Associations Of Serum 25-Hydroxyvitamin D Concentration With Risk Of Cardiovascular Events And Mortality Among Individuals With Type 2 Diabetes

With the rapid development of economy,the improvement of living standards and change of lifestyle,the number of people with diabetes has increased dramatically in recent years.Cardiovascular disease(CVD)is the most common complication of diabetes and one of the main causes of death among patients with type 2 diabetes(T2D).Therefore,it is of importance to prevent or delay CVD occurrence among individuals with T2 D.Dietary factors play an important role in the prevention of CVD among T2 D patients.Vitamin D is an essential micronutrient,mainly produced by the skin through ultraviolet radiation.Vitamin D combines with vitamin D receptor(VDR)in specific tissues and organs to play its biological roles.Epidemiological studies among general population suggested that serum vitamin D concentrations inversely associated with risk of CVD.However,Recent randomized controlled trials(RCTs)found null effect of vitamin D supplementation,possibly because that subjects enrolled in these RCTs had relatively adequate vitamin D levels.In recent years,studies have proposed the "threshold effect" theory,which indicating that vitamin D may have health benefits only in people with low vitamin D levels.The prevalence of vitamin D deficiency or insufficiency among T2 D patients were higher than non-diabetes population.However,studies exploring the associations of vitamin D with CVD and mortality among indivuduals with T2 D were limited,and the conclusions were inconsistent.To fill these knowledge gaps,we included adults with T2 D from the UK Biobank(UKB)study,to assess the associations of serum 25-hydroxyvitamin D(25[OH]D)levels and the VDR gene polymorphisms with CVD,allcause and CVD mortality,to explore the potential mediating effect of serum biomarkers in the above associations,and to investigate the causal relationship between 25(OH)D concentration and outcomes using mendelian randomization.The present study includes the following three parts:Part 1 Association of serum 25(OH)D concentrations and genetic variants in VDR with cardiovascular events and mortality among patients with type 2 diabetesObjectives: To investigate the associations of serum 25(OH)D concentrations and gene polymorphism in VDR with risks of CVD and mortality among patients with T2 D.Methods: A total of 15226 T2 D patients with serum 25(OH)D measurements,free of CVD and cancer at baseline were enrolled from the UKB.Serum 25(OH)D concentrations were divided into four groups: <25.0,25.0-49.9,50.0-74.9 and ≥75.0 nmol/L,according to the Endocrine Society Clinical Practice Guidelines.Cox proportional hazards regression model was used to calculate the hazard ratio(HR)and 95% confidence interval(CI)of CVD events,all-cause and CVD mortality in the other three groups with <25.0 nmol/L as the reference group.Restricted cubic spline(RCS)models were further used to evaluate the shape of the above associations.The associations of four VDR gene polymorphisms(Apa I [rs7975232],Bsm I [rs1544410],Taq I [rs731236],and Fok I [rs2228570])with outcomes was also analyzed using Cox regression.The interaction between serum 25(OH)D and VDR gene polymorphisms was assessed using likelihood ratio tests.Results: the mean(SD)of serum 25(OH)D concentration was 43.4±20.4 nmol/L in UKB.With a median follow-up of 11.1 years,the UKB study documented 3464 new CVD events and 1598 deaths.Compared with those with serum 25(OH)D concentration <25.0 nmol/L,risks of CVD(HR=0.81,95% CI: 0.70-0.95,Ptrend=0.001)and all-cause mortality(HR=0.61,95% CI: 0.48-0.76,Ptrend<0.001)were significantly decreased.Restricted cubic spline(RCS)results suggested that the above correlations are all nonlinear,risks of CVD and mortality were the lowest when serum 25(OH)D was around 50 nmol/L.We found no significant associations of VDR gene polymorphism with the risks of CVD and mortality(all P>0.01).In addition,we found significant associations of additive interaction of Bsm I(rs1544410),Taq I(rs731236)and serum 25(OH)D with risk of all-cause mortality in T2 D patients.Conclusions: Among individuals with T2 D,serum 25(OH)D concentrations were negatively and non-linearly associated with the risk of CVD and mortality.In addition,there are additive interactions between VDR gene polymorphisms and 25(OH)D on the risk of all-cause mortality.the association of serum 25(OH)D concentration with the risk of cardiovascular events and mortality among patients with type 2 diabetes Part 2 Potential mediating role of cardiovascular metabolism related biomarkers inObjectives: To explore the mediating role of cardiovascular metabolism related biomarkers in the association of 25(OH)D concentrations with the risks of cardiovascular events and mortality among individuals with T2 D.Methods: Participants included in this part are the same as that in the first part.First,linear regression models were used to screen biomarkers those were associated with serum 25(OH)D concentrations.Then,cox regression models were used to screen biomarkers that were also significantly associated with risks of CVD and mortality among individuals with T2 D.Finally,mediation analyses were performed to assess the mediating effect of these biomarkers in the association of 25(OH)D concentrations with the risks of CVD and mortality among patients with T2 D.Results: The baseline characteristics of the 15226 T2 D patients included in this study were shown in Part one.After adjusting for multiple confounders,the combined mediating effect of low density lipoprotein(LDL),Apolipoprotein(APOB),albumin(ALB),alkaline phosphatase(ALP),and cystatin C(CYSC)on the association between 25(OH)D concentration and CVD was 36.6%(95% CI: 21.6%-54.9%,P<0.001);the combined mediating effect of ALB,ALP,gamma-glutamyl transferase(GGT)and CYSC accounted for 32.9%(95% CI: 21.1%-47.2%,P<0.001)of the total association between 25(OH)D concentration and the risk of all-cause mortality.In the subgroup of 25(OH)D<50 nmol/L,the mediating effects of these indicators still existed significantly,except for ALB.However,in the subgroup of 25(OH)D>50 nmol/L,no mediating effects was observed.Conclusions: Among T2 D patients with 25(OH)D<50 nmol/L,the association between serum 25(OH)D and the risk of CVD and mortality might be partially mediated by LDL,APOB,CYSC,ALP,GGT and Hb A1 c,indicating that 25(OH)D might decrease the risks of CVD and mortality by improving the lipids profile,liver and kidney function,and glycemic metabolic status.Part 3 Mendelian randomization analyses of the associations between serum 25(OH)D concentration and cardiovascular events and mortality among patients with type 2 diabetesObjectives: To explore the causalities between serum 25(OH)D levels and risks of CVD and mortality among individuals with T2 D.Methods: A total of 11260 patients with T2 D,free of CVD and cancer,and had genegtic data,were enrolled.Single nuclear polymorphisms(SNPs)significantly associated with serum 25(OH)D concentrations were used to calculate a weighted genetic risk score(GRS).To assess the causalities of genetically-determined serum 25(OH)D concentrations with CVD and its subtypes(IHD,stroke,and heart failure),all-cause and CVD mortality in T2 D patients,ratio method was used in one-sample linear mendelian randomization study.Besides,the fractional polynomial method of nonlinear mendelian randomization study was used to assess whether the above associations were nonlinear.Results: Sixty-two SNPs significantly associated with serum 25(OH)D concentration were used to calculate the weighted GRS.Linear mendelian randomization suggested that genetically predicted 25(OH)D concentration might causally associated with CVD mortality(HR=0.50,95% CI: 0.25-0.97,P=0.04).Non-linear mendelian randomization studies suggest a trend of nonlinear causalities between genetically predicted 25(OH)D concentrations and IHD(Pnon-linear =0.06),a significantly nonlinear causality between genetically predicted 25(OH)D concentrations and risks of all-cause and CVD mortality(Pnon-linear<0.05).When 25(OH)D < 50 nmol/L,as its concentration increased,the risk of the above outcomes decreased,and the slope of the decline slowed down and gradually reached a plateau at about 50 nmol/L.Conclusions: Our study suggested a trend of non-linear causal association of genetically predicted 25(OH)D concentrations with risk of IHD,and a non-linear causality with risks of all-cause and CVD mortality among patients with T2 D.