Positron Emission Tomography: A New Method for Determination of Renal Function

0022-5347 /93/1503-1064$03.00/0

Vol. 150, 1064-1068, September 1993 Printed in U. S. A.

THE J OURNAL OF UROLOGY Copyright © 1993 by AMERICAN UROLOGICAL ASSOCIATION, INC.

POSITRON EMISSION TOMOGRAPHY: A NEW METHOD FOR DETERMINATION OF RENAL FUNCTION DAVID KILLION, EGBERT NITZSCHE, YONG CHOI, HEINRICH SCHELBERT J. THOMAS ROSENTHAL*

AND

From the Divisions of Urology and Nuclear Medicine, University of California, Los Angeles, California ABSTRACT

Positron emission tomography (PET) is a newly evolving diagnostic modality that has been widely used in many facets of clinical medicine, but whose use in the diagnosis and management of disorders of the kidney has not been previously described. Employing the radiotracer N-13 ammonia, flow­ dependent extraction of this compound after intravenous injection was used to measure renal blood flow (RBF) in a swine model (N = 10) . A mean baseline value of 3.16 ml./min./gm. kidney was obtained with this method, in close agreement with values previously reported using established invasive techniques. Four conditions known to affect RBF were also studied to determine the ability of PET to detect changes in RBF. Kidneys were subjected to varying durations of warm ischemia, demonstrating a progressive decrease in RBF with increasing ischemic insult, with return to normal significantly impaired in animals exposed to the greatest degree of ischemia ( 180 minutes versus 150 or 120 minutes ischemia) . Cross-transplant between animals produced acute allograft rejection and a corresponding marked decrease in RBF that failed to normalize. After unilateral nephrectomy, RBF increased two-fold in the remaining kidney by 7 days (R = 0.79) , as predicted for compensatory renal hypertrophy. Lastly, there was an inverse, linear relationship between toxic cyclosporine level and RBF (R = 0.68) , indicative of vascular-mediated cyclosporine nephrotoxicity. Positron emission tomography is safe and efficient, and yields an accurate measurement of RBF in several important physiologic states. The development of PET as a quantitative measure of renal function is promising. KEY WORD S :

tomography; emission computed; renal circulation

Positron emission tomography (PET) has recently emerged as an important diagnostic modality in clinical medicine. It is currently employed in the evaluation of central nervous system disorders (such as Alzheimer's and Parkinson's disease); as a method to map micrometastatic disease, as in metastatic mel­ anoma; and as a means to assess in vivo myocardial perfusion. 1 Since PET involves cross-sectional imaging of a particular organ via quantitation of the distribution of a radioactive indicator, its clinical usefulness would appear to be in those organ systems most intimately regulated by blood flow, for example the heart, brain, or kidney. However, in contrast to the heart and brain, the use of PET in the diagnosis and management of renal disorders has not been previously de­ scribed. The current study was undertaken to determine whether the flow-dependent tissue extraction of the radiotracer N-13 am­ monia could be used to estimate renal blood flow (RBF) as a measure of renal function. Since adequate renal function is critically dependent on the integrity of RBF, attempts have been made to assess changes in renal circulation and microcir­ culation to improve understanding of the pathophysiology of various renal conditions. However, there is no accepted means to measure RBF rapidly, reliably and noninvasively, either experimentally or clinically. 2 The availability of a noninvasive functional imaging method that would provide quantitative estimates of renal perfusion in a variety of clinically relevant situations (advanced arterial hypertension, acute tubular neAccepted for publication May 7, 1993. * Requests for reprints: UCLA Division of Urology and Renal Trans­ plantation, 10833 LeConte Ave., Los Angeles, California 90024. This work was supported by the UCLA Division of Urology Research Fund and by Research Grants HL 29845 and HL 33177, National Institutes of Health, Bethesda, Maryland, and by an Investigative Group Award by the Greater Los Angeles Affiliate of the American Heart Association, Los Angeles, California.

crosis, acute rejection of a renal allograft, or cyclosporine toxicity) would enhance understanding and provide an im­ proved objective measure of renal function in these conditions. The purpose of this study was to determine the utility of PET as an accurate, reproducible and clinically relevant diag­ nostic measure of renal function. Renal blood flow was meas­ ured in the normal kidney, in addition to four clinical conditions known to influence RBF: ischemia,3• 4 acute rejection,5 • 6 cyclo­ sporine nephrotoxicity7-9 and compensatory renal hypertrophy after unilateral nephrectomy. 10-12 The pig was employed as an experimental subject, due to its similarity to humans in terms of retroperitoneal and abdominal anatomy, renal anatomy and renal physiology. 13 • 14 MATERIALS AND METHODS

Surgical Procedure, Autotransplant. Young adult, female

Duroc farm pigs (25 to 30 kg. body weight) were acclimated to the vivarium for 14 days before surgery. Animals were fed standard pig chow and given free access to water until 6 hours prior to surgery. Animals (N = 8) were preanesthetized with ketamine 20 mg./kg. body weight and atropine 0.4 mg./kg. intramuscularly, and were maintained during the procedure on halothane gas, 0.5 to 2.0% ET with oxygen at 5 1. per minute. Antibiotic prophylaxis, Kefzol, 500 mg. intramuscularly, was given at time of surgery. A left neck incision was performed for placement of a central venous line into the external jugular vein for intraoperative fluid administration, central venous pressure monitoring and eventual N-13 ammonia delivery. In addition, an indwelling arterial line was placed into the left carotid artery for blood pressure monitoring and for drawing blood samples. Both of these lines were tunnelled subcutane­ ously to the posterior neck and secured there for further use. The left kidney was mobilized and removed through a midline abdominal incision. It was immediately flushed with 200 to 300

1064

1065

P O S ITRON EMISSION T O M O G RAPHY AND RENAL BLOOD F L O W

mL of ambient Lactated Ringer's solution (supplemented with heparin 10 U/ml.) at a height of 100 cm. The ureter was left intact and externally occluded with umbilical tape. Autotrans­ plantation was performed at body temperature, consisting of an end-to-end arterial anastomosis to the left external iliac artery and end-to-side venous anastomosis to the left external iliac vein. Differing amounts of warm ischemia were used prior to reperfusion (120 minutes, N = 2; 150 minutes, N = 2, 180 minutes, N = 3). The contralateral kidney was left undisturbed. Autotransplanted kidneys were then removed 7 days after the initial procedure. Heterotransplant. Preoperative and intraoperative conditions were similar to those in the autotransplant group. Two pairs of animals were used. The left kidney with ureter in pig A was removed and flushed with 200 to 300 ml. of ice-cold Collins' solution (supplemented with 50 ml. of 50% dextrose and 5000 U heparin per 1.). It was immediately transferred to the left pelvis of pig B, and an anastomosis to the iliac vessels was performed with the kidney wrapped in an ice-filled laparotomy pad. The ureter was secured with an extravesical ureteroneo­ cystostomy, and the kidney was placed in a subcutaneous pocket to facilitate percutaneous biopsies. A Foley catheter remained for 48 hours. After 7 days, the heterotransplant was removed, and a reciprocal transplant (left kidney of pig B into pig A) was performed. No immunosuppression was adminis­ tered in either animal to promote allograft rejection. Special care was taken to standardize intraoperative condi­ tions in both groups. Blood pressure was maintained between 70 to 100 mm.Hg systolic; central venous pressure was kept at 10 cm. H 20 or greater at all times; animals were generously hydrated and a brisk diuresis was noted from the left kidney prior to removal; all kidneys were perfused until efflux was clear; and all kidneys developed good color upon reperfusion, implying adequate arterial inflow. Evaluation of cyclosporine effect. Cyclosporine was adminis­ tered either prior to transplant (N = 3) or at a delayed time point after transplant nephrectomy (mean 11 days) to eliminate the effect of compensatory hyp ertrophy on RBF (N = 5). It was given daily as both an intravenous and oral dose, titrated to a serum trough level >500 ng./ml. (range: 446 to >8000 ng./ ml.). Positron Emission Tomography Scan. A total of 58 scans were obtained (mean 5.8 scans/animal). They were performed under light general anesthesia as described in Surgical Proce­ dures, and were approximately 45 minutes in length. Baseline scans were performed on most animals (N = 8). Animals were scanned at various times during the first 7 days after auto­ transplant or heterotransplant. Scans performed during the period of renal hyp ertrophy were obtained in 6 animals during the first 9 days after transplant nephrectomy. In animals re­ ceiving daily cyclosporine, cyclosporine-toxicity scans were per­ formed only on those days when a toxic level was confirmed by laboratory analysis. Data Acquisition. All scans were performed on a Siemens/ CTI 931/08-12 tomograph. This device produces 15 simulta­ neous image planes encompassing a 10.8 cm. field-of-view. Transmission images using a Ge-68 filled circular ring source were obtained for correction of photon attenuation. The acqui­ sition protocol included twelve 10 second, six 20 second and four 240 second frames. A calibration factor was determined by comparing the activity concentration as measured by PET in a cylinder containing a uniform concentration of Ge-68 with the concentration obtained by counting an aliquot of the cylinder solution in a well counter. Emission scan was performed follow­ ing intravenous bolus injection of 10 mCi N-13 ammonia. To determine input function (adequacy of N-13 ammonia delivery to the animal), blood samples obtained every 10 seconds for 2 minutes from an arterial line were used to calculate the amount of recoverable tracer and metabolites sequentially over a 7minute period.

Reconstruction of cross-sectional regional imData ages employed a Shepp-Logan filter with a cutoff frequency of 0.30 mm.- 1 • Renal cortex activity was determined from the reconstructed images by isolating regions of interest (ROI). These were then used to calculate renal time-activity curves, which were adjusted for the recovery of radiotracer. To deter­ mine regional renal blood flow, these renal time-activity curves were fitted to a two-compartment model, which has been vali­ dated using microsphere RBF measurements.1 5 Patlak graphi­ cal analysis, employed in cardiac N-13 ammonia studies for estimating myocardial blood flow, 16 was employed to obtain flow (expressed as ml./min./gm. kidney). All analyses were performed by one of us (E.N.) who was blinded to the experi­ mental condition. Radioactive Microspheres. To validate dynamic PET imaging with N-13 ammonia as an estimate of RBF, an additional animal underwent radiolabeled microsphere (MS) injection to compare the results of PET with simultaneously acquired mi­ crosphere blood flow measurement. Positron emission tomog­ raphy has been shown to correlate closely with MS determi­ nation of RBF in the dog; MS injection, kidney harvesting and data acquisition were performed as previously described.15 In addition, one high-flow study induced by continuous intrave­ nous infusion of dopamine at 5 µg./min./kg. and one low-flow study induced by continuous intravenous infusion of norepi­ nephrine at 2 µ. per minute were performed to ascertain the adequacy of PET to determine RBF accurately during states of high and low flow. To attain varying degrees of arteriolar occlusion, microspheres of differing diameter were used in each of the three flow states. Histologic A nalysis. Tissue was obtained from transplant kidneys at time of transplant nephrectomy. Interval percuta­ neous biopsies of the heterotransplants were obtained with a 14 G biopsy gun. All tissues were fixed in 4% paraformaldehyde in 0.075 M. sodium phosphate, pH 7.4, processed for routine histology and evaluated for evidence of tubular necrosis or acute rejection by a renal pathologist. Contralateral kidney sections were taken at time of sacrifice (10 days after transplant nephrectomy), preserved and processed in a similar fashion, and examined for evidence of cyclosporine toxicity. RESULTS

N-13 ammonia yielded high quality cross-sectional images of the kidney with PET imaging. These were observed in states of normal RBF (fig. 1, normal kidney) in which uniform distri­ bution of radiotracer is observed in representative slices of the kidney. In contrast, images obtained in the setting of markedly altered RBF (fig. 2, allograft rejection) yield a patchy and diminished distribution of the radiotracer throughout all re­ gions of the kidney. Cranial

Caudal

A00557 U C LA

Fm. 1. Representative PET scan images from normal subject (base­ line).

POSITRO N EMISSION TOMOGRAPHY AND RENAL BLOOD FLOW

1066

FIG. 2. Representative PET scan images from subject with allograft rejection.

Method

REF estimation in the swine Flow (ml./min./gm. kidney)

Author

3.63 PAR/extraction % Gyrd-Hansen, 196814 2.87 Helin, et al., 197517 Dye dilution 3.69 Friis, 197918 PAR/extraction % DTPA/Hippuran clearance 3.77 Robbins, 19851 9 2.53 Radiolabelled microspheres Poulsen, et al., 198820 N-13 ammonia/PET 3.18 Killion, et al., 1993 Values are corrected for a) renal extraction of PAH in normal pig kidney and b) kidney weight in relation to body weight.14

Baseline RBF was determined in 8 animals prior to any surgical intervention, yielding a value of 3.16 ml./min./gm. kidney (standard deviation = 1.1). This value closely parallels that reported for swine in the literature, using other well­ established methods,14• -20 as reported in the table. Moreover, when RBF was determined simultaneously using PET and the established microsphere technique, PET was found to correlate closely with MS determination, differing by less than 8% during periods of normal (PET, 3.16; MS, 3.40), high (dopamine: PET, 4.76; MS, 5.19) and low (norepinephrine: PET, 1.72; MS, 1.95) renal flow. Figure 3 illustrates calculated RBF in four different experi­ mental settings. Figure 3, A reveals RBF over time in autotrans­ planted kidneys exposed to varying amounts of ischemia. Com­ paring the three groups, there was a progressive decrease in RBF through post-transplant day 3, which was greater for those animals exposed to longer ischemic insult. Animals exposed to 120 and 150 minutes of ischemia showed a return to normal RBF by post-transplant day 7 (albeit a slower return in the 150 minute ischemia animals), whereas those animals receiving 180 minutes of ischemia demonstrated both the greatest fall in RBF and the least improvement in RBF by post-transplant day 7. Figure 3, B demonstrates RBF over time for animals undergoing heterotransplant and demonstrates a marked decrease in RBF immediately after transplant that persisted through post-trans­ plant day 7 (despite an apparent rise on post-transplant day 2, probably prior to the onset of rejection). Figure 3, C illustrates the effect of unilateral nephrectomy on RBF in the remaining kidney. As the period since nephrectomy increased, there was an increase in RBF in the remaining kidney, implying compen­ satory hypertrophy (R = 0.79). By 9 days after nephrectomy, RBF reached a level of 6.1 ml./min./gm., which is nearly double the baseline value (3.16 ml./min./gm.). This value of 6.1 ml./ min./gm. persisted until post-nephrectomy day 27, as measured in one animal. Figure 3, D demonstrates the effect of cyclo­ sporine level on RBF. As the level of cyclosporine increased above the therapeutic range, there was a linear, progressive decrease in RBF (R = 0.68). 17

All kidneys were perfused at the time of transplant nephrec­ tomy, and no technical complications were encountered. No animals showed evidence of deleterious effects from the PET scanning. Two complications were encountered. One animal expired on post-transplant day 1 from an anesthesia-related pulmonary arrest, and post-operative data from that animal are omitted. Another animal died 14 days after a successful 180 minute autotransplant from bowel ischemia. Histologic evaluation of tissue slices from autotransplanted kidneys removed 7 days after ischemic insult demonstrated evidence of resolving acute tubular necrosis (ATN), which was slightly increased for those kidneys exposed to greater degrees of ischemia. These changes included peritubular mononuclear infiltrate, mild tubular degeneration and vascular congestion. Kidneys subject to high levels of cyclosporine showed a pattern consistent with cyclosporine toxicity, which included perivas­ cular lymphocytic infiltrate, capillary engorgement and large vessel intimal edema and luminal narrowing. Histology from cross-transplant kidneys demonstrated profound graft rejec­ tion, with diffuse lymphocytic infiltrate, interstitial edema, vascular thrombosis and ischemic necrosis. DISCUSSION

Positron emission tomography technology is currently find­ ing applications in many areas of medicine, and although it has become a mainstay diagnostic modality in several disease processes, its potential use in the evaluation of renal disorders has not been explored. The flow-dependent extraction of an available radiotracer, such as N-13 ammonia, coupled with its short half-life and proven safety in humans, makes for an ideal method for the estimation of renal perfusion. The availability of a noninvasive means to accurately quantitate renal perfusion has never been attained, as most methods are either time­ consuming and invasive (such as, para-aminohippuric acid ( [PAH]), technically complex and difficult to interpret ( 86 Rb, albumin and erythrocyte accumulation),2 or require postmor­ tem analysis of tissue radioactivity (radiolabeled micro­ spheres).20• 21 Washout of inert gases, such as Xe-133, has been used to measure RBF; however, this technique requires an intraarterial bolus of the agent and is subject to numerous limitations.2 2 The advantages of PET over other nuclear medicine tech­ nologies in the evaluation of renal perfusion is apparent. 99Tc­ DTPA and 99Tc-MAG3 have been developed to measure renal perfusion, yet they primarily estimate glomerular filtration rate. Their images offer excellent information regarding col­ lecting system anatomy, but only a visual assessment of relative index of renal perfusion can be made.23 Hippuran is eliminated from plasma in a single pass through the kidney and has been used to measure effective renal plasma flow. Its use, however, is limited in patients with asymmetric renal uptake or poorly functioning kidneys because its slow clearance and relatively high liver activity yield poor image quality.24 Compared with these modalities, PET duration is short (45 minutes as opposed to 3 hours), produces functional, not merely anatomical, images of the kidney, and N-13 ammonia possesses excellent imaging quality. Specific regions within a kidney can be evaluated for a more detailed estimate of cortical function, and areas of im­ paired activity can be identified. The necessity of arterial blood sampling for calculation of input function in PET imaging is awkward, yet alternative methods are now available which will eliminate this component by utilizing dynamic PET measure­ ments of abdominal aortic activity to estimate radiotracer and metabolite activity.25 Positron emission tomography determination of RBF in nor­ mal swine kidney correlated well with both microsphere meas­ urement and values obtained by other investigators using a variety of techniques.14• 17-2° Four clinical situations known to affect RBF were therefore investigated to determine the ability of this method to detect changes in RBF. Ischemic damage,

1067

P O SITRON EMI S S I O N T O M OGRAPHY AND RENAL BLOOD FLO "\V 5

4

E .!:

� E

u. OJ a:

3

� .E C:

u. OJ a:

--®- 1 2 0 " i s c h , N•2 -I0-- 1 5 0 " isch, N•2 -+- 18 0 " isch, N•3

A

2

'E

2

0

3

E

0

Immediately after transplant

0

2

3

4

6

5

8

7

0

B

.

1

.E

'E LL

OJ

a:

C

Iii

4

5

6

7

e

Ill!

Iii

5

E

-'2? C:

4

E

4 ill

3

"

E

3

u. ID a:

2

0

3

5

6

E �

2

D a y s a f t e r t r a n s p l a nt

Days after transplant 7

Immediately after t r ansplant

ft

ii!

2

l!l li!l Iii

ill Ii Ii

o

3

4

6

7

8

D a y s after t r a n s p l a nt

0 9

10

11

12

D

0

1000 2000 3 0 0 0 4000 5 0 0 0 6000 7000 8 0 0 0 9000

Cyc losporine level

FIG. 3. A, mean RBF in post-transplant period for animals exposed to varying amounts of ischemia (120 minutes, N = 2; 150 minutes, N = 2; 180 minutes, N = 3). B, mean RBF after heterotransplant (N = 2). C, mean RBF after unilateral nephrectomy (N = 9 scans, x = 6 animals) . R = 0.79. D , RBF at various cyclosporine levels (N = 1 5 scans; x = 8 animals). R = 0.68.

either from prolonged hyp otension or in the transplant setting of excessive warm exposure or hypothermic preservation, causes a progressive deterioration in renal circulation and mi­ crocirculation as a result of tissue and endothelial swelling. 3 Positron emission tomography demonstrated that exposure to increasing amounts of warm ischemia produced a measureable difference in RBF over time, but that RBF returned to normal in 7 days after injury in animals exposed to 120 or 150 minutes of ischemia. Renal blood flow in animals exposed to 180 minutes of renal ischemia failed to normalize after 7 days, suggesting that this degree of ischemic insult may be incompatible with eventual renal recovery. These findings suggest that PET may be a useful adjunct in providing an objective assessment of allograft viability during prolonged graft nonfunction, or like­ wise, in evaluating the potential for renal recovery after a variety of nephrotoxic insults. Acute rejection of a renal allograft is evident pathologically as direct renal vascular damage. 5 • In this study, animals who underwent cross-transplant without the aid of immunosuppres­ sion developed profound graft rejection, as predicted and evi­ denced by the histologic results. Renal blood flow remained near normal by post-transplant day 2, but then rapidly fell by transplant day 5 as rejection ensued, without any further im­ provement. Since vascular changes are evidenced early in the course of allograft rejection, often preceding any measureable alteration in renal function, it is possible that PET technology may be applied in the diagnosis of early rejection through detection of changes in renal perfusion prior to clinical evidence 6

of azotemia. In other words, sequential PET scanning in the post-transplant period may offer additional, more subtle infor­ mation regarding dynamic allograft function than is available with serum creatinine determination alone. Positron emission tomography readily demonstrated the physiologic processes involved in compensatory renal hyp ertro­ phy after unilateral nephrectomy, wherein glomeruli and blood vessels increase not in number, but in size, leading to increased RBF. 1 1 • 1 2 In this study, a near-doubling in RBF by post-ne­ phrectomy day 9 was observed and persisted through post­ nephrectomy day 27. The close correlation between time after nephrectomy and increase in RBF suggests the potential use­ fulness of PET in both the experimental laboratory and the clinical environment as a tool in the evaluation of the regen­ eration potential of a kidney after contralateral or direct renal insult. Furthermore, PET was useful in discriminating differential renal toxicity after cyclosporine administration. Despite its profound benefit as an immunosuppressive agent, cyclosporine has been recognized as inherently nephrotoxic when adminis­ tered at supratherapeutic levels. 7-9 , 2 This effect is a vascular­ mediated phenomenon that causes a fall in RBF. With acute cyclosporine toxicity, there is preglomerular vasoconstriction, whereas with chronic toxicity, an arteriolopathy results leading to a further decrease in RBF and eventual tubulopathy and graft failure. 27 To evaluate this vascular effect, subjects received cyclosporine in toxic amounts to determine whether PET could detect a dose-related change in renal perfusion. An inverse 6

1068

POSITRO N EMISSION TOMOGRAPHY AND RENAL BLOOD FLOW

relationship was noted between cyclosporine level and RBF, indicative of the vasculogenic activity of cyclosporine when present at toxic levels. In contrast to well established, invasive and time-consuming methods for the determination of RBF, the use of PET is safe, efficient and capable of yielding an accurate measure of RBF in normal, pathologic and supraphysiologic conditions. How­ ever, the sensitivity of this particular application of PET as a single diagnostic tool in differentiating various renal patholo­ gies is questionable. The three nephrotoxic conditions all caused a decrease in RBF from baseline, yet the magnitude of this fall was fairly similar for these three conditions. Therefore, this particular use of PET may represent merely the first step in defining the optimal application of this new technology to the study of renal disorders. In particular, PET determination of RBF may prove to be a useful adjunct in the evaluation of true hemodynamic renal flow in the patient with renal artery stenosis, and help to determine who would most benefit from intervention. Additionally, the use of radiotracers targetted to specific renal receptors (such as, catecholamines, angiotensin converting enzyme inhibitors) may aid in improving under­ standing of renal hypertension and its treatment. Furthermore, the development and use of radiotracers whose extraction is dependent not on flow, but rather, on the specific metabolism of a target organ, will provide a better assessment of the true functional state of the kidney than do conventional means that are currently available. Ongoing work has verified the safety and accuracy of this technique in human subjects, and further work will be directed at broadening the capability of PET as a new diagnostic tool in the study of renal disease. REFERENCES 1. Schelbert, H., Phelps, M., Hoffman, E., Huang, S., Selin, C. and Kuhl, D.: Regional myocardial perfusion assessed with N-13 labelled ammonia and positron emission computerized axial to­ mography. Am. J. Cardiol., 43: 209, 1979. 2. Aukland, K.: Methods for measuring renal blood flow: total flow and regional distribution. Ann. Rev. Physiol., 42: 543, 1980. 3. Anaise, D., Bachvaroff, R., Sato, K., Walzer, W., Asari, H., Pollack, W., Oster, Z., Atkins, H. and Rapaport, F.: Enhanced resistance to the effects of hypothermic ischemia in the preserved canine kidney. Transplantation, 3 8 : 570, 1984. 4. Shabtai, M., Anaise, D., Shabtai, E., Raisbeck, A., Malinowski, K., Oster, z., Atkins, H., Walzer, W. and Rapaport, F.: Renal blood flow in the immediate post-transplant period as an index of the efficacy of organ procurement and preservation. Transplant. Proc., 2 2 : 2366, 1990. 5. Porter, K.: Pathologic changes in transplanted kidneys. In: Expe­ rience in Renal Transplantation. Edited by T. E. Starzl. Phila­ delphia: Saunders, 1964 6. Tourville, D., Kim, D., Viscuso, R., Jacobs, M., Schoen, S. and Filipone, D.: Anticipation of renal transplant failure by postan­ astamosis biopsy and immunofluorescence. Transplant. Proc., 9: 91, 1977. 7. Myers, B.: Cyclosporine nephrotoxicity. Kidney Int., 30: 964, 1986.

8. Fry, W., Dawidson, I., Alway, C. and Rooth, P.: Cyclosporine induces decreased blood flow in cadaveric kidney transplants. Transplant. Proc., 20, suppl. 3: 222, 1988. 9. Youngleman, D., Kahng, K., Dresner, L., Munshi, I. and Wait, R.: Cyclosporine-induced alterations in renal, intrarenal, and pan­ creatic blood flow. Transplant. Proc., 23: 718, 1991. 10. Halliburton, I. and Thomson, R.: Chemical aspects of compensa­ tory renal hypertrophy. Cancer Res., 25: 1882, 1965. 11. Platt, R.: Structural and functional adaptation in renal failure. Br. Med. J., 1: 1313, 1952. 12. Preuss, H. and Goldin, H.: Humoral regulation of compensatory renal growth. Med. Clin. North Am., 59: 771, 1975. 13. Sack, W.: Essentials of pig anatomy. Ithaca: Veterinary Textbooks, 1982. 14. Gyrd-Hansen, N.: Renal clearance in pigs. Acta Vet. Scand., 9 : 183, 1968. 15. Chen, B., Germano, G., Huang, S., Hawkins, R., Hansen, H., Robert, M., Buxton, D., Schelbert, H., Kurtz, I. and Phelps, M.: A new noninvasive quantification of renal blood flow with N-13 ammonia, dynamic PET and a two-compartment model. J. Am. Soc. Nephrol., 3: 1295, 1992. 16. Choi, Y., Huang, S., Hawkins, R., Kuhle, W., Dahlbom, M., Hoh, C., Czernin, J., Phelps, M. and Schelbert, H.: A simplified method for the quantification of myocardial blood flow using N13 ammonia and dynamic PET. J. Nucl. Med., 34: 488, 1993. 17. Helin, I., Okmian, L. and Olin, T.: RBF and function during neurolept-anaesthesia and elevated intravesical pressure. An experimental study in the pig. Scand. J. Urol. Neph., suppl., 2 8 : 3 1 , 1975. 18. Friis, C.: Postnatal development of renal function in pigs: GFR, clearance of PAH and PAH extraction. Biol. Neonate, 35: 180, 1979. 19. Robbins, M.: Single injection techniques in determining age-related changes in porcine renal function. Int. J. Appl. Radiat. Isot., 3 5 : 85, 1984. 20. Poulsen, E., Jorgensen, J., Madsen, F. and Djurhuus, J.: Cortical blood flow in the porcine kidney: a radioactive microsphere study. Urol. Res., 1 1 6: 385, 1988. 21. Slotkoff, M., Logan, A., Jose, P., D'Avella, J. and Eisner, G.: Microsphere measurement of intrarenal circulation in dogs. Circ. Res., 28: 158, 1971. 22. Moore, C. and Gewertz, B.: Measurement of renal blood flow. J. Surg. Res., 32: 85, 1982. 23. Hilson, A., Maisey, N., Brown, C., Ogg, C. and Bewick, M.: Dynamic renal transplant imaging with Tc-99m DTPA supplemented by a transplant perfusion index in the management of renal trans­ plant. J. Nucl. Med., 1 9: 994, 1978. 24. Dubovsky, E. and Russell, C.: Quantitation of renal function with glomerular and tubular agents. Sem. Nucl. Med., 12: 308, 1982. 25. Germano, G., Chen, B., Huang, S., Gambhir, S., Hoffman, E. and Phelps, M.: Use of the abdominal aorta for arterial input function determination in hepatic and renal PET studies. J. Nucl. Med., 33: 613, 1992. 26. Abraham, J., Bentley, F., Garrison, R. and Cryer, H.: The influence of the cyclosporine vehicle, cremophor EL, on renal microvas­ cular blood flow in the rat. Transplantation, 52 : 101, 1991. 27. Youngleman, D., Kahng, K., Rosen, B., Dresner, L. and Wait, R.: Effects of chronic cyclosporine administration on renal blood flow and intrarenal blood flow distribution. Transplantation, 5 1 : 503, 1991.