Longitudinal assessment of myocardial function in childhood chronic kidney disease, during dialysis, and following kidney transplantation

  • Rawan K. Rumman
  • Ronand Ramroop
  • Rahul Chanchlani
  • Mikaeel Ghany
  • Diane Hebert
  • Elizabeth A. Harvey
  • Rulan S. Parekh
  • Luc Mertens
  • Michael Grattan
Original Article

DOI: 10.1007/s00467-017-3622-7

Cite this article as:
Rumman, R.K., Ramroop, R., Chanchlani, R. et al. Pediatr Nephrol (2017). doi:10.1007/s00467-017-3622-7

Abstract

Background

Childhood chronic kidney disease (CKD) and dialysis are associated with increased long-term cardiovascular risk. We examined subclinical alterations in myocardial mechanics longitudinally in children with CKD, during dialysis, and following renal transplantation.

Methods

Forty-eight children with CKD (stage III or higher) who received kidney transplants from 2008 to 2014 were included in a retrospective study and compared to 192 age- and sex-matched healthy children. Measurements of cardiac systolic and diastolic function were performed, and global longitudinal strain (GLS) and circumferential strain (GCS) were measured by speckle-tracking echocardiography at CKD, during dialysis, and 1 year following kidney transplantation. Mixed-effects modeling examined changes in GLS and GCS over different disease stages.

Results

Children with CKD had a mean age of 10 ± 5 years and 67% were male. Eighteen children received preemptive transplantation. Children with CKD had increased left ventricular mass, lower GLS, and impaired diastolic function (lower E/A ratio and E′ velocities) than healthy children. Changes in left ventricular diastolic parameters persisted during dialysis and after renal transplantation. Dialysis was associated with reduced GLS compared to CKD (β = 1.6, 95% confidence interval 0.2–3.0); however, this was not significant after adjustment for systolic blood pressure and CKD duration. Post-transplantation GLS levels were similar to those at CKD assessment. GCS was unchanged during dialysis but significantly improved following transplantation.

Conclusions

There are differences in diastolic parameters in childhood CKD that persist during dialysis and after transplantation. Systolic parameters are preserved, with significant improvement in systolic myocardial deformation following transplantation. The impact of persistent diastolic changes on long-term outcomes requires further investigation.

Keywords

Chronic kidney disease Dialysis Kidney transplant Myocardial mechanics Systolic strain Children 

Introduction

Children with chronic kidney disease (CKD) have a significantly higher risk of cardiovascular morbidity than healthy children of the same age, with the highest cardiovascular risk in those on dialysis [1, 2]. The pathogenesis of cardiovascular disease in CKD and dialysis patients is complex as it is attributable to both cardiovascular as well as kidney-specific risk factors. Kidney transplantation positively modifies some of the risk factors for cardiomyopathy, such as volume and pressure overload combined with uremia [3], thereby attenuating cardiovascular risk; however transplanted children continue to experience a high prevalence of cardiovascular morbidity [4, 5]. Therefore, a better understanding of the effect of kidney disease and the effect of treatment on cardiac function may facilitate the prevention and management of cardiovascular disease in children with CKD.

In both children and adults with CKD, left ventricular hypertrophy (LVH) and subclinical diastolic impairment are present even at early stages of CKD, and both conditions will progress during dialysis treatment [6, 7, 8, 9]. While systolic dysfunction is noted in adults with CKD, initial studies using conventional echocardiographic parameters in children report preserved systolic function [2, 10]. Recent studies in adult CKD patients have suggested that myocardial strain, measured using tissue Doppler imaging (TDI) or two-dimensional speckle tracking echocardiography (STE), can be useful for detecting early subclinical ventricular abnormalities [11, 12, 13, 14, 15, 16, 17]. Myocardial strain is the percentage change in length of a myocardial segment in systole, where longitudinal and circumferential shortening during myocardial contraction are represented by negative strain values. Systolic strain is considered to be a reliable predictor of prognosis [13, 18, 19] and provides a more accurate method of detecting early subclinical LV dysfunction compared to conventional measurements of systolic function [20].

Data relating to the use of strain imaging in children with CKD and dialysis are sparse, likely due to lack of normal strain values available for comparison in the pediatric group. Few cross-sectional studies have evaluated myocardial strain in pediatric patients with CKD and receiving dialysis [21, 22]. However, due to lack of serial data in children, it is not known whether kidney transplantation improves myocardial function. Thus, we sought to provide a comprehensive assessment of ventricular function in children with CKD compared to healthy controls, and to assess the change in these functional parameters longitudinally after the initiation of dialysis and following kidney transplantation. We hypothesized that STE would enable the identification of subclinical abnormalities in systolic function in patients with CKD and that these abnormalities would worsen with dialysis and improve after kidney transplantation.

Material and methods

Study population and inclusion criteria

A total of 48 children with CKD who underwent kidney transplantation at the Hospital for Sick Children (Toronto, Canada) from 2008 to 2014 were included in this retrospective cohort study. Inclusion criteria were a diagnosis of CKD stage III or higher {estimated glomerular filtration rate (GFR) <60 ml/min/1.73 m2 using the Schwartz IDMS creatinine GFR formula [23]}, age <18 years throughout the study period, and available echocardiographic assessments prior to dialysis or kidney transplantation, as well as 6–18 months following first kidney transplantation. Exclusion criteria included multiple organ transplants, graft loss, cardiomyopathy other than hypertensive cardiomyopathy, and lack of sufficient echocardiographic data.

CKD patients were compared to 192 age- and sex-matched controls (using 1:4 matching), who were selected from our control database. Control patients were healthy, with no known cardiovascular risk factors (including hypertension), no hemodynamically significant cardiovascular abnormalities, no active diseases with the potential to interfere with cardiac function, and no cardiac or vasoactive medications. The study protocol was approved by the Research Ethics Board at the Hospital for Sick Children.

Echocardiographic assessment

Measurements were made from three echocardiograms per patient, performed at different disease stages (CKD, dialysis, and 6–18 months after kidney transplantation). If more than one study was available during a particular stage, the most complete study was chosen for analysis. All echocardiograms were performed using a GE Vivid 7 or Vivid E9 system (General Electric Medical Systems, Waukesha, WI), using a standardized comprehensive functional echocardiographic protocol as previously described [24]. Brachial blood pressure was measured non-invasively in the echocardiography laboratory using an arm size-appropriate cuff as the average of three readings with an automated Dinamap sphygmomanometer (Critikon, Tampa, FL). Images were stored in RAW data format, and all measurements were performed offline using EchoPAC version 110.1.3 (General Electric Medical Systems). All measurements were performed prospectively by a single reader (RR). Measurements of ventricular systolic function (shortening fraction, biplane Simpson method ejection fraction, tissue Doppler S′ wave) and diastolic function [mitral valve (MV) and pulmonary vein Doppler, myocardial tissue Doppler imaging, and left atrial volume] were performed [25]. LV mass was calculated using Devereux’s method. Given the ongoing controversy regarding LV mass allometry in the pediatric population, LV mass was adjusted for body surface area, adjusted for height2.7 and converted to the Boston Z score [26] prior to statistical analysis.

Measurement of myocardial strain was performed by STE as previously described [25]. Briefly, the endocardial surface of the left ventricle was manually traced and the algorithm for longitudinal or circumferential strain applied. Myocardial tracking was automatically performed and the tracking was also visually inspected to ensure it was appropriate. In cases where tracking was inadequate, the endocardial surface of the ventricle was retraced. If tracking was still inadequate, the particular segment was excluded. If more than two segments needed to be excluded, the image was considered inadequate and excluded from the analysis. Longitudinal strain was measured from the apical two-, three- and four-chamber views, and global longitudinal strain was calculated as the mean of the measurements from these three views. Circumferential strain was measured from parasternal short axis images at the basal, mid-ventricular, and apical levels, and global circumferential strain was calculated as the mean of the strain values from these three levels. Intraobserver and interobserver variability of the STE-derived longitudinal and circumferential strain measurements was evaluated in 20 randomly selected healthy subjects. Intraobserver and interobserver variability was also assessed in the diseased population in 15 randomly selected patients at the post-transplantation time point.

Renal assessment

Medical records were reviewed for etiology of renal disease, type of dialysis, and duration of dialysis. Clinical and laboratory data were collected on the day of each of the three echocardiographic assessments, including blood pressure, height, weight, body mass index (BMI), serum creatinine, albumin, and hemoglobin. Standard deviation scores (Z score) for systolic and diastolic blood pressures were derived from the Fourth Task Force Report , and BMI Z scores were derived from the Centre for Disease and Control growth charts [27, 28]. Systolic hypertension was defined as systolic blood pressure of ≥95th percentile for patients of the same age, sex, and height [27].

Statistical analysis

Categorical data are reported as frequency and percentage. Continuous variables are expressed as the mean ± standard deviation or median and interquartile range [IQR] depending on the distribution of the continuous variable. Student t tests were used to assess the differences between children with CKD and matched healthy controls. For differences between the different disease stages (CKD, dialysis, and post-transplantation) in children who received dialysis, a repeated-measures analysis of variance test was performed to correct for the lack of independence of the observations. Differences in frequencies were assessed using a chi-square or Fischer exact test, as appropriate. Linear mixed-effects models (with random intercepts and slopes and unstructured covariance) were used to assess the change in global longitudinal strain over the different disease stages (CKD, dialysis, post-transplantation) to account for serially correlated data within an individual. Potential confounders were chosen a priori based on previously published work and the biology of disease. Confounders assessed included systolic blood pressure and BMI Z scores. As the time from CKD diagnosis to echocardiographic assessment varied between patients, the duration of CKD was added to the multivariable model. The Akaike information criterion and likelihood ratio test were used to assess the model fit. A two-tailed p value of 0.05 was considered to be statistically significant. All statistical analyses were performed using Stata 13.0 (StataCorp LP, College Station, TX).

Results

Patient characteristics

Clinical characteristics of the study cohort are summarized in Table 1. Mean age at the baseline CKD III evaluation was 10.2 ± 5.2 years, and 67% were male. Median time from diagnosis of CKD III to baseline echocardiographic exam was 1.1 years [IQR 0.4–3.3]. The primary etiology of CKD included urinary tract anomalies (renal dysplasia, posterior urethral valves, and vesicoureteral reflux) in 25 (52%) patients, glomerular disorders in 13 (27%) patients, and genetic disorders (tuberous sclerosis, cystinosis, and autosomal dominant polycystic kidney disease) in ten (21%) patients. At the time of CKD assessment, 6% of children had CKD stage III, 27% had CKD stage IV, and 67% had CKD stage V. Patients with significant congenital heart disease were excluded from the study. Of the 48 included patients, one patient had a bicuspid aortic valve with no valve dysfunction (no aortic stenosis or aortic insufficiency), and one patient had a very small muscular ventricular septal defect of no hemodynamic significance. No patient had significant aortic stenosis or aortic insufficiency at the time of the baseline echocardiographic evaluation. A total of 18 patients had preemptive kidney transplantation, and the remaining 30 underwent dialysis prior to transplantation. This consisted of hemodialysis in 20 (67%) patients and peritoneal dialysis in ten (33%). Of the patients receiving hemodialysis, the echocardiogram was performed immediately following the dialysis session in 45%, immediately prior to dialysis in 35%, and on a non-dialysis day in 20% of children. Mean time between the dialysis echocardiogram and the last dialysis session for children imaged on non-dialysis days was 2 ± 1 days. Mean age at the time of dialysis evaluation was 11.3 ± 5.7 years. The median duration of dialysis was 0.9 [IQR 0.4–1.7] years, and the median time between dialysis initiation and the second echocardiographic assessment was 0.7 [IQR 0.2–1.1] years. All 48 patients received a kidney transplant, 21 (44%) from living related donors, and 27 (56%) from deceased donors. Mean age was 11.5 ± 5.1 years at time of renal transplantation. Median time from transplantation to post transplantation assessment was 1.1 [IQR 0.8–1.3] years.
Table 1

Clinical characteristics of children with chronic kidney disease at the time of echocardiographic imaging

Variable

Healthy controls (n = 192)

CKD (n = 48)

Dialysis prior to renal transplantation (n = 30)a

Post transplantation (n = 48)

pb

pc

Age at imaging (years)

10.1 ± 5.2

10.2 ± 5.2

11.3 ± 5.7

12.1 ± 4.7

0.9

0.2

Male sex

116 (60.0%)

32 (67.0%)

20 (67.0%)

32 (67.0%)

0.8

0.9

Body surface area ( m2)

1.2 ± 0.5

1.1 ± 0.5

1.2 ± 0.5

1.3 ± 0.4

0.3

0.5

Body mass index ( Z score)

0.2 ± 1.1

0.3 ± 1.1

0.3 ± 1.3

0.5 ± 1.2

0.5

0.4

Systolic blood pressure ( Z score)

0.01 ± 0.8

1.1 ± 1.4

1.3 ± 1.3

0.1 ± 1.2

<0.001

<0.001

Diastolic blood pressure ( Z score)

−0.3 ± 0.9

0.6 ± 1.0

0.9 ± 0.9

−0.1 ± 0.9

<0.001

<0.001

Hypertension

0 (0%)

23 (47.9%)

14 (46.7%)

17 (35.4%)

<0.001

Medications

    Antihypertensives

20 (41.7%)

20 (66.7%)

26 (54.2%)

0.5

    Calcium channel blockers

15 (31.3%)

18 (60.0%)

17 (35.4%)

0.9

    ACEi

3 (6.3%)

8 (26.7%)

13 (27.1%)

0.05

    Beta-blockers

5 (10.4%)

3 (10.0%)

3 (6.3%)

0.6

    Alpha-blockers

5 (10.4%)

3 (10.0%)

2 (4.2%)

0.4

    Anticoagulation/antiplatelet

1 (2.1%)

3 (10.0%)

2 (4.2%)

0.8

Laboratory data

    Creatinine (μmol/L)

390 [254–510]

63 [53–86]

<0.001

    Hemoglobin (g/L)

107.1 ± 13.7

108.8 ± 14.3

124.5 ± 16.5

<0.001

    Albumin (g/L)

40 [33–44]

39 [35–43]

43 [40–46]

0.003

    Calcium (mmol/L)

2.3 ± 0.3

2.4 ± 0.3

2.4 ± 0.1

0.002

    Phosphate (mmol/L)

1.9 ± 0.4

1.9 ± 0.5

1.4 ± 0.3

<0.001

Values in table are presented as the mean ± standard deviation, the median with the interquartile range (IQR) in square brackets, or a number with the percentage in parenthesis

CKD, Chronic kidney disease; ACEi, angiotensin converting enzyme inhibitor

a18 patients did not receive dialysis

bp for comparison of CKD patients to matched controls

cp for comparison of CKD patients to those undergoing dialysis prior to renal transplantation (n = 30) using repeated-measures analysis of variance (ANOVA)

The BMI Z-score was similar between children with CKD and healthy controls and did not change significantly on dialysis or following kidney transplantation. Compared to matched controls, children with CKD had significantly higher systolic and diastolic blood pressure and blood pressure Z scores at the time of echocardiographic assessment. Blood pressures increased significantly on dialysis and decreased significantly following renal transplantation compared to pre-dialysis pressures. At the time of CKD evaluation, 48% of patients were hypertensive, and 42% were on antihypertensive medications. At the time of dialysis, 47% had hypertension, and 67% were receiving antihypertensive medications. Following kidney transplantation, 35% had hypertension, 54% were receiving antihypertensive therapy, and all were on immunosuppressive medications. Laboratory measurements at each of the three study time points, including creatinine and hemoglobin, are summarized in Table 1.

Cardiac geometry and systolic function

Atrial and ventricular geometry and LV systolic functional parameters are summarized in Table 2. LV posterior wall dimensions and LV mass Z score were significantly higher in the CKD group than in the age- and sex- matched controls. Chamber dimensions and interventricular septum thickness were similar in both groups. Dialysis was associated with a decrease in left atrial volume index, with no change in LV dimensions compared to the CKD group. After transplantation, there were no significant changes in chamber dimension or wall thickness compared to the CKD or dialysis group. Following transplantation, the LV mass Z score was slightly reduced, but it was not significantly different from levels in the CKD or dialysis groups (Fig. 1a). There were no significant differences in shortening fraction or ejection fraction between the CKD, dialysis, transplantation, or control groups. Systolic TDI velocities were lower at the level of the MV lateral annulus in children with CKD than in the healthy controls, but they were not significantly different at the septum or tricuspid valve lateral annulus (Table 3). There were no differences in systolic TDI velocities between the CKD, dialysis, and transplantation groups. There were no differences in LV geometry or systolic parameters between children receiving hemodialysis compared to those receiving peritoneal dialysis. CKD measurements did not differ between the subset of children receiving a pre-emptive transplant and the overall group; this was also the case for post-transplantation measurements. Measurements for the subgroup receiving a pre-emptive transplant are summarized in Electronic Supplementary Material (ESM) Table 2.
Table 2

Traditional measures of atrial and ventricular geometry and left ventricle systolic function in healthy controls and in children at three disease stages (CKD, receiving dialysis, and following kidney transplantation)

Variable

Healthy controls (n = 192)

CKD (n = 48)

Dialysis (n = 30)a

Post transplantation (n = 48)

pb

pc

LAD (cm)

2.7 ± 0.5

2.7 ± 0.7

2.8 ± 0.7

2.9 ± 0.5

0.6

0.3

RVEDD (cm)

1.7 ± 0.5

1.7 ± 0.6

1.8 ± 0.5

2.0 ± 0.6

0.6

0.04

IVSEDD (cm)

0.7 ± 0.2

0.7 ± 0.2

0.8 ± 0.2

0.7 ± 0.2

0.2

0.3

LVEDD (cm)

4.2 ± 0.7

4.2 ± 0.9

4.2 ± 0.9

4.3 ± 0.7

0.9

0.8

LVESD (cm)

2.6 ± 0.5

2.6 ± 0.7

2.8 ± 0.8

2.7 ± 0.6

0.7

0.6

LVPWD (cm)

0.6 ± 0.1

0.7 ± 0.2

0.7 ± 0.2

0.7 ± 0.2

0.002

0.8

LVM (g)

76.7 ± 38.2

88.6 ± 51.4

94.2 ± 49.2

90.8 ± 41.8

0.08

0.9

LVM Z score

−1.4 ± 1.4

0.1 ± 1.6

0.2 ± 1.8

−0.4 ± 1.4

<0.001

0.2

LVM adjusted for BSA (g/m2)

62.2 ± 12.9

79.2 ± 24.5

80.3 ± 29.1

70.6 ± 18.0

<0.001

0.1

LVM adjusted for height × 2.7 g/m2.7

32.1 ± 10.8

43.7 ± 17.3

40.0 ± 13.0

35.9 ± 10.9

<0.001

0.03

LA volume (mL)

36.1 ± 23.5

31.0 ± 14.7

34.1 ± 12.1

0.6

LA volume index (mL/m2)

31.1 ± 11.4

24.4 ± 8.0

26.0 ± 5.3

0.02

Shortening fraction (%)

37.6 ± 4.2

36.7 ± 5.8

34.9 ± 7.1

36.5 ± 7.3

0.3

0.4

Ejection fractio (%)

67.8 ± 7.5

66.6 ± 7.5

63.8 ± 10.1

65.9 ± 9.9

0.2

0.4

Values in table are presented as the mean ± SD

LAD, Left atrial (LA) dimension during systole; RVEDD, right ventricular (RV) end-diastolic dimension; IVSEDD, intraventricular septal end-diastolic dimension; LVEDD, left ventricular (LV) end-diastolic dimension; LVESD, LV end-systolic dimension; LVPWD, LV posterior wall dimension in diastole; LVM, LV mass, BSA, body surface area

a18 patients did not receive dialysis; CKD and post-dialysis echocardiographic measurements for this subset are compared in ESM Table 2

bp for comparison of CKD patients to matched controls

cp for comparison of CKD patients to those undergoing dialysis prior to renal transplantation (n = 30) using repeated-measures ANOVA; CKD and post-transplant values did not differ between children receiving preemptive transplant and the overall cohort

Fig. 1

Box plots of left ventricular mass Z score (a) and early mitral valve inflow velocity/late mitral valve inflow velocity (E/A ratio) (b) among age- and sex matched healthy children (Healthy Controls) and children with chronic kidney disease (CKD), receiving dialysis (Dialysis), and post-transplantation (Post-Transplant). P1 comparison of children at CKD to healthy controls, P2 comparison of children at CKD, dialysis, and post-transplantation

Table 3

Pulsed Doppler, pulsed-wave tissue Doppler, and myocardial mechanics in healthy controls and in children at three disease stages (CKD, receiving dialysis, and following kidney transplantation)

Variable

Healthy controls (n = 192)

CKD (n = 48)

Dialysis (n = 30)a

Post transplantation (n = 48)

pb

pc

Mitral inflow

   E-wave velocity (cm/s)

103.4 ± 17.9

104.1 ± 17.3

103.1 ± 23.4

101.9 ± 21.8

0.8

0.9

   A-wave velocity (cm/s)

47.9 ± 13.7

60.8 ± 20.3

69.1 ± 29.1

62.1 ± 18.3

<0.001

0.3

   E-wave deceleration time (ms)

139.3 ± 20.5

142.7 ± 37.3

149.0 ± 28.9

149.1 ± 30.9

0.4

0.6

   E/A ratio

2.3 ± 0.7

1.9 ± 0.8

1.7 ± 0.6

1.7 ± 0.5

<0.001

0.4

   A-wave duration (ms)

111.2 ± 20.8

125.3 ± 31.3

123.4 ± 28.1

128.3 ± 46.6

0.007

0.9

   IVRT (ms)

68.7 ± 10.0

70.4 ± 20.8

69.0 ± 13.6

69.2 ± 17.0

0.5

0.9

Pulmonary vein

   Systolic wave velocity (cm/s)

44.6 ± 11.0

53.4 ± 11.8

50.9 ± 13.0

49.5 ± 12.5

<0.001

0.3

   Diastolic wave velocity (cm/s)

60.7 ± 10.8

54.5 ± 12.7

55.4 ± 11.3

55.2 ± 11.2

0.001

0.9

   A-wave velocity (cm/s)

18.3 ± 5.4

23.0 ± 6.9

26.0 ± 11.9

20.5 ± 6.1

0.007

0.4

   A-wave duration (cm/s)

96.4 ± 21.5

113.4 ± 37.0

121.2 ± 34.5

93.2 ± 24.8

0.05

0.5

LV tissue Doppler

   Lateral S′ (cm/s)

10.1 ± 2.3

9.2 ± 2.3

9.4 ± 3.2

9.4 ± 2.4

0.01

0.9

   Lateral E′ (cm/s)

18.0 ± 3.3

15.5 ± 4.3

14.4 ± 4.3

14.7 ± 3.3

<0.001

0.4

   Lateral A′ (cm/s)

6.2 ± 1.5

7.6 ± 2.4

6.9 ± 3.3

6.9 ± 1.7

<0.001

0.3

   Lateral E/E′ ratio

5.92 ± 1.5

7.3 ± 5.9

7.1 ± 6.4

6.5 ± 4.1

0.004

0.5

   Septal S′ (cm/s)

8.43 ± 1.0

8.2 ± 1.4

8.20 ± 1.5

8.3 ± 1.4

0.2

0.9

   Septal E′ (cm/s)

14.7 ± 2.5

11.8 ± 3.0

11.8 ± 2.8

11.9 ± 2.9

<0.001

0.9

   Septal A′ (cm/s)

6.2 ± 1.3

7.2 ± 1.9

7.4 ± 2.4

7.1 ± 1.8

<0.001

0.8

   Septal E/E′ ratio

7.2 ± 1.5

9.7 ± 3.9

9.1 ± 2.5

9.2 ± 3.2

<0.001

0.7

RV tissue Doppler

   Lateral S′ (cm/s)

13.1 ± 20.

13.4 ± 3.4

12.8 ± 2.9

12.8 ± 2.5

0.5

0.7

   Lateral E′ (cm/s)

15.9 ± 2.8

14.6 ± 3.8

14.3 ± 4.1

13.6 ± 3.3

0.02

0.4

   Lateral A′ (cm/s)

8.9 ± 2.5

11.4 ± 3.5

9.9 ± 2.9

9.9 ± 2.7

<0.001

0.05

Myocardial mechanicsd

   Global longitudinal strain (%)

−20.2 ± 1.6

−21.4 ± 2.9

−19.7 ± 2.8

−21.1 ± 2.2

0.003

0.05

   Global circumferential strain (%)

−19.7 ± 1.4

−20.7 ± 5.0

−21.1 ± 3.9

−24.6 ± 5.9

0.03

0.01

Values in table are presented as the mean ± SD

E, Early mitral inflow velocity; A, late mitral inflow velocity; IVRT, isovolumic relaxation time; PV-A, pulmonary vein atrial reversal wave velocity; S′, systolic velocity with tissue Doppler imaging; E′, early diastolic myocardial velocity; A′, late diastolic myocardial velocity

a18 patients did not receive dialysis; CKD and post-dialysis echocardiographic measurements for this subset are compared in ESM Table 2

bp for comparison of CKD patients to matched controls

cp for comparison of CKD patients to those undergoing dialysis prior to renal transplantation (n = 30) using repeated-measures ANOVA; CKD and post-transplant values did not differ between children receiving preemptive transplant and the overall cohort

dStrain analysis was feasible in 60% of patients at the time of CKD (n = 29), 87% of patients at the time of dialysis (n = 26), and 88% of patients following transplantation(n = 42)

Diastolic function assessment

Pulsed Doppler and pulsed-wave tissue Doppler measurements are summarized in Table 3. Early mitral inflow velocity (E), mitral inflow deceleration time, and isovolumic relaxation time were not different between children with CKD and healthy controls. Children with CKD had higher mitral A-wave velocity (A) and A-wave duration and a reduced E/A ratio compared to controls (Fig. 1b). Pulmonary venous A-wave velocities were higher in children with CKD than in controls. The relationship between MV and pulmonary vein A-wave duration was similar in CKD patients and healthy controls. At the level of the MV lateral annulus, the septum, or tricuspid valve lateral annulus, tissue Doppler early diastolic myocardial velocities (E′′) were lower in the CKD group than in the control group, and late diastolic myocardial velocities (A′′) were higher in the CKD group than in the control group. E/E′ ratios were significantly higher in the CKD group than in the controls at MV lateral and septal annular levels. MV, pulmonary vein, and tissue Doppler velocities did not change significantly on dialysis or after transplantation compared to baseline CKD levels. Diastolic parameters did not differ by type of dialysis.

Strain measurements

Assessments of the global longitudinal strain (GLS) and circumferential strain (GCS) are summarized in Table 3. Analysis of strain was possible in 60% of CKD studies, 87% of dialysis studies, and 88% of post-transplantation studies. This variation is mainly reflective of the changes in storage of echocardiographic data in RAW (rather than DICOM) format over time. There were no significant differences between the subset of CKD patients analyzed for strain and the remaining study cohorts. The calculated intraobserver and interobserver coefficients of variation were 6% for GLS, and 4 and 9% for GCS in the healthy controls. In the diseased group, intraobserver and interobserver coefficients of variation were 2 and 7%, respectively, for GLS, and 3 and 5% for GCS. LV longitudinal strain and circumferential strains were higher in children with CKD than in controls. On dialysis, there was a decrease in longitudinal strain, while the circumferential strain remained unchanged from CKD levels. Longitudinal and circumferential strains did not differ by type of dialysis received. Longitudinal strain improved following transplant compared to dialysis, but was not significantly different from CKD levels (Fig. 2a). Circumferential strain measurements improved significantly from CKD levels (Fig. 2b).
Fig. 2

Box plots of global longitudinal strain (a), and global circumferential strain (b) among age- and sex-matched healthy children and children with chronic kidney disease (CKD), receiving dialysis, and post-transplantation. P1 Comparison of children at CKD to healthy controls, P2 comparison of children at CKD, dialysis, and post-transplant. See Fig. 1 caption for definition of CKD groups

The results of the linear mixed-effects model used to assess the longitudinal change in GLS and GCS are summarized in Table 4. Being on dialysis was associated with a decrease in average GLS compared to CKD values [β = 1.6, 95% confidence interval (CI) 0.2–3.0], whereas the post-transplantation GLS did not differ significantly from baseline CKD levels. After adjusting for systolic blood pressure Z score and duration of CKD, being on dialysis was no longer associated with reduced GLS compared to CKD levels (β = 0.9, 95% CI −0.5 to 2.4). Post-transplantation GLS was not significantly different from CKD levels. In the univariable and multivariable analyses, GCS during dialysis was not significantly different from CKD levels. Post-transplant GCS was significantly higher than baseline CKD levels (β = −2.4, 95% CI −4.4 to −0.4); however, this was no longer significant after adjusting for systolic blood pressures and duration of CKD.
Table 4

Linear mixed-effects model assessing the change in global longitudinal strain and global circumferential strain at the three disease stages of chronic kidney disease, dialysis treatment and post-transplantation

Disease stage

Univariable analysis

Multivariable analysis a

β

95% CI

p

β

95% CI

p

Global longitudinal strain

  CKD

Reference

Reference

Reference

Reference

Reference

Reference

   Dialysis

1.6

0.2–3.0

0.02

0.9

−0.5 to 2.4

0.2

   Post transplantation

0.3

−1.0 to 1.5

0.7

0.5

−0.8 to 1.7

0.5

Global circumferential strain

  CKD

Reference

Reference

Reference

Reference

Reference

Reference

   Dialysis

−0.5

−2.8 to 1.8

0.7

−0.04

−2.4 to 2.4

0.9

   Post transplantation

−2.4

−4.4 to −0.4

0.01

−1.0

−3.2 to 1.1

0.4

aMultivariable models adjusted for systolic blood pressure Z scores and duration of CKD

We recognize that children who received dialysis prior to transplantation may have had more severe kidney disease compared to those who did not receive dialysis; therefore, we examined the longitudinal change in myocardial deformation parameters among the subset of 30 children who received dialysis prior to transplant as a subgroup analysis (ESM Table 1). Dialysis was associated with a reduction in GLS (β = 2.0, 95% CI 0.4–3.6); however the association was no longer significant after adjusting for systolic blood pressure and duration of CKD. Post-transplantation GLS was not significantly different from CKD levels. In the univariable and multivariable analyses, GCS was not significantly different between the dialysis or post-transplantation groups compared to the CKD group.

Discussion

In this study we examined the changes in cardiac function and myocardial mechanics longitudinally using STE in children with CKD as they progress to dialysis and then after kidney transplantation. The results of this study demonstrate that changes in ventricular diastolic function develop early in the course of CKD and persist up to 18 months following kidney transplantation. Changes in ventricular longitudinal strain also occur in children on dialysis. The clinical significance of these changes in children requires further prospective investigation in larger pediatric cohorts.

Several pediatric CKD studies have reported preserved LV systolic function. While this may be the case for overt systolic functional abnormalities assessed by ejection and shortening fractions, subclinical systolic changes as detected by TDI or STE are already present in children at relatively early stages of CKD [21, 29]. In our cohort, patients with CKD had similar ejection and shortening fractions and similar strain measurements as the controls. However, children who underwent dialysis had lower GLS compared to CKD levels, possibly suggesting more severe disease in this subgroup of patients prior to dialysis, or an additional insult due to uremic factors attributable to the dialysis treatment itself. The observed changes in strain during dialysis were attenuated after adjustment for systolic blood pressures, which suggests that hypertension may account in part for the worsening strain values. This observation underscores the importance of adequate blood pressure control in children with CKD and on dialysis. Furthermore, children were on dialysis for only a short period of time (<1 year), which may account for smaller changes in strain values on dialysis compared to values reported in other studies [17]. Following kidney transplantation, longitudinal strain was restored to pre-dialysis levels, and there was a significant improvement in circumferential strain. This improvement in strain measurements following renal transplantation is promising and supports the strategy of avoiding long-term dialysis treatment in children.

Children with CKD had significantly higher LV mass compared to healthy controls, which could be due to hypertension [30] or uremic cardiomyopathy resulting in progressive change in myocardial composition and function [14, 31]. Approximately half of children with CKD were hypertensive, with a similar proportion of patients having hypertension during dialysis. The chronic increase in systemic afterload can cause concentric hypertrophy and can be associated with changes in systolic and diastolic function [30]. Although patients did show some regression of LV mass following transplantation, values were not restored to normal levels and may be related to persistent hypertension following transplantation, the effect of medications used in the post-transplantation period, such as steroids, or delayed/incomplete ventricular remodeling.

Children with CKD have diastolic functional changes, as represented by the decreased E/A ratio, decreased tissue Doppler E′ velocities, and increased E/E′ ratio. These findings may represent changes in myocardial relaxation and compliance. These results are consistent with previous cross-sectional studies demonstrating diastolic abnormalities in children early on in the course of CKD, which worsen with declining renal function [29, 32, 33, 34, 35]. Our findings extend these studies and show that these diastolic changes persist during dialysis and are present over 1 year following kidney transplant. The observed diastolic dysfunction and subtle changes in systolic function merit longer term prospective evaluation to determine their clinical significance in later life.

This study has some limitations. The study cohort is relatively small and may have insufficient statistical power to detect smaller differences in functional cardiac parameters. Certain patients were excluded due to missing echocardiographic measurements, particularly longitudinal and circumferential strain, due to insufficient image quality for STE analysis or to lack of images in RAW data format. Patients with no available echocardiographic studies were excluded from this analysis, which may affect the generalizability of our findings. Due to the retrospective study design it was not possible to standardize the timing of echocardiograms relative to the onset of CKD, dialysis, and kidney transplantation. Variability regarding the timing of echocardiogram relative to dialysis may have influenced baseline volume status and the diastolic parameters during the cardiac assessment. Although we were able to address short-term outcomes following transplantation, the limited follow-up after kidney transplantation did not allow us to make inferences regarding the effect of dialysis and transplant on longer term (>2 years) outcomes. Lastly, since very few patients had available ambulatory blood pressure reports, this study did not assess the presence of nocturnal or masked hypertension.

Despite these limitations, this longitudinal analysis provides a comprehensive assessment of conventional and STE-derived cardiac parameters using a standardized protocol and a large cohort of healthy children. Our results demonstrate early functional cardiac changes in children with CKD and during dialysis, some of which persist after kidney transplantation. Longer term echocardiographic assessment will determine whether these functional abnormalities normalize in later years. Future studies should explore the benefit of further blood pressure lowering in this patient population.

Conclusion

Children with CKD have diastolic functional changes compared to healthy children, and these changes persist during dialysis as well as after kidney transplantation. Systolic function is generally preserved in childhood CKD, with significant improvement in myocardial strain after transplantation. Further longitudinal investigation is required to determine the long-term clinical significance and reversibility of the persistent changes in diastolic parameters.

Compliance with ethical standards

Disclosure

The authors have no conflicts of interest to declare.

Ethical approval

The study protocol was approved by the Research Ethics Board at the Hospital for Sick Children. For this type of study formal consent is not required.

Supplementary material

467_2017_3622_MOESM1_ESM.doc (48 kb)
Supplementary Table 1(DOC 48 kb)
467_2017_3622_MOESM2_ESM.doc (72 kb)
Supplementary Table 2(DOC 71 kb)

Copyright information

© IPNA 2017

Authors and Affiliations

  • Rawan K. Rumman
    • 1
    • 2
  • Ronand Ramroop
    • 3
  • Rahul Chanchlani
    • 2
    • 4
    • 5
  • Mikaeel Ghany
    • 2
  • Diane Hebert
    • 6
  • Elizabeth A. Harvey
    • 6
  • Rulan S. Parekh
    • 6
    • 7
  • Luc Mertens
    • 3
    • 7
  • Michael Grattan
    • 3
    • 8
  1. 1.Institute of Medical Science, and the Cardiovascular Sciences Collaborative ProgramUniversity of TorontoTorontoCanada
  2. 2.Child Health Evaluative Sciences, Research InstituteThe Hospital for Sick ChildrenTorontoCanada
  3. 3.Division of Cardiology, Labatt Family Heart CenterThe Hospital for Sick ChildrenTorontoCanada
  4. 4.Division of Nephrology, Department of PediatricsMcMaster Children’s Hospital–McMaster UniversityHamiltonCanada
  5. 5.Institute of Health Policy, Management and EvaluationUniversity of TorontoTorontoCanada
  6. 6.Division of NephrologyThe Hospital for Sick Children–University of TorontoTorontoCanada
  7. 7.Department of Pediatrics, Hospital for Sick Children and MedicineUniversity Health Network–University of TorontoTorontoCanada
  8. 8.Department of Pediatrics, Children’s Hospital, London Health Sciences Centre University of Western OntarioLondonCanada

Personalised recommendations