Feasibility of using automated insufflated carbon dioxide (CO2) for luminal distension in 3.0T MR colonography

Abstract

Purpose: Primary aim of our study was to prospectively evaluate the feasibility of automated carbon dioxide (CO(2)) delivery as luminal distending agent in 3.0T MR colonography.

Materials and methods: Rectally insufflated CO(2) was evaluated in four groups with different bowel preparation (A-D). Bowel preparation regimes were: gadolinium-based tagging (A), bowel purgation (B), barium-based tagging (C) and iodine-based tagging (D). Supine (3D)T1w-FFE and (2D)T2w-SSFSE series were acquired. Each colon was divided into six segments (cecum S1-rectum S6). Two observers independently assessed the presence of artefacts, diagnostic confidence and segmental colonic distension. Also characteristics of the residual stool (presence, composition and signal-intensity) were assessed per segment. Discomfort was assessed with questionnaires.

Results: Fourteen healthy subjects were included. Colonic distension by means of rectally insufflated CO(2) was not associated with susceptibility artefacts. Overall image quality was affected by the presence of bowel motion-related artefacts: none of the segments in 3DT1w-series and 10/84 (12%) colon segments in 2DT2w-series were rated artefact-free by both observers. Diagnostic confidence ratings were superior for the 2DT2w-SSFSE series. Overall bowel distension was rated adequate to optimal in 312/336 (93%) colon segments.

Conclusion: MR colonography at 3.0T using carbon dioxide (CO(2)) for colonic distension is technically feasible. The presence of intraluminal CO(2) did not result in susceptibility artefacts, although overall image quality was influenced by artefacts.