Cerebral O2 Extraction Research Articles

Fetal responses to acute fetal cocaine injection in sheep

Maternal cocaine injection causes fetal hypoxemia, hypertension, and increased cerebral blood flow (CBF) in sheep. To test the hypothesis that increased fetal CBF is not due solely to fetal hypoxemia, we injected cocaine directly into a fetal vein. A single dose of cocaine [1 (group 1; n = 7) or 2 (group 2; n = 8) mg/kg i.v.] was administered to chronically catheterized, unanesthetized, near-term fetal sheep. Fetal CBF (microspheres), arterial blood pressure (BP), O2 content, and cerebral O2 consumption (CMRo2) were measured at baseline, 30 s, and 2, 5, and 15 min after fetal cocaine injection. Fetal CBF increased 27 +/- 9% (SE) at 5 min in group 1 and returned to baseline by 15 min, whereas fetal CBF increased 57 +/- 8% at 5 min and remained elevated at 15 min in group 2. Fetal BP increased at 30 min in both groups and remained increased at 2 min in group 1 and at 5 min in group 2. Cerebrovascular resistance increased at 30 s in both groups and then decreased at 5 min only in group 2. Fetal hypoxemia was observed in group 2 5 min after cocaine injection (arterial PO2 decreased 24 +/- 5%), whereas no hypoxemia was noted in group 1. CMRO2 was unchanged in group 1 but increased in group 2 at 5 min (41 +/- 10%) and was associated with an increase in cerebral O2 extraction. Increases in myocardial and adrenal blood flows and reductions in both small and large intestinal blood flows were noted at 5 min in both groups.(ABSTRACT TRUNCATED AT 250 WORDS)

Hypothermic circulatory arrest and other methods of cerebral protection during operations on the thoracic aorta.

Current surgical techniques in operations on the thoracic aorta frequently require exclusion of the cerebral circulation for varying periods. During these periods, hypothermic circulatory arrest (HCA), selective cerebral perfusion (SCP), and retrograde cerebral perfusion (RCP) can be used for cerebral protection. Hypothermia is the principle component of these methods of protection. The main protective effect of hypothermia is based on reduction of cerebral energy expenditures and largely depends on adequate suppression of cerebral function. It is most effective at deep hypothermic levels (13 degrees C to 15 degrees C). Measures that preserve autoregulation of cerebral blood flow help increase the margin of safety with all methods of protection. There is solid experimental and clinical data indicating the safe limits and outcome following HCA. Current applications of SCP and RCP are fairly recent developments and do not have comparable supporting data. SCP can be used without deep hypothermia and allows prolonged periods of cerebral protection, but is complex in application. RCP is simpler, but always requires deep hypothermia. Present clinical data do not allow separation of its protective effect from that of HCA alone. Recent modifications in the application of HCA include monitoring of cerebral O2 extraction, and selective use of supplemental SCP to limit arrest times to less than 50 minutes, or RCP to prevent embolic strokes, as indicated. These changes appear to have reduced the overall mortality, the severity of embolic strokes, and stroke-related mortality.

128 ALTERATION OF CEREBRAL OXYGENATION AND HEMODYNAMICS CAUSED BY OPENING OF THE BYPASS BRIDGE DURING ECMO

Repeated opening of the bypass bridge during extracorporeal membrane oxygenation (ECMO) is necessary to prevent clot formation. Using near infrared spectrophotometry concentration changes of oxyhemoglobin (ΔcO2Hb), deoxyhemoglobin (ΔcHHb) and total hemoglobin (ΔctHb) in cerebral tissue were measured during a 10 s opening of the bypass bridge during ECMO in 12 anesthetized piglets. ΔcO2Hb and ΔcHHb reflect changes in cerebral O2 supply, while ΔctHb reflects changes in cerebral blood volume (CBV). Simultaneously heart rate (HR), arterial O2 saturation (saO2), mean arterial blood pressure (MABP), central venous pressure (CVP), intracranial pressure (ICP) and mean blood flow in left common carotid artery (CaBF) were also continuously measured. Most variables showed biphasic changes. The peak values are shown in table (*=p<0.05). HR and saO2 were unchanged.Conclusion: opening of the bypass bridge results initially in decreased CaBF, CBV and cerebral O2 supply which is compensated by increased cerebral O2 extraction and vasodilatation, resulting in secondary increased CBV and ICP. The increased CVP might be caused by retrograde bloodflow through the venous cannula. Repeated opening of the bypass bridge during ECMO might contribute to cerebral hemorrhage.

Methylene blue inhibits hypoxic cerebral vasodilation in awake sheep.

Cerebral vasodilation in hypoxia may involve endothelium-derived relaxing factor-nitric oxide. Methylene blue (MB), an in vitro inhibitor of soluble guanylate cyclase, was injected intravenously into six adult ewes instrumented chronically with left ventricular, aortic, and sagittal sinus catheters. In normoxia, MB (0.5 mg/kg) did not alter cerebral blood flow (CBF, measured with 15-microns radiolabeled microspheres), cerebral O2 uptake, mean arterial pressure (MAP), heart rate, cerebral lactate release, or cerebral O2 extraction fraction (OEF). After 1 h of normobaric poikilocapnic hypoxia (arterial PO2 40 Torr, arterial O2 saturation 50%), CBF increased from 51 +/- 5.8 to 142 +/- 18.8 ml.min-1 x 100 g-1, cerebral O2 uptake from 3.5 +/- 0.25 to 4.7 +/- 0.41 ml.min-1 x 100 g-1, cerebral lactate release from 2 +/- 10 to 100 +/- 50 mumol.min- x 100 g-1, and heart rate from 107 +/- 5 to 155 +/- 9 beats/min (P < 0.01). MAP and OEF were unchanged from 91 +/- 3 mmHg and 48 +/- 4%, respectively. In hypoxia, 30 min after MB (0.5 mg/kg), CBF declined to 79.3 +/- 11.7 ml.min-1 x 100 g-1 (P < 0.01), brain O2 uptake (4.3 +/- 0.9 ml.min-1 x 100 g-1) and heart rate (133 +/- 9 beats/min) remained elevated, cerebral lactate release became negative (-155 +/- 60 mumol.min-1 x 100 g-1, P < 0.01), OEF increased to 57 +/- 3% (P < 0.01), and MAP (93 +/- 5 mmHg) was unchanged. The sheep became behaviorally depressed, probably because of global cerebral ischemia. These results may be related to interference with a guanylate cyclase-dependent mechanism.(ABSTRACT TRUNCATED AT 250 WORDS)

Epinephrine Dosage Effects on Cerebral and Myocardial Blood Flow in an Infant Swine Model of Cardiopulmonary Resuscitation

Although epinephrine increases cerebral blood flow (CBF) and left ventricular blood flow (LVBF) during cardiopulmonary resuscitation (CPR), the effects of high dosages on LVBF and CBF and cerebral O2 uptake have not been examined during prolonged CPR. We determined whether log increment dosages of epinephrine would enhance LVBF and CBF and cerebral O2 uptake in an infant swine CPR model. We compared these responses with epinephrine to those with the alpha-adrenergic agonist, phenylephrine. CPR was performed in five groups (n = 6) of pentobarbital-anesthetized piglets (3.5-5.6 kg) receiving a continuous epinephrine infusion (0, 1, 10, and 100 micrograms.kg-1.min-1) or phenylephrine infusion (40 micrograms.kg-1.min-1). Plasma epinephrine concentrations increased 10-100-fold in the control group during CPR and in a stepwise manner such that concentrations were increased by more than 10(4) in the 100 micrograms.kg-1.min-1 epinephrine group. In the control group with no epinephrine infusion, LVBF decreased to less than 10 ml.min-1.100 g-1 by 5 min of CPR. With epinephrine in dosages of 10 and 100 micrograms.kg-1.min-1, LVBF at 5 min was 75 +/- 19 and 44 +/- 15 ml.min-1.100 g-1, respectively, which was significantly greater than values in the control group. With more prolonged CPR, LVBF remained significantly greater than that in the control group but only at 10 micrograms.kg-1.min-1 of epinephrine. Phenylephrine also increased LVBF for 10 min of CPR when compared with the control group. All dosages of epinephrine and phenylephrine maintained CBF close to prearrest values for 20 min of CPR. With prolonged CPR, 10 and 100 micrograms.kg-1.min-1 epinephrine resulted in significantly greater CBF than that in the control group. Incremental dosages of epinephrine did not statistically increase cerebral O2 uptake or lower the cerebral fractional O2 extraction when compared with the control group, despite the higher CBF that was generated. In this immature animal CPR model, 10 micrograms.kg-1.min-1 epinephrine is an optimal dosage for maximizing both CBF and LVBF, a dosage that substantially exceeds the current recommended epinephrine dosage for human infant CPR. In addition, for short periods of CPR, 40 micrograms.kg-1.min-1 phenylephrine increases CBF and LVBF to levels similar to those generated by high dosages of epinephrine.

Intracranial pressures and O2 extraction in conscious sheep during 72 h of hypoxia

High-altitude cerebral edema may occur within several days after a rapid ascent to altitude. However, the mechanisms that produce this potentially lethal condition are unclear. This experiment assessed systemic arterial and intracranial pressures (n = 22) and cerebral blood flow per unit cerebral O2 consumption (Qc/cVo2) and cerebral O2 extraction fraction (cEo2) (n = 9) in conscious sheep before and during 72 h of normobaric hypoxia (arterial O2 tension approximately 40 mmHg, oxygen saturation in arterial blood approximately 50%). Qc/cVo2 and cEo2 were calculated from systemic arterial and cerebral venous O2 contents. Wet-to-dry brain weight ratios were calculated during normoxia (n = 4) or after 72 h of hypoxia (n = 5) in additional sheep. Intracranial pressures did not change during hypoxia (+ 1.3 to +1.8 mmHg, P = 1.0); however, estimated intracranial capillary hydrostatic pressure may have increased 1-20 mmHg depending on the arterial-to-downstream resistance ratio. During the 72 h of hypoxia, Qc/cVO2 doubled (P = 0.02) and cEo2 tended to decrease (5% absolute, P = 1.0). Regional wet-to-dry brain weight ratios after 72 h of hypoxia were 4-13% greater than their respective ratios during normoxia (0.001 less than P less than or equal to 0.05); indirect evidence suggests that this increased brain water content was extravascular. The sheep may be an appropriate model for the further study of high-altitude cerebral edema.

Effect of salbutamol on regional cerebral oxygen consumption, flow and capillary and arteriolar perfusion

This study quantitatively determined the effect of salbutamol (1 microgram kg-1), a beta 2-adrenoceptor agonist, on the perfusion of the brain microvasculature, cerebral O2 consumption, O2 extraction and cerebral blood flow (CBF) in conscious rat. Indices of arteriolar and capillary structure and the percentage of the total cerebral microvascular volume/mm3 (% Vv) and number/mm2 (% Na) perfused were determined. These parameters were obtained from the perfused microvessels, identified by the presence of fluorescein isothiocyanate (FITC) - dextran, and compared with the entire microvascular bed, identified by alkaline phosphatase stain. Cerebral O2 extraction was determined microspectrophotometrically and CBF was determined using 14[C]iodoantipyrine in another group of salbutamol-treated rats. The acute administration of salbutamol did not alter systemic arterial blood pressure. Significant tachycardia was noted in the salbutamol-treated rats. Salbutamol resulted in a significant increase in the percentage of arterioles perfused. Average percentage perfused capillary Na increased significantly from 46 +/- 2 to 88 +/- 1%; %Vv increased significantly and similarly in the arteriolar and capillary beds in all brain regions examined. Average cerebral O2 consumption increased significantly from 3.0 +/- 0.2 to 7.4 +/- 0.7 ml O2 min-1 100 g-1 with salbutamol, while cerebral O2 extraction was unchanged. Average CBF increased from 50 +/- 2 to 142 +/- 9 ml min-1 100 g-1 with salbutamol. Salbutamol may increase the perfusion of the regional microvasculature by increasing cerebral O2 consumption (metabolic vasodilation) and CBF and microvascular perfusion secondarily, although a direct effect of salbutamol on cerebral microvessels cannot be ruled out.

Oxygen delivery rate and sufficiency of oxygenation during ECMO in newborn baboons.

Minimum acceptable O2 delivery (DO2) during extracorporeal membrane oxygenation (ECMO) remains to be defined in a newborn primate model. The right atrium, carotid artery, and femoral artery were cannulated, and the ductus arteriosus, aorta, and pulmonary artery ligated in neonatal baboons (Papio cynocephalus) under a combination of ketamine, diazepam, and pancuronium. The internal jugular vein was also cannulated retrograde to the level of the occipital ridge. We measured hemoglobin, pH, arterial and venous PO2 (both from the pump circuit and from the cerebral venous site), serum lactate and bicarbonate concentrations, and pump flow, and we calculated hemoglobin saturations, (DO2), O2 consumption (VO2), systemic O2 extraction, and cerebral O2 extraction. Six baboons were studied during each of two phases of the experiment. In the first, flow rates were varied sequentially from 200 to 50 ml.kg-1.min-1 with saturation maximized. In the second, flow was maintained at 200 ml.kg-1.min-1 and saturation was reduced sequentially from 100 to 38%. VO2 fell significantly below baseline at a flow rate of 50 ml.kg-1.min-1 and a DO2 of 8 +/- 2 (SE) ml.kg-1.min-1 in phase 1 and at DO2 of 12 +/- 5 in phase 2. Both systemic and cerebral O2 extraction rose significantly at a flow of 100 ml.kg-1.min-1 and DO2 of 17 +/- 4 ml.kg-1.min-1 in phase 1, whereas neither rose with decreasing DO2 in phase 2. In fact, cerebral extraction fell significantly DO2 of 16 +/- 6 ml.kg-1.min-1.(ABSTRACT TRUNCATED AT 250 WORDS)

Cerebral blood flow and O2 metabolism after asphyxia in neonatal lambs.

A neonatal lamb model has been developed to examine the regulation of cerebral blood flow (CBF) and oxygen metabolism during the critical period after an asphyxial insult. Nine newborn lambs had control measurements and timed measurements after asphyxia of CBF (radioactive microsphere technique), arterial and cerebral venous (sagittal sinus) blood gases and oxygen contents performed. Immediately after resuscitation from asphyxia, there was a marked increase in CBF compared to control (239 +/- 22 versus 82 +/- 7 ml X 100 g-1 X min-1, mean +/- SEM; p less than 0.01). Cerebral oxygen delivery (CBF X arterial O2 content) increased from 12.87 +/- 1.20 to 37.40 +/- 3.40 ml X 100 g-1 X min-1 (p less than 0.01), while cerebral O2 consumption was significantly decreased compared to control (4.75 +/- 0.42 to 3.42 +/- 0.46 ml X 100 g-1 X min-1, p less than 0.05). Cerebral fractional O2 extraction, the relationship between oxygen uptake and delivery fell from 0.38 +/- 0.03 to 0.09 +/- 0.02; p less than 0.01. This reactive hyperemia was followed in all animals by a period of hypoperfusion. CBF (52 +/- 4 ml X 100 g-1 X min-1), O2 delivery (7.94 +/- 0.50 ml X 100 g-1 X min-1), and cerebral O2 consumption (3.34 +/- 0.24 ml X 100 g-1 X min-1) were all significantly depressed when compared to control. These data demonstrate important changes in CBF and O2 metabolism after neonatal asphyxia that may be important to the pathogenesis of brain injury.

Effect of hematocrit on cerebral blood flow.

Cerebral blood flow (CBF) falls as hematocrit (Hct) rises. Investigators have differed on the relative importance of the increases in arterial O2 content (CaO2) and red blood cell concentration in mediating the fall. Our experimental protocol attempted to determine the independent effects of these two variables. In 13 unanesthetized lambs (less than 7 days old) we measured arterial and sagittal sinus blood gases, and O2 contents, and CBF (microsphere technique) at oxyhemoglobin Hcts of approximately 20 and 40% and after an isovolemic exchange transfusion with a mixture of normal and pure methemoglobin (MHb) containing red cells. Following MHb exchange, Hct rose (19.7 +/- 0.3 vs. 38.2 +/- 0.4%, mean +/- SEM) with little change in CaO2 (9.3 +/- 0.2 vs. 10.0 +/- 0.3 vol%). Arterial PCO2, pH, mean arterial blood pressure, and cerebral O2 consumption (CMRO2) did not change. However, CBF fell (153 +/- 11 vs. 110 +/- 7 ml . 100 g-1 . min-1). CBF declined further when CaO2 rose (17.3 +/- 0.5 vol%) at the higher oxyhemoglobin Hct (36.9 +/- 0.8%). We calculated that the increase in red cell concentration accounted for 56% of the decrease in CBF that ordinarily occurs as Hct rises from 20 to 40%. The effect of red cell concentration on CBF varied among individual animals. It correlated closely (r = -0.77) with the initial cerebral fractional O2 extraction [E = CMRO2/(CBF X CaO2)]. Animals with the most luxuriant O2 supply (CBF X CaO2) relative to demand (CMRO2) had the greatest decrements in CBF as red blood cell concentration rose.

165 CEREBRAL BLOOD FLOW AUTOREGULATION WITH ELEVATED INTRACRANIAL PRESSURE DURING NORMOXIA AND HYPOXIA

We assessed the ability to maintain cerebral blood flow (CBF) when intracranial pressure (ICP) is raised under normoxic and isocapnic hypoxic conditions in pentobarbital-anesthetized neonatal lambs (3-9 days old). ICP was increased by infusion of artificial CSF in the lateral ventricle to produce 8 mmHg stepwise decrements in cerebral perfusion pressure (CPP) from baseline (&asymp 66 mmHg) down to a CPP of approximately 26 mmHg. In one group (n=6) of normoxic lambs (arterial O2 content (CaO2) = 14.8 ± .7 ml/dl), CBF (ml·min “100g−1; microspheres) was unchanged from baseline (49 ± 7) down to a CPP of 34 mmHg (47 ± 6). At a CPP of 26 mmHg, CBF was decreased to 37 ± 5. In another group (n=6) of hypoxic lambs (CaO2 = 7.6 ± .9 ml/dl; 49 ± 4% arterial O2 saturation) baseline CBF (87 ± 11) was nearly twice that of the normoxic group. CBF was not significantly changed down to a CPP of 34 mmHg (78 ± 8), but was decreased at a CPP of 26 mmHg (73 ± 8). Cerebral fractional O2 extraction (measured at sagittal sinus) was also maintained down to a CPP of 34 mmHg in both groups before it rose at 26 mmHg. Cerebral O2 uptake was not different between groups and was not diminished with elevations of ICP in either group. We conclude that a) neonatal lambs are capable of CBF autoregulation with increasing ICP, and b) hypoxia sufficient to double baseline CBF but not diminish O2 uptake does not impair CBF autoregulation.

1416 VISCOSITY REGULATES CEREBRAL BLOOD FLOW INDEPENDENTLY OF ARTERIAL OXYGEN CONTENT

Cerebral blood flow (CBF) falls as hematocrit (Hct) rises. Investigators have differed on the relative importance of the increases in arterial oxygen content (CaO2) and blood viscosity (η) in mediating the fall in CBF. In a previous abstract (Pediatr. Res. 18:377A, 1984) we reported preliminary work in newborn lambs and concluded that the increase in η as Hct rose from 20% to 40% accounted for 45% of the decrease in CBF. Subsequent studies have confirmed this finding. In 18 lambs we performed an isovolemic exchange transfusion with pure methemoglobin containing adult sheep red cells and raised Hct (19.8±0.3 vs 39.0±0.5%; X±SEM) and η (2.6±0.1 vs 4.3±0.2 centipoise: shear rate=230/sec) with minimal increase in CaO2 (9.3±0.2 vs 10.2±0.2 vol%). Other variables known to influence CBF (PaCO2, pH, P50, mean arterial blood pressure) did not change. Similarly, cerebral O2 consumption (CMRO2) was constant. The fall in CBF (microsphere technique: 141±8 vs 103±5 ml/100g/min) must be attributed to the increase in η alone. On the basis of 18 lambs we calculate that the increase in η accounts for 64% of the decrease in CBF as Hct rises from 20% to 40%. The effect of η varied widely among animals, but correlated closely (r = 0.81) with the baseline cerebral fractional O2 extraction E = [CMRO2/(CBFxCaO2)]. Animals with the most luxuriant oxygen supply (CBFxCaO2) relative to demand (CMRO2) had the greater decrements in CBF.

164 EFFECT OF ELEVATED INTRACRANIAL PRESSURE ON CERE BRAL BLOO D FLOW AND EVOKED POTENTIAL RESPONSES

We studied the relationship of cerebral blood flow (CBF) and cerebral O2 uptake (CMRO2) to somatosensory evoked potential (SEP) and brainstem auditory evoked response (BAER) under conditions of elevated intracranial pressure (ICP). Sheep were anesthetized with pentobarbital and pancuronium and ventilated. ICP was increased to a fixed level by infusing mock CSF into the lateral ventricle. ICP was raised to a calculated cerebral perfusion pressure of either 0, 20-25, or 50 mmHg. CBF was measured using the radiolabelled microsphere technique. CMRO2 was calculated with sagittal sinus blood samples. When CBF fell, cerebral O2 extraction increased. However, with CBF below 70% of baseline, CMRO2 was not sustained by increased extraction. BAER interwave I to V latency increased below a mid-brain blood flow of 15 ml·min−1. 100g−1. SEP central conduction time (CCT) determined from the latency differences between N1 of foreleg SEP and C2 increased below a CBF threshold of 15-20 ml·min−1·100g−1 (50-65% reduction from baseline CBF). Changes in CCT were associated with a 25% decrease in CMRO2 from baseline. Therefore, under conditions of elevated ICP, cerebral ischemia as defined by CMRO2 appears to correlate with changes in evoked potential responses. The large threshold observed when evoked potentials are related to regional CBF is probably a function of the O2 extraction reserve.

EFFECTS OF NaHCO3 ON CEREBRAL BLOOD FLOW (CBF) DURING HYPOXIA AND ACIDOSIS IN NEWBORN PIGLETS

A reciprical relationship exists between CBF and arterial O2 content (CaO2) during hypoxic hypoxia. Prolonged hypoxia may be complicated by metabolic acidosis further decreasing CaO2. To investigate if NaHCO3 lowers CBF by increasing pH and CaO2 CBF (microspheres), osmolarity, and arteriovenous differences of O2 content and blood gases were obtained in 14 ventilated piglets. Measurements were obtained during control (C), prolonged isocapnic hypoxia (50 min) to produce metabolic acidosis (H+A), and following infusion of saline (Grl, n=6) or 2mEq/kg NaHCO3 (Gr2, n=8) during hypoxia. At C pH was 7.44±.01(X±SE) in Grl and 7.41±.01 in Gr2. H+A resulted in similar reductions in pH (Grl=7.21±.02 and Gr2=7.21±.03), but differences occurred after the infusions, (Grl=7.15±.03 vs. Gr2=7.30±.03, p<.05). In each Gr CaO2 paralleled pH but did not differ between Gr. CBF (ml/min·100g) in Grl was 120±17 at C, increased to 287±60* during H+A and 259±31* during saline. Similarly, CBF rose in Gr2 from C of 116±12 to 251±23* during H+A, and 221±34* with NaHCO3. Although cerebral O2 delivery, extraction, and uptake changed (due to the extent of CaO2 reduction), both Gr were comparable. No changes in osmolarity occurred. It appears that the magnitude of change in pH induced by 2 mEq/kg NaHCO3 does not alter CBF. It is speculated that even larger increases in pH produced by NaHCO3 may not affect CBF due to changes in O2 affinity. (*p<.05 vs. control).

PATTERN OF CEREBRAL REPERFUSION FOLLOWING BRAIN ISCHEMIA

Systemic hypotension in adults results in hypoxic-ischemic injury to the brain, which may be exacerbated by the post-ischemic pattern of cerebral reperfusion. Therefore, we studied cerebral O2 delivery (OD), uptake (CMRO2), and extraction (OE) following nemorrhagic, transient global ischemia in 12 ventilated newborn piglets. We measured cerebral blood flow (CBF) and cardiac output (CO) with microspheres and arteriovenous differences of O2 content and blood gases during control (C), after 15 min of ischemia, and 10 and 90 min after rapid reperfusion (R) with whole blood. Perfusion pressure (mmHg) fell from 81±2 (±SE) to 31±2* during ischemia, and was 80±3 and 77±2 after R. Hemodynamic and metabolic data are:

Role of O2-hemoglobin affinity on cerebrovascular response to carbon monoxide hypoxia.

Our previous studies showed that, in contrast to hypoxic and anemic hypoxia, CO hypoxia increased cerebral O2 delivery and decreased cerebral fractional O2 extraction. These changes were correlated with the accompanying decrease in P50 (PO2 at 50% saturation of non-CO bound sites on hemoglobin). To assess directly the role of P50 in the cerebrovascular response to CO, we first performed isovolemic exchange transfusions on unanesthetized newborn lambs, replacing their high-O2-affinity hemoglobin with low-affinity adult sheep donor blood. Exchange transfusion resulted in an average increase in P50 of 10 Torr and in a uniform decrease of regional cerebral blood flow and cerebral O2 delivery of 14%. Thus shifts in P50 can produce cerebrovascular changes during normoxia, implying that the mechanism regulating cerebral blood flow does not have a discrete threshold to an hypoxic stimulus. Induction of CO hypoxia (20-40% carboxyhemoglobin) after the exchange transfusion returned P50 to the control level, and with it restored both cerebral O2 delivery and fractional O2 extraction to the pretransfusion values. We conclude that the fall in P50, rather than a direct tissue effect of CO, is responsible for the relative cerebral overperfusion during CO hypoxia. The importance of the position of oxyhemoglobin dissociation curve as a determinant of cerebral blood flow supports the presence of a highly sensitive, tissue O2-dependent mechanism regulating the cerebral circulation.

256 HYPOXIC HYBOXIA (HH) AND CEREBRAL O2 DELIVERY IN THE FETAL AND NEWBORN SHEEP

The effects of HH on cerebral O2 delivery and on the ratio between O2 delivery and cerebral O2 consumption, i.e. the cerebral fractional O2 extraction, has been studied in 20 unanesthetized fetal sheep in utero and 7 newborn lambs. Cerebral O2 delivery was calculated as the product of cerebral blood flow (CBF), measured with the radioactive microsphere technique, and arterial O2 content (CaO2). Fetuses were studied as PO2 varied from 12 to 35mmHg; lambs, as PaO2 varied from 30 to 150mmHg. All data were corrected to a constant arterial PCO2. Neither cerebral O2 delivery nor fractional O2 extraction varied with PaO2 (unless hypoxia was extreme) in either fetuses or lambs, despite wide differences in their physiologic circumstances. This suggests that the regulated variable during HH is (CBFX CaO2), not CBF. Since PaO2 and CaO2 are not linearly related, CBF must respond to changes in CaO2 rather than PaO2. This indicates regulation of cerebral blood flow at the tissue level. Cerebral fractional O2 extraction is relatively constant between fetuses and lambs despite considerable differences in cerebral O2 consumption and delivery. Thus, the most important determinant of cerebral O2 delivery is not developmental stage, but O2 need.