Abstract
BACKGROUND & AIMS: Total bilirubin (TB) has been regarded as a key prognostic factor for patients with primary biliary cholangitis (PBC), yet the prognostic significance of direct bilirubin (DB) is unclear. This study aimed to investigate the role of DB in predicting outcomes in patients with PBC. METHODS: We retrospectively reviewed the medical records of patients with PBC, and collected the follow-up data on liver biochemistry and clinical outcomes including decompensation, hepatocellular carcinoma, liver transplantation (LT), and liver-related deaths through regular outpatient visits and phone interviews. RESULTS: A total of 553 patients with PBC were enrolled and grouped by TB and DB after 12 months of ursodeoxycholic acid treatment (TB12 and DB12) : Group 1 (elevated TB12 and DB12, n = 145), Group 2 (elevated TB12 and normal DB12, n = 88), and Group 3 (normal TB12 and DB12, n = 308). After a median follow-up of 5.42 years, 63 patients (11.4%) developed LT or liver-related deaths. Patients in Group 2 exhibited comparable transplant-free survival rates and hepatic events incidence to those in Group 3, yet demonstrated significantly better transplant-free survival rates and lower hepatic events incidence than Group 1(both p < 0.001). Multivariate Cox analyses identified DB12, rather than TB or TB12, as an independent predictor of long-term outcomes. Finally, patients with normal DB12 experienced improved survival compared to those with elevated DB12, irrespective of ALP levels. CONCLUSION: DB is a superior predictor of long-term outcomes than TB in PBC patients, attaining normal DB and ALP levels are associated with lower risk for LT or liver-related death. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12876-026-04667-y.