Associations of remnant cholesterol with mental disorders: findings from retrospective cohort and Mendelian randomization analyses

Background Remnant cholesterol (RC) is an independent risk factor for atherosclerotic cardiovascular disease. The relationship between RC and mental disorders is not clear. This study sets out to evaluate associations between RC and mental disorders via retrospective cohort and bidirectional Mendelian randomization (MR) approaches. Methods This study used data on anxiety and depression from NHANES and the Second Affiliated Hospital of Nanchang University. Correlation analysis, restricted cubic splines, logistic regression, and stratified regression were employed to examine the associations between RC and these mental health conditions. Bidirectional MR analyses evaluated RC and 20 mental disorders. Instruments were selected at P < 5 × 10 -6 for the main analysis, with a complementary sensitivity analysis using genome-wide significant instruments (P < 5 × 10 -8 ). The Bonferroni threshold for 40 directional tests was P < 0.00125. Results Retrospective analysis of clinical EHR showed no association between RC and depression; instead, RC was significantly negatively correlated with anxiety. The NHANES-based retrospective analysis revealed no association between RC levels and either anxiety or depression. In the MR analysis, genetically predicted RC was positively associated with LBD (OR 1.45, 95% CI 1.19–1.77; P = 0.0002) and APOE4-positive LBD (OR 2.64, 95% CI 1.64–4.28; P = 0.000073), and both associations remained below the Bonferroni threshold. The corresponding estimates remained directionally consistent in the genome-wide-significance sensitivity analysis. Conclusions This study provides complementary evidence from clinical, population-based, and genetic analyses regarding the relationship between RC and mental disorders. Lower RC was associated with anxiety in the hospital cohort, while genetically predicted RC was positively associated with LBD-related outcomes. Further investigation in independent populations is warranted.