Thyroid glands have plentiful lymphatic vessels; thus, cervical lymph node metastasis is regularly observed and specific to PTC. Generally, PTC first metastasizes from the primary tumor site to the pretracheal, paratracheal and upper mediastinal lymph nodes, then to the ipsilateral LLNs, and finally to the contralateral region [5]. Skip metastasis occurs when tumor cells bypass the CLNs, and metastasis first occurs in the LLNs. This phenomenon has been reported in several studies, affecting approximately 21.8–23.5% of patients [6, 7]. In this study, 13 of 203 patients with lateral metastasis had skip metastasis. Lateral metastasis is suggestive of a poor prognosis. In this study, the probability of lymph node metastasis in the lateral area was 39.9% at level IIa, 84.2% at level III, 60% at level IV and 8.9% at level Vb. This is consistent with the 53% probability of metastasis at level II, 71% at level III, 66% at level IV and 25% at level V reported in the literature [8].
The ATA first published guidelines for the treatment of thyroid nodules and differentiated thyroid cancer in 1996 and updated these guidelines in 2006, 2009 and 2015. The range of central neck dissection was defined in the 2009 ATA guidelines [9] and has not been revised to date. The upper boundary is the hyoid bone, the bilateral boundary is the medial margin of the common carotid artery, and the lower boundary is the innominate artery, including lymph nodes at levels VI and VII. For the treatment of lymph nodes in the lateral cervical region, George Crile first proposed radical neck lymph node dissection in 1906, including lymph nodes at levels I-V and resection of the internal jugular vein, accessory nerve and sternocleidomastoid muscle. Because of the severe surgical trauma that can occur with this approach, along with many complications and poor postoperative quality of life, modified cervical lymph node dissection is now widely used. For differentiated PTC, the ATA recommended in 2012 that patients with LLN metastasis should undergo lymph node dissection at levels IIa, III, IV and Vb. Dissection of lymph nodes at level IIb or Va should be performed only if there is clear evidence of metastasis at these two levels [4]. The latest National Comprehensive Cancer Network (NCCN) guidelines [10] for the range of lateral neck lymph node dissection are consistent with the ATA guidelines. The patients enrolled in this study were diagnosed and treated from 2014 to the present, and the surgical procedures were performed in strict adherence with the standards outlined in the 2012 ATA guidelines.
In this study, the number of positive lymph nodes in the central region was a risk factor for lymph node metastasis at level IV. Previous studies have suggested that CLN metastasis can predict the possibility of overall lateral metastasis [11,12,13]; however, no studies have yet reported the predictive effect of CNL metastasis at each level of the lateral cervical region. The results of this study suggest that the number of positive CNLs has predictive value for level IV metastasis. For each additional positive lymph node in the central region, the risk of metastasis at level IV increased by 1.126 times. However, the number of positive lymph nodes in the central region could not be accurately confirmed by preoperative imaging [14, 15]. Moreover, all the patients included in this study had pathological evidence of LLN metastasis before lateral dissection. Level IV is a conventional dissection region. The frozen pathology results of the number of positive lymph nodes in the central region during surgery could not provide additional guidance for selection of the surgical strategy. For patients without pathological evidence of lateral metastasis before lateral dissection, if the number of suspected positive CLNs during the operation is high, and this suspicion is confirmed by frozen pathology, does this indicate a risk for metastasis at level IV? Should this guide the surgeon in evaluating the lateral region? To answer these questions, further confirmation by a prospective study is needed.
The lateral cervical lymph node dissection range has been modified several times. The reduction in the surgical scope has been brought on by the application of precision therapy according to evidence-based medicine. Prolonging the overall survival of patients, reducing postoperative complications and improving the quality of life of patients have always been the goals of surgery. Investigation into whether the LLN dissection scope can be further reduced, such as in selective level dissection or super-selective level dissection, is ongoing [16, 17]. Some researchers believe that super-selective lymph node dissection is feasible, especially in robot-assisted and endoscopic surgery [16, 18], while other researchers doubt the safety of this approach. Doctor Piccin [19] suggests that patients with preoperatively proven PTC LLN metastasis should undergo dissection including levels II to V and that the transoral robotic approach may not be the ideal surgical technique for neck dissection. In a study conducted by Doctor Zhao [16], the experimental group underwent selective dissection at levels III and IV, and the control group underwent standardized dissection at levels II through V. The rate of lymph node metastasis at level II in the control group was 33% (42/147). Although 1/3 of the patients in the experimental group were spared from level II dissection, the risk of recurrence could not be ruled out without close long-term follow-up. Our results revealed that the risk of metastasis of multifocal tumors was higher than that of unifocal tumors by 1.958 times at level IIa (P = 0.021, OR = 1.958) and 2.929 times at level Vb (P = 0.049, OR = 2.929). The risk of metastasis at levels IIa and Vb is significantly higher for multifocal tumors than unifocal tumors. In cases of multifocal tumors detected on preoperative imaging, more attention should be given to these two levels when performing LLN dissection. Can patients with a unifocal tumor be spared from level IIa and Vb dissection? The data in the present study showed that, in cases of unifocal tumors, 5 patients were positive for metastasis at level Vb; 33 patients, level IIa; and 2 patients, levels IIa and Vb. Among them, only 2 patients were pathologically diagnosed with level IIa or Vb metastases before the operation. Other diagnoses were made incidentally after the surgery. Therefore, with selective dissection, 16.7% (34/203) of patients with a unifocal tumor would have been at risk of residual metastatic lymph nodes at levels IIa and Vb. We suggest that for patients with unifocal tumors, if selective dissection is considered, careful evaluation by ultrasonography and enhanced CT of levels IIa and Vb is critical. If any suspicious lymph node is observed, FNA should be performed to guide the dissection range and avoid residual positive lymph nodes.
This study has the following limitations. First, the time span of this study was long; some surgeons used monopolar electrosurgical units, some used ultrasonic knives, and some used bipolar electrosurgical units. Bipolar electrosurgical units are more accurate anatomically and may allow the removal of more lymph nodes. Second, because the number of positive lymph nodes at each level varied widely across patients, the metastatic status of lymph nodes in each lateral region was recorded as negative or positive for research and analysis. If the absolute number of positive lymph nodes or the proportion of positive lymph nodes is recorded and analyzed, different results may be obtained [20], but further evidence is needed to demonstrate that a certain data recording method is more clinically objective. In addition, this study was retrospective in nature. Further prospective studies with higher evidence levels are needed in the future.
