CKD-EPI creatinine-cystatin C glomerular filtration rate estimation equation seems more suitable for Chinese patients with chronic kidney disease than other equations

BACKGROUND: The aim of this study was to identify the optimal equation that accurately estimates the glomerular filtration rate (GFR) and the chronic kidney disease (CKD) stage in the Chinese population. METHODS: A total of 1296 Chinese patients aged 18–65 years old were enrolled in this study. The estimated GFRs (eGFRs) calculated separately by three Diet in Renal Disease (MDRD) equations and three Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations were compared with the reference GFR (rGFR) measured by the (99)Tc(m)-DTPA renal dynamic imaging method. RESULTS: By Bland-Altman analysis, eGFR(cys) and eGFR(scr_cys) performed similarly, showing the tightest limits of agreement among the six equations. They also achieved the first and second highest 30% and 50% accuracies. Using a combination of the serum creatinine and cystatin C levels (eGFR(scr_cys)) could improve the bias (−0.3 for eGFR(scr_cys)) of the equation and achieve the highest diagnostic accuracy for renal insufficiency (AUC(60), 0.953; P < 0.05, except for eGFR_(MDRD)). All equations predicted stage 3 CKD with moderate accuracy (49.7–51.4%) and stage 5 CKD with good accuracy (90.2–96.4%). For stage 1 CKD, eGFR(cys) showed a higher percentage of misclassification than the other equations. All equations seemed to perform poorly at predicting stage 2 and 4 CKD, as compared to the other CKD stages. eGFR(scr_cys) was the best-performing equation in terms of accurate classification of the CKD stage based on the overall performance (kappa value, 0.423). CONCLUSION: For a Chinese population, the CKD-EPI(scr_cys) equation seems more suitable for estimating the GFR than the other equations. Each equation had its own advantages in predicting different CKD stages.