Non-invasive hepatic steatosis and fibrosis indices predict differentially mortality in the adult Korean population

Abstract

Since the association of hepatic fibrosis and steatosis non-invasive indices with mortality remain controversial, their association with all-cause, cardiovascular, cancer and liver-related mortality was evaluated in the Korean population. In this registry-based, cohort study, data were retrieved from the Korea National Health and Nutrition Examination Survey and mortality data from the Korean Cause of Death data registry; 40,491 individuals followed-up for 8.6 years (median). Hepatic fibrosis was evaluated with alanine aminotransferase (AST)-to-platelet ratio index (APRI), body mass index-AST-to-alanine aminotransferase (ALT) ratio-diabetes mellitus (BARD) and metabolic dysfunction-associated fibrosis-5 (MAF-5) score, and steatosis was evaluated with liver fat score (LFS) and lipid accumulation product (LAP). Cox regression analysis showed that APRI (<1.0 vs. ≥1.0) was independently associated with all-cause (hazard ratio [HR] 3.84, 95% confidence interval [CI] 2.30-6.43, C-index 0.870), cancer (HR 4.21, 95%CI 1.88-9.45, C-index 0.866) and liver-related (HR 25.36, 95%CI 11.02-58.38, C-index 0.909) mortality. MAF-5 (<1.0 vs. ≥1.0) was independently associated with all-cause mortality (HR 1.50, 95%CI 1.10-2.03, C-index 0.868) and liver-related mortality (HR 8.35, 95%CI 3.58-19.46, C-index 0.911). LFS (<1.257 vs. ≥1.257), was independently associated with all-cause (HR 1.55, 95%CI 1.14-2.12, C-index 0.872) and liver-related (HR 7.00, 95%CI 1.63-29.96, C-index 0.887) mortality. LAP (<38.05 vs. ≥38.05) was independently associated with cardiovascular mortality (HR 2.23, 95%CI 1.40-3.58, C-index 0.898). BARD was not associated with mortality. APRI, MAF-5, LFS were independently associated with all-cause mortality, LAP (cut-off 38.05) with cardiovascular mortality, APRI with cancer mortality, and APRI, MAF-5, LFS with liver-related mortality in the adult Korean population.