Predictive risk-scoring model for lateral lymph node metastasis in papillary thyroid carcinoma

This study aims to evaluate candidate risk factors for lateral lymph node metastasis (LLNM) and develop a predictive model to identify high-risk groups among patients with papillary thyroid carcinoma (PTC). Additionally, we identified risk factors for recurrence to inform postoperative therapeutic decisions and follow-up for physicians and patients. A total of 4107 patients (4884 lesions) who underwent lymph node dissection at our hospital from 2005 to 2014 were evaluated. LLNM risk was stratified, and a risk-scoring model was developed based on identified independent risk factors for LLNM. Cox's proportional hazards regression model was used to investigate the risk factors for recurrence. Lateral Lymph Node (LLN) metastasis was observed in 10.49% (431/4107) of patients. Multivariate analysis identified the following independent risk predictors for LLN metastasis: Age ≤ 35 years (P = 0.002), tumor size > 1.0 cm (P = 0.000), lobe dissemination (+) (P = 0.000), and CLNM (+) (P = 0.000). A 12-point risk-scoring model was constructed to predict stratified LLNM in PTC patients, with an area under the receiver operating characteristic curve (AUROC) of 0.794 (95% CI: 0.774-0.814) (P < 0.01). The Cox regression model indicated that tumor size > 1.0 cm, lobe dissemination (+), multifocality, Central Lymph Node Metastasis (CLNM), and LLNM were significant risk factors associated with poor outcomes. Based on the risk scoring model, additional investigations and comprehensive considerations are recommended for patients with a total score greater than 5, and prophylactic cervical lymph node dissection is performed if necessary. Additionally, more aggressive treatment and more frequent follow-ups should be considered for patients with tumor size > 1.0 cm, lobe dissemination (+), multifocality, CLNM, and LLNM.