Kidney Week

Abstract: SA-PO171

Comparison of Risk Prediction Using the CKD-EPI Equations or the MDRD Study Equation for Estimated Glomerular Filtration Rate in Chinese Patients with Diabetic Nephropathy

Session Information

• Diabetic Kidney Disease: Clinical - II
  October 27, 2018 | Location: Exhibit Hall, San Diego Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Diabetic Kidney Disease

• 602 Diabetic Kidney Disease: Clinical

Author

• Wang, Yiting, Division of Nephrology, West China Hospital of Sichuan University, Chengdu, Sichuan, China

Yiting Wang,

Background

To evaluate risk implications of estimated glomerular filtration calculated by CKD-EPI equations (eGFR_CKD-EPI-CysC, eGFR_{CKD-EPI-Cr-CysC}) compared to eGFR by eGFR_CKD-EPI-Cr and eGFR_MDRD in Chinese patients with diabetic nephropathy (DN).

Methods

A retrospective study was conducted 363 patients with renal-biopsy DN, 227 patients of them were followed up at least one year. The stages of chronic kidney disease (CKD) were classified by different equations. The eGFR concordance between each two equations was assessed by Bland–Altman plots. The correlation between eGFR and pathological findings was using Spearman analysis. The end point was defined by receiving renal replacement therapy (RRT). Log Rank and Cox regression were employed to evaluate risk implications of eGFR_CKD-EPI and eGFR_MDRD.

Results

Overall, the proportion of CKD 3-5 stages (eGFR ,60mL/min/1.73 m²) was 57.2%, 56.1%, 72.1%, and 64.1% by the eGFR_MDRD, eGFR_CKD-EPI-Cr, eGFR_CKD-EPI-CysC and eGFR_{CKD-EPI-Cr-CysC} respectively. The concordance between eGFR_CKD-EPI-CysC and eGFR_{CKD-EPI-Cr-CysC} (mean bias: 1.20), and between eGFR_MDRD and eGFR_CKD-EPI-Cr (mean bias: 4.21) were superior to others. However, the bias of eGFR_MDRD and eGFR_CKD-EPI-Cr enlarged when eGFR was great than 60mL/min/1.73 m². Compared to eGFR_CKD-EPI-Cr, 40% and 18% of patients were reclassified from CKD 1-2 stages to CKD 3-5 by CKD-EPI-CysC and CKD-EPI-Cr-CysC, respectively, and reached the end point quicker than not reclassified patients (P<0.01); 67.6% and 55.3% of patients with eGFR_CKD-EPI-Cr 45-60 were reclassified to CKD 3b-5 stages by CKD-EPI-CysC and CKD-EPI-Cr-CysC, respectively, however, no significant difference of renal survival time was observed compared to not reclassified patients. The results were similar when compared to eGFR_MDRD. The hazard ratio was 0.960 (95% CI 0.949-0.971), 0.961 (95% CI 0.950-0.971), 0.940 (95% CI 0.925-0.955) and 0.947 (95% CI 0.934-0.960) for eGFR_MDRD, eGFR_CKD-EPI-Cr, eGFR_CKD-EPI-CysC and eGFR_{CKD-EPI-Cr-CysC} respectively, when eGFR was regarded as continuous variables.

Conclusion

eGFR_CKD-EPI-CysC and eGFR_{CKD-EPI-Cr-CysC} classified more patients to CKD 3-5 stages and more accurately categorized the risk of RRT than did eGFR_CKD-EPI-Cr in Chinese DN patients.

Funding

• Government Support - Non-U.S.