Sage Journals: Discover world-class research

Abstract

Hyperbaric oxygen (HBO₂) increases oxygen tension (PO₂) in blood but reduces blood flow by means of O₂-induced vasoconstriction. Here we report the first quantitative evaluation of these opposing effects on tissue PO₂ in brain, using anesthetized rats exposed to HBO₂ at 2 to 6 atmospheres absolute (ATA). We assessed the contribution of regional cerebral blood flow (rCBF) to brain PO₂ as inspired PO₂ (PiO₂) exceeds 1 ATA. We measured rCBF and local PO₂ simultaneously in striatum using collocated platinum electrodes. Cerebral blood flow was computed from H₂ clearance curves in vivo and PO₂ from electrodes calibrated in vitro, before and after insertion. Arterial PCO₂ was controlled, and body temperature, blood pressure, and EEG were monitored. Scatter plots of rCBF versus pO₂ were nonlinear (R² = 0.75) for rats breathing room air but nearly linear (R² = 0.88–0.91) for O₂ at 2 to 6 ATA. The contribution of rCBF to brain PO₂ was estimated at constant inspired PO₂, by increasing rCBF with acetazolamide (AZA) or decreasing it with N-nitro-l-arginine methyl ester (l-NAME). At basal rCBF (78 mL/100 g min), local PO₂ increased 7- to 33-fold at 2 to 6 ATA, compared with room air. A doubling of rCBF increased striatal PO₂ not quite two-fold in rats breathing room air but 13- to 64-fold in those breathing HBO₂ at 2 to 6 ATA. These findings support our hypothesis that HBO₂ increases PO₂ in brain in direct proportion to rCBF.

Keywords

brain cerebral blood flow hyperbaric oxygenation tissue oxygen tension

Introduction

The O₂ content of blood is the mathematical product of hemoglobin concentration and arterial hemoglobin O₂ saturation, the latter being a nonlinear function of arterial oxygen partial pressure (PaO₂). The small amount of O₂ dissolved in plasma is usually negligible. However, breathing hyperbaric oxygen (HBO₂), that is oxygen at pressures greater than 1 atmosphere absolute (ATA), raises PaO₂ beyond the point at which hemoglobin is fully saturated, so that the dissolved fraction becomes the main source of O₂ available to cells. But this does not assure enhanced O₂ delivery to brain, because tissue PO₂ also depends on regional blood flow.

Tissue oxygen tension (PO₂) is a dynamic balance between O₂ delivery and consumption. Many authors have reported increased brain PO₂ in HBO₂ (Jamieson and Van Den Brenk, 1962; Bennett, 1965; Bean et al, 1971; Torbati et al, 1978; Hunt et al, 1978), and reviews are available (Jamieson, 1989; Camporesi et al, 1996; Dean et al, 2003). It is also known that total or regional cerebral blood flow (rCBF) decreases in HBO₂ as a function of pressure and time. But the contribution of CBF to brain PO₂ had never been quantified. Cerebral vasoconstriction and decreased total or rCBF have been shown in healthy volunteers and patients breathing O₂ at 3.5 ATA for brief periods (Lambertsen et al, 1953; Visser et al, 1996; Omae et al, 1998). In animals, in which HBO₂ is maintained for longer times and at higher pressures, the rCBF response is biphasic: the initial decrease in rCBF is followed by a secondary rise to control levels and above (Bean et al, 1971; Torbati et al, 1978; Demchenko et al, 2000a), which heralds the onset of O₂-induced EEG spikes or convulsions (Bean et al, 1971; Torbati et al, 1978; Chavko et al, 1998; Demchenko et al, 1998, 2000a; Sato et al, 2002) and may be important in the etiology of central nervous system (CNS) oxygen toxicity.

In our previous studies in mice and rats at 5 ATA, rCBF first fell as brain PO₂ rose modestly, but rCBF increased after 40 to 60 mins accompanied by a dramatic rise in brain PO₂ (Demchenko et al, 2001; Atochin et al, 2003). This led us to attempt quantitative measurements of the contribution of rCBF to changes in brain PO₂ during HBO₂, which can only be assessed in the same brain region, because CBF and brain PO₂ may differ in adjacent microregions (see Kuschinsky and Paulson, 1992; Lübbers et al, 1994). We used collocated hydrogen clearance and oxygen microelectrodes to measure these two variables.

Our goals were to: (a) surmount the technical challenges of measuring CBF and PO₂ simultaneously in the same microregion of brain in HBO₂; (b) to assess the contribution of inspired PO₂ (PiO₂) to brain PO₂ in HBO₂; (c) quantify the rCBF contribution to brain PO₂ in HBO₂; (d) test our hypothesis that CBF determines brain PO₂ in HBO₂; and (e) to mathematically model the rCBF contribution to brain PO₂.

Materials and methods

Animal Preparation

Adult, male Sprague—Dawley rats (Charles River, NC, USA n = 87) weighing 304±15 g were used as approved by the Duke University Institutional Animal Care and Use Committee, as described (Demchenko et al, 1998). Anesthesia was induced with urethane (1.2 g/kg, intraperitoneal), the trachea was intubated and both femoral arteries and one femoral vein were catheterized to measure blood pressure, take samples, and infuse drugs. Animals were secured in a stereotaxic frame and ventilated with 30% O₂ (balance N₂). The skull was drilled over the left and right caudate putamen 0.5 mm anterior and 2.5 mm lateral/medial to bregma (Paxinos and Watson, 1986). After stripping the dura mater, oxygen- and hydrogen-sensitive electrodes were inserted 5.4 mm below the brain surface and fixed in position. Anatomic locations of the electrodes were confirmed postmortem. For EEG recording, two stainless-steel screws were driven into the skull symmetrically over the left and right parietal cortex.

Fully anesthetized rats were given pancuronium bromide (0.5 mg/kg, intravenous) to prevent voluntary respiratory motion and to permit maintaining PaCO₂ at 35 to 38 mm Hg by adjusting respiratory volume. Anesthesia and immobility were maintained by intravenous urethane (0.2 g/kg h) and pancuronium bromide (0.1 mg/kgh). Adequacy of anesthesia was verified by testing corneal reflexes during surgical preparation and, after paralysis, by observing blood pressure responses to toe pinch. Control studies indicated that this regimen of supplemental anesthesia is adequate for the HBO₂ exposures.

Physiologic Measurements

Arterial blood pressure was measured continuously and integrated to obtain mean arterial blood pressure (MABP). The PaO₂, PCO₂, and pH were determined before and after HBO₂ exposure using an IL 1306 Blood Gas/pH analyzer. Rectal temperature was monitored throughout and held at 37°C±0.5°C using a heating pad. EEG was recorded continuously and EEG spikes equated with CNS oxygen toxicity.

Oxygen Tension and Cerebral Blood Flow Measurements

Brain PO₂ and rCBF were measured simultaneously and ipsilaterally in striatum using collocated electrodes (Figure 1) of platinum wire (100 μm diameter) insulated with epoxy, except for 1 mm at the tip. The electrically sharpened 10 μm tips were coated with Nafion® (Aldrich Chemical Co., Milwaukee, WI) to prevent protein fouling and were placed in the same micro-region of striatum, separated by distances from 20 to 120 μm (Demchenko et al, 1998). Oxygen electrodes were calibrated before and after each experiment in three artificial cerebral spinal fluid (ACSF) buffers at 37°C, equilibrated with: 100% nitrogen (0% O₂), air (21% O₂), and 100% O₂. Actual PO₂ in the ACSF was measured by the blood gas analyzer. If initial and final calibrations diverged by more than 10%, that experiment was not included in the series. Brain PO₂ under HBO₂ was determined by linear regression.

Figure 1

Collocated electrodes simultaneously measure regional cerebral blood flow (rCBF) and brain oxygen tension (PO₂). (A) The oxygen/hydrogen sampling regions of the paired platinum electrodes correspond to cone-like tissue volumes of approximately 0.58 mm³. The estimated volume of tissue sampled in common by the O₂ and H₂ working electrodes is approximately 0.41 mm³.(B, C) To verify the consistency of the electrode measurements, either CBF alone or PO₂ alone was measured simultaneously by both electrodes in rat striatum during hyperbaric oxygen (HBO₂) exposure at 3, 5, and 6 atmospheres absolute (ATA). Linear regression and correlation coefficients show that rCBF or PO₂ are measured in the same brain region.

To measure CBF, 2.5% H₂ in air was introduced through the respirator for 60 sec, then H₂ washout curves were captured using WINDAQ software (D-1200 AC, DATAQ Instruments). Absolute CBF (mL/100 g min) was calculated by the initial slope method using Mathematica 3.0 software (Wolfram Research) with minor modifications (Demchenko et al, 1998).

Hyperbaric Oxygen Exposures

Rats were placed in a hyperbaric chamber (volume 36 m³) with the stereotaxic frame, respirator, blood pressure transducer, heating pad, and infusion pump. Electrodes were connected through hermetic wall penetrations to amperometric amplifiers outside the chamber. After 2 h of stabilization in which the animal breathed 30% O₂ and blood gas values remained physiologic, three H₂ clearance curves were recorded to calculate control rCBF. The respirator was then supplied with 100% O₂, and the chamber atmosphere raised at 0.6 ATA per minute with air to 2, 3, 5 or 6 ATA. Hyperbaric oxygen exposures lasted 60 to 75 min while CBF was measured every 15 mins. Immediately after decompression blood gases and pH were determined as the rats continued to breathe 100% O₂.

Study Design

In a first series of experiments (1 ATA controls), we correlated CBF with PO₂ using three groups of rats breathing 30% O₂ (balance N₂) at 1 ATA (n = 10 each group). Brain PO₂ and CBF were measured at the steady state and after infusion of either the vasodilator acetazolamide (AZA, 35 mg/kg, intraperitoneal), the vasoconstrictor N-nitro-l-arginine methyl ester (l-NAME, 30 mg/kg, intraperitoneal) or saline (Demchenko et al, 2000b, 2001; Gutsaeva et al, 2004).

In a second series (HBO₂ without vasoactive drugs), CBF and PO₂ were measured in four groups of rats at 2, 3, 5, or 6 ATA (n = 8 to 10 each group) to assess the effect of PiO₂ on brain PO₂ and CBF.

In a third series (HBO₂ with vasoactive drugs), CBF versus PO₂ was studied in eight groups of rats treated with AZA or l-NAME 30 mins before exposure to 2, 3, 5, or 6 ATA (n = 8 each group) to evaluate the contribution of CBF to brain PO₂ in HBO₂.

Data from all these studies were applied to our model of O₂ diffusion at the capillary—tissue interface when hemoglobin (Hb) is fully saturated, based on the Krogh cylinder (see Appendix).

Data Analysis

Grouped data are reported as means±standard deviations (s.d.). Cerebral blood flow, brain PO₂, and physiologic variables over time were compared using ANOVA and Fisher's post hoc analysis. Paired t-tests enabled comparison of absolute and percent changes in rCBF or PO₂ in HBO₂, with or without l-NAME or AZA, with respect to their control values. Relationships between CBF and PO₂ were displayed on scatter plots of rCBF versus brain PO₂, with and without AZA or l-NAME. Differences were considered significant if P<0.05.

Results

Cerebral Blood Flow and Oxygen Tension Measurements in the Same Microregion of Striatum

Our individual working electrodes sample a cone-like tissue volume of 0.58 mm³, since a 90% oxygen diffusion field encompasses a tissue radius 6.3 times that of the cathode (Baumgartl et al, 1974). And two electrodes ~0.2 mm apart at their tips sample a shared cone of 0.43 mm³ for PO₂ and PH₂ (Figure 1A). Because the volume of the rat striatum in one hemisphere is approximately 36 mm³, the sampled region constitutes less than 2% of striatum volume. To show that CBF and PO₂ recorded from the same brain region were highly correlated, we compared measurements in which both Pt electrodes in one hemisphere recorded only PO₂ or only CBF (Figures 1C and Figure 1B).

Cerebral Blood Flow and Oxygen Tension in Control Conditions

At 1 ATA (series I, n = 31), striatal CBF was stable at 78±5 mL/100 g min (Figure 2A) and striatal PO₂ ranged from 29 to 38 mm Hg, averaging 34±5.5 mm Hg (Figure 2B), throughout 90 mins of observation in rats ventilated on 30% O₂ with arterial PCO₂ at 35 to 38 mm Hg. For rats exposed to HBO₂ (series II and III, n = 88), average initial control values for CBF at 1 ATA were 80±6 mL/100 g min and for brain PO₂ were 35±5 mm Hg. Switching to room air decreased striatal PO₂ to 25±4 mm Hg (P<0.05) without changing rCBF significantly (not shown).

In rats in series I (breathing 30% O₂ at 1ATA), treated with l-NAME or AZA, CBF changed with time. After administration of l-NAME, rCBF decreased by a maximum of 35%±14% (P<0.05) after 45 mins and then remained close to 75% of control throughout the remaining 30 mins of the experiment (Figure 2A). By contrast, 45 mins after AZA administration, rCBF increased by 50%±21% (P<0.05) and remained near this level throughout the remaining 30 mins (Figure 2A).

Figure 2

Striatal regional cerebral blood flow (rCBF) (A) and oxygen tension (PO₂) (B) responses to acetazolamide (AZA) or l-NAME are shown in rats breathing 30% O₂ (balance N₂) at 1 atmospheres absolute (ATA) or breathing 100% O₂ at 2 to 6 ATA (C, D). Blood flow and PO₂ were measured in hyperbaric oxygen (HBO₂) every 15 mins for 75 mins, and were compared with measurements in the same rats breathing 30% O₂ (balance N₂), before hyperbaric exposure. Decreases in CBF at 2 to 3 ATA for 15 to 75 mins, and at 5 ATA for the first 30 mins, are statistically significant. But in HBO₂ at 6 ATA, vasoconstriction is absent. Brain PO₂ increased significantly in all groups during compression to 2 to 6 ATA and remained high in spite of hyperoxic vasoconstriction. (*P<0.05 versus pretreatment values (A, B) or pre-exposure to HBO₂ (C, D). Data are expressed as mean±s.d.).

Changes in brain PO₂ paralleled CBF in the same brain region after administration of l-NAME or AZA (Figure 2B). Acetazolamide increased PO₂ from 36 to 45 mm Hg, but l-NAME decreased PO₂ from 35 to 24 mm Hg, both 45 mins after administration. Control physiologic variables including MABP, pH, PaO₂, and PaCO₂ were similar in all groups except for an increase in MABP after l-NAME and a slight rise in PaCO₂ after AZA (Table 1).

Table 1

Gas-exchange and mean arterial blood pressure in control rats breathing 30% O₂ (balance N₂), before and after receiving l-NAME or AZA

	Before l-NAME or AZA				75 min after l-NAME or AZA
Group	MABP	PaO₂	PaCO₂	pH	MABP	PaO₂	PaCO₂	pH
0.3 ATA (Control)	116±12	123±18	35±3	7.41±0.03	117±13	119±14	37±2	7.40±0.04
0.3 ATA (l-NAME)	113±12	113±17	35±2	7.41±0.02	131±16*	117±15	35±2	7.41±0.05
0.3 ATA (AZA)	121±14	108±14	36±3	7.40±0.02	111±14	118±14	39±3	7.39±0.06

MABP: mean arterial blood pressure (mm Hg), PaO₂ or PaCO₂ (mm Hg).

Values are means±s.d. P<0.05 versus pretreatment.

Cerebral Blood Flow and Oxygen Tension in Hyperbaric Oxygen

Cerebral blood flow responses to HBO₂ were strongly PiO₂-dependent. At 2 and 3 ATA, striatal blood flow decreased by 18%±5% (P<0.01) and 24%±13% (P<0.05), respectively, over 30 mins of HBO₂ and then remained near these levels until decompression (Figure 2C). In these animals, no EEG evidence of O₂ toxicity was observed throughout the 75 mins exposure. However, after 15 mins at 5 ATA, CBF decreased 34%±11% (P<0.01), then gradually rose and by 45 mins returned to pre-exposure levels, and over the next 30 mins increased by 81%±31% (P<0.01) above initial control values (Figure 2C). Paroxysmal EEG spikes exceeding 100 μV were observed at a mean time of 62±7 mins after initiating HBO₂ and always followed increased rCBF. Hyperbaric oxygen at 6 ATA produced a different pattern: striatal blood flow rose immediately, reaching 128%±47% (P<0.01) of control after 45 mins (Figure 2C).

Brain PO₂ responses to HBO₂ also depended on PiO₂ and time. At 1 ATA, ventilation with 100% O₂ increased mean striatal PO₂ to 98±7 mm Hg, whereas during steady-state HBO₂ at 2 or 3 ATA, brain PO₂ reached stable values of 181 to 199 and 255 to 319 mm Hg, respectively (Figure 2D). During compression to 5 ATA, brain PO₂ rose, despite vasoconstriction, to an average of 554±106 mm Hg, and when vasoconstriction was abolished and CBF rebounded to pre-exposure levels, PO₂ increased to 849±141 mm Hg. After 45 mins, PO₂ climbed dramatically to more than 1,500 mm Hg, and EEG spikes appeared at 900 to 1000 mm Hg. Hyperbaric oxygen at 6 ATA produced even greater changes: striatal PO₂ reached 836±161 mm Hg after compression and after the next 45 mins exceeded 2,000 mm Hg (Figure 2D). Because EEG spikes appeared at 37±8 mins, O₂ exposure at 6 ATA was limited to 60 mins.

Figure 3 illustrates the effect of the vasoconstrictor l-NAME on striatal blood flow and PO₂ during HBO₂ at 3, 5, or 6 ATA. After 30 mins of HBO₂, CBF decreased by 32%±6% (P<0.01) at 3 ATA, 36%±11% (P<0.01) at 5 ATA, and 19%±11% (P<0.05) at 6 ATA. These lower rCBF levels persisted until decompression, except at 6 ATA, where CBF returned to control levels (Figure 3A). Hyperbaric oxygen-induced EEG spikes were only observed in 2 of 8 l-NAME-treated rats exposed to 6 ATA. In these rats, brain PO₂ rose in proportion to inspired oxygen pressure, but during HBO₂, striatal PO₂ at equilibrium was maintained in the ranges of 238 to 301 mm Hg at 3 ATA and 455 to 625 mm Hg at 5 ATA (Figure 3B). During the first 45 mins at 6 ATA, PO₂ values were from 691 to 825 mm Hg, and over the next 30 mins rose above 1,000 mm Hg (Figure 3B). After AZA, HBO₂ augmented both CBF and PO₂ without cerebral vasoconstriction (Figure 3C). Cerebral blood flow increased by 32% to 62% at 2 ATA, by 23% to 62% at 3 ATA and by 16% to 99% at 5 ATA (Figure 3C). Brain PO₂ was 283 to 346 mm Hg at 2 ATA, and 402 to 499 mm Hg at 3 ATA. At 5 ATA, PO₂ varied in the first 15 mins from 876 to 993 mm Hg but then rose to approximately 1,400 mm Hg (Figure 3D).

Figure 3

Effects of hyperbaric oxygen (HBO₂) on striatal cerebral blood flow (CBF) and oxygen tension (PO₂) are shown in rats pretreated with l-NAME (A, B) or AZA (C, D) administrated 30 mins before hyperbaric exposure. Cerebral blood flow and PO₂ were measured in HBO₂ for 75 mins and compared with initial control CBF and PO₂ from the same rats breathing 30% O₂ (balance N₂)at 1 ATA. (*P<0.05 compared with CBF or PO₂ values before l-NAME or AZA administration. Data are expressed as mean±s.d.).

In all HBO₂-exposed animals, baseline MABP, pH, PaO₂, and PaCO₂ values were similar (Table 2). After HBO₂ exposure, arterial PaCO₂ and pH were slightly higher than initial values, but well within the physiologic range for rats. The PaO₂ was greater than 400 mm Hg for animals breathing 100% O₂ at 1 ATA, before and after HBO₂, indicating that normal pulmonary gas exchange was preserved and very high arterial PO₂ values were attained when blood gases could not be measured during the HBO₂ exposures. Arterial blood pressure in rats exposed to 2, 3, 5, or 6 ATA rose by 17% to 31% during compression and the first few minutes of HBO₂ and increased again immediately before or with the appearance of EEG spikes.

Table 2

Gas-exchange and mean arterial blood pressure in rats exposed to oxygen at 2 to 6 ATA

Group	Before HBO₂ exposure				After HBO₂ exposure
	MABP	paO₂	PaCO₂	pH	MABP	paO₂	PaCO₂	pH
2 ATA	115±14	110±17	35±2	7.41±0.04	117±14	453±34*	38±3	7.40±0.04
2 ATA (AZA)	122±15	120±22	37±4	7.41±0.04	109±13	423±34*	43±4	7.39±0.04
3 ATA	118±12	120±17	36±3	7.40±0.03	122±17	463±34*	37±3	7.41±0.05
3ATA (l-NAME)	109±9	117±17	36±2	7.40±0.04	129±19*	447±34*	39±4	7.39±0.04
3 ATA (AZA)	117±11	121±24	37±3	7.41±0.02	107±12	456±34*	44±5*	7.38±0.04
5 ATA	116±12	103±17	38±5	7.39±0.06	119±13	433±34*	41±5	7.39±0.04
5 ATA (l-NAME)	112±14	104±13	36±3	7.40±0.05	133±35*	476±55*	41±4	7.39±0.04
5 ATA (AZA)	120±10	119±19	35±3	7.42±0.02	115±21	441±55*	45±5*	7.36±0.06
6 ATA	127±15	117±16	37±3	7.39±0.05	131±18	415±65*	44±5*	7.37±0.06
6 ATA (l-NAME)	119±13	99±15	37±5	7.40±0.03	137±29*	421±74*	42±3	7.39±0.05

MABP: mean arterial blood pressure (mm Hg); PaO₂ and PaCO₂ (mm Hg).

Values are means±s.d.P<0.05 compared with pre-HBO₂ exposure.

Cerebral Blood Flow—Oxygen Tension Correlation During Hyperbaric Oxygen

Figure 4 shows scatter plots of PO₂ versus rCBF measured simultaneously in the same brain region in 30% O₂ at 1 ATA and in 100% O₂ at 2, 3, 5, and 6 ATA, with and without l-NAME or AZA. In 30% O₂, the relationship of absolute CBF to PO₂ was nonlinear with a coefficient of determination of 0.75. By contrast, the scatter plots at 2, 3, 5 and 6 ATA approached linearity, with coefficients of determination of 0.88 to 0.91. In 30% O₂, linear regression indicates that a doubling of rCBF (from 80 to 160 mL/100 g min) effects a change in PO₂ of 17 mm Hg, whereas in HBO₂ at 2, 3, 5, or 6 ATA the same increase in CBF elevates brain PO₂ by 139, 315, 657, or 770 mm Hg, respectively, an increase of 13- to 64-fold relative to normobaric normoxia.

Figure 4

Relationships between absolute CBF and PO₂ in rats exposed to O₂ at 0.3 to 6 ATA were obtained from the data presented in Figures 2 and Figure 3 using the linear regressions and determination coefficients shown.

Simulation of Cerebral Blood Flow—Oxygen Tension Relationships

A Krogh-cylinder model—a cylindrical blood vessel surrounded by a cylinder of perfused tissue (Figure 6 and Appendix-)–simulates CBF—PO₂ relationships for variations in HBO₂ or blood velocity (BV). With a BV assumed between 250 to 1000 μm/sec, tissue cylinder PO₂ in the model varies almost linearly with PiO₂ between 2 and 6 ATA (Figure 5A). The model's predictions were compared with real data from direct measurements of CBF and brain PO₂. The modeled HBO₂-dependent changes in brain PO₂ were very similar to our experimentally determined results (Figure 5B).

Figure 5

Modeled simulations (A) of inspired PO₂ (PiO₂) versus brain PO₂ at different levels of blood perfusion are compared with measured data (B). With blood velocity (BV) assumed between 250 and 1000 μm/sec, modeled tissue cylinder PO₂ varied almost linearly with PiO₂ in the 2 to 6 ATA range and was similar to our experimentally determined results. Modeled simulations (C) of the effect of blood flow on brain PO₂ at different levels of blood perfusion are compared with measured data (PiO₂). The effect of blood flow on brain PO₂ is shown at different levels of inspired PO₂. Simulations of BV versus tissue PO₂ are close to linear when BV changed in the range of 250 to 800 μm/sec. Increases in BV up to 1,500 μm/sec caused a smaller, nonlinear increase in tissue cylinder PO₂. (D) Experimentally determined CBF—PO₂ data plotted in higher order polynomial solutions reveal overall similarities to the model.

Figure 6

Diagram of a simple capillary/tissue oxygen diffusion model based on Krogh geometry. (See Appendix for explanation.)

Our simulations of BV versus tissue PO₂ (Figure 5C) approached linearity for BV from 250 to 800 μm/sec. However, further increases up to 1500 μm/sec caused smaller, nonlinear increases in tissue cylinder PO₂. Our experimentally determined CBF—PO₂ data fitted to higher order polynomial solutions (Figure 5D) revealed overall similarities to the model. When brain PO₂ values, experimentally determined at various values of CBF, were scaled as percent changes, linear approximations showed a CBF/brain PO₂ ratio of ~1: 0.9 for HBO₂ exposures of 2 to 6 ATA. And modeled BV—PO₂ relationships showed overall similarity to the experimental data (not shown).

Discussion

We have quantified CBF—PO₂ relationships in HBO₂ to a degree not previously achieved. Moreover, by comparing our experimental data to a mathematical model of O₂ transport from blood to tissue, the CBF dependence of brain PO₂ is accurately predicted when PiO₂ exceeds 2 ATA. However, several additional points need to be discussed.

Advantages and Limitations of Cerebral Blood Flow—Oxygen Tension Measurements in the Same Brain Region

Under physiologic conditions, blood flow in the rat brain is distributed heterogeneously and may differ in adjacent microregions (Leniger-Follert and Lübbers, 1976; Kuschinsky and Paulson, 1992; Iadecola et al, 1994). Because CBF affects both O₂ delivery and brain tissue oxygenation, quantitative CBF—PO₂ relationships must be based on the absolute values of these parameters in the same brain region. For this reason, we selected electrochemistry—which permits highly localized measurements—to record regional blood flow and brain PO₂ using paired microelectrodes inserted together in close proximity.

We chose the striatum for our study because our previous work showed that its vascular responses to HBO₂ are pronounced and because it is vulnerable to HBO. Electrophysiologic studies by others confirm that bioelectrical abnormalities take place in the striatum and then spread to reticular and cortical regions (Rucci et al, 1967). And morphologic studies have shown that cortical areas are not affected by the neuronal necrosis found mainly in striatum and subcortical structures of post-HBO convulsed brains (Balentine, 1982). In addition, electrode measurements can be made more precisely in striatum because it is a relatively large structure.

For O₂ polarography, Baumgartl et al (1974, 2002) and Lübbers et al (1994) reported that 90% of the O₂ signal is produced in a tissue volume whose radius is 6.3 times that of the electrode. Because diffusion coefficients and solubility for O₂ and H₂ are so similar (Young, 1980), this can also be applied to the hydrogen electrode. Our electrodes sample from calculated tissue volumes of 0.58 mm³. The common tissue volume sampled by the O₂ and H₂ electrode pair is about 0.41 mm³ (see Figure 1). From data of Laursen and Diemer (1980), we calculate that the striatum contains about 2,500 microvessels/mm³. Therefore, our electrodes measure CBF and PO₂ in a region containing ~1,000 microvessels. We confirmed that CBF and PO₂ are recorded from the same brain region in initial experiments in which both Pt electrodes in one hemisphere recorded only PO₂ or only CBF (Figures 1C and Figure 1B).

In the PO₂ distributions around microvessels found by others using polarographic microelectrodes (typically <25 μm in diameter), values near zero coexist with those near that of arterial blood (Feng et al, 1988; Lübbers et al, 1994; Erecinska and Silver, 2001). Our 100 μm diameter electrodes however, with conical sensing surfaces of about 0.9 mm², integrate PO₂ measurements from all sites in a tissue volume of 0.58 mm³. Therefore, in rats ventilated with 30% O₂ we found a tighter range of basal PO₂ (29 to 38 mm Hg).

Control rCBF and rCBF measured after l-NAME and AZA accord with earlier estimates from anesthetized rats but are less than found in conscious animals (Ohata et al, 1981; Fellows and Boutelle, 1993). In our study, 7% of the rats had low resting CBF in striatum (<40 mL/100 g min) and weak CBF responses to HBO₂, l-NAME, or AZA, which we attributed to small-vessel damage found postmortem. Trauma from puncturing is a disadvantage of needle electrodes, and these animals were excluded from the statistical analysis. Such damage was minimized by carefully shaping the platinum tip.

Temporal Profile of Cerebral Blood Flow and Oxygen Tension During Hyperbaric Oxygen

Absolute CBF and brain PO₂ have been recorded in HBO₂ in only a few studies because of the technical difficulties involved. Data from various animal species show that brain PO₂ increases similarly with blood oxygen content at sea level, but show marked variations during HBO₂ exposure, even under the same conditions. For example, Jamieson and Van Den Brenk (1962) report average cortical PO₂ values in anesthetized rats of 452, 917, and 1293 mm Hg in HBO₂ at 3, 5, and 6 ATA, respectively. However, other authors have published mean values in rat striatum of 253 to 480 mm Hg (maximums were 490 to 1,010 mm Hg) during 60-mins HBO₂ at 5 ATA (Hunt et al, 1978). Individual cortical PO₂ measurements as high as ~2,000 mm Hg have been described in anesthetized rats at 2.8 ATA (Thom et al, 2002).

Reasons for these discrepancies have been discussed previously (Dean et al, 2003) and include electrode size and location as well as proximity to blood vessels: near the external vessel wall of precapillary cerebral arterioles, tissue PO₂ approaches that of arterial blood (Duling et al, 1979; Ivanov et al, 1999; Vovenko, 1999). But we think the most important source of discrepancies is that PO₂ and CBF have not been quantatively measured simultaneously and collocally. Indeed, striatal PO₂ in rats we exposed to HBO₂ at 3 ATA ranged between 255 and 319 mm Hg when vasoconstriction was observed, but increased by 23% to 62% after AZA-induced increases in CBF.

At 5 ATA, striatal PO₂ was 554 mm Hg during oxygen-induced cerebral vasoconstriction at the onset of HBO₂; but when CBF returned to control levels, PO₂ was 849 mm Hg and then when CBF rose by 81%, PO₂ increased to 1,600 mm Hg. By contrast, l-NAME decreased CBF by 35% at 5 ATA and depressed brain PO₂ below 450 mm Hg (see Figure 3).

Both here and in our previous work (Demchenko et al, 2000b, 2001), simultaneous measurements of rCBF and PO₂ in the same brain region in HBO₂ show that PO₂ follows blood flow in a time-dependent manner. At 3 and 5 ATA, we found that CBF in the striatum fell and remained stable for 75 mins at 3 ATA, and for at least 30 mins at 5 ATA. Cerebral vasoconstriction has been shown using different methods in anesthetized and awake animals after relatively brief periods of HBO₂ (Bean et al, 1971; Torbati et al, 1978; Bergo and Tyssebotn, 1992; Demchenko et al, 1998 2000a). Vasoconstriction presumably protects the brain from excess tissue oxygenation, and is in part related to interruption of NO⁻ release from S-nitrosohemoglobin (Stamler et al, 1997) and to inactivation of NO⁻ by superoxide anion (Demchenko et al, 2000b). Indeed, the temporal profile of CBF responses to HBO₂ at 5 ATA are biphasic: striatal blood flow initially decreases but then returns to or rises above control levels. Similar rCBF responses have been found in rats at 5 or 6 ATA by others (Bean et al, 1971; Torbati et al, 1978; Chavko et al, 1998).

Apparent exceptions to hyperoxic vasoconstriction in rats exposed to HBO₂ have been observed in laser Doppler measurements of cortical blood flow (Zhang et al, 1995; Sato et al, 2002; Thom et al, 2002). Laser-Doppler probes determine red cell velocity, not blood flow, and these two parameters may not correlate well during vasoconstriction in extreme hyperoxia. And again, the timing of these measurements is important, because initial declines in CBF due to vasoconstriction may be quite brief, whereas secondary increase in CBF above basal levels at 4 ATA or above may be protracted (Demchenko et al, 2000b, 2003).

Contribution of inspired Oxygen Tension and Cerebral Blood Flow to brain Oxygen Tension

When hemoglobin is fully saturated with oxygen and O₂ delivery greatly exceeds the cerebral metabolic rate for O₂ (CMRO₂), only three factors contribute to brain PO₂: PaO₂ (which with normal lungs approaches PiO₂), CBF, and changes in cerebral metabolic rate for O₂. By assuming constant CMRO₂, we obtained data to estimate the importance of the other two factors. Cerebral metabolic rate for O₂ is stable at HBO₂ pressures of at least 3 ATA (Clark and Thom, 2003). Although rats exposed to HBO₂ at 5 ATA show elevations in cerebral metabolic rate for glucose (CMRgl), a correlate of CMRO₂ (Torbati et al, 1982), this reaches statistical significance only in the immediate preconvulsive period (Torbati and Lambertsen, 1983). To alter CBF, we used the well-known vasoactive NOS inhibitor l-NAME as well as the carbonic anhydrase inhibitor AZA. Although these compounds decrease CMRO₂ slightly (Iadecola et al, 1994), we believe that they do not significantly influence CBF—PO₂ relationships in extreme hyperoxia, because both the availability and delivery of extracellular O₂ are very high.

We estimated the effect of inspired O₂ on brain oxygenation by comparing striatal PO₂ in rats at different HBO₂ pressures but at a constant rCBF of 80 mL/100 g min, because resting CBF values measured in different groups of rats were in the range of 76–83 mL/100 g min in this and other studies (Ohata et al, 1981; Fellows and Boutelle, 1993; Lowry et al, 1997; Demchenko et al, 2000b). At this rCBF, using the regression equations of Figure 4, we calculated striatum PO₂ values of 357, 694, and 831 mm Hg in HBO₂ at 3, 5, and 6 ATA, respectively, yielding ratios of inspired PO₂ to brain PO₂ of 6.3, 5.5, and 5.5.

To compute the ratio of arterial to brain PO₂, we need accurate measurements of arterial PO₂ in rats. In this and our previous studies, arterial PO₂ in rats ventilated with 100% O₂ at 1 ATA were in the range of 420 to 510 mm Hg (Demchenko et al, 2000b). Preliminary data for this study found PaO₂ values up to 1497±75 mm Hg in HBO₂ at 3 ATA (unpublished data). Actual arterial PO₂ in rats exposed to more than 3 ATA apparently has not been determined, but anesthetized cats ventilated with oxygen at 1, 2, 3, 4, 5, and 6 ATA had average PaO₂ values of 421, 1,003, 1,561, 2,247, 2,730 and 3,206 mm Hg, respectively (Selivra, 1979). Assuming similar values, ratios of PaO₂ to brain PO₂ (PaO₂/PbO₂) are 4.2, 3.6, and 3.9 for HBO₂ at 3, 5, and 6 ATA, respectively. Hence, simple calculations suggest that brain PO₂ is related to inspired PO₂ by a factor of ~6 (with normal lungs) and to arterial PO₂ by a factor of ~4. These factors allow estimation of brain PO₂ at various levels of HBO₂, within the constraints of the CBF response, which shows individual variability in HBO₂ and is affected by injury and disease. However, the fact that we consistently observed these ratios is intriguing and suggests an interesting new avenue of research.

Again using the regression equations of Figure 4, we found that decreasing CBF by 50%, in steady-state HBO at 3, 5, or 6 ATA, decreases brain PO₂ by 43%, 46%, and 42%, respectively. And increasing CBF by 100% elevates brain PO₂ by 88%, 94%, and 98%, respectively. Thus, under these conditions, brain PO₂ is almost directly proportional to CBF.

How Does Cerebral Blood Flow Elevate Brain Oxygen Tension in Hyperbaric Oxygen?

Hyperbaric oxygen raises brain PO₂ by increasing the volume of oxygenated tissue around small vessels, establishing a steeper O₂ diffusion gradient between blood and tissue. But how an increase in CBF brings this about is unclear. We modeled CBF-dependent brain tissue oxygenation in HBO₂ using a Krogh-cylinder model (see Appendix), which reflects general rather than specific morphologic features, but allows examination of the effects of both radial diffusion and convective flux of plasma oxygen on tissue oxygen distribution (see review by Popel, 1989). Theoretical work with this model shows that the oxygen pressure field surrounding the tissue cylinder within which local PO₂ varies under hyperbaric conditions depends on both input PO₂ and BV, but in different ways. When input PO₂ is raised, tissue cylinder PO₂ increases mainly at the arterial end of the capillary, whereas increases in velocity lead to increases in PO₂ at the venous end. However, physiologic conditions that increase CBF may also elevate PO₂ at the capillary entrance, which during 100% oxygen breathing is lower than in the aorta due to precapillary loss of oxygen because the walls of precapillary arterioles are gas permeable (Duling et al, 1979; Ivanov et al, 1999; Vovenko, 1999) and because of countercurrent diffusive shunting between arterioles and venules (Popel, 1989).

Our modeling calculations also show nonlinear BV—PO₂ relationships for blood velocities over 800 μm/sec. Our in vivo studies also show nonlinear CBF—PO₂ relationships when striatal CBF is increased more than 100% in HBO₂ (Figure 5). There are at least two reasons for nonlinearity: the high blood velocity shown in the model decreases time for oxygen diffusion during capillary transit; and increases in CMRO₂ during preconvulsive neuronal activation, which despite concomitant increases in CBF, might increase O₂ extraction as suggested by Buxton and Frank (1997) and other (Mintun et al, 2001). We do not exclude the latter possibility, but the ratio of brain O₂ delivery to CMRO₂ at 5 and 6 ATA is high and any falloff in striatal PO₂ by preconvulsive activation should be small.

In summary, these studies indicate that brain oxygenation in HBO₂ is equally affected by inspired oxygen pressure and CBF. The ratio of PiO₂ to brain PO₂ is approximately 6:1 under normal physiologic conditions and does not change at constant CBF during HBO₂ in the range of 3 to 6 ATA. Our results also indicate that patterns of tissue PO₂ response are greatly affected by local alterations in CBF under HBO₂, and that PO₂ directly tracks CBF when hemoglobin is fully saturated. If brain PO₂ needs to be raised to a specific level, this will be best achieved by combining HBO₂ with agents that counteract O₂ vasoconstriction.

Footnotes

Acknowledgements

The authors are also grateful to Albert Boso and Craig Marshall for excellent technical assistance.

Appendix A

The Krogh-cylinder model (Figure 6) assumes a central capillary surrounded by a homogeneously respiring tissue cylinder. Oxygen diffuses radially outwards from capillary plasma, and its content in the blood decreases linearly from the arterial to the venous end of the capillary, provided: (a) hemoglobin is 100% saturated; (b) PO₂ distribution in the tissue cylinder is symmetrical about its axis, (c) tissue oxygen consumption is homogeneous and constant, and (d) the solubility coefficient for oxygen is constant.

Given these conditions, the radial distribution of tissue PO₂ in terms of capillary and tissue cylinder radii is expressed by the Krogh—Erlang equation

where U (x, y) is the PO₂ in the point x, y located on the radius of tissue cylinder; P(y) is the PO₂ in the point y on the capillary wall; R_c and R_t are the radii of the capillary and tissue cylinder, respectively, and equal to 3 and 30 μm according to Grunewald and Sowa (1977); D and α are the diffusion and solubility coefficients for O₂, respectively, and equal 1.7 × 10⁻⁵ cm²/sec and 3 × 10⁻⁵ mLO₂/g mm Hg (Kreuzer, 1982); and m is an O₂ consumption of 1 × 10⁻³ mLO₂/g sec (Gleichmann et al, 1962; Thews, 1968).

Equation (A.1) defines PO₂ at any point of the tissue cylinder from R_c to R_t, if PO₂ on any point of the capillary wall is known. To find the capillary wall PO₂, the following equation (A.2) was used:

Here P_o is the blood PO₂ at the arterial end of capillary, L is the capillary length of 1,000 mm (Ma et al, 1974; Seylaz et al, 1999), V is the blood velocity. By combining equations (A.1) and (A.2) it is possible to compute a PO₂ value for any point of the tissue cylinder:

But to compare experimental and modeling data, the average PO₂ in the tissue cylinder must be determined. For this, in equation (A.3), U was integrated for radius, angle and length (x, L, Ψ) of the tissue cylinder and divided by the square of cylinder cross-sectional area and length of the cylinder: (A.4)

\begin{aligned} (\int_{0}^{1} \int_{0}^{2 π} \int_{R c}^{R 1} U (x) \cdot x \cdot d x \cdot d ψ \cdot d l) / (π \cdot l \cdot (R_{t}^{2} - R_{c}^{2})) \\ = \frac{m}{D \cdot α} \cdot [\frac{3 \cdot R_{t} - R_{c}^{2}}{8} - \frac{R_{t}^{4}}{2 \cdot (R_{t}^{2} - R_{c}^{2})} \cdot \ln (\frac{R_{t}}{R_{c}})] \\ + P_{0} + \frac{m}{2 \cdot α} \cdot \frac{R_{c}^{2} - R_{t}^{2}}{R_{c}^{2}} \cdot \frac{L}{V} \end{aligned}

Equation (A.4) enables us to model the relationship between CBF and PO₂ under a range of conditions: (a) PiO₂ from 0.3 to 6 ATA, equivalent to normobaric and hyperbaric hyperoxia; (b) capillary BV from 250 to 1,000 μm/sec, simulating hyperoxic vasoconstriction or hyperemia during HBO₂, as previously reported (Ma et al, 1974; Grunewald and Sowa, 1977; Wei et al, 1993; Kleinfeld et al, 1998); (c) blood PO₂ at the arterial terminus of the capillary was assumed to be 53% of PaO₂ in the aorta, based on studies in rat brain showing that external surface PO₂ on the arterial end of the capillaries averaged 58±10 mm Hg in normoxia (Vovenko, 1999; Ivanov et al, 1999) and 154±11 mm Hg while breathing 100% oxygen (Ivanov et al, 1999). In these studies, systemic PaO₂ were 86±10 and 345±75, respectively. We also evaluated the ability of this model to realistically simulate the CBF—PO₂ relationships we found in vivo (see Results).

References

Atochin

Demchenko

Astern

Boso

Piantadosi

Huang

(2003) Contribution of endothelial and neuronal nitric oxide synthase to cerebrovascular responses to hyperoxia. J Cereb Blood Flow Metab 23:1219–26

Balentine

(1982) Pathology of oxygen toxicity. New York: Academic Press, pp 378

Baumgartl

Grunewald

Lübbers

(1974) Polarographic determination of the oxygen partial pressure field by Pt microelectrodes using the O₂ field in front of a Pt macroelectrode as a model. Pflugers Arch 347:49–61

Baumgartl

Zimelka

Lübbers

(2002) Evolution PO₂ profiles to describe the oxygen pressure field within the tissue. Comp Biochem Physiol Pat A 132:75–85

Bean

Lignell

Coulson

(1971) Regional cerebral blood flow, O₂ and EEG in exposures to O₂ at high pressure. J Appl Physiol 31:235–42

Bennett

(1965) Cortical CO₂ and O₂ at high pressures of argon, nitrogen, helium, and oxygen. J Appl Physiol 20:1249–52

Bergo

Tyssebotn

(1992) Cerebral blood flow distribution during exposure to 5 bar oxygen in awake rats. Undersea Biomed Res 19:339–45

Buxton

Frank

(1997) A Model for the coupling between cerebral blood flow and oxygen metabolism during neural stimulation. J Cereb Blood Flow Metab 17:64–72

Camporesi

Mascia

Thom

(1996) Physiological principles of hyperbaric medicine. In: Handbook of hyperbaric medicine ( Oriani

Marroni

Wattel

, eds), Berlin: Springer-Verlag, 35–58

10.

Chavko

Braisted

Ousta

Harabin

(1998) Role of cerebral blood flow in seizures from hyperbaric oxygen exposure. Brain Res 781:75–82

11.

Clark

Thom

(2003) Oxygen under pressure. In: Bennett and Elliott's physiology and medicine of diving ( Brubakk

Newman

, eds), New York: Best Publishing Company, 358–432

12.

Dean

Mulkey

Garcia

Putnam

Henderson

(2003) Neuronal sensitivity to hyperoxia, hypercapnia, and inert gases at hyperbaric pressures. J Appl Physiol 95:883–909

13.

Demchenko

Boso

Natoli

Doar

O'Neill

Piantadosi

(1998) Measurement of cerebral blood flow in rats and mice by hydrogen clearance during hyperbaric oxygen exposure. Undersea Hyperbaric Med 25:147–53

14.

Demchenko

Boso

Bennett

Whorton

Piantadosi

(2000a) Hyperbaric oxygen reduces cerebral blood flow by inactivating nitric oxide. Nitric Oxide 4:597–608

15.

Demchenko

Boso

O'Neil

Bennett

Piantadosi

(2000b) Nitric oxide and cerebral blood flow responses to hyperbaric oxygen. J Appl Physiol 88:1381–9

16.

Demchenko

Boso

Whorton

Piantadosi

(2001) Nitric oxide production is enhanced in rat brain before oxygen-induced convulsions. Brain Res 917:253–61

17.

Demchenko

Atochin

Boso

Astern

Huang

Piantadosi

(2003) Oxygen seizure latency and peroxynitrite formation in mice lacking neuronal or endothelial nitric oxide synthases. Neurosci Lett 344:53–6

18.

Duling

Kushinscky

Wahl

(1979) Measurements of the perivascular PO₂ in the vicinity of pial vessels of the cat. Pflugers Arch 383:29–34

19.

Erecinska

Silver

(2001) Tissue oxygen tension and brain sensitivity to hypoxia. Respir Physiol 128:263–76

20.

Fellows

Boutelle

(1993) Rapid changes in extracellular glucose levels and blood flow in the striatum of the freely moving rat. Brain Res 604:225–31

21.

Feng

Roberts

Jr Sick

Rosenthal

(1988) Depth profile of local oxygen tension and blood flow in rat cerebral cortex, white matter and hippocampus. Brain Res 445:280–8

22.

Gleichmann

Ingvar

Lassen

Lübbers

Siesjo

Thews

(1962) Regional cerebra cortical metabolic rate of oxygen and carbon dioxide related to the EEG in the anesthetized dog. Acta Physiol Scand 55(1):82–94

23.

Grunewald

Sowa

(1977) Capillary structures and O₂ supply to tissue. Rev Physiol Biochem Pharmacol 77:140–209

24.

Gutsaeva

Moskvin

Zhilyaev

S Yu

Kostkin

Demchenko

(2004) Contribution of nitric oxide in CO₂-induced cerebral hyperemia in rats under extreme hyperoxia. Russian J Physiology 90:428–36

25.

Hunt

Blackburn

Ogilvie

Balentine

(1978) Oxygen tension in the globus pallidus and neostriatum of unanesthetized rats during exposure to hyperbaric oxygen. Exp Neurol 62:698–707

26.

Iadecola

Pelligrino

Moskowitz

Lassen

(1994) Nitric oxide synthase inhibition and cerebrovascular regulation. J Cereb Blood Flow Metab 14:175–192

27.

Ivanov

Sokolova

Vovenko

(1999) Oxygen transport in the rat brain cortex at normabaric hyperoxia. Eur J Appl Physiol Occup Physiol 80:582–7

28.

Jamieson

Van den Brenk

(1962) Measurement of oxygen tensions in cerebral tissues of rats exposed to high pressures of oxygen. J Appl Physiol 18:869–76

29.

Jamieson

(1989) Oxygen toxicity and reactive oxygen metabolites in mammals. Free Radic Biol Med 7:87–108

30.

Kleinfeld

Mitra

Helmchen

Denk

(1998) Fluctuations and stimulus-induced changes in blood flow observed in individual capillaries in layers 2 through 4 of rat neocortex. Proc Natl Acad Sci USA 95:15741–6

31.

Kreuzer

(1982) Oxygen supply to tissues: The Krogh model and its assumptions. Experientia 38:1415–25

32.

Kuschinsky

Paulson

(1992) Capillary circulation in the brain. Cerebrovasc Brain Metab Rev 4:261–86

33.

Lambertsen

Krough

Cooper

(1953) Oxygen toxicity. Effects in man of oxygen inhalation at 1 and 3.5 atmospheres upon blood gas transport, cerebral circulation and cerebral metabolism. J Appl Physiol 5:471–86

34.

Laursen

Diemer

(1980) Capillary size, density and ultrastructure in brain of rats with urease-induced hyperammonaemia. Acta Neurol Scandinav 62:103–15

35.

Leniger-Follert

Lübbers

(1976) Behavior of micro flow and local PO₂ of the brain cortex during and after direct electrical stimulation. A contribution to the problem of metabolic regulation of microcirculation in the brain. Pflugers Arch 366:39–44

36.

Lowry

Boutelle

Fillenz

(1997) Measurement of brain oxygen at a carbon paste electrode can serve asan index of increases in regional cerebral blood flow. J Meuroscience Methods 71:177–82

37.

Lübbers

Baumgärtl

Zimelka

(1994) Heterogeneity and stability of local pO₂ distribution within the brain tissue. Adv Exp Med Biol 345:567–74

38.

Kao

Kwan

Cheng

(1974) On-line measurement of the dynamic velocity of erythrocytes in the cerebral microvessels in the rat. Microvasc Res 8:1–13

39.

Mintun

Lundstrom

Snyder

Vlassenko

Shulman

Raichle

(2001) Blood flow and oxygen delivery to human brain during functional activity: theoretical modeling and experimental data. Proc Natl Acad Sci USA 98:6859–64

40.

Ohata

Fredericks

Sundaram

Rapoport

(1981) Effects of mmobilization stress on regional cerebral blood flow in the conscious rat. J Cereb Blood Flow Metab 1:187–94

41.

Omae

Ibayashi

Kusuda

Nakamura

Yagi

Fujishima

(1998) Effects of high atmospheric pressure and oxygen on middle cerebral blood flow velocity in humans measured by transcranial Doppler. Stroke 29:94–7

42.

Paxinos

Watson

(1986) The rat brain in stereotaxic coordinates. San Diego: Academic Press, 272

43.

Popel

(1989) Theory of oxygen transport to tissue. Clin Rev Biomed Eng 17:257–321

44.

Rucci

Giretti

La Costa

(1967) Changes in electrical activity of the cerebral cortex and of some subcortical centers in hyperbaric oxygen. EEG Clin Neurophysiol 22:231–8

45.

Sato

Takeda

Hagioka

Zhang

Hirakawa

(2002) Dynamic changes in nitric oxide production and cerebral blood flow before development of hyperbaric oxygen-induced seizures in rats. Brain Res 918:131–40

46.

Stamler

Jia

McMahon

Demchenko

Bonaventura

Gernet

Piantadosi

(1997) Blood flow regulation by S-nitrosohemoglobin in the physiological oxygen gradient. Science 276:2034–7

47.

Seylaz

Charbonne

Nanri

Von Euw

Borredon

Kacem

Meric

Pinard

(1999) Dynamic in vivo measurement of erythrocyte velocity and flow in capillaries and of microvessel diameter in the rat brain by co-focal laser microscopy. J Cereb Blood Flow Metab 19:863–70

48.

Selivra

(1979) Influence of the hyperoxia on the oxygen tension in blood and cerebral tissue. Physiol J USSR 65:513–20

49.

Thews

(1968) The theory of oxygen transport. In: Oxygen transport in blood and tissue ( Lubbers

Luft

Thews

Witzleb

, eds), Stuttgart: Georg Thieme Verlag, 1–10

50.

Thom

Bhopale

Fisher

Manevich

Huang

Buerk

(2002) Stimulation of nitric oxide synthesis in cerebral cortex due to elevated partial pressure of oxygen: an oxidative stress response. J Neurobiol 51:85–100

51.

Torbati

Parolla

Lavy

(1978) Blood flow in rat brain during exposure to high oxygen pressure. Aviat Space Environ Med 49:963–7

52.

Torbati

Greenberg

Lambertsen

(1982) Correlation of brain glucose utilization and cortical electrical activity during development of brain oxygen toxicity. Brain Res 262:267–73

53.

Torbati

Lambertsen

(1983) Regional cerebral metabolic rate for glucose during hyperbaric oxygen-induced convulsions. Brain Res 279:382–6

54.

Visser

Van Hulst

Wieneke

Van Huffelen

(1996) Transcranial Doppler sonographic measurements of middle cerebral artery flow velocity during hyperbaric oxygen exposures. Undersea Hyper Med 23:157–65

55.

Vovenko

(1999) Distribution of oxygen tension on the surface of arterioles, capillaries and venules of brain cortex and in tissue in normoxia: an experimental study on rats. Pflugers Arch 437:617–23

56.

Wei

Otsuka

Acuff

Bereczki

Pettigrew

Patlak

Fenstermacher

(1993) The velocities of red cell and plasma flows through parenchymal microvessels of rat brain are decreased by pentobarbital. J Cereb Blood Flow Metab 13:487–97

57.

Young

(1980) H₂ clearance measurement of blood flow: a review of techniques and polarographic principles. Stroke 11:552–64

58.

Zhang

Sam

Klitzman

Piantadosi

(1995) Inhibition of nitric oxide synthase on brain oxygenation in anesthetized rats exposed to hyperbarc oxygen. Undersea Hyper Med 22:377–82

Cerebral Blood Flow and Brain Oxygenation in Rats Breathing Oxygen under Pressure

Abstract

Keywords

Introduction

Materials and methods

Animal Preparation

Physiologic Measurements

Oxygen Tension and Cerebral Blood Flow Measurements

Hyperbaric Oxygen Exposures

Study Design

Data Analysis

Results

Cerebral Blood Flow and Oxygen Tension Measurements in the Same Microregion of Striatum

Cerebral Blood Flow and Oxygen Tension in Control Conditions

Cerebral Blood Flow and Oxygen Tension in Hyperbaric Oxygen

Cerebral Blood Flow—Oxygen Tension Correlation During Hyperbaric Oxygen

Simulation of Cerebral Blood Flow—Oxygen Tension Relationships

Discussion

Advantages and Limitations of Cerebral Blood Flow—Oxygen Tension Measurements in the Same Brain Region

Temporal Profile of Cerebral Blood Flow and Oxygen Tension During Hyperbaric Oxygen

Contribution of inspired Oxygen Tension and Cerebral Blood Flow to brain Oxygen Tension

How Does Cerebral Blood Flow Elevate Brain Oxygen Tension in Hyperbaric Oxygen?

Footnotes

Acknowledgements

Appendix A

References