Sexual maturation and metabolic profile among adolescents and children of the Health Worker Cohort Study in Mexico

Madurez sexual y perfil metabólico de los adolescentes y niños del estudio de cohorte de trabajadores de la salud en México

Jodie G Katon, MS^{I, II}; Yvonne N Flores, PhD^{II, III}; Jorge Salmerón, MD, DrSc^II

^IUniversity of California, Los Angeles (UCLA) Department of Epidemiology, Los Angeles, CA, USA
^IIIMSS Epidemiology and Health Services Research Unit, Morelos, Mexico ]]> ^IIIUCLA Department of Health Services, School of Public Health and Jonsson Comprehensive Cancer Center, Los Angeles, CA, USA

ABSTRACT

OBJECTIVE:Our objective was to investigate the associations between level and timing of sexual development with metabolic profile in a cohort of Mexican adolescents in central Mexico.
MATERIAL AND METHODS:Baseline data from children between the ages of 7 and 17 years (n= 582) who participated in the Health Worker Cohort Study, was used. The study participants included children of workers at the Mexican Institute of Social Security (IMSS) and the National Institute of Public Health, both located in Cuernavaca, in addition to children of workers at the Universidad Autónoma del Estado de México in Toluca who were enrolled between March 2004 and April 2006. Multiple linear regressions with robust estimates of variance, were used adjusting for specific covariates.
RESULTS:Both pubertal boys and girls, compared to their pre-pubertal counterparts, had higher body mass index (girls: 4.59 kg/m², p<0.0001; boys: 1.12 kg/m², p= 0.05) and percent body fat (girls: 3.61, p<0.0001; boys: 1.48, p= 0.0001). A significant difference in level of insulin resistance (homeostasis model assessment, HOMA) was detectable among girls (0.92, p<0.0001).
CONCLUSIONS:Timing and levels of sexual development were significantly associated with adverse differences in several critical anthropometric and metabolic parameters.

Key words: sexual maturation; metabolism; adolescents; Mexico

RESUMEN

Palabras clave: madurez sexual; metabolismo; adolescentes; México

Puberty is a stage of rapid physiologic changes that may impact metabolic outcomes later in life. During adolescence, over the course of pubertal development, various metabolic and physiologic changes occur in parameters such as insulin sensitivity^1,2 and lipid levels.³ In a recent euglycemic clamp study of 357 children, it was found that insulin sensitivity decreased with the onset of puberty (Tanner stage II), reached a nadir at Tanner stage III, and returned to near pre-pubertal levels by Tanner stage V.¹

Mexican-Americans have higher rates of obesity⁴ and metabolic syndrome (MS),⁵ which are predictive of type 2 diabetes and poor cardiovascular health in adulthood.⁶ Adolescents in Mexico experience a similarly high prevalence of obesity and metabolic syndrome. A recent study reported a prevalence of MS from 10.6% to 19.6% among a group of Mexican adolescents depending on the definition used.⁷ Further, compared to Caucasian children, Hispanic children have lower insulin sensitivity and higher insulin secretion throughout childhood and adolescence.⁸ The elevated prevalence of obesity in this population, combined with lower insulin sensitivity and the high prevalence of low high density lipoprotein (HDL) and high triglycerides puts Mexican adolescents at a higher risk for metabolic syndrome. The exact mechanism responsible for the pubertal drop in insulin sensitivity remains unknown. Moreover, the impact these changes in insulin sensitivity might have on the risk of developing metabolic syndrome is also not known.

Timing of sexual maturation is another important factor when assessing metabolic parameters in pubertal children. Mexican-American adolescents generally experience earlier sexual maturation than their Caucasian peers.⁹ Multiple studies have found that early sexual maturation is associated with a higher risk of obesity and MS.^10-13 This association may convey additional risk of developing metabolic disorders among Mexican-American and Mexican adolescents who experience earlier sexual maturation,⁹ and have lower insulin sensitivity,⁸ a higher prevalence of obesity,⁴ lower HDL, and higher triglycerides¹⁰ than their Caucasian peers.

Materials and Methods

Study population

The study population consisted of a sample of children and adolescents who participated, along with their parents, in the Health Worker cohort study. The study participants included children of workers at the Mexican Institute of Social Security (IMSS, per its abbreviation in Spanish) and the National Institute of Public Health, both located in Cuernavaca, in addition to children of workers at the Universidad Autonoma del Estado de Mexico in Toluca, the capital of the neighboring state. All participants are being followed in an on going cohort study of life style and health and were enrolled between March 2004 and April 2006. The specifics regarding the study design, methodology and baseline characteristics of participants have been detailed elsewhere,⁷ and the ethical committees of all participating institutions approved the study protocol and consent forms for the cohort study.

A total of 1349 children and adolescents aged 7-17 years old were enrolled in the Health Worker cohort study. In order to examine the differences in metabolic profile by level of sexual development, we restricted our analysis to participants aged 7-17 who reported being in Tanner stages I-IV (n=1000). Those in Tanner stage V were excluded from this analysis since sexual development ends during Tanner V and we wished to eliminate children from our sample who had completed development prior to our study. Most participants had complete clinical data at the time of this analysis; however, only 551 of these participants had fasting insulin and glucose data in addition to percent body fat. To examine the differences in metabolic profile associated with timing of sexual maturation we expanded our sexual development criteria to include those in Tanner stages I-V, but further restricted the age range to those 7-14. The age restriction was done in order to eliminate children who had already completed sexual development at the time of enrollment, for whom we could not determine the timing of their development. Sexual development is completed during Tanner stage V. According to Sun et. al. the median age at entry to Tanner stage V,⁵ for Mexican-American girls was 14.70. Therefore, excluding those 14 and older was a conservative approach that ensured that we only included those in our analysis for whom timing of puberty could be established with some confidence. Of these participants, 425 met these inclusion criteria and had complete glucose and insulin data.

Assessment of pubertal stage and metabolic parameters

We classified pubertal stage based on Tanner staging. Tanner staging uses the appearance of secondary sexual characteristics that include four separate indexes: breast development and pubic hair for girls and testicular development and pubic hair for boys.¹⁴ Participants were shown sex-specific pictures of adolescents in each stage and asked to identify which picture most accurately matched their current stage of sexual development. Breast stage for girls and testicular development stage for boys were used in this analysis. For both sexes, those in Tanner I were classified as pre-pubertal, those in Tanner II-IV were classified as pubertal, and those in Tanner V were classified as post-pubertal.

We defined early maturation as being below the lower limit of the 95% confidence interval of the mean age for a given Tanner stage and sex as indicated by the Third National Health and Nutrition Examination Survey (NHANES III) for Mexican-Americans.⁹ We used the mean age in a Tanner stage and the standard error reported by Sun et al.⁹ to calculate the lower limits of the 95% confidence intervals for the mean age in a stage. This was a modification of the method used by Wang, who classified children who were below the median age for a given gender and stage.¹³ Rather than simply defining half of our population as early maturers we used the lower limit of the 95% confidence interval for the mean age in a stage in order to give the definition of early maturer more physiologic and clinical meaning. Specifically, the lower limits we calculated for girls were: 8 years old (Tanner I), 10.29 years old (Tanner II), 12.22 years old (Tanner III), 13.50 years old (Tanner IV), and 15.97 years old (Tanner V). The lower limits for boys were: 9 years old (Tanner I), 10.76 years old (Tanner II), 12.42 years old (Tanner III), 15.01 years old (Tanner IV), and 16.60 years old (Tanner V).

Insulin sensitivity was measured using the homeostasis model assessment (HOMA).¹⁵ HOMA was calculated from the fasting insulin and glucose measurements using the standard formula: (glucose (mmol/L) x insulin (μU/mL))/22.5. Similarly, HOMA-B was used as a measure of beta-cell function and calculated using the standard formula: (20 x insulin (μU/mL))/( glucose (mmol/L)-3.5).¹⁶ Serum glucose concentration was determined by the oxidize glucose method; insulin was measured in serum by solid phase RIA (Coat-A-Count, Diagnostic Products, Los Angeles, CA, USA).¹⁷ To remain consistent with previous analyses among adolescents from the National Health and Nutrition Examination Survey (NHANES)^18-21 and other Health Worker cohort studies,⁷ a fasting time of eight hours or greater was used for all assays. Serum triglyceride concentrations were analyzed with a colorimetric method after enzymatic hydrolysis with lipases. HDL/LDL was analyzed by the elimination of chylomicron and VLDL-C, LDL/HDL-C by cholesterol esterase, cholesterol oxidize, and subsequently catalase, respectively.⁷

Physical activity was assessed using the International Physical Activity Questionnaire (IPAQ) instrument^22,23 which was adapted for the Mexican population. Participants were asked about their leisure activity, recreational activity, daily activity, and any physical labor associated with employment. The IPAQ section on leisure time activity included 16 items on the amounts of time spent weekly performing exercises such as walking, running and cycling. Each activity was given a value in metabolic equivalents (MET) and total METS per week were calculated. Three categories were defined: <21 METS per week, > 21 METS per week and < 110 METS per week, and >110 METS per week.

Participants were shown a series of graded silhouettes, numbered 1-9, of men and women and asked to mark the body type that best described their mother's and father's current physical appearance. The use of this scale was previously validated for use within this population.²⁴ A selection of an image numbered four or greater was interpreted as a parent being overweight or obese. Participants were categorized as having one, two, or no overweight or obese parents.

Statistical analysis

Multiple regression models with robust variance estimates were used to determine the means and 95% confidence intervals for BMI, waist circumference, percent body fat, insulin, glucose, HOMA, HOMA-B, HDL, triglycerides, and low density lipoprotein (LDL) (treated as dependent variables) according to levels of sexual maturation for girls and boys, adjusting for age, percent body fat (except where percent body fat was the dependent variable), physical activity and number of obese or overweight parents. The association of being pubertal versus pre-pubertal with differences in metabolic parameters was allowed to vary by age. Student's t-tests were used to identify differences between pre-pubertal versus pubertal status.

Analyses of the association between early sexual maturation and differences in metabolic components and anthropometric parameters were carried out in a similar fashion. These analyses were restricted to participants aged 7-14 years in order to exclude individuals who could not be identified conclusively as early-maturers. The cut-point of age 14 was chosen based on the average age of entry into Tanner stage V for Mexican-American adolescent girls reported by Sun et al. (mean age at entry for Tanner V for Mexican-American girls: 14.70).⁹ While this was a conservative estimate, using this cut point gave us confidence that we were excluding from our analysis those children who had completed puberty before enrolling in the cohort and for whom we could not adequately establish timing of sexual development. To evaluate the association of early sexual maturation and the change in the metabolic components of interest we used our definitions of early versus not early. All analyses were adjusted for age, percent body fat (except where percent body fat was the dependent variable), physical activity and parental obesity or overweight. Student's t-tests were used to identify differences between early and not early maturers. Analyses were conducted using STATA 9 software.^*

Results

The characteristics of the sample covariates are shown in Table I by level of sexual maturation and timing of sexual maturation. The pre-pubertal children differed significantly from the pubertal children in age, mean BMI, and percent body fat. The early maturing children did not differ significantly from the non-early maturers in any of the covariates.

Being pubertal versus pre-pubertal among girls was associated with higher BMI (4.59 kg/m², p<0.0001), waist circumference (15.26 cm, p<0.0001), percent body fat (3.61%, p<0.0001), glucose (0.45 mg/dL, p=0.01), insulin (4.47 μU/mL, p<0.0001), HOMA (0.92, p<0.0001), HOMA-B (67.57, p<0.0001), and triglycerides (10.43 mg/dL, p<0.0001), as well as lower HDL (-3.25 mg/dL, p<0.0001) and LDL (-3.94 mg/dL, p<0.0001). Among boys, being pubertal versus pre-pubertal was associated with higher BMI (1.12 kg/m², p= 0.05), waist circumference (4.74 cm, p= 0.01), and percent body fat (1.48%, p=0.0001), as well as lower HDL (-0.95 mg/dL, p= 0.03), triglycerides (-6.99 mg/dL, p=0.01) and LDL (-11.20 mg/dL, p<0.0001) (Table II).

Discussion

In this cross-sectional survey of 582 Mexican adolescents in Cuernavaca and Toluca, Mexico we observed significant differences in metabolic profiles according to levels of sexual development. Among girls, being pubertal (as opposed to pre-pubertal) was associated with unfavorable levels of BMI, waist circumference, percent body fat, fasting insulin, fasting glucose, HOMA, HDL, and triglycerides. Among boys, being pubertal (as compared to pre-pubertal) appeared to be associated with adverse levels of BMI, waist circumference, percent body fat and HDL. Early versus not early sexual maturation was associated with higher levels of BMI, waist circumference, HOMA, HOMA-B, fasting insulin, and fasting glucose among girls, in addition to lower LDL. For boys, early versus not early sexual maturation was associated with lower levels of BMI, waist circumference, percent body fat, fasting glucose, triglycerides, and LDL, in addition to higher HDL.

These findings indicated differences by sex in the associations between the measured metabolic parameters and level of sexual maturation. In general, the differences in metabolic components associated with increasing levels of sexual maturation were smaller among boys than among girls. A difference in the average fasting glucose, fasting insulin, HOMA, and HOMA-B for pre-pubertal versus pubertal girls was detected; however, no differences were detected for boys. Our inability to detect differences in HOMA by pubertal level in boys may corroborate Ball's finding that among overweight Hispanic children with a family history of type 2 diabetes the pubertal drop in insulin sensitivity is not detectable.²⁵ While only a small percentage of the children in our study were overweight, roughly 90% of our entire cohort reported having at least one parent who was obese or overweight. Since obesity is the main determinant of insulin resistance, then our findings suggest that it is the shared genetic and obesogenic environment rather than the obesity of the child which may be instrumental in masking or eliminating the pubertal drop in insulin sensitivity in our population. Both genetic predisposition and the environment may be impacting insulin sensitivity before either puberty or obesity can occur. This finding is largely unprecedented in the literature and warrants further investigation. In particular, longitudinal data with more exact information regarding timing of sexual maturation is needed to verify these findings and explore possible mechanisms.

The observed decrease in HDL between pubertal and pre-pubertal children has been previously described. Morrison et al. reported that the observed drop in total cholesterol across adolescence is due to reductions in HDL.³ While the expected drop in HDL was observed in both boys and girls, a rise in triglycerides was not uniformly observed for both sexes. Boys displayed a decrease in triglycerides while girls displayed the expected increase in triglycerides. This is a new finding that indicates that the changes in lipid profile associated with puberty are different than those associated with age and that they vary by sex.

The differences in anthropometric parameters for both boys and girls, observed in this study appear to reflect the normal reported patterns of growth.¹³ BMI, waist circumference and percent body fat were lower, on average, for boys and girls who were pre-pubertal versus pubertal. Since this data was cross-sectional it was not possible to interpret this finding as implying either that increased BMI is necessary for sexual maturation or that sexual maturation causes increased BMI. However, the association between increased BMI and puberty was consistent with the existing literature, particularly regarding girls.¹³

The greater magnitude of the differences in BMI, waist circumference, and percent body fat by level of sexual maturation among girls was most likely due to differences in growth patterns between the sexes. On average, girls experience greater increases in fat mass relative to lean mass during adolescence, and boys experience greater increases in lean mass relative to fat mass.¹³ This is probably part of the mechanism underlying the fact that, on average, adult women have 22% body fat compared to adult men who, on average, have only 15% body fat.²⁶ Although in our study, percent body fat was, on average, lower for girls who were pre-pubertal versus pubertal, this cannot be interpreted as implying a causal relationship or being supportive of a particular direction of causality. However, this observation was consistent with the existing literature.²⁶

The association of early sexual maturity with obesity and metabolic syndrome is fairly well established for girls.^10-12 Therefore, it is not surprising that our findings show that higher BMI, waist circumference, fasting insulin, fasting glucose, HOMA, and HOMA-B are associated with early sexual maturation among girls. Wang reported that early sexual maturation was inversely associated with BMI for boys and our results are consistent with this;¹³ however; our study is, to the best of our knowledge, the first to examine insulin sensitivity with respect to timing of sexual maturation among boys. Among the boys in our study population, early versus not early sexual maturation was associated with lower BMI, waist circumference, and percent body fat. In our study population, lower HOMA was associated with early versus not early sexual maturation among boys, but this decrease was not detectable at the 0.05 significance level. Early sexual maturation has been reported to be associated with adverse changes in serum lipids,¹² but our results indicate that early maturation is associated with lower LDL for girls and boys.

The cross-sectional design of this study makes it difficult to examine a potential causal relation of sexual maturation and early sexual maturation with metabolic outcomes such as insulin resistance and/or lipid profile, as temporality of events cannot be established using this type of data. The follow-up data on this cohort should shed more light on whether the differences that we have reported between early and not early maturers do impart excess risk for type 2 diabetes or the metabolic syndrome. Also, this cohort was drawn from the children and adolescent relatives of working, seemingly healthy individuals. While these children cannot be considered representative of the Mexican population as a whole, they may be considered representative of middle to low income adolescents residing in the urban areas of central Mexico.

Acknowledgements

Financial support for this study was provided by the Instituto Mexicano del Seguro Social (IMSS) and the Consejo Nacional para la Ciencia y Tecnología (CONACYT) grant #26267M. The authors would like to thank Dr. Simin Liu, Dr. Leo Morales, and Dr. Anna Haddal for their insight and suggestions and the investigators at IMSS for the use of their data. We would also like to acknowledge the Program on Genomics and Nutrition, Department of Epidemiology UCLA. Finally, the authors wish to thank the Neuman-Drabkin and Monica Salinas Funds for their generous financial support for Ms. Katon.

References

1. Moran A, Jacobs DR Jr, Steinberger J, Hong CP, Prineas R, Luepker R, et al. Insulin resistance during puberty: results from clamp studies in 357 children. Diabetes 1999;48:2039-2044.

2. Goran MI, Gower BA. Longitudinal study on pubertal insulin resistance. Diabetes 2001;50:2444-2449.        [ Links ]

3. Morrison JA, deGroot I, Edwards BK, Kelly KA, Mellies MJ, Khoury P, et al. Lipids and Lipoproteins in 927 School Children, Ages 6-17 Years. Pediatrics 1978;62:990-995.        [ Links ]

4. Ogden CL, Flegal KM, Carroll MD, Johson CL. Prevalence and trends in overweight among US children and adolescents, 1999-2000. JAMA 2002;288:1728-1732.        [ Links ]

5. Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults: Findings from the Third National Health and Nutrition Examination Survey. JAMA 2002;287:356-359.        [ Links ]

6. National Institutes of Health. 2001. The Third Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Bethesda (MD): National Institutes of Health. NIH Publication 01-3670.        [ Links ]

7. Halley Castillo E, Borges G, Talavera JO, Orozco R, Vargas-Alemán C, Huitrón-Bravo G, et al. Body mass index and the prevalence of metabolic syndrome among children and adolescents in two Mexican populations. J Adolesc Health 2007;40(6):521-526.        [ Links ]

8. Goran MI, Bergman RN, Cruz ML, Watanabe R. Insulin resistance and associated compensatory responses in African-American and Hispanic children. Diabetes Care 2002;25:2184-2190.        [ Links ]

9. Sun SS, Schubert CM, Chumlea WC, Roche AF, Kulin HE, Lee PA, et al. National estimates of the timing of sexual maturation and racial differences among US children. Pediatrics 2002;110:911-919.        [ Links ]

10. Frontini MG, Srinivasan SR, Berenson GS. Longitudinal changes in risk variables underlying metabolic syndrome X from childhood to young adulthood in female subjects with a history of early menarche: The Bogalusa Heart Study. Int J Obes 2003;27:1398-404.        [ Links ]

11. Himes JH. Examining the evidence for recent secular changes in the timing of puberty in US children in light of increases in the prevalence of obesity. Mol Cell Endocrinol 2006;254-255:13-21.        [ Links ]

12. Ibanez L, Potau N, Chacon P, Pascual C, Carrascosa A. Hyperinsulinaemia, dyslipaemia and cardiovascular risk in girls with a history of premature pubarche. Diabetologia 1998;41:1057-1063.        [ Links ]

13. Wang Y. Is obesity associated with early sexual maturation? A comparison of the association in American boys versus American girls. Pediatrics 2002;110:903-910.        [ Links ]

14. Tanner JM. Normal growth and techniques of growth assessment. Clin Endocrinol Metab 1986;15(3):411-451.        [ Links ]

15. Keskin M, Kurtoglu S, Kendirci M, Atabek ME, Yazici C.. Homeostasis model assesment is more reliable than the fasting glucose/insulin ratio and quantitative insulin sensitivity check index for assessing insulin resistance among obese children and adolescents. Pediatrics 2005;115:e500-e503.        [ Links ]

16. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and b-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412-419.        [ Links ]

17. Saucedo R, Basurto L, Zarate A, Martínez C, Hernandez M, Galván R. Effect of estrogen therapy on insulin resistance and plasminogen activator inhibitor type 1 concentrations in postmenopausal women. Gynecol Obstet Invest 2007;64(2):61-64.        [ Links ]

18. Weiss R, Dziura J, Burgert TS, Tamborlane WV, Taksali SE, Yeckel CW, et al. Obesity and the metabolic syndrome in children and adolescents. N Engl J Med 2004;350:2362-2374.        [ Links ]

19. Glen ED, Sierra ML, Xiao-Hua Z. Prevalence and trends of a metabolic syndrome phenotype among U.S. adolescents, 1999-2000. Diabetes Care 2004;2:2438-2443.        [ Links ]

20. Ferranti SD, Gauvreau K, Ludwig DS, Neufeld EJ, Newburger JW, Rifai N. Prevalence of the metabolic syndrome in American adolescents. Findings from the Third National Health and Nutrition Examination Survey. Circulation 2004;110:2494 -2497.        [ Links ]

21. Da Silva RCQ, Miranda W, Chacra AR, Dib SA. Metabolic syndrome and insulin resistance in normal glucose tolerant Brazilian adolescents with family history of type 2 diabetes. Diabetes Care 2005;28;716-718.        [ Links ]

22. The International Physical Activity Questionnaire 2005. Available: http://www.ipaq.ki.se        [ Links ]

23. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003;35:1381-1395.        [ Links ]

24. Osuna-Ramírez I, Hernández-Prado B, Campuzano JC, Salmerón J. Body mass index and body image perception in a Mexican adult population: the accuracy of self-reporting. Salud Publica Mex 2006;48(2):94-103.        [ Links ]

25. Ball G, Weigensberg MJ, Cruz ML, Shaibi GQ, Kobaissi H, Goran MI. Insulin sensitivity, insulin secretion and b-cell function during puberty in overweight Hispanic children with a family history of type 2 diabetes. Int J Obes 2005;29:1471-1477.        [ Links ]

26. Mahan LK, Escott-Stump S. Krause's food, nutrition, and diet therapy. 9^th ed. Philadelphia (PA): WB Saunders, 1996.        [ Links ]

Received on: August 19, 2008
Accepted on: February 27, 2009

Address reprint requests to: Dra. Yvonne Flores. Unidad de Investigación Epidemiológica y en Servicios de Salud. Instituto Mexicano del Seguro Social (IMSS). Av. Plan de Ayala Esq. Calle Central S/N col. Chapultepec. 62450 Cuernavaca, Morelos, México. E-mail: ynflores@ucla.edu
* StataCorp. Stata Statistical Software: Release 9. College Station (TX, USA): Statacorp LP, 2005.