Lean Body Mass May Explain Apparent Racial Differences in Carotid Intima-Media Thickness in Obese Children


Racial differences in carotid intima-media thickness (cIMT) have been suggested to be associated with the disproportionally high prevalence of cardiovascular disease in black adults. The objective of this study was to evaluate the effects of cardiovascular risk factors on the racial differences seen in cIMT in obese children.


Obese subjects aged 4 to 21 years were recruited prospectively. Height, weight, blood pressure, fasting insulin, glucose, lipid panel, high-sensitivity C-reactive protein, and body composition by dual-energy x-ray absorptiometry were obtained. B-mode carotid imaging was analyzed by a single blinded physician.


A total of 120 subjects (46 white, 74 black) were enrolled. Black subjects exhibited greater cIMT (0.45 ± 0.03 vs 0.43 ± 0.02 cm, P < .01) and higher lean body mass index (19.3 ± 3.4 vs 17.3 ± 3.2 kg/m 2 , P = .02) than white subjects. Simple linear regression revealed modest associations between mean cIMT and race ( R = 0.52, P < .01), systolic blood pressure ( R = 0.47, P < .01), and lean body mass ( R = 0.51, P < .01). On multivariate regression analysis, lean body mass remained the only measure to maintain a statistically significant relationship with mean cIMT ( P < .01).


Black subjects demonstrated greater cIMT than white subjects. The relationship between race and cIMT disappeared when lean body mass was accounted for. Future studies assessing the association of cardiovascular disease risk factors to cIMT in obese children should include lean body mass in the analysis.

In the United States, adults of black African descent have a higher prevalence of obesity and an increased risk for cardiovascular disease than whites of European origin. Carotid intima-media thickness (cIMT) is a strong predictor of cardiovascular disease; it has been shown to be higher in black than in white healthy adults. Traditional risk factors for cardiovascular disease, such as hypertension, contribute to higher cIMT and increased risk for cardiovascular disease in black adults.

Similar to adults, cardiovascular disease risk factors in childhood and adolescence also show racial differences. In fact, racial differences in cIMT have been reported in healthy nonobese children. The etiology behind the racial differences in cIMT in children is not clear, and no studies have examined whether such differences persist in obese children, a group at high risk for future cardiovascular disease. If racial differences are found in cIMT in obese patients, the cardiovascular risk factors associated with these differences may provide targets for intervention in future studies. The primary objectives of this study were (1) determine if racial differences exist in cIMT between white and black obese children and, if such differences are present, (2) to identify measures of body composition and markers of cardiovascular risk that contribute to these differences. We hypothesized that black obese children would have higher cIMT than whites and that blood pressure, race, and lean body mass would be associated with cIMT.


This was a prospective, cross-sectional study. All tests were conducted during a single assessment using a standardized protocol. The protocol was approved by the institutional review board. Informed consent was obtained from the parents or legal guardians of minors or from participants aged ≥18 years.

Subject Population

Patients were recruited from the Medical University of South Carolina’s childhood obesity management clinic. Inclusion criteria were (1) body mass index (BMI) > 95th percentile, (2) age 4 to 21 years, and (3) white or black race. Patients of Hispanic ethnicity were not included in the analysis. Subjects who were pregnant, were taking insulin, or were taking oral steroids were excluded. Patients were enrolled consecutively as long as they were not of Hispanic ethnicity and did not have one of the three exclusion factors listed. Study visits were rescheduled if patients had experienced febrile illnesses within 72 hours of the planned study date.


Patients’ anthropomorphic assessments were performed at the Clinical and Translational Research Center. Blood pressure was measured using an automatic cuff (Dinamap; GE Healthcare, Little Chalfont, United Kingdom) with an appropriately sized cuff after remaining seated for 5 min. The average of two blood pressure measurements, one taken at the beginning of the visit and one at the end, was used in the analysis. Patients’ fasting status was confirmed before phlebotomy. Laboratory values obtained included serum insulin, glucose, low-density lipoprotein (LDL), high-density lipoprotein, triglycerides, and high sensitivity C-reactive protein. Body composition (total body fat, percentage body fat, and lean body mass) was quantified using dual-energy x-ray absorptiometry.

Carotid arteries were studied with a duplex scanner using a 7.5-MHz linear-array transducer (iE33, 4–8 MHz; Philips Medical Systems, Andover, MA). All B-mode carotid imaging was performed by a single sonographer. Participants were positioned supine with the neck rotated at 45° to expose an area from the clavicle to the angle of the jaw. Recommendations from the American Society of Echocardiography’s consensus statement on cIMT were followed; that is, measurements from both carotid arteries were used, and cIMT was measured only from the far wall of the artery. Measurements from the near wall were not used, as recommendations from the Measuring Effects on Intima-Media Thickness: An Evaluation of Rosuvastatin study group were not published before the start of the study. Right and left common cIMTs were imaged longitudinally 1 cm proximal to the carotid bifurcation with the transducer placed both in the lateral and anterior-posterior windows. Three 5-sec acquisitions were recorded, and three magnified end-diastolic frames of the far wall were selected and analyzed at each position ( Figure 1 ). All studies were read offline by a single physician blinded to the clinical and laboratory data using QLAB version 8.1 (Phillips Medical Systems, Bothell, WA) with automatic detection of cIMT by the software ( Figure 2 ). For each subject, the mean cIMT was calculated as the average of the 12 measurements from the left and right common carotid arteries (3 frames for each position × 2 positions × 2 carotid arteries = 12 measurements). A subgroup of 30 studies (the initial 15 and the final 15) was reanalyzed at a 4-week time interval to assess for intraobserver and interobserver variability. All 12 measurements were repeated and averaged. Observers were free to choose the image and frame to remeasure.

Figure 1

Common carotid artery ultrasound. Representative still frame at end-diastole from a right common carotid artery 1 cm below the carotid bifurcation with the ultrasound beam directed anterior to posteriorly. The yellow arrows bracket the intima-media of the far wall of the common carotid artery, where measurements will be made. The distance between the depth markers (major and minor hash marks) to the right of the figure is 5 mm. The two orange opposing triangles indicate the focal depth.

Figure 2

Measurement of cIMT. This figure shows the offline software analysis of the common carotid ultrasound from Figure 2. The software requires the user to manually place a blue box 1 cm in length over the far wall of the carotid artery. The software then automatically traces the intima-media borders ( yellow lines ). The reported carotid-intima media thickness in this patient was 0.51 mm. The distance between the depth markers ( major and minor hash marks ) to the right of the figure is 5 mm.


BMI was calculated as weight (kg)/height 2 (m 2 ). Lean BMI was calculated as lean body mass (kg)/height 2 (m 2 ) and fat mass index as fat mass (kg)/height 2 (m 2 ). Body surface area was calculated using the method of Haycock et al . as 0.024265 × height 0.3964 × weight 0.5378 . The quantitative insulin sensitivity check index (QUICKI) was calculated as 1/{log[fasting insulin (μU/mL)] + log[fasting glucose (mg/dL)]}. Insulin level was divided dichotomously as normal (< 20 μIU/mL) or abnormal (≥20 μIU/mL).

Statistical Analysis

To assess for differences between white and black patients, two-group t tests were used for parametric data, and Mann-Whitney U tests were used for nonparametric data. Simple linear regression was used to assess the individual effects of cardiovascular risk factors on cIMT. Multivariate regression was used to model the relationship between two or more independent variables and cIMT. Intraobserver and interobserver variability was assessed using intraclass correlation coefficients using a random-effects model measuring absolute agreement. On the basis of clinical relevance and previous studies investigating the correlation of lean body mass to cIMT, an effect size of r = 0.25 was chosen to base the sample-size calculation. This resulted in a sample size of 120 subjects, giving a power of 80% at α = 0.05 to detect the chosen effect size. All statistical analyses were performed using SPSS Statistics version 20 (IBM, Armonk, NY).


From September 2009 to December 2011, 142 patients were eligible for inclusion. Nineteen declined participation, and three met exclusion criteria. Therefore, 120 obese children (46 white [72% female] and 74 black [64% female]) were enrolled. Figure 3 demonstrates the age distribution of subjects. Differences by race in clinically derived anthropometrics and laboratory data can be found in Table 1 . Five patients were on antihypertensive medicines. Eight other patients had blood pressure values above the 95th percentile for age, sex, and height at the time of the visit without the diagnosis of hypertension. No patients were diagnosed with diabetes mellitus. Two patients were smokers. Black patients had higher insulin levels and evidence of insulin resistance by QUICKI compared with white patients, whereas white patients had increased triglycerides compared with blacks. Body composition and cIMT results are summarized in Table 2 . Intraobserver and interobserver variability for cIMT measures had intraclass correlation coefficients of r = 0.92 and r = 0.86, respectively. Although BMI was higher in black subjects than in white subjects by about 13%, fat mass index and percentage body fat were not significantly different between groups. This was consistent with the finding that lean body mass was significantly higher in black subjects by almost 16%. Also, black patients demonstrated higher cIMT compared with white patients. There were no statistically significant differences in anthropomorphic, laboratory, or imaging data between men and women.

Figure 3

Age distribution of subjects.

Table 1

Differences by race in clinically derived anthropometrics and laboratory data

Measure White Black P value ( t test or Mann-Whitney U test)
Age (y) 12.5 ± 3.6 11.8 ± 3.3 .30
Height (cm) 158 ± 12.8 155 ± 13.7 .30
Weight (kg) 79.5 ± 25.7 88.1 ± 32.3 .13
Body surface area (m 2 ) 1.8 ± 0.3 1.9 ± 0.4 .43
Systolic blood pressure (mm Hg) 113 ± 13 112 ± 17 .63
Diastolic blood pressure (mm Hg) 63 ± 8 61 ± 9 .15
Insulin (μIU/mL) 22 (7–119) 28 (11–116) .03
Glucose (mg/dL) 91 ± 7.6 92 ± 8.2 .53
QUICKI 0.30 ± 0.03 0.29 ± 0.02 .02
LDL (mg/dL) 103 ± 24 104 ± 24 .73
High-density lipoprotein (mg/dL) 41 ± 11 42 ± 9.2 .49
Triglycerides (mg/dL) 95 (32–297) 65 (29–175) <.01
High-sensitivity C-reactive protein (mg/dL) 0.25 (0.02–2.41) 0.38 (0.01–1.98) .11

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

May 31, 2018 | Posted by in CARDIOLOGY | Comments Off on Lean Body Mass May Explain Apparent Racial Differences in Carotid Intima-Media Thickness in Obese Children

Full access? Get Clinical Tree

Get Clinical Tree app for offline access