Thyroid cancer is becoming one of the most common malignant human tumors. The incidence of thyroid cancer has continued to rise around the world in the past few decades (1, 2). Studies have shown that papillary thyroid carcinoma (PTC) has high rates of central compartment lymph nodes metastasis (CLNM), ranging from 20% to 90% (3). Prophylactic central lymph node dissection remains controversial (4–6). Some investigators believe that micrometastasis to lymph nodes does not affect the survival time of patients, while other studies have pointed out that prophylactic lymph node dissection could increase the recurrence free-survival rate and reduce postoperative serum thyroglobulin levels, yielding accurate staging to assist further treatment. Moreover, due to the use of nerve monitoring and carbon nanoparticle negative imaging (7, 8), permanent damage to the recurrent laryngeal nerve and parathyroid gland is rarely found during dissection of the initial central compartment lymph nodes. Because PTC has a good postoperative survival time, more attention is paid to the probability of postoperative recurrence. when PTC is recurrent and reoperated, it is more likely to cause permanent damage to the recurrent laryngeal nerve and parathyroid gland due to changes in anatomical structure.

CLNM is a manifestation of tumor aggressiveness and an indicator of tumor prognosis. There are many studies assessing the risk factors for CLNM (9–12). However, studies assessing CLNM of thyroid cancer are based on the overall population without age stratification. It is well known that age is a significant factor affecting the biological behavior of thyroid cancer (13). The 8^th edition of American Joint Committee on Cancer guidelines indicate elderly patients have decreased disease-free survival (14). Due to different hormone levels and immune states in patients of different ages (15–18), distinct patterns of CLNM may be present.

The purpose of this study was to explore central compartment lymph nodes that have distinct metastatic patterns in different age groups and to assist in clinical decision-making. At the same time, we examined the potential differences in the biological behavior of thyroid cancer in different ages.

In the clinical treatment of thyroid papillary carcinoma, there are still many controversies in certain aspects (19, 20), including the scope of thyroid papillary gland resection and the choice between central and lateral neck prophylactic dissection. In the latest ATA guidelines (20), it was pointed out that in cases with metastasis lymph nodes before the operation, therapeutic central compartment lymph node dissection is recommended, but in cases without central compartment lymph node metastasis before the operation, preventive lymph node dissection is only conducted in progressive disease. The guideline was based on the survival time of patients, not postoperative recurrence, and some data (4) showed that preventive lymph node dissection can reduce the postoperative recurrence of thyroid papillary carcinoma. Because of the shielding of bone and trachea, the detection rate of central compartment lymph nodes by preoperative ultrasound is very low, with a sensitivity of less than 30% (21). Metastatic lymph nodes in PTC will have the characteristics of increased size, rounded shape, absence of a visible hilum, irregular reflection pattern, unsharp borders, cystic change, calcifications, and nonhilar vasculature (22, 23). However, if PTC is complicated with Hashimoto’s thyroiditis, we may obtain false-positive results. In our center, all patients diagnosed with thyroid papillary carcinoma underwent routine prophylactic central compartment lymph node dissection at least on one side. In all of patients, only 21 patients described the abnormal changes of central compartment lymph node at ultrasonography, of which 16 were proved to have CLNM. In the context of this debate, a more sensitive and accurate method for identifying preoperative central compartment lymph nodes becomes important.

Three systematic reviews (9, 11, 12) of papillary thyroid carcinoma showed gender, age, tumor size, multifocality, capsular invasion, lymphovascular invasion, tumor location, lymphocytic thyroiditis, bilateral tumor, extrathyroidal extension, lateral cervical lymph node metastasis, histological subtype of the tumor and BRAF gene mutations may be risk factors for central compartment lymph node metastasis. There are other studies on proteomics and fibronectin expression in CLNM (24, 25). In almost all studies (26–28), age, tumor size, gender, multifocality and extrathyroidal extension were considered risk factors for CLNM, but all these reports were based on the overall population without age stratification. It is well-known that age is an important factor affecting the biological behavior of the thyroid, and an important prognostic indicator in thyroid cancer (13). Differentiated thyroid carcinoma is the only tumor with age included in TNM staging. In the 8^th edition of AJCC (14), elderly patients are considered to have a higher death risk; however, PTC shows more aggressive biological behavior in younger groups (29, 30). Studies have shown that the types of gene mutations in PTC are different in distinct age stratification (31). The importance of age is also reflected in other risk predictive models, including AMES, DAMES and MACIS. In this study, we found that the proportion of CLNM was higher in the younger group, and the proportion of males (P<0.05) and Hashimoto’s thyroiditis incidence (P<0.05) in younger patients were higher than in older patients.

Based on different age groups, we found that papillary thyroid cancer has different metastatic patterns. In the multi-factor study, it was found that in the subgroup with age <45 years, tumor size, extrathyroidal extension, Hashimoto’s thyroiditis and serum thyroglobulin levels were independent risk factors for CLNM. However, in the subgroup with age ≥45 years, tumor size, extrathyroidal extension, unclear margin under ultrasonography and male gender were independent risk factors for CLNM. Previous studies based on the overall population without age stratification also found that in different studies, factors affecting CLNM are different, which may be caused by the uneven distribution of the population. In this study, as in the overall population, tumor size, extrathyroidal extension, male gender, younger age and calcification were independent risk factors for CLNM. In the present study, we divided the overall population into different subgroups according to age, and the predictive model had a higher area under the ROC curve than the model constructed according to the overall population. Whether this difference is caused by different tumor subtypes or the body’s immune status remains to be further studied (15–17). This would enable more accurate clinical decisions in patients of different ages.

In addition, we found that thyroglobulin level was an important factor associated with CLNM in the subgroup with age <45 years. In previous studies (32, 33), serum thyroglobulin could generally be assessed by lymph node puncture before surgery to determine the status of lymph node metastasis or tumor recurrence after surgery. Besides, previous studies (27, 28) have shown that male patients are more likely to have CLNM in PTC, which was reflected in patients older than 45 years, but this trend was not obvious in younger patients. It is well known that differences in sex hormones are more common in younger patients, so the above finding is intriguing. It is worth mentioning that in the younger group, Hashimoto’s thyroiditis may be a protective factor of CLNM, but not in older cases. Hashimoto’s thyroiditis represents an autoimmune inflammation of the thyroid gland that is common in young patients. Studies have shown that Hashimoto’s thyroiditis is a stimulating factor in the occurrence of thyroid papillary carcinoma through immune activation. There are similar findings in existing studies (34–36), with no indication of the role of age.

Our research had limitations as a single-center study with no external validation. Further multi-center studies are required for verifying these findings. In addition, due to the limited number of patients, low AUCs were obtained, and more data are needed to make our results more stable. In addition, the focus of our research is to predict whether there is CLNM, but it cannot be used to predict the number and size of lymph node metastasis, for which we need do further research. Furthermore, long-term follow-up is needed to compare the postoperative recurrence risk of pN1 patients in different age groups, in order to further clarify the impact of age on tumor metastasis and recurrence.

Central compartment lymph nodes of papillary thyroid carcinoma in different age groups may have distinct patterns. The effects of extrathyroidal extension and tumor size in CLNM are reflected in all age groups. In young patients, preoperative serum thyroglobulin levels and the presence of Hashimoto’s thyroiditis may affect CLNM, while gender is not significantly important. In older cases, males were more likely to have CLNM. These differences may be explained by different hormone levels and immune states in patients of different ages, which may inform decision-making for individual patients in the clinic.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

CY, XH, ZC, and YW contributed to the conception and design of the study. CY, ZC, and YW organized the database. CY performed the statistical analyses, wrote the first draft of the manuscript, and wrote sections of the manuscript. All authors contributed to manuscript revision and read and approved the submitted version. All authors contributed to the article and approved the submitted version.

This work was partially supported by the National Natural Science Foundation of China (81672641).

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.