Evaluation of Transplanted Kidneys Using Blood Oxygenation Level Dependent MRI at 3 T: A Preliminary Study

Transcription

1 Genitourinary Imaging Original Research Park et al. OLD MRI of Renal llografts Genitourinary Imaging Original Research Sung Yoon Park 1 Chan Kyo Kim 1 yung Kwan Park 1 Wooseong Huh 2 Sung Ju Kim 3 ohyun Kim 4 Park SY, Kim CK, Park K, Huh W, Kim SJ, Kim Keywords: blood oxygenation level dependent (OLD) MRI, functional MRI, kidney, renal transplantation DOI:1.2214/JR Received May 2, 211; accepted after revision September 23, 211. This research was supported by a 29 grant from the Korean Society of Radiology provided by ayer Schering Pharma, Korea. 1 Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, 5 Ilwon-dong, Kangnam-gu, Seoul , Republic of Korea. ddress correspondence to C. K. Kim (chankyokim@skku.edu). 2 Department of Nephrology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea. 3 Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea. 4 Department of Radiology, Mayo Clinic College of Medicine, Rochester, MN. JR 212; 198: X/12/ merican Roentgen Ray Society Evaluation of Transplanted Kidneys Using lood Oxygenation Level Dependent MRI at 3 T: Preliminary Study OJECTIVE. The objective of our study was to investigate the feasibility and reproducibility of blood oxygenation level dependent (OLD) MRI using different gradient echoes at 3 T in patients with renal allografts and healthy volunteers and to evaluate whether OLD MRI can be used to distinguish between cases of acute allograft rejection and normally functioning allografts. SUJECTS ND METHODS. OLD MRI at 3 T was performed of eight patients with normal allografts, four patients with acute allograft rejection, and 1 healthy volunteers. Multiple fast-field echo sequences were performed at gradient echoes of 8, 16, and 2 to obtain T2*-weighted images. The reproducibility of OLD MRI was evaluated in patients with normal allografts. RESULTS. Cortical and medullary R2* values were not significantly different between healthy volunteers and patients with normal allografts, but medullary R2* values were significantly greater than cortical R2* values in both groups for all three protocols (p <.1). Medullary R2* values were significantly lower in cases of acute allograft rejection than in normal allografts for all three protocols (p <.1). The mean difference in cortical or medullary R2* values was 3.8% or less in all protocols. CONCLUSION. OLD MRI performed using different gradient echoes at 3 T is feasible and reproducible in patients with renal allografts and can show significant changes in medullary oxygenation in patients with acute rejection. renal transplant is one of the most effective treatment options for impaired renal function and can improve the quality of life and the survival rate of patients with chronic renal failure [1 3]. However, acute allograft rejection is still a common complication of renal transplant and affects long-term graft survival, although the development of more effective and less toxic immunosuppressive protocols has improved long-term outcomes [4]. ccurate early detection and treatment of acute allograft rejection are crucially important for graft survival. Currently, percutaneous renal biopsy is a cornerstone of the workup for allograft dysfunction. lthough the introduction of sonographic guidance has improved the renal biopsy technique, it is still an invasive technique that can result in several complications including bleeding, arteriovenous fistula, infection, or even graft loss [5, 6]. To detect and evaluate renal allograft dysfunction, the development of a noninvasive method alternative to biopsy may reduce the incidence of bi- opsy-related complications and result in earlier initiation of treatment. Since the first assessment of the renal oxygenation state using blood oxygenation level dependent (OLD) MRI in an animal model in 1996 [7], several studies have investigated the potential of OLD MRI to identify various pathologic conditions of the kidneys such as renal artery stenosis [8, 9], ureteral obstruction [1], diabetes mellitus [11], renal allograft rejection, and acute tubular necrosis [12 14]. OLD MRI is a noninvasive method that can assess regional tissue oxygen concentrations on the basis of the paramagnetic properties of deoxyhemoglobin as an endogenous contrast agent [7]. decrease in the oxygen concentration in the regional tissue results in an increase in the concentration of deoxyhemoglobin, which results in increased magnetic spin dephasing of blood water protons and decreased signal intensity on T2*-weighted MRI sequences. ccordingly, the apparent relaxation rate, denoted as R2* (= 1 / T2*), is inversely elevated as tissue oxygenation 118 JR:198, May 212

2 OLD MRI of Renal llografts decreases [15]. In a normally functioning kidney, the oxygen concentration of the cortex is greater than that of the medulla; this corticomedullary difference in oxygenation states can be detected using OLD MRI [16, 17]. In cases of acute allograft rejection, the bioavailable oxygen concentration in the medulla increases, and OLD MRI can detect the significant decrease in medullary R2* [13, 14]. ecause OLD MRI is affected by magnetic field strength, a high magnetic field strength allows better detection of the paramagnetic effect of deoxyhemoglobin in the kidney [18]. Several studies have shown that OLD MRI at 3 T resulted in better delineation of the cortex and medulla of the kidneys compared with MRI at a magnetic field strength of 1.5 T [9, 16, 17]. However, to our knowledge, 3 T has not been used to evaluate the intrarenal oxygenation states in renal allografts. We therefore prospectively investigated the feasibility and reproducibility of OLD MRI under different gradient echoes at 3 T in patients with renal allografts, compared these findings with those of healthy volunteers with normal kidneys, and evaluated whether OLD MRI at 3 T can distinguish between acute allograft rejection and normal allograft cases. Subjects and Methods Patients This prospective study was approved by the ethics committee of our institute. Written informed consent was obtained from healthy volunteers and from all patients who underwent renal transplant. etween May 29 and pril 21, 28 subjects, comprising 1 healthy volunteers and 18 consecutive patients who received a renal allograft within 3 months, were referred for a renal OLD MRI examination at 3 T. Ten healthy age- and sex-matched volunteers (five men and five women ranging in age from 2 to 6 years; median age, 44.5 years; age range, years) without a history of chronic diseases, such as primary renal disease or diabetes mellitus, and who were not taking medications that could influence renal function were recruited. Renal allograft function was determined by two medical and surgical nephrologists with 15 years of experience, respectively. In patients with normally functioning allografts, serum creatinine level was within normal limits (creatinine < 1.3 mg/dl) after renal transplant. Of the 18 patients with a renal allograft, one patient was excluded from the study because a mass was detected in the allograft on a routine sonographic examination just before OLD MRI was performed; the mass was eventually diagnosed as posttransplant lymphoproliferative disease on the basis of histopathologic findings. The remaining Fig. 1 Flowchart of study group. PTLD = posttransplant lymphoproliferative disease. Normal functioning allografts, n = 8 Pathologic finding: cute rejection, n = 4 17 patients consisted of eight subjects with normally functioning allografts and nine subjects with dysfunctioning allografts. For the nine subjects with dysfunctioning allografts, sonography-guided renal biopsy was performed and the pathologic results were as follows: normal findings (n = 2), minimal tubulointerstitial changes (n = 3), and acute rejection (n = 4). The four cases of acute rejection were grade I in two subjects, grade I in one subject, and grade II in one subject. The five subjects with pathologically normal findings or minimal tubulointerstitial changes were excluded from our study. Therefore, a total of 22 subjects were finally included in our study: 1 healthy volunteers, eight patients with normally functioning allografts, and four patients with acute allograft rejection (Fig. 1). Four of the eight patients with normally functioning allografts underwent two sessions of OLD MRI twice, separated by a 1-month interval, to evaluate the reproducibility of OLD MRI. Renal allograft recipients, n = 18 Enrolled subjects, n = 28 Dysfunctioning allografts, n = 1 Pathologic findings: Normal, n = 2 Tubulointerstitial change, n = 3 ge- and sex-matched healthy volunteers, n = 1 No MRI Pathologic finding: PTLD, n = 1 Sonographic examinations were performed routinely by two genitourinary radiologists with 6 and 8 years of dedicated experience and training in genitourinary imaging, respectively, before OLD MRI to exclude hydronephrosis, a perirenal fluid collection, a vascular occlusion, and a tumor. If acute allograft rejection was suspected clinically by the two nephrologists, OLD MRI was performed before the renal allograft biopsy. The interval between the MRI examination and biopsy ranged from 1 to 6 hours. No oral intake or IV fluid administration was allowed at least 8 hours before the MR examination because hydration status can influence the R2* value on OLD MRI [19]. Three or four 18-gauge core biopsy samples were obtained at biopsy by the two genitourinary radiologists. Pre- and Posttransplant Regimens ll patients had received induction therapy using basiliximab with or without antithymocyte globulin before transplant. fter the transplant, three of TLE 1: Characteristics of Healthy Volunteers, Patients With Normally Functioning llografts, and Patients With Dysfunctioning llografts llograft Recipients Parameter Healthy Volunteers (n = 1) Normal Function (n = 8) cute Rejection (n = 4) Donor, no. of grafts Cadaveric 1 1 Living related 7 3 ge (y), mean ± SD 44.9 ± ± ± 9.3 Creatinine (mg/dl), mean ± SD.75 ± ± ±.13 Hemoglobin (g/dl), mean ± SD ± ± ±.73 Hematocrit (%), mean ± SD ± ± ± 2.2 JR:198, May

3 Park et al. 17 patients received triple immunosuppression therapy (methylprednisolone plus mycophenolic acid or mycophenolate mofetil plus tacrolimus or cyclosporine), and 1 of 17 patients received two immunosuppressants (methylprednisolone or mycophenolic acid or mycophenolate mofetil plus tacrolimus or cyclosporine). MRI Technique Renal OLD MRI was performed with a 3-T clinical MR system (Intera chieva, Philips Healthcare) using torso phased-array coils with 16 elements. OLD MRI was performed using a multiple fast-field echo sequence and three different gradient-echo protocols to obtain T2*-weighted images in the coronal planes. The three gradient echoes used in the different protocols were 8, 16, and 2. The coronal planes were chosen on the basis of our preferences. For the protocol using a gradient echo of 8, the TR was 6 ms and TE ranged from 1 to 26 ms. For the protocol using a gradient echo of 16, the TR was 6 ms and the TE ranged from 1 to 42 ms. For the protocol using a gradient echo of 2, the TR was 6 ms and the TE ranged from 1 to 5 ms. For all three protocols, scanning began at 1 ms with a 2-ms increment. The remaining parameters for all three protocols were as follows: flip angle, 27 ; matrix, ; sensitivity-encoding (SENSE) factor, 2; number of signals acquired, 1; FOV, 4 cm; slice thickness, 5 mm; interslice gap, 1 mm; and slice number, 5. Images for each protocol were obtained within one 17.3-second breath-hold. T2-weighted single-shot fast spin-echo sequences (TR range/te, /8; slice thickness, 5 mm; interslice gap, 1 mm; FOV, 36 4 cm; matrix, ; SENSE factor, 4) were performed in the transverse and coronal planes before performing the OLD MRI sequence to avoid inaccurate measurements of R2* in renal lesions. R2* Measurements Images acquired using OLD MRI were analyzed using a software tool (PRIDE Relaxation Maps Tool, version 2.1.1, Philips Healthcare). This tool generated a set of color-coded parametric images of R2* (= 1 / s) on a voxel basis according to an exponential function describing the expected signal decay as a function of TE and solving for the unknown value of R2*. On the map, red represents the highest R2* levels, indicating the highest concentration of deoxyhemoglobin, whereas blue represents the lowest R2* levels, indicating the lowest concentration of deoxyhemoglobin. Quantitative analyses were performed by a single genitourinary radiologist. The biopsy results were not available at the time of R2* analysis. The clinical status of the patients was unknown to the TLE 2: Rates of Spin Dephasing (R2*) in Renal Cortex and Medulla as Measured on lood Oxygenation Level Dependent MRI Using Three Gradient-Echo Protocols in Healthy Volunteers, Patients With Normally Functioning llografts, and Patients With Dysfunctioning llografts R2* (s 1 ) Gradient-Echo Protocol and Location ± SD 95% CI Healthy volunteers Gradient echo of 8 Cortex 11.1 ± Medulla 31.2 ± Gradient echo of 16 Cortex 14.6 ± Medulla 31. ± Gradient echo of 2 Cortex 15.2 ± Medulla 31.4 ± Normal allograft cases Gradient echo of 8 Cortex 11.6 ± Medulla 32.2 ± Gradient echo of 16 Cortex 14.9 ± Medulla 31.5 ± Gradient echo of 2 Cortex 14.6 ± Medulla 31.8 ± cute allograft rejection cases Gradient echo of 8 Cortex 11.8 ± Medulla 23.4 ± Gradient echo of 16 Cortex 13.4 ± Medulla 23.1 ± Gradient echo of 2 Cortex 13. ± Medulla 22.5 ± reader during the quantitative OLD MRI analysis. The analysis was performed by placing circular regions of interest (ROIs) on the color-coded parametric maps of R2*. The ROIs were manually drawn within the cortex and medulla at each upper pole, midportion, and lower pole of the two different coronal planes; then, six cortical and medullary R2* values were averaged to obtain the final cortical and medullary R2* value for a particular patient. The ROIs included at least 3 pixels (mean, 41.5 pixels; range, 3 57 pixels). Gray scales of T2*-weighted images and T2-weighted images were used as anatomic references to avoid adjacent structures or focal lesions such as a renal cyst. The reader also made an effort to exclude susceptibility artifacts, such as bowel gas, from the ROIs. Statistical nalysis We determined whether the R2* values of the cortex or medulla for different protocols had a normal distribution using the Kolmogorov-Smirnov test. The repeated-measures analysis of variance test was used to compare the R2* values in the cortex and medulla of healthy volunteers, patients with normal 111 JR:198, May 212

4 OLD MRI of Renal llografts renal allografts, and patients with acute allograft rejection for each of the three protocols. The cortical and medullary R2* results were compared using the analysis of variance with onferroni correction according to the protocol used. The ltman-land test was used to assess the reproducibilities of R2* measurements within subjects. Statistical analyses were performed with statistical software (PSW, version 18., SPSS). twosided p value of <.5 was considered to indicate a significant difference. Results The characteristics of the patients with normally functioning allografts and of those with acute allograft rejection are summarized in Table 1. The R2* values of the cortex and medulla in healthy volunteers, patients with normally functioning allografts, and patients with Fig. 2 Coronal color-coded maps from blood oxygenation level dependent MRI show rates of spin dephasing, denoted as R2*, of normally functioning renal allograft in 41-year-old woman. Red represents the highest R2* levels, indicating the highest concentration of deoxyhemoglobin, whereas blue represents the lowest R2* levels, indicating the lowest concentration of deoxyhemoglobin. C, Maps at gradient echoes of 8 (), 16 (), and 2 (C). ll protocols yielded good corticomedullary differentiation. cortical versus medullary R2* values at echo gradients of 8, 16, and 2 were 8.7 versus 33.7, 14.2 versus 36.8 s 1, and 13.1 versus 34.2 s 1, respectively. Fig. 3 Coronal color-coded maps from blood oxygenation level dependent MRI performed at gradient echo of 16 show rates of spin dephasing, denoted as R2*, in renal allograft recipients. Red represents the highest R2* levels, indicating highest concentration of deoxyhemoglobin, whereas blue represents lowest R2* levels, indicating lowest concentration of deoxyhemoglobin., Normally functioning allograft in 31-year-old woman. Good differentiation between cortex (blue or purple) and medulla (green) is visible in normal case. medullary R2* is 31.5 s 1., Dysfunctioning allograft in 52-year-old man. Map shows loss of corticomedullary differentiation in acute allograft rejection. medullary R2* is 2.7 s 1. Compared with normal allograft (), medullary R2* value in acute rejection case is decreased. acute allograft rejection for the three different protocols are presented in Table 2. Cortex Versus Medulla The medullary R2* values were significantly greater than the cortical R2* values in healthy volunteers, patients with normal allografts, and patients with acute allograft rejection for all three protocols (p <.1) (Fig. 2). Healthy Volunteers Versus Patients With Normal llografts The cortical and medullary R2* values were not significantly different between healthy volunteers and patients with normal allografts regardless of the MRI protocol used (p >.5). Three Different Protocols The cortical and medullary R2* values in the various groups were not significantly different among the three different protocols except that the cortical R2* values at a gradient echo of 8 were significantly lower than those obtained at gradient echoes of 16 and 2 (p <.1). Normal llografts Versus cute llograft Rejection For all protocols, medullary R2* values were significantly lower in cases of acute allograft rejection than in normal allografts (p <.1) (Fig. 3). In contrast, cortical R2* values in acute rejection cases were not significantly C JR:198, May

5 Park et al. 4 4 Difference in Cortical R2* at Gradient Echo of 8 (%) Difference in Cortical R2* at Gradient Echo of 16 (%) Cortical R2* at Gradient Echo of Cortical R2* at Gradient Echo of 16 different from those in normal allograft cases (p >.5). Reproducibility of R2* Measurement In the four patients with a normally functioning allograft, the mean difference in cortical or medullary R2* values between two separate MRI examinations at a 1-month interval was 3.8% or less for all protocols: 1.8% for the cortex and 1.1% for the medulla at a gradient echo of 8 (Fig. 4), 3.8% each for the cortex and medulla at a gradient echo of 16 (Fig. 5), and.6% for the cortex and 1.7% for the medulla at a gradient echo of 2 (Fig. 6) Difference in Medullary R2* at Gradient Echo of 8 (%) Discussion Our results show that OLD MRI at 3 T is a feasible and reproducible technique that can be used to assess the oxygenation status of a renal allograft. Medullary R2* values were significantly lower in acute allograft rejection cases than they were in normal allograft cases, whereas cortical R2* values were not significantly different between acute allograft rejection cases and normal allograft cases for the three different protocols. Furthermore, the cortical and medullary R2* values were reproducible in all protocols. These findings suggest that OLD Medullary R2* at Gradient Echo of 8 Fig. 4 ltman-land plots show reproducibility of rates of spin dephasing, denoted as R2* values, on blood oxygenation level dependent MRI. and, Cortical () and medullary () R2* values at gradient echo of Medullary R2* at Gradient Echo of 16 Fig. 5 ltman-land plots show reproducibility of rates of spin dephasing, denoted as R2* values, on blood oxygenation level dependent MRI. and, Cortical () and medullary () R2* values at gradient echo of 16. Difference in Medullary R2* at Gradient Echo of 16 (%) MRI at 3 T is a noninvasive method that can reproducibly evaluate cortical and medullary oxygenation in renal allografts. Furthermore, using this technique, we found significant elevation of intramedullary oxygenation levels in acute allograft rejection cases compared with normal allograft cases. High magnetic field strength can provide high spatial resolution because of an improved signal-to-noise ratio (SNR) [2]. The increase in the availability of 3-T MR scanners in clinical practice has resulted in several studies that have compared 3-T MRI results for assessment of intrarenal oxygenation in JR:198, May 212

6 OLD MRI of Renal llografts 15 3 Difference in Cortical R2* at Gradient Echo of 2 (%) Cortical R2* at Gradient Echo of 2 healthy volunteers with the results obtained using 1.5-T MRI [9, 16, 17]. t a magnetic field strength of 3 T, the paramagnetic effects of deoxyhemoglobin increase and, accordingly, corticomedullary differentiation on the R2* map is better than that achieved using a magnetic field strength of 1.5 T [9, 16, 17]. We found two- to threefold differences in R2* values between the cortex and medulla in both healthy volunteers and patients with normal allografts using 3-T MRI; this fold difference is greater than that reported in previous studies that used 1.5 T [13, 21, 22]. Our results suggest that 3-T MRI can provide better corticomedullary differentiation on an R2* map than 1.5 T, although we did not directly compare 3 T and 1.5 T in our study. few studies have used 1.5 T to assess the oxygenation status of the kidney in acute allograft rejection cases. These studies reported that medullary R2* values were significantly lower in acute allograft rejection cases than in normally functioning allograft cases, but the results reported for cortical R2* values in acute allograft rejection cases are conflicting [13, 14, 23, 24]. No previous study that we are aware of has reported the 3-T MRI findings at different gradient echoes for acute human allograft rejection cases. In accordance with previous findings at 1.5 T, our 3-T results indicate that R2* values in the medulla of acute allograft rejection cases are significantly lower than the R2* values of normal allografts. These findings indicate that the oxygen concentrations in the medullae of acute allograft rejection cases are Difference in Medullary R2* at Gradient Echo of 2 (%) 1 increased. This increase in oxygen concentration may result from altered corticomedullary hemodynamics during acute rejection, with preferential blood shunting toward the medulla and increased oxygen content [13]. In animal studies, increases in oxygen partial pressure in the renal medulla have been shown to be associated with decreases in glomerular filtration rates [25, 26]. recent study showed that, in human allograft rejections, a decrease in medullary R2* (i.e., decreased deoxyhemoglobin) on OLD MRI and a decrease in medullary blood flow according to perfusion MRI indicated an increase in medullary oxygenation [23]. In our study, three different gradient echoes (i.e., 8, 16, and 2 echoes) were applied at 3 T. Signal changes due to OLD effects scaled with field strength, but susceptibility, motion, and chemical-shift artifacts also increased at the higher field strength. Therefore, optimization of renal OLD MRI at 3 T may be necessary. Multiple gradient echoes after each excitation pulse (e.g., 8 or 16 echoes) are generally used for OLD MRI at 1.5 or 3 T [9, 12, 13, 23]. For optimal SNR, the theoretic maximum TE is approximately 5 ms at 1.5 T and 25 ms at 3 T [18]. However, there is no consensus as to which TE better reflects oxygenation status in human kidneys. We investigated whether different gradient echoes affect R2* values in healthy volunteers and patients with normal allografts. We found no significant differences in cortical or medullary R2* values between healthy kidneys and normal allografts for gradient echoes of Medullary R2* at Gradient Echo of 2 Fig. 6 ltman-land plots show reproducibility of rates of spin dephasing, denoted as R2* values, on blood oxygenation level dependent MRI. and, Cortical () and medullary () R2* values at gradient echo of 2. (maximum TE, 26 ms), 16 (maximum TE, 42 ms), or 2 (maximum TE, 5 ms). These results suggest that the use of the recommended maximum TE at 3 T is not mandatory for assessing the intrarenal oxygenation status of human kidneys in clinical practice. For OLD MRI to be clinically useful, the results have to be reproducible. The reproducibility of OLD MRI has been confirmed for both short-term periods (5-minute interval) and long-term periods ( 3 months) [14, 27]. In line with these previous studies, our study showed that OLD MRI findings were reproducible over a 1-month period; we found that the mean difference in cortical or medullary R2* values was 3.8% or less for all protocols. study reported that medullary R2* values were lower in renal allograft recipients than in healthy volunteers [14]. However, we found no significant difference in R2* values between healthy volunteers and patients with normally functioning allografts. We propose two explanations for this discrepancy. First, the timing of OLD MRI after transplant may influence medullary R2* values [18]. We performed OLD MRI less than 3 months after transplant, whereas OLD MRI in the previous study was performed more than 4 months after transplant. Second, the degree of renal function was different in our study compared with that reported in the previous study. Serum creatinine levels in renal allograft patients in our study were similar to those in healthy volunteers (< 115 μmol/l or 1.3 mg/dl), whereas in the previous study, patients with JR:198, May

7 Park et al. renal allografts had serum creatinine levels of less than 2 μmol/l or 2.26 mg/dl. Our study had several limitations. First, we enrolled a relatively small population of patients, especially patients with acute allograft rejection (n = 4). Nevertheless, we observed a significant difference in mean medullary R2* values between acute allograft rejection cases and normal allograft cases. Moreover, most of our data are consistent with the results reported in previous investigations, such as the significant difference in R2* values between the cortex and medulla in healthy volunteers and patients with normal allografts. Second, we did not include patients with other pathologic kidney diseases such as biopsy-proven acute tubular necrosis. larger future study including patients with acute tubular necrosis should be performed. Third, we did not evaluate interobserver variability in R2* measurements. In conclusion, our preliminary experience suggests that OLD MRI under different gradient echoes at 3 T is feasible, can provide reproducible findings in patients with renal allografts, and can be used to detect significant changes in medullary oxygenation in acute rejection cases. References 1. Wolfe R, shby V, Milford EL, et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999; 341: Simmons RG, bress L. Quality-of-life issues for end-stage renal disease patients. m J Kidney Dis 199; 15: Port FK, Wolfe R, Mauger E, erling DP, Jiang K. Comparison of survival probabilities for dialysis patients vs cadaveric renal transplant recipients. JM 1993; 27: Cosio FG, Pelletier RP, Falkenhain ME, et al. Impact of acute rejection and early allograft function on renal allograft survival. Transplantation 1997; 63: Wilczek HE. Percutaneous needle biopsy of the renal allograft: a clinical safety evaluation of 1129 biopsies. Transplantation 199; 5: Nyman RS, Cappelen-Smith J, al Suhaibani H, lfurayh O, Shakweer W, khtar M. Yield and complications in percutaneous renal biopsy: a comparison between ultrasound-guided gun-biopsy and manual techniques in native and transplant kidneys. cta Radiol 1997; 38: Prasad PV, Edelman RR, Epstein FH. Noninvasive evaluation of intrarenal oxygenation with OLD MRI. Circulation 1996; 94: Textor SC, Glockner JF, Lerman LO, et al. The use of magnetic resonance to evaluate tissue oxygenation in renal artery stenosis. J m Soc Nephrol 28; 19: Gloviczki ML, Glockner J, Gomez SI, et al. Comparison of 1.5 and 3 T OLD MR to study oxygenation of kidney cortex and medulla in human renovascular disease. Invest Radiol 29; 44: Thoeny HC, Kessler TM, Simon-Zoula S, et al. Renal oxygenation changes during acute unilateral ureteral obstruction: assessment with blood oxygen level dependent MR imaging initial experience. Radiology 28; 247: Epstein FH, Veves, Prasad PV. Effect of diabetes on renal medullary oxygenation during water diuresis. Diabetes Care 22; 25: Djamali, Sadowski E, Muehrer RJ, et al. OLD-MRI assessment of intrarenal oxygenation and oxidative stress in patients with chronic kidney allograft dysfunction. m J Physiol Renal Physiol 27; 292:F513 F Sadowski E, Fain S, lford SK, et al. ssessment of acute renal transplant rejection with blood oxygen level-dependent MR imaging: initial experience. Radiology 25; 236: Thoeny HC, Zumstein D, Simon-Zoula S, et al. Functional evaluation of transplanted kidneys with diffusion-weighted and OLD MR imaging: initial experience. Radiology 26; 241: Thulborn KR, Waterton JC, Matthews PM, Radda GK. Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field. iochim iophys cta 1982; 714: Li LP, Vu T, Li S, Dunkle E, Prasad PV. Evaluation of intrarenal oxygenation by OLD MRI at 3. T. J Magn Reson Imaging 24; 2: Li LP, Ji L, Lindsay S, Prasad PV. Evaluation of intrarenal oxygenation in mice by OLD MRI on a 3.T human whole-body scanner. J Magn Reson Imaging 27; 25: Li LP, Halter S, Prasad PV. lood oxygen level dependent MR imaging of the kidneys. Magn Reson Imaging Clin N m 28; 16: , viii 19. Epstein FH, Prasad P. Effects of furosemide on medullary oxygenation in younger and older subjects. Kidney Int 2; 57: Reichenbach JR, Venkatesan R, Yablonskiy D, Thompson MR, Lai S, Haacke EM. Theory and application of static field inhomogeneity effects in gradient-echo imaging. J Magn Reson Imaging 1997; 7: Prasad PV, Chen Q, Goldfarb JW, Epstein FH, Edelman RR. reath-hold R2* mapping with a multiple gradient-recalled echo sequence: application to the evaluation of intrarenal oxygenation. J Magn Reson Imaging 1997; 7: Zuo CS, Rofsky NM, Mahallati H, et al. Visualization and quantification of renal R2* changes during water diuresis. J Magn Reson Imaging 23; 17: Sadowski E, Djamali, Wentland L, et al. lood oxygen level dependent and perfusion magnetic resonance imaging: detecting differences in oxygen bioavailability and blood flow in transplanted kidneys. Magn Reson Imaging 21; 28: Han F, Xiao W, Xu Y, et al. The significance of OLD MRI in differentiation between renal transplant rejection and acute tubular necrosis. Nephrol Dial Transplant 28; 23: rezis M, gmon Y, Epstein FH. Determinants of intrarenal oxygenation. I. Effects of diuretics. m J Physiol 1994; 267:F159 F rezis M, Heyman SN, Epstein FH. Determinants of intrarenal oxygenation. II. Hemodynamic effects. m J Physiol 1994; 267:F163 F Simon-Zoula SC, Hofmann L, Giger, et al. Non-invasive monitoring of renal oxygenation using OLD-MRI: a reproducibility study. NMR iomed 26; 19: JR:198, May 212