The Association Between Chronic Low-level Cadmium Exposure And Vascular Endothelial Dysfunction

Objective:This study aimed to determine the status of cadmium exposure and the distribution of vascular endothelial dysfunction,and to evaluate the association between chronic low-level cadmium exposure and vascular endothelial dysfunction.Methods:Cross-sectional analysis was conducted in 702 adult residents from urban and rural of Guangxi.Information on sociodemographic factors,cigarette smoking,drinking,physician diagnosis of disease,as well as use of medication was collected by a self-reported questionnaire.Height,weight,blood pressure and Flow-mediated dilatation(FMD)were measured.And during the physical examinations,the spot urine samples and fasting venous blood were collected for the urine cadmium and urine creatinine measurings and the serological tests.The urine cadmium concentrations were determined by a graphite furnace atomic absorption spectrophotometer.The urine creatinine concentrations were determined with the use of an ultraviolet spectrophotometer.All the serologicaltests were performed using Hitachi 7600-120 or 7600-020 analyzer.All statistical analyses listed below were stratified by gender and conducted in STATA or R software.First,we described the status of cadmium exposure and the distribution of vascular endothelial dysfunction by different subgroup.Second,we explored the influencing factors of urine cadmium concentrations and FMD levels by using simple linear regression and stepwise multiple linear regression.Third,the linear does-response relationship of urine cadmium concerntrations and vascular endothelial dysfunction was estimated by using the multivariable logistic regression.Fourth,the nonlinear dose-response relationship of urine cadmium concentratons and vascular endothelial dysfunction was estimated by using the restricted quadratic spline.Fifth,the multivariable logistic regression models were used to assessed the odds ratio for vascular endothelial dysfunction by comparing the 75 th with 25 th percentile of urine cadmium distribution as well as to estimate the interactions between urine cadmium and subgroup variables for vascular endothelial dysfunction.Results:1.The final sample in this study was 702,which including 241 men and461 women.Mean age of overall participants was 54.4±0.4 years.Median urine cadmium concentrations for overall,men,and women were 1.25 ?g/g creatinine[Interquartile range(IQR): 3.50 ?g/g creatinine],1.04 ?g/g creatinine(IQR: 3.07?g/g creatinine),and 1.38 ?g/g creatinine(IQR: 3.96 ?g/g creatinine),respectively.Median FMD levels for overall,men,and women were 8.5%(IQR:5.1%),7.4%(IQR: 4.2%),and 9.3%(IQR: 5.2%),respectively.2.For men,the stepwise multiple linear regression showed that the influencing factors of urine cadmium concentrations were race/ethnicity,age,cigarette smoking,and high-density lipoprotein cholesterol concentration.Andthe influencing factors of FMD levels were baseline brachial artery diameter,age,body mass index,and total cholesterol concentration.For women,the stepwise multiple linear regression showed that the influencing factors of urine cadmium were race/ethnicity,and age.And the influencing factors of FMD levels were baseline brachial artery diameter,postmenopausal status,and race/ethnicity.3.No linear trend was found in the analyses of linear dose-response relationship between urine cadmium concentration and vascular endothelial dysfunction.And the fully adjusted odds ratio for vascular endothelial dysfunction by comparing the highest tertile to the lowest tretile was 2.65(95%confidence interval: 1.08,6.52)for men,and 0.53(95% confidence interval:0.28,1.00)for women.4.In the analyses of nonlinear dose-response relationship between urine cadmium concentration and vascular endothelial dysfunction,the spline regression model for men showed that at the low levels of urine cadmium the dose-response relation was gradually increased,and followed by a latform trend.The spline regression model for women showed that at the low levels of urine cadmium the dose-response relation was gradually decreased,and followed by a latform trend.5.For the interaction analyses,none statistically significant interacton was observed between urine cadmium and the following covariates: age,race/ethnicity,cigarette smoking(for men),postmenopausal status(for women),hypertension,dyslipidemia,fasting blood glucose,body mass index,C-reactive protein for vascular endothelial dysfunction in both men and women.Conclusions:1.The study population were chronically exposed to low-level cadmium.And there are 45.2% men and 26.7% women with vascular endothelial dysfunction.2.In both men and women,age and race/ethnicity were the influencing factors for urine cadmium concentration.Furthermore,cigarette smoking was an important source for cadmium exposure in men.Baseline brachial artery diameter was the influencing factor for FMD in both men and women.In addition,postmenopausal status was a major factor for FMD in women.3.No linear dose-response relationship between chronic low-level cadmium exposure and vascular endothelial dysfunction.However,these analyses implied that the positive association between low-level cadmium exposure and vascular endothelial dysfunction in men but not in women,at least in a specific range of cadmium concentration.Nonlinear dose-response relationship further demonstrated that chronic low-level cadmium increased the risk of vascular endothelial dysfunction in men.4.The interaction analyses suggested that there was no interaction in both men and women.