Prospective Comparison of Equations Based on Creatinine and Cystatin C for the Glomerular Filtration Rate Estimation in Chinese Renal Transplant Recipients

摘要

Abstract Background Currently, there is no dedicated equation to estimate glomerular filtration rate (GFR) for transplanted kidneys. This study aimed to compare the performance of serum creatinine (Scr)- and cystatin C (CysC)-based equations in Chinese renal transplant recipients. Methods A total of 252 stable renal transplant recipients were enrolled in this study. The plasma clearance of 99m Tc-DTPA (rGFR) was used as a reference standard. The Scr, CysC, and rGFR of the patients were measured on the same day. The bias, precision, accuracy (percentage of estimates within 10%, 30%, and 50% of rGFR), and agreements of 8 Scr and 5 CysC eGFR equations were assessed. The factors affecting the accuracy were also evaluated. Results Among the Scr-based equations, the Japanese Society of Nephrology-Chronic Kidney Disease Initiatives (JSN-CKDI) equation had the best overall performance with a bias of??6.2 mL/min/1.73 m 2 , and 96.1% of its estimates were within 30% of the rGFR. For the CysC-based equations, the Filler equation had the best performance with a bias of??3.9 mL/min/1.73 m 2 , and 93.7% of its estimates were within 30% of the rGFR. Overall, the CysC-based equations showed better performance than the Scr-based equations. In addition, significant differences were observed between bias and gender and between bias and rGFR value in some equations, whereas transplantation time and immunosuppressive regimens were not correlated with the bias. Conclusion The JSN-CKDI equation provides the best estimation of the GFR equations, and the CysC-based equations performed better than the Scr-based equations in this population. Highlights ? In this work, we evaluated the performance of Scr- and CysC-based equations for GFR estimation in Chinese renal transplant recipients. To the best of our knowledge, this is the first evaluation of the GFR estimation equations in renal transplant recipients with a large population. ? Among all the equations included in our study, the Scr-based JSN-CKDI equation had the best estimation; the genetic and environmental similarities between Chinese and Japanese people may contribute to the current result. ? Our results were similar with several previous studies. These results supported the use of CysC-based GFR equations in renal transplant recipients. ? We observed that equations involving females generally had larger bias than those in males. Moreover, these differences were significant in the JSN-CKDI, Larsson, Rule, Hoek, and Filler equations. This finding may be attributed to the absence or inadequate gender calibration in these equations. ? In our research, similar results were observed that nearly all the equations have increased net biases when the GFR value is >60 mL/min/1.73 m 2 , which indicates that GFR equations are not suited for renal transplant recipients whose GFR values are at high level.