| Objective:Hypertension is the most common chronic cardiovascular disease and an important risk factor for other cardiovascular diseases.In the study of risk factors of hypertension,both hyperuricemia and microalbuminuria are associated with hypertension,and there is also a correlation between hyperuricemia and microalbuminuria.However,the sequence of occurrence and the interaction among them have not been clarified.This study intends to analyze the relationship between hyperuricemia,microalbuminuria and hypertension through a longitudinal study,providing a new way to explore the risk factors of hypertension and a reference for the prevention of hypertension and screening of high-risk population.Study methods:Our study was a community-based longitudinal study.The subjects of the study were residents over 30 years old in 8 communities in Suzhou,China's economically developed area.The baseline and follow-up surveys of cardiovascular disease and its risk factors were conducted in 2010 and 2014 respectively.The baseline and follow-up survey were completed by the community health service staff and the graduate students majoring in epidemiology.The content of the on-site investigation included general demographic characteristics,behavior habits,measurement of physical indicators,prevalence of related diseases and family history of related diseases(including hypertension,coronary heart disease,stroke,diabetes,tumor and other diseases).The fasting venous blood of the subjects was collected on site and stored by anticoagulation,separation and freezing.Body measurements include height,weight,waist circumference,systolic blood pressure(SBP)and diastolic blood pressure(DBP).The stored blood samples were used for laboratory tests,including total cholesterol(TC),triglyceride(TG),high-density lipoprotein cholesterol(HDL-C),low-density lipoprotein cholesterol(LDL-C),fasting blood glucose(FPG),uric acid(UA),urine creatinine(UCr),urinary albumin(UAlb),urinary albumin to creatinine ratio(UACR).A total of 2706 subjects were included in the baseline data,while 1981 subjects were included in the follow-up survey due to the lack of some data and the loss of interviews.Therefore,1981 subjects were finally included in the longitudinal study.The data of baseline and follow-up survey were recorded into the computer database with EpiData software package,checked and consolidated.All statistical analysis were completed with sas9.4 software package and R software package.A two-tailed p-value less than 0.05 was considered statistically significant.Suitable statistics were used to describe the baseline characteristics of the study population,and the baseline indexes of hypertension group and normal blood pressure group,hyperuricemia group and normal uric acid level group,microalbuminuria group and normal urinary albumin group were compared by t test(or Wilcoxon rank test)and ?2 normal urinary albumin group were compared by t test(or Wilcoxon rank sum test)and ?2 test(or Fisher exact probability method).The temporal relationship of hyperuricemia,microalbuminuria and hypertension was analyzed by cross-lagged panel analysis performed by the "lavaan" R package with baseline and follow-up data;The causal mediation model was used to confirm the effect relationship among hyperuricemia,microalbuminuria and hypertension with R package"mediation".Age,gender,education,smoking,drinking,BMI,FPG,LDL-C and HDL-C were adjusted for regression analysis both in cross-lagged panel analysis and in causal mediation model.Results:1.Baseline characteristics of subjectsA total of 1981 subjects were included in the final statistical analysis,with an average age of 53 years,722 subjects were men,accounting for 39%;902 subjects with hypertension,the prevalence rate of 45.5%;212 subjects with hyperuricemia,the prevalence rate of 10.7%;403 subjects with microalbuminuria,the prevalence rate of 20.3%.Compared with the subjects with normal blood pressure,the levels of PFG,TC,TG,LDL-C,uric acid and UACR were higher in the subjects with hypertension(all P<0.01).2.Temporal relationship between hyperuricemia,microalbuminuria and hypertension After adjusting the age,gender,education level,smoking,drinking,BMI,LDL-C,HDL-C and FPG,the cross-lagged panel analysis of the temporal relationship of any two diseases was carried out.The results were as follows:(1)Temporal relationship between microalbuminuria and hyperuricemiathe path coefficient from baseline logUACR_z to follow-up UA_z(?2)was 0.010(P=0.535),which was significantly smaller than that from baseline UA_z to follow-up logUACR_z(?1)of 0.054(P=0.017),with a P less than 0.001 for the difference between,?1 and ?2.(2)Temporal relationship between hyperuricemia and hypertensionThe path coefficient from baseline UA_z to follow-up SBP_z(?1)was 0.057(P=0.005),which was significantly higher than that from baseline SBP_z to follow-up UA_z(?2)of 0.011(P=0.546),with a P less than 0.001 for the difference between ?1 and ?2 The results indicated that the increase of UA occur before the increase of SBP.The path coefficient ?1=0.037(P=0.098)was from baseline UA_z to follow-up DBP_z,and?2=0.020(P=0.235)was from baseline DBP_z to follow-up UA_z.Both p-values were greater than 0.05,so their temporal sequence of UA_z and DBP_z could not be obtained.(3)Temporal relationship between microalbuminuria and hypertensionThe path coefficient ?1 from baseline logUACR_z to follow-up SBP_z was 0.092(P<0.001),which was significantly higher than that from baseline SBP_z to follow-up logUACR_z(?2=0.031,P<0.01).The results indicated that the increase of UACR might occur before the increase of SBP.The path coefficient from baseline logUACR_z to follow-up DBP_z was 0.096(P<0.001),which was significantly higher than that from baseline DBP_z to follow-up logUACR_z(?2=0.015,P=0.468),indicating that the increase of UACR might occur before the rise of DBP.In general,hyperuricemia was more likely to precede microalbuminuria and elevated urinary albumin excretion was more likely to precede hypertension.3.The mediating effect of microalbuminuria on the relationship between hyperuricemia and hypertensionCross-lagged panel analysis showed that microalbuminuria might occur before hypertension but after hyperuricemia.Our causal mediation model further confirmed this temporal sequence.After adjusting the baseline age,gender,education level,smoking,drinking,BMI,LDL-C,HDL-C and FPG,for SBP,the mediating effect of logUACR_z accounted for 9.14%(95%confidence interval,CI:1.58%-23.00%).After controlling the influence of mediating variable logUACR_z,the direct effect of baseline UA_z on follow-up SBP_z was 1.598(95%CI:0.754-2.380),and the effect of baseline UA z on mediating variable logUACR_z was 0.053(P=0.017).The effect of logUACR_z on SBP_z was 2.903(P<0.001).The mediating effect of logUACR_z on the rise of SBP_z induced by baseline UA_zwas 0.149(95%CI:0.024-0.290).For DBP,the mediating effect of logUACR_z accounted for 7.38%(95%CI:1.05%-19.00%).After controlling the influence of mediating variable logUACR_z,the direct effect of baseline UA_z on follow-up DBP_z was 0.898(95%CI:0.411-1.400),and the effect of baseline UA_z on mediating variable logUACR_z was 0.053(P=0.017),After controlling the effect of baseline UA_z,the effect of the intermediate variable logUACR_z on DBP_z was 1.310(P<0.001).The mediating effect of logUACR_z on the rise of DBP_z induced by baseline UA_z was 0.066(95%CI:0.010-0.130).Conclusions:Microalbuminuria may precede hypertension,and hyperuricemia may precede microalbuminuria.Microalbuminuria mediates the relationship between hyperuricemia and hypertension. |
PDF Full Text Request