| Hypertension and obesity are both global public health issues.Studies of the relationship between obesity and hypertension can be traced back to the 1920s,and then Landsberg et al.began to describe hypertension associated with obesity as "obesity-related hypertension".Although the biological mechanism of hypertension caused by obesity is not yet fully understood,the relationship between the two has been basically established.However,the association between another risk factor with much more public health significance-"weight change" and incidence of hypertension and blood pressure changes should be paid more attention of researchers.As far as we know,most studies only observe a general index of weight change(defined as the difference of weight between baseline or last follow-up),in fact,weight change has a variety of meaning,and thus the indicator cannot clarify the effect of different indexes of weight change on hypertension.In addition,some studies only focus on a specific gender group or very narrow age-range population,which may lead to some limitations when promoting the research conclusions.Furthermore,some studies are based on cross-sectional designs or self-reported anthropometric indicators,which may lead to some bias.Materials and methods:Based on Shandong Multi-center Health Management Large Database and we included subjects who took health check-up between 2004 and 2016,18 years and above to compose the hypertesnsion cohort.Participants were eligible inclusion if they possessed at least two year’s records.We excluded participants whose height,weight,SBP and DBP were missing and who had hypertension or cardiovascular disease at first check-up.Eventually,87296 subjects(45482 men and 41814 women)entered the cohort.Therefore,in this study we first defined five dimensions of weight change indicators including "absolute weight change(defined as the difference of weight between baseline and hypertension onset or last follow-up)","relative weight change(absolute weight change divided by baseline weight)","average weight change per year(absolute weight change divided by follow-up years)","absolute BMI change(defined as the difference of BMI between baseline and hypertension onset or last follow-up)" and "BMI slope(the slope was obtained by simple linear regression analysis with the follow-up year as dependent variable and BMI as independent variable)",each of which had different epidemiological and statistical significance.Then,based on the Chinese obesity criteria,Cox proportional harzard model was used to explore the effect of BMI and five weight change indicators on hypertension among Chinese Han population,and generalized estimation equation(GEE model)was used to investigate the relationship between four weight change and blood pressure change(since the repeated measurement data were used to calculate the rate change of BMI,so only the first four indicators were included in the GEE model).Finally,the standardized regression coefficients of the correlation between different weight change indicators and hypertension and blood pressure were obtained,thus providing guidelines for people to lose weight scientifically.The results are as follows:1.Incidence density of hypertension of BMI and five weight change indicators and their comparison.① Both men and women,different BMI groups(including underweight group,normal weight group,overweight group and obese group)had different incidence density and the incidence density increased with increasing BMI.②Both men and women,incidence density of each group of absolute weight change group,relative weight change group,average weight change group and absolute BMI change group were different.In addition to the weight gain>10kg group of absolute weight change found no significant difference between different genders(u=0.384,P=0.701),the incidence density of men were higher than women in the other groups.2.Analysis of baseline indicators.All study indicators found significant difference between genders(P<0.001).In addition to HDL(P<0.001),men were higher than women of the other indicators(P<0.001).3.The result of the Cox model estimating the relationship between BMI and incidence of hypertension was as follows:RRs increased with the increase of BMI.Overweight and obesity were risk factors of hypertension for both men and women(men:RR=1.254,1.554;women:RR=1.245,1.700).Underweight was a protective factor of hypertension for men(RR=0.780).4.The results of the Cox models estimating the relationship between five weight change indicators and incident hypertension were shown as follows:① Absolute weight change(with absolute weight change≤5kg as reference group).For men,loss>10kg,5kg<loss≤10kg and 5kg<gain<10kg were protective factors of men(RR:0.680,0.585 and 0.943),when gaining weight more than 10kg,the risk of hypertension increased(RR=1.212).For women,weight loss was a protective factor of hypertension(RR of loss>10kg and 5kg<loss<10kg were 0.716 and 0.652)and weight gain was a risk factor(RR of 5kg<gain<10kg and gain>10kg were 1.124 and 4.005).②Relative weight change(with relative weight change<5%as reference group).Both men and women,loss>10%,5%<loss≤10%decreased the risk of hypertension(men:RR=0.681,0.633.women:RR=0.744,0.694)and gain>10%increased the risk of hypertension(men:RR=1.083.women:R=2.100).③Average weight change per year(with average weight change≤1 kg as reference group).Whether losing or gaining weight>lkg per year averagely,the RRs were greater than 1 in both genders.④Absolute BMI change(with BMI change≤1kg/m2 as reference group).For both genders,BMI loss>2kg/m2,1kg/m2<BMI<2kg/m2 were protective factors(men:RR=0.520,0.703.women:RR=0.637,0.761).While BMI>2kg/m2 increased the riks of women(RR=1.870).⑤BMI slope.BMI slope was a risk factor of hypertension in both men and women(men:RR=1.474.women:RR=1.530).⑥According to the results of standardized regression coefficients,for both men and women,the greatest impact indicator on the incidence of hypertension was BMI slope(standardized regression coefficient were 0.393 and 0.524).5.Correlation analysis and GEE model investigating the relationship between four weight change indicators and blood pressure change.(1)Correlation analysis betweenweight change and blood change.①For both men and women,there was a positive correlation relationship between weight change and SBP change,DBP change.②For both genders,the correlation coefficient of weight change and SBP change was higher than that of DBP change.(2)GEE model investigating the relationship between weight change and blood pressure change.①For men,in addition to absolute BMI change,the effect of other weight change indicators on SBP change and DBP change were little.Each additional absolute weight change,relative weight change,average weight change of one level were associated with increase of 0.0021mmHg,0.018mmHg,0.0011mmHg of SBP and 0.0017mmHg,0.014mmHg,0.0007 of DBP.Each additional absolute BMI change of one level was associated with increase of 0.0026mmHg of SBP and 0.0021mmHg of DBP.②For women,in addition to absolute BMI change,the association of weight change and DBP change was higher than that of SBP change.Each additional absolute weight change,relative weight change,average weight change of one level were associated with increase of 0.0026mmHg,0.0020mmHg,0.0007mmHg of SBP and 0.4650mmHg,0.3014mmHg,0.1810mmHg of DBP,Each additional absolute BMI change of one level was associated with increase of 0.0022mmHg of SBP and 0.3236mmHg of DBP.(3)According to the results of standardized regression coefficients,for both men and women,the greatest impact indicator on SBP change and DBP change was absolute BMI change.Conclusion:1.Weight change had an effect on blood pressure,but the relationship between different weight change indicators and blood pressure were not the same.2.According to the results of standardized regression coefficients of Cox proportional harzard model.①For men,the importance of different weight change indicators were BMI slope,average weight change per year,absolute BMI change,absolute weight change and relative weight change.②For women,the importance of different weight change indicators were BMI slope,average weight change per year,relative weight change,absolute weight change and absolute BMI change.3.According to the results of standardized regression coefficients of GEE model,for both men and women,the most important weight change indicator of SBP change,DBP change was absolute BMI change. |
PDF Full Text Request