Bowel preparation for capsule endoscopy (CE) has not been standardized. This study aimed to compare CE images between patients prepared by simethicone and those prepared by magnesium citrate. Retrospective analysis of case series of our hospital from 2004 to 2007. Single center. CE images of 75 patients receiving bowel preparation either by 200 mg of simethicone (n=39) or by 34 g of magnesium citrate (n=36) were retrospectively investigated. Grades of fluid transparency and mucosal invisibility by air bubbles and food residue were compared between the 2 preparations. Capsule transit time, frequency of positive findings, and interobserver variations between 2 observers were also investigated. Image quality and diagnostic yield of CE. Fluid transparency in the first and the third time segments of the small intestine was better in patients prepared by magnesium citrate than in those prepared by simethicone (P= .001 and P= .03, respectively). On the other hand, mucosal invisibility was not different in any part of the small intestine between the 2 groups. Neither gastric transit time nor small-bowel transit time was different between the 2 groups. The diagnostic yield of CE correlated significantly with fluid transparency (P= .04), but it did not correlate with mucosal invisibility. Single-center retrospective study. Magnesium citrate seems to be a recommended preparation for CE compared with simethicone. The fluid transparency, rather than the mucosal invisibility, may be a factor associated with the diagnostic yield of CE.