Abstract
This study aimed to explore the risk factors for mediastinal lymph node metastases (MLNM) in patients with early-stage non-small-cell lung cancer (NSCLC) and to establish a predictive model. A retrospective analysis was conducted on the clinical data from NSCLC patients treated at the Second Affiliated Hospital of Guangzhou Medical University and the First Affiliated Dongguan Hospital of Guangdong Medical University between March 2021 and March 2023. Baseline clinical data, laboratory parameters, and pathological features were collected and analyzed. Univariate and multivariate logistic regression identified several independent risk factors for MLNM, including Cyfra21-1, D-dimer (D-D), tumor size, percentage of tumor solid, and lesion location. These risk factors were incorporated into a Nomogram model to visually assess the likelihood of MLNM. The model demonstrated excellent diagnostic accuracy with an area under the curve (AUC) of 0.904, a specificity of 73.85%, and a sensitivity of 93.68%. Cyfra21-1 and D-D were particularly significant predictors of MLNM. This Nomogram model provides an effective and practical tool for assessing MLNM risk in early-stage NSCLC, aiding clinical decision-making and optimizing treatment strategies.