Serum 25(OH)D3 and advanced liver fibrosis in type 2 diabetes: a Chinese hospital cohort with NHANES validation and mortality follow-up.

BACKGROUND

Liver fibrosis is an important predictor of mortality in patients with type 2 diabetes mellitus (T2DM). Current clinical practice relies on total 25-hydroxyvitamin D, which does not distinguish between its two main circulating isoforms, 25(OH)D2 and 25(OH)D3, although these metabolites may differ in their associations with liver fibrosis and long-term prognosis.

METHODS

We analyzed two independent T2DM cohorts: hospitalized Chinese patients with T2DM (n = 1,978) and US adults with T2DM from the National Health and Nutrition Examination Survey (NHANES) 2007-2018 (n = 4,616). Serum 25(OH)D2 and 25(OH)D3 were quantified by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Isoform-specific associations with advanced fibrosis, defined by the NAFLD Fibrosis Score (NFS) and Fibrosis-4 index (FIB-4), were examined using multivariable logistic regression and restricted cubic spline (RCS) analyses. In NHANES, multivariable Cox proportional-hazards models were used to evaluate all-cause, cardiovascular, and diabetes-related mortality according to vitamin D status.

RESULTS

In both cohorts, lower serum 25(OH)D3 was non-linearly associated with a higher risk of advanced liver fibrosis. By contrast, 25(OH)D2 showed no protective association; in NHANES, the highest tertile was associated with a higher risk of NFS-defined fibrosis (OR = 1.41, P = 0.003). RCS analyses suggested cohort-specific apparent inflection regions for 25(OH)D3 at 11.88 ng/mL in the Chinese cohort and 23.86 ng/mL in NHANES; these exploratory estimates were not intended as clinical cut-offs. In longitudinal NHANES analyses, lower 25(OH)D3 was associated with higher cause-specific mortality among participants with advanced fibrosis; however, estimates within the FIB-4 high-risk stratum were based on sparse events and were therefore regarded as exploratory and interpreted with caution. Bioinformatic analysis implicated the AGE-RAGE and PI3K-Akt signaling pathways as candidate mechanisms.

CONCLUSION

Lower serum 25(OH)D3, but not 25(OH)D2, was associated with advanced liver fibrosis in T2DM after multivariable adjustment. Among participants with advanced fibrosis, lower 25(OH)D3 was also associated with higher cardiovascular mortality, although cause-specific estimates based on sparse events were exploratory. These findings may support further investigation of isoform-specific 25(OH)D3 assessment for risk stratification in this population.