Validity of Preoperative CT Scan and Histopathological Staging in Resected Gastric Cancer
Authors
Keywords
Abstract
Background: Preoperative computed tomography (CT) is widely used for clinical staging in gastric cancer, yet the concordance between radiological and histopathological staging remains variable across populations. Accurate preoperative staging is critical for guiding treatment decisions, particularly in resource-constrained settings where multimodal imaging may be limited. Objective: To evaluate the validity of preoperative contrast-enhanced CT staging against histopathological staging in patients with resected gastric cancer. Methods: This prospective cohort study was conducted at the BSMMU, Dhaka, Bangladesh, from January to December 2021. Forty-three patients with histologically confirmed gastric cancer who underwent preoperative contrast-enhanced multidetector CT followed by curative and potentially curative gastrectomy were enrolled purposively. CT-based TNM staging was compared with postoperative histopathological staging as the reference standard. Data were analyzed using SPSS version 23.0, with agreement assessed using Cohen's kappa coefficient. Results: Preoperative CT revealed notable discrepancies in tumor and nodal classification compared with histopathology. For T staging, it overclassified the T2 (25.6% vs 18.6%) and T3 (41.9% vs 32.6%) categories while underclassifying the T4 (20.9% vs 34.9%). For nodal assessment, it overestimated the N1 (34.9% vs 27.9%) and N2 (32.6% vs 25.6%) but underestimated the N3 (11.6% vs 23.3%). Conclusion: Preoperative CT demonstrates moderate-to-fair agreement with histopathological staging in gastric cancer; however, significant misclassification persists, characterized by overstaging of intermediate T and N categories and underestimation of advanced T4 and N3 disease. Standardized high-resolution protocols and dedicated radiological review are recommended to improve staging reliability in routine clinical practice.
Journal of Paediatric Surgeons of Bangladesh (2026) Vol. 17 (2): 64–70
