Background: Several organizations recommend using estimated glomerular filtration rate (eGFR) in kidney function monitoring, preferably calculated with Modification of Diet in Renal Disease (MDRD) formula. The role of this formula is not clear in the risk stratification of contrast induced acute kidney injury (CIAKI) in nonsteady state patients. Aim: Comparative evaluation of the MDRD eGFR in risk stratification of CIAKI. Method. GFR was measured twice (pre- and post-examination) by Tc-99m-DTPA, along with serum levels of urea nitrogen and creatinine in 32 patients (mean age +/- SD, 60.1 +/- 13.2 years) needing hospital care for various reasons and underwent to x-ray examination with contrast media (mean; 90.2 +/- 16.8 ml). eGFR was calculated by the dedicated formula. Agreement between measured GFR (mGFR) and MDRD eGFR was assessed and patients were scored and stratified for CIAKI by using first mGFR, then eGFR and results were compared. Results: A moderate correlation was obtained between mGFR and eGFR (r = 0.47, p <0.001) and the difference was not significant. However, Bland&Altman analysis revealed large limits of agreement between mGFR and eGFR (-80.3 to 55.2) with a mean difference of -12.5 ml/min/1.73m(2). In ROC analysis, when mGFR values were classified as normal (>60 ml/min/1.73m(2)) and decreased (<60ml/min/1.73m(2)), AUC was 0.80 (95%CI; 0.62-0.92) for eGFR, with a sensitivity of 29% and specificity of 100%. Furthermore, the risk group categorization, using eGFR instead of mGFR was resulted in a group change for four patients (13%); from moderate to low risk group. Conclusion: It seems that MDRD eGFR differs from mGFR. In nonsteady state patients CIAKI classification using eGFR should be considered with caution.