Risk stratification for in-hospital mortality after heart transplantation using the modification of diet in renal disease and the chronic kidney disease epidemiology collaboration equations for estimated glomerular filtration rate.

Abstract

A new equation for estimating glomerular filtration rate (GFR)-the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation-is better at predicting outcomes in the general population than the Modification of Diet in Renal Disease (MDRD) equation. We compared risk stratification of heart transplant (HT) recipients for early post-HT mortality using estimated GFR from the MDRD and the CKD-EPI equations. We identified all patients 18 years or older who underwent their first HT in the United States between January 2007 and October 2010 (n=6,564). We compared risk stratification for posttransplant in-hospital mortality by GFR estimated by the CKD-EPI equation versus that estimated by the MDRD equation. Posttransplant in-hospital mortality was 4.6%. Lower GFR (mL/min/1.73 m2, MDRD) was associated with higher in-hospital mortality in adjusted analysis (GFR, 60-89; odds ratio [OR], 1.5; 95% confidence interval [CI], 1.0-2.3; GFR, 30-59; OR, 2.2; 95% CI, 1.4-3.3; GFR<30; OR, 3.3; 95% CI, 1.8-6.1; vs. GFR ≥ 90). Glomerular filtration rate estimated using the CKD-EPI equation reclassified 10%, 15%, and 18% of patients, respectively, in GFR categories lower than 30, 30-59, and 60-89 to the next higher GFR category. Using the CKD-EPI equation was not associated with a significant net reclassification improvement for mortality risk in the overall cohort or in GFR subgroups. The risk models of in-hospital mortality developed using the two GFR equations performed similarly for discrimination and calibration. Estimated GFR using the CKD-EPI equation is comparable to estimated GFR using the MDRD equation in risk stratification of HT recipients for early posttransplant mortality.