Abstract
Left ventricular hypertrophy (LVH) is prevalent among hypertensive children; however, blood pressure (BP) does not predict its presence. The authors conducted a 1‐year prospective cohort study to examine the hypothesis that obesity‐related risk factors are associated with left ventricular mass index (LVMI) in hypertensive children, and the association between adiposity and LVMI is mediated by BP‐dependent and ‐independent pathways. A total of 49 hypertensive children were enrolled: 51% were overweight/obese and 41% had LVH at baseline. Children overweight/obese at baseline and follow‐up had a greater LVMI increase than those of healthy weight at each visit: mean change of 6.4 g/m2.7 vs 0.95 g/m2.7. Baseline body mass index z score was independently associated with LVMI change (β=4.08, 1.54–6.61; P=.002). Only pulse pressure and serum aldosterone partially mediated this relationship. Hypertensive youth manifest multiple cardiovascular disease risk factors that worsen over time despite treatment. Of these, adiposity is most associated with LVH and increasing LVMI.
The prevalence of cardiovascular disease (CVD) risk factors such as obesity and hypertension in children continues to rise. While the burden of CVD risk factors and associated morbidity and mortality in adulthood are widely known, the burden of hypertension and the extent of end‐organ damage in children have been underappreciated.
Left ventricular hypertrophy (LVH), a pathological remodeling of the heart associated with heart disease and mortality in adults, is a common manifestation of early CVD in children.1, 2 Hypertension has been considered the main cause of LVH, presumably as a response to increased left ventricular (LV) afterload. As such, echocardiography is now recommended for all hypertensive children.1, 3
LVH is diagnosed by an elevated LV mass index (LVMI; LV mass [g]/height [m2.7]), defined as ≥95th percentile. Up to 41% of children have LVH at initial diagnosis of hypertension1, 2 but the degree of blood pressure (BP) elevation is not associated with its presence.2, 4 Further, LVH can occur in normotensive children.5, 6 These findings imply that mechanisms independent of BP may also be responsible for the development of LVH among hypertensive children.
Based on earlier work,2 we hypothesized that obesity and obesity‐related risk factors are associated with LVMI in hypertensive children over time. We also hypothesized that adiposity leads to increased LVMI via several pathways: BP‐dependent pathways (mediated by elevated BP itself and by hormonal regulators of BP) and BP‐independent pathways (mediated by metabolic dysregulation, increased intravascular volume, and inflammation).
Methods
Study Design
We conducted a prospective, observational study of hypertensive children referred for care at the pediatric nephrology clinic at Johns Hopkins University. Hypertensive children aged 3 to 22 years with either a systolic BP (SBP) or diastolic BP (DBP) ≥95th percentile for their age/sex/height1 or taking antihypertensive medication for the treatment of hypertension at the time of their clinic visit were eligible. Children with a history of congenital heart disease, cancer, or chronic kidney disease stage ≥2 were ineligible. The Johns Hopkins University School of Medicine institutional review board approved this study.
Data Collection and Variables
All enrolled participants 18 years and older provided informed consent and those younger than 18 years provided assent to participate in addition to the parental consent that was obtained. Standardized assessments were conducted at baseline and at 12 months where demographic and clinical information were collected. Children provided a 3‐day diet history that was reviewed for accuracy and enhanced recall.
Anthropometric measurements including height, measured to the nearest 0.1 cm using a wall‐mounted stadiometer, and weight, measured to the nearest 0.1 kg using a calibrated balance scale, were obtained. Waist circumference measurements were taken 1 cm above the patients’ navel on exhale using a Gulick tape measure, which applies known tension. All measurements were taken in triplicate and the mean recorded to the nearest 0.1 cm.
Clinic BP (CBP) was measured by manual auscultation after 5 minutes of rest in the right arm with a calibrated aneroid sphygmomanometer according to standardized methods.7 Three seated measurements, 30 seconds apart, were averaged for one composite measurement. Individuals responsible for CBP measurement underwent training and passed yearly certification evaluations.
All children also underwent 24‐hour oscillometric ambulatory BP monitoring (ABPM)8 with BP measurements obtained every 20 minutes (Spacelabs 90201, Spacelabs Inc, Redmond, WA). Awake and sleep cutoffs were determined by diary. As normative BPs for children vary based on age/sex/height, each child's SBP and DBP was divided by his/her age‐/sex‐/height‐specific 95th percentile BP to compare BP elevations between children.1, 8 This provides a 95th percentile BP index; values ≥1 indicate measurements ≥95th percentile. Pulse pressure was also calculated (SBP – DBP in mm Hg).
Images from the standardized two‐dimensional guided M‐mode assessment for left ventricular mass (LVM) were obtained by echocardiography at baseline and at 12 months and were digitally recorded. Three measurements were taken for each patient and averaged by a single study cardiologist (KWH). LVM was calculated based on diastolic measurements of the LV dimension, intraventricular septum, and LV posterior wall.9, 10 LVMI was calculated as LVM (g)/height (m2.7) and LVH was defined as LVMI ≥95th percentile.11
Laboratory Assessments
Random blood samples were obtained at baseline and at 12 months. Serum aldosterone (ng/dL) and plasma renin activity (ng/mL/h) were measured at Quest Diagnostics using liquid chromatography tandem mass spectrometry (LC/MS/MS), with an analytical sensitivity of 1.0 ng/dL and 0.03 ng/mL/h, respectively. High‐sensitivity C‐reactive protein (hs‐CRP; mg/dL) was determined using highly sensitive immunonoturbidimetry, and glycated hemoglobin (%) was measured by high‐performance liquid chromatography. Serum uric acid concentration (mg/dL) was determined by enzymatic spectrophotometry. Additional laboratory assessments included 25‐hydroxyvitamin D (ng/mL); lipids (total cholesterol, low‐density lipoprotein, high‐density lipoprotein, triglycerides; all mg/dL); lipoprotein (a) (nmol/L); and pro‐B natriuretic peptide (pg/mL). Twenty‐four–hour urinary sodium excretion was calculated by multiplying the total urine volume by the measured sodium concentration. Urine collections were deemed appropriate for inclusion when the 24‐hour urine excretion of creatinine was between 14 mg/kg/24 h and 25 mg/kg/24 h.
Demographic and Clinical Characteristics
Children were categorized as African American or non–African American based on self‐report. Body mass index (BMI) was calculated as kg/m2 and age‐sex specific BMI percentiles and z scores were determined.12, 13 Children were defined as overweight/obese when BMI was ≥85th percentile.14
Statistical Analyses
Demographic and clinical variables were compared between children with and without LVH using Student t tests with unequal variances for continuous variables that were normally distributed, Wilcoxon rank‐sum for continuous variables not normally distributed, and Fisher exact tests for categorical variables. There were no missing data for LVMI, BMI, or BMI z score. As several other variables had missing values, we conducted multiple imputation utilizing all available information on related covariates. This imputed data set was used to conduct the analyses detailed below. Analyses using the data set with missing data were compared with those conducted using the imputed data set to assess the appropriateness of imputation.
To assess for longitudinal associations between participant characteristics and LVMI over time, delta variables were calculated (12‐month measurement – baseline measurement). Linear regression quantified the association between change in LVMI and both baseline characteristics and change over time, adjusting for age, sex, and race.
To investigate the association between adiposity and LVMI over time, we conducted linear regression analysis with baseline BMI z score as the independent variable and delta LVMI as the dependent variable, adjusting for age, race, sex, and baseline LVMI. To then investigate the relative contribution of each hypothesized mediating pathway between adiposity and LVMI, we conducted several additional linear regression analyses. These separate analyses started with the base model described above and further adjusted for each hypothesized mediator. In these analyses, we expected an attenuation of the association between BMI z score and LVMI for any variable that mediated this relationship. For risk factors that did not mediate this association, we expected unchanged or similar point estimates and P values despite the addition of the variable to the model.
A two‐sided P value <.05 was considered statistically significant. Statistical analyses were conducted using STATA 11.2 (StataCorp LP, College Station, TX). Data were managed using REDCap (Research Electronic Data Capture) hosted at Johns Hopkins University.15
Results
Characteristics
Forty‐nine of 53 enrolled children completed both baseline and 12‐month assessments (92% retention; two disenrolled, two lost to follow‐up). Forty‐one percent had LVH at baseline and 53.1% had LVH at follow‐up (P=.31). The mean age of participants at baseline was 13.8 years (standard deviation [SD], 3.9; range, 3–19 years). The average time between baseline and follow‐up assessments was 12.8 months and the mean (SD) change in LVMI was 2.89 g/m2.7 (8.4 g/m2.7). Of the measures of adiposity, the mean change in BMI was 1.6 kg/m2 (2.1 kg/m2), mean change in BMI z score was 0.11 (0.32), and the mean change in waist circumference was 4.6 cm (5.9 cm). The percentage of overweight/obese children was 51% at baseline and 57% at follow‐up (P=.69). At follow‐up, 19 children had a lower BMI z score than at baseline. Those who had improved BMI z score over time had a greater BMI z score at baseline (compared with those who did not): 1.33 (SD, 0.91) vs 1.05 (SD, 1.02), but this was not significant (P=.33). Further, only 11 of the 19 who had improved BMI z score were overweight or obese to start, and at follow‐up almost all of them (n=10) remained overweight or obese. BP changed to a minimal degree over the course of the study: the mean change in clinic SBP was −0.14 mm Hg (12.5), in clinic SBP index was −0.02 (0.09), in awake SBP was 0.52 mm Hg (9.5), and in awake SBP index was −0.001 (0.07). Forty‐four children were prescribed antihypertensive medications at baseline, with 43 prescribed antihypertensive medications at follow‐up.
Baseline Characteristics
Children with LVH had greater LVMI compared with those without LVH (Table 1). There were no age, sex, or race differences between the two groups, but there was a greater proportion of children with a positive obstructive sleep apnea screen (defined as >30% positive responses on validated questionnaire16, 17) among those with LVH. Children with LVH also had a greater BMI z score, a higher serum uric acid level, a lower serum lipoprotein (a) level, and a greater pro‐B natriuretic peptide level than those without LVH. hs‐CRP and urinary sodium excretion were nonsignificantly higher in those with LVH (1.1 mg/dL vs 0.40 mg/dL for hs‐CRP; 4.1 g vs 3.4 g/d for urinary sodium excretion).
Table 1.
Baseline Characteristics of 49 Hypertensive Youth With and Without LVH
| Characteristic | Overall | LVH (n=20) | No LVH (n=29) | P Value |
|---|---|---|---|---|
| LVMI, g/m2.7 | 37.7 (9.5) | 45.4 (7.4) | 32.3 (6.6) | <.001 |
| Demographics | ||||
| Age, y | 13.8 (3.9) | 14.3 (4.1) | 13.4 (3.8) | .48 |
| Male | 29 (59) | 14 (70) | 15 (52) | .25 |
| African American | 19 (39) | 2 (25) | 14 (48) | .14 |
| Medical history | ||||
| Positive OSA screen | 10 (21) | 7 (37) | 3 (10) | .04 |
| Duration of hypertension, mo | 39.4 (32.3) | 49.2 (40.6) | 32.6 (23.5) | .11 |
| Measures of adiposity | ||||
| BMI, kg/m2 | 25.3 (7.6) | 27.8 (9.0) | 23.5 (6.0) | .07 |
| BMI z score | 1.2 (0.98) | 1.5 (0.7) | 0.9 (1.1) | .03 |
| Overweight/obese | 25 (51) | 13 (65) | 12 (41) | .15 |
| Waist circumference, cm | 81.5 (18.6) | 85.5 (20.1) | 78.7 (17.3) | .22 |
| Inflammatory markers | ||||
| hs‐CRP, mg/L | 0.5 (0.2–1.4) | 1.1 (0.4–2.1), 18 | 0.4 (0.2–1.2), 28 | .13 |
| 25‐OH vitamin D, ng/mL | 25 (10.8) | 26.2 (10.1), 19 | 24.2 (11.3), 29 | .53 |
| Markers of metabolic dysregulation | ||||
| Glycated hemoglobin | 5.2 (0.39) | 5.1 (0.42), 17 | 5.3 (0.37), 28 | .31 |
| Non‐HDL cholesterol, mg/dL | 105.9 (32.8) | 102.8 (23.9), 18 | 107.9 (37.6), 29 | .57 |
| Lipoprotein (a), nmol/L | 55 (17–100) | 43 (5–75), 19 | 65 (28.3–137.5), 28 | .05 |
| Metabolic syndromea | 11 (23) | 5 (25) | 6 (21) | .74 |
| Serum uric acid, mg/dL | 5.6 (1.7) | 6.4 (1.9), 16 | 5.0 (1.4), 25 | .02 |
| Dietary risk factors and markers of intravascular volume | ||||
| Dietary sodium intake, mg/24 h | 3282 (1100) | 3161 (953), 14 | 3350 (1188), 25 | .59 |
| Urinary sodium excretion, mg/24 h | 3658 (1973) | 4118 (2885), 10 | 3351 (1037), 15 | .40 |
| Pro‐B natriuretic peptide, pg/mL | 16 (10.54) | 29 (10.84), 19 | 10 (10.38), 28 | .04 |
Abbreviations: hs‐CRP, high‐sensitivity C‐reactive protein; LVH, left ventricular hypertrophy; LVMI, left ventricular mass index; OSA, obstructive sleep apnea. Values are expressed as mean (standard deviation), median (interquartile range), or number (percentage). aDefined as having three or more of the following: systolic or diastolic blood pressure ≥90th percentile, body mass index (BMI) >97th percentile, triglycerides >110 mg/dL, high‐density lipoprotein (HDL) <40 mg/dL, and waist circumference >90th percentile for boys and ≥90th percentile for girls.34
There was no difference in BP between those with and those without LVH when measured at rest or by 24‐hour ABPM (Table 2). The median plasma renin activity was also no different between the groups. Serum aldosterone, however, was higher among children with LVH who were prescribed an angiotensin‐converting enzyme (ACE) inhibitor or angiotensin II receptor blocker (ARB) when compared with those without LVH who were also prescribed one of those medications.
Table 2.
Baseline Clinic BP, 24‐Hour Ambulatory BP, and Related Measures in Children With and Without Left Ventricular Hypertrophy
| Characteristic | Overall | LVH (n=20), No. | No LVH (n=29), No. | P Value |
|---|---|---|---|---|
| Systolic BP | ||||
| Clinic systolic BP, mm Hg | 122 (11.7) | 121 (12.3), 11 | 123 (11.7), 24 | .69 |
| Clinic 95th Systolic BP indexa | 0.97 (0.08) | 0.93 (0.08), 11 | 0.98 (0.08), 24 | .1 |
| Awake systolic BP, mm Hg | 128 (10.9) | 126 (9.4), 18 | 129 (11.8), 27 | .34 |
| Awake 95th systolic BP index | 0.97 (0.08) | 0.95 (0.06), 18 | 0.99 (0.09), 27 | .11 |
| 24‐h Systolic BP, mm Hg | 123 (10.5) | 122 (9.9), 18 | 124 (11), 27 | .63 |
| 24‐h 95th Systolic BP index | 0.98 (0.07) | 0.97 (0.07), 18 | 0.99 (0.08), 27 | .32 |
| Systolic dipb | 9.7 (5.5) | 9.3 (4.1), 17 | 10.0 (6.3) | .68 |
| Diastolic BP | ||||
| Clinic diastolic BP, mm Hg | 73 (9.6) | 75 (10.8), 11 | 73 (9.2), 24 | .59 |
| Clinic 95th diastolic BP indexb | 0.89 (0.11) | 0.89 (0.14), 11 | 0.89 (0.10), 24 | .98 |
| Awake diastolic BP, mm Hg | 72 (7.9) | 70 (7.1), 18 | 73 (8.3), 27 | .15 |
| Awake 95th diastolic BP index | 0.88 (0.1) | 0.85 (0.09), 18 | 0.89 (0.1), 27 | .11 |
| 24‐h Diastolic BP, mm Hg | 67 (7.3) | 66 (7.4), 18 | 68 (7.2), 27 | .38 |
| 24‐h 95th Diastolic BP index | 0.88 (0.1) | 0.86 (0.1), 18 | 0.89 (0.09), 27 | .29 |
| Diastolic dipb | 15.4 (11.4) | 13.7 (7.3), 17 | 16.5 (13.2), 27 | .37 |
| Pulse pressure | ||||
| Clinic pulse pressure | 49.1 (9.4) | 46.5 (12.4), 11 | 50.3 (7.7), 24 | .36 |
| Awake pulse pressure | 56.2 (7.8) | 56.4 (8.6), 18 | 56.1 (7.5), 27 | .89 |
| BP control | ||||
| Systolic and diastolic wake BP <95th percentile | 27 (60) | 13 (72), 18 | 14 (52), 27 | .22 |
| Hormonal regulators of BP | ||||
| Plasma renin activity, ng/mL/h | 4.13 (2.37–9.77) | 7.28 (2.56–13.11), 20 | 3.68 (2.37–5.94), 28 | .14 |
| Taking ACE inhibitor/ARB | 10.73 (5.36–18.95) | 10.73 (8.77–18.95), 13 | 10.75 (4.71–20.04), 8 | .56 |
| Not taking ACE inhibitor/ARB | 2.89 (1.35–3.78) | 2.1 (0.94–2.96), 7 | 2.91 (2.2–3.9), 20 | .20 |
| Serum aldosterone, ng/dL | 2.5 (1–6) | 3 (2–6), 20 | 2.5 (1–7), 28 | .47 |
| Taking ACE inhibitor/ARB | 2 (0.5–6) | 2 (2–9), 13 | 0.5 (0.5–1.5), 8 | .03 |
| Not taking ACE inhibitor/ARB | 4 (2–6) | 4 (1–5), 7 | 4 (2–8.5), 20 | .60 |
Abbreviations: ACE, angiotensin‐converting enzyme; ARB, angiotensin receptor blocker. Values are expressed as mean (standard deviation), median (interquartile range), or number (percentage). aBlood pressure (BP) index was defined as the measured BP/95th percentile BP for age/sex/height, with a value ≥1 denoting BP ≥95th percentile. bBP dip=1−(BP during sleep/BP while awake)×100%.
Longitudinal Associations
In multivariable linear regression analyses, baseline waist circumference and change in glycated hemoglobin were associated with change in LVMI (Table 3). Compared with those who did not have improved BMI z score, those children who did have improved BMI z score had an overall lower increase in LVMI over the course of follow‐up (2.56 [SD, 7.7] vs 4.88 [SD, 10.1]), but this was not significant (P=.37). Children who were overweight/obese at both study visits experienced the greatest increase in LVMI over time: mean change in LVMI was 6.4 g/m2.7 (95% confidence interval [CI], 2.4–10.5] among those overweight or obese at each visit, vs 0.95 g/m2.7 (95% CI, −3.2 to 5.1) among children who were of healthy weight at each visit (P=.056; Figure 1). In addition, when stratified by weight status (healthy weight vs overweight/obese) and LVH status at baseline, overweight/obese children with and without LVH demonstrated a larger increase in LVMI compared with healthy‐weight children. In fact, healthy‐weight children with LVH were the only ones to demonstrate a decrease in LVMI over time (Figure 2).
Table 3.
Association of Change in LVMI With Baseline Characteristics and Change in Baseline Characteristics, Adjusted for Age, Sex, and Race
| Characteristica | Baseline Variables | Delta Variablesb | ||||
|---|---|---|---|---|---|---|
| ΔLVMI, g/m2.7 | 95% CI | P Value | ΔLVMI, g/m2.7 | 95% CI | P Value | |
| Demographics | ||||||
| Age, y | 1.0 | 0.4–1.6 | .003 | |||
| African American vs non–African American | −0.1 | −5.2 to 5.0 | .97 | |||
| Male vs female | 0.1 | −5.0 to 5.1 | .97 | |||
| Measures of adiposity | ||||||
| BMI z score | 2.6 | −0.04 to 5.2 | .05 | 3.8 | −4.2 to 11.7 | .35 |
| Overweight/obesec | 4.4 | 0.5–9.4 | .08 | |||
| Waist circumference, cm | 0.2 | 0.06–0.4 | .008 | 0.2 | −0.3 to 0.6 | .40 |
| Blood pressure | ||||||
| Awake systolic BP indexd | −1.0 | −32.0 to 30.0 | .95 | 11.8 | −21.0 to 44.2 | .46 |
| Awake diastolic BP indexd | −17.4 | −40.0 to 5.6 | .13 | −1.21 | −26.5 to 24.09 | .92 |
| Awake pulse pressure | 0.4 | −0.002 to 0.8 | .05 | 0.1 | −0.1 to 0.4 | .32 |
| Hormonal regulators of BP | ||||||
| Serum aldosterone, ng/dL | −0.1 | −0.3 to 0.2 | .65 | 0.03 | −0.2 to 0.2 | .78 |
| Plasma renin activity, ng/mL/h | 0.1 | −0.04 to 0.3 | .12 | −0.02 | −0.2 to 0.1 | .81 |
| Inflammatory markers | ||||||
| hs‐CRP, mg/L | −0.7 | −2.3 to 0.9 | .40 | 0.6 | −0.4 to 1.5 | .22 |
| 25‐OH Vitamin D, ng/mL | 0.1 | −0.1 to 0.4 | .30 | −0.3 | −0.5 to 0.003 | .05 |
| Markers of metabolic dysregulation | ||||||
| Glycated hemoglobin, % | −5.8 | −12.6 to 1.0 | .09 | 5.5 | 0.3–10.8 | .04 |
| Non‐HDL cholesterol, mg/dL | −0.03 | −0.1 to 0.1 | .4 | 0.03 | −0.1 to 0.2 | .59 |
| Lipoprotein (a), nmol/L | −0.008 | −0.04 to 0.02 | .55 | 0.01 | −0.1 to 0.1 | .75 |
| Serum uric acid, mg/dL | 0.3 | −1.8 to 2.4 | .77 | −0.2 | −4.3 to 3.9 | .92 |
| Dietary risk factors and markers of intravascular volume | ||||||
| Urinary sodium excretion, mg/24 h | 0.0003 | −0.001 to 0.001 | .65 | −0.0003 | −0.001 to 0.001 | .52 |
| Pro‐B natriuretic peptide, pg/mL | 0.02 | −0.1 to 0.1 | .56 | −0.1 | −0.1 to 0.03 | .24 |
Abbreviations: CI, confidence interval; HDL, high‐density lipoprotein; hs‐CRP, high‐sensitivity C‐reactive protein; LVMI, left ventricular mass index. aEach characteristic individually assessed in a separate model that adjusted for age, sex, and race. bDelta variables (Δ) are the change values for each characteristic listed. They represent the difference in measurement between the two study visits (value at 12‐month follow‐up – value at baseline). cDefined as body mass index (BMI) ≥85th percentile. dBlood pressure (BP) index is defined as measured BP/95th percentile BP for age/sex/height, with a value ≥1 denoting a BP ≥95th percentile.
Figure 1.
Mean change in left ventricular mass index by weight status.
Figure 2.
Mean change in left ventricular mass index by weight and left ventricular hypertrophy status at baseline.
Mediator Analyses
Baseline‐adjusted analyses revealed a significant relationship between BMI z score at baseline and change in LVMI over time (Table 4). Sequential adjustment for BP, markers of metabolic dysregulation, and markers of inflammation revealed essentially no change in the point estimate or P values between BMI z score and change in LVMI. The models adjusting for serum aldosterone with ACE inhibitor/ARB use and pulse pressure resulted in a decrease in both the point estimate and P value, suggesting partial mediation.
Table 4.
Association of Baseline BMI z Score With Change in LVMI Investigating for Mediating Pathway
| Multivariable Models | Adjusting for the Following Variables | Δ LVMI by BMI z Score at Baseline | 95% CI | P Value |
|---|---|---|---|---|
| Model 1 | Age, race, sex | 2.6 | −0.04 to 5.2 | .05 |
| Model 2 (base model) | Age, race, sex, LVMI at baseline | 4.1 | 1.5–6.6 | .002 |
| Blood pressure | ||||
| Model 3a | Model 2 covariates + awake SBP indexa at baseline and follow‐up | 3.9 | 1.3–6.5 | .005 |
| Metabolic dysregulation | ||||
| Model 3b | Model 2 covariates + non‐HDL cholesterol at baseline and follow‐up | 3.7 | 1.1–6.4 | .007 |
| Model 3c | Model 2 covariates + glycated hemoglobin at baseline and follow‐up | 3.9 | 1.4–6.5 | .003 |
| Inflammation | ||||
| Model 3d | Model 2 covariates + hs‐CRP at baseline and follow‐up | 4.2 | 1.4–7.0 | .004 |
| Model 3e | Model 2 covariates + 25‐hydroxy vitamin D at baseline and follow‐up | 3.8 | 1.2–6.4 | .005 |
| Hormonal regulators of blood pressure | ||||
| Model 3f | Model 2 covariates + serum aldosterone at baseline and follow‐up | 4.2 | 1.6–6.8 | .003 |
| Model 3g | Model 3f covariates + ACE inhibitor/ARB use at baseline and follow‐up | 3.7 | 1.0–6.5 | .01 |
| Increased intravascular volume | ||||
| Model 3h | Model 2 covariates + pulse pressure at baseline and follow‐up | 3.2 | 0.3–6.1 | .03 |
Abbreviations: Δ, change; ACE, angiotensin‐converting enzyme; ARB, angiotensin receptor blocker; BMI, body mass index; CI, confidence interval; HDL, high‐density lipoprotein; hs‐CRP, high‐sensitivity C‐reactive protein; LVMI, left ventricular mass index. aSystolic blood pressure (SBP) index was defined as awake SBP/95th percentile awake SBP.
Discussion
In this year‐long observational study of hypertensive children, we demonstrated a high burden of comorbid CVD risk factors at baseline and an increase in both their prevalence and severity over time. Most striking was the presence and substantial degree of overweight and obesity, with more than 50% of the cohort overweight or obese––much more than the overall prevalence in the US population (32%).18 Several CVD risk factors and cardiometabolic abnormalities, specifically uric acid, measures of adiposity, lipids, glycated hemoglobin, and hs‐CRP, along with target organ damage felt to be secondary to hypertension itself, increased/worsened over time despite relatively good BP control as evidenced by the mean BP index being <1 at baseline and decreasing over time. Over time, adiposity, not BP, was demonstrated as the greatest risk factor for LVH and increasing LVMI among these hypertensive children.
While several other studies in children have described the impact of CVD risk factors on target organ damage, ours is the first to report longitudinal data on a racially diverse population of hypertensive children without kidney disease. Litwin and colleagues19 demonstrated that among European children with incident primary hypertension, successful treatment of hypertension was not independently associated with decrease in LVMI. In fact, there was no difference in the prevalence of LVH between those children who “responded” to antihypertensive therapy and those who did not. Instead, adiposity was most strongly associated with LVMI over time, with decreasing waist circumference the main predictor of decreasing LVMI. 19
Studies in adults and normotensive children also provide evidence for the greater role of adiposity on LVM when compared with BP. Obese adults who lost weight after bariatric surgery had concomitant reductions in BP, LVM, and relative wall thickness. Interestingly, their BP reduction was not associated with decreased LVM or improved LV structure.20 A dietary intervention trial in Finnish children followed yearly, from 7 months of age to adolescence, revealed the strongest determinant of LVM to be an adolescent's concurrent weight.21 While children in the intervention group showed significant improvements in diet, lipids, BP, and endothelial function, they did not experience significant changes in weight, LVM, or LVH.21, 22 So, while improved diet quality was able to decrease BP among these normotensive children, lower BP did not impact LVM or the presence/absence of LVH.
One of the indications for initiation or intensification of antihypertensive medication among hypertensive children is the presence of LVH. Recommendations suggest aiming for a more aggressive BP treatment goal of <90th percentile when LVH is present1 to promote LVH regression. Given these recommendations, the pattern of change in LVMI among healthy‐weight children with LVH as demonstrated in Figure 2 is what would be expected to occur with successful antihypertensive treatment. The increase in LVMI over time experienced by the healthy‐weight children without LVH, however, was not necessarily expected, but if these changes did not result in the development of LVH (ie, did not increase to a degree that placed the child at/above the 95th percentile LVMI) the clinical significance of this change is unclear.
Conversely, the pattern of change in LVMI among children who were overweight/obese is concerning. Specifically, overweight/obese children without LVH demonstrated an increase in LVMI that was 2.5 times higher than that experienced by the healthy‐weight children without LVH. Further, those overweight/obese children with LVH experienced almost as large of an increase as those without LVH, not the expected decrease over time.
Perhaps the most striking finding in our study was the strong independent association between adiposity and change in LVMI over time that remained despite sequential adjustments for multiple mediating pathways. The two variables that partially mediated this association, serum aldosterone (independent of ACE inhibitor or ARB use) and pulse pressure, are potentially modifiable risk factors for increasing LVMI. The adipocyte is an endocrine organ that secretes neurohumoral factors that influence cardiac remodeling directly and contribute to intravascular volume expansion. These neurohumoral factors lead to increased, and possibly inappropriate, aldosterone secretion. In the setting of excess sodium intake, this excess aldosterone secretion contributes to the target organ damage found in hypertensive individuals23, 24 and can cause increased intravascular volume. The role of this neurohumoral pathway was also demonstrated in the bariatric surgery intervention study described earlier, where the authors concluded that the effect of weight loss on cardiac geometry was primarily mediated by the associated decrease in intravascular volume and reversal in hormonal abnormalities, not by a decrease in BP.20
Along the same lines, while increased pulse pressure is thought to occur primarily as a result of an increase in stroke volume among younger individuals, in our study it may be a marker of increased intravascular volume. In fact, several findings we report suggest that the children with LVH may have increased intravascular volume––those with LVH had a greater 24‐hour urinary sodium excretion (a marker for dietary sodium intake the day prior) and a greater pro‐B natriuretic factor (a peptide thought to represent intravascular fluid volume). Another possible explanation for how pulse pressure might mediate the relationship between adiposity and increasing LVMI is that it may instead represent greater aortic stiffness. Increasing pulse pressure is associated with this finding in older individuals, and hypertensive individuals with the metabolic syndrome have higher pulse pressure than those without the metabolic syndrome.25 It is known that children as young as 2 years have evidence of atherosclerosis, and the presence of these fibrous plaques is related to the presence of traditional CVD risk factors such as obesity, BP, and dyslipidemia.26 We also know that elevated pulse pressure can lead to intimal damage, atherosclerosis, increased left ventricular stress, and hypertrophy.27 More studies are clearly needed to elucidate this further.
The strong association between BMI z score and change in LVMI that remains despite adjusting for these factors suggests either an independent association between adiposity and LVMI or, more likely, an interaction of LVMI with many or all of these mediators. Lending further support to this complex association are the multiple studies that describe different patterns of abnormal cardiac geometry among obese hypertensive compared with nonobese hypertensive individuals.28 Specifically, obese hypertensive individuals tend to exhibit eccentric hypertrophy,28 a geometric pattern associated with greater dietary sodium intake and larger intravascular volume.29, 30 Nonobese hypertensive individuals tend to have concentric hypertrophy, which is associated with more severely elevated BP in adults.28, 29, 31
Study Limitations
Our study has several limitations. These include its relatively small sample size and observational design, which makes us unable to infer causality. As children were recruited from a pediatric nephrology clinic, there may be selection bias in that included children may have more severe hypertension than those who might have been recruited from a nonreferral‐based office setting. In addition, the majority of the children were not fasting at the time of their study visit. As a result we were unable to fully ascertain the degree of metabolic dysfunction among the cohort. Finally, BP was a controlled variable, as well as a variable with a high degree of intraindividual variability. Hence, the absence of associations of LVMI with BP should be interpreted cautiously, especially in relation to obesity‐related variables, which are measured extremely precisely.
Study Strengths
Our study also has several strengths. It provides data on a racially diverse population of hypertensive children at considerable CVD risk. Its longitudinal, prospective design allows us to minimize information bias and evaluate the relationship of various mediators of obesity and LVH on the change in LVMI over time. The standardized echocardiograms read by a study cardiologist provide us with a robust estimate of the LV size and prevalence of LVH. Our use of 24‐hour ABPM provides a more precise assessment of BP than one set of clinic BP measurements.32, 33
Conclusions
Our study emphasizes the substantial contribution of obesity on CVD risk among hypertensive children. Further, we demonstrate that once a child is overweight or obese, his/her associated comorbidities appear to worsen over time despite standard‐of‐care guidance. These findings suggest that children who are both hypertensive and overweight/obese should be considered at extremely high CVD risk. Improved lifestyle interventions are urgently needed to achieve cardiovascular health in children.
Acknowledgments
We would like to thank Sara Boynton, BA, and Sanjay Jumani, BA, who conducted the data collection that made this study possible. Neither of these individuals have any conflicts of interest to disclose.
Conflict of Interest
The authors have no conflicts of interest to report.
Funding Source
This study is supported by grants from the American Society of Nephrology, National Kidney Foundation of MD, American Heart Association, Thomas Wilson Sanitarium for Children of Baltimore City, and the National Institutes of Health (NIH)/National Heart, Lung, and Blood Institute (K23HL119622) and was also made possible by grant number UL1 RR 025005 from the National Center for Research Resources (NCRR), a component of the NIH, and NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of the NCRR or NIH. None of the sponsors had any role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript. The first draft of this manuscript was written by Tammy M. Brady and there was no honorarium, grant, or other form of payment given to anyone to produce the manuscript.
J Clin Hypertens (Greenwich). 2016;18:625–633. DOI: 10.1111/jch.12717 © 2015 Wiley Periodicals, Inc.
References
- 1. National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents . The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics. 2004;114(suppl 4th report):555–576. [PubMed] [Google Scholar]
- 2. Brady TM, Fivush B, Flynn JT, Parekh R. Ability of blood pressure to predict left ventricular hypertrophy in children with primary hypertension. J Pediatr. 2008;152:73–78, 78 e71. [DOI] [PubMed] [Google Scholar]
- 3. Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents; National Heart, Lung, and Blood Institute . Expert panel on integrated guidelines for cardiovascular health and risk reduction in children and adolescents: summary report. Pediatrics. 2011;128(suppl 5):S213–S256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Brady TM, Fivush B, Parekh RS, Flynn JT. Racial differences among children with primary hypertension. Pediatrics. 2010;126:931–937. [DOI] [PubMed] [Google Scholar]
- 5. Stabouli S, Kotsis V, Rizos Z, et al. Left ventricular mass in normotensive, prehypertensive and hypertensive children and adolescents. Pediatr Nephrol. 2009;24:1545–1551. [DOI] [PubMed] [Google Scholar]
- 6. Falkner B, Deloach S, Keith SW, Gidding SS. High risk blood pressure and obesity increase the risk for left ventricular hypertrophy in African‐American adolescents. J Pediatr. 2012;162:94–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Furth SL, Cole SR, Moxey‐Mims M, et al. Design and methods of the chronic kidney disease in children (CKiD) prospective cohort study. Clin J Am Soc Nephrol. 2006;1:1006–1015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Urbina E, Alpert B, Flynn J, 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. 2008;52:433–451. [DOI] [PubMed] [Google Scholar]
- 9. Devereux RB, Alonso DR, Lutas EM, et al. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol. 1986;57:450–458. [DOI] [PubMed] [Google Scholar]
- 10. Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr. 2005;18:1440–1463. [DOI] [PubMed] [Google Scholar]
- 11. Khoury PR, Mitsnefes M, Daniels SR, Kimball TR. Age‐specific reference intervals for indexed left ventricular mass in children. J Am Soc Echocardiogr. 2009;22:709–714. [DOI] [PubMed] [Google Scholar]
- 12. CDC . Centers for Disease Control Clinical Growth Charts. August 4, 2009; http://www.cdc.gov/growthcharts/clinical_charts.htm. Accessed March 15, 2010.
- 13. Prevention CfDCa . A SAS Program for the CDC Growth Charts. June 27, 2011; http://www.cdc.gov/nccdphp/dnpao/growthcharts/resources/sas.htm. Accessed March 18, 2013.
- 14. Krebs NF, Himes JH, Jacobson D, et al. Assessment of child and adolescent overweight and obesity. Pediatrics. 2007;120(suppl 4):S193–S228. [DOI] [PubMed] [Google Scholar]
- 15. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)–a metadata‐driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Mindell JA, Owens JA. A Clinical Guide to Pediatric Sleep; Diagnosis and Management of Sleep Problems. Philadelphia, PA: Lippincott Williams & Wilkins; 2003. [Google Scholar]
- 17. Chervin RD, Hedger K, Dillon JE, Pituch KJ. Pediatric sleep questionnaire (PSQ): validity and reliability of scales for sleep‐disordered breathing, snoring, sleepiness, and behavioral problems. Sleep Med. 2000;1:21–32. [DOI] [PubMed] [Google Scholar]
- 18. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA. 2014;311:806–814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Litwin M, Niemirska A, Sladowska‐Kozlowska J, et al. Regression of target organ damage in children and adolescents with primary hypertension. Pediatr Nephrol. 2010;25:2489–2499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Karason K, Wallentin I, Larsson B, Sjostrom L. Effects of obesity and weight loss on left ventricular mass and relative wall thickness: survey and intervention study. BMJ. 1997;315:912–916. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Hietalampi H, Pahkala K, Jokinen E, et al. Left ventricular mass and geometry in adolescence: early childhood determinants. Hypertension. 2012;60:1266–1272. [DOI] [PubMed] [Google Scholar]
- 22. Niinikoski H, Jula A, Viikari J, et al. Blood pressure is lower in children and adolescents with a low‐saturated‐fat diet since infancy: the special turku coronary risk factor intervention project. Hypertension. 2009;53:918–924. [DOI] [PubMed] [Google Scholar]
- 23. Rossi GP, Seccia TM. Changes in aldosterone and obesity‐related cardiometabolic risk factors with a 1‐year weight loss intervention in normotensive overweight and obese young adults. Hypertens Res. 2013;36:856–858. [DOI] [PubMed] [Google Scholar]
- 24. Rossi GP, Sechi LA, Giacchetti G, et al. Primary aldosteronism: cardiovascular, renal and metabolic implications. Trends Endocrinol Metab. 2008;19:88–90. [DOI] [PubMed] [Google Scholar]
- 25. Mule G, Nardi E, Cottone S, et al. Relationship of metabolic syndrome with pulse pressure in patients with essential hypertension. Am J Hypertens. 2007;20:197–203. [DOI] [PubMed] [Google Scholar]
- 26. Berenson GS, Srinivasan SR, Bao W, et al. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. The Bogalusa Heart Study. N Engl J Med. 1998;338:1650–1656. [DOI] [PubMed] [Google Scholar]
- 27. O'Rourke M, Frohlich ED. Pulse pressure: is this a clinically useful risk factor? Hypertension. 1999;34:372–374. [DOI] [PubMed] [Google Scholar]
- 28. Messerli FH. Cardiovascular adaptations to obesity and arterial hypertension: detrimental or beneficial? Int J Cardiol. 1983;3:94–97. [DOI] [PubMed] [Google Scholar]
- 29. Ganau A, Arru A, Saba PS, et al. Stroke volume and left heart anatomy in relation to plasma volume in essential hypertension. J Hypertens Suppl. 1991;9:S150–S151. [PubMed] [Google Scholar]
- 30. Blake J, Devereux RB, Borer JS, et al. Relation of obesity, high sodium intake, and eccentric left ventricular hypertrophy to left ventricular exercise dysfunction in essential hypertension. Am J Med. 1990;88:477–485. [DOI] [PubMed] [Google Scholar]
- 31. Ganau A, Devereux RB, Roman MJ, et al. Patterns of left ventricular hypertrophy and geometric remodeling in essential hypertension. J Am Coll Cardiol. 1992;19:1550–1558. [DOI] [PubMed] [Google Scholar]
- 32. Gimpel C, Wuhl E, Arbeiter K, et al. Superior consistency of ambulatory blood pressure monitoring in children: implications for clinical trials. J Hypertens. 2009;27:1568–1574. [DOI] [PubMed] [Google Scholar]
- 33. Vollmer WM, Appel LJ, Svetkey LP, et al. Comparing office‐based and ambulatory blood pressure monitoring in clinical trials. J Hum Hypertens. 2005;19:77–82. [DOI] [PubMed] [Google Scholar]
- 34. Pediatrics AAo . Promoting Healthy Weight. Bright Futures Guidelines for Health Supervision of Infants, Children and Adolescents. http://brightfutures.aap.org/pdfs/Guidelines_PDF/5-Promoting_Healthy_Weight.pdf. Accessed March 18, 2013.