Cardiovascular Disease Risk Factors in Chronic Kidney Disease in Children

Abstract

Over the last two decades, cardiovascular disease (CVD) has been recognized as one of the most important complications of chronic kidney disease (CKD) and one of the leading causes of death in children with advanced CKD and in young adults who developed CKD during childhood. Cardiovascular (CV) abnormalities develop early and progress during the course of CKD in children. Characterization of the prevalence and evolution of CVD risk factors in progressive CKD is one of the primary aims of the Chronic Kidney Disease in Children (CKiD) study. In this review, we summarize up-to-date findings from the CKiD study with the focus on traditional and CKD-related CV risk factors and early, subclinical markers of cardiac and vascular structure and function. We also discuss the effect of CV risk factors on progression of CKD.

Cardiovascular risk factors

One of the most important initial findings of the CKiD study was the high prevalence of under-recognized and under-treated systemic hypertension in children with chronic kidney disease (CKD). The diagnosis of hypertension in CKiD is based on having a systolic and/or diastolic blood pressure (BP) ≥ 95^th percentile for age, sex and height, or a self-reported history of high BP and current treatment with antihypertensive medications. Using the above definition, Flynn et al, in a cross-sectional study showed that upon enrollment, about onehalf of the participants had hypertension with approximately 25% having BP ≥ 90^th percentiles.¹ Only 68% of children with BP ≥ 90^th percentile received antihypertensive medications, and among them, 48 % still had elevated BP. Absence of angiotensin-converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB) use was associated with uncontrolled BP.

Baseline ambulatory blood pressure monitoring (ABPM) among CKiD participants previously demonstrated that only half of the children had a normal ABPM evaluation, with about 32% having an elevated mean BP and 52% an abnormally elevated BP load (e.g. > 25%).² A higher proportion of children had elevated BP during the sleep compared to the wake period. Similar to the casual BP analysis, those using an ACEI/ARB were 89% more likely to have a normal ABPM study than those who did not report using an ACEI. During a 4-year follow up, only 46% of hypertensive patients achieved controlled clinic BP; least likely to demonstrate controlled BP were those children with nephrotic-range proteinuria, black children, and children with a baseline glomerular filtration rate (GFR) < 40ml/min per 1.73m².³

In the study comparing ambulatory BP control over two time periods (2005-2008 versus 2010-2013), there was a significantly higher prevalence of masked hypertension and a lower prevalence of normotension in the latest period.⁴ The most recent published CKiD data indicate that about 20% of children with non-glomerular disease and about 50% of children with glomerular disease had persistent BP ≥ 90^th percentiles at all study visits over a median follow up of four years.⁵ The study also demonstrated that children with non-glomerular disease and BP ≥ 90^th percentiles were significantly younger than children with normal BP. Thus, despite publication of the initial CKiD hypertension data in 2008-2010, plus published recommendations and guidelines for stricter BP control in patients with CKD, hypertension has remained under-recognized and under-treated in children with CKD in the CKiD cohort.

As in prior studies, the CKiD study has confirmed that elevated BP is one of the strongest risk factors for CKD progression: it was independently associated with significantly reduced times to a composite event (initiation of kidney replacement therapy or 50% decline in GFR), being reduced by 38% in children with non-glomerular and by 67% in children with glomerular disease.⁶ A more recent analysis by Ku et al suggested that clinic BP measurements may be as effective as ABPM in predicting risk for CKD progression.⁷ These findings are clinically important in situations when ambulatory monitors are not available, when patients are too young or too small to wear the monitor, when monitoring is incomplete, or when the patient refuses to cooperate with the ABPM. The key point about the clinic BP measurements in this study was that they were obtained using a strict, standardized BP measurement protocol.

Another traditional CV risk, dyslipidemia, was initially found in 43% of the CKiD cohort: 32% had triglycerides levels > 130 mg/dL; 21% had HDL-cholesterol levels < 40 mg/dL and 21% had total cholesterol > 200 mg/dl.⁸ The follow-up study was the first to examine the relationship between within-subject longitudinal change in GFR and proteinuria to within-subject longitudinal change in lipids. Dyslipidemia remained persistent (51%) 6.5 years later and worsened in those with a faster declining GFR, worsening proteinuria, and increasing BMI.⁹ The results of the study suggested that better control of proteinuria and the prevention of obesity might slow the development of dyslipidemia and subsequent CVD.

Wilson et al determined the presence of multiple traditional CV risk factors in the CKiD participants.¹⁰ In addition to hypertension and dyslipidemia, this study also reported rates of obesity and abnormal glucose metabolism. On enrollment, about 30% of the CKiD cohort were either overweight or obese.¹⁰ Although diabetes is very rare in children with CKD, hyperinsulinemia and insulin resistance was present in 9% and 19% of the CKiD cohort, respectively.¹⁰ Importantly, almost one half of the CKiD participants had more than one traditional CV factor,¹⁰ with nearly one-quarter having two or three cardiovascular risk factors. In multivariate analysis, the likelihood of having multiple CV risks was two-fold higher in patients with glomerular disease and nephrotic-range proteinuria. Remarkably, even lean patients had a high prevalence of multiple CV risk factors, with nearly one-quarter having two or three risk factors (Figure 1). This is different from healthy children, in whom the coexistence of multiple CV risk factors (metabolic syndrome) is very rare, and usually found in those who are obese. In the follow-up study, looking specifically at overweight and obese children, classical metabolic syndrome was diagnosed in 40% of overweight and in 60% of obese children with CKD.¹¹ Importantly, children with persistent metabolic syndrome had approximately twice the odds of experiencing significant kidney function decline (>10% per year) compared to those with normal BMI and without multiple CV risk factors.¹¹

Figure 1.

Distribution of the number of cardiovascular risk factors (CVRF) by category of age-gender-specific body mass index percentile lean (0 to 85%), overweight (85 to 95%), and obese (95 to 100%), n=250. There was a significant overall association between these two variables (PCMH _ 0.001). BMI, body mass index. Reprinted with permission.¹⁰

Cardiac structure and function

Left ventricular hypertrophy (LVH) is one of the most common cardiac abnormalities found in children with CKD. The baseline data on LVH from the CKiD cohort indicated a prevalence of 17%.¹² In this study, those with masked and confirmed hypertension had a significantly higher prevalence of LVH than those children with a normal BP (Figure 2). The likelihood of having LVH was four times higher in children with masked hypertension compared to children with normal clinic and ambulatory BP. The study concluded that casual BP measurements alone were insufficient to predict the presence of LVH in children with CKD. The high prevalence of masked hypertension and associated LVH in this study supported the case for early echocardiographic evaluation and ABPM as a part of standard care to screen for the above cardiovascular risks in children with CKD. The follow-up study to assess cardiac structure determined that over 4-years, the prevalence of LVH diminished from 15.3% to 12.6% in a systolic BP model and from 15.1% to 12.6% in a diastolic BP model.¹³ As in the initial cross-sectional study, this analysis suggested that a decline in BP was a significant predictor of a decline in LVH.

Figure 2.

LVH by casual and ambulatory BP status (n=198) is shown. Overall P=0.003. Reprinted with permission.²¹

While hypertension is considered to be the strongest predictor of LVH, adiposity is a known CV risk factor and has been shown to be independently associated with LVH among children with non-CKD hypertension.¹⁴ To address this issue in the CKD population, Brady et al examined the longitudinal association of adiposity with LVH among the CKiD participants and confirmed that, as in general pediatric population, increased BMI was associated with LVH independently of hypertension.¹⁵ The associations were strongest among females, possibly due to abnormal hormonal levels in females with obesity.¹⁵ The study results emphasized that in addition to treatment of hypertension, weight control is important in reducing CV risk in children with CKD.

Studies over last decade have identified fibroblast growth factor 23 (FGF23), a bone-derived circulating peptide, as a novel CKD-related risk factor in the development of LVH, premature death and cardiovascular events.^{16,17,18,19,20} The CKiD study determined that among children and adolescents with GFR ≥45 ml/min per 1.73m², higher plasma FGF23 concentrations were associated with a higher prevalence of LVH (Figure 3), whereas with advanced CKD, factors including hypertension and high BMI were relatively stronger determinants of LVH than was FGF23.²¹ This association was strongest in children with FGF23 levels ≥170 RU/ml in whom the odds of LVH was three times higher than in those with FGF23 levels <100 RU/m. The results of this study raised the question of whether therapeutic interventions that reduce or attenuate the increase in FGF23 can prevent or delay the onset of LVH in children with CKD.

Figure 3.

Higher prevalence of left ventricular hypertrophy associates with higher category of plasma FGF23 concentration in children with GFR ‡45 ml/min per 1.73 m2. FGF23, fibroblast growth factor 23; LVH, left ventricular hypertrophy. Reprinted with permission.²¹

Brady et al investigated the role of cystatin C on CV structure and function in the CKiD cohort.²² Whereas cystatin C is an endogenous proteinase inhibitor widely utilized as a surrogate marker of kidney function, it is also emerging as a biomarker for CVD.^23-26 In a longitudinal analysis, a higher cystatin C level was independently associated with increasing left ventricular mass index (LVMI) and worsening diastolic function (lower E’A’ ratio), even after adjusting for kidney function as assessed by iohexol GFR. This suggested that cystatin C may have an independent role in CVD risk stratification among children with CKD.

Vascular structure and function

To assess vascular structure, the CKiD study has examined carotid intima-media thickness (cIMT), a known early marker of atherosclerosis and a valid predictor of CV events in adults.^27,28 The CKiD study added to the evidence that children with mild to moderate CKD have increased cIMT when compared to healthy children, even after adjusting for differences in age, sex and race (Figure 4).²⁹ In this cross-sectional study, dyslipidemia and hypertension were the only CV risk factors found to be significantly associated with increased cIMT.²⁹ Other CV risk factors, notably BMI, calcium, and phosphorus were not found to be associated with cIMT. The results of this study differentiate between children with mild to moderate CKD in whom traditional CV risk factors predicted increased cIMT, versus children on maintenance dialysis in whom abnormal mineral and bone disease with increased serum phosphorus and calciumphosphorus product are the main predictors of high cIMT.

Figure 4.

Box-percentile plots illustrating the distribution of cIMT measurements in 101 children in the CKiD study and in 97 healthy controls. The numerical values of the 50th percentile are shown. The 2.5th, 5th, 10th, 25th, 75th, 90th, 95th, and 97.5th percentiles of the distribution are denoted. cIMT values < 2.5^th percentile or >97^th percentile are denoted by closed circles. cIMT, carotid artery intima-medial thickness; CKiD, Chronic Kidney Disease in Children study. Reprinted with permission.²⁹

One of the novel CV related observations of the CKiD study has been the presence of aortic dilatation.³⁰ Dilatation of aortic dimensions can be a precursor to aortic dissection and has been reported previously in children on maintenance dialysis and after kidney transplantation.³¹ The prevalence of aortic dilatation in the CKiD study (6%) was higher than would be expected for the general population (2%) or children with primary uncontrolled hypertension (2.8%).³² A higher prevalence of aortic dilatation was associated with a lower BMI and was 2.4 times more likely to occur in those with protein energy wasting (PEW), a marker of poor nutrition. The reason for such associations is not clear, but suggests that primary modulators of PEW (e.g. acidosis, inflammation, etc.)³³ might play role in the development of aortic dilatation in children with CKD.