The Chronic Kidney Disease Epidemiology Collaboration equation improves the detection of hyperfiltration in Chinese diabetic patients.

OBJECTIVE Hyperfiltration confers an increased risk of diabetic nephropathy. Early detection can ensure timely intervention and improved treatment outcomes. Because GFR is known to be affected by hyperglycemia, the aim of this study was to compare the influence of hyperglycemia on GFR estimations calculated by the CKD-EPI equation, the CG equation, and the MDRD equations in estimating hyperfiltration in Chinese diabetic patients. MATERIALS AND METHODS The performance of the equations, compared with the measured (99)mTc-DTPA glomerular filtration rate was analyzed in 3492 diabetic patients. Bias, precision, and accuracies were compared with respect to HbA1c status. The Bland-Altman method was used to evaluate the agreement among the equations with respect to the mGFR, and the receiver-operating characteristic curve method was used to evaluate diagnostic value of the three equations with respect to the detection of moderate renal failure and hyperfiltration. RESULTS The mean absolute bias was the smallest for the CKD-EPI equation in the HbA1c < 7.2% cohort, and the highest accuracy within ± 15% and ± 30% was also reached with the CKD-EPI equation in both cohorts. For the detection of hyperfiltration, the CKD-EPI equation exhibited the best performance with the greatest combination of sensitivity and specificity. The biases of the three equations were significantly higher in the HbA1c ≥ 10.5% subgroup compared with the HbA1c < 7.2% cohort. CONCLUSION The CKD-EPI equation can be used as a screening tool for hyperfiltration and appears to be a more generalizable and accurate equation for estimating GFR in Chinese diabetic patients.