Antioxidant-Related Biomarkers and Long-Term Mortality in Probable MASLD: A Joint Analysis of Albumin and Bilirubin

Albumin and bilirubin contribute to endogenous antioxidant defenses, but their circulating concentrations also reflect hepatic and systemic processes. We examined their mutually adjusted associations with mortality in probable metabolic dysfunction-associated steatotic liver disease (MASLD). This UK Biobank analysis included 73,099 participants with fatty liver index ≥ 60, cardiometabolic risk, quantifiable low alcohol intake, and complete covariates. Albumin and total bilirubin were modeled simultaneously using restricted cubic splines. During 1,079,873 person-years, 10,467 deaths occurred. The fully adjusted hazard ratio for albumin 42 versus 46 g/L was 1.21 (95% confidence interval 1.15–1.27). For bilirubin 12 versus 6 µmol/L, it was 0.96 (0.91–1.02), attenuated from 0.89 (0.84–0.94) after age and sex adjustment. The low-albumin/low-bilirubin profile had a hazard ratio of 1.18 (1.10–1.28) relative to the middle/middle profile, without evidence of interaction (p = 0.810). Albumin associations with cardiovascular and non-liver cancer mortality survived correction for six secondary comparisons; the liver mortality association did not. Sensitivity analyses, including a post-review model allowing adjustment-covariate effects to vary across follow-up, supported the albumin finding. Lower albumin was consistently associated with mortality, whereas an independent bilirubin association was not established. Antioxidant function was not measured; these observational associations do not demonstrate antioxidant protection or therapeutic benefit.