Đề ôn thi thử môn hóa (87)

30  The Cardiovascular Status of Pediatric Dialysis Patients

tion is common in children with CKD, affecting
up to 40% of pediatric HD patients. Using measurements of midwall shortening fraction
(mwSF), similar systolic function abnormalities
also have been identified in early CKD, albeit at
lower frequency [96]. The mwSF is thought to be
a more accurate marker of systolic function than
eSF, particularly in those patients with LVH, as
eSF tends to overestimate systolic function in this
group. One of relatively new markers of subclinical myocardial dysfunction is abnormal LV systolic strain [97]. It is considered to be an early
predictor of LV dysfunction [98]. In the 4C study
of children with advanced CKD in Europe, LV
strain was assessed by echocardiography [99].
While there was no difference in LV ejection
fraction (EF) between CKD patients and healthy
controls, children with CKD were found to have
a higher prevalence of reduced global circumferential strain components [99]. Rumman et  al.
using echocardiography, also showed global longitudinal strain (GLS) to be lower in dialysis
patients compared to CKD patients [100]. These
patients were followed longitudinally and found
to have improvement of their GLS to their pre-­
dialysis CKD levels [101].
Cardiac MRI has recently been utilized in the
assessment of CV structure and function in children with CKD. Malatesta-Muncher et al. showed
that children on maintenance dialysis had significantly lower circumferential strain than in children after transplantation; strain was inversely
correlated with LVMI [102]. Hothi et al. assessed
myocardial stunning in children during hemodialysis treatment [103]. In their study, 11 of 12
patients developed myocardial stunning while
maintaining LV EF throughout hemodialysis.

 ascular Structure, Function,
V
and Coronary Artery Calcification
(CAC)
A number of cross-sectional observational studies in pediatric dialysis patients or young adult
survivors of pediatric dialysis programs have
described surrogate measures of vascular damage
and sought to identify associated risk factors.

569

Children provide a good opportunity to study
uremic influences on the vasculature as they have
fewer confounding pro-atherosclerotic risk factors such as diabetes and dyslipidemia that are
major confounders in similar adult studies. Since
the initial study of CAC in pediatric patients on
dialysis published in 2000 by Goodman et  al.
[104], virtually all studies conducted in children
on maintenance dialysis consistently have shown
increased carotid artery IMT and increased arterial stiffness (e.g., increased PWV). Many of
these studies also detected CAC.  Key pediatric
studies are shown in Table 30.2.
Although all of the available pediatric studies
are small, often single-center and cross-sectional,
they show remarkably similar risk factors associated with cardiovascular damage. A key risk factor highlighted by virtually all of the studies is
the strong linear association between deteriorating vascular measures and time spent on dialysis
[2, 25, 104, 107, 110, 113]. Prolonged exposure
to the uremic milieu with high, and often worsening Ca–P–PTH control, exposure to pro-­
inflammatory agents such as advanced glycation
end-products and oxidative stress, and reduced

levels of the circulating calcification inhibitors all
contribute toward deleterious structural and functional changes in the vasculature. To support this,
vascular measures have consistently and significantly correlated with Ca, P [2, 25, 69, 78, 104,
107–110], and PTH levels [2, 78, 107, 110]. The
vascular changes have also correlated with medication dosages of calcium-based P binders and
vitamin D compounds, suggesting that dysregulated mineral metabolism is central to the vasculopathy of CKD, and that these modifiable risk
factors require careful monitoring and strict control from the earliest stages of CKD.
An increase in cIMT and PWV have been
shown to begin even in the first decade of life in
children on dialysis [78] and in pre-dialysis CKD
stages 2–4 as well [25, 107, 114]. The 4C study
[17] determined the vascular phenotype in 737
children with advanced CKD: cIMT was elevated
in 41.6%, with only 10.8% of patients displaying
measurements below the 50th percentile; PWV
was increased in 20.1%. The office systolic BP was
the single independent factor significantly associ-


570

ated with all surrogate markers of cardiovascular
disease. Importantly, although structural vascular
changes are found in pre-dialysis patients, the vessel retains its normal compliance and distensibility
properties as compared to controls [107]. However,
with progressive duration and severity of uremic
damage as found in dialysis patients, a further deterioration in cIMT ­coupled with increased vascular
stiffness occurs. Interestingly, an increase in the
vessel wall thickness or cIMT is coupled with a
remodeling of the vessel such that an increase in

the carotid artery lumen occurs, possibly to counter
the stiffness or loss of compliance of the vessel
[25]. It may be this compensatory remodeling in
the early stages of CKD and the more plastic vessels of children that protect them against the deleterious consequences of vascular damage.
Direct evidence of calcification in the coronary vessels has been shown in 15–20% of pediatric chronic dialysis patients [78, 110, 112, 115,
116] and correlates with many of the above listed
risk factors. However, despite the presence of
these risk factors and of CAC, none of the patients
in these studies had overt CVD.
None of the studies in children with CKD
have reported the presence of intimal plaques in
the cardiac or carotid arteries, and although ultrasound is not an accurate means of assessing intimal versus medial changes in the vessel wall, it
appears that uremic vasculopathy, at least in children, is a predominantly medial process.

R. Shroff and M. M. Mitsnefes

Calcification progression on CT scan has also
been shown by Civilibal et  al., with the time-­
averaged serum Ca × P product and serum albumin levels predicting the final CAC score and
change in CAC score, respectively [69]. This
suggests that in the pro-calcific and pro-­
inflammatory uremic milieu “calcium begets calcium,” so our efforts must be directed at the
prevention of calcification starting in the earliest
stages CKD. It is fascinating that in all studies,
patients who did not have baseline calcification
continued to remain free of calcification despite
exposure to similar uremic conditions.
Importantly, the presence of CAC is strongly predictive of myocardial infarction, heart failure and
stroke in adult pre-dialysis CKD patients [117],
and it is an independent predictor of all-cause

mortality, cardiovascular events, and cardiovascular mortality in adult dialysis patients [118].
By ameliorating the uremic milieu, renal
transplantation might intuitively be thought of as
a procedure that might lead to a reversal of some
of the cardiovascular damage from dialysis.
However, there is increasing evidence from adult
studies to show that CVD remains a significant
problem posttransplantation, a problem that may
be driven by hypertension, obesity, and related
risk factors and possibly by immunosuppressive
agents. Krmar et al. have shown that there is no
increase in cIMT following renal transplantation
when there is good blood pressure control [119].
As cIMT progressively increases with age, this
can be interpreted as a regression in cIMT when
hypertension is ameliorated after transplantation
Progression of Vascular
[120]. Litwin et al. have shown that cIMT thickCalcification Through Different
ening and remodeling of the vessel wall begins
Stages of CKD
early in CKD and progress rapidly on dialysis,
Despite a plethora of observational cross-­ correlating with the blood pressure and mean
sectional studies, there are very few longitudinal serum phosphorus levels. Successful transplantastudies that have followed children through pre-­ tion can improve the cIMT toward pre-dialysis
dialysis–dialysis–transplantation phases and values, but cannot normalize it [113].
described changes in vascular markers at different stages of uremia. Calcification progresses
rapidly in patients on dialysis as shown in a sys- Physiological Inhibitors of
tematic follow-up study by Goodman et al. [78, Calcification
104]. When a repeat CT scan was performed after
a mean interval of 20  months, the calcification Vascular calcification occurs in the majority of
score almost doubled in the 10 patients who had patients with CKD, but as noted above, a subset of

evidence of initial calcification [78, 104]. patients do not develop calcification despite expo-


30  The Cardiovascular Status of Pediatric Dialysis Patients

sure to a similar uremic environment [63]. There is
now a growing body of evidence showing that calcification is a highly regulated cell-­mediated process, involving a complex interplay of promoters
and inhibitors of calcification. Animal knockout
models and human single-gene defects have confirmed the role of physiological inhibitors in regulating vascular calcification [121].
Fetuin-A (α2-Heremans–Schmid protein) is a
key circulating calcification inhibitor that contributes to ∼50% of the calcification inhibitory
capacity of human plasma and walls off the nidus
of calcification, thereby preventing further crystal growth. Fetuin-A is a negative acute phase
reactant, and in the pro-inflammatory dialysis
milieu its production may be reduced [63].
Several studies have reported that adults on dialysis have significantly lower fetuin-A levels than
controls. Interestingly, whereas a protective
upregulation of fetuin-A has been reported in
pediatric dialysis patients, with increasing dialysis vintage and in the associated pro-calcific and
pro-inflammatory uremic milieu, fetuin-A levels
are decreased [84]. At the VSMC level, fetuin-A
can inhibit apoptosis, enhance phagocytosis, and
protect the smooth muscle cell from calcifying
[84, 122]. Another group has reported lower
fetuin-A levels in pediatric transplant recipients,
but did not find an association with vascular measures [123].
An important local inhibitor of calcification,
matrix Gla [γ-carboxyglutamic acid] protein
(MGP), is expressed in the media of arteries
where it acts as an inhibitor of Ca–P precipitation

[97, 124]. The γ-carboxylation of MGP is vitamin K dependent, and drugs such as warfarin
may inhibit this process, resulting in the accumulation of inactive under-carboxylated MGP and
ectopic calcification [84, 124]. Osteoprotegerin
and pyrophosphate are other potent calcification
inhibitors that are shown to be perturbed in children with CKD [84]. The importance of circulating calcification inhibitors was recently confirmed
by studies using an in vitro test (T50 test) for the
determination of calcification propensity in
blood. The T50 test quantifies the calcification
inhibition of serum by treatment with supersaturated calcium and phosphate solutions, which

571

results in the formation of primary calciprotein
particles that mainly contain fetuin [125].
Calcification propensity was significantly associated with cardiovascular events in pre-dialysis
CKD and hemodialysis patients [125, 126]. In
incident adult dialysis patients, OPG and fetuinA were significantly associated with all-cause
­
and cardiovascular mortality during follow-up
[127]. While further longitudinal studies are
required to fully characterize these circulating
biomarkers in children, they may prove to be a
useful and convenient measure of an individual
patient’s susceptibility to vascular calcification.

 he Role of Vitamin D
T
in Cardiovascular Health in CKD
Virtually all studies in dialysis patients have
reported the prevalence of 25-hydroxyvitamin D

[25(OH)D] and 1,25-dihydroxyvitamin D
[1,25(OH)2D] deficiency to be on the order of
50–90% [128, 129], and have shown that deficiency begins early in the course of renal decline
[129]. CKD patients can have low 25(OH)D levels for several reasons: they may have less sunlight exposure, the endogenous synthesis of
vitamin D in the skin is reduced in CKD, ingestion of foods that are natural sources of vitamin D
may be diminished, and proteinuria may be
accompanied by high urinary losses of vitamin
D-binding protein [129]. In addition, when the
GFR falls to <50  mL/min/1.73  m2, the kidney
cannot convert “nutritional” 25(OH)D to the biologically active 1,25(OH)2D [130]. The synthesis, metabolism, and interactions of vitamin D in
the Ca–P–PTH axis are shown in Fig.  30.3. A
recent report of nearly 700 children with CKD
across Europe has shown that disease-related factors and vitamin D supplementation are the main
correlates of vitamin D status in children with
CKD, whereas variations in the vitamin
D-binding protein showed only a weak association with the vitamin D status [131]. A core
working group of the European Society for
Paediatric Nephrology has developed recommendations for the evaluation, treatment, and prevention of native vitamin D deficiency and active


R. Shroff and M. M. Mitsnefes

572
Sunlight

Food

oily fish plant/yeast

7-dehydro-cholesterol

UVB

Skin

vitamin D3

vitamin D2
Liver

have shown that there is a bimodal association of
vitamin D levels with vascular measures such
that both low and high levels of vitamin D are
associated with abnormal cIMT and CAC [66].
These effects may be determined by the effects of
vitamin D on Ca–P homeostasis, as well as its
pro-inflammatory effect [66].

25 hydroxylase
Blood

25OHD

2-

PO4

The Vascular Biology of Calcification
Kidney

1-alpha hydroxylase

Parathyroids

PTH

1,25 (OH)2 D

Bone
Blood

Calcium

Intestine

Fig. 30.3  The synthesis, metabolism, and interactions of
vitamin D in the Ca–P–PTH axis

vitamin D analog therapy in children with CKD
stages 2 to 5 and on dialysis [132, 133].
Most tissues and cells in the body have a vitamin D receptor and also have the enzymatic
machinery to convert 25(OH)D to the active form
1,25(OH)2D [128]. In the cardiovascular system,
vitamin D acts as a negative endocrine regulator
of the renin–angiotensin system [134], inhibits
atrial natriuretic peptide [135], increases myocardial contractility, and reduces cardiomyocyte
hypertrophy [136]. Several large epidemiological
studies have consistently shown that hemodialyzed patients receiving any activated vitamin D
treatment have a significant survival advantage
on the order of 20–25% as compared to untreated
patients [80–82].
Clinical studies in children have examined the

effects of vitamin D therapy on vascular measures and calcification (Table  30.2) and shown
that both CAC and cIMT correlated with a higher
calcitriol dosage [25, 66, 78, 107]. However, the
association of vitamin D levels with vascular
measures is more interesting. In a recent study of
children on maintenance dialysis, Shroff et  al.

In recent years, converging evidence from in vitro
studies, molecular genetic techniques, and human
single-gene defects has shown that vascular calcification is an active, highly regulated process
and not merely a passive deposition of Ca and P
in dead or dying cells [50]. In response to raised
extracellular Ca and P levels, vascular smooth
muscle cells (VSMCs) undergo specific phenotypic changes including apoptosis, osteo-/chondrocytic differentiation, and the release of small
membrane-bound bodies called vesicles that
form a nidus for the deposition of basic Ca–P in
the form of hydroxyapatite [50]. Transformation
of VSMC to an osteo-/chondrocytic phenotype is
characterized by the upregulation of bone-specific transcription factors and matrix proteins,
including Runx2/Cbfa1, osterix, and alkaline
phosphatase, that in turn lead to accelerated calcification. Raised serum P has been shown to be
a key factor that triggers osteoblastic differentiation of the VSMC [49, 50, 137]. A schematic diagram depicting key events in the calcification
process is shown in Fig. 30.4.
Using intact arteries from children, Shroff
et al. have shown that calcification in the vessel
wall begins in pre-dialysis CKD stages 4 and 5,
but is significantly greater in dialysis patients
[121]. The calcium load in the vessel wall
increases linearly with time on dialysis and is
strongly correlated with the mean time-averaged

serum Ca x P product. Dialysis vessels showed
VSMC apoptosis with significantly fewer
VSMCs as compared to pre-dialysis or healthy
control vessels, and this may be a key event that
triggers accelerated calcification in dialysis
patients. Importantly, the vessel Ca load did
not result in an increase in cIMT and only the


30  The Cardiovascular Status of Pediatric Dialysis Patients
Fig. 30.4  A schematic
diagram showing
disease-related and
treatment-related risk
factors that lead to
vascular smooth muscle
cell (VSMC) damage
and the mechanisms that
drive accelerated
calcification

573

Renal
osteodystrophy

Ø GFR

Disease – related factors
↑Ca, ↑P, ↑PTH

Inflammation
Hypertension
AGEs
Oxidized lipids

Treatment – related factors
Vitamin D
Ca based P binders
? Vitamin D analogs
Warfarin

VSMC Damage

Osteo/chondrocytic
Differentiation

Apoptosis

↑ Local Ca levels
↑cbfa-1/runx2
↑Alkaline phosphatase
Oseoblastic conversion of SMC

Release of matrix vesicles
and apoptotic bodies
Loss of inhibitors
↓ Fetuin-A & MGP

Calcification

most severely affected patients had coronary
calcification on CT scan, implying that the currently available vascular measures are not sensitive enough to detect early calcification. Shroff
et al. cultured vessel rings from healthy subjects
and pre-dialysis and dialysis patients in graded
concentrations of Ca and P and showed that normal and pre-dialysis vessels were resistant to calcification, while dialysis vessels showed
accelerated calcification in high Ca and P media
[138]. This suggests that dialysis vessels have
lost protective mechanisms; exposure to the uremic milieu has “primed” them to calcify. In the
presence of a high P, even a small increase in Ca

in the culture medium significantly increased calcification, implying that Ca may be a key mediator of VSMC damage and calcification [138], and
careful attention must be paid to even transient
increases in calcium levels such as are seen after
HD, or with the use of calcium-containing phosphate binders and vitamin D analogs.
Recent research suggests that premature aging
in dialysis vessels may drive the process of accelerated calcification. Accumulation of the aging
biomarker prelamin A has been shown in the calcified arteries of children on dialysis [139].
Prelamin A interferes with DNA damage repair
leading to accelerated VSMC senescence [140].


574

This toxic nuclear protein also accumulates in the
calcified vasculature of aged adults and is causal
in the induction of accelerated vascular calcification and stiffening in children with the premature
aging disorder Hutchinson–Gilford progeria
­syndrome [141]. In a recent study, Shanahan
et al. have shown that vessels from children with
CKD undergo oxidative DNA damage and have

elevated senescence markers that may drive
osteogenic differentiation and calcification [142].

The Bone-Vascular Link
There is a growing awareness that mineral dysregulation in CKD is closely linked to abnormal
bone pathology, and that this in turn leads to
extraskeletal calcification. Hormones such as
PTH and vitamin D that closely regulate calcium–phosphate metabolism affect skeletal mineralization and can lead to ectopic soft-tissue
calcification. Key factors produced by osteocytes
(e.g., FGF-23), osteoblasts (e.g., alkaline phosphatase), and osteoclasts (e.g., osteoprotegerin)
also influence vascular calcification [49].
It is a matter of debate whether vascular calcification and bone loss are simultaneously occurring but largely independent processes, or
whether poor bone health predisposes to vascular
calcification, especially in patients with kidney
disease. In a prospective study in 213 adult HD
patients, bone mineral density of the spine was
inversely related to the coronary artery calcification score, and CAC progression was associated
with the severity of osteoporosis [143]. A further
study in adults on HD has shown that low trabecular bone volume and decreased cortical bone
density are associated with coronary artery calcification [144].
The K/DIGO (Kidney Disease Improving
Global Outcomes) working group has proposed
a broader and more encompassing term to
describe this clinical disorder: CKD-mineral
and bone disorder (CKD-MBD) [145]. They
proposed three primary components of CKD-

R. Shroff and M. M. Mitsnefes

MBD: (1) biochemical abnormalities in calcium, phosphorus, PTH, or vitamin D

metabolism; (2) changes in bone histology
(bone turnover or mineralization), linear growth,
and fractures; and (3) vascular or other soft-tissue calcification. The broader definition of
CKD-MBD is an improvement on historical
practice in which renal ostedystrophy and its
management was thought only to affect skeletal
health and growth. Recognition of the importance of the full spectrum of CKD-MBD also
highlights the need for more cautious use of
calcium-­
containing P binders and vitamin D
analogs to minimize the risk of vascular disease,
as will be discussed in detail below.

 valuation and Management of CV
E
Risks in CKD
Primary among all management strategies in
childhood CKD/ESRD is the avoidance of long-­
term dialysis, with preference given to preemptive transplantation when feasible, as the strongest
evidence for cardiovascular risk reduction is that
associated with avoiding dialysis [5]. Although
far from perfect with regard to cardiovascular
risk, successful transplantation can eliminate or
significantly improve uremia-related risk factors
and increase predicted life expectancy by
20–30 years when compared to long-term dialysis. Otherwise, management strategies should be
specific to the stage of CKD (pre-dialysis, dialysis, or transplant) as each has a unique subset of
common risk factors. For those patients who
must have long-term dialysis, the strategy is
directly linked to the achievement of adequate

dialysis outcomes, which include aggressive
monitoring and management of hypertension,
dyslipidemia, calcium–phosphorus metabolism,
anemia, nutrition, systemic inflammation, and
other dialysis complications. Current recommendations for the management of the most common
individual risk factors are summarized below.