Objective: The purpose of this study was to evaluate the usefulness of CT in determining the causes of intestinal obstruction in patients who have undergone abdominal surgery for malignancy.
Materials and methods: We analyzed the CT scans of 55 patients with benign (n = 26) or malignant (n = 29) intestinal obstruction that developed after abdominal surgery for malignancy. After calculating the diagnostic accuracy of interpretations by three radiologists, we compared CT findings for benign and malignant intestinal obstructions with respect to peritoneal involvement patterns and other ancillary findings. Multivariate logistic regression analysis was used to determine the diagnostic performance of CT in revealing causes of obstruction.
Results: Diagnostic accuracies of the three radiologists were 67%, 75%, and 78%. CT findings indicating malignant obstruction were a mass at the site of obstruction or prior surgery, lymphadenopathy, or an abrupt transition zone and irregular bowel wall thickening at obstructed sites (p < .05). Conversely, the chance for benign obstruction increased when CT revealed mesenteric vascular changes, a large amount of ascites, or a smooth transition zone and smooth bowel wall thickening at the obstructed site (p < .05). With multivariate logistic regression analysis using two variables (a mass at the site of obstruction or prior surgery and lymphadenopathy), we calculated the overall accuracy of CT as 84% (46/55 patients).
Conclusion: CT is useful in differentiating benign from malignant intestinal obstructions in patients who have undergone abdominal surgery for malignancy. However, CT has limitations in patients not having a demonstrable peritoneal mass.