Effect of Android to Gynoid Fat Ratio on Insulin Resistance in Obese Youth

ARTICLE Effect of Android to Gynoid Fat Ratio on Insulin Resistance in Obese Youth

Julien Aucouturier, MSc; Martine Meyer, MD; David Thivel, MSc; Michel Taillardat, MD; Pascale Duche´, PhD

Background: Upper body fat distribution is associ- energy x-ray absorptiometry. Insulin resistance was as- ated with the early development of insulin resistance in sessed by the homeostasis model of insulin resistance obese children and adolescents. (HOMA-IR) index.

Objective:: To determine if an android to gynoid fat ra- Results: There were no differences in weight, body mass tio is associated with the severity of insulin resistance in index, and body fat percentage between tertiles. Values obese children and adolescents, whereas peripheral sub- of HOMA-IR were significantly increased in the 2 higher cutaneous fat may have a protective effect against insu- tertiles (mean [SD], tertile 2, 2.73 [1.41]; tertile 3, 2.89 lin resistance. [1.28]) compared with the lower tertile (tertile 1, 1.67 [1.24]) of android to gynoid fat ratio (PϽ.001). The Setting: The pediatric department of University Hos- HOMA-IR value was significantly associated with an- pital, Clermont-Ferrand, France. droid to gynoid fat ratio (r=0.35; PϽ.01). Design: A retrospective analysis using data from medi- Conclusions: Android fat distribution is associated with cal consultations between January 2005 and January 2007. an increased insulin resistance in obese children and adolescents. An android to gynoid fat ratio based on dual- Participants: Data from 66 obese children and adolescents coming to the hospital for medical consultation were energy x-ray absorptiometry measurements is a useful and used in this study. simple technique to assess distribution of body fat associated with an increased risk of insulin resistance. Main Outcome Measures: Subjects were stratified into tertiles of android to gynoid fat ratio determined by dual- Arch Pediatr Adolesc Med. 2009;163(9):826-831

HE RISING PREVALENCE OF pendently of total fat mass and subcutane- childhood obesity repre- ous abdominal adipose tissue.4 Conversely, sents an early risk factor for several studies have also shown that pe- the development of meta- ripheral subcutaneous fat may represent a bolic and cardiovascular “metabolic sink” for the storage of excess diseasesT in adults. Among obese children energy.5 When peripheral fat depots fail to and adolescents, there is also an in- expand to store excess energy intake, fat ac- creased number of cases of type 2 diabe- cumulates by default in liver, pancreas, and tes mellitus, which was once considered skeletal muscles, thereby inducing meta- as an adult-onset disease. bolic alterations leading to type 2 diabe- Since Vague,1 it has been well estab- tes.6 Indeed, there is some evidence show- Author Affiliations: Laboratory lished that the development of insulin re- ing that in individuals of similar waist of Exercise Biology (BAPS), sistance and the risk of cardiovascular dis- circumference, those with a high thigh fat Blaise Pascal University, Aubière eases are associated with excess body fat in accumulation have a lipoprotein-lipid pro- (Drs Aucouturier, Thivel, and abdominal rather than in peripheral fat de- file associated with lower risks of cardio- Duche´), Department of pots.2 Imaging techniques, such as mag- vascular diseases.7 In addition, a protec- Pediatrics, Hotel Dieu, University Hospital, netic resonance imaging and computed to- tive effect of peripheral fat depots may even Clermont-Ferrand (Dr Meyer), mography, have enabled differentiation be expected owing to the secretion of adi- and Children’s Medical Center, between subcutaneous and visceral fat. The ponectin, which enhances insulin sensitiv- Romagnat (Dr Taillardat), visceral fat area has been shown to be cor- ity and has antiatherogenic properties.8 In France. related with glucose intolerance3,4 inde- agreement with results in adults, obese ado-

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 163 (NO. 9), SEP 2009 WWW.ARCHPEDIATRICS.COM 826

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 lescents with a high visceral to subcutaneous fat ratio ex- BODY COMPOSITION hibit an impaired glucose tolerance in comparison with those with a low ratio.9 Moreover, there are several re- Body composition was determined by DXA scan (QDR 4500 ports in obese children and adolescents showing that ec- x-ray bone densimeter; Hologic, Waltham, Massachusetts) and topic fat storage and hypertriglyceridemic waist are asso- version 9.10 of total body scans software (Hologic Inc, Bed- ciated with declined insulin sensitivity.9,10 Weiss et al9 have ford, Massachusetts). Children were asked to lie down in a su- pine position on the DXA table and to stay still until the end of shown that the development of severe peripheral insulin the scanning procedure. They were also instructed to keep their resistance in obese children and adolescents with im- arms separated from their trunk and their legs separated from paired glucose tolerance was closely associated with in- one another. tramyocellular and intra-abdominal lipid accumulation. Percentage of abdominal fat was determined manually by A high intramyocellular lipid deposition has been shown an experienced experimenter by drawing a rectangular box to occur early during childhood and adolescence in asso- around the region of interest between vertebral bodies L1 ciation with peripheral insulin resistance.10,11 Most stud- and L4. The upper limit was set with the horizontal line ies investigating metabolic alterations in relation with re- going through the T2/L1 vertebral space and the lowest limit was set with a horizontal line going through the L4/L5 verte- gional fat deposition have focused on subcutaneous 18 abdominal, visceral, or thigh fat depots. Dual-energy x-ray bral space. Data were analyzed with Hologic QDR Software for Windows (version 12.6), which integrates whole-body absorptiometry (DXA) measurements have been used in measurement and standard body regions, such as the trunk, several studies to assess regional body fat distribution in arms, and legs, delineated by specific anatomical landmarks. 12-14 children and the association with cardiovascular risk Gynoid fat deposition was assessed by lower limb fat per- factors.13 Moreover, there is a good agreement between centage. Android to gynoid fat ratio was determined by magnetic resonance imaging measurements of abdomi- using fat percentage in lower limbs and in the abdominal nal adipose tissue and DXA fat measurements. Little at- region. To test the hypothesis that an android to gynoid fat tention has been paid to the association between gynoid ratio is associated with an impairment of insulin sensitivity, fat storage and insulin resistance in obese children.9,15 In study subjects were grouped into tertiles. We used tertiles to this study, we attempted to determine if the ratio be- ensure a number of subjects in each subgroup sufficient to tween abdominal, or android fat pattern, and lower limb give meaningful results. fat percentage, or gynoid fat pattern, determined by DXA scan was associated with the severity of insulin resistance assessed by the homeostasis model of insulin resistance BLOOD SAMPLES (HOMA-IR) index. We hypothesized that children with a high android to gynoid fat ratio would exhibit an in- Blood samples were drawn between 8 AM and 10 AM in a fasted state from an antecubital vein. Samples were centri- creased insulin resistance. fuged (at 4000g for 10 minutes at 4°C) and plasma was transferred into plastic tubes and kept at −80°C until analy- METHODS sis. The plasma glucose concentration was determined by enzymatic methods (Modular P900; Roche Diagnostics, Meylan, France). Plasma insulin concentration was assayed SUBJECTS by a chemiluminescent enzyme immunoassay on an Immu- lite 2000 (Diagnostic Products Corporation, Los Angeles, California). Participants in this study were 66 obese children and adoles- Two indexes of insulin resistance were calculated from glu- cents (31 girls and 35 boys) and their parents coming to the cose and insulin concentrations. The HOMA-IR index was cal- Department of Pediatrics, University Hospital, Clermont- culated as (fasting insulin levelϫfasting glucose level)/22.519 Ferrand, France, for medical consultation. Parents and chil- and quantitative insulin-sensitivity check index, as 1/(log fast- dren who agreed to take part to the study signed an ing insulin levelϩlog fasting glucose level).20 informed consent. The experimental protocol of this study was approved by the local ethics committee (Comite´ de Pro- tection des Personnes, Sud Est IV). Children included in this study were higher than the 95th percentile of body mass STATISTICAL ANALYSIS index (BMI) for age and sex defined by the International Obesity Task Force.16 Results are expressed as mean (SD). Normality of the distri- Medical examination and anthropometric measurements bution was checked with the Kolmogorov-Smirnov test for each were performed for each subject by a pediatrician. Body variable. Dependent variables were compared between the 3 mass was measured to the nearest 0.05 kg with a digital scale groups by using a 1-way analysis of variance. Android to gy- (model 873; Seca Omega, Hamburg, Germany). Height was noid fat ratio and abdominal fat percentage were similar be- measured with a standing stadiometer and recorded with a tween boys and girls in the 3 groups. Hence, boys and girls were precision of 1 mm. Body mass index was calculated as grouped together in each tertile. weight in kilograms divided by height in meters squared. Spearman correlation coefficients were used to describe as- Body mass index and waist circumference z scores were cal- sociations between continuous variables. We also used a mul- culated for age and sex reference values.16,17 Waist circumfer- tiple stepwise regression to explain the variance of HOMA-IR ence was measured in a standing position with a nonelastic values. Age, waist circumference z score, BMI, body fat per- tape that was applied horizontally midway between the cos- centage, and the android to gynoid fat ratio were included as tal arch and the iliac crest. All subjects were free of medica- independent variables. tion known to affect energy metabolism and none of the All statistical analyses were carried out with Statview soft- subjects had evidence of significant disease, non–insulin- ware, version 5.0 (Abacus Concepts, Berkeley, California). Sta- dependent diabetes mellitus, or other endocrine disease. tistical significance was set at PϽ.05.

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 163 (NO. 9), SEP 2009 WWW.ARCHPEDIATRICS.COM 827

Tertile, Mean (SD)

123P Value Sex, No. M 9 11 11 F 141212 Age, y 9.61 (2.70) 11.21 (2.27) 11.11 (2.37) .07 Weight, kg 57.00 (29.42) 65.33 (24.43) 61.65 (19.03) .31 BMI 24.46 (5.82) 27.94 (5.74) 27.23 (4.94) .09 BMI z score 1.91 (0.40) 2.05 (0.38) 1.94 (0.47) .51 Body fat, % 36.84 (5.38) 38.75 (6.34) 36.83 (4.48) .41 Waist circumference, cm 78.95 (16.50) 89.23 (17.15) 90.15 (12.73) .07 Waist circumference z score 1.94 (0.74) 2.35 (0.57) 2.55 (0.56) .05a Abdominal fat, % 30.36 (4.18) 37.24 (6.62) 40.43 (5.11) .001b Truncal fat mass, % 32.38 (5.53) 36.22 (5.94) 34.81 (5.27) .08 LBM, kg 33.37 (15.24) 37.86 (11.34) 40.81 (14.21) .21 Lower limb fat, % 43.82 (6.28) 43.63 (7.79) 40.00 (5.01) .09 Android to gynoid fat ratio 0.70 (0.07) 0.86 (0.05) 1.01 (0.04) .001c

Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); LBM, lean body mass. a Protected least significant difference Fisher test for comparison between groups: significant difference between tertile 1 and tertile 3, PϽ.05. b Protected least significant difference Fisher test for comparison between groups: significant difference between tertile 1 and tertile 2, PϽ.001; tertile 1 and tertile 3, PϽ.001; and tertile 2 and tertile 3, PϽ.05. c Protected least significant difference Fisher test for comparison between groups: significant difference between tertile 1 and tertile 2, PϽ.001; tertile 1 and tertile 3, PϽ.001; and tertile 2 and tertile 3, PϽ.001.

5.0 0.5

4.0 0.4

3.0 0.3 QUICKI HOMA-IR 2.0 0.2

1.0 0.1

0.0 0.0 Tertile 1 Tertile 2 Tertile 3 Tertile 1 Tertile 2 Tertile 3

Figure 1. Mean (SD) homeostasis model of insulin resistance (HOMA-IR) Figure 2. Mean (SD) quantitative insulin-sensitivity check index (QUICKI) index values in tertiles of android to gynoid fat ratio. *PϽ.001. values in tertiles of android to gynoid fat ratio. *PϽ.001.

RESULTS INDEXES OF INSULIN RESISTANCE: FASTING GLUCOSE AND INSULIN CONCENTRATIONS

DESCRIPTIVE STATISTICS Mean (SD) HOMA-IR values were significantly higher in OF THE SAMPLE tertiles 2 (2.73 [1.41]) and 3 (2.89 [1.28]) than in tertile 1 (1.67 [1.24]) (P=.009 and .003, respectively). Mean Descriptive results of the population are presented for (SD) quantitative insulin-sensitivity check index values boys and girls in Table 1. Body mass, percentage of were also significantly higher in tertile 1 (0.37 [0.04]) body fat, and lean body mass were similar in the 3 ter- than in tertiles 2 (0.34 [0.03]) and 3 (0.33 [0.02]) (P=.005 tiles. Tertiles were also similar for the number of boys and P=.001, respectively). Differences were not signifi- and girls. Because of the study design, the android to cant between tertiles 2 and 3. Results are shown in gynoid fat ratio was significantly different between the Figure 1 and Figure 2. 3 tertiles (PϽ.001 for all comparisons). There were Mean (SD) fasting plasma glucose level was not sig- significant differences for percentage of abdominal fat nificantly different between tertiles (tertile 1, 85.79 [4.30] between tertiles 1 and 2 (PϽ.001) and tertiles 1 and 3 mg/dL; tertile 2, 88.79 [6.59] mg/dL; tertile 3, 89.28 [7.50] (PϽ.001). There was no significant difference for per- mg/dL), whereas mean (SD) insulin concentration was centage of fat mass in lower limbs between tertiles. significantly higher in tertiles 2 (12.39 [6.12] mU/L) and

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 163 (NO. 9), SEP 2009 WWW.ARCHPEDIATRICS.COM 828

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 Table 2. Correlation Coefficients for Association Between Fat Distribution Variables and Markers of Insulin Resistance

HOMA-IR QUICKI Plasma Glucose Level Plasma Insulin Level Abdominal fat, % 0.34a −0.36a 0.12 0.34a Lower limb fat, % 0.04 0.01 0.08 0.05 Android to gynoid fat ratio 0.35a −0.41a 0.20 0.34a BMI 0.45b 0.46a 0.13 0.47a BMI z score 0.22 0.22 0.24 0.22 Waist circumference 0.44a 0.52c 0.06 0.48c Waist circumference z score 0.33b 0.45a 0.11 0.35a Body fat, % 0.20 −0.19 0.00 0.22

Abbreviation: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HOMA-IR, homeostasis model of insulin resistance index; QUICKI, quantitative insulin-sensitivity check index. a PϽ.01. b PϽ.05. c PϽ.001.

3 (13.04 [5.20] mU/L) than in tertile 1 (7.91 [5.68] mU/L) quantitative insulin-sensitivity check index calculated (P=.01 and P=.004, respectively). Differences were not from fasting samples have been shown to be valid to as- significant between tertiles 2 and 3 (P=.70). sess insulin resistance during puberty when compared with direct measurement with a glucose clamp.21,22 The CORRELATION COEFFICIENT main finding was that insulin resistance was increased in children with central rather peripheral fat depots in Relationships between fat distribution variables and in- groups matched for body mass and percentage of body sulin sensitivity variables are shown in Table 2. Per- fat. Furthermore, insulin resistance was associated with centage of abdominal fat correlated positively with abdominal adiposity without distinction between sub- HOMA-IR value (r=0.34; PϽ.01). The android to gy- cutaneous and visceral fat depots. However, although noid fat ratio was positively correlated with HOMA-IR HOMA-IR values increased from the lowest tertile to ter- value (r=0.35; PϽ.01). Android to gynoid fat ratio was tiles 2 and 3, whereas there was no significant differ- also significantly and positively correlated with fasting ence between tertiles 2 and 3, a linear regression be- insulin concentration (r=0.34; PϽ.01). Waist circum- tween the android to gynoid fat ratio and HOMA-IR value ference and waist circumference z score were signifi- did not provide a threshold value of android to gynoid cantly correlated with HOMA-IR value (r=0.44; PϽ.001 fat ratio above which obese children have an increased and r=0.33; PϽ.01). Body mass index was correlated with risk of insulin resistance. Ͻ HOMA-IR value (r=0.45; P .001) but not BMI z score. Indeed, in the present study, there was no signifi- Neither body fat percentage nor lower limbs fat percent- cant association between percentage of body fat and age were significantly correlated with insulin sensitivity insulin resistance. Previous studies have shown in variables or glucose and insulin concentrations. None of young subjects that the degree of obesity is associated the fat distribution variables had significant correlation with a worsening of all the components of the meta- with fasting glucose concentration. bolic syndrome, including insulin resistance.23 Several points can explain the lack of correlations between MULTIPLE STEPWISE REGRESSION percentage of body fat and indexes of insulin resistance in the present study. Despite a similar degree of The multiple stepwise regression showed that age and obesity, a lower prevalence of impaired glucose toler- the android to gynoid fat ratio were significant predic- ance and type 2 diabetes have been reported in Euro- tors of HOMA-IR value (␤ coefficients were 0.26 and 2.28, 11,24 2 pean than in American children. Indeed, even respectively). Adjusted R was 0.30. Body mass index, though impaired fasting glucose concentration may waist circumference z score, and body fat percentage were not be sensitive enough to detect impaired glucose tol- not significant predictors of HOMA-IR value. erance,25 only 2 children had a fasting glucose concentration higher than 100 mg/dL. Hence, together with a COMMENT reduced number of subjects with severe obesity in comparison with other studies, only mild alterations Our hypothesis was that a preferential fat storage at the of insulin sensitivity may explain the lack of associa- abdominal level rather than in the lower limbs would be tion between percentage of body fat and insulin resis- associated with increased insulin resistance. To this aim, tance. The development of abdominal obesity during we calculated a simple index of android to gynoid fat dis- puberty may be favored by pubertal insulin resistance tribution as a ratio between percentage of abdominal fat and its consequent hyperinsulinemia.15 A limitation of and percentage of lower limbs fat based on DXA mea- this study is that data analysis was based on age rang- surements. Insulin resistance was estimated by using ing between 6 and 17 years and not on direct assess- simple indexes based on fasting plasma glucose and in- ment of pubertal stages. Logically, age was a signifi- sulin concentrations. Indexes such as HOMA-IR and the cant predictor of insulin resistance. Moreover, the

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 163 (NO. 9), SEP 2009 WWW.ARCHPEDIATRICS.COM 829

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 effect of puberty was partly controlled by the use of Accepted for Publication: March 4, 2009. age- and sex-specific BMI and waist circumference Correspondence: Pascale Duche´, PhD, Laboratory of growth charts. Exercise Biology (BAPS), Blaise Pascal University, Several studies have already used DXA to provide mea- Baˆtiment de Biologie B, Complexe Universitaire des surements of abdominal fat mass.18,26 Measurement by Ce´zeaux, 63177 Aubière CEDEX, France (pascale.duche DXA of central adiposity from L1 to L4 has previously @univ-bpclermont.fr). been shown to be associated with insulin resistance in Author Contributions: Study concept and design: Aucou- adults and to be a valid alternative to other tech- turier, Meyer, and Duche´. Acquisition of data: Aucoutu- niques.26 However, a limitation of the present study is rier, Thivel, and Taillardat. Analysis and interpretation of that abdominal fat determined by DXA does not allow data: Aucouturier, Meyer, Thivel, and Duche´. Drafting the distinction between visceral and subcutaneous ab- of the manuscript: Aucouturier. Critical revision of the dominal tissues. Bacha et al27 observed that in 2 groups manuscript for important intellectual content: Aucoutu- of obese adolescents with a similar percentage of body rier, Meyer, Thivel, Taillardat, and Duche´.Statistical analy- fat (42.5 and 44.2) those who had the lowest visceral fat sis: Aucouturier, Thivel, Taillardat, and Duche´. Admin- area and the lowest subcutaneous fat area at L4-L5 ex- istrative, technical, and material support: Thivel and hibited only a moderate insulin resistance. On the other Taillardat. Study supervision: Aucouturier, Meyer, and hand, Maffeis et al28 recently showed that visceral adi- Duche´. pose tissue area was not associated with insulin sensi- Financial Disclosure: None reported. tivity, but may rather alter insulin sensitivity through its effect on liver fat content, which explained 16% of the REFERENCES variation in insulin sensitivity in obese children. Hence, questions remain about the importance of visceral fat for 1. Vague J. Diffe´renciation sexuelle, facteur de´terminant des formes de l’obe´site´. the development of insulin resistance. Finally, signifi- Presse Med. 1947;55:339-340. cant correlations between waist circumference or waist 2. Despre´s JP. Cardiovascular disease under the influence of excess visceral fat. circumference z score and HOMA-IR confirm that simple Crit Pathw Cardiol. 2007;6(2):51-59. 3. Fujioka S, Matsuzawa Y, Tokunaga K, Tarui S. Contribution of intra-abdominal anthropometric measurements are also reliable to as- fat accumulation to the impairment of glucose and lipid metabolism in human sess an association between upper body adiposity and in- obesity. Metabolism. 1987;36(1):54-59. sulin resistance.29 4. Despre´s JP, Nadeau A, Tremblay A, et al. Role of deep abdominal fat in the as- We did not observe any association between lower sociation between regional adipose tissue distribution and glucose tolerance in obese women. Diabetes. 1989;38(3):304-309. body fat percentage and insulin resistance. This result 5. Okura T, Nakata Y, Yamabuki K, Tanaka K. Regional body composition changes 30,31 is similar to previous findings in adults. Although exhibit opposing effects on coronary heart disease risk factors. Arterioscler Thromb subcutaneous adipose tissue stores approximately 90% Vasc Biol. 2004;24(5):923-929. of thigh fat, it is other compartments quantitatively 6. Danforth E Jr. Failure of adipocyte differentiation causes type II diabetes mellitus? Nat Genet. 2000;26(1):13. minor for fat storage, such as intramuscular triglycer- 7. Terry RB, Stefanick ML, Haskell WL, Wood PD. Contributions of regional adi- ides, that are significantly associated with insulin pose tissue depots to plasma lipoprotein concentrations in overweight men and resistance.31 In obese children, both intramyocellular women: possible protective effects of thigh fat. Metabolism. 1991;40(7):733- and extramyocellular triglycerides stores are associ- 740. 8. Berg AH, Combs TP, Scherer PE. ACRP30/adiponectin: an adipokine regulat- ated with central adiposity, suggesting that deposition ing glucose and lipid metabolism. Trends Endocrinol Metab. 2002;13(2):84- 10 of fat in these tissues is interdependent. Meanwhile, 89. an important leg fat storage, almost at the subcutane- 9. Weiss R, Dufour S, Taksali SE, et al. Prediabetes in obese youth: a syndrome of ous level, reflects the ability of this depot to be a meta- impaired glucose tolerance, severe insulin resistance, and altered myocellular bolic sink for excess energy intake, thereby preventing and abdominal fat partitioning. Lancet. 2003;362(9388):951-957. 10. Sinha R, Dufour S, Petersen KF, et al. Assessment of skeletal muscle triglycer- 32 ectopic accumulation. Together, android to gynoid ide content by (1)H nuclear magnetic resonance spectroscopy in lean and obese ratio and age explained 30% of the variance of insulin adolescents: relationships to insulin sensitivity, total body fat, and central adiposity. resistance. Fitness level, which was not assessed in the Diabetes. 2002;51(4):1022-1027. present study, has important effects on indexes of 11. Weiss R, Caprio S. The metabolic consequences of childhood obesity. Best Pract 33 Res Clin Endocrinol Metab. 2005;19(3):405-419. insulin sensitivity even in obese children and may be 12. Dencker M, Thorsson O, Linden C, Wollmer P, Andersen LB, Karlsson MK. a factor that could also explain an important part of BMI and objectively measured body fat and body fat distribution in prepubertal variability of insulin resistance in our population. children. Clin Physiol Funct Imaging. 2007;27(1):12-16. To conclude, the present study showed that an an- 13. Daniels SR, Morrison JA, Sprecher DL, Khoury P, Kimball TR. Association of body fat distribution and cardiovascular risk factors in children and adolescents. droid rather than gynoid fat distribution was associated Circulation. 1999;99(4):541-545. with an increased insulin resistance in obese children and 14. Novotny R, Going S, Teegarden D, et al; ACT Research Team. Hispanic and Asian adolescents. Hence, an android to gynoid fat ratio based pubertal girls have higher android/gynoid fat ratio than whites. Obesity (Silver on DXA measurement may be a useful and simple tech- Spring). 2007;15(6):1565-1570. 15. Caprio S, Hyman LD, McCarthy S, Lange R, Bronson M, Tamborlane WV. Fat nique to assess a pattern of body fat distribution associ- distribution and cardiovascular risk factors in obese adolescent girls: impor- ated with an increased insulin resistance. This study also tance of the intraabdominal fat depot. Am J Clin Nutr. 1996;64(1):12-17. confirmed that the severity of insulin resistance is asso- 16. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for ciated with abdominal obesity, which can be assessed by child overweight and obesity worldwide: international survey. BMJ. 2000;320 (7244):1240-1243. waist circumference measurement, whether fat is lo- 17. Eisenmann JC. Waist circumference percentiles for 7- to 15-year-old Australian cated essentially in visceral or subcutaneous adipose tis- children. Acta Paediatr. 2005;94(9):1182-1185. sue in children and adolescents. 18. Glickman SG, Marn CS, Supiano MA, Dengel DR. Validity and reliability of dual-

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 163 (NO. 9), SEP 2009 WWW.ARCHPEDIATRICS.COM 830

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 energy X-ray absorptiometry for the assessment of abdominal adiposity. J Appl ment of fat mass distribution and its association with the insulin resistance Physiol. 2004;97(2):509-514. syndrome. Diabetes Care. 1999;22(8):1310-1317. 19. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. 27. Bacha F, Saad R, Gungor N, Arslanian SA. Are obesity-related metabolic risk fac- Homeostasis model assessment: insulin resistance and beta-cell function from tors modulated by the degree of insulin resistance in adolescents? Diabetes Care. fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985; 2006;29(7):1599-1604. 28(7):412-419. 28. Maffeis C, Manfredi R, Trombetta M, et al. Insulin sensitivity is correlated with 20. Katz A, Nambi SS, Mather K, et al. Quantitative insulin sensitivity check index: a subcutaneous but not visceral body fat in overweight and obese prepubertal simple, accurate method for assessing insulin sensitivity in humans. J Clin En- children. J Clin Endocrinol Metab. 2008;93(6):2122-2128. docrinol Metab. 2000;85(7):2402-2410. 29. Taylor RW, Jones IE, Williams SM, Goulding A. Evaluation of waist circumfer- 21. Uwaifo GI, Fallon EM, Chin J, Elberg J, Parikh SJ, Yanovski JA. Indices of insulin ence, waist-to-hip ratio, and the conicity index as screening tools for high trunk action, disposal, and secretion derived from fasting samples and clamps in nor- fat mass, as measured by dual-energy X-ray absorptiometry, in children aged mal glucose-tolerant black and white children. Diabetes Care. 2002;25(11): 3-19 y. Am J Clin Nutr. 2000;72(2):490-495. 2081-2087. 30. Carey DG, Jenkins AB, Campbell LV, Freund J, Chisholm DJ. Abdominal fat and 22. Conwell LS, Trost SG, Brown WJ, Batch JA. Indexes of insulin resistance and insulin resistance in normal and overweight women: direct measurements re- secretion in obese children and adolescents: a validation study. Diabetes Care. 2004;27(2):314-319. veal a strong relationship in subjects at both low and high risk of NIDDM. Diabetes. 23. Weiss R, Dziura J, Burgert TS, et al. Obesity and the metabolic syndrome in chil- 1996;45(5):633-638. dren and adolescents. N Engl J Med. 2004;350(23):2362-2374. 31. Goodpaster BH, Thaete FL, Simoneau JA, Kelley DE. Subcutaneous abdominal 24. Invitti C, Guzzaloni G, Gilardini L, Morabito F, Viberti G. Prevalence and con- fat and thigh muscle composition predict insulin sensitivity independently of vis- comitants of glucose intolerance in European obese children and adolescents. ceral fat. Diabetes. 1997;46(10):1579-1585. Diabetes Care. 2003;26(1):118-124. 32. Lemieux I. Energy partitioning in gluteal-femoral fat: does the metabolic fate of 25. Go´mez-Dı´az R, Aguilar-Salinas CA, Moran-Villota S, et al. Lack of agreement be- triglycerides affect coronary heart disease risk? Arterioscler Thromb Vasc Biol. tween the revised criteria of impaired fasting glucose and impaired glucose tol- 2004;24(5):795-797. erance in children with excess body weight. Diabetes Care. 2004;27(9):2229- 33. Eisenmann JC, DuBose KD, Donnelly JE. Fatness, fitness, and insulin sensitivity 2233. among 7- to 9-year-old children. Obesity (Silver Spring). 2007;15(8):2135- 26. Paradisi G, Smith L, Burtner C, et al. Dual energy X-ray absorptiometry assess- 2144.

Call for Papers

Archives of Pediatrics and Adolescent Medicine will de- vote their May 2010 issue to papers on the effects of life experiences occurring during the critical window of birth to age 5 years on the emotional and psychological health and development—or ill health—of children both during that age and at later ages during childhood and adolescence. Papers submitted by September 30, 2009, have the best chance of acceptance.

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 163 (NO. 9), SEP 2009 WWW.ARCHPEDIATRICS.COM 831