Diagnostic accuracy of MR colonography with fecal tagging Michael Patrick Achiam, 1,2 Vibeke Berg Løgager, 1 Elizaveta Chabanova, 1 Bodil Eegholm, 1 Henrik Segelcke Thomsen, 1 Jacob Rosenberg 2 1 Department of Diagnostic Radiology, Copenhagen University Hospital Herlev, Herlev Ringvej, 2730 Herlev, Denmark 2 Department of Surgical Gastroenterology D, Copenhagen University Hospital Herlev, Herlev Ringvej, 2730 Herlev, Denmark Abstract Background: MR colonography (MRC) is a non-invasive method of examining the colon, but it is still only used in a few specialized centers on a daily basis. The purpose of this study was to evaluate the performance of MRC with improved fecal tagging vs. conventional colonoscopy (CC). Methods: Between March 2006 and February 2007, consecutive patients who received first-time referrals to CC were asked to participate in the study. Two days prior to MRC, a new contrast mixture (barium/ferumoxsil) was ingested together with four meals each day. Standard bowel cleansing was performed before CC. MRC was evaluated by two blinded observers. MRC results were compared with CC. Results: A total of 56 patients were included. The per- polyp sensitivity was 85.7% for polyps 6–10 mm and 81.3% for polyps >10 mm. The per-patient sensitivity/ specificity was 100%/80% for polyps >6 mm and 100%/ 91.4% for polyps >10 mm. Conclusion: MRC showed acceptable per-patient sensitiv- ities, but the per-polyp sensitivities are still compromised by the fecal tagging. MRC may be implemented, but at the present time MRC with fecal tagging is not ready for widespread use, although it remains a very promising diagnostic tool. Key words: MR colonography—Fecal tagging—Conventional colonoscopy—Polyps—Colorectal cancer In the last 10 years, MR- and CT-colonography have emerged as a new non-invasive diagnostic tool for examining and screening the colon for polyps and cancer. Due to the natural history of polyps and the adenoma/ carcinoma sequence theory [1, 2], the great potential of colorectal screening has led many to turn their attention to colonography as a future diagnostic screening tool. At the present time conventional colonoscopy (CC) is the gold standard for colorectal screening, but the invasive nature of the procedure with potentially severe compli- cations and mortality combined with the need for seda- tion and low acceptance have given room for speculation as to whether colonography would be a better tool for screening in the future. In order to further improve patient acceptance, fecal tagging, a method of avoiding colonic cleansing prior to examination which is rated as one of the most uncomfortable parts of a colonic examination [3, 4], has been developed [5]. Another method to improve patient acceptance and reduce pro- cedural difficulties that have been described is the use of air for colonic distension instead of water [6–10]. Although earlier studies have reported high sensitivity using fecal tagging [11, 12], recent studies have reported low sensitivity and low patient acceptance and most studies come to the conclusion that the quality of the fecal tagging agent needs to be optimized [13–16]. In a recent study, our group described a new contrast mix- ture, which seemed advantageous for a small number of patients [17]. The purpose of this study was to evaluate the performance of MRC with the improved fecal tagging vs. conventional colonoscopy (CC). Methods Study group Between March 2006 and February 2007, consecutive patients with first-time referrals to CC at our surgical department were asked to participate. Patients younger than 18 years, with colostomy, or contraindications to MRI (pregnancy, claustrophobia etc.) were excluded Correspondence to: Michael Patrick Achiam; email: achiam1@dadl- net.dk ª Springer Science+Business Media, LLC 2008 Published online: 2 May 2008 Abdominal Imaging Abdom Imaging (2009) 34:483–490 DOI: 10.1007/s00261-008-9402-x