Quantitative measurement of renal oxygen metabolism level is of central importance in understanding and treating renal diseases. In the context of Blood Oxygenation Level Dependent (BOLD) contrast, Gradient Echo (GRE) based sequences have been employed to estimate the normal human renal R2* non-invasively [1]. Moreover, according to the biophysical analytical model [2], R2′ is measured by using Asymmetric Spin Echo [3]. R2′ is linearly dependent on tissue oxygenation [2], however, R2′ contains the contributions from multiple factors including blood volume, oxygenation and renal tissue composition. Renal metabolic rate of oxygen (RMRO2) is critically important to access the tissue oxygen metabolism under both normal and disease states [4], therefore, RMRO2 able to provide a more specific and direct evaluation of renal oxygenation. In this study, a qBOLD sequence, multi-echo gradient and spin echo (MEGSE) [5] and FAIR-ASL sequence [7], were implemented to estimate RMRO2 based on the Fick principle of arteriovenous oxygen difference [8] as a direct indication of the renal oxygenation level in rabbits. This approach was used to assess the RMRO2 of normal health rabbits. Furthermore, we evaluated whether the approach can reliably detect renal oxygenation changes under a pathological/physiological condition induced by the renal acute artery stenosis.Ten New Zealand rabbits weighting 2.8-3.5kg were study with an approved animal protocol by the Ethics Committee. Left Renal Artery Stenosis (RAS) was induced surgically in all animals. MR images were acquired on a 3.0T whole-body MR scanner (Signa ExciteTM; GE Medical Sysgems, Milwaukee, Wisconsin, USA) with 8 Channel Phase Array KNEE coil in rabbits. Three sequential MR scans were acquired pre- -20 and -30, post-RAS operation 20, 30, 40, 50, 60,70, 80 and 90 minutes respectively. T2-weighted Fast Spin Echo, qBOLD sequence (Multi-echo gradient and spin echo, MEGSE) and Arterial Spin Labeling (ASL), were acquired at all ten time points (tps) in this sequential order. As in brain OEF applications[4,6], a 2D multi-echo gradient and spin echo (MEGSE) was used for the acquisition of the renal OEF signal. FAIR-ASL was used for the acquisition of the renal RBF signal [7]. The MRI parameters for MEGSE images were: TR=1500ms; TE=56ms, # of echo = 32, echo spacing = 3.748ms, readout bandwidth = 62.5kHz, FOV=256*256mm2, matrix size = 128*128, slice thickness = 5mm. Free hand ROIs were defined to encompass the renal cortex and medulla region(CORTEX, outer stripes of the outer medulla: OSOM, inner stripes of the outer medulla: ISOM and inner medulla: IM) of rabbits to obtain OEF and RBF at all ten time points. According to the Fick principle of arteriovenous oxygen difference, the RMRO2 can be estimated [8]. Paired student t test was employed to test whether measurements of renal RMRO2 was significantly different pre- and post renal artery stenosis.

Selection of Free hand regions of interest (ROIs) in the representative baseline axial T2-weighted images. Red region identify cortex, green region identify OSOM, yellow region identify ISOM, and black region identify IM and the time scale shows the MRI protocol performed before and after left renal artery stenosis operation was shown in Figure 1. Representative Spin Echo images and RMRO2 maps acquired pre-, post-RAS operation 30, 60 and 90 minutes in the same rabbit are shown in Figure 2. At the Post RAS time points , RMRO2 firstly decrease and then maintain steady-state in the occluded cortex and OSOM region. As shown in Figure 3, significant reductions of RMRO2 in the renal cortex and OSOM were obtained (Cortex, RMRO2 = 890.4±420.7 baseline vs. 385.2±164.5 post-RAS 30 min, 355.3±149.4 post-RAS 60 min, 412.5±237.5 post-RAS 90 min, P < 0.05; OSOM, RMRO2 = 609.0±223.0 baseline vs. 275.0±132.9 post-RAS 30 min, 325.0±217.3 post-RAS 60 min, 200.2±80.0 post-RAS 90 min, P < 0.05), suggesting an significantly decrease of oxygen metabolism level in the cortex and OSOM region after the renal artery stenos. In addition, RMRO2 in the ISOM and IM regions decreased slightly, but not statistically significantly.

Renal artery stenosis decreases the supply blood flow to kidney tissue and may lead to a hypoxic state. It is expected that intra-renal RMRO2 may decrease due to insufficient blood flow under this pathological condition. In agreement of this concept, our results demonstrated that a consistent and significant increase of renal RMRO2 in rabbits post renal artery stenosis, suggesting that the technique can be utilized to noninvasively detect pathophysiological changes in intra-renal oxygen metabolism level during an acute reduction of RBF, which may be potentially applicable in humans.