摘要

Aim: To develop algorithms for detecting significant fibrosis and cirrhosis in chronic hepatitis B (CHB) patients with the aim of reducing unwarranted liver biopsy. @@@ Methods: For 307 CHB patients, the aspartate aminotransferase-to-platelet ratio index (APRI), the fibrosis index based on four factors (FIB-4), and the result of transient elastography with FibroScan (FS) were obtained when a liver biopsy was carried out. All patients were classified based on APRI or FIB-4 score and further assessed by FS results. Patients who remained unclassified after two steps of evaluation were considered to need liver biopsy. @@@ Results: Algorithm implementation found that APRI + FS significantly lowered the requirement for liver biopsy for the detection of significant fibrosis compared to either individual APRI or FS screening (65.1% vs 75.9% or 78.5%, P=0.003 or < 0.001, respectively). The combination of FIB-4 + FS significantly reduced the need for liver biopsy compared to single FIB-4 or FS (58.3% vs 67.4% or 78.5%, P=0.019 or < 0.001, respectively). The FIB-4 + FS algorithm also reduced the need for liver biopsy for detection of significant fibrosis in patients > 50 years old compared to APRI + FS (22.6% vs 56.5%, P < 0.001), with a relatively lower accuracy (83.9% vs 98.4%, P=0.004). Only 3.6% or 1.3% of patients needed liver biopsy for diagnosis of cirrhosis after screening with APRI + FS or FIB-4 + FS, respectively. @@@ Conclusion: The APRI + FS and FIB-4 + FS algorithms could significantly reduce the need for liver biopsy with high accuracy, sensitivity, and positive predictive value for diagnosis of significant fibrosis and cirrhosis in CHB patients.