Effects of ethanol- and saline-base PGE1 on the canine cerebral circulation

Transcription

1 J Neurosurg 48: , 1978 Effects of ethanol- and saline-base PGE1 on the canine cerebral circulation DANIEL RADAWSKI, PH.D., M.D., ROBERT M. DAUGHERTY, JR., M.D., PH.D., AND THOMAS E. EMERSON, JR., PH.D. Departments of Physiology and Medicine, Michigan State University, East Lansing, Michigan The effects of ethanol- and saline-base prostaglandin Ex (PGEx) on systemic arterial blood pressure (ABP), cerebral blood flow (CBF), cerebral vascular resistance (CVR), and cerebrospinal fluid (CSF) pressure were determined in anesthetized dogs. Progressively greater carotid intra-arterial infusions of ethanol-base PGE~ moderately decreased systemic ABP and CVR while perfusion of the CSF system with PGE~ moderately increased ABP and CVR; CBF was unaffected by either route of administration and CSF pressure was constant except for a slight decrease at the lowest intraventricular perfusion rate. Similar infusions of PGE~ were administered in salinebase solutions in another group of dogs. Carotid intra-arterial infusion decreased ABP and CBF moderately at the highest infusion rate and caused a transient increase in CSF pressure. Cerebrospinal fluid system perfusion increased ABP moderately but did not affect the other parameters. These data indicate that PGE~ does not have a significant effect on cerebral hemodynamics when infused via the CSF system, but may produce slight cerebral vasodilation when infused into the carotid arteries in an ethanol base. This vasodilation may be due to autoregulation. KEY WORDS 9 cerebral hemodynamics 9 prostaglandin E~ 9 saline base 9 ethanol base p ROSTAGLANDIN El (PGEI) is a potent vasodilator that decreases vascular resistance in peripheral organs z,s and lowers total peripheral resistance e during systemic infusion. However, PGEx has been reported to have no effect on, ~ to dilate, 4,x~ or constrict ~5,16 cerebral blood vessels. The vasoconstriction produced by PGE~ occurs during intracarotid artery administration of PGE1 dissolved totally in saline (saline base), while cerebral vasodilation or no effect occurs during similar administration of PGEa dissolved initially in ethanol and diluted with saline (ethanol base). Research with salinebase PGE1 suggests an important role for this agent in the genesis of cerebral vasospasm associated with subarachnoid hemorrhage? 5,xe The aim of the present study was to determine and compare cerebral hemodynamic effects of close intra-arterial and cerebrospinal fluid (CSF) administration of salinebase and ethanol-base PGEx. Materials and Methods Preparation of Animals Mongrel dogs of both sexes ( kg) were anesthetized with Nembutal (pentobarbital) (30 mg/kg), heparinized (10 mg/kg), and artificially ventilated with a constantvolume respirator. Cerebral blood flow (CBF) was measured using the technique of Rapela and Green. TM Briefly, the technique in- 724 J. Neurosurg. / Volume 48 / May, 1978

2 PGE1 and cerebral circulation volved cannulation of the confluence of the cerebral venous sinuses. Venous outflow was directed into a 50-ml plastic reservoir and returned continuously to a cannulated femoral vein via a blood pump; CBF was measured with a cylinder and stopwatch. Extracranial contamination was prevented by blockage of the transverse canals with bone wax inserted through two small burr holes. This procedure measures blood flow from the cortex and thalamus, representing 50% to 67% of total CBF),~,t2 Cerebral venous pressure from the outflow tubing near the sinus confluens, cerebrospinal fluid (CSF) pressure from the cisterna magna, and arterial blood pressure (ABP) from a cannulated femoral artery were recorded continuously via Statham pressure transducers and a Sanborn recorder.* Cerebral vascular reactivity was tested by challenging the animal with 7.5% CO2 in the breathing gas (7.5% CO2, 21% 02 and 71.5% N~) for 5 minutes. All animals responded appropriately to this challenge; the average increase in CBF was 100%, due to a decrease in cerebral vascular resistance (CVR) of 50%. Additionally, before each experiment the cerebral venous outflow cannula was briefly occluded while cerebral venous pressure upstream from this site was monitored. Experiments were conducted in those animals that exhibited a high responsiveness to COs, and in which venous pressure rose rapidly to at least 40 mm Hg. Intracarotid Artery Infusions Local infusions were made bilaterally via cannulas inserted retrogradely into side branches of both common carotid arteries. The rate of infusion was increased progressively by increasing the infusion volume every 10 minutes. Cerebrospinal Fluid Perfusion For CSF perfusion a needle was inserted into the right lateral ventricle through a burr hole drilled in the skull; CSF outflow was *Statham pressure transducer manufactured by Statham Instruments Co., 2230 Statham Boulevard, Oxnard, California; Sanborn recorder manufactured by Sanborn Co., Waltham, Massachusetts. through a needle placed in the cisterna magna. The CSF outflow rate was determined during the last 2 minutes of each infusion period with a graduated cylinder and stopwatch. The artificial CSF consisted of a solution similar in composition and ph to dog CSF) The solutions were bubbled with 5% CO2, 95% O2 for 30 minutes just before the infusion. Four solutions were prepared, each one containing one concentration of PGE~. In addition, a control solution with no PGE1 was infused. The infusion of every CSF was maintained at 1 ml/min for at least 10 minutes before making measurements. The infusion syringe was then changed to the next higher concentration or to the control solution. It is important to mention that India ink was added to the perfusate at the termination of several experiments to check the CSF perfusate distribution. Autopsy showed that the entire brain surface, all sulci, and the circle of Willis were always deeply stained black. Experimental Groups Group 1: Ethanol Base. Group 1 contained seven dogs. Ethanol-base PGE1 was prepared fresh for each experiment by dissolving 1 mg of recently thawed PGE, solution at room temperature in 0.1 ml of 95% ethanol and then adding 0.9 ml of sodium carbonate solution. This stock solution was subsequently diluted in normal saline to a concentration of 10 #g/ml. Local infusions were made at 4, 9, and 20 ~tg/min; the CSF system was perfused with PGE1 concentrations of 0.01, 0.05, 0.5, and 1.0 #g/ml. Group l-a: Ethanol Control. Group 1-A consisted of six dogs. Ethanol was diluted with normal saline in an identical fashion as in Group 1, except no PGE~ was added and infusions were made at the same rate to determine whether ethanol had any cerebrovascular effects at the concentrations used. Group 2: Saline-Base PGEa. Group 2 included eight dogs. Saline-base PGE1 was prepared identically to ethanol-base PGE1, except that no ethanol was used. Local infusions were made at 0.4, 2 or 4, and 10 or 20 ug/min, while CSF perfusions were given at 1 ml/min at concentrations of 0.1 and 1.0 #g/ml. In four of these eight animals the results were compared to similar infusions of ethanol-base PGE1. Additionally, arterial and CSF ph, pco2, and po~ were determined J. Neurosurg. / Volume 48 / May,

3 D. Radawski, R. M. Daugherty, Jr. and T. E. Emerson, Jr. TABLE 1 Average effects of ethanol- and saline-base PGEI infused locally and intraventricularly on cerebral hemodynamics in four dogs* Ethanol-Base PGE1 Saline-Base PGEt Infusion CVR CVR Rate PA CBF (mm Hg/ PA CBF (mm Hg/ (3,/min) (mm Hg) (ml/min) ml/min) (mm Hg) (ml/min) ml/min) local infusion :~ I a= ~ t ~ 0.6t 114 =~ :~ ~= 4t 19 :~ :~ 0.9t 91 7t intraventricular infusion ~= 7 25 :~ =~ ~ =~ =~ =~ lit =~ 2.7 *Px = systemic arterial blood pressure; CBF = cerebral venous flow; CVR = cerebral vascular resistance. Values are mean SE. tp = < 0.05 relative to control. with an Astrup blood gas analyzert before infusion and during the last minute of infusion of artificial CSF in four dogs. Ethanol-base PGE~ was also tested in four of these dogs. In both groups of dogs cerebral vascular resistance (CVR) was calculated by dividing the pressure gradient (ABP-cerebral venous pressure) by cerebral venous outflow. Statistical evaluation was by the Student t-test modified for paired replicates. Differences at the p < 0.05 level were considered to be significant. Standard errors of the mean were also computed. A CSF infusion of ethanol-base PGE1 produced a slight rise in ABP while having no effect on the other data, except for a slight but significant increase in CSF pressure at the lowest infusion rate and a slight increase in CVR (Fig. 2). Measured CSF outflow rate did not change during PGE1 infusion with PA (mm Hg) 120~ * Results Local, bilateral intracarotid artery infusion of ethanol-base PGE1 produced a fall in ABP (Fig. 1). Despite this fall CBF was not significantly affected. This maintenance in CBF was accomplished by a reduction in CVR. There was a slight but significant rise in CSF pressure during local infusion. Intracarotid artery infusion of ethanol diluted with saline to the same concentration used in Group 1, but which contained no PGEx, did not alter ABP, CBF, or CVR. tastrup blood gas analyzer made by Radiometer, Emdrupvej 72, DK 2400, Copenhagen NV, Denmark, and distributed by The London Co., 811 Sharon Drive, Cleveland, Ohio, CSF PRESSURE 1 5 ~ i " (mm Hg) I O ' i CBF 5[ I (ml/min) I CVR 7 ~,. (mm Hglml/min) 1 ~ I" ~0 4 9 ~) INFUSION RATE (~g/min) FIG. 1. Average effects of local, bilateral carotid artery infusion of ethanol-base PGE~ on cerebral hemodynamics in seven dogs. PA = systemic arterial blood pressure, CSF = cerebrospinal fluid; CBF = cerebral venous outflow; CVR = cerebral vascular resistance. *p < 0.05 relative to control. (0 infusion rate). 726 J. Neurosurg. / Volume 48 / May, 1978

4 PGE1 and cerebral circulation (mm Hg) I ~ ~ (mm PA Hg) iloi 90c CSF PRESSUREISs (mm Hg) IOL CBF 20[ (ml/min) 15 f t t! [ t t t i-----t" t" t (mm CVR Hg/ml/min) 8 f 6 i i i to INFUSION RATE (vg/min) FIG 2. Average effects of intraventricular infusion of ethanol-base PGE1 on cerebral hemodynamics in seven dogs. Symbols are the same as in Fig. 1. average respective values of , and ml/min during each infusion level. Local, bilateral intracarotid artery infusion of saline-base PGEx resulted in a fall in ABP and a fall in CBF at the highest infusion rate (Fig. 3). There was no significant change in CVR. A CSF infusion of saline-base PGEI increased ABP while not affecting the other data (Fig. 4). The values for CSF gases and ph in four of the eight dogs are also shown in Fig. 4. A CSF infusion with artificial CSF solution did not affect blood gases or ph. Of the eight dogs which were infused with both ethanol- and saline-base PGE1, the results were compared in four (Table 1). Ethanol-base PGE1 given locally decreased CVR while saline-base PGE~ did not. Salinebase PGEx given by CSF infusion increased systemic pressure. The other data were not significantly affected by CSF infusion of ethanol- or saline-base PGEx. Neither arterial nor CSF ph, pco2, or pox changed during PGE~ administration. Discussion Administration of ethanol-base or salinebase PGE1 by local or CSF infusion at several different concentrations had little, if any, direct effect on cerebral hemodynamics. Infusions of ethanol-base PGEx into the carotid CSF(mmHg)PRESSURE ISft 5 (ml/min) 20 IOL CVR 6f[ ~ " ~ I (mm Hg/ml/min) i ] 4 i i i 0 Q4 d-4 10"20 INFUSION RATE (~g/min) FIG. 3. Average effects of local, bilateral carotid artery infusion of saline base PGE~ on cerebral hemodynamics in eight dogs. Symbols are same as in Fig. 1. PA (mm Hg) CSF(mmPRESSUREHg) '51~ (mllmin) 30 ph -'~ Z34 Pc02= P02 = = 37 mo$ ~! 5L 2O cvr t t (ram Hg/ml/min) 4L b i 0 OJ Ii0 INFUSION RATE (pg/min) FIG. 4. Average effects of intraventricular infusion of saline-base PGE1 on cerebral hemodynamics in eight dogs. Symbols are the same as in Fig. 1. The ph, pco= and po= of the cerebrospinal fluid before and during infusion of PGE1 are shown above the graph. J. Neurosurg. / Volume 48 / May,

5 D. Radawski, R. M. Daugherty, Jr. and T. E. Emerson, Jr. arteries decreased CVR slightly, while a similar infusion of saline-base PGEI decreased CBF slightly at the highest infusion rate. None of these changes occurred independent of change in systemic ABP. Thus, a fall in CVR or CBF was always associated with a fall in systemic ABP, or conversely when CVR rose, systemic ABP increased. The observation of a variable effect of PGE~ on ABP is readily explainable. The fall in systemic pressure during intracarotid infusion most likely resulted from the direct effect of PGE~ on the carotid sinus 9 as well as the systemic vasculature?,s,e This effect may have been partially antagonized by a direct effect of PGE~ on the renin-angiotensin system? s The increase in ABP during CSF infusion most likely resulted from direct stimulation of the central nervous system?,9 The observation of a decrease in CVR during local infusion of ethanol-base PGE~ most likely resulted from cerebral autoregulation since the fall in systemic pressure was proportional to the decrease in resistance, and CBF remained constant. These findings agree with those of Denton, et al., 4 who observed that 5 to 30 ug/kg/min of ethanol-base PGE~ infused into one pump-perfused carotid artery of "carotid sinus obliterated" monkeys had no effect on perfusion pressure independent of systemic ABP. However, our findings do not agree with these same authors/ who found that infusion of 0.1 to 1.0 t~g/kg/min of ethanol-base PGE~ into the pump-perfused carotid arteries of dogs with debuffered carotid sinus produced an average 25% fall in perfusion pressure. This was not accompanied by a coincidental fall in systemic pressure, suggesting that CVR had decreased. However, this interpretation is valid only if blood flow through the intracranial circulation remained constant with no redistribution of blood between the cerebral and cranial skin and muscle vasculature. In the dog, the extensive anastomotic connections between the supplying arteries of the head along with the possibility of alterations in blood flow through the vertebral system make interpretations of cerebral arterial pressure changes difficult at best. Also, our findings do not agree with the data of Pelofsky, et al., x~ who reported that intracarotid infusion of 1 to 100 mg/kg/min of an unstated but probably ethanol-base PGE~ dramatically increased blood flow in the internal carotid artery in five of seven baboons with intact carotid sinus. However, it is uncertain whether extracranial "head" flow may have been totally or partially responsible for this increase in internal carotid artery blood flow, especially since intracerebral angiography performed while carotid arterial blood flow was increased did not reveal significant vasodilation when compared with control angiograms. Finally, there may well not have been a statistically significant increase in carotid artery blood flow if the two groups of animals were pooled. The lack of fall in CVR concomitant with the significant fall in systemic ABP during local infusion of saline-base PGEI in the present study may indicate a slight cerebrovascular constriction, which inhibited autoregulation. These findings agree qualitatively but not quantitatively with those of Yamamoto and associates, 15,18 who observed a reduction of epicranial, microcirculatory blood flow during infusion of 0.5 tzg/min saline-base PGE1 into the carotid artery of dogs. It is possible that many of the effects of PGE1 may be dose-related. While apparently comparable doses have been infused in all of the above studies the effective concentration in the intracranial blood vessel wall is not known. Allen, et al.? have shown that small doses (< VM) of ethanol-base PGE~ constrict canine basilar and middle cerebral artery strips in vitro while higher doses ( VM to 1 10-SM) relax these same vessels. It is surprising that no changes in the cerebral vasculature independent of systemic blood pressure occurred during intraventricular infusion of either ethanol- or saline-base PGE~ if this agent is to be strongly implicated as a spasmogenic agent in subarachnoid hemorrhage? 5,1e Profound changes in CVR should occur when PGEa is infused into the CSF system since this most closely simulates the mode of admittance into the cerebral vasculature during intracranial bleeding. The present findings agree with the recent study of White, et al., ~" who showed that 0.01 to 50 /sg/kg of ethanol-base PGEx did not produce arteriographic changes when injected into the cisterna magna of dogs. The findings of a slight but significant increase in CSF pressure during local infusion of ethanol-base PGE1 confirms the observa- 728 J. Neurosurg. / Volume 48 / May, 1978

6 PGE1 and cerebral circulation tion of others. ',x~ Most likely the rise in CSF pressure resulted from an increase in intravascular volume in the cerebrum since it was associated with a decrease in CVR and CSF outflow was constant. This study does not support the hypothesis that PGEx alone causes cerebral vasospasm or that it is responsible for the vasospasm often associated with subarachnoid hemorrhage. However, our data provide no information concerning the possible interaction of PGE1 with other substances, for example, blood components, in causing cerebral vasospasm in certain clinical disorders. Acknowledgments The authors gratefully appreciate the technical assistance of Scott Underwood and Mr. Todd Burns. We thank Dr. John Pike of the Upjohn Company for the generous supply of prostaglandin. References 1. Allen GS, Henderson LM, Chou SN, et al: Cerebral arterial spasm. Parts 1 through 3. J Neurosurg 40: , Burns TD, Radawski D, Underwood R, et al: Effects of PGE1 on vascular resistances and weight of the jejunum. Proceedings of the Fifth International Congress of Pharmacology, July 23-28, 1972, San Franclseo, p 34 (Abstract) 3. Daugherty RM Jr: The effects of IV and IA prostaglandin E1 on the dog forelimb skin and muscle blood flow. Am J Physiol 220: , Denton IC Jr, White RP, Robertson JT: The effects of prostaglandins El, A1 and F2~ on the cerebral circulation of dogs and monkeys. J Neurosurg 36:34-42, Emerson TE Jr, Havran JC, Radawski D, et al: Cerebral vascular effects of increased cerebrospinal fluid concentrations of naturally occurring vasoactive agents in the dog, in Cerv6s-Navarro J, Betz E, Matakas F, et al (eds): The Cerebral Vessel Wall. New York: Raven Press, 1976, pp Emerson TE Jr, Jelks GW, Daugherty RM Jr, et al: Effects of prostaglandins E1 and F2~ on venous return and other parameters in the dog. Am J Physiol 220: , Emerson TE Jr, Radawski D, Veenendaal M, et al: Effects of cerebral ventricular systemic, and local administration of prostaglandin F2~ on canine cerebral hemodynamics. Prostaglandins 8: , Gyang EA, Deuben RR, Buckley JP: Interaction of prostaglandin E1 and angiotensin II on centrally mediated pressor activities in the cat. Proc Soc Exp Biol Med 142: , Kaplan HR, Grega G J, Sherman GP, et al: Central and reflexogenic cardiovascular actions of prostaglandin El. Int J Neuropharmacol 8:15-24, Nakano J, Chang ACK, Fisher RG: Effects of prostaglandin El, E2, A1, As, and F2~ on canine carotid arterial blood flow, cerebrospinal fluid, and intraocular pressure. J Neurosurg 38: 32-39, Pelofsky S, Jacobson ED, Fisher RG: Effects of prostaglandin E1 on experimental cerebral vasospasm. J Neurosurg 36: , Rapela CE, Green HD: Autoregulation of canine cerebral blood flow. Circ Res 14 (Suppl): , Werning C, Vetter W, Weidmann P, et al: Effects of prostaglandin E1 on renin in the dog. Am J Physiol 220: , White RP, Hagen AA, Morgan H, et al: Experimental study on the genesis of cerebral vasospasm. Stroke 6:52-56, Yamamoto YL, Feindel W, Wolfe LS, et al: Effects of prostaglandins on cerebral blood flow. Eur Neurol 6: , Yamamoto YL, Feindel W, Wolfe LS, et al: Experimental vasoconstriction of cerebral arteries by prostaglandins. J Neurosurg 37: , 1972 This research was supported in part by a grant from the Michigan Heart Association. Present address for Dr. Radawski: Mayo Clinic, Rochester, Minnesota Present address for Dr. Daugherty: College of Human Medicine, University of Wyoming, Laramie, Wyoming Address reprint requests to: Thomas E. Emerson, Jr., Ph.D., Department of Physiology, Michigan State University, East Lansing, Michigan J. Neurosurg. / Volume 48 / May,