- Email: [email protected]
Assessment of Intrarenal Oxygenation in Renal Donor With Blood Oxygenation Level–dependent Magnetic Resonance Imaging
Basic and Translational Science Assessment of Intrarenal Oxygenation in Renal Donor With Blood Oxygenation Leveledependent Magnetic Resonance Imaging Turun Song, Tao Lin, Zixing Huang, Lei Fu, Shaofeng He, Bin Song, and Qiang Wei OBJECTIVE METHODS
RESULTS
CONCLUSION
To examine change of the apparent relaxation rate R2* values in living kidney donors after uninephrectomy using blood oxygenation leveledependent magnetic resonance imaging. Between July 2011 and January 2012, 45 kidney donors were enrolled into this study. Blood oxygenation leveledependent magnetic resonance imaging scanning was performed before surgery, 3 and 7 days postoperatively. Participants were followed up for 1 year. The R2* values in medulla (mR2*) were significantly greater than that of cortex (cR2*), both in resected kidney and remaining one. cR2* values of the remaining kidney was 17.52 1.36 s1 and then decrease significantly by 8.97% to 15.95 1.14 s1 at 3 days (P <.001) and by 7.82% to 16.15 of 1.05 s1 at 7 days. No significant modification occurred in mR2* after surgery. Multivariate regression analysis showed that the decrease in cR2* values of the remaining kidney was positively associated with sex (r ¼ 0.418), body surface area (r ¼ 0.307), and preoperative cR2* values (r ¼ 0.659). Comparing with glomerular filtration rate at 7 days, a further increment in the glomerular filtration rate was noted at 1 year in patients with cR2* values decrease of 10% at 1 week (62.63 11.69 vs 56.97 7.51 mL/min/1.73 m2, P ¼ .02) but not in the other patients (66.43 10.89 vs 62.78 13.74, P ¼ .064). Kidney donation will induce early, profound oxygenation modification within the renal cortex of the remaining kidney. Donors with cR2* value decrease of 10% at 1 week have a more favorable renal function compensation at 1 year. UROLOGY 83: 1205.e1e1205.e5, 2014. 2014 Elsevier Inc.
T
he surgical removal of a normal kidney leads to dramatic hemodynamic, morphologic, and functional adaptation for the remaining kidney, including early increase in glomerular filtration rate (GFR), effective renal plasma flow, and filtration fraction.1-3 This adaptation implies a physiology alternation and altered tissue oxygen bioavailability status.4 The are few reports about the early effects of uninephrectomy in humans, particularly in pathologic condition, mostly renal tumors.5,6 Comparing with general population, kidney donors enjoy a similar long-term survival and comparable risk of developing into an end stage of renal diseases.7-10 However, the early changes in kidney parenchyma oxygen bioavailability status after renal Turun Song and Tao Lin contributed equally. Financial Disclosure: The authors declare that they have no relevant financial interests. Funding Support: This project was supported by the National Natural Science Foundation of China (grant no. 30872579). From the Department of Urology, West China Hospital, Sichuan University, Chengdu, Sichuan, People’s Republic of China; and the Department of Radiology, West China Hospital, Sichuan University, Chengdu, Sichuan, People’s Republic of China Reprint requests: Qiang Wei, M.D., Department of Urology, West China Hospital, Sichuan University, Guoxue Xiang #37, Chengdu, Sichuan 610041, People’s Republic of China. E-mail: [email protected] Submitted: July 19, 2013, accepted (with revisions): January 7, 2014
ª 2014 Elsevier Inc. All Rights Reserved
donation and their effects on renal function in the long term are far from being well addressed. Blood oxygenation leveledependent magnetic resonance imaging (BOLD MRI) is a noninvasive method to assess intrarenal oxygen bioavailability using deoxyhemoglobin as an endogenous contrast agent.11 Oxyhemoglobin is a diamagnetic molecule that creates no magnetic moment as oxygen molecules are bound to iron, whereas deoxyhemoglobin is a paramagnetic molecule that generates magnetic moments by its unpaired iron electrons.12 Changes in deoxyhemoglobin concentration involve generation of phase incoherence of magnetic spins, leading to reduction of the T2* relaxation time which in turn attenuate signal in gradient echo sequences and an increase in the apparent spin-spin relaxation rate denoted as R2* (R2* ¼ 1/T2*). As presented in previous study that R2* was directly proportional to the amount of deoxyhemoglobin in blood,11 the increased R2* level implies an increased deoxyhaemoglobin level. This behavior, combined with the fact that oxygen pressure (pO2) in blood is in equilibrium with pO2 in tissue, potentates BOLD MRI to be an attractive tool for high resolution mapping of the intrarenal oxygenation. 0090-4295/14/$36.00 1205.e1 http://dx.doi.org/10.1016/j.urology.2014.01.015
MRI at higher field strengths such as 3.0 T has higher inherent signal-to-noise ratios.13,14 For BOLD MRI, there is increased sensitivity to susceptibility effects at higher field strengths in distinguishing discrete cortical and inner medullary regions of the kidney and approximate measured differences in oxygen tension.15 In this study, 3.0-T BOLD MRI was applied to noninvasively measure intrarenal oxygenation modifications induced by unilateral nephrectomy in kidney donors.
MATERIALS AND METHODS Patients The study was approved by the ethics committee of the West China Hospital, Sichuan University (REC reference number 07/ Q1405/21), and all patients gave fully informed consent before inclusion. Between July 2011 and January 2012, we prospectively evaluated 45 living kidney donors in a consecutive fashion as they presented to transplant clinic for transplantation. Potential participants had to be aged at least 18 years, not pregnant, and without a contraindication to an MRI examination. In addition, all patients undergo a thorough preoperative evaluation process to ensure they are medically fit for donation and their kidney is suitable for transplantation. Results from BOLD MRI are not considered as criteria in selection of patients. A 99mTc-mercaptoacetyltriglycine renal scintigraphy was performed in all patients, and effective renal plasma flow was calculated using a camera-based technique. Serum creatinine levels were determined preoperatively and at postoperative day 7 in all patients and 1 year postoperatively in 34 patients. The GFR was estimated by Cockoroft-Gault formula.16 The preoperative single-kidney GFR was calculated as the preoperative total GFR multiplied by the differential ratio of the singlekidney effective renal plasma flow. To rule out the difference in body surface area (BSA) among included patients, preoperative and postoperative GFR were adjusted to the standard BSA (1.73 m2). BSA was calculated as described in the literature.17
BOLD Methodology All patients were designed to undergoing 3 times of MRI scanning, within 5 days before surgery and at the third and seventh day after the surgery. MRI was performed with a 3.0-T MR scanner (Siemens Magnetom Trio Tim, Germany), using a multiple gradient-echo breath-hold acquisition. An 8-channel phased-array coil was used for signal reception. The MRI parameters were as follows: 6 oblique-coronal slices through both kidneys and the remnant one; range of echo time 2.5-59.32 ms, field of view ¼ 380 380 mm, 256 256 matrix, slice thickness ¼ 5 mm, space ¼ 1 mm, flip angle ¼ 33 , time of repetition ¼ 75 ms, band-width, 540 Hz per pixel. Twelve T2*-weighted images corresponding to 12 different gradient echoes were acquired for each section within one 10-second breath hold. A color T2* map of the kidney was generated using Functools on the MR working station. On the color map, bright green represents the highest T2* levels, indicating the lowest concentration of deoxyhaemoglobin, whereas blue represents the lowest T2* levels, indicating the highest concentration of deoxyhaemoglobin. T2* levels were measured using the regions of interest (ROIs) tool. For each examination, 6 ROIs were drawn in the cortex and the medulla, respectively, using regular T2-weighted images as anatomic reference to determine the 1205.e2
Table 1. The characteristics of included patients Characteristics
Value
P
Sex Male 11 Female 34 Age (y) 48.24 8.14 Weight (kg) 58.07 6.69 Height (m) 1.60 0.06 BSA (m2) 1.60 0.10 Serum creatinine (mmol/L) Preoperative 65.34 11.41 <.001 7 d postoperative 105.07 22.62 1 y postoperative 97.54 19.09 GFR of the resected kidney 48.87 10.23 2 (mL/min/1.73 m ) GFR of the remaining kidney (mL/min/1.73 m2) Preoperative 48.75 8.93 <.001 7 d postoperative 61.61 13.41 1 y postoperative 65.40 11.82 Preoperative R2* of resected kidney (s1) Cortex 17.50 1.52 <.001 Medulla 38.84 4.43 Preoperative R2* of remaining kidney (s1) Cortex 17.52 1.36 <.001 Medulla 37.99 4.92 BSA, body surface area; GFR, glomerular filtration rate; R2*, the apparent spin-spin relaxation rate.
cortical or medullary region. Each ROI consisted of 20-60 pixels. Then, T2* values were converted to R2* value (R2* ¼ 1/T2*).
Data Analysis All continuous data are presented as mean standard deviation and compared using Student t test, chi-square test for categorical data. When necessary, 1-way analysis of variance with posthoc test was applied to perform the pairwise multiple comparisons of continuous data between groups. Univariate and multivariate stepwise regression analyses were used to test the relationship between R2* value and patient characteristics. All tests were 2-sided, and P <.05 was considered statistically significant. All statistical analyses were performed using Statistical Package for Social Sciences software (SPSS 18.0, Chicago, IL).
RESULTS We included 45 donors among whom 11 were male and 34 were female donors. Twenty-five patients underwent laparoscopic nephrectomy, and the other 20 patients received open nephrectomy. The demographic characteristics of the included patients are displayed in Table 1. The preoperative serum creatinine level was 65.34 11.41 mmol/L, and it increased to 105.07 22.62 mmol/L at 7 days, and then stabilized at a slightly lower level of 97.54 19.09 mmol/L at 1 year. The preoperative GFR was 97.62 18.08 mL/min/1.73 m2, and the GFR of resected kidney was 48.87 10.23 mL/min/1.73 m2. The GFR of remaining kidney increased by 26.38% from 48.75 8.93 to 61.61 13.41 mL/min/1.73 m2 at 1 week (P <.001) and by 34.15% to 65.40 11.82 mL/min/ UROLOGY 83 (5), 2014
Table 2. Correlation between patient characteristics and decrease of cR2* values of remaining kidney
Patient Characteristics
Figure 1. The R2* values of remaining kidneys. R2*, the apparent spin-spin relaxation rate.
1.73 m2 at 1 year (P <.001). The R2* values in medulla (mR2*) were significantly greater than that of cortex (cR2*), both for resected kidney and remaining one (38.84 4.43 vs 17.50 1.52 s1, P <.001; 37.99 4.92 vs 17.52 1.36 s1, P <.001, respectively). At 3 and 7 days, the discrepancy in R2* values between cortex and medulla remain significant (P <.001). There were no differences in mR2* values at different measuring time. cR2* values decrease significantly by 8.97% from 17.52 1.36 to 15.95 1.14 s1 at 3 days (P <.001) and by 7.82% to 16.15 of 1.05 s1 at 7 days (P <.001; Fig. 1). Multivariate regression analysis showed that there was no association between preoperative mR2* and cR2* values of remaining kidney and patients’s characteristics. In terms of postoperative R2* values of the remaining kidney, cR2* values were negatively associated with sex, and mR2* values were positively associated with GFR of the resected kidney. Furthermore, no statistically significant association was noted for other factors. Univariate analysis demonstrated that the decrease in cR2* values at 7 days was positively associated with preoperative cR2* values of the resected (r ¼ 0.486) and remaining kidney (r ¼ 0.771). Multivariate regression noted the decrease in cR2* values was positively associated with sex (r ¼ 0.418), BSA (r ¼ 0.307), and preoperative cR2* values of the remaining kidney (r ¼ 0.659). There is no statistically significant association for age, single kidney GFR, and preoperative mR2* values (Table 2). To investigate the changes in renal function, the study population was stratified into 2 groups according to the decrease in cR2* values at 7 days. Group 1 included patients with cR2* values decrease of >10%, and group 2 included those with cR2* values decrease of 10%. Both groups bear comparable baseline characteristics except the preoperative cR2 values that cR2* values of the remaining kidney in group 1 is significantly greater than that of group 2 (18.76 1.50 vs 16.92 1.08 s1, P <.001). The GFR of the remaining kidney increased significantly 7 days after surgery in both groups (group 1: 48.08 7.05 vs 56.97 7.51 mL/min/1.73 m2, P <.001; group 2: 51.08 8.77 vs 62.78 13.74 mL/min/1.73 m2, UROLOGY 83 (5), 2014
Univariate Analysis
b
P
Multivariate Analysis
b
P
Sex 0.266 .098 0.418 <.001* Age 0.017 .917 0.046 .733 BSA 0.183 .295 0.307 .005* GFR of the resected 0.166 .306 0.084 .632 kidney GFR of remaining 0.031 .849 0.056 .763 kidney (preoperative) GFR of remaining 0.069 .672 0.141 .324 kidney (7 d postoperative) cR2* of resected 0.486 <.001* 0.109 .329 kidney (preoperative) mR2* of resected 0.159 .328 0.069 .543 kidney (preoperative) cR2* of remaining 0.771 <.001* 0.659 <.001* kidney (preoperative) mR2* of remaining 0.243 .131 0.000 .999 kidney (preoperative) cR2*, R2* of cortex; mR2*, R2* of medulla; other abbreviations as in Table 1. *P <.05.
P <.001; Table 3). However, at 1 year, group 1 got further improvement in GFR significantly (P ¼ .02), whereas the GFR in group 2 remains stable and comparable with that at 7 days (P ¼ .064).
COMMENT As is known, unilateral nephrectomy results in a dramatic change of the hemodynamics and function in the remaining kidney.1-3 The hemodynamic adaptation implies a physiology change, especially the oxygenation alternation. Although the long-term follow-up has yielded favorable results,7-10 the oxygen bioavailability alternation in kidney donor after kidney donation remains unclear. To the best of our knowledge, this is the first patient series discussing how the oxygenation gradient evolves in kidney donors, and we did find some interesting results. First, we confirmed what other authors have already described: R2* values were significantly higher in the medulla than that in the cortex in normally functioning kidneys,11,18,19 indicating the medulla works in a hypoxic condition. Second, we observed that after the kidney removal, GFR of the remaining kidney increased significantly, and the cR2* values of remaining kidney decreased dramatically, whereas that in the medulla produced no sensible modification. Animal model has shown that the effective renal blood flow of the remnant kidney increased by 40% immediately after uninephrectomy, from 255.55 27.8 to 357.77 38.9 mL/min.2 Kidney donors presented the same phenomenon, with renal blood flow increased by 25%-32.5% at 1 week and 29%-30.1% at 1 year after 1205.e3
Table 3. Patient characteristics stratified by decrease of cR2* value of the remaining kidney at 7 d Decrease of cR2* Values Patient Characteristics Sex Male Female Age (y) Weight (kg) Height (m) BSA (m2) Serum creatinine (mmol/L) Preoperative 7 d postoperative 1 y postoperative GFR of the resected kidney (mL/min/1.73 m2) GFR of the remaining kidney (mL/min/1.73 m2) Preoperative 7 d postoperative 1 y postoperative cR2* value of the remaining kidney (s1) Preoperative 3 d postoperative 7 d postoperative mR2* value of the remaining kidney (s1) Preoperative 3 d postoperative 7 d postoperative
>10% 2 10 47.83 58.66 1.62 1.62
6 16 46.95 57.18 1.60 1.59
P .402
9.41 6.14 0.06 0.10
.778 .545 .356 .379
9.62 21.03 13.32 10.16
.092 .334 .305 .023
48.08 7.05 56.97 7.51 62.63 11.69
51.08 8.77 62.78 13.74 66.43 10.89
.316 .185 .349
18.76 1.50 16.14 1.07 15.80 0.89
16.92 1.08 15.80 1.07 16.50 1.08
<.001 .413 .062
41.26 5.96 34.64 6.54 37.37 4.08
37.74 3.93 36.37 6.66 39.57 7.67
.084 .502 .346
68.67 111.30 102.55 44.37
6.82 6.79 0.07 0.11
10%
15.21 27.28 26.52 9.01
61.29 103.07 95.47 52.76
Abbreviations as in Tables 1 and 2.
kidney donation.1,3 It is reasonable that increased renal blood flow results in increased GFR, and greater amount of oxygenated blood would improve the tissue oxygenation status. Our findings corroborated the facts that cR2* values decrease significantly by 8.97% from 17.52 to 15.95s1 at 3 days (P <.001) and by 7.82% to 16.15 s1 at 7 days (P <.001), indicating an overall increase in cortical oxygenation. However, there is no significant alternation in the mR2* values, which is in line with the observation by Malvezzi et al,4 suggesting the medulla remains hypoxic. Medulla is believed to consume the main part of oxygen supply to kidney because of electrolyte and water reabsorption. Inhibition of the Na-K-Cl cotransportor by forosemide attenuates deoxyhemoglobin levels and increases medullary tissue pO2, leading to a decrement of mR2* values.20,21 Increased renal blood flow after uninephrectomy consequently induces production of more ultrafiltrate22,23 that the greater solute load to the medullary portions of the tubules forces active oxygen consuming reabsorption mechanism. Consequently, the increase of blood oxygen supply to the medulla was offset. BOLD MRI is widely being used as a noninvasive method to assess tissue oxygen bioavailability. Sadowski et al24 found in transplanted kidneys, there was a significant difference between mR2* values in the group with acute rejection (R2* ¼ 16.2/s) compared with allografts with acute tubular necrosis (R2* ¼ 19.8/s; P ¼ .047) and normal-functioning allografts (R2* ¼ 24.3/s; P ¼ .0003). In addition, combined with single kidney GFR, Chrysochou et al25 found that R2*:GFR was an effective 1205.e4
predictive biomarker for positive renal functional response to revascularization in patient with stenosed renal artery. However, as demonstrated by Michaely et al26 from a study of 368 chronic kidney disease that there was no association of the R2* values of cortex and medulla with the GFR (r ¼ 0.007) and BOLD MRI fails to discriminate between patients with various chronic kidney disease stages. This is confirmed by our results that no significant association was detected between single kidney GFR and the cR2* and mR2* values. Nonetheless, we found that patients with decrease in cR2* values of remaining kidney >10% at 7 days have a more favorable renal function adaptation at 1 year. As indicated by Sigmon et al, increased renal blood flow after uninephrectomy is attributed to reduction in renal vascular resistance, which is induced by enhanced nitric oxide synthesis.27 Thus, we speculated greater improvement in cortex oxygenation status may reflect a greater compensatory potential in vascular dilation chemicals production, and this potential may be an important determinant of the renal function compensation in the long term. This study is not without limitations, some of which are difficult to overcome. First is the small number of patients who were studied. Second, water diuresis is known to reduce the corticomedullary oxygenation. To minimize the effect of patient hydration, patients were told to abstain from food or water from the night before examination, but no certainty of each individual can be made. Third, we only performed a BOLD MRI examination within 7 days after surgery and get the oxygen UROLOGY 83 (5), 2014
bioavailability alteration at the very early phase. How does the oxygenation gradient evolve as the contralateral kidney adapting to its new status and its impact on the long-term renal function are unknown. Further follow-up is required. Finally, our results indicated BOLD MRI is helpful in donor’s postoperative follow-up, rather than patient selection, because we do not know who will get better renal function compensation from the preoperative BOLD MRI evaluation. Thus a BOLD MRIebased novel technique, which may imitate the physiological change after kidney removal is required, such as using vasodilator and measuring the renal blood flow simultaneously.
CONCLUSION In conclusion, the present study has demonstrated that BOLD MRI is a noninvasive tool to study renal physiology. Kidney donation will induce early, profound oxygenation modification within the renal cortex of the remnant kidney, which probably reflects the adaptive mechanism. In addition, greater improvement in cortex oxygenation status suggests better kidney function compensation in moderate term. References 1. Anderson RG, Bueschen AJ, Lloyd LK, et al. Short-term and longterm changes in renal function after donor nephrectomy. J Urol. 1991;145:11-13. 2. Ziada G, Youseif H, Khalil M. Compensatory changes in the function of the remaining kidney immediately after unilateral nephrectomy in sheep. Tohoku J Exp Med. 2009;219:165-168. 3. Indudhara R, Kenney PJ, Bueschen AJ, et al. Live donor nephrectomy in patients with fibromuscular dysplasia of the renal arteries. J Urol. 1999;162:678-681. 4. Malvezzi P, Bricault I, Terrier N, et al. Evaluation of intrarenal oxygenation by blood oxygen level dependent magnetic resonance imaging in living kidney donors and their recipients: preliminary results. Transplant Proc. 2009;41:641-644. 5. Funahashi Y, Hattori R, Yamamoto T, et al. Ischemic renal damage after nephron sparing surgery in patients with normal contralateral kidney. Eur Urol. 2009;55:209-216. 6. Funahashi Y, Hattori R, Yamamoto T, et al. Change in contralateral renal parenchymal volume 1 week after unilateral nephrectomy. Urology. 2009;74:708-712. 7. Ramcharan Thiagarajan, Matas Arthur J. Long-term (20e37 years) follow-up of living kidney donors. Am J Transplant. 2002;2:959-964. 8. Ibrahim HN, Foley R, Tan L, et al. Long-term consequences of kidney donation. N Engl J Med. 2009;360:459-469. 9. Australian and New Zealand Dialysis and transplant Registry. ANZDATA Registry Rep, 2007, pp30.
UROLOGY 83 (5), 2014
10. Oppenheimer SF. Short-, medium-, and long-term follow-up of living donors. Nefrologia. 2010;30(Suppl. 2):100-105. 11. Prasad PV, Edelman RR, Epstein FH. Noninvasive evaluation of intrarenal oxygenation with BOLD MRI. Circulation. 1996;94: 3271-3275. 12. Gomori JM, Grossman RI, Yu-Ip C, et al. NMR relaxation times of blood: dependence on field strength, oxidation state, and cell integrity. J Comput Assist Tomogr. 1987;11:684-690. 13. Haacke EBR, Brown RW, Thompson M, et al. Magnetic Resonance Imaging Physical Principles and Sequence Design. New York: John Wiley & Sons; 1999:343-387. 14. Edelstein WA, Glover GH, Hardy CJ, et al. The intrinsic signalto-noise ratio in NMR imaging. Magn Reson Med. 1986;3: 604-618. 15. Gloviczki ML, Glockner J, Gomez SI, et al. Comparison of 1.5 and 3 T BOLD MR to study oxygenation of kidney cortex and medulla in human renovascular disease. Invest Radiol. 2009;44: 566-571. 16. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16:31-41. 17. Mosteller RD. Simplified calculation of body-surface area. N Engl J Med. 1987;317:1098. 18. Djamali A, Sadowski EA, Muehrer RJ, et al. BOLD-MRI assessment of intrarenal oxygenation and oxidative stress in patients with chronic kidney allograft dysfunction. Am J Physiol Ren Physiol. 2007; 292:513-522. 19. Han F, Xiao W, Xu Y, et al. The significance of BOLD MRI in differentiation between renal transplant rejection and acute tubular necrosis. Nephrol Dial Transpl. 2008;23:2666-2672. 20. Warner L, Glockner JF, Woollard J, et al. Determination of renal cortical and medullary oxygenation using blood level-dependent magnetic resonance imaging and selective diuretics. Invest Radiol. 2011;46:41-47. 21. Li LP, Vu AT, Li BS, et al. Evaluation of intrarenal oxygenation by BOLD MRI at 3.0 T. J Magn Reson Imaging. 2004;20:901-904. 22. Hosteller TH, Olson JL, Rennke HG, et al. Hyperfiltration in remnant nephrons: a potentially adverse response to renal ablation. J Am Soc Nephrol. 2001;12:1315-1325. 23. Bohlouli A, Tarzamni MK, Zomorodi A, et al. Remnant kidney function and size in living unrelated kidney donors after nephrctomy. Saudi J Kidney Dis Transpl. 2010;21:246-250. 24. Sadowski EA, Djamali A, Wentland AL, et al. Blood oxygen leveldependent and perfusion magnetic resonance imaging: detecting differences in oxygen bioavailability and blood flow in transplanted kidneys. Magn Reson Imaging. 2010;28:56-64. 25. Chrysochou C, Mendichovszky IA, Buckley DL, et al. BOLD imaging: a potential predictive biomarker of renal functional outcome following revascularization in atheromatous renovascular disease. Nephrol Dial Transpl. 2012;27:1013-1019. 26. Michaely HJ, Metzger L, Attenberger UI, et al. Renal BOLD-MRI does not reflect renal function: a prospective study in 368 patients. Proc Intl Soc Mag Reson Med. 2011;19:441. 27. Sigmon DH, Gonzalez-Feldman E, Cavasin MA, et al. Role of nitric oxide in the renal hemodynamic response to unilateral nephrectomy. J Am Soc Nephrol. 2004;15:1413-1420.
1205.e5
RESULTS
CONCLUSION
To examine change of the apparent relaxation rate R2* values in living kidney donors after uninephrectomy using blood oxygenation leveledependent magnetic resonance imaging. Between July 2011 and January 2012, 45 kidney donors were enrolled into this study. Blood oxygenation leveledependent magnetic resonance imaging scanning was performed before surgery, 3 and 7 days postoperatively. Participants were followed up for 1 year. The R2* values in medulla (mR2*) were significantly greater than that of cortex (cR2*), both in resected kidney and remaining one. cR2* values of the remaining kidney was 17.52 1.36 s1 and then decrease significantly by 8.97% to 15.95 1.14 s1 at 3 days (P <.001) and by 7.82% to 16.15 of 1.05 s1 at 7 days. No significant modification occurred in mR2* after surgery. Multivariate regression analysis showed that the decrease in cR2* values of the remaining kidney was positively associated with sex (r ¼ 0.418), body surface area (r ¼ 0.307), and preoperative cR2* values (r ¼ 0.659). Comparing with glomerular filtration rate at 7 days, a further increment in the glomerular filtration rate was noted at 1 year in patients with cR2* values decrease of 10% at 1 week (62.63 11.69 vs 56.97 7.51 mL/min/1.73 m2, P ¼ .02) but not in the other patients (66.43 10.89 vs 62.78 13.74, P ¼ .064). Kidney donation will induce early, profound oxygenation modification within the renal cortex of the remaining kidney. Donors with cR2* value decrease of 10% at 1 week have a more favorable renal function compensation at 1 year. UROLOGY 83: 1205.e1e1205.e5, 2014. 2014 Elsevier Inc.
T
he surgical removal of a normal kidney leads to dramatic hemodynamic, morphologic, and functional adaptation for the remaining kidney, including early increase in glomerular filtration rate (GFR), effective renal plasma flow, and filtration fraction.1-3 This adaptation implies a physiology alternation and altered tissue oxygen bioavailability status.4 The are few reports about the early effects of uninephrectomy in humans, particularly in pathologic condition, mostly renal tumors.5,6 Comparing with general population, kidney donors enjoy a similar long-term survival and comparable risk of developing into an end stage of renal diseases.7-10 However, the early changes in kidney parenchyma oxygen bioavailability status after renal Turun Song and Tao Lin contributed equally. Financial Disclosure: The authors declare that they have no relevant financial interests. Funding Support: This project was supported by the National Natural Science Foundation of China (grant no. 30872579). From the Department of Urology, West China Hospital, Sichuan University, Chengdu, Sichuan, People’s Republic of China; and the Department of Radiology, West China Hospital, Sichuan University, Chengdu, Sichuan, People’s Republic of China Reprint requests: Qiang Wei, M.D., Department of Urology, West China Hospital, Sichuan University, Guoxue Xiang #37, Chengdu, Sichuan 610041, People’s Republic of China. E-mail: [email protected] Submitted: July 19, 2013, accepted (with revisions): January 7, 2014
ª 2014 Elsevier Inc. All Rights Reserved
donation and their effects on renal function in the long term are far from being well addressed. Blood oxygenation leveledependent magnetic resonance imaging (BOLD MRI) is a noninvasive method to assess intrarenal oxygen bioavailability using deoxyhemoglobin as an endogenous contrast agent.11 Oxyhemoglobin is a diamagnetic molecule that creates no magnetic moment as oxygen molecules are bound to iron, whereas deoxyhemoglobin is a paramagnetic molecule that generates magnetic moments by its unpaired iron electrons.12 Changes in deoxyhemoglobin concentration involve generation of phase incoherence of magnetic spins, leading to reduction of the T2* relaxation time which in turn attenuate signal in gradient echo sequences and an increase in the apparent spin-spin relaxation rate denoted as R2* (R2* ¼ 1/T2*). As presented in previous study that R2* was directly proportional to the amount of deoxyhemoglobin in blood,11 the increased R2* level implies an increased deoxyhaemoglobin level. This behavior, combined with the fact that oxygen pressure (pO2) in blood is in equilibrium with pO2 in tissue, potentates BOLD MRI to be an attractive tool for high resolution mapping of the intrarenal oxygenation. 0090-4295/14/$36.00 1205.e1 http://dx.doi.org/10.1016/j.urology.2014.01.015
MRI at higher field strengths such as 3.0 T has higher inherent signal-to-noise ratios.13,14 For BOLD MRI, there is increased sensitivity to susceptibility effects at higher field strengths in distinguishing discrete cortical and inner medullary regions of the kidney and approximate measured differences in oxygen tension.15 In this study, 3.0-T BOLD MRI was applied to noninvasively measure intrarenal oxygenation modifications induced by unilateral nephrectomy in kidney donors.
MATERIALS AND METHODS Patients The study was approved by the ethics committee of the West China Hospital, Sichuan University (REC reference number 07/ Q1405/21), and all patients gave fully informed consent before inclusion. Between July 2011 and January 2012, we prospectively evaluated 45 living kidney donors in a consecutive fashion as they presented to transplant clinic for transplantation. Potential participants had to be aged at least 18 years, not pregnant, and without a contraindication to an MRI examination. In addition, all patients undergo a thorough preoperative evaluation process to ensure they are medically fit for donation and their kidney is suitable for transplantation. Results from BOLD MRI are not considered as criteria in selection of patients. A 99mTc-mercaptoacetyltriglycine renal scintigraphy was performed in all patients, and effective renal plasma flow was calculated using a camera-based technique. Serum creatinine levels were determined preoperatively and at postoperative day 7 in all patients and 1 year postoperatively in 34 patients. The GFR was estimated by Cockoroft-Gault formula.16 The preoperative single-kidney GFR was calculated as the preoperative total GFR multiplied by the differential ratio of the singlekidney effective renal plasma flow. To rule out the difference in body surface area (BSA) among included patients, preoperative and postoperative GFR were adjusted to the standard BSA (1.73 m2). BSA was calculated as described in the literature.17
BOLD Methodology All patients were designed to undergoing 3 times of MRI scanning, within 5 days before surgery and at the third and seventh day after the surgery. MRI was performed with a 3.0-T MR scanner (Siemens Magnetom Trio Tim, Germany), using a multiple gradient-echo breath-hold acquisition. An 8-channel phased-array coil was used for signal reception. The MRI parameters were as follows: 6 oblique-coronal slices through both kidneys and the remnant one; range of echo time 2.5-59.32 ms, field of view ¼ 380 380 mm, 256 256 matrix, slice thickness ¼ 5 mm, space ¼ 1 mm, flip angle ¼ 33 , time of repetition ¼ 75 ms, band-width, 540 Hz per pixel. Twelve T2*-weighted images corresponding to 12 different gradient echoes were acquired for each section within one 10-second breath hold. A color T2* map of the kidney was generated using Functools on the MR working station. On the color map, bright green represents the highest T2* levels, indicating the lowest concentration of deoxyhaemoglobin, whereas blue represents the lowest T2* levels, indicating the highest concentration of deoxyhaemoglobin. T2* levels were measured using the regions of interest (ROIs) tool. For each examination, 6 ROIs were drawn in the cortex and the medulla, respectively, using regular T2-weighted images as anatomic reference to determine the 1205.e2
Table 1. The characteristics of included patients Characteristics
Value
P
Sex Male 11 Female 34 Age (y) 48.24 8.14 Weight (kg) 58.07 6.69 Height (m) 1.60 0.06 BSA (m2) 1.60 0.10 Serum creatinine (mmol/L) Preoperative 65.34 11.41 <.001 7 d postoperative 105.07 22.62 1 y postoperative 97.54 19.09 GFR of the resected kidney 48.87 10.23 2 (mL/min/1.73 m ) GFR of the remaining kidney (mL/min/1.73 m2) Preoperative 48.75 8.93 <.001 7 d postoperative 61.61 13.41 1 y postoperative 65.40 11.82 Preoperative R2* of resected kidney (s1) Cortex 17.50 1.52 <.001 Medulla 38.84 4.43 Preoperative R2* of remaining kidney (s1) Cortex 17.52 1.36 <.001 Medulla 37.99 4.92 BSA, body surface area; GFR, glomerular filtration rate; R2*, the apparent spin-spin relaxation rate.
cortical or medullary region. Each ROI consisted of 20-60 pixels. Then, T2* values were converted to R2* value (R2* ¼ 1/T2*).
Data Analysis All continuous data are presented as mean standard deviation and compared using Student t test, chi-square test for categorical data. When necessary, 1-way analysis of variance with posthoc test was applied to perform the pairwise multiple comparisons of continuous data between groups. Univariate and multivariate stepwise regression analyses were used to test the relationship between R2* value and patient characteristics. All tests were 2-sided, and P <.05 was considered statistically significant. All statistical analyses were performed using Statistical Package for Social Sciences software (SPSS 18.0, Chicago, IL).
RESULTS We included 45 donors among whom 11 were male and 34 were female donors. Twenty-five patients underwent laparoscopic nephrectomy, and the other 20 patients received open nephrectomy. The demographic characteristics of the included patients are displayed in Table 1. The preoperative serum creatinine level was 65.34 11.41 mmol/L, and it increased to 105.07 22.62 mmol/L at 7 days, and then stabilized at a slightly lower level of 97.54 19.09 mmol/L at 1 year. The preoperative GFR was 97.62 18.08 mL/min/1.73 m2, and the GFR of resected kidney was 48.87 10.23 mL/min/1.73 m2. The GFR of remaining kidney increased by 26.38% from 48.75 8.93 to 61.61 13.41 mL/min/1.73 m2 at 1 week (P <.001) and by 34.15% to 65.40 11.82 mL/min/ UROLOGY 83 (5), 2014
Table 2. Correlation between patient characteristics and decrease of cR2* values of remaining kidney
Patient Characteristics
Figure 1. The R2* values of remaining kidneys. R2*, the apparent spin-spin relaxation rate.
1.73 m2 at 1 year (P <.001). The R2* values in medulla (mR2*) were significantly greater than that of cortex (cR2*), both for resected kidney and remaining one (38.84 4.43 vs 17.50 1.52 s1, P <.001; 37.99 4.92 vs 17.52 1.36 s1, P <.001, respectively). At 3 and 7 days, the discrepancy in R2* values between cortex and medulla remain significant (P <.001). There were no differences in mR2* values at different measuring time. cR2* values decrease significantly by 8.97% from 17.52 1.36 to 15.95 1.14 s1 at 3 days (P <.001) and by 7.82% to 16.15 of 1.05 s1 at 7 days (P <.001; Fig. 1). Multivariate regression analysis showed that there was no association between preoperative mR2* and cR2* values of remaining kidney and patients’s characteristics. In terms of postoperative R2* values of the remaining kidney, cR2* values were negatively associated with sex, and mR2* values were positively associated with GFR of the resected kidney. Furthermore, no statistically significant association was noted for other factors. Univariate analysis demonstrated that the decrease in cR2* values at 7 days was positively associated with preoperative cR2* values of the resected (r ¼ 0.486) and remaining kidney (r ¼ 0.771). Multivariate regression noted the decrease in cR2* values was positively associated with sex (r ¼ 0.418), BSA (r ¼ 0.307), and preoperative cR2* values of the remaining kidney (r ¼ 0.659). There is no statistically significant association for age, single kidney GFR, and preoperative mR2* values (Table 2). To investigate the changes in renal function, the study population was stratified into 2 groups according to the decrease in cR2* values at 7 days. Group 1 included patients with cR2* values decrease of >10%, and group 2 included those with cR2* values decrease of 10%. Both groups bear comparable baseline characteristics except the preoperative cR2 values that cR2* values of the remaining kidney in group 1 is significantly greater than that of group 2 (18.76 1.50 vs 16.92 1.08 s1, P <.001). The GFR of the remaining kidney increased significantly 7 days after surgery in both groups (group 1: 48.08 7.05 vs 56.97 7.51 mL/min/1.73 m2, P <.001; group 2: 51.08 8.77 vs 62.78 13.74 mL/min/1.73 m2, UROLOGY 83 (5), 2014
Univariate Analysis
b
P
Multivariate Analysis
b
P
Sex 0.266 .098 0.418 <.001* Age 0.017 .917 0.046 .733 BSA 0.183 .295 0.307 .005* GFR of the resected 0.166 .306 0.084 .632 kidney GFR of remaining 0.031 .849 0.056 .763 kidney (preoperative) GFR of remaining 0.069 .672 0.141 .324 kidney (7 d postoperative) cR2* of resected 0.486 <.001* 0.109 .329 kidney (preoperative) mR2* of resected 0.159 .328 0.069 .543 kidney (preoperative) cR2* of remaining 0.771 <.001* 0.659 <.001* kidney (preoperative) mR2* of remaining 0.243 .131 0.000 .999 kidney (preoperative) cR2*, R2* of cortex; mR2*, R2* of medulla; other abbreviations as in Table 1. *P <.05.
P <.001; Table 3). However, at 1 year, group 1 got further improvement in GFR significantly (P ¼ .02), whereas the GFR in group 2 remains stable and comparable with that at 7 days (P ¼ .064).
COMMENT As is known, unilateral nephrectomy results in a dramatic change of the hemodynamics and function in the remaining kidney.1-3 The hemodynamic adaptation implies a physiology change, especially the oxygenation alternation. Although the long-term follow-up has yielded favorable results,7-10 the oxygen bioavailability alternation in kidney donor after kidney donation remains unclear. To the best of our knowledge, this is the first patient series discussing how the oxygenation gradient evolves in kidney donors, and we did find some interesting results. First, we confirmed what other authors have already described: R2* values were significantly higher in the medulla than that in the cortex in normally functioning kidneys,11,18,19 indicating the medulla works in a hypoxic condition. Second, we observed that after the kidney removal, GFR of the remaining kidney increased significantly, and the cR2* values of remaining kidney decreased dramatically, whereas that in the medulla produced no sensible modification. Animal model has shown that the effective renal blood flow of the remnant kidney increased by 40% immediately after uninephrectomy, from 255.55 27.8 to 357.77 38.9 mL/min.2 Kidney donors presented the same phenomenon, with renal blood flow increased by 25%-32.5% at 1 week and 29%-30.1% at 1 year after 1205.e3
Table 3. Patient characteristics stratified by decrease of cR2* value of the remaining kidney at 7 d Decrease of cR2* Values Patient Characteristics Sex Male Female Age (y) Weight (kg) Height (m) BSA (m2) Serum creatinine (mmol/L) Preoperative 7 d postoperative 1 y postoperative GFR of the resected kidney (mL/min/1.73 m2) GFR of the remaining kidney (mL/min/1.73 m2) Preoperative 7 d postoperative 1 y postoperative cR2* value of the remaining kidney (s1) Preoperative 3 d postoperative 7 d postoperative mR2* value of the remaining kidney (s1) Preoperative 3 d postoperative 7 d postoperative
>10% 2 10 47.83 58.66 1.62 1.62
6 16 46.95 57.18 1.60 1.59
P .402
9.41 6.14 0.06 0.10
.778 .545 .356 .379
9.62 21.03 13.32 10.16
.092 .334 .305 .023
48.08 7.05 56.97 7.51 62.63 11.69
51.08 8.77 62.78 13.74 66.43 10.89
.316 .185 .349
18.76 1.50 16.14 1.07 15.80 0.89
16.92 1.08 15.80 1.07 16.50 1.08
<.001 .413 .062
41.26 5.96 34.64 6.54 37.37 4.08
37.74 3.93 36.37 6.66 39.57 7.67
.084 .502 .346
68.67 111.30 102.55 44.37
6.82 6.79 0.07 0.11
10%
15.21 27.28 26.52 9.01
61.29 103.07 95.47 52.76
Abbreviations as in Tables 1 and 2.
kidney donation.1,3 It is reasonable that increased renal blood flow results in increased GFR, and greater amount of oxygenated blood would improve the tissue oxygenation status. Our findings corroborated the facts that cR2* values decrease significantly by 8.97% from 17.52 to 15.95s1 at 3 days (P <.001) and by 7.82% to 16.15 s1 at 7 days (P <.001), indicating an overall increase in cortical oxygenation. However, there is no significant alternation in the mR2* values, which is in line with the observation by Malvezzi et al,4 suggesting the medulla remains hypoxic. Medulla is believed to consume the main part of oxygen supply to kidney because of electrolyte and water reabsorption. Inhibition of the Na-K-Cl cotransportor by forosemide attenuates deoxyhemoglobin levels and increases medullary tissue pO2, leading to a decrement of mR2* values.20,21 Increased renal blood flow after uninephrectomy consequently induces production of more ultrafiltrate22,23 that the greater solute load to the medullary portions of the tubules forces active oxygen consuming reabsorption mechanism. Consequently, the increase of blood oxygen supply to the medulla was offset. BOLD MRI is widely being used as a noninvasive method to assess tissue oxygen bioavailability. Sadowski et al24 found in transplanted kidneys, there was a significant difference between mR2* values in the group with acute rejection (R2* ¼ 16.2/s) compared with allografts with acute tubular necrosis (R2* ¼ 19.8/s; P ¼ .047) and normal-functioning allografts (R2* ¼ 24.3/s; P ¼ .0003). In addition, combined with single kidney GFR, Chrysochou et al25 found that R2*:GFR was an effective 1205.e4
predictive biomarker for positive renal functional response to revascularization in patient with stenosed renal artery. However, as demonstrated by Michaely et al26 from a study of 368 chronic kidney disease that there was no association of the R2* values of cortex and medulla with the GFR (r ¼ 0.007) and BOLD MRI fails to discriminate between patients with various chronic kidney disease stages. This is confirmed by our results that no significant association was detected between single kidney GFR and the cR2* and mR2* values. Nonetheless, we found that patients with decrease in cR2* values of remaining kidney >10% at 7 days have a more favorable renal function adaptation at 1 year. As indicated by Sigmon et al, increased renal blood flow after uninephrectomy is attributed to reduction in renal vascular resistance, which is induced by enhanced nitric oxide synthesis.27 Thus, we speculated greater improvement in cortex oxygenation status may reflect a greater compensatory potential in vascular dilation chemicals production, and this potential may be an important determinant of the renal function compensation in the long term. This study is not without limitations, some of which are difficult to overcome. First is the small number of patients who were studied. Second, water diuresis is known to reduce the corticomedullary oxygenation. To minimize the effect of patient hydration, patients were told to abstain from food or water from the night before examination, but no certainty of each individual can be made. Third, we only performed a BOLD MRI examination within 7 days after surgery and get the oxygen UROLOGY 83 (5), 2014
bioavailability alteration at the very early phase. How does the oxygenation gradient evolve as the contralateral kidney adapting to its new status and its impact on the long-term renal function are unknown. Further follow-up is required. Finally, our results indicated BOLD MRI is helpful in donor’s postoperative follow-up, rather than patient selection, because we do not know who will get better renal function compensation from the preoperative BOLD MRI evaluation. Thus a BOLD MRIebased novel technique, which may imitate the physiological change after kidney removal is required, such as using vasodilator and measuring the renal blood flow simultaneously.
CONCLUSION In conclusion, the present study has demonstrated that BOLD MRI is a noninvasive tool to study renal physiology. Kidney donation will induce early, profound oxygenation modification within the renal cortex of the remnant kidney, which probably reflects the adaptive mechanism. In addition, greater improvement in cortex oxygenation status suggests better kidney function compensation in moderate term. References 1. Anderson RG, Bueschen AJ, Lloyd LK, et al. Short-term and longterm changes in renal function after donor nephrectomy. J Urol. 1991;145:11-13. 2. Ziada G, Youseif H, Khalil M. Compensatory changes in the function of the remaining kidney immediately after unilateral nephrectomy in sheep. Tohoku J Exp Med. 2009;219:165-168. 3. Indudhara R, Kenney PJ, Bueschen AJ, et al. Live donor nephrectomy in patients with fibromuscular dysplasia of the renal arteries. J Urol. 1999;162:678-681. 4. Malvezzi P, Bricault I, Terrier N, et al. Evaluation of intrarenal oxygenation by blood oxygen level dependent magnetic resonance imaging in living kidney donors and their recipients: preliminary results. Transplant Proc. 2009;41:641-644. 5. Funahashi Y, Hattori R, Yamamoto T, et al. Ischemic renal damage after nephron sparing surgery in patients with normal contralateral kidney. Eur Urol. 2009;55:209-216. 6. Funahashi Y, Hattori R, Yamamoto T, et al. Change in contralateral renal parenchymal volume 1 week after unilateral nephrectomy. Urology. 2009;74:708-712. 7. Ramcharan Thiagarajan, Matas Arthur J. Long-term (20e37 years) follow-up of living kidney donors. Am J Transplant. 2002;2:959-964. 8. Ibrahim HN, Foley R, Tan L, et al. Long-term consequences of kidney donation. N Engl J Med. 2009;360:459-469. 9. Australian and New Zealand Dialysis and transplant Registry. ANZDATA Registry Rep, 2007, pp30.
UROLOGY 83 (5), 2014
10. Oppenheimer SF. Short-, medium-, and long-term follow-up of living donors. Nefrologia. 2010;30(Suppl. 2):100-105. 11. Prasad PV, Edelman RR, Epstein FH. Noninvasive evaluation of intrarenal oxygenation with BOLD MRI. Circulation. 1996;94: 3271-3275. 12. Gomori JM, Grossman RI, Yu-Ip C, et al. NMR relaxation times of blood: dependence on field strength, oxidation state, and cell integrity. J Comput Assist Tomogr. 1987;11:684-690. 13. Haacke EBR, Brown RW, Thompson M, et al. Magnetic Resonance Imaging Physical Principles and Sequence Design. New York: John Wiley & Sons; 1999:343-387. 14. Edelstein WA, Glover GH, Hardy CJ, et al. The intrinsic signalto-noise ratio in NMR imaging. Magn Reson Med. 1986;3: 604-618. 15. Gloviczki ML, Glockner J, Gomez SI, et al. Comparison of 1.5 and 3 T BOLD MR to study oxygenation of kidney cortex and medulla in human renovascular disease. Invest Radiol. 2009;44: 566-571. 16. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16:31-41. 17. Mosteller RD. Simplified calculation of body-surface area. N Engl J Med. 1987;317:1098. 18. Djamali A, Sadowski EA, Muehrer RJ, et al. BOLD-MRI assessment of intrarenal oxygenation and oxidative stress in patients with chronic kidney allograft dysfunction. Am J Physiol Ren Physiol. 2007; 292:513-522. 19. Han F, Xiao W, Xu Y, et al. The significance of BOLD MRI in differentiation between renal transplant rejection and acute tubular necrosis. Nephrol Dial Transpl. 2008;23:2666-2672. 20. Warner L, Glockner JF, Woollard J, et al. Determination of renal cortical and medullary oxygenation using blood level-dependent magnetic resonance imaging and selective diuretics. Invest Radiol. 2011;46:41-47. 21. Li LP, Vu AT, Li BS, et al. Evaluation of intrarenal oxygenation by BOLD MRI at 3.0 T. J Magn Reson Imaging. 2004;20:901-904. 22. Hosteller TH, Olson JL, Rennke HG, et al. Hyperfiltration in remnant nephrons: a potentially adverse response to renal ablation. J Am Soc Nephrol. 2001;12:1315-1325. 23. Bohlouli A, Tarzamni MK, Zomorodi A, et al. Remnant kidney function and size in living unrelated kidney donors after nephrctomy. Saudi J Kidney Dis Transpl. 2010;21:246-250. 24. Sadowski EA, Djamali A, Wentland AL, et al. Blood oxygen leveldependent and perfusion magnetic resonance imaging: detecting differences in oxygen bioavailability and blood flow in transplanted kidneys. Magn Reson Imaging. 2010;28:56-64. 25. Chrysochou C, Mendichovszky IA, Buckley DL, et al. BOLD imaging: a potential predictive biomarker of renal functional outcome following revascularization in atheromatous renovascular disease. Nephrol Dial Transpl. 2012;27:1013-1019. 26. Michaely HJ, Metzger L, Attenberger UI, et al. Renal BOLD-MRI does not reflect renal function: a prospective study in 368 patients. Proc Intl Soc Mag Reson Med. 2011;19:441. 27. Sigmon DH, Gonzalez-Feldman E, Cavasin MA, et al. Role of nitric oxide in the renal hemodynamic response to unilateral nephrectomy. J Am Soc Nephrol. 2004;15:1413-1420.
1205.e5