| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2006;48:40.)
© 2006 American Heart Association, Inc.
Original Articles |
From the Divisions of Pediatric Nephrology (M.B.L.) and Pediatric Cardiology (C.C.M.), Department of Pediatrics, Department of Radiology (N.L.C.), and Department of Biostatistics and Computational Biology (J.R.), University of Rochester Medical Center, NY.
Correspondence to Marc B. Lande, 601 Elmwood Ave, Box 777, Rochester, NY 14642. E-mail marc_lande{at}urmc.rochester.edu
| Abstract |
|---|
|
|
|---|
85th percentile). The median carotid intima media thickness of hypertensive subjects was greater than that of matched controls (0.67 versus 0.63 mm; P=0.045). In the hypertensive subjects, carotid intima media thickness correlated strongly with several ambulatory blood pressure monitoring parameters, with the strongest correlation for daytime systolic blood pressure index (r=0.57; P=0.003). In the hypertensive group, the prevalence of left ventricular hypertrophy was 32%, but unlike carotid intima media thickness, left ventricular mass index did not correlate with ambulatory blood pressure monitoring. Together, the findings that hypertensive subjects had increased carotid intima media thickness compared with matched controls and that higher carotid intima media thickness correlated with more severe hypertension by ambulatory blood pressure monitoring provide strong evidence that carotid intima media thickness is increased in childhood primary hypertension, independent of the effects of obesity.
Key Words: carotid arteries children hypertension, obesity
| Introduction |
|---|
|
|
|---|
Previous reports have shown that cIMT is increased in several childhood diseases that increase cardiovascular risk, including diabetes,8 familial hypercholesterolemia,9 and end-stage kidney disease.10,11 Recent studies have investigated whether cIMT is also increased in childhood primary hypertension.1217 However, these reports have been confounded by the high proportion of hypertensive subjects with obesity, itself an independent risk factor for increased cIMT.1822 Consequently, after adjusting for body mass index (BMI), previous studies have been inconsistent in showing increased cIMT in hypertensive children and have failed to demonstrate office blood pressure (BP) as an independent predictor of cIMT. Given such limitations, the recent Fourth Report from The Working Group on High Blood Pressure in Children and Adolescents3 stated that further research is needed to determine the clinical use of cIMT as a measure of hypertensive end-organ damage in children.
The objectives of the current study were as follows: (1) to determine whether cIMT is increased in children with primary hypertension after prospectively controlling for obesity by pairing hypertensive subjects with normotensive controls matched for BMI; and (2) to investigate the relationship among cIMT, left ventricular mass index (LVMI), and hypertension severity as measured by 24-hour ambulatory BP monitoring (ABPM).
| Methods |
|---|
|
|
|---|
95th percentile for age, gender, and height on
3 occasions (office hypertension).3 The diagnosis of hypertension was confirmed by ABPM, where hypertension was defined as average daytime and/or nighttime BP
95th percentile for gender and height for ABPM pediatric norms.23 Children with average BP <95th percentile for both day and night were considered to have white-coat hypertension (WCH) and were not included in the hypertension group. Subjects were limited to children aged 10 to 18 years of age without evidence of secondary hypertension. Secondary causes of hypertension were excluded by history, physical examination, urinalysis, serum chemistries, renal ultrasonography, and other tests as indicated, according to guidelines from the Working Group on High Blood Pressure in Children and Adolescents.3 For each hypertensive subject, a matched healthy, normotensive control subject was recruited from the General Pediatric Clinic at the University of Rochester. Control and hypertensive subjects were matched pairwise for gender, age (±1 year), and BMI (±10%). Control subjects were required to have 2 office BP readings with systolic and diastolic BP <90th percentile3 within the preceding 6 months. The study protocol was approved by the Research Subjects Review Board at the University of Rochester. All of the parents and participants >12 years of age gave informed, written consent.
ABPM
Office hypertension subjects had ABPM to confirm hypertension and to characterize the severity of BP elevation. An appropriate cuff size was placed on the nondominant arm. Spacelabs monitors 90217 (Spacelabs Medical) were used. Monitors were programmed to obtain BP readings every 20 minutes during the day (8:00 AM to 10:00 PM) and every 30 minutes at night (10:00 PM to 8:00 AM).23 Wake/sleep periods for ABPM analysis were determined by the patients self-report. BPs were analyzed using Spacelabs software (ABP Report Management System, version 1.03.16). To be considered adequate, ABPM needed to have a minimum of 40 readings with
2 hours between successful readings.23 BP load was defined as the percent of BP readings
95th percentile, and BP index was defined as the average BP divided by the 95th percentile. BP load and index were calculated for systolic and diastolic BP for 24-hour, daytime, and nighttime intervals. The percentage of daynight systolic BP fall (dip) was defined as the percentage decrease in average systolic BP from day to night.
Carotid Ultrasonography
Carotid IMT was determined in both hypertensive and control subjects using a high-resolution M12L matrix transducer on the GE Logic 700 ultrasound machine. On longitudinal 2D ultrasound images of the carotid artery, the near and far arterial walls are displayed as 2 echogenic lines, the adventitia and intima, separated by the hypoechoic media. The distance between the leading edge of the first bright line on the far wall (lumenintima interface) and the leading edge of the second bright line (mediaadventitia interface) indicates the IMT.24 Measurements of the far wall were made 1 cm proximal to the bifurcation of the left carotid artery, with the subjects head turned 45° toward the right.25 The mean of 3 measurements was used for each subject. All of the measurements were performed by a single experienced sonographer (N.L.C.) who was blinded to the subjects BP status. Carotid ultrasonography was performed within 1 month of echocardiography and before the initiation of antihypertensive therapy. Intraobserver reproducibility of cIMT measurements was determined by measuring cIMT twice, 2 weeks apart, in 6 healthy children, aged 10 to 16 years. The average difference between cIMT measures was <4%.
Echocardiography
All of the hypertensive patients underwent a complete 2D echocardiogram with M-mode and Doppler study (Acuson Sequoia or Aspen, Siemens). No patients were found to have coarctation of the aorta, and all had structurally normal hearts. 2D left ventricular mass by the arealength method was measured according to standards published by the American Society of Echocardiography.26 Measurements obtained were reviewed on digitally stored images and confirmed by an experienced physician echocardiographer (C.C.M.) blinded to the BP and cIMT data from the patients using a Siemens KinetDx workstation 3000, version 4.0.0. Left ventricular mass was indexed to height2.7 to correct for the effect of body size on assessment of left ventricular hypertrophy (LVH).27 LVH was defined as LVMI
95th percentile, which was 39.36 g/m2.7 for boys and 36.88 g/m2.7 for girls, respectively.1
Data Analysis
Data are expressed as mean±SD or median and range, where appropriate. Statistical analysis was performed using SAS for Windows 9.1 (SAS Institute, Inc). McNemars test was used to evaluate the difference in group proportions. Pearson correlation coefficients were used for correlations with the hypertensive group. Spearman correlation coefficients were used for correlations with the groups combined, because the control group was not normally distributed. The Wilcoxon rank sum test was used to examine differences between the hypertensive and control groups. P values <0.05 were considered significant for all analyses.
| Results |
|---|
|
|
|---|
95th percentile. These BMI proportions were identical in the matched normotensive group. Table 1 shows matched demographic characteristics and average office BP for hypertensive and normotensive subjects. Table 2 shows ABPM parameters for the 25 hypertension subjects with ABPM.
|
|
Echocardiography
All of the hypertension subjects had echocardiography. Mean LVMI was 36.0±7.1 g/m2.7 (range, 24.0 to 51.0 g/m2.7). Nine (32%) of 28 had LVH. Only 1 subject had an LVMI of 51 g/m2.7, and no subject had LVMI above this value. LVMI did not correlate with office systolic or diastolic BP, ABPM parameters, age, weight, BMI, or BMI z score.
Carotid Ultrasonography
For all of the subjects combined (hypertension, 28; control, 28; WCH, 7), cIMT correlated with BMI (r=0.33; P=0.008) and BMI z score (r=0.26; P=0.04) but not office SBP, diastolic BP (DBP), age, height, or lean body mass. The lower and upper quartiles for cIMT, based on measurements of the normotensive subjects, was
0.58 mm and
0.67 mm, respectively.
Carotid IMT in hypertension subjects (n=28) did not correlate with age, height, weight, BMI, BMI z score, office systolic or diastolic BP, or office pulse pressure. Carotid IMT did not correlate with LVMI (P=0.92). By contrast, among hypertension subjects with ABPM (n=25), there was significant positive correlation between cIMT and several ABPM parameters, with the strongest correlation for daytime systolic index (r=0.57; P=0.003; Figure). Other significant correlations between cIMT and ABPM parameters included daytime systolic load (r=0.54; P=0.005), daytime diastolic load (r=0.56; P=0.004), daytime diastolic index (0.54; P=0.005), daytime mean DBP (r=0.54; P=0.005), 24-hour SBP load (r=0.51; P=0.009), 24-hour DBP load (r=0.50; P=0.01), daytime mean SBP (r=0.43; P=0.03), and nighttime SBP index (r=0.40; P=0.04). Fifteen (60%) of 25 hypertension subjects with ABPM had 24-hour systolic load
50%, a value associated previously with higher risk of cardiovascular end-organ damage in pediatric primary hypertension.29 Fourteen hypertension subjects with ABPM had cIMT in the upper quartile. Of these, 9 (64%) had 24-hour systolic load
50%.
|
Comparison of groups showed that the median cIMT was significantly greater for hypertensive subjects compared with that of normotensive control subjects, 0.67 mm versus 0.63 mm, respectively (P=0.045). The range for cIMT for hypertension subjects was 0.53 to 0.77 mm and for normotensive controls, 0.53 to 0.93 mm. Seventeen (61%) of 28 hypertensive subjects had cIMT measurements at or above the upper quartile compared with only 9 (32%) of 28 control subjects (P=0.032). The control group contained 2 subjects with cIMT measurements greater than the maximum value for the hypertensive group. Both of these control subjects were significantly overweight, with BMI values >98th percentile for age and gender (BMI z scores of 2.13 and 2.31).
| Discussion |
|---|
|
|
|---|
The current study was designed to control directly for the strong confounding effect of obesity on cIMT. It is the first study of cIMT to match hypertensive and control subjects closely for BMI. Our results showed that cIMT was increased in hypertensive children compared with normotensive controls. This finding is particularly remarkable given that 82% of all subjects were either overweight or obese, a factor that would tend to overshadow any potential difference in groups because of hypertension. The current finding of increased cIMT in the hypertensive group provides further evidence that primary hypertension can lead to vascular abnormalities in childhood. The result is consistent with previous autopsy studies that showed early atherosclerotic changes of the aorta and coronary arteries associated with hypertension in adolescence.30,31 The current study also demonstrates the limitations of using cIMT as a marker of hypertensive end-organ damage in children with primary hypertension who are also overweight. There was considerable overlap in the range of cIMT of the hypertensive and normotensive study groups, and the highest individual cIMT measurements were in 2 obese normotensive subjects. These findings are in agreement with previous studies demonstrating the association between obesity and increased cIMT1822 and underscore the importance of childhood obesity as an independent cardiovascular risk factor.
Studies in both adults and children with hypertension have shown that ABPM is superior to office BP as a predictor of hypertensive end-organ damage.28 Consistent with this observation, previous studies of cIMT in childhood primary hypertension failed to show a correlation between office systolic or diastolic BP and cIMT, after adjusting for BMI. The current study also failed to show a correlation between office BP and cIMT. By contrast, we found a strong correlation between cIMT and several ABPM parameters. The relationship was particularly strong for daytime systolic BP load and daytime systolic BP index, parameters that measure the percentage of BP readings over normal and the degree of BP elevation over the normal range, respectively. Considered together, BP load and BP index provide an ABPM expression of the severity of hypertension that has been shown previously to correlate with end-organ damage in hypertensive children.29 The finding of significant correlation between increased hypertension severity by ABPM and higher cIMT suggests a dose effect of hypertension on cIMT and provides further evidence that primary hypertension can lead to vasculature end-organ damage in childhood. However, because of the cross-sectional nature of the current study, one cannot assume that the association between hypertension severity and increased cIMT is causative.
In the current study, 32% of the hypertensive subjects had LVH by echocardiography, but there was no correlation between LVMI and severity of hypertension by office BP or ABPM. These findings are consistent with a previous report by Belsha et al32 in which 35% of children with primary hypertension had LVH, but ABPM parameters did not differ between hypertensive children with and without LVH. By contrast, Sorof et al29 reported a strong correlation between LVMI and ABPM parameters in pediatric primary hypertension. In that study, a relatively high proportion of children had LVMI >51 g/m2.7, a value that is predictive of cardiovascular morbidity in hypertensive adults and greater than the 99th percentile for LVMI in normal children. Regardless, the current finding of a positive correlation of ABPM parameters with cIMT, but not with LVMI, has potentially important implications. This result suggests that, in the subjects studied, the presence of increased cIMT was complementary to the presence of LVH in the detection of children with hypertensive end-organ damage. In addition, the current study found that cIMT did not correlate with LVMI, a result that differs from a previous report by Sorof et al.12 This disparity in results may be because of potential differences between studies in the severity of hypertension, the duration of hypertension before referral, or differences in subject genetic predisposition to hypertensive end-organ damage. The difference in results may also be because of potential differences between studies in the type of hypertension present (systolic, diastolic, or combined systolic and diastolic). Further study of the vascular effects of primary hypertension in childhood may help clarify the relationship between the level of BP elevation, the duration of hypertension, and the development of hypertensive end-organ damage.
The current study has several limitations. The sample size is relatively small, limiting power for subgroup analysis. For example, BMI was significantly correlated with cIMT for all of the subjects combined, but not for the HTN subjects and control subjects when analyzed separately. Furthermore, ABPM was not performed in the control subjects to document normotension. Although 2 normal office BP readings were required, some of the control subjects may have had masked hypertension (normal office BP but elevated ABPM), an entity that may represent increased cardiovascular risk.33 In addition, there were potential differences in the quality of the office BP assessments. Hypertensive subjects had more office BP readings and had their BP measured in a hypertension subspecialty clinic, where the method for BP measurement may have been more standardized. Control subjects also did not have echocardiography, so the differences between hypertensive and control LVMI could not be assessed. Lastly, the majority of subjects were overweight, limiting analysis of the effects of hypertension on cIMT in normal weight children.
Perspectives
The current study showed that cIMT was increased in hypertensive children compared with controls matched for BMI and that higher cIMT correlated with more severe hypertension as determined by ABPM. These findings provide further evidence that childhood primary hypertension is associated with vascular pathology, independent of the effects of obesity. Used together with echocardiography, measures of vascular damage may help identify hypertensive children who are at increased cardiovascular risk. Our findings confirm that hypertension-associated vascular pathology can occur in childhood and suggest that future research directed at the mechanisms of vascular damage associated with childhood primary hypertension may lead to improved cardiovascular outcomes.
| Acknowledgments |
|---|
Sources of Funding.
This research was supported by a grant from the National Kidney Foundation of Upstate New York. M.B.L. is supported, in part, by National Heart, Lung, and Blood Institute grant K23 HL080068.
Disclosures
None.
Received February 16, 2006; first decision March 6, 2006; accepted March 28, 2006.
| References |
|---|
|
|
|---|
Related Article:
This article has been cited by other articles:
![]() |
E. Urbina, B. Alpert, J. Flynn, L. Hayman, G. A. Harshfield, M. Jacobson, L. Mahoney, B. McCrindle, M. Mietus-Snyder, J. Steinberger, et al. Ambulatory Blood Pressure Monitoring in Children and Adolescents: Recommendations for Standard Assessment: A Scientific Statement From the American Heart Association Atherosclerosis, Hypertension, and Obesity in Youth Committee of the Council on Cardiovascular Disease in the Young and the Council for High Blood Pressure Research Hypertension, September 1, 2008; 52(3): 433 - 451. [Full Text] [PDF] |
||||
![]() |
B. Williams The year in hypertension. J. Am. Coll. Cardiol., May 6, 2008; 51(18): 1803 - 1817. [Full Text] [PDF] |
||||
![]() |
J. Gonzalez, J. C. Wood, F. J. Dorey, T. A. L. Wren, and V. Gilsanz Reproducibility of Carotid Intima-Media Thickness Measurements in Young Adults Radiology, May 1, 2008; 247(2): 465 - 471. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Puato, P. Palatini, M. Zanardo, F. Dorigatti, C. Tirrito, M. Rattazzi, and P. Pauletto Increase in Carotid Intima-Media Thickness in Grade I Hypertensive Subjects: White-Coat Versus Sustained Hypertension Hypertension, May 1, 2008; 51(5): 1300 - 1305. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Genovesi, F. Pieruzzi, M. Giussani, V. Tono, A. Stella, A. Porta, M. Pagani, and D. Lucini Analysis of Heart Period and Arterial Pressure Variability in Childhood Hypertension: Key Role of Baroreflex Impairment Hypertension, May 1, 2008; 51(5): 1289 - 1294. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. W. Rooke Controversies in vascular screening art versus science Vascular Medicine, August 1, 2007; 12(3): 235 - 242. [Abstract] [PDF] |
||||
![]() |
J. T. Flynn What Is the Significance of Increased Carotid Intima Media Thickness in Hypertensive Adolescents? Hypertension, July 1, 2006; 48(1): 23 - 24. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2006 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |