Vitamin D and oral disease relationships: Insights from a bidirectional Mendelian randomization investigation

The causal relationship between serum 25-hydroxyvitamin D levels and oral diseases remains uncertain due to confounding and reverse causation in observational studies. This bidirectional 2-sample Mendelian randomization (MR) study aims to elucidate the causal relationship between serum 25-hydroxyvitamin D levels and multiple oral pathologies while addressing confounding factors and reverse causation observed in conventional observational studies. Utilizing summary-level genome-wide association study data from the IEU OpenGWAS project, we analyzed serum 25-hydroxyvitamin D levels ("ebi-a-GCST90000618," n = 4,96,946 Europeans) and oral disease phenotypes from Finnish biobank registries. Instrumental variables for serum 25-hydroxyvitamin D levels were selected under genome-wide significance (P < 5 × 10-8) with strict clumping thresholds (r2 < 0.001, kb = 10,000) and F-statistics > 30. For oral disease exposures, relaxed criteria (P < 1 × 10-5, r2 < 0.001, F-statistic > 10) were applied to ensure sufficient instrumental variables. Primary analyses employed inverse-variance weighted and MR-Egger methods, supplemented by sensitivity analyses including Cochran's Q test, leave-one-out validation, funnel plots, and MR-PRESSO for horizontal pleiotropy assessment. Forward MR analysis demonstrated significant inverse causal relationships between genetically elevated serum 25-hydroxyvitamin D levels and risks of perioral dermatitis and lip, oral cavity, pharyngeal malignancies. No statistically significant associations were observed with other oral diseases. Reverse MR analysis revealed a nominal positive association between lichen planus and serum 25-hydroxyvitamin D levels that did not persist after multiple testing correction, with no other reverse causal effects detected. Our findings provide genetic evidence supporting a protective role of serum 25-hydroxyvitamin D levels against specific oral disorders, particularly perioral dermatitis and oropharyngeal malignancies. The bidirectional design effectively mitigated reverse causation concerns, underscoring the robustness of these causal inferences. These results warrant further investigation into vitamin D supplementation as a potential preventive strategy for targeted oral pathologies.