Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Am Coll Cardiol. Author manuscript; available in PMC 2010 June 16.
Published in final edited form as:
PMCID: PMC2747309

Risk Factors Associated with Aortic and Carotid Intimal-Medial Thickness in Adolescents and Young Adults: the Muscatine Offspring Study

Jeffrey D. Dawson, Sc.D.,* Milan Sonka, Ph.D., Mary Beth Blecha, R.V.T., Wenjiao Lin, M.S.,*§ and Patricia H. Davis, M.D.



To determine whether cardiovascular risk factors are associated with aortic and carotid intimal-medial thickness (aIMT and cIMT) in adolescents and young adults.


Atherosclerotic lesions begin developing in youth, first in the distal abdominal aorta and later in the carotid arteries. Knowledge of how risk factors relate to aIMT and cIMT may help in the design of early interventions to prevent cardiovascular disease.


Participants were 635 members of the Muscatine Offspring cohort. The mean aIMT and cIMT were measured using an automated reading program.


The means (SDs) of aIMT and cIMT were 0.63 (0.14) mm and 0.49 (0.04) mm, respectively. In adolescents (ages 11 to 17), aIMT was associated with triglycerides, systolic blood pressure (SBP), diastolic blood pressure (DBP), body mass index (BMI), and waist/hip ratio, after adjusting for age, gender, and height. In young adults (ages 18 to 34), aIMT was associated with those same five risk factors, plus HDL-cholesterol and pulse pressure. In adolescents, cIMT was associated with SBP, pulse pressure, heart rate, BMI, and waist/hip ratio. In young adults, cIMT was associated total cholesterol, LDL-cholesterol, triglycerides, SBP, .DBP, BMI, waist/hip ratio, and HbA1C. In both age groups, aIMT and cIMT were significantly correlated with the PDAY coronary artery risk score.


Both aIMT and cIMT are associated with cardiovascular risk factors. Using aIMT in adolescents gives information beyond that obtained from cIMT alone. Measurement of aIMT and cIMT may help identify those at risk for premature cardiovascular disease.

Keywords: Atherosclerosis, Ultrasound, Preclinical disease, Abdominal aorta, IMT

Atherosclerosis begins in childhood with the accumulation of lipid in the intima of arteries to form fatty streaks. Fatty streaks occur in the aorta in almost every child over the age of 3 years (1). Data from the Pathobiological Determinants of Atherosclerosis in Youth Study (PDAY), an autopsy study of individuals aged 15 to 34 years, demonstrated a propensity to develop atherosclerotic lesions within certain segments of the affected arteries. In the abdominal aorta, lesions were most likely to develop in the dorsolateral vessel just proximal to the bifurcation (2). Atherosclerotic lesions occur later in life in the coronary and carotid arteries than in the aorta but there is a strong association between atherosclerosis in the abdominal aorta and coronary arteries (3). In the PDAY Study, several risk factors were strongly associated with atherosclerotic lesions of all grades (2).

Early atherosclerotic lesions may also be detected in living asymptomatic individuals non-invasively by using ultrasound to measure the intimal-medial thickness (IMT) of the affected vessel. In older adults, this method has been widely used to assess the carotid arteries and increased carotid IMT (cIMT) predicts the risk of developing myocardial infarction (MI), stroke and peripheral vascular disease (4). In young and middle-aged adults, increased cIMT is associated with cardiovascular risk factors (5-7). In case-control studies, high-risk children who have hypertension, obesity, familial hyperlipidemia, type 1 diabetes, or a parental history of early coronary artery disease (the cases) have significantly higher cIMT compared to controls (8). Serial cIMT measurements have been used to monitor the effect of statin therapy in children with familial hyperlipidemia (9).

Since the atherosclerotic changes occur earlier in the abdominal aorta than in the carotid arteries, it has been suggested that measurement of aortic IMT (aIMT) may be a more sensitive indicator in younger individuals. Detection of the earliest manifestations of atherosclerosis could allow risk factor modification at an even younger age when the lesions are less advanced. McGill et al. recently suggested that risk factor control in youth would be the most effective strategy to prevent heart disease in the 21st century (2). The American Academy of Pediatrics recommends that starting as early as 2 to 10 years of age, children with risk factors or a family history of premature heart disease or hyperlipidemia should be screened using a fasting lipid profile. This approach was recommended because there was no currently available noninvasive method to assess atherosclerosis in children (10). In the study reported herein, we measured aIMT and cIMT in adolescents and young adults who are members of the Muscatine Study Offspring cohort, to identify the relationship of concurrently measured cardiovascular risk factors with both of these ultrasound-derived IMT measures.


From 1996 to 2001, 788 members of the Muscatine Study Longitudinal Adult Cohort participated in a baseline examination of cIMT at ages 33 to 46 (5). At the time of the exam, we obtained information about their children, enabling us to ask them to participate in the current study. We invited all eligible offspring of each contacted family, with priority given to children in the age range of 11 to 20. Eligibility requirements included being at least 11 years old, not pregnant, and without a medical condition that would preclude examination (two were excluded due to traumatic neurological disorders). Data were obtained from 635 offspring. This study was approved by the Institutional Review Board of the University of Iowa, and all participants provided signed informed consent.

Risk Factor Measurements

Height was recorded to the nearest 0.5 cm and weight was recorded to the nearest 0.1 kg. Systolic and diastolic blood pressures (SBP and DBP) were recorded on each participant following a five-minute seated rest and by measurement of pulse obliteration pressure. Glycosylated hemoglobin (HbA1C) and lipids were measured in the University of Iowa Core Pathology Laboratory. Total cholesterol and triglycerides were measured by automated colorimetric, enzymatic assays using the Spectrum High Performance Diagnostic System (Abbott Laboratories). HDL-C was measured in the supernatant after dextran sulfate - MgCl2 precipitation of VLDL- and LDL-cholesterol (LDL-C) (Abbott Laboratories). LDL-C was calculated as total cholesterol - HDL-C - triglycerides/5. Smoking history was obtained using data from health questionnaires but pack-years were calculated for young adults only as no adolescents smoked more than a cigarette a day. Coronary artery PDAY scores were calculated as a composite measure of cardiovascular disease risk, based on age, gender, HDL-C, non-HDL cholesterol, smoking status, blood pressure, BMI, and HbA1C (2).

Obtaining and Measuring IMT

A Biosound Technos/MPX ultrasound system along with a linear and a convex array transducer was used to obtain the images. Fasting for 8 hours before the examination was necessary to visualize the aorta. Two images of the far wall IMT were obtained in the distal 10 mm of the abdominal aorta proximal to the iliac. For the carotid arteries, the near and far wall of the left and right internal, bifurcation and common carotid arteries were imaged at three angles using a Plexiglas neck collar following published guidelines (11).

Aortic and carotid wall borders were identified in ultrasound B-mode image data using a globally optimal graph search border detection approach within the Carotid Analyzer 5 software system (12). For aorta images, aIMT was calculated as the mean thickness along the 10 mm length, and a mean IMT measure was then computed from the two images to obtain the overall aIMT value used for analysis. For carotid IMT, the mean across angles of the measures was first calculated to obtain location-specific means, with the average of these 12 measures computed as the overall cIMT value. If measures were obtained at less than 12 locations, the mean of the available locations was used.

Statistical Methods

Descriptive statistics were obtained for demographic, risk factor, and IMT measures. For associations between risk factors and IMT measures, mixed effects models were fit to accommodate and estimate within-family correlations, using SAS Proc Mixed Version 9.1.3. Partial correlation coefficients (R) were obtained based on Wald scores, and partial coefficients of determination (R2) were based on likelihood ratio test (13). Olkin's method (14) was used to compare correlation estimates. In all risk factor analyses, age and gender were included as covariates or stratification variables. Height was also included as a covariate in adolescents, as a surrogate for maturity. After examining the adjusted effects of risk factors individually, a manual stepwise procedure was used to build predictive models with multiple risk factors.


Data were obtained for 313 males and 322 females from 365 families. There were 161 families with one participating offspring, 155 with two, and 49 with three or more. The average (SD) age was 20.4 (5.8), with a range from 11 to 34 years; 228 offspring (36%) were less than 18 years old. Nearly all participants (98.5%) were Caucasian, corresponding to the racial composition of the Muscatine Study Longitudinal Adult Cohort. Of the 635 offspring, 606 (95%) had ultrasound images that allowed aIMT measurement. Carotid IMT was measured in all participants, with a mean of 11.5 walls measurable. Means (SDs) for aIMT were 0.63 (0.14) mm and 0.61 (0.13) mm for males and females, respectively. For cIMT, the mean (SD) was 0.50 (0.04) mm for males and 0.49 (0.04) mm for females. Figure 1 displays boxplots illustrating the sample distributions of aIMT and cIMT for adolescents (ages 11 to 17) and young adults (ages 18 to 34). Note that compared to cIMT, aIMT tends to be higher, has more variability, and increases more with age. A partial correlation of 0.35 was observed between aIMT and cIMT values. Stratifying by gender and age group, descriptive statistics for demographic, risk factor, and IMT measures were calculated (Table 1). As expected, the older offspring have higher levels of most risk factors, as well as higher IMT.

Figure 1
Boxplots showing distributions of aIMT and cIMT for adolescents and young adults
Table 1
Mean (SD) or %, for demographic, risk factor, and intimal-medial thickness measures, by gender and age group.

Partial correlations of individual risk factors and IMT are shown in Table 2. In adolescents, aIMT was associated with triglycerides, SBP, DBP, body mass index (BMI), and waist/hip ratio. In young adults, aIMT was associated with those same five risk factors, plus HDL-cholesterol and pulse pressure. In adolescents, cIMT was associated with SBP, pulse pressure, BMI, and waist/hip ratio. In young adults, cIMT was associated total cholesterol, LDL-C, triglycerides, SBP, DBP, BMI, waist/hip ratio, and HbA1C. All of these associations were in the anticipated direction. There was a borderline indication (p=0.045) of an unexpected negative association between heart rate and cIMT in adolescents. Both aIMT and cIMT were positively correlated with the PDAY coronary artery risk score. Additional analysis showed that aIMT was associated with the PDAY score in adolescents after adjusting for cIMT (R=0.283, p<0.001), while cIMT was associated with the PDAY score in young adults after adjusting for aIMT (R=0.199, p=0.001). Triglycerides, DBP, BMI, and waist/hip ratio had stronger correlations with aIMT than with cIMT in adolescents, while HDL had a stronger correlation with cIMT than with aIMT in young adults (p<0.05).

Table 2
Partial correlation coefficients of risk factors individually predicting aortic and carotid intimal-medial thickness, adjusting for age and gender in all subjects, and for height in those 11−17 years of age.

The association analyses summarized in Table 2 assumed no interaction between risk factors and gender. This assumption was tested and found to be violated in four cases (p<0.05). Table 3 shows the gender-specific associations for those four models. Note that in each case, the association between the risk factor and cIMT was estimated to be higher in males than in females.

Table 3
Gender-specific partial correlation coefficients of risk factors from multivariable models predicting carotid intimal-medial thickness, where gender-risk factor interaction effects were significant, adjusting for age in all subjects, and for height in ...

Finally, we examined the simultaneous effects of multiple risk factors, while still adjusting for age, gender, and (in adolescents) height (Table 4). BMI and waist/hip ratio tended to be the predominant risk factors in multivariable models. However, blood pressure was significant in three out of the four models, and cholesterol was significant in one model, even while including body size as a predictor. In these four models, within-family correlations were estimated to be higher for cIMT vs. aIMT in adolescents (0.22 vs. 0.15), as well as in young adults (0.30 vs. 0.00).

Table 4
Partial correlation coefficients of risk factors modeled simultaneously, adjusting for age and gender in all subjects, and for height in those 11−17 years of age.


Both aIMT and cIMT were found to be associated with several cardiovascular risk factors, suggesting that both methods may be effective in detecting atherosclerosis in adolescents and young adults. In adolescents, triglycerides, DBP, BMI, and waist/hip ratio had significantly stronger associations with aIMT than with cIMT. In young adults, HDL-C had a significantly stronger association with cIMT than with aIMT. Though not significant; the correlation of PDAY scores with aIMT appeared higher than with cIMT in adolescents, with a reverse trend in young adults. Hence, stronger risk factor associations appear to be present with aIMT in adolescents, and with cIMT in young adults.

We previously reported the association of childhood risk factors with subclinical atherosclerotic disease measured as a young or middle-aged adult in the parents of this cohort. Elevated childhood BMI was associated with coronary artery calcification in both genders and with increased carotid IMT in women. (5, 15) In the Bogalusa Study, childhood BMI was associated with adult cIMT and persisted after adjusting for the adult BMI, although the association was reduced. (16) The current study provides further support for the importance of childhood obesity in premature atherosclerosis.

There are only a few other studies that measured aIMT and these were predominantly conducted in neonates and children. Using abdominal ultrasound, increased aIMT has been associated with low birth weight (17), intrauterine growth restriction (18), and maternal smoking (19) in studies of neonates. One case-control study of aIMT and cIMT has been conducted in children at high risk for atherosclerosis. Sixteen children with hypercholesterolemia and 44 with type 1 diabetes were compared to 28 healthy controls. There was a greater increase in aIMT than cIMT in high-risk children, although both were significantly thicker than in the control children. Age and DBP were associated with aIMT in the children with diabetes or hyperlipidemia (20). In a second case-control study, seropositivity to c pneumoniae was significantly associated with aIMT but not cIMT in healthy children age 7 to 11 years. (21) Finally, in 512 healthy, 13-year-old, Finnish children, the mean aIMT was associated with BMI, blood pressure, C-reactive protein, and triglycerides on univariate analysis. The mean cIMT was significantly associated with blood pressure, total cholesterol, LDL-C, and male gender (22). We found that aIMT is associated with BMI, waist/hip ratio, blood pressure, and triglycerides in adolescents. In addition, aIMT is associated with HDL-C and pulse pressure in the young adult group. However, in the current study, the cIMT in those under age 18 years was associated only with blood pressure, BMI, and waist/hip ratio and an association with total cholesterol and LDL-C was seen only in the young adults. Both the Finnish study and our study suggest that there are differences in risk factors according to the vascular bed studied. While both aIMT and cIMT are related to blood pressure and BMI, aIMT seems to have a stronger association with risk factors associated with the metabolic syndrome, such as abdominal obesity, high triglycerides and low HDL-C, while cIMT is more strongly related to total and LDL-cholesterol. The mechanisms underlying these differences are unclear.

In older populations, different imaging modalities have been used to detect more advanced atherosclerosis in the aorta, with the results compared to the measurement of carotid IMT (23,24). In a sample of the Framingham Heart Study offspring cohort, the correlation between aortic plaque found on cardiovascular magnetic resonance imaging and cIMT was low, but the proportion of subjects with positive results on multiple tests of subclinical atherosclerosis increased as the Framingham risk score increased (23). In the Rotterdam Study, the presence of aortic atherosclerosis as identified by calcifications on lateral abdominal x-rays was correlated with carotid IMT, but aortic atherosclerosis remained a significant predictor of subsequent MI even after adjustment for cIMT (24). These findings suggest that additional prognostic information may be obtained by imaging the aorta as well as the carotid artery.

There are a limited number of studies of cIMT measurement in adolescents from the general population (25-28). In 247 individuals aged 10 to 20 years, cIMT increased with height and age. In univariate analysis, cIMT was associated with BMI, SBP, pulse pressure, and smoking, while on multivariable analysis, only pulse pressure and BMI were significant (27). In the Stanislaus cohort of 193 participants aged 10 to 24 years, cIMT was not associated with gender or BMI. In males only, there was a borderline significant association with SBP (28). In 60 Japanese children aged 5 to 14 years, cIMT was not associated with any cardiovascular risk factors but did increase with age (26). In a study of 216 children all aged 9 years, cIMT was associated with gender, BMI, SBP and a lower maternal energy intake during pregnancy on univariate analysis (25). Our study further confirms the association of blood pressure and BMI with carotid IMT in adolescents. In prior studies of young adults (ages 18 to 30 years) in the general population cIMT was significantly associated with male gender (30), BMI (29,30), blood pressure (29,30), LDL-C (30), and smoking (31) and inversely with HDL-C (31) and alcohol intake (31). In our current study, in those over age 18 years, cIMT was associated with total cholesterol, LDL-C, triglycerides, SBP, DBP, BMI, waist/hip ratio, and HbA1C. The lack of association with smoking in our cohort may be due to the low number of pack-years.

One limitation of our study is that we did not examine the participants in a manner that would enable us to obtain Tanner stages of maturity, which may be a confounder in the relationship between risk factors and IMT in adolescents. However, in our analyses we adjusted for age and height as surrogates for maturity. Furthermore, we found an association between DBP and aIMT, even after additional adjustment for BMI and waist/hip ratio. This result suggests that our associations between risk factors and aIMT are not just due to the maturation process.

Our study had other limitations, as well. Our list of potential risk factors was not comprehensive, and did not include such variables as birth weight, maternal smoking, or intrauterine growth retardation. Conversely, we did evaluate a number of risk factors, so that some spurious results could have occurred due to multiple comparisons (e.g., perhaps the negative association involving heart rate). Finally, the associations we saw were based on cross-sectional data, which weakens the argument for causality. We have previous risk factor data, obtained four to seven years prior to the current study, on 342 of our 635 participants (mostly current young adults, with mean age 24.6). Preliminary analysis of these data (details not shown) found that cIMT was associated with previous values of total cholesterol, LDL-C, triglycerides, DBP, SBP, BMI, and waist/hip ratio, while aIMT was associated with previous levels of BMI and waist/hip ratio.

One advantage of aIMT over cIMT is that the time involved in scanning and in measuring the vessels is much shorter, due to the fact that only two images are used. The procedure was well-accepted by this offspring cohort. There is a low rate of missing data even though our cohort included obese participants, a group that is at high-risk of early atherosclerosis. A disadvantage is that participants must be fasting and be scanned early in the morning in order for images of adequate quality to be obtained.

We found that aIMT is changing more rapidly than cIMT in the age range we studied and the observed correlation (R=0.35) between aIMT and cIMT was moderate. These findings are consistent with either the pathological evidence that atherosclerotic lesions begin in the aorta or physiologic differences with more rapid growth of the aorta. The former hypothesis is supported by the stronger associations of aortic IMT with risk factors in adolescents, compared to cIMT. As in older adults, additional prognostic information may be found by measuring aIMT in addition to cIMT in adolescents, but long-term follow-up until clinical endpoints occur would be needed to confirm this.

In summary, our results suggest that measurement of aIMT should be considered along with cIMT as a noninvasive method to objectively detect early atherosclerosis in adolescents. These could be important complementary tools to identify those at high risk, given the recent emphasis on prevention of atherosclerosis beginning in childhood (2,10). Further studies are needed to assess aIMT in other population-based studies of adolescents, and to see whether aIMT progression can be measured, so that the effect of interventions to retard the atherosclerotic process can be monitored.


The authors would like to thank Ms. Jeni Michelson for study coordination, Ms. Rebecca Mastbergen for statistical programming assistance, and the rest of the Muscatine Study research team for their contributions towards this study. We are especially indebted to the Muscatine Study participants

All authors received financial support from NIH Award HL54730.


Aortic intimal medial thickness
Carotid intimal medial thickness
Diastolic blood pressure
Glycosylated hemoglobin
High-density lipoprotein cholesterol
Low-density lipoprotein cholesterol
Myocardial infarction
Pathobiological Determinants of Atherosclerosis in Youth
Systolic blood pressure
Standard deviation


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Relationship with industry: Dr. Milan Sonka: co-founder of Medical Imaging Applications, LLC


1. Holman RL, McGill HC, Jr., Strong JP, Geer JC. The natural history of atherosclerosis; the early aortic lesions seen in New Orleans in the middle of the 20th century. Am J Pathol. 1958;34:209–235. [PubMed]
2. McGill HC, Jr., McMahan CA, Gidding SS. Preventing heart disease in the 21st century: implications of the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Study. Circulation. 2008;117:1216–1227. [PubMed]
3. Guzman MA, McMahan CA, McGill HC, Jr., et al. Selected methodologic aspects of the International Atherosclerosis Project. Lab Invest. 1968;18:479–497. [PubMed]
4. Lorenz MW, Markus HS, Bots ML, Rosvall M, Sitzer M. Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and meta-analysis. Circulation. 2007;115:459–467. [PubMed]
5. Davis PH, Dawson JD, Riley WA, Lauer RM. Carotid intimal-medial thickness is related to cardiovascular risk factors measured from childhood through middle age. Circulation. 2001;104:2815–2819. [PubMed]
6. Li S, Chen W, Srinivasan SR, et al. Childhood cardiovascular risk factors and carotid vascular changes in adulthood: The Bogalusa Heart Study. JAMA. 2003;290:2271–2276. [PubMed]
7. Raitakari OT, Juonala M, Kahonen M, et al. Cardiovascular risk factors in childhood and carotid artery intima-media thickness in adulthood: The Cardiovascular Risk in Young Finns Study. JAMA. 2003;290:2277–2283. [PubMed]
8. Slyper AH. What vascular ultrasound testing has revealed about pediatric atherogenesis, and a potential clinical role for ultrasound in pediatric risk assessment. J Clin Endocrinol Metab. 2004;89:3089–3095. [PubMed]
9. Wiegman A, Hutten BA, de Groot E, et al. Efficacy and safety of statin therapy in children with familial hypercholesterolemia: a randomized controlled trial. JAMA. 2004;292:331–337. [PubMed]
10. Daniels SR, Greer FR, the Committee on Nutrition Lipid screening and cardiovascular health in childhood. Pediatrics. 2008;122:198–208. [PubMed]
11. Stein JH, Korcarz CE, Hurst RT, et al. Use of carotid ultrasound to identify subclinical vascular disease and evaluate cardiovascular disease risk: a consensus statement from the American Society of Echocardiography Carotid Intima-Media Thickness Task Force endorsed by the Society for Vascular Medicine. J Am Soc Echocardiography. 2008;21:93–111. [PubMed]
12. Mancini GB, Abbott D, Kamimura C, Yeoh E. Validation of a new ultrasound method for the measurement of carotid artery intima medial thickness and plaque dimensions. Can J Cardiol. 2004;20:1355–1359. [PubMed]
13. Lipsitz SR, Leong T, Ibrahim J, Lipshultz T. A partial correlation coefficient and coefficient of determination for multivariate normal repeated measures data. The Statistician. 2001;50:87–95.
14. Kleinbaum DG, Kupper LL, Muller KE, Nizam A. Applied Regression Analysis and Other Multivariable Methods. 4th Edition Duxbury; Belmont CA: 2008. p. 103.
15. Mahoney LT, Burns TL, Stanford W, et al. Coronary risk factors measured in childhood and young adult life are associated with coronary artery calcification in young adults: The Muscatine Study. J Am Coll Cardiol. 1996;27:277–284. [PubMed]
16. Freedman DS, Patel DA, Srinivasan SR, et al. The contribution of childhood obesity to adult carotid intima-media thickness: the Bogalusa Heart Study. Int J Obes. 2008;32:749–756. [PubMed]
17. Koklu E, Kurtoglu S, Akcakus M, Yikilmaz A, Coskun A, Gunes T. Intima-media thickness of the abdominal aorta of neonate with different gestational ages. J Clin Ultrasound. 2007;35:491–497. [PubMed]
18. Skilton MR, Evans N, Griffiths KA, Harmer JA, Celermajer DS. Aortic wall thickness in newborns with intrauterine growth restriction. Lancet. 2005;365:1484–1486. [PubMed]
19. Gunes T, Koklu E, Yikilmaz A, et al. Influence of maternal smoking on neonatal aortic intima-media thickness, serum IGF-I and IGFBP-3 levels. Eur J Pediatr. 2007;166:1039–1044. [PubMed]
20. Jarvisalo MJ, Jartti L, Nanto-Salonen K, et al. Increased aortic intima-media thickness: a marker of preclinical atherosclerosis in high-risk children. Circulation. 2001;104:2943–2947. [PubMed]
21. Volanen I, Jarvisalo MJ, Vainionpaa R, et al. Increased aortic intima-media thickness in 11-year-old healthy children with persistent chlamydia pneumoniae seropositivity. Arterioscler Thromb Vasc Biol. 2006;26:649–655. [PubMed]
22. Volanen I, Kallio K, Saarinen M, et al. Arterial intima-media thickness in 13-year-old adolescents and previous antichlamydial antimicrobial use: a retrospective follow-up study. Pediatrics. 2008;122:e675–e681. [PubMed]
23. Kathiresan S, Larson MG, Keyes MJ, et al. Assessment by cardiovascular magnetic resonance, electron beam computed tomography, and carotid ultrasonography of the distribution of subclinical atherosclerosis across Framingham risk strata. Am J Cardiol. 2007;99:310–314. [PubMed]
24. van der Meer IM, Bots ML, Hofman A, Iglesias del Sol A, van der Kuip DAM, Witteman JCM. Predictive value of noninvasive measures of atherosclerosis for incident myocardial infarction: The Rotterdam Study. Circulation. 2004;109:1089–1094. [PubMed]
25. Gale CR, Jiang B, Robinson SM, Godfrey KM, Law CM, Martyn CN. Maternal diet during pregnancy and carotid intima-media thickness in children. Arterioscler Thromb Vasc Biol. 2006;26:1877–1882. [PubMed]
26. Ishizu T, Ishimitsu T, Yanagi H, et al. Effect of age on carotid arterial intima-media thickness in childhood. Heart Vessels. 2004;19:189–195. [PubMed]
27. Jourdan C, Wuhl E, Litwin M, et al. Normative values for intima-media thickness and distensibility of large arteries in healthy adolescents. J Hypertens. 2005;23:1707–1715. [PubMed]
28. Sass C, Herbeth B, Chapet O, Siest G, Visvikis S, Zannad F. Intima-media thickness and diameter of carotid and femoral arteries in children, adolescents and adults from the Stanislas cohort: effect of age, sex, anthropometry and blood pressure. J Hypertens. 1998;16:1593–1602. [PubMed]
29. Gonzalez J, Wood JC, Dorey FJ, Wren TAL, Gilsanz V. Reproducibility of carotid intima-media thickness measurements in young adults. Radiology. 2008;247:465–471. [PubMed]
30. Oren A, Vos LE, Uiterwaal CSPM, Grobbee DE, Bots ML. Cardiovascular risk factors and increased carotid intima-media thickness in healthy young adults: The Atherosclerosis Risk in Young Adults (ARYA) Study. Arch Int Med. 2003;163:1787–1792. [PubMed]
31. Knoflach M, Kiechl S, Kind M, et al. Cardiovascular risk factors and atherosclerosis in young males: ARMY Study (Atherosclerosis Risk-Factors in Male Youngsters). Circulation. 2003;108:1064–1069. [PubMed]