Validation of the Virga GFR Equation in a Renal Transplant Population

Abstract

Background: Virga and colleagues derived a glomerular filtration rate (GFR) equation which demonstrated a superior performance over Cockcroft-Gault (C-G) and modified diet in renal disease-isotope dilution mass spectrometry (MDRD-IDMS) formulas in chronic kidney disease (CKD) patients. Aim: To validate the performance of the Virga equation on 103 renal transplant patients. Methods: We compared the performances of the MDRD-IDMS, C-G and Virga equations using inulin clearance as a reference test. Error, accuracy, relative accuracy, precision, scatter, and coefficient of variance of each equation were tested. Results: The mean absolute percentage error in estimated GFR by the new equation was 39.8 ± 36.34% (mean ± SD). Relative accuracy at 10, 30 and 50% range were 18.44, 48.54 and 73.78%, respectively. It has a bias of 0.09 ± 0.169 and a precision of 19.69. Inulin clearance (GFR) in stages 1–4 were 106.19 ± 14.11, 71.17 ± 7, 42.37 ± 8.40 and 22.92 ± 3.48 ml/min/1.73 m², respectively. Comparative statistics in the overall population and in patients with transplant CKD stage 3T showed that the MDRD-IDMS equation had better accuracy. The performance of MDRD-IDMS over the Virga equation was clearly superior for males. In patients with CKD stage 2T, the Virga equation showed superiority over MDRD-IDMS. In the overall and subpopulations, the Virga equation performed better than the C-G equation. Conclusion: Among renal transplant patients, the results suggest that the best GFR estimate is probably obtained using the MDRD-IDMS equation in moderate kidney failure whilst the Virga formula was superior to MDRD-IDMS for patients with mild kidney failure. As in untransplanted patients, estimating GFR with the MDRD-IDMS equation is not advisable in the range of normal renal function because of its known underestimation of renal function.