Determination of bile acids from human gallbladder by 1 H-MRS-Protocol optimization and estimation of reproducibility.

Bile exerts multiple functions in the liver and gut and is involved in multiple disease processes. It is secreted continuously from the liver and stored in the gallbladder until needed, and closely reflects the available bile acid pool. The study objective was therefore to develop a reliable MRS protocol and to assess variability of bile acid determination in human gallbladder. MRS measurements were performed on a 3 T MR scanner with 20 subjects to optimize protocols (26 measurements) and conduct a prospective reproducibility study (18 measurements). Measurements were carried out with subjects lying in either supine (23 scans) or prone positions (21 scans) to compare results from the two positions. For reproducibility determination, six of the 20 volunteers (three males, three females, age = 34.9 ± 10.9 years, BMI = 23.4 ± 2.1 kg/m2 ) were measured three times: back to back to assess technical variability and once again after three weeks to assess total variability, including additional physiological variability. A single voxel was measured in the gallbladder with respiratory triggering. For quantification, apparent T2 times were determined and a non-water-suppressed spectrum was acquired. Total bile acids, glycine and taurine conjugated bile acids, and lipids including choline-containing phospholipids were determined. Higher quality and reliability of gallbladder spectra were obtained with subjects measured in prone compared with supine position. All measurements of the reproducibility sub-study were of sufficient quality to be included in the analysis. Average coefficients of variation within subjects for the main compounds were 37% for total variation (including physiological and technical variation) and 24% for technical variation alone. These values were much smaller than those between subjects, which were >54% for both back-to-back and three weeks separated measurements. These results suggest diagnostic applicability of the method, especially for longitudinal studies aiming at non-invasive characterization of bile composition in humans with various diseases and/or interventional maneuvers.