Childhood obesity is increasing at a fast pace, together with its complications. The aim of the present study is to assess several candidate triggering agents, mechanisms and intermediate phenotypes of atherosclerosis during the post-prandial phase in the obese insulin-resistant child/adolescent.

Some of earlier outlines of the relationships between enlarged fat mass and insulin resistance have not been proven experimentally in man. For instance, most of the relationships shown in humans between cytokines and insulin sensitivity are correlational observations.

In the obese child, fatty liver is quite prevalent and is more closely associated than other fat depots to insulin resistance, dyslipidemia, high blood pressure and to high levels of inflammation markers. The entire picture, therefore, is consistent with an accelerated drive to atherosclerosis in the obese child/adolescent. Indeed, several reports have demonstrated that intermediate phenotypes of atherosclerosis can be detected in the obese child. Furthermore, the post-prandial phase is thought to play a role of its own in atherogenesis.

The investigators hypothesize that in the obese insulin-resistant child mechanisms of atherogenesis, such as hyperlipemia, oxidant stress and sub-inflammation, are activated in the post-prandial phase resulting in impairments of vascular function.

Parameters under study:

1. Candidate triggering agents

   • Post-prandial glucose
   • Post-prandial hyperlipemia
   • Post-prandial small dense LDL
   • Post-prandial insulin
   • Candidate mechanisms
   • Complement activation
   • C3

2. Inflammation

   • TNF-alpha
   • Interleukin-6 (IL-6)
   • Interleukin-10 (IL-10)
   • C-reactive protein (CRP)
   • Oxidant Stress
   • Urinary iso-PGF2-alpha

3. Intermediate Phenotypes

   • Carotid artery intima-media thickness (IMT) (only at baseline)
   • Endothelial Function (flow-mediated vasodilation) (FMD)
   • Arterial Stiffness (pulse-wave velocity) (PWV)

Experimental protocol Subjects will arrive at the Division of Endocrinology and Metabolic Diseases of the Verona City Hospital at 0800 h after an overnight fast. Weight, height, waist circumference and blood pressure will be measured. During the entire study subjects will be lying in bed in a quiet, temperature controlled (22 °C) room. A Teflon venous catheter will be inserted into an antecubital vein for blood sampling and will be kept patent with a normal saline slow-drip infusion. Baseline state will be considered attained after at least 20 minutes of rest in bed following vein puncture.

Subjects must have been on no drugs known to interfere with vascular function for at least 2 weeks and will be asked to refrain from caffeine containing beverages for at least 24 hours before study. Systolic blood pressure, diastolic blood pressure, mean arterial pressure and heart rate will be monitored at time intervals with Cardiocap II (Datex, Finland) throughout the study.

Carotid-femoral PWV will be assessed using Complior (Colson, Garges les Genosse, France) . Carotid artery IMT will be assessed by high resolution US color-doppler scan. Then, flow-mediated endothelium-dependent vasodilation of the nondominant common femoral artery will be assessed by high resolution echo-doppler (Esaote Biomedica AU6, Genova, Italy), with a 10 MHz linear vascular probe with axial resolution of 0.01 mm. Baseline diameter and blood flow velocity, a proxy of shear stress, will be assessed at least in triplicate. After this, a sphygmomanometric cuff will be put under the knee and inflated 50 mm Hg higher than systolic blood pressure and maintained for 5'. Vessel diameter and flow velocity will be measured at time 0.5', 2', 4', 6', and 8' after deflating the cuff, according to investigators' well established protocol . Under these conditions, when the cuff is deflated postischemic hyperemia will take place in the tissues which are distal to the sphygmomanometric cuff. This hyperemia will induced an increase in blood flow velocity of the common femoral artery, which is located proximally to the cuff. The increase in flow velocity, and the ensuing increase in shear stress, elicit an endothelium-dependent vasodilation. It has been shown that this dilation is NO-dependent. A global quantitative index of flow-dependent vasodilation will be obtained by calculating the area under the curve of change in vessel diameter as a function of time (8 min of observation), expressed both as absolute values and as percent changes over the baseline vessel diameter. In investigators' hands, the between-day coefficient of variation of this technique (n=15, covering a wide range of vascular response) is 23.0±3.8%. Endothelium-independent vasodilation will not be evaluated because glyceryl-trinitrate administration for research purpose is deemed ethically unjustifiable in children by the local Institutional Ethical Committee.

In investigators' current experience (n=42 severely obese children), average FMD after an overnight fast is 1.40±0.91% per 8 minutes (mean±SD), thereby leaving substantial room to show dynamic changes. In these severely obese children these investigators found no correlation between FMD and BMI. An ad hoc mechanical US probe holder will be employed in the present paired studies, with the goal to help increasing reproducibility of the technique. Furthermore, a specific automatic contour tracking technique for the automatic calculation of the changes in arterial diameter will be used to minimize the operator dependency of this method.

Baseline (-10 min and 0') blood samples will be collected to measure glucose, insulin, C-peptide, standard lipid profile (total cholesterol, HDL-cholesterol, and triglycerides), oxidized-LDL, adiponectin, leptin, C3, TNF-alpha, IL-6, IL-10, and CRP by micro-methods to limit blood loss. A separate sample will be collected to measure LDL and HDL subfractions. Baseline urines will be collected to measure iso-PGF2-alpha, a product derived from nonenzymatic lipid peroxidation, catalyzed by oxygen free radicals on cell membranes and LDL particles (43).

At time 0', subjects will ingest a mixed liquid meal (Ensure Plus, 1.5 kcal/cc) of known caloric content (300 kcal per m2 of body surface area) and composition (53.8% carbohydrates, 29.5% lipids, 16.7% proteins) over 10 minutes. Glucose, insulin and C-peptide will be measured at +15', +30', +45', +60', +90', +120', +150', +180', +210', +240' and +300'. Standard lipids and PWV will be measured at +60', +120', +180', +240' and +300'. Adiponectin, leptin, C3, TNF-alpha, IL-6, IL-10, CRP and LDL and HDL subfractions will be measured at +180' and +300'. At time +180' assessment of FMD will be repeated. At the end of the meal test urines will be collected to assess 8-iso-PGF2-alpha.

Blood samples will be quickly spun at 1500 g at +4 °C, plasma/serum will be collected and stored at -80° C. Urine samples will be stored at -80° C. Total blood loss will be kept below 100 cc.

Analytical Methods Except for LDL/HDL subfraction assessment and urine PGF2alpha, methods specifically optimized to use tiny amounts of plasma/serum will be used to minimize blood loss. Glucose will be measured at bedside with a Yellow Spring Glucose Analyzer by the glucose oxidase method. C-peptide and insulin will be measured by in-house ELISA micro-methods. Total and HDL cholesterol and total triglycerides will be assessed by standard in-house enzymatic micro-methods. Adiponectin, leptin, C3, TNF-alpha, IL-6, IL-10, CRP will be assessed by in-house immunometric micro methods. 8-Iso-PGF2alpha will be assessed by an immunometric assay after extraction from urines.

LDL/HDL subfractions will be measured in the laboratory of prof. Alberto Zambon, at University of Padua School of Medicine. Briefly, after creating a discontinuous salt density gradient in an ultracentrifuge tube, the samples will be centrifuged at 65,000 rpm for 90 min at 10°C in a Sorvall TV-865B vertical rotor. Thirty-eight 0.45-ml fractions are then collected from the bottom of the centrifuge tube. Cholesterol is measured in each fraction. The relative flotation rate (Rf), which characterizes LDL peak buoyancy and is a measure of LDL size/density, is obtained by dividing the fraction number containing the LDL cholesterol peak by the total number of fractions collected.

Assessment of beta-cell function

Beta-cell function will be evaluated by analyzing the glucose and C-peptide curves during the mixed meal test, according to the general strategy proposed by several laboratories with some slight modifications. Briefly, insulin secretion is the sum of three components:

1. basal (postabsorptive) insulin secretion rates;
2. insulin secretion in response to the rate of increase in plasma glucose ("dynamic" or "derivative" secretion component);
3. insulin secretion in response to the actual glucose levels above the postabsorptive glucose concentration ("static" or "proportional" secretion component).

A complete description of the modeling strategy can be found in investigators' earlier publications. Parameters will be estimated by implementing this minimal model of C-peptide secretion in the SAAM 1.1.2 software (SAAM Institute, Seattle, WA). Numerical values of the unknown parameters will be estimated by using nonlinear least squares.

Weights are chosen optimally, i.e., equal to the inverse of the variance of the measurement errors, which are assumed to be additive, uncorrelated, with zero mean, and a constant coefficient of variation (CV) of 8%.

This analysis will provide the hormonal scenario in which the postprandial atherogenetic processes take place.

Statistical analysis Data will be summarized as means±SEM. The size of the sample (n=20 in each group) has been selected with the aim of having 80% chance of attaining statistical significance for differences of 20-40% (the minimum detectable difference varies from variable to variable under study). The general estimating equation procedure (GEE) will be used to run 2-way ANOVA for repeated measures and to adjust for covariates, if needed. Gaussian distribution will be tested in all variables, and, if statistically significant deviations from normality are found, logarithmic (or other) transformation will be applied. Correlations will be sought by the Spearman's rank correlation coefficient. Statistical significance will be declared at p<0.05. Statistical analyses will be performed using SPSS 12.0 (or higher) for MacOS X.

Reaching the goals of the present research project will:

1. enlarge the knowledge on the relationship between childhood obesity, inflammation, insulin resistance and vascular dysfunction;
2. elucidate whether postprandial hyperlipemia, postprandial inflammation and/or postprandial oxidant stress are associated to intermediate phenotypes of atherosclerosis in the obese child/adolescent;
3. establish potential novel targets of prevention of obesity complications in the child/adolescent and potentially indicate novel tools for the treatment of obesity associated disturbances.

Inclusion Criteria:

• Male and female
• 6 ≤ age ≤14 years
• BMI ≥ 90th percentile for sex and age
• Normal glucose tolerance

Exclusion Criteria:

• Any overt disease other than obesity
• Drugs known to interfere with vascular function for at least 2 weeks before study
• Inflammatory diseases (assessed clinically and by routine blood tests)
• 1.8 < HOMA-IR < 2.5

Total number: 40 children with a BMI > 90th percentile for sex and age. Selection criteria: group A, n=20, less insulin resistant: HOMA-insulin resistance index (HOMA-IR) < 1.8; group B, n=20, severely insulin-resistant: HOMA-IR > 2.5.