Background: The renal dynamic imaging method (modified Gate's method) with (99m)Tc-diethylene triamine pentaacetic acid ((99m)Tc-DTPA) is simple and less time consuming for glomerular filtration rate (GFR) estimation than other methods. However, its diagnostic performance as a surrogate marker of GFR is questioned increasingly. Recently, the modified Modification of Diet in Renal Disease (MDRD) study equation based on data from Chinese patients of chronic kidney disease (CKD) showed significant performance improvement. In the present study, the renal dynamic imaging methods and the modified abbreviated MDRD equation were compared with the plasma clearance method.
Methods: Four hundred and eighty two patients with CKD were selected. GFR were estimated simultaneously using three methods: (i) modified Gate's method (gGFR); (ii) the modified abbreviated MDRD equation (c-aGFR) and (iii) dual plasma sampling method (rGFR). Using rGFR as the reference method, gGFR and c-aGFR were compared with rGFR in each stage of CKD.
Results: Both gGFR and c-aGFR were correlated well with rGFR (r(gGFR) = 0.81 and r(c-aGFR) = 0.90, P < 0.001). In the overall performance, c-aGFR had less bias (849.5 vs 933.1 arbitrary units), higher precision (57 vs 78.4 ml/min/1.73 m(2)) and higher accuracy than gGFR. For gGFR, the 15, 30 and 50% accuracies were 32.4, 56.0 and 79.1%, respectively; for c-aGFR, the corresponding accuracy rose to 43.2%, 75.5% and 90.9%, respectively. In each stage of CKD, the modified abbreviated MDRD equation also outperformed the modified Gate's method in the GFR estimation.
Conclusion: Our results indicated that the performance of the renal dynamic imaging in total GFR estimation was not better than the modified abbreviated MDRD equation in our patient group, and should not be used as a surrogate marker of GFR, especially in clinical trials. We presume that the dynamic renal imaging methods for estimation of GFR can be improved by using proper reference GFR, more adequate background subtraction and soft-tissue attenuation correction, in a relatively larger sample size.