Cystatin C and Cardiac Measures in Children and Adolescents With CKD

Abstract

Background: Cardiovascular disease (CVD) is highly prevalent among children with chronic kidney disease (CKD). Cystatin C is an established marker of kidney function and an emerging biomarker for CVD events. We quantified the relationship between cystatin C level and cardiac structure and function over time among children with CKD and assessed whether cystatin C level and diastolic function retained an association after accounting for kidney function.

Study design: Prospective cohort study.

Setting & participants: 678 children and adolescents with mild to moderate CKD enrolled in the CKD in Children (CKiD) Study with 1,228 echocardiographically obtained cardiac structure and function measurements.

Predictor: Serum cystatin C (mg/L) measured annually.

Outcomes: Cardiac structure (left ventricular mass index [g/m2.7]) and cardiac function (shortening fraction; E/A, E'/A', E/E' ratios) measured every other year.

Measurements: Demographics and anthropometrics, measured glomerular filtration rate (mGFR), heart rate, blood pressure, hemoglobin z score, serum albumin level, and calcium-phosphorus product.

Results: Independent of time, each 1-mg/L increase in cystatin C level was independently associated with a concurrent 7.7% (95% CI, 5.3%-10.0%) increase in left ventricular mass index, a -4.7% (95% CI, -7.0% to -2.4%) change in E/A ratio, a -6.6% (95% CI, -9.0% to -4.2%) change in E'/A' ratio, and a 2.5% (95% CI, 0.3%-4.7%) increase in E/E' ratio. mGFR was also independently associated with E'/A' ratio. When cystatin C level and mGFR were included in the same model, cystatin C level remained independently associated with E'/A' ratio, whereas mGFR was not.

Limitations: 24% of the cohort was missing data for outcomes of interest or measurements; study population includes only children and adolescents with mild to moderate CKD.

Conclusions: In this study of children and adolescents with mild to moderate CKD, cystatin C level was independently associated with cardiac structure and diastolic function. Cystatin C level remained able to predict diastolic function decline via E'/A' ratio even after adjusting for mGFR, suggesting that cystatin C level may have an independent role in CVD risk stratification among children and adolescents with CKD.

Keywords: Biomarker; adolescents; cardiac function; cardiac structure; cardiovascular disease (CVD); children; chronic kidney disease (CKD); cystatin C; echocardiograph; left ventricular hypertrophy (LVH); left ventricular mass (LVM); pediatrics; renal function; systolic function.

Copyright © 2016 National Kidney Foundation, Inc. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

Flow chart outlining the number of visits and number of participants contributing data to the primary and subsequent analyses

Figure 2. Partial residual plot of adjusted E/A Ratio versus Cystatin C (mg/L)

aAdjusted for baseline age, gender, African-American race, age-sex specific BMI z score, age-/sex-/height-specific systolic and diastolic blood pressure z scores, calcium-phosphorus product, and follow-up time. The outcome variable was log transformed for analysis. Multiple points for each subject were included in the longitudinal analysis and are included in the plot, with adjustment for the individual values of the random intercept.

Figure 3. Partial residual plot of adjusted E′/A′ Ratio versus Cystatin C (mg/L)

aAdjusted for baseline age, gender, African-American race, age-sex specific BMI z score, age-/sex-/height-specific systolic and diastolic blood pressure z scores, calcium-phosphorus product, and follow-up time. The outcome variable was log transformed for analysis. Multiple points for each subject were included in the longitudinal analysis and are included in the plot, with adjustment for the individual values of the random intercept.

Figure 4. Partial residual plot of adjusted E/E′ Ratio versus Cystatin C (mg/L)

aAdjusted for baseline age, gender, African-American race, age-sex specific BMI z score, age-/sex-/height-specific systolic and diastolic blood pressure z scores, calcium-phosphorus product, and follow-up time. The outcome variable was log transformed for analysis. Multiple points for each subject were included in the longitudinal analysis and are included in the plot, with adjustment for the individual values of the random intercept.