Comparison of adiposity measures in the identification of adolescents with hypertension

Background: Hypertension in children is becoming a growing heath problem, with increasing prevalence. Objectives: To examine the relationship of body mass index (BMI), waist circumference (WC), waist hip ratio (WHR), waist stature ratio (WSR) and conicity index (CI) with blood pressure and to differentiate the screening potential of BMI, WC, WHR, WSR and CI for the presence of hypertension in children. Method: A school-based cross-sectional study was carried out in 2017 on 13-16 year old Bengali children in two schools (a boys’ school and a girl’s school) from Howrah city, West Bengal where stature (ST), weight (WT), WC and hip circumference (HC) were measured, and BMI, WHR, WSR and CI calculated. Schools were randomly selected from the list of affiliated schools under the West Bengal Board of Secondary Education. All 13-16 year old children from the two schools were recruited after obtaining informed consent from parents and assent from the children. Children with pre-existing serious illnesses and deformities were excluded from the study. Bio-social information was obtained using an open-ended schedule. Age was obtained from the school register. Hypertension was defined as age and stature adjusted systolic blood pressure (SBP) and/or diastolic blood pressure (DBP) ≥95th percentile. Partial correlations, logistic regressions and area under the receiver operating characteristic curves (AUC) were estimated to understand the relationship of adiposity measure with blood pressure and hypertension. _________________________________________ 1Research Scholar, 2Assistant Professor, Department of Anthropology, Visva-Bharati University, West Bengal, India *Correspondence: jrghosh@rediffmail.com (Received on 24 April 2018: Accepted after revision on 29 June 2018) The authors declare that there are no conflicts of interest Funding: Research scholar is getting UGC fellowship from Visva-Bharati, Govt. of India Open Access Article published under the Creative Commons Attribution CC-BY License Results: A total of 535 adolescents, comprising 282 boys and 253 girls aged 13–16 years, participated in this study. There was no statistically significant sex difference in age. Overall prevalence of hypertension was 8.5%. All adiposity measures were positively correlated with blood pressures. Logistic regression showed that BMI had the strongest associations with hypertension in both sexes. The AUC for BMI (0.82, 95% CI 0.751 to 0.895) was higher than those of other adiposity measures in assessing hypertension in boys. However, in girls, both BMI (AUC 0.75, 95% CI 0.644 to 0.856) and WSR (AUC 0.75, 95% CI 0.638 to 0.865) had similar higher assessment ability for hypertension compares to other measures. Conclusions: BMI had the highest odds ratio for hypertension, and was the best adiposity measure to assess hypertension. DOI: http://dx.doi.org/10.4038/sljch.v48i1.8650 (


Introduction
Today, in developing countries, there is a marked shift from communicable to non-communicable diseases 1 .Studies show a high prevalence of obesity among children globally 2 .Obesity is the most significant risk factor for childhood hypertension 3,4 .Obese children have a threefold greater risk for high blood pressure compared to non-obese children 5 .Hypertension is identified as the third major risk factor of deaths among children, and the prevalence of hypertension in children is increasing 6 .Moreover, among children, especially those with obesity, hypertension is associated with end-organ damage 7,8 .However, little attention has been paid to the problem of hypertension in the young.Increasing evidence suggests blood pressure tracking from childhood to adulthood 9 .High blood pressure in children is associated with increased cardiac morbidity and mortality 10 .However, remarkably, the prevalence of hypertension in children is generally underestimated 8 .Body fat distribution, rather than the extent of obesity, determines the hypertension risk 11 .Rational assessment of body fat distribution involves measurement of body mass index (BMI), waist circumference (WC) and waist hip ratio (WHR) 3,11,12 .However, there is little data on the relationship between body fat distribution and the risk of high blood pressure in Indian children.Studies that examined the relationships between adiposity measures and hypertension have yielded different findings 7,13 and thus, the relationships are not fully understood.

Objectives
To examine the relationship of BMI, WC, WHR, waist stature ratio (WSR) and conicity index (CI) with blood pressure and to differentiate the screening potential of BMI, WC, WHR, WSR and CI for the presence of hypertension in Indian children.

Method
A school-based cross-sectional study was carried out in 2017 on 13-16 year old Bengali children in two schools (a boys' school and a girl's school) from Howrah city, an urban area adjacent to Kolkata, the capital of the state West Bengal.Schools were randomly selected from the list of affiliated schools under the West Bengal Board of Secondary Education.All 13-16 year old children from the two schools were recruited after obtaining informed consent from parents, permission from the schools and assent from the children.Children with pre-existing serious illnesses and deformities were excluded from the study.Bio-social information was obtained using an open-ended schedule.Age was obtained from the school register.Ethical clearance for the study was obtained from the Research Board, Visva-Bharati University, West Bengal, India.
Children enrolled in study were physically examined to obtain stature (ST), weight (WT), WC and hip circumference (HC) in accordance with standard guidelines 14 .ST was measured to the nearest 0.1 cm using an anthropometer.WT was measured in light clothing without shoes to the nearest 0.1 kg using an electronic scale.WC and HC were measured to the nearest 0.1 cm in standing position using a non-elastic flexible tape.BMI [WT (kg)/ST (m 2 )], WHR [WC (cm)/HC (cm)], WSR [WC (cm)/ST (cm)] and CI = WC (m)/ {0.109 × √ [WT (kg)/HT (m)]} were calculated subsequently.
Systolic blood pressure (SBP) and diastolic blood pressure (DBP) in left arms were measured in each student using a sphygmomanometer with appropriate cuff.Each student was requested to sit quietly for 5 minutes before taking the blood pressure to relieve any stress.The mean of two measurements was taken.Hypertension was defined as age and ST adjusted SBP and/or DBP ≥95 th percentile 15 .
Descriptive statistics were computed by sex for age, anthropometric variables and blood pressures, and expressed as mean and standard deviation (SD).Student-t test was used to compare anthropometric and blood pressure variables between boys and girls.Partial correlations, adjusted for age and ST were estimated separately for both sexes to examine association of adiposity measures with blood pressures.Logistic regressions adjusted for age and ST were used to examine relationship of adiposity measures with hypertension.The area under each receiver operating characteristic curves (AUC) and 95% confidence intervals (95% CI) were estimated to understand the relative ability of various adiposity measures to assess hypertension.Statistical analysis was conducted with software package IBM SPSS (version 22.0).A p value less than 0.05 was considered of significance.

Results
A total of 535 apparently healthy adolescents comprising 282 boys and 253 girls were included in the study.Characteristics of the studied population are presented in Table 1.The age and ST adjusted odds ratio (OR) along with 95% confidence intervals of BMI, WC, WHR, WSR and CI for hypertension are shown in Table 3.All anthropometric variables were significantly and positively associated with hypertension (Table 3).
The AUC in assessing hypertension for adiposity measures are shown in Table 4.

Table 4: Area under the ROC curves of different adiposity measures in assessing hypertension
Variable

AUC: Area under each receiver operating characteristic curves, 95% CI: 95% confidence interval
Apart from the CI in boys, AUCs of all adiposity measures were significantly higher than the noneffect value 0.5.The AUC for BMI (0.82, 95% CI 0.751 to 0.895) was larger than those of other measures in assessing hypertension in boys (Figure 1).However, in girls, both BMI (AUC 0.75, 95% CI 0.644 to 0.856) and WSR (AUC 0.75, 95% CI 0.638 to 0.865) had similar higher assessment ability for hypertension compared to other measures (Figure 2).

Discussion
Hypertension in adults is now recognized globally as a major public health problem 16 .Unexpectedly, hypertension in children is also becoming a growing heath problem, with increasing prevalence 17 .Interestingly, the complications of hypertension not only occur in adult patients, but can involve children as well 4 .
This cross sectional study explored the relationship of various adiposity measures with blood pressures in adolescents living in an urban community.In addition, we studied the screening potential of BMI, WC, WHR, WSR and CI for the presence of hypertension.We found that, even though participants were of similar age, males tended to have higher mean anthropometric measurements compared to females, except for BMI and WSR.The prevalence of hypertension in boys and girls was 8.2% and 8.7%, respectively.Similar to the present study, higher prevalence of hypertension in girls was observed by Nkeh-Chungag et al. 8 .
Durrani et al. also found a higher prevalence of hypertension in girls as compared to boys 18 .Contrary to that, Gamboa-Delgado et al. 13 observed that prevalence of hypertension was more in boys than in girls 13 .Similar findings were also reported by other studies 3 .However, the overall prevalence of hypertension in the study was 8.5% which is higher than in the recently published studies by Hansel et al. 19 (3.6%), Patel et al. 20 (5.4%) and Vohra et al. 21(3.1%).Compared to the present study, a lower prevalence of hypertension was also observed in children of Wardha (5.75%) 22 and Patna (4.6%) 23 .However, Durrani and Waseem 24 reported that 9.4% of children were hypertensive in their study population.
The results of the present study revealed that all adiposity measures were positively correlated with blood pressures, though the association of CI with SBP and DBP were not significant.Similar positive associations were also observed in recent studies 3,23 .In this study, logistic regression analysis showed a very strong positive association of BMI with risk of hypertension prevalence in these children.Zhang et al. also reported a strong positive association of high blood pressure with BMI in children of Hainan province of China 25 .Similar positive association between hypertension and BMI was also observed by Sun et al. 26 .Some studies reported an association between childhood obesity and hypertension 23,27 .Study also demonstrated that children with hypertension had higher mean BMI compared with the prehypertensive and normal weight participants 4 .Another recent study demonstrated that being in an early onset obese or overweight trajectory was associated with increased risk of childhood hypertension 9 .Moreover, overweight children had a twofold greater risk of hypertension whilst obese children had a fourfold risk 28 .Other studies also demonstrated that increased BMI was associated with increased blood pressure in children 29,30 .ROC curve analysis in the present study also confirmed that BMI had comparatively better screening potential for hypertension risk in children.This is similar to the findings in a recent study by Ma et al. 31 , who demonstrated that WC and WSR were no better than BMI to identify the risk of hypertension in children.However, Liang et al. 32 showed a similar discriminatory capacity of BMI, WC and WSR for hypertension risk.
The strong positive association between increasing BMI and elevated blood pressure might be due to the fact that obesity progressively raises blood pressure by changing cardiac output, cardiac systolic and diastolic function, renal pressure natriuresis 33 , sympathetic nervous system, reninangiotensin-aldosterone system activation 34 and aortic stiffness .However, the possible mechanism whereby obesity lead to higher blood pressure might be enhanced adipocyte secretion of adipokines and pro-inflammatory cytokines 36 .In a recent study, Pierce et al. found that adolescents with raised BMI had increased left ventricular mass, high aortic wave amplitude and SBP, compared with their peers with normal weight 37 .This might be the reason that BMI has been recommended as an index of adiposity for epidemiological studies / clinical practice 38 .Moreover, amount of visceral adipose tissue in young adults was also associated with BMI changes during adolescence 39 .
The cross-sectional design of our study allows us to demonstrate the associations between adiposity measures and blood pressure in children.Further studies with larger samples to determine a cut-off value of adiposity measures should be conducted.A limitation of the present study was unavailability of biochemical measurements so that it was not possible to examine the usefulness of these indices for screening other cardio-metabolic risk factors.

Table 1 : Characteristics of the studied population
There were no significant sex differences in age, HC and WSR.However, boys had significantly higher mean values than girls in ST, WT, WC, BMI, WHR, CI, SBP and DBP (Table1).The prevalence of hypertension in boys and girls was 8.2% and 8.7%, respectively with an overall prevalence of 8.5%.Gender wise partial correlation coefficients of SBP and DBP with anthropometric variables are presented in Table2.

Table 2 : Partial correlation between adiposity measures and blood pressure
All correlations were positive and significant (p<0.05),except for CI with SBP and DBP in boys.

Table 3 : Odds ratios of adiposity measures for hypertension
BMI had the highest odds ratio for hypertension, and was the best adiposity measure to assess hypertension.