Lymph Node Staging by CT Texture Analysis in Patients with Sigmoid Cancer

Abstract

Objective: Accurate radiological staging of colon cancer is essential for appropriately selecting patients who might benefit from neoadjuvant chemotherapy. Lymph node staging and the detection of metastatic lymph nodes form an integral part of the local staging. This study aimed to determine the efficacy of computed tomography texture analysis (CTTA) in characterizing lymph nodes in patients with sigmoid cancers. Materials andMethods: Forty-five patients diagnosed with sigmoid adenocarcinoma, who underwent computed tomography (CT) scans, were included in this retrospective study. Based on post-surgery histopathological results, 25 patients were classified as stage N1-2, and 20 patients were classified as stage N0. CTTA was conducted on 51 metastatic lymph nodes from the 25 N1-2 patients and 30 benign lymph nodes from the 20 N0 patients. Histogram analysis was employed to calculate texture features, and the texture features of both groups were statistically compared. Receiver operating characteristic (ROC) analysis was used to evaluate the predictive performance of the parameters.Results: The maximum of the histogram, 99th percentile of the histogram, and entropy values were significantly higher in the metastatic lymph nodes. Conversely, skewness, uniformity, and kurtosis values were significantly higher in benign lymph nodes (p<0.05). ROC analysis for uniformity and skewness revealed area under the curve (AUC) values of 0.904 and 0.909, respectively.Conclusion: In patients with sigmoid cancer, CTTA can serve as a useful tool for lymph node staging.