Differential associations of vitamin D metabolites with metabolic dysfunction-associated steatotic liver disease: a cross-sectional study in a northern Chinese population

Background Metabolic dysfunction-associated steatotic liver disease (MASLD) is a common chronic liver disease. Previous studies have primarily focused on total 25-hydroxyvitamin D [25(OH)D], whereas the associations of 25-hydroxyvitamin D₂ [25(OH)D₂] and 25-hydroxyvitamin D₃ [25(OH)D₃] with MASLD remain incompletely understood. This study aimed to evaluate the associations of serum 25(OH)D₂, 25(OH)D₃, and total 25(OH)D concentrations with the prevalence of MASLD. Methods This cross-sectional study included 41,074 adults undergoing routine health examinations in northern China from the Dalian Health Management Cohort. Multivariable logistic regression models were used to evaluate the associations of vitamin D metabolites with MASLD. Restricted cubic spline (RCS) analyses and two-piece logistic regression models were used to assess nonlinear associations and estimate exploratory turning points. Prespecified subgroup analyses were conducted to examine the consistency of the associations across population subgroups. Sensitivity analyses evaluated the associations under alternative approaches to data handling and covariate adjustment. Results In the primary adjusted model, each one-standard-deviation increase in total 25(OH)D and 25(OH)D₃ was associated with 11% (OR = 0.889, 95% CI: 0.864–0.915) and 10% (OR = 0.897, 95% CI: 0.871–0.923) lower odds of MASLD, respectively. For 25(OH)D₂, the association was small and not statistically significant in the primary adjusted model (OR = 0.980, 95% CI: 0.953–1.007; p = 0.152), although it reached statistical significance in some other models. An inverted U-shaped association was observed for 25(OH)D₃, with an exploratory turning point at 14.32 ng/mL. For 25(OH)D₃, subgroup associations were consistently inverse in direction, although their magnitudes differed (all P for interaction < 0.001). The adjusted inverse associations of total 25(OH)D and 25(OH)D₃ with MASLD persisted across sensitivity analyses. Conclusion Total 25(OH)D and 25(OH)D₃ were inversely associated with prevalent MASLD after adjustment, with a nonlinear, inverted U-shaped pattern observed for 25(OH)D₃. Associations for 25(OH)D₂ were small in magnitude, and their clinical relevance remains uncertain. Further prospective studies and randomized trials are needed to clarify the causal nature and clinical implications of these associations.