Role of blood lipids in mediating the effect of dietary factors on gastroesophageal reflux disease: a two-step mendelian randomization study.

BACKGROUND: Growing studies have indicated an association between dietary factors and gastroesophageal reflux disease (GERD). However, whether these associations refer to a causal relationship and the potential mechanism by which dietary factors affect GERD is still unclear.

METHODS: A two-step mendelian randomization analysis was performed to obtain causal estimates of dietary factors, blood lipids on GERD. Independent genetic variants associated with 13 kinds of dietary factors and 5 kinds of blood lipids at the genome-wide significance level were selected as instrumental variables. The summary statistics for GERD were obtained from European Bioinformatics Institute, including 129,080 cases and 473,524 controls. Inverse variance weighted was utilized as the main statistical method. MR-Egger intercept test, Cochran's Q test, and leave-one-out analysis were performed to evaluate possible heterogeneity and pleiotropy. And the potential reverse causality was assessed using Steiger filtering.

RESULTS: The results of the inverse variance weighted method indicated that genetically predicted total pork intake (OR = 2.60, 95% CI: 1.21-5.58, p = 0.0143), total bread intake (OR = 0.68, 95% CI: 0.46-0.99, p = 0.0497), total cereal intake (OR = 0.42, 95% CI: 0.31-0.56, p = 2.98E-06), and total cheese intake (OR = 0.41, 95% CI: 0.27-0.61, p = 1.06E-05) were associated with the risk of GERD. Multivariable Mendelian randomization analysis also revealed a negative association between total cereal intake, total cheese intake and the risk of GERD, but the effect of total pork intake and total bread intake on GERD disappeared after adjustment of smoking, alcohol consumption, use of calcium channel blockers, BMI, physical activity levels, and biological sex (age adjusted). Furthermore, the concentration of low-density lipoprotein cholesterol (LDL-C) is negatively correlated with total cheese intake, which mediates the impact of total cheese intake on GERD. The proportion mediated by LDL-C is 2.27% (95%CI: 1.57%, 4.09%).

CONCLUSIONS: This study provides evidence that an increase in total cereal intake and total cheese intake will decrease the risk of GERD. Additionally, LDL-C mediates the causal effect of total cheese intake on GERD. These results provide new insights into the role of dietary factors and blood lipids in GERD, which is beneficial for disease prevention.