Influence of Obesity Parameters on Different Regional Patterns of Lymph Node Metastasis in Papillary Thyroid Cancer

Objective Obesity increases risk of thyroid cancer. However, the association between obesity and the progression of papillary thyroid cancer (PTC) remains controversial. This retrospective study aimed to explore the relationship between obesity and regional patterns of lymph node metastasis (LNM) in PTC. Methods We retrospectively reviewed data from 1015 patients with PTC. We calculated obese parameters, such as body mass index (BMI), body fat percentage (BFP), and body surface area (BSA). Logistic regression models were used to assess associations between obese parameters and the rate of lymph node metastasis (LNM), number of LNM, pattern of LNM, and lymph node ratio (LNR). Results Higher BMI was not associated with different regional patterns of LNM in PTC. In men with PTC, high BFP was an independent predictor of total LNM, central lymph node metastasis (CLNM), total lateral lymph node metastasis (LLNM), multiple lateral lymph node metastasis, and simultaneous metastasis in lateral compartment. In addition, male patients with high BFP had higher central LNR and higher number of CLNM. For women, high BSA was an independent predictor of LLNM and level IV metastasis. Female patients with high BSA had higher number of CLNM. Conclusion BFP and BSA, possibly influenced by gender, were positively associated with the number and risk of LNM in different regions of PTC patients. However, BMI was not the predictor for aggressiveness of PTC in terms of LNM. Clinical decision-making for regional LNM in PTC patients should consider the factor of obesity.


Introduction
Excessive accumulation of body fat can lead to obesity. In recent decades, the incidence of obesity has risen dramatically and has become a major global public health problem [1]. Cancer and obesity are closely related, especially for thyroid disease, previous studies have found a positive relationship between obesity and thyroid cancer [1][2][3]. A recent systematic review reported that obese people have a 55% increased risk of developing thyroid cancer compared with lean people [1]. Schmid et al. [4] reported that obesity only increased the risk of papillary thyroid cancer (PTC), follicular thyroid cancer, and anaplastic thyroid cancer, indicating a potential dependence on tumor type and histological specifcity. In addition to morbidity, some authors have noted that obesity is associated with aggressive cancer histopathological features and adverse outcomes, especially in breast cancer and prostate cancer [5]. However, whether obesity may contribute to aggressive histopathological features of thyroid cancer remains controversial.
Obesity is usually indicated by a higher body mass index (BMI). Tere is a signifcant association between increased BMI and increased incidence of PTC [6]. However, little is known about the association between BMI and clinicopathological features or outcomes of PTC. Paes et al. [7] showed that higher BMI was not associated with more aggressive tumor characteristics of PTC, such as the absence of lymph node metastasis (LNM). In contrast, Li et al. [8] demonstrated that higher BMI was associated with poor prognosis in PTC patients. Tis paradoxical result may be due to limitations of BMI, such as the inability to distinguish between fat mass and lean mass [9], and the failure to take into account the diferences in body composition related to age, gender, etc., [10]. As we all know, obesity can be quantifed not only by BMI, but also by body fat percentage (BFP) and body surface area (BSA), which can be obtained by simple measurement and calculation. According to the clinical defnition of obesity, its assessment should preferably be based on percent fat content. Hence, some investigators believe that BFP rather than BMI is a more efective indicator of actual obesity [11]. In addition, BSA is positively proportional to basal metabolism, refecting the ratio of body fat to nonfat components in individuals with the same BMI [12].
Te status of lymph nodes determines the prognosis of patients with PTC [13][14][15]. Hence, lymph nodes status can infuence clinical decisions. Te association between obesity and the status of lymph nodes in PTC has not been thoroughly investigated. Terefore, we assessed the relationship between obesity indicators (BMI, BFP, and BSA) and patterns of lymph node involvement in diferent regions through a retrospective analysis of large-scale data, focusing on the incidence of LNM, number of LNM, pattern of LNM, and lymph node ratio (LNR).

Study Design.
Our study was a retrospective cohort study of patients with PTC. Te Institutional Review Board of CHangzhou First People's Hospital approved this retrospective study. Te approval number of Institutional Review Board was (2022) Teach No. 024. Te need for informed consent was waived due to the retrospective nature of this study. We retrospectively reviewed the medical records of 1143 patients with pathologically proven PTC who underwent primary surgical treatment at our institution between March 2019 and October 2021. Patients were excluded for any of the following factors: nonPTCs or other subtypes than classic PTC, history of prior treatment for head and neck cancer, history of cervical radiation exposure in childhood, family history of thyroid cancer, history with other malignancy, incomplete clinical data, loss to follow-up, and patients who underwent noncurative surgery (residual tumor or lymph node detected within 6 months of initial surgery). A total of 1015 patients were eventually enrolled in this study. Te recruitment pathway for patients in this study was displayed in Figure 1.

Surgical Strategy.
Fine needle aspiration cytology (FNAC) was conducted to confrm the histopathologic diagnosis preoperatively for suspicious thyroid nodules. For cervical lymph nodes were considered suspicious by ultrasound, FNAC was used to confrm the diagnosis preoperatively. All patients with PTC routinely underwent prophylactic central neck dissection (CND) according to the Chinese guidelines for diagnosis and treatment of diferentiated thyroid carcinoma. Ipsilateral CND included the removal of prelaryngeal, pretracheal, and ipsilateral paratracheal lymph nodes, whereas bilateral CND included the removal of prelaryngeal, pretracheal, and bilateral paratracheal lymph nodes [16]. Prophylactic modifed radical neck dissection (MRND) was not recommended. Ipsilateral therapeutic MRND was performed for patients with LLNM confrmed by preoperative FNAC or intraoperative frozen pathological examination. MRND referred to the removal of the lateral lymph nodes, including level II to V, while preserving one or more nonlymphatic structures, such as the spinal accessory nerve, internal jugular vein, or sternocleidomastoid muscle. Unless indicated, level I dissection was not routinely performed.

Clinicopathological Characteristics.
Two or more experienced pathologists examined the surgical specimens microscopically. We considered a patient to have chronic lymphocytic thyroiditis (CLT) if any of the following conditions were met: (i) elevated levels of thyroid peroxidase antibodies, (ii) the discovery of difuse heterogeneity on ultrasonography [22]. Extrathyroidal extension (ETE) was defned as the primary tumor extending through the thyroid capsule to perithyroidal soft tissue such as perithyroidal fat, or involving strap muscles, or extending to surrounding structures such as larynx, trachea, esophagus, recurrent laryngeal nerve, subcutaneous soft tissue, skin, internal jugular vein, or carotid artery [23]. Tumor size was determined as the largest diameter of the primary tumor in multifocal tumors [24]. LNR was defned as metastatic lymph nodes divided by the number of dissected lymph nodes [25,26]. Pattern of LNM included skip metastasis, multiple lateral lymph node metastasis (MLLNM), and simultaneous metastasis in lateral compartment. Skip metastasis was defned as negative CLNM with positive LLNM [27]. MLLNM referred to the number of metastatic lymph nodes in the lateral compartment more than two [28]. In contrast, solitary lateral lymph node metastasis (SLLNM) was defned as patients with only one metastatic lymph nodes in the lateral compartment regardless of CLNM or not [28]. Simultaneous metastasis referred to more than 1-level metastasis in the lateral compartment [29].

Statistical
Analysis. Statistical analysis was performed using SPSS v 25.0 software (Chicago, IL, USA). Mean and standard deviation were used to represent continuous variables. Numbers and percentages were used to represent categorical variables. Continuous variables were examined by the Independent t-test or one-way analysis of variance (ANOVA). Categorical variables were analyzed by using Pearson's chi-square test or Fisher's exact test. Signifcant variables in the univariate analysis were included in the multivariate analysis by using the binary logistic regression test. Te characteristics of diferent regional lymph node involvement were regarded as dependent variables, and BMI, BFP, and BSA group were used as a covariate, respectively, and age (per 10 years), glucose, thyroid stimulating hormone (TSH), cholesterol, triglyceride, tumor size, multifocality, ETE, and CLT were used as adjustment variables.

Baseline Clinicopathological Characteristics of PTC
Patients. Our database included 1143 PTC patients, of which 1015 patients were included in the fnal analysis ( Figure 1). As shown in Table 1, the 1015 patients consisted of 743 women (73.2%) and 272 men (26.8%), with mean age of 43.4 ± 12.2 years. Te mean BMI was 24.03 ± 3.88. Overweight and obesity accounted for 30.5% and 5.7% of patients, respectively. Te mean BFP and BSA were 24.30 ± 10.49 and 1.70 ± 0.18, respectively. BFP and BSA were divided into male and female groups by gender.
Obesity accounted for 15.9% and 20.7% of the male patient group by BFP and BSA criteria, respectively. In the female patient group, obesity accounted for 25.8% in BFP and 15.5% in BSA, respectively.

Impact of BMI on Diferent Regional Lymph Node
Involvement. We frst analyzed the impact of BMI on the rate of LNM. As shown in Table 2, no statistical diferences were observed in rates of total LNM, CLNM, LLNM, and level II, III, and IV metastasis in normal, overweight, and obese patients. Te rate of level V metastasis in overweight patients with PTC was signifcantly higher than that in normal-weight patients (P � 0.013). However, binary logistic regression showed that overweight was not anindependent risk factor for level V metastasis (OR � 1.758, 95% CI: 0.957-3.229, P � 0.069). We then analyzed the efect of BMI on the number of LNM, LNR, and pattern of LNM, and found that BMI was not the risk factor for above variables. Table 3   . Univariate analysis showed that the rate of level IV metastasis in obese male patients was higher than that in normal male patients, but multivariate analysis showed that obesity was not an independent risk factor for level IV lymph node metastasis.

Impact of BFP on Diferent Regional Lymph Node Involvement in Men. As shown in
In obese male patients, the number of CLNM was higher than in normal-weight patients (3.1 vs. 1.9, P < 0.001). It was found that BFP in obese men was an independent risk factor for more than two CLNM (OR � 2.034, 95% CI: 1.183-3.497, P � 0.010).
Te central LNR of obese male patients was signifcantly higher than that of normal-weight patients (0.40 vs. 0.30, P � 0.025). It was found that BFP in obese men was an independent risk factor for a central LNR greater than 0.09 (OR � 2.442, 95% CI: 1.471-4.054, P � 0.001).
We fnally analyzed the pattern of LNM. MLLNM and simultaneous metastasis in lateral compartment presented the signifcant association with obesity (all P < 0.05). We further conducted a binary logistic regression and found that BFP in obese men remained the independent risk factor for MLLNM (OR � 2.564, 95% CI: 1.387-4.737, P � 0.003) and simultaneous metastasis in lateral compartment (OR � 2.096, 95% CI: 1.110-3.958, P � 0.023).

Impact of BFP on Diferent Regional Lymph Node Involvement in Women.
We then analyzed the impact of BFP on diferent regional lymph node involvement in female patients. As shown in Table 4, there were no statistically signifcant diferences in rates of LNM, number of LNM, LNR, and pattern of LNM between normal and obese female patients.
3.5. Impact of BSA on Diferent Regional Lymph Node Involvement in Men. According to gender, we divided BSAs into male and female groups. After analyzing the relationship between BSA and diferent regional lymph node involvement in male patients, we found that BSA in obese men was not a risk factor for rates of LNM, number of LNM, LNR, or pattern of LNM. Details are shown in Table 5.

Discussion
Body fat has been identifed as a compelling risk factor for several cancers, including PTC [1][2][3]. One of the hypotheses about the mechanism by which obesity may increase the risk of thyroid cancer is that it is mediated by increases in thyroid volume, thus an increase in the number of thyroid cells. Tis hypothesis is supported by the fact that thyroid volume increases with height, weight, BMI, and BSA [30]. Unlike previous studies that focused on the association between  International Journal of Endocrinology 5 obesity and PTC behavior [7,8], our study has unique factors. First, this is a large study of more than 1000 cases, using strict exclusion and inclusion criteria for accurate analysis. In addition to anthropometric parameters of BMI, we also assessed BFP and BSA. Due to sexual dimorphism in obesity, BFP and BSA were analyzed separately for men and women.
We frst analyzed the efect of BMI on cervical lymph nodes and found that BMI was not a risk factor for the rate of LNM, number of LNM, pattern of LNM, and LNR. Our results are consistent with previous studies that the increased risk associated with higher BMI did not vary with the presence or absence of adverse cancer histopathological  Te categorical variables were expressed as n (%). Te continuous variables were expressed as the mean (standard deviations). Multivariate analysis was expressed as adjusted odds ratio (95% CI). BSA, body surface area; LNM, lymph node metastasis; LLNM, lateral lymph node metastasis; CLNM, central lymph node metastasis; MLLNM, multiple lateral lymph node metastasis; LNR, lymph node ratio; and LC, lateral compartment. Age (per 10 years), glucose, thyroid stimulating hormone, cholesterol, triglyceride, tumor size, multifocality, extrathyroidal extension, and chronic lymphocytic thyroiditis as covariates to adjust OR value. 6 International Journal of Endocrinology features [31,32]. Notably, clinical BMI as the sole criterion for assessing obesity has certain limitations. First, BMI does not measure the body fat directly. Second, the composition of body adipose tissue changes with age. Tird, the relationship between BMI and body fat difers between men and women. Furthermore, studies even showed that patients assigned to be overweight according to BMI parameters had the same or better health outcomes than patients with normal BMI [33]. Tese limitations may be important reasons for the above results. BMI cannot explain the large diferences in body composition and fat content between individuals with similar body types, such as adipose tissue having a much lower metabolic rate than fat-free mass. Hence, some other indicators such as BFP and BSA have been proposed as supplements for obesity assessment [11,12]. As for BFP, it takes into account the efects of age and gender to estimate body fat content. BSA explains diferences between muscle and fat better than BMI, especially for individuals with the same BMI. To date, only few studies have examined the association of BSA and BFP with thyroid cancer [34][35][36][37], but no studies have investigated the efect of BFP and BSA on PTC behavior. We then analyzed the efect of BFP and BSA on lymph nodes status in both sexes. In men, high BFP was an independent predictor of LNM (including total LNM, CLNM, and total LLNM), MLLNM, and simultaneous metastasis in lateral compartment. In addition, male patients with high BFP had higher central LNR and higher number of CLNM. But in women, there was no signifcant association between lymph node status and BFP. For women, higher BSA was an independent predictor of LLNM and level IV metastasis. Moreover, female patients with high BSA had higher number of CLNM. However, no association was found between lymph node status and BFP in men. Tis study demonstrates gender dimorphism in the association between lymph node status and anthropometric parameters of PTC patients. Te reason for these diferences may be that women have more body fat than men, even with the same relative BMI [38]. Furthermore, unlike men who mainly store fat predominately in the abdomen, women tend to accumulate fat subcutaneously [39]. Terefore, we believe that BFP can be used as a supplementary indicator of obesity to assess the status of cervical lymph nodes in male PTC patients, while BSA can be used in female PTC patients.
According to the molecular mechanism of obesitytumor association, obesity can promote tumor invasion and metastasis through various obesity-related factors and metabolic pathways [40]. Certain adipokines, such as leptin and adiponectin, have also been implicated as mediators of the efects of obesity on the progression of thyroid cancer. For example, obesity leads to a decrease in adiponectin, which can increase the activity of tumor suppressors such as P53, inhibiting tumor growth and survival [41]. Fan and Li [42] found a strong correlation between leptin and the aggressiveness of thyroid cancer. Hormonal changes, such as TSH and insulin, are also associated with obesity [43]. Te growth and diferentiation of thyroid cells may be infuenced by TSH [44]. About 30% of the rest energy expenditure are regulated by thyroid hormones, which even generated the hypothesis that thyroid hormone substitution with TSHtitration into the lower reference levels may prevent body weight gain. Te growth and diferentiation of thyroid cells are also infuenced by TSH. Previous study found that BSA was independently associated to serum peak TSH concentrations [45,46]. Tis is the reason why BSA is proportional to basal metabolism with thyroid cancer. We believe that the detection of aggressive features of PTC, especially the status of cervical lymph nodes, is crucial in the management of patients with PTC from preoperative examination, surgical approach, to postoperative follow-up. Since the association between BMI and certain histopathological parameters is controversial in PTC patients, we applied other anthropometric factors, namely, BFP and BSA, to study obesity and lymph node status in PTC patients. Tis is the frst analysis demonstrating gender dimorphism in which histopathological parameters of lymph nodes are strongly associated with higher BFP and BSA. Tese fndings may aid clinical decision-making for regional LNM in PTC patients. For example, given the higher risk of LNM, male patients with high BFP or female patients with high BSA should undergo a detailed preoperative examination by an experienced sonographer. Furthermore, given that obese patients may have more positive lymph nodes and that obesity may increase the risk of inadequate lymph node dissection during surgery, surgeons should exercise more caution when performing CND for these patients. Te presence of high BFP and BSA should be reported in the specimen's referral as a reminder to pathologists to be more meticulous in the lymph nodes and tumor evaluation. In addition, increased vigilance for occult LLNM may be warranted in women with high BSA or men with high BFP if receiving CND only. Tese patients should be followed up more closely postoperatively. If suspicious lymph nodes are detected in the lateral compartment after surgery, FNAC should be actively conducted to confrm the histopathological diagnosis, and therapeutic MRND should be performed if necessary.
Tere were several potential limitations to our study. First, our study was a retrospective study, which is based on singlecenter data, and tended to have selection biases. Despite of strictly following the inclusion and exclusion criteria, we still could not rule out the possibility of residual confounding variables of measured or unmeasured factors, such as type of diet. Second, our study lacked long-termfollow-up data from patients. Hence, we could not analyze relationship between the risk of obesity and PTC recurrence. Tird, while BMI, BFP, and BSA were most commonly used as measures of body mass and fat, other measures such as neck circumference, skin-fold thickness, waist to and hip ratio circumference and regional fat content were not applicable in this study. Fourth, the relationship between weight change and PTC risk has not been uniformly concluded. It has been suggested that changes in obesity over time were unlikely to signifcantly afect the relationship between obesity and PTC risk [35], while others have shown that weight gain increased the risk of developing TC, and weight loss decreased the risk [47,48]. Because the measure of obesity was based on records at frst admission, we lacked a dynamic measure of changes in obesity. Terefore, the interpretation of our fndings is based on the assumption that anthropometric parameters maintain long-term balance and lack dynamic observation. Whether changes in body weight afect the aggressiveness of PTC remains to be further investigated. Finally, the patients enrolled in our study were all Chinese. We will conduct prospective multicenter institutional trials in subsequent studies to obtain more objective conclusions.
In summary, high BMI does not refect the aggressiveness of PTC in terms of lymph nodes. BFP and BSA can be used as supplementary indicators of obesity to assess the status of cervical lymph nodes in male and female PTC patients, respectively. Te possibility of the higher risk of LNM and higher number of CLNM in obese patients should be considered during the decision-making process.

Data Availability
Te data that support the fndings of this study are available on request from the corresponding author. Te data are not publicly available due to privacy or ethical restrictions.

Ethical Approval
Tis study has been approved by the Institutional Review Board of Changzhou First People's Hospital ethics committee, and has been performed according to the ethical standards laid down in the 1964 Declaration of Helsinki. Our study is a retrospective study, approved by the hospital ethics committee, and does not involve endangering the privacy and rights of patients. In addition, our research has not been carried out on human bodies or taken from human specimens. We retrospectively reviewed the medical records of patients at our institution. Te Institutional Review Board of Changzhou First People's Hospital approved this retrospective study. Te approval number of Institutional Review Board was (2022) Teach No. 024. Terefore, our study does not require trial registration.

Consent
Te need for informed consent was waived due to the retrospective nature of this study.