| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the Department of Biological Sciences, Pittsburgh NMR Center for Biomedical Research, Carnegie Mellon University, 4400 Fifth Ave, Pittsburgh, PA 15213. From the 2002 RSNA scientific assembly. Received November 27, 2002; revision requested January 30, 2003; final revision received October 7; accepted October 20. Supported by contract grants P41EB-001977 and R01EB/AI-00318 from the National Institutes of Health. Address correspondence to C.H. (e-mail: chienho@andrew.cmu.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Spin-echo diffusion-weighted MR imaging measurements were performed in 20 normal rats and nine rats that had undergone transplantation (six rats had received allografts; three had received isografts) at 7 T. To evaluate the effect of alteration in blood flow and water transport function, angiotensin II was infused in six normal rats and a series of spin-echo diffusion-weighted MR images was obtained at five time points. Transplanted kidneys were monitored by obtaining spin-echo diffusion-weighted MR images and gradient-echo MR images every 2 hours for 8 hours on postoperative day 4. Statistical analysis was performed with repeated-measures multivariate analysis of variance and the paired t test.
RESULTS: No significant differences in ADC values were observed between right and left kidneys in all three orthogonal directions; however, a small difference was observed between the cortex and medulla. ADC values in the cephalocaudal and mediolateral directions were higher than those in the anteroposterior direction (P < .01 for all). ADC values in the cortex and medulla decreased significantly (by >35%, P < .01) during angiotensin II–induced reduction in renal blood flow. No significant signal intensity change was observed between native and transplanted kidneys on gradient-echo MR images. Allografts exhibited decreased ADC values (P < .01) and isografts exhibited similar ADC values compared with native kidneys.
CONCLUSION: These findings suggest that reproducible renal ADC maps can be obtained in rats by using spin-echo diffusion-weighted MR imaging at 7 T. Spin-echo diffusion-weighted MR imaging may have potential as a noninvasive tool for monitoring early graft rejection after kidney transplantation.
© RSNA, 2004
Index terms: Animals • Experimental study • Kidney, MR, 81.121411, 81.121412, 81.121415, 81.121416, 81.12143 • Kidney, transplantation, 81.4551, 81.4552 • Magnetic resonance (MR), diffusion study, 81.12144
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Attempts have been made to quantify the combined effects of capillary perfusion and water diffusion in vivo by using an apparent diffusion coefficient (ADC) (2,3). Quantitative ADC mapping can be achieved by recording a series of diffusion-weighted MR images with different diffusion-sensitizing gradient factors (b values) (2,4). Because the ADC directly reflects the microenvironment (ie, the brownian motion and capillary perfusion) of the diffusing water molecules in tissue, this modality can yield information on the normal structure of tissues and on organ disease.
The kidney is a particularly interesting organ to study with diffusion-weighted MR imaging because of its high blood flow and water transport functions (4). Transport function in the nephrons and countercurrent mechanisms in the loop of Henle may have a substantial influence on ADC values in the kidneys. Diffusion-weighted MR imaging of the kidneys could provide a measure of perfusion according to the intravoxel incoherent motion model (5).
Unfortunately, diffusion-weighted MR imaging of kidneys in vivo has been technically challenging because the kidney is subject to a large amount of respiration-induced motion. Breath holding is often required to reduce respiratory motion during diffusion-weighted MR imaging of the kidney in humans. However, this approach limits the acquisition time, and both the signal-to-noise ratio and the spatial resolution can be compromised.
Development of suitable pulse sequences for diffusion-weighted MR imaging of the kidney is an active area of research. Until recently, only a few studies have involved measurement of water diffusion in the kidneys (4–9). Some of these studies (7,9) involved the use of echo-planar imaging techniques to minimize the effect of motion in renal diffusion-weighted MR imaging because echo-planar imaging has a clear advantage in that it has no reliance on phase consistency between excitations.
However, some serious drawbacks have been reported for the echo-planar diffusion imaging techniques. For example, section misregistrations have been described between echo-planar images acquired with different b values (9). Other disadvantages of echo-planar diffusion-weighted MR imaging are limited spatial resolution (because of B0 field inhomogeneities) and strong magnetic susceptibility effects (10) that result in geometric distortion artifacts that tend to be more severe with increasing b values. At high magnetic fields, which are typically used in animal imaging units, the strong susceptibility effects at the boundaries between the kidney and the ambient air preclude the use of echo-planar imaging for renal imaging in animals.
Although conventional spin-echo diffusion techniques may not be ideal for clinical renal diffusion-weighted MR imaging due to a long acquisition time and bulk motion artifacts, in animal studies, the use of large gradients and respiratory gating may allow the acquisition of high-spatial-resolution renal diffusion-weighted MR images with fewer susceptibility artifacts.
Thus, the purposes of our study were (a) to investigate the feasibility of obtaining reproducible ADC maps of normal rat kidney by using respiratory-triggered spin-echo diffusion-weighted MR imaging, (b) to investigate the sensitivity of ADC maps in the evaluation of renal blood flow, and (c) to use this technique to monitor acute rejection in transplanted rat kidney.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Before diffusion-weighted MR imaging examinations were performed, 20 normal rats and nine rats that had undergone transplantation (mean weight, 262 g ± 18.7 [SD]; weight range, 240–300 g) were anesthetized with inhaled methoxyflurane (Metofane; Medical Developments, Melbourne, Australia). The rats were orally intubated and then mechanically ventilated with 2% isoflurane (Isoflo; Abbott Laboratories, North Chicago, Ill) in a 2:1 oxygen–nitrous oxide mixture administered through an endotracheal tube. The ventilation rate was 60 cycles per minute with a 2.5-mL stroke volume. The body temperature of the rats was maintained at 37°C by using a heated water pad placed on the back of the head and the chest.
Renal blood flow was altered in six normal rats with the infusion of 300 pmol/min of angiotensin II (Sigma Chemical, St Louis, Mo) (11) for 30 minutes through a jugular vein catheter. This dosage of angiotensin II is known to reduce renal blood flow in rats by approximately 37% (11). The blood pressures of the six rats were monitored during the experiments by one of the authors (D.Y.), who used a femoral arterial catheter and a pressure transducer (Becton Dickinson Critical Care Systems, Singapore).
Both the native and the transplanted kidneys of the nine rats that had undergone transplantation (six rats had received allografts; three had received isografts) were monitored by using spin-echo diffusion-weighted MR imaging. All kidney transplantations were performed by an experienced microtransplant surgeon (Q.Y.). Brown Norway rats were used as recipients throughout the experiment and as donors for syngeneic transplants (isografts), and Dark Agouti rats were used as donors for allogeneic transplants (allografts).
Before transplantation, the animals were anesthetized with inhaled methoxyflurane. Left nephrectomy was performed in the recipient rats before left kidney transplantation. The graft (ie, the donated left kidney) was flushed with lactated Ringer solution (Abbott Laboratories) containing 1,000 units per milliliter of heparin (Elkins-Sinn, Cherry Hill, NJ) and then stored in the same solution at 4°C. The ischemia time was about 25 minutes (12,13).
MR Imaging
All examinations were performed with a 7-T, 21-cm-horizontal-bore AVANCE DRX MR instrument (Bruker Instruments, Billerica, Mass) equipped with a 72-mm-diameter volume coil and 12-cm shielded gradients with a maximum gradient strength of 200 mT/m. Pilot images (repetition time msec/effective echo time msec, 400/20; number of acquisitions, four) and T2-weighted (1,600/80; number of acquisitions, four) fast spin-echo images were initially acquired so that we could determine the optimal coronal plane. A coronal plane with a 2-mm section thickness passing through the central parts of both kidneys was chosen as the final section position for the diffusion study.
To minimize bulk motion artifacts, data acquisitions were gated to respiration by using a mechanical switch activated by the ventilator piston at inspiration. A posttrigger delay time of 700 msec in the pulse sequence allowed images to be gated to the middle phase of expiration. A respiratory-gated spin-echo diffusion-weighted sequence with six b values (5, 20, 42, 72, 142, and 260 sec/mm2) was used to perform diffusion-weighted MR imaging in all normal and transplanted rat kidneys.
The imaging parameters for the spin-echo diffusion sequence were as follows: interval between centers of diffusion gradients (
), 20 msec; diffusion duration (
), 6 msec; echo time, 37.7 msec; field of view, 6.4 x 6.4 cm; matrix size, 128 x 128; section thickness, 2 mm; and number of acquisitions, three. The repetition time at diffusion-weighted MR imaging was limited by the respiratory rate of 60 breaths per second and was 1,000 msec.
For evaluation of anisotropic diffusion in the kidneys of normal rats, a series of diffusion-weighted MR images was acquired with the diffusion-sensitizing gradient oriented cephalocaudally, mediolaterally, and anteroposteriorly for readout, phase, and section directions, respectively; corresponding directional ADC values (ie, ADCc, ADCm, ADCa) were then calculated. Imaging time for acquisition of an ADC data set was 38.4 minutes for one diffusion direction.
In the rats with renal blood flow that had been altered by angiotensin II administration, diffusion-weighted MR images (matrix size, 64 x 64; number of acquisitions, two) were acquired with the diffusion-sensitizing gradient in the phase direction at five time points. Imaging time was 12.8 minutes per ADC data set. The five time points were as follows: before the infusion of angiotensin II, 17 minutes after the infusion was started, immediately after the infusion was stopped, 30 minutes after the infusion was stopped, and 60 minutes after the infusion was stopped.
At imaging of all normal rats, a water phantom was placed within the field of view as a control. In the group of nine rats that had undergone transplantation, the spin-echo diffusion-weighted MR images in the phase (mediolateral) direction—along with gradient-echo images (100/7.3; number of acquisitions, 32; matrix, 256 x 256; field of view, 6.4 cm) of the kidneys—were acquired every 2 hours for a total of 8 hours on postoperative day 4. For the gradient-echo sequence, the flip angle was adjusted to maximize the total signal from the whole section. The b values, matrix size, and field of view for the spin-echo diffusion-weighted MR images acquired in these rats were the same as those for the spin-echo diffusion-weighted MR images acquired in the group of normal rats.
Data and Statistical Analysis
The ADC maps were generated by means of pixel-by-pixel linear regression analysis of the natural log of signal intensity versus b values (14). Data analysis was performed by using regions of interest at different locations (cortex and medulla) within the kidneys. Two irregular regions of interest (one over the cortex and one over the medulla) of at least 50 pixels were drawn by one of the authors (D.Y.). All values were reported as means ± SDs. A repeated-measures multivariate analysis of variance was used to examine the influence of location (cortex vs medulla; left kidney vs right kidney) on ADC values and the sensitivity of ADC values to each of the three orthogonal directions of diffusion (ie, cephalocaudal, mediolateral, and anteroposterior). Statistical analyses of ADC measurements for control versus angiotensin II infusion states and for native versus transplanted kidney were performed with a paired t test. P 
.05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The difficulties associated with minimizing macroscopic motion in live animals and humans have placed practical limits on the application of quantitative diffusion measurements to many in vivo situations. A common approach is to use fast imaging techniques such as echo-planar imaging. Another remedy consists of measuring and correcting the phase errors in the raw data by using the navigator-echo technique before image calculation (16). This method allows the use of conventional spin-echo sequences, yielding high signal-to-noise ratios.
In contrast to clinical examinations in humans, in animal experiments, the overall imaging time is not a limiting parameter; furthermore, respiration rate and volume can easily be controlled with mechanical ventilation. In this case, the repetition time and the overall imaging time are limited by the ventilation rate. High-spatial-resolution ADC maps, which compare favorably with published renal ADC maps obtained by using the diffusion-weighted echo-planar imaging technique, have been obtained by using spin-echo diffusion-weighted MR imaging (4,6,9).
The kidney is a challenging organ to study with diffusion-weighted imaging since it consists of many different structural components and the ADC is influenced by flow dynamics. Some investigators have reported higher values in the medulla than in the cortex (4,9), whereas others have reported the opposite (7,8). The range of ADC values reported for the cortex includes values as low as 1.38 x 10–3 mm2/sec (17) to values as high as 5.1 x 10–3 mm2/sec (4), with the majority of the values near 2.5 x 10–3 mm2/sec (7–9). Comparison of these results is difficult because of the different imaging strategies employed in these studies. For example, the most important consideration for an optimal echo-planar imaging sequence at diffusion-weighted MR imaging is that the imaging be limited to a single breath hold. Since at least two b values are required, diffusion experiments always involve the acquisition of multiple images. This can lead to some section misregistration at echo-planar imaging–based diffusion-weighted MR imaging when several acquisitions with different b values are used.
In this study, we used a respiratory-gated spin-echo technique to measure ADC values in rat kidneys. The mean measured ADC values ranged from 2.39 to 3.09 x 10–3 mm2/sec for the cortex and from 2.07 to 2.94 x 10–3 mm2/sec for the medulla. The ADC values that we measured in normal kidneys were larger than those measured in the water phantom. We determined that the water phantom had a diffusion coefficient value of 2.65 x 10–3 mm2/sec ± 0.06 at body temperature; this is in good agreement with another reported pure water diffusion coefficient value (2.5 x 10–3 mm2/sec) (2). These results are not surprising and indicate that other factors must have influenced our measurement of water diffusion (ie, ADC value) in the kidney.
Yamada et al (5) calculated the true diffusion coefficient and perfusion fraction in 78 patients and found that perfusion contributed highly to the ADC values. However, the diffusion of water in the kidney is complicated because of the influence of many other factors. For example, glomerular filtration, tubular resorption, tubular secretion, and urine flow may cause some degree of signal variability due to slow coherent flow (9). These factors may account for the ADC values of the kidneys being higher than the diffusion coefficient of water at body temperature.
To the best of our knowledge, there is only one previous report of renal diffusion-weighted MR imaging in the rat (6). That report indicated that higher ADC values were obtained in the medulla than in the cortex; this implies that perfusion is not the only mechanism acting on the diffusion measurement. In the present study, we found that higher ADC values were observed in the cortex in the 14 normal rats. In the additional six normal rats given an infusion of angiotensin II, decreased ADC values in the kidneys were observed during the infusion, and the ADC values returned to normal 30 minutes after the infusion was stopped. Our results strongly suggest that high ADC values in the renal cortex are a direct consequence of blood perfusion. The mechanism for decreased ADC values in the medulla is generally considered to be the mesangial contraction caused by angiotensin II that reduces the glomerular filtration surface area and the filtration coefficient. The end result of the effect of angiotensin II on mesangial cells is a decline in glomerular capillary plasma flow, which leads to a decrease in glomerular filtration rate and tubular urine flow, so the ADC values of the medulla are decreased.
Results of many studies involving the use of the echo-planar imaging sequence have implied that the structural characteristics of the kidney result in anisotropic ADC values (4,7,8,17). In our diffusion study, in which we used coronal section selection, we observed that the in-plane ADC values were higher than those in the section (anteroposterior) direction. The reason for the higher ADC values in the cephalocaudal and mediolateral directions may be the local structure of the kidney (ie, that there is more small vascular and tubular flow along these two directions than along the section direction). Siegel et al (8) reported that renal arterial pulsation waves are expected to propagate along the right-left direction of the diffusion-sensitizing gradient, which may cause elevated ADC values.
The range of maximal b values for obtaining ADC values for kidneys has been reported to be from 55 to 1,300 sec/mm2 (4–9,17,18). It is important to choose the proper b value for diffusion-weighted MR imaging. The b values used in this study were chosen so that they would yield a sufficient signal-to-noise ratio and enable us to examine the ADC in the flow-sensitive regime (2).
In some MR images obtained in the present study, a white band was observed on the upper pole of the normal and transplanted kidneys. This chemical shift artifact can be a problem with high-field-strength MR imaging. Two methods that can help alleviate the problem are (a) using a wider receiver bandwidth and (b) suppressing the signal from fat by using either inversion recovery to nullify the fat signal or frequency-selective fat saturation. Both wider bandwidth and inversion recovery will lead to a reduced signal-to-noise ratio, and fat saturation often appears nonuniform across the section or imaging volume. However, in the present study, because chemical shift artifacts were seen in only some rats, and for consistency in the imaging protocol across the entire study, no attempt was made to suppress these artifacts. In the cases in which chemical shift artifacts were present, ADC values were determined with irregular regions of interest in the cortex and medulla that avoided the problematic regions.
Practical application: In the kidney, structure and function are closely linked. Blood and urine are transient. Water transport is the predominant phenomenon in the kidney owing to the kidney’s major role in both water reabsorption and concentration-dilution functions. Thus, measurement of the diffusion characteristics of a kidney may provide useful insights into the events leading to acute renal rejection.
In the kidney transplant model that we chose, acute renal rejection occurs within a few days after kidney transplantation (12,13). A significant decrease in ADC values was observed in the allografts compared with those in the native kidneys. However, no difference in the ADC values was found between the isografts and the native kidneys. These results indicate that diffusion-weighted MR imaging may be a useful tool for evaluating acute allograft rejection. Diffusion-weighted MR imaging involving the use of small b values can be used to measure ADC values that reflect brownian motion and capillary perfusion. According to our results with angiotensin II infusion, the ADC values in the mediolateral direction of the kidney are decreased when the renal blood flow is markedly reduced. Clearly, renal perfusion might play an important role in the reduction of ADC values. Due to the complex architecture of the kidney, it would be useful in the future to compare the ADC values in the in-plane directions with those in the anteroposterior direction by using the same diffusion parameters used in our rat transplant model.
Although renal biopsy with histopathologic assessment is the present standard of reference for diagnosing graft rejection, it is an invasive procedure, has associated risks, and is prone to sampling errors. Thus, noninvasive techniques for detecting graft rejection are desirable. Several MR imaging techniques have been used to detect renal graft rejection in both experimental animal models (12,13,19–21) and clinical studies (22–24). In particular, MR angiography has been shown to be a useful method for assessing abnormalities of renal arteries and veins in humans (24).
In our laboratory, we have applied a number of MR imaging techniques to the detection of renal graft rejection after kidney transplantation in our rat models. Our previous MR imaging measurements with arterial spin labeling showed that the renal cortical perfusion rate in allograft kidneys in acute rejection is greatly reduced compared with that in isograft kidneys (19). We have also found that there is an excellent correlation between the accumulation of macrophages labeled with dextran-coated ultrasmall superparamagnetic iron oxide particles and the MR imaging signal intensity reduction in kidneys during rejection (13,20). We previously found that allograft kidneys can be differentiated from normal native and isograft kidneys by using first-pass dynamic MR imaging techniques on day 4 after kidney transplantation in a rat model (involving Dark Agouti donor and Brown Norway recipient rat pairs) (12). Results of the present study show that kidney transplant rejection has an effect on ADC values. Thus, there are a number of promising MR imaging methods that can be developed for detection of graft rejection clinically.
In the present study, reproducible high-spatial-resolution renal ADC maps were obtained in rats at 7 T by using a spin-echo diffusion-weighted MR imaging sequence. Motion was well controlled with respiratory gating. Reduction in renal blood flow with the use of a renal vasoconstrictor, angiotensin II, resulted in a concomitant decrease in ADC values. Spin-echo diffusion-weighted MR imaging provides a useful method for measurement of water mobility and perfusion in rat kidneys. Diffusion-weighted MR imaging offers a potential noninvasive method for monitoring the early signs of renal rejection after kidney transplantation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, D.Y., C.H.; study concepts and design, D.Y., D.S.W., C.H.; literature research, D.Y.; experimental studies, D.Y., Q.Y.; data acquisition, D.Y., Q.Y.; data analysis/interpretation, all authors; statistical analysis, D.Y., T.K.H.; manuscript preparation, D.Y.; manuscript definition of intellectual content, revision/review, and final version approval, all authors; manuscript editing, D.Y., D.S.W., T.K.H., C.H.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Han, W. Xiao, Y. Xu, J. Wu, Q. Wang, H. Wang, M. Zhang, and J. Chen The significance of BOLD MRI in differentiation between renal transplant rejection and acute tubular necrosis Nephrol. Dial. Transplant., August 1, 2008; 23(8): 2666 - 2672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. C. Thoeny, D. Zumstein, S. Simon-Zoula, U. Eisenberger, F. De Keyzer, L. Hofmann, P. Vock, C. Boesch, F. J. Frey, and P. Vermathen Functional Evaluation of Transplanted Kidneys with Diffusion-weighted and BOLD MR Imaging: Initial Experience Radiology, December 1, 2006; 241(3): 812 - 821. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
