Preoperative Prediction of Lymph Node Metastasis in Esophageal Squamous Cell Carcinoma using Multiple Quantitative Parameters from Dual-Source Energy Spectrum CT.
Abstract
Objectives
This study evaluated the diagnostic performance of quantitative parameters derived from dual-source spectral CT (DECT) for preoperative prediction of lymph node metastasis (LNM) in esophageal squamous cell carcinoma (ESCC).
Methods
From January 2024 to August 2025, a total of 117 ESCC patients meeting the inclusion and exclusion criteria were recruited. Based on postoperative pathology, patients were categorized into two groups: LNM-positive (n=55) and LNM-negative (n=62). Preoperative DECT was performed on all patients. Measurable parameters of the original tumor, including iodine concentration (IC), normalized iodine concentration (NIC), effective atomic number (Zeff), electron density (Rho), spectral curve slope (λHU, 40-70 keV), dual-energy index (DEI), and virtual noncontrast (VNC), were acquired during the arterial phase (AP) and the venous phase (VP). Clinicopathological data were collected. T-tests and chi-square tests were used for between-group comparisons; logistic regression analysis was employed to identify independent risk factors for lymph node metastasis; and receiver operating characteristic (ROC) curves were used to evaluate their predictive performance.
Results
Primary tumor length (L) differed significantly between groups (P<0.05; AUC=0.68). Statistical analysis revealed that within both AP and VP, the LNM-negative group had significantly elevated IC, NIC, λHU, Zeff, and DEI relative to the LNM-positive group (all P<0.05). Multivariate analysis showed that VP-NIC is an independent predictor of lymph node metastasis, with an AUC value of 0.86. The combination of VP-NIC and tumor length (L) achieved the highest predictive efficacy (AUC=0.88; sensitivity=95.2%; specificity=69.1%).
Discussion
This is the first study to demonstrate that primary tumor-derived DECT parameters can predict ESCC LNM preoperatively. VP-NIC independently predicted nodal status; its combination with tumor length enhanced predictive performance.
Conclusion
DECT-derived multiparametric analysis, especially the combined model, may serve as a non-invasive supplementary method for preoperative LNM assessment in ESCC, though further validation is warranted.