Association of smoking with coronary artery disease and myocardial infarction: A Mendelian randomization study

Observational studies have suggested that smoking is associated with coronary artery disease (CAD).1,2 Compared with non-smokers, there is a 29% to 80% increased risk of mortality due to CAD in people who smoke less than one pack per day.3 However, the exact relationship between some smoking phenotypes, such as the age at which one began to smoke, and CAD is unclear. Furthermore, the association of smoking with CAD in observational studies is limited to confounding factors such as obesity, hypertension, diabetes and hyperlipidemia. The causal relationship between smoking and CAD will be biased for causal inference such as weight gain accompanied by smoking cessation.1 Mendelian randomization analysis is a method that uses exposure-associated instrumental variables to assess the causality between exposure and outcome, which can reduce the bias in observational research. To examine the potential role of smoking in CAD and myocardial infarction risk, we conducted this Mendelian randomization study, using single-nucleotide polymorphisms (SNPs) associated with smoking phenotypes as instrumental variables at the genome-wide significance level (p < 5 × 10−8) in genome-wide association studies (GWASs).4 Summary statistics for the association of smoking-related SNPs with CAD and myocardial infarction were obtained from the Coronary Artery Disease Genome-Wide Replication and Meta-analysis plus the Coronary Artery Disease Genetics (CardiogramplusC4D) consortium.5 A standard Mendelian randomization analysis (fixed-effects inverse-variance weighted (IVW) method) was used to combine the estimates. A two-sided p-value of <0.025 (=0.05/2 outcomes) was set as the threshold for significance. The GWAS on smoking initiation (current or former smoker) identified 378 SNPs, accounting for 7.8% of the heritability in up to 1,232,091 individuals.3 While 13 SNPs were unavailable in both the CAD and myocardial infarction databases, eight proxy SNPs in linkage disequilibrium (LD) (R2 ≥ 0.8) with the specified SNPs were used. Finally, for our inclusion criteria, we chose 373 out of 378 SNPs; similarly, nine, 24 and 54 SNPs were included in the Mendelian randomization analyses of the age at initiation of regular smoking, smoking cessation, and cigarettes per day, respectively. The analytic sample included 60,801 CAD patients, of which 43,676 had a myocardial infarction, and 123,504 were referents. Genetically predicted smoking initiation was associated with a higher risk of CAD in the standard IVW analysis, with an odds ratio (OR) of 1.26 (95% confidence interval (CI), 1.19–1.34; p = 3.9 × 10–16) (Figure 1). Similarly, the genetically instrumented smoking initiation increased the risk of myocardial infarction (OR, 1.28; 95% CI, 1.20–1.36; p = 6.0 × 10–15). The age at initiation of regular smoking was negatively associated with myocardial infarction but not CAD, with the OR of 0.58 (95% CI, 0.37–0.92; p = 0.02) per standard deviation (SD) increase. The OR per SD increase in cigarettes per day was 1.17 (95% CI, 1.03–1.33; p = 0.01) for myocardial infarction, and 1.08 (95% CI, 0.96–1.21; p = 0.18) for CAD, respectively. The standard IVW estimate revealed a 40% and 44% increased risk of CAD and myocardial infarction in current smokers, compared with that in former smokers, respectively. There was no evidence of directional pleiotropy (all p > 0.05). Mendelian randomization association between genetically predicted smoking and coronary artery disease and myocardial infarction. Odds ratios are scaled to per standard deviation increase in genetically predicted smoking initiation (about 10–12% increase in the probability of being a regular smoker), age at initiation of regular smoking (about additional 0.31–0.50 years), cigarettes per day (about 2–3 additional cigarettes daily) and smoking cessation (about 3–5% increase in the probability of being a current smoker). *Current smoker versus former smoker. CI: confidence interval; OR: odds ratio The association of genetically predicted smoking with CAD and myocardial infarction may be caused by increasing systemic catecholamine, reducing oxygen-carrying capacity and promoting atherosclerosis through effects on endothelial function, oxidative stress and thrombosis.6 Smoking also increases the risk of other cardiovascular diseases, such as heart failure.7 This study includes a large sample of up to 60,801 CAD and 43,676 myocardial infarction patients and can reduce potential biases in observational studies. However, there are several limitations. First, instrumental variables may affect the outcome through other risk factors although we found no evidence of pleiotropy. Second, there are two studies included in both the smoking database and CardiogramplusC4D consortium, in spite of accounting for a low proportion (approximately 1%) in smoking GWASs. In conclusion, our Mendelian randomization study indicates that smoking initiation and current smokers are associated with an increased risk of CAD and myocardial infarction. People who started smoking at a younger age and smoke more cigarettes per day are more prone to myocardial infarction but not to CAD. LYL and ZW contributed equally to this study. LYL and LRZ contributed to the concept and design of the work. LYL and ZW contributed to the statistical analysis of the work. All authors contributed to the acquisition, analysis, or interpretation of data. LYL and ZW drafted the manuscript and LRZ critically revised the manuscript. All authors gave final approval and agreed to be accountable for all aspects of the work in terms of integrity and accuracy. The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Key Research and Development Program of China (grant no. 2016YFC1301003); the National Science Foundation of China (grant no. 81873484); the Natural Science Foundation of Zhejiang Province, Zhejiang, China (grant no. LZ16H020001).