Creatinine-based GFR predicting equations in renal transplantation: reassessing the tubular secretion effect

The real utility of blocking the tubular secretion of creatinine with cimetidine in order to ameliorate the prediction of renal graft function is questionable, particularly in the context of an increasing diffusion of the Modification of Diet in Renal Disease (MDRD) study equation. We have compared the impact of cimetidine on the performances of the Cockcroft-Gault (C-G) and MDRD equations in 56 renal transplant patients with an estimated glomerular filter rate (GFR) >30 mL/min/1.73 m(2) for whom true GFR was directly measured by inulin clearance.

Serum creatinine concentration (SCr) was measured [isotope dilution mass spectrometry (IDMS) traceable enzymatic assay] at the beginning of the inulin clearance procedure and 2 days later, after three oral cimetidine doses of 800 mg every 12 h. Predictive and diagnostic performances of the re-expressed MDRD and C-G formulas were compared before and after cimetidine intake.

Mean SCr (+/-SD) increased from 120 micromol/L (+/-34) before to 154 micromol/L (+/-47) after cimetidine. The beneficial effect of cimetidine was significant only on the accuracy of the C-G formula (accuracy 30% post-cimetidine of 93 and 79% for the C-G and MDRD equations, respectively). Likewise, while a higher proportion of patients were correctly staged using the chronic kidney disease classification after cimetidine with the C-G equation (59% before and 68% after), no improvement was seen with the MDRD formula (59 vs 57%). For both equations, receiver operating characteristic curves analysis showed only a marginal gain in GFR prediction.

Our data do not support the use of a cimetidine-based strategy for the evaluation of renal graft function in the clinic, particularly when the GFR is estimated by the MDRD equation.