Paris II and Rotterdam criteria are the best predictors of outcomes in patients with primary biliary cholangitis in Japan.

BACKGROUND Biochemical response to treatment in patients with primary biliary cholangitis (PBC) reflects prognosis. However, the best predictive criteria to detect biochemical response remain undetermined. In addition, because these criteria need > 6 months until definition, parameters that can estimate its results before initiating treatment are needed. METHODS We conducted a single-center retrospective study on 196 patients with PBC, followed up for at least 12 months after initiating treatment. RESULTS Kaplan-Meier analysis showed that Paris II (p = 0.002) and Rotterdam criteria (p = 0.001) could estimate the overall survival of PBC patients, whereas Paris II (p = 0.001), Rotterdam (p = 0.001), and Rochester criteria (p= 0.025) could estimate liver-related deaths. Cox hazard analysis revealed Paris II and Rotterdam criteria as significantly independent predictors of overall survival (hazard ratio (HR) 3.948, 95% CI 1.293-12.054, p = 0.016 and HR 6.040, 95% CI 1.969-18.527, p = 0.002, respectively) and liver-related deaths (HR 10.461, 95% CI 1.231-88.936, p = 0.032 and HR 10.824, 95% CI 1.252-93.572, p = 0.032, respectively). The results of Paris II criteria could be estimated by serum prothrombin time (Odds ratio (OR) 1.052, 95% CI 1.008-1.098, p = 0.021) and alanine transaminase level (OR 0.954, 95% CI 0.919-0.991, p = 0.014) whereas, those of Rotterdam criteria could be estimated by serum albumin level (OR 3.649, 95% CI 1.098-12.128, p = 0.035) at the time of diagnosis. CONCLUSIONS This study highlights the best prediction criteria and pre-treatment parameters that facilitate the prognosis of PBC patients.