Abstract
BACKGROUND: We applied a network Mendelian randomization (MR) framework to determine the causal association between body mass index (BMI) and coronary heart disease (CHD) and explored whether glycated hemoglobin (HbA1c) and lipid parameters (total cholesterol, TC; low-density lipoprotein cholesterol, LDL; high-density lipoprotein cholesterol, HDL; triglycerides, TG) serve as causal mediators from BMI to CHD by integrating summary-level genome-wide association study data. METHODS: Network MR analysis, an approach using genetic variants as the instrumental variables for both the exposure and mediator to infer causality was performed. Summary statistics from the GIANT consortium were used (n = 152,893) for BMI, CARDIoGRAMplusC4D consortium data were used (n = 184,305) for CHD, Global Lipids Genetics Consortium data were used (n = 108,363) for TC, LDL, HDL and TG, and MAGIC consortia data were used (n = 108,363) for HbA1c. RESULTS: The inverse-variance-weighted-method estimate indicated that the odds ratio (95% confidence interval) for CHD was 1.562 (1.391-1.753) per 1 standard deviation (kg/m(2)) increase in BMI. Results were consistent in MR Egger method and weighted-median methods. MR estimate indicated that BMI was positively associated with HbA1c and TG, and negatively associated with HDL, but was not associated with TC or LDL. Moreover, HbA1c, TC, LDL, and TG were positively associated with CHD, yet there was no causal association between HDL and CHD. HbA1c was positively associated with TC, LDL, and HDL, but was not associated with TG. CONCLUSIONS: Higher BMI conferred an increased risk of CHD, which was partially mediated by HbA1c and lipid parameters. HbA1c and TG might be the main mediators in the link from BMI to CHD.