Prediction of morbidity and mortality in patients with chronic hepatitis C by non-invasive liver fibrosis models
Abstract
Background & Aims
Liver fibrosis is prognostic of outcomes among patients with chronic hepatitis C (CHC). We evaluated the accuracy of non-invasive markers and liver biopsy in predicting morbidity and mortality in CHC patients.
Methods
Compensated CHC patients were evaluated over a 10-year period. Non-invasive markers including Hepascore, FIB-4, APRI and liver biopsy results were retrospectively collated. Follow-up morbidity and mortality data were obtained from the Western Australian Data Linkage System. The prognostic significance of baseline non-invasive markers and biopsy were assessed using Kaplan–Meier analysis.
Results
A total of 406 subjects (64% male, mean age 48 ± 11 years) were followed up for 2385 person-years, during which there were 22 (5.4%) deaths including 14 (3.4%) who died from liver disease or required liver transplantation. Sixteen (3.9%) subjects developed liver decompensation. Hepascore and liver biopsy (P < 0.005) but not APRI or FIB-4 were predictive of overall and liver-related mortality as well as liver decompensation. A Hepascore>0.5 was associated with increased overall mortality [Hazard Ratio (95%CI) 6.7 (2.6–17), P < 0.001], liver-related mortality [32.8 (4.3–250), P = 0.001] and risk of future decompensation [11.8 (3.3–41), P < 0.001], whereas a Hepascore ≤0.5 was associated with a 99% probability of not dying from liver-related causes over 10 years. Hepascore had comparable accuracy with liver biopsy in predicting liver-related mortality with AUROC of 0.86 (95%CI 0.80–0.90) and 0.87 (0.79–0.96), respectively.
Conclusion
Hepascore is predictive of overall and liver-related mortality and morbidity in CHC patients with comparable accuracy to liver biopsy. Hepascore may be a useful prognostic marker in clinical practice.
