The performance of three estimates of glomerular filtration rate before and after live donor nephrectomy

Summary Serum creatinine‐based estimates of glomerular filtration rate (GFR) are inaccurate in healthy individuals. Therefore, their use in assessment prior to live donor nephrectomy has been restricted. There are less data on their use postdonor nephrectomy. This study assessed three GFR estimates against Cr51 EDTA radioisotope GFR (iGFR) in the same cohort of patients before and after donor nephrectomy. A total of 206 patients underwent iGFR measurement prior to donor nephrectomy and this was repeated in 187 patients 6–8 weeks postsurgery. The iGFR was compared with the modification of diet in renal disease (eGFR), Cockcroft–Gault (cgGFR) and Mayo Clinic equation (mcGFR) estimates of GFR. Preoperatively, all GFR estimates performed poorly against iGFR; however, mcGFR provided the most reliable estimate. The eGFR underestimated iGFR by 23.60 ± 16.43 ml/min/1.73 m2, cgGFR by 15.54 ± 18.13 ml/min/1.73 m2 and mcGFR overestimated by 0.72 ± 18.11 ml/min/1.73 m2. Postdonation, all estimates again performed poorly, but eGFR and mcGFR outperformed cgGFR. The eGFR underestimated iGFR by 9.13 ± 10.11 ml/min/1.73 m2, mcGFR by 9.44 ± 13.80 ml/min/1.73 m2 and cgGFR overestimated by 6.42 ± 14.49 ml/min/1.73 m2. No GFR estimate performed sufficiently well to supersede iGFR measurement prior to donor nephrectomy. Performance postdonation was little better. In addition, there was no correlation between fall in iGFR and fall in GFR estimates postdonation.