The independent effects of fatigue and UDCA therapy on mortality in primary biliary cirrhosis: Results of a 9 year follow-up

Background & Aims: Long-term outcome in primary biliary cirrhosis (PBC) remains unclear. Whilst response to ursodeoxycholic acid (UDCA) is associated with good outcome, this effect is not universal. Early data from our group have suggested that one factor associated with a poorer outcome in PBC is fatigue. The aim of this study was to explore the inter-relationship between UDCA use, response, and fatigue in determining outcome over 9 years in a unique, comprehensive patient cohort. Methods: Longitudinal prospective study of a geographically-defined complete cohort of PBC patients in North-East England and matched community controls. Results: Survival to death or transplant was significantly lower in PBC patients than in the case-control population (88/136 (65%) v 114/136 84% (p <0.001 by log-rank test), with better survival in UDCA responders (defined using the Paris criteria) than in patients not treated with UDCA at study outset. Compared to the whole control group survival was reduced in PBC patients fatigued at study outset but not in those without fatigue (p <0.0001); an effect independent of the beneficial effect of UDCA response and of conventional parameters of liver disease severity. UDCA responders without fatigue at the study outset had a 9 year survival which was identical to controls. Patients without fatigue at the study outset who developed fatigue during follow-up had significantly worse survival than patients who remained without fatigue throughout (p <0.05). Fatigued controls had worse survival than non-fatigued controls (p = 0.05). Conclusions: Survival in a comprehensive cohort of PBC patients is substantially reduced compared with case-matched community controls. Development of fatigue and non-treatment with UDCA were specifically (and independently) associated with increased risk of death in PBC. (C) 2010 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.