Assessment of Validity of Estimated Glomerular Filtration Rate (GFR) from Cross-Sectional Regression Equations When Evaluating Risk Factors In Longitudinal Studies Of Renal Disease

• Activity Number: 219
• Type: Contributed
• Date/Time: Tuesday, August 5, 2003 : 9:00 AM to 10:50 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Xuelei Wang*+ and Tom H. Greene and Harold Feldman
• Companies: Cleveland Clinic Foundation and Cleveland Clinic Foundation and University of Pennsylvania
• Keywords: validation ; mixed effects model ; renal disease

Abstract

The glomerular filtration rate measured by clearance of 125I-iothalamate (iGFR) is a reference standard for determining kidney function. However, because measurement of iGFR is difficult, cross-sectionally derived regression equations based on serum creatinine (Scr) and other easily ascertained factors have been used to obtain estimates of GFR (eGFR). The African American Study of Hypertension and Kidney Disease (AASK) is a recently completed clinical trial with biannual Scr and iGFR measurements over four years in 1,094 patients. We apply the AASK database to examine the validity of widely used equations for eGFR as a surrogate for GFR in longitudinal studies. Analyses included comparisons of effects of 32 potential risk factors on eGFR and iGFR slope and on time-to-event outcomes defined by 50% reduction in either iGFR or eGFR. Results for eGFR generally agreed with iGFR for time-to-event outcomes. Effects on slope were similar for most factors, but certain discrepancies were too large to explain by chance. Pending confirmation in other studies, this study supports cautious use of eGFR in longitudinal studies for time-to-event outcomes defined by large changes in renal function.