Findings of cecal volvulus at CT

To assess the diagnostic performance of computed tomographic (CT) and radiographic (as seen on CT topograms) signs of cecal volvulus.

In this institutional review board-approved, HIPAA-compliant retrospective study, the CT and CT topogram findings in 11 patients (one man, 10 women; age range, 26-100 years) with surgically confirmed cecal volvulus and 12 control patients were reviewed. The control subjects had suspicious radiographs, had undergone CT within 24 hours of radiography, and had received a clinical diagnosis other than cecal volvulus. Three radiologists independently evaluated the CT topograms for cecal distention, the coffee bean sign, cecal apex location, and distal colon decompression. CT images were analyzed for cecal distention, cecal apex location, distal colon decompression, and presence or absence of the whirl, ileocecal twist, transition point(s), the X-marks-the-spot, and the split wall. Sensitivity, specificity, and predictive values were computed. Baseline statistical values for the cecal volvulus and control groups were analyzed by using a two-tailed Z test to compare proportions with a threshold confidence interval of 95%. CT findings of bowel ischemia (free air or fluid, pneumatosis intestinalis, portal venous gas, mesenteric stranding) were correlated with pathology report findings.

On CT topograms, greater than 10-cm cecal distention, coffee bean sign, and left upper quadrant cecal apex had sensitivities of 45% (five of 11 patients), 27% (three of 11 patients), and 45% (five of 11 patients), respectively, and specificities of 100% (12 of 12 control subjects), 92% (11 of 12 control subjects), and 100% (12 of 12 control subjects), respectively. Distal colon decompression had sensitivities and specificities of 91% (10 of 11 patients) and 83% (10 of 12 control subject), respectively, on topograms and of 91% (10 of 11 patients) and 92% (11 of 12 patients), respectively, on CT images. On cross-sectional CT images, greater than 10-cm cecal distention, left upper quadrant cecal apex, whirl, ileocecal twist, transition point(s), X-marks-the-spot, and split wall had sensitivities of 45% (five of 11 patients), 36% (four of 11 patients), 73% (eight of 11 patients), 54% (six of 11 patients), 82% (nine of 11 patients), 27% (three of 11 patients), and 54% (six of 11 patients), respectively; each had 100% specificity. Pneumatosis intestinalis and free air had 100% (four of four control subjects) specificity. Overall, CT signs of bowel ischemia correlated poorly with pathology report findings.

When cecal volvulus is suspected, the absence of distal colonic decompression on CT topograms makes the diagnosis very unlikely. Whirl, ileocecal twist, transition points, X-marks-the-spot, and split wall have high specificity for cecal volvulus.