MR colonography in patients with incomplete conventional colonoscopy.

PURPOSE: To assess dark-lumen magnetic resonance (MR) colonography for the evaluation of colonic segments in patients in whom conventional colonoscopy could not be completed. MATERIALS AND METHODS: Institutional review board approval and written informed consent were obtained. Within 24 hours of incomplete conventional colonoscopy, 37 patients (22 women, 15 men; age range, 25-63 years) underwent MR colonography. Contrast material-enhanced T1-weighted three-dimensional images were collected after rectal administration of water for colonic distention. Data from MR colonography were evaluated by two radiologists. With a three-point scale, image quality was characterized in terms of colonic distention (1 = good; 2 = moderate, diagnostic; and 3 = poor, nondiagnostic) and presence of artifacts (1 = none; 2 = moderate, diagnostic; and 3 = extensive, nondiagnostic). Depiction of colorectal disease was assessed according to the following colonic segments: cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum. RESULTS: Four patients had history of colorectal cancer, and each had undergone partial colonic resection of two segments. Hence, 214 segments were evaluated. Conventional colonoscopy failed in assessment of 127 of 214 potentially visible colonic segments in the 37 patients. MR image quality was rated diagnostic in 35 patients and permitted assessment of 206 of 214 potentially visible segments. Nondiagnostic image quality in two patients was attributed to inadequate distention of prestenotic colonic segments owing to high-grade tumor stenosis. All inflammation- and tumor-induced stenoses and all five polyps identified at conventional colonoscopy in poststenotic segments were correctly detected at MR colonography. However, MR-based assessment of prestenotic segments revealed two lesions suspected of being carcinoma, five polyps, and four segments affected by colitis. CONCLUSION: MR colonography proved reliable in evaluating the majority of colonic segments inaccessible with conventional colonoscopy. The identification of additional disease at MR colonography underscores the need for a second diagnostic step in the setting of incomplete conventional colonoscopy.

MR colonography: how does air compare to water for colonic distention?

PURPOSE: To prove the feasibility of air-distended magnetic resonance colonography (MRC) and compare it with water-based distention. MATERIALS AND METHODS: In five volunteers, the colon was imaged twice: once after distending the colon with air and a second time after distending the colon with water. A total of 50 patients, who had been referred to colonoscopy for a suspected colorectal pathology were randomized into water-distention (N = 25) and air-distention (N = 25) groups. A contrast-enhanced T1-weighted three-dimensional volume interpolated breath-hold (VIBE) sequence was collected. Comparative analysis was based on qualitative ratings of image quality and bowel distention, as well as contrast-to-noise ratio (CNR) measurements for the colonic wall with respect to the colonic lumen. In addition, patient acceptance was evaluated. RESULTS: Inflammatory changes and colorectal masses were correctly identified on MRC in eight patients each. One 4-mm polyp identified at colonoscopy was missed on water-distended MRC. There were no false positive findings. No significant differences were found between air- and water-distention regarding discomfort levels and image quality. The presence of air in the colonic lumen was not associated with susceptibility artifacts. CNR of the contrast-enhanced colonic wall, as well as bowel distention, were superior on air-distended three-dimensional data sets. CONCLUSION: MRC can be performed using either water or air for colonic distention. Both techniques permit assessment of the colonic wall and identification of colorectal masses. While discomfort levels are similar for both agents, MRC with air provides higher CNR and better colonic distention.