Objective: We assessed the value of the recently developed aspartate aminotransferase to platelet ratio index (APRI) for predicting significant fibrosis or cirrhosis in patients with chronic hepatitis C or HBeAg-negative chronic hepatitis B.
Methods: In total, 489 patients (chronic hepatitis C, 284 patients; HBeAg-negative chronic hepatitis B, 205 patients) were included. APRI values of 0.50 or less and greater than 1.50 were evaluated for predicting significant fibrosis, and APRI values of 1.00 or less and greater than 2.00 for predicting cirrhosis. Liver biopsies were evaluated according to the Ishak's classification. Fibrosis was considered to be significant in cases with scores 3-6, and cirrhosis to be present in cases with fibrosis scores of 5 and 6.
Results: Significant fibrosis was observed in 56/148 (38%) patients with APRI< or = 0.50, 130/227 (57%) patients with 0.50<APRI< or = 1.50 and 84/114 (74%) patients with APRI>1.50 (P<10). Cirrhosis was observed in 47/311 (15%) patients with APRI< or = 1.00, 29/93 (31%) patients with 1.00<APRI< or = 2.00 and 37/85 (44%) patients with APRI>2.00 (P<10). The areas under receiver-operating characteristic curves were 0.65 and 0.70 for prediction of significant fibrosis or cirrhosis, respectively. The combination of APRI< or = 0.50 and APRI>1.50 classified correctly 36% of patients with or without significant fibrosis, while the combination of APRI< or = 1.00 and APRI>2.00 classified correctly 62% of patients with or without cirrhosis. There was no significant difference in the predictive values of APRI between patients with chronic hepatitis C and chronic hepatitis B.
Conclusions: APRI is significantly associated with the extent of fibrosis, but it does not classify correctly 40-65% of patients with chronic hepatitis C or HBeAg-negative chronic hepatitis B, and thus it cannot replace liver biopsy.