Ethnic Differences in Lipid Profiles of Overweight,Obese,and Severely Obese Children and Adolescents 6–19 Years of Age

Background

Pediatric obesity is prevalent and continues to be a major public health concern.^1–3 Results from the 2011 to 2012 National Health and Nutrition Examination Survey (NHANES) indicate that 16.9% of children and adolescents aged 2–19 years were obese and another 14.9% were overweight.^3,4 This means that nearly one third of youth in the United States (US) were overweight or obese in 2012.^2,3 These statistics are worrisome because childhood obesity has both immediate and long-term health effects. Even more alarming is that severe obesity is the fastest growing subcategory of obesity and currently affects between four and six percent of US children. ⁵

Obesity is defined as a body mass index (BMI) greater than the 95th percentile up to 120% of the 95th percentile.^5–7 Subcategories of severe obesity exist and are used to define Class 2 and Class 3 cases beyond the cutoff of 120% of the 95th percentile.^3,5 Identifying and treating severely obese youth are critical because, compared with overweight and obese youth, those with severe obesity are at a higher risk of developing chronic illnesses, including Type 2 diabetes mellitus (T2DM) and cardiovascular disease (CVD).^5,8

A cluster of cardiometabolic risk factors, known as the metabolic syndrome, was designed to identify those at high risk for developing CVD and T2DM.^5,9 In the presence of the metabolic syndrome, there is a twofold increased risk of CVD and a fivefold increased risk of diabetes. ⁹ The diagnosis of the metabolic syndrome requires the presence of three out of five characteristics. The five characteristics are increased triglyceride (TG) levels, low high-density lipoprotein (HDL) levels, central obesity, hypertension, and fasting hyperglycemia. ⁹ Analysis of NHANES III data found that, among 1960 youth aged 12 through 19 years, non-Hispanic blacks (NHBs) had a lower prevalence of metabolic syndrome compared to non-Hispanic whites (NHWs) and Mexican Americans (MA) (2.5%, [95% CI 1.3%–3.7%]; 10.9%, [95% CI 8.4%–13.4%]; and 12.9%, [95% CI 10.4%–15.4%], respectively). ⁶ Results of an age-adjusted study found that, when compared to NHWs, NHBs and Hispanics were more likely to exhibit obesity and were at a higher risk for developing T2DM. ¹ Although NHDs are more likely to be obese and have hypertension, CVD, and diabetes, when compared to NHWs, they are less likely to be diagnosed with the metabolic syndrome because NHBs are less likely to have either elevated TG or low HDL levels.^8,9

Whether ethnic differences in lipid profiles are observed at severe levels of obesity in children and adolescents is not well known.^8,9 With this in mind, the purpose of this study was to determine if ethnic variations in lipid profiles present in overweight and obese youth are also found in severely obese youth. The findings could explain the lower prevalence of the metabolic syndrome in NHB youth and determine if it is also likely at severe levels of obesity.

Methods

Data were obtained from NHANES for the years 2001 through 2012. ¹⁰ NHANES surveys are conducted by the National Center for Health Statistics (NCHS) and the Centers for Disease Control and Prevention (CDC) and are a series of cross-sectional, nationally representative surveys of the noninstitutionalized US population. Data were collected from household interviews as well as from physical and laboratory examinations. Parents provided informed written consent for children <18 years. Hemophiliacs and individuals who had received chemotherapy within the last 4 weeks were excluded from having venous samples drawn. In addition, individuals with rashes, gauze dressings, casts, edema, paralysis, tubes, open sores or wounds, withered arms or limbs missing, damaged, sclerosed, or occluded veins, allergies to cleansing reagents, burned or scarred tissue, and shunt or intravenous lines on both arms were excluded. ¹¹

Race and ethnicity were self-reported and answer choices included an open-ended response option. Participants aged 16 years and older reported their own race/ethnicity. For children <16 years, a family member reported race/ethnicity. Race and ethnicity were categorized as NHW, NHB, and MA.

The NCHS approved the protocol used to obtain, handle, and analyze participant venous samples. ¹¹ The NHANES Laboratory/Medical Technologists Procedures Manual details procedures used for collecting, labeling, processing, preserving, and transporting samples. Participants in the examination component of the survey were randomly assigned to either a morning or an afternoon session. Those assigned to the morning session were at least 12 years of age and were asked to fast for at least 8 hours before the examination. Total cholesterol (TC) and HDL measurements were obtained from all examined participants. Low-density lipoprotein (LDL) and TG measurements were only taken from participants in the morning session, who had fasted for a minimum of 8 hours. Analyses involving TC and HDL measurements were conducted for respondents 6–19 years of age. Since those less than 12 years of age were not requested to fast, analyses for LDL and TG were limited to participants 12–19 years of age.

A calibrated equipment was used to obtain height and weight measurements to calculate BMI. Participants were divided into groups according to BMI classification. Respondents with a BMI less than the fifth percentile were excluded. Normal weight was defined as a BMI less than the 85th percentile. Overweight was defined as a BMI between the 85th and 95th percentile. Class 1 obesity was defined as a BMI greater than the 95th percentile up to 120% of the 95th percentile. Class 2 obesity was defined as a BMI between 120% and 140% of the 95th percentile. Class 3 obesity was defined as a BMI above 140% of the 95th percentile.

Analysis of variance (ANOVA) tests were conducted to assess differences in lipid profiles among ethnic groups across BMI categories. Tukey Honest Significant Difference (HSD) tests for post-hoc comparisons were performed where there were statistically significant intergroup differences. Statistical analysis was performed using SPSS version 22.0. Statistical significance was set at an alpha level of 0.05.

Results

The sample included 14,481 subjects and Table 1 shows the sample's demographic composition. Of the three groups, NHB had the greatest mean weight (49.9 kg), while MA had the greatest mean waist circumference (72 cm). Approximately 8.6% of NHB subjects were severely obese compared to 7.8% of MA and 4.1% of NHW participants.

Mean TC, TG, LDL, and HDL levels by race/ethnicity and BMI category are shown in Table 2. The sample included 10,380 subjects with a reported TC level. TG levels were included for 4498 subjects, LDL levels were reported for 4406 subjects, and 5881 subjects had a reported HDL level. The mean TC level for the Normal Weight group was 159.3 mg/dL. There were significant differences (p < 0.0001) between the TC level means of the race/ethnic groups with the NHB group having the highest at 161.6 mg/dL. Significantly higher (p < 0.05) mean TC levels were observed for the NHW group compared to the NHB group within the Class 1 obesity category. There were no significant differences in mean TC levels between any of the race/ethnic groups within the Overweight, Class 2, and Class 3 obesity categories.

Across all BMI categories, NHBs had the lowest mean TG levels. Significantly higher TG levels (p < 0.0001) were observed for NHWs compared to NHBs and for MA compared to NHBs in all BMI categories. There were no differences in mean TG or HDL levels between NHWs and MA in the Overweight, Class 1, Class 2, and Class 3 obesity categories. Mean HDL levels were highest for NHBs across all BMI categories. Significantly higher mean HDL levels (p < 0.0001) were observed for the NHB group compared to the NHW group and for NHB group compared to the MA group in all BMI categories

Discussion

Although recent data suggest that the rate of increase in obesity among children and adolescents has slowed and the prevalence of obesity has begun to plateau, a troubling trend in the form of severe pediatric obesity has emerged.^3,5 The prevalence of severe obesity among US youth is increasing. ¹² Wang et al. used three NHANES (NHANES II: 1976–1980, NHANES III: 1988–1994, and NHANES 1999–2006) to examine trends in the prevalence of severe obesity among US children. ¹³ The study found that the age-standardized prevalence of severe obesity increased from 1.1% to 1.3% (boys/girls) in NHANES II, to 2.9% to 3.1% in NHANES III, and 5.1% to 4.7% in NHANES 1999–2006 (p-values for trend <0.001). ¹³ Data have consistently shown higher prevalence of obesity among Hispanic or MA children and NHB or African American youth.^2–5 Findings of our study are consistent with this.

Severe pediatric obesity has grave immediate and long-term health consequences. Compared with overweight and obese youth, severely obese youth demonstrate early signs of vascular dysfunction and subclinical atherosclerosis and have a more adverse cardiometabolic risk factor profile. ⁵ Severe pediatric obesity is associated with obstructive sleep apnea syndrome, fatty liver disease, musculoskeletal complications, and psychosocial problems. ⁵ The tracking of adiposity from childhood into adulthood is stronger in the severely obese, and high BMI during childhood is associated with an increase in the risk of CVD, T2DM, and premature death. ⁵

Although several definitions have been proposed to identify severe obesity in youth, BMI ≥99th percentile has been the most commonly used. ⁵ Recently, it has been found that 120% of the 95th percentile of BMI for age was similar to the unsmoothed 99th percentile. ⁵ This finding led to the recommendation that severe obesity in adolescents and children ≥2 years be defined as a BMI ≥120% of the 95th percentile or an absolute BMI ≥35 kg/m², whichever value is lower based on age and sex. ⁵ The 2000 CDC growth charts cannot accurately characterize severe obesity beyond the 97th percentile.^5–7 Newer growth charts exist and allow clinicians to track BMI values in severely obese children. ⁷ To justify the need for the newer growth charts, Gulati et al. use the hypothetical example of a 15-year-old boy who weighs 250 lbs, is 5′8″ tall, and has a BMI of 38.1 kg/m², which is at the 99th percentile. ⁷ If the same boy were to gain 70 lbs, his BMI would increase to 48.8 kg/m², which is also still at the 99th percentile. ⁷ If the newer growth charts were used, the boy's BMI percentile would increase from 142% of the 95th percentile to 182% of the 95th percentile. ⁷ This characterization is a more accurate representation of the drastic change in the boy's weight. ⁷ Use of the newer charts is recommended to help clinicians avoid the false assumption that patients have not changed BMI percentiles. ⁷

Severe obesity includes a large spectrum of BMIs. Stratifying classes of severe obesity facilitates the subcategorization of an overgeneralized group. ⁷ The methodology used to define Class 2 and 3 obesity in our study is similar to the methods used in research conducted by Skinner & Skelton to assess the prevalence of severe obesity in US children between 1999 and 2012. ² Other studies have used similar methodology to measure prevalence of severe obesity in youth by race and ethnicity. Our research is different in that we not only defined and measured prevalence of severe obesity by race and ethnicity but we also analyzed ethnic differences in lipid profiles of normal weight, overweight, obese, and severely obese youth to determine if the low prevalence of the metabolic syndrome often seen in NHB youth is likely to be observed in NHB youth with Class 2 and 3 obesity.

The metabolic syndrome is used to identify individuals at high risk for the development of T2DM and CVD. ⁹ Although NHBs have higher rates of diabetes and CVD, when compared to other ethnic groups, they have a lower prevalence of the metabolic syndrome because they are less likely to have elevated TG or low HDL levels.^8–9,12,14 Research conducted by Park et al. found significantly lower frequencies of high TG and low HDL cholesterol levels in blacks compared to whites and MAs. ¹⁴ When compared to NHW and MA youth, NHB youth in our study had significantly lower TG and greater HDL levels (p < 0.0001) in all of the BMI categories. Our findings are similar to results of a study conducted by Johnson et al., which found high TG concentrations to be significantly greater in Hispanic and NHW adolescents than in NHB adolescents. ¹² Johnson et al. also reported a greater prevalence in low HDL-C concentration in Hispanic and NHW adolescents than in NHB adolescents. ¹²

The presence of high TG levels and low HDL levels simultaneously is known as the dyslipidemia associated with insulin resistance. ⁹ Insulin resistance refers to an increase in the amount of insulin that is needed to stimulate glucose uptake. ⁸ The literature shows that, although African Americans have lower rates of hypertriglyceridemia than seen in NHWs or Hispanics, African American youth have a higher degree of insulin resistance when compared to NHW youth.^8,9 TGs are cleared from circulation by lipoprotein lipase (LPL), an enzyme whose activity decreases in the presence of insulin resistance. ⁸ Lower levels of LPL activity are, therefore, associated with higher levels of TG. High TG and low HDL levels occur together because they are cleared from circulation reciprocally. ⁹ The enzyme that clears HDL is hepatic lipase (HL) and increased levels of HL activity are associated with lower levels of HDL. Theoretically, greater insulin resistance is associated with less LPL activity and greater HL activity. When compared to NHW individuals, African Americans have higher LPL activity and decreased HL activity.^8,9 If the effect of insulin resistance on lipid profiles were the same in NHW and NHB individuals, then TG levels would be higher and HDL levels would be lower in NHB individuals since they have a higher degree of insulin resistance, but this is not the case. ⁹ The reverse is true. Insulin resistance does not seem to adversely affect the lipid profile of NHB individuals. ⁹

It appears that, when compared to those from other ethnic groups, NHB individuals start with lower baseline levels of TG and are, therefore, less likely to exceed the cutoff values specified in criteria set for diagnosis of the metabolic syndrome. ⁸ Results from our study support this theory because, compared to NHWs and MA, significantly lower TG levels were observed for normal-weight NHB individuals and this trend was observed across all BMI categories.

Studies conducted by l'Allemand-Jander, Orenes-Piñero et al., and Shamai et al. report no association between BMI and LDL.^15–17 Morris & Ferdinand report that LDL levels in African American men are similar or lower than those of white men. ¹⁸ Elevated LDL levels pose a major risk factor for coronary artery disease (CAD) and NHB individuals are at increased risk for CAD and obesity.^18,19 Despite this, significant differences were not found in mean LDL levels between any of the ethnic groups in any of the BMI categories in our study.

Diagnosis with the metabolic syndrome is an early warning sign for diabetes and CVD, yet, NHB individuals who are at risk for diabetes and CVD have a lower prevalence of the metabolic syndrome.^6,9 The lower tendency for the dyslipidemia of insulin resistance is the main cause of lower rates of diagnosis.^8,9 This study did not take fasting glucose or insulin levels into account. In addition, waist circumference, physical activity levels, age, and sexual maturity were not controlled for. These are limitations and future research should consider the inclusion of these factors to determine the effect they may have on lipid level differences among ethnic groups across BMI categories. Despite these limitations, our study had strengths. One major strength was the use of a large sample. In addition, stratification by levels of severe obesity allowed for the comparison of ethnic differences in lipid profiles for normal weight, overweight, obese, severely obese, and very severely obese youth. To the best of our knowledge, that has not been done before.

Conclusion

Significantly low TG and high HDL levels were observed for NHB individuals across all BMI categories. Ethnic differences in lipid levels may explain why the dyslipidemia associated with the metabolic syndrome is observed less frequently in NHBs even though the conditions the metabolic syndrome is supposed to predict are more prevalent in this group. ⁹ Because metabolic syndrome in childhood likely tracks into adulthood, early identification may help target interventions to improve future cardiovascular health. ⁶ TG and HDL thresholds currently defined by the metabolic syndrome guidelines as abnormal may not be appropriate for NHB individuals. Ethnic-specific guidelines may be necessary for improved identification of those at risk, especially for severely obese youth.