| Asymptomatic organ damage in cardiovascular system is an important intermediate stage in the development of cardiovascular disease.Arterial stiffness（AS）and lower-extremity peripheral artery disease（LEAD）,are common asymptomatic blood vessel damages,which are diagnosed according to brachial-ankle pulse wave velocity（ba PWV）and ankle-brachial index（ABI）,respectively.Systolic blood pressure（SBP）,low-density lipoprotein cholesterol（LDL）,body mass index（BMI）,and fasting blood glucose（FBG）were the top 4 metabolic risk factors that contributed to the global burden of cardiovascular diseases.However,most previous studies assessed level of metabolic factors with only one-single measurement,which could not consider the characteristics over time.Accumulating studies have revealed that not only the level of metabolic factors,but other long-term change patterns have a significant impact on cardiovascular health,such as variability and trajectory.Objective:The aim of this study was to assess the long-term change patterns of metabolic factors as level,variability,and trajectory,and explored the associations of long-term change patterns of SBP,FBG,LDL and BMI with AS and LEAD,as well as ba PWV and ABI in cross-sectional and prospective cohort study.Methods:This study included 20 752 participants,which wwere derived from Kailuan cohort study in Tangshan,Hebei.Kailuan cohort study recruited 101 510 employees in Kailuan company in 2006,and followed them up every two years.The first health assessment of blood vessels was performed during the investigation in 2010,and a series of follow-up assessments were performed later.We considered baseline as the time of investigation in2010.The long-term change patterns of metabolic factors were assessed based on three measurements in 2006,2008,and 2010.The level of metabolic factor was assessed using time-weighted mean.The variability was assessed with four indexes of variability（standard deviation,coefficient of variation,average real variability,and variability independent of the mean）.The trajectories of metabolic factors were assess using group-based trajectory model.The outcomes in the present study were ba PWV and ABI measurements and prevalence of AS and LEAD at baseline,as well as progression and incident of AS and LEAD during follow-up.AS was diagnosed as ba PWV≥1800 cm/s,LEAD was diagnosed as ABI≤0.9.The progression of AS or LEAD was assessed using the absolute difference between first and second measurements of the ba PWV or ABI divided by the follow-up time in years.We used multiple linear regression model,binary logistic regression model,and Cox proportional hazards model to evaluate the associations of long-term change patterns of metabolic factors with AS（ba PWV）and LEAD（ABI）.Restricted cubic spline model was also applied to explore the potential nonlinear relationship between level of metabolic factors and AS and LEAD.Results:Part Ⅰ:Relationships of level of metabolic factors with AS and LEADCovariates in fully adjusted models were age,sex,educational level,smoking status,dinking status,estimated glomerular filtration rate,other history of other chronic diseases and medication use.Notably,analyses about AS and LEAD progression were additionally adjusted for ba PWV and ABI at baseline,respectively.SBP level was positively associated with ba PWV（β:8.37;95%CI:8.06,8.67）,prevalence risk of AS（OR:1.05;95%CI:1.05,1.06）,annual change of ba PWV（β:1.19;95%CI:0.96,1.41）,and incident risk of AS（HR:1.03;95%CI:1.03,1.04）.FBG level was also positively associated with ba PWV（β:33.37;95%CI:29.34,37.39）,prevalence risk of AS（OR:1.23;95%CI:1.19,1.28）,AS progression（β:6.87;95%CI:4.52,9.23）,and incident risk of AS（HR:1.18;95%CI:1.12,1.24）.Higher LDL level was only related with higher incident risk of AS（HR:1.15;95%CI:1.06,1.26）.BMI level were negatively associated with ba PWV（β:-3.9;95%CI:-5.32,-2.61）and prevalence risk of AS（OR:0.97;95%CI:0.96,0.98）.The relationships of level of metabolic factors with LEAD and ABI were mostly nonlinear.After fully adjustment,restricted cubic spline model showed“U-shaped”relationships of SBP level with ABI,prevalence risk of LEAD,and progression of LEAD(all Poverall<0.05,and Pnonlinear<0.05).An inverted“U-shaped”association was also observed between FBG level and ABI(Poverall=0.027,Pnonlinear=0.038).FBG level was negatively associated with ABI only when FBG level higher than approximately 8.00mmol/L.FBG level had positive linear relationship with LEAD incident risk(Poverall<.001,Pnonlinear=0.540).LDL level was also showed nonlinear relationships with ABI and prevalence risk of LEAD.Compared with median level of LDL(all Poverall<.001,Pnonlinear<0.05).BMI level also has“U-shaped”relationship with ABI and LEAD progression(all Poverall<.005,Pnonlinear<0.05).Part Ⅱ:Relationships of variability of metabolic factors with AS and LEADThe variability（standard deviation）of metabolic factors were divided into four groups based on quartiles（Q）,and variability increased in the order of Q1,Q2,Q3 and Q4.Covariates included in Part II were those in Part I plus the level of metabolic factors（time-weighted mean）.Compared with Q1 of SBP variability,Q4 had higher ba PWV（β:29.04;95%CI:18.11,39.96）,higher prevalence risk of AS（OR:1.23;95%CI:1.10,1.38）,and faster progression of AS（β:11.06;95%CI:3.66,18.47）.Among subgroups of FBG variability,Q4 had higher ba PWV than Q1（β:23.27;95%CI:11.29,35.25）.Among subgroups of LDL variability,compared to Q1,Q3 and Q4 had 22.77 cm/s（95%CI:11.68,33.85）and 50.27 cm/s（95%CI:38.84,61.70）higher ba PWV,and had 1.18 times（95%CI:1.06,1.33）and 1.43 times（95%CI:1.28,1.60）prevalence risk of AS,respectively.Among subgroups of BMI variability,Q2（β:15.93;95%CI:4.92,26.94）,Q3（β:17.45;95%CI:6.41,28.49）,and Q4（β:20.46;95%CI:9.31,31.61）had higher ba PWV than Q1,but only Q3（OR:1.12;95%CI:1.01,1.25）showed a significant higher prevalence risk of AS compared with Q1,and the linear trend was not significant(Ptrend=0.398).Among subgroups of SBP variability,only Q2 showed a significant higher incident risk of LEAD（HR:1.30;95%CI:1.03,1.63）than Q1,and no significant linear trend was observed（Ptrend=0.736）.FBG variability has significant association with ABI（Ptrend=0.015）.The Q4 of FBG variability had 57.81×10-4（95%CI:-98.40,-17.23）lower ABI than Q1.Compared with Q1,subgroup with highest variability of BMI（Q4）had70.66×10-4（95%CI:-108.31,-33.02）lower ABI and 1.49 times（95%CI:1.21,1.83）prevalence risk of AS,Q3 of BMI variability had 37.55×10-4（95%CI:-74.82,-0.28）lower ABI.BMI variability also showed significant linear trend with ABI and prevalence risk of AS(all Ptrend<.001).Part Ⅲ:Relationships of trajectories of metabolic factors with AS and LEADThis study assessed the trajectories of metabolic factors using the Group-based trajectory model.The best fitted model was identified according to the Bayesian Information Criteria,average posterior probability,and tiwce of the difference in Bayesian Information Criteria of complex model and simple model.The shape of each trajectory was identified by the highest within the range of 0～3 order with significant statistical significance（P<0.05）.The numbers of subgroup in the best fitted models for SBP,FBG,LDL,and BMI were five,four,five and five.After adjusting for covariates included in Part I,as well as the level（time-weighted mean）and variability（standard deviation）of metabolic factors,participants with other SBP trajectories all had higher ba PWV compared with normal SBP-stable trajectory（P<0.05）.Except for II stage abnormal SBP-stable trajectory,all other trajectories were also related with higher prevalence risk of AS,faster AS progression,and higher incident risk of AS（all P<0.05）.In FBG trajectory groups,participants with abnormal FBG-stable trajectory had significantly higher ba PWV than those with normal FBG-stable trajectory（β:31.87;95%CI:9.69,54.05）.In LDL trajectory groups,participants with normal LDL-increasing trajectory had 1.22 times（OR:1.22;95%CI:1.01,1.47）higher prevalence risk of AS than those with high normal LDL-stable trajectory（P=0.040）.Compared with normal SBP-stable trajectory,elevated SBP-stable was related with higher ABI(β:53.69×10-4;95%CI:5.69,101.69),while elevate SBP-increasing(β:-115.97×10-4;95%CI:-209.50,-22.44)and II stage abnormal SBP-stable(β:-211.31×10-4;95%CI:-347.41,-75.21)trajectories were related with lower ABI.In FBG trajectory groups,subjects with abnormal FBG-stable trajectory had significant lower ABI in comparison with those with normal FBG-stable trajectory(β:-83.77×10-4;95%CI:-158.88,-8.65).Compared with low normal BMI-stable trajectory,elevated BMI-stable(β:85.56×10-4;95%CI:5.66,165.46)and I stage abnormal BMI-stable(β:117.39×10-4;95%CI:1.34,233.43)trajectories were significantly related with faster progression of LEAD.Conclusion:Results from this study showed that SBP,FBG,and LDL could affect AS by all three long-term change patterns,namely level（time-weighted mean）,variability and trajectory,while two change patterns of level and variability of BMI could affect AS.In addition,all three change patterns of FBG and BMI were associated with LEAD,level and variability of SBP were associated with LEAD,while the significantly relayionship was only observred between LDL level and LEAD.This study indicated that multiple long-term change patterns of cardiovascular metabolic factors were significantly related to AS and LEAD,as well as ba PWV and ABI,suggesting monitoring of long-term change patterns of metabolic factors could help identify individuals with higher risk of CVD,and benefit early prevention of CVD. |
PDF Full Text Request