Iohexol as a marker of glomerular filtration rate in patients with diabetes: comparison of multiple and simplified sampling protocols

Aims To report on the reproducibility of iohexol glomerular filtration rate (GFR) estimation, to compare the plasma clearance of iohexol with that of [Cr-51]EDTA and to evaluate the reliability of reduced sampling schedules in estimating GFR in Type 1 and Type 2 diabetes mellitus. Methods Agreement was assessed in 15 Type 1 and 26 Type 2 diabetics with creatinine ranging from 53 to 564 mmol/l. Results The regression between multiple-sample iohexol and [Cr-51]EDTA clearances was 0.999 in Type 1 and 0.987 in Type 2 diabetes (P < 0.0001 for both). A seven-sample design and the three-sample approach by Brochner-Mortensen were validated by comparison with the full-sample schedule in 87 patients (51 Type 1, 36 Type 2). Full-sample GFR was 80.3 +/- 43.8, seven-sample 79.5 +/- 43.9 (r = 0.990) and three-sample 79.8 +/- 45.2 ml.min(-1) .1.73 m(-2) (r= 0.972). The coefficients of variation of GFR were 2.7 +/- 1.4% m(-2) and 3.8 +/- 1.9% for the full-sample and the seven-sample approaches, respectively, and significantly higher for the three-sample design (6.9 +/- 3.4%, P=0.0001). Conclusions After iohexol injection, the Brochner-Mortensen schedule does not provide an accurate estimate of GFR. The seven-sample approach gives acceptable errors and allows a good estimate of GFR throughout a wide range of renal function.