PURPOSE: To evaluate the usefulness of CT for determining the causes of intestinal obstruction followingsurgery for abdominal cancer. MATERIALS AND METHODS: CT scans were performed in 54 patients with benign (n = 25)or malignant (n = 29) bowel obstruction after surgery for abdominal malignancies ; the causes of obstruction wereconfirmed pathologically (n = 34) or clinically (n = 20). Three radiologists interpreted the CT scans andevaluated their accuracy, sensitivity, and specificity. Through analysis of CT findings, malignant and benignobstruction was compared with regard to the presence of mass or lymphadenopathy, bowel change, mesenteric change,ascites, and other ancillary findings. RESULTS: In distinguishing malignant from benign obstruction, thediagnostic accuracy achieved by three radiologists was 67%, 74%, and 78%. When there was a mass at the obstructedor prior surgical site, lymphadenopathy, an abrupt transitional zone, or irregular wall thickening at anobstructed site, malignant obstruction was suspected (P < .05) ; in the absence of a mass, CT findings ofmesenteric vascular engorgement, extensive ascites, a smooth transition zone and normal or smooth wall thickeningat the obstructed site suggested benign obstruction (P < .05). The presence of omental infiltration, mesentericinfiltration and metastatic lesions at other sites did not always indicate malignant obstruction. In addition,bowel wall thckness, luminal diameter and the interval between surgery and bowel obstruction were notstatistically significant. CONCLUSION: CT is useful for determining the causes of obstruction following surgeryfor abdominal cancer, though particularly when definite peritoneal mass is not demonstrated this usefulness inlimited.