Associations Of Plasma Coagulation Factors Ⅹ And Ⅺ Levels With The Risk Of Incident Stroke,Acute Coronary Syndrome,and Their Influencing Factors

Cardiovascular disease(CVD)is the major killer of human beings,especially stroke and acute coronary syndrome(ACS).Although the pathogenesis of stroke and ACS were complex,the process of thrombosis,rupture of unstable plaque,and disorder of the fibrinolytic system,might be important mechanisms.Some studies proposed that coagulation factor X(FX)and coagulation factor XI(FXI),which involved in the coagulation cascade,might be novel therapeutic targets of CVD.However,most previous studies were case-control design conducted in European and American populations,which were limited by small sample sizes and generated inconsistent conclusions.Moreover,the existing evidences mostly focused on ischemic stroke and acute myocardial infarction,two subtypes of stroke and ACS,but rarely evaluated the associations of plasma FX and FXI levels with the risk of hemorrhagic stroke and unstable angina.Therefore,it is necessary to explore the associations of plasma FX and FXI levels with the risk of incident stroke,ACS,and their subtypes among Chinese,and to evaluate the predictive efficacy of coagulation factors on these diseases.The study may have important public health implications for the prevention of these diseases.In addition,although diabetes mellitus,hypertension,hyperlipidemia,drinking status,smoking status,shift work,and plasma metals/metalloids exposure were reported to be associated with higher risk of incident CVD,only a few studies found hypercoagulable states could be acquired from hyperlipidemia and hyperglycemia.It remains unknown whether the above-mentioned other factors are linked with FX and FXI levels.Thus,the associations of cardiovascular risk factors with FX and FXI levels need to be further explored,so as to provide scientific basis for effective control of plasma FX and FXI levels.Therefore,we conducted two nested case-control studies based on the DongfengTongji cohort to explore the prospective associations of FX and XI with incident stroke,ACS,and their subtypes,and to evaluate the predictive efficacy of coagulation factors.In addition,a cross-sectional study was performed to explore associations of cardiovascular risk factors with FX and FXI levels.This study mainly includes the following two parts:Part I Associations of coagulation factors X and XI with incident stroke and acute coronary syndrome: two nested case-control studiesObjective: Based on two nested case-control studies within the Dongfeng-Tongji cohort,we prospectively evaluated the associations of plasma FX and FXI levels with the risk of incident stroke,ACS,and their subtypes(ischemic stroke,hemorrhagic stroke,unstable angina,and acute myocardial infarction).Methods: The Dongfeng-Tongji cohort in 2013 was taken as the research baseline(n=38295),and a total of 23477 participants were eligible after excluding those with cancer,CVD,abnormal electrocardiogram,medical history of thrombolysis and anticoagulation,and insufficient plasma samples at baseline.Finally,374 and 549 participants were diagnosed with incident stroke(295 ischemic stroke and 79 hemorrhagic stroke)and incident ACS(432 unstable angina and 117 acute myocardial infarction)until December 31,2016.The same number of controls were randomly selected from eligible participants by incidence density sampling.Each case was matched a control according to age(within 1 year),gender,and sampling date(within 1 month).Therefore,a total of 374 incident stroke,549 incident ACS,and their matched controls were included in the two nested case-control studies.The antigen levels of FX and FXI in plasma samples were detected by enzyme-linked immuno-sorbent assay.Participants were divided into low groups(<25%),medium groups(25%~75%),and high groups(>75%)according to distribution characteristics of plasma FX and FXI levels.The Kaplan-Meier curve was used to estimate the cumulative percentages of incident stroke,ACS,and their subtypes among FX and FXI groups.We estimated odds ratios(ORs)and 95% confidence intervals(CIs),and evaluated the associations of plasma FX and FXI levels with the risk of incident stroke and ACS by conditional logistic regression.In addition,the predictive efficacy of coagulation factors was evaluated by receiver-operating characteristic curve.Results: During a median follow-up of 2.7 years,the cumulative percentage of incident stroke in the high group of FXI levels(>5.63 mg/L)was higher than that in the medium or low group(P<0.05),but without significant difference for incident ACS.Meanwhile,the cumulative percentages of incident stroke and ACS were no significantly difference among FX groups.After multivarible adjustment,the ORs(95% CIs)of incident stroke in high groups of plasma FX and FXI levels were 1.01(0.63,1.62)and 1.72(1.14,2.60),respectively,and the corresponding ORs(95% CIs)of incident ACS were 1.11(0.79,1.56)and 0.96(0.68,1.36),respectively.In the subgroup analysis of stroke and ACS,only higher FXI levels were associated with higher risk of incident ischemic stroke(OR: 1.66,95% CI: 1.05,2.65).All results remained robust after additional adjustment for platelets,white blood cells,and coagulation factors.The predictive efficacy for incident stroke and ischemic stroke was not significantly improved after adding FXI levels into predictive models(P>0.05).Conclusions: Higher plasma FXI levels were significantly associated with higher risk of incident stroke,especially ischemic stroke,but not with that of incident hemorrhagic stroke and ACS.Plasma FX levels were not significantly associated with the risk of incident stroke,ACS,and their subtypes.These findings need to be further verified in larger sample sizes and multi-center prospective studies.Part II Associations of cardiovascular risk factors with plasma coagulation factors X and XI levelsObjective: To investigate the associations of cardiovascular risk factors with plasma FX and FXI levels.Methods: Based on the nested case-control studies in the part I,a total of 1836 participants were included in this part.The associations of cardiovascular risk factors with plasma FX and FXI levels were evaluated by generalized linear regression models.Participants without enough plasma samples for metal detection were further excluded,thus,787 participants were included to investigate the associations of plasma metals/metalloids with FX and FXI levels.The concentrations of 23 plasma metals/metalloids were detected by inductively coupled plasma-mass spectrometry,and six plasma metals/metalloids(iron,cadmium,chromium,antimony,tin,and tungsten)were excluded after quality control.The associations of other 17 plasma metals/metalloids with FX and FXI levels were estimated by generalized linear regression models and LASSO penalty regression models.In addition,principal component analysis was performed to extract principal components of metals/metalloids,and evaluated the associations of polymetallic exposure of plasma metals/metalloids with plasma FX and FXI levels.Results: Age was negatively associated with FX and FXI levels in the generalized linear regression models after multivariable adjustment.Current drinking,hypertension,history of aspirin usage,hyperlipidemia,platelets,and white blood cells were all positively associated with plasma FX levels,and white blood cells and total cholesterol had higher effects(all P<0.05).The corresponding plasma FX levels increased by 3.107 mg/L and 0.622 mg/L with per interquartile range(IQR)increase in white blood cells and total cholesterol,respectively.Meanwhile,female,family history of CVD,hyperlipidemia,diabetes mellitus,platelets and white blood cells were all positively associated with plasma FXI levels,and white blood cells and platelets had higher effects(all P<0.05).The corresponding FXI levels increased by 1.659 mg/L and 0.442 mg/L with per IQR increase in platelets and white blood cells,respectively.Plasma copper and thallium were positively associated with FXI levels after multivariable adjustment in single metal model.The corresponding FXI levels increased by 0.094 mg/L and 0.117 mg/L with per IQR increase in plasma copper and thallium concentrations,respectively(P<0.05).These associations were verified in the LASSO penalty regression model.Only the association between plasma thallium and FXI level was still significant in the multiple-metals model(P for trend<0.05),and the corresponding β(95% CI)of FXI levels was 0.101(0.016,0.186).In addition,we observed significant multiplicative interaction between plasma copper and thallium on plasma FXI levels,and higher plasma thallium concentrations increased the positive association between plasma copper and FXI levels(P for interaction<0.05).Moreover,the first principal component,which mainly represented copper,rubidium,selenium,titanium,and zinc,was positively associated with FX and FXI levels in the principal component analysis.Compared with the lowest quartile,the β(95% CI)of the highest quartile was 0.836(0.349,1.323)and 0.470(0.225,0.714)for plasma FX and FXI levels,respectively(all P for trend<0.05).Conclusions: Elderly age was negatively associated with FX and FXI levels,whereas current drinking,hyperlipidemia,platelets,and white blood cells were positively associated with plasma FX and FXI levels.Plasma copper and thallium were positively associated with FXI levels,and higher plasma thallium increased the positive association between copper and FXI levels.The polymetallic exposure levels of the first principal component(copper,rubidium,selenium,titanium,and zinc)were positively associated with plasma FX and FXI levels.