| BackgroundThe Global Burden of Disease Study 2019 that consisted of 369 diseases and injuries in 204 countries and regions showed that cardiovascular diseases and cancers occupy the top two positions respectively.China has a large number of patients with cardiovascular diseases and cancers.Facing the threat of a major disease burden,how to reasonably and effectively prevent cardiovascular diseases and cancers is currently an urgent problem for China.Although some observational epidemiological studies have shown that recognized health risk factors including obesity,blood glucose and insulin related indicators are risk factors for the occurrence and development of cardiovascular diseases and cancers,while general inconsistencies are observed among the results of previous studies.In observational epidemiological studies,the association between exposure and outcome is difficult to avoid deviations caused by confounding factors and reverse causal effects,leading to false associations,and it is hard to obtain the true causal relationship and exact effect between exposure and outcome.Therefore,whether obesity,blood glucose and insulin related indicators are causally related to the occurrence and development of cardiovascular diseases and cancers is still controversial.Mendelian Randomization is a method that uses SNP(single nucleotide polymorphism)to unbiasedly test or estimate the causal association between exposure and outcome.With the continuous development of genome-wide association studies,Mendelian Randomization has been widely used to identify the causality in medical research.Research MethodsThis present study used a two-sample Mendelian Randomization method to test and evaluate blood glucose related indicators including fasting blood glucose(FBG),2-hour postprandial blood glucose(2hGlu),glycosylated hemoglobin(HbA1c),and insulin related indicators including fasting insulin(FI)and Homeostasis Model Assessment-insulin resistance index(HOMA-IR)and obesity indicators including body mass index(BMI,surrogate measures of overall obesity)and BMI-adjusted waist-tohip ratio(WHRadjBMI,surrogate measures of abdominal obesity)for 12 types of cardiovascular diseases(hypercholesterolemia,hypertension,coronary heart disease,ischemic stroke,deep vein thrombosis,myocardial infarction,angina pectoris,hypertrophic cardiomyopathy,thrombophlebitis,varicose veins,atrial fibrillation and flutter and cardiogenic stroke)and 26 cancers(basal cell carcinoma,gallbladder cancer,malignant melanoma,oropharyngeal cancer,ER+ breast cancer,ER-breast cancer,endometrial cancer,endometrioid ovarian cancer,high-grade serous ovarian cancer,low-grade serous ovarian cancer,invasive mucinous ovarian cancer,ovarian clear cell carcinoma,cervical cancer,lung squamous cell carcinoma,lung adenocarcinoma,renal cell carcinoma,prostate cancer,neuroblastoma,rectal cancer,small bowel cancer,pancreatic cancer,colon cancer,thyroid cancer,glioma,Paget's disease and bladder cancer).In addition,the causal effect of obesity on blood glucose and insulin related indicators was also evaluated.All data were obtained from the European population GWAS database.Firstly,based on the genome-wide association study summary statistics data of FBG(N=140,595),2hGlu(N=42,854),HbA1c(N=123,665),FI(N=98,210),HOMA-IR(N=37,037),BMI(N=806,834)and WHRadjBMI(N=694,649),the instrumental variables of exposure factors(P?5×10-8,r2<0.001,genetic distance?10000 kb)were identified and their strength were tested using the F-statistics;Secondly,a series of Mendelian Randomization methods(Wald ratio method,inverse variance weighting method,MR-Egger method,weighted median method,and MRPRESSO method)were used to infer causal effects;Thirdly,Cochran's Q test and I2 statistics are conducted to assess heterogeneity in Mendelian Randomization analysis;Both MR-Egger method and MR-PRESSO method were used to evaluate the horizontal pleiotropy;Finally,sensitivity analysis(leave-one-out method)was conducted to test the robustness of the results.Research Results(1)A total of 44,7,37,13,8,546 and 330 SNPs served as instrumental variables for FBG,2hGlu,HbA1c,FI,HOMA-IR,BMI and WHRadjBMI,and their F values are 63.58,112.71,16.67,14.89,7.59,88.22 and 98.38,respectively.Except for HOMA-IR,the weak instrumental variable bias of the other six exposure factors was negligible.(2)Causal effects of blood glucose related indicators on cardiovascular diseases and cancers.Among blood glucose related indicators,FBG and 2hGlu have limited effects on cardiovascular diseases.The elevated FBG and 2hGlu only have a weak causal link to the increased risk of angina pectoris[OR:1.005;95%CI:1.000-1.009 vs.OR:1.007;95%CI:1.002-1.013],while have no causal association with the other cardiovascular diseases.However,elevated HbA1c is associated with the increased risk of several cardiovascular diseases including coronary heart disease[OR:1.662;95%CI:1.298-2.129],myocardial infarction[OR:1.54;95%CI:1.199-1.979],hypercholesterolemia[OR:1.054;95%CI:1.03-1.08],hypertension[OR:1.046;95%CI:1.011-1.083]and angina pectoris[OR:1.013;95%CI:1.004-1.022].Among the Mendelian Randomization analysis between blood glucose related indicators and 26 types of cancers,only th e elevated 2hGlu is causally associated with the increased risk of ER-breast cancer[OR:1.216;95%CI:1.075-1.375].Neither FBG nor HbAlc has a significant causal relationship with these 26 types of cancers.(3)Causal effects of insulin related indicators on cardiovascular diseases and cancers.Among insulin related indicators,elevated HOMA-IR is significantly associated the decreased risk of hypercholesterolemia[OR:0.966;95%CI:0.963-0.969],and has no no causal relationship with other cardiovascular diseases.Compared with HOMA-IR,FI has a greater causal relationship with cardiovascular diseases.Elevated FI is causally associated with increased coronary heart disease[OR:2.159;95%CI:1.433-3.253],myocardial infarction[OR:1.873;95%CI:1.224-2.866],hypertension[OR:1.144;95%CI:1.058-1.238],deep vein thrombosis[OR:1.018;95%CI:1.0051.030]and hypertrophic cardiomyopathy[OR:1.001;95%CI:1.000-1.003].Unlike blood glucose,insulin related indicators have a significant causal relationship with the risk of several cancers.Among them,both elevated FI and HOMA-IR significantly increased the risk of high-grade serous ovarian cancer[OR:1.873;95%CI:1.006-3.486 vs.OR:1.958;95%CI:1.07-3.583].In addition,up-regulated FI and HOMA-IR has significant causality with the increased risk of lung squamous cell carcinoma[OR:3.171;95%CI:1.118-8.994]and ER+ breast cancer[OR:1.354;95%CI:1.014-1.808],respectively.(4)Causal effect of obesity on cardiovascular diseases and cancers.The causal link between obesity and cardiovascular diseases and cancers is greater than that of blood glucose and insulin.The elevated BMI and WHRadjBMI are causally associated with a variety of cardiovascular diseases including hypertension[OR:1.125;95%CI:1.1151.135 vs.OR:1.065;95%CI:1.053-1.078],coronary heart disease[OR:1.538;95%CI:1.44-1.644 vs.OR:1.332;95%CI:1.221-1.453],hypercholesterolemia[OR:1.039;95%CI:1.033-1.045 vs.OR:1.049;95%CI:1.041-1.056],angina[OR:1.018;95%CI:1.015-1.02 vs.OR:1.012;95%CI:1.008-1.015],myocardial infarction[OR:1.515;95%CI:1.41-1.628 vs.OR:1.331;95%CI:1.214-1.459].It is noteworthy that the increased risk of ischemic stroke[OR:1.288;95%CI:1.115-1.488]is only causally associated with the increase in WHRadjBMI,but not with BMI.It suggests that the distribution of body fat affects the risk of cardiovascular diseases.The causal inference analysis of obesity and cancer shows that elevated BMI is causally related to the risk of more cancers.Among them,elevated BMI is causally associated with increased risk of lung squamous cell carcinoma[OR:1.462;95%CI:1.236-1.729],endometrioid ovarian cancer[OR:1.239;95%CI:1.034-1.485],endometrial cancer[OR:1.001;95%CI:11.002]and renal cell carcinoma[OR:1.001;95%CI:1-1.002]and decreased risk of ER+breast cancer[OR:0.868;95%CI:0.812-0.927],ER-breast cancer[OR:0.857;95%CI:0.784-0.936],prostate cancer[OR:0.999;95%CI:0.997-1]and malignant melanoma[OR:0.998;95%CI:0.997-1].The elevated WHRadjBMI also has a causal association with reduced risk of ER+breast cancer[OR:0.889;95%CI:0.824-0.959].However,unlike BMI,the increase of WHRadjBMI has a causal link to decreased risk of lung squamous cell carcinoma[OR:0.787;95%CI:0.643-0.962].It suggests that the distribution of body fat has different causal effects on the risk of cancers.(5)Causal effect of obesity on blood glucose and insulin related indicators.Both BMI and WHRadjBMI have a significant causal relationship with the increased blood glucose and insulin related indicators.It is noteworthy that BMI has a greater effect on blood glucose related indicators FBG[OR:1.081;95%CI:1.062-1.099 vs.OR:1.024;95%CI:1.004-1.045]and insulin related indicators including FI[OR:1.181 95%CI:1.157-1.207 vs.OR:1.128 95%CI:1.100-1.157]and HOMA-IR[OR:1.223;95%CI:1.190-1.258 vs.OR:1.129;95%CI:1.091-1.169]than that of WHRadjBMI.However,causal effect of WHRadjBMI on blood glucose related indicators including 2hGlu[OR:1.282;95%CI:1.076-1.527 vs.OR:1.192;95%CI:1.043-1.363]and HbA1c[OR:1.054;95%CI:1.037-1.072 vs.OR:1.049;95%CI:1.036-1.063]are greater than that of BMI.It suggests that the distribution of body fat has different causal effects on the regulation of blood glucose and insulin related indicators.Research ConclusionIn summary,this present study systematically explored the causal effect of blood glucose,insulin related indicators,and different types of obesity on cardiovascular diseases and cancers through a two-sample Mendelian Randomization method,and evaluated the causal relationship between obesity and both blood glucose and insulin related indicators as well.1.Elevated HbAlc is causally associated with the increased risk of a variety of cardiovascular diseases;except the causal association between elevated 2hGlu and increased risk of ER-breast cancer,neither FBG nor HbAlc has a significant causal association with the risk of cancers.2.Elevated insulin levels are found to be causally associated with increased risk of various cardiovascular diseases and cancers.3.Both BMI and WHRadjBMI have a significant causal association with the increased risk of various cardiovascular diseases.Increased WHRadjBMI had a particularly significant causal effect on the increased risk of ischemic stroke.4.BMI is also causally associated with increased risk of various cancers,while the elevated BMI and WHRadjBMI are both causally associated with reduced risk of breast cancer.5.Both BMI and WHRadjBMI have a significant causal association with the elevated blood glucose and insulin.Therefore,it is suggested that there is no direct causal relationship between FBG and 2hGlu and most cardiovascular diseases and cancers,their correlations are mostly mediated by metabolic changes and other risk factors caused by high blood glucose.In addition,hyperinsulinemia is an independent risk factor and plays an important role in etiology.The association between diabetes and the risk of some cardiovascular diseases and cancers found in observational epidemiological studies may be driven by insulin resistance or hyperinsulinemia.However,the risk effect of insulin has been ignored in clinical treatment for a long time.The results of this study suggest that alleviating IR and avoiding hyperinsulinemia should be considered as important measures for the treatment and prevention of diabetic vascular complications.We should pay attention to the influence of both endogenous and exogenous insulin on health,which contributes for the prevention of cardiovascular diseases and cancers.Obesity is not only causally associated with increased risk of various cardiovascular diseases and cancers,but also significantly associated with increased blood glucose and insulin.From a new perspective,it has been proved that obesity is the source risk factor of hyperinsulinemia,diabetes,cardiovascular diseases and cancers.On the one hand,it plays a direct role through adipokines and mediating immune cell responses.On the other hand,it has an indirect impact on cardiovascular diseases and cancers by causing metabolic disorders,insulin resistance,etc.For the treatment of obese patients combined with cardiovascular diseases and cancers,weight loss is the first priority.This study has linked blood glucose and insulin with the increased risk of various cardiovascular diseases and cancers.It evaluates the wide range of effects of obesity on hyperinsulinemia,diabetes,cardiovascular diseases and cancers from a causal perspective,which provides strong evidence for the important role of obesity,blood glucose and insulin related traits in the etiology of cardiovascular diseases and cancers. |
PDF Full Text Request