A Cohort Study On The Association Between Different Blood Pressure Indexes And All-Cause Mortality

Objectives1.To evaluate the relation between blood pressure?BP?indexes,including systolic BP?SBP?,diastolic BP?DBP?,pulse pressure?PP?,mean arterial pressure?MAP?,mean/mid BP?MBP?,and mean proportional arterial pressure?MPAP?,and the risk of all-cause mortality.2.To assess the predictive utility and differences of six BP indexes on the risk of all-cause mortality.MethodsCluster random sampling was used to recruit 20 194 study participants aged?18years in Xin'an County,Henan Province,China during 2007-2008 based on a prospective cohort study.We selected Cijian and Tiemen towns from Xin'an County,and conducted the questionnaire interview,physical examination,fasting plasma glucose and lipid profile measurements for all the study participants.Follow-up examination of 17 265 study participants was performed during 2013-2014?response rate 85.50%?.Study participants were excluded from the current study if they had incomplete data for SBP or DBP?n=1?or used antihypertensive medication?n=3 504?at baseline examination.Ultimately,13 760 eligible study participants were included in the present analyses.The relations between SBP,DBP,PP,MAP,MBP,and MPAP and the risk of all-cause mortality were examined by Cox proportional-hazards regression models,calculating hazard ratio?HR?or adjusted HR?aHR?and 95%confidence interval?CI?,according to these six BP indexes for per standard deviation?SD?increase.Restricted cubic splines were used to model the dose–response relation.We used receiver operating characteristic?ROC?curve to compare the predictive utility of six BP indexes for the risk of all-cause mortality.Also,we did subgroup analyses across sex and age?<60 and?60 years?.In addition,sensitivity analyses were used to assess the results robustness:1)excluding baseline smokers;2)further excluding people died within the first two years of follow-up;3)further excluding people with cardiovascular disease?myocardial infarction,heart failure,and stroke?,diabetes mellitus,or cancer at baseline.Results1.A total of 13 760 people?5 661 men and 8 099 women?were included in this study,with a cumulative follow-up of 81 855.96 person-years?mean 5.95 years?,710deaths?death density:8.67/1000 person-years?during follow-up,including 428 men?death density:12.72/1000 person-years?and 282 women?death density:5.85/1000person-years?.2.After adjusting for potential confounding factors,including sex,age,marital status,mean individual income?monthly?,education level,smoking,drinking,physical activity,family history of hypertension,body mass index,fasting plasma glucose,total cholesterol,triglycerides,high-density lipoprotein-cholesterol,and low-density lipoprotein-cholesterol,the BP–mortality relations were analyzed by Cox proportional-hazards regression models with per SD increase as a unit for each BP index,and the results showed that all six BP indexes significantly increased the risk of all-cause mortality,the aHR?95%CI?were 1.13?1.06-1.21?,1.10?1.03-1.19?,1.11?1.04-1.19?,1.12?1.05-1.21?,1.13?1.05-1.21?,and 1.13?1.05-1.21?for SBP,DBP,PP,MAP,MBP,and MPAP,respectively.3.After adjusting for possible confounders,restricted cubic splines showed the dose–response relations of SBP,DBP,PP,MAP,MBP,and MPAP with the risk of all-cause mortality were U-shaped(P_{non-linearity}on-linearity value<0.001),and the lowest risk of all-cause mortality was at about 113,70,41,85,93,and 97 mmHg for SBP,DBP,PP,MAP,MBP,and MPAP,respectively.4.In stratified analysis,Cox proportional-hazards regression models showed that after adjusting for possible confounders,per SD increase in SBP,DBP,PP,MAP,MBP,and MPAP increased the risk of all-cause mortality for men,women,people aged<60and?60 years.The results were consistent after excluding baseline smokers,people died within the first two years of follow-up,and cardiovascular disease,diabetes mellitus,or cancer patients at baseline.5.ROC curve showed that area under the curve of the risk of all-cause mortality predicted by different BP indexes decreased successively?P value<0.05?:PP?0.675[0.667-0.682]?,SBP?0.630[0.622-0.638]?,MPAP?0.618[0.610-0.626]?,MBP?0.596[0.588-0.605]?,MAP?0.577[0.569-0.586]?,DBP?0.522[0.514-0.531]?.The corresponding sensitivity and specificity were 50.28%and 76.80%,41.69%and79.35%,39.86%and 79.40%,39.15%and 77.55%,36.06%and 77.98%,and 22.96%and 83.82%,respectively.In stratified analysis and sensitivity analysis,the results were consistent.Conclusions1.In this study,there were U-shaped relations of SBP,DBP,PP,MAP,MBP,and MPAP with the risk of all-cause mortality.2.The predictive power of different BP indexes for the risk of all-cause mortality was arranged in descending order:PP,SBP,MPAP,MBP,MAP,and DBP,with poor predictive power.