We sought to determine whether childhood obesity is associated with increased aortic stiffness by measuring the biophysical properties of the aorta in obese children using a noninvasive echocardiographic Doppler method. Increased aortic stiffness is a strong predictor of future cardiovascular events and mortality in adults. Obesity is known to be associated with increased aortic stiffness and arterial disease in adults. We prospectively evaluated a cohort of obese children (n = 61) and compared them to normal-weight controls (n = 55). The anthropometric data were recorded. The pulsewave velocity (PWV), aortic input impedance (Zi), characteristic impedance (Zc), arterial pressure-strain elastic modulus (Ep), arterial wall stiffness index (B index), and peak aortic velocity were calculated. We correlated our echocardiographic Doppler findings with the lipid levels. We assessed the left ventricular (LV) dimensions and standard measures of cardiac function. Cardiopulmonary exercise testing was performed on all obese children. Compared to normal-weight children, obese children had a greater PWV, Zc, B index, Ep, and peak aortic velocity. Obese children had greater systolic blood pressure than normal-weight children but no difference in diastolic blood pressure. The LV dimensions and standard measures of cardiac systolic function were similar in the 2 groups, but the obese children had altered diastolic properties. The LV mass was greater in the obese children. No association was found between the lipid levels and the biophysical properties of the aorta. The relative oxygen consumption was 68% predicted in obese children. In conclusion, measures of the biophysical properties of the aorta are already abnormal in obese children, reflecting increased aortic stiffness at this early stage of disease. Obese children also had an increased LV mass, altered diastolic properties, and an abnormal exercise capacity. PWV might be useful in monitoring the progression of arterial disease or the effect of therapeutic interventions.
The normal aorta is elastic and acts as a cushion and conduit to buffer the pulsatile blood flow; the stiffer the aorta, the faster the propagation of the pulsatile flow wave. Thus, the pulsewave velocity (PWV) is an indirect measure of aortic stiffness. Direct measures of aortic stiffness, including the arterial pressure-strain elastic modulus (Ep) and the arterial wall stiffness index (B index), assess the relation of pulsatile flow changes in the aorta with blood pressure. These measures, as well as measures of impedance, can be assessed using an echocardiographic Doppler technique. The primary aim of the present study was to determine whether childhood obesity is associated with abnormal biophysical properties of the aorta. Our secondary aims were to determine whether childhood obesity is associated with altered left ventricular (LV) dimensions, indexes of LV function, and exercise capacity.
Methods
We prospectively recruited a cohort of obese children. They were referred from the Shapedown Program at the Centre for Healthy Weights at British Columbia Children’s Hospital for the evaluation of cardiovascular risk factors. Children were eligible for inclusion if they were ≤18 years old and had a body mass index ≥95th percentile for age. We excluded children in whom a systemic illness was the primary cause of their obesity and children for whom we were unable to obtain echocardiographic and Doppler data because of poor image quality. We compared these children to a cohort of normal-weight children who were ≤18 years old, had a body mass index of ≤95th percentile for age and gender, and no known systemic illness.
All children had their height, weight, and blood pressure (BP) measured, underwent a complete cardiovascular examination, had a standard 12-lead electrocardiogram recorded, and underwent standard 2-dimensional and M-mode echocardiography and Doppler imaging. The LV dimensions during systole and diastole were measured and corrected for height (in m 2.7 ). Standard measures of cardiac size and function were obtained. All obese children also underwent cardiopulmonary treadmill exercise testing. Lipid profiles were obtained.
The biophysical properties of the aorta were derived using an echocardiographic-Doppler method previously described. In brief, in a suprasternal long-axis view, a pulsed wave Doppler trace was obtained from the proximal ascending aorta. The interval from the Q wave on the electrocardiogram tracing to the onset of the Doppler envelope was measured (time 1). Maintaining the same transducer position, a pulsed wave Doppler trace was obtained in the descending aorta. The interval from the Q wave to the onset of the Doppler envelope was measured again (time 2). The transit time was then calculated as the difference in time between these 2 locations. The length of the arch was measured between these 2 sample volumes with electronic calipers, and the PWV was determined by dividing the distance of the arch by the transit time. In the high left or right parasternal view, M-mode recordings of the ascending aortic diameter were made, and the end-diastolic and maximal systolic dimensions were recorded. All measurements were averaged for 3 consecutive cardiac cycles.
The following calculations were performed: pulse pressure = systolic BP − diastolic BP (mm Hg); peak aortic flow = peak aortic velocity × aortic annulus cross-sectional area (cm 3 /s); transit time = time 2 − time 1 (seconds); PWV = aortic arch length/transit time (cm/s); Ep = pulse pressure/[(maximal systolic dimension − end-diastolic dimension)/end-diastolic dimension] 2 (mm Hg); β index = ln (systolic BP/diastolic BP)/[(maximal systolic dimension − end-diastolic dimension)/end-diastolic dimension]; aortic input impedance, Zi = pulse pressure/peak aortic flow (dyne × s/cm 5 , 1 mm Hg = 1,333 dyne/cm 2 ); and characteristic impedance, Zc = PWV × ρ/aortic annulus cross-sectional area (dyne × s/cm 5 , blood density ρ = 1.06 g/cm 3 , 1 dyne = 1 g cm/s 2 ). We have previously shown that the interobserver variability for these measurements is ≤12%. BP z scores were derived as described in the fourth report on hypertension. We corrected the PWV by dividing by the systolic BP, and we compared the corrected PWV results.
Cardiopulmonary treadmill exercise testing was performed using an institutional protocol (British Columbia Children’s Hospital Protocol). The initial treadmill speed and grade were 2.0 mph and 0% grade, respectively. This was increased by 0.5 mph every minute with no changes in grade. The children were encouraged to exercise until volitional fatigue. A heart rate ≥190 beats/min and/or a respiratory exchange ratio >1.0 were criteria for a maximal test.
Univariate analysis was performed on all continuous variables. The mean ± SD are presented. Group differences were determined using a 1-way analysis of variance. The association between continuous variables was determined using Pearson product moment correlations. Because of multiple comparisons, p <0.01 was considered statistically significant. All statistical analyses were performed using IBM SPSS Statistics, version 19.0 (SPSS, IBM, Somers, New York).
The University of British Columbia Children’s and Women’s Research ethics board approved the present study.
Results
We evaluated 74 obese children, of whom 11 were excluded because of poor image quality on the echocardiogram, 1 becaues of a bicuspid aortic valve, and 1 because of co-morbid polycystic ovarian syndrome. We compared the 61 obese children who met the inclusion criteria to 55 normal-weight controls. The clinical data are listed in Table 1 . The systolic BP was greater in the obese children than in the normal-weight children (p = 0.002).
| Characteristic | Obese (BMI ≥95th Percentile) (n = 61) | Controls (n = 55) | p Value |
|---|---|---|---|
| Age (years) | 13.8 ± 2.3 | 13.8 ± 4.0 | NS |
| Body mass index (kg/m 2 ) | 32.6 ± 4.4 | 21.2 ± 4.7 | <0.001 |
| Boys (n) | 34 (56%) | 22 (40%) | NS |
| Weight (kg) | 88.3 ± 20.5 | 53.0 ± 18.0 | <0.001 |
| Height (cm) | 163 ± 12 | 156 ± 16 | 0.006 |
| Systolic blood pressure (mm Hg) | 114 ± 13 | 107 ± 11 | 0.002 |
| Systolic blood pressure z score | 0.56 ± 1.28 | 0.04 ± 0.86 | 0.012 |
| Diastolic blood pressure (mm Hg) | 62 ± 9 | 64 ± 8 | NS |
| Pulse pressure (mm Hg) | 52 ± 13 | 43 ± 8 | <0.001 |
Of the 61 obese children, 58 had normal electrocardiographic findings at rest. One patient met the voltage criteria for LV hypertrophy, one had first-degree atrioventricular block, and one had left axis deviation. Standard 2-dimensional echocardiograms demonstrated a normal cardiac structure in all children. The PWV, arterial pressure-strain elastic modulus, β index, and characteristic impedance were all greater in the obese children, indicating increased aortic stiffness compared to the normal-weight children ( Table 2 ). The aortic input impedance was not significantly different between the 2 groups. No correlation was found between PWV and systolic BP or pulse pressure. The PWV corrected for systolic BP was greater in the obese children than in the normal-weight controls (4.38 cm/s/mm Hg, 95% confidence interval 4.04 to 4.72, vs 3.32 cm/s/mm Hg, 95% confidence interval 3.16 to 3.49; p <0.001).
| Variable | Obese (BMI ≥95th Percentile) | Controls | p Value |
|---|---|---|---|
| Pulsewave velocity (cm/s −1 ) | 492 ± 136 | 351 ± 50 | <0.001 |
| Arterial wall stiffness index | 3.61 ± 1.25 | 2.98 ± 0.61 | 0.001 |
| Arterial pressure-strain elastic modulus (mm Hg) | 305 ± 111 | 248 ± 52 | 0.001 |
| Characteristic impedance (dyne s/cm −5 ) | 185 ± 72 | 133 ± 31 | <0.001 |
| Aortic input impedance (dyne s/cm −5 ) | 150 ± 54 | 141 ± 27 | NS |
The lipid profile of the obese children was compared to normative adult data and cutoffs for the 95th percentile in adolescent children ( Table 3 ). The mean values for high-density lipoprotein, low-density lipoprotein, and total cholesterol were normal. The mean triglycerides were elevated in the obese children. No correlation was found between high-density lipoprotein, low-density lipoprotein, triglycerides, or total cholesterol and any of the biophysical properties of the aorta.
| Variable | Obese Children (BMI ≥95th Percentile) | Adult Range | 95th Percentile for 14 Years Old |
|---|---|---|---|
| High-density lipoprotein(mmol/L) [mg/dl] | 1.15 ± 0.25 [45 ± 10 mg] | >1.0 [39] | 0.96 (5th percentile) |
| Low-density lipoprotein(mmol/L) [mg/dl] | 2.48 ± 0.62 [96 ± 24] | 1.50–2.79 [58–108] | 3.5 [135] |
| Total cholesterol (mmol/L) [mg/dl] | 4.30 ± 0.82 [166 ± 32] | 2.00–4.39 [77–170] | 5.3 [205] |
| Triglycerides (mmol/L) [mg/dl] | (1.69 ± 1.08) [149 ± 95] | 0.40–1.29 [35–113] | 1.3 [114] |
The LV dimensions and mass were significantly greater in the obese children compared to the normal-weight controls ( Table 4 ). When indexed to height, the LV dimensions were similar in the obese and normal-weight children ( Table 4 ). Differences in the LV mass persisted even after indexing for height (p <0.001). The standard measures of LV systolic function were similar in the obese children and normal-weight controls, but the diastolic indexes were altered in the obese children ( Table 5 ).
| Obese Children (BMI ≥95th Percentile) | Controls | p Value | |
|---|---|---|---|
| Left ventricular end-diastolic dimension (cm) | 5.0 ± 0.4 | 4.6 ± 0.5 | <0.001 |
| Left ventricular end-diastolic dimension indexed to height (cm/m 2.7 ) | 1.4 ± 0.2 | 1.4 ± 0.3 | NS |
| Left ventricular end-systolic dimension (cm) | 3.1 ± 0.4 | 2.9 ± 0.4 | <0.001 |
| Left ventricular end-systolic dimension indexed to height (cm/m 2.7 ) | 0.84 ± 0.21 | 0.88 ± 0.21 | NS |
| Left ventricular posterior wall in diastole (cm) | 0.81 ± 0.12 | 0.74 ± 0.13 | <0.005 |
| Left ventricular posterior wall in diastole indexed to height (cm/m 2.7 ) | 0.21 ± 0.05 | 0.23 ± 0.05 | NS |
| Left ventricular posterior wall in systole (cm) | 1.33 ± 0.17 | 1.21 ± 0.21 | 0.001 |
| Left ventricular posterior wall in systole indexed to height (cm/m 2.7 ) | 0.36 ± 0.07 | 0.37 ± 0.08 | NS |
| Left ventricular mass (g) | 178 ± 44 | 139 ± 53 | <0.001 |
| Left ventricular mass indexed to height (g/m 2.7 ) | 47.0 ± 8.8 | 40.8 ± 10.5 | 0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
