Abstract
Introduction Magnetic resonance imaging (MRI) serves as a pivotal tool in the preoperative assessment of rectal cancer. This study aims to evaluate the accuracy of preoperative MRI pelvis in rectal cancer for locoregional staging, circumferential margin (CRM+), and vascular invasion (V1) with postoperative histopathological findings. Methods All patients who underwent preoperative staging MRI pelvis scanning for histology-proven primary rectal adenocarcinoma between January 2020 and July 2022 were included in this study. Preoperative MRI assessment data, including tumor (T), nodal (N) staging, and other prognostic factors, including circumferential margin and vascular invasion, were compared with histopathological reports. Sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and accuracy were calculated. Agreement between MRI and pathology reports was evaluated using the weighted kappa statistic. Results The study included 143 patients. The accuracy of MRI is at its peak for T0 and T4 tumors (T0-97.9%, T4-93.37). The weighted kappa statistic for T staging is 0.401, and for N staging, it is 0.286. The standard kappa statistic for extramural vascular invasion (V) and circumferential resection margin (CRM) is 0.269 and 0.225, respectively. Conclusion MRI demonstrates fair agreement with pathology regarding T and N staging and exhibits high specificity for CRM and V1.