Macronutrient intakes and cardio metabolic risk factors in high BMI African American children

Body weight and height were measured to the nearest 0.1 kg and 0.1 cm using a digital electronic scale (BWB 800, Tanita, Japan), and a portable stadiometer, respectively. Body mass index (BMI), BMI percentiles and BMI z-scores were generated using an age and gender specific CDC calculator http://​www.​cdc.​gov/​nccdphp/​dnpa/​growthcharts/​resources/​sas.​htm. For simplicity, the term "high BMI" has been used in this paper to collectively refer to children defined by an expert committees, and based on CDC growth charts [17‐19], as being either "at risk of overweight". (BMI's between 85th and 95th percentil) or "Overweight/Obese" (BMI's at or above 95th percentile).

Using a plastic non-elastic measuring tape, waist circumference was measured just above the iliac crest with child the standing position. Measurements were taken twice and if agreement between repeats was greater than 0.4 cm, a third measurement was taken and the mean calculated using the closest two values [20]. Total body resistance was measured using bioelectrical impedance analysis (Model BIA-101Q; RJL systems, Detroit, MI) according to manufacturer's instructions [20], and percent body fat was computed using Horlick's equation [21]. This method and equation have been validated in AA children previously.

Blood pressure measurements were taken between 0800 and 1200 hr. Children were asked to rest for a duration of 15-20 minutes in a sitting position after which they were measured with a manual sphygmomanometer by highly trained individuals. Measurements were repeated until two consecutive systolic and diastolic measurements agreed within 4 and 2 mm Hg, respectively. This series of measurements was repeated twice in children, with repeats spaced at least 2 hr apart, and the second series of measurements used for analyses. Data were analyzed using both systolic and diastolic pressures, and after conversion to blood pressure percentiles (matched for age, height and gender) using regression equations developed and reported elsewhere [22].

Subjects reported to Children's Hospital and Research Center, Oakland CA after a 12 hr overnight fasting when their blood was drawn. Plasma lipids were measured by a comprehensive lipoprotein analysis performed by a commercial lab (LabCorp). Using the vertical auto profile (VAP)-cholesterol method a modified density gradient centrifugation technique, concentrations of the cholesterol content of all lipoprotein subclasses (total cholesterol, HDL-cholesterol, LDL-cholesterol, IDL-cholesterol, VLDL-cholesterol) and triglycerides were determined on a single sample [23].

Fasting plasma glucose was determined using a hexokinase- peroxidase method (glucose HK-60 assay; Diagnostic Chemicals, Oxford, CT). Reference standards of normal and elevated glucose concentration (DC-Trol;Diagnostic Chemicals, Oxford, CT) were analysed with each assay for quality-control purposes.

Fasting serum insulin concentrations were determined using enzyme-immunoassay (Linco, St. Charles, MO). The insulin assay has <1% cross-reactivity with proinsulin. The intra-assay variation was 4.5% and inter-assay variation was 10%

Fasting glucose and insulin values were used to calculate insulin resistance [24] using the homeostasis model assessment of insulin resistance (HOMA-IR), defined as fasting glucose (mmol/l) × insulin (μU/ml)/22.5.

Serum non esterified free fatty acids (NEFA) were measured using an in vitro enzymatic colorimetric assay (NEFA, Wako Diagnostics, USA) and used to calculate Fatty acid insulin sensitivity (ISI-FFA) by the formula:{2/(insulin × NEFA) +1 } [25].

Pubertal development was assessed by measurements of serum luteinizing hormones (LH) in boys and estradiol and LH in girls. Children were classified into stages 1 through 5 using literature-derived values [26‐31]. The hormone concentrations included in the 5-pubertal stages were as follows: 0-11.6, 11.61-26.2; 26.21-50.7; 50.71-64.9 and >64.91 pg/mL for E₂ in girls; 0-0.4, 0.41-3.4; 3.41-4.1; 4.11-5.9 and >5.91 IU/L for LH in girls; and 0-0.7, 0.71-1.6; 1.61-2.6; 2.61-3.1 and >3.11 IU/L for LH in boys. When E₂ and LH hormone concentrations in girls suggested different pubertal stages, the higher stage was assigned to that girl [31].

Three-day food diaries were used to assess intakes of nutrients as they have been shown to be a more accurate tool for diet assessment than 24-h diet recall or food frequency questionnaire methods [32]. Before completing a 3-day food diary (2 weekdays and 1 weekend day), participating children and a parent/guardian were trained to record portion sizes, brand names, place consumed, and time of day food was consumed. Trained staff reviewed the questionnaires with the child, expanding on descriptions of foods and portion sizes when needed.

Macronutrient intakes were determined using the USDA nutrient database [33]. To do this, the foods listed on the 3-day food diaries were labeled according to the 8-digit food codes, portion sizes or weights were entered into this software, and computer programs were used to calculate the 3-day average intakes of energy and specific macronutrients. Analyses were carried out on the data set as a whole and, following the protocol for the National Health and Nutrition Examination Survey (NHANES), no quantification or exclusion for underreporting or over reporting was made. Average daily intakes of energy (kcal/day) and of the macronutrients, carbohydrates, fat and protein (g/day) were used for analyses.

Statistical procedures were performed using SPSS for Windows version 16.0 (SPSS Inc, Chicago, IL). Statistical significance was defined to be p ≤ 0.05. Results with 0.05 < p < 0.10 were also noted to show close associations. Differences in the anthropometric and lipoprotein profiles in boys versus girls were performed using independent t tests. Dixon's test for outliers was used to identify unusual values. When identified, all data for that participant were excluded from further analyses. Using Dixon's test, data for 4 children were excluded. Data for the variables of interest were not significantly skewed. Correlations were used to evaluate bivariate associations among dependent and independent variables including intakes of total energy, fat, carbohydrate and protein, and biochemical and blood pressure variables of interest. As associations between these variables were similar for Pearsons and Spearmans correlations, values are reported only for Pearsons correlations, and parametric tests were used in subsequent analyses. Multiple linear regression analyses were used to assess the relationship of total energy intake (unadjusted for source of energy), or, individually, of fat, carbohydrate and protein to TC, VLDL-C, IDL-C, LDL-C, HDL-C, TG, HOMA-IR, NEFA, ISI-FFA, sBPz and dBPz after adjusting for gender, pubertal stage and waist circumference of the participating child. Child age was not included as a covariate since it was not significantly correlated with dependent or independent variables.

As might be expected BMIz and WC were highly correlated (r = 0.87, p < 0.001). In addition, WC and BMI were both associated with markers of cardio metabolic risk factors including TG (BMIz r = 0.34; WC r = 0.30), HDL-C (BMIz r = -0.40; WC r = -0.46), HOMA-IR (BMIz r = 0.60; WC r = 0.57), ISI-FFA (BMIz r = -0.57; WC r = -0.55), sBPz (BMIz r = 0.25; WC r = 0.28) dBPz (BMIz r = 0.23; WC r = 0.87). Because WC is included in the National Cholesterol Education Program (NCEP) ATP III recommendations [34], WC was chosen for results presentation and discussion.