- Email: [email protected]
Risk Factors for Type 2 Diabetes and Cardiovascular Diseases in Hispanic Adolescents
Journal of Adolescent Health 43 (2008) 444 – 450
Original article
Risk Factors for Type 2 Diabetes and Cardiovascular Diseases in Hispanic Adolescents Gianna Perez Gomez, Ph.D. and Fatma G. Huffman, Ph.D.* Department of Dietetics and Nutrition, Florida International University, Miami, Florida Manuscript received December 6, 2007; manuscript accepted March 18, 2008
Abstract
Purpose: To investigate the associations of overweight with risk factors for type 2 diabetes (T2DM), cardiovascular diseases (CVD), and presence of metabolic syndrome (MetS) in Hispanic adolescents. Methods: A total of 100 adolescents (50 overweight, 50 nonoverweight) aged 12 to 16 years were included. Fasting plasma glucose (FPG), insulin sensitivity (IS), total cholesterol (TC), triacylglycerols (TG), LDL, HDL, and blood pressure (BP) were evaluated as risk factors for T2DM and CVD. IS was determined using the HOMA calculator. Abnormal levels were: FPG ⱖ100 and ⬍126 mg/dL (impaired fasting glucose [IFG]), IS ⬍100%, TC ⱖ200 mg/dL, TG ⱖ130 mg/dL, LDL ⱖ130 mg/dL, HDL ⬍35 mg/dL, and BP ⱖ95th percentile. MetS was defined as having equal to or more than three of the following: abdominal obesity (waist-to-height ratio ⬎0.5), high BP, high TG, low HDL, and IFG. Results: Overweight adolescents had higher TC, TG, and LDL, and lower IS, ps ⬍ .001. They were more likely to have abnormal levels of IS, TG, and LDL, one and two or more risk factors compared to having zero, and two or more risk factors compared to having one, ps ⬍ .05. Only 9% of adolescents met the criteria of MetS, and they were all overweight. Abdominal obesity (51%), low HDL (21%), and high TG (16%) were the most predominant components of the MetS. Conclusions: Overweight Hispanic adolescents are at increased risk of developing T2DM or CVD. Preventive programs targeting Hispanic adolescents are needed to manage the overweight epidemic and its consequences. © 2008 Society for Adolescent Medicine. All rights reserved.
Keywords:
Hispanic adolescents; Risk factors; Cardiovascular diseases; Diabetes; Metabolic syndrome
Type 2 diabetes mellitus (T2DM) was for many years considered a disease of the middle aged. However, its incidence has been increasing among children of populations at high risk, such as Hispanics [1,2]. In addition, overweight, one of the main risk factors for T2DM, has more than doubled in the last 2 decades in Hispanic adolescents [3]. One of the most life-threatening complications of T2DM is heart disease, and studies have indicated that there is a Funding for this research was provided through an NIH-sponsored training grant (FREA), a Foundation grant from Florida International University, and a Florida International University Graduate School Dissertation Year Fellowship. *Address correspondence to: Fatma G. Huffman, Ph.D., Department of Dietetics and Nutrition, Florida International University, 11200 SW 8th Street, HLS II-450, Miami, FL 33199. E-mail address: [email protected]
high cardiovascular risk prior to the clinical diagnosis of T2DM [4]. Type 2 diabetes in children is usually diagnosed after age 10. These patients have a high prevalence of overweight and other risk factors for cardiovascular diseases (CVD), with their body mass index (BMI) ranging from 26 to 38 kg/m2, the prevalence of high blood pressure (BP) ranging from 17% to 32%, and high triacylglycerols (TG) ranging from 4% to 32% [5]. In addition, ethnic differences in blood insulin and glucose, BMI, and other CVD risk factors occur as early as age 8 to 10 years, with Hispanic children having more unfavorable profiles than non-Hispanic white children [6]. The metabolic syndrome (MetS) is defined as the clustering of multiple risk factors that increase the risk of T2DM and CVD [7]. Components of the MetS are also found in children and adolescents [8]. However, a worldwide con-
1054-139X/08/$ – see front matter © 2008 Society for Adolescent Medicine. All rights reserved. doi:10.1016/j.jadohealth.2008.03.010
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
sensus for the definition of MetS has not been established in the pediatric population. The National Cholesterol Education Program Adult Treatment Panel III (ATP III) identifies five components of the MetS in adults, and defines it as having three or more of them: abdominal obesity, high TG, low high-density lipoprotein cholesterol (HDL), hypertension, and impaired glucose metabolism [9]. Researchers have modified the adult criteria in different ways to create their own definition for children and adolescents [10 –13]. All of the definitions have been consistent with the general features of the MetS in adults, including a combination of central obesity, hypertension, dyslipidemia, and impaired glucose metabolism. Hispanics represent 14.2% of U.S. households [14]. The increasing prevalence of overweight in the Hispanic pediatric population will represent a major challenge to our society and the healthcare system. Thus, this study responds to the need for more information on the epidemiology of T2DM and CVD in the pediatric population by exploring the associations of being overweight with risk factors for these diseases. In addition, because of the importance of the MetS as a prelude to T2DM and CVD in the adult population, the study also explored presence of MetS components in this population. This study is important for four reasons: first, T2DM is increasing in children and adolescents of all ethnic backgrounds; second, there is limited amount of information regarding the epidemiology of the disease in the pediatric population; third, Hispanics are at increased risk for T2DM; and fourth, there is evidence that individuals at risk for T2DM are also at risk for CVD. Methods Inclusion criteria Included were apparently healthy Hispanic adolescents 12 to 16 years of age, who signed an assent form and whose parents signed a consent form. Exclusion criteria Adolescents receiving hormonal treatment or any other drug (i.e., steroids) that could confound the results of the biochemical parameters, adolescents with chronic diseases like diabetes, CVD, inherited lipid disorders, etc., underweight (BMI-for-age ⬍5th percentile) adolescents or any other disease that could impede their participation in the study were not included. At risk of overweight (BMI-forage ⱖ85th to ⬍95th percentile) adolescents were not included in the study to better contrast overweight versus nonoverweight. Study population This was a cross-sectional study comprising of 117 adolescents of various Hispanic origins. Of those, 57 were overweight (BMI-for-age ⱖ95th percentile) and 60 were
445
nonoverweight (BMI-for-age ⬍85th percentile). Subjects were selected by convenience sampling technique from three different Pediatric clinics in Miami–Dade County, South Florida, until a minimum of 50 subjects that completed the study was reached in each group. All subjects meeting the inclusion criteria were approached and invited to participate in the study. Subjects invited to participate in the study were apparently healthy children who had gone to the clinic for a regular well-check or vaccination for school purposes. A total of 14 subjects (5 overweight, 9 nonoverweight) did not complete the biochemical analyses and were not included in the statistical analyses (listwise deletion). Of the subjects that completed the study, three were excluded for meeting the exclusion criteria after the biochemical analyses were received. One overweight subject had TG levels ⬎900 mg/dL and was diagnosed with Familial Hyperlipidemia afterward. Another overweight subject met the diagnosis criteria for diabetes (FPG ⬎126 mg/dL). The nonoverweight subject was taking Accutane (can cause hyperlipidemia and increased glucose levels as side effects), but this was reported to us after the results of the biochemical analyses were received. The study was approved by Florida International University institutional review board for the use of human subjects. Parental consent and child assent were both needed and obtained through institutional review board-approved consent and assent forms given to parents and adolescents, respectively. Assessment and measures Demographic questionnaires were given to parents and adolescents to complete. The physician and/or the medical assistant from the clinic where data were collected conducted all physical examinations, which included assessment of Tanner stage, measure of BP, and anthropometrics (weight, height, and waist circumference). A certified phlebotomist collected blood samples after an overnight fasting of no less than 12 hours for the biochemical analyses. BMI was calculated from weight and height and plotted in the Center for Disease Control’s gender- and age-specific growth charts [15]. Risk factors for T2DM and CVD. Fasting plasma glucose (FPG) (mg/dL) was classified according to the American Diabetes Association (ADA) as normal (FPG ⬍100 mg/ dL), impaired fasting glucose (IFG) (FPG ⱖ100 mg/dL and ⬍126 mg/dL), and provisional diagnosis of diabetes (FPG ⱖ126 mg/dL) [16]. Insulin sensitivity (IS) (%) was determined from FPG and insulin levels (IU/mL) using the HOMA Calculator© (The University of Oxford) [17]. This measure corresponds well to other estimates of -cell function and IS derived from hyperinsulinaemic and hyperglycemic clamps and intravenous and oral glucose tolerance tests [18]. The HOMA Calculator uses the HOMA2 model to estimate -cell function (%B) and IS (%S) for an indi-
446
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
vidual [19]. The HOMA approach has been used in research to determine IS and insulin resistance index in children and adolescents [20,21]. Normal IS was defined as IS ⱖ100% and low IS as IS ⬍100% [19]. High total cholesterol (TC) was considered as ⱖ200 mg/dL, high TG as ⱖ130 mg/dL, high LDL as ⱖ130 mg/ dL, and low HDL as ⬍35 mg/dL [9,22–24]. Measurement of BP (mmHg) followed the guidelines in the Physician Examiner’s Training Manual and used a standard clinical sphygmomanometer (auscultatory measurement) and appropriate size cuff [25]. Values of BP were plotted in the gender-, age-, and height-specific standard tables to find the precise percentile for systolic BP (SBP) and diastolic BP (DBP), that were published by the National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents [26]. Classification of BP followed the National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents definition: high BP (SBP and/or DBP ⱖ95th percentile), high-normal BP (SBP or DBP ⱖ90th and ⬍95th percentile), and normal BP (SBP and DBP ⬍90th percentile). Adolescents with BP ⱖ120/80 mmHg were also considered as having high-normal BP. Clustering of risk factors was defined as having greater than or equal two of the following: IFG, low IS, high TG, high LDL, low HDL, and high BP. Because TC contains HDL and LDL, it was excluded from the analysis of clustering of risk factors. Definition of metabolic syndrome. In agreement with the general definition of the MetS in adults using ATP III criteria [16], we included in our definition the general features of the MetS (central obesity, hypertension, dyslipidemia, and impaired glucose metabolism). There are no guidelines for assessing obesity-related risk factors in children and adolescents by using waist circumference (WC). Waistto-height ratio (W/Ht ratio), a predictor of cardiovascular risk in the adult population [27], was used in our study to determine abdominal obesity. A cutoff value of 0.5 established for adults [28] and also explored in children and adolescents [29] was used in our study to define abdominal obesity. We defined the MetS in our sample as having three or more of the following risk factors: abdominal obesity (W/Ht ratio ⬎0.5), high BP (either SBP or DBP ⱖ95th percentile), high TG (TG ⱖ130 mg/dL), low HDL (HDL ⬍35 mg/dL), and IFG. Data analysis Data analyses used SPSS 14.0 for Windows (SPSS Inc., Chicago, IL) and included descriptive statistics, logistic regression, multinomial logistic regression, two sample means tests, correlations, and chi-square. All variables were first tested for normality. All tests were considered statistically significant if p ⬍ .05. In all the analyses, gender,
family history of T2DM, and family history of CVD were controlled. Due to the importance of sexual development in puberty-related insulin resistance [30], Tanner stage was controlled in all the analyses of IS. Results A total of 100 adolescents (50 overweight, 50 nonoverweight) that completed the study were included in the statistical analyses for the hypotheses testing. Of those, 44% were Cuban descendants, 15% were Nicaraguan descendants, 13% had combined descendant, and the remaining % Table 1 General characteristics of the study subjects by BMI category BMI category*
p
Characteristics
Overweight
Nonoverweight
n Male/female (n) Age (years) Tanner stage† Category 2 3 4 5 M (⫾SD) Cuban descendant Born in the U.S. Family history of: Obesity T2DM CVD Gestational diabetes Blood pressure‡ Normal High-normal High FPG (mg/dL) IS (%) TC (mg/dL) TG (mg/dL) LDL (mg/dL) HDL (mg/dL)
50 26/24 14.1 (⫾1.3)
50 26/24 14.4 (⫾1.5)
.151
4% 34% 46% 16% 3.7 (⫾0.8) 34% 72%
4% 38% 42% 16% 3.7 (⫾0.8) 54% 34%
.977 .767 .044 ⬍.001
74% 51% 53% 8%
42% 49% 47% 4%
36% 46% 18% 90 (⫾11) 69 (⫾38) 167 (⫾33) 106 (⫾49) 105 (⫾28) 44 (⫾11)
92% 8% 0% 88 (⫾6) 170 (⫾68) 143 (⫾20) 66 (⫾31) 87 (⫾18) 46 (⫾10)
.001 .841 .362 .400
⬍.001 .812 ⬍.001 ⬍.001 ⬍.001 ⬍.001 .100
Note. Values are n, M (⫾SD), and %. Mann-Whitney U test was conducted for age, FPG, IS, TC, TG, LDL, and HDL and 2 test for Tanner stage, Cuban descendant, born in the U.S. family history, and blood pressure. FPG ⫽ fasting plasma glucose, IS ⫽ insulin sensitivity, TC ⫽ total cholesterol, TG ⫽ triacylglycerols, LDL ⫽ low-density lipoprotein cholesterol, HDL ⫽ high-density lipoprotein cholesterol. BP ⫽ either SBP or DBP ⱖ90th and ⬍95th percentile, or BP ⱖ120/80 mmHg, High BP ⫽ either SBP or DBP ⱖ95th percentile. * Overweight ⫽ BMI ⱖ95th percentile for age and gender, nonoverweight ⫽ BMI ⬍85th percentile for age and gender. † Tanner stage ranges from 1 (prepubertal) to 5 (adult like). ‡ High-normal and high blood pressure (BP) were combined into one category to conduct the 2 test due to small cell counts. Normal BP ⫽ SBP and DBP ⬍90th percentile, high-normal BP ⫽ either SBP or DBP ⱖ90th and ⬍95th percentile, or BP ⱖ120/80 mmHg, high BP ⫽ either SBP or DBP ⱖ95th percentile.
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
447
Table 2 Likelihood of overweight adolescents of having adverse levels for each of the risk factors for T2DM and CVD Risk factor
Overweight (%)
Nonoverweight (%)
p
OR
95% CI
IFG Low IS High TG High LDL Low HDL High BP
18 84 28 20 22 18
4 18 4 2 20 0
.125 ⬍.001 .011 .015 .513 .996
2.45 25.78 8.89 72.10 0.96 —
0.69 to 20.11 7.79 to 85.32 1.65 to 47.90 2.32 to 2243.09 0.33 to 2.76 — to —
Note. This analysis used logistic regression and controlled for gender, family history of T2DM, and family history of CVD. For Low IS, Tanner stage was also controlled. IFG (FPG ⱖ100 mg/dL and ⬍126 mg/dL), model 2(4, N ⫽ 100) ⫽ 7.45, p ⫽ .11. Low IS (IS ⬍100%), model 2(5, N ⫽ 100) ⫽ 48.43, p ⬍ .001. High TG (TG ⱖ130 mg/dL), model 2(4, N ⫽ 100) ⫽ 19.695, p ⫽ .001. High LDL (LDL ⱖ130 mg/dL), model 2(4, N ⫽ 100) ⫽ 12.20, p ⫽ .02. Low HDL (HDL ⬍35 mg/dL), model 2(4, N ⫽ 100) ⫽ .44, p ⫽ .98. High BP (either SBP or DBP ⱖ95th percentile for age, gender, and height), the test statistic was unstable due to large standard error; model 2(5, N ⫽ 100) ⫽ 13.50, p ⫽ .01.
were of other Hispanic origins. Table 1 summarizes characteristics of the study subjects by BMI category. No statistically significant differences were found for age or Tanner stage between overweight and nonoverweight adolescents, ps ⬎ .05. Family history of T2DM, CVD, and gestational diabetes were not associated with being overweight, ps ⬎ .05. Family history of obesity was significantly associated with being overweight, p ⫽ .001. A higher percentage of nonoverweight adolescents were of Cuban ancestry, p ⫽ .04. A higher percentage of adolescents born in the United States were overweight, p ⬍ .001. Blood pressure was significantly associated with being overweight, p ⬍ .001. Overweight adolescents had higher TC, TG, and LDL, and lower IS, than nonoverweight adolescents, ps ⬍ .001. A greater percentage of overweight adolescents had IFG, low IS, high TG, high LDL, and high BP, ps ⬍ .05. Table 2 presents the likelihood of overweight adolescents of having adverse levels for each of the risk factors, after controlling for gender and family history of T2DM and CVD. Overweight adolescents were more likely to have low IS, high TG, and high LDL, ps ⬍ .05. IFG and high BP were not significant after controlling for these confounding factors. A higher percentage of overweight adolescents had one and two or more risk factors for T2DM and CVD compared to nonoverweight adolescents, 2(2, N ⫽ 100) ⫽ 43.84, p ⬍ .001. Table 3 presents the likelihood of overweight adolescents of having one and two or more risk factors compared to having zero, and the likelihood of having two or more risk factors compared to having one, after controlling for gender, family history of T2DM, and CVD. Overweight adolescents were more likely to have one and two or more risk factors compared to having zero, ps ⬍ .001. In addition, overweight adolescents were more likely to have two or more risk factors compared to having one, p ⫽ .03. Mean W/Ht ratio was 0.6 (⫾0.1) for overweight and 0.4
(⫾0.04) for nonoverweight adolescents, z ⫽ ⫺8.55, p ⬍ .001. A total of 98% of overweight and 4% of nonoverweight adolescents had W/Ht ratio ⬎0.5, 2(1, N ⫽ 100) ⫽ 88.4, p ⬍ .001. Adolescents with higher W/Ht ratio or W/Ht ratio ⬎0.5 had lower IS and higher BP percentiles, TC, TG, and LDL, ps ⬍ .05. Only 9% of the adolescents met the criteria definition of the MetS, and they were all overweight. Abdominal obesity (51%), low HDL (21%), and high TG (16%) were the most predominant components of the MetS for the entire sample. Figure 1 shows percentage of Hispanic adolescents having the individual components of the MetS for each group, and Figure 2 shows percentage of overweight and nonoverweight adolescents with zero, one, two, and three or more components of the MetS. Discussion The problem of overweight children is worsening at a dramatic rate in the United States for all ethnicities [3], including Hispanics, who are one of the fastest growing minority group. Excess body fat carries additional conse-
Table 3 Likelihood of overweight adolescents of having one or two or more risk factors for T2DM and CVD Number of risk factors*

SE
Wald
p
OR
95% CI
1 vs. 0 2 ⫹ vs. 0 2 ⫹ vs. 1
3.42 4.75 1.34
0.89 0.93 0.63
14.59 26.38 4.53
⬍.001 ⬍.001 .033
30.42 115.79 3.81
5.27 to 175.43 18.89 to 709.78 1.11 to 13.04
Note. Multinomial logistic regression was used, controlling for gender, family history of T2DM, and family history of CVD. Model 2(8, N ⫽ 100) ⫽ 65.86, p ⬍ .001. * Overweight adolescents: 4% had 0 risk factors, 36% had one risk factor, and 60% had two or more risk factors; nonoverweight adolescents: 62% had 0 risk factors, 28% had one risk factor, and 10% had two or more risk factors.
448
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
Figure 1. Percentage of overweight and nonoverweight Hispanic adolescents with individual components of the metabolic syndrome.
quences that increase the risk for developing chronic diseases like T2DM and CVD [31]. Addressing the question of whether overweight Hispanic adolescents have an increased risk for these chronic diseases, this study explored the associations of being overweight with risk factors for T2DM and CVD. Significant differences were found for BP, TC, TG, LDL, and IS between overweight and nonoverweight adolescents, with overweight adolescents always having the worst profiles. These findings suggest that overweight Hispanic adolescents are at increased risk of developing T2DM or CVD. These data are consistent with the results of other studies that found an increased risk for T2DM and CVD in overweight adolescents of other ethnicities [32,33]. Most of the studies in adolescents evaluate the linear associations between BMI and risk factors for T2DM and CVD. Only few studies have assessed the likelihood of overweight adolescents of having abnormal levels of each of the risk factors and their likelihood of having clustering of multiple risk factors [23]. By using published cutoff points for each risk factor, this study explored the probability that an overweight Hispanic adolescent will have high
BP and abnormal levels of FPG, IS, TC, TG, LDL, and HDL, as well as their likelihood of having associated clustering of risk factors. It was found that overweight adolescents were more likely to have adverse levels of IS, TG, and LDL after controlling for possible confounding factors. Also, risk factors tended to cluster in overweight adolescents, who were more likely to have one and two or more risk factors for T2DM and CVD compared to having zero, and more likely to have two or more risk factors compared to having just one. Odds ratios and 95% confidence intervals for the likelihood of having one and two or more risk factors compared to having zero (one vs. zero and two ⫹ vs. zero) were relatively large, and this might be attributable to the small sample size. Despite that, our results are in agreement with other studies of other ethnicities that used larger sample sizes [23,34]. Unlike these studies, our study was done in Hispanic adolescents, who belong to one of the fastest growing minority group in the United States. In addition, although the few studies assessing the relationship of overweight with risk factors for T2DM and CVD in Hispanic adolescents have mainly involved Mexican Americans, our study is distinctive, as it comprised Hispanic adolescents of various origins, including Cuban Americans, who are the largest group in the area and also the least studied. It is worth noting that risk factors for T2DM and CVD were not limited to overweight adolescents. Of the nonoverweight adolescents, 28% had presence of one risk factor and 10% had presence of two or more risk factors. This might be due to the influence that ethnicity and family history of disease play in the development of T2DM and CVD. Based on ADA criteria, being Hispanic and having positive family history of diabetes in a first- or second-degree relative, constitute risk factors for the development of the disease [35]. Of the nonoverweight adolescents, 49% had family history of T2DM and 47% had family history of CVD.
Figure 2. Percentage of overweight and nonoverweight Hispanic adolescents with zero, one, two, and three or more components of the metabolic syndrome. Note: components of the metabolic syndrome were abdominal obesity (W/Ht ratio ⬎0.5), high TG (TG ⱖ130 mg/dL), low HDL (HDL ⬍35 mg/dL), high BP (either SBP or DBP ⬎95th percentile for age, gender, and height), and IFG (FPG ⱖ100 mg/dL or ⬍126 mg/dL).
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
In this study, various risk factors for T2DM and CVD were defined using published cutoff points. In addition, it was demonstrated that W/Ht ratio ⬎0.5, a measure of central obesity, was associated with several of these risk factors (BP, IS, TC, TG, and LDL). In agreement with the general definition of the MetS adopted in adults and in the pediatric population (using ATP III criteria), we included in our definition the general features of the MetS (central obesity, hypertension, dyslipidemia, and impaired glucose metabolism). We found that 9% of the entire sample met the proposed definition criteria for the MetS, and these were all overweight adolescents. Similar to Cook et al [10] and de Ferranti et al [12], abdominal obesity, low HDL, and high TG were the most predominant components of the MetS in this population. Among overweight adolescents, 98% had central obesity. Interestingly, 4% of nonoverweight adolescents also had central obesity. Although this is a low percentage, it confirms the insensitivity of BMI to measure fat distribution [36] and correctly identify those truly at risk. Cook et al [10] was one of the first research teams estimating the prevalence of the MetS in U.S. adolescents, using data from NHANES III (1988 –1994). They used the following criteria for the different components of the MetS: abdominal obesity (WC ⱖ90th percentile for age and gender), TG ⱖ110 mg/dL, HDL ⱕ40 mg/dL, BP ⱖ90th percentile (for age, gender, and height), and fasting glucose ⱖ110 mg/dL. They found that about 4% of all U.S. adolescents and approximately 28.7% of overweight adolescents (BMI ⱖ95th percentile) met their criteria for MetS. This same population was evaluated by de Ferranti et al [12] using different criteria. The different components of the MetS were defined as: abdominal obesity (WC ⬎75th percentile for age and gender), TG ⱖ100 mg/dL (ⱖ1.1 mmol/L), HDL ⬍45 mg/dL (⬍1.2 mmol/L) for boys aged 15–19 years and HDL ⬍50 mg/dL (⬍1.3 mmol/L) for the rest, SBP ⬎90th percentile (for age, gender, and height), and fasting glucose ⱖ110 mg/dL (’6.1 mmol/L). They found that about 9.2% of all U.S. adolescents and 31.2% of adolescents with BMI ⱖ85th percentile (at risk of overweight or overweight) met their criteria for MetS. Cook et al [10] used more limited cutoff points for the definition of the MetS, and hence, the prevalence was lower than the one found by de Ferranti et al [12]. These two studies clearly capture the problem that rises from the lack of a clear definition of the MetS in children and adolescents. Two recent studies have proposed definition criteria of the MetS in children and adolescents [37,38]. However, cutoff points for risk factors have differed between the two studies, emphasizing the importance of a consistent and clear definition. It is worth mentioning that we did not attempt to propose a definition of the MetS for adolescents. Our primary goal was to identify those Hispanic adolescents at increased risk for T2DM and CVD, using as baseline the ATP III criteria definition of MetS adopted for adults. Some of the studies evaluating the associations of being overweight and adverse risk for T2DM and CVD have used
449
cutoff points for risk factors different from the ones used in our study. We used cutoff points similar to those in Freedman et al [23], with the difference being that IS ⬍100% was used instead of insulin levels ⱖ95th percentile. The use of BMI as a dichotomous variable and the use of the term “overweight” can also differ across studies. However, independently of the classification system used, overweight children and adolescents have been shown to be at greater risk for T2DM and CVD in various studies [23,32,33], as well as in this study. In addition, the limitations of BMI as to discriminate between fat mass and fat-free mass are well recognized. However, it seems that the association of BMI with body fatness is much stronger at higher BMI levels than it is at lower levels [39]. Although the Center for Disease Control’s criteria to define overweight in children and adolescents is not universally accepted, a BMI level above the 95th percentile has been shown to be a good predictor of body fatness [40]. BMI is an inexpensive method that can easily be obtained in any clinical setting, and in our study a strong association with risk factors for T2DM and CVD at levels above the 95th percentile was clearly evident. This study joins in the efforts made by other researchers to understand the risk factors that contribute to the onset of T2DM and CVD in the pediatric population. However, the cross-sectional design of the study, which measures exposure and outcome at the same time, not only limits its utility for studying the etiology of the risk factors for T2DM and CVD, but also confounds the temporal association between the predictors and the outcomes. Also, the study used a convenience sampling method, which limits generalization of our findings to the entire Hispanic adolescent population (external validity). However, given its descriptive nature, this type of study is able to provide quantitative estimates of the magnitude of a health problem like risk factors for T2DM and CVD in Hispanic adolescents, as well as important characteristics of the study population. In addition, it can serve to generate new hypotheses that can be tested in future studies and assist in planning interventional health programs. References [1] Centers for Disease Control and Prevention. National Diabetes Fact Sheet: General Information and National Estimates on Diabetes in the United States, 2005. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2005. [2] Fagot-Campagna A. Emergence of type 2 diabetes mellitus in children: epidemiological evidence. J Pediatr Endocrinol Metab 2000; 13(Suppl 6):1395– 402. [3] National Center for Health Statistics. Health, United States, 2006, With Chartbook on Trends in the Health of Americans. Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2006. [4] Hu FB, Stampfer MJ, Haffner SM, et al. Elevated risk of cardiovascular disease prior to clinical diagnosis of type 2 diabetes. Diabetes Care 2002;25:1129 –34.
450
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
[5] Fagot-Campagna A, Pettitt DJ, Engelgau MM, et al. Type 2 diabetes among North American children and adolescents: an epidemiologic review and a public health perspective. J Pediatr 2000;136:664 –72. [6] Tortolero SR, Goff DC Jr, Nichaman MZ, et al. Cardiovascular risk factors in Mexican-American and non-Hispanic white children: The Corpus Christi Child Heart Study. Circulation 1997;96:418 –23. [7] Laaksonen DE, Lakka HM, Niskanen LK, et al. Metabolic syndrome and development of diabetes mellitus: application and validation of recently suggested definitions of the metabolic syndrome in a prospective cohort study. Am J Epidemiol 2002;156:1070 –7. [8] Jones KL. The dilemma of the metabolic syndrome in children and adolescents: disease or distraction? Pediatr Diabetes 2006;7:311–21. [9] National Cholesterol Education Program. 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). National Heart, Lung, and Blood Institute. National Institutes of Health. NIH Publication No. 02-5215, September 2002. [10] Cook S, Weitzman M, Auinger P, et al. Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 1988 –1994. Arch Pediatr Adolesc Med 2003;157:821–7. [11] Cruz ML, Weigensberg MJ, Huang TT, et al. The metabolic syndrome in overweight Hispanic youth and the role of insulin sensitivity. J Clin Endocrinol Metab 2004;89:108 –13. [12] de Ferranti SD, Gauvreau K, Ludwig DS, et al. Prevalence of the metabolic syndrome in American adolescents: findings from the Third National Health and Nutrition Examination Survey. Circulation 2004;110:2494 –7. [13] Weiss R, Dziura J, Burgert TS, et al. Obesity and the metabolic syndrome in children and adolescents. N Engl J Med 2004 350: 2362–74. [14] U.S. Census Bureau. The American Community—Hispanics: 2004. American Community Survey Reports. U.S. Department of Commerce. Economics and Statistics Administration. U.S. Census Bureau, 2007. Available at http://www.census.gov/prod/2007pubs/ acs-03.pdf. Accessed November 28, 2007. [15] Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC growth charts for the United States: Methods and development. National Center for Health Statistics. Vital Health Stat 11(246), 2002. [16] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2005;28(Suppl 1):S37– 42. [17] The University of Oxford. HOMA Calculator, 2004. The Oxford Centre for Diabetes Endocrinology & Metabolism. Diabetes Trials Unit. Downloaded on June 16, 2005. Available at http://www.dtu. ox.ac.uk/. [18] Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21:2191–2. [19] Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004;27:1487–95. [20] Keskin M, Kurtoglu S, Kendirci M, et al. Homeostasis model assessment 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 –3. [21] Lee JM, Okumura MJ, Davis MM, et al. Prevalence and determinants of insulin resistance among U.S. adolescents: a population-based study. Diabetes Care 2006;29:2427–32. [22] American Academy of Pediatrics. National Cholesterol Education Program: report of the expert panel on blood cholesterol levels in children and adolescents. Pediatrics 1992;89:525– 84. [23] Freedman DS, Dietz WH, Srinivasan SR, et al. The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics 1999;103:1175– 82.
[24] National Cholesterol Education Program Expert Panel on Blood Cholesterol Levels in Children and Adolescents. National Cholesterol Education Program (NCEP): highlights of the report of the expert panel on blood cholesterol levels in children and adolescents. Pediatrics 1992;89:495–501. [25] National Health and Nutrition Examination Survey III. Cycle 2. Physician Examiner’s Training Manual. Westat, Inc., 1991. Retrieved October 13, 2004, from http://www.cdc.gov/nchs/data/nhanes/ nhanes3/cdrom/nchs/manuals/phys.pdf. [26] National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents [NHBPEP]. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics 2004;114:555–76. [27] Hsieh SD, Yoshinaga H. Waist/height ratio as a simple and useful predictor of coronary heart disease risk factors in women. Intern Med 1995;34:1147–52. [28] Ashwell M, Lejeune S, McPherson K. Ratio of waist circumference to height may be better indicator of need for weight management. BMJ 1996;312:377. [29] Li C, Ford ES, Mokdad AH, et al. Recent trends in waist circumference and waist-height ratio among US children and adolescents. Pediatrics 2006;118:e1390 – 8. [30] Moran A, Jacobs DR Jr, Steinberger J, et al. Insulin resistance during puberty: results from clamp studies in 357 children. Diabetes 1999; 48:2039 – 44. [31] Steinberger J, Daniels SR; American Heart Association Atherosclerosis, Hypertension, and Obesity in the Young Committee (Council on Cardiovascular Disease in the Young); American Heart Association Diabetes Committee (Council on Nutrition, Physical Activity, and Metabolism). Obesity, insulin resistance, diabetes, and cardiovascular risk in children: an American Heart Association scientific statement from the Atherosclerosis, Hypertension, and Obesity in the Young Committee (Council on Cardiovascular Disease in the Young) and the Diabetes Committee (Council on Nutrition, Physical Activity, and Metabolism). Circulation 2003;107:1448 –53. [32] Sharp TA, Grunwald GK, Giltinan KE, et al. Association of anthropometric measures with risk of diabetes and cardiovascular disease in Hispanic and Caucasian adolescents. Prev Med 2003;37:611– 6. [33] Srinivasan SR, Myers L, Berenson GS. Rate of change in adiposity and its relationship to concomitant changes in cardiovascular risk variables among biracial (black-white) children and young adults: The Bogalusa Heart Study. Metabolism 2001;50:299 –305. [34] Freedman DS, Mei Z, Srinivasan SR, et al. Cardiovascular risk factors and excess adiposity among overweight children and adolescents: the Bogalusa Heart Study. J Pediatr 2007;150:12–17.e2. [35] American Diabetes Association. Type 2 diabetes in children and adolescents. Diabetes Care 2000;23:381–9. [36] Kahn HS, Imperatore G, Cheng YJ. A population-based comparison of BMI percentiles and waist-to-height ratio for identifying cardiovascular risk in youth. J Pediatr 2005;146:482– 8. [37] Joliffe CJ, Janssen I. Development of age-specific metabolic syndrome criteria that are linked to the Adult Treatment Panel III and International Diabetes Federation Criteria. J Am Coll Cardiol 2007; 49:891– 8. [38] Zimmet P, Alberti G, Kaufman F, et al. International Diabetes Federation Task Force on Epidemiology and Prevention of Diabetes. The metabolic syndrome in children and adolescents. Lancet 2007; 369(9579):2059 – 61. [39] Bray GA, DeLany JP, Harsha DW, et al. Evaluation of body fat in fatter and leaner 10-y-old African American and white children: the Baton Rouge Children’s Study. Am J Clin Nutr 2001;73:687–702. [40] Freedman DS, Ogden CL, Berenson GS, et al. Body mass index and body fatness in childhood. Curr Opin Clin Nutr Metab Care 2005;8: 618 –23.
Original article
Risk Factors for Type 2 Diabetes and Cardiovascular Diseases in Hispanic Adolescents Gianna Perez Gomez, Ph.D. and Fatma G. Huffman, Ph.D.* Department of Dietetics and Nutrition, Florida International University, Miami, Florida Manuscript received December 6, 2007; manuscript accepted March 18, 2008
Abstract
Purpose: To investigate the associations of overweight with risk factors for type 2 diabetes (T2DM), cardiovascular diseases (CVD), and presence of metabolic syndrome (MetS) in Hispanic adolescents. Methods: A total of 100 adolescents (50 overweight, 50 nonoverweight) aged 12 to 16 years were included. Fasting plasma glucose (FPG), insulin sensitivity (IS), total cholesterol (TC), triacylglycerols (TG), LDL, HDL, and blood pressure (BP) were evaluated as risk factors for T2DM and CVD. IS was determined using the HOMA calculator. Abnormal levels were: FPG ⱖ100 and ⬍126 mg/dL (impaired fasting glucose [IFG]), IS ⬍100%, TC ⱖ200 mg/dL, TG ⱖ130 mg/dL, LDL ⱖ130 mg/dL, HDL ⬍35 mg/dL, and BP ⱖ95th percentile. MetS was defined as having equal to or more than three of the following: abdominal obesity (waist-to-height ratio ⬎0.5), high BP, high TG, low HDL, and IFG. Results: Overweight adolescents had higher TC, TG, and LDL, and lower IS, ps ⬍ .001. They were more likely to have abnormal levels of IS, TG, and LDL, one and two or more risk factors compared to having zero, and two or more risk factors compared to having one, ps ⬍ .05. Only 9% of adolescents met the criteria of MetS, and they were all overweight. Abdominal obesity (51%), low HDL (21%), and high TG (16%) were the most predominant components of the MetS. Conclusions: Overweight Hispanic adolescents are at increased risk of developing T2DM or CVD. Preventive programs targeting Hispanic adolescents are needed to manage the overweight epidemic and its consequences. © 2008 Society for Adolescent Medicine. All rights reserved.
Keywords:
Hispanic adolescents; Risk factors; Cardiovascular diseases; Diabetes; Metabolic syndrome
Type 2 diabetes mellitus (T2DM) was for many years considered a disease of the middle aged. However, its incidence has been increasing among children of populations at high risk, such as Hispanics [1,2]. In addition, overweight, one of the main risk factors for T2DM, has more than doubled in the last 2 decades in Hispanic adolescents [3]. One of the most life-threatening complications of T2DM is heart disease, and studies have indicated that there is a Funding for this research was provided through an NIH-sponsored training grant (FREA), a Foundation grant from Florida International University, and a Florida International University Graduate School Dissertation Year Fellowship. *Address correspondence to: Fatma G. Huffman, Ph.D., Department of Dietetics and Nutrition, Florida International University, 11200 SW 8th Street, HLS II-450, Miami, FL 33199. E-mail address: [email protected]
high cardiovascular risk prior to the clinical diagnosis of T2DM [4]. Type 2 diabetes in children is usually diagnosed after age 10. These patients have a high prevalence of overweight and other risk factors for cardiovascular diseases (CVD), with their body mass index (BMI) ranging from 26 to 38 kg/m2, the prevalence of high blood pressure (BP) ranging from 17% to 32%, and high triacylglycerols (TG) ranging from 4% to 32% [5]. In addition, ethnic differences in blood insulin and glucose, BMI, and other CVD risk factors occur as early as age 8 to 10 years, with Hispanic children having more unfavorable profiles than non-Hispanic white children [6]. The metabolic syndrome (MetS) is defined as the clustering of multiple risk factors that increase the risk of T2DM and CVD [7]. Components of the MetS are also found in children and adolescents [8]. However, a worldwide con-
1054-139X/08/$ – see front matter © 2008 Society for Adolescent Medicine. All rights reserved. doi:10.1016/j.jadohealth.2008.03.010
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
sensus for the definition of MetS has not been established in the pediatric population. The National Cholesterol Education Program Adult Treatment Panel III (ATP III) identifies five components of the MetS in adults, and defines it as having three or more of them: abdominal obesity, high TG, low high-density lipoprotein cholesterol (HDL), hypertension, and impaired glucose metabolism [9]. Researchers have modified the adult criteria in different ways to create their own definition for children and adolescents [10 –13]. All of the definitions have been consistent with the general features of the MetS in adults, including a combination of central obesity, hypertension, dyslipidemia, and impaired glucose metabolism. Hispanics represent 14.2% of U.S. households [14]. The increasing prevalence of overweight in the Hispanic pediatric population will represent a major challenge to our society and the healthcare system. Thus, this study responds to the need for more information on the epidemiology of T2DM and CVD in the pediatric population by exploring the associations of being overweight with risk factors for these diseases. In addition, because of the importance of the MetS as a prelude to T2DM and CVD in the adult population, the study also explored presence of MetS components in this population. This study is important for four reasons: first, T2DM is increasing in children and adolescents of all ethnic backgrounds; second, there is limited amount of information regarding the epidemiology of the disease in the pediatric population; third, Hispanics are at increased risk for T2DM; and fourth, there is evidence that individuals at risk for T2DM are also at risk for CVD. Methods Inclusion criteria Included were apparently healthy Hispanic adolescents 12 to 16 years of age, who signed an assent form and whose parents signed a consent form. Exclusion criteria Adolescents receiving hormonal treatment or any other drug (i.e., steroids) that could confound the results of the biochemical parameters, adolescents with chronic diseases like diabetes, CVD, inherited lipid disorders, etc., underweight (BMI-for-age ⬍5th percentile) adolescents or any other disease that could impede their participation in the study were not included. At risk of overweight (BMI-forage ⱖ85th to ⬍95th percentile) adolescents were not included in the study to better contrast overweight versus nonoverweight. Study population This was a cross-sectional study comprising of 117 adolescents of various Hispanic origins. Of those, 57 were overweight (BMI-for-age ⱖ95th percentile) and 60 were
445
nonoverweight (BMI-for-age ⬍85th percentile). Subjects were selected by convenience sampling technique from three different Pediatric clinics in Miami–Dade County, South Florida, until a minimum of 50 subjects that completed the study was reached in each group. All subjects meeting the inclusion criteria were approached and invited to participate in the study. Subjects invited to participate in the study were apparently healthy children who had gone to the clinic for a regular well-check or vaccination for school purposes. A total of 14 subjects (5 overweight, 9 nonoverweight) did not complete the biochemical analyses and were not included in the statistical analyses (listwise deletion). Of the subjects that completed the study, three were excluded for meeting the exclusion criteria after the biochemical analyses were received. One overweight subject had TG levels ⬎900 mg/dL and was diagnosed with Familial Hyperlipidemia afterward. Another overweight subject met the diagnosis criteria for diabetes (FPG ⬎126 mg/dL). The nonoverweight subject was taking Accutane (can cause hyperlipidemia and increased glucose levels as side effects), but this was reported to us after the results of the biochemical analyses were received. The study was approved by Florida International University institutional review board for the use of human subjects. Parental consent and child assent were both needed and obtained through institutional review board-approved consent and assent forms given to parents and adolescents, respectively. Assessment and measures Demographic questionnaires were given to parents and adolescents to complete. The physician and/or the medical assistant from the clinic where data were collected conducted all physical examinations, which included assessment of Tanner stage, measure of BP, and anthropometrics (weight, height, and waist circumference). A certified phlebotomist collected blood samples after an overnight fasting of no less than 12 hours for the biochemical analyses. BMI was calculated from weight and height and plotted in the Center for Disease Control’s gender- and age-specific growth charts [15]. Risk factors for T2DM and CVD. Fasting plasma glucose (FPG) (mg/dL) was classified according to the American Diabetes Association (ADA) as normal (FPG ⬍100 mg/ dL), impaired fasting glucose (IFG) (FPG ⱖ100 mg/dL and ⬍126 mg/dL), and provisional diagnosis of diabetes (FPG ⱖ126 mg/dL) [16]. Insulin sensitivity (IS) (%) was determined from FPG and insulin levels (IU/mL) using the HOMA Calculator© (The University of Oxford) [17]. This measure corresponds well to other estimates of -cell function and IS derived from hyperinsulinaemic and hyperglycemic clamps and intravenous and oral glucose tolerance tests [18]. The HOMA Calculator uses the HOMA2 model to estimate -cell function (%B) and IS (%S) for an indi-
446
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
vidual [19]. The HOMA approach has been used in research to determine IS and insulin resistance index in children and adolescents [20,21]. Normal IS was defined as IS ⱖ100% and low IS as IS ⬍100% [19]. High total cholesterol (TC) was considered as ⱖ200 mg/dL, high TG as ⱖ130 mg/dL, high LDL as ⱖ130 mg/ dL, and low HDL as ⬍35 mg/dL [9,22–24]. Measurement of BP (mmHg) followed the guidelines in the Physician Examiner’s Training Manual and used a standard clinical sphygmomanometer (auscultatory measurement) and appropriate size cuff [25]. Values of BP were plotted in the gender-, age-, and height-specific standard tables to find the precise percentile for systolic BP (SBP) and diastolic BP (DBP), that were published by the National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents [26]. Classification of BP followed the National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents definition: high BP (SBP and/or DBP ⱖ95th percentile), high-normal BP (SBP or DBP ⱖ90th and ⬍95th percentile), and normal BP (SBP and DBP ⬍90th percentile). Adolescents with BP ⱖ120/80 mmHg were also considered as having high-normal BP. Clustering of risk factors was defined as having greater than or equal two of the following: IFG, low IS, high TG, high LDL, low HDL, and high BP. Because TC contains HDL and LDL, it was excluded from the analysis of clustering of risk factors. Definition of metabolic syndrome. In agreement with the general definition of the MetS in adults using ATP III criteria [16], we included in our definition the general features of the MetS (central obesity, hypertension, dyslipidemia, and impaired glucose metabolism). There are no guidelines for assessing obesity-related risk factors in children and adolescents by using waist circumference (WC). Waistto-height ratio (W/Ht ratio), a predictor of cardiovascular risk in the adult population [27], was used in our study to determine abdominal obesity. A cutoff value of 0.5 established for adults [28] and also explored in children and adolescents [29] was used in our study to define abdominal obesity. We defined the MetS in our sample as having three or more of the following risk factors: abdominal obesity (W/Ht ratio ⬎0.5), high BP (either SBP or DBP ⱖ95th percentile), high TG (TG ⱖ130 mg/dL), low HDL (HDL ⬍35 mg/dL), and IFG. Data analysis Data analyses used SPSS 14.0 for Windows (SPSS Inc., Chicago, IL) and included descriptive statistics, logistic regression, multinomial logistic regression, two sample means tests, correlations, and chi-square. All variables were first tested for normality. All tests were considered statistically significant if p ⬍ .05. In all the analyses, gender,
family history of T2DM, and family history of CVD were controlled. Due to the importance of sexual development in puberty-related insulin resistance [30], Tanner stage was controlled in all the analyses of IS. Results A total of 100 adolescents (50 overweight, 50 nonoverweight) that completed the study were included in the statistical analyses for the hypotheses testing. Of those, 44% were Cuban descendants, 15% were Nicaraguan descendants, 13% had combined descendant, and the remaining % Table 1 General characteristics of the study subjects by BMI category BMI category*
p
Characteristics
Overweight
Nonoverweight
n Male/female (n) Age (years) Tanner stage† Category 2 3 4 5 M (⫾SD) Cuban descendant Born in the U.S. Family history of: Obesity T2DM CVD Gestational diabetes Blood pressure‡ Normal High-normal High FPG (mg/dL) IS (%) TC (mg/dL) TG (mg/dL) LDL (mg/dL) HDL (mg/dL)
50 26/24 14.1 (⫾1.3)
50 26/24 14.4 (⫾1.5)
.151
4% 34% 46% 16% 3.7 (⫾0.8) 34% 72%
4% 38% 42% 16% 3.7 (⫾0.8) 54% 34%
.977 .767 .044 ⬍.001
74% 51% 53% 8%
42% 49% 47% 4%
36% 46% 18% 90 (⫾11) 69 (⫾38) 167 (⫾33) 106 (⫾49) 105 (⫾28) 44 (⫾11)
92% 8% 0% 88 (⫾6) 170 (⫾68) 143 (⫾20) 66 (⫾31) 87 (⫾18) 46 (⫾10)
.001 .841 .362 .400
⬍.001 .812 ⬍.001 ⬍.001 ⬍.001 ⬍.001 .100
Note. Values are n, M (⫾SD), and %. Mann-Whitney U test was conducted for age, FPG, IS, TC, TG, LDL, and HDL and 2 test for Tanner stage, Cuban descendant, born in the U.S. family history, and blood pressure. FPG ⫽ fasting plasma glucose, IS ⫽ insulin sensitivity, TC ⫽ total cholesterol, TG ⫽ triacylglycerols, LDL ⫽ low-density lipoprotein cholesterol, HDL ⫽ high-density lipoprotein cholesterol. BP ⫽ either SBP or DBP ⱖ90th and ⬍95th percentile, or BP ⱖ120/80 mmHg, High BP ⫽ either SBP or DBP ⱖ95th percentile. * Overweight ⫽ BMI ⱖ95th percentile for age and gender, nonoverweight ⫽ BMI ⬍85th percentile for age and gender. † Tanner stage ranges from 1 (prepubertal) to 5 (adult like). ‡ High-normal and high blood pressure (BP) were combined into one category to conduct the 2 test due to small cell counts. Normal BP ⫽ SBP and DBP ⬍90th percentile, high-normal BP ⫽ either SBP or DBP ⱖ90th and ⬍95th percentile, or BP ⱖ120/80 mmHg, high BP ⫽ either SBP or DBP ⱖ95th percentile.
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
447
Table 2 Likelihood of overweight adolescents of having adverse levels for each of the risk factors for T2DM and CVD Risk factor
Overweight (%)
Nonoverweight (%)
p
OR
95% CI
IFG Low IS High TG High LDL Low HDL High BP
18 84 28 20 22 18
4 18 4 2 20 0
.125 ⬍.001 .011 .015 .513 .996
2.45 25.78 8.89 72.10 0.96 —
0.69 to 20.11 7.79 to 85.32 1.65 to 47.90 2.32 to 2243.09 0.33 to 2.76 — to —
Note. This analysis used logistic regression and controlled for gender, family history of T2DM, and family history of CVD. For Low IS, Tanner stage was also controlled. IFG (FPG ⱖ100 mg/dL and ⬍126 mg/dL), model 2(4, N ⫽ 100) ⫽ 7.45, p ⫽ .11. Low IS (IS ⬍100%), model 2(5, N ⫽ 100) ⫽ 48.43, p ⬍ .001. High TG (TG ⱖ130 mg/dL), model 2(4, N ⫽ 100) ⫽ 19.695, p ⫽ .001. High LDL (LDL ⱖ130 mg/dL), model 2(4, N ⫽ 100) ⫽ 12.20, p ⫽ .02. Low HDL (HDL ⬍35 mg/dL), model 2(4, N ⫽ 100) ⫽ .44, p ⫽ .98. High BP (either SBP or DBP ⱖ95th percentile for age, gender, and height), the test statistic was unstable due to large standard error; model 2(5, N ⫽ 100) ⫽ 13.50, p ⫽ .01.
were of other Hispanic origins. Table 1 summarizes characteristics of the study subjects by BMI category. No statistically significant differences were found for age or Tanner stage between overweight and nonoverweight adolescents, ps ⬎ .05. Family history of T2DM, CVD, and gestational diabetes were not associated with being overweight, ps ⬎ .05. Family history of obesity was significantly associated with being overweight, p ⫽ .001. A higher percentage of nonoverweight adolescents were of Cuban ancestry, p ⫽ .04. A higher percentage of adolescents born in the United States were overweight, p ⬍ .001. Blood pressure was significantly associated with being overweight, p ⬍ .001. Overweight adolescents had higher TC, TG, and LDL, and lower IS, than nonoverweight adolescents, ps ⬍ .001. A greater percentage of overweight adolescents had IFG, low IS, high TG, high LDL, and high BP, ps ⬍ .05. Table 2 presents the likelihood of overweight adolescents of having adverse levels for each of the risk factors, after controlling for gender and family history of T2DM and CVD. Overweight adolescents were more likely to have low IS, high TG, and high LDL, ps ⬍ .05. IFG and high BP were not significant after controlling for these confounding factors. A higher percentage of overweight adolescents had one and two or more risk factors for T2DM and CVD compared to nonoverweight adolescents, 2(2, N ⫽ 100) ⫽ 43.84, p ⬍ .001. Table 3 presents the likelihood of overweight adolescents of having one and two or more risk factors compared to having zero, and the likelihood of having two or more risk factors compared to having one, after controlling for gender, family history of T2DM, and CVD. Overweight adolescents were more likely to have one and two or more risk factors compared to having zero, ps ⬍ .001. In addition, overweight adolescents were more likely to have two or more risk factors compared to having one, p ⫽ .03. Mean W/Ht ratio was 0.6 (⫾0.1) for overweight and 0.4
(⫾0.04) for nonoverweight adolescents, z ⫽ ⫺8.55, p ⬍ .001. A total of 98% of overweight and 4% of nonoverweight adolescents had W/Ht ratio ⬎0.5, 2(1, N ⫽ 100) ⫽ 88.4, p ⬍ .001. Adolescents with higher W/Ht ratio or W/Ht ratio ⬎0.5 had lower IS and higher BP percentiles, TC, TG, and LDL, ps ⬍ .05. Only 9% of the adolescents met the criteria definition of the MetS, and they were all overweight. Abdominal obesity (51%), low HDL (21%), and high TG (16%) were the most predominant components of the MetS for the entire sample. Figure 1 shows percentage of Hispanic adolescents having the individual components of the MetS for each group, and Figure 2 shows percentage of overweight and nonoverweight adolescents with zero, one, two, and three or more components of the MetS. Discussion The problem of overweight children is worsening at a dramatic rate in the United States for all ethnicities [3], including Hispanics, who are one of the fastest growing minority group. Excess body fat carries additional conse-
Table 3 Likelihood of overweight adolescents of having one or two or more risk factors for T2DM and CVD Number of risk factors*

SE
Wald
p
OR
95% CI
1 vs. 0 2 ⫹ vs. 0 2 ⫹ vs. 1
3.42 4.75 1.34
0.89 0.93 0.63
14.59 26.38 4.53
⬍.001 ⬍.001 .033
30.42 115.79 3.81
5.27 to 175.43 18.89 to 709.78 1.11 to 13.04
Note. Multinomial logistic regression was used, controlling for gender, family history of T2DM, and family history of CVD. Model 2(8, N ⫽ 100) ⫽ 65.86, p ⬍ .001. * Overweight adolescents: 4% had 0 risk factors, 36% had one risk factor, and 60% had two or more risk factors; nonoverweight adolescents: 62% had 0 risk factors, 28% had one risk factor, and 10% had two or more risk factors.
448
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
Figure 1. Percentage of overweight and nonoverweight Hispanic adolescents with individual components of the metabolic syndrome.
quences that increase the risk for developing chronic diseases like T2DM and CVD [31]. Addressing the question of whether overweight Hispanic adolescents have an increased risk for these chronic diseases, this study explored the associations of being overweight with risk factors for T2DM and CVD. Significant differences were found for BP, TC, TG, LDL, and IS between overweight and nonoverweight adolescents, with overweight adolescents always having the worst profiles. These findings suggest that overweight Hispanic adolescents are at increased risk of developing T2DM or CVD. These data are consistent with the results of other studies that found an increased risk for T2DM and CVD in overweight adolescents of other ethnicities [32,33]. Most of the studies in adolescents evaluate the linear associations between BMI and risk factors for T2DM and CVD. Only few studies have assessed the likelihood of overweight adolescents of having abnormal levels of each of the risk factors and their likelihood of having clustering of multiple risk factors [23]. By using published cutoff points for each risk factor, this study explored the probability that an overweight Hispanic adolescent will have high
BP and abnormal levels of FPG, IS, TC, TG, LDL, and HDL, as well as their likelihood of having associated clustering of risk factors. It was found that overweight adolescents were more likely to have adverse levels of IS, TG, and LDL after controlling for possible confounding factors. Also, risk factors tended to cluster in overweight adolescents, who were more likely to have one and two or more risk factors for T2DM and CVD compared to having zero, and more likely to have two or more risk factors compared to having just one. Odds ratios and 95% confidence intervals for the likelihood of having one and two or more risk factors compared to having zero (one vs. zero and two ⫹ vs. zero) were relatively large, and this might be attributable to the small sample size. Despite that, our results are in agreement with other studies of other ethnicities that used larger sample sizes [23,34]. Unlike these studies, our study was done in Hispanic adolescents, who belong to one of the fastest growing minority group in the United States. In addition, although the few studies assessing the relationship of overweight with risk factors for T2DM and CVD in Hispanic adolescents have mainly involved Mexican Americans, our study is distinctive, as it comprised Hispanic adolescents of various origins, including Cuban Americans, who are the largest group in the area and also the least studied. It is worth noting that risk factors for T2DM and CVD were not limited to overweight adolescents. Of the nonoverweight adolescents, 28% had presence of one risk factor and 10% had presence of two or more risk factors. This might be due to the influence that ethnicity and family history of disease play in the development of T2DM and CVD. Based on ADA criteria, being Hispanic and having positive family history of diabetes in a first- or second-degree relative, constitute risk factors for the development of the disease [35]. Of the nonoverweight adolescents, 49% had family history of T2DM and 47% had family history of CVD.
Figure 2. Percentage of overweight and nonoverweight Hispanic adolescents with zero, one, two, and three or more components of the metabolic syndrome. Note: components of the metabolic syndrome were abdominal obesity (W/Ht ratio ⬎0.5), high TG (TG ⱖ130 mg/dL), low HDL (HDL ⬍35 mg/dL), high BP (either SBP or DBP ⬎95th percentile for age, gender, and height), and IFG (FPG ⱖ100 mg/dL or ⬍126 mg/dL).
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
In this study, various risk factors for T2DM and CVD were defined using published cutoff points. In addition, it was demonstrated that W/Ht ratio ⬎0.5, a measure of central obesity, was associated with several of these risk factors (BP, IS, TC, TG, and LDL). In agreement with the general definition of the MetS adopted in adults and in the pediatric population (using ATP III criteria), we included in our definition the general features of the MetS (central obesity, hypertension, dyslipidemia, and impaired glucose metabolism). We found that 9% of the entire sample met the proposed definition criteria for the MetS, and these were all overweight adolescents. Similar to Cook et al [10] and de Ferranti et al [12], abdominal obesity, low HDL, and high TG were the most predominant components of the MetS in this population. Among overweight adolescents, 98% had central obesity. Interestingly, 4% of nonoverweight adolescents also had central obesity. Although this is a low percentage, it confirms the insensitivity of BMI to measure fat distribution [36] and correctly identify those truly at risk. Cook et al [10] was one of the first research teams estimating the prevalence of the MetS in U.S. adolescents, using data from NHANES III (1988 –1994). They used the following criteria for the different components of the MetS: abdominal obesity (WC ⱖ90th percentile for age and gender), TG ⱖ110 mg/dL, HDL ⱕ40 mg/dL, BP ⱖ90th percentile (for age, gender, and height), and fasting glucose ⱖ110 mg/dL. They found that about 4% of all U.S. adolescents and approximately 28.7% of overweight adolescents (BMI ⱖ95th percentile) met their criteria for MetS. This same population was evaluated by de Ferranti et al [12] using different criteria. The different components of the MetS were defined as: abdominal obesity (WC ⬎75th percentile for age and gender), TG ⱖ100 mg/dL (ⱖ1.1 mmol/L), HDL ⬍45 mg/dL (⬍1.2 mmol/L) for boys aged 15–19 years and HDL ⬍50 mg/dL (⬍1.3 mmol/L) for the rest, SBP ⬎90th percentile (for age, gender, and height), and fasting glucose ⱖ110 mg/dL (’6.1 mmol/L). They found that about 9.2% of all U.S. adolescents and 31.2% of adolescents with BMI ⱖ85th percentile (at risk of overweight or overweight) met their criteria for MetS. Cook et al [10] used more limited cutoff points for the definition of the MetS, and hence, the prevalence was lower than the one found by de Ferranti et al [12]. These two studies clearly capture the problem that rises from the lack of a clear definition of the MetS in children and adolescents. Two recent studies have proposed definition criteria of the MetS in children and adolescents [37,38]. However, cutoff points for risk factors have differed between the two studies, emphasizing the importance of a consistent and clear definition. It is worth mentioning that we did not attempt to propose a definition of the MetS for adolescents. Our primary goal was to identify those Hispanic adolescents at increased risk for T2DM and CVD, using as baseline the ATP III criteria definition of MetS adopted for adults. Some of the studies evaluating the associations of being overweight and adverse risk for T2DM and CVD have used
449
cutoff points for risk factors different from the ones used in our study. We used cutoff points similar to those in Freedman et al [23], with the difference being that IS ⬍100% was used instead of insulin levels ⱖ95th percentile. The use of BMI as a dichotomous variable and the use of the term “overweight” can also differ across studies. However, independently of the classification system used, overweight children and adolescents have been shown to be at greater risk for T2DM and CVD in various studies [23,32,33], as well as in this study. In addition, the limitations of BMI as to discriminate between fat mass and fat-free mass are well recognized. However, it seems that the association of BMI with body fatness is much stronger at higher BMI levels than it is at lower levels [39]. Although the Center for Disease Control’s criteria to define overweight in children and adolescents is not universally accepted, a BMI level above the 95th percentile has been shown to be a good predictor of body fatness [40]. BMI is an inexpensive method that can easily be obtained in any clinical setting, and in our study a strong association with risk factors for T2DM and CVD at levels above the 95th percentile was clearly evident. This study joins in the efforts made by other researchers to understand the risk factors that contribute to the onset of T2DM and CVD in the pediatric population. However, the cross-sectional design of the study, which measures exposure and outcome at the same time, not only limits its utility for studying the etiology of the risk factors for T2DM and CVD, but also confounds the temporal association between the predictors and the outcomes. Also, the study used a convenience sampling method, which limits generalization of our findings to the entire Hispanic adolescent population (external validity). However, given its descriptive nature, this type of study is able to provide quantitative estimates of the magnitude of a health problem like risk factors for T2DM and CVD in Hispanic adolescents, as well as important characteristics of the study population. In addition, it can serve to generate new hypotheses that can be tested in future studies and assist in planning interventional health programs. References [1] Centers for Disease Control and Prevention. National Diabetes Fact Sheet: General Information and National Estimates on Diabetes in the United States, 2005. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2005. [2] Fagot-Campagna A. Emergence of type 2 diabetes mellitus in children: epidemiological evidence. J Pediatr Endocrinol Metab 2000; 13(Suppl 6):1395– 402. [3] National Center for Health Statistics. Health, United States, 2006, With Chartbook on Trends in the Health of Americans. Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2006. [4] Hu FB, Stampfer MJ, Haffner SM, et al. Elevated risk of cardiovascular disease prior to clinical diagnosis of type 2 diabetes. Diabetes Care 2002;25:1129 –34.
450
G. Perez Gomez and F.G. Huffman / Journal of Adolescent Health 43 (2008) 444 – 450
[5] Fagot-Campagna A, Pettitt DJ, Engelgau MM, et al. Type 2 diabetes among North American children and adolescents: an epidemiologic review and a public health perspective. J Pediatr 2000;136:664 –72. [6] Tortolero SR, Goff DC Jr, Nichaman MZ, et al. Cardiovascular risk factors in Mexican-American and non-Hispanic white children: The Corpus Christi Child Heart Study. Circulation 1997;96:418 –23. [7] Laaksonen DE, Lakka HM, Niskanen LK, et al. Metabolic syndrome and development of diabetes mellitus: application and validation of recently suggested definitions of the metabolic syndrome in a prospective cohort study. Am J Epidemiol 2002;156:1070 –7. [8] Jones KL. The dilemma of the metabolic syndrome in children and adolescents: disease or distraction? Pediatr Diabetes 2006;7:311–21. [9] National Cholesterol Education Program. 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). National Heart, Lung, and Blood Institute. National Institutes of Health. NIH Publication No. 02-5215, September 2002. [10] Cook S, Weitzman M, Auinger P, et al. Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 1988 –1994. Arch Pediatr Adolesc Med 2003;157:821–7. [11] Cruz ML, Weigensberg MJ, Huang TT, et al. The metabolic syndrome in overweight Hispanic youth and the role of insulin sensitivity. J Clin Endocrinol Metab 2004;89:108 –13. [12] de Ferranti SD, Gauvreau K, Ludwig DS, et al. Prevalence of the metabolic syndrome in American adolescents: findings from the Third National Health and Nutrition Examination Survey. Circulation 2004;110:2494 –7. [13] Weiss R, Dziura J, Burgert TS, et al. Obesity and the metabolic syndrome in children and adolescents. N Engl J Med 2004 350: 2362–74. [14] U.S. Census Bureau. The American Community—Hispanics: 2004. American Community Survey Reports. U.S. Department of Commerce. Economics and Statistics Administration. U.S. Census Bureau, 2007. Available at http://www.census.gov/prod/2007pubs/ acs-03.pdf. Accessed November 28, 2007. [15] Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC growth charts for the United States: Methods and development. National Center for Health Statistics. Vital Health Stat 11(246), 2002. [16] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2005;28(Suppl 1):S37– 42. [17] The University of Oxford. HOMA Calculator, 2004. The Oxford Centre for Diabetes Endocrinology & Metabolism. Diabetes Trials Unit. Downloaded on June 16, 2005. Available at http://www.dtu. ox.ac.uk/. [18] Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21:2191–2. [19] Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004;27:1487–95. [20] Keskin M, Kurtoglu S, Kendirci M, et al. Homeostasis model assessment 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 –3. [21] Lee JM, Okumura MJ, Davis MM, et al. Prevalence and determinants of insulin resistance among U.S. adolescents: a population-based study. Diabetes Care 2006;29:2427–32. [22] American Academy of Pediatrics. National Cholesterol Education Program: report of the expert panel on blood cholesterol levels in children and adolescents. Pediatrics 1992;89:525– 84. [23] Freedman DS, Dietz WH, Srinivasan SR, et al. The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics 1999;103:1175– 82.
[24] National Cholesterol Education Program Expert Panel on Blood Cholesterol Levels in Children and Adolescents. National Cholesterol Education Program (NCEP): highlights of the report of the expert panel on blood cholesterol levels in children and adolescents. Pediatrics 1992;89:495–501. [25] National Health and Nutrition Examination Survey III. Cycle 2. Physician Examiner’s Training Manual. Westat, Inc., 1991. Retrieved October 13, 2004, from http://www.cdc.gov/nchs/data/nhanes/ nhanes3/cdrom/nchs/manuals/phys.pdf. [26] National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents [NHBPEP]. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics 2004;114:555–76. [27] Hsieh SD, Yoshinaga H. Waist/height ratio as a simple and useful predictor of coronary heart disease risk factors in women. Intern Med 1995;34:1147–52. [28] Ashwell M, Lejeune S, McPherson K. Ratio of waist circumference to height may be better indicator of need for weight management. BMJ 1996;312:377. [29] Li C, Ford ES, Mokdad AH, et al. Recent trends in waist circumference and waist-height ratio among US children and adolescents. Pediatrics 2006;118:e1390 – 8. [30] Moran A, Jacobs DR Jr, Steinberger J, et al. Insulin resistance during puberty: results from clamp studies in 357 children. Diabetes 1999; 48:2039 – 44. [31] Steinberger J, Daniels SR; American Heart Association Atherosclerosis, Hypertension, and Obesity in the Young Committee (Council on Cardiovascular Disease in the Young); American Heart Association Diabetes Committee (Council on Nutrition, Physical Activity, and Metabolism). Obesity, insulin resistance, diabetes, and cardiovascular risk in children: an American Heart Association scientific statement from the Atherosclerosis, Hypertension, and Obesity in the Young Committee (Council on Cardiovascular Disease in the Young) and the Diabetes Committee (Council on Nutrition, Physical Activity, and Metabolism). Circulation 2003;107:1448 –53. [32] Sharp TA, Grunwald GK, Giltinan KE, et al. Association of anthropometric measures with risk of diabetes and cardiovascular disease in Hispanic and Caucasian adolescents. Prev Med 2003;37:611– 6. [33] Srinivasan SR, Myers L, Berenson GS. Rate of change in adiposity and its relationship to concomitant changes in cardiovascular risk variables among biracial (black-white) children and young adults: The Bogalusa Heart Study. Metabolism 2001;50:299 –305. [34] Freedman DS, Mei Z, Srinivasan SR, et al. Cardiovascular risk factors and excess adiposity among overweight children and adolescents: the Bogalusa Heart Study. J Pediatr 2007;150:12–17.e2. [35] American Diabetes Association. Type 2 diabetes in children and adolescents. Diabetes Care 2000;23:381–9. [36] Kahn HS, Imperatore G, Cheng YJ. A population-based comparison of BMI percentiles and waist-to-height ratio for identifying cardiovascular risk in youth. J Pediatr 2005;146:482– 8. [37] Joliffe CJ, Janssen I. Development of age-specific metabolic syndrome criteria that are linked to the Adult Treatment Panel III and International Diabetes Federation Criteria. J Am Coll Cardiol 2007; 49:891– 8. [38] Zimmet P, Alberti G, Kaufman F, et al. International Diabetes Federation Task Force on Epidemiology and Prevention of Diabetes. The metabolic syndrome in children and adolescents. Lancet 2007; 369(9579):2059 – 61. [39] Bray GA, DeLany JP, Harsha DW, et al. Evaluation of body fat in fatter and leaner 10-y-old African American and white children: the Baton Rouge Children’s Study. Am J Clin Nutr 2001;73:687–702. [40] Freedman DS, Ogden CL, Berenson GS, et al. Body mass index and body fatness in childhood. Curr Opin Clin Nutr Metab Care 2005;8: 618 –23.