Obesity and associated cardiovascular risk factors in iranian children: a cross-sectional study

Pediatrics International (2006) 48, 566–571
Obesity and associated cardiovascular risk factors in Iranian
children: A cross-sectional study
ANAHITA HAMIDI , HOSSEIN FAKHRZADEH , ALIREZA MOAYYERI , RASSUL POUREBRAHIM , RAMIN HESHMAT , MASOUMEH NOORI , YALDA REZAEIKHAH AND BAGHER LARIJANI Endocrinology and Metabolism Research Center, Doctor Shariati Hospital, Tehran University of Medical Science, Tehran, Iran Abstract
Background : Obesity is a growing public health problem in developing countries considering its association
with cardiovascular risk factors. Relationship between childhood obesity and these risk factors has not been
attested in the Iranian population before. The aim of the present study was to investigate frequency of cardiovas-
cular risk factors and their association with severity of obesity in a sample of Iranian obese children.
Methods : A total of 13 086 children aged 7 – 12 years were screened and those with waist circumference = 90th
percentile of their age were invited for further evaluations. Participants were divided into two groups of over-
weight or obese according to International Obesity Task Force criteria. Cardiovascular risk factors were defi ned
as high fasting total cholesterol, high low density lipoprotein, low high density lipoprotein, high triglycerides,
and systolic or diastolic hypertension. These factors were compared between obese and overweight children and
their correlations with body mass index and other measures of obesity were tested.
Results : Of 532 children (274 boys, mean age 9.5 ± 1.3) enrolled in the study, 194 were overweight and 338
were obese. Mean levels of triglyceride and Apo-lipoprotein B in obese children were signifi cantly higher than
overweight participants. A total of 81.9% of obese children and 75.4% of overweight children had at least one
cardiovascular risk factor. There were signifi cant correlations between body mass index and systolic blood pres-
sure, diastolic blood pressure, serum triglyceride, and Apo-lipoprotein B levels ( P values <0.01).
Conclusion : The high prevalence of cardiovascular risk factors in overweight and obese children and positive
correlation of these factors with severity of obesity emphasizes the need for prevention and control of childhood
obesity from early stages.
Key words
cardiovascular risk factors , childhood obesity , dyslipidemia , hypertension , prevention . Obesity is a growing global epidemic affecting all age groups. Health Organization Monica Project, over 30% of the people It is currently regarded as the leading cause of preventable in the Middle East are overweight. 6 A survey on secondary death worldwide and escalating obesity would be the greatest school students in Bahrain revealed that 15.6% of boys and health threat the world will face in the twenty-fi rst century. 1,2 14.7% of girls were overweight. 7 Like many other countries, Childhood obesity has also become a major global health con- the trend for obesity in Iranian children has doubled between cern in recent decades. 3 It is associated with increasing risk of atherosclerosis, hyperinsulinemia, hypertension and psycho- Even in childhood, obesity accompanies other risk factors social problems in the affected population. 4 for future cardiovascular diseases (CVD), including hyperten- The prevalence of obesity among children appears to be sion and dyslipidemia. This set of risk factors has been proved rising rapidly in developing countries, which could be attrib- to accelerate progression of atherosclerotic lesions in the cor- uted to changes in lifestyle. 5 Based on the data of the World onary arteries of young people. 8 Obese children are at increased risk for adult mortality and morbidity particularly from CVD. 9 Limited number of studies have been performed on child- Professor Bagher Larijani, Endocrinology and hood obesity and accompanying cardiovascular risk factors in Metabolism Research Center, Doctor Shariati Hospital, North the developing world. For a better understanding on the issue Kargar Avenue, Tehran 14114, Iran. Email: emrc@sina.tums.ac.ir and more effi cient programs for worldwide prevention of Received 13 March 2005; revised 25 June 2005; accepted 2 CVD mortality, more ethnic-specifi c studies are needed. Obesity and cardiovascular risk factors 567 Moreover, the relationship between obesity and other newly Iran) using automatic serum auto analyzer (Hitachi 902; Boe- suggested CVD risk factors (such as Apo-lipoprotein A and hringer Manneheim, Germany). Low Density Lipoprotein B) is unconfi rmed so far. We aimed in this study to determine (LDL) cholesterol, High Density Lipoprotein (HDL) choles- frequency and assortment of these factors and their relation terol, Apo-lipoprotein A, and Apo-lipoprotein B levels were with severity of obesity in a sample of primary school Iranian measured directly by Immunoturbidometric assay (Pars Azmoon kit; Tehran, Iran). As there is scarce evidence in hand about the borderline cholesterol levels to be considered as nor-mal or risky for children, and considering that all of our par- ticipants were obese and at-risk children, cut-offs proposed as acceptable by the expert panel of National Cholesterol Educa- This cross-sectional study was conducted from January to tion Program on blood cholesterol in children and adolescents May 2004 in the sixth district of Tehran. We screened 13 089 were used to defi ne normal values. 14 Total cholesterol = 170 pupils aged 7 – 12 years in all of the 65 primary schools of this mg/dL, LDL =110 mg/dL, HDL < 35 mg/dL, and triglyceride area. Waist circumference was used as the screening tool to levels = 150 mg/dL were considered as cardiovascular risk fac- tors. These four lipid profi le variables and systolic and diasto- equal or greater than 90th percentile for their ages were invited lic hypertension comprised six study target variables. The for further evaluations. A total of 1644 students had high waist crude values of Apo-lipoproteins A and B were also entered circumference, of whom 563 subjects attended a school-based clinic and enrolled in the study. The study protocol was ap- Pearson correlation test was used to assess the relationships proved by the ethics committee of Tehran University of between anthropometric measurements and cardiovascular Medical Sciences, Tehran, Iran, and written informed consent risk factors. To control for the possible effects of age and was obtained from students and their parents. height on the relationship between BMI and hypertension, par- Bodyweight was recorded to the nearest 0.5 kilogram (kg) tial correlation analysis was used. Differences between two using a standard beam balance scale with the subject barefoot groups of obese and overweight children were investigated and wearing light dresses. Body height was recorded to the nearest 0.5 centimeters (cm) according to the following proto- necessity of type I error correction for multiple tests used in col: no shoes, heels together and head touching the ruler with this study, P values <0.01 were considered as signifi cant. line of sight aligned horizontally. Body Mass Index (BMI) was calculated as the ratio of body weight (kg) to body height (m) squared. Participants were classifi ed as normal, over- weight or obese using the international age and gender spe-cifi c BMI cut-offs recommended by the International Obesity In total, 532 children (274 boys, mean age 9.5 ± 1.3) were en- Task Force (IOTF). 12 In total, 31 students (5%) were normal rolled in the study. According to IOTF criteria, 194 participants were overweight and the other 338 were obese. Anthropometric Waist circumference was measured to the nearest cen ti- measurements, blood pressures, and lipid profi les of the study meter with elastic-band meter with the student in the standing population are summarized in Table 1. Mean levels of triglyc- position at the level of the distal third of the distance from the eride in obese children were signifi cantly higher than over- xyphoid process to the umbilicus. Hip circumference was weight group (160.3 ± 46.0 vs 145.0 ± 36.0, measured 4 cm below the anterior superior iliac spine and Difference between the two groups regarding Apo-lipoprotein waist-to-hip ratios (WHR) were calculated as an index of B was marginally signifi cant ( P = 0.009). abdominal obesity. One experienced physician measured As indicated in Table 2, the percentage of participants with blood pressure using a mercury sphyngmomanometer with a any of the given risk factors was higher in the obese group. The cuff size suitable for each subject. Systolic blood pressure difference between two groups reached a statistical signifi - (Korotkoff phase I) and diastolic blood pressure (Korotkoff cance level only for hypertriglyceridemia ( P = 0.003). A total phase V) were measured twice with a 5 min interval from the of 18.1% of obese children and 24.6% of overweight children right arm in supine position and the average of the two meas- had no cardiovascular risk factor. A total of 12.9% of obese urements was taken as the student’s blood pressure. We used children had four or fi ve risk factors, while this fi gure was 8.3% recommendations of Task Force for High Blood Pressure for overweight children. The percentages of participants with in Children and Adolescents to defi ne systolic or diastolic different numbers of risk factors are schematically presented in Figures 1 and 2 for different genders. Difference between boys Overnight fasting venous blood samples were taken from and girls for none of the risk factors was signifi cant. the children. Triglyceride and cholesterol levels were meas- As shown in Table 3, different obesity scales were highly ured through calorimetric assays (Pars Azmoon kit, Tehran, correlated with each other. This statement was also correct for Table 1 Physical and paraclinical characteristics of studied children †
† Data are presented as mean ± standard deviation. HDL, High Density Lipoprotein; LDL, Low Density Lipoprotein. systolic and diastolic blood pressures as well as different com- severity of obesity and different cardiovascular risk factors. ponents of lipid profi le. Correlations between elements of lipid We observed that about 80% of our obese children are suffer- profi le and obesity scales reached to statistical signifi cance ing from at least one established cardiovascular risk factor and only for triglyceride and Apo-lipoprotein B ( Table 3 ). these risk factors are interconnected with the severity of obes- Systolic and diastolic blood pressures were signifi cantly ity. We found signifi cant correlation between crude measures correlated with anthropometric variables. The correlation of obesity (including BMI, waist circumference, and WHR) between Systolic blood pressure and BMI ( r = 0.297, P < and systolic and diastolic blood pressures of children. However, 0.001) remained signifi cant after controlling for age ( r = 0.240, the only elements of lipid profi le that showed a signifi cant cor- P < 0.001) and height ( r = 0.229, P < 0.001). Partial correla- relation with obesity measures were serum triglycerides and tion analysis also showed an independent relationship between Apo-lipoprotein B. This could have some applications for risk diastolic blood pressure and BMI controlling for age ( r = 0.193, assessment of CVD in children and adolescents. P < 0.001) and height ( r = 0.199, P < 0.001). The results of our study are consistent with other major studies that have examined the relationship between obesity and cardiovascular risk factors in children and adolescents. 9,15 – 22 Discussion
The high prevalence of hypertension in this study is in con-cordance with the results of other studies indicating a strong This is the fi rst study in Iran trying to explicate characteristics relationship between obesity and hypertension in children. 16,17 of obese children and to search for relationships between The signifi cant and independent association between BMI and Table 2 Number of participants with different cardiovascular risk
factors †
† Data are presented as frequency (percentage in parentheses).
‡ Defi ned according to Task Force for High Blood Pressure in Fig. 1 Distribution of boys with different numbers of cardiovas-
HDL, High Density Lipoprotein; LDL, Low Density Lipoprotein. Obesity and cardiovascular risk factors 569 Fig. 2 Distribution of girls with different numbers of cardiovas-
cular risk factors.
blood pressure (controlling for both child age and height) sup- ports the hypothesis that early-onset obesity is accompanied by many health risks including hypertension. 18 We found a signifi cant association between BMI and trig- lyceride levels in overweight or obese children. This was not true for the association of BMI with total cholesterol and LDL. Li et al. have also found that degree of obesity has no strong correlation with total cholesterol and LDL. 19 Only 2.8% of the children showed low HDL levels and there was no signifi cant relationship between HDL and anthro-pometric variables in this study, which supports the hypothesis that reduced HDL level may require years of history of obesity to develop and the degree of obesity would affect HDL levels later in life. 19,20 However, there was a signifi cant relationship between BMI and Apo-lipoprotein B levels in our sample. These fi ndings support the idea that measurement of the Apo- lipoprotein B levels is a useful initial approach to cardiovascu-lar risk assessment in obese children. 21 In contrast, we found no association between BMI and Apo-lipoprotein A levels which is similar to the fi ndings of other studies on obese chil- dren. 22 According to some authors, reduced Apo-lipoprotein A and increased Apo-lipoprotein B levels have a strong correla- tion with the development and progression of Atherosclerosis, more so than their equivalent lipoproteins HDL and LDL and in children these changes are associated with the presence of CVD in their parents. 23,24 Given the more fl uctuating and unre- liable nature of triglyceride levels, Apo-lipoprotein B could be the best choice for assessment of dyslipidemia and cardiovas-cular risk in obese children. , blood pressure; HDL, High Density Lipoprotein; LDL, Lo It has been suggested that both fat mass and fat distribution are important factors in assessing cardiovascular risk even However, the lack of correlation between WHR and dyslipidemia of obese children in our study could be interpreted as this index may be less appropriate for evalu- ation of cardiovascular risk in children and adolescents than in adults. We also found no association between hip circum- ference and dyslipidemia in this study. This is similar to the results of the Bogalusa Heart Study, which emphasizes that the hip circumference provides little information on cardio- ers and health-care professionals to prevent progression of association between waist circumference and triglycerides, In summary, we detected a high prevalence of cardiovascu- Apo-lipoprotein B and blood pressure in this study is in agree- lar risk factors in overweight and obese children and positive ment with the results of other studies which suggested that correlation of these factors with severity of obesity. Future waist circumference may be better measures of fat distribu- studies evaluating the possible role of Apo-lipoprotein B in the pathogenesis and prognosis estimation of obesity are rec- A combination of major cardiovascular risk factors (includ- ommended. Hypertension and dyslipidemia in overweight and ing hypercholesterolemia, hypertriglyceridemia, and hyper- obese children indicate a serious health risk and emphasize the tension) as well as insulin resistance are now known as importance of prevention and control of obesity from early ‘metabolic syndrome’. 26 Studies on prevalence and character- childhood to improve present and future health status. istics of this syndrome in children and adolescents also have shown similar results to this study. It has been indicated that the prevalence of ‘metabolic syndrome’ increases signifi cantly Acknowledgments
in relation with bodyweight and severity of obesity. 27,28 The main limitation of our study was the poor response The authors are thankful to Mr. Peyman Shooshtarizadeh in the rate, as from 1644 invited students only 563 (34%) particip- laboratory of Endocrine and Metabolism Research Center. ated in the study. The primary reason for nonparticipation They also appreciate the support from the Ministry of Education, could be the necessity of blood sampling in our study, which Bureau of Tehran, and the pupils of the primary schools in the decreases the compliance of children. As just about one-third sixth district of Tehran for their collaboration. of invited children attended the clinic, we should be cautious to generalize obvious characteristics of these children to the Iranian obese children, however, the paraclinical nature of cardiovascular risk factors assessed in this study decreases References
the impact of potential selection bias on our fi ndings. 1 Braddon FE , Rodgers B , Wadsworth ME , Davies JM . Onset of Number of obese participants was notably higher than obesity in a 36 year birth cohort study . BMJ 1986 ; 293 : 299 – 303 .
overweight children in our sample, however, this could not be 2 Burton BT , Foster WR , Hirsch J , Van Italie TB . Health implica- the same as the distribution of these children across the com- tions of obesity:an NIH consensus development conference . Int. munity. Two potential explanations for this observation could J. Obes. Relat. Metab. Disord . 1985 , 9 : 155 – 69 .
be, fi rst, the more anxiety and apprehension of parents of more 3 Zwiauer KF . Prevention and treatment of overweight and obes- ity in children . Eur. J. Pediatr . 2000 ; 159; (Suppl 1) : S56 – S68 . obese children regarding health status of their children (which 4 Bellizzi MC , Horgan GW , Guillaume M , Dietz WH . Prevalence forced them to attend with their children to the clinic), and of childhood and adolescent overweight and obesity in Asian second, the method of screening used in this study. We used and European countries . In : Chen C , Dietz WH ( eds ). Obesity in the cut-off of 90th percentile for waist circumference to fi nd Childhood and Adolescence . Lippincott Williams & Wilkins , obese children. This approach seemed to be highly sensitive 5 Kelishadi R , Pour MH , Sarraf-Zadegan N et al . Obesity and (as 95% of our participants were obese or overweight), but the associated modifi able environmental factors in Iranian adoles- specifi city of this screening method is questionable and some cents: Isfahan Healthy Heart Program – Heart Health Promotion from Childhood . Pediatr Int . 2003 ; 45 : 435 – 42 .
Another point is that, although most of the attended children 6 Gurney M , Gorstein J . The global prevalence of obesity: an ini- were overweight or obese, a minority of 5% were in the normal tial overview of available data . World Health Stat. Q . 1988 ; 41 :
251 – 4 .
range of weight according to IOTF criteria. Given the low 7 Musaiger AO , Matter AM , Alekri SA , Mahdi AR . Obesity number of these children and considering our purpose to fi nd among secondary school students in Bahrain. Nutr relationships between various risk factors and severity of obesity, 1993 ; 9 : 25 – 32 .
we excluded these children from the study. It is also of interest to 8 Berenson GS , Srinivasan SR , Nicklas TA . Atherosclerosis: a nu- evaluate the relationship between nutritional and biochemical tritional disease of childhood . Am. J. Cardiol . 1998 ; 82 : 22T – 9 .
9 Freedman DS , Dietz WH , Srinivasan SR , Berenson GS . The re- markers and severity of obesity, which was behind the scope of lation of overweight to cardiovascular risk factors among chil- this article. Further studies in this regard are warranted. dren and adolescents: the Bogalusa Heart Study The high percentage of obese children with two or more 1999 ; 103 : 1175 – 82 .
cardiovascular risk factors (55.2%) supports the notion that 10 Maffeis C , Pietrobelli A , Grezzani A , Provera S , Tato L . Waist excess adiposity is related to a poor cardiovascular risk profi le circumference and cardiovascular risk factors in prepubertal
children . Obes. Res . 2001 ; 9 : 79 – 87 .
even in children. Given the growing rate of obese children in 11 Higgins PB , Gower BA , Hunter GR , Goran MI . Defi ning health- urbanizing communities like Iran, identifi cation and manage- . Obes. Res . 2001 ; 9 :
ment of obese children is a critical issue for health policymak- Obesity and cardiovascular risk factors 571 12 Cole TJ , Bellizzi MC , Flegal KM , Dietz WH . Establishing a 20 Ho TF , Paramsothy S , Aw TC , Yip WC . Serum lipid and lipo- standard defi nition for child overweight and obesity worldwide: protein profi les of obese Chinese children . Med. J. Malaysia international survey . BMJ 2000 ; 320 : 240 – 43 .
1996 ; 51 : 68 – 74 .
13 Update on the 1987 Task Force Report on High Blood Pressure 21 Lynch JF , Marshall MD , Wang XL , Wilcken DE . Apolipopro- in Children and Adolescents: a working group report from the tein screening in Australian children: feasibility and the effect National High Blood Pressure Education Program of age, sex, and ethnicity . Med. J. Aust . 1998 ; 168 : 61 – 4 .
High Blood Pressure Education Program Working Group on 22 Glowinska B , Urban M , Koput A , Galar M . New atherosclerosis Hypertension Control in Children and Adolescents . Pediatrics risk factors in obese, hypertensive and diabetic children and 1996 ; 8 : 649 – 58 .
adolescents . Atherosclerosis 2003 ; 167 : 275 – 86 .
14 American Heart Associaton . Cholesterol in Children . [Cited 24 23 Brunzell JD , Sniderman AD , Albers JJ , Kwiterovich PO Jr. June 2005.] Dallas, TX, USA. Available from URL Apoproteins B and A-I and coronary artery disease in humans . www.americanheart.org/presenter.jhtml?identifi er= 556 Arteriosclerosis 1984 ; 4 : 79 – 83 .
15 Katzmarzyk PT , Tremblay A , Perusse L , Despres JP , Bouchard 24 Srinivasan SR , Berenson GS . Serum apolipoproteins A-I and B C . The utility of the international child and adolescent over- as markers of coronary artery disease risk in early life: the weight guidelines for predicting coronary heart disease risk fac- Bogalusa Heart Study . Clin. Chem . 1995 ; 41 : 159 – 64 .
tors . J. Clin. Epidemiol . 2003 ; 56 : 456 – 62 .
25 Freedman DS , Serdula MK , Srinivasan SR , Berenson GS . Rela- 16 Adeyanju M , Creswell WH , Stone DB , Macrina DM . A three- tion of circumferences and skin-fold thicknesses to lipid and year study of obesity and its relationship to high blood pressure insulin concentrations in children and adolescents: the Bogalusa in adolescents . J. Sch. Health 1987 ; 57 : 109 – 13 .
Heart Study . Am. J. Clin. Nutr . 1999 ; 69 : 308 – 17 .
17 McMurray RG , Harrel JS , Levine AA , Gansky SA . Childhood 26 James PT , Rigby N , Leach R .; International Obesity Task Force. obesity elevates blood pressure and total cholesterol independ- The obesity epidemic, metabolic syndrome and future preven- . Int. J. Obes. Relat. Metab. Disord . tion strategies . Eur. J. Cardiovasc. Prev. Rehabil . 2004 ; 11 :
1995 ; 19 : 881 – 6 .
18 Chu NF , Wang DJ , Shieh SM . Obesity, leptin and blood pres- 27 Weiss R , Dziura J , Burgert TS et al . Obesity and the metabolic sure among children in Taiwan: the Taipei Children’s Heart syndrome in children and adolescents . N. Engl. J. Med . 2004 ; Study . Am. J. Hypertens . 2001 ; 14 : 135 – 40 .
350 : 2362 – 74 .
19 Li S , Liu X , Okada T , Iwata F , Hara M , Harada K . Serum lipid 28 Vanhala M , Vanhala P , Kumpusalo E , Halonen P , Takala J . Rela- profi le in obese children in China. Pediatr Int . 2004 ; 46 :
tion between obesity from childhood to adulthood and the meta- bolic syndrome: population based study . BMJ 1998 ; 317 : 319 .

Source: http://emrii.tums.ac.ir/upfiles/54215642.pdf