Innocuity of agents for radiocontrast enhancement of urinary tract stones

1NNOCWl-Y OF AGENTS FOR RADIOCONTRAST ENHANCEMENT OF URINARY TRACT STONES* JACQUES

CORCOS, M.D., CHANTAL SAVOIE, M.Sc., PROMETE MADARNAS, LYNE PICARD, M.D., AND EMANUEL ESCHER, PH.D.

M.D.,

ABSTRACT-Objectives. Mineral kidney stones are frequently difficult to detect due to their radiotranslucency. We have recently developed a method that enhances the visibility of such stones by retrograde infusions of certain heavy metal salt solutions such as cesium or lanthanide gadolinium. This report describes toxicologic studies carried out on the use of those contrast agents to introduce this technique eventually into clinical trials. Methods. Systemic absorption was assessedin dogs through infusion of radioactive contrast agent into the renal pelvis with or without ureteral obstruction. Radioactivity in urine and blood was monitored. Local toxicity was studied in animals infused with the contrast agent at intervals up to 4 weeks. Results. Reabsorption studies under high intrapelvic pressures (70 cm H,O or higher), demonstrated reabsorption of cesium. However, at normal intrapelvic pressures, only a moderate reabsorption of cesium was observed. No gadolinium reabsorption was detected even at high intrapelvic pressures. Histopathologic studies showed no major urothelial lesions but only a transient inflammatory reaction that was undetectable 4 weeks following exposure to gadolinium salts. Conclusions. Gadolinium salt solutions are good positive radiocontrast agents for mineral kidney stones without having serious toxic effects or systemic reabsorption. UROLGCV 46: 643-647, 1995.

F

luoroscopy and ultrasonography are the most frequently used methods to visualize urinary tract (UT) stones for all therapeutic approaches. For extracorporeal shock-wave lithotripsy (ESWL), ureteroscopy, percutaneous surgery, and chemical dissolution, adequate radiologic visualization of stones and stone fragments is mandatory. Poor visualization can result in incomplete extraction after fragmentation, a longer duration of surgery, added morbidity, and complications. l-3 We have proposed a simple, effective, and inexpensive method to enhance the radiopacity of UT stones4 We used incubation of struvite fragments in vitro and in vivo animal models to demonstrate that lanthanides (gadolinium, ytterbium, europium) and cesium significantly enhance the radiopacity of UT stones. The direct contact of the metal salt solution with the stone resulted in a specific incorporation of metal ion, enhancing stone radiopacity. The present study was under*Supported by grants from the Kidney Foundation of Canada, the Medical Research Council of Canada, and the Centre de Recherches Cliniques du CHLJS. From the Department of Urology, McGill University, Montreal, and Departments of Pathology and Pharmacology, University of Sherbrooke, Faculty ofMedicine, Sherbrooke, Quebec, Canada Reprint requests: Jacques Corcos, M.D., Jewish General Hospital, 3755, Cbte Sainte-Catherine, E-211, Montreal, Quebec, Canada H3T 1E2 Submitted: July 22, 1994, accepted (with revisions): May 16, 1995

taken to assessthe biologic risks to the UT associated with such treatments. In a first series of experiments, we investigated the degree of systemic absorption of the contrast agents through the lining of the upper UT. Two putative situations were explored: first, infusion was performed at normal intrapelvic hydrostatic pressure; second, infusion was done under increased intrapelvic pressure, simulating a partial or total obstruction of the urinary outflow. Two contrast agents were tested-cesium and gadolinium-because of their proven properties as contrast agents4 and because of their well-documented systemic toxicology. In a second series of experiments, we evaluated the toxic effects to the urothelium following an infusion of gadolinium salt solutions over time. This part of the study was limited to gadolinium because cesium was found to be reabsorbed, to a considerable degree, under conditions corresponding to obstruction of the UT. MATERIAL

AND METHODS

ABSORPTION STUDIES Experiments to measure the retrograde absorption of heavy metal salts during and after intrapelvic infusions were carried out on female, nonestrous, and nongravid mongrel dogs (weighing 23 to 35 kg). The animals were fasted for 24 hours with water ad libitum prior to surgery, sedated with xylazine (1.1 mg/kg) and an-opine (0.05 mg/kg) subcutaneously, and anesthetized with pentobarbital (15 mg/kg) delivered via the cephalic vein. After intubation, the animals were mechanically ventilated and were kept normovolemic by continuous

643

infusion of 0.9% sodium chloride (NaCl; 250 mm). A sternum-to-pubis laparotomy was performed, the ureters were located, mobilized, and sectioned 10 cm below the kidneys. A dual lumen catheter was inserted into the ureters in a retrograde fashion until the catheter tips were positioned within the renal pelvis. The catheters were secured in place by ureteral ligation and sutured to the peritoneal wall. The abdominal wound was closed with skin clips. Urine outflow was collected below the operating table from the external catheter, which was branched to a 120 cm vertical tube that was open at its end. The side part was placed at the level of the kidney to assure zero hydrostatic pressure within the renal pelvis. A butterfly catheter was placed into the cephalic vein for blood sampling. The volume and radioactivity of the urinary outflow was recorded independently for each kidney. Blood samples (3 mL) were drawn at 15-minute intervals and their radioactivity was measured in a gamma counter with window settings between 500 and 1000 keV for 13’Cs and fully open windows for ‘53Gd. A slow infusion (0.94 mL/min) of radioactively labeled contrast solution into the left renal pelvis was continued for 30 minutes. The cesium-containing solution comprised 25 mL of 5% (w/v) aqueous cesium chloride (CsCl) containing 100 t.Ki of 13’Cs (Amersham) and the gadolinium solution was 5% (w/v) aqueous gadolinium nitrate [Gd(NO,),l containing 100 pCi of 153Gd (Amersham). Blood and urine sampling were continued for 4 hours after the termination of the infusion until at least 98% of the initial radioactivity had been excreted. Anesthesia was maintained throughout the experiment by intermittent 15 mg/kg doses of pentobarbital as required (ocular reflex). At the end of the experiment, the animal was sacrificed by a rapid intravenous infusion of pentobarbital (200 mg/kg). Experiments simulating UT obstruction were carried out in an identical manner except that a smaller volume of contrast agent (0.5 mL) containing 100 pCi of tracer in 5% (w/v> CsCl or Gd(NO,), was injected as a bolus into the renal pelvis while urine flow was blocked immediately afterward by a hemostat below the vertical, open-ended tube. Intrapelvic pressure was measured by the height of the urine column in the vertical tube. Sampling of urine from the contralateral kidney and blood was carried out continuously as already described. Intrapelvic pressure in the left kidney increased spontaneously to 70 to 80 cm/urine pressure, where it stabilized and remained at this pressure for 30 minutes. If, however, a pressure of 120 cm/urine was reached, as in 1 case, an internal expansion device was used to stabilize the pressure at this level. After this delay, the obstruction was removed and the left urinary tract was emptied, the urine volume measured, and its radioactivity assessed. Independent continuous sampling was obtained from both kidneys until the end of the experiment.

LOCALTOXICOLOGY The local toxicity of the gadolinium salt solutions to the UT was studied in acute and chronic animal models, with 3 animals in each group. For the acute toxicity studies, anesthesia, operative access, and ureteral intubation were carried out as already described. The infusion of 5% (w/v) of Gd(NO,), into the renal pelvis was performed with a Harvard syringe pump at the rate of 0.94 ml/min over a l-hour period. The urinary output was monitored continuously and the experiment was stopped immediately if the urine output fell below input volume. Once the perfusion was completed, the renal pelvis was flushed with 0.9% NaCl at a rate of 0.94 mumin for 15 minutes. Subsequently, the animals were sacrificed as already described. The kidneys and the proximal ureters were removed and samples were collected from calices and pelvis. Those specimens had not been in direct con-

644

tact with the catheters and were therefore free from mechanical injury. Tissue samples were fixed in 10% buffered form01 solution for 2 days, subsequently dehydrated with alcohol and toluene. The tissues were then embedded in paraffin and prepared as 3 to 5 pm sections, and stained with hematoxylin and eosin for light microscopy. For the chronic toxicity studies, anesthesia and surgical access were carried out as already described. A stab wound was made in the middle of the ureters, allowing placement of a dual lumen catheter into the renal pelvis. Only the nonlabeled gadolinium solution was used in these experiments; the infusion rate, time of infusion, monitoring of urine output, and washing of the pelvis were the same as for the reuptake protocol. After the infusion, the catheters were removed and the ureteral wound was carefully closed with 6-O Vicryl stiches. The abdominal wall was closed with a running suture of 3-O Vicryl and the skin closed. To prevent infectious complications, the dogs received a topic antibiotic (furazoladone [Topazone]) and intramuscular long-acting penicillin (Pendistrep) 0.114 mLIkg the first day, 0.228 mL/kg the second day, then 0.228 mL/kg every 2 days for a week. The animals received a normal diet beginning the day following the surgery and were sacrificed at 1 to 4 weeks postoperatively. Tissue sampling and fixation were carried out as already described. Initially, in some animals hydronephrosis and subsquent obstruction developed due to ureteral scarring; they were excluded from the study. Every experimental group consisted finally of at least 3 admissible animals.

RESULTS ABSORPTIONUNDERNORMALINTRAPELVIC PRESSURE

153Gd-related radioactivity was not present in the blood samples following infusion of gadolinium salt solutions into the renal pelvis of 3 dogs under normal intrapelvic pressure (Fig. 1A). The urine from the contralateral kidney was also free of radioactivity. The specific radioactivity of the infused gadolinium solution was of 36.2 lKi/mmol and therefore trace amounts below 1.5 PM concentrations of the contrast agent would not be detected. Clinically, levels below 10 ~.LM are known to be nontoxic; therefore, this minimal amount is of little significance.5 In contrast, reabsorption of cesium solutions occurred to a significant extent (close to 1% of totally presented cesium) even in the absence of an abnormally elevated intrapelvic pressure in an equivalent group of 3 dogs (Fig. 1B). Shortly after the appearence of 137Cs in the blood, the tracer was also found in the urine from the contralateral kidney, following apparent first order kinetics and eliminating 0.11% of the totally injected cesium. ABSORPTIONUNDERELEVATEDINTRAPELVICPRESSURE

The possibility of an accidental obstruction of the urinary flow and consequent complications has to be considered when performing retrograde kidney pelvis perfusion for diagnostic purposes. Therefore, a fixed amount of radioactively labeled contrast agent, either gadolinium or cesium (three independent experiments for each element) was injected as bolus into the pelvis and the urinary UROLOGY@ 46 (5), 1995

0

60

A

M

Right

(F----O

Right

a--O

Left kidney

R-----U

Left kidney

u

Blood

n----n

Blood

160 120 Time (minutes)

kidney

240

300

0

60

120

180

kidney

240

75

75

390

3,O 503

(D

fl 2,0 x if

0

3100

Time (minutes)

B

w

.g

Pressure

I,0

;

n

0

60

120

180

240

zc

25

5 P

D

190

0

60

120

160

240

.E 3

Pressure

x

2 0

300

Time (minutes)

50

z 2,0 u

252

C

(D

300

Time (minutes)

FIGURE 1. Elimination of heavy metals: (A and B) without urinary obstruction, (C and D) with urinary obstruction (pressure around 70 cm H,O for 30 minutes). (A) Infusion of 5% gadolinium solution into the right kidney. There is no trace of 153Gd-related radioactivity in blood and urine samples from left kidney. (B) Infusion of 5% cesium chloride with 137Cs into the right kidney. Significant reabsorption of labeled Cs visible in blood. (C) Bolus injection of 5% Cd solution into the right kidney with obstruction of right ureter. No reabsorption is observed. (D] Bolus injection of 5% Cs solution into the right kidney with obstruction of right ureter. Major reabsorption is noticed in blood as well as in urine output of the left kidney [around 4000 cpm/mL not visible in the graphic).

flow was interrupted. The intrapelvic volume and pressure increased over a period of 15 minutes to reach a stable hydrostatic pressure. The experiment with gadolinium showed absence of serum radioactivity or urine radioactivity from the contralateral kidney (Fig. 1C) above the detection limit of 1.5 p,M. In contrast, the experiments with cesium under elevated intrapelvic pressure showed a highly significant reabsorption of this heavy alkali ion (7% of the totally injected amount) (Fig. 1D). The comparison of cesium concentration in urine and blood from the occluded kidney indicates an efficient retrograde absorption of cesium. Cesium excretion through the contralateral kidney follows again first order kinetics, similar to the normal physiologic situation.6 After release of the urinary obstruction of the left kidney and initial voiding, 137Cs elimination from this kidney also follows first order kinetics. Homologues of potassium, cesium, and rubidium are cell membrane permeable, and this is well demonstrated by the cesium uptake even at low intrapelvic pressures. The experience with increased intrapelvic pressure enhances this reabsorption approximately lo-fold in a relatively UROLOGY~ 46 (5), 1995

steady manner, suggestive of increased reabsorption and not of fornical rupture. The latter can be excluded by the parallel experiences with gadolinium, for even at high intrapelvic pressures no measurable reabsorption could be detected. Fornical rupture would have led to an appearance of gadolinium in the vascular space. Tissue permeability of gadolinium is quite low. It behaves as a biologic homologue of calcium, and with much slower membrane transport kinetics. Under normal infusion conditions as well as with elevated intrapelvic pressure, gadolinium is not reabsorbed systemically. Cesium as a homologue of potassium has pronounced cardiotoxic effects.7,8 Since our results show that cesium is reabsorbed to a considerable amount even at normal intrapelvic pressure, this cardiotoxicity potential was judged too high for further consideration of cesium as a contrast agent. Thus the local toxicity study of potential contrast agents was therefore carried out with gadolinium only LOCAL TOXICITYSTUDYANDPATHOLOGY

Care was taken in this part of the study to investigate only UT tissues from the calices and 645

FIGURE 2. Histologic evaluation of urothelium exposed to radiocontrast agent. (A) Normal urothelium from renal pelvis of control animal, occasional lymphocytes are observable in the submucosa. (B] Urothelium from renal pelvis 3 hours after exposure to gadolinium solution. There is loss of the most superficial epithelial layer with fragmented cytoplasmic debris of disintegrated cells. There are also some cells with cytoplasmic vacuolation indicating mild cellular damage of the upper layers. The subepithelial fibrovascular tissue presented some edema and occasional granulocytes, which are indicative of a mild acute inflammation. [Clear areas in subepithelial tissue are sectioning artefacts.) (Original magnifications x 400.)

from the pelvis that came into direct contact with the contrast solution only We avoided evaluating all tissue with possible mechanical alterations as a consequence of catheter injury so that the changes observed would reflect the effect of the contrast solution on the surface epithelium and the underlying fibrovascular tissue. Figure 2A shows the control tissues and Figure 2B the acute effect to gadolinium solution 3 hours after exposure, showing only a mild inflammatory reaction. Chronic changes are illustrated in Figure 3, at 1 week (A), 2 weeks (B), and 4 weeks (0. The results show a progressing healing process with some residual inflammation present after 1 week, a complete recovery after 2 weeks, and a rebound effect with some epithelial hypoplasma after a few weeks, The tissue damage induced by gadolinium salt infusions is very minor compared with lesions induced by other frequently used infusions, such as 646

FIGURE 3. Long-term evolution after contrast agent exposure. (A) Urothelium from the renal pelvis 1 week after exposure to gadolinium solution. Superficial flat cells are missing. The urothelium has hyperchromatic nuclei and rare mitoses revealing regenerating features. (B) Renal pelvis epithelium 2 weeks after exposure to gadolinium solution. Normal number of cell layers and flat surface cells indicate complete recovery. [C) Urothelium of renal pelvis 4 weeks after exposure to gadolinium solution. Number of cell layers is increased although cell features are normal with normal maturation. (Original magnifications x400.1

Renacidin. This stone dissolution procedure produces focal necrosis and considerable inflammation extending into the pelvic well and even into the renal medulla.9J0 COMMENT The clinical utilization of radiopacity enhancers for localization of renal calculi as a clinical tool for lithotripsy has to be closely scrutinized to evaluate even minor potential risks. On the other hand, careful analysis of the risks versus the potential benefits indicates that a successful radiopacity enhancement technique is probably more than a mere accessory to stone extraction. Accurate radiologic localization of poorly visible stones should improve all lithotripsy or stone extraction techniques because it will permit reduction of operating time, yield smaller stone fragments, UROLOC?I”

46 (51, 1995

help achieve a stone-free kidney, and thus postoperative complications.11 radiopacity enhancers are particularly valuable for ESWL because it will permit precise targeting of the shock-wave focus. The enhancers4 incorporate into the stone surface or their fragments; therefore, in many ESWL cases, preoperative and postoperative opacification has to be foreseen. The emergence of medical magnetic resonance imaging (MRI) instrumentation into general use will also potentially profit from this particular contrast technique, since the most versatile radiopacity agent is gadolinium, which, as an MRIshift reagent, should make even the smallest residual stones visible by MRI-shift enhancement. Our toxicology study did not address the repetitive use of enhancers in a short period or in combination with ESWL. This point remains to be studied before actual clinical studies with ESWL can be undertaken. Accidental retroperitoreal or even intravascular infusion was not specifically addressed in this study However, the systemic toxicity in man is quite well documented in the literature, showing that Gd-containing preparations are widely used as MRI contrast agents with a high safety profile.3 This aspect of accidental dislodgement of the infusion catheters will have to be addressed in future animal work prior to clinical studies. CONCLUSIONS Gadolinium salt solutions appear to be well tolerated by the lining of the UT and clinical studies with this radiopacity enhancer could be undertaken after the worst-case scenario of direct retroperitoneal infusion has been addressed in animal models. ACKNOWLEDGMENT. To Jacques Rousseau for the nuclear imaging procedures. This work is in partial requirement for the degree of M.Sc. to L.P. and to C.S.

REFERENCES 1. Davidson AJ: Intraluminal Abnormalities. Toronto, WB Saunders, 1985, pp 424-436. 2. Kellum CD, Tegtmeyer CJ,Jenkins AD, BarrJD, Gillenwater JY, Wyker AW, and Lippert MC: The role of radiology in extracorporeal shock wave therapy. Radiology 165: 431-438. 1987.

UROLCGY’

46 (51, 1995

3. Morris TW, and Fischer HW: The pharmacology of intravascular radiocontrast media. Annu Rev Pharmacol Toxicol 26: 143-160, 1986. 4. Corcos J, Picard L, Groleau S, Madarnas P, and Escher E: Radio-contrast enhancement of urinary tract stones. J Urol 145: 618-623, 1991. 5. Cupples WA: Renal medullary blood flow: its measurement and physiology. Can J Physiol Pharmacol 64: 873-880, 1986. 6. Lortie M, Regoh D, Rhaleb NE, and Plante GE: The role of Bt- and B,-kinin receptors in the renal tubular and hemodynamic response to bradykinin. Am J Physiol 262: R72-R76, 1992. 7. Johnson GT, Lewis TR, and Wagner WD: Acute toxicity of cesium and rubidium compounds. Toxic01 Appl Pharmacol32: 239-245, 1975. 8. Barrera H, and Gomez-Puyou A: Characteristics of the movement of K+ across the mitochondrial membrane and the inhibitory action of Tl+. J Biol Chem 250: 5370-5374, 1975. 9. Auerbach S, Mainwaring R, and Schwartz F: Renal and ureteral damage following clinical use of Renacidin. JAMA 183: 61-63, 1963. 10. Kohler FP: Renacidin and tissue reaction. J Urol 87: 102-105, 1962. 11. Chaussy CG, and Fuchs GJ: Current state and future developments of non invasive treatment of human urinary stones with extracorporeal shock wave lithotripsy. J Urol 141: 782-789, 1989. EDITORIAL COMMENT The success rate of endourologic techniques for stone removal are dependent to a large degree on accurate fluoroscopic localization of the stone for shock-wave lithotripsy or endoscopic removal. The authors have previously demonstrated in vitro the ability to increase the radiopacity of renal calculi by bathing the stones in a heavy metal salt solution (reference 4). This article is a toxicity report on two such heavy metal compounds, namely, cesium and gadolinium. In this model, cesium demonstrated significant systemic reabsorption, however, no gadolinium reabsorption was detected even at high intrapelvic pressures and effects on the urothelium were minimal. This is an intriguing concept with the obvious clinical implication being an enhancement of our existing techniques of stone therapy. As the authors have indicated, prior to this technique being introduced into the clinical realm, further toxicity studies will be necessary to examine the possible local and systemic effects that may occur when heavy metal compounds are utilized during shock-wave lithotripsy or endoscopic surgery where the possibility of direct vascular or retroperitoneal infusion may occur. John Denstedt, M.D. St. Joseph’s Health Centre London, Ontario, Canada N6A 4V2

647