How to estimate kidney function in kidney transplant recipients with mild to moderate kidney impairment: comparison of estimated glomerular filtration (eGFR) values between creatinine-based GFR equations and cystatin C-based GFR equations for Japanese population.

backgroundWith the recent increase in renal transplantations in Japan, accurate assessment of renal function is required.

methodsThis study included 73 patients who had undergone renal transplantation at Nagoya Daini Red Cross Hospital at least 6 months previously and had stable renal function for >3 months. Glomerular filtration rates (GFRs) were measured by inulin clearance (mGFR) and compared with estimated cystatin C-based GFRs (eGFRcys), estimated creatinine-based GFRs (eGFRcre) and their average values (eGFRave).

resultsmGFR was 43.3 ± 14.1 mL/min/1.73 m(2), eGFRcre was 39.6 ± 11.7, eGFRcys was 56.0 ± 17.1, and eGFRave was 47.8 ± 13.7 mL/min/1.73 m(2). Serum cystatin C was 1.39 ± 0.37 mg/L and serum creatinine was 1.58 ± 0.51 mg/dL. The correlation coefficients between mGFR and eGFRcre, eGFRcys, and eGFRave were 0.768, 0.831, and 0.841, respectively (P < 0.001, for all).The intraclass correlation coefficients were 0.754, 0.816, and 0.840, respectively (P < 0.001, for all).The mean differences between measured and estimated GFR values were 3.74 mL/min/1.73 m(2) with a root-mean square error (RMSE) of 9.06 for eGFRcre, +12.64 with RMSE of 9.48 for eGFRcys, and +4.45 with RMSE of 7.86 for eGFRave. Bland-Altman plots showed that eGFRcys overestimated GFR values compared with mGFR values in most cases and that eGFRave overestimated GFR values in 53 of 73 cases, whereas eGFRcre underestimated the values in 53 of 73 cases.

conclusioneGFRave may be the best marker to estimate kidney function in Japanese renal transplant recipients with mildly reduced or normal kidney function.