PURPOSE: To evaluate the accuracy of spiral hydro-CT in the diagnosis and preoperative staging of colo-sigmoideal neoplasms and the impact of this technique on therapeutic decisions. MATERIALS AND METHODS: Sixty patients with colo-sigmoideal neoplasms, 16 in caecum-ascending colon, 2 in transverse colon, 14 in descending colon, and 28 in sigmoid colon, diagnosed with fiberoptic colonoscopy and histopathological specimen, under-went spiral hydro-CT with IV contrast medium administration. Local invasion (T factor) and nodal involvement (N factor) were evaluated. In all cases, the CT reports were compared with the result of the histopathological examination of the surgical specimen. TNM and Dukes staging classifications were applied. RESULTS: Local invasion was correctly identified in 80% of cases. There was overstaging in 18.3% and understaging in 1.7% of the patients. Sensitivity, specificity and accuracy for nodal involvement were 57.9%, 67.5% and 63.3%, respectively. There was overstaging in 23.3% and understaging in 15% of patients. With Dukes classification, CT correctly staged 38.8% of patients. Significant differences were detected in different stages. The stage with the highest concordance was A/B1 with 58.8% (10/17), whereas the one with the lowest concordance was B2 with 9.1% (2/22). All of the three patients in stage D were correctly staged and none of the patients with stage C was overstaged. CONCLUSIONS: Spiral hydro-CT with intravenous contrast medium administration, despite the limits in the correct evaluation of local depth invasion and nodal involvement, proved useful for selecting patients likely to benefit from surgery.
PURPOSE: To evaluate the accuracy of spiral hydro-CT in the diagnosis and preoperative staging of colo-sigmoideal neoplasms and the impact of this technique on therapeutic decisions. MATERIALS AND METHODS: Sixty patients with colo-sigmoideal neoplasms, 16 in caecum-ascending colon, 2 in transverse colon, 14 in descending colon, and 28 in sigmoid colon, diagnosed with fiberoptic colonoscopy and histopathological specimen, under-went spiral hydro-CT with IV contrast medium administration. Local invasion (T factor) and nodal involvement (N factor) were evaluated. In all cases, the CT reports were compared with the result of the histopathological examination of the surgical specimen. TNM and Dukes staging classifications were applied. RESULTS: Local invasion was correctly identified in 80% of cases. There was overstaging in 18.3% and understaging in 1.7% of the patients. Sensitivity, specificity and accuracy for nodal involvement were 57.9%, 67.5% and 63.3%, respectively. There was overstaging in 23.3% and understaging in 15% of patients. With Dukes classification, CT correctly staged 38.8% of patients. Significant differences were detected in different stages. The stage with the highest concordance was A/B1 with 58.8% (10/17), whereas the one with the lowest concordance was B2 with 9.1% (2/22). All of the three patients in stage D were correctly staged and none of the patients with stage C was overstaged. CONCLUSIONS: Spiral hydro-CT with intravenous contrast medium administration, despite the limits in the correct evaluation of local depth invasion and nodal involvement, proved useful for selecting patients likely to benefit from surgery.