Creatinine-based equations to estimate glomerular filtration rate should be used with caution in patients with neurogenic bladder.

Abstract

Patients with neurogenic bladder (NGB) are at an increased risk of developing chronic kidney disease (CKD). However, data related to the real performance of the serum creatinine (Cr)-based estimated glomerular filtration rate (eGFR) equation in patients with NGB are limited. This study is to evaluate the performance of new Cr-based Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation without race and the GFR estimation equation for Chinese CKD patients for the estimation of GFR in Chinese patients with NGB. GFR was determined simultaneously by three methods: a) GFR measured by renal dynamic imaging with 99mTc-DTPA (G-GFR), which was used as the reference GFR; b) GFR estimated by the new Cr-based Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation without race (EPI-GFR); and c) GFR estimated by the equation for Chinese CKD patients (C-GFR). Pearson correlation and linear regression were used to compare eGFR and G-GFR. Differences, absolute differences, precision, and accuracy were compared to identify which equation showed better performance in evaluating GFR in patients with NGB. A total of 171 patients with NGB, including 121 men and 50 women from 20 provinces, 4 autonomous regions, and 3municipalities in China, were enrolled in the final analysis, and the average age was 31.3 ± 11.9 years. Both C-GFR and EPI-GFR were moderately correlated with G-GFR and overestimated G-GFR. The difference between EPI-GFR and G-GFR was similar to that between C-GFR and G-GFR (median of 9.97 vs. 9.95 mL/min/1.73m2 for difference, Wilcoxon signed ranks test, Z = -1.704, p = 0.088), but the absolute difference between EPI-GFR and G-GFR was significantly lower than that between C-GFR and G-GFR (median of 22.3 vs. 25.1 mL/min/1.73m2 for absolute difference, Wilcoxon signed ranks test, Z = -4.806, p < 0.001). Both EPI-GFR and C-GFR displayed similar results of 15, 30, and 50% accuracies (χ2-test, p > 0.05), and there were no significant differences between EPI-GFR and C-GFR in misclassification percentages at different G-GFR levels (χ2-test, p > 0.05). Our study indicated that for patients with NGB in China, Cr-based eGFR equations, which include the new CKD-EPI equation without race and the Chinese GFR estimation equation, showed suboptimal performance, and limited their application in GFR estimation. Further studies are needed to investigate whether incorporating additional biomarkers, such as cystatin C, could improve their performance of GFR estimating equations in patients with NGB.