Updated Reference Limits for Liver Blood Tests With Validation Against Long-Term Liver-Related Outcomes

Liver blood tests are widely used, but their reference limits may not reliably predict long-term liver-related risks. Previous studies to establish reference limits often lacked rigorous exclusion of individuals with undiagnosed liver disease or associated risk factors. This study aims to establish more precise reference limits in a rigorously defined healthy middle-aged population and validate their clinical relevance against major adverse liver outcomes (MALOs).

We analysed 5412 participants from the Finnish population-based Health 2000 Survey, systematically excluding individuals with baseline or future liver disease during the next 10 years, advanced fibrosis, liver-related risk factors, hepatotoxic medication use, or systemic conditions to define a truly healthy reference population. Alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl transferase (GGT), alkaline phosphatase (ALP), and bilirubin were measured using standard methods. External validation was conducted in three independent FINRISK cohorts (n = 20 423) against incident MALOs.

The updated reference limits (97.5th percentile) in the healthiest subpopulation for men and women, respectively, were: ALT 57 and 35 U/L, AST 49 and 33 U/L, GGT 48 and 75 U/L, ALP 108 and 93 U/L, and total bilirubin 28 and 25 μmol/L. A 10-year MALO risk of > 5% was observed for AST > 2× and GGT > 3× the upper reference limit, whereas no ALT elevation exceeded this risk level. At AST > 2× the upper reference limit, the MALO risk surpassed or equalled competing mortality risk.

Updated reference limits for liver blood tests were established in a well-characterised population and validated against liver outcomes, with AST and GGT proving to be stronger predictors than ALT.