Abstract
Objective:
To identify independent risk factors for lymph node metastasis posterior to the right recurrent laryngeal nerve (RLN) in cN0 papillary thyroid carcinoma and develop a predictive model for preoperative risk stratification.
Methods:
A retrospective analysis was conducted on 721 cN0 papillary thyroid carcinoma patients that underwent lymph node dissection posterior to the right RLN between 2016 and 2019. Univariable and multivariable logistic regression analyses were used to assess clinical and pathological characteristics, and a nomogram was constructed. The model was validated using receiver operating characteristic curves and calibration curves.
Results:
The incidence of lymph nodes posterior to the right was 19.8%. Multivariable analysis revealed three independent predictors: the number of paratracheal lymph node metastases (OR = 1.360, p = 0.021), combined prelaryngeal and paratracheal lymph node metastases (1–2 involved nodes: OR = 3.433; ⩾ 3 involved nodes: OR = 4.402, p < 0.05), and extrathyroidal extension (OR = 0.580, p = 0.021).
Conclusion:
This model provides a quantitative tool for intraoperative risk assessment and guides the development of individualized surgical plans for high-risk patients with cN0 papillary thyroid carcinoma.
Keywords
Introduction
The incidence of thyroid cancer in China has steadily increased over the past decade. According to the National Cancer Centre, there were 466,100 new cases of thyroid cancer in 2022, representing a growing global health challenge and the third most common cancer in women. 1 Papillary thyroid carcinoma (PTC) is the most prevalent type and accounts for approximately 90% of all thyroid cancer. 2 It should be noted that the rising incidence of PTC is partly attributed to the increased detection of small, indolent tumors, including subcentimeter carcinomas. Thus, while incidence rates are increasing, mortality rates remain stable.3,4 PTC is however prone to metastasize to the central lymph nodes (CLN), with 30%–60% of patients experiencing CLN metastasis.5–7 Micrometastasis to the CLN compartment is frequently observed in PTC, with reported rates as high as 80%, yet the prognostic implication of this finding is unclear.8,9
The CLN consist of three subgroups of lymph nodes. Prelaryngeal lymph nodes (PLLN), pretracheal lymph nodes (PRLN), and paratracheal lymph nodes (PALN). Anatomically, the esophagus is located posterior to the left side of the trachea, and there is a concealed space between the right recurrent laryngeal nerve (RLN), prevertebral fascia, and esophagus, which can accommodate lymphatic and adipose tissue. As a result, lymph nodes can still be present posterior to the right RLN (termed LN-prRLN). Studies have shown that when metastasis occurs in this region, the likelihood of lateral lymph node metastasis (LLNM) is five times higher compared to patients without LN-prRLN involvement.10,11
Currently, therapeutic CLN dissection (CLND) is widely accepted in PTC treatment. Opinions however remain divided regarding the value of prophylactic CLND,12–14 as it has not been conclusively shown to improve survival and carries risks such as hypoparathyroidism and RLN injury.15,16 There are no specific recommendations for LN-prRLN dissection and therefore the decision relies mainly on intraoperative findings. Although ultrasound and computed tomography can assist in detecting LLN metastasis, their sensitivity for detecting CLN, especially LN-prRLN, is limited (20%–30%). 17 Due to the unclear anatomical structure, the LN-prRLN area is also often overlooked. Although several studies have recently explored the risk factors for LN-prRLN metastasis,11,17,18 there is still a lack of effective predictive models for assessing the risk of metastasis in LN-prRLN.
This study aimed to develop a risk assessment model based on pre- and intraoperative frozen section results to predict the risk of LNM-prRLN and to guide surgical strategies for thyroid surgeons.
Materials and methods
Patient inclusion and data processing
This retrospective study was approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University (approval number: 2020-181) and included the clinical case records of 3116 patients with PTC treated at the hospital from January 2016 to December 2019. Data were accessed from January 2020 to June 2020 for research purposes. All data were de-identified before analysis, and authors had no access to personally identifiable information. The study included all patients over 18 years of age (due to distinct tumor biology and management guidelines) with a preoperative biopsy (fine needle aspiration) that confirmed PTC. All patients were clinically diagnosed as lymph node-negative (cN0), underwent at least a CLND, and had complete clinical and pathological data. Patients with a history of previous neck surgery and/or radiation therapy, diagnosis of other types of thyroid cancers, cancers located in the isthmus or bilateral cancers, and incomplete clinical or pathological data were excluded. Postoperative pathology results were used as the gold standard for diagnosing LNM-prRLN.
Data collection
Clinical variables
Sex was defined as male and female. Age was divided into two groups based on the 8th edition of the American Joint Committee on Cancer (AJCC). 19 Body mass index (BMI) was categorized according to both the Chinese and the World Health Organization (WHO) standards into three groups as follows: BMI < 18.5 (underweight), 18.5 ⩽ BMI < 24 (normal weight), and BMI ⩾ 24 (overweight).
Histopathology variables
Tumor size was defined as the largest diameter of the tumor, and extrathyroidal extension (ETE) was defined as tumor invasion into tissues beyond the thyroid capsule, including both microscopic and gross (maximum ETE involving the trachea, larynx, and recurrent laryngeal nerve (RLN)) invasion. ETE was assessed by intraoperative paraffin sectioning. The diagnostic criteria of Hashimoto’s thyroiditis (HT) included any of the following conditions: (1) thyroid peroxidase antibody level > 50 IU/mL, (2) ultrasound findings of diffuse heterogeneity, or (3) histopathological examination showing diffuse lymphocytic thyroiditis. Tumor location and multifocality were confirmed by ultrasound and intraoperative paraffin sectioning.
Intraoperative frozen pathology variables
These included lymph nodes from the PLAN, PRLN, PALN, and LN-prRLN, all of which were treated as binary variables (metastasis yes/no). The lymph node metastasis (LNM) status was determined based on the results of the final paraffin-embedded sections.
Surgical complications
Recurrent laryngeal nerve injury was assessed based on the presence of postoperative hoarseness confirmed by laryngoscopy. Postoperative hypoparathyroidism was defined as a parathyroid hormone (PTH) level below the normal reference range (as per the institutional laboratory standard).
Surgical method and intraoperative frozen pathology feature extraction
The surgical approach at the authors’ center for patients with PTC includes unilateral thyroid lobectomy and ipsilateral CLND for tumors smaller than 4 cm. If CLN metastasis (CLNM) is confirmed intraoperatively, total thyroidectomy is performed. The excised lymph node specimens are divided into categories; the PLAN, PRLN, PALN, and LN-prRLN. These lymph nodes are sequentially removed, labeled, and immediately sent for intraoperative frozen section analysis. Postoperatively, all excised specimens were subjected to histopathological examination by three independent pathologists.
Statistical analysis
The medical records and relevant data of all patients were analyzed using SPSS 25.0 statistical software (version 25.0; SPSS Inc., Chicago, IL, United States). Categorical data were expressed as proportions, whereas continuous data were processed according to their distribution. The median (interquartile range, IQR) was used for not normally distributed continuous data, and intergroup comparisons were performed by non-parametric tests (Mann–Whitney U test). For comparisons of categorical data between groups, the chi-square test (χ2 test) or Fisher’s exact test was used. For variables showing significant associations, binary logistic regression analysis was conducted, with a significance level set at p < 0.05. With this model, a nomogram was constructed using R 4.1.2, and the model’s discriminatory ability (AUC) and predictive accuracy of the model were evaluated using ROC curves. Internal validation of the predictive model was performed using calibration curves.
Results
Clinicopathologic characteristics
After strictly applying the inclusion and exclusion criteria (Fig. 1), a total of 721 patients with cN0 PTC were included. The basic characteristics of the patients are shown in Supplement Table 1. The cohort was predominantly female (71.6%), with a median age at diagnosis of 42 years (range, 18–80 years) and the majority of patients (86.4%) being under 55 years of age. Most patients had a normal BMI (68.9%), while 26.5% were overweight. The median tumor size was 14.0 mm, with 53.8% of tumors larger than 10 mm. Tumors were more frequently located in non-upper regions of the thyroid (73.2%). The LNM-prRLN rate was 19.8%. Postoperative complications were rare, with recurrent laryngeal nerve injury occurring in 0.4% and hypoparathyroidism in 2.8% of patients.

Flowchart of this retrospective study for LNM-prRLN.
Univariable analysis
In univariable analysis (Table 1), there were no significant differences between the LN-prRLN positive and negative groups in terms of tumor location, HT, BMI, or multifocality. The median tumor size in the LNM-prRLN group was 14.0 mm, while in the control group, it was 11.0 mm (p < 0.001). The proportion of male patients in the LNM-prRLN group was greater than that in the control group (36.4% vs 26.5%, p < 0.019). LNM-prRLN was associated with ETE (32.4% vs 15.6%, p < 0.001), prelaryngeal LNM (40.6% vs 20.6%, p < 0.001), pretracheal LNM (69.9% vs 45.8%, p < 0.001), and paratracheal LNM (86.0% vs 42.2%, p < 0.001). When the PLLN, PRLN, and PALN regions were combined in pairs, it was shown that LNM-prRLN was associated with the number of metastases in the PLLN and PRLN regions, the PLLN and PALN regions, and the PRLN and PALN regions separately. All these differences were statistically significant (p < 0.001).
Univariable analysis of risk factors for LNM-prRLN.
LNM-prRLN, lymph node metastasis posterior to the right recurrent laryngeal nerve; BMI, body mass index; HT, Hashimoto’s thyroiditis; ETE, extrathyroidal extension; LNM, lymph node metastasis; PLLN, prelaryngeal lymph nodes; PRLN, pretracheal lymph nodes; PALN, paratracheal lymph nodes.
Multivariable analysis
Factors with statistical significance in the univariable analysis were included in the multivariable analysis. This showed that metastasis of PLLN and PALN 1-2 (OR = 3.433, 95% CI: 1.407–8.379, p = 0.007), metastasis of PLLN and PALN ⩾ 3 (OR = 4.402, 95% CI: 1.080–15.124, p = 0.038), and the number of PALN metastases (OR = 1.360, 95% CI: 1.047–1.765, p = 0.021) were independent risk factors for LNM-prRLN, whereas ETE (OR = 0.58, 95% CI: 0.365–0.921, p = 0.021) was an independent protective factor for LNM-prRLN (Table 2) (Fig. 2).
Multivariable analysis of risk factors for LNM-prRLN.
LNM-prRLN: lymph node metastasis posterior to the right recurrent laryngeal nerve; ETE: extrathyroidal extension; LNM: lymph node metastasis; PLLN: prelaryngeal lymph nodes; PALN: paratracheal lymph nodes; OR: odds ratio; CI: confidence interval.

Forest plots of risk factors for LNM-prRLN.
Construction and validation of a risk prediction model
A nomogram was constructed based on the above risk factors (Fig. 3). The area under the ROC curve (AUC) was 0.790 (95% CI: 0.751–0.829), with a specificity of 0.578 and sensitivity of 0.860 (Fig. 4A). The nomogram revealed that the number of PALN metastases and the presence of both PALN and PLLN metastases had the greatest impact on the risk of LNM-prRLN. In addition, internal validation of the established predictive model was performed. In Fig. 4B, the solid line represents the actual incidence rate of LNM-prRLN, while the dashed line represents the predicted incidence rate. The closer the solid line is to the dashed line, the more accurate the prediction model is. This suggests that the predicted risk of LNM-prRLN closely matched the actual outcomes.

Nomogram for LNM-prRLN risk prediction.

(A) ROC curve for the logistic regression model of LNM-prRLN. (B) Calibration curve for the predictive model of LNM-prRLN.
Discussion
In recent years, the incidence of thyroid cancer has been steadily increasing due to changes in environmental factors and lifestyles, making it a growing global public health concern. 20 As the most common malignant thyroid tumor, PTC generally has a high 5-year survival rate, and most patients achieve favorable outcomes after surgical treatment. However, 14% to 30% of patients with PTC require reoperation for various reasons. 10 Notably, LNM recurrence in the LN-prRLN accounts for 60% of recurrence in the CLN. 21 The necessity of routine dissection of LN-prRLN remains controversial. Some researchers argue that the metastasis rate in this region is relatively low and that meticulous intraoperative dissection is technically challenging, increases the risk of RLN injury, parathyroid damage, and chyle leakage.17,22 On the other hand, the reported incidence of LNM-prRLN in PTC patients ranges from 9.36% to 38.27% and thus it can be argued that LN-prRLN dissection should be performed.10,23,24 In the present study, the LNM-prRLN rate was 19.8%, which is consistent with previous findings.
Multiple studies have sought to identify predictors of LNM‑prRLN. Zhu et al. 18 demonstrated that metastasis in prelaryngeal (PLLN), pretracheal (PRLN), and paratracheal (PALN) nodes serve as independent risk factors. Anatomically, these lymph node groups are more accessible during surgery and can often be dissected without substantially prolonging operative time or increasing complication rates.25,26 The present results similarly identified PALN metastasis as an independent predictor, with each additional positive PALN increasing the odds of LN‑prRLN involvement by 1.360‑fold. Luo et al. 27 further suggested that LN‑prRLN dissection should be considered when ⩾2 PALN metastases are present. Extending these findings, this study examined the combined predictive value of CLN subgroup involvement. It was observed that metastasis in both PLLN and PALN, particularly when the total number of metastases in these two regions reaches three or more, heightens the risk of LNM‑prRLN by 4.402 times, offering a practical intraoperative guide for surgical decision‑making.
While macroscopic LNM is widely recognized as a risk factor for local recurrence and potentially worse prognosis,28,29 the significance of micrometastases (⩽2 mm) remains less clear and continues to be debated. 30 This distinction is especially pertinent when interpreting predictive models such as ours. In general, CLN micrometastasis is often considered to have minimal prognostic impact and may not alter staging or management in otherwise low‑risk patients.3,8 The unique anatomical location posterior to the RLN, with its proximity to critical neurovascular and visceral structures, may confer distinct surgical and prognostic implications, regardless of metastasis size, due to the technical difficulty of achieving complete resection in this region. Hence, the clinical interpretation of a positive LN‑prRLN finding, particularly when involving micrometastasis, requires careful contextualization. Future studies incorporating long‑term follow‑up and standardized pathologic size assessment are needed to clarify whether LN‑prRLN micrometastases warrant more aggressive management.
Extrathyroidal extension (ETE) is a well-established risk factor for LNM PTC.27,31 Recent studies have shown that ETE is an independent predictive factor for LNM-prRLN, suggesting that ETE is associated with tumor aggressiveness and poor prognosis.27,32 Nevertheless, its role as a prognostic factor remains controversial, with some evidence contradicting this association.23,33–35 One unanticipated finding in this study was that the difference between LNM-prRLN with or without ETE (OR = 0.58, p = 0.021) was not the same as previously reported. A possible explanation is that, in our data, ETE was defined as any extent of invasion, ranging from minimal to widespread. As highlighted in other studies,35–37 minimal capsular invasion carries little to no risk of nodal spread, whereas gross ETE is strongly correlated with LNM. Future research should subclassify ETE to better delineate its association with metastasis in specific anatomical subsites such as the LN‑prRLN.
We acknowledge that prophylactic dissection of clinically uninvolved lymph nodes carries inherent risks, principally RLN injury and hypoparathyroidism, that must be balanced against the potential consequences of leaving behind nodal disease. In this series, rates of RLN injury (0.4%) and transient hypoparathyroidism (2.8%) were notably low despite routine LN‑prRLN dissection. These figures compare favorably with some published RLN injury rates of approximately 6.0% in thyroidectomy patients without routine CLD, 38 suggesting that in high‑volume centers with routine intraoperative nerve monitoring, the incremental risk of LN‑prRLN dissection may be minimal. Given that up to 60% of central‑compartment recurrences occur posterior to the RLN 21 and that reoperation in this region is associated with significantly higher morbidity, a selective, nerve‑monitored dissection during the initial surgery, guided by imaging or intraoperative findings, may prevent the need for more hazardous revision procedures. This approach should be implemented in centers with appropriate surgical expertise and monitoring capabilities to optimize risk‑benefit balance.
Limitations
This study has several limitations. First, its single‑center retrospective design introduces inherent selection bias, and the relatively modest sample size may affect generalizability to the broader PTC population. Second, the prediction model was only internally validated, exhibits relatively low specificity, and lacks long‑term follow‑up data. Third, we did not differentiate between micro‑ and macro‑metastases in LN‑prRLN, which may influence the clinical interpretation of the results. Finally, the model’s reliance on intraoperative frozen‑section analysis may limit its applicability in centers where this technique is not routinely available and as such it is not intended for preoperative risk assessment. Therefore, future prospective, multi‑center studies with larger cohorts and external validation are warranted to further refine and confirm risk factors for LNM‑prRLN in patients with cN0 PTC.
Conclusion
LN-prRLN metastasis is not uncommon in cN0 PTC patients. LN-prRLN dissection is necessary when a patient has high-risk factors affecting LNM-prRLN (the number of metastases in the PALN or combined metastasis in the PLLN and PALN). This model, that integrates intraoperative frozen pathology findings with valuable predictive information from subgroups of CLN and clinical risk factors, provides rapid and accurate data during the operation. It allows surgeons to adjust intraoperative decision-making, including whether to dissect the LN-prRLN, based on available frozen section results.
Supplemental Material
sj-docx-1-sjs-10.1177_14574969261440429 – Supplemental material for Predicting lymph node metastasis posterior to the right recurrent laryngeal nerve in papillary thyroid carcinoma: A risk model based on prelaryngeal and paratracheal lymph nodes
Supplemental material, sj-docx-1-sjs-10.1177_14574969261440429 for Predicting lymph node metastasis posterior to the right recurrent laryngeal nerve in papillary thyroid carcinoma: A risk model based on prelaryngeal and paratracheal lymph nodes by Chang Deng, Zhixin Yang, Yijia Cao, Chun Huang, Jing Zhou and Xinliang Su in Scandinavian Journal of Surgery
Footnotes
Acknowledgements
The authors appreciate all participants, including surgeons, pathologists, and hospitals, for their important contributions to the establishment of the cohort and the resulting work.
Author contributions
C.D., J.Z., and X.S. designed this study. C.D., Z.Y., Y.C., C.H., and J.Z. collected the data. C.D. and Z.Y. analyzed the data. All authors contributed to the article and approved the submitted version.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Bethune Charitable Foundation (BCF) 2024, Z04J2024E107-B-09, China, and the Science and Technology Research Program of Chongqing Municipal Education Commission of China (Grant No. KJQN202501121).
Ethics statement
These human subject studies were approved by the ethics committee/institutional review board (IRB) of the First Affiliated Hospital of Chongqing Medical University. They were conducted in accordance with the ethical standards outlined in the 1964 Declaration of Helsinki and its subsequent amendments. The study was approved by the ethics committee/IRB under the approval number: 2020 Research Ethics 2020-181.
Data availability statement
The datasets generated or analyzed during the current study are available from the corresponding author on reasonable request.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
