Plain abdominal radiographic features are not reliable markers of disease extent in active ulcerative colitis

The aims of this study were to assess the reliability of the distribution of fecal residue, mucosal irregularity, and colonic wall thickening on plain abdominal x-ray as markers of disease extent assessed by technetium-99 m hexamethylpropylene amine oxine-labeled leukocyte scans in active ulcerative colitis (UC).

Plain abdominal radiographs were blindly assessed for the distribution of fecal residue, mucosal irregularity, and colonic wall thickening (>3 mm) in 30 patients with active UC. Most patients were too sick to safely allow total colonoscopy. In 11 patients for whom details of disease extent were available from total colonoscopy or surgery undertaken at the time, there was a close correlation with disease extent defined by technetium-99 m hexamethylpropylene amine oxine-labeled leukocyte scans. Contemporaneous radiolabeled leukocyte scans were therefore used to assess disease extent in comparison with plain abdominal radiographs.

Of 30 patients, 15 had pancolitis and 15 had subtotal or distal disease as indicated by radiolabeled leukocyte scans. The distribution of fecal residue on plain abdominal radiographs correctly identified disease extent defined on radiolabeled leukocyte scans in 40% of patients, overestimating it in 13% and underestimating it in 47%. There was no significant correlation between distribution of fecal residue on plain abdominal radiographs and disease extent on radiolabeled leukocyte scans or colonoscopy or surgery. Of patients with pancolitis, 60% had fecal residue present on plain abdominal radiograph with 40% showing stool distal to the hepatic flexure as well as in the right colon. For total UC on radiolabeled leukocyte scanning, the sensitivity and specificity of absence of fecal residue on plain abdominal radiographs were 40% and 80% respectively. Irregularity of mucosal edge and colonic wall thickening were even less accurate than fecal residue in defining disease extent.

The distribution of fecal residue, irregularity of mucosal edge, and colonic wall thickening on plain abdominal radiography do not provide a reliable guide to disease extent in active UC.