Development of the Metabolic Syndrome in Black and White Adolescent
Girls: A Longitudinal Assessment
John A. Morrison, PhD*; Lisa Aronson Friedman, ScM‡; William R. Harlan, MD§; Linda C. Harlan, PhD㛳; Bruce A. Barton, PhD‡; George B. Schreiber, DSc¶; and David J. Klein, MD, PhD#
ABSTRACT. Background. The metabolic syndrome, associated with increased risk of type 2 diabetes mellitus and cardiovascular disease, begins to develop during adolescence.
Objective. We sought to identify early predictors of the presence of the syndrome at the ages of 18 and 19 years in black and white girls.
Methods. Using longitudinal data on participants from 2 centers in the National Heart, Lung, and Blood Institute Growth and Health Study, a 10-year cohort study, we applied cutoffs from the Adult Treatment Panel III to document changes in the prevalence of ab-normal syndrome elements and the syndrome in girls aged 9 and 10 years, when cases were rare, and those aged 18 and 19 years, when prevalence had reached 3%. Lon-gitudinal regression models identified early predictors for the presence of the syndrome.
Results. Only 1 girl of each race had>3 factors at ages 9 and 10 (0.2%), but 20 black girls (3.5%) and 12 white girls (2.3%) had the syndrome 10 years later. Low high-density lipoprotein cholesterol was prevalent throughout the period in both black and white girls. The prevalence of other variables was low at enrollment but increased during follow-up, except for abnormal triglyceride levels in black girls, which remained low throughout follow-up. In multivariate models, early measures of waist cir-cumference and triglyceride level were significant pre-dictors for development of the syndrome.
Conclusion. The strong association of central adipos-ity with the development of the metabolic syndrome suggests that early interventions aimed at managing pre-teen obesity could reduce risk of developing the syndrome.Pediatrics 2005;116:1178–1182;metabolic syn-drome, adolescence, overweight, insulin resistance, ethnic-ity.
ABBREVIATIONS. ATP III, Adult Treatment Panel III; HDL-C, high-density lipoprotein cholesterol; NHANES, National Health and Nutrition Examination Survey; NGHS, National Heart, Lung, and Blood Institute Growth and Health Study; HOMA-IR, ho-meostasis model assessment, insulin resistance.
T
he metabolic syndrome, a constellation of fac-tors including abdominal obesity, high blood pressure, dyslipidemias, and high fasting glu-cose levels, confers increased risk for diabetes melli-tus and cardiovascular disease.1 The Adult Treat-ment Panel III (ATP III) defined the syndrome as the conjoint presence ofⱖ3 of the following factors: (1) a waist circumference ⱖ102 cm in men and 88 cm in women; (2) serum triglyceride level ⱖ 150 mg/dL; (3) high-density lipoprotein cholesterol (HDL-C) level of ⱕ40 mg/dL in men and ⱕ50 mg/dL in women; (4) blood pressureⱖ130/85 mm Hg; and (5) fasting glucose level ⱖ 110 mg/dL.2 None of the cutoffs used to define these factors is extreme, rep-resenting instead the 80th to 85th percentiles of the factors as ascertained in epidemiologic surveys.3,4Ford et al4reported that⬃22% of adults have the syndrome based on data from the representative sample of the US population in the National Health and Nutrition Examination Survey (NHANES) III. Other reports identified important roles for obesity and insulin resistance in the development of the syndrome.5–7Cook et al8reported on the prevalence of the syndrome phenotype in adolescents 12 to 19 years of age by using NHANES III data and redefin-ing the cutoffs based on age-specific percentiles. This analytic strategy provided a sufficient number of cases to compare the prevalence across gender-race-ethnic groups but could not describe changes in the prevalence of the syndrome during adolescence be-cause adjusted cutoffs were used. The Bogalusa Heart Study had begun before Kissebah et al9 iden-tified central adiposity as a key predictor of meta-bolic disorders, but recent analyses of data from multiple Bogalusa surveys between 1978 and 1996, using BMI instead of waist circumference, showed that the clustered presence in childhood of top-quar-tile values for BMI, blood pressure, insulin level, and triglyceride level or low HDL-C levels predicted their clustered presence in young adulthood, ⬃11 years later.10 BMI and insulin were the strongest predictors in multiple logistic models, with BMI be-ing the stronger of these 2.10
The dramatic rise in the prevalence of obesity in the years after NHANES II (1976 –1980) and the at-tendant increase in the incidence of diabetes11 sug-gest that a prospective evaluation of the appearance of the syndrome during adolescence could provide insights into its pathogenesis and identify potential interventions. The National Heart, Lung, and Blood
From the Department of Pediatrics, University of Cincinnati Medical Center and the Divisions of *Cardiology and #Endocrinology, Cincinnati Chil-dren’s Hospital Medical Center, Cincinnati, Ohio; ‡Maryland Medical Re-search Institute, Baltimore, Maryland; §National Institute of Mental Health, 㛳National Cancer Institute, Bethesda, Maryland; and ¶Westat, Rockville, Maryland.
Accepted for publication Jan 24, 2005. doi:10.1542/peds.2004-2358 No conflict of interest declared.
Address correspondence to John A. Morrison, PhD, Division of Cardiology, Children’s Hospital Medical Center, OSB 4, 3333 Burnet Ave, Cincinnati, OH 45229. E-mail: john.morrison@cchmc.org
Institute Growth and Health Study12(NGHS), a 10-year longitudinal study of adolescent obesity in black and white girls, provides an opportunity for such a prospective evaluation. Two NGHS clinical centers obtained fasting insulin and glucose mea-sures at baseline and year 10 in addition to NGHS measures of anthropometry, lipids, and blood pres-sure.12The purpose of this report is to identify early predictors of the syndrome’s presence at ages 18 and 19 years. We hypothesized that early childhood obe-sity and fat distribution would predict the presence of the syndrome in early adulthood.
METHODS
The NGHS has been described in detail previously.12Briefly,
the study recruited 9- and 10-year-old girls in 1987–1988 who identified themselves as black or white, whose parents or guard-ians identified themselves as black or white, and who lived in racially concordant households. The Cincinnati, Ohio, and Wash-ington, DC, clinics collected blood for the measurement of insulin and glucose at baseline and year 10 as part of an ancillary study. The institutional review boards at the 2 clinical sites approved the study, and signed, informed consent was obtained from the girls’ parents or guardians.
Clinical and Laboratory Measures
Trained, certified staff took all clinical measures according to standard protocols.12At annual visits, 2 measurements of height,
weight, and, starting in year 2, the minimum waist circumference were made. A third measurement was taken if the first 2 differed by a preset amount. The mean of all measures was used in anal-yses. The BMI was used to evaluate overweight. Three blood pressure measurements were made at each visit by using a mer-cury sphygmomanometer and an appropriate-size cuff based on arm circumference.12The mean of the second and third
measure-ments of systolic or K-5 diastolic blood pressure was used in these analyses. Fasting HDL-C and triglyceride levels were measured by the Central Lipid Laboratory at Johns Hopkins University Medical Center (Baltimore, MD) at baseline and year 10,12In years
1 and 10, plasma was obtained for insulin and glucose levels as reported previously.13Homeostasis model assessment
(HOMA-IR) was used as an index of insulin resistance.14
Statistical Analyses
The prevalence of abnormal levels of each component of the metabolic syndrome and the syndrome itself was calculated as the percent of participants with risk-factor values outside ATP III cutoffs. The use of age-specific 10th and 90th percentiles as cutoffs results in ⬃10% of participants with abnormal levels at both
baseline and ages 18 and 19 years by design and, thus, could not demonstrate the emergence of the metabolic syndrome from being extremely rare or nonexistent to being a relatively common find-ing. When the ATP III cutoffs were used at year 10 only, the prevalence of high blood pressure and high triglyceride level was actually lower than at baseline by age-specific cutoffs. Because use of the ATP III cutoffs is appropriate at ages 18 and 19 years, these cutoffs were used at both visits, thus retaining a constant metric across the period and showing the appearance and development of the syndrome. General estimating equations were used to eval-uate changes in the prevalence of individual syndrome compo-nents over time with race as an explanatory variable. Logistic-regression models were used to calculate the risk (odds ratios) of having the syndrome at ages 18 and 19 years that was associated with the presence of the elements that define the syndrome (waist circumference, glucose, triglycerides, HDL-C, and blood pressure) early in the study and with indices of obesity and insulin resis-tance, BMI, and HOMA-IR. All statistical analyses were per-formed with SAS 9.1.15
RESULTS
Table 1 presents baseline and year-10 descriptive data by race for the analysis cohort, including age, BMI, insulin, HOMA-IR, and the ATP III factors that define the metabolic syndrome. Black girls at base-line were slightly older (2 months) and had greater mean BMI, higher insulin, and higher HOMA-IR (all P⬍.0001) and higher HDL-C (P⬍ .01) and glucose (P ⫽.02) levels than white girls. As reported previ-ously, multivariate analyses indicated that the differ-ences in age and pubertal status did not explain the racial differences in obesity or insulin sensitivity.13 Black girls had higher systolic blood pressure (P ⫽ .01) and significantly lower triglyceride levels (P ⬍ .0001) at baseline. The measurement of waist circum-ference was begun in year 2; black girls had signifi-cantly larger mean waist circumference than white girls (P⬍.0001).
As expected, there were significant increases in the means for BMI and the metabolic syndrome elements during follow-up, except for triglycerides, in black girls. The increases were 42.4% and 34.6%, respec-tively, for BMI in black and white girls and 20.3% and 18.8%, respectively, for waist circumference, thus amplifying baseline ethnic differences. Consis-tent with the increases in BMI and waist circumfer-ence, there were increases in mean systolic and
dia-TABLE 1. Age, BMI, Insulin, Glucose, HOMA-IR, Triglycerides, HDL-C, Systolic and Diastolic Blood Pressure, and Waist Circum-ference in Black and White Girls at Ages 9 and 10 and 18 and 19 Years According to Race and Age
White Girls Black Girls
Ages 9 and 10 y Ages 18 and 19 y Ages 9 and 10 y Ages 18 and 19 y
N Mean SD N Mean SD N Mean SD N Mean SD
Age, y 608 10.0 0.6 511 19.0 0.7 584 10.1 0.5 567 19.2 0.6
BMI 608 17.6 3.2 510 23.7 5.2 583 19.3 4.4 564 27.5 7.8
Insulin,U/mL 491 9.8 6.8 412 9.0 7.0 489 14.7 9.6 469 13.0 10.3 Glucose, mg/dL 419 93.0 6.1 458 86.3 7.8 492 94.0 6.8 504 90.1 24.1
HOMA-IR 416 2.3 1.7 398 1.9 1.6 478 3.5 2.4 451 3.0 2.6
Triglycerides, mg/dL 568 80.8 38.9 346 94.1 50.1 517 72.1 30.3 440 72.1 34.4 HDL-C, mg/dL 554 52.7 11.4 345 50.8 10.5 515 54.7 13.1 440 52.9 11.3 Blood pressure
Systolic, mm Hg 606 100.6 8.9 510 107.1 8.2 579 101.8 8.8 561 110.1 9.4 Diastolic, mm Hg 585 57.1 11.5 510 65.7 8.6 556 57.8 12.1 561 67.3 9.4 Waist circumference, cm* 579 62.6 8.0 507 74.2 10.9 570 66.7 10.0 564 80.3 15.1 All comparisons between black and white girls at baseline and year 10 are significant atP⬍.0001 except glucose (P⬍.03关year 1兴and
P⬍.01关year 10兴), HDL-C (P⬍.01 at both years), systolic blood pressure (P⬍.01 at baseline), and diastolic blood pressure (nonsignificant at baseline andP⬍.03 at year 10).
stolic blood pressure in both groups, but they were greater in black girls than white girls (systolic: P ⬍ .001; diastolic:P⫽.09) (Table 1). There was a signif-icant increase in the mean triglyceride level in white girls (P⬍.001) but not in black girls. Decreases in the mean HDL-C level did not differ between groups. With completion of pubertal development, mean in-sulin level decreased in both groups; mean glucose level also decreased in both groups, but the decrease was greater in white girls (P ⫽ .02). Although HOMA-IR decreased in black girls and increased slightly in white girls, the absolute value remained lower in the white girls, indicating increased insulin resistance in black girls. Thus, at ages 18 and 19 years, black girls had significantly greater BMI, waist circumference, insulin level, HOMA-IR, and systolic blood pressure and lower triglyceride level (allP ⬍ .0001) and higher glucose level (P⫽ .001) (Table 1). HDL-C level and diastolic blood pressure were also higher in black girls (P ⫽ .006). In addition to the changes in mean values during adolescence, there were significant increases in the dispersion of values for BMI and waist circumference in both groups and for triglyceride level in white girls, as shown by significantly larger variances (all P ⬍ .01). Thus, a larger percentage of girls had extreme (elevated) val-ues for waist circumference and BMI and (white girls only) for triglycerides.
Table 2 presents the prevalence of the metabolic syndrome and its constituent factors at ages 9 and 10 and 18 and 19 years. As expected, the prevalence of the syndrome was low at baseline; at ages 9 and 10 years, only 1 girl (0.2%) in each racial group hadⱖ3 abnormal factors. The prevalence of abnormal factors was also low except for HDL-C, which had a preva-lence of 37.3% in black girls and 40.6% in white girls. As a consequence of the increases in the mean and variances of anthropometric variables and the con-stituent factors discussed above, the prevalence of a large waist circumference and high blood pressure increased in both black and white girls (allP⬍.01). The prevalence of low HDL-C levels, already high in
both groups at baseline, increased further. The prev-alence of elevated triglyceride levels increased in white girls only (P ⬍ .01). The prevalence of an impaired fasting glucose level (ⱖ110 mg/dL) in-creased little in either group. As a consequence of these changes, the prevalence of the syndrome at year 10 was 3.5% in black girls and 2.4% in white girls. Prevalence estimates for the syndrome were recalculated by using the new American Diabetic Association definition of an impaired fasting glucose level (ⱖ100 mg/dL), which increased the prevalence of the syndrome from 20 to 21 (3.6%) in black girls and from 12 to 15 (3%) in white girls.
Early measures of the elements that define the syndrome plus BMI and HOMA-IR were all signifi-cant predictors of the development of the syndrome at ages 18 and 19 years in univariate logistic-regres-sion models except for diastolic blood pressure (data not shown). When all these predictors were included in 1 multivariate model, multicollinearity was an issue. Therefore, separate models with BMI or waist circumference, race, and the other elements of the syndrome, 1 at a time, were used. As shown in Table 3, results indicated that BMI was significant in all models except when waist circumference was in-cluded, that waist circumference (P⬍.001) was sig-nificant in all models, that baseline triglyceride level was significant (P ⬍ .002) with either waist circum-ference or BMI in the model, and that HDL-C level was of borderline significance with BMI but not waist circumference. No other variable was signifi-cant when either BMI or waist circumference was in the model. For every increase of 1 cm in waist cir-cumference at year 2, the risk of developing the syndrome increased 7.4%, and for every increase of 1 mg/dL in triglyceride level at baseline, the risk in-creased 1.3%. Race was not significant in any model. To clarify the relationship between central adipos-ity in year 2 and the metabolic syndrome at ages 18 and 19 years, we examined the degree to which girls maintained their relative rank for waist circumfer-ence (ie, the degree to which waist circumfercircumfer-ence
TABLE 2. Prevalence of Metabolic Syndrome and Its Component Factors in Black and White Girls According to Visit (Age) Components of Metabolic Syndrome Black Girls Year 1
(Ages 9 and 10 y)
Black Girls Year 10 (Ages 18 and 19 y)
White Girls Year 1 (Ages 9 and 10 y)
White Girls Year 10 (Ages 18 and 19 y)
N* %† N % N % N %
Glucoseⱖ110 mg/dL‡ 11 2.2 15 3.0 4 1.0 6 1.3
SBP⬎130 mm Hg or DBP⬎85 mm Hg‡§ 1 0.2 23 4.1 2 0.3 8 1.6
Triglycerides⬎150 mg/dL储¶ 17 3.3 13 3.0 24 4.2 34 9.8
HDL-C⬍50 mg/dL§¶ 192 37.3 185 42.1 225 40.6 184 53.0
Waist circumference⬎88 cm** NA NA 145 25.7 NA NA 60 11.8
Presence of metabolic syndrome (ⱖ3 factors)§†† 1 0.2 20 3.5 1 0.2 12 2.4 SBP indicates systolic blood pressure; DBP, diastolic blood pressure.
* Number of subjects with an abnormal factor.
† Percent of all participants having that factor measured with an abnormal value. Denominators vary because of missed visits or missing data.
‡ Comparison of races adjusted for visit:P⬍.05. § Comparison of visits adjusted for race:P⬍.0001.
储Comparison of visits adjusted for race:P⬍.05. ¶ Comparison of races adjusted for visit:P⬍.01.
tracked from year 2 to 10). Adjusting for pubertal stage and race, the tracking coefficient was 0.83, in-dicating a very strong tendency for girls who had high central obesity in year 2 to continue having high central obesity through adolescence. Because a waist circumference of 72 cm in year 2 represented the same percentile in the distribution as a waist circum-ference of 88 cm in year 10, we compared the prev-alence of the syndrome in girls with waist circum-ferences above these cutoffs at both years 2 and 10 and at year 2 only. There were 233 girls with a waist circumferenceⱖ72 cm in year 2, 167 of which had a waist circumference ⱖ 88 cm in year 10 and 66 of which did not. The prevalence of the metabolic syn-drome in the 167 girls was 12.1% compared with 0% in the girls who had increased waist circumference at year 2 but not at year 10.
DISCUSSION
We report here the first large-scale, longitudinal assessment of the development of the metabolic syn-drome during adolescence in girls by using the ATP III definition of the syndrome, a strategy that high-lights the increase in prevalence during adolescence and factors related to it.
Although there was already 1 case in each group at baseline, we have shown the emergence of the syn-drome during adolescence from (virtually) zero to a combined prevalence of 3% at young adulthood (3.4%, using the new American Diabetic Association definition of high glucose level). We have also iden-tified the factors at years 1 and 2 that predict its presence 10 years later. The marked increase in the prevalence of the syndrome reflects increases in syn-drome variables resulting (in part) from adolescent growth, but the increases in the dispersion of values for syndrome elements (larger variances) suggest that there is a subset of girls in whom the increases exceed normal growth.
Although the high prevalence of low HDL-C level in this sample is consistent with the NHANES III data in adult black and white women (34.0% and 39.3%, respectively4), the findings from this and the other studies raise questions about the appropriate-ness of this cutoff. Although the HDL-C cutoff of 40 mg/dL used in males is⬃25th percentile, the cutoff
of 50 mg/dL in women is closer to the 50th percentile and may identify too many cases of low HDL-C level. This issue needs to be reconsidered by the expert panels.
There were striking racial differences in the prev-alence of individual components of the syndrome. The adolescent increases in body size and central adiposity were greater in black than white girls, as were the associated increases in systolic and diastolic blood pressure. At the same time, the increase in insulin sensitivity, reflected in lower glucose levels and HOMA-IR scores, which normally occurs with completion of pubertal development,16,17 reduced the risk of the syndrome due to this factor. Finally, there was a striking increase in abnormal triglyceride levels in white girls but not in black girls, despite the greater increases in BMI and waist circumference in black girls and the strong association between obe-sity and triglyceride levels in both black and white girls.18This difference is part of the continuing puz-zle of racial differences in triglyceride metabolism. It could reflect more subcutaneous adiposity and less visceral adiposity in black girls19or a greater preva-lence of the polymorphism of the hepatic lipase gene associated with low triglyceride levels, as reported by Vega et al,20or both. Consequently, the contribu-tion of elevated triglyceride levels to the presence of the syndrome is less in black girls. Reflecting all the changes in these individual factors, the prevalence of the syndrome increased in both races to 3.5% in black girls and 2.4% in white girls at ages 18 and 19 years. These observed prevalence levels are consonant with the percentages of affected females reported by Ford et al4and represent levels midway between the esti-mates for ages 9 and 10 years and for ages 20 to 29 years (6%).
The results of this study underscore the impor-tance of early central adiposity in the development of the metabolic syndrome. BMI can serve as a surro-gate variable if waist circumference is not available, as shown by its significant relation to the syndrome in multivariate models except when waist circumfer-ence was included. The data show that the preva-lence of the syndrome, which was 3% overall at ages 18 and 19 years, was 12.1% in the girls with persis-tently increased central adiposity and 0% in the girls TABLE 3. Increased Risk of Metabolic Syndrome at Ages 18 and 19 Years From Multivariate Logistic-Regression Analysis Using Early Predictors, With BMI and Waist Circumference in the Models
With Race and BMI in Models With Race and Waist Circumferencein Models
Odds Ratio 95% Confidence Interval P Odds Ratio 95% Confidence Interval P
Predictor
BMI at year 1* 0.86 0.71–1.05 .14 0.86 0.71–1.05 .14
Waist circumference at year 2† 1.16 1.07–1.27 ⬍.001 1.16 1.07–1.27 ⬍.001
HOMA-IR at year 1 0.96 0.77–1.21 NS 0.91 0.72–1.14 NS
Triglyceride level at year 1 1.12 1.004–1.25 ⬍.002 1.01 1.005–1.024 ⬍.002 HDL-C level at year 1 0.96 0.92–1.00 ⬍.05 0.96 0.92–1.01 .1 Systolic blood pressure at year 1 1.01 0.96–1.05 NS 1.00 0.96–1.04 NS Diastolic blood pressure at year 1 0.99 0.97–1.03 NS 0.99 0.96–1.03 NS
Glucose level at year 1 1.03 0.97–1.10 NS 1.02 0.96–1.09 NS
Year 1 indicates ages 9 and 10 years; year 2, ages 10 and 11 years.
* Odds ratios for BMI ranged from 1.2 with diastolic blood pressure and glucose level to 1.19 with HOMA-IR and systolic blood pressure (allP⬍.009), but BMI was not significant with waist circumference in the model).
who had increased waist circumference at year 2 but not at year 10. The tracking coefficient for waist circumference was high (82.3%), and ⬃72% of girls with increased central adiposity at ages 10 and 11 years by age-specific cutoffs had increased central adiposity at 18 and 19 years according to ATP III standards. These findings are consistent with earlier reports that waist circumference is the primary pre-dictor of the metabolic consequences of obesity9,21–23 and extend these findings by showing that what happens in childhood central adiposity is critical.
Strengths of the current report include its prospec-tive design, relaprospec-tively large sample size, and careful, collaborative protocol. Weaknesses of the report in-clude the fact that the NGHS was designed to un-derstand factors associated with racial differences in obesity and not to address questions concerning the evolution of insulin resistance, metabolic syndrome, and diabetes; hence, girls could participate in the NGHS without having blood drawn. Although par-ticipants with and without laboratory measures did not have mean BMI significantly different at baseline or in black girls in year 10, white girls who had blood drawn at ages 18 and 19 years had significantly lower mean BMI than those who did not. This could have affected prevalence estimates of elevated glu-cose and triglyceride levels and of the syndrome in white girls.
The development of abnormal levels of the indi-vidual elements constituting the metabolic syndrome can have its origins and onset in adolescence and develop through adulthood. Although very rare in prepuberty when adult standards are used, the ponents of the syndrome become increasingly com-mon during adolescence, and the constellation ofⱖ3 components occurs in 3% of young adult women. The early predictors of metabolic syndrome can be identified before the complete syndrome is seen. Early recognition of central adiposity is critical, and as this study illustrates, preadolescent central adi-posity that does not persist is not associated with increased incidence of the syndrome. Thus, taking action in adolescence could provide major health benefits.
ACKNOWLEDGMENTS
This research was supported in part by National Institutes of Health, National Heart, Lung, and Blood Institute contracts HC-55023-26, HL 48941, and MO1 RR 08084
REFERENCES
1. Isomaa B, Almgren P, Tuomi T, et al. Cardiovascular morbidity and mortality associated with the metabolic syndrome.Diabetes Care.2001; 24:683– 689
2. Expert Panel on Detection, Evaluation, and Treatment of High Choles-terol in Adults. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III).JAMA.2001;285:2486 –2497
3. US Department of Health and Human Services. The Lipid Research Clinics Population Studies Data Book. Vol I: The Prevalence Study. Bethesda, MD: US Department of Health and Human Services, Public Health Service: 1980. National Institutes of Health publication 80 –1527 4. Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome
among US adults: findings from the Third National Health Examination Survey.JAMA.2002;287:356 –359
5. Haffner SM, Valdez RA, Hazuda HP, Mitchell BD, Morales PA, Stern MP. Prospective analysis of the incidence of the insulin-resistance syn-drome (synsyn-drome X).Diabetes.1992;41:715–722
6. Liese AD, Mayer-Davis EJ, Tyroler HA, et al. Development of the multiple metabolic syndrome in the ARIC cohort: joint contribution of insulin, BMI, and WHR: atherosclerosis risk in communities.Ann Epi-demiol.1997;7:407– 416
7. Schmidt MI, Duncan BB, Watson RL, Sharrett AR, Brancati FL, Heiss G. A metabolic syndrome in whites and African Americans: the Athero-sclerosis Risk in Communities Study.Diabetes Care.1996;19:414 – 418 8. Cook S, Weitzman M, Auinger P, Nguyen M, Dietz WH. Prevalence of
metabolic syndrome phenotype in adolescents: findings from the Third National Health and Nutrition Examination Survey, 1988 –1994.Arch Pediatr Adolesc Med.2003;157:821– 827
9. Kissebah AH, Vydelingum N, Murray R, et al. Relation of body fat distribution to metabolic complications of obesity.J Clin Endocrinol Metab.1982;54:254 –260
10. Srinivasan SR, Myers L, Berenson GS. Predictibility of childhood adi-posity and insulin for developing insulin resistance syndrome (syn-drome X) in young adulthood: the Bogalusa Heart Study.Diabetes.
2002;51:204 –209
11. Pinhas-Hamiel O, Dolan LM, Daniels SR, Standiford D, Khoury PR, Zeitler P. Increased incidence of non-insulin-dependent diabetes melli-tus among adolescents.J Pediatr.1996;128:608 – 615
12. The NHLBI Growth and Health Study Research Group. Obesity and cardiovascular disease risk factors in 9-and 10-year-old black and white girls: the NHLBI Growth and Health Study.Am J Public Health.1992; 82:1613–1620
13. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man.Diabetologia.1985;28:412– 419
14. SAS Institute.SAS/STAT User’s Guide. Version 8. Cary, NC: SAS Insti-tute, Inc; 1999
15. Klein DJ, Aronson Friedman L, Harlan WR, et al. Obesity and the development of insulin resistance and impaired fasting glucose in black and white adolescent girls: a longitudinal study.Diabetes Care.2004;27: 378 –383
16. Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC growth charts for the United States: methods and development. National Center for Healh Statistics.Vital Health Stat 11.2002;(246):1–190
17. Moran A, Jacobs DR Jr, Steinberger J, et al. Insulin resistance during puberty: results from clamp studies in 357 children.Diabetes.1999;48: 2039 –2044
18. Morrison JA, McMahon RP, Sprecher DL, Simon JA. Obesity and high-density lipoprotein cholesterol in black and white 9- and 10-year-old girls: the National Heart, Lung, and Blood Institute Growth and Health Study.Metabolism.1996;45:469 – 475
19. Morrison JA, Barton BA, Obarzanek E, et al. Racial differences in sums of skinfolds and percent body fat estimated from impedance in black and white girls, 9 –19 years of age: the NHLBI Growth and Health Study [published correction appears inObes Res. 2001;9:510].Obes Res.2001; 9:297–305
20. Vega GL, Clark LT, Tang A, Marcovina S, Grundy SM, Cohen JC. Hepatic lipase is lower in African American men than in white Amer-ican men: effects of 5⬘flanking polymorphism in the hepatic lipase gene (LIPC).J Lipid Res.1998;39:228 –232
21. Hartz AJ, Rupley DC, Rimm AA. The association of girth measurements with disease in 32,856 women.Am J Epidemiol.1984;119:71– 80 22. Stevens J, Keil JE, Rust PF, Tyroler HA, Davis CE, Gazes PC. Body mass
index and body girths as predictors of mortality in black and white women.Arch Intern Med.1992;152:1257–1262
DOI: 10.1542/peds.2004-2358
2005;116;1178
Pediatrics
Bruce A. Barton, George B. Schreiber and David J. Klein
John A. Morrison, Lisa Aronson Friedman, William R. Harlan, Linda C. Harlan,
Longitudinal Assessment
Development of the Metabolic Syndrome in Black and White Adolescent Girls: A
Services
Updated Information &
http://pediatrics.aappublications.org/content/116/5/1178
including high resolution figures, can be found at:
References
http://pediatrics.aappublications.org/content/116/5/1178#BIBL
This article cites 20 articles, 7 of which you can access for free at:
Subspecialty Collections
http://www.aappublications.org/cgi/collection/obesity_new_sub Obesity
http://www.aappublications.org/cgi/collection/nutrition_sub Nutrition
icine_sub
http://www.aappublications.org/cgi/collection/adolescent_health:med Adolescent Health/Medicine
ub
http://www.aappublications.org/cgi/collection/metabolic_disorders_s Metabolic Disorders
following collection(s):
This article, along with others on similar topics, appears in the
Permissions & Licensing
http://www.aappublications.org/site/misc/Permissions.xhtml
in its entirety can be found online at:
Information about reproducing this article in parts (figures, tables) or
Reprints
http://www.aappublications.org/site/misc/reprints.xhtml
DOI: 10.1542/peds.2004-2358
2005;116;1178
Pediatrics
Bruce A. Barton, George B. Schreiber and David J. Klein
John A. Morrison, Lisa Aronson Friedman, William R. Harlan, Linda C. Harlan,
Longitudinal Assessment
Development of the Metabolic Syndrome in Black and White Adolescent Girls: A
http://pediatrics.aappublications.org/content/116/5/1178
located on the World Wide Web at:
The online version of this article, along with updated information and services, is
by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.
