Aim: To understand the agreement, precision, and accuracy between other estimated glomerular filtration rate (eGFR) equations and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) creatinine-cystatin C equation (EPI_Cr_CysC).
Materials and methods: We conducted a cross-sectional study of 1,913 CKD patients. The eGFRs were calculated separately by creatinine clearance rate and Cockcroft-Gault equation corrected for standard body surface area (Ccr_BSA and eCcr_BSA); CKD-EPI creatinine equation (EPI_Cr); CKD-EPI cystatin C equation (EPI_CysC); EPI_Cr_CysC equation; Modification of Diet in Renal Disease (MDRD) Study equation with standardized serum creatinine; and full-age spectrum creatinine equation (FAS). The EPI_Cr_CysC equation was used as the reference.
Results: When compared with the EPI_Cr_CysC equation, the EPI_Cr equation achieved the highest agreement in eGFRs (Lin's concordance correlation coefficient = 0.936, 95% confidence interval (CI) = 0.930, 0.941). eCcr_BSA and EPI_Cr equations achieved the first and second highest percentage agreement in the accurate classification of CKD stage (72.55 vs. 71.25%). The MDRD equation had minimal bias and was closely followed by the EPI_Cr equation (median difference = -1.3, 95% CI = -2.0, -0.8 vs. median difference = 2.5, 95% CI = 1.7, 3.3 mL/min/1.73m2). The EPI_CysC and EPI_Cr equations achieved the first and second highest precision (interquartile range (IQR) of the difference = 12.2, 95% CI = 11.6, 12.9 vs. IQR of the difference = 15.5, 95% CI = 14.7, 16.3 mL/min/1.73m2). The EPI_Cr and MDRD equations performed similarly and both had the highest accuracy at 30% (1 - P30 = 18.6, 95% CI = 16.9, 20.4 vs. 1 - P30 = 18.6, 95% CI = 16.8, 20.3%).
Conclusion: For assessment of renal function, the EPI_Cr equation performed the best and remained an acceptable alternative to the EPI_Cr_CysC equation in the absence of cystatin C. .