TY - JOUR T1 - Mapping of Local Renal Blood Flow with PET and H<sub>2</sub><sup>15</sup>O JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 470 LP - 475 VL - 43 IS - 4 AU - Nathaniel M. Alpert AU - Carlos A. Rabito AU - D. John A. Correia AU - John W. Babich AU - Bruce H. Littman AU - Ronald G. Tompkins AU - Nina T. Rubin AU - Robert H. Rubin AU - Alan J. Fischman Y1 - 2002/04/01 UR - http://jnm.snmjournals.org/content/43/4/470.abstract N2 - We developed a noninvasive method for the mapping of regional renal blood flow in humans using PET and H215O. Methods: Fifteen subjects participated in the study, 5 with normal renal function and 10 with renal disease. The protocol used a whole-body PET scanner, intravenous bolus injection of 1,110–1,850 MBq H215O and sequential imaging at 3 s per frame. 131I-Iodohippuran was used to independently assess effective renal plasma flow in each subject. Hippuran clearance and renal blood flow (RBF) were measured twice, before and after treatment with probenecid, to verify that RBF is not affected. Flow analysis was based on the Kety model, according to the operational equation: C(t) = F ∫ Ca(u)du − k ∫ C(u)du, where F is the RBF, k is the tissue-to-blood clearance rate, C is the PET concentration, and Ca is the tracer concentration in the abdominal aorta. F and k were estimated by linear least squares on a pixel-by-pixel basis to produce quantitative maps (parametric images) of RBF. The flow maps were analyzed by regions of interest (largely excluding the medulla and collecting system) for each kidney on each slice and pooled to yield mean RBF. Results: In the 5 healthy subjects, mean RBF was 3.4 ± 0.4 mL/min/g. There was no difference in flow between kidneys (t = −0.59; n = 11; P &gt; 0.95). Before treatment with probenecid, RBF was linearly related to hippuran clearance (r2 = 0.92). Probenecid treatment significantly reduced hippuran clearance (P &lt; 0.003), but RBF was unchanged (P &gt; 0.17). Compared with healthy control subjects, RBF was significantly decreased in patients with renal disease (P &lt; 0.002). Flow maps were of good quality in all subjects, exhibiting characteristic patterns, with higher values in regions composed largely of renal cortex. Conclusion: Parametric mapping of RBF with PET and H215O provides a straightforward, noninvasive method for quantitative mapping of RBF, which may prove useful in research applications and in the management of patients whose therapy alters renal tubular transport. ER -