Abstract
BACKGROUND: The liver plays critical roles in human health. Circulating level of liver function biomarkers may associate with the long-term and short-term mortality in general population. METHODS: We used data from US National Health and Nutrition Examination Survey 1988-94 and 1999-2014. People aged ≥ 20 years with measured serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyltransferase (GGT), alkaline phosphatase (ALP), total bilirubin (TB), and albumin (ALB) at baseline were included. All-cause and cause-specific mortality was identified from the National Death Index through 31 December 2015. Additive Cox regression models were applied to assess the correlation patterns between the serum level of these analytes and mortality risk. RESULTS: A total of 44,508 participants were included; among them, 9,721 deaths occurred during a mean follow-up of 12.5 years. A "J-shaped" correlation was found between serum levels of ALT, AST, and TB and all-cause mortality. The risk of mortality monotonically increased with increasing GGT and ALP levels when their levels exceeded the valley points. A "L-shaped" correlation was found between the serum level of ALB and all-cause mortality. The correlation patterns were comparable among deaths from different causes and were consistent in subgroup and sensitivity analyses. While the integration of all six liver function biomarkers did not yield an optimal predictive performance for mortality (area under ROC curve = 0.706), it demonstrated a significantly better performance compared to any single biomarker.. CONCLUSION: Circulating liver function biomarkers showed diverse nonlinear correlations with mortality and may be utilized as part of a screening process to help identify individuals who may be at elevated risk of mortality.