Autoregulation of Spinal Cord Blood Flow: Is The Cord A
Microcosm Of The Brain?
ROSEMARY HICKEY, M . D . , MAURICE S. ALBIN, M . D . , M . S c , ( A N E S . ) LEONID BUNEGIN, B . S . , AND JERRY GELINEAU, B . S .
SUMMARY The autoregulatory capability of regional areas of the brain and spinal cord was demonstrat-ed in 18 rats anesthetizdemonstrat-ed with a continuous infusion of intravenous pentothal. Blood flow was measurdemonstrat-ed by the injection of radioactive microspheres (Co57, Sn113, Ru103, Sc4*). Blood flow measurements were made at varying levels of mean arterial pressure (MAP) which was altered by neosynephrine to raise MAP or trimethaphan to lower MAP. Autoregulation of the spinal cord mirrored that of the brain, with an autoregulatory range of 60 to 120 mm Hg for both tissues. Within this range, cerebral blood flow (CBF) was 59.2 ± 3.2 ml/100 g/min (SEM) and spinal cord blood flow (SCBF) was 61.1 ± 3.6. There was no significant difference in CBF and SCBF in the autoregulatory range. Autoregulation was also demonstrated regionally in the left cortex, right cortex, brainstem, thalamus, cerebellum, hippocampus and cervical, thoracic and lumbar cord. This data provides a coherent reference point hi establishing autoregulatory curves under barbiturate anesthesia. Further investigation of the effects of other anesthetic agents on autoregulation of the spinal cord is needed. It is possible that intraspinal cord compliance, like intracranial compliance, might be adversely affected by the effects of anesthetics on autoregulation.
Stroke Vol 17, No 6, 1986 HEMODYNAMIC AUTOREGULATION, a
mecha-nism intrinsic to the cerebrovascular system, and also present in many other tissues, maintains tissue blood flow within a narrow range despite changes in perfu-sion pressure. It has been firmly established as the primary blood flow regulatory mechanism in the brain and constitutes a physiological adjustment that is of major importance in maintenance of a homeostatic in-ternal environment of the brain. The first observation of cerebral autoregulation was made by Fog in 1934.1-2 He observed the responses of pial vessels of cats through a cranial window under conditions of varying arterial blood pressure and noted dilation with a fall in blood pressure and constriction with a rise in blood pressure. Lassen, in his review in 1959, established the concept of cerebral autoregulation based on actual measurements of blood flow during blood pressure alterations.3 Since that time, cerebral autoregulation has been extensively investigated and it has been found that under normotensive conditions mean arterial blood pressure can be varied from a lower limit of 60 mm Hg to an upper limit of 130 mm Hg without any measurable change in blood flow.4 Below these limits, cerebral blood flow markedly falls, and above these limits, forced dilation of arterioles occurs which is associated with disruption of the blood brain barrier and edema formation.5
Spinal cord autoregulation has been less extensively investigated. Many of the therapeutic principles ap-plied to spinal cord injured patients have been made on the basis of a parallelism between brain and spinal cord
From the Department of Anesthestiology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, Texas 78284.
This work was supported in part by a grant from the Moody Founda-tion and an NIH InstituFounda-tional Grant.
Address correspondence to: Maurice S. Albin, M.D., Department of Anesthesiology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, Texas 78284.
Received July 17, 1985; revision #1 accepted January 29, 1986.
vascular dynamics. However, a comparison between the autoregulatory capabilities of these two tissues has not been demonstrated.
The purpose of this study was to simultaneously measure spinal cord and cerebral blood flow under conditions of varying mean arterial blood pressure and to determine if this parallelism, indeed does exist. Au-toregulatory capabilities of regional areas of the brain and spinal cord will also be determined.
Eighteen adult Sprague Dawley rats (450-555 g) were anesthetized with intraperitoneal pentobarbital (75 mg/kg). The animals were intubated and ventilated with air and oxygen to maintain PaO2 (greater than 60
mm Hg) and PaCO2 (33—44 mm Hg) within normal
physiologic limits. ECG was monitored continuously and temperature (measured rectally) was maintained with warming lights. An incision was made in the left groin and a femoral artery catheter was inserted to monitor arterial blood pressure and for the sampling of arterial blood. A femoral vein catheter was inserted for drug administration. An incision was then made in the neck and a catheter was inserted into the left ventricle via the right carotid artery. Placement was confirmed by observation of the left ventricular pressure pattern. The ventricular catheter was used for microsphere in-jections in order to measure blood flows. A catheter was also inserted into the left jugular vein and ad-vanced into the superior vena cava for measurement of central venous pressure. An additional dose of intra-peritoneal pentobarbital, 25 mg/kg was given during line insertion. After completion of line insertion, the animals were paralyzed with pancuronium bromide 0.1 mg/kg/hr and maintained on a continuous intrave-nous infusion of pentothal (250 mg of pentothal mixed in 250 ml of normal saline) at a rate of 4 mg/kg/hr.
Blood flow measurements were then made in each animal at varying levels of mean arterial pressure
1184 STROKE VOL 17, No 6, NOVEMBER-DECEMBER 1986
(MAP). MAP was altered by a continuous intravenous infusion of neosynephrine (160 /Ag/ml) to raise MAP or trimethaphan (4 mg/ml) to decrease MAP. No ani-mal received both neosynephrine and trimethaphan. Highly concentrated solutions of neosynephrine and trimethaphan were selected to minimize alterations in volume status of the rats. MAP was allowed to stabilize for thirty minutes between each microsphere injection.
The microspheres used for measurement of blood flow were of 15 ± 5 /i.m diameter and labeled with four gamma-emitting isotopes (37Co,"3SN,103Ru, ^Sc). Approximately 0.3 ml of microsphere solution (500,000 microspheres per ml) was injected into the left ventricle and immediately flushed with 0.3 ml of saline. Reference arterial blood samples were with-drawn from a catheter in the femoral artery beginning immediately prior to injecting and continuing for 1 minute after initiation of the injection. The order of isotope injection was randomized for each animal.
After completion of the microsphere injections, the animals were sacrificed with intravenous KCL. The brain and spinal cord were removed and the following sections dissected by anatomical landmarks: left cor-tex, right corcor-tex, cerebellum, hippocampus, thalamus, brainstem, cervical, thoracic, and lumbar spinal cord. Radioactivity of the tissue and reference blood samples were measured by counting for 1 minute on a Packard® multichannel AutoGamma Scintillation Spectrometer. Blood flows were obtained by computer calculation (as has been previously described6), using the following formula:
F = (CT/CB) X 100 F = tissue blood flow (ml/100 g/min) CT = counts/gram of tissue
CB = counts/ml/minute of reference arterial blood. Data was then plotted against perfusion pressure for the total cerebral and spinal cord flows and for each of the individual sections. A best fit curve obtained by polynomial regression analysis was drawn through the data with calculation of a correlation coefficient and an F-ratio statistic for evaluation of statistical significance of the relationship between blood flow and perfusion pressure. Significance was taken to bep < .05. Curves for each brain and cord section were based on analysis of 51 blood flow measurements.
Within the autoregulatory range established by the best fit curves, the mean brain and mean spinal cord flows were compared. Flows in the autoregulatory range of each of the individual brain and cord sections were also compared. Data comparisons were made using a two-way analysis of variance followed by a student Neuman-Keuls (normal data) or Friedman's test (nonnormal data). Significance was again taken to bep < .05.
Arterial blood gases remained within normal phys-iological limits. Mean PaCO2 was 38.02 ± 2.6 (SD)
and mean ph was 7.35 ± .05. Mean PaO2 was 135.0
Average brain blood flow was determined by pool-ing data from the individual brain sections. Cerebral blood flow (CBF) remained relatively constant within the range of 60 to 120 mm Hg at 59.2 ± 3.2 (SEM) ml/100 g/min. Above 120 mm Hg, increases in CBF paralleled increases in perfusion pressure (PP) to a maximum flow of 214 ml/100 g/min at the highest PP measured (187 mm Hg). Below 60 mm Hg, changes in CBF also paralleled reductions in PP to a minimum flow of 5.0 ml/100 g/min at a PP of 30 mm Hg.
Average spinal cord blood flow was determined by pooling data from the three cord sections. Spinal cord blood flow (SCBF) remained relatively constant with-in the range of 60 to 120 mm Hg at 61.1 ± 3.6 ml/100 g/min. Above 120 mm Hg, SCBF increased with fur-ther increases in PP to a maximum flow of 229 ml/100 g/min at a PP of 187 mm Hg. Below 60 mm Hg, SCBF decreased with further lowering in PP to a minimum flow of 3 ml/100 g/min at a PP of 30 mm Hg.
A graphic comparison of the established autoregula-tory curves for the brain and spinal cord was made. These curves appear to be virtually identical (fig. 1). Individual autoregulatory curves for each of the brain and cord sections were established (figs. 2 and 3). Left cortical blood flow remained relatively con-stant within the range of 60 to 120 mm Hg. Blood flow in this range of PP was 54.2 ± 3.8 ml/100 g/min.
Right cortical blood flow was also relatively con-stant throughout the same range and the mean blood flow in this range was 45.1 ± 3.3 ml/100 g/min.
Brainstem, thalamic, cerebellar, and hippocampal blood flows also did not appear to vary over the range of 60 to 120 mm Hg. Mean blood flows in this range were 67.7 ± 3.3,56.5 ± 4.8, 80.1 ± 4.9, and 51.0
± 4.0 ml/100 g/min respectively.
Cervical, thoracic and lumbar blood flow also did
240-| 180- 140- 100- 60- 20--20
Autoregulation of CNS Blood Flow o Brain r=B58 p<.0002
A Cord r=B5O p<XX)02
0 40 80 120 160 190 PP(mmHg)
FIGURE 1. Autoregulation of the spinal cord mimics the brain, with an autoregulatory range of 60-120 mm Hg perfu-sion pressure.
Autoregulation of Cerebral Blood Flow E o> O O \ P low i • 220-• 180 - 140- 100- 60- 20- -20-Left Cortex r Right Cortex r Brainstem r Cerebellum r Thalamus r1 —— Hippocampus r« •.888 •.827 •.680 .800 .721 .788 1 • p<.0002 p<.0002 p<.0002 p<.0002 p<XX)02 p<JXX)2
/ffi0 40 80 120 PP(mmHg) 160 190
FIGURE 2. The left cortex, right cortex, brainstem, cerebel-lum, thalamus and hippocampus all demonstrated similar auto-regulatory patterns. Best fit curves have been obtained by poly-nomial regression analysis of the individual data points.
not vary significantly within the range of 60 to 120 mm Hg. Blood flows in this range of PP were 43.7 ± 4.8, 35.3 ± 3.1 and 56.4 ± 3.7 ml/100 g/min respective-
ly-No significant difference was demonstrated between the mean brain and spinal cord blood flows in the autoregulatory range. Left cortical blood flow was higher than right cortical blood flow in the autoregula-tory range (p < .05). Cerebellar blood flow appears to be higher than any of the other brain regions at a significance level of p < .01. Brainstem blood flow in the autoregulatory range was higher than each of the brain regions except the cerebellum (p < .05). Analy-sis of individual cord flows demonstrated that lumbar flow was higher than thoracic flow (p < .01) and cervical flow (p < .05).
A summary of regional blood flows in the autoregu-latory range is presented in table 1.
Although autoregulation has been extensively inves-tigated in the cerebral vasculature, considerably less information is available on autoregulation of spinal cord blood flow. This may be, in part, due to limita-tions in the methodology of blood flow measurement. Some of the techniques traditionally used to measure cerebral blood flow are unsuitable for the spinal cord. The Kety-Schmidt inhalation technique7 is based on measuring the mean transit time for tracer molecules to traverse the brain by analysis of systemic arterial and cerebral venous blood. This technique is unsuitable for measurement of spinal cord blood flow because of the difficulty in obtaining representative venous blood samples.8 The Lassen-Ingvar9 method of monitoring the brain washout of a radioactive inert gas following its bolus injection into the carotid artery is also
unsuit-able because it is not possible to obtain selective label-ing and recordlabel-ing of the cord.8 As a result of the limita-tions of these methods, many investigators have used diffusable indicators or hydrogen clearance techniques to estimate spinal cord blood flow. Autoradiographic analysis of tissue distribution of diffusable indicators (14C-antipyrine10 or CF313II'') has the disadvantage that
only a single measurement of blood flow can be made in the same animal. The hydrogen clearance tech-nique12' 13 involves the insertion of platinum electrodes into the cord to measure the rate of washout of inhaled hydrogen gas. This method is disadvantageous in that it requires invasion of the spinal canal. It has also yielded lower values for spinal cord blood flow than other techniques. Other methods used have included the use of a surface flow device14 which permits the measurement of qualitative blood flow changes only and Xenon washout,13 which necessitates a direct in-jection of 133Xe into the parenchyma of the spinal cord, and may increase the possibility of cord damage with multiple flow determinations.
A more recent method of measuring spinal cord blood flow is the microsphere technique.6'16 Labeled microspheres are injected into the left ventricle of ex-perimental animals to enable complete mixing with arterial blood. The microspheres distribute throughout the body and are trapped in the microcirculation, being too large to pass through the capillary bed. Their distri-bution to the organs of the body is in direct proportion to their blood flow. Microspheres labelled with differ-ent gamma emitting isotopes can be separated with differential spectroscopy. Although this technique does not readily differentiate white and gray matter flow, it does have the advantage of allowing repeated measurements of blood flow in the same animal, is neurologically noninvasive, and permits measurement of regional blood flows.
Using the microsphere technique, our values for
Autoregulation of Spinal Cord Blood Flow
c ILU /
\E $ o 1 IVJ 140- 120- 1008 0 6 0 - 40- 200 -Cervical Thoracic Lumbar
, , , , , ,r=.658 r=.632 r=624 p<.0002 / p<.002 /V p<.002 / / — >;'' 40 80 120 PP(mmHg) 160 190
FIGURE 3. Autoregulatory curves for the cervical, thoracic and lumbar spinal cord have been obtained by polynomial re-gression analysis of the individual blood flows.
1186 STROKE VOL 17, No 6, NOVEMBER-DECEMBER 1986
TABLE 1 Autoregulation of CNS Blood Flow
Region Brain (n = 31) Spinal cord (n = 31) Left cortex (n = 31) Right cortex (n = 31) Brainstem (n = 31) Thalamus (n = 31) Cerebellum (n = 31) Hippocampus (n = 31) Cervical cord (n = 31) Thoracic cord (n = 31) Lumbar cord (n = 31) Autoregulation Range (PP in mm Hg) 60-120 60-120 60-120 60-120 60-120 60-120 60-120 60-120 60-120 60-120 60-120 Blood Flow (ml/100 g/min) values given as mean + SEM 59.2±3.2 61.1+3.6 54.2±3.8* 45.1 ±3.3* 67.7±3.3t 56.5±4.8 80.1+4.9* 51.0±4.0 43.7 ±4.8 35.3±3.1 56.4±3.7§ n = number of blood flow measurements made in this range of perfusion pressure.
•Difference between left and right cortical blood flow significant atp < .05.
tDifference between brainstem and other regional brain flows significant at p < .05.
tDifference between cerebellar and other regional brain flows significant at p < .01.
§Difference between lumbar and other regional cord flows sig-nificant at p < .05.
TABLE 2 Cerebral Blood Flow in the Rat
cerebral blood flow of 59.2 ± 3.2 ml/100 g/min were comparable to those of other investigators using this technique in rats (table 2). De Ley et al17 using halo-thane (1 %) and N2O (70%) anesthesia found that mean cerebral blood flow in the rat was 128 ml/100 g/min. Pannier et al,26 also using the microsphere technique, reported a mean cerebral blood flow of 75 ml/100 g/min in pentobarbital anesthetized rats. Our studies, like those of Pannier,26 were carried out using barbitu-rate anesthesia. It is not unexpected, therefore, that our values and those reported by Pannier26 would be lower than found by De Ley,17 in which halothane and N2O anesthesia was used. Halothane, a potent cerebral va-sodilator, has repeatedly been shown to markedly in-crease cerebral blood flow. The effect of N2O on cere-bral blood flow is more controversial.
Regional cerebral blood flow in our study demon-strated a significant difference between left and right cortical flows. Although collateral and posterior circu-lation would be expected to compensate for loss of carotid flow, the most likely explanation for the corti-cal flow differences was right carotid artery occlusion during catheter insertion. Regional cerebral blood flow in the rat has also been investigated by Nystrom and Norl6n. '8 They found the following values for regional flows: cerebellar flow of 58 ml/100 g/min; brainstem flow of 54 ml/100 g/min; right hemisphere flow of 33
Author De Ley et al, 198517 NystrSm et al, 198318 Ishitsuka et al, 198219 Nilsson et al, 197620 Hernandez et al, 197621 Matsumoto et al, 197622 Gjedde et al, 197523 Eklofetal, 197324 Goldman et al, 197325 Pannier et al, 197326 Haining et al, 196827 Method microspheres microspheres hydrogen clearance I33 Xe inhalation Kety-Schmidt 133 Xe inhalation Kety-Schmidt l33 Xe tissue direct count AV shunt 133 Xe inhalation Kety-Schmidt l33 Xe inhalation Kety-Schmidt 14 C-antipyrine indicator fractionation technique 14l Ce microspheres hydrogen clearance Anesthesia halothane 70% N2O thiobutabarbital amobarbital, 70% N2O halothane. 70% N2O 70% N2O 70%N2O pentobarbital 70% N2O 70%N2O Awake Pentobarbital Pentobarbital Awake CBF (ml/100 gm/min) mean ± SEM 128±12 141±17 146±14 54±6 58 ± 8 33±5 38±5 64.5±5.5 85.0±5.6 79.5 ±2.4 103 + 22 (SD) 86±3 41 ± 2 98±6 100±3 102 ± 3 92 ± 3 82 + 3 82±2 89±2 85±2 75 79 ±22 (SD) 81 ±25 (SD) Area of brain total brain Right hemisphere left hemisphere brainstem cerebellum Rt hemisphere left hemisphere cortex thalamus whole brain cortex hemisphere whole brain cortex cortex cerebellum pons and medulla cortex
cerebellum pons and medulla whole brain frontal cortex cerebellum
TABLE 3 Spinal Cord Blood Flow in the Rat Author Nystrom et al, 198318 Rivlin ct al, 197828 Method microspheres 14 C-antipyrine Anesthesia thiobutabarbital pentobarbital SCBF (ml/100 gm/min) 48±6 42±6 74 ±10 98±9 61.4±2.9 17.7±1.0
Area of SpinaJ Cord cervical upper thoracic lower thoracic lumbar T-l gray matter T-l dorsal columns
ml/100 g/min and left hemisphere flow of 38 ml/100 g/min. Although the values reported are lower than those found in our study, the same tendency was ob-served, namely higher flows in the cerebellum and brainstem than in the cortical regions. Goldman et al23 also observed similar regional flow variations in pento-barbital anesthetized rats.
Fewer studies have investigated spinal cord blood flow in the rat (table 3). Rivlin and Tator28 measured blood flow at the T-l level using the 14C-antipyrine technique. They reported values for gray and white matter of 61 and 15 ml/100 g/min respectively. Nys-trom and Norl6n18 reported values for the cervical, upper thoracic, lower thoracic and lumbar regions of 48, 42, 74, and 98 ml/100 g/min respectively. It is apparent, therefore, that our values for spinal cord blood flow are comparable to those found by others. The anatomic distribution of blood supply to the spinal cord is in concordance with findings of regional flow variations. The blood supply can be conceived as consisting of three divisions of arterial pathways.29 The primary division consists of the vertebral, costocervi-cal, and lumbar arteries. The intermediate division consists of the radicular branches of the arterial trunks forming the primary divisions. The anterior and pos-terolateral spinal arteries make up the terminal divi-sion. It is the rudimentary development of the radicular branches of the intermediate division that renders the cord subject to ischemia. The branches are few in number: two or three for the cervical region, one for the superior thoracic region, and one large feeder called the artery of Adamkiewicz for the thoracolum-bar cord.30 The wide spacing of the radicular feeders leaves large "watershed" areas along the course of the anterior spinal artery.29 In these watershed areas, the arterial inflow is precarious and interruption of any of the feeding radicular branches may have serious ische-mic consequences.29 Their distribution provides an anatomic basis for regional flow variations.
The effects of variations of blood pressure on spinal cord blood flow has been investigated by Kindt,31 Grif-fiths,13 Kobrine et al,32 and Flohr et al.33 Kindt,31 using a surface flow device which measured qualitative changes only, reported that spinal cord blood flow did not change with systemic blood pressure elevation. Using the Xenon washout curve after direct intraparen-chymal spinal cord injection, Griffiths" demonstrated no significant change in white matter flow between 60-150 mm Hg. Below 60 mm Hg, flow decreased
with further reductions in blood pressure. Kobrine et al,32 using hydrogen clearance, also demonstrated that spinal cord blood flow in the white matter remained constant in the range of 50-135 mm Hg, indicating the presence of autoregulation. Similarly, Flohr et al,33 demonstrated with the particle distribution method that autoregulation was in effect between MAP 60-160 mm Hg.
Our study, which measured cerebral and spinal cord blood flows simultaneously, demonstrated that there was essentially no difference in the autoregulatory ca-pacities of these two tissues. Autoregulation was func-tional between MAP of 60-120 mm Hg in both brain and spinal cord. Regional blood flows demonstrated similar autoregulatory patterns. Although this similar-ity has been implied by many investigators, little or no data exists which actually compares autoregulation in regional areas of both the brain and spinal cord.
Four theories have attempted to explain autoregula-tion in the cerebral vasculature. The myogenic theory, suggested by Bayliss in 1902,M suggests that vascular smooth muscle has an inherent ability to contract in response to a rise in pressure within the vessel and relax in response to a fall in pressure. The tissue perfu-sion theory suggests that increased arterial blood pres-sure causes outward filtration of fluid which increases tissue pressure and decreases flow. A lower arterial pressure decreases the outward filtration fluid, reduc-ing tissue pressure and increasreduc-ing flow.35 The neuro-genic theory proposes that autoregulation is mediated by periadventitial nerves in the cerebral vessels.36 Evi-dence suggests however, that these perivascular fibers play only a limited role in autoregulation. The blood flow changes in response to neurogenic stimulation are small. Maximal stimulation of the sympathetic nerves reduces cerebral blood flow by only 5-10%.8 Sympa-thetic stimulation has been shown to shift the limits of the autoregulatory curve. During hemorrhagic hypo-tension, the lower limit of autoregulation is shifted towards higher pressures.37 Neurogenic influences, therefore, although not necessary for the autoregula-tory response to occur, do play a role in its modifica-tion. The metabolic theory suggests that changes in periarteriolar concentrations of metabolites mediate autoregulation. According to this theory, changes in perfusion pressure are initially reflected in altered flow. With altered flow, concentrations of perivascu-lar substances, particuperivascu-larly CO2 and lactic acid tend to
media-1188 STROKE VOL 17, No 6, NOVEMBER-DECEMBER 1986
nisms postulated, the myogenic38 and metabolic theor-ies are the most widely accepted. There is evidence to suggest that spinal cord autoregulation is also a local phenomenon. The autoregulatory response of spinal cord white matter is preserved after severing the cord high in the neck, suggesting a local regulatory mecha-nism.39
Autoregulation of cerebral blood flow has been demonstrated to be altered in a variety of circum-stances. In chronic hypertension, cerebral blood flow autoregulation is adapted to higher pressures with a shift of the entire autoregulatory curve to the right.40 Autoregulation is also impaired in patients with head injury,41 brain tumors,42 and cerebral artery vaso-spasm.43 Hypoxia and hypercarbia also abolish cere-bral autoregulation.40 Less information exists on alter-ations of spinal cord autoregulation. It has been shown to be impaired by hypoxia and hypercarbia.15 Conflict-ing results of the effects of spinal cord trauma have been reported, with some investigators reporting mar-kedly increased blood flows44 and others reporting re-duction in flows after cord injury.45 Little or no data exists on the effects of different anesthetic agents on spinal cord autoregulation. In this study, autoregula-tory curves were established using a continuous intra-venous infusion of pentothal in animals paralyzed with pancuronium. The need remains to establish autoregu-latory curves for different anesthetic agents in both normal and cord injured animals.
The data presented here has demonstrated a coherent reference point in establishing an autoregulatory curve under barbiturate anesthesia. It has been established46 that experimental impact injury to the spinal cord will cause spinal cord edema, a condition that is also noted radiologically in man. It is possible that intraspinal cord compliance, like intracranial compliance, might be adversely affected because of the effects of anesthe-sia on autoregulation of spinal cord blood flow. This is an important area needing further clarification.
Autoregulation was demonstrated to be operative throughout the central nervous system. The parallel-ism between brain and spinal cord vascular dynamics was established. Autoregulatory curves for brain and spinal cord were virtually identical, with an autoregu-latory range of 60—120 mm Hg perfusiori pressure. Regional areas of the brain also had similar autoregula-tory patterns. Further investigation of the effects of anesthetic agents on spinal cord blood flow both in the normal and cord-injured state is needed.
1. Fog M: Cerebral Circulation. The reaction of the pial arteries to a fall in blood pressure. Arch Neurol Psychiatry 37: 351-364, 1937 2. Fog M: Cerebral Circulation II. Reaction of pial arteries to increase in blood pressure. Arch Neurol Psychiatry 41: 260-268, 1939 3. Lassen NA: Cerebral blood flow and oxygen consumption in man.
Physiol Rev 39: 183-238, 1959
4. Lassen NA: Cerebral circulation and anaesthesia. Acta Anaesthiol Scand [Suppl] 70: 53-55, 1978
5. Larsen B, Lassen NA: Regulation of cerebral blood flow in health and disease. Adv Neurol 25: 7-22, 1979
6. Smith DR, Smith HJ, Rajjoub RK: Measurement of spinal cord blood flow by the microsphere technique. Neurosurgery 2: 27-30, 1978
7. Kety SS, Schmidt CF: The nitrous oxide method for the quantita-tive determination of cerebral blood flow in man: theory, procedure and normal values. J Clin Invest 27: 476-483, 1948
8. Lassen NA: Cerebral and spinal cord blood flow. In: Cottrell JE, Turndorf H (eds.): Anesthesia and neurosurgery. CV Mosby, St. Louis, 1-24, 1980
9. Lassen NA, Ingvar DH: The blood flow of the cerebral cortex determined by radioactive Krypton 85. Experentia 17:42-43, 1961 10. Sandier AN, Tator CH: Effect of acute spinal cord compression injury on regional spinal cord blood flow in primates. J Neurosurg 45: 660-676, 1976
11. Landau WM, Freygang WH Jr, Roland LP, Sokoloff L, Kety SS: The local circulation of the living brain; values in the unanesthe-tized and anestheunanesthe-tized cat. Trans Am Neurol Assoc 80: 125-129,
12. Kobrine AI, Doyle TF, Martins AN: Spinal cord blood flow in the rhesus monkey by the hydrogen clearance method. Surg Neurol 2: 197-200, 1974
13. Griffiths IR, Rowan JO, Crawford RA: Spinal cord blood flow measured by a hydrogen clearance technique. J Neurol Sci 26: 529-544, 1975
14. Kindt GW, Ducker TB, Huddlestone J: Regulation of spinal cord blood flow. In Russell RWR (ed.). Brain and blood flow. Pitman, London, 401-405, 1971
15. Griffiths IR: Spinal cord blood flow in dogs: the effect of blood pressure. Journal of Neurol Neurosurg and Psychiatry 36: 914-929, 1973
16. Marcus ML, Heistad DD, Ehrhardt JC, Abbout FM: Regulation of total and regional spinal cord blood flow. Circ Res 41: 128-134,
17. De Ley G, Nshimyumuremyi JB, Leusen I: Hemispheric blood flow in the rat after unilateral common carotid occlusion: evolution with time. Stroke 16: 69-73, 1985
18. Nystrom B, Norton K: Regional spinal cord and brain blood flows in the rat. Neurol Res 5(1): 91-101, 1983
19. Ishitsuka T, Fujishima M, Nakatomi Y, Tamaki K, Omae T: Ef-fects of hyperventilation on cerebral blood flow and brain tissue metabolism in normotensive and spontaneously hypertensive rats. Stroke 13(5): 687-692, 1982
20. Nilsson B, Siesjo BK: A method for determining blood flow and oxygen consumption in the rat brain. Acta Physiol Scand 96: 72-82, 1976
21. Hernandez MJ, Brennan RW, Bowman GS: Cerebral blood flow autoregulation in the rat. Stroke 9(2): 150-154, 1978
22. MatsumotoA, NamonR, Utsunomiya Y, KogureK, ScheinbergP, Reinmuth OM: The measurement of cerebral blood flow in the rat. Stroke 6: 630-637, 1975
23. Gjedde A, Caronna JJ, Hindfelt B, Plum F: Whole-brain blood flow and oxygen metabolism in the rat during nitrous oxide anes-thesia. J of Physiol 229(1): 113-118, 1975
21. Ekl6f B, Lassen NA, Nillson L, Norberg K, SiesjoBK: Blood flow and metabolic rate for oxygen in the cerebral cortex of the rat. Acta Physiol Scand 88: 587-589, 1973
25. Goldman H, Sapirstein LA: Brain blood flow in the conscious and anesthetized rat. Amer J Physiol 224: 122-126, 1973
26. Pannier JL, Leusen I: Circulation to the brain of the rat during acute and prolonged respiratory changes in the acid-base balance. Pflugers Arch 338: 347-359, 1973
27. Haining JL, Turner MD, Pantall RM: Measurement of local cere-bral blood flow in the unanesthetized rat using a hydrogen clear-ance method. Circ Res 23: 313-324, 1968
28. Rivlin AS, Tator CH: Regional spinal cord blood flow in rats after severe cord trauma. J Neurosurg 49: 844-853, 1978
29. Szilagyi DE, Hageman JH, Smith RF, Elliot JP: Spinal cord dam-age in surgery of the abdominal aorta. Surgery 83: 38-56, 1978 30. Djindjian R, Dublin A, Djindjian M: Angiography of the spinal cord. In: Youmas JR (eds): Neurological Surgery. W.B. Saunders, Philadelphia, 551-616, 1982
31. Kindt GW: Autoregulation of spinal cord blood flow. Europ Neurol 6: 19-23, 1971/2
affected by changes in systemic arterial blood pressure. J Neuro-surg 44: 12-15, 1976
33. Flohr H, Poll W, Brock M: Regulation of spinal cord blood flow. 40. In: Russell RWR (ed.): Brain and blood flow. Pitman, London,
406-409, 1971 41. 34. Bayliss WM: On the local reaction of the arterial wall to changes of
internal pressure. J Physiol 28: 220-231, 1902
35. Youmans JR: Cerebral blood flow in clinical problems. In: You- 42. mans JR (ed.): Neurological surgery. W. B.
Saunders,"Philadel-phia, 786-845, 1972 43. 36. Olesen J: Perfusion pressure and the cerebrovascular resistance
(CVR). Acta Neurol Scand Suppl 50: 19-30, 1974 44. 37. Fitch W, Ferguson GG Sengupta D, Garibi J, Harper AM:
Auto-regulation of cerebral blood flow during controlled hypotension in
baboons. J Neurol Neurosurg Psychiatry 39: 1014-1022, 1976 45. 38. Folkow B: Description of the myogenic hypothesis. Cir Res
[Suppl] 15: 279-287, 1964 46. 39. Kobrine AI, Doyle TF, Newby N, Rizzoli HV: Preserved
autoregu-lation in the rhesus spinal cord after high cervical cord section. J Neurosurg 44: 425-428, 1976
Lassen NA, Christensen MS: Physiology of cerebral blood flow. Br J Anaesth 48: 719-734, 1976
Enevoldsen EM, Jensen FT: Autoregulation and CO2 responses of
cerebral blood flow in patients with acute severe head injury. J Neurosurg 48: 689-703, 1978
Palvolgyi R: Regional cerebral blood flow in patients with intra-cranial tumors. J Neurosurg 31: 149-162, 1969
Fein JM, Boulos R: Local cerebral blood flow in experimental middle cerebral artery vasospasm. J Neurosurg 39: 337-347, 1973 Kobrine AI, Doyle TF, Martins AN: Local spinal cord blood flow in experimental traumatic myelopathy. J Neurosurg 42: 144-149, 1975
Griffiths IR: Spinal cord blood flow after acute experimental cord injury in dogs. Journal of the Neurol Sci 27: 247-259, 1976 Albin MS, White RJ, Yashon D, Harris LS: Effects of localized cooling on spinal cord trauma. J Trauma 9: 1000-1007, 1969
Evidence That Vasoactive Intestinal Polypeptide (VIP)
Mediates Neurogenic Vasodilation of Feline Cerebral Arteries
JOSEPH E. B R A Y D E N , P H . D . , AND JOHN A. B E V A N , M . D .
SUMMARY In this study the magnitude of non-sympathetic, non-cholinergic neurogenic vasodilation of feline cerebral arteries in vitro was correlated with the extent of innervation by VLP-immunoreactive nerves. Well-innervated arteries underwent nerve-mediated relaxation whereas those that are not supplied with VIP-containing axons did not relax to transmural nerve stimulation. The relaxation of cerebral arteries that are well endowed with VlP-immunoreactive nerves was selectively and reversibly inhibited by VIP-specific antiserum. Substance P-VIP-specific antiserum did not affect the dilator responses. We conclude that VIP is a functional neurodilator transmitter in the cerebral circulation.
Stroke Vol 17, No 6, 1986
ALTHOUGH VIP was first isolated from porcine small intestine and classified as a hormone,1'2 VIP-like immunoreactivity was soon discovered within nerves that distribute to a number of organ systems.w In 1976 Larsson and his colleagues6 described perivascular nerves that were immunoreactive for VIP within the cerebral circulation of the cat and this observation' has since been confirmed by many investigators in a num-ber of animal species.7"9 These findings imply that VIP is a cerebrovascular neurotransmitter but convincing evidence for such a possibility has been slow in forth-coming, primarily due to lack of a specific antagonist against the action of this peptide. However, the potent dilator action of VIP has led to suggestions of a neuro-vasodilator function.'Oi " In the present study we have used two separate approaches to study the possible role of VIP-immunoreactive (VIP-IR) nerves in the cere-bral circulation. First, using in vitro techniques, we
From the Department of Pharmacology, College of Medicine, The University of Vermont, Burlington, Vermont 05405.
This work has been supported by USPHS grants HL 32383 and HL 06804.
Address correspondence to: Joseph E. Brayden, Ph D., Department of Pharmacology, College of Medicine, The University of Vermont, Burlington, Vermont 05405.
Received September 6, 1985; revision #1 accepted February 10, 1986.
have studied non-cholinergic dilator responses of fe-line cerebral arteries that receive differing degrees of innervation by VIP-IR nerves, from dense to none, to elucidate possible correlations between innervation and physiological response. In addition, we have as-sessed the inhibitory actions of VIP-specific antiserum on non-adrenergic, non-cholinergic neurogenic vaso-dilator responses of the feline middle cerebral artery, a blood vessel that is richly endowed with VIP-IR nerves. Our findings support and confirm the previous suggestions of a specific vasodilator transmitter role for VIP in the cerebral circulation.
Adult, mongrel cats of either sex and weighing 4-5 kg were anesthetized with sodium pentobarbital and exsanguinated. The brain was quickly removed and placed in Krebs bicarbonate buffer that was gassed with 95% O2 and 5% CO2 (pH 7.4). In a first series of
experiments, the anterior inferior cerebellar artery and a second order branch of the middle cerebral artery (lumen diameter: —0.3 mm) were dissected from the brain and placed in Krebs solution. The cerebellar ar-tery was separated into two segments, one proximal (diameter: —0.3 mm) and one distal (diameter: —0.25 mm) to the first major side branch of the artery. Cylin-drical segments of these arteries (3 mm in length) were