We read with interest the articles by Stering et al. 1 and ValletPichard et al. 2 The former authors developed the FIB-4 index, a noninvasive method for assessing liver fibrosis in patients with HIV/HCV coinfection. The variables used are age, aspartate aminotransferase (AST), alanine aminotransferase (ALT) and platelet (PLT) count, and the formula is as follows:). They showed that over 70% of patients could be classified into either absence or presence of advanced fibrosis by cutoff of Ͻ1.45 or Ͼ3.25 respectively, with diagnostic accuracy of 87%. The latter authors expanded the applicability of the FIB-4 index to HCVmonoinfected patients and showed that 73% of patients were classified with diagnostic accuracy of 93%, an excellent performance in both classification and accuracy of diagnosis.Because the mean age of patients was young in these studies (40 years 1 and 44 years 2 ), we wondered whether this index could also fit to Japanese patients who are rather older than the Western patients. We validated the FIB-4 index in a retrospective cohort of 1,405 patients who underwent liver biopsy at our hospital. The mean age was 55 Ϯ 12 years. The distribution of METAVIR fibrosis scores was as follows: 1.6% showed no fibrosis (F0), 44.8% showed mild fibrosis (F1), 29.5% showed moderate fibrosis (F2), 20.2% showed severe fibrosis (F3), and 3.9% showed cirrhosis (F4). The proportion of advanced fibrosis (F3 or F4) was slightly higher in our population compared to the former studies (24.1% vs. 20.7% 1 and 17.2% 2 ). As shown in Table 1, only 53% of patients were classified to either Ͻ1.45 or Ͼ3.25, a much lower rate than previous reports. The diagnostic accuracy was excellent in patients with a FIB-4 index Ͻ1.45 (94%), however, it was relatively poor in patients with a FIB-4 index Ͼ3.25 (50%) making the overall accuracy as low as 67%.We supposed this discordance with previous reports may be derived from the older age of our populations and thus we categorized patients into three groups according to age and analyzed separately. In patients with age Յ50 years, 64% of patients were classified, and the diagnostic accuracy was 94% for a FIB-4 index Ͻ1.45 and 68% for a FIB-4 index Ͼ3.25 making the overall accuracy of 90%, a result comparable to previous reports. In older patients, however, diagnostic accuracy was significantly low compared to those with age Յ50 years (56% for age 51-60 years, P Ͻ 0.0001 and 51% for age Ն60 years, P Ͻ 0.0001). Because patients with a FIB-4 index Ͼ3.25 increased according to age (6%, 34%, and 53% for ages Յ50, 51-60 and Ͼ60 years), and the diagnostic accuracy was low in these patients (48% to 50%), these results suggest that, in elderly patients, a variable "age" generates excessively high FIB-4 index leading to misclassification of no-moderate fibrosis (F0-F2) into a FIB-4 index Ͼ3.25.In conclusion, the FIB-4 index could accurately differentiate advanced fibrosis in young Japanese patients with chronic hepatitis C but the diagnostic accuracy is limited in the elderly. Thus, in elderly patients,...