A nomogram-based risk stratification strategy to optimize central lymph node dissection in cN1a papillary thyroid cancer with Hashimoto’s thyroiditis

Introduction

Hashimoto’s thyroiditis (HT) and papillary thyroid carcinoma (PTC) frequently coexist, both conditions contributing to central lymph node enlargement in the neck, which is classified as cN1a stage through ultrasound or other imaging modalities (1,2). However, current diagnostic approaches are inadequate in differentiating whether the enlarged lymph nodes are a result of inflammatory enlargement due to HT or lymph node metastasis from PTC (3). Although numerous guidelines advocate for fine-needle aspiration (FNA) biopsy for lymph nodes exceeding 8–10 mm, many patients present with lymph nodes smaller than this threshold or are unable to undergo FNA, leading to misdiagnosis and unnecessary central lymph node dissection (CLND) in cN1a PTC patients with HT (4,5).

CLND poses risks of complications, including hypoparathyroidism, recurrent laryngeal nerve injury, and lymphatic leakage (6). The presence of HT, characterized by an increase in the number and volume of lymph nodes, significantly heightens the risk of these complications (7,8). Consequently, enhancing the accurate identification of central lymph node metastasis (CLNM) in patients with concurrent HT is crucial to avoid unnecessary CLND.

While most guidelines recommend CLND for patients with clinical node-positive stage cN1a, clinicians often overlook the possibility that cN1a may be a false positive for CLNM due to coexisting HT (4,9). Notably, the 2025 American Thyroid Association (ATA) guidelines explicitly advise against prophylactic CLND for low-risk clinical node-negative (cN0) PTC patients (5). However, there is currently no research addressing whether low-risk cN1a patients, who may have false positive CLNM attributable to coexisting HT, can be managed similarly to cN0 PTC patients.

This study investigates the influence of various risk factors on CLNM by analyzing how coexisting HT affects the false-positive rate of CLNM and lymph node metastatic burden in cN1a PTC patients. In addition, this study established a predictive model and risk stratification, and proposed providing a scientific basis for refining the indications for CLND, thereby facilitating a more personalized and precise management approach for cN1a PTC patients with HT. We present this article in accordance with the TRIPOD reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0065/rc).

Methods

Patient information

We conducted a retrospective study on patients with cN1a PTC at Guangdong Provincial People’s Hospital from October 2015 to December 2025, partitioned into a training set (Oct 2015–Oct 2023) and a subsequent temporal validation set (Nov 2023–Dec 2025) to evaluate the predictive performance of the nomogram. All cN1a PTC patients underwent primary thyroidectomy and at least one side of CLND. Patients who were older than 70 years or younger than 18 years, who had already undergone preoperative central lymph node FNA, those with incomplete important information such as preoperative ultrasound reports, and those who had undergone preoperative I¹³¹, ablation, or previous head and neck tumor surgery were excluded. The same inclusion and exclusion criteria were used for both the training and validation sets. Excluding FNA cases ensures that the model is tested only on a purely imaging-based cohort of cN1a patients, which is more representative of the diagnostic challenges faced by patients with HT.

The collected patient information included general details such as age and gender, ultrasound examination-related information including tumor lesion characteristics and lymph node characteristics, as well as postoperative pathological information such as the presence of CLNM. The ultrasound examinations were performed using HI Vision 900, HI Vision Ascendus, and HI Vision Preirus color US units (with US elasticity imaging capability) from Hitachi. The ultrasound imaging features of each patient were retrospectively reviewed by two independent radiologists with over 10 years of experience in thyroid ultrasound imaging.

The diagnostic criteria for HT include ultrasound findings of lymphocytic thyroiditis, accompanied by at least one of serum levels of thyroid peroxidase antibody (TPOAb) and thyroglobulin antibody (TgAb) exceeding the upper limit of normal. This diagnosis further verified through postoperative pathological examination, based on the presence of diffuse lymphoplasmacytic infiltration, germinal centers, and enlarged epithelial cells with large nuclei and eosinophilic cytoplasm. The criteria for CN1a defined in this study are lymph nodes visible on ultrasound plus any of the following high-risk factors: disappearance of the fat hilum, round shape, hyperechoic, cystic changes, calcification, and rich peripheral vascularity [according to the 2015/2025 ATA guidelines (4,5)]. For all patients with preoperative ultrasound diagnosis of suspected CLNM (cN1a), our center routinely performs therapeutic CLND. The area where cN1a located is in the central neck compartment (level VI). The standardized surgical boundaries are: upper boundary is the hyoid bone, lower boundary is the suprasternal notch, and lateral boundaries are the common carotid arteries. The diagnosis of CLNM (pN1) was confirmed using postoperative paraffin-embedded pathological sections. Indicators with an unknown value accounting for more than 30% were excluded from univariate and multivariate analyses.

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (KY-2024-246-01) (Guangzhou, China). Informed consent was obtained from all individual participants included in the study.

Statistical analysis

Statistical analysis was conducted using R software (version 4.2.2; https://www.r-project.org). Differences were considered statistically significant when two-tailed P<0.05. Categorical variables were presented as numbers and percentages. The rates between two groups of categorical variables were assessed using a Chi-squared test or Fisher’s exact test for univariate analysis. Risk factors were evaluated using a logistic regression model. The rms package was utilized for univariate and multivariate analyses as well as for building predictive models. The predictive value of these factors was measured using the area under the receiver operating characteristic (ROC) curve (AUC). The calibration curve was drawn using the bootstrap 1,000 resampling method. Decision curve analysis (DCA) was employed to assess the feasibility and application value of this model in clinical practice. We determined the optimal critical value for lymph node size by using the optimal critical value of the ROC curve.

Results

HT increases false-positive cN1a diagnosis but doesn’t affect lymph node metastatic burden

This study included 627 cN1a PTC patients, of whom 150 (23.9%, 150/627) had coexisting HT. Based on the final paraffin pathology results, we categorized cN1a patients into those with true CLNM (pN1a, 459 cases) and those with false-positive CLNM (pN0, 168 cases). Our findings indicate that among cN1a PTC patients with coexisting HT, the proportion of false-positive CLNM was one-third, which was significantly higher than that observed in patients without HT [33.3% (50/150) vs. 24.7% (118/477), P<0.05]. The total number of CLNs dissected was significantly higher in the HT group compared to the non-HT group {7 [interquartile range (IQR), 4–11] vs. 6 (IQR, 3–10), P=0.04}. The number of metastatic lymph nodes did not differ significantly between the two groups [1 (IQR, 0–3.75) vs. 2 (IQR, 1–5), P=0.28], and the metastatic ratio of CLN [0.2 (IQR, 0–0.6) vs. 0.5 (IQR, 0.08–0.83), P=0.03] was lower in the HT group. Therefore, we conclude that coexisting HT significantly increases the false-positive rate of CLNM, but did not increase the number or proportion of metastatic lymph nodes in cN1a PTC patients (Table 1).

Analysis of risk factors and nomogram for CLNM in cN1a PTC patients with HT

To further investigate the clinical characteristics of CLNM in cN1a PTC patients with HT and identify risk factors for CLNM, we conducted both univariate and multivariate analyses of the general clinical and ultrasound characteristics (including thyroid nodules and lymph nodes) of these 150 patients. The results from the univariate analysis indicated that CLNM was significantly associated with patients under the age of 55 years, male gender, larger tumor size, bilateral gland involvement, and lymph nodes size larger than 0.7 cm, round shape, cystic change, and microcalcification. Moreover, the multivariate analysis confirmed that younger age, male gender, larger tumor size, bilateral gland involvement, lymph node round shape, cystic change, and microcalcification were independent risk factors for CLNM (Table 2).

To predict the risk of developing CLNM in cN1a PTC patients with HT, we constructed a nomogram for independent risk factors based on a training set analysis of these 150 patients (Figure 1A). We divided the cohort into very low-risk, low-risk, and medium-high-risk groups based on risk coefficients. The very low-risk group corresponds to a CLNM risk of less than 5% and a nomogram score of less than 107.4. Approximately 25 out of 150 individuals in this group, and none of them actually developed CLNM, resulting in a negative predictive value (NPV) of 100%. The low-risk group corresponds to a CLNM risk of 5–15% and a nomogram score of 107.4–133.4. Approximately 36 out of 150 individuals in this cohort, among whom only 2 actually developed CLNM, resulting in an NPV of 94.4%. The medium-high-risk group corresponds to a CLNM risk of greater than 15% and a nomogram score greater than 133.4. Approximately 44 individuals in this cohort, with more than 7 false-negative cases, resulting in an NPV of less than 84.1% (Table 3).

Figure 1 Nomogram construction for predicting CLNM in cN1a PTC patients with HT. (A) Clinical factor-based nomogram used for predicting the CLNM in cN1a PTC patients with HT. The ROC curves (B,E), calibration curves (C,F), and DCA (D,G) for the training and validation cohorts, respectively. AUC, area under the ROC curve; CLN, central lymph node; CLNM, central lymph node metastasis; DCA, decision curve analysis; HT, Hashimoto’s thyroiditis; PTC, papillary thyroid carcinoma; ROC, receiver operating characteristic.

Subsequently, we further evaluated the model and collected a subsequent temporal validation set (n=42) for verification. The baseline characteristics of patients in the validation set were not significantly different from those in the training set (Table 4). The AUC of the training set was 0.948 [95% confidence interval (CI): 0.914–0.982], indicating good clinical prediction ability of the model (Figure 1B). The calibration curve of the training set showed a close alignment between the predicted results of the nomogram model and the actual results. The mean absolute error was 0.036, indicating good calibration of the model (Figure 1C). Additionally, the DCA of the training set further confirmed the strong clinical practicality of the model (Figure 1D). The validation set results further confirmed the robustness of the model. Its AUC (0.942; 95% CI: 0.8579–1) was highly similar to that of the training set (Figure 1E), and the calibration curves showed excellent consistency (Figure 1F). Furthermore, the DCA results on the validation set indicated that the model has promising applications in clinical practice and can effectively identify low-risk cN1a patients (Figure 1G).

Discussion

In recent years, several guidelines have recommended simple thyroid lobectomy for low-risk cN0 PTC patients without prophylactic CLND (4,10,11). The 2025 ATA guidelines further advise against routine prophylactic CLND in T1–2 PTC patients due to the low recurrence rate and favorable prognosis (5,12). For low-risk patients, there is an increasing emphasis on preserving function while ensuring complete tumor resection to minimize damage to the recurrent laryngeal nerve and parathyroid glands (13).

Currently, CLND is a consensus procedure for patients diagnosed with cN1a stage PTC (10,11). However, the incidence of false positives and the subsequent identification and management of patients with false positives in cN1a patients with HT remain poorly understood (14). Previous studies have indicated that PTC patients with HT experience a higher false-positive rate and lower specificity in preoperative central lymph node ultrasound examinations compared to PTC patients without HT (15). Furthermore, some studies have reported an increased proportion of non-metastatic central lymph nodes lacking a fat hilum in PTC patients with HT, suggesting that HT may interfere with the preoperative ultrasound assessment of central lymph nodes and elevate the incidence of non-fat hilum in benign lymph nodes (16). Our findings also suggest that HT increases the number of lymph nodes dissected, but does not increase the number or proportion of metastatic lymph nodes; furthermore, in cN1a PTC patients with HT, lymph node size and the disappearance of the lymphohepatic hilum are not independent risk factors for CLNM. Given the substantial number of PTC patients, this suggests that a considerable proportion of false-positive diagnoses of CLNM may be present among cN1a PTC patients undergoing CLND.

To better identify patients with false-positive CLNM, we conducted univariate and multivariate analyses on patients with cN1a PTC and HT. Subsequently, we established a nomogram to predict the probability of CLNM occurrence. The results indicated that gender, age, tumor size, and bilateral gland involvement were independent risk factors for CLNM in cN1a PTC patients with HT. Our data indicate that up to 27 out of 29 males with coexisting HT exhibit true CLNM. We hypothesize that sex steroids play a crucial role in this process, as compelling evidence supports their involvement in various animal models of autoimmunity (17). Female sex hormones have been shown to inhibit the function of dendritic cells in initiating immune responses, reduce T cell apoptosis in patients, and induce polyclonal activation of human B cells. Conversely, testosterone exerts different effects by inhibiting the production of immunoglobulin G (IgG) and immunoglobulin M (IgM), as well as alleviating induced chronic autoimmune thyroiditis (18,19). These hormonal differences may elucidate why females constitute the majority of false-positive central node cN1a PTC patients, and why HT significantly increases the false-positive rate of CLNM in female PTC patients.

Two significant predictors of CLNM in patients with cN1a PTC and coexisting HT are tumor size exceeding 2 cm and bilateral gland involvement. Prior studies have demonstrated that the presence of HT correlates with upregulation of interleukin-2 (IL-2) and human leukocyte antigen class I (HLA-I) molecules, subsequently enhancing anti-tumoral T cell immunity. This phenomenon is one of the reasons many studies have established that HT may exert an inhibitory effect on tumor growth (20). Consequently, we hypothesize that in the context of HT, larger tumor size and bilateral lobe involvement may indicate a more significant disease burden, potentially increasing the likelihood of CLNM in these patients. In addition, our multivariate analysis results also indicated that round, cystic, and microcalcified lymph nodes on ultrasound were independent risk factors for CLNM, excluding lymph node size, absence of the fat hilum, and hyperechogenicity. This suggests that both inflammatory and metastatic lymph nodes may exhibit lymph node enlargement, absence of the fat hilum, and hyperechogenicity, and we should pay more attention to ultrasound features unique to other metastatic lymph nodes.

In addition to preoperative assessment, intraoperative adjunctive techniques play a crucial role in determining the necessity and extent of CLND. Direct visualization and palpation during surgery are the most straightforward methods for assessing lymph nodes (21). However, in the context of HT, inflammatory lymph nodes often appear rubbery or enlarged, which can be very similar in texture to metastatic lymph nodes. This clinical ambiguity often leads to over-dissection in patients, and our model can serve as a necessary preoperative screening tool to help surgeons interpret these intraoperative findings more objectively. Sentinel lymph node biopsy (SLNB) is a promising tool for identifying the first station of lymphatic drainage (22). Although its application in PTC is still under investigation, it has the potential to complement our nomogram. For patients classified as low-risk in our model, a negative intraoperative SLNB result provides greater assurance. Frozen section analysis (FSA) remains the gold standard for intraoperative decision-making (23). We believe that for patients with borderline risk scores on the nomogram, intraoperative FSA analysis of the most suspicious lymph nodes found during exploration can provide clear pathological evidence to guide the scope of surgery.

This study has several limitations that warrant attention. First, we must acknowledge that as a single-center retrospective study, this cohort lacks external validation. Although the reliability of the results has been increased by including time-validated cohorts, conducting multi-center collaborative studies in the future is a necessary step for clinical promotion. Secondly, being a retrospective single-center study conducted at a teaching hospital introduces potential selection bias, and future research needs to incorporate more pathological information (such as micrometastatic lymph nodes) and follow-up data to validate the model’s effectiveness. Furthermore, it is important to clarify the clinical positioning of our proposed model. Although the nomogram offers a valuable non-invasive tool for risk stratification in patients with PTC and HT, it is designed to function as a clinical auxiliary rather than a replacement for established diagnostic standards. FNA and intraoperative frozen section biopsy remain the cornerstones of definitive diagnosis. Our model aims to provide surgeons with supplementary decision support, enabling a more nuanced and individualized approach to central lymph node management.

Conclusions

In summary, our findings suggest that incorporating HT complicates preoperative central lymph node assessment during PTC. While further validation in conjunction with clinical presentation is needed, this study provides a nomogram as a predictive framework that can serve as an effective tool for assessing the risk of CLNM and guiding decisions regarding CLND.

Acknowledgments

We would like to express thanks to all participants.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0065/rc

Data Sharing Statement: Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0065/dss

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0065/prf

Funding: This work was supported by the National Natural Science Foundation of China (No. 82303152 to L.W.), the Science and Technology Program of Guangzhou (No. 2024A04J5132 to L.W.), and the Natural Science Foundation of Guangdong Province (No. 2020A1515010126 to Z.W.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0065/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (KY-2024-246-01) (Guangzhou, China). Informed consent was obtained from all individual participants included in the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Yao S, Zhang H. Papillary thyroid carcinoma with Hashimoto’s thyroiditis: impact and correlation. Front Endocrinol (Lausanne) 2025;16:1512417. [Crossref] [PubMed]
2. Báez Berríos AI, Monaghan M, Brandwein-Weber M, et al. Investigating the Association Between Hashimoto’s Thyroiditis and Papillary Thyroid Cancer. Head Neck 2025;47:1214-22. [Crossref] [PubMed]
3. Wang L, Zhang L, Wang D, et al. Predicting central cervical lymph node metastasis in papillary thyroid carcinoma with Hashimoto’s thyroiditis: a practical nomogram based on retrospective study. PeerJ 2024;12:e17108. [Crossref] [PubMed]
4. Haugen BR, Alexander EK, Bible KC, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid 2016;26:1-133. [Crossref] [PubMed]
5. Ringel MD, Sosa JA, Baloch Z, et al. 2025 American Thyroid Association Management Guidelines for Adult Patients with Differentiated Thyroid Cancer. Thyroid 2025;35:841-985. [Crossref] [PubMed]
6. Atallah K, Awny S, Abdelwahab K, et al. Morbidity patterns and long-term outcomes of central lymph node dissection in thyroid cancer patients. Sci Rep 2025;15:23527. [Crossref] [PubMed]
7. Baud G, Jannin A, Marciniak C, et al. Impact of Lymph Node Dissection on Postoperative Complications of Total Thyroidectomy in Patients with Thyroid Carcinoma. Cancers (Basel) 2022;14:5462. [Crossref] [PubMed]
8. Huang Y, Zeng F, Song J, et al. Construction and application of a nomogram model for early stage central lymph node metastasis in papillary thyroid cancer combined with Hashimoto’s thyroiditis. Braz J Med Biol Res 2026;58:e14881. [Crossref] [PubMed]
9. Filetti S, Durante C, Hartl D, et al. Thyroid cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up†. Ann Oncol 2019;30:1856-83. [Crossref] [PubMed]
10. Patel KN, Yip L, Lubitz CC, et al. The American Association of Endocrine Surgeons Guidelines for the Definitive Surgical Management of Thyroid Disease in Adults. Ann Surg 2020;271:e21-93. [Crossref] [PubMed]
11. Orloff LA, Wiseman SM, Bernet VJ, et al. American Thyroid Association Statement on Postoperative Hypoparathyroidism: Diagnosis, Prevention, and Management in Adults. Thyroid 2018;28:830-41. [Crossref] [PubMed]
12. Kim BC, Pak SJ, Kwon D, et al. Clinically Significant Central Lymph Node Metastasis is Not Common in Patients with Clinically N0 Papillary Thyroid Carcinoma on Both Ultrasonography and CT. Thyroid 2025;35:415-23. [Crossref] [PubMed]
13. Zocchi J, Giugliano G, Mossinelli C, et al. Central Neck Dissection in Papillary Thyroid Carcinoma: Benefits and Doubts in the Era of Thyroid Lobectomy. Biomedicines 2024;12:2177. [Crossref] [PubMed]
14. Wei Y, Sun P, Chang C, et al. Ultrasound-based Nomogram for Predicting the Pathological Nodal Negativity of Unilateral Clinical N1a Papillary Thyroid Carcinoma in Adolescents and Young Adults. Acad Radiol 2023;30:2000-9. [Crossref] [PubMed]
15. Song E, Jeon MJ, Park S, et al. Influence of coexistent Hashimoto’s thyroiditis on the extent of cervical lymph node dissection and prognosis in papillary thyroid carcinoma. Clin Endocrinol (Oxf) 2018;88:123-8. [Crossref] [PubMed]
16. Tan HL, Nyarko A, Duan SL, et al. Comprehensive analysis of the effect of Hashimoto’s thyroiditis on the diagnostic efficacy of preoperative ultrasonography on cervical lymph node lesions in papillary thyroid cancer. Front Endocrinol (Lausanne) 2022;13:987906. [Crossref] [PubMed]
17. Forsyth KS, Jiwrajka N, Lovell CD, et al. The conneXion between sex and immune responses. Nat Rev Immunol 2024;24:487-502. [Crossref] [PubMed]
18. Fairweather D, Beetler DJ, McCabe EJ, et al. Mechanisms underlying sex differences in autoimmunity. J Clin Invest 2024;134:e180076. [Crossref] [PubMed]
19. Hoffmann JP, Liu JA, Seddu K, et al. Sex hormone signaling and regulation of immune function. Immunity 2023;56:2472-91. [Crossref] [PubMed]
20. Hu JQ, Lei BW, Wen D, et al. IL-2 enhanced MHC class I expression in papillary thyroid cancer with Hashimoto’s thyroiditis overcomes immune escape in vitro. J Cancer 2020;11:4250-60. [Crossref] [PubMed]
21. Yuan Q, Yang Y, Li C, et al. Prophylactic Central Neck Dissection Based on Preoperative Imaging and Intraoperative Surgeon’s Palpation Versus Total Thyroidectomy Alone for Papillary Thyroid Cancer. J Surg Res 2023;290:126-32. [Crossref] [PubMed]
22. Yan X, Zeng R, Ma Z, et al. The Utility of Sentinel Lymph Node Biopsy in Papillary Thyroid Carcinoma with Occult Lymph Nodes. PLoS One 2015;10:e0129304. [Crossref] [PubMed]
23. Kim MJ, Kim HJ, Park CS, et al. Frozen section analysis of central lymph nodes in papillary thyroid cancer: the significance in determining the extent of surgery. Gland Surg 2022;11:640-50. [Crossref] [PubMed]