Cerebral ischemia alters cerebral reactivity in newborn pigs

CHARLES Department

microvascular

W. LEFFLER, DONATHAN G. BEASLEY, AND DAVID W. BUSIJA of Physiology and Biophysics, University of Tennessee, Memphis, Tennessee 38163

LEFFLER$HARLES W., DONATHANG.BEASLEY,ANDDAVID W. BUSIJA. Cerebral ischemia alters cerebral microvascular reactivity in newborn pigs. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H266-H271, 1989.-The effects of cerebral ischemia on cerebral microvascular reactivity and prostanoid synthesis were examined in chloralose-anesthetized newborn pigs. Microvascular responses and periarachnoid cerebrospinal fluid (CSF) prostanoid concentrations were determined between 10 and 140 min after a 2O-min period of total cerebral ischemia, as well as in sham-control piglets without cerebral ischemia. After cerebral ischemia, the decrease in pial arteriolar diameter in response to topical norepinephrine (low4 M) was similar in sham (-27 t 6%) and postischemic (-25 t 5%) piglets. However, the increase in pial arteriolar diameter in response to hypercapnia (10% CO:! ventilation) that was observed in sham piglets (+2l t 5%) was absent after ischemia (-2 t 3%). In contrast, dilations of pial arterioles in response to topical prostaglandin (PG)E, (at 100 ng PGEJml: sham, +I3 t 3%; postischemia, +2l k 4%) and topical isoproterenol (low6 M) (sham, +29 t 4%; postischemia, +23 k 3%) were not decreased by prior cerebral ischemia. In sham piglets, norepinephrine and hypercapnia produced increases in cortical periarachnoid prostanoid concentrations, whereas after cerebral ischemia, neither stimulus increased cortical periarachnoid prostanoid concentrations. The results are consistent with the hypothesis that failure of hypercapnia to dilate pial arterioles after cerebral ischemia results from the inability of this stimulus to increase cerebral vasodilator prostanoid synthesis. hypercapnia; prostanoids; norepinephrine; isoproterenol

and 3) pial arteriolar (PGEd.

METHODS

responses to prostaglandin

Ez

CEREBRAL

MICROVASCULAR RESPONSES to physiological and pharmacological stimuli may be altered subsequent to cerebral ischemia. In adult animals and humans, decreased cerebral vasodilator responses to hypercapnia and hypotension have been observed after cerebral ischemia (6, 18). In newborn pigs, vasodilator prostanoids in cortical periarachnoid fluid are increased during hypotension (9) and hypercapnia-hypoxia (7). Furthermore, vasodilation in response to hypercapnia (21) and hypotension (9) is attenuated by prostaglandin cyclooxygenase inhibition with indomethacin. Therefore, we hypothesize that certain alterations in cerebral microvascular responses to physiological and pharmacological stimuli subsequent to cerebral ischemia result from an inability to produce dilator prostanoids. This study addresses this hypothesis in newborn pigs by investigating the effects of total cerebral ischemia on I) pial arteriolar responses to hypercapnia, norepinephrine, and isoproterenol; 2) cerebral prostanoid synthesis;
0363-6135/89 $1.50 Copyright

Cranial window implantation. Newborn pigs (l-3 days old) were anesthetized with ketamine hydrochloride (33 mg/kg im) and acepromazine (3.3 mg/kg im) and maintained on cu-chloralose (50 mg/kg iv initially, plus 10 mg* kg- h-l). The animals were intubated and ventilated with air. Catheters were inserted in the femoral vein for maintenance of anesthesia and blood withdrawal and in the femoral artery to record blood pressure and draw samples for blood gas and pH analysis. Body temperature was maintained between 37 and 38C by wrapping the piglet in plastic film and a water-circulating heating pad. The scalp was retracted, and a Z-cm diameter hole was made in the skull over the parietal cortex. The dura was cut without touching the brain, and all cut edges were retracted over the bone so that the periarachnoid space was not exposed to damaged bone or damaged membranes. A stainless steel and glass cranial window was placed in the hole and cemented into place with dental acrylic. The space under the window was filled with artificial cerebrospinal fluid (CSF) (150 Na meq/l, 3 K+ meq/l, 2.5 Ca2 meq/l, 1.2 Mg2+ meq/l, 132 Cl- meq/l, 3.7 mM glucose, 6 mM urea, 25 HCO; meq/l; pH, 7.33; 46 mmHg Pco~, 43 mmHg Po2) through needles incorporated into the sides of the window. The volume of fluid directly under the window was 500 ~1 and was contiguous with the periarachnoid space. At the same time that the windows were implanted, at a site remote from the window, a hollow stainless steel bolt was implanted in the skull of each piglet without damaging the dura. The hollow bolts allowed infusion of artificial CSF into the cranium to increase intracranial pressure, producing ischemia. The bolts also were secured with dental acrylic. After implantation of the window and the hollow bolt, 20 min were allowed for exchange and equilibration of fluid under the window with the periarachnoid fluid before experimentation was begun. Pial arterioles were observed with a trinocular stereomicroscope. Pial arteriolar diameter was measured with a video micrometer coupled to a television camera mounted on the microscope and a video monitor. Using a stage micrometer, we determined that the scale is linear over the range of O-1,000 pm. Cerebral surface CSF (300 ~1) was collected by placing a l-ml syringe on an injection port of the cranial window. CSF was collected by slowly infusing artificial CSF into
l

H266

0 1989 the American

Physiological

Society

VASCULAR

EFFECTS

OF

CEREBRAL

ISCHEMIA

HZ67

one side of the window, and the CSF under the window was allowed to drip freely into a collection tube on the opposite side. Experimental design. Piglets were divided into two groups of six piglets each. In the sham group, the hollow bolt was implanted, but intracranial pressure was not altered. In the ischemia group, 20 min of total brain ischemia was produced by increasing the intracranial pressure. Artificial CSF (37C) was infused into the hollow bolt in the skull to maintain intracranial pressure 15 mmHg above mean arterial pressure. Arterial blood was withdrawn as necessary to maintain the mean arterial pressure no higher than 100 mmHg. We found that this procedure results in reduction of blood flow throughout the brain and spinal cord to a level that is not detectable using radioactively labeled microspheres. At the end of the 20-min ischemia period, the intracranial pressure was returned to atmospheric, the infusion tube removed from the hollow bolt, and the bolt sealed with bone wax. Measurements of pial arteriolar diameter and arterial pressure were made before the ischemic period and throughout the first 10 min of reperfusion. At that time, the experimental protocol for investigating microvascular reactivity was begun. The experimental protocol for investigating microvascular reactivity and for collecting CSF for prostanoid analysis involved topical application of norepinephrine, arterial hypercapnia, topical application of isoproterenol, and topical application of PGEZ. Ten minutes after the cerebral ischemia or 50 min after completion of surgery in the sham group, the window was flushed with artificial CSF and a timed lo-min collection of cortical periarachnoid fluid was made. Then, norepinephrine dissolved in artificial CSF (10m4 M) was injected under the cranial window while pial arteriolar diameter and arterial pressure were measured. Ten minutes later, the CSF under the cranial window was collected for prostanoid analysis. The window was then flushed with artificial CSF two times at 5-min intervals, followed by a repeat lo-min control period, during which pial arteriolar diameter and arterial pressure were measured. Then the CSF from under the cranial window was collected for prostanoid analysis. Hypercapnia was produced by ventilating the piglet with a 10% CO,-21% Oz-69% N2 mixture, and pial arteriolar diameter and arterial pressure were measured. After 10 min of hypercapnia, the CSF from under the cranial window was collected for prostanoid analysis, and an arterial blood sample for blood gas and pH analysis was drawn. The ventilation mixture was returned to air, and the window was flushed with artificial CSF four times at 5-min intervals. Next, the ability of the pial arterioles to dilate in response to topical application of PGE2 was examined. Responses of pial arterioles to 100 ng of PGEJml and 1,000 ng PGEJml dissolved in the artificial CSF were determined. The maximum response to the lower dose of PGE2 within a lo-min period was recorded; then the higher dose was applied, and the maximal response to this dose was recorded. The CSF from under the cranial window was flushed four times at 5-min intervals, and control measurements of pial arteriolar diameters were recorded. Isoproterenol (10e6 M) dissolved in artificial CSF was then injected under the

cranial window and the maximal response within 10 min of application recorded. Prostanoid analysis. Prostanoids [6-keto-prostaglandin F1, (6-keto-PGF1,), thromboxane B2 (TXBZ), PGEZ, and PGF2, in cortical periarachnoid CSF were analyzed by radioimmunoassay against an artificial CSF matrix as described previously (7). All unknowns were processed at three dilutions, with parallelism between the unknown dilution curve and the standard curve required before the result was used. Sample dilutions used in the present study allowed analysis of prostanoid concentrations between 100 and 50,000 pg/ml. Previously, using this assay, we demonstrated large proportional increases in prostanoids examined after topical application of arachidonic acid and greater than 90% decreases in concentrations of all prostanoids examined in the cortical periarachnoid fluid following treatment with 10 mg/kg iv indomethacin in basal conditions and when stimulated with exogenous arachidonic acid (8). Our antibodies cross-react minimally (Cl%) with other prostanoids studied. Furthermore, target ligands are not displaced from the antibodies by arachidonic acid (20 pg/ml); 5-hydroxyeicosatetraenoic acid (HETE), 12-HETE, or 15-HETE (1 pg/ml); leukotriene (LT)B4, LTC4, LTD4, or LTE4 (5 pg/ml); or lipoxin A4 or lipoxin B4 (10 rig/ml). Statistical analyses. All values are presented as means t SE. Comparisons between two populations were made using t tests for planned comparisons (paired or unpaired, as appropriate).
RESULTS

During the first 10 min of reperfusion after cerebral ischemia, the diameters of the pial arterioles increased (Table 1). The time between the onset of reperfusion and the attainment of maximal pial arteriolar diameter was inconsistent, occurring as early as 1 min or as late as 9 min of reperfusion (mean 5 t 2 min). The duration of the increase in pial arteriolar diameter above the preischemia diameter, when it occurred, was short (3.3 t 1.7 min). By 10 min of reperfusion, both pial arteriolar diameter and mean arterial pressure were the same as before the ischemia (Table 1). After cerebral ischemia, pial arteriolar constriction in response to topical application of 10D4M norepinephrine was similar to the constriction observed in the sham piglets (Table 2 and Fig. 1). In contrast, the response to hypercapnia in the two groups was much different (Table 2 and Fig. 1). In sham piglets, pial arterioles dilated in
TABLE 1. Pial arteriolar diameter and mean arterial pressures before and after 20 min of cerebral ischemia
Preischemia Postischemia*
10 min Postischemia

Pial arteriolar diameter, pm Mean arterial pressure, mmHg

132klO 58t5

169+19t 65t10

138k17 60&7 diameter with preis-

Values are means k SE; n = 5 arterioles. * Maximum during first 10 min of reperfusion. t P = 0.055 compared chemia.

HZ68
TABLE

VASCULAR

EFFECTS

OF

CEREBRAL

ISCHEMIA

2. Pial arteriolar diameters, arterial pressures, blood gases,and pH

Control Norepinephrine,
1O-4 M

Control

Hypercapnia

Sham Pial arteriolar Mean arteriolar Pm,, mm@ Pace,, mmHg PH Postischemia Pial arteriolar Mean arteriolar Pm,, mmHg Pace,, mmHg
PH

diameter, pressure,

pm mmHg

191t13 61t7 91-e7 3421 7.41kO.01 150t16 55t4 94t6 34t1 7.45kO.02 Pao,, arterial O2 partial pressure

139t15* 63k7

188H9 59t6

224&17* 53t3 96t4 64tl* 7.22t0.05* 151t7 47t2 88k8 62*4* 7.21kO.04' *

diameter, pressure,

pm mmHg

113*17* 59t4

154k5 5Ok3

Values appropriate

are means control.

t SE; n = 6 arterioles.

Pace,,

arterial

COz partial

pressures.

P < 0.05 compared

with

co 2

SHAM POSTISCHEMIA
FIG. 1. Change in pial arteriolar diameter in response to topical 10e4 M norepinephrine (NE) and ventilation with 10% CO2 in newborn pigs not exposed to cerebral ischemia (sham) (n = 6) and those exposed to 20 min of cerebral ischemia (Postischemia) (n = 6). Values are means k SE. * P < 0.05 compared with sham.

response to ventilation with 10% COZ, which produced an increase in arterial CO2 partial pressure (Pa& from 34 to 64 mmHg. In contrast, the same increase in Pace,, from 34 to 62 mmHg, failed to produce an increase in pial arteriolar diameter after cerebral ischemia. To determine whether pial arterioles had the capability of dilating after ischemia, pial arteriolar responses to topical PGEZ and topical isoproterenol were examined in sham and postischemic piglets. Topical PGEZ produced similar increases in pial arteriolar diameter in sham and postischemic piglets (Fig. 2). In the sham-control piglets, the pial arteriolar diameter increased from 186 t 15 to 211 t 19 pm (P c 0.05) with 100 ng PGEJml and increased to 238 t 23 pm (P < 0.05) with 1,000 ng PGEJ ml. Similarly, after cerebral ischemia, pial arteriolar diameter increased from 139 t 11 to 167 t 11 pm (P < 0.05) with 100 ng PGEJml and increased to 173 t 14 pm (P < 0.05) with 1,000 ng PGEJml. Isoproterenol (10e6 M) produced similar increases in pial arteriolar diameter in the sham piglets [145 t 3 to 177 t 3 pm (P

< 0.05); 29 t 4% increase in diameter] and postcerebral ischemia piglets [183 t 19 to 237 t 25 pm (P < 0.05); 23 t 3% increase in diameter]. Ten minutes after a ZO-min period of total cerebral ischemia, total periarachnoid prostanoid concentrations tended to be increased when compared with sham-control piglets (Table 3), but the differences were not statistically significant. There was little variation in periarachnoid prostanoid concentrations in sham piglets but considerable variability in the piglets after cerebral ischemia. The one piglet with prostanoid levels about 10 times the sham-control levels was the one with the latest increase in pial arteriolar diameter, i.e., greatest diameter at 9 min of reperfusion. In the sham piglets, topical application of 10D4 M norepinephrine caused an increase in cortical periarachnoid prostanoid concentrations (Table 3). In contrast, after cerebral ischemia, topical norepinephrine did not increase prostanoid concentrations in cortical periarachnoid CSF. Similarly, in sham piglets, ventilation with

VASCULAR

EFFECTS

OF

CEREBRAL

ISCHEMIA

HZ69

a w 30 IW r

SHAM POSTISCHEMIA

a 5
K

20

a ml 0 E 10
W IK

0 100 1000

FIG. 2. Change in pial arteriolar diameter in response to topical prostaglandin Ez (PGEZ) in newborn pigs not exposed to cerebral ischemia (sham) (n = 5) and those exposed to 20 min of cerebral ischemia (postischemia) (n = 5). Mean arterial pressures in two groups were similar (sham 60 t 5 mmHg, postischemia 56 t 5 mmHg) and were not affected by topical PGEZ. Values are means t SE.

PGE 2 CONCENTRATION

TABLE

3. Cortical periarachnoid CSF prostanoids

Concentration, 6-keto-PGF,, TXB, pg/ml PGE, PGL

Sham Control Topical NE (1O-4 M) Postischemia Control NE

1,892+398 2,919*394*

330t127 388t85

2,422+441 6,116+1,988?

1,751&460 2,692+754*

4,346+2,302 5,520&2,790

271k140 390t243

6,185+4,437 10,366+7,859 *

4,832+2,434 4,616+1,878 with

Values are means t SE; n = 6 arterioles. control; t P = 0.055 compared with control.

P < 0.05 compared

10% CO2 increased cortical periarachnoid concentrations of all the prostanoids examined (Fig. 3). After cerebral ischemia, ventilation with 10% CO:! did not change the concentration of any of the prostanoids examined in cortical periarachnoid CSF.
DISCUSSION

Results of the present experiments indicate that 20 min of total cerebral ischemia produces surprisingly little
SHAM POSTISCHEMIA

change in basal pial arteriolar diameter after a transient dilation (Table 1) but selectively alters microcirculatory reactivity. Specifically, pial arteriolar constriction in response to norepinephrine (Table 2, Fig. 1) and pial arteriolar dilation in response to topical PGEZ or isoproterenol are unaltered (Fig. 2), whereas pial arteriolar dilation in response to hypercapnia is abolished (Table 2, Fig. 1) during the immediate reperfusion period. Furthermore, neither topical norepinephrine (Table 2) nor hypercapnia (Fig. 3) stimulate cerebral prostanoid synthesis after cerebral ischemia, whereas both these stimuli do so in sham piglets. The changes in pial arteriolar diameter occurring during the immediate reperfusion period after 20 min of cerebral ischemia in newborn pigs were relatively small and short in duration. Cerebral hyperemia, after complete and relative ischemia of various durations, has been reported in adults of numerous species (6, 17, 18). Conversely, in gerbils, no hyperemia was observed in the cerebrum either 3 or 15 min after 3 or 15 min of cerebral ischemia (20). Also, in rats, no significant increase in flow through the parietal neocortex was observed during

7000 6000 E ; 5000 0 6-KETO

n q

P-&C

4000 e - 3000 2000

TXB2 PGE2 PGFzcx

FIG. 3. Effect of hypercapnia (Con) on cortical periarachnoid prostanoid concentrations in newborn pigs without (sham) and with (postischemia) previous cerebral ischemia (20 min). n = 6 in both with congroups. * P < 0.05 compared trol. 6-keto-PGF,,, 6-ketoprostaglandin Fi,; TXB2, thromboxane Bz; PGEZ, prostaglandin EZ.

0 a e
CONT CO2 CONT CO2

HZ70

VASCULAR

EFFECTS

OF

CEREBRAL

ISCHEMIA

reperfusion after 30 min of cerebral ischemia, although no apparent increase in flow was observed at 5 min of reperfusion. Early work of Wolff and Forbes (22) using cranial windows in cats indicates that, after 5-10 min of incomplete cerebral ischemia produced by raising intracranial pressure, the already dilated pial arteries dilated further when the intracranial pressure was reduced and remained dilated for lo-15 min. Figure 10 of that paper (22) reveals a 43% increase in pial artery diameter at some point during reperfusion after 9 min of ischemia. We detected a smaller increase in diameter (26 t 14%) of very short duration. We did not monitor cerebral blood flow but, instead, measured the diameters of pial arterioles that reflect conductance of one important segment of the cerebral microvascular bed. The maximal increase in diameter observed during the first 10 min after ischemia (26 t 14%) represents an increase in approximate conductance through that pial arteriolar segment of 233 t 150%, assuming the length of the segment does not change. If upstream and downstream pressures did not change, a proportionate change in flow would occur. Dilation or constriction downstream or upstream affects flow through the pial arterioles, as well, by altering upstream and downstream pressures. Gourley and Heistad (5), during very short periods of ischemia (30 s to 2 min) and in larger vessels than we used, found increases in pial arterial diameter of -10% in adult cats. During this period, blood flow through the pial artery (red cell velocity x cross-sectional area) increased -ZSO%, although the diameter only increased lo%, which would produce approximately a 46% increase in conductance. This emphasizes the concept that pial arterial diameters reflect resistance of only one segment of the cerebral microcirculation, and conductance changes elsewhere also affect flow through that segment. Therefore, pial arteriolar diameter measurements are ideal for studying microvascular reactivity but may not predict overall changes in total vascular conductance. Recently, Mayhan et al. (11) reported that vasodilatory responses of feline pial arterioles to acetylcholine and adenosine were impaired after 30 min of cerebral ischemia, whereas vasoconstrictor responses to serotonin and angiotensin II were not. Although the results using the constrictors in cats are similar to the present experiments, those obtained with adenosine, in particular, appear to be somewhat different than would be predicted from our experiments on piglets. In piglets, isoproterenol- and PGEz-induced pial arteriolar dilation was not altered after cerebral ischemia. Therefore, one would guess that dilation caused by adenosine would not be affected either. However, we did not use adenosine and, therefore, we cannot be sure responses to this dilator were not affected. Furthermore, age and species differences certainly could be involved in producing stimulusdependent variation in experiments using adult cats and newborn pigs. The failure of hypercapnia to produce dilation of pial arterioles occurs very early in the reperfusion period. In addition, the failure of both topical norepinephrine and hypercapnia to increase cortical periarachnoid synthesis also appears to occur within 20 min of reperfusion.

Previously, we have found that prostanoids represent an important vasodilator component in cerebral circulation of the newborn pig (10). These dilator prostanoids act to attenuate constriction in response to low doses of norepinephrine and to mediate vasodilation in response to asphyxia or hypercapnia. In addition, maintenance of the cerebral circulation during hypotension in newborn pigs involves prostanoids. Thus the failure of hypercapnia to produce dilation of pial arterioles following cerebral ischemia could result from the failure of the hypercapnia to stimulate cortical vasodilator prostanoid synthesis. The failure of appropriate stimuli to increase cortical prostanoid synthesis may be caused by inactivation or inhibition of cyclooxygenase during high levels of arachidonic acid metabolism in the immediate reperfusion period. Such inhibition could result from activated oxygen species generated during this burst of cyclooxygenase activity (1), by cyclooxygenase self-deactivation (IZ), or by increased production of 15-HETE, which may inhibit cyclooxygenase directly (16). Alterations in cellular metabolism that occurred during the ischemia period could affect the ability of the stimuli to produce increased prostanoid synthesis either because of diminished phospholipase activity or damage to stimulus-effector coupling. Furthermore, lipid peroxidation in membranes caused by activated oxygen species could alter the release of free arachidonic acid in response to appropriate stimuli. Prostanoid production could have been increased during reperfusion, although the cortical periarachnoid CSF concentration during the first postischemic collection period was not increased. The means of the first cortical periarachnoid CSF prostanoid concentrations obtained after ischemia (lo-20 min postischemia) were much higher than the sham means, and the variation in these concentrations was considerable in contrast to the sham group, producing no significant difference between the two groups (Table 3). The inconsistency in the time between the onset of reperfusion and maximal pial arteriolar dilation and the considerable variation in cortical periarachnoid prostanoid concentrations at lo-20 min of reperfusion could be related. The highest cortical periarachnoid prostanoid concentration was obtained in the piglet in which the increase in pial arteriolar diameter occurred 9 min after the end of the ischemic period. If increased cyclooxygenase activity occurs as a burst and then decreases rapidly, the proximity of the collection to that burst would determine the level of prostanoids obtained, with only those collections made during the increased activity being increased greatly. Therefore, it is not surprising that the lo- to ZO-min collections did not reveal a significant increase in CSF prostanoids. This concept is reinforced further by the observation that the later CSF collection made during the control period before hypercapnia (50-60 min reperfusion) revealed much lower mean prostanoid levels and smaller variation (Fig. 3). Clearly, the failure of pial arterioles to dilate in response to hypercapnia did not result from damage to the vessels that prevented vasodilation. When a nonprostanoid-dependent cerebrovasodilator, isoproterenol (3), was

VASCULAR

EFFECTS

OF

CEREBRAL REFERENCES

ISCHEMIA

HZ71

applied to the pial arterioles of postischemia piglets, the dilation produced was identical to that observed in the sham piglets. Furthermore, topical application of the predominant prostanoid found in cortical periarachnoid CSF, PGEz (7) produced similar vasodilation in the sham and postischemia piglets. Therefore, it appears that the lack of dilation results from the failure to produce the vasodilators, probably prostanoids, rather than from vessel damage that prevents vasodilation. The similarity in pial arteriolar diameters and dilatory responses to exogenous PGEZ in the two groups might seem surprising, since the endogenous PGE, concentration in the cortical periarachnoid CSF appeared to be higher, although not significantly, in postischemic piglets than in controls. However, from previous research (9), an increase in cortical periarachnoid CSF PGEZ concentration from 2,422 pg/ml of 6,185 pg/ml would be expected to produce only a 2% increase in pial arteriolar diameter, which would not be detectable. The responses to 100 rig/ml and 1,000 rig/ml of PGEZ would probably not be altered by the much lower endogenous concentrations. We believe such levels of PGEz as were applied topically are physiologically relevant. The pocket of CSF under the window must act to dilute prostanoids entering the CSF from adjacent tissues and to dampen fluctuations in prostanoid concentrations that occur near sites of synthesis and, presumably, action. We believe the changes in periarachnoid CSF prostanoid concentrations measured with treatments in these experiments should be viewed as indications of changes in rates of production. Certainly, much higher concentrations and greater fluctuations in concentration are obtained near sites of synthesis. The prostanoids can thereby act locally to alter vascular tone. Therefore, the absolute concentration of prostanoids in cortical periarachnoid CSF may not reflect the effect of those prostanoids on pial arteriolar tone as much as the direction of change. Rising prostanoid concentrations represent increased synthesis and, therefore, high local concentrations, whereas decreasing prostanoid concentrations represent decreased synthesis and, therefore, low local concentrations. Babies experiencing difficult gestations and deliveries may sustain periods of cerebral ischemia and anoxia. After such periods, abnormal cerebral vascular responses to respiratory complications could produce further cerebral damage and complicate the postnatal course markedly.
We acknowledge the excellent technical assistance of D. Hardy, L. Doti, M. Jackson, J. Giddens, T. Nilson, and N. Leffler. The research was supported by grants-in-aid from the National Institutes of Health. The animal protocols used were reviewed and approved by the Animal Care and Use Committee of the University of Tennessee, Memphis. Address for reprint requests: C. W. Leffler, Dept. of Physiology and Biophysics, University of Tennessee, Memphis, 894 Union Ave., Memphis, TN 38163. Received 19 September 1988: accented in final form 3 March 1989.

1. ARMSTEAD, W. M., R. MIRRO, D. W. BUSIJA, AND C. W. LEFFLER. Post-ischemic generation of superoxide anion by newborn pig brain. Am. J. Physiol. 255 (Heart Circ. Physiol. 24): H401-H403, 1988. 2. ARMSTEAD, W. M., M. POURCYROUS, R. MIRRO, C. W. LEFFLER, AND D. W. BUSIJA. Platelet activating factor: a potent constrictor of cerebral arterioles in newborn pigs. Circ. Res. 62: l-7, 1988. 3. BUSIJA, D. W., AND C. W. LEFFLER. Eicosanoid synthesis elicited by norepinephrine in piglet parietal cortex. Brain Res. 403: 243248, 1987. 4. CROCKARD, H. A., F. IANNOTTI, AND G. LADDS. Cerebrovascular effects of prostaglandin inhibitors in the gerbil. J. Cereb. Blood Flow Metab. 2: 67-72, 1982. 5. GOURLEY, J. K., AND D. D. HEISTAD. Characteristics of reactive hyperemia in the cerebral circulation. Am. J. Physiol. 246 (Heart Circ. Physiol. 15): H52-H58, 1984. 6. KAGSTROM, E., M. L. SMITH, AND B. K. SIESJ~. Cerebral circulatory responses to hypercapnia and hypoxia in the recovery period following complete and incomplete ischemia in the rat. Acta Physiol. Stand. 118: 281-291, 1983. 7. LEFFLER, C. W., AND D. W. BUSIJA. Prostanoids in cortical subarachnoid cerebrospinal fluid and pial artery diameter in newborn pigs. Circ. Res. 57: 689-694, 1985. C. W., AND D. W. BUSIJA. Arachidonate metabolism on 8. LEFFLER, the cerebral surface of newborn pigs. Prostaglandins 30: 811-818, 1985. C. W., AND D. W. BUSIJA. Prostanoids and pial arteriolar 9. LEFFLER, diameter in hypotensive newborn pigs. Am. J. Physiol. 252 (Heart Circ. Physiol. 21): H687-H691, 1987. C. W., AND D. W. BUSIJA. Arachidonic acid metabolites 10. LEFFLER, and perinatal cerebral hemodynamics. Sem. Perinatol. 11: 31-42, 1987. W. G., S. M. AMUNDSEN, F. M. FARACI, AND D. D. 11. MAYHAN, HEISTAD. Responses of cerebral arteries after ischemia and reperfusion in cats. Am. J. Physiol. 252 (Heart Circ. Physiol. 21): H879H884, 1987. P., J. TURK, B. A. JAKSCHIK, A. R. MORRISON, AND 12. NEEDLEMEN, J. B. LEFKOWITH. Arachidonic acid metabolism. Annu. Reu. Biochem. 55: 69-102, 1986. T. AMANO, H. OKAYASU, AND K. MORTEL. 13. OKABE, T., J. S. MEYER, Prostaglandin inhibition and cerebrovascular hemodynamics in normal and ischemic human brain. J. Cereb. Blood Flow Metab. 3: 115-121, 1983. J. D., AND E. T. MACKENZIE. Inhibition of prostaglandin 14. PICKARD, synthesis and response of baboon cerebral circulation to carbon dioxide. Nature Lond. 245: 187-188, 1973. T., AND B. K. SIESJ~. The effect of indomethacin on the 15. SAKABE, blood flow metabolism couple in the brain under normal, hypercapnia and hypoxic conditions. Acta Physiol. Stand. 107: 283-284, 1979. 15 hydroxy 5,8,11,13 eicosa16. SETTY, B. N. Y., AND M. J. STUART. tetraenoic acid inhibits human vascular cyclooxygenase. J. Clin. Invest. 77: 202-211, 1986. 17. SNYDER, J. V., E. M. NEMOTO, R. G. CARROLL, AND P. SAFAR. Global ischemia in dogs: intracranial pressures, brain blood flow and metabolism. Stroke 6: 21-27, 1975. 18. SUNDT, T. N., JR., AND A. G. WALTZ. Cerebral ischemia and reactive hyperemia. Studies of cortical blood flow and microcirculation before, during and after temporary occlusion of the middle cerebral artery of squirrel monkeys. Circ. Res. 28: 426-433, 1971. 19. TAKAGI, S., L. COCITO, AND K. A. HOSSMAN. Blood recirculation and pharmacological responsiveness of the cerebral vasculature following prolonged ischemia of the cat brain. Stroke 8: 707-712, 1977. 20. TOMIDA, S., T. S. NOWAK, JR., K. VASS, J. M. LOHR, AND I. KLATZO. Experimental model for repetitive ischemic attacks in the gerbil: the cumulative effect of repeated ischemic insults. J. Cereb. Blood Flow Metab 7: 773-782, 1987. L. C., AND 0. P. MISHRA. Mechanisms of CO2 response 21. WAGERLE, in cerebral arteries of the newborn pig: role of phospholipase, cyclooxygenase, and lipoxygenase pathways. Circ. Res. 62: 10191026, 1988. 22. WOLFF, H. G., AND H. S. FORBES. The cerebral circulation. V. Observations of the pial circulation during changes in intracranial pressure. Arch. Neurol. Psychiat. 20: 1035-1047, 1928.