Pediatric Cardiology

, Volume 34, Issue 5, pp 1218–1225

Echo-Doppler Assessment of the Biophysical Properties of the Aorta in Children With Chronic Kidney Disease

Authors

  • Mohammed Alghamdi
    • Division of Cardiology, Department of Pediatrics, British Columbia Children’s HospitalThe University of British Columbia
    • Division of Pediatric Cardiology, King Fahad Cardiac CentreCollege of Medicine, King Saud University
  • Astrid M. De Souza
    • Division of Cardiology, Department of Pediatrics, British Columbia Children’s HospitalThe University of British Columbia
  • Colin T. White
    • Division of Nephrology, Department of Pediatrics, British Columbia Children’s HospitalThe University of British Columbia
  • M. Terri Potts
    • Division of Cardiology, Department of Pediatrics, British Columbia Children’s HospitalThe University of British Columbia
  • Bradley A. Warady
    • Division of Nephrology, Department of Pediatrics, Children’s Mercy HospitalThe University of Missouri-Kansas City School of Medicine
  • Susan L. Furth
    • Division of Nephrology, Department of Pediatrics, Children’s Hospital of PhiladelphiaThe University of Pennsylvania
  • Thomas R. Kimball
    • Division of Cardiology, Department of Pediatrics, Cincinnati Children’s Hospital Medical CenterThe University of Cincinnati
  • James E. Potts
    • Division of Cardiology, Department of Pediatrics, British Columbia Children’s HospitalThe University of British Columbia
    • Division of Cardiology, Department of Pediatrics, British Columbia Children’s HospitalThe University of British Columbia
    • Children’s Heart Centre
Original Article

DOI: 10.1007/s00246-013-0632-5

Cite this article as:
Alghamdi, M., De Souza, A.M., White, C.T. et al. Pediatr Cardiol (2013) 34: 1218. doi:10.1007/s00246-013-0632-5

Abstract

Chronic kidney disease (CKD) is known to cause increased arterial stiffness, which is an important independent risk factor for adverse cardiovascular events. The purpose of this study was to assess the vascular properties of the aorta (AO) in a group of children with CKD using a noninvasive echocardiography (echo)-Doppler method. We studied 24 children with stages 2 through 5 CKD and 48 age-matched controls. Detailed echocardiographic assessment and echo-Doppler pulse wave velocity (PWV) was performed. Indices of arterial stiffness, including characteristic (Zc) and input (Zi) impedances, elastic pressure-strain modulus (Ep), and arterial wall stiffness index, were calculated. CKD patients underwent full nephrology assessment, and an iohexol glomerular filtration rate was performed, which allowed for accurate assignment of the CKD stage. CKD patients had greater median systolic blood pressure (114 vs. 110 mmHg; p < 0.04) and pulse pressure (51 vs. 40 mmHg; p < 0.001) compared with controls. PWV was similar between groups (358 vs. 344 cm s−1; p = 0.759), whereas Zi (182 vs. 131 dyne s cm−5; p < 0.001), Zc (146 vs. 138 dyne s cm−5; p = 0.05), and Ep (280 vs. 230 mmHg; p < 0.02) were significantly greater in CKD than in controls. Although load-dependent measures of arterial stiffness were greater in non-dialysis dependent CKD patients, PWV was not increased compared with controls. This suggests that the increased arterial stiffness may not be permanent in these pediatric patients with kidney disease.

Keywords

Kidney diseaseChildrenAortaArterial stiffnessPulse wave velocityEchocardiography

Abbreviations

AO

Aorta or aortic

AOcsa

Aortic cross-sectional area

AOD

(Dd in Fig. 1) Aortic diameter in diastole

AOL

Aortic length

AOS

(Ds in Fig. 1) Aortic diameter in systole

β-index

Arterial wall stiffness index

BMI

Body mass index

BPdia

Diastolic blood pressure

BPsys

Systolic blood pressure

BSA

Body surface area

BUN

Blood urea nitrogen

Ca-T

Total calcium

cIMT

Carotid artery intima-media thickness

CKD

Chronic kidney disease

CKiD

Chronic kidney disease in children study

Ep

Elastic pressure-strain modulus

ESRD

End-stage renal disease

ET

Ejection time

ETc

Ejection time corrected for heart rate

GFR

Glomerular filtration rate

Hb

Hemoglobin

IVSDi

Interventricular septal thickness in diastole indexed to body surface area

LDL-C

Low density lipoprotein cholesterol

Ln

Natural logarithm

LV

Left ventricular

LVEDi

Left ventricular end-diastolic dimension indexed to body surface area

LVESi

Left ventricular end-systolic dimension indexed to body surface area

LVH

Left ventricular hypertrophy

LVMassi

Left ventricular mass indexed to body surface area

MVCFc

Mean velocity of circumferential fiber shortening corrected for heart rate

σPS

Peak systolic wall stress

π (pi)

Constant = 3.14

PkAOFlow

Peak aortic flow

PkAOV

Peak aortic velocity

PO4

Phosphate

PP

Pulse pressure

PTH

Parathyroid hormone

PWDi

Posterior wall thickness in diastole indexed to body surface area

PWSi

Posterior wall thickness in systole indexed to body surface area

PWV

Pulse wave velocity

ρ (rho)

Blood density = 1.06 g cm−5, 1 dyne = 1 g cm−2 s−1

SCr

Serum creatinine

SF

Shortening fraction

TC

Total cholesterol

TG

Triglycerides

TT

Transit time

Zc

Characteristic impedance

Zi

Input impedance

Introduction

Arterial stiffening is an important and independent risk factor for future cardiovascular events in both healthy adults and those with diseases known to affect the stiffness of the large arteries, such as diabetes, hypertension, and renal disease [38, 42]. In patients with chronic kidney disease (CKD), increased arterial stiffening may be the result of fluid overload, hypertension, altered lipid metabolism, vascular calcification, and uremia leading to inflammation and oxidative stress injury [19, 35]. The overall increase in arterial stiffness eventually leads to left-ventricular (LV) hypertrophy (LVH), increased myocardial after-load, and altered coronary perfusion [16]. Previous studies in adults with CKD have shown that they are at an increased risk of developing atherosclerosis and arterial stiffening [5, 20, 28, 40]. Despite the fact that children may be at a similar risk of developing arterial disease, limited data are available on the arterial stiffness of children with CKD, especially in those not yet on dialysis.

The biophysical properties of the aorta (AO) and arterial stiffness can be assessed using a number of methods, both invasive and noninvasive, each with a unique set of limitations. Noninvasive methods rely on the assessment of either regional or systemic indices of arterial stiffness. These include measures of aortic/carotid distensibility, characteristic impedance (Zc), input impedance (Zi), elastic pressure-strain modulus (Ep), arterial wall stiffness index (β-index), and pulse wave velocity (PWV) [42]. PWV, defined as the speed with which a pulsatile blood wave travels along a length of artery, is a simple, reproducible, load-independent measure of arterial health [17]. It is considered the best measure of arterial health and the earliest predictor of cardiovascular risk [10, 22, 36].

Different methods have been used to measure PWV in both clinical and epidemiological settings. The carotid–femoral PWV incorporates both the elastic and muscular properties of the arterial system by measuring various segments of the arterial pathway starting from the more proximal elastic carotid and thoracic AO to the more distal muscular abdominal AO and femoral artery [9, 15, 17, 24]. Limitations of this method include inaccurate measurement of the traveling distance of the pulse wave and the difficulty in obtaining the carotid and femoral pulse waves [33]. A simpler echocardiography (echo)-Doppler technique has been developed and used to assess the biophysical properties of the AO in adult renal disease patients and several pediatric cohorts with other types of disease [1, 2, 11, 12, 26, 30]. The purpose of this study was to assess the biophysical properties of the AO in children with CKD using an echo-Doppler PWV technique.

Materials and Methods

Study Design

This was a cross-sectional study performed in the Children’s Heart Centre at British Columbia Children’s Hospital, Vancouver, Canada. We studied 24 children (18 male) with CKD (stages 2 through 5) who were part of an ongoing National Institutes of Health (NIH)—sponsored multicentre prospective cohort study [the Chronic Kidney Disease in Children (CKiD) study]. No patients were on dialysis at the time of the study. Comprehensive echocardiographic and vascular assessment and blood work were performed on the same day as the renal clinic visit.

The glomerular filtration rate (GFR) was measured using a previously validated, nonradioactive iohexol plasma clearance method per the NIH-sponsored CKiD study [31, 32]. Blood work was performed and reported from the central CKiD study laboratory. For the purposes of this study, and based on American Academy of Pediatrics guidelines [7, 34], lipid profiles were classified into three categories: (1) acceptable (<75th percentile, e.g., triglycerides [TG], TC < 170 mg dL−1, low-density lipoprotein cholesterol [LDL-C] < 110 mg dL−1); (2) borderline (75th to 95th percentile, e.g., TC = 170–199 mg dL−1, LDL-C = 110–130 mg dL−1); and (3) increased (>95th percentile, e.g., TC >200 mg dL−1, LDL-C > 130 mg dL−1). The control group consisted of 48 children (21 male). These data have previously been published [30]. Ethics approval for this study was obtained from the Children’s and Women’s Hospitals’ Research Ethics Board and the University of British Columbia’s Clinical Research Ethics Board as part of the CKiD study.

Echocardiography Technique

All echocardiographic measurements were performed by one experienced echocardiographer (M. T. P.) who was blinded to the severity of the renal disease in the CKD patients. Measurements of the biophysical properties of the AO were independently performed by two physicians (M. A. G., G. G. S. S.) on two separate occasions and the measurements averaged. Intraobserver and short- and long-term variability for this technique has been previously reported for our laboratory [1]. Both physicians were blinded to the severity of the renal disease in the CKD patients.

Standard M-mode, two-dimensional (2D) echo, and Doppler with simultaneous blood pressure (sphygmomanometer using phase I [systolic blood pressure; BPsys] and phase V [diastolic blood pressure; BPdia] Korotkoff’s sounds were performed. Standard measures of LV function were performed [14]. The annulus of the AO was measured at the valve leaflets using 2D echocardiography. The technique for obtaining echo-Doppler PWV has been previously described [30]. In brief, from the M-mode of the ascending AO, the AO dimension in end-diastole and the peak dimension in systole (AOS) were measured using the inner edge method, keeping the ascending AO as close to a right angle as possible (Fig. 1a). An ascending AO Doppler tracing was recorded at a speed of 100 mm s−1 with the sample volume in the center of the AO at the valve leaflets. The sample volume was then placed as distally as possible in the centre of descending AO, and a Doppler tracing was recorded at a speed of 100 mm s−1. These two recordings were performed within 10 s of each other, thus eliminating the confounding effect of a change in heart rate (HR). From the same suprasternal window, the AO arch length (AOL) was measured between these two points (Fig. 1b). The time from the onset of the QRS to the onset of the ascending (Fig. 1c) and descending (Fig. 1d) AO Doppler envelope was measured.
https://static-content.springer.com/image/art%3A10.1007%2Fs00246-013-0632-5/MediaObjects/246_2013_632_Fig1_HTML.jpg
Fig. 1

a M-mode of ascending AO. Dd maximum dimension taken at the end of diastole (i.e. correspond to R wave on the surface electrocardiogram). Ds minimum dimension taken at the end of systole. Methods used to measure echo-Doppler PWV. b AOL measured between the Doppler sampling area. c Pulse wave Doppler of the ascending AO. T1 = time from QRS to onset of the ascending AO Doppler envelope. d Pulse wave Doppler of the descending AO. T2 = time from QRS to onset of the descending AO Doppler envelope. TT was calculated as (T2 − T1). PWV was measured as (AOL/TT). Reprinted with permission from Bradley et al. [1]

Calculations

Calculations for anthropometric and echocardiographic measurements and the biophysical properties of the AO are listed in Table 1.
Table 1

List of formulae for calculations

Variable

Formula

AOcsa

π·[(AO annulus/2)2] (cm2, π: constant = 3.14)

β-index

In(BPsys/BPdia)/[(AOS − AOD)/AOD]

BMI

Weight (kg)/height (m)2 (kg m−2)

BSA

0.20247·height (m)0.725·weight (kg)0.425 (m2)

Zc

[(PWV·ρ)/AOcsa]; (dyne s cm−5); ρ: blood density = 1.06 g cm−5, 1 dyne = 1 g cm−2 s−1)

ETc

\( {\text{ET}}/\sqrt {{\text{R - R}};} ({\text{s}}) \)

Ep

(BPsys − BPdia)/[(AOS − AOD)/AOD] (mmHg)

Zi

[(BPsys − BPdia)/PkAOFlow]; (dyne s cm−5); (1 mmHg = 1,333 dyne cm−2)

LVMassi

{[(LVEDi + (2·PWDi)]3 − (LVEDi)3}/BSA·1.05 (g m−2)

MVCFc

SF/ETc (circ s−1)

Peak aortic flow

[PkAOV·AOcsa] (cm3 s−1)

σPS

(1.36·BPsys·LVESi)/[4·(PWSi·(1 + PWSi/LVESi)] (g cm−2)

PP

BPsys − BPdia (mmHg)

PWV

AOL/TT; (cm s−1)

SF

(LVEDi − LVESi)/LVEDi (%)

TT

[T2 − T1]; T2 (s) = time from onset of the QRS to onset of descending AO Doppler envelope; T1 (s) = time from onset of QRS to onset of ascending AO Doppler envelope

AOcsa AO cross-sectional area, AOD AO diameter in diastole, AOS AO diameter in systole, LVEDi LV end-diastolic dimension indexed to BSA; constant 3.14, ρ (rho) blood density = 1.06 g cm−5, 1 dyne = 1 g cm−2 s−1

Statistical Analysis

An a priori power analysis found that 22 patients were needed to find a difference of 50 cm s−1 in PWV given an α = 0.05, 1−β = 0.80 and a 1:2 case–control ratio. Frequency distributions were generated to determine if data were normally distributed. A univariate procedure was used to describe the data. Medians and ranges were calculated for all continuous data. A nonparametric Mann–Whitney U test was used to determine statistical differences between groups, and p < 0.05 was considered statistically significant. All statistical analysis was performed using SPSS 16.0 Statistical Software (SPSS, Chicago, IL).

Results

CKD Patient Characteristics

Of the 24 CKD patients enrolled in this study, 3 of 24 (12 %) had glomerular kidney disease, whereas 21 of 24 (88 %) had nonglomerular disease. Ten of 24 (42 %) CKD patients were being treated with angiotensin-converting enzyme inhibitors, and 4 of 24 (16.7 %) were receiving calcium-channel blockers for treatment of their hypertension at the time of the study.

Physical Characteristics

Table 2 lists the demographics of the CKD patients and the controls. Groups were comparable in age, height, weight, body surface area (BSA), and body mass index (BMI). HR was similar in both groups. Systolic blood pressure and pulse pressure (PP) were significantly greater in the CKD patients compared with controls [114 vs. 110 mmHg (p < 0.04) and 51 vs. 40 mmHg (p < 0.001), respectively].
Table 2

Subject demographics

Demographics

CKD (n = 24)

Controls (n = 48)

p

Age (y)

13.9 (5.9–17.5)

14.0 (7.4–19.7)

NS

Height (cm)

156.8 (106.6–188.0)

161.0 (119.2–188.7)

NS

Male/female

18/6

21/27

Weight (kg)

47.7 (17.2–94.9)

52.2 (22.0–94.7)

NS

BSA (m2)

1.43 (0.71–2.18)

1.56 (0.86–2.14)

NS

BMI (kg m−2)

19.2 (14.7–28.0)

20.6 (14.4–36.1)

NS

BPsys (mmHg)

114 (96–147)

110 (90–142)

<0.04

BPdia (mmHg)

62 (40–93)

64 (54–90)

NS

PP (mmHg)

51 (36–68)

40 (26–69)

<0.001

HR (bpm)

69 (56–102)

71 (46–110)

NS

Data are reported as median (range)

NS not statistically significant

Ventricular Size and Function

Table 3 lists the LV function measurements of both groups. CKD patients had greater shortening fraction (SF) than controls as well as greater interventricular septal thickness in diastole indexed to BSA (IVSDi), posterior wall thickness in diastole indexed to BSA (PWDi), posterior wall thickness in systole indexed to BSA (PWSi), and lower LV peak systolic wall stress (σPS). LV mass indexed to BSA was similar between groups. The mean velocity of circumferential fiber shortening corrected for HR (MVCFc) and cardiac index were similar in both groups as was the diameter of the AO annulus.
Table 3

LV size and function

Parameters

CKD (n = 24)

Controls (n = 48)

p

AO annulus (cm)

1.9 (1.2–2.1)

1.9 (1.5–2.4)

NS

LVEDi (cm)

3.2 (2.2–4.8)

3.0 (2.4–4.8)

NS

LVESi (cm)

1.8 (1.2–2.8)

1.9 (1.5–3.2)

NS

SF (%)

40 (25–47)

36 (31–46)

<0.001

MVCFc (circ s−1)

1.3 (1.0–1.5)

1.2 (1.0–1.6)

NS

σPS (g cm−2)

47 (34–76)

64 (36–96)

<0.001

PWDi (cm)

0.59 (0.34–0.85)

0.50 (0.35–0.86)

<0.001

PWSi (cm)

0.93 (0.69 –1.41)

0.81 (0.57–1.11)

0.002

IVSDi (cm)

0.60 (0.41–0.81)

0.49 (0.33 –0.65)

0.001

LVMassi (g m−2)

96 (64–158)

93 (34–144)

NS

CI (L min−1 m−2)

2.6 (1.6–4.6)

2.9 (2.1–5.4)

NS

Data are reported as median (range)

CI Cardiac Index, LVEDi LV end-diastolic dimension indexed to BSA, NS not statistically significant

Biophysical Properties of the Aorta

The indices of arterial stiffness are listed in Table 4. PWV (358 vs. 344 cm s−1; p > 0.05) and β-index were similar between groups; however, Zi (182 vs. 131 dyne s cm−5; p < 0.001), Zc (146 vs. 138 dyne s cm−5; p < 0.05), and Ep (280 vs. 230 mmHg; p < 0.02) were significantly greater in CKD patients.
Table 4

Biophysical properties of the AO

Properties

CKD (n = 24)

Controls (n = 48)

p

PWV (cm s−1)

358 (248–560)

344 (260–580)

NS

Zi (dyne s cm−5)

182 (134–347)

131 (80–206)

<0.001

Zc (dyne s cm−5)

146 (84–272)

138 (72–248)

<0.05

Ep (mmHg)

280 (115–477)

230 (136–362)

<0.02

β-index

2.98 (1.51–5.99)

2.67 (1.43–4.06)

NS

Data are reported as median (range)

NS not statistically significant

GFR

GFR results for the CKD patients are listed in Table 5. A total of 5 patients had CKD stages 4 or 5, whereas the remainder had measured GFRs that placed them in disease-stage categories 2 and 3. No patient was on dialysis at the time of the echocardiographic studies.
Table 5

Classification of disease and laboratory results for CKD patients

CKD patient no.

Disease type

CKD stage

GFR ml min−1  1.73 m−2

BUN mg dL−1

SCr mg dL−1

Ca-T mg dL−1

PO4 mg dL−1

PTH pg mL−1 (N: 1.6-9.3)

TC mg dL−1

LDL-C mg dL−1

TG mg dL−1

Hb

1

Glomerular

4

16

61

5.8

8.3

6.4

19.4a

254a

157a

202

11.8

2

Nonglomerular

4

24

48

3.2

9.2

4.8

7.4b

158b

86b

116

11.7

3

Nonglomerular

3

30

39

2.3

10.2

4.6

13.4a

122b

56b

104

12.6

4

Nonglomerular

2

79

24

0.8

9.3

4.9

6.4b

147b

67b

67

13.0

5

Nonglomerular

4

25

79

2.9

9.3

5.3

18.0a

100b

51b

104

10.4

6

Nonglomerular

2

68

15

1.1

9.9

4.0

4.7b

175c

116c

129

15.7

7

Nonglomerular

2

71

22

0.9

9.9

4.0

3.6b

186c

115c

95

15.7

8

Nonglomerular

3

52

27

1.2

9.3

4.5

7.9b

211

124c

129

12.2

9

Glomerular

5

10

65

9.1

6.7

7.2

38.4a

244a

144a

344

11.8

10

Nonglomerular

3

44

35

1.9

9.5

4.5

9.6b

170c

110c

75

12.5

11

Nonglomerular

2

63

18

1.3

9.2

5.0

3.6b

130b

68b

106

15.3

12

Glomerular

2

65

28

1.1

9.8

4.6

6.2b

175c

110c

125

15.5

13

Nonglomerular

3

59

25

1.3

9.2

3.6

3.6b

159b

87b

70

14.5

14

Nonglomerular

2

68

16

0.8

9.9

4.8

6.3b

203a

113c

200

14.0

15

Nonglomerular

2

75

17

0.6

9.6

4.2

2.8b

162b

104b

45

13.5

16

Nonglomerular

3

44

31

1.3

9.5

5.2

11.3a

263a

174a

206

13.8

17

Nonglomerular

3

38

56

3.5

9.5

5.2

2.9b

149b

96b

107

11.0

18

Nonglomerular

3

43

47

1.1

10.4

5.3

7.4b

198c

98b

106

12.7

19

Nonglomerular

2

63

22

1.0

9.4

3.8

8.6b

160b

93b

122

15.1

20

Nonglomerular

4

25

37

1.8

10.1

4.8

17.7a

207a

127c

125

12.4

21

Nonglomerular

3

51

27

1.7

9.5

4.1

7.5b

177c

104b

127

12.5

22

Nonglomerular

3

44

23

1.0

9.6

4.3

11.4a

196c

81b

94

11.7

23

Nonglomerular

3

56

22

0.9

9.9

4.2

7.2b

165b

85b

82

13.6

24

Nonglomerular

2

77

16

1.1

10.1

4.7

5.4b

174c

103b

108

16.8

BUN blood urea nitrogen, Ca-T total calcium

aIncreased

bNormal

cBorderline

Lipid Profiles

Table 5 lists the total cholesterol (TC), LDL-C, and hemoglobin (Hb) values for the CKD group. No similar data were available for the controls. Six of 24 (25 %) CKD patients had increased, 9 of 24 (37.5 %) had borderline, and 10 of 24 (41.5 %) had acceptable TC values. Increased LDL-C levels were seen in 3 of 24 (12.5 %) patients, whereas borderline levels were present in 7 of 24 (29.2 %). The remainder of CKD patients had acceptable LDL-C levels. There was no difference in PWV when we compared those with increased, borderline, or normal LDL-C levels, GFR, or the use of any specific blood pressure medications.

Discussion

Arterial stiffening has been shown to be an important and independent risk factor for future cardiovascular events [38, 42]. Among the different parameters used to assess arterial stiffening, PWV is considered the best measure and the earliest predictor of cardiovascular risk [10, 22, 36]. In adults with CKD, assessment of arterial stiffness using both the carotid–femoral PWV and echo-Doppler PWV methods has found stiffer central arteries as reflected by an increased PWV [5, 20, 28, 40]. Wang et al. [37] showed a stepwise increase in carotid-femoral PWV in adults as kidney function worsens. Patrianakos et al. [26], using a similar method to ours, evaluated echo-Doppler PWV in 71 adult patients with end-stage renal disease (ESRD) undergoing hemodialysis. The echo-Doppler PWV was faster in the ESRD group, suggesting that these patients have impaired central aortic function, a finding that is consistent with the results found in pediatric patients with ESRD on dialysis [6, 37].

To date, few studies have evaluated carotid-femoral PWV in a pediatric CKD population. These have been limited to children postrenal transplant or on dialysis at the time of the studies and showed that carotid–femoral PWV was significantly greater in both groups compared with controls [3, 6, 13]. In this study of children with non–dialysis dependent CKD, we measured echo-Doppler PWV and showed that PWV was similar to that of age-matched healthy controls. The difference between their study and ours is that our patients were younger, and none were on dialysis at the time of the studies. Our results and those reported in previous pediatric studies suggest that factors associated with arterial stiffening include age, worsening kidney function, and ESRD [3, 6, 13].

Other indices of arterial stiffness (Zi and Zc) were increased in the CKD patients. These are load-dependent measures. Increased Zi suggests that the LV has an increased ejection load, whereas increased Zc represents increased “resistance to ejection” in a pulsatile arterial system and may be caused by altered hemodynamic factors. Ep was also greater in the CKD group, suggestive of a stiffer central AO, but Ep is also a load-dependent measure; β-index, which is less load-dependent, was similar. These findings, taken in context with a normal PWV, suggest that the arterial changes at less advanced stages of CKD may be reversible.

LVH is a well-known complication in patients with CKD [18, 25]. In our study, LV mass was also similar between groups, although the CKD group had thicker IVSDi, PWDi, and PWSi. The exact mechanism responsible for LVH in CKD patients is not well understood, but it is likely to be multifactorial. Hypertension has been linked to the development of LVH in renal failure patients, but LVH may occur without hypertension [19]. The renin–angiotensin–aldosterone system is more likely to play a role in the development of LVH [39]. A recent randomized control trial showed a beneficial decrease of LVH and arterial stiffness in 112 adults with stages 2 to 3 CKD on established treatment with angiotensin-converting enzyme (ACE) inhibitors or with angiotensin-receptor blockers who were in the aldosterone-antagonist treatment arm of the study [8]. In their systematic review and meta-analysis, Mallareddy et al. [21] concluded that ACE inhibitors have a modest beneficial effects on arterial stiffness. Our study was not designed to answer this question.

When we stratified our CKD group according to their lipid profiles or GFR, no differences were seen in PWV. Children with CKD are considered at greater risk of dyslipidemia, which, in addition to other risk factors, may trigger the development of accelerated atherosclerosis seen in adults with CKD [27, 29]. Atherosclerotic changes are suspected to begin during childhood [4, 41]. Consistent with our study results, Mitsnefes et al. showed that carotid artery intima-media thickness (cIMT) increased in children with CKD even in mild to moderate stages of CKD [23]. However, in their study, arterial stiffness (as measured by distensibility and β-index) was abnormal only in patients on dialysis, and no change in arterial stiffness was observed in CKD patients before they began dialysis, which is consistent with our findings [23].

Limitations

Due to the small sample size of our study group, it is difficult to generalize our findings to all non–dialysis dependent pediatric CKD patients. The CKD group was heterogeneous regarding their primary diagnosis, duration of illness, current stage of CKD, medical therapy, and laboratory abnormalities. In addition, Tanner staging was not performed, and we cannot exclude the possibility that different stages of sexual maturity might have affected our results. Our patients were enrolled as a part of a larger ongoing observational study; therefore, stratification of these factors was not possible. A larger study of CKD patients with more severe degrees of renal dysfunction would help answer these clinical questions.

Various technical considerations must be considered when measuring PWV using this echo-Doppler method in a clinical setting. Using current echo technology, measurement of curved structures, such as the AO arch, is not feasible. We summed the serial distance measurements around the AO arch to minimize this problem. Varying velocity profiles within an artery remain a problem. We avoided this by ensuring that the sample volume was as close to the center of the artery as possible and used the onset of flow rather than the peak velocity for timing. Finally, the transit time (TT) of the central pulse wave is rapid; therefore, a 5-ms error may potentially make a significant difference in each patient’s results. We accounted for these issues by having one experienced echocardiographer perform all studies, by having two physicians independently measure variables to minimize intraobserver variability, and by averaging a large number (6–10) of samples for each patient. We previously showed that interobserver variability for echo-Doppler PWV ranges from 4 % for PWV to 12 % for EP, whereas measurements taken 2–4 weeks apart may vary from 2 % for PWV to 16 % for EP [1].

Conclusion

The results of our study showed that central aortic arterial stiffness is normal in children with less severe forms of CKD. Because adults with CKD are known to develop abnormal arterial function, early identification, careful follow-up, and possible intervention at a younger age may decrease adverse cardiovascular events associated with this condition in later life.

Acknowledgments

The CKiD Study is funded by the National Institute of Diabetes and Digestive and Kidney Diseases with additional funding from the National Institute of Neurologic Disorders and Stroke; the National Institute of Child Health and Human Development; and the National Heart, Lung, and Blood Institute (Grants No. UO1-DK-66143, UO1-DK-66174, UO1-DK-66116, and U01-DK-082194). GE Health Care, Amersham Division, provided iohexol for the GFR measurements. The CKID web site is located at http://www.statepi.jhsph.edu/ckid. We thank Josephine Chow for invaluable assistance in coordinating all of the CKiD investigations locally.

Conflict of interest

None.

Copyright information

© Springer Science+Business Media New York 2013