As several studies indicated that oncologic outcomes of lobectomy are comparable to those of a near-total or total thyroidectomy for low-risk papillary thyroid carcinoma (PTC) (1–3), thyroid lobectomy alone is recommended as an alternative for low-risk PTC (4–6). However, some adverse pathological features, which would upgrade the cases from low risk to intermediate risk or high risk, could only be assessed in postoperative examination (7, 8). Some of these adverse pathological features are recommended as indications of completion thyroidectomy following lobectomy, which may lead to more severe anxiety for the patients and increase expenditures for medical care and public health, compared to one-stage near-total or total thyroidectomy (7, 9). In NCCN clinical practice guidelines in oncology (thyroid carcinoma, version 3.2021), large volume pathological N1a metastasis is recommended as one of the adverse pathological features indicating completion thyroidectomy following lobectomy. It refers to equal to or more than 5 involved lymph nodes or any nodal metastasis >2 mm in largest dimension, which reveals high risk of PTC recurrence (7, 10, 11), whereas the diagnostic performance of imaging techniques, such as ultrasonography (US) or computer tomography, is poor in detecting involved lymph nodes in the central compartment preoperatively (12). In this research, we tried to build a preoperative nomogram to predict high-volume central lymph node metastasis (HVCLNMs, equal to or more than 5 involved lymph nodes in central compartment) to reduce reoperation rate after thyroid lobectomy.

In our study, a total of 423 patients with age of 47.0 ± 12.2 years were included. Among the patients, 77 (18.2%) were male. The BRAF mutation test results showed that 372 (87.9%) had a V600E mutation. A total of 112 (26.5%) patients were diagnosed as Hashimoto’s thyroiditis preoperatively according to circulating autoantibodies and ultrasonic features. The characteristics of the thyroid nodules in ultrasound examination and postoperative pathological features are presented in Table 1 . The intra-observer agreements of the two researchers in assessing US characteristics were 0.773–0.966 and 0.839–0.977, respectively, whereas the inter-observer agreements between the two researchers were 0.759–0.947 ( Table S1 ).

The univariate logistic regression analysis revealed that age, BRAF V600E mutation, multifocality, and calcifications in US examination were significantly associated with HVCLNM. Nodule diameter, TIRADS classification, and suspicious malignant focus at the inferior part of thyroid lobe might be associated with HVCLNMs with a p-value between 0.05 and 0.10. They were all included in multivariate logistic regression analysis. As shown in Table S2 and Supplementary Figure S1 , the risk of HVCLNM in patients with PTC decreased dramatically when the patients were older than 35 years. Thus, age was included in multivariate logistic regression analysis as a dichotomous variable (</=35 years vs. >35 years). Multivariate logistic regression analysis showed that age (</=35 years vs. >35 years), BRAF with V600E mutated, nodule diameter, and calcification in US examination independently predicted HVCLNMs ( Table 2 ).

The nomogram that incorporated the independently predictive factors was developed and presented ( Figure 1 ). The mean AUC for the nomogram was 0.821 (95% CI, 0.768–0.875) ( Figure 2 ). Setting a cutoff point of 0.50, sensitivity, specificity, positive predictive value, and negative predictive value were 40.4%, 96.2%, 62.2%, and 91.2%, respectively. The calibration curve of the nomogram presented good agreement between the predicted and observed probability of HVCLNMs, especially for the observed HVCLNMs probability >0.5 ( Figure 3 ).

In the recent two decades, several studies indicated that oncologic outcomes of lobectomy are comparable to those of a near-total or total thyroidectomy for low-risk papillary thyroid carcinoma. Davies et al. analyzed 35,223 cases of PTC confined to the thyroid gland from 1973 to 2005 and found that the prognosis of the patients with either hemithyroidectomy or total thyroidectomy would be the same (1). Bilimoria et al. analyzed 52,173 patients having underwent surgery for PTC from 1985 to 1998 and concluded that extent of surgery (lobectomy versus total thyroidectomy) did not impact recurrence or survival for PTC <1 cm (2). Jonklaas et al. drew a similar conclusion that the extent of thyroid surgery was not associated with overall survival, disease-specific survival, or disease-free survival in stage I patients (3). In the 2015 ATA guidelines, thyroid lobectomy alone was recommended as an alternative for the initial treatment of low-risk PTC (8). A similar opinion was presented in NCCN clinical practice guidelines in oncology (thyroid carcinoma, version 3.2021) (7). However, on account of the unsatisfying diagnostic performance of imaging techniques, some adverse features, which indicate high risk of recurrence and reveal unfavorable prognosis, could only be assessed in postoperative pathological examination. Completion thyroidectomy following lobectomy is recommended to reduce risk of recurrence and improve prognosis for these patients, whereas completion thyroidectomy following lobectomy may lead to increasing complications, severe anxiety of the patients, and extra expenditures for medical care and public health, compared with one-stage surgery (near-total or total thyroidectomy) (9, 20).

HVLNMs (equal to or more than 5 involved lymph nodes) was one of the risk factors indicating high recurrence rate and poor prognosis in several studies (21). Sugitani et al. reported that the patients with > 5 LN metastases demonstrated a significantly higher risk of recurrence (19% versus 8%) (11). Likewise, in the research by Leboulleux et al., the patients with >10 positive LNs (21%) or 6–10 LN metastases (7%) significantly predicted higher 10-year risk of recurrence compared with the patients with <5 LN metastases (3%) (10). In the 2015 ATA guidelines, pathologic N1 with more than 5 involved lymph nodes was one of the criteria to upgrade low-risk PTC to intermediate-risk PTC, which makes risk of structural disease recurrence increase by about 15% (8). Furthermore, HVLNMs is recommended as one of the adverse pathological features indicating completion thyroidectomy following lobectomy in NCCN clinical practice guidelines in oncology (thyroid carcinoma, version 3.2021) (7). However, the diagnostic performance of preoperative US or computer tomography in predicting lymph node metastasis (LNM) in the central compartment is unsatisfactory for now, which is the most frequently involved level in PTC (8, 12, 22). Thus, it may be valuable to develop a predictive model in evaluating the status of lymph nodes and predicting HVLNM in the central compartment, which may help surgeons and patients draw up individual treatment plans preoperatively and reduce reoperation risk. In our research, we developed a nomogram based on preoperative clinical features, in which the AUC was 0.821 (95% CI, 0.768–0.875) in predicting HVCLNMs. These results validate the idea that our nomogram provides good discrimination ability for detecting HVCLNMs in PTC patients (23, 24). Furthermore, the nomogram is easy to use, repeated, and economical, especially in the undeveloped area. In the 2015 ATA guidelines, lateral neck compartmental lymph node dissection was recommended for the patients with biopsy-proven metastatic lateral cervical lymphadenopathy (8). Thus, in this cohort, lateral neck dissection was not performed for the patients. During the follow-up period, the lateral neck lymph node should be actively surveilled for the patients with HVCLNMs. If lateral neck lymph node recurrence increases significantly in patients with HVCLNMs, elective neck dissection should be considered in these cases.

V-raf murine sarcoma viral oncogene homolog B1 (BRAF) mutations play roles in tumor cell proliferation and differentiation (25). BRAF V600E is the most common oncogene in PTC, which is reported to be correlated with some clinicopathological features in some studies (26, 27). In the research by Shi et al., a cohort of 126 patients was retrospectively analyzed and BRAF mutation status was found to be significantly associated with tumor size and LNM (28). A meta-analysis, including 3,437 patients, showed that the BRAF V600E mutation was correlated with some aggressive clinicopathological features, such as tumor multifocality, extrathyroidal extension, lymph node metastases, and advanced stage of PTMC (29). Similarly, Chen et al. reported that BRAF (V600E) mutation was an independent risk factor for central lymph node metastasis (CLNM), together with size </=10mm, microcalcification, internal vascularity, and capsule contact or involvement in preoperative ultrasound examination (30). Although several studies reported the association between BRAF (V600E) mutation and CLNM, no study focused on the predictive value of BRAF (V600E) mutation for HVCLNMs and the assistance in clinical decision-making of thyroidectomy. In our research, positive BRAF V600E mutation in PTC was proved to be able to predict high-volume CLNM independently with an odds ratio of 14.363 (95% CI, 2.847–72.463), which indicated that the preoperative BRAF V600E mutation test is useful in risk stratification of HVCLNMs and could assist in the management of PTC.

Age was reported to be associated with LNM in several studies previously (21, 31–35). Liu et al. included a total of 48,166 PTC and concluded that LNM rate decreased with age, especially women (p < 0.0001) (31). Luo et al. included 1,031 patients with PTMC and reported that male, age (</= 40 years), tumor largest diameter (>/= 5 mm), multifocal, non-uniform echoic distribution, the sum of the maximum diameter of multifocal in a unilateral lobe (>/= 8.5 mm), and tumors in the lower pole location were risk factors of CLNM in papillary thyroid microcarcinoma (PTMC) (32). Another study about PTMC similarly reported that young (<40 years old) and male patients were independent risk factors for large‐volume LNM (>5 metastatic lymph nodes) (21). In our study, the analysis drew a similar conclusion that young age (</=35 years) was an independent risk factor for high-volume CLNM, which could effectively predict HVCLNMs, together with BRAF V600E mutation, nodule diameter, and nodule calcification in US.

Our study has several limitations. Firstly, in our study, we only focused on high-volume CLNM. For the patients with high risk of HVCLNMs, total thyroidectomy should be recommended more strongly. Otherwise, the risk of other adverse features, such as macroscopic multifocal disease and vascular invasion, should be assessed at the same time, which would assist in determining surgical procedure, total thyroidectomy versus thyroid lobectomy, for the patients with low risk of HVCLNMs. Secondly, the nomogram was assessed only using the internal validation method, and this might have some adverse impact on the application in other centers. In the near future, the diagnostic performance of the nomogram may be further improved via multicentral research with a large sample.

The preoperative nomogram performed well in discriminating the patients with or without high-volume CLNM (equal to or more than 5 lymph nodes) in papillary thyroid carcinoma, which is one of the indications for completion thyroidectomy after thyroid lobectomy.

The datasets used during the current study are available in Mendeley Data, V1, doi: 10.17632/sd59nt5dty.1.

The studies involving human participants were reviewed and approved by the Ethics Committee of Sun Yat-sen Memorial Hospital, Sun Yat-sen University. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

PL: Conceptualization, Methodology, Software, Formal analysis, Investigation, Data Curation, and Writing—Original Draft. FL: Methodology, Formal analysis, Investigation, Resources, and Writing—Original Draft. JR: Conceptualization, Formal analysis, Investigation, Resources, Data Curation, and Writing—Original Draft. PH: Validation, Investigation, Writing—Original Draft, and Funding acquisition. JL: Conceptualization, Validation, Investigation, Resources, Writing—Original Draft, and Funding acquisition. RC: Validation, Formal analysis, Investigation, Writing—Original Draft, and Visualization. BL: Methodology, Resources, Writing—Review and Editing, Supervision, and Project administration. NO: Conceptualization, Methodology, Writing—Review and Editing, Supervision, and Project administration. XH: Conceptualization, Methodology, Resources, Data Curation, Writing—Review and Editing, Supervision, and Funding acquisition. All authors contributed to the article and approved the submitted version.

This study was supported by the Grants from the Sun Yat-Sen University Clinical Research 5010 Program (Grant 2010008), the National Natural Science Foundation of China (Nos. 81872193, 81702697, and 81903043), and the Natural Science Foundation of Guangdong Province (No. 2018A030310086). The sponsors were not involved in research activities, such as study design, analysis, and interpretation of data.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.