Recent Clinical and Translational Advances in Pediatric Hypertension

Introduction

Hypertension increases risks for adverse cardiovascular outcomes in adults. Due to lack of long term outcome data to link a blood pressure (BP) threshold in childhood that predicts elevated risk for future cardiovascular events, hypertension is defined differently in pediatric patients. Despite the difference in definition, which is statistically based (≥95^th percentile according to sex, age, and height percentile), hypertension is present is a substantial number of asymptomatic children and adolescents. Although outcome data on benefit versus risk of treatment in children with established hypertension are limited, available clinical trial data indicate that lowering BP in hypertensive children with anti-hypertensive medications is generally effective and safe. Hypertension that is secondary to an underlying cause is thought to be more common in childhood which raises evaluation issues, especially as primary hypertension is now commonly found in childhood. A recent report by Flynn et al.¹ provides observational clarity on secondary versus primary hypertension in childhood. The authors analyzed contemporary demographic and clinical characteristics at baseline of 351 hypertensive children and adolescents enrolled from multiple sites in a clinical trial. Compared to children in mid-childhood (age 6 to <12 years) and adolescents (age 12 to <17 years), the younger children (<6 years of age) were significantly more likely to have secondary hypertension, were less likely to be obese, and had significantly higher diastolic BP. Thus secondary hypertension is more likely to be detected in non-obese younger children with higher BP, whereas, primary hypertension is more commonly found in late childhood and adolescence and is associated with overweight/obesity and modest BP elevations.

Recent pediatric reports provide other tools that can be applied in clinical evaluation of hypertensive children. Ambulatory BP monitoring (ABPM) has become a very useful diagnostic procedure in evaluation of hypertensive children as well as adults. A Scientific Statement From the American Heart Association² provides guidance on the use of ABPM measurement in children and adolescents. Target organ damage (TOD) is detectable in some hypertensive children. The evidence for TOD in childhood had been limited to small studies that compared left ventricular mass index (LVMI) and carotid intimal media thickness (cIMT) in hypertensive children to in normotensive children. There are now published normative reference data on LVMI³ and more recently on cIMT⁴ in children and adolescents. These tools represent considerable advancement for clinical research and clinical management of children with hypertension and also children with prehypertension.

Increasing BP level and prevalence of high BP in childhood

An examination of epidemiologic studies on BP in children and adolescents conducted over the past decade demonstrates a significant increase in BP level and an increase in prevalence of hypertension.⁵ An analysis of data from the National Health and Nutrition Examination Survey (NHANES), by Munter et al.⁶ in 2004 reported that the increase in childhood BP level was largely, but not entirely, attributable to the increase in the prevalence of obesity in children and adolescence. Large school based BP screening studies.⁷ and analysis of BP data in electronic medical records from large primary pediatric care clinics,⁸ report a prevalence of childhood hypertension (based on repeated BP measurement) at 3.5% and a prevalence of prehypertension also at 3.5%. However, a much lower prevalence was reported in an analysis of electronic medical data of BP in a large cohort of children receiving primary care within the Kaiser Permanente health system.⁹ Consistent among all reports is the observation that rates of high BP were higher among overweight and obese children. Childhood BP trends were recently re-examined by Rosner et al.¹⁰ The authors analyzed a population-based sample of 3248 children in NHANES III (1988–1994) and 8388 children in continuous NHANES (1999–2008), aged 8 to 17 years of age and determined the prevalence of high BP (prehypertension and hypertension combined). Within this 10 year period the prevalence of high BP (based on single measurement) increased from 15.8% to 19.2% among boys and the prevalence of high BP increased from 8.2% to 12.6% among girls. Although the analyses were limited to one BP measurement session, these data document a striking increase in the prevalence of high risk BP among otherwise healthy children. The increasing prevalence of childhood obesity was again determined to be the major, but not the only, determinant of the population increase in high BP prevalence among children. Other countries are also experiencing marked increases in childhood obesity. Investigators in China reported concordant upward trends in body mass index (BMI) and BP level in Chinese children from age 7 to 17 years. Data obtained from two separate cross-sectional surveys in 2005 and 2010 on a total of 391,982 children were compared. The data demonstrate a population increase in BP among both boys and girls. In the same five year interval there was also a significant increase in BMI. Following adjustment for BMI the mean increase in systolic BP was reduced by 40.5% indicating that increasing obesity was a leading determinant of the child population increase in BP.¹¹ However, the BMI adjustment indicated that some other factor could also be contributing to the childhood increase in BP level.

The childhood obesity epidemic and concurrent increasing rates of high BP, strongly suggest an increase in risk for premature cardiovascular disease in adulthood. A pooled analysis of four prospective cohort studies that include data on over 6,000 individuals from childhood into young adulthood found the relative risk for hypertension as adults among those who were obese in childhood was increased (relative risk 2.7; 95%CI, 2.2 to 3.3) with similar increases in risk for other metabolic cardiovascular risk factors. However, individuals who were overweight or obese during childhood but became non-obese as adults had outcome risks that were similar to individuals who were in a normal weight range in childhood and as adults.¹² Other reports describe similar outcomes for cIMT in young adulthood relative to high BP and visceral fatness in childhood.^13,14 In another cohort, Suglia et al.¹⁵ analyzed prospective data from the National Longitudinal Study of Adolescent Health. This cohort included data on individuals from adolescence to adulthood, including Black, Hispanic, and Caucasian participants. The results demonstrate a higher risk for hypertension among those who were chronically overweight from adolescence to adulthood as well as those who became obese as adults. The results also demonstrate that the risk for hypertension in adulthood is lower among those who were overweight in adolescence and subsequently lost weight. An exception was observed among Black men, who have a higher risk for hypertension regardless of weight gain or weight loss. These reports from longitudinal cohort data support the potential long term health benefits of interventions to prevent and reverse childhood obesity. However, the observation that the risk for hypertension is not lower among Black men who lose weight as adults, indicates additional health disparity among Black men.

Dietary Salt and Blood Pressure in Childhood

Epidemiologic and clinical evidence consistently support a strong association of dietary salt intake with BP level and hypertension among adults. Data in support of this relationship have been less clear in childhood. A meta-analysis on 13 randomized clinical trials in healthy adolescents on effects of reduction in dietary salt intake detected a small but statistically significant effect in lowering BP.¹⁶ The same investigators recently conducted a cross-sectional study to quantify dietary salt intake among children in the South London. The study stratified healthy children into three age groups; 5–6 years, 8–9 years, and adolescents age 13–17 years. Salt intake was computed from a combination of 24-hour urinary sodium excretion and a 24-hour photographic food diary. Valid 24-hour urine collections were obtained in 340 children. Mean salt intake, which increased with age, was 3.75 g/day in 5–6 year olds, 4.72 g/day in 8–9 year olds, and 7.44 g/day in adolescents. Overall the salt intake was greater than the maximum daily intake recommendations in 66% of 5–6 year olds, 73% of 8–9 year olds, and 73% of adolescents. Analysis of diet quality determined that the predominant source of dietary salt was from processed foods.¹⁷ Very similar results were reported on sodium intake in US children, based on analysis of recent NHANES data. Average daily sodium intake for children 6–10 years was 2,903 mg, and for children 11–13 years was 3194 mg, and for adolescents (14–18 years) was 3672 mg. The predominant source of dietary sodium was also processed and fast foods.¹⁸

The epidemiologic studies cited above demonstrate that childhood overweight and obesity explain much, but not all, of the increasing BP levels in children.^6,11 There appears to be something else that is contributing to the BP increase in children, and there is now evidence that something else may be salt intake. Two recent reports add further insights on the effect of high salt intake BP in children. Yang et al.¹⁹ examined recent NHANES 2003–2008 data to determine if there was an association between dietary sodium intake and BP level in children age 8 to 18 years of age. The study sample included 6,235 children of whom 37% had BMI ≥85th percentile indicating overweight or obese. Based on analysis of multiple 24-hour dietary recall measures the average sodium intake was estimated at 3.89 g/day. Overall, each 1,000 mg/day of sodium intake was associated with about 1.0 mm Hg increase in systolic BP. However, among overweight/obese children, systolic BP increased by 1.5 mm Hg for each 1,000 mg/day sodium intake. The investigators then stratified the children by quartiles of sodium intake. For the entire sample, the adjusted odds ratios (OR) comparing the risk for prehypertension and hypertension combined in the highest sodium intake quartile to the lowest sodium intake quartile was 2.0 (95% CI = 0.95–4.1; P=0.062). Among overweight/obese children, the adjusted OR for prehypertension/hypertension in the high sodium intake quartile increased to 3.5 (95% CI = 1.2–9.2; P=0.013). These findings were further advanced in the study by Rosner et al.¹⁰ As described above, NHANES data from 1988 to 2008 were analyzed for child BP trends and risk factors. A significant increase in BP among children over this time period was demonstrated. An analysis on associated risk factors determined that BMI, waist circumference, and dietary sodium intake were each independently associated with the prevalence of prehypertension/hypertension among children 8 to 17 years of age. Together these data indicate that the effect of a high sodium intake on BP is amplified among children who are also overweight or obese.

An association of serum uric acid levels with hypertension, which has been commonly observed in adults, is now being reported in childhood. In a sample of children referred for evaluation of hypertension, Feig and Johnson²⁰ reported significantly higher mean serum uric acid levels in children with primary hypertension compared to children with secondary hypertension or white coat hypertension. Subsequent clinical studies in hypertensive children have replicated these observations.^21,22 The effect of uric acid is not limited to hypertension as data from the Bogalusa Heart Study demonstrate a significant correlation childhood uric acid levels with both childhood BP and later BP in adulthood.²³ Further support of an independent effect of uric acid on BP in the young were reported from a randomized placebo-controlled study to determine if lowering uric acid lowers BP in adolescents with stage 1 hypertension. In this study, treatment with allopurinol 200 mg twice daily for four weeks resulted in a significant reduction in BP.²⁴ The mechanism through which uric acid effects an increase in BP, especially in the young, is not clear. As reviewed by Feig,²⁵ a plausible mechanistic pathway is metabolic consequences of high fructose consumption. This concept is supported by both experimental and clinical studies that demonstrate a relationship of high fructose consumption with elevated uric acid levels and also other components of metabolic syndrome. Therefore, a high intake of processed foods sweetened with fructose, especially among obese children, may also contributes to increasing BP levels in the young.

These reports contribute clarity on the early phase of primary hypertension beginning in childhood. BP levels in the young are increasing largely due to childhood obesity with concurrent secular diet changes including salt and possibly fructose exposure. Considering the rapid changes in BMI and diet patterns in childhood, it would seem unlikely that genetics would play a significant role in the increasing BP levels observed among children. However, a recent report from a study by Bo XI et al.²⁶ adds additional insights on obesity associated hypertension in childhood. The investigators genotyped 610 hypertensive children and 2,458 normotensive children from the Beijing Child and Adolescent Metabolic Syndrome Study in a case-control study on genetic variants thought to be associated with hypertension. Based on previous genome wide association studies (GWAS) for hypertension in adults, six single nucleotide polymorphisms (SNPs) were selected for study. There were no significant associations of SNPs with BP in normal weight children regardless of BP status. Among obese children, three SNPs were significantly associated with higher systolic BP. There were also significant associations of four SNPs with hypertension in obese children. Thus the investigators identified a significant association of hypertension susceptibility loci with BP level and with hypertension in Chinese children, but this was found only among obese children. It is also striking that three of the four SNPs associated with hypertension in obese children were reported to be linked with renal sodium regulation. This genetic study connects SNPs on renal sodium regulation with two environmental exposures; obesity and salt intake.

Birth Weight and Intrauterine Effects on Childhood BP

The fetal programming theory, wherein certain adverse intrauterine exposures during fetal development can set the stage for chronic diseases in later life, is a topic of research interest. Epidemiologic studies link low birth weight with adverse outcomes in later adulthood; and experimental research have developed plausible mechanistic pathways. However, reports from clinical studies in children have been inconsistent. Recent reports describe potential maternal stressors that may have an effect on subsequent BP and other metabolic risk factors in childhood. Fraser et al.²⁷ conducted a study to determine if maternal hypertensive disorders of pregnancy, including preeclampsia and gestational hypertension are associated with BP and metabolic risk factors in adolescent offspring. Their study, on mother-offspring pairs from the Avon Longitudinal study of Parents and Children, measured BP and metabolic parameters in offspring at age 17 years. Compared to offspring of normotensive pregnancies, there were no difference in insulin, glucose, or lipid values in offspring of both preeclampsia and gestational hypertension pregnancies. However, BP was significantly higher in offspring of both hypertensive disorders of pregnancy. The BP in offspring of hypertensive pregnancies remained significantly higher following adjustment for potential confounders suggesting the association may be driven by genetic or familial risk factors. Another secondary analysis of data from the same Avon Longitudinal Study cohort examined the relative contribution of different growth periods to BP level at age 10 years. An inverse association of birth weight with systolic BP was found. In subsequent child growth periods all growth parameters, including weight, height, and weight-for-height (an adiposity measure) were all positively associated with systolic BP. The authors concluded that development of excess adiposity during early child growth periods was a modifiable determinant of later BP.²⁸ In another study, the effect of maternal and paternal obesity on child cardiovascular risk factors were examined in 6 year old children in the Generation R cohort. The investigators reported that higher maternal and paternal pre-pregnancy BMI were associated with higher childhood BMI, abdominal fat mass, systolic BP, and insulin levels with lower high density lipoprotein cholesterol levels. The associations were stronger for maternal BMI than Paternal BMI. Although birth weight did not appear to be different between mothers with and without pre-pregnancy obesity, the authors suggest these results may indicate that maternal pre-pregnancy BMI may influence later cardiometabolic health status of offspring through direct intrauterine mechanisms.²⁹ These reports, based on secondary analyses of data from relatively large prospective cohorts, describe relationships between maternal status, birth parameters and child cardiovascular risk status. Wolfenstetter et al.³⁰ used an alternative approach in a study designed to determine if children born with low birth weight have altered cardiovascular rhythmicity (an indirect estimate of sympathetic nervous system activity). The investigators examined healthy children (mean age 8 year) born with low birth weight and control children matched for age and sex with 24-hour ABPM. The 24-hour, daytime, and night BP levels were higher in low birth weight children compared to controls. BP rhythmicity, computed by fourier analysis was different between the groups with blunted circadian and ultradian BP rhythmicity detected in the children with low birth weight. These unique observations suggest possible intrauterine programing with subtle alterations in cardiovascular regulation in children having low birth weight.

Findings in the above reports were based on analysis of data in existing cohorts or studies in children selected for a history of low or normal birth weight. Prospective studies on offspring of normal pregnancies beginning in the newborn period are limited. Lurbe et al.³¹ conducted a small but rigorous prospective study on a sample of healthy full-term newborn infants stratified by birth weight as small (SGA), appropriate (AGA), or large (LGA) for gestational age. BP measured at 2 days of age was positively associated with birth weight. Subsequent BP and growth parameters were measured at 6 months, 2 years, and 5 years of age; and at the 5 year exam a blood sample was obtained for metabolic parameters including glucose, insulin, and lipids. Each birth weight group gained a similar amount of weight between each exam interval. SGA infants remained the smallest and LGA infants remained the largest at subsequent exams. Following 6 months current weight and weight gain were positively associated with birth weight, and birth weight was not associated with BP level. The metabolic measures delineate interesting findings at age 5 years when the birth weight groups were further stratified according to current weight status. Fasting insulin levels were higher in all infants who became heavy at age 5 years, and highest among the SGA group. However, the homostatic model assessment index (HOMA), an estimate of insulin resistance, is higher in the entire SGA group regardless of relative weight status at age 5 years. In addition to relative insulin resistance, the SGA group also had lower high-density lipoprotein-cholesterol and higher uric acid levels compared to the other birth weight groups. These findings suggest that intrauterine factors related to lower birth weight may have induced metabolic programming for relative insulin resistance that is sustained, at least in early childhood, regardless of later weight status. The metabolic measures also indicate subtle emerging features of metabolic syndrome. It is possible that high BP may emerge later as a consequence of the insulin resistance or metabolic syndrome.

Several potential mechanisms have been examined, in experimental studies, to explain perinatal programming, or the effect of the intrauterine environment on later health or chronic disease. In response to suboptimal intrauterine nutrition or other stresses there could be changes in size of various organs, neuroendocrine changes, or other changes that involve epigenetic modifications. Epigenetic alterations indicate modifications in DNA without changes in DNA sequence through DNA methylation, post-translational histone modifications, modification of nuclear receptors, and microRNAs.^32,33 Clinical studies that examine epigenetic modification and clinical outcomes in childhood are limited. An example of a recent study that applied epigenetic strategies to cardiovascular pathophysiology in the young was reported by Breton et al.³⁴ These investigators measured DNA methylation of nitric oxide synthase (NOS) and identified an association of percent DNA methylation of NOS1 with carotid inima-media thickness in children. Additional studies, beginning in childhood, that apply similar molecular strategies are needed to delineate the pathway from intrauterine experiences and epigenetic modification to evolution of chronic disease markers.

BP and Target Organ Damage in Childhood

A recent publication that resulted in strong responses from pediatric hypertension specialists was the US Preventive Services Task Force (USPSTF) recommendation statement on Screening for Hypertension in Children and Adolescents.³⁵ Based on an evidence review, the USPSTF stated that “current evidence is insufficient to assess the balance of benefits and harms of screening for primary hypertension in asymptomatic children and adolescents to prevent subsequent cardiovascular disease in childhood and adulthood.” As summarized in a response by Urbina et al.³⁶ the USPSTF created 8 key questions and performed an evidence based review to answer the questions. From a literature review of 6435 potentially relevant articles only 35 articles met USPSTF criteria of acceptable evidence, all of which were small relatively short randomized controlled clinical trials. All reports based on observational studies were excluded, including prospective cohort studies that demonstrate a significant relationship of high BP in childhood with hypertension in young adulthood and also evidence of target organ damage (TOD) in young adulthood. Cross-sectional studies that describe TOD in hypertensive children were also not considered. The issues not considered by the USPSTF, especially hypertension related TOD in childhood, represent substantial advancements in pediatric hypertension.

Several reports over the past decade demonstrate that TOD is detectable in hypertensive children and is associated with other cardiovascular and metabolic risk factors. Left ventricular hypertrophy (LVH) on echocardiographic measurement of left ventricular mass (LVM) has been described in adolescents with mild untreated high BP.^37,38 Based on evidence of an association of LVM with BP level in children, the NHLBI-sponsored Fourth Report on high BP in children and adolescents³⁹ recommended including an evaluation for TOD in children with confirmed hypertension. Since publication of that 2004 report, additional data have emerged from cross-sectional studies, confirming that TOD is detectable in the young,^40,41 and is not limited to increases in LVM. Compared to normotensive children, a measurable increase in carotid intima-media thickness (cIMT), a surrogate marker for preclinical atherosclerosis in adults, has been reported in children with high BP, as well as in children with diabetes and familial hypercholesterolemia.^42,43 Increased arterial stiffness, or loss of elasticity, a change generally associated with aging, has been detected in pediatric patients with high BP and other conditions linked with CV disease including obesity, diabetes, and dyslipidemia.^44,45 Retinal arteriolar narrowing, an established consequence of hypertension in adults, has recently been described in children with higher BP.⁴⁶ Hypertensive adults have heightened risk for developing cognitive impairment. In recent studies, Lande et al.^47–49 detected a comparable link between high BP and cognitive function in children. In both the NHANES III data, and in subsequent small clinical studies, children with high BP perform at a lower level in measures of executive function compared with age matched normotensive children. Findings of TOD in childhood years may not be limited to patients with established hypertension, as recent reports describe evidence of TOD among prehypertensive adolescents. In a cohort that included type 2 diabetic adolescents, Urbina et al.⁵⁰ reported greater left ventricular mass index (LVMI) among prehypertensive participants compared to normotensives. The investigators also found higher pulse wave velocity, indicative of vascular stiffness, in prehypertensives compared to normotensives.⁵⁰ In another study on a cohort of African American adolescents, the effects of prehypertension and obesity were examined. The investigators reported significantly higher LVMI among prehypertensive adolescents compared to normotensives. LVMI was highest, with more LVH, among those with both prehypertension and obesity. In this study the effects of prehypertension and obesity on LVMI were found to be additive.⁵¹ Overall, recent reports on hypertension related TOD in the young demonstrate evidence of underlying vascular pathology among children with primary hypertension. Recent reports that detected LVH and vascular stiffness among prehypertensive adolescents strongly suggest that the BP level for heightened risk for TOD may be lower than the 95^th percentile which is the current definition of hypertension in childhood.

Summary

Epidemiologic reports describe a child population increase in BP level and an increase in prevalence of hypertension, that is largely, but not entirely, driven by a concurrent increase in childhood obesity. Given current estimates, approximately 10% of adolescents have hypertension or prehypertension. In addition to obesity, dietary salt intake and waist circumference, a marker of visceral obesity, are found to be independently associated with the rise in BP among children and adolescents. Dietary salt intake in urban children is well above recommended levels largely due to consumption of processed and fast foods. Childhood exposures such as stress,⁵² salt, and fructose as well as lifestyles including food sources, sleep patterns, and reductions in physical activity may have a role in obesity-high BP associations. In addition, clinical and translational evidence is mounting that intrauterine exposures alter can effect changes in fetal development that have an enduring effect on cardiovascular and metabolic function later in life. These effects can be detected even in children who are products of a term otherwise normal pregnancy.

Hypertension in childhood has been defined statistically (BP ≥95^th percentile) due to lack of outcome data that links a BP level with heightened risk for future cardiovascular events. Therefore, primary hypertension had been considered a risk factor for later hypertension in adulthood. Intermediate markers of TOD, including cardiac hypertrophy, vascular stiffness, and increases in carotid intimal medial thickness (cIMT) are detectable in adolescents with primary hypertension. Evidence that vascular injury is present in the early phase of hypertension and even in prehypertension warrants consideration on the current definition of pediatric hypertension. With further studies on TOD and other risk factors in addition to high BP it may be possible to shift from a statistical definition to a definition of childhood hypertension that is evidence based.

Preventing or reducing childhood obesity would have substantial benefit in countering the documented increase in BP levels and prevalence of high BP in childhood. Weight control in overweight and obese children, along with dietary changes⁵³ and increases in physical activity⁵⁴ have benefit on BP levels in childhood. Prevention of childhood obesity and BP risk will require multiple levels of intervention, including public health, health policy, and attention to food supply to foster the necessary lifestyle changes to prevent and reduce childhood obesity.