A Cross-sectional Study On The Association Between Serum Magnesium And Risks Of Metabolic Syndrome In Adults

BackgroundMetabolic syndrome (MetS), as an epidemic in the word, is widely considered as the early symptom of type2diebetes and cardiovascular disease. Some studies on the relationship between dietary magnesium intake and MetS along with relative diseases have been reported. While few studies focus on the serum magnesium (Smg) and risks of MetS, especially in chinease subjects. We conducted this study to investigate following aspects:(1) the association between Smg and MetS risks in Chinese adults;(2) the association between Smg and risks of each MetS component;(3) the appropriate range of Smg for Chinese adults with MetS.Objectives and MethodsThe data analysis was based on the cross-sectional study of China Health and Nutrition Survey in2009(CHNS2009). According to the criteria for inclusion or exclusion,1357MetS patients and5918individuals without MetS were obtained. And we constructed a case-control study with the1357MetS patients and1893health controls who had no MetS components. The IDF2005criterion was used to diagnose the MetS individuals. All the subjects were divided into four groups according to quartiles of Smg in controls, including reference (-2.14mg/dl), low (2.14～2.26mg/dl), median (2.26～2.38mg/dl) and high level (2.38mg/dl～) groups. Logistic regression was performed to test the association between Smg and the risks of MetS in adults. Some potential confounders such as age and gender were adjusted in model1, and further adjustments of BMI, urbanization index, education status, smoking, drinking, percentage of fat calories in food were included in model2. We explored the dose-response relationship between Smg and MetS along with the MetS components with B-spline regression model and generalized additive model (GAM). The change trends and95%medical reference ranges of Smg among different ages and genders were investigated.ResultsThe MetS prevalence reached to18.65%,14.47%and22.38%in total individuals, males and females, respectively. The median level of Smg was higher in MetS than that in controls (2.31vs.2.26mg/dl, p<0.001). And with the increase of Smg quartiles, the proportion of subjects who had3～5components increased, while the percentage of health controls without any components decreased from65.95%to51.86%. We also observed that Smg and Triglyceride was positively correlated in simple correlation analysis (r=0.26, p<0.001) and partial correlation analysis (r=0.22, p<0.001). And insignificant or weak correlations were found between Smg and other MetS components.Compared with reference concentration of Smg in model1, the low, median and high level groups showed increased ORs of1.21(0.96-1.53),1.53(1.22-1.93) and1.83(1.47-2.26), gradually. In model2the ORs were1.10(0.76-1.59),1.26(0.88-1.82) and1.47(1.04-2.06) corresponding to the low, median and high dose groups, respectively. And the trend test showed statistical significance (p<0.001). Besides, the best fitted B-spline regression model with freedom degree of1and2knots indicated that the MetS risks might rise rapidly when the Smg was more than2.4mg/dl.Additionally, the risks of abnormal Triglyceride increased from low to high quartiles of Smg. Compared with reference group, the ORs were1.09(0.85-1.41),1.45(1.14-1.85) and1.96(1.56-2.46) of low, median and high level groups in model1, and0.96(0.70-1.32),1.24(0.91-1.68) and1.73(1.30-2.30) in model2, respectively. A significant trend was observed (p<0.001). While we did not find obvious associations between Smg and other MetS components including fasting glucose, blood pressure and high density lipoprotein. In the GAM analysis, it indicated that the risks of abnormal triglyceride increased dramatically when Smg beyond the point of2.5mg/dl.In the sensitivity analysis, significant association between Smg and risks of MetS was observed in the subgroup with abnormal triglyceride. While the results in the normal triglyceride group were insignificant. And when all the5918individuals without MetS were regarded as controls, similar results that increased Smg might lead to more MetS risks were observed.The95%medical reference range of Smg among health controls without any components were1.78～2.65mg/dl. And the reference values were1.68-2.60and1.75～2.62mg/dl in females and males respectively.ConclusionsIn this study, the prevalence of MetS reached to18.65%in Chinese adults, with14.47%of males and22.38%of females. There was a dose-response relationship between Smg level and the risks of MetS. When the Smg level was higher than2.4mg/dl the MetS risks increased rapidly. And we did not find significant association between Smg and the risks of MetS components except for triglyceride. The risks of abnormal triglyceride increased with the rise of Smg. And the triglyceride concentration increased dramatically when the Smg reached to2.5mg/dl. The95%medical reference level of Smg ranged from1.75to2.62mg/dl in health controls without any MetS components.