Comparison of predictive performance of renal function estimation equations for all-cause and cardiovascular mortality in an elderly hypertensive population

Abstract

BACKGROUND: The Modifications of Diet in Renal Disease (MDRD) and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) are 2 equations commonly used to estimate glomerular filtration rate (eGFR). The predictive performance offered by these equations, particularly in relation to clinical outcomes in elderly hypertensive patients, is not clear. METHODS: The Second Australian National Blood Pressure Study cohort was used to investigate the predictive performance of these 2 equations for long-term outcomes (median 10.8 years) in elderly treated hypertensive patients. Both equations were used to calculate eGFR in 6,083 patients aged =65 years and classified as having chronic kidney disease (CKD) or no CKD (eGFR =60ml/min/1.73 m2). RESULTS: More patients were classified as having no CKD using the CKD-EPI equation compared with the MDRD equation (72.1% vs. 69.4%; P = 0.001). Both equations performed similarly in risk prediction of all-cause and cardiovascular mortality with decreased eGFR, except for patients with baseline eGFR of 45-59ml/min/1.73 m2, where the CKD-EPI equation predicted higher risk of all-cause mortality compared with those with no CKD. However, the magnitude of difference in risk prediction was too small to be clinically meaningful. Both equations showed similar predictive performance. However, we observed longer survival and no higher risk in those who were reclassified as having no CKD using the CKD-EPI equation, but these patients were classified earlier as having CKD using the MDRD equation. CONCLUSIONS: There was no clinically relevant difference in predictive performance for long-term survival by eGFR calculated using either of these equations in this elderly hypertensive population.