Page 31 - Human Bile Acid Metabolism: a Postprandial Perspective

ursodeoxycholic acid (UDCA), as well as their corresponding glycine and taurine
amidates (conjugates). Total bile acid (TBA) concentrations were determined by
calculating the molar sum of all 12 bile acids. Bile acids were measured using an ultra-performanceliquidchromatography ionization tandem mass spectrometry 2 method (12) (Supplemental Data). A sandwich ELISA kit was used for colorimetric determination of FGF19 in plasma (FGF19 Quantikine ELISA kit, catalog no.
DF1900; R&D Systems), following the manufacturer´s instructions. Intra- and interassay coefficients of variation were 6.0 and 7.5%, respectively.
Presentation of bile acid data
Bile acid concentrations are presented according to biological groupings (TBA, primary bile acids [CA, CDCA], and secondary bile acids [DCA, UDCA]) in Table 1 and Figures 1 and 2. Postprandial area under the curve (AUC) calculations for bile acids and FGF19 concentrations are shown in Table 1, and postprandial profiles and AUCs are displayed graphically in Figures 1 and 2. Time courses for individual bile acids are found in mental Figures 1–4. Fasting concentrations of bile acids and FGF19 are presented in Supplemental Table 1. Bile acids conjugated with glycine or taurine are termed amidates in tables andfigures.
Statistical analysis
Results are reported as medians and interquartile ranges (25th– 75th percentiles). Baseline values are defined as the mean of the three values obtained before consumption of the meals, and AUC values were calculated using the trapezoidal rule. The effect of group and meal type was analyzed by repeated measures ANOVA in a linear mixed-effect model using group and meal as fixed effects and subjects as random effect. The assumptions of a Gaussian distribution of residuals and homogeneity of variances were assessed visually by drawing histograms, residual plots, and probability plots. If assumptions could not be met, continuous variables were transformed using Box-Cox transformations. We chose to calculate type III sums of squares for the fixed effects. Between-group differences were tested using Student’s t test and Holm-Sidak’s adjustment for multiple comparisons. Correlations were assessed with the Spearman rank correlation test. A two-sided P value of .05 was used to indicate significant differences. Statisticalanalyses were performed using GraphPad Prism version 6.0b forWindows/Mac (GraphPad Software) and R (3.2.1) for Windows/Mac (http://cran.r-project.org).
Postprandial bile acid concentrations
29