Glomerular Filtration Rate Estimated by a New Equation is a Better Predictor for Cardiovascular Events in Early-Stage Chronic Kidney Disease Patients Hospitalized with Acute Heart Failure: Data from KorHF (korean Acute Heart Failure) Registry

Introduction: Renal dysfunction is a well-known prognostic marker in patients (pts) with acute heart failure (AHF). Generally, glomerular filtration rate (GFR) is used for assessing renal dysfunction. A new equation to estimate GFR from Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) has been recently introduced to be more accurate than Modification of Diet in Renal Disease (MDRD) equation. Hypothesis: We hypothesized that estimation of GFR by CKD-EPI equation would have a better prognostic power in predicting cardiovascular (CV) events than MDRD equation in early-stage CKD pts with AHF. Methods: We retrospectively analyzed data from the Korean acute Heart Failure (KorHF) Registry from its inception in November 2005 to November 2009, which enrolled hospitalized pts with treatment for AHF (n = 3200, 67.6 ± 14.3 years old). Only patients available with both serum creatinine and N-terminal pro-brain natriuretic peptide (NT-proBNP) were included in this study (n = 1312, 687 males, 68.1 ± 14.5 years old, 555 ischemic origin). The CV events were composite of CV mortality and HF rehospitalization. Early-stage CKD was defined as CKD stage 2 (60 ml/min/1.73m2 ≤ GFR < 90 ml/min/1.73m2). Results: The median NT-proBNP was 4972.5 (IQR 2110.5-11413.5) pg/ml and mean left ventricular ejection fraction (LVEF) was 38.8 ± 16.1%. Mean GFR by CKD-EPI and by MDRD were 57.4 ± 27.8 and 58.8 ± 33.9 ml/min/1.73m2, respectively. The prevalence of CKD stage 2 by CKD-EPI (30.8%) was significantly different from that by MDRD (25.5%) (p < 0.001). During follow-up period (median 339 days, IQR 73-732), 248 (18.9%) pts died and 287 (21.9%) pts were rehospitalized. In multiple Cox proportional hazard analysis, CKD stage 2 by CKD-EPI was an independent predictor of CV events after adjusting age, sex, systolic blood pressure, LVEF, hemoglobin and log NT-proBNP (hazard ratio: 1.49, 95% CI 1.06– 2.11, p = 0.022), whereas not by MDRD (hazard ratio: 1.18, 95% CI 0.87– 1.59, p = 0.284). Conclusion: We found that estimated GFR by CKD-EPI was more predictive of CV events in early-stage CKD pts than that by MDRD. Therefore, CKD-EPI equation could be more helpful for CV risk stratification in AHF pts with early-stage CKD.