Epidemiological survey of hemoglobin A1c and liver fibrosis in a general population with non-alcoholic fatty liver disease.

ABSTRACT

AIM: The association between glycemia and liver fibrosis was analyzed using hemoglobin A1c (HbA1c) and the Fibrosis-4 (FIB-4) index in a large general population cohort that underwent a health checkup. METHODS: A total of 6927 subjects without hepatitis B or C virus infection or habitual alcohol intake were enrolled. Non-alcoholic fatty liver disease (NAFLD) was diagnosed by ultrasonography and potential liver fibrosis (FIB-4 index ≥1.3) in NAFLD was analyzed in relation to HbA1c level. Factors associated with potential liver fibrosis of NAFLD were also analyzed. RESULTS: The overall frequency of NAFLD was 27.9% (1935 subjects) and the frequency of NAFLD by HbA1c level (<4.9%, 5.0-5.9%, 6.0-6.9%, 7.0-7.9%, ≥8.0%) was 16%, 27%, 54%, 53%, and 54%, respectively. Among the 1935 NAFLD cases, the frequency of potential liver fibrosis was 25.2% (487 subjects) overall and 19%, 22%, 30%, 52%, and 31%, respectively, by HbA1c category. From multivariate analysis, an HbA1c level ≥6.5% was significantly associated with potential liver fibrosis (P = 0.017, hazard ratio = 1.7). CONCLUSIONS: The prevalence of NAFLD and liver fibrosis of NAFLD increased according to glycemia, up to 8.0% HbA1c. Measuring HbA1c and calculating the FIB-4 index in health checkups could help to identify potential cases of liver fibrosis of NAFLD, which should then be further evaluated using other techniques to confirm liver fibrosis.