Acute Kidney Injury: Arterial Spin Labeling to Monitor Renal Perfusion Impairment in Mice Comparison with Histopathologic Results and Renal Function 1

Transcription

1 Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at Katja Hueper, MD Marcel Gutberlet, PhD Song Rong, MD Dagmar Hartung, MD Michael Mengel, MD Xia Lu, MD Hermann Haller, MD Frank Wacker, MD Martin Meier, PhD Faikah Gueler, MD Acute Kidney Injury: Arterial Spin Labeling to Monitor Renal Perfusion Impairment in Mice Comparison with Histopathologic Results and Renal Function 1 Purpose: Materials and Methods: To determine if arterial spin-labeling (ASL) magnetic resonance (MR) imaging can show serial changes in renal perfusion in mice with ischemia-induced acute kidney injury (AKI) and to compare imaging results with those of renal histologic examination and inulin and para-aminohippuric acid (PAH) clearance. In this animal care committee approved study, AKI was induced in C57Bl/6 mice (n = 26) by clamping the right renal pedicle for 35 minutes for moderate (n = 16) or 45 minutes (n = 11) for severe AKI. Renal perfusion was measured in 10 animals with moderate and seven animals with severe AKI before and at five time points 1 28 days after surgery by using ASL with a 7-T MR imaging unit. Kidney volume loss and histologic evidence of acute tubular injury were assessed. Inulin and PAH clearance was determined in four animals with moderate and six animals with severe AKI to evaluate renal function and plasma flow for statistical analysis. Repeated measures analysis of variance, unpaired t tests, and correlation analysis were used. Original Research n Experimental Studies 1 From the Department of Radiology (K.H., M.G., D.H., F.W.), Department of Nephrology (S.R., X.L., H.H., F.G.), and Institute for Animal Science (M. Meier), Hannover Medical School, Carl-Neuberg-Str 1, Hannover, Germany; and Alberta Transplant Applied Genomics Centre, University of Alberta, Edmonton, Alberta, Canada (M. Mengel). Received February 12, 2013; revision requested April 2; revision received April 29; accepted June 6; final version accepted June 6. Address correspondence to K.H. ( hueper.katja@mh-hannover.de). q RSNA, 2013 Results: Conclusion: Renal perfusion values at day 7 were significantly reduced after moderate (56% 6 8; P,.01) and severe (33% 6 6; P,.001) AKI compared with presurgery values. Renal perfusion had returned to baseline levels at day 21 after moderate (96% 6 14) and remained compromised until day 28 after severe (46 % 6 9; P,.05) AKI. At day 28, for moderate versus severe AKI, kidney volume (84% 6 6 vs 60% 6 5; P,.05), degree of tubular injury (5.6% vs 15.8% 6 2.4; P,.01), and inulin and para-aminohippuric acid clearance (47.5 µl/min vs 7.3 µl/min 6 2.7; P,.001 and µl/min vs 4.8 µl/min 6 1.0; P,.001, respectively) were significantly different. Relative renal perfusion at days 7 28 significantly correlated with kidney volume loss (P,.01) and tubular injury (P,.05) 4 weeks after AKI. ASL allows evaluation of renal perfusion impairment associated with kidney volume loss and histologic changes after AKI in mice and may serve as a noninvasive biomarker for AKI. q RSNA, 2013 Radiology: Volume 270: Number 1 January 2014 n radiology.rsna.org 117

2 Acute kidney injury (AKI) is a common medical problem that is characterized by sudden loss of renal function (1,2). It occurs in various clinical settings, particularly in critically ill patients. Depending on its severity, AKI is associated with subsequent development of chronic kidney disease, poor prognosis, and increased mortality (3 6). Renal ischemia reperfusion injury is a major cause of AKI and leads to inflammation, decreased renal perfusion, loss of renal function, and progressive renal fibrosis (7 9). Clinical diagnosis and staging of AKI are routinely based on measurement of serum creatinine and urine output (1). However, these parameters are only altered if substantial, potentially irreversible kidney damage is present. Functional magnetic resonance (MR) imaging may allow noninvasive detection of the presence and severity of renal abnormalities associated with AKI. The decrease in renal perfusion after an ischemic event is thought to be critical to the pathophysiology of ischemia-induced AKI (8,10,11). Noninvasive measurement Advances in Knowledge nn Arterial spin-labeling MR imaging allows noninvasive quantitation of renal perfusion impairment after acute kidney injury (AKI) in mice. nn Renal perfusion impairment is more pronounced after severe than after moderate AKI, with severe AKI causing a 33% reduction in renal perfusion at day 7 and persisting impairment until day 28 and moderate AKI causing only transient impairment compared with renal perfusion before surgery. nn Renal perfusion changes at day 7 after AKI significantly correlate with subsequent tubular injury measured at histologic examination (r = 20.54) and kidney volume (r = 0.68) 4 weeks after AKI. of renal perfusion changes could allow early detection of the problem. With arterial spin labeling (ASL), tissue perfusion can be quantified without administration of a contrast agent by using water protons as the endogenous tracer (12). For examination of perfusion in the kidney, flow-alternating inversion-recovery techniques have been used. These are based on the acquisition of one section-selective and one nonselective inversion-recovery image. Tissue perfusion is directly related to the signal intensity difference between these two images (13,14). ASL of the kidney has been used to quantify renal perfusion impairment in renal allografts in human and animal studies (15,16), in patients with renal artery stenosis (17), and in hypertensive rats after administration of iodinated contrast agent (18). Our hypothesis was that renal perfusion measured by means of ASL may serve as a noninvasive biomarker to determine the severity of AKI and to predict the extent of subsequent histologic alterations of the kidney early after the organ is damaged. Therefore, the purpose of this study was to determine if ASL can show serial changes in renal perfusion in a mouse model of ischemia-induced AKI, and to compare imaging results with those of renal histologic examination and inulin and para-aminohippuric acid (PAH) clearance. Materials and Methods Animals The local animal protection committee approved the experimental protocol. Male C57Bl/6N mice were supplied by Charles River (Sulzfeld, Germany). Mice weighing g at 8 10 weeks of age were used for all experiments. Implication for Patient Care nn Renal perfusion measured by means of arterial spin-labeling MR imaging may be useful as a noninvasive biomarker to determine the presence and severity of AKI soon after it occurs. Renal Ischemia Reperfusion Injury AKI was induced by means of transient unilateral clamping of the right renal pedicle for either 35 minutes for moderate AKI (n = 16) or 45 minutes for severe AKI (n = 11) as described previously (19). Mice were anesthetized with isoflurane, and after median laparotomy, the renal pedicle was bluntly dissected and a nontraumatic vascular clamp was applied to the renal pedicle for 35 or 45 minutes. After removal of the clamp and closure of the abdomen, the mice were returned to their cages and monitored until they were fully awake. Animals received a standard diet with free access to tap water. MR Imaging Protocol Ten animals with moderate and seven animals with severe AKI underwent MR imaging examinations before surgery (day 0) and at five different time points after surgery (days 1, 7, 14, 21, and 28) by using a 7-T small-animal MR imaging unit (Bruker, Pharmascan, PHS701, Ettlingen, Germany) and a circularly polarized volume coil (Bruker T10327 V3). Animals were anesthetized by using isoflurane inhalation and their respiration was monitored and kept constant between 30 and 50 breaths per minute during the entire examination. Respiratory-triggered, fatsaturated T2-weighted sequences were Published online before print /radiol Content code: Radiology 2014; 270: Abbreviations: AKI = acute kidney injury ASL = arterial spin labeling GFR = glomerular filtration rate PAH = para-aminohippuric acid RPF = renal plasma flow Author contributions: Guarantors of integrity of entire study, K.H., M.G., F.G.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, K.H., M.G., D.H., M. Meier, F.G.; experimental studies, K.H., M.G., S.R., M. Mengel, X.L., M. Meier, F.G.; statistical analysis, K.H., D.H., M. Meier, F.G.; and manuscript editing, K.H., M.G., S.R., D.H., M. Mengel, H.H., F.W., M. Meier, F.G. Conflicts of interest are listed at the end of this article. 118 radiology.rsna.org n Radiology: Volume 270: Number 1 January 2014

3 performed in axial and coronal planes that covered both kidneys. The coronal plane was adjusted to the long axis of the kidney. Sequence parameters were repetition time msec/echo time msec, 1500/33; averages, two; matrix, ; field of view, mm 2 ; section thickness, 1 mm. For quantification of renal perfusion, a fat-saturated flow-alternating inversionrecovery ASL sequence with an echo-planar readout was performed in a central coronal plane by using the following parameters: /16.4; inversion times, 30, 100, 200, 300, 500, 700, 1000, 1200, 1500, 2000, 3000, 5000, and 8000 msec, matrix, ; field of view, mm 2 ; section thickness, 2 mm; number of sections, one. Total time of the MR imaging examination was approximately 30 minutes. For section-selective inversion, an adiabatic frequency-selective inversion pulse (hyperbolic secant-shaped pulse) was applied with a bandwidth of 5190 Hz and a section thickness of 5 mm. The inversion section was positioned in a coronal plane matching the imaging plane. For quantification of renal perfusion, the optimized inversion time of 2000 msec was used, and 10 pairs of ASL images with selective and nonselective inversion pulses were acquired to improve signalto-noise ratio. The other inversion time values were used to quantify apparent relaxation time of renal tissue, which is necessary to calculate renal perfusion. MR Imaging Data Analysis To correct for respiratory motion, ASL images were coregistered by using Elastix (open source software, uu.nl/) and a rigid registration algorithm (20). Renal perfusion was then quantified according to an equation described by Kim et al and Kwong et al (13,14): λ ΔM 1/ T1 a 1/ T 1 1app f =, 2α M exp( TI / T ) exp( TI / T ) 0 0 1app 1a where f is renal perfusion, DM t is the signal intensity difference between control and labeled images and M 0 is the signal intensity of the fully relaxed longitudinal magnetization, T 1a is the longitudinal relaxation time of arterial blood, and T 1app is the apparent relaxation time. T 1a was set to 2200 msec (21), and T 1app was calculated on a pixel-by-pixel basis by using a least-squares fit of the section-selective inversion recovery data to the following model: T 1app 5 M 0 [122exp(2T1/T 1app )]. The blood-tissue partition coefficient for water was assumed to be 80 ml/100 g and the labeling efficiency was set to unity as described previously (22 24). The model used for calculation of renal perfusion assumes negligible delay time for the transit of arterial blood from outside the section-selective inversion slab to renal tissue in the imaging plane (13,25). Regions of interest were placed manually on M 0 images by one reader (K.H., with 3 years of experience in small-animal MR imaging of the kidney, who was blinded to the identity of the animal group) covering the entire renal cortex of each kidney (Fig 1). Regions of interest were copied to parameter maps of renal perfusion and mean values were recorded separately for each kidney. Furthermore, relative perfusion values were calculated as the percentage of renal perfusion compared with the value of the same kidney before surgery. Total kidney volume was determined separately for each kidney by means of manual segmentation of axial T2-weighted images. Renal Plasma Flow and Renal Function Measurement by PAH and Inulin Clearance Renal plasma flow (RPF) and glomerular filtration rate (GFR) were determined by measuring PAH and inulin clearance, respectively, and were performed 4 weeks after AKI in separate animals that were not examined with MR imaging (four animals with moderate AKI and six animals with severe AKI). In one animal with moderate AKI, quantification of PAH was unsuccessful, so this animal was excluded. Mice were anesthetized by using isoflurane, and contralateral nephrectomy was performed to measure only the RPF and GFR of the postischemic kidney. A catheter was placed in the bladder for urine collection, and a cannula was placed in the tail vein to infuse inulin and PAH. After a 1-hour equilibration period, urine was collected for 90 minutes and blood was drawn at the end of the infusion. Infusion solutions for the measurement of the GFR and RPF were 15% inulin, 3.75% PAH, and 1% bovine albumin in a 0.9% saline solution at a rate of 5 ml/ min. Inulin concentrations in urine and plasma aliquots were measured by using the standard colorimetric methods. Renal Histologic Evaluation Renal morphology was assessed in all animals after the last MR imaging examination (10 with moderate AKI and seven with severe AKI), 4 weeks after unilateral AKI. For organ retrieval, mice were anesthetized and perfused with ice-cold saline solution. Kidney tissue was immediately fixed in buffered formalin and embedded in paraffin. We cut 3 µm paraffin sections and stained them with periodic acid Schiff stain. Evaluation of acute tubular injury was performed by a nephropathologist with more 15 years of experience (M.Mengel), who did not know the identity of the animal group. Acute tubular injury was assessed as the percentage of tubuli affected. Statistical Analysis Software (SPSS 20; IBM Corporation, Armonk, New York) was used for statistical evaluation. Differences in relative cortical renal perfusion between groups of moderate and severe AKI were assessed by using the unpaired t test. Longitudinal changes in renal perfusion and kidney volumes between time points were determined by using analysis of variance for repeated measurements followed by post hoc multiple comparison. Adjustment for multiple comparisons was performed by using the Sidak method, and adjusted P values were calculated. The correlations of relative renal perfusion at different time points with relative kidney volume at day 28 and renal histologic results at day 28 were evaluated with linear regression and the Pearson coefficient of correlation. Differences in renal tissue perfusion measured at MR imaging, histologic differences, and differences between RPF and GFR between groups with moderate and severe AKI were analyzed by using unpaired t tests. Data were presented as Radiology: Volume 270: Number 1 January 2014 n radiology.rsna.org 119

4 Figure 1 Figure 1: Renal perfusion maps and T2-weighted images after moderate and severe unilateral AKI. Renal perfusion changes before surgery, or day 0 (d0) and at different time points (d7, d14, d28) after moderate (top row) and severe (bottom row) AKI are shown. Region of interest covering the entire renal cortex is shown in the animal with moderate AKI at d0. Note that image size and window level and width are similar for all perfusion maps. T2-weighted images of both kidneys at day 28 (right images) show (bottom) marked kidney volume loss of right kidney after severe AKI and (top) minimal volume loss after moderate AKI compared with that of contralateral normal kidney. mean 6 standard error of the mean. P values less than.05 were considered to indicate a significant difference. Results Changes in Kidney Volume Relative kidney volume compared with values before surgery was significantly decreased at day 14 after severe AKI (73% 6 4; range, 65% 92%) compared with the volume before surgery (P,.01) and further decreased to 60% 6 5 (range, 48% 89%) at day 28 (P,.01). After moderate AKI, there was a small, statistically insignificant kidney volume loss at days compared with the baseline measurement. At day 28, for example, kidney volume was reduced to 84% 6 6 (range, 51% 111%; P =.28). At day 21 and day 28, relative kidney volume was significantly lower in animals with severe than in those with moderate AKI (P,.05 and P,.01, respectively). Relative volume of the contralateral kidney without AKI significantly increased to 119% 6 3 (range, 110% 132%) after moderate and to 125% 6 3 (range, 114% 140%) after severe AKI by day 28 (P,.001; Fig 2a). Changes in Cortical Renal Perfusion Significant impairment of cortical renal perfusion measured by means of ASL compared with values before surgery was detectable at day 7 after moderate AKI (247 ml/min per 100 g 6 36, corresponding to a relative renal perfusion of 56% 6 8, with a range of 12% 92%; P,.01) and severe AKI (158 ml/min per 100 g 6 19, corresponding to a relative renal perfusion of 33% 6 6; range, 17% 60%; P,.001). After moderate AKI, renal perfusion nearly returned to baseline by day 21 (96% 6 14; range, 28% 159%). After severe AKI, perfusion impairment persisted until day 28 (46% 6 9; range, 18% 75%; P,.01; Fig 2b). Renal perfusion values in the severe AKI group were significantly lower compared with those of the group with moderate AKI at days 7 28 (P,.01, Table 1). Examples of renal perfusion maps at different time points after AKI are shown in Figure 1. Renal Function and Renal Plasma Flow (Inulin and PAH Clearance) GFR 4 weeks after severe AKI was significantly lower than that after moderate AKI (7.3 µl/min 6 2.7; range, µl/min vs 47.5 µl/min 6 5.6; range, µl/min; P,.001) in the injured kidney. Similarly, significant differences were observed for RPF, with values of 4.8 µl/min (range, µl/min) in severe and µl/min (range, µl/min) in moderate AKI (P,.001; Fig 3). Histologic Changes Transient unilateral ischemia caused histologic changes and kidney damage within 4 weeks in both the moderate and severe AKI groups. Acute tubular injury showing casts and loss of brush borders of the tubular epithelium was significantly more pronounced after severe than after moderate ischemia (5.6% vs 15.8% 6 2.4, respectively; P,.01; Fig 4) and most prominent in the corticomedullary junction. Early atrophy of single nephrons in the cortex and perivascular and interstitial inflammatory cell infiltration 120 radiology.rsna.org n Radiology: Volume 270: Number 1 January 2014

5 Table 1 Changes in Cortical Renal Perfusion Measured with MR Imaging after AKI Renal Perfusion, Moderate AKI* Renal Perfusion, Severe AKI Day Contralateral (ml/min/100 g) After AKI (ml/min/100 g) After AKI (%) Contralateral (ml/min/100 g) After AKI (ml/min/100 g) After AKI(%) P Value ( ) ( ) ( ) ( ) ( ) ( ) (41 115) ( ) ( ) (65 111), ( ) (60 488) (12 92) ( ) ( ) (17 60), ( ) ( ) (23 105) ( ) ( ) (23 56), ( ) ( ) (28 159) ( ) (87 337) (19 64), ( ) ( ) (44 145) ( ) (69 393) (18 75),.01 Note. Unless otherwise indicated, data are mean 6 standard error of the mean, with the range in parentheses. * Ischemia time of 35 minutes. Ischemia time of 45 minutes. Data are percentages compared with the same kidney before surgery, with the range in parentheses. Difference of relative renal perfusion after moderate versus severe AKI Figure 2 Figure 2: Graphs show (a) mean 6 standard error of the mean for relative changes in kidney volume and (b) cortical renal perfusion over time after moderate AKI (green curve), severe AKI (purple curve), and in the contralateral kidney (black curve). P values indicate significant differences compared with baseline (d0) after adjustment for multiple comparisons with the Sidak method. 5 P,.05, 5 P,.01, 5 P,.001. were detectable. These changes were more severe in animals with an ischemia time of 45 minutes compared with animals with an ischemia time of 35 minutes. The contralateral kidney showed unremarkable renal histologic results in both groups. Correlation of Relative Renal Perfusion with Kidney Volume and Histologic Evidence of Acute Kidney Injury Relative renal perfusion at day 7 significantly correlated with kidney volume at day 28 as a measure of chronic kidney damage (r = 0.68, P,.01). Significant correlations between relative renal perfusion at day and relative kidney volume were observed (Fig 5). Relative renal perfusion at days 7 28 was negatively correlated with the extent of acute kidney injury at day 28 (Table 2). For example, at day 7, the coefficient of correlation was r = (P,.05) and at day 28, it was r = (P,.01). Discussion We showed that ASL allows noninvasive detection and monitoring of renal perfusion impairment after ischemia-induced AKI in mice. The degree of perfusion impairment measured by means of MR imaging was related to kidney volume loss, the severity of histopathologic alterations of renal tissue, and impairment of renal function. The mouse model of unilateral ischemia reperfusion injury we used for our experiments is a well-established and characterized model of AKI (19). As we demonstrated by means of quantification of kidney volumes from T2-weighted images, significant kidney volume loss occurred in animals with severe AKI beginning at day 14. This volume loss is associated with irreversible kidney damage and may be interpreted as a sign of chronic kidney disease. After moderate AKI, only a minimal reduction of kidney volume was observed, but it was not statistically significant. As a compensatory mechanism after acute decline of renal function of one kidney, the volume of the contralateral kidney without AKI increased significantly in both groups, as would be expected on the basis of prior reports of unilateral nephrectomy (26) and of a mouse model of unilateral ischemia reperfusion injury (27). ASL allowed quantification of renal perfusion in all animals and at all time points. We used rigid registration to compensate for respiratory motion because respiratory triggering was not sufficient to reduce motion artifacts due to long repetition time of the ASL sequence. Radiology: Volume 270: Number 1 January 2014 n radiology.rsna.org 121

6 Figure 3 Figure 3: Bar graphs show GFR and RPF determined by measuring inulin and PAH clearance. (a) GFR as evaluated by measuring inulin clearance and (b) RPF evaluated by measuring PAH clearance were significantly more pronounced after severe than after moderate AKI ( 5 P,.001). Figure 4 Figure 4: Photomicrographs (magnification, 3200) show representative kidney samples of the renal corticomedullary junction stained with periodic acid Schiff stain at renal histologic examination 4 weeks after (a) moderate (35-min ischemia time) and (b) severe AKI (45-min ischemia time) compared with (c) the contralateral kidney without ischemia. Ischemia time of 35 minutes caused moderate focal AKI at corticomeduallary junction with small areas of injured nephrons showing loss of brush borders of the tubular epithelium (arrows). Ischemia time of 45 minutes caused more severe AKI, with multifocal features of tubular injury and perivascular and interstitial inflammatory cell infiltration (dashed line) and extensive cast formation at corticomeduallary junction (white stars, b). Contralateral kidney showed unremarkable renal histologic results. Oostendorp et al found in mice comparable values of renal perfusion of 481 ml/ min per 100 g by using dynamic contrastenhanced MR imaging (28). The accuracy of ASL to measure renal tissue perfusion was previously validated compared with microspheres in a swine model (22). Furthermore, ASL has been compared with PAH clearance as the standard clinical and experimental technique to determine renal plasma flow. Although a significant correlation of renal perfusion measured by means of ASL and RPF determined by measuring PAH clearance was observed (29) and, in our study, PAH clearance measurement showed differences in RPF between groups of mice with moderate and severe AKI, there are important differences between the techniques. PAH clearance is a global measurement of RPF including that of the renal cortex and the medulla of both kidneys and is defined as the volume of plasma that is delivered to the kidneys per unit of time. In comparison, ASL allows calculation of parameter maps of renal tissue perfusion normalized to 100 g of renal tissue, so that regional perfusion differences can be detected separately for each kidney. We observed that RPF measured by using PAH clearance is decreased by approximately 95%, but renal perfusion measured by means of ASL is decreased by approximately 60% in the group with severe AKI compared with that of the group with moderate AKI. There are different reasons for this discrepancy. First, renal perfusion measured by means of ASL is normalized to 100 g of renal tissue, but RPF is a global measurement that is not normalized for the size of the kidney. Because the volume loss in animals with severe AKI is more severe than that in animals with moderate AKI (40% vs 16% volume loss, respectively), the normalization of perfusion values measured by means of ASL for kidney weight contributes to the smaller difference between groups compared with PAH clearance values. Second, only renal perfusion in the cortex was determined by means of ASL, but RPF includes measurement of the cortex and the medulla. PAH clearance measurement is the standard technique in experimental nephrology, but it is invasive and requires general anesthesia for approximately three hours. The advantage of ASL is that it is noninvasive, does not require the administration of a gadolinium-based contrast agent, and can be performed in approximately 30 minutes of MR imaging. Thus, repeated examinations are possible and patients with renal insufficiency may be examined without the risk of nephrogenic systemic fibrosis. In our study, cortical renal perfusion measured by means of ASL was significantly decreased at day 7 in both groups. After moderate AKI, renal perfusion 122 radiology.rsna.org n Radiology: Volume 270: Number 1 January 2014

7 Figure 5 Figure 5: Scatterplots show correlation of relative renal perfusion at different time points with kidney volume loss at day 28. Significant correlations of relative renal perfusion are shown at (a) day 7, (b) day 14, and (c) day 28, with relative kidney volume at day 28. Results for 17 animals, 10 animals with ischemia time of 35 minutes and seven with ischemia time of 45 minutes represent moderate and severe AKI, respectively. Results indicate that impairment of renal perfusion measured at early MR imaging is predictive for loss of kidney volume and chronic kidney damage. Table 2 Correlation of Relative Renal Perfusion and Acute Tubular Injury Day Coefficient of Correlation, r P Value , , , ,.01 Note. Relative renal perfusion was measured on all days; acute tubular injury was measured on day 28. nearly recovered at later time points, whereas after severe AKI, it remained significantly reduced compared with the values before surgery until the end of the observation period at day 28. Furthermore, the degree of perfusion impairment measured by means of ASL differed significantly depending on the severity of AKI. Early changes of functional MR imaging parameters have previously been investigated in rat models of ischemiainduced AKI. By using diffusion-tensor imaging, Cheung et al observed significant restriction of diffusion and diffusion anisotropy and a reduction of the perfusion fraction of diffusion 5 hours after ischemia-induced AKI (30). Oostendorp et al found decreased oxygenation during ischemia by using blood oxygen level dependent imaging and significant reduction of filtration and renal perfusion measured by means of dynamic contrast-enhanced MR imaging 24 hours after AKI (28). Because we measured renal perfusion during a period of 4 weeks after the ischemic episode, perfusion changes associated with AKI could be evaluated over time. As suggested previously, persistent perfusion changes after injury appear to be an important factor in the pathogenesis of AKI and may contribute to impairment of renal function and development of acute tubular injury and chronic kidney disease (8 10). Thus, the degree of renal perfusion impairment measured by means of ASL at days 7 28 was significantly related to the severity of kidney volume loss and histopathologic alterations 4 weeks after AKI. At this time point, histologic examination showed severe tubular injury after severe AKI and minor changes after moderate AKI. Impairment of PAH and inulin clearance, representing GFR and global RPF, was also more pronounced 4 weeks after severe AKI compared with those from 4 weeks after moderate AKI. Significant perfusion impairment measured by means of ASL clearly preceded kidney volume loss; a reduction in cortical renal perfusion at day 7 significantly correlated with kidney volume loss at day 28. A limitation of our study was the relatively small number of study animals included and the variation in renal function among animals in the groups. Furthermore, because of local animal protection regulations, clearance measurements were performed in separate animals with moderate and severe AKI. Therefore, direct comparison between renal perfusion measured by means of ASL was only available with renal histologic results and kidney volume loss, but not with RPF determined by measuring PAH clearance. In conclusion, ASL allows noninvasive detection of the severity of AKI and monitoring of renal perfusion impairment over time in a mouse model of ischemia-induced AKI. Cortical renal tissue perfusion measured by means of ASL was related to impairment of renal function and chronic kidney damage, indicating the predictive value of early perfusion impairment for chronic kidney disease. Thus, ASL may be very valuable for clinical follow-up of patients at risk for AKI and for drug development in experimental renal disease models. Practical application: Renal perfusion changes measured by means of ASL may be an early and noninvasive predictor of the severity of AKI and the subsequent risk of chronic kidney disease. In the field of experimental animal studies, renal perfusion measured by means of ASL could be of great interest for drug development in vascular medicine, because therapeutic effects Radiology: Volume 270: Number 1 January 2014 n radiology.rsna.org 123

8 of new compounds could be detected early and could be monitored over time without sacrificing animals. Moreover, monitoring of renal perfusion by means of ASL could be beneficial in clinical practice, because renal perfusion impairment is an early and often critical parameter in the pathogenesis of AKI and chronic kidney disease. Early diagnosis of AKI would facilitate therapy decisions to prevent further kidney damage by reducing nephrotoxic medication. Acknowledgments: We thank Petra Gürtler, Herle Chlebusch, and Hayet Richi for excellent technical assistance. Disclosures of Conflicts of Interest: K.H. No relevant conflicts of interest to disclose. M.G. No relevant conflicts of interest to disclose. S.R. No relevant conflicts of interest to disclose. D.H. No relevant conflicts of interest to disclose. M. Mengel Financial activities related to the present article: none to disclose. Financial activities not related to the present article: consultancy for Bertech Pharma, payment for lectures from Recline Pharma, patents and royalties from Invitrogen, travel expenses from Novartis Canada and Astellas Canada. Other relationships: none to disclose. X.L. No relevant conflicts of interest to disclose. H.H. Financial activities related to the present article: none to disclose. Financial activities not related to the present article: board membership for Bayer, CVRx and Alexion; payment for lectures from Bayer CVRx, Takeda, and Astra Zeneca. Other relationships: none to disclose. F.W. Financial activities related to the present article: none to disclose. Financial activities not related to the present article: grants/grants pending from Siemens and Pro Medicus. Other relationships: none to disclose. M. Meier No relevant conflicts of interest to disclose. F.G. No relevant conflicts of interest to disclose. References 1. Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007;11(2):R Srisawat N, Kellum JA. Acute kidney injury: definition, epidemiology, and outcome. Curr Opin Crit Care 2011;17(6): Mehta RL, Pascual MT, Soroko S, et al. Spectrum of acute renal failure in the intensive care unit: the PICARD experience. Kidney Int 2004;66(4): Metnitz PG, Krenn CG, Steltzer H, et al. Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients. Crit Care Med 2002;30(9): Druml W. Long term prognosis of patients with acute renal failure: is intensive care worth it? Intensive Care Med 2005;31(9): Levy MM, Macias WL, Vincent JL, et al. Early changes in organ function predict eventual survival in severe sepsis. Crit Care Med 2005;33(10): Wen X, Murugan R, Peng Z, Kellum JA. Pathophysiology of acute kidney injury: a new perspective. Contrib Nephrol 2010;165: Sutton TA, Fisher CJ, Molitoris BA. Microvascular endothelial injury and dysfunction during ischemic acute renal failure. Kidney Int 2002;62(5): Molitoris BA, Sutton TA. Endothelial injury and dysfunction: role in the extension phase of acute renal failure. Kidney Int 2004;66(2): Biology of acute renal failure: therapeutic implications. Kidney Int 1997;52(4): Basile DP. The endothelial cell in ischemic acute kidney injury: implications for acute and chronic function. Kidney Int 2007;72(2): Martirosian P, Boss A, Schraml C, et al. Magnetic resonance perfusion imaging without contrast media. Eur J Nucl Med Mol Imaging 2010;37(Suppl 1):S52 S Kim SG. Quantification of relative cerebral blood flow change by flow-sensitive alternating inversion recovery (FAIR) technique: application to functional mapping. Magn Reson Med 1995;34(3): Kwong KK, Chesler DA, Weisskoff RM, et al. MR perfusion studies with T1-weighted echo planar imaging. Magn Reson Med 1995;34(6): Lanzman RS, Wittsack HJ, Martirosian P, et al. Quantification of renal allograft perfusion using arterial spin labeling MRI: initial results. Eur Radiol 2010;20(6): Wang JJ, Hendrich KS, Jackson EK, Ildstad ST, Williams DS, Ho C. Perfusion quantitation in transplanted rat kidney by MRI with arterial spin labeling. Kidney Int 1998;53(6): Fenchel M, Martirosian P, Langanke J, et al. Perfusion MR imaging with FAIR true FISP spin labeling in patients with and without renal artery stenosis: initial experience. Radiology 2006;238(3): Liu YP, Song R, Liang Ch, Chen X, Liu B. Arterial spin labeling blood flow magnetic resonance imaging for evaluation of renal injury. Am J Physiol Renal Physiol 2012;303(4):F551 F Lu X, Li N, Shushakova N, et al. C57BL/6 and 129/Sv mice: genetic difference to renal ischemia-reperfusion. J Nephrol 2012;25(5): Klein S, Staring M, Murphy K, Viergever MA, Pluim JP. elastix: a toolbox for intensity-based medical image registration. IEEE Trans Med Imaging 2010;29(1): Dobre MC, Uğurbil K, Marjanska M. Determination of blood longitudinal relaxation time (T1) at high magnetic field strengths. Magn Reson Imaging 2007;25(5): Artz NS, Wentland AL, Sadowski EA, et al. Comparing kidney perfusion using noncontrast arterial spin labeling MRI and microsphere methods in an interventional swine model. Invest Radiol 2011;46(2): Boss A, Martirosian P, Graf H, Claussen CD, Schlemmer HP, Schick F. High resolution MR perfusion imaging of the kidneys at 3 Tesla without administration of contrast media. Rofo 2005;177(12): Martirosian P, Klose U, Mader I, Schick F. FAIR true-fisp perfusion imaging of the kidneys. Magn Reson Med 2004;51(2): Buxton RB, Frank LR, Wong EC, Siewert B, Warach S, Edelman RR. A general kinetic model for quantitative perfusion imaging with arterial spin labeling. Magn Reson Med 1998;40(3): Lopes GS, Lemos CC, Mandarim-De-Lacerda CA, Bregman R. Effect of unilateral nephrectomy on renal function of diabetic rats. Histol Histopathol 2004;19(4): Jang HS, Kim JI, Kim J, Na YK, Park JW, Park KM. Bone marrow derived cells and reactive oxygen species in hypertrophy of contralateral kidney of transient unilateral renal ischemia-induced mouse. Free Radic Res 2012;46(7): Oostendorp M, de Vries EE, Slenter JM, et al. MRI of renal oxygenation and function after normothermic ischemia-reperfusion injury. NMR Biomed 2011;24(2): Ritt M, Janka R, Schneider MP, et al. Measurement of kidney perfusion by magnetic resonance imaging: comparison of MRI with arterial spin labeling to para-aminohippuric acid plasma clearance in male subjects with metabolic syndrome. Nephrol Dial Transplant 2010;25(4): Cheung JS, Fan SJ, Chow AM, Zhang J, Man K, Wu EX. Diffusion tensor imaging of renal ischemia reperfusion injury in an experimental model. NMR Biomed 2010;23(5): radiology.rsna.org n Radiology: Volume 270: Number 1 January 2014