PTC demonstrates a propensity for early lymph node metastasis, with the central lymph node typically being the first area affected. Metastatic rates in this region have been reported to reach as high as 90% [9]. In patients diagnosed with UPTC of clinical N0 status, BCLNM is documented in up to 50% of patients. Moreover, an incidence rate of 6.7% has been recorded for ‘skip metastasis’ to the contralateral central lymph nodes [10]. In the present study, we demonstrated a CCLNM rate of 35.3% (78/221). Hence, for patients with UPTC, surgery confined to ipsilateral central lymph node dissection (ICLND) may not provide adequate coverage. Undetected positive lymph nodes during initial surgery could increase the tumor recurrence rate and complicate any subsequent surgeries. Although preventive BCLND can help us to accurately evaluate pathological staging, lymph nodes guide subsequent treatment but also increase the risk of postoperative parathyroid function disorder [11, 12]. Therefore, when delineating the suitable extent of lymph node dissection, surgeons must carefully consider strategies to minimize postoperative complications and tumor recurrence. This underscores the importance of rigorous preoperative and intraoperative evaluations to identify risk factors related to CCLNM in patients with UPTC, thereby defining a precise surgical roadmap.

In this study, we retrospectively examined the clinicopathological data of 221 patients and reported a CCLNM rate of 35.3% (78/221). Among these patients, 33.0% (73/221) had BCLNM, while 2.26% (5/221) had metastasis absent in the ipsilateral central lymph node but present in the contralateral central lymph node (skip metastasis). These findings are in line with the majority of the academic literature [6, 13]. Nonetheless, studies led by Roh and Sun et al. reported a lower CCLNM rate ranging between 3.88% and 30.63% [9, 14], which is in contrast with our recorded rate of 35.3%. This discrepancy could be attributed to two main factors: (1) our study did not include clinical lymph node (cN) data according to our enrollment criteria, and (2) our study had a relatively small sample size. With a high incidence of HT (26.9%) [15], cN assessment could be affected, leading to bias in our enrolled patients. Consequently, we did not confine our UPTC patient inclusion criteria to only cN0. Furthermore, by diligently excluding patients with isthmic thyroid tumors, we were able to enhance the reliability of our conclusions.

The prior literature has implicated several factors as independent risk factors for CCLNM in UPTC patients. These included male sex, age under 45 years, BMI exceeding 25, external thyroid invasion, tumor diameter greater than 1 cm, presence of vascular cancer thrombus, multifocal tumors, BRAF V-600E gene mutation, and ICLNM [9, 16,17,18,19,20,21]. Our multivariate analysis corroborated these findings, revealing male sex (P = 0.010, OR: 3.790, 95% CI: 1.373–10.465), external thyroid invasion (P = 0.040, OR: 6.364, 95% CI: 1.083–37.381), tumor diameter (P = 0.01, OR: 3.674, 95% CI: 1.372–9.839), and ICLNM (P < 0.001, OR: 38.552, 95% CI: 2.675–27.342) as independent predictors of CCLNM. In line with our findings, Kang SK’s study also integrated the number of ICLNMs as a variable and utilized a cutoff score of 5, demonstrating that while ICLNM served as an independent risk factor for CCLNM, there was no correlation with the number of ICLNMs [18]. Consequently, it may be beneficial to consider BCLND for UPTC patients exhibiting confirmed ICLNM. The focus should prioritize ICLNM presence over attempting to determine surgical scope by the quantity of ICLNM.

While PTC is more prevalent in the female population [1], it is markedly more aggressive in male patients, often manifesting as cervical lymph node metastasis [9, 22, 23]. Our results align with this observation, indicating that male patients diagnosed with UPTC tend to be at a greater risk of CCLNM.

Sun W et al., in their meta-analysis, suggested that a tumor diameter exceeding 2 cm serves as an independent risk factor for CCLNM in cN0 UPTC patients [9]. Consistent with most previous studies, our research also indicated that a tumor diameter greater than 1 cm is an independent predictor of CCLNM. However, the correlation between tumor diameter and CCLNM warrants further exploration. An investigation by Yan S suggested that BCLND may be an appropriate surgical approach for UPTC patients with a tumor diameter greater than 1 cm. On the other hand, it also increases the risk of short-term postoperative complications such as vocal cord paralysis and decreased parathyroid hormone (PTH) levels [6]. Hence, accurate preoperative evaluation of the surgical scope is indispensable for minimizing postoperative complications. We do not advocate prophylactic contralateral central lymph node dissection for UPTC patients with tumors less than 1 cm in diameter.

In a meta-analysis by Kim DH et al., it was found that the presence of ICLNM (odds ratio = 13.9118; 95% CI: 8.7096 ~ 22.2213) showed the strongest correlation with CCLNM in UPTC patients [22], a finding that aligns with the conclusions drawn in this paper. Consequently, it is recommended that UPTC patients with confirmed ICLNM through preoperative US-FNAB or intraoperative frozen pathology undergo BCLND.

Moreover, the study revealed a heightened risk of cervical lymph node metastasis in patients where the tumor infiltrates the surrounding thyroid tissue, with ETE emerging as an independent risk factor for CCLNM in UPTC patients. These results are in line with the findings of Feng JW and Kim DH et al. [21][23]. Currently, preoperative ultrasonography has demonstrated significant predictive value in assessing the extrathyroidal invasion of thyroid tumors, and the widespread use of intraoperative frozen pathological examination can serve as a valuable tool for determining the appropriate surgical scope. Furthermore, ETE is indicative of the aggressive nature of PTC and plays a pivotal role in predicting patient prognosis. Upon the occurrence of ETE, the risk of tumor recurrence and mortality escalates [24].

Due to the inclusion of a multicenter sample in this study, uniform testing for BRAF V-600E gene mutations could not be conducted and hence was not included in the analysis. Furthermore, discrepancies with previous studies in factors such as age, multifocality, HT, vascular cancer thrombus, and BMI are likely attributable to the limitations posed by the small sample size.

In essence, this retrospective study comprehensively analyzed clinicopathological data from 221 samples collected across multiple centers to develop a predictive model for determining CCLNM in UPTC patients based on four distinct risk factors: sex, tumor diameter, external thyroid invasion, and ICLNM. The resulting model demonstrated robust differentiation and calibration, yielding tangible clinical benefits for decision-making processes. By integrating the nomogram with preoperative and intraoperative assessment parameters such as male sex, tumor diameter > 1 cm, external thyroid invasion, and ICLNM in UPTC patients, personalized and precise surgical delineation can be achieved, optimizing treatment strategies. However, it is important to note the limitations of this study, which include its relatively small sample size and the absence of external validation. Therefore, there is a clear imperative to bolster data supplementation to enhance the generalizability and applicability of the prediction model.