Cardiovascular Disease in CKD in Children: Update on Risk Factors, Risk Assessment, and Management

Abstract

In young adults with onset of chronic kidney disease in childhood, cardiovascular disease is the most common cause of death. The likely reason for increased cardiovascular disease in these patients is high prevalence of traditional and uremia-related cardiovascular disease risk factors during childhood chronic kidney disease. Early markers of cardiomyopathy, such as left ventricular hypertrophy and left ventricular dysfunction and early markers of atherosclerosis, such as increased carotid artery intima-media thickness, carotid arterial wall stiffness and coronary artery calcification are frequently found in this patient population. The purpose of this review is to provide an update of recent advances in the understanding and management of cardiovascular disease risks in this population.

Introduction

Despite significant advances in the care of children and adolescents with chronic kidney disease (CKD), long-term survival of children with CKD in the U.S. remains far lower than for the general population. For children on dialysis, anticipated lifespan is reduced by 40–60 years, and for transplant patients by 20–25 years, when compared to an age- and race-matched population^1–3. Further, the most likely cause of this reduction in survival is an excessive burden of cardiovascular mortality, related to both accelerated ischemic heart disease and premature development of dilated cardiomyopathy in young adult survivors of childhood-onset CKD. The burden of cardiovascular morbidity and mortality in this population has been reviewed in detail previously⁴. The purpose of this review is to provide an update of recent advances in the understanding and management of cardiovascular disease (CVD) risks in this population. Specifically, we have focused on a discussion of traditional CVD risk factors, and their evaluation and management, although those uremia-related risk factors with the most robust evidence for association with CV abnormalities in children with CKD are also reviewed.

Risk Factors for Cardiovascular Disease

As in adults, the risk factors thought to be responsible for accelerated CVD in children with CKD can be divided into two primary groups: traditional risk factors for atherosclerotic disease (e.g. dyslipidemia, diabetes, hypertension, smoking) and uremia-related risk factors which are unique to or far more prevalent among patients with CKD. A majority of adults who develop end-stage renal disease (ESRD) do so as a complication of diabetes or generalized atherosclerosis. The frequency of both traditional and uremia-related risk factors for CVD is high, and cardiac disease is already well established at the onset of ESRD in these patients. Although children have neither diabetes nor symptomatic atherosclerosis at the time of starting maintenance dialysis, they unfortunately share similarly high rates of CVD risk factors (Table 1).

Table 1.

Key Risk Factors for CVD in Children With CKD & Relevance after Kidney Transplant

Risk Factor	Prevalence	Impact of Kidney Transplant
Hypertension	48–87%	Prevalence increases significantly after kidney transplant, to ~80–90% at 1 year
Dyslipidemia	50–90%	Highest rates occur after kidney transplant, largely iatrogenic
Obesity	10–30%	Rates double in 1st year after kidney transplant
Abnormal glucose metabolism	Up to 30%	Studied almost exclusively in kidney transplant ; minimal data in children with CKD or end-stage renal disease
Sedentary lifestyle/poor overall fitness	Near 100%	Contrary to expectations, does not improve significantly after kidney transplant
Coexistence of multiple Risk factors	Up to 66%	Emerging data indicates prevalence rises sharply after kidney transplant
Anemia	18–86%	Previously thought less common after kidney transplant; new data shows prevalence up to 86% after kidney transplant
Hyperparathyroidism	30–60%	Improves with kidney transplant

Abbreviations: CKD, chronic kidney disease; CVD, cardiovascular disease.

Traditional Risk Factors

The high prevalence of traditional CVD risk factors in children with CKD has been appreciated for over a decade. Little progress has been made in reducing the frequency of these risk factors in the intervening years.

Hypertension

Hypertensionis the most common traditional risk factor. North American Pediatric Renal Transplant Cooperative Study (NAPRTCS) data demonstrate uncontrolled hypertension rates of 48% in early CKD, increasing to 50–75% in ESRD^5–8; unfortunately, this is not improved with transplantation, with reported prevalence of 50–87%^9–13. Recent data from the Chronic Kidney Disease in Children (CKiD) Study supports that hypertension remains frequent (54%) in patients with early stages of CKD, with up to 25% of study participants having blood pressure in excess of current recommendations¹⁴. Even more concerning, when examining only those children with hypertension who were on antihypertensive medication, the investigators found that 48% still had elevated blood pressure.

Dyslipidemia

Dyslipidemia is also frequent, particularly following kidney transplantation. Reported rates of dyslipidemia range from 50–90%, with elevated total cholesterol and elevated low-density lipoprotein (LDL) cholesterol reported most frequently, and low high-density lipoprotein (HDL) cholesterol reported least often^13,15–23. The highest rates of dyslipidemia are after renal transplant. However, it is also frequent in CKD stage 2–4: hypertriglyceridemia is most common (31%), followed by high total cholesterol, high LDL cholesterol, and low HDL cholesterol (13%, 13%, and 8%, respectively)²⁴.

Obesity

Obesity in children with CKD has been recognized for some time now; unfortunately, the trends in prevalence are mirroring secular obesity trends, with rates rising from 8% (1995) to 12% (2002) in patients on dialysis (NAPRTCS)²⁵. Obesity becomes more common following renal transplant, with both single- and multi-center studies demonstrating doubling of obesity prevalence (from approximately 15% to 30%) in the first year after renal transplant^12,26.

Abnormalities in Insulin and Glucose Metabolism

The frequency of abnormalities in insulin and glucose metabolism, particularly following kidney transplantation, is increasingly appreciated. Based on the current American Diabetes Association guidelines²⁷ for impaired fasting glucose and provisional diagnosis of diabetes mellitus, recent publications examining children receiving kidney transplants demonstrated rates of 16–18% and 7–13%, respectively, with a total of nearly 30% having frank abnormalities in glucose metabolism^23,28. Small studies in children with CKD stage 2–4 or on dialysis have demonstrated rates of hyperinsulinemia as high as 33%, and rates of abnormal insulin resistance (as measured by elevated homeostasis model assessment for insulin resistance [HOMA-IR]) in up to 16% of patients^29,30, indicating that abnormalities in insulin and glucose metabolism may be present earlier than previously believed.

Sedentary Lifestyle

In adults with CKD, sedentary lifestyle and low levels of physical function have been well documented. Painter et al. demonstrated that children with CKD (specifically, those on dialysis and with functioning kidney transplants) are also quite physically inactive, with less than 10% of non-school time spent participating in physical activity³¹. Furthermore, these study participants had peak oxygen uptake and muscle strength measurements that were uniformly below norms for age and sex; only one out of 40 patients tested achieved a “healthy fitness zone” score on field tests of physical fitness. This corresponds well with data demonstrating that aerobic capacity is already diminished in children and adolescents with early stages of CKD (stage 3), and that it does not (in cross-sectional study) improve in kidney transplant recipients³². Given established evidence for the value of regular physical activity and improved fitness in reducing CVD risk in the general population, a reasonable next question is whether similar risk reduction could be achieved in a population with the medical comorbid conditions common in adult survivors of childhood CKD; this remains to be studied.

Frequency of Traditional Risk Factors

Also new to the pediatric literature is data on the frequency with which individual traditional risk factors for CVD coexist in children with CKD. Silverstein et al. assessed 45 children who received kidney transplants, all with stage 2–4 CKD at time of study; two-thirds of patients had at least two risk factors for CVD, and one-third had at least 3 risk factors²³. A multicenter study of over 200 kidney transplant recipients (age 1–21 years) demonstrated that 37% met at least 3 (of a possible 5) diagnostic criteria for the metabolic syndrome at one year post-transplant¹².

Uremia-Related Risk Factors

In addition to traditional risk factors, there are a host of uremia-related risk factors that uniquely increase risk for atherosclerotic CVD in patients with CKD. These are primarily evident in patients on maintenance dialysis rather than in those with early CKD or following successful transplantation. Although maintenance dialysis is typically a temporary state prior to transplantation in children, the consequences of even relatively short-term dialysis persist well beyond successful transplant into adulthood³. We will focus here on those risk factors which are well defined and for which the ability exists to modify those risk factors in the course of routine treatment.

Volume overload

Volume overload among patients on maintenance dialysis is frequent. This is primarily related to failure to achieve and maintain a true dry weight, and is a primary mechanism for the development and persistence of hypertension in the maintenance dialysis population, particularly for patients treated with intermittent hemodialysis rather than peritoneal dialysis. Vandevoorde et al. demonstrated that in children on hemodialysis, those who were hypertensive had significantly higher excess weight postdialysis, and increased normalized intradialytic weight gain than did those who were normotensive, with volume overload identified as the main cause of hypertension³³. This is concordant with previous findings⁶ in a European population. Furthermore, long-term volume overload in children is associated with higher rates of cardiac structural and functional abnormalities that will be discussed later in this article^34–37.

Anemia

Anemia is another highly-prevalent uremia-related CV risk factor in children and adolescents with advanced CKD; unlike the other major uremia-related risk factors, it appears relatively early in the course of CKD. Despite the wide use of recombinant erythropoeisis stimulating agents (ESAs), it remains common. Recent data from the CKiD cohort demonstrated that below a measured GFR of 43ml/min/1.73m², hemoglobin decreased by 0.3g/dL for every 5ml/min/1.73m² drop in GFR³⁸. Data from the NAPRTCS Chronic Renal Insufficiency (CRI) registry supports that anemia is common in children with CKD (increasing from 18.5% in stage 2 CKD to 68% in stage 5 predialysis patients); furthermore, patients with anemia were 55% more likely to be hospitalized than those with normal hemoglobin³⁹. Anemia remains a significant risk for both morbidity and mortality^39,40. Until recently, post transplant anemia has been under-appreciated. However, with introduction of more potent immunosuppression therapy, recently reported anemia rates range from 61% to 86%^41–43.

Abnormalities in calcium-phosphorous metabolism

Abnormalities in calcium-phosphorous metabolism are universal among patients with advanced CKD, and they appear to be among the most important drivers of cardiac and vascular disease progression in this population. Hyperparathyroidism affects approximately 30–45% of children with CKD stage 2–4 and nearly 60% of children on dialysis⁴⁴. Among Turkish children with CKD, nearly 30% had a calcium-phosphorus product above target range, and 40% had intact parathyroid hormone >300pg/ml⁴⁵. Associations between abnormal mineral metabolism and changes in vascular structure in children with CKD have also been documented (discussed in detail below).

Although supplementation of 1,25(OH)₂D₃ in ESRD is an established therapy for control of secondary hyperparathyroidism, there is evidence that1,25(OH)₂D₃ has a direct promoter effect on calcium deposition in vascular smooth muscle cells⁴⁶. Nutritional vitamin D deficiency is also quite common in children with CKD, with only 23% of children with CKD stage 2–4 having adequate levels of 25(OH)-D₃⁴⁷. Further, nutritional vitamin D deficiency is associated with hyperparathyroidism even among those patients with normal activated vitamin D levels⁴⁷.

Malnutrition and Inflammation

Finally, the constellation of malnutrition and inflammation appears to be intimately related to the development of vascular disease in adults with CKD. The same inflammatory markers have been identified at high levels in children on maintenance dialysis^48–50. However, there is no strong evidence to guide inflammation-related risk-modification in children with CKD, although Goldstein et al. did demonstrate reduction in pro-inflammatory cytokine levels with aspirin therapy in children with ESRD⁵¹. For a more complete review of this subject, the reader is referred to recent reviews by Nurmohamed and Nube⁵² and Silverstein⁵³.

Spectrum of Cardiovascular Findings in Children With CKD

Clinically evident cardiovascular lesions (symptomatic coronary artery disease, myocardial infarction, cerebrovascular accident) are fortunately rare in children and adolescents with CKD. However, there is increasing evidence demonstrating significant subclinical cardiovascular abnormalities in this population, all of which are independent predictors of CVD morbidity and mortality (both in the general population and in adults with CKD).

Left ventricular structure and function

As in adults, left ventricular hypertrophy (LVH) develops relatively early in the course of CKD in children, and becomes more common as kidney function declines. Although some small retrospective studies demonstrate regression of LVH with better blood pressure and volume control while on dialysis, others have demonstrated advancement of LVH. Unfortunately, LVH also remains common after pediatric kidney transplant (see Table 2). LVH is also common in adult survivors of ESRD in childhood (47% of male patients and 39% of female patients), as is diastolic dysfunction (13%)⁷⁷.

Table 2.

LVH in children with CKD

Study	Type	Population	Frequency of LVH	Geometry*	LVH associations
Studies in patients with CKD Stage 2–4
Johnstone et al, 1996 (54)	Cross-sectional	Mean eGFR not reported.	22% (n=32)	All concentric
Mitsnefes et al. 2003 (55)	Cross-sectional	Mean GFR 39 ml/min/1.73m2	24% (n=25)	Not reported	Contractile reserve during exercise similar to control group.
Poyrazoglu et al. 2004 (56)	Cross-sectional	eGFR 17–48 ml/min/1.73m2)	50%(n=10)	Not reported	Lower hemoglobin
Malikenas et al. 2005 (57)	Cross-sectional	Mean eGFR 34 ml/min/1.73m2	36% (n=56)	23%E, 13%C	More common at lower eGFR
Matteucci et al. 2006 (58)	Cross-sectional	Mean eGFR 49 ml/min/1.73m2	33%(n=156)	12%C, 21%E	Males, anemia, obesity; BP not.
Mitsnefes et al. 2006 (59)	Prospective cohort, followed 2 yrs	Mean measured GFR 49 ml/min/1.73m2 at echo 1, 43ml/min/1.73m2 at echo 2	19% at echo 1; 39% at echo 2 (n=31)	Echo 1: 13%E, 6%C; Echo 2: 23%E, 16%C	Initial LVMI, lower hemoglobin, rise in iPTH, night SBP load all associated w/change in LVMI
Mitsnefes et al. 2008 (60)	Cross-sectional	Stage 2–4 CKD; mean GFR 42ml/min/1.73m2	19% (n=293)	13%E/6%C	Both confirmed & masked HTN
Weaver et al. 2008 (61)	Cross-sectional	Stage 2–4 CKD; mean GFR 46ml/min/1.73m2	17% (n=45)	Not reported	Higher cardiac output
Studies in patients on maintenance dialysis
Drukker et al, 1981 (62)	Cross-sectional	HD	40% (n=10)	Not reported	“moderate to severe” HTN
Morris et al, 1993 (63)	Cross-sectional	Dialysis (12 PD, 1 HD)	92% (n=13)	Not reported	50% also had diastolic dysfunction
Loirat et al, 1994 (64)	Cross-sectional	PD and HD	59% HD (n=85); 29% PD (n=198)
Johnstone et al, 1996 (54)	Cross-sectional	PD	30% (n=10)	All concentric
Mitsnefes et al, 2000 (65)	Cross-sectional	PD and HD	85% HD (n=26); 68% PD (n=38)	Of 48 w/LVH, 25E, 23C.
Mitsnefes et al, 2001 (66)	Retrospective cohort	13 HD, 16PD	69% at initiation & at follow-up	Of 20 w/LVH, 11E, 9C	Change in LVMI over time correlated w/change in SBP index
Poyrazoglu et al. 2004 (56)	Cross-sectional	9 PD, 8 HD	88% (n=17)	Not reported	Lower hemoglobin
Lumpaopong et al. 2005 (67)	Retrospective	11 HD, 3PD	57% (n=14)	7E, 1C	Higher BP
Ulinski et al, 2006 (34)	Prospective cohort	HD	Start dialysis: 82%; 41% at 16mof/u (n=17)	Not reported	Higher BP is only independent predictor
Mitsnefes et al. 2006 (35)	Retrospective cohort	HD at least 2 years’ duration (n=17)	Initiation: 82% 2 years: 82%	Initial: 3C/11E 2yr: 5C/9E
Ucar et al. 2008 (68)	Cross-sectional	Maintenance PD	76% (n=25)	14C/5E	Higher LVMI associated with increased global LV dysfunction
Longitudinal studies of patients prior to (on dialysis) and following rena) transplantation
Mitsnefes et al, 2001 (69)	Prospective cohort	16 PD; 7HD w/subsequent RT	52% while on dialysis vs 56% after transplant	Not reported	Mean LVMI decreased but not significant
Becker-Cohen et al. 2008 (70)	Prospective cohort	PD and HD, then RT	D: 50% (n=12); RT: 54%–>8% (n=13)	Not reported	Reduction in LVMI after RT; no change if stayed on dialysis
Guizar-Mendoza et al. 2006 (71)	Prospective cohort, ages 8–35 yrs	PD and HD, then RT (n=40)	Pre-RT: 82.5%; Post-RT57.5%	Not reported	No patient with increase in LVM after RT: mean LVMI pre-RT 73g/m2,7
Studies in patients with successful renal transplantation
Johnstone et al, 1996 (54)	Cross-sectional		63% (n=30)	All concentric
Matteucci et al, 1999 (72)	Cross-sectional		82% (n=28)	Not reported	Higher mean 24hr SBP
El-Husseini et al, 2004 (73)	Cross-sectional		48% (n=73); median of 4.6 yrs post-RT	Not reported	Also showed 14% prevalence of LV systolic dysfunction
Kitzmueller et al, 2004 (74)	Retrospective cohort		48% (n=39); 20 w/2 yr repeat	Not reported	Subgroup of 20: 60% w/LVH at initial echo; 70% w/LVH on f/u
Bullington et al. 2006 (75)	Retrospective cohort	Up to 3 echos per pt.	54% in echo 1 (n=47)	59%E/41%C	14 pts w/3 echos: LVH in 57% at echo 1 vs 64% at echo 3.
Silverstein et al. 2008 (23)	Cross-sectional	Mean GFR 88ml/min/1.73m2	18.2% (n=45)	Not reported
Wilson et al. 2009 (12)	Cross-sectional		40% (n=113)	53%C, 47%E.	Metabolic syndrome
Muscheites et al. 2008 (76)	Cross-sectional	CKD 2–4, ESRD, & RT	50% (n=26); 8 w/severe LVH	Not reported

C, concentric; E, eccentric; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; HD, hemodialysis; HTN, hypertension; iPTH, intact parathyroid hormone; LVMI, left ventricular mass index; RT, renal transplant; PD, peritoneal dialysis.

To convert GFR in mL/min/1.73m² to mL/s/1.73m², x0.01667.

Diastolic dysfunction is thought to be the initial functional LV abnormality evident in children with CKD. Historically, the most widely used method of assessment of impaired LV relaxation has been the use of Doppler measurement of the mitral inflow velocity (E/A ratio). By this method, reduced and/or frankly abnormal E/A ratios have been seen in patients with CKD, ESRD, and renal transplant^54,63,78. However, left atrial pressure and preload significantly affect the E/A ratio. Given the chronic hypervolemia associated with advanced CKD, E/A ratio may not be an ideal means of assessing diastolic function. More recently, tissue Doppler imaging (TDI) was introduced as a less load dependent and more accurate means of evaluating diastolic function in CKD. Diastolic dysfunction has also been documented by this method^79–81, thus supporting that the cardiac dysfunction identified in earlier studies was not merely an artifact of the methodology used. Overall, children on maintenance dialysis have worse diastolic dysfunction than those with either CKD stage 2–4 or functioning kidney transplants. Diastolic dysfunction was recently demonstrated to be independently associated with reduced maximal aerobic capacity (VO₂max) in children with CKD⁸².

Systolic function has classically been thought of as preserved in children with CKD. While that appears to be true of overt systolic function abnormalities (as assessed by LV contractility or endocardial shortening fraction, eSF), recent studies have demonstrated that subclinical systolic dysfunction is common in children with CKD, affecting up to 40% of hemodialysis patients. Further, such abnormalities have been identified relatively early in the course of CKD, albeit at lower frequency^83,84. These abnormalities have been identified through utilization of midwall shortening fraction (mwSF), which is thought to more accurately estimate systolic function than eSF, particularly in those patients with LVH, as eSF tends to overestimate systolic function in this group.

Large vessel disease

Large vessel disease occurs at disturbingly early ages among children with CKD. Known abnormalities range from early changes in endothelial function (as assessed by flow mediated dilatation of the brachial artery) to increased carotid intimal media thickness (cIMT), coronary calcifications, and frank atherosclerotic changes. Major risk factors for early vascular disease identified include longer duration of dialysis dependence, higher calcium-phosphorous product, cumulative dose of calcium-based phosphate binders, and cumulative dose of activated vitamin D (specifically calcitriol). Findings of major studies of cIMT and coronary artery calcifications (CAC) in children with CKD published to date are summarized in Tables 3 and 4. Particular attention should be paid to three recently published studies.

Table 3.

Intima-media thickness (IMT) in children with CKD

Study	Type	Population	Findings	IMT Associations	Notes
Groothoff et al. 2002 (85)	Cross-sectional	130 adult survivors of childhood ESRD	IMT similar to controls but higher stiffness & lower distensibility
Oh et al. 2002 (86)	Cross-sectional	19–39 yr old survivors of childhood CKD	IMT significantly higher than controls	Dialysis time; Ca X P
Mitsnefes et al. 2004 (87)	Cross-sectional	31 pediatric RT vs. healthy matched control	Higher carotid IMT, lower distensibility	SBP load, number of BP meds	eGFR 44–128 ml/min/1.73m2
Litwin et al. 2005 (88)	Cross-sectional	55 CKD 2–4; 37 on dialysis; 34 post-RT; 270 healthy control	cIMT, femoral artery IMT increased in all pt groups	Mean past Ca x P, young age, dialysis at time of study	Dialysis patients significantly worse than CKD or RT
Mitsnefes et al. 2005 (89)	Cross-sectional	44 CKD 2–4; 16 on dialysis; 35 healthy control	cIMT & stiffness higher in both CKD & ESRD	Increased Ca x P and dialysis;	Arterial compliance similar in CKD 2–4 and controls; worse in dialysis
Briese et al. 2006 (90)	Cross-sectional	40 young adults with childhood ESRD	IMT similar to healthy controls	n/a	Relatively low Ca intake & Vit. D compared to other study populations
Poyrazoglu et al. 2007 (91)	Cross-sectional	34 children & young adults with ESRD; 20 controls	Higher IMT in ESRD than control	Increased BP, LVMI negative correlation w/iPTH
Shroff et al. 2007 (92)	Cross-sectional	85 children on dialysis >6 months	Higher IMT in ESRD than control	Increased iPTH & Ca x P	If iPTH < 2x upper limit of normal, vasculature comparable to controls
Bilginer et al. 2007 (93)	Cross-sectional	24 children & young adults with RT; 20 controls	Carotid IMT higher in RT than control	Duration of renal failure, duration of dialysis, mean past Ca x P
van Summeren et al. 2008 (94)	Cross-sectional	29 children with RT; 54 healthy controls	IMT& elasticity of carotid significantly higher than in controls	No association with calcification inhibitors (MGP & fetuin-A)	Fetuin-A decreased in RT vs. control; no differences in MGP levels
Muscheites et al. 2008 (76)	Cross-sectional	26 children with CKD 2–4, ESRD, or RT; 24 controls	IMT SDS score significantly higher than controls	Impaired FMD
Litwin et al. 2008 (95)	Observational cohort (12 months f/u)	24 CKD 2–4; 32 ESRD (19 had RT between 1 & 2 evaluation)	Mean IMT-SDS above norms in all group	Total dialysis vintage, BP, phosphate	Patients with RT showed stabilization/partial regression of IMT
Ziolkowska et al. 2008 (96)	Cross-sectional	32 CKD 2–4; 28 ESRD; 43 controls	37 patients w/abnl IMT(18 CKD, 19 CSRD)	Negative correlation with fetuin-A	15 patients also w/abnormal vessel wall echogenicity & 2 with calcifications
Krmar et al. 2008 (97)	Observational cohort (mean 4.1 year f/u interval)	31 RT; 21 controls	Baseline & f/u IMT both higher than controls	No significant associations between BP& IMT	Controls only studied at baseline; IMT in RT patients was stable over time
Delucchi et al. 2008 (98)	Cross-sectional	8 children on dialysis; 12 RT; 30 controls	Both pt groups with IMT higher than control	Duration of dialysis

BP, blood pressure; CKD, chronic kidney disease; Ca X P, calcium-phosphorous product; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; FMD, flow-mediated vasodilatation; IMT, intima-media thickness; iPTH, intact parathyroid hormone; LVMI, left ventricular mass index; MGP, matrix Gla protein; RT, renal transplant.

Table 4.

Calcification in children with CKD

Study	Type	Population	Findings	CAC Associations	Notes
Milliner et al. 1990 (99)	Retrospective autopsy series	120 pediatric CKD, dialysis, RT	10% w/CAC; soft tissue calcification in 60%	Vitamin D, age at onset of ESRD, Ca x P, male sex	Autopsy study; 4 patients w/evidence of MI or calcification of myocardium
Goodman et al. 2000 (100)	Cross- sectional	39 dialysis pts (age 7–30); 60 controls	No CAC in pts <20 yrs; 14/16 pts>20 years w/CAC	Age, serum P, duration of dialysis, Ca intake, Ca x P	10 patients w/CAC had f/u scan at mean 20 mos; calcification score nearly doubled
Oh et al. 2002 (86)	Cross- sectional	19–39 yr old childhood CKD survivors (n=37)	92% w/CAC	Current dialysis, dialysis duration, iPTH, CRP, Ca x P, C pneumoniae seropositivity, homocysteine	34% also w/cardiac valve calcification; 32% w/aortic calcification
Ishitani et al. 2005 (101)	Cross- sectional	21–48 yr old pediatric RT recipients (n=19)	47% w/CAC, all in pts >30 yrs old	None identified as significant	12 pts never dialyzed; 6 w/≤6 months total dialysis time; 1 w/24 months total dialysis
Briese et al. 2006 (90)	Cross- sectional	40 young adults w/childhood ESRD	10% w/moderate to severe CAC.	Age, ESRD duration; time on dialysis; BP; cIMT; total Ca dose, Vit. D intake	Relatively low Ca intake & Vit D compared to other study populations
Civilibal et al. 2006 (102)	Cross- sectional	Children w/ESRD (15 HD, 24 PD, 14 RT)	15% w/CAC (20% HD, 12.5% PD, 14% RT)	Dialysis duration, serum P, Ca x P, iPTH, & vitB12; Hgb; total Ca & Vit D doses	3 patients w/severe (high probability of stenosis) lesions.
Shroff et al. 2007 (92)	Cross-sectional	85 children on dialysis>6 mos	20% w/CAC (27% among those w/iPTH ≥ 2x upper limit of normal)	Time integrated iPTH, serum phosphorous	2 w/severe CAC; only high iPTH group had moderate/severe CAC
Civilibal et al. 2009 (103)	Prospective cohort	48 children w/ESRD; 8 w/baseline CAC	0% CAC incidence. Among those with baseline CAC, median score increased	Final Ca x P; inverse association w/serum albumin	1 CAC resolution; 3 w/mild decrease; 3 w/moderate increase; 1 w/large increase

BP, blood pressure; Ca X P, calcium-phosphorous product; CAC, coronary artery calcification; cIMT, carotid intima media thickness; CKD, chronic kidney disease; CRP, c-reactive protein; ESRD, end-stage renal disease; HD, hemodialysis; Hgb, hemoglobin; PD, peritoneal dialysis; Hgb, hemoglobin; RT, renal transplant.

First, Krmar et al.⁹⁷ concluded, based on their observational study of two groups of kidney transplant recipients (one with strict ambulatory normotension throughout the study, and one with routinely treated hypertension), that blood pressure is not a contributing factor to increased IMT, in that cIMT was not significantly different between the two groups at either baseline or follow-up. However, it should be noted that exceptional blood pressure control was associated with stable cIMT over the 4 year follow-up period despite the fact that population studies have demonstrated a 0.01–0.02mm/yr increase in cIMT among healthy patients¹⁰⁴. For a more comprehensive review of the use of IMT in children, the reader is referred to a recent review by Litwin and Niemirska¹⁰⁵.

The second study is the recently published longitudinal study of CAC by Civilibal et al¹⁰³. This is the only study to date regarding the natural history of CAC in children with ESRD. Briefly, among a cohort of 48 children studied with multidetector spiral computed tomography at baseline and again 2 years later, 8 were found to have CAC at study inception. None of the 40 patients without CAC at baseline developed CAC in the intervening 2 years. Among the eight with CAC at baseline, the median score increased significantly over the interval, although CAC in one patient did resolve completely. Time averaged mean calcium-phosphorus product was independently associated with final CAC score, and the serum albumin at time of the second scan was the only independent predictor of the change in CAC score over time.

The final study, by Shroff et al. ¹⁰⁶, sheds light into the mechanisms responsible for the development of vascular calcification in children with ESRD. In this study, 34 children with CKD (10 predialysis, 24 dialysis) underwent examination of medium-sized muscular arteries. Vessels from both predialysis and dialysis patients had higher calcium load than control vessels. Dialysis patients had significantly higher calcium load than predialysis patients, despite similar age and kidney function in the two groups; vascular calcification strongly correlated with mean serum calcium-phosphorus product. No predialysis patient (despite increased calcium load) had reduced numbers of vascular smooth muscle cells per unit area. However, among dialysis patients, there was a significant reduction in vascular smooth muscle cell number, even when controlling for vessel calcium load. In summary, it appears that calcium accumulation begins well in advance of initiation of maintenance dialysis, but accelerates and leads to overt changes in vascular structure only following initiation of dialysis. Neither cIMT nor CAC was a sensitive indicator of increased calcium load.

In addition to measuring IMT and CAC, decreased flow-mediated dilation of the brachial artery has been proposed as a marker of vascular dysfunction that may precede frank structural changes. Early work in children indicated promise in this regard, with several small studies demonstrating decreased FMD in patients with advanced CKD (predialyisis, dialysis, and post-transplant)^107–110. However, a more recent study of children with less severe kidney impairment (CKD stage 2–4) was less promising regarding the promise of FMD as a marker of early vascular changes, with median FMD similar to that of healthy controls¹¹¹.

Evaluation and Management of Cardiovascular Risk in Children With CKD

The most significant change in paradigm regarding management of CV risk in children with CKD came with the 2006 release of the American Heart Association guidelines for CV risk reduction in high-risk pediatric patients¹¹². There, for the first time, children with CKD were stratified as “high risk” for the development of CVD, with associated “pathological and/or clinical evidence for manifest coronary disease before 30 years of age.” This group is otherwise limited to patients with homozygous familial hypercholesterolemia, type 1 diabetes, heart transplant recipients, and Kawaski disease patients with current coronary aneurysms. The panel’s treatment recommendations for children with CKD refer the reader to several National Kidney Foundation Kidney Dialysis Outcomes and Quality Initiative (NKF KDOQI) clinical practice guidelines.

Primary among management strategies in childhood CKD/ESRD is the avoidance of long-term dialysis, with preference for preemptive transplantation when feasible, as the strongest evidence for cardiovascular risk reduction is that associated with avoiding dialysis³. Although far from perfect with regard to CV risk, successful transplantation can eliminate or significantly improve uremia-related risk factors and increases predicted life expectancy by 20–30 years compared to long-term dialysis. Otherwise, management strategies are specific to the stage of CKD (predialysis, dialysis, or transplant) as each has a unique subset of common risk factors. Current recommendations for management of the most common risk factors are summarized in Table 5, with comments below.

Table 5.

Summary of Treatment Goals for CVD Risk Factors in Children With CKD

Risk factor	Goal	Reference
Hypertension	Blood pressure <90th percentile for age, sex, and height, or <120/80mmHg, whichever is lower.	NKF KDOQI Clinical Practice Guidelines on Hypertension and Antihypertensive Agents in Chronic Kidney Disease113 and the Fourth report on blood pressure in children114.
Dyslipidemia	Maintenance of total cholesterol, LDL cholesterol, HDL cholesterol, and triglycerides in age-appropriate ranges for high cardiovascular disease risk patients.	NKF KDOQI Clinical Practice Guidelines for Managing Dyslipidemias in Chronic Kidney Disease115 & for Managing Dyslipidemias in Kidney Transplant Patients116 (note: these guidelines contain recommendations only for patients older than 13 years, with end-stage renal disease & transplant, respectively). For patients with stage 2–4 CKD, and those younger than 13 years, see clinical report by Daniels et al.117 for age-appropriate guidelines in high-risk patients.
Obesity & Physical Inactivity	Maintenance of normal body habitus (Body mass index between 5th and 85th percentile for age, sex).	NKF KDOQI Clinical Practice Guidelines for Nutrition in Chronic Renal Failure118.
Hyperinsulinemia/Hyperglycemia	Maintenance of normoglycemia and normal insulin sensitivity.	No pediatric-specific guidelines exist.
Abnormal mineral metabolism	Phosphorous, Ca x P, and intact parathyroid hormone goals vary by age and CKD stage.	NKF KDOQI Clinical Practice Guidelines for Bone Metabolism and Disease in Children with Chronic Kidney Disease119.
Anemia	Maintenance of hemoglobin in the range of 11.0 to 12.0gm/dL in patients receiving erythropoeisis stimulating agent (ESA) therapy, along with maintenance of adequate iron stores.	KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Anemia in Chronic Kidney Disease120.
LVH	Maintenance of normal left ventricular mass and function; in those patients with pre-existing left ventricular hypertension, normalization of echocardiographic findings.	KDOQI Clinical Practice Guidelines for Cardiovascular Disease in Dialysis Patients121.

Abbreviations: NKF, National Kidney Foundation; KDOQI, Kidney Disease Outcomes Quality Initiative; LDL, low-density lipoprotein; HDL, high-density lipoprotein; CKD, chronic kidney disease; Ca x P, calcium-phosphorus product.

Hypertension

Hypertension therapy is determined by CKD stage. Angiotensin-converting enzyme inhibition (ACE-Inhibition) and/or angiotensin-receptor blockade (ARB) is preferred in stage 2–4 CKD. Among patients on maintenance dialysis, meticulous attention to management of volume overload is essential, which may necessitate more frequent or intense dialysis than a standard thrice-weekly regimen. Although data in children is limited³⁷, it is consistent with evidence in adult hemodialysis patients that significant improvements in volume control, hypertension, and regression in left ventricular hypertrophy can be achieved with either daily in-center dialysis or nocturnal hemodialysis performed 5–7 nights per week^122–124. Following transplant, in addition to judicious use of antihypertensive agents, minimization/avoidance of iatrogenic causes of hypertension should be considered.

Guidelines suggest that ambulatory blood pressure monitoring “be considered” to assess for nocturnal hypertension, particularly after renal transplant, but stop short of recommending its routine use. However, newer literature suggests that use of annual ambulatory blood pressure monitoring leads to improved rates of achievement of goal blood pressure compared to standard therapy in children with kidney transplants¹²⁵. Recent data from the CKiD cohort demonstrate that masked hypertension (defined as ambulatory hypertension in the presence of casual blood pressure <95^th percentile) is present in 25% of patients with stage 2–4 CKD, and that the presence of masked hypertension is associated with a two-fold increase in frequency of LVH at time of initial examination⁶⁰. Therefore, routine use of ambulatory blood pressure monitoring in children with CKD (beginning in early CKD & continuing thereafter) may significantly improve ascertainment and management of hypertension.

Dyslipidemia

The latest guidelines for lipid screening and treatment in childhood state that for children 8 years and older with diabetes, statin therapy should be considered for those with LDL cholesterol ≥ 130mg/dL¹¹⁷. Although these guidelines do not specifically address children with CKD, given that the 2006 guidelines from the American Heart Association clearly place patients with CKD in the same high CV risk group as those with diabetes and heart transplant, it could be argued that statin therapy be standard of care for patients with CKD who are 8 years and older with LDL cholesterol ≥ 130mg/dL, although this is not explicitly stated in any guidelines. There exist no data regarding changes in long-term outcome related to tight lipid control in children with CKD; however, emerging data suggest a survival benefit associated with statin use in adult kidney transplant recipients¹²⁶. Statins are also considered standard of care in all patients >1 year of age following pediatric heart transplantation (with the goal of reducing LDL cholesterol to below 100mg/dL), due to the associated reduction in incidence of transplant vasculopathy. Emerging evidence in adults suggests potential for similar improvements in renal transplant. Serón et al. demonstrated a significant reduction in incidence of renal transplant vasculopathy (33% versus 7%) in the first 6 months after transplant with use of statin therapy compared to placebo¹²⁷. Although it remains unclear what (if any) benefit such treatment would have on long-term graft survival, one might speculate that strategies improving graft survival could improve cardiovascular outcomes over decades.

Obesity & Physical Inactivity

Evidence from adult maintenance dialysis populations suggests a survival advantage conferred by higher body mass index^128–130; however, this appears to be limited to those with low body fat and high muscle mass¹³¹. Per KDOQI, “the safety and efficacy of weight loss in the overweight dialysis patient is unknown… Therefore, weight loss in the dialysis patients should be approached with close monitoring by a registered dietician and physician¹²¹.” However, increased physical activity is encouraged. Whether obesity in children with ESRD confers survival benefits while on maintenance dialysis is unknown. After pediatric kidney transplantation obesity is associated with higher rates of graft loss and graft dysfunction^132,133. There is evidence in children that obesity reduces efficacy of antihypertensive therapy¹³⁴.

Hyperinsulinemia/Hyperglycemia

No formal recommendations exist for frequency and type of glucose monitoring or assessment of insulin resistance in children with CKD without preexisting diabetes. For those children with known type I diabetes and CKD of any cause, KDOQI guidelines defer to the American Diabetes Association standards of care for children and adolescents¹³⁵. Following transplantation, serum glucose should be monitored routinely given the high incidence of post-transplantation diabetes. There are no formal guidelines for assessments of insulin resistance in these patients. However, as abnormal insulin resistance clearly exists prior to the development of frank hyperglycemia (see Risk Factors, above), and given that such data could serve as impetus for more aggressive lifestyle interventions, there may be a role for regular assessments even in patients with earlier stages of CKD.

Abnormal mineral metabolism

Although there is evidence suggesting that calcium and phosphorous load are associated with significant increase in risk for early CVD, there is little hard evidence supporting improved CV outcomes associated with achievement of these treatment goals. More research is needed regarding optimal clinical management of calcium-phosphorous metabolism, including that regarding the role of non-calcium containing phosphorous binders and calcimimetics in children with ESRD.

Anemia

Notably, anemia in kidney transplant recipients may be more difficult to treat successfully related to relative ESA-resistance due to multiple factors (immunosuppression-related bone marrow suppression, infections, malignancy, inflammation related to rejection, etc). Also of note, patients in the early pretransplant period are at somewhat higher risk for iron deficiency than at other stages of disease, related to low pretransplant iron stores in the face of increased iron utilization following resumption of erythropoiesis after successful transplant.

Left Ventricular Hypertrophy

Guidelines recommend screening echocardiography within 3 months of beginning dialysis, with follow-up examinations every six months for those with abnormal studies or annually if normal. As noted previously, meticulous attention to volume control is the treatment strategy most likely to result in improvement of LVH in the maintenance dialysis patient, although there is certainly a role for anti-hypertensive therapy (specifically ACE-Inhibition/ARB) as well. Following transplant, antihypertensive therapy and ACE-Inhibition/ARB become the main therapeutic approach. Given the high frequency of LVH in kidney transplant recipients (Table 2), our center continues with routine echocardiography following transplantation as well, with studies at 1 year post-transplant and then every 2 years thereafter if normal, or annually (at a minimum frequency) if abnormal. There is not enough evidence to recommend a routine evaluation for other intermediate CV outcomes (e.g. IMT or CAC) in children with CKD.

Conclusions

Morbidity and mortality associated with cardiovascular disease remain a major threat to the long-term survival of children and adolescents with CKD. Cardiovascular risk factors and early cardiovascular changes are common even in very young patients, and much remains to be done to define and achieve optimal management of these patients.