[en] Full text: A retrospective study was performed to compare 3 different methods of estimating GFR: DTPA clearance, 24-hour urine collection and serum creatinine using the Cockcroft-Gault equation. 40 patients (22 female, 18 male, mean age 61 (range 23-82) years) referred between 1997 and 2000 for DTPA renal scans and assessment of GFR were studied. The DTPA renal scan used an administered dose of 413-874MBq. Blood samples were drawn at 60 minutes and 150 minutes. Blood samples were centrifuged and two 1 ml aliquots of each sample were pipetted and counted in a Wallac well counter. A 3.7 kBq Cobalt 57 and technetium dilution standards were used to calibrate the well counter against the dose calibrator. Urinary 24-hour collections and serum creatinine levels were measured in a NATA accredited pathology laboratory. GFR estimated by DTPA clearance gave results in the range 42-168 ml/min/1.73m2. Analysis of GFR estimates normalised for body area for all 3 methods gave correlation coefficients of 0.79 for DTPA clearance vs serum creatinine, 0.72 for DTPA clearance vs 24- hour urine collection and 0.80 for 24-hour urine collection vs serum creatinine. The correlation was stronger for patients with impaired renal function and estimated GFR less than 80 ml/min/1.73m2 (r= 0.87, 0.86, 0.86 respectively). The DTPA clearance correlates well with other routine measurements used to estimate GFR. Copyright (2003) The Australian and New Zealand Society of Nuclear Medicine Inc