0%)	97 (21.5%)
Bilateral
Positive	76 (7.1%)	12 (2.0%)	64 (14.2%)	0.001
Negative	990 (92.9%)	602 (98.0%)	388 (85.8%)
Subtype
Ordinary	1054 (98.9%)	605 (98.5%)	449 (99.3%)	0.220
Follicular variant	12 (1.1%)	9 (1.5%)	3 (0.7%)
Extrathyroidal extension of the primary tumor
Positive	210 (19.7%)	56 (9.1%)	154 (34.1%)	0.001
Negative	856 (80.3%)	558 (90.0%)	298 (65.9%)
Lateral neck metastasis
Positive	100 (9.4%)	24 (3.9%)	76 (16.8%)	0.001
Negative	966 (90.6%)	590 (96.1%)	376 (83.2%)

−, Negative; +, positive; P value, the statistically difference between CLNM+ group and CLNM− group.

^a

The P value means the difference between the group solitary lesion and multifocal lesions.

^b

The P value means the difference among the upper third, middle third, lower third, and isthmus in the group of solitary lesion.

^c

The P value means the difference between the group multifocal in both lobes and in affected lobe in the group of multifocal lesions.

Risk factors for CLNM

Lee et al. (20) reported a tumor size of greater than 7 mm indicated a risk factor for CLNM in PTMC by using a multiple test. We used the ROC analysis to determine the cutoff point for the primary tumor size (1–9 mm). As a result, a tumor size of 6.5 mm was the cutoff point for CLNM in our population (Supplemental Table 1 and Supplemental Fig. 1, published on The Endocrine Society's Journals Online web site at http://jcem.endojournals.org).

Tumor size greater than 6 mm and greater than 7 mm were both significantly associated with CLNM according to a univariate logistic regression. Other variables, such as younger age (≤45 yr of age), male gender, extrathyroidal extension, multifocal, bilateral, were also significantly associated with CLNM by univariate logistic regression. Histological subtype was not associated with CLNM. However, when these variables were included in multivariate logistic regression models, a tumor size greater than 7 mm was not a statistically significant predictor of CLNM. Thus, we analyzed the risk of a tumor size greater than 6 mm, which was a statistically significant predictor (Table 2). The results of the logistic regression for tumor size greater than 7 mm were not shown.

Table 2.

Univariate and multivariate logistic regression for CLNM

Independent variable	Univariate		Multivariate
	OR (95% CI)	P value	OR (95% CI)	P value
Gender (male vs. female)	1.949 (1.464–2.591)	0.001	1.957 (1.427–2.681)	0.001
Age (≤45 vs. >45 yr)	1.348 (1.056–1.718)	0.016	1.406 (1.071–1.848)	0.014
Size (>6 vs. ≤6 mm)	1.581 (1.238–2.020)	0.001	1.360 (1.029–1.798)	0.028
Extrathyroidal extension (positive vs. negative)	5.149 (3.678–7.210)	0.001	4.181 (2.908–6.010)	0.001
Multifocal (positive vs. negative)	5.678 (4.053–7.954)	0.001	4.476 (2.975–6.735)	0.001
Bilateral (positive vs. negative)	8.275 (4.409–15.531)	0.001	1.893 (0.901–3.978)	0.092
Subtype (follicular vs. ordinary)	0.449 (0.121–1.669)	0.232	0.249 (0.053–1.161)	0.177

Tumor location as a risk factor for CLNM

The association between primary tumor location (upper third, middle third, or lower third) and CLNM risk was analyzed using data from the 829 patients with solitary primary lesions limited to one lobe. There were 278 patients with a primary lesion in the upper third of the thyroid lobe, 339 patients in the middle third, and 212 patients in the lower third (Table 1).

Table 3 shows that there was an increased risk according to the location of the tumor when the location was adjusted for the upper third, which indicates that patients with primary tumor in the lower third had a greater probability of suffering from CLNM than did those with a primary tumor in the upper third. This association persisted in the multivariate logistic analysis (Table 4).

Table 3.

The risk of location in the solitary primary tumor for CLNM adjusted for the factor of upper third

Variable	Adjusted OR	95% CI	P value
Location upper third	1
Middle third	3.016	1.984–4.586	0.001
Lower third	13.637	8.676–21.434	0.001

Table 4.

Univariate and multivariate logistic regression for CLNM in the solitary primary tumor

Independent variable	Univariate		Multivariate
	OR (95% CI)	P value	OR (95% CI)	P value
Location
Upper third	1 (reference)		1 (reference)
Middle third	3.016 (1.984–4.586)	0.001	3.049 (1.957–4.751)	0.001
Lower third	13.637 (8.676–21.434)	0.001	14.881 (9.184–24.112)	0.001
Gender
Female	1 (reference)		1 (reference)
Male	1.803 (1.294–2.51)	0.001	1.795 (1.215–2.653)	0.003
Age (yr)
>45	1 (reference)		1 (reference)
≤45	1.477 (1.107–1.972)	0.008	1.742 (1.237–2.451)	0.001
Size (mm)
≤6	1 (reference)		1 (reference)
>6	1.876 (1.400–2.514)	0.001	1.623 (1.146–2.299)	0.006
Subtype
Ordinary	1 (reference)		1 (reference)
Follicular	3.133 (0.375–26.154)	0.291	4.017 (0.365–44.215)	0.256
Extrathyroidal extension
Negative	1 (reference)		1 (reference)
Positive	4.825 (3.255–7.152)	0.001	4.292 (2.722–6.767)	0.001

Multivariate analyses were adjusted for all factors listed in table.

Risk factors for LLNM

The risk factors for LLNM were evaluated using univariate and multivariate logistic regression analyses for patients who had positive LLNM, according to the final pathological result. The ROC analysis indicated that a tumor size of 6.5 mm was the threshold for LLNM in our population in a single-factor analysis (Supplemental Table 2 and Supplemental Fig. 2). The univariate analyses showed that a tumor size greater than 6 mm, extrathyroidal extension, CLNM, multifocal, and bilateral were risk factors for LLNM upon final pathological result, whereas a tumor size greater than 7 mm was not significantly associated with LLNM (data not shown). A multivariate analysis revealed that extrathyroidal extension, CLNM, and multifocal were significantly associated with LLNM, as shown in Table 5. Moreover, a logistic regression analysis was performed for the solitary primary tumor group, and we found that tumor location in the upper third conferred a greater risk of LLNM (Table 6).

Table 5.

Univariate and multivariate logistic regression for LLNM

Independent variable	Univariate		Multivariate
	OR (95% CI)	P value	OR (95% CI)	P value
Gender (male vs. female)	1.281 (0.807–2.031)	0.293	0.975 (0.577–1.648)	0.926
Age (≤45 vs. >45 yr)	1.162 (0.770–1.755)	0.474	1.147 (0.724–1.817)	0.558
Size (>6 vs. ≤6 mm)	1.282 (0.846–1.943)	0.041	0.771 (0.472–1.257)	0.297
Extrathyroidal extension (positive vs. negative)	9.486 (6.096–14.764)	0.001	7.925 (4.779–13.114)	0.001
Multifocal (positive vs. negative)	3.777 (2.464–5.789)	0.001	3.560 (2.083–6.086)	0.001
Bilateral (positive vs. negative)	1.148 (0.535–2.464)	0.722	0.194 (0.079–0.476)	0.001
Subtype (follicular vs. ordinary)	0.877 (0.112–6.864)	0.900	1.004 (0.081–18.583)	0.997
CLNM (positive vs. negative)	4.969 (3.085–8.005)	0.001	2.334 (1.352–4.028)	0.002

Table 6.

Multivariate logistic regression for LLNM in the solitary primary tumor

Variable	Adjusted OR	95% CI	P value
Location
Upper third	1 (reference)
Middle third	0.297	0.140–0.630	0.002
Lower third	0.264	0.114–0.610	0.002
Gender
Female	1 (reference)
Male	0.486	0.221–1.067	0.072
Age (yr)
>45	1 (reference)
≤45	1.028	0.562–1.879	0.928
Size (mm)
≤6	1 (reference)
>6	1.103	0.576–2.110	0.768
Subtype
Ordinary	1 (reference)
Follicular	0.004	0.001–20.112	0.997
Extrathyroidal extension
Negative	1 (reference)
Positive	9.468	4.898–18.304	0.001
CLNM
Negative	1 (reference)
Positive	2.710	1.317–5.575	0.007

Multivariate analyses were adjusted for all factors listed in table.

Complications, follow-up, and recurrence

Among 1066 patients, 80 (7.5%) had transient hypoparathyroidism and three patients (0.28%) had permanent hypoparathyroidism. Recurrent laryngeal nerve injury occurred in 12 patients (1.1%). One of these 12 cases was transient, and 11 of the 12 cases were permanent.

Postsurgical physical examinations were performed every 3–6 months. All patients adhered to a 12-month follow-up period, and many patients had a longer follow-up period. The mean length of follow-up was 33.5 ± 15.9 months, ranging from 12 to 72 months. During the follow-up period, 10 patients (0.94%) had locoregional recurrence; no patient demonstrated distant metastasis or died. Among 990 patients who did not undergo TT, only seven (0.81%) suffered from a malignant recurrence in the contralateral lobe.

Discussion

There were 1124 patients (38.24%, 1124 of 2939) histologically diagnosed with PTMC at our hospital from 2005 to 2009, which represents a frequency of microcarcinoma in PTC similar to previous reports (3, 24–26). Excluding the 58 incidental PTMC patients, we systematically analyzed the pattern and frequency of lymph node metastases in 1066 patients with PTMC. The prevalence of CLNM and LLNM in our population was 42.4% (452 of 1066) and 9.4% (100 of 1066), respectively. In previous reports, the CLNM prevalence ranged from 24.1 to 64.1%, and the LLNM prevalence ranged between 3.7 and 44.5% (5–13). This large range in the prevalence of LLNM may be due to differences in the number of patients and the performance of prophylactic LLND.

In our study, univariate and multivariate analyses were used to evaluate the risk factors for CLNM. We found that male gender, younger age (≤45 yr), multifocal lesions, extrathyroidal extension, and primary tumor greater than 6 mm were risk factors for CLNM.

Age is known to be an important prognostic factor for patients with PTC greater than 1 cm; however, its prognostic value in PTMC was uncertain (5, 6, 13, 27). In our study, younger age (≤45 yr of age) was associated with an increased risk of CLNM, which was also reported by previous studies (28, 29). In addition, in agreement with a previous report (13), we found that male gender was a risk factor for CLNM. The follicular variant is the most common subhistological variant in PTC. In a retrospective study, Passler et al. (30) found that the prevalence of CLNM was lower in the follicular variant PTC group, although this difference was not statistically significant. In our study, the follicular variant subtype was also not significantly associated with CLNM; however, this subtype was rare because it was found in only 1.1% of patients, which limited our ability to assess its association with CLNM.

Previous studies suggested that clinicopathological and prognostic differences are associated with PTMC tumor size (13, 19, 20, 28, 31–33). Most studies used a tumor size greater than 5 mm as the size threshold. However, thresholds greater than 7 mm and greater than 8 mm were also assessed. In the univariate analysis, we found that a tumor size greater than 7 mm was associated with the greatest risk of CLNM. However, when the size was adjusted for other factors, in the multivariate analysis, only a tumor size greater than 6 mm was significantly associated with CLNM. In contrast to previous reports, our results lead us to believe that a threshold greater than 5 mm is too small a size threshold and greater than 6 mm may be more appropriate.

Wada et al. (5) first reported that the location of PTMC within the thyroid was related to the prevalence of CLNM and LLNM, although these differences were not statistically significant. Kwak et al. (6) found that the upper polar location, determined by US, was significantly associated with LLNM. In our study of patients with a solitary primary tumor, we found that location in the upper third of the thyroid lobe conferred a lower risk for CLNM and a higher risk for lateral cervical metastases. These results suggest that CLND may not be necessary for patients with only one tumor located in the upper third of the thyroid.

Furthermore, in the multivariate logistic regression analysis of the risk factors for LLNM, extrathyroidal extension, multifocal lesions, and CLNM were statistically significant factors. These results imply that patients with these types of PTMC have a greater frequency of LLNM. However, one limitation in our study was that the LLND was performed in only 105 patients. Therefore, the rate of LLNM (9.4%) in our study may be underestimated considering many patients with PTC suffer from subclinical LLNM. In fact, although prophylactic LLND is not recommended in the American Thyroid Association guidelines, it is still performed in some centers (5). The rate of LLNM was reported to range from 39.5 to 64.2% (5, 11, 34), whereas the data based on prophylactic LLND were more accurate and meaningful. However, in PTMC, the complications of LLND cannot be weighed against its undefined effect on recurrence and survival (11, 12). Our study may offer some useful information for selecting PTMC patients who will be more likely to develop LLNM. Therefore, our analysis should not be disregarded solely because prophylactic LLND was not beneficial in PTMC patients (5).

An interesting finding of the current study was that there were 24 patients (24 of 100, 24%) with skip metastasis [LLNM without CLNM (5)]. The prevalence of skip metastasis that we observed in PTMC was similar to that reported for all sizes of PTC, ranging from 5.4 to 37.5% (5, 23, 35). Moreover, our results indicate an association with the location of the solitary primary tumor. For example, the presence of a primary tumor in the upper third of the thyroid lobe indicates a lower risk for CLNM and a higher risk for LLNM.

The optimal surgical treatment for PTMC still remains controversial. Neither the American Thyroid Association nor the American Joint Committee on Cancer guidelines provides recommendations for PTMC treatment. In contrast to other reports, most PTMC patients in our study underwent lobectomy for the initial surgical management instead of TT or near-total thyroidectomy. TT for traditional PTC was recommended in the latest version of the National Comprehensive Cancer Network guidelines; however, in the context of PTMC, thyroid lobectomy is considered adequate for patients without a history of radiation or any other risk factors (36). After nearly 3 yr of follow-up, only seven patients (0.81%, seven of 990) developed contralateral malignant lesions. The proportion of patients who required additional surgery in the contralateral lobe was very low. Recently Koo et al. (37) reported a 16.7% rate of contralateral carcinoma, which did not justify routine TT in all patients with unilateral PTMC.

The existence of multifocal lesion is recognized as a risk factor for CLNM in PTC patients (37). Likewise, multifocal lesions consistently demonstrated an effect on the risk of CLNM in our study. However, we were limited by underestimation of the prevalence rate of multifocal lesions, which was a consequence of not performing TT without the detection of a suspected contralateral nodule by US and FS. The minimum nodule diameter that can be detected by US is approximately 2 mm. Based on the final pathology results, there were 38 patients (38 of 1066, 3.6%) with tumors 2 mm or smaller in diameter, which is similar to the results obtained in a previous study (15 of 259, 5.8%) (10). In our population, there were 217 patients (20.4%) who were found to have multifocal lesions after their first operation, which was similar to that found in previous reports (ranging from 19.1 to 44.3%) (5, 6, 8, 11, 13, 19, 20, 32). Therefore, the extent to which the presence of multifocal lesions was underestimated in our study seems to be reasonable.

Recent studies of CLND report the development of permanent hypoparathyroidism in 0–4% of patients and the development of recurrent laryngeal nerve injury in 0–6% of patients, respectively; the temporary complication rates for these conditions were 47 and 5%, respectively (38, 39). Moreover, the complication rate resulting from CLND plus TT was compared with that for TT alone (40), and only the rate of transient hypoparathyroidism differed significantly between TT plus CLND group and the TT group. The rates of transient or permanent recurrent laryngeal nerve injury and permanent hypoparathyroidism did not differ significantly between the groups. Hence, the advantage of prophylactic CLND should be weighed against the associated complications. In our study, permanent hypoparathyroidism and recurrent laryngeal nerve injury occurred in three (0.3%) and 12 (1.1%) of the patients, respectively, which suggests that routine ipsilateral level VI dissection can be performed safely and with low morbidity by experts. Thus, in our opinion, it may be better to perform CLND in selected patients with high risk factors.

Finally, for a more thorough determination of the rates of locoregional recurrence and disease-free survival in PTMC, we recommend a prospective long-term follow-up analysis for in this study population. We also suggest the implementation of a multicenter research study to ascertain the optimal surgical approach for patients with PTMC.

Conclusion

The results of our study showed that male gender, younger age (≤45 yr of age), the presence of multifocal lesions, extrathyroidal extension, and larger sized primary tumor (>6 mm) were risk factors for CLNM. For LLNM, the factors of extrathyroidal extension, multifocal lesions, and CLNM were factors that conferred an increased risk. In patients with a solitary primary tumor, the location of the tumor had an important relationship with neck lymph node metastasis status. The patients with a tumor located in the upper third of the thyroid lobe had a lower risk of CLNM and a higher risk of LLNM. We recommend additional multicenter research and long-term follow-up to better understand the risk factors and surgical management of PTMC.

Acknowledgments

This research was supported by Grant 30872958 from the National Natural Science Foundation of China.

Disclosure Summary: The authors have nothing to disclose.

^*

L.Z. and W.W. contributed equally to the paper.

For editorial see page 1169

Abbreviations

 
• CI

  Confidence interval

 
• CLND

  central lymph node dissection

 
• CLNM

  central lymph node metastasis

 
• CT

  computed tomography

 
• FNA

  fine-needle aspiration

 
• FS

  frozen section

 
• LLND

  lateral lymph node dissection

 
• LLNM

  lateral neck lymph node metastasis

 
• OR

  odds ratio

 
• PTC

  papillary thyroid carcinoma

 
• PTMC

  papillary thyroid microcarcinoma

 
• ROC

  receiver-operating characteristic

 
• TT

  total thyroidectomy

 
• US

  ultrasound.