Cortical oxidative metabolism under conditions of ischemia, hypoxia, and asphyxia in the rabbit

Transcription

1 J Neurosurg 59:57-62, 1983 Cortical oxidative metabolism under conditions of ischemia, hypoxia, and asphyxia in the rabbit HIROAKI KOGA, M.D., AND GEORGE AUSTIN, M.D. Department of Neurosurgery, University of Nagasaki, Nagasaki, Japan, and California Stroke Foundation, and Department of Environmental Studies, University of California, Santa Barbara, California ~," The purpose of this investigation was to compare the effects of hypoxia, asphyxia, and ischemia on brain cortical oxidative metabolism. This study was carried out using 14 New Zealand White rabbits. The effects of episodic stress were measured simultaneously on brain functional metabolism by monitoring cortical oxygen tension (brain po2), cortical cerebral blood flow (ccbf), cortical blood volume, and mitochondrial oxidative metabolism. During hypoxia (when the fraction of inspired 02 (FiO2) was reduced to 10%) and asphyxia (induced by turning the respirator off), there was a decrease of brain po2 but an increase of ccbf and blood volume. Similarly, there was a reduction of cortical oxidative metabolism. In post-asphyxic conditions, an overshoot of brain po2 and post-asphyxic oxidation of cytochrome (Cyt.) aa3 were usually shown. Under ischemic conditions (induced by sudden severe hypotension plus bilateral common carotid occlusion), ccbf and blood volume were decreased. There was also a decrease of brain po2 and a reduction of Cyt. aa3 following ischemia. These techniques are applicable in intraoperative monitoring of patients. KEY WORDS 9 cortical metabolism 9 ischemia 9 asphyxia 9 hypoxia 9 oxygen tension 9 cytochrome aa3 I X is known that transient ischemic attacks (TIA s) can be caused by a hemodynamic effect. Hemodynamic TIA s are thought to be due to background factors consisting of a subthreshold flow-limiting lesion of the arteries leading to the brain, causing some degree of ischemia. Triggering factors include a transient drop in perfusion pressure due to decreased cardiac output. These factors together produce a critical reduction in cerebral blood flow (CBF) to the point that brain oxygen tension (brain p02) is lowered to a threshold level for TIA s. 4 However, hypoxia (low PaO2) and asphyxia (low PaO2 plus high PaCO2) are also known to produce a similarly decreased brain P02 and to cause clinical symptoms. To investigate these mechanisms, we used a hemodynamic ischemic model consisting of sudden severe hypotension plus transient bilateral common carotid occlusion, and observed the effects on brain cortical oxidative metabolism. In addition, we compared the effects of hypoxia and asphyxia. Materials and Methods Protocol Fourteen adult New Zealand White rabbits of both sexes were used in this investigation. Each rabbit weighed between 3 and 5 kg. After a preliminary dose of ketamine, a tracheostomy was performed. All animals were anesthetized throughout the studies with nitrous oxide and oxygen in a 3:1 mixture. Rectal temperature was maintained between 36 ~ and 38~ The animals were ventilated by respirator, and were immobilized with gallamine triethiodide (Flaxedil). Changes in gas mixtures were made by altering the input to the respirator. The femoral artery was cannulated to monitor the systemic blood pressure and sample the blood for arterial ph, PaO2, and PaCO2. The mean values were: arterial ph 7.34 _+ 0.02, PaO _ 5.5 mm Hg, and PaCO _+ 1.2 mm Hg (Table 1). The control blood pressure was 92.0 _+ 8.6 mm Hg. The femoral vein was cannulated for administering additional Flaxedil and saline, and for injections of trimethaphan camsylate (Arfonad) or phenylephrine hydrochloride (Neo-Synephrine). Silk ligatures were placed loosely around both common carotid arteries for temporary occlusion. The head of the animal was fixed in a stereotaxic frame for performing bilateral craniotomies. Hypoxia was induced by reducing the fraction of inspired oxygen (FiO2) from 25% to 10% for 2 minutes, while continuously ventilating the rabbit at the same J. Neurosurg. / Volume 59 / July,

2 H. Koga and G. Austin TABLE 1 Analysis of blood gas during hypoxia and asphyxia* Blood Gas Asphyxia Resting State Hypoxia (Respirator (25% FiO2) (10% FiO2) Off) ph PaCO2 (mm Hg) PaO2 (ram Hg) _ _ _ * FiO2 = Fraction of inspired oxygen. FIG. 1. Response of brain cortical oxygen tension (bpo2), cytochrome (Cyt) aa3, cortical cerebral blood flow (ccbf), and cortical blood volume (B1 Vol) during hypoxia. FiO2 = fraction of inspired oxygen; Red. = reduction; F.S. = percent of the full-scale optical signal. rate and tidal volume. Asphyxia was produced by turning off the respirator and allowing no respiratory exchange in the animal for a 2-minute period. Ischemic insults were caused by briefly inducing hypotension (usually a decrease in mean blood pressure of 30 mm Hg) with an injection of Arfonad, and then temporarily occluding both common carotid arteries. Monitoring Techniques Cortical Oxygen Tension. Cortical oxygen tension was continuously recorded by the polarographic method, as modified by Yonekura and Austin (unpublished data), using a 25-u platinum electrode and 250-u silver-silver chloride indifferent electrode. The oxygen microelectrode used in these experiments and the techniques were essentially the same as those described by Silver The electronic circuitry for the polarographic measurement of brain po2 was provided by a Schema Versatae Model 400 polarographic current amplifier.* The platinum electrode was polarized at -700 mv; this electrode recorded the change in oxygen availability at its tip. The electrode calibration was checked before each experiment by dipping the tip in saline and bubbling 100% N2, room air, and 100% 02 through it. The probe rested lightly on the pulsating cortical surface and was supported by a coiled spring, avoiding small arteries or veins. Cortical Cerebral Blood Flow. Cortical CBF (ccbf) was measured polarographically in the normal resting state by the hydrogen-clearance method, using the same electrode as was used for measuring brain po2. The theoretical considerations are based on Kety and Schmidt s approach to blood tissue exchange of inert gases. 17,18 The basic method of hydrogen-clearance measurement consisted of resting a positively polarized * Schema Versatae Model 400 polarographic current amplifier manufactured by Tegal Scientific Inc., Berkeley, California. (+200 mv) electrode on the tissue and administering hydrogen by respiration. As the hydrogen was cleared from the arterial blood, the exponential clearance rate of hydrogen from the tissue was monitored. 26 At another cortical site, ccbf was monitored continuously by a thermal diffusion flow probe for dynamic quantitative measurement. The probe consisted of a Peltier stack with an t_-shaped gold plate and thermocouple attached to the back of a portion of the gold plate. This technique has been described previously by Brawley 7 and Carter and Atkinson. 8 9 Carter and Erspamer ~ reported that the thermal gradient between the thermocouple in microvolts was highly correlated with cortical blood flow as determined by xenon-133 clearance. This thermal diffusion technique was used to monitor cortical blood flow under conditions of ischemia, asphyxia, and hypoxia. Cortical Mitochondrial Oxidative Metabolism. Cortical mitochondrial oxidative metabolism was monitored by two noninvasive optical methods, based on the original work of Chance, et al, II lz and J6bsis, et al ~5 ~6 The first method involved the ratio of reduced cytochrome (Cyt.) aa3 to oxidized Cyt. aa3, as recorded continuously by a dual-beam dual-wavelength reflectance spectrophotometer, modified by J6bsis, et al. 5"16 This measurement is based on the observation that Cyt. aa3 in its reduced form absorbs light at a specific wavelength (605 nm). As a comparison, to rule out hemoglobin absorption, the cortex was also illuminated with light at 590 nm, a so-called "equibestic point" where the relative amounts of reduced and oxygenated hemoglobin are the same as at 605 nm. The spectrophotometric difference tracing was calibrated as a percentage change of total light intensity. The relative blood volume was estimated from the change in total amount of hemoglobin, monitored at the equibestic wavelength (590 nm). 15 The second method of measurement involved the redox level of mitochondrial flavoprotein, monitored by a real-time flying-spot fluorometer designed by Chance, et al ~ ~2 Flavoprotein in oxidized form becomes fluorescent at 540 nm when excited by light at 442 nm. Flavoprotein excitation is obtained with a 10- mw helium-cadmium laser.t A rectangular region of t Liconix Model 902 helium-cadmium laser manufactured by Liconix, Mountain View, California. 58 J. Neurosurg. / Volume 59 / July, 1983

3 Factors in cortical metabolism TABLE 2 Comparison of brain p02, ccbf, CBV, cytochrome aa3, and flavoprotein during oxygen deficiency* Oxygen Deficiency Brain po2 ccbf CBV Cytochrome aa 3 Flavoprotein hypoxia decrease increase slight increase reduction reduction asphyxia decrease large increase slight increase reduction reduction ischemia decrease decrease decrease reduction reduction * Brain po2 = cortical oxygen tension; ccbf = cortical cerebral blood flow; CBV = cerebral blood volume. cortex was swept over an area of approximately 2 x 4 mm with a 0.1-mm spot. The fying-spot fluorometer has the advantage of providing a percentage change in oxidized heterogeneity of cortex. Results We observed cortical oxidative metabolism under conditions ofischemia, hypoxia, and asphyxia from the same cortical sites in all subjects. The results are summarized in Table 2. The mean values of brain po2 under normoxic steady conditions, was 30.6 _+ 4.3 mm Hg from 42 measurements of these 14 cortical sites. Hypoxia Arterial po2 was decreased by reducing the FiO2 from 25% to 10% for 2 minutes while continuing to ventilate the rabbit at the same rate and tidal volume. There was no significant difference in the PaCO2 and ph between blood samples drawn during ventilation at a 25% FiO2 and at a 10% FiO2 (Table 1), but there was a relationship between arterial po2 and FiO2. During a short period of hypoxia, the brain po2 initially decreased gradually and then leveled out for a short period of time. On return to normoxia from hypoxia, the brain po2 values at each of the 14 sites did not increase above the control value (Fig. 1). During hypoxia, brain po2 decreased an average 7.1 mm Hg from the initial value of 32.2 _+ 7.6 mm Hg at the 14 sites (Table 3). There was an increase of blood flow and blood volume following the period of decreased brain po2. The average increase in ccbf was 47.7% from the resting value of 24.3 _+ 3.2 ml/100 gm/min. On the other hand, there was a reduction of Cyt. aa3, which paralleled the time course of decreased brain po2 during hypoxia (Fig. 1). It was also shown that there was a shift toward reduction of the other member of the mitochondrial respiratory chain, flavoprotein. Asphyxia Asphyxia was induced by turning off the respirator for 2 minutes, and was accompanied by a large decrease in PaO2 and an increase in PaCO2 compared with normal levels (Table 1). The brain poe decreased abruptly after the respirator was turned off(fig. 2), for an average decrease of 7.3 mm Hg from the initial value TABLE 3 Comparison of average decrease in brain p02 at same cortical sites during oxygen deficiency* Oxygen Resting Value Decrease of Brain po2 of Brain po2 Deficiency (mm Hg) (mm Hg) hypoxia _+ 1.9 asphyxia 31.4 _ isehemia 28.1 _ _+ 0.6 * Brain po2 = cortical oxygen tension. of 31.4 _+ 7.4 mm Hg in the 14 sites. Comparison of brain poz values recorded at the same 14 sites during hypoxia and asphyxia showed no significant differences (Table 3). With the onset of asphyxia, there was an immediate increase in the level of reduced Cyt. aa3 (Fig. 2 upper). Also the fuorescence histogram of flavoprotein shifted to a lower intensity, indicating that it was reduced (Fig. 2 lower). On the other hand, ccbf was remarkably increased and there was a slight increase in blood volume during asphyxia, when the cortex was in good condition (Fig. 2 upper). After the respirator was restarted, there was a further increase in ccbf (Fig. 2 upper). The average increase in ccbf was 100% above the resting value of 24.4 _+ 3.0 ml/100 gm/min, which was a greater increase than that found in hypoxia (Table 4). There was usually an overshoot of brain po2 after restarting the respirator and also a post-asphyxic oxidation of Cyt. aa3 (Fig. 2 upper). The increased brain po2 then returned to its initial value, and paralleled the normalization of ccbf and redox level of Cyt. aa3 (Fig. 2 upper). Ischemia When the blood pressure was abruptly reduced by Arfonad to a value of 30 mm Hg, on the average, the ccbf and blood volume were also decreased. Similarly, brain po2 was decreased, and a reduction of Cyt. aa3 was observed during hypotension (Fig. 3). Subsequent bilateral common carotid occlusion, carried out while maintaining this severe hypotension, resulted in only slight additional changes of brain po2, Cyt. aa3 and flavoprotein (Fig. 3). During ischemia, brain po2 decreased an average of 3.9 mm Hg from the resting J. Neurosurg. / Volume 59 ~July,

4 H. Koga and G. Austin value of mm Hg in the 14 sites (Table 3). The average decrease in ccbf was 27.5% from the resting value ( ml/100 gm/min, Table 4). After the ischemic stress, blood po2, and ccbf showed a slight tendency to return toward baseline. As soon as the blood pressure was raised by injecting 1% Neo-Synephrine, ccbf and brain po2 increased dramatically, coupled with the oxidation of Cyt. aa3 (Fig. 3). FIG. 2. Upper: Response of brain cortical oxygen tension (bpo2), cytochrome (Cyt) aa3, cortical cerebral blood flow (ccbf), and cortical blood volume (B1 Vol) during asphyxia. FiO2 = fraction of inspired oxygen; Red. = reduction; F.S. = percent of the full-scale optical signal; BP = blood pressure. Lower: Response of cortical oxygen tension and flavoprotein by flying spot scanner during asphyxia. TABLE 4 Comparison of average changes at the same cortical sites of ccbf during oxygen deficiency* Oxygen Resting Values of ccbf Change in Deficiency (ml/100 gm/min) ccbf (%) hypoxia _ 14.6 asphyxia 24.4 _ _ ischemia _ 5.8 * ccbf = cortical cerebral blood flow. Discussion Since the reports of polarographic measurement of oxygen tension in cortex by Davies and Brink, 13 numerous reports have been published on oxygen tension in the brain cortex measured by microelectrodes, z~ Silver TM and Bicher and Knisely 6 reported that the cortical po2 at a specific microarea is remarkably constant, and can be changed only through major alterations in blood supply or the composition of respiratory gases. In general, a correlation was shown between PaCO2 and the inspiratory 02 concentration.19 In these experiments, the brain po2 during a short period ofhypoxia initially decreased gradually, and then leveled out for a short period of time (Fig. 1). There was an increase of blood flow and blood volume following the period of decreased cortical oxygen tension (Fig. 1). These changes suggest an autoregulation mechanism to maintain a constant brain cortical oxygen environment in the intact brain. Bicher 5 introduced the concept of maintaining a constant brain-cell oxygen microenvironment, which he called "oxygen autoregulatory mechanisms." He used four criteria to identify the oxygen autoregulation mechanism in brain tissue following a short period of anoxia" 1) a short period of reoxygenation; 2) an increase in CBF; 3) an overshoot; and 4) electrical silence paralleling the period of tissue po2 depression. In our studies, we observed either no overshoot or only a small one. The difference between our experimental procedure and Bicher s protocol is that we exposed our animals to a 10% FiO2, whereas he obtained hypoxia by nitrogen respiration. There were shifts toward reduction of members of the mitochondrial respiratory chains (Cyt. aa3 and flavoprotein) similar to the changes of brain po2. However, Schutz, et al., 22 have shown that there was no change in isolated mitochondrial function after 37 minutes of severe systemic hypoxic normotension in the rabbit. This suggests that there is a marked difference in the mitochondrial respiration in isolated mitochondria and in the in vivo preparation. Rosenthal, et al., 21 already showed that increasing inspired oxygen levels produced an oxidation of Cyt. aa3, while there was a sharp reduction in response to lowering the FiO2. Our data confirm their results. The time course of brain po2 during and after asphyxia showed three phases: 1) a decrease; 2) a leveling off; and 3) a return to the initial pressure, accompanied by an overshoot. The leveling off phase could be ex- 60 J. Neurosurg. / Volume 59 ~July, 1983

5 Factors in cortical metabolism plained by decreased oxygen utilization, with reduced oxygen supply. After the respirator was restarted, there was a noticeable overshoot of brain po2 paralleling the increased systemic blood pressure and further increased ccbf (Fig. 2). This is best explained by tissue hyperemia from high tissue carbon dioxide. The increased blood pressure and ccbf, coupled with the delayed return of the vasodilated cerebral vessels to their original state, rapidly resaturates the tissue with oxygen and leads to an overshoot of brain po2. The efficiency of oxygen supply and utilization can be tested by measuring the redox state of the mitochondrial respiratory chain. The time course of redox level of Cyt. aa3 under conditions of asphyxia showed a change parallel to the brain po2. This would support the idea that mitochondrial respiration is the main consumer of oxygen, and that reduction of Cyt. aa3, is associated with a change of oxygen utilization, as shown by Austin, et all 3 The drop in blood pressure produced by Arfonad followed by the brief bilateral carotid occlusion during ischemia was accompanied by a decrease in brain po2, ccbf, and Cyt. aa3. Before autoregulation occurred, the abrupt drop in the mean blood pressure produced a drop in perfusion pressure and caused a decrease in CBF. The reduction in blood flow decreased oxygen delivery to the cortex. In our ischemic model (hypotension and bilateral carotid occlusion in the rabbit), there were no overshoots of brain po2 and postocclusion oxidation of Cyt. aa3 after releasing the occluding ligatures. We attribute this finding to the still low perfusion pressure and a lack of development of collateral circulation. This suggested that circulatory perfusion has an important role in oxygen autoregulation mechanisms under severe conditions of decreased oxygen delivery. Ekl6f and Siesj614 reported that energy and ratios of nicotinamide adenine dinucleotide in its reduced to its oxidized form (NADH:NAD + ratios) were grossly impaired at cerebral venous po2 values of 32 mm Hg, but there was also a reduction of the CBF to 45 % of normal, induced by carotid ligation and hypotension. The data we have presented suggest that members of the cortical mitochondrial respiratory chains may be more sensitive to changes in tissue oxygen tension and may be in need of a much higher concentration of oxygen in order to be fully oxidized. On the other hand, it may be that fewer mitochondria are being oxidized when cortical blood flow is reduced, or under conditions of reduced PaO2, due to simple hypoxia or asphyxia. None of these mechanisms can at present be excluded. The increase of relative oxidation of members of the electron transport chain under conditions of increased cortical po2 suggests that there may be mitochondria in a borderline region of hypoxia that only respond under conditions that we regard as hyperoxic. This study is important for the further understanding of human cortical oxidative metabolism by noninvasive monitoring. Austin, et al,1 2"4 have reported that prelim- FIG. 3. Response of brain cortical oxygen tension (bpo2), cytochrome (Cyt) aa3, cortical cerebral blood flow (ccbf), and cortical blood volume (B1 Vol) during ischemia. Red. = reduction; F.S. = percent of the full-scale optical signal; BP = blood pressure; Bil.C.Car.Occ. = bilateral common carotid occlusion. inary results ofnoninvasive monitoring in the operating room showed not only a significant increase in ccbf following superficial temporal artery-middle cerebral artery anastomosis, but an associated increase of brain oxygen availability and also an increase in cortical oxidative metabolism. The changes of cortical po2 and cortical oxidative metabolism in the rabbit, cat, and human cortex therefore appear to vary in rather similar fashion in response to return to reoxygenation. References 1. Austin G, Jutzy R, Chance B, et al: Noninvasive monitoring of human brain oxidative metabolism. Front Biol Energet 2: , Austin G, Schuler W, Haugen G, et al: Brain metabolism in the cat during brief transient ischemia, in Austin GM (ed): Contemporary Aspects of Cerebrovaseular Disease Dallas, Texas: Professional Information Library, 1976, pp Austin G, Schuler W, Willey J: In vivo studies of mitochondrial respiration, in Silver IA, Erecifiska M, Bicher HI (eds): Oxygen Transport to Tissue III. New York: Plenum Press, 1978, pp Austin GM, Haugen G, Lichter E, et al: Microneurosurgical anastomosis: A biochemical basis for improvement. Acta Med Nagasaki 22:35-49, Bicher HI: Brain oxygen autoregulation: a protective reflex to hypoxia? Microvasc Res 8: , Bicher HI, Knisely MH: Brain tissue reoxygenation time, demonstrated with a new ultramicro oxygen electrode. J Appl Physiol 28: , Brawley BW: Construction of a regional cerebral blood flow probe using a Peltier device. J Sorg Res 9: , Carter LP, Atkinson JR: Autoregulation and hyperemia of cerebral blood flow as evaluated by thermal diffusion. Stroke 4: , Carter LP, Atkinson JR: Cortical blood flow in controlled hypotension as measured by thermal diffusion. J Neurol Neurosurg Psychiatry 36: , 1973 J. Neurosurg. / Volume 59 ~July,

6 H. Koga and G. Austin 10. Carter LP, Erspamer R: Dynamic quantitative assessment of cortical blood flow. Neurol Res 2: , Chance B: Spectrophotometric and kinetics studies of flavoproteins in tissue, cell suspensions, mitochondria and their fragments, in Slater EC (ed): Flavins and Flavoproreins. Amsterdam/New York: Elsevier, 1966, pp Chance B, Barlow C, Nakase Y, et al: Heterogeneity of oxygen delivery in normoxic and hypoxic states: a fluorometer study. Am J Physiol 235:H809-H820, Davies PW, Brink F Jr: Microelectrodes for measuring local oxygen tension in animal tissues. Rev Sci Instrum 13: , Ekl6f B, Siesj6 BK: Cerebral blood flow and cerebral energy state. Acta Physiol Scand 82: , J6bsis FF, Keizer JH, LaManna JC, et al: Reflectance spectrophotometry of cytochrome aa3 in vivo. J Appl Physiol 43: , J6bsis FF, Rosenthal M, LaManna JC, et al: Metabolic activity in epileptic seizures, in Ingvar DH, Lassen NA (eds): Brain Work: The Coupling of Function, Metabolism and Blood Flow in the Brain. Copenhagen: Munksgaard, 1975, pp Kety SS: The theory and applications of the exchange of inert gas at the lungs and tissues. Pharmacol Rev 3:1-41, Kety SS, Schmidt CF: The determination of cerebral blood flow in man by the use of nitrous oxide in low concentrations. Am J Physiol 143:53-66, Metzeger H, Erdmann W, Thews G: Effect of short periods of hypoxia, hyperoxia, and hypercapnia on brain 02 supply. J Appl Physiol 31: , Nair P, Whatlen WJ, Buerk D: PO2 of cat cerebral cortex: response to breathing N2 and 100% O2. Mierovasc Res 9: , Rosenthal M, LaManna JC, J6bsis FF, et al: Effects of respiratory gases on cytochrome A in intact cerebral cortex: is there a critical PO2? Brain Res 108: , Schutz H, Silverstein PR, Vapalahti M, et al: Brain mitochondrial function after ischemia and hypoxia. II. Normotensive systemic hypoxemia. Arch Neuro129: , Silver IA: The measurement of oxygen tension in tissues, in Payne JP, Hill DW (eds): Oxygen Measurements in Blood and Tissues and Their Significance. Boston: Little, Brown and Co, 1966, pp Silver IA: Some observations on the cerebral cortex with an ultramicro, membrane-covered, oxygen electrode. Med Electron Biol Eng 3: , Smith RH, Guilbeau E J, Reneau DD: The oxygen tension field within a discrete volume of cerebral cortex. Microvasc Res 13: , Young W: HE clearance measurement of blood flow: a review of technique and polarographic principles. Stroke 11: , 1980 Manuscript received April 12, Accepted in final form January 14, Address reprint requests to: Hiroaki Koga, M.D., Department of Neurosurgery, Nagasaki University School of Medicine, 7-1 Sakamoto-machi, Nagasaki 852 Japan. 62 J. Neurosurg. / Volume 59 / July, 1983