Optimizing the Estimation of Renal Function in Lung Transplant Candidates

Abstract

Purpose The Modified Diet in Renal Disease (MDRD) equation is commonly used for GFR estimation, but has been questioned due to inaccuracy for GFR greater than 60ml/ (min * 1.73m2). The CKD-EPI and Wright equations have shown improved predictive capabilities in some patient populations. We compare these equations to determine which correlates best with direct GFR measurement in lung transplant candidates. Methods and Materials We conducted a retrospective cohort analysis of 274 lung transplant cases at a single, high-volume lung transplant program. GFR was directly measured using intravenous Tc-99m DTPA. Results from the following equations were compared to direct GFR measurements: Normalized and BSA adjusted versions of MDRD and CKD-EPI, and Wright. Results The BSA dependent CKD-EPI equation provided the most consistent results with low Bias (Median Standard Error = 1.34), Relative Standard Error (Median RSE = 0.02), Mean Percentage Error (MPE = 4.14), Mean Absolute Percentage Error (MAPE = 15.72%) and precision (IQR = 24.24). R^2 was highest for this equation at 0.45 and accuracy was 98.33%. MDRD parameters poorly correlated with the direct GFR measurements with R^2 = 0.36, accuracy = 98.13, Median SE = 0.77, Median RSE = 0.01, MPE = 7.15, MAPE = 18.60 and precision = 25.18. The Wright equation was least accurate with: R^2 = 0.40 and accuracy = 92.64. Bland-Altman analyses provide visual confirmation of results. [ figure 1 ] Conclusions In patients being evaluated for lung transplantation, the CKD-EPI equation is a better estimator of GFR than the more popular MDRD equation. Both CKD-EPI and MDRD perform better than the Wright formula. Pre-transplant evaluation and early postoperative management of lung recipients may be improved by the use of the CKD-EPI to estimate GFR.