Long-Term Prognostic Value for Patients with Chronic Heart Failure of Estimated Glomerular Filtration Rate Calculated with the New CKD-EPI Equations Containing Cystatin C E. Zamora, J. Lupón, M. de Antonio, J. Vila, J. Peñafiel, A. Galán, A. Urrutia, M. Domingo, and A. Bayes-Genis March © Copyright 2014 by the American Association for Clinical Chemistry

© Copyright 2009 by the American Association for Clinical Chemistry Introduction  Background  Correct estimation of renal function is crucial in assessing prognosis of patients with heart failure (HF).  Recently, two new equations have been proposed to calculate estimated glomerular filtration rate (eGFR) with cystatin C alone or both creatinine and cystatin C:  CKD-EPI-cystatin C  CKD-EPI-creatinine-cystatin C  Aims  To assess the prognostic value of eGFR estimated by these new equations in outpatients with HF.

© Copyright 2009 by the American Association for Clinical Chemistry Measurements of performance  Discrimination  The area under the ROC curve (AUC) summarized the diagnostic discrimination.  We used the index of rank correlation, Somers D, which incorporates information from censored data.  AUCs between models were compared with 10,000 bootstrapping replicates.  Calibration  The D’Agostino–Nam version of the Hosmer–Lemeshow calibration test was used to calculate 2 values.  In addition, the Bayesian information criterion (BIC), the Akaike information criterion (AIC), and the Brier score were calculated for each model.  Reclassification  Two main statistics were used to assess reclassification. The integrated discrimination improvement (IDI) and the net reclassification improvement (NRI).

© Copyright 2009 by the American Association for Clinical Chemistry QUESTION  Is it important to assess in heart failure the prognostic value of eGFR calculated with the newer equations?

© Copyright 2009 by the American Association for Clinical Chemistry Table 1. Patient demographics, baseline clinical data, and treatments. Data are n (%) or median (25th, 75th percentile) unless noted otherwise

© Copyright 2009 by the American Association for Clinical Chemistry Figure 1. Long-term Kaplan–Meier survival curves according to National Kidney Foundation eGFR categories for all equations. (A) MDRD-4-IDMS. (B) CKD-EPI. (C) CKD-EPI-cystatin C. (D) CKD-EPI-creatinine- cystatin C.

© Copyright 2009 by the American Association for Clinical Chemistry Table 2. Multivariable Cox regression analysis for risk of death based on eGFR National Kidney Foundation categories a from all equations. a Categories IV and V of the National Kidney Foundation have been merged into 1 category for reasons of sample size. b I–II vs III–IV.

© Copyright 2009 by the American Association for Clinical Chemistry Figure 2. Hazard ratios for death from any cause according to eGFR as calculated with each of the 4 equations. The graphs show associations by plotting the hazard ratio versus a reference eGFR, which is indicated by diamond at 95 mL·min -1 ·(1.73 m 2 ) -1. Coefficients per 1-mL·min -1 ·(1.73 m 2 ) -1 eGFR increment have been calculated in each eGFR interval (i.e., 30, 30 to 60, 60 to 90, 90+), and then a coefficient resulting from the addition of these coefficients has been created for all eGFR values.

© Copyright 2009 by the American Association for Clinical Chemistry QUESTION  What is the likely explanation for the large differences in HR (prognostic implication) found in the different equations to estimate a given eGFR?

© Copyright 2009 by the American Association for Clinical Chemistry Table 3. Performance for risk of death prediction at 4 years of CKD-EPI-cystatin C and CKD-EPI-creatinine-cystatin C compared to MDRD-4-IDMS equation. a P < 0.001; b P =0.51; c P =0.11; d P =0.76; e P =0.045 Table 3

© Copyright 2009 by the American Association for Clinical Chemistry Table 4. Performance for risk of death prediction at 4 years of CKD-EPI-cystatin C and CKD-EPI-creatinine-cystatin C compared to CKD-EPI equation. a P < 0.001; b P =0.68; c P =0.11; d P =0.76; e P =0.003; f P =0.360 Table 4

© Copyright 2009 by the American Association for Clinical Chemistry Figure 3. AUC for risk of death prediction at 4 years according to eGFR calculated with the 4 equations. The best AUC was obtained with CKD-EPI-cystatin C equation.

© Copyright 2009 by the American Association for Clinical Chemistry QUESTION  Why do you think CKD-EPI-cystatin C was the best-performing prognostically, despite prior studies that found it is not the more accurate equation for estimating real eGFR (compared with isotopic GFR)?

© Copyright 2009 by the American Association for Clinical Chemistry Conclusions  The two new CKD-EPI equations containing cystatin C are useful for HF risk stratification and show better prognostic performance than creatinine-only based eGFR equations, mostly in patients with intermediate eGFR.  These equations seem appropriate for assessing prognosis of HF patients with moderate renal insufficiency.

© Copyright 2009 by the American Association for Clinical Chemistry Thank you for participating in this month’s Clinical Chemistry Journal Club. Additional Journal Clubs are available at Download the free Clinical Chemistry app on iTunes for additional content! Follow us