Associations Between Tri-ponderal Mass Index And Blood Pressure Among Children In Yantai:A Longitudinal Study

Background:Hypertension is one of the most common chronic diseases and the most important risk factor for cardiovascular disease and death.More than 1 billion people worldwide suffer from hypertension,and about 17.7 million people die from cardiovascular diseases every year,of which 9.4 million die from hypertension.Children and adolescents with elevated blood pressure,if not controlled,may develop coronary heart disease,stroke,kidney failure,and cognitive impairment in adulthood.Epidemiological studies have shown that childhood obesity is closely related to adult hypertension,dyslipidemia,diabetes and other diseases.Therefore,the control and intervention of childhood obesity should begin from childhood.Body mass index（BMI）is widely used to measure the body growth and development in populations worldwide.However,weight and height squared are not proportional during adolescent growth,thus reducing the accuracy of BMI.Therefore,Peterson et al.proposed that tri-ponderal mass index（TMI）could replace BMI as a new index for body fat screening among children and adolescents in 2017,which was calculated as weight divided by height cubed.To date,no longitudinal studies have explored the association between TMI and blood pressure in children,and only one cross-sectional study in Italian adolescents found that TMI was superior to BMI in assessing hypertension.Objective:The objectives of this study were as follows:（1）analyze the causal relationship between TMI and blood pressure;（2）explore the grouping of children’s TMI trajectories;（3）analyze the relationship between different TMI trajectory groups and the incidence of hypertension;（4）evaluate the effect of weight status change based on TMI on the changes of blood pressure and hypertension over time in children.Methods:Four districts of Yantai（Zhifu District,Laiyang city,Laizhou city and Longkou city）were randomly selected to establish the Yantai Children Health examination Project cohort in 2014.All 7 years old children in the first grade of primary schools were included in the study.Children’s demographic information（name,school,class,sex,region）and physical examination information（height,weight,blood pressure）were collected（Visit 1）.Follow-up was conducted annually from 2015 to 2019 with the same content as baseline measurements（Visit 2-Visit 6）.Children who had been followed up less than four times and had abnormal TMI data were excluded.TMI was calculated by dividing weight by the cube of height.Blood pressure was measured using an electronic sphygmomanometer.Repeated measures analysis of variance was used to assess longitudinal changes in BMI and TMI between the six consecutive years.We investigated temporal associations between TMI and BP with a cross-lagged panel model.The group-based trajectory model was used to analyze the trajectory grouping of children’s TMI over time.Cox proportional hazard model was used to evaluate the risk of hypertension in different TMI trajectories.Generalized estimating equation model was fitted to examine the main and interaction effects of weight status change based on TMI and time with blood pressure and the incidence of hypertension.We considered P<0.05to be statistically significant.Results:1.Characteristics of study population The study at baseline included 20,856participants（10,810 boys and 10,046 girls）aged 7 years in 2014.There were significant differences between males and females in region,height,weight,BMI,TMI,systolic blood pressure（SBP）,diastolic blood pressure（DBP）.Compared with the previous visit,height,weight,BMI,SBP,and DBP were higher for each follow-up visit.2.Stability of TMI and BMI with Age BMI increased rapidly between ages 7 and12（P<0.001）,while TMI was almost stable throughout childhood,with population means hovering at approximately 13 kg/m³.3.The causal relationship between TMI and BP In the cross-lagged panel analysis,TMI at Visit 1 was a significant predictor of BP at Visit 3（β=0.715 for SBP andβ=0.310for DBP,P<0.01）,and TMI at Visit 3 was a significant predictor of SBP at Visit 5（β=1.016,P<0.01）.Meanwhile,BP at Visit 1 was significantly associated with TMI at Visit 3（β=0.011 for SBP andβ=0.008 for DBP,P<0.01）,and BP at Visit 3 was significantly associated with TMI at Visit 5（β=0.003 for SBP andβ=0.008 for DBP,P<0.01）.In addition,the cross-lagged path coefficients indicated that the influence of TMI on BP was stronger than that of BP on TMI.4.TMI trajectory grouping and their association with hypertension The results of TMI group-based trajectory model showed that in the total population,the model was optimal when five distinct trajectories in TMI were identified and were labeled as low-decreasing group（n=5,625,27.0%）,moderate-stable group（n=6,932,33.1%）,moderate-increasing group（n=4,628,22.1%）,high-increasing group（n=2,500,12.1%）,and extreme-high increasing group（n=1,171,5.7%）.Compared with the moderate-stable group,the hazard ratio（HR）and 95%confidence interval（CI）of hypertension in low-decreasing group was 0.88（0.83-0.94）,indicating that the decreased level of TMI was a protective factor for hypertension;while moderate-increasing group,high-increasing group and extreme-high increasing group increased the risk of hypertension（moderate-increasing group:HR=1.25,95%CI:1.18-1.33;high-increasing group:HR=1.56,95%CI:1.46-1.68;extreme-high increasing group:HR=1.63,95%CI:1.49-1.80）.5.Effect of weight status changes based on TMI on blood pressure changes over time Significant interactions between change in weight status and time with SBP and DBP were identified(χ_interaction²=361.22,df=15,P<0.001 for SBP,andχ_interaction²=195.88,df=15,P<0.001 for DBP).Compared with participants who remained normal weight,participants who transited from normal weight to abnormal weight had higher SBP and DBP from Visit 2（β=0.627,P=0.011 for SBP andβ=0.457,P=0.03 for DBP）to Visit 6（β=5.931,P<0.001 for SBP andβ=2.786,P<0.001 for DBP）.Meanwhile,remained abnormal weight group and transited from normal weight to abnormal weight group maintained a higher blood pressure level,and the rate of increase tended to be faster overtime.6.The association between change in weight status based on TMI and the risk of hypertension Compared with participants who remained normal weight,the odd ratio（OR）and 95%CI of hypertension in transited from normal weight to abnormal weight group and remained abnormal weight group were 1.53（1.45-1.63）and 2.57（2.34-2.82）,suggesting that children who become or remain overweight or obese are at higher risk for high blood pressure.The OR（95%CI）of the transited from abnormal weight to normal weight group was 1.81（1.57-2.10）,which was significantly lower than that of the remained abnormal weight group,suggesting that weight loss could reduce the risk of hypertension.Conclusions:（1）In elementary school children,TMI remained stable with age compared to BMI;（2）The present study found a bidirectional causal relationship between TMI and BP,but the effect of TMI on BP was stronger than that of BP on TMI;（3）Five distinct trajectories in TMI were identified,namely,low-decreasing group,moderate-stable group,moderate-increasing group,high-increasing group,and extreme-high increasing group;（4）The decrease of TMI level is a protective factor of hypertension in children,while the increase of TMI level is a risk factor of hypertension in children;（5）Weight loss could control or even reduce the risk of hypertension in children.This study suggests that we should pay attention to the long-term weight changes in children,maintaining a normal weight level may help reduce the risk of hypertension in adolescence.