Magnetic resonance colonography with limited bowel preparation: a comparison of three strategies.

Purpose: To prospectively compare three strategies of magnetic resonance colonography (MRC) with fecal tagging. Materials and Methods: Three strategies were compared: (S1) gadolinium as oral tagging agent and a gadolinium-water mixture for rectal filling (bright lumen), (S2) oral barium and water rectally, and (S3) oral barium and air rectally. In S2 and S3 (both dark lumen) gadolinium was injected intravenously. Three-dimensional (3D) T1-weighted and two-dimensional (2D) T2-weighted sequences were used. Two observers scored diagnostic confidence and image quality (contrast, homogeneity, artifacts), analyzed by chi-squared and Fisher's exact test. Patient experience and preference were determined by questionnaire (Mann-Whitney test). Results: A total of 45 patients were included, 15 were randomly assigned per strategy. Diagnostic confidence of S1 and S3 is significantly better than for S2. S1 has the additional advantage of showing significantly better contrast between bowel wall and lumen, and showing significantly better homogeneity on both T1- and T2-weighted sequences, but with significantly more artifacts on the T1 weighted sequences. S3 showed significantly better contrast and homogeneity than S2 on the T2-weighted sequences. Bowel preparation of S1 was rated significantly better. Patient preference was comparable. Conclusion: Image quality was best using the bright lumen strategy or the dark lumen strategy using air for rectal filling. Although bowel preparation was rated best using the bright lumen strategy, patient preference was comparable.