Abstract
BACKGROUND/OBJECTIVES: Coffee consumption is associated with metabolic syndrome (MetS), but the findings have been inconsistent because of residual confounding. This study investigated the relationship between coffee consumption and MetS and its components using Mendelian randomization (MR) with genetic instrumental variables (IVs). SUBJECTS/METHODS: Coffee consumption was calculated from a food-frequency questionnaire using the 12-mon frequency and portion size, with cups converted to grams (2.7 g/cup) in the Korean Genome and Epidemiology Study-Health Examinees cohort (n = 58,144) for genetic instrument selection. MetS was defined by the National Cholesterol Education Program Adult Treatment Panel III guidelines in the Ansan and Ansung cohort (n = 5,351) for the outcome estimation. Genome-wide association analysis (GWAS) for coffee consumption (g/day) was conducted using linear regression, and logistic regression was used for MetS and its components. IVs were single-nucleotide polymorphisms associated with coffee consumption at genome-wide significance (P < 5 × 10(-6)) from GWAS after linkage disequilibrium clumping (r(2) < 0.05 within ± 500 kb). MR analyses of the association between coffee consumption and MetS and its components were conducted using 5 approaches (inverse variance weighted [IVW], MR-Egger, weighted median, weighted mode, and simple mode). Sensitivity analyses were conducted using Cochran's Q and the horizontal pleiotropy test. RESULTS: IVW indicated a negative association with MetS in males (odds ratio, 0.702; 95% confidence interval, 0.613-0.805). In females, MetS and its components had no association (P < 0.05). Cochran's Q detected heterogeneity in females for elevated waist circumference, and MR-pleiotropy residual sum and outlier analysis identified outliers, but the results remained unchanged. CONCLUSION: This MR study revealed sex-specific causal relationships between coffee consumption and MetS in Korean adults, suggesting protection against MetS and its components in males.