Assessment of Renal Viability by Phosphorus-31 Magnetic Resonance Spectroscopy

_0022-534 7/86/1354-0866$02.00/0 Vol. 135, April Printed in U.S.A.

THE JOURNAL OF UROLOGY

Copyright © 1986 by The Williams & Wilkins Co.

ASSESSMENT OF RENAL VIABILITY BY PHOSPHORUS-31 MAGNETIC RESONANCE SPECTROSCOPY PETER N. BRETAN, JR.,* DANIEL B. VIGNERON, THOMAS L. JAMES, RICHARD D. WILLIAMS, HEDVIG HRICAK, KLAUS-PETER JUENEMANN,t T. S. BENEDICT YEN AND EMIL A. TANAGHO From the Departments of Urology, Radiology, Pharmaceutical Chemistry and Pathology, University of California, San Francisco, and The Veterans Administration Medical Center, San Francisco, California

ABSTRACT

To assess the applicability of phosphorus-31 magnetic resonance spectroscopy (31 P-MRS) in the analysis of renal transplant viability and preservation techniques with respect to pre-transplant ischemia, we studied two rat groups. Twenty-five rat kidneys were subjected to various time increments of warm ischemia (Group A), and 31 P-MRS was performed on each kidney at time intervals of up to 72 hours during simple hypothermic storage. We correlated findings of 31 P-MRS with simultaneous findings of electron microscopic (EM) ultrastructural viability parameters (in Group A) and subsequent survival and renal function in 30 rats (Group B) subjected to similar amounts of variable ischemia. Intracellular phosphorus metabolite levels were nondestructively monitored by 31 P-MRS via spectral peaks of NAD, sugar monophosphates (SP), and inorganic phosphate (PJ. We concluded: 1) SP/Pi and NAD/Pi ratios decay in a time-dependent manner for both warm and cold ischemia, although this process is much slower during cold storage; 2) EM viability parameters correlate with the development of acute tubular necrosis (irreversible damage) versus nonviability (gross cell death) on a qualitative basis only; and 3) 31 P-MRS enables a quantitative assessment of renal viability and ischemic renal damage and can predict the degree of acute tubular necrosis and post-ischemic renal function. 31 P-MRS is potentially a noninvasive, nondestructive method of assessing viability during simple hypothermic storage of the rat kidney. Preliminary evidence shows that this MRS method can be applied to human kidney viability studies for clinical renal transplantation and urologic research concerning renal preservation. Although renal viability is difficult to quantify, it is generally thought to be inversely related to the degree of ischemic injury to the kidney. Renal ischemia is commonly associated with cadaver donor nephrectomy and subsequent cold storage (both simple and perfusion hypothermic preservation) of the cadaveric kidneys. 1 Prediction of renal viability is currently based on estimates of total operative ischemia times (time from a significant hypoperfusion of the kidney until subsequent cold perfusion) and duration of cold storage of the organ. 1 In the major transplant centers (>50 transplants per year), these inexact estimates have led to organ wastage ranging from five to 20 per cent. 1 • 2 Mitochondrial membrane and matrix stability, measured by electron microscopy 3 as well as intracellular adenine phosphate assay,4 correlates with intact metabolic pathways and can predict renal viability. 5 • 6 However, this is not practical for clinical transplantation 7 or clinical estimation of viability. 8 At present, there is no clinically useful, noninvasive method to assess the metabolic competence of an ischemically injured site within the body or a cold-stored kidney awaiting transplant to a selected recipient. 9 Magnetic resonance spectroscopy (MRS) is a noninvasive diagnostic technique that will quantify intracellular metabolic parameters and has been studied in muscle, brain, kidney, liver and heart of both intact animal models and humans, 10-19 Tissue viability has been studied extensively, making MRS useful in Accepted for publication October 31, 1985. * Requests for reprints: Dept. of Urology, U-518, University of California, San Francisco, CA 94143. Winner of the Third Prize for Laboratory Research, 1985 AUA Essay Contest. t Supported in part by the B. Braun Foundation of the Federal Republic of Germany.

866

renal transplantation. Heretofore, there has been no noninvasive or nondestructive method to measure intracellular metabolic chemical changes accurately. Although MRS is historically a much older research tool than magnetic resonance imaging (MRI), and both depend on powerful magnetic systems, they are quite different. The imaging methods exploit magnetic characteristics of protons in water to produce a tom ographic image similar to that obtained from x-ray computed tomography. However, one of the greatest potentials of magnetic resonance spectroscopy is to determine the concentration of compounds containing such nuclei as phosphorus-31 (31 P), carbon-13, and fluorine-19 in selected volumes of intact animal and human tissues. 14 A representative in vivo spectrum of a normal rat kidney shows relative concentrations of specific adenine nucleotides and other phosphorus metabolites (fig. 1). 14' 18 Chemical shifts are expressed numerically in parts per million (ppm) relative to a reference signal. The area under the signal gives a measure of the number of nuclei contributing to the peak; therefore, the amounts of specific compounds in the tissue can be measured.10· 11 A comprehensive description of MRS is beyond the scope of this discussion, and there are many complete reviews. 12- 14• 19 The greatest single advantage of MRS is its ability to measure noninvasively intracellular phosphorus metabolites with 31 P levels of the kidney in vivo and ex vivo, enabling us to study metabolic parameters and to establish a strong viability correlation. To determine if the use of MRS is feasible and practical in clinical renal transplantation, we designed a prospective study in which intracellular phosphorus metabolite levels were correlated with variable amounts of warm renal ischemia time (25

867

ASSESSMENT OF RENAL VIABILITY Rat kidney

+20

0

-20

Chemical shift (PPM) FlG. 1. Phosphorus-31 spectrum (97 MHz) of in vivo rat kidney. Chemical shifts measured in parts per million (ppm) from MDPA reference peak. Peak assignments used were previously described by Koretsky et al.18 1 = beta-ATP; 2 = alpha-ATP + NAD(H) + NADP(H); 3 = gamma-ATP; 4 = phosphocreatine (P-Cr); 5 = phosphodiester; 6 = 7 = inorganic phosphate (P;); 8 = sugar monophosphates (SP); 9 = unknown; 10 = dimethylphosphate acid reference (MDPA).

to 37C) and subsequent cold storage (4C) time in the rat kidney. MRS parameters were also correlated with electron microscopic ischemia findings and subsequent renal function and survival after ischemic injury.

MATERIALS AND METHODS

Induced warm ischemia followed by simp/,e hypothermic storage. Mature male Sprague-Dawley rats (180 to 250 gm.) were anesthetized with intraperitoneal injection ofpentobarbital (40 mg./kg.) for all studies. Surgical exposures were made via a midline transperitoneal incision from xyphoid to genitalia. Rats whose kidneys were removed for 31 P-MRS analysis (Group A; N = 25) previously had their aorta cannulated with a 24-gauge angiocatheter (just above the iliac bifurcation), which was secured with a 5-0 silk suture. The inferior vena cava was vented and the animals were exsanguinated. The thorax was simultaneously opened through the anterior diaphragmatic wall and immediate clamping of the suprarenal aorta and inferior was administered. Sixty milliliters of 4 to 5C sterile lactated Ringer's solution with five per cent dextrose (pH 5.0) was delivered through the aortic cannula. The kidneys were removed immediately after specific warm ischemia times and kept in simple cold storage in magnetic resonance spectroscopy test tubes at 4 to 5C. Group I rats (7) had no ischemia before cold storage; Group II (8), Group III (6), and Group IV (4) underwent 20, 60 and 120 minutes, respectively, of warm ischemia before flushing and simple hypothermic storage. of warm ischemia before flushing and simple hypothermic storage. Magnetic resonance spectroscopy monitoring 31 P nuclear magnetic resonance was accomplished on all kidneys (without interrupting continuous cold storage) at intervals of approximately 2, 24, 48 and 72 hours after cold renal flushing. Samples from kidneys were taken during each 31 P-MRS scan period for electron microscopic ultrastructure analysis. Renal function and survival after ischemia. A separate group of rats was studied for the effects of varying amounts of ischemia on subsequent renal function and survival (Group B; N = 30). In this group, the right ureter was isolated and ligated with 5-0 dexon ties, followed by systemic heparinization with 300 units aqueous heparin. The left renal vascular pedicle was isolated and clamped for varying lengths of time: Group I rats (7) had no ischemia (pedicle not clamped); Group II (8) 20 minutes' ischemia; Group III (6) 60 minutes'; and Group IV (9) 120 minutes' warm ischemia time. After the set length of ischemia time had elapsed, the vascular clamps were removed and perfusion of kidneys immediately resumed in all rats. The

abdomen was closed. Blood samples were acquired from rat tail veins for serum creatinine levels (by colorimetric alkaline picrate analysis) at 72 hours in Groups I to III and at 24 hours for Group IV (depending on animal survival). 31 P magnetic resonance spectroscopy. At the end of the warm ischemia period, the kidneys (Group A) were flushed with the cold perfusate and immediately placed in 10 mm. MRS tubes, stored at 4C for the remainder of the experiment. A sealed capillary tube containing 1 M methylenediphosphoric acid (MDPA; Sigma), buffered to a pH of 9.0, was placed in each MRS tube as a ref~rence standard. The 31 P nuclear magnetic resonance, or 31 P-MRS, was obtained at 97.6 MHz oh a homebuilt spectrometer using a 5.6 Tesla superconducting magnet with a 7.6 cm. diameter bore (Cryomagnet Systems) and a Nicolet 1180/293B data system. The spectra were obtained from the Fourier transform of 500 acquisitions with a delay of two seconds and a tilt angle of 45 °. When compared with spectra obtained with a delay of 10 seconds, no change in peak intensity was observed. All spectra were collected at various times over a 72-hour period. Chemical shifts were referenced (in ppm) to the MDPA peak, and the peak assignments described by Koretsky et al. (fig. 1) were used. 18 All peak intensity ratios were obtained with the Nicolet computer line fitting program. Statistical analysis was done with the unpaired Student's t test. Electron microscopic analysis of variably ischemic renal tissue. Immediately upon removal of the kidneys (Group A), 1 mm.thick shavings of renal cortex and medulla were dropped in five ml. of two per cent glutaraldehyde, and 0.1 M sodium cacodylate, pH 7.4, at 4C. After 24 hours the tissue fragments were rinsed, post-fixed in 2 per cent osmium tetroxide for one hour, and then embedded. Ultrathin sections were cut from each block, stained with two per cent uranyl acetate and lead citrate, and examined in a Phillips EM-201 microscope. Proximal tubular epithelial cells were examined for evidence of ischemic damage. 8 RESULTS

P-MRS scans showed measurable sugar monophosphate and NAD peaks throughout the first 50 hours of cold storage time in all kidneys in Group A (figs. 2-4). The spectral peaks labelled NAD may have contributions from both oxidized and reduced forms of NAD and NADP. The 31 P spectra gave no evidence of ATP or ADP in the ex vivo cold-stored kidney; 31

O Minutes'ischemia time P;

MOPA

72

Hours cold

48

storage 24

5

0

-10

-20

-30

PPM

Chemical shift

FIG. 2. In kidneys subjected to no warm ischemia (0-minute), timedependent decay is noted up to 48 hours in both sugar phosphate and NAD peak intensities. PD= phosphodiester.

868

BRETAN AND ASSOCIATES

for more than 48 hours (N = 9) and all serum creatinines were measured at 24 hours post-ischemia. The average creatinine in this group was 4.7 mg./dl. The 4-week survival for Groups I through III was 100 per cent (table 1). Non-fatal azotemia associated with Group III corresponded to an NAD/P; ratio of 0.17 and sugar monophosphate/P; ratio of 0.45 (figs. 5 and 6). Electron microscopic findings were also correlated with the amount of ischemia and cold storage time (table 2). Established irreversible ischemic cell damage findings in Group A were found in Group I at 48 hours' cold storage time, Group II at 24 hours, and Group III immediately. This corresponded to an NAD/P; ratio of 0.17 and a sugar monophosphate/P; ratio of 0.45 (table 1, figs. 5 and 6).

20 Minutes· ischemia time P;

SP

70

46 Hours cold storage

22

DISCUSSION 4

-10

0

-20

-30

PPM

As cells become anoxic, oxidative phosphorylation ceases. Adenosine triphosphate (ATP) stores are rapidly depleted and virtually all energy-dependent functions cease. 20 Electrochemical gradients that depend on energy-requiring ion-exchange pumps cannot be maintained (for example, intracellular calcium increases with ischemia). 4 Glycolysis is initially acceler-

Chemical shift

FIG. 3. 20-minute warm ischemia kidneys have time-dependent decay of SP and NAD peak intensities, as does 0-minute group; however, they start with significantly lower amounts. PD= phosphodiester.

1.0

.

60 Minutes· ischemia time

Warm ischemia time A--

None

t----

60mm

P;

MOPA

SP

,

70

Hours

~

50

cold storage

11----· 120 m I n

~

. Sugar monophosphate

P; .5

\ ~

24

~

~~~~~

4

0

-10

-20

-30

3

P~t-i

Chemical shift

FIG. 4. In 60-minute warm ischemia group, SP and NAD peak intensities were significantly diminished at 24 hours (p < 0.005 compared with less ischemic groups). PD = phosphodiester.

thus, contributions from alpha-ADP or alpha-ATP to the NAO peak are negligible. The inorganic phosphorus (P;) chemical shift, which corresponded to a pH of approximately 7.5, changed only slightly during the study. Using the sugar monophosphate/P; ratio (fig. 5) and the NAD/P; ratio (fig. 6), a time-dependent decay was found from O to 72 hours of simple cold storage of kidneys. All groups were significantly (p < 0.05) different from each other at approximate scanning times of 2, 24 and 48 hours (table 1), which helped establish a separate but parallel decay curve of NAO /P; and sugar monophosphate/ P; ratios (figs. 5 and 6). Renal function studies of solitary functioning post-ischemic kidneys (Group B) and subsequent survival revealed no difference in serum creatinine between Groups I and II; both the zero and 20-minute ischemic groups had an average serum creatinine of 0.6 mg./dl. However, Group 1II was noted at 72 hours postoperatively to have a significantly (p < 0.05) elevated creatinine (2.1 mg./dl.) when compared with Groups I and II. For the 120-minute ischemia group (Group IV), no rat survived

• 10

20

30

40

• 50

60

70

HRS

Simple hypothermic (4°C) storage

FIG. 5. 31 P-MRS of rat kidney subjected to variable warm and cold ischemia times. Sugar monophosphate/inorganic phosphate (SP/P,) ratio plotted against time of simple (4C) hypothermic storage (SHS). All curves are derived from median values, with significant (p < 0.05) separation of each curve from others up to 50 hours of cold storage. Time-dependent decay of simple hypothermically stored kidney is noted. Ischemic injury associated with acute tubular necrosis and azotemia correlates with SP /P, ratio of 0.45. This level of cell energy depletion is noted at following times: immediately for 60 minutes of warm ischemia; at 13 hours of cold storage for 20 minutes' warm ischemia; and at 23 hours for O minutes' warm ischemia. This level corresponds with development of irreversible ischemic changes noted in EM (table 2). Level of SP/P, = 0.3 corresponds with EM stage 4 or gross cell death and 0% survival of rats (table 1) post-ischemic injury. The greater the warm ischemia time, the faster depletion levels of SP/ P, are reached during cold storage.

869

ASSESSMENT OF RENAL VIABILITY

ated and cells produce large amounts of lactic acid (a decrease in intracellular pH is predicted, although at 4C all glycolytic pathways are inhibited). 20 However, inadequate assay techniques make it impossible to correlate with certainty the depletion of energy stores with irreversible ischemic cell damage. 20 Nevertheless, previous intracellular assay studies for adenine nucleotides established, for rat kidneys, some basic trends in ischemic metabolic changes: after 15 minutes of warm ischemia time, ATP falls to 13 per cent of normal controls; total adenine nucleotides, however, fall at a much lower rate- 72 per cent at 15 minutes and 18 per cent at 120 minutes (per cent of normal nonischemic control). 4 Diminished ATP reflects uncoupling of oxidative phosphorylation, but this process is reversible if a sufficient amount of total adenine nucleotides is available for regeneration of ATP. 21 The ability to regenerate ATP has been .5

Warm ischemia time . & . - None

• - - 20min. .4

• - - - - 60min. .,____, 120 min.

.3

t

NAO pi

. ..

shown to correlate best with tissue survival; 20 however, in some studies, total adenine nucleotides have correlated with survival. 6 A difficulty shared by many of these reports is that free concentrations of ATP, ADP and AMP cannot be precisely measured by usual chemical analysis because these metabolites are stripped from protein-binding sites during tissue extraction,22 causing an overestimation of active mobile metabolites. 23 In fact, total adenine nucleotides measured by 31P-MRS, which monitors only free nucleotides because of the magnetic characteristics of their nuclei, is 10 mM below stated assay measurements, reflecting this artifact. 22 Studies have suggested that there is a correlation between diminished total adenine nucleotides and tissue viability: 23 loss of total adenine nucleotides occurs on the same time scale as irreversible tissue damage in cold storage in both rat and human kidneys. 10• 11 However, variable warm ischemia has not been concurrently studied or post-ischemic renal function correlated with MRS parameters in a prospective manner as in this report. ATP has been found to disappear after eight minutes of ischemia at 37C and after 90 minutes of ischemia at OC. 11 These findings are in accordance with our data. In vivo studies support the clinical use of MRS in brain studies. Topical magnetic resonance spectroscopy studies of the brains of newborn infants with birth asphyxia have found a reduced phosphocreatine concentration and ip.creased P;, which reverted to normal as the clinical condition improved. 12·13· 15 These findings are entirely in accord with studies (here at our institution) on rat and rabbit organs in vivo that were subjected to ischemia or hypoxia. 14·16• 18 · Preliminary studies have shown that in vitro ATP and P; correlated with early human graft function in 11 patiep.ts. 24 These preliminary findings have indicated that MRS can non-

'

.2

••

'

&

Comparison of electron microscopic viability parameters with variable warm and cold ischemia in group A rat kidneys

TABLE 2.

· · · · · · · · · · ·,,., · ·.................... ·~·· , .............. Nonfatal azotemia

.. ' . . ····--.•.:.¥

~::.~~··········~···· .... ..... •

..........

............

••• •

...

· .•~............... ...~ .•.•.. Nonfunctioning kidney &

,,

..

........

••

&

',,, ........~ ........ ................... ............................ ' .................... _ --.............. ............:::-..-::::::---

_~

Intracellular EM Pathologic Ischemic Staging" 1 1 2 3

20

30

40

50

60

immediate

immediate

3 3 4 4 4

20 20 20 20 60 60 60 60 60

4

120

immediate

2

70

2 3

Simple hypothermic (4°C) storage

4

FIG. 6. 31 P-MRS of rat kidney subjected to variable warm and cold ischemia times. NAD/inorganic phosphate (NAD/P,) ratio plotted against time of simple hypothermic storage. Similar changes of timedependent decay are noted in NAD/P, ratio as for SP/P, ratios. Moderate ischemia associated with nonfatal azotemia is reached at NAD/ P, = 0.17. Level of 0.13 corresponds to fatal uremia EM stage 4 (gross cell death).

TABLE 1.

I II III IV

4 24

48 72 4 24

48 immediate 4 24

48 72

* For staging criteria, see table 1.

Comparison of 31P-magnetic resonance spectroscopy and electron microscopy with post-ischemic renal function and survival All

Group

Simple Hypothermic Storage Time (Hours)

0 0 0 0 0

3

0

10

Warm Ischemia Time (Minutes)

Warm Ischemia Time (Minutes) 0 20 60 120

Group A* Peak Intensity Ratios Median Values (N)§ NAD/P; SP/P; 0.80 0.63 0.45 0.30

0.40(7) 0.36 (8) 0.17 (6) 0.13 (4)

Group Bt Stage of EM Ischemic Damages 1 2 3 4

Survival at 4 Weeks

Average Creatinine

(%)

(N)

(mg./dl.)

(N)

100 100 100 0 at 48 hrs

(7) (8) (6)

0.6 0.6 2.0 4.7

(7) (7) (5) (5)

(9)

* Hypothermically stored rat kidneys (see METHODS). t The manipulated rat (see METHODS). § All groups are significantly different (p < 0.005) from each other. Criteria for EM Ultrastructural Staging: Stage 1) No evidence of intracellular damage. The nuclei and mitochondria are completely normal. No evidence of ischemic damage. Stage 2) Mild ischemic damage. Some condensation of mitochondrial matrix and mild mitochondrial swelling noted. Stage mostly associated with reversible changes. Stage 3) Moderate ischemic damage associated with irreversible changes. Frequent floculent densities in inner compartment of mitochondria with disruption of cell membrane noted. Stage 4) Gross cell death with intracellular calcifications and karyolysis.

870

BRETAN AND ASSOCIATES

destructively measure intracellular phosphorus metabolites that correlate with ischemic damage. However, further investigation is needed. MRS parameters were correlated with established intracellular electron microscopic ischemic changes. In previous studies of the rat kidney, 3 • 8 electron microscopic changes caused by ischemia were: loss of proximal tubule brush border, swelling, and mitochondrial and matrix condensation with large floculent densities, intracellular calcification, and karyolysis. Of these parameters, the last three have been correlated with irreversible cell injury and correspond to a warm ischemia time of between 60 and 120 minutes in previous rat studies 3 • 4-in agreement with our findings. As previously discussed, electron microscopy parameters are not practical for clinical assessment of viability. In this study, viability is assessed qualitatively at best. For example (tables 1 and 2), EM stage III (irreversible cell damage) correlates with the subsequent development of acute tubular necrosis (ATN), and EM stage IV (gross cell death) with nonviability of the kidney. However, the LDso (the dose causing a 50 per cent fatality) of warm ischemia time lies between 60 and 120 minutes in the rat kidney, and 31 P-MRS (table 1, figs, 5 and 6) continues to show a steady decay of SP/ P; as well as NAD/P; ratios between these two limits. Thus, 31 P-MRS can assess quantitatively the ischemic damage associated with ATN versus nonviability. Preliminary data suggest that these findings are all reproducible in the dog model and humans. Viability of renal and other tissue could not be assessed instantaneously and nondestructively heretofore. Not only is MRS emerging as a useful research tool in urology and renal transplantation, but it can potentially be incorporated with MRI in clinical diagnosis. With present more powerful magnets, it is now possible to use the same instrument for MRI and MRS (with surface coils). 25 For example, the 5.6 and 2.0 Tesla magnets in this laboratory can do both. With these capabilities, renal transplant patients may be assessed for renal viability perioperatively. Preoperative assessment of the harvested cadaver kidney viability may be quite useful. For example, relative amounts of ATN (which may be as high as 45 per cent in cadaver transplant patients 1 ) caused by ischemic injury may be quantitatively predicted preoperatively by MRS analysis. Although it does not affect long-term graft survival, the presence of ATN may affect subsequent nephrotoxicity if cyclosporine is used to prevent rejection. 26 Thus, MRS may aid in lowering the toxicity of immunosuppressive therapy. As the technology of MRS rapidly advances and in vivo spectra are obtained with animals, 18 it is conceivable that postoperative assessment of renal transplant patients by MRS may help differentiate between ATN, rejection, and cyclosporine nephrotoxicity.

SUMMARY

In the isolated ex vivo rat kidney subjected to variable amounts of warm and cold ischemia, intracellular phosphorus metabolite levels, monitored by 31 P magnetic resonance spectroscopy, were correlated with previously established electron microscopic viability parameters and parallel studies of postischemic renal function. We found a strong correlation of the sugar monophosphate/P; and the NAD/P; ratios with subsequent renal function and viability. A ratio of sugar monophosphate/P; of 0.30 to 0.45 and an NAD/P; ratio of 0.13 to 0.17 correspond with nonfatal azotemia; ratios above this range are associated with completely normal renal function; ratios less than these limits are associated with irreversible ischemic renal damage, inadequate to sustain life in the rat model. A timedependent decay of sugar monophosphate/P; and NAD/P; ratios has been established in the simple cold-stored rat kidney. Acknowledgments. We are indebted to Ann Halili and Noel Taylor for technical assistance.

REFERENCES

1. Goldszer, R. C., Strom, T. B. and Tilney, N. L.: Acute renal failure

2. 3.

4. 5.

associated with renal transplantation. In: Acute Renal Failure. Edited by B. M. Brenner and J. M. Lazarus. Philadelphia: W. B. Saunders, chapter 21, pp. 555-566, 1983. Kootstra, G.: Organ preservation. Transplant. Proc., 16: 185, 1984. Trump, B. F., Mergner, W. J., Kahng, M. W. and Saladino, A. J.: Studies of the subcellular pathophysiology of ischemia. Circulation, 53(suppl. 1): 17, 1976. Humes, H. D. and Weinberg, J. M.: Alterations of renal tubular cell metabolism in acute renal failure. Mineral Electrolyte Metab., 9: 290, 1983. Feinberg, H.: Energetics and mitochondria. In: Organ Preservation: Basic and Applied Aspects. Edited by D. E. Pegg, I. A. Jacobsen and N. A. Halasz. Lancaster: MTP Press Limited, pp. 3-15, 1982.

6. Calman, K. C.: The prediction of organ viability. I. An hypothesis. Cryobiology, 11: 1, 1974. 7. Nunnally, R. L.: In vivo monitoring of metabolism with nuclear magnetic resonance spectroscopy. Semin. Nucl. Med., 13: 377, 1983.

8. Venkatachalam, M.A., Bernard, D. B., Donohoe, J. F. and Levinsky, N. G.: Ischemic damage and repair in the rat proximal tubule: differences among the 8 1, 8 2, and 8 3 segments. Kidney Int., 14: 31, 1978. 9. Jacobsen, I. A. and Pegg, D. E.: Kidney. In: Organ Preservation for Transplantation. Edited by A. M. Karrow, Jr. and D. E. Pegg. New York: Marcel Dekker, Inc., pp. 553-575, 1981. 10. Sehr, P. A., Radda, G. K., Bore, P. J. and Sells, R. A.: A model kidney transplant studied by phosphorus nuclear magnetic resonance. Biochem. Biophys. Res. Comm., 77: 195, 1977. 11. Sehr, P. A., Bore, P. J., Papatheofanis, J. and Radda, G. K.: Nondestructive measurement of metabolites and tissue pH in the kidney by 31 P nuclear magnetic resonance. Br. J. Exp. Pathol., 60: 632, 1979. 12. Gordon, R. E., Hanley, P. E. and Shaw, D.: Topical magnetic resonance. Prog. NMR Spectroscopy, 15: 1, 1982. 13. Iles, R. A., Stevens, A. N. and Griffiths, J. R.: NMR studies of the metabolites in living tissue. Prog. NMR Spectroscopy, 15: 49, 1982. 14. James, T. L.: In vivo nuclear magnetic resonance spectroscopy. In:

NMR, CT, and Interventional Radiology. Edited by A. A. Moss, E. J. Ring and C. B. Higgins. San Francisco: University of California Press, pp. 235-244, 1984. 15. Cady, E. B., Costello, A. M. de L., Dawson, M. J., Delphy, D. T., Hope, P. L., Reynolds, E. 0. R., Tofts, P. S. and Wilkie, D.R.: Non-invasive investigation of cerebral metabolism in newborn infants by phosphorus nuclear magnetic resonance spectroscopy. Lancet, 1: 1059, 1983. 16. Bore, P. J., Sehr, P.A., Chan, L., Thulborn, K. R., Ross, B. D. and Radda, G. K.: The importance of pH in renal preservation. Transplant. Proc., 13: 707, 1981. 17. Rosen, B. R. and Brady, T. J.: Principles of nuclear magnetic resonance for medical application. Sero. Nucl. Med., 13: 308, 1983. 18. Koretsky, A. P., Wang, S., Murphy-Boesch, J., Klein, M. P., James,

T. L. and Weiner, M. W.: 31 P NMR spectroscopy of rat organs, in situ, using chronically implanted radiofrequency coils. Proc. Natl. Acad. Sci., 80: 7491, 1983. 19. Gadian, D. G.: Nuclear Magnetic Resonance and Its Applications to Living Systems. Oxford: Clarendon Press, 1982. 20. Farber, J. L., Chien, K. R. and Mittnacht, S. Jr.: The pathogenesis of irreversible cell injury in ischemia. Am. J. Pathol., 102: 271, 1981. 21. Southard, J. H., Senzig, K. A., Hoffmann, R. M. and Belzer, F. 0.:

Energy metabolism in kidneys stored by simple hypothermia. Transplant. Proc., 9: 1535, 1977. 22. Dawson, M. J. and Wilkie, D. R.: Muscle and brain metabolism studied by 31 P nuclear magnetic resonance. In: Recent Advances in Physiology. Edited by P. Baker. Edinburgh: Churchill-Livingstone, 1983. 23. Wong, G. G. and Ross, B. D.: Application of phosphorus nuclear magnetic resonance to problems of renal physiology and metabolism. Mineral Electrolyte Metab., 9: 282, 1983. 24. Chan, L., French, M. E., Gadian, D. G., Morris, P. J., Radda, G. K., Bore, P. J., Ross, B. D. and Styles, P.: Study of human kidneys prior to transplantation by phosphorus nuclear magnetic

ASSESSMENT OF RENAL VIABILITY resonance. In: Organ Preservation: Basic and Applied Aspects. Edited by D. E. Pegg, I. A. Jacobsen and N. A. Halasz. Lancaster: MTP Press, pp. 113-119, 1982. 25. Bottomley, P.A., Hart, H. R. Jr., Edelstein, W. A., Schenck, J. F., Smith, L. S., Leue, W. M., Mueller, 0. M. and Redington, R. W.:

871

Anatomy and metabolism of the normal human brain studied by magnetic resonance at 1.5 Tesla. Radiology, 150: 441, 1984. 26. The Canadian Multicentre Transplant Study Group: A randomized clinical trial of cyclosporine in cadaveric renal transplantation. N. Engl. J. Med., 309: 809, 1983.