Patients received UDCA at a dose of 28 to 30 mg/kg/day (Axcan Pharma, Mont-St. Hiliare, Canada) in divided doses given with meals or an identical placebo. Serum samples were obtained at entry into the study and at the end of treatment. All available paired samples were retrieved and used for analysis of the bile acid composition. Disease progression to cirrhosis, development of varices, cholangiocarcinoma, liver transplantation, and death during the trial were considered clinical endpoints. Serum bile acids were analyzed qualitatively by conventional gas chromatography-mass

spectrometry (GCMS) and quantitatively by isotope-dilution GCMS, as described previously, with the following modification: alkaline hydrolysis with 2 M sodium hydroxide in 90% (vol/vol) ethanol LY2835219 (1 hour at

67°C) was performed instead of enzymatic hydrolysis with cholylglycine hydrolase.14 Deuterium-labeled LCA, deoxycholic acid (DCA), chenodeoxycholic acid (CDCA), cholic acid (CA), and UDCA as internal standards were obtained from CDN Isotopes (Pointe-Claire, Canada). Baseline characteristics were calculated as medians and ranges for continuous variables. The number and percent in each group were tabulated for categorical variables. Bile acid concentrations were calculated as means and standard deviations. The chi-square test of independence was used to determine Pifithrin-�� statistical significance for categorical data. For the continuous variables, the Wilcoxon rank sum test was used. Baseline characteristics with

a significance of P ≤ 0.2 were entered into a multivariate model of multiple linear regression analysis to explore possible correlations per bile acid. From 2001 to 2005, 150 patients with PSC were entered into the study. Serum for bile acid analysis was available at the baseline and at the end of the study (mean treatment duration Dimethyl sulfoxide = 2.38 ± 0.56 years) for 56 of these patients. The baseline characteristics of this subset of patients versus the remaining patients (n = 94) are shown in Table 1. Patients analyzed for their bile acid composition were younger (41.8 versus 49.3 years), were more likely to have concomitant inflammatory bowel disease (93% versus 68%), and had a lower Mayo risk score (0.015 versus 0.58). The baseline characteristics of this cohort of patients by treatment were comparable as well (Table 2). At the baseline, the bile acid pool consisted of CA (32%), LCA (13%), DCA (21%), UDCA (11%), and CDCA (23%). GCMS spectra indicating the prevalence of significant amounts (>0.05 μmol/L) of uncommon bile acids were not observed. Only traces of hydroxylation products of UDCA were occasionally seen.15 The GCMS systems that were used did not separate UDCA from isoUDCA. Patients who had undergone colectomy (n = 12) had significantly lower levels of DCA (P < 0.