Aim: The use of glomerular filtration rate (GFR) estimating equations is not always reliable, especially in specific populations, such as patients with transplanted kidney. The purpose of this study was to improve the performance of GFR equations by taking into account dry lean body mass.
Materials and methods: A prospective clinical study included 100 patients with kidney graft. Estimated GFR (eGFR) with Chronic Kidney Disease Epidemiology Collaboration equations with serum creatinine concentration (CKD-EPI Cr), serum cystatin C concentration (CKD-EPI CysC), or both (CKD-EPI Cr-CysC) were compared with measured GFR with 51Cr-EDTA clearance (mGFR 51Cr-EDTA). Dry lean body mass (body mass without fat mass and water) was measured with bioimpedance analysis.
Results: All of the eGFRs overestimated mGFR 51Cr-EDTA by a significant degree (shown as bias ± SD in mL/min/1.73m2 with 30% accuracy in parentheses): CKD-EPI Cr 15.1 ± 15.3 (50%), CKD-EPI CysC 8.0 ± 16.6 (56%), CKD-EPI Cr-CysC 10.3 ± 13.4 (55%). Dry lean body mass significantly correlated with mGFR 51Cr-EDTA (R = 0.241; p = 0.016) and all biases except the bias of CKD-EPI CysC. Considering the dry lean body mass and preexisting equations with creatinine, we developed two new equations with better performance and statistically insignificant bias: Corrected CKD-EPI Cr -1.43 ± 13.6 (67%) and Corrected CKD-EPI Cr-CysC -1.64 ± 13.4 (77%).
Conclusion: Dry lean body mass improves the performance of GFR equations in our kidney transplant cohort.