Impact of diet on stool signal in dark lumen magnetic resonance colonography

Purpose: To examine the magnetic resonance (MR) properties of different foods and their effect on the colonic stool signal to potentially support fecal tagging strategies for dark lumen MR colonography (MRC). Materials and Methods: T1 relaxation times of 120 different foods (partially diluted with sufficient water) were determined by use of a multi-flip-angle two-dimensional gradient echo (GRE) sequence and correlated to the foods' signal in a three-dimensional GRE volumetric interpolated breath-hold examination (VIBE) sequence. Different dilutions of six foods were examined. VIBE stool signal was determined in six volunteers under two different conditions: after a three-day diet of short T1 food and of long T1 food, respectively. Results: Most foods exhibit short to very short T1 relaxation times. T1 correlates well with the fat-saturated VIBE signal except for fatty products. Diluted food exhibits T1 times similar to water; concentrated food strongly varies according to their T1 values. No significant difference in stool signal could be found in the in vivo examination comparing the two diets. Conclusion: According to our results, a restricted diet strategy to reduce fecal signal for dark lumen MRC is unlikely to be successful. Moreover, the stool signal reduction found in the other fecal tagging studies can be explained at least to a great extent by the relative content of other material with long T1 relaxation times, such as water or oral barium.