Obesity and Its Diagnostic Methods in Turkish Children

Objective: The application of body mass index (BMI) for obesity classification in all population studies has been questioned by the scientific community. It has been found that the bioelectrical impedance analysis (BIA) is an accurate method for screening obesity. This study was conducted to evaluate the accuracies of BMI, skinfold thickness (SFT), leptin and BIA measurements in obesity classification and to find correlations between BIA and the other indicators for obesity. Materials and Methods: This case-control study included 178 children of whom 90 were in the obese group and 88 in the control group. The study measured BMI, SFT, leptin level and BIA-mediated body fat percentage (BIA BFP) in each child. Results: The BMIs, leptin levels, SFTs and BIA BFPs of children in the obese group were found to be higher than those in the control group (p<0.001). The measurement of BIA BFP strongly correlated with BMI, SFT and waist circumference, whereas BIA BFP measurement showed weak-moderate correlation with leptin level. Conclusion: Bioelectrical impedance analysis was found to be an accurate measure of BFP in obesity. In addition, BIA may prevent the incorrect diagnosis of obesity as determined by BMI alone, especially in boys during the pubertal period.


Introduction
The prevalence of primary obesity, which occurs upon excessive deposition of fat in the body and is not caused by a defined disease, has been increasing worldwide [1].Obesity is a community problem with social and economic ramifications, and it significantly affects morbidity and mortality [2,3].
The first important step in the treatment of obesity is assessing the body composition of individuals and thus helping to prevent obesity.There exist various methods for the determination of body fat [4,5], such as body mass index (BMI), skinfold thickness (SFT), bioelectrical impedance analysis (BIA), isotope dilution method, measurement of body K, measurement of dual energy X-ray absorption (DEXA), neutron activation, computed tomography and magnetic resonance imaging.BMI is the most commonly used measure for determining obesity.However, in the clinic, the measure of BMI alone is insufficient, as it does not distinguish between increases in body weight attributed to increased muscle mass and those attributed to fat [6].Therefore, in recent obesity studies various anthropometric measurements and the measurement of body fat distribution have been used frequently.The measurement of SFT, a low cost and easy method, has a poor accuracy rate for determining obesity.Another easily applicable method, DEXA, has a high accuracy rate but it involves a high cost even though it uses low radiations.One of the most effective methods to assess the body fat ratio is BIA [7].This method is based on the body's resistance to an applied electrical current [8].It determines a body's fluid volume (total, intracellular and extracellular), body fat and muscle mass.It is an easily applicable, low cost method with high accuracy.
Leptin, the protein product of the ob gene, is secreted by the white adipose tissue cells in response to changes in the metabolic state.This hormone regulates body fat mass and body weight by inhibiting dietary intake and stimulating energy expenditure [9,10].In several past studies, leptin has been shown to be an important marker of obesity in adults [11,12].The detection of increased leptin levels in obese individuals shows a resistance to leptin [13,14].
The aim of this study was to investigate the role of BMI, SFT, leptin and BIA measurements in the diagnosis of obesity and to compare BIA with other diagnostic methods.

Materials and Methods
In this study, school-attending subjects between the ages 6 and 16 years were considered.Of the 178 children, 90 children weighed above the 97 th percentile and the remaining 88 children weighed below the 97 th percentile, constituting the obese and control groups, respectively.Children with secondary obesity were excluded from the study.Permission for the study was obtained from the ethics committee.Signed consent forms were taken from the families of the children and a questionnaire was filled out by the families.The height, weight, and waist circumference of each child were measured using the Seca scale at precision levels of 1 mm and 0.1 kg for height and weight measurements, respectively.These measurements were taken in children who had fasted for a period of 12 hours and were wearing thin clothes and no shoes.The BMI (kg/m 2 ) was calculated by dividing body weight (kg) by the square of height (m).According to the BMI values specified by the Centers for Disease Control (CDC, 2000) based on age and sex, the children with BMI measurements above the 95th percentile were identified as obese [15].
The measurement of SFT was performed using the Holtain caliper (Holtain Ltd.Crymch, UK) and was taken from the biceps; triceps; subscapular, abdominal and suprailiac regions; thigh and calf of the right side of the body.Each measurement was repeated three times and the average was taken.The waist circumference was also measured.
Blood samples for measuring leptin levels were taken when the children were hungry and were stored at -80°C.Plasma leptin levels were measured using Enzyme Amplified Sensitivity Immuno Assay (EASIA) method, and the result was recorded as ng/mL.Each child's body fat percentage (BFP) was measured by a BIA device (BIA 101) using the conventional tetrapolar technique, a 50 kHz fixed frequency, and the passage of a fixed sinusoidal current.The BFP values were calculated on a computer using the Bodygram-Akern S.r.l 1,3 program.
Statistical analysis was performed with SPSS 11.5, and p<0.05 was considered significant.
Student's t test and chi-square test were used for the comparisons of parametric data.A Pearson correlation test was used to determine the correlations between various measurements.

Results
A total of 178 children (64 of whom were boys) with a mean age of 10.36±2.44 years were studied.The mean body weight, height and BMI were calculated in all children as 49.43±18.09kg, 144.6±12.7 cm, 22.84±5.51kg/m 2 , respectively.
There was no significant difference in the ages, heights and gender distribution between the obese and control groups.In 47% of the children in the obese group, the obesity had started by the beginning of primary school.Moreover, in the obese group, 86.5% of the children had family histories of obesity, whereas this rate was 67% in the control group.An increase in BMI was determined as age and weight increased in all children (r=0.422,r=0.902).
All SFT values were two times higher in the obese group compared with those in the control group (p=0.000)(Table 1).The anthropometric measurements of the girls were higher than those of the boys (p>0.05).The waist circumferences in the obese group were significantly higher than those in the control group (Table 1).In our study, waist circumference had a good correlation with BMI, serum leptin level, and BIA BFP (r=0.813,r=0.548, and r=0.500, respectively).Also subscapular (r=0.598),suprailiac (r=0.585),biceps (r=0.582),triceps (r=0.575),abdomen (r=0.584),thigh (r=0.549),leg SFT (r=0.554) and waist circumference (r=0.642)measurements were significantly correlated with the serum leptin levels.Plasma leptin levels were significantly higher in the obese group as compared with those in the control group (Table 1).For the obese group, the leptin levels of girls were 1.5 times higher than those of boys (13.6 ng/mL and 8.29 ng/mL, p=0.019).In our study, plasma leptin levels were increased moderately by age both in girls (r=0.284) and boys (r=0.266).Although the leptin levels between girls and boys showed no differences in the prepubertal period (3.84±3.76,3.46±3.76ng/mL), during the pubertal period, the leptin levels were significantly higher in girls than in boys (10.66±9.51,6.16±6.48ng/mL) (Figure 1).In addition, the leptin levels of girls were significantly higher in the pubertal period than in the prepubertal period (10.66±9.51,3.84±3.76,p<0.001).However, the increase in the leptin levels of boys in the pubertal period was not significant compared with those in the prepubertal period.(6.16±6.48,3.46±3.76,p=0.083) (Fig. 1).We found that leptin levels increased with increases in body weight for both sexes during the prepubertal period.

Discussion
The concepts of obesity and overweight in the pediatric age group are not fully understood all over the world including Turkey.The difficulties associated with the diagnostic methods used to identify obesity and overweight have been challenging.In addition, comprehensive studies in this area are limited.For this reason, we analyzed the methods (BMI, BIA, anthropometric measurements, and serum leptin level) that are used in the evaluation of obesity using Turkish children between the ages 6 and 16 years.
Our studies indicated that the BMIs of Turkish children who were considered obese were significantly higher than those of the control group.In addition, there was no significant difference in the above finding between the sexes.Therefore, we suggest that the measurement of BMI for the evaluation of obesity still remains a useful, reproducible and easily performed method.
The advantages of the SFT method, which is used in the diagnosis and follow-up of obesity, are that it is a simple, easily performed and inexpensive method.However, the greatest limitation of this method lies in the high variability in measurements obtained by different people, and thus the SFT is a low reliability method.Because the measurement variability in millimeters can lead to large differences in the derived parameters, it is important that the same person perform and repeat all measurements for reliability [16].With the repeated measurements taken by the same person, we found that the subcutaneous fat tissue, total body fat and waist circumference measurements were significantly higher in the obese group as compared with those in the control group (p<0.001).Hence, the usage of anthropometric measurements, such as subcutaneous fat tissue and BIA BFP, along with the measurement of BMI, would enable better diagnosis and follow-up treatment for obesity.Leptin regulates body weight by inhibiting dietary intake and stimulating energy expenditure [10].In several previously published studies, leptin has been found to be an obesity marker in adults [11,12].Increased leptin levels in human obesity indicate a resistance to leptin.Similarly, we observed higher leptin levels in obese children, corresponding to a resistance to leptin during the pathogenesis of obesity [14].
It is thought that the lower serum leptin levels in boys during the pubertal period might be related to testosterone [13].Eventually, serum leptin levels might become related to adipose tissue in the pubertal period.Additional prospective studies are required to understand the role of leptin in growth and reproductive function during childhood and adolescent period.
The serum leptin level was increased upon weight gain in malnourished children, and its level was positively correlated with BMI [17].We determined good correlations between BIA BFP and BMI, SFT and anthropometric measurements, whereas a weak-moderate correlation between BIA BFP and serum leptin level was obtained.These data support a role for leptin in regulating visceral fat tissue [14].
Body composition, especially fat mass assessment, is an important parameter in obese patients.But densitometrydual methods, such as X-ray absorptiometry, are expensive, difficult to use and impractical.However, the measurement of BIA is not harmful to the patient, and it is repeatable and easy to use in the clinic [18].In a 2001 study, measurements of body fat in 1139 female and 1243 male children between the ages of 7 and 16 years were made, and BIA was determined to be the alternative method for DXA [19].In another study, it was stated that the usage of BIA combined with BMI is more significant in the evaluation of obesity [20].In a 2000 study performed on 98 Native American children, Lohman et al. [21] suggested the usage of both anthropometric and BIA measurements in estimating total body fat, and they stated that measurements of SFT alone are not significant.In children between the ages 8 and 12 years, the amount of body fat detected with MRI was compared with those from BIA-SFT-BMI measurements, and a significant correlation was found between the MRI values and those of BIA-BMI-triceps and subscapular SFT measurements.In addition, the combination of BMI and BIA measurements was found to be the closest to the result obtained by MRI [22].Moreover, in another study, it was determined that the usage of BIA alone was not adequate [23].
It is known that BMI measurement might be affected by muscle and bone structure.In girls, there was an increase in both BMI and BIA BFP during the pubertal period compared with the prepubertal period.However, in boys, only BMI, not BIA BFP, increased in the pubertal period compared with the prepubertal period, and this observation was thought to be related to an increase in muscle mass.
In conclusion, our studies showed that BIA is an accurate indicator for evaluating BFP in obese children.In addition, we observed that BIA may prevent the incorrect diagnosis of obesity, which is possible when using BMI measurement alone, especially in boys during the pubertal period.

8 Figure 1 .
Figure 1.The comprasion of all parameters investigated in prepubertal and pubertal children.

Table 1 . Comparison of parameters investigated in children from the obese and the control groups
*: Body mass index, † : Skinfold thickness II, ‡ : Bioelectrical impedance analysis, § : Body fat percentage