- Email: [email protected]
Hepatic and renal blood flow responses to a clinical dose of intravenous cyclosporine in the pig
Immnophaimacology ELSEVIER
Immunopharmacology 28 (1994) 87-94
Hepatic and renal blood flow responses to a clinical dose of intravenous cyclosporine in the pig Kenneth A. Andreoni*, Christopher P. O'Donnell, James F. Burdick, James L. Robotham Departments of Surgery and Anesthesiology and Critical Care Medicine, Pulmonary Anesthesiology Laboratory, School of Medicine, Johns Hopkins University, Baltimore, MD 21215, USA (Revision received 7 April 1994; accepted 8 April 1994)
Abstract
The immunosuppressant Cyclosporine A (CsA) is considered to induce nephrotoxicity in part by causing vasoconstriction of the glomerular afferent arterioles. Although CsA is widely used in hepatic transplantation, little is known concerning its effects on hepatic blood flow. We used ultrasonic flow probes in an anesthetized swine model to measure the effects of a single 60 min infusion of a clinically comparable dose of CsA (5 mg/kg per h) on hepatic, renal, and supraceliac descending aortic blood flows (n = 7 swine). To account for any changes in systemic output or systemic vascular resistance during the 60 min CsA infusion that may non-specifically affect hepatic and renal blood flows, the total hepatic (portal vein plus hepatic artery) and renal blood flows were reported relative to the supraceliac descending aortic blood flow (termed 'fractional' total hepatic and renal blood flows). The fractional total hepatic blood flow decreased significantly (p < 0.05) by 40 min of CsA infusion vs baseline, and continued to decrease throughout the infusion (baseline = 0.38 + 0.03 units vs 0.28 +_0.05 units by 60 min of CsA infusion). During the recovery period, the fractional total hepatic blood flow increased to a value which was not different from baseline (recovery = 0.38 _+0.03 units). Fractional right renal artery blood flow did not change significantly from baseline at any time during the CsA infusion or during the recovery period. We conclude that a single, clinically comparable dose of CsA results in a significant decrease in total hepatic blood flow, and that this decrease is greater than that seen in renal blood flow.
Key words: Cyclosporine A; Toxicity; Blood flow; Liver; Pig
I. Introduction
Cyclosporine A (CsA) is currently considered a very effective drug to suppress the immune system
* Corresponding author, c/o James F. Burdick, Johns Hopkins Hospital, Blalock Bldg., Rm. 606, Baltimore, MD 21215, USA. Tel: + 1 (410)955-6875. Fax: + 1 (301)955-4462. Abbreviations: CsA, Cyclosporine A; Qda, supraceliac descending aorta blood flow; Qha, common hepatic artery blood flow; Qpv, portal vein blood flow; Qra, right renal artery blood flow. 0162-3109/94/$7.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0 1 6 2 - 3 1 0 9 ( 9 4 ) 0 0 0 1 8 - B
following kidney or liver transplantation (Bantle et al., 1990; Perico et al., 1990). Although CsA is widely used in renal transplantation, it can lead to acute and chronic nephrotoxicity (Greenberg, 1990). This nephrotoxicity is considered at least partially due to vasoconstriction of the glomerular afferent arterioles (Mason, 1989). M a n y transplantation centers routinely delay starting CsA therapy until adequate renal function is demonstrated after renal transplantation. In contrast, the administration of CsA to liver graft recipients is a routine
88
K.A. Amtreoni et al.
lmmunopharmacology 28 (1994) 87-94
peri-operative strategy despite the goal of optimization of blood flow during this critical period. CsA has been shown to be hepatotoxic (Greenberg, 1990; Mason, 1989; Leunissen, 1987). There is qualitative evidence that this impaired liver function may be due to alterations in hepatic blood flow (Mason, 1989; McKenzie et al., 1985). Studies using radiolabelled erythrocytes in humans (Mason, 1989) and microspheres in dogs (McKenzie et al., 1985) suggest that CsA can decrease hepatic blood flow. However, to date there have been no systematic studies investigating the absolute changes in hepatic blood flow following CsA administration. Therefore, we have investigated the acute effects of CsA on hepatic and renal blood flow m an anesthetized porcine model. CsA was infused at 5 mg/kg over a 1 h period. This dose in the pig produces comparable blood levels to a human clinical dose of 2 mg/kg given over 1 h (Frey et al., 1988). We controlled for possible non-specific effects of CsA on hepatic and renal blood flow, due to changes in cardiac output, by measuring hepatic and renal blood flow relative to supraceliac descending aortic blood flow (termed fractional hepatic and renal blood flow). In order to produce a more relevant comparison to the transplanted liver, the porcine liver was denervated by dividing the periarterial nerve bundle (Bennett et al., 1982; Daniel and Prichard, 1951; Mathie and Blumgart, 1983).
artery was cannulated to record systemic blood pressure and for arterial blood sampling. A puhnonary artery catheter (Edwards Swan-Ganz Pediatric Thermodilution Catheter, size 7 Fr, Baxter, Irvine, CA) was advanced through the right internal jugular vein into the pulmonary artery for thermodilution cardiac output measurement and body temperature monitoring. A catheter placed in the right external jugular vein was used for intravenous fluid and either cyclosporine or cremophor (diluent) administration. After midline laparotomy, the bladder was drained and a catheter was left in the bladder for decompression. The common bile duct was cannulated and then divided distal to the cannulation. The hepatic artery was dissected and the gastroduodenal and right gastric arteries were ligated and divided. The periarterial nerve bundle was identified, ligated and divided. The portal vein was dissected and a small side branch was catheterized (PE 60 tubing) for pressure measurement. Ultrasonic flow probes were placed around the portal vein (Qpv) and proximal common hepatic artery (Qh~). A third flow probe was placed around the supraceliac descending aorta (Qd~0 through an incision made in the left posteromedial aspect of the diaphragm. The right kidney was exposed by blunt dissection and a fourth flow probe was placed around the renal artery (Q,.~). The abdominal skin was reapproximated with towel clamps and dry towels were placed to minimize heat loss.
2. Materials and methods
2.2. Methods qf measurement
2.1. Surgical procedure
Blood flows were measured by the use of an ultrasonic transit time flow meter (T201 2-channel blood flow meter, Transonic Systems Inc., Ithaca, NY). Vascular pressures were measured with Isotec disposable pressure transducers (Healthdyne, Irvine, CA). Signals from the flow meters and pressure transducers were recorded on a pen recorder (Gould Co., Cleveland, OH). All vascular pressure measurements were zeroed to the level of the portal vein. Arterial blood gas measurements of pO> pCO~ and pH were performed on an ABL 30 analyzer (Radiometer, Copenhagen). Cardiac output was measured by thermodilution in triplicate and averaged. In one animal, samples of EDTA treated whole blood were analyzed for CsA levels using a mono-
An approved animal care protocol was followed. Female juvenile swine weighing 35-45 kg were fasted for 24 h, except for water ad libitum. Anesthesia was induced with intramuscular ketamine (25 mg/kg). A catheter was placed in an ear vein and pentobarbital sodium (10 mg/kg i.v.) was given to allow for tracheostomy and mechanical ventilation (Harvard Apparatus, South Natick, MA). Anesthesia was continued with isoflurane (1.2~o in 50°Jo O2), and adjusted to maintain stable heart rate and blood pressure. The animal was placed on a heating blanket and heat lamps were used to maintain core body temperature above 36 ° C. The right common carotid
K.A. Andreoni et al. / Immunopharmacology 28 (1994) 87-94
clonal RIA (Cyclotrac, INCSTAR Corp., Stillwater, MN) performed by the Johns Hopkins clinical laboratories.
2.3. Experimental protocol The preparation was allowed 30 min to stabilize after the completion of surgery before baseline hemodynamic values were obtained. Each of the seven CsA-treated animals received a 5 mg/kg dose of cyclosporine (Sandimmune, 50 mg/ml in cremophor, Sandoz Pharmaceuticals). The CsA was dissolved in 5 ~o dextrose to a total volume of 20 ml, and was infused over a 60 min period. After the infusion period, monitoring continued until the fractional hepatic blood flow recovered to baseline (90 + 30 min). A time control study was performed in three animals using an identical protocol to that described above, except that cremophor (vehicle; CremophorEL, a gift of Sandoz Research Institution, East Hanover, N J) was substituted for CsA. These animals were followed to 90 min post-infusion. Arterial blood gas evaluation was performed at baseline, near the end of the infusion period, and then at least every 60 min during the recovery period. Ventilatory rate and tidal volume were adjusted to maintain p C O 2 at 35 to 45 mmHg (sodium bicarbonate was not added at any time). Sodium chloride (0.9 ~o) was infused via the right external jugular vein at a rate of 5 ml/kg/h and acted as a carrier for the CsA and cremophor. One animal was treated with a series of CsA bolus infusions of 1 mg/kg, 3 mg/kg, 5 mg/kg, and 10 mg/kg. This was performed to compare hepatic and renal blood flow changes in our model to the more widely reported models which utilize bolus CsA infusions. At the end of each study, the pig was sacrificed with an overdose of i.v. potassium chloride and a bilateral thoracotomy was performed.
2.4. Test of hepatic artery denervation A stimulating electrode (Grass Inc., Quincy, MA) was used to stimulate (40 V, 10 Hz) the hepatic periarterial nerve bundle and demonstrate the liver's denervation after transection of the nerve bundle.
89
2.5. Data analysis Data are presented as means + SE. A one-way analysis of variance (ANOVA) was used to analyze changes across time within a treatment, and the Neuman-Keuls test was used to identify differences between means within a treatment group. Differences were considered significant for p < 0.05. This allowed us to determine differences in flow at various time points during CsA infusion within the treated group. The cremophor-treated animals were not used for statistical comparison with the CsA group, but are presented to demonstrate the lack of effect of cremophor on either hemodynamics or the stability of the preparation over time. The renal arterial blood flow measurements presented are from the right renal artery only, and do not represent the total bilateral renal blood flow.
2.6. Definitions Total hepatic blood flow is the sum of the hepatic arterial and portal venous blood flows. 'Fractional hepatic artery blood flow' is the blood flow of the hepatic artery divided by the descending aorta blood flow (Qha/Qda)" 'Fractional portal vein blood flow' is the blood flow of the portal vein divided by the descending aorta blood flow (Qpv/Qda). 'Fractional hepatic blood flow' is the sum of the hepatic artery and portal vein flows divided by the descending aorta blood flow ([Qha + Qpv]/Qda). 'Fractional renal artery blood flow' is the blood flow of the right renal artery divided by the descending aorta blood flow (Qra/Qaa). 3. Results
3.1. Test of hepatic artery denervation The stimulating electrode was placed on the distal end of the transected periarterial nerve bundle, and stimulation produced a large and sustained decrease in hepatic arterial blood flow. The periarterial nerve bundle was transected a second time, now between the stimulating electrode and the liver. Restimulation at the original position on the periarterial nerve bundle now failed to change hepatic arterial
90
K.A. Andreoni et al. ,/lmmunopharmacology 28 (1994) 87-94
blood flow. This demonstrates that transection of the periarterial nerve bundle is an effective method for hepatic arterial denervation. For completeness, the periarterial nerve bundle distal to the second point of transection (i.e. closer to the liver) was stimulated. This resulted in a similar decrease in hepatic arterial blood flow as was demonstrated in the first stimulation (data not shown).
3.2. Effects of 60 rain CsA infusion Hemodynamic and blood gas parameters The mean + SE for the hemodynamic and arterial blood gas values for the seven animals studied at baseline, at the end of 60 min of CsA infusion, and after 90 min of recovery are shown in Table I. Arterial blood gas values remained within the normal range and heart rate and portal venous pressure remained constant throughout the experiment. The baseline MAP of 80 _+8 mmHg did not change during the 60 min CsA infusion, but did decrease significantly (p<0.05) to 67 + 7 mmHg by the end of the recovery period. Cardiac output and descending aortic blood flow both decreased significantly by the end of the CsA infusion and remained at this level throughout the recovery period.
Time course of blood flow changes relative to descending aortic.flow To account for any non-specific changes in hepatic and renal blood flows during the CsA infusion (e.g. decreased cardiac output (McKenzie et al., 1985; Toung et al., 1992) or increased systemic vascular resistance (Mason, 1989; Leunissen et al.. 1987)), the hepatic and renal blood flows are plotted relative to the descending aortic blood flow in Figs. 1 and 2 (fractional total hepatic [{Qp~ + Qh~J,' Qda] and fractional renal blood flows [Q~a/Q&]). The fractional total hepatic blood flow decreased during CsA infusion, first reaching significance after 40 min (p < 0.05; Fig. 1). During the recovery period, the fractional total hepatic blood flow increased significantly ( p < 0.05) returning to the baseline value (Fig. 1). The fractional renal blood flow did not change significantly from baseline at any' time during the CsA infusion or during the recovery' period (Fig. 2). Time course of hepatic and renal blood flow changes The total hepatic blood flow decreased significantly (p < 0.05) by 10 rain of the CsA infusion, and continued to fall throughout the CsA infusion
Table 1 H e m o d y n a m i c a n d arterial b l o o d g a s results for 60 rain C s A - t r e a t e d a n i m a l s
Q CO (ml/min) Q Desc Ao (ml/min) Q Renal A (ml/min) Q H e p A (rnl/min) Q Portal V ( m l / m i n ) Q Hepatic (ml/min) F r a c t i o n a l Q portal vein F r a c t i o n a l Q h e p a t i c artery MAP (mmHg) P portal V (mmHg) H e a r t rate ( b p m ) pO 2 (mmHg) pCO: (mmHg) pH p < 0.05 vs baseline. b p < 0 . 0 5 end o f C s A vs recovery. Q = flow; p = p r e s s u r e . M A P = m e a n arterial p r e s s u r e .
Baseline
End of CsA
Recover)
3497 + 377 2214 + 235 196 + 28 27(1 _+ 53 591 + 117 861 + 127 0.261 ,+ 0.026 O. 124 ,+ {1.023 8(} ,+ 8 9.8 + 1.4 145 -+ 20 298 _+ 51 38 + 0.9 7.38 + 0.01
2781 + 335 ~' 1699 + 218" 164 _+ 26 ~' 168 + 55 ~' 345 + 10& 513,+ 135 ~ 0.186 _+0.036 '~ 0.094 ,+ (}.031 77 ,+ 9 10.3 + 1.4 135-+ 17 258 + 60 42 + 2.2 7.35 ,+ 0.01
2771 + 404 ~' 1764 + 253 ~' 153 _+ 29" 259 -+ 64 ~' 442 + 98 ~'h 701 + 139 ~'h 0.243 ,+ 0.(/22 0.142 _4__0.029 67 + 7 :'h 1(1-- 1.7 135 _+ 14 218 + 35 4(} + 1.3 7.36 + 0.02
K.A. Andreoni et al. / Immunopharmacology 28 (1994) 87-94 0.60
91
1.00 4-
! 0
0.76
n-
0
0.50
U
'!
=I. m -r "8 I-0.00
I
I 10
I 20
I 30
I 40
I 50
I 60
0.28
0.00 !
I
i 10
20
[ ' ~ 1
I 30
1 I 40
I 50
CSA InfuMon (mkn,
I 60
l I~-~"
~
Fig. 3. Mean ( + SE; n = 7) of the total hepatic blood flow (hepatic artery + portal vein blood flows) during baseline, after 60 rain of 5 mg/kg CsA infusion, and after recovery (90 + 30 rain). * p < 0.05 vs baseline. + p < 0 . 0 5 vs 60 min CsA infusion.
Fig. 1. 'Fractional total hepatic blood flow' is the ratio of the total hepatic blood flow (hepatic artery plus portal vein) relative to supraceliac descending aortic blood flow for each animal, then averaged (+ SE, n = 7 ) . * p < 0 . 0 5 vs baseline; + p < 0 . 0 5 vs 60 min CsA infusion.
Effects of 60 min cremophor (vehicle) infusion (Fig. 3). The total hepatic blood flow significantly increased (p < 0.05) during the recovery period, but remained statistically less ( p < 0.05) than the baseline value. In contrast to total hepatic blood flow, the renal blood flow did not significantly decrease until 60 min of CsA infusion and did not significantly increase during the recovery period (Fig. 4).
~,~
0,12 -
0.30
l.
11 j
<
0.20
0.o6
0 0
,! ~'*
The CsA vehicle, cremophor, was infused over 60 min into three animals to demonstrate the stability of the preparation; these results were not used for statistical comparison with the CsA-treated group. Fig. 5 demonstrates that there was only a slight, insignificant change due to cremophor on the fraction of blood flow to the liver or kidney during either the 60 min infusion or the 90 min recovery period.
] I m
0.10
0.00 I ~
i 10 1
i 20
i 30
I 40
I 50
Cr,mophor Infusion (mini
I 60
i I I Rec°vlry I
Fig. 2. 'Fractional renal artery blood flow' is the ratio of right renal artery blood flow relative to supraceliac descending aortic blood flow for each animal, then averaged ( + SE, n = 7). No statistical difference between means (p>0.05) was observed before, during, or after the 60 min CsA infusion.
0.00 I
I 111/
[-~"~
I 10
I
I 20
I 30
! 40
I 50
C S A I n f u ~ o n (min)
! 60
I
~
Fig. 4. Mean ( _+SE; n = 7) of the right renal artery blood flow during baseline, after 60 min of 5 mg/kg CsA infusion, and after recovery (90 _+30 min). * p < 0 . 0 5 vs baseline.
I
K.A. Andreoni et al. /hnmunopharmacology 28 (1994) 87-94
92
Effects of bolus CsA inJusion
0.60
I--
D
,,.
_=
0 0
1"0 e. 41
~
Fig. 6 demonstrates the effect of CsA boluses of 1 mg/kg, 3 mg/kg, 5 mg/kg and 10 mg/kg on the total hepatic and renal artery blood flows. Renal blood flow was diminished to a greater extent by the lower CsA doses than was hepatic flow. The highest dose of 10 mg/kg profoundly diminished both flows with eventual recovery of the flows through both organs.
•
0.00
[ ~
0.12
"
0.06
-
0.00
-
i 10
i 20
l
i 30
C ....
i 40
i0
i 60
6
phor Infusion (mln)
4. Discussion
I l[~ov.~l
0 "1= 0 e. •
¢j
¢: 0
=, ee
cO ° o• ,,'t
I
i 10
|
i
i
i
20
30
40
60
~0
i
CSA I n f u s l ~ ' l ( r a i n )
Fig. 5. Fractional total hepatic and fractional right renal blood flows before, during and after Cremophor infusion. Cremophor was infused at the same concentration and rate as that contained with the 5 mg/kg CsA infusion in three animals.
0.75
Bolus of Cyclo~orlne 3 mglkg ! 5 mglkg
I mglkg:
06
"~
-
I
A :
',,
:/
10 mglkg
I
\
.o.0
m 0
:
; 3o
oo
9o
Toted Hepabc Flow
............
"
|
120
1go
i 1~
Time (mini
Renld ktewy Flow
Fig. 6. Total hepatic and right renal artery blood flows in one animal after subsequent bolus injections of CsA at 1, 3, 5 and 10 mg/kg. See text for details.
This study is the first to quantitatively demonstrate a reversible decrease in total hepatic blood flow during a 60 min infusion of a clinical dose of CsA. This decrease in total hepatic blood flow was independent of changes in descending aortic blood flow. There was an absolute decrease in renal arterial blood flow during CsA infusion, but no change was observed in the proportion of descending aortic blood flow to the kidney. This greater effect of CsA infusion on total hepatic blood flow than on renal arterial blood flow is surprising when considering the well established acute and chronic nephrotoxicity, secondary to renal vasoconstriction, associated with CsA use in man (Greenberg, 1990; McKenzie et al., 1985; Carrier et al., 1991) and other mammals (Murray et al., 1985; Perico et al., 1990; Greenberg, 1990; McKenzie et al., 1985; Carrier et al., 1991). Such a decrease in hepatic blood flow during CsA infusion may have implications for maintaining normal hepatic function and for maintaining hepatic artery flow to prevent vascular thrombosis. There is evidence that cremophor, the vehicle for CsA, may produce an anaphylactoid systemic response when infused intravenously into humans (Friedman et al., 1985; Howrie et al., 1985; Kahan et al., 1984; Leunissen et al., 1985; Ptachcinski et al., 1985) and into dogs (Bowers et al., 1991). In our study, however, there was no evidence of such a systemic response to cremophor in the three animals given this vehicle over 1 h. Furthermore, crem o p h o r failed to produce any change in either hepatic or renal blood flow during the 60 min perfusion period or the 90 rain recovery period. The absence of changes in hepatic and renal blood flows during cremophor infusion simultaneously
K.A. Andreoni et aL / Immunopharmacology 28 (1994) 87-94
confirms the stability of the preparation over time. Additional evidence for the stability of our model is demonstrated by the return to baseline of the fractional renal and hepatic blood flows after CsA infusion. The 5 mg/kg dose of CsA infused over 1 h was chosen to duplicate the blood levels of CsA observed when humans receive the routine clinical dose of 2 mg/kg i.v. Frey et al. (1988) infused an identical 5 mg/kg dose of CsA over 60 min in swine, and measured blood levels of CsA over time. They reported CsA blood levels of approximately 3000 ng/ml at the end of the 60 min infusion period, and 60 and 120 min later, values of approximately 800 and 600 ng/ml. We measured CsA blood levels in one animal at equivalent time points and found values of 4152, 1258, and 848 ng/ml respectively. The CsA blood levels obtained in our study are comparable to those following a CsA infusion of 2 mg/kg over 60 min in humans (Frey et al., 1988). All seven animals displayed a decrease in total hepatic blood flow by the end of 60 min of CsA infusion. However, the pattern of the response between the decrement in hepatic arterial flow versus portal venous flow showed individual variation. This suggests that CsA does not necessarily vasoconstrict the superior mesenteric or hepatic/celiac arterial systems to the same degree. Nevertheless, 60 min of CsA caused significant decreases (p < 0.05) in both hepatic artery and portal vein blood flows, which both increased significantly (p<0.05) during the recovery period (Table I). In contrast to previously published reports of CsA-induced renal vasoconstriction (Perico et al., 1990; Carrier et al., 1991), we did not observe a significant decrease in corrected renal arterial blood flow over the 1 h infusion or the 90 min recovery period. Since many published studies were performed using boluses of CsA, we measured renal and hepatic blood flow in one pig following a series of CsA bolus infusions (Fig. 6). In contrast to the 60 min infusion, a bolus infusion of 5 mg/kg of CsA resulted in a complete, but transient cessation of renal blood flow. The 10 mg/kg bolus of CsA resulted in an abrupt cessation of total hepatic and
93
renal blood flows, followed by an equally dramatic reperfusion in these organs. There is evidence that CsA can stimulate the production and release of endothelin, a potent vasoconstrictor synthesized by endothelial cells (Perico et al., 1990; McKenzie et al., 1985; Brooks et al., 1991; Bunchman and Brookshire, 1991; Deray et al., 1991; Yanagisawa and Masaki, 1989a,b; Rubanyi and Parker Botelho, 1991). The above described pattern of blood flow changes observed following the 10 mg/kg bolus of CsA is consistent with the systemic administration of a potent vasoconstrictor, such as endothelin. However, we do not have direct evidence to support this hypothesis as plasma levels of endothelin were not measured in this study. PGE2, a naturally occurring vasodilator, has been shown to be decreased in isolated glomeruli and renal cortical and medullary slices after CsA exposure (Copeland and Yatscoff, 1991). Many clinical centers routinely use an infusion of prostaglandin E 1 when either hepatic arterial blood flow is shown to be below normal or when the post-transplant liver grafts display generalized poor function (Grieg et al., 1989). PGE1 is a potent hepatic artery vasodilator and may aid post-transplant liver function by partially reversing the vasoconstriction from CsAinduced endothelin release. Our results suggest that a single infusion of a clinical dose of CsA may significantly decrease the total hepatic blood flow. Optimal maintenance of hepatic perfusion is considered crucial for liver survival in the early post-hepatictransplant period. In summary, a 60 min infusion of CsA reduces both hepatic and renal blood flows, with a relatively larger decrease in hepatic perfusion. Further evaluation of the effects of administration of CsA on acute and chronic hepatic perfusion and function in human transplant patients would appear to be indicated.
Acknowledgements We would like to thank the Sandoz Research Institution for generous gifts of cyclosporine (Sandimmune) and Cremophor-EL.
94
K.A. Andreoni et al, / Immunopharmacology 28 (1994) 87-94
References Bantle JP, Paller MS, Bourdreau RJ, Olivari MT, Ferris TF. Long-term effects of cyclosporine on renal function in organ transplant recipients. J Lab Clin Med 1990; 115: 233-240. Bennett TD, MacAnespie CL, Rothe CF. Active hepatic capacitance responses to neural and humoral stimuli in dogs. Am J Physiol 1982; 242: H1000-H1009. Bowers VD, Locker S, Ames S, et al. The hemodynamic effects of cremophor-EL. Transplantation 1991; 51 : 847-850. Brooks DP, Ohlstein EH, Contino LC, et al. Effect of nifedipine on cyclosporin A-induced, urinary endothelin excretion and renal endothelin receptor number. Eur J Pharmacol 1991; 194: 115-117. Bunchman TE, Brookshire CA. Cyclosporine-induced synthesis of endothelin by cultured human endothelial cells. J Clin Invest 1991; 88: 310-314. Carrier M, Tronc F, Stewart D, Pelletier LC. Dose-dependent effect of cyclosporin on renal arterial resistance in dogs. Am J Physiol 1991; 261: H1791-1796. Copeland KR, Yatscoff RW. Comparison of the effects of cyclosporine and its metabolites on the release of prostacyclin and endothelin from mesangial cells. Transplantation 1991 : 53: 640-645. Daniel PM, Prichard MML. Effects of stimulation of the hepatic nerves and of adrenaline upon the circulation of the portal venous blood within the liver. J Physiol 1951; 114: 538-548. Deray G, Carayon A, LeHoang P. Increased endothelin level after cyclosporine therapy (letter to the editor). Ann Int Med 1991; 114: 809. Frey BM, Sieber M, Mettler D, Gaenger H, Frey FJ. Marked interspecies differences between humans and pigs in cyclosporine and prednisolone disposition. Drug Met Disp 1988; 16: 285-289. Friedman LS, Dienstag JL, Nelson PW, Russell PS, Cosimi AB. Anaphylactic reaction and cardiopulmonary arrest following intravenous cyclosporine. Am J Med 1985; 78: 343-345. Greenberg A. Discussion. In: Puschett JB (moderator): The Spectrum of Cyclosporin Nephrotoxicity. Am J Nephrol 1990; 10: 296-309. Grieg PD, Woolf GM, Abecassis M, et al. Prostaglandin E l for primary nonfunction following liver transplantation. Transplant Proc 1989; 21: 3360-3361.
Howrie DL, Ptachcinski RJ, Griffith BP, et al. Anaphylactoid reactions associated with parental cyclosporine use: possible role of cremophor. Drug Intel Clin Pharm 1985; 19: 425. Kahan BD, Wideman CA, Flecher S, Van Buren CT. Anaphylactic reaction to intravenous cyclosporin (Letter to the editor). Lancet 1984: 1: 52, 13. Leunissen KML, Teule J, Degenaar CP, Kho TL, Frenken LAM, van Hooff JP. Impairment of liver synthetic function and decreased liver flow during cyclosporine A therapy. Transplant Proc 1987; 19: 1822-1824. Leunissen KML, Waterval PWG, van Hooff JP. Anaphylactic reaction to intravenous cyclosporin (Letter to the editor). Lancet 1985; 1: 636. Mason J. Pharmacology of cyclosporine (Sandimmune) VII. Pathophysiology and toxicology ofcyclosporine in humans and animals. Pharmacol Rev 1989; 42: 423-434. Mathie RT, Blumgart LH. Effect of denervation on the hepatic haemodynamic response to hypercapnia and hypoxia in the dog. Pfluegers Arch 1983; 397: 152-157. McKenzie N, Deveneni R, Vezina W, Keown P, Stiller C. The effect of cyclosporine on organ blood flow. Transplant Proc 1985; 17: 1973-1975. Murray BM, Patler MS, Ferris TF. Effect of cyclosporine administration on renal hemodynamics in conscious rats. Kidney Int 1985: 28: 767-774. Perico N, Dadan J, Remuzzi G. Endothelin mediates the renal vasoconstriction induced by cyclosporine in the rat. J Am Soc Nephrol 1990: 1: 76-83. Ptachcinski RJ, Gray J, Venkataramanan R, Burckart G J, Van Thielm DH, Rosenthal JT. Anaphylactic reaction to intravenous cyclosporin (Letter to the editor). Lancet 1985: 1: 636637. Rubanyi GM, Parker Botelho LH. Endothelins. FASEB J 1991: 5: 2713-2719. Toung TJK, Bunke FJ, Grayson RF, et al. Effects ofcyclosporinc on cerebral blood flow and metabolism in dogs. Transplantation 1992; 53: 1082-1088. Yanagisawa M, Masaki T. Endothelin, a novel endotheliumderived peptide. Biochem Pharmacol 1989; 38: 1877-1883. Yanagisawa M, Masaki T. Molecular biology and biochemistry of the endothelins. Trends Physiol Sci 1989: 10: 374-378.
Immunopharmacology 28 (1994) 87-94
Hepatic and renal blood flow responses to a clinical dose of intravenous cyclosporine in the pig Kenneth A. Andreoni*, Christopher P. O'Donnell, James F. Burdick, James L. Robotham Departments of Surgery and Anesthesiology and Critical Care Medicine, Pulmonary Anesthesiology Laboratory, School of Medicine, Johns Hopkins University, Baltimore, MD 21215, USA (Revision received 7 April 1994; accepted 8 April 1994)
Abstract
The immunosuppressant Cyclosporine A (CsA) is considered to induce nephrotoxicity in part by causing vasoconstriction of the glomerular afferent arterioles. Although CsA is widely used in hepatic transplantation, little is known concerning its effects on hepatic blood flow. We used ultrasonic flow probes in an anesthetized swine model to measure the effects of a single 60 min infusion of a clinically comparable dose of CsA (5 mg/kg per h) on hepatic, renal, and supraceliac descending aortic blood flows (n = 7 swine). To account for any changes in systemic output or systemic vascular resistance during the 60 min CsA infusion that may non-specifically affect hepatic and renal blood flows, the total hepatic (portal vein plus hepatic artery) and renal blood flows were reported relative to the supraceliac descending aortic blood flow (termed 'fractional' total hepatic and renal blood flows). The fractional total hepatic blood flow decreased significantly (p < 0.05) by 40 min of CsA infusion vs baseline, and continued to decrease throughout the infusion (baseline = 0.38 + 0.03 units vs 0.28 +_0.05 units by 60 min of CsA infusion). During the recovery period, the fractional total hepatic blood flow increased to a value which was not different from baseline (recovery = 0.38 _+0.03 units). Fractional right renal artery blood flow did not change significantly from baseline at any time during the CsA infusion or during the recovery period. We conclude that a single, clinically comparable dose of CsA results in a significant decrease in total hepatic blood flow, and that this decrease is greater than that seen in renal blood flow.
Key words: Cyclosporine A; Toxicity; Blood flow; Liver; Pig
I. Introduction
Cyclosporine A (CsA) is currently considered a very effective drug to suppress the immune system
* Corresponding author, c/o James F. Burdick, Johns Hopkins Hospital, Blalock Bldg., Rm. 606, Baltimore, MD 21215, USA. Tel: + 1 (410)955-6875. Fax: + 1 (301)955-4462. Abbreviations: CsA, Cyclosporine A; Qda, supraceliac descending aorta blood flow; Qha, common hepatic artery blood flow; Qpv, portal vein blood flow; Qra, right renal artery blood flow. 0162-3109/94/$7.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0 1 6 2 - 3 1 0 9 ( 9 4 ) 0 0 0 1 8 - B
following kidney or liver transplantation (Bantle et al., 1990; Perico et al., 1990). Although CsA is widely used in renal transplantation, it can lead to acute and chronic nephrotoxicity (Greenberg, 1990). This nephrotoxicity is considered at least partially due to vasoconstriction of the glomerular afferent arterioles (Mason, 1989). M a n y transplantation centers routinely delay starting CsA therapy until adequate renal function is demonstrated after renal transplantation. In contrast, the administration of CsA to liver graft recipients is a routine
88
K.A. Amtreoni et al.
lmmunopharmacology 28 (1994) 87-94
peri-operative strategy despite the goal of optimization of blood flow during this critical period. CsA has been shown to be hepatotoxic (Greenberg, 1990; Mason, 1989; Leunissen, 1987). There is qualitative evidence that this impaired liver function may be due to alterations in hepatic blood flow (Mason, 1989; McKenzie et al., 1985). Studies using radiolabelled erythrocytes in humans (Mason, 1989) and microspheres in dogs (McKenzie et al., 1985) suggest that CsA can decrease hepatic blood flow. However, to date there have been no systematic studies investigating the absolute changes in hepatic blood flow following CsA administration. Therefore, we have investigated the acute effects of CsA on hepatic and renal blood flow m an anesthetized porcine model. CsA was infused at 5 mg/kg over a 1 h period. This dose in the pig produces comparable blood levels to a human clinical dose of 2 mg/kg given over 1 h (Frey et al., 1988). We controlled for possible non-specific effects of CsA on hepatic and renal blood flow, due to changes in cardiac output, by measuring hepatic and renal blood flow relative to supraceliac descending aortic blood flow (termed fractional hepatic and renal blood flow). In order to produce a more relevant comparison to the transplanted liver, the porcine liver was denervated by dividing the periarterial nerve bundle (Bennett et al., 1982; Daniel and Prichard, 1951; Mathie and Blumgart, 1983).
artery was cannulated to record systemic blood pressure and for arterial blood sampling. A puhnonary artery catheter (Edwards Swan-Ganz Pediatric Thermodilution Catheter, size 7 Fr, Baxter, Irvine, CA) was advanced through the right internal jugular vein into the pulmonary artery for thermodilution cardiac output measurement and body temperature monitoring. A catheter placed in the right external jugular vein was used for intravenous fluid and either cyclosporine or cremophor (diluent) administration. After midline laparotomy, the bladder was drained and a catheter was left in the bladder for decompression. The common bile duct was cannulated and then divided distal to the cannulation. The hepatic artery was dissected and the gastroduodenal and right gastric arteries were ligated and divided. The periarterial nerve bundle was identified, ligated and divided. The portal vein was dissected and a small side branch was catheterized (PE 60 tubing) for pressure measurement. Ultrasonic flow probes were placed around the portal vein (Qpv) and proximal common hepatic artery (Qh~). A third flow probe was placed around the supraceliac descending aorta (Qd~0 through an incision made in the left posteromedial aspect of the diaphragm. The right kidney was exposed by blunt dissection and a fourth flow probe was placed around the renal artery (Q,.~). The abdominal skin was reapproximated with towel clamps and dry towels were placed to minimize heat loss.
2. Materials and methods
2.2. Methods qf measurement
2.1. Surgical procedure
Blood flows were measured by the use of an ultrasonic transit time flow meter (T201 2-channel blood flow meter, Transonic Systems Inc., Ithaca, NY). Vascular pressures were measured with Isotec disposable pressure transducers (Healthdyne, Irvine, CA). Signals from the flow meters and pressure transducers were recorded on a pen recorder (Gould Co., Cleveland, OH). All vascular pressure measurements were zeroed to the level of the portal vein. Arterial blood gas measurements of pO> pCO~ and pH were performed on an ABL 30 analyzer (Radiometer, Copenhagen). Cardiac output was measured by thermodilution in triplicate and averaged. In one animal, samples of EDTA treated whole blood were analyzed for CsA levels using a mono-
An approved animal care protocol was followed. Female juvenile swine weighing 35-45 kg were fasted for 24 h, except for water ad libitum. Anesthesia was induced with intramuscular ketamine (25 mg/kg). A catheter was placed in an ear vein and pentobarbital sodium (10 mg/kg i.v.) was given to allow for tracheostomy and mechanical ventilation (Harvard Apparatus, South Natick, MA). Anesthesia was continued with isoflurane (1.2~o in 50°Jo O2), and adjusted to maintain stable heart rate and blood pressure. The animal was placed on a heating blanket and heat lamps were used to maintain core body temperature above 36 ° C. The right common carotid
K.A. Andreoni et al. / Immunopharmacology 28 (1994) 87-94
clonal RIA (Cyclotrac, INCSTAR Corp., Stillwater, MN) performed by the Johns Hopkins clinical laboratories.
2.3. Experimental protocol The preparation was allowed 30 min to stabilize after the completion of surgery before baseline hemodynamic values were obtained. Each of the seven CsA-treated animals received a 5 mg/kg dose of cyclosporine (Sandimmune, 50 mg/ml in cremophor, Sandoz Pharmaceuticals). The CsA was dissolved in 5 ~o dextrose to a total volume of 20 ml, and was infused over a 60 min period. After the infusion period, monitoring continued until the fractional hepatic blood flow recovered to baseline (90 + 30 min). A time control study was performed in three animals using an identical protocol to that described above, except that cremophor (vehicle; CremophorEL, a gift of Sandoz Research Institution, East Hanover, N J) was substituted for CsA. These animals were followed to 90 min post-infusion. Arterial blood gas evaluation was performed at baseline, near the end of the infusion period, and then at least every 60 min during the recovery period. Ventilatory rate and tidal volume were adjusted to maintain p C O 2 at 35 to 45 mmHg (sodium bicarbonate was not added at any time). Sodium chloride (0.9 ~o) was infused via the right external jugular vein at a rate of 5 ml/kg/h and acted as a carrier for the CsA and cremophor. One animal was treated with a series of CsA bolus infusions of 1 mg/kg, 3 mg/kg, 5 mg/kg, and 10 mg/kg. This was performed to compare hepatic and renal blood flow changes in our model to the more widely reported models which utilize bolus CsA infusions. At the end of each study, the pig was sacrificed with an overdose of i.v. potassium chloride and a bilateral thoracotomy was performed.
2.4. Test of hepatic artery denervation A stimulating electrode (Grass Inc., Quincy, MA) was used to stimulate (40 V, 10 Hz) the hepatic periarterial nerve bundle and demonstrate the liver's denervation after transection of the nerve bundle.
89
2.5. Data analysis Data are presented as means + SE. A one-way analysis of variance (ANOVA) was used to analyze changes across time within a treatment, and the Neuman-Keuls test was used to identify differences between means within a treatment group. Differences were considered significant for p < 0.05. This allowed us to determine differences in flow at various time points during CsA infusion within the treated group. The cremophor-treated animals were not used for statistical comparison with the CsA group, but are presented to demonstrate the lack of effect of cremophor on either hemodynamics or the stability of the preparation over time. The renal arterial blood flow measurements presented are from the right renal artery only, and do not represent the total bilateral renal blood flow.
2.6. Definitions Total hepatic blood flow is the sum of the hepatic arterial and portal venous blood flows. 'Fractional hepatic artery blood flow' is the blood flow of the hepatic artery divided by the descending aorta blood flow (Qha/Qda)" 'Fractional portal vein blood flow' is the blood flow of the portal vein divided by the descending aorta blood flow (Qpv/Qda). 'Fractional hepatic blood flow' is the sum of the hepatic artery and portal vein flows divided by the descending aorta blood flow ([Qha + Qpv]/Qda). 'Fractional renal artery blood flow' is the blood flow of the right renal artery divided by the descending aorta blood flow (Qra/Qaa). 3. Results
3.1. Test of hepatic artery denervation The stimulating electrode was placed on the distal end of the transected periarterial nerve bundle, and stimulation produced a large and sustained decrease in hepatic arterial blood flow. The periarterial nerve bundle was transected a second time, now between the stimulating electrode and the liver. Restimulation at the original position on the periarterial nerve bundle now failed to change hepatic arterial
90
K.A. Andreoni et al. ,/lmmunopharmacology 28 (1994) 87-94
blood flow. This demonstrates that transection of the periarterial nerve bundle is an effective method for hepatic arterial denervation. For completeness, the periarterial nerve bundle distal to the second point of transection (i.e. closer to the liver) was stimulated. This resulted in a similar decrease in hepatic arterial blood flow as was demonstrated in the first stimulation (data not shown).
3.2. Effects of 60 rain CsA infusion Hemodynamic and blood gas parameters The mean + SE for the hemodynamic and arterial blood gas values for the seven animals studied at baseline, at the end of 60 min of CsA infusion, and after 90 min of recovery are shown in Table I. Arterial blood gas values remained within the normal range and heart rate and portal venous pressure remained constant throughout the experiment. The baseline MAP of 80 _+8 mmHg did not change during the 60 min CsA infusion, but did decrease significantly (p<0.05) to 67 + 7 mmHg by the end of the recovery period. Cardiac output and descending aortic blood flow both decreased significantly by the end of the CsA infusion and remained at this level throughout the recovery period.
Time course of blood flow changes relative to descending aortic.flow To account for any non-specific changes in hepatic and renal blood flows during the CsA infusion (e.g. decreased cardiac output (McKenzie et al., 1985; Toung et al., 1992) or increased systemic vascular resistance (Mason, 1989; Leunissen et al.. 1987)), the hepatic and renal blood flows are plotted relative to the descending aortic blood flow in Figs. 1 and 2 (fractional total hepatic [{Qp~ + Qh~J,' Qda] and fractional renal blood flows [Q~a/Q&]). The fractional total hepatic blood flow decreased during CsA infusion, first reaching significance after 40 min (p < 0.05; Fig. 1). During the recovery period, the fractional total hepatic blood flow increased significantly ( p < 0.05) returning to the baseline value (Fig. 1). The fractional renal blood flow did not change significantly from baseline at any' time during the CsA infusion or during the recovery' period (Fig. 2). Time course of hepatic and renal blood flow changes The total hepatic blood flow decreased significantly (p < 0.05) by 10 rain of the CsA infusion, and continued to fall throughout the CsA infusion
Table 1 H e m o d y n a m i c a n d arterial b l o o d g a s results for 60 rain C s A - t r e a t e d a n i m a l s
Q CO (ml/min) Q Desc Ao (ml/min) Q Renal A (ml/min) Q H e p A (rnl/min) Q Portal V ( m l / m i n ) Q Hepatic (ml/min) F r a c t i o n a l Q portal vein F r a c t i o n a l Q h e p a t i c artery MAP (mmHg) P portal V (mmHg) H e a r t rate ( b p m ) pO 2 (mmHg) pCO: (mmHg) pH p < 0.05 vs baseline. b p < 0 . 0 5 end o f C s A vs recovery. Q = flow; p = p r e s s u r e . M A P = m e a n arterial p r e s s u r e .
Baseline
End of CsA
Recover)
3497 + 377 2214 + 235 196 + 28 27(1 _+ 53 591 + 117 861 + 127 0.261 ,+ 0.026 O. 124 ,+ {1.023 8(} ,+ 8 9.8 + 1.4 145 -+ 20 298 _+ 51 38 + 0.9 7.38 + 0.01
2781 + 335 ~' 1699 + 218" 164 _+ 26 ~' 168 + 55 ~' 345 + 10& 513,+ 135 ~ 0.186 _+0.036 '~ 0.094 ,+ (}.031 77 ,+ 9 10.3 + 1.4 135-+ 17 258 + 60 42 + 2.2 7.35 ,+ 0.01
2771 + 404 ~' 1764 + 253 ~' 153 _+ 29" 259 -+ 64 ~' 442 + 98 ~'h 701 + 139 ~'h 0.243 ,+ 0.(/22 0.142 _4__0.029 67 + 7 :'h 1(1-- 1.7 135 _+ 14 218 + 35 4(} + 1.3 7.36 + 0.02
K.A. Andreoni et al. / Immunopharmacology 28 (1994) 87-94 0.60
91
1.00 4-
! 0
0.76
n-
0
0.50
U
'!
=I. m -r "8 I-0.00
I
I 10
I 20
I 30
I 40
I 50
I 60
0.28
0.00 !
I
i 10
20
[ ' ~ 1
I 30
1 I 40
I 50
CSA InfuMon (mkn,
I 60
l I~-~"
~
Fig. 3. Mean ( + SE; n = 7) of the total hepatic blood flow (hepatic artery + portal vein blood flows) during baseline, after 60 rain of 5 mg/kg CsA infusion, and after recovery (90 + 30 rain). * p < 0.05 vs baseline. + p < 0 . 0 5 vs 60 min CsA infusion.
Fig. 1. 'Fractional total hepatic blood flow' is the ratio of the total hepatic blood flow (hepatic artery plus portal vein) relative to supraceliac descending aortic blood flow for each animal, then averaged (+ SE, n = 7 ) . * p < 0 . 0 5 vs baseline; + p < 0 . 0 5 vs 60 min CsA infusion.
Effects of 60 min cremophor (vehicle) infusion (Fig. 3). The total hepatic blood flow significantly increased (p < 0.05) during the recovery period, but remained statistically less ( p < 0.05) than the baseline value. In contrast to total hepatic blood flow, the renal blood flow did not significantly decrease until 60 min of CsA infusion and did not significantly increase during the recovery period (Fig. 4).
~,~
0,12 -
0.30
l.
11 j
<
0.20
0.o6
0 0
,! ~'*
The CsA vehicle, cremophor, was infused over 60 min into three animals to demonstrate the stability of the preparation; these results were not used for statistical comparison with the CsA-treated group. Fig. 5 demonstrates that there was only a slight, insignificant change due to cremophor on the fraction of blood flow to the liver or kidney during either the 60 min infusion or the 90 min recovery period.
] I m
0.10
0.00 I ~
i 10 1
i 20
i 30
I 40
I 50
Cr,mophor Infusion (mini
I 60
i I I Rec°vlry I
Fig. 2. 'Fractional renal artery blood flow' is the ratio of right renal artery blood flow relative to supraceliac descending aortic blood flow for each animal, then averaged ( + SE, n = 7). No statistical difference between means (p>0.05) was observed before, during, or after the 60 min CsA infusion.
0.00 I
I 111/
[-~"~
I 10
I
I 20
I 30
! 40
I 50
C S A I n f u ~ o n (min)
! 60
I
~
Fig. 4. Mean ( _+SE; n = 7) of the right renal artery blood flow during baseline, after 60 min of 5 mg/kg CsA infusion, and after recovery (90 _+30 min). * p < 0 . 0 5 vs baseline.
I
K.A. Andreoni et al. /hnmunopharmacology 28 (1994) 87-94
92
Effects of bolus CsA inJusion
0.60
I--
D
,,.
_=
0 0
1"0 e. 41
~
Fig. 6 demonstrates the effect of CsA boluses of 1 mg/kg, 3 mg/kg, 5 mg/kg and 10 mg/kg on the total hepatic and renal artery blood flows. Renal blood flow was diminished to a greater extent by the lower CsA doses than was hepatic flow. The highest dose of 10 mg/kg profoundly diminished both flows with eventual recovery of the flows through both organs.
•
0.00
[ ~
0.12
"
0.06
-
0.00
-
i 10
i 20
l
i 30
C ....
i 40
i0
i 60
6
phor Infusion (mln)
4. Discussion
I l[~ov.~l
0 "1= 0 e. •
¢j
¢: 0
=, ee
cO ° o• ,,'t
I
i 10
|
i
i
i
20
30
40
60
~0
i
CSA I n f u s l ~ ' l ( r a i n )
Fig. 5. Fractional total hepatic and fractional right renal blood flows before, during and after Cremophor infusion. Cremophor was infused at the same concentration and rate as that contained with the 5 mg/kg CsA infusion in three animals.
0.75
Bolus of Cyclo~orlne 3 mglkg ! 5 mglkg
I mglkg:
06
"~
-
I
A :
',,
:/
10 mglkg
I
\
.o.0
m 0
:
; 3o
oo
9o
Toted Hepabc Flow
............
"
|
120
1go
i 1~
Time (mini
Renld ktewy Flow
Fig. 6. Total hepatic and right renal artery blood flows in one animal after subsequent bolus injections of CsA at 1, 3, 5 and 10 mg/kg. See text for details.
This study is the first to quantitatively demonstrate a reversible decrease in total hepatic blood flow during a 60 min infusion of a clinical dose of CsA. This decrease in total hepatic blood flow was independent of changes in descending aortic blood flow. There was an absolute decrease in renal arterial blood flow during CsA infusion, but no change was observed in the proportion of descending aortic blood flow to the kidney. This greater effect of CsA infusion on total hepatic blood flow than on renal arterial blood flow is surprising when considering the well established acute and chronic nephrotoxicity, secondary to renal vasoconstriction, associated with CsA use in man (Greenberg, 1990; McKenzie et al., 1985; Carrier et al., 1991) and other mammals (Murray et al., 1985; Perico et al., 1990; Greenberg, 1990; McKenzie et al., 1985; Carrier et al., 1991). Such a decrease in hepatic blood flow during CsA infusion may have implications for maintaining normal hepatic function and for maintaining hepatic artery flow to prevent vascular thrombosis. There is evidence that cremophor, the vehicle for CsA, may produce an anaphylactoid systemic response when infused intravenously into humans (Friedman et al., 1985; Howrie et al., 1985; Kahan et al., 1984; Leunissen et al., 1985; Ptachcinski et al., 1985) and into dogs (Bowers et al., 1991). In our study, however, there was no evidence of such a systemic response to cremophor in the three animals given this vehicle over 1 h. Furthermore, crem o p h o r failed to produce any change in either hepatic or renal blood flow during the 60 min perfusion period or the 90 rain recovery period. The absence of changes in hepatic and renal blood flows during cremophor infusion simultaneously
K.A. Andreoni et aL / Immunopharmacology 28 (1994) 87-94
confirms the stability of the preparation over time. Additional evidence for the stability of our model is demonstrated by the return to baseline of the fractional renal and hepatic blood flows after CsA infusion. The 5 mg/kg dose of CsA infused over 1 h was chosen to duplicate the blood levels of CsA observed when humans receive the routine clinical dose of 2 mg/kg i.v. Frey et al. (1988) infused an identical 5 mg/kg dose of CsA over 60 min in swine, and measured blood levels of CsA over time. They reported CsA blood levels of approximately 3000 ng/ml at the end of the 60 min infusion period, and 60 and 120 min later, values of approximately 800 and 600 ng/ml. We measured CsA blood levels in one animal at equivalent time points and found values of 4152, 1258, and 848 ng/ml respectively. The CsA blood levels obtained in our study are comparable to those following a CsA infusion of 2 mg/kg over 60 min in humans (Frey et al., 1988). All seven animals displayed a decrease in total hepatic blood flow by the end of 60 min of CsA infusion. However, the pattern of the response between the decrement in hepatic arterial flow versus portal venous flow showed individual variation. This suggests that CsA does not necessarily vasoconstrict the superior mesenteric or hepatic/celiac arterial systems to the same degree. Nevertheless, 60 min of CsA caused significant decreases (p < 0.05) in both hepatic artery and portal vein blood flows, which both increased significantly (p<0.05) during the recovery period (Table I). In contrast to previously published reports of CsA-induced renal vasoconstriction (Perico et al., 1990; Carrier et al., 1991), we did not observe a significant decrease in corrected renal arterial blood flow over the 1 h infusion or the 90 min recovery period. Since many published studies were performed using boluses of CsA, we measured renal and hepatic blood flow in one pig following a series of CsA bolus infusions (Fig. 6). In contrast to the 60 min infusion, a bolus infusion of 5 mg/kg of CsA resulted in a complete, but transient cessation of renal blood flow. The 10 mg/kg bolus of CsA resulted in an abrupt cessation of total hepatic and
93
renal blood flows, followed by an equally dramatic reperfusion in these organs. There is evidence that CsA can stimulate the production and release of endothelin, a potent vasoconstrictor synthesized by endothelial cells (Perico et al., 1990; McKenzie et al., 1985; Brooks et al., 1991; Bunchman and Brookshire, 1991; Deray et al., 1991; Yanagisawa and Masaki, 1989a,b; Rubanyi and Parker Botelho, 1991). The above described pattern of blood flow changes observed following the 10 mg/kg bolus of CsA is consistent with the systemic administration of a potent vasoconstrictor, such as endothelin. However, we do not have direct evidence to support this hypothesis as plasma levels of endothelin were not measured in this study. PGE2, a naturally occurring vasodilator, has been shown to be decreased in isolated glomeruli and renal cortical and medullary slices after CsA exposure (Copeland and Yatscoff, 1991). Many clinical centers routinely use an infusion of prostaglandin E 1 when either hepatic arterial blood flow is shown to be below normal or when the post-transplant liver grafts display generalized poor function (Grieg et al., 1989). PGE1 is a potent hepatic artery vasodilator and may aid post-transplant liver function by partially reversing the vasoconstriction from CsAinduced endothelin release. Our results suggest that a single infusion of a clinical dose of CsA may significantly decrease the total hepatic blood flow. Optimal maintenance of hepatic perfusion is considered crucial for liver survival in the early post-hepatictransplant period. In summary, a 60 min infusion of CsA reduces both hepatic and renal blood flows, with a relatively larger decrease in hepatic perfusion. Further evaluation of the effects of administration of CsA on acute and chronic hepatic perfusion and function in human transplant patients would appear to be indicated.
Acknowledgements We would like to thank the Sandoz Research Institution for generous gifts of cyclosporine (Sandimmune) and Cremophor-EL.
94
K.A. Andreoni et al, / Immunopharmacology 28 (1994) 87-94
References Bantle JP, Paller MS, Bourdreau RJ, Olivari MT, Ferris TF. Long-term effects of cyclosporine on renal function in organ transplant recipients. J Lab Clin Med 1990; 115: 233-240. Bennett TD, MacAnespie CL, Rothe CF. Active hepatic capacitance responses to neural and humoral stimuli in dogs. Am J Physiol 1982; 242: H1000-H1009. Bowers VD, Locker S, Ames S, et al. The hemodynamic effects of cremophor-EL. Transplantation 1991; 51 : 847-850. Brooks DP, Ohlstein EH, Contino LC, et al. Effect of nifedipine on cyclosporin A-induced, urinary endothelin excretion and renal endothelin receptor number. Eur J Pharmacol 1991; 194: 115-117. Bunchman TE, Brookshire CA. Cyclosporine-induced synthesis of endothelin by cultured human endothelial cells. J Clin Invest 1991; 88: 310-314. Carrier M, Tronc F, Stewart D, Pelletier LC. Dose-dependent effect of cyclosporin on renal arterial resistance in dogs. Am J Physiol 1991; 261: H1791-1796. Copeland KR, Yatscoff RW. Comparison of the effects of cyclosporine and its metabolites on the release of prostacyclin and endothelin from mesangial cells. Transplantation 1991 : 53: 640-645. Daniel PM, Prichard MML. Effects of stimulation of the hepatic nerves and of adrenaline upon the circulation of the portal venous blood within the liver. J Physiol 1951; 114: 538-548. Deray G, Carayon A, LeHoang P. Increased endothelin level after cyclosporine therapy (letter to the editor). Ann Int Med 1991; 114: 809. Frey BM, Sieber M, Mettler D, Gaenger H, Frey FJ. Marked interspecies differences between humans and pigs in cyclosporine and prednisolone disposition. Drug Met Disp 1988; 16: 285-289. Friedman LS, Dienstag JL, Nelson PW, Russell PS, Cosimi AB. Anaphylactic reaction and cardiopulmonary arrest following intravenous cyclosporine. Am J Med 1985; 78: 343-345. Greenberg A. Discussion. In: Puschett JB (moderator): The Spectrum of Cyclosporin Nephrotoxicity. Am J Nephrol 1990; 10: 296-309. Grieg PD, Woolf GM, Abecassis M, et al. Prostaglandin E l for primary nonfunction following liver transplantation. Transplant Proc 1989; 21: 3360-3361.
Howrie DL, Ptachcinski RJ, Griffith BP, et al. Anaphylactoid reactions associated with parental cyclosporine use: possible role of cremophor. Drug Intel Clin Pharm 1985; 19: 425. Kahan BD, Wideman CA, Flecher S, Van Buren CT. Anaphylactic reaction to intravenous cyclosporin (Letter to the editor). Lancet 1984: 1: 52, 13. Leunissen KML, Teule J, Degenaar CP, Kho TL, Frenken LAM, van Hooff JP. Impairment of liver synthetic function and decreased liver flow during cyclosporine A therapy. Transplant Proc 1987; 19: 1822-1824. Leunissen KML, Waterval PWG, van Hooff JP. Anaphylactic reaction to intravenous cyclosporin (Letter to the editor). Lancet 1985; 1: 636. Mason J. Pharmacology of cyclosporine (Sandimmune) VII. Pathophysiology and toxicology ofcyclosporine in humans and animals. Pharmacol Rev 1989; 42: 423-434. Mathie RT, Blumgart LH. Effect of denervation on the hepatic haemodynamic response to hypercapnia and hypoxia in the dog. Pfluegers Arch 1983; 397: 152-157. McKenzie N, Deveneni R, Vezina W, Keown P, Stiller C. The effect of cyclosporine on organ blood flow. Transplant Proc 1985; 17: 1973-1975. Murray BM, Patler MS, Ferris TF. Effect of cyclosporine administration on renal hemodynamics in conscious rats. Kidney Int 1985: 28: 767-774. Perico N, Dadan J, Remuzzi G. Endothelin mediates the renal vasoconstriction induced by cyclosporine in the rat. J Am Soc Nephrol 1990: 1: 76-83. Ptachcinski RJ, Gray J, Venkataramanan R, Burckart G J, Van Thielm DH, Rosenthal JT. Anaphylactic reaction to intravenous cyclosporin (Letter to the editor). Lancet 1985: 1: 636637. Rubanyi GM, Parker Botelho LH. Endothelins. FASEB J 1991: 5: 2713-2719. Toung TJK, Bunke FJ, Grayson RF, et al. Effects ofcyclosporinc on cerebral blood flow and metabolism in dogs. Transplantation 1992; 53: 1082-1088. Yanagisawa M, Masaki T. Endothelin, a novel endotheliumderived peptide. Biochem Pharmacol 1989; 38: 1877-1883. Yanagisawa M, Masaki T. Molecular biology and biochemistry of the endothelins. Trends Physiol Sci 1989: 10: 374-378.