Acta physiol. scand. 1975. 93. 515-524 From the Brain Research Laboratory, E-Blocket and the Department of Surgery, University of Lund, Sweden
Brain Energy Metabolism in Anesthetized Rats in Acute Anemia BY
HALLDOR JOHANNSSON and Bo. K. SIESJO
Received 28 October 1974
Abstract H. and B. K. SIESJO.Brain energy metabolism in anesthetized rats in acute anemia. Acta physiol. scand. 1975. 93. 515-524.
JOHANNSSON,
In order to evaluate if pronounced anemic hypoxia gives rise to signs of cerebral oxygen lack the blood hemoglobin content was reduced to 6 and 3 g.(IOO mI)-l. Cerebral blood flow increased in spite of the fact that there was a moderate reduction in blood pressure, and in mean tissue COa tension, and in the absence of signs of an increased glycolytic rate in the tissue. With a reduction in a hemglobin content to 3 g.(IOO mI)-l there was a moderate increase in tissue lactate content and associated changes in other carbohydrate metabolites and in amino acids of the type seen in hypoxic hypoxia, suggesting that tissue hypoxia was present. However, since the concentrations of phosphocreatine and adenine nucleotides remained constant this hypoxia must have been slight. It is concluded that there is cerebral vasodilation in the brain in pronounced anemic hypoxia, and that this vasodilatation, in combination with the reduced viscosity, creates favourable conditions for cerebral oxygenation.
In a preceding communication (Borgstrom et al. 1974 b) results were reported which showed that the cerebral blood flow (CBF) of lightly anaesthetized rats was progressively increased when the blood hemoglobin content was reduced from a normal value of 14-16 g.(lOO mI)-l to 12, 9, 6 and 3 gS(100 ml)-l. At hemoglobin contents of 12 and 9 g.(IOO ml)-I the increase in CBF could, at least theoretically, be attributed to reduction in blood viscosity but with more pronounced degrees of anemia the increase in CBF was far in excess of that which could have been due to viscosity changes. This suggests that the cerebral hyperemia occurring at marked degrees of normovolemic anemia is related to the fall in arterial oxygen content (Toy),i.e. to hypoxia. However, since neither the cerebral venous Po$ nor the venous oxygen saturation decreased significantly (Borgstrom et al. 1974 b) the data failed to give evidence of the presence of cerebral hypoxia. In the present communication we report on experiments in which the blood hemoglobin content was decreased to 6 and 3 g.(IOO ml)-* respectively, with subsequent measurements 515
516
H A L L D ~ RJ ~ H A N N S S O NAND BO K. SIESJO
of organic phosphates, glycolytic and citric acid cycle intermediates and of amino acids in the cerebral cortex. The objective of the study was to find out whether or not signs of tissue hypoxia develop at pronounced degrees of anemia. A priori, the presence of a normal venous Po*, and a normal venous oxygen saturation, would seem to exclude the possibility of cerebral hypoxia. However, experiments on arterial hypotension combined with moderate hypercapnia show that signs of energy failure in the brain may appear at normal, or supernormal venous Po, values (Eklof et al. 1973). Such paradoxical results suggest that under certain conditions cerebral blood flow may become grossly inhomogeneous and that an elevated venous Po, may reflect hyperemia in only part of the tissue (Siesjo et at. 1973). It is evident that provided such conditions arise only direct analyses of labile cerebral metabolites can reveal the presence of tissue hypoxia.
Methods As in the previous study (Borgstrom et al. 1974 b) the experiments were performed on male Wistar rats (310-410 g) that were initially anesthetized with 2-3 I% halothane and then maintained artificially ventilated on 704: N,O and 30% O,, with halothane administration discontinued. The body temperature was adjusted to 37°C and the arterial Pco, to 35-40 mm Hg. Both femoral arteries and one femoral vein were cannulated, allowing measurements of arterial Po,, Pco,, pH, oxygen content (To,), hemoglobin content, glucose, lactate and pyruvate, as well as bleeding from the artery and isovolemic substitution with fresh homologous plasma intravenously. There were two experimental series. I n one (series A) the posterior part of the superior sagittal sinus was exposed through a small burr hole, allowing sampling of cerebral venous blood. In these animals the blood hemoglobin content was reduced to about 6 g.(IOO ml)-*.Two and thirty min after that the procedure of bleeding and substitution had been completed arterial and cerebral venous blood were sampled whereafter the tissue was frozen in situ with liquid nitrogen (see Pontkn et al. 1973). Blood was analysed for Po., PCO, and pH and cortical tissue for organic phosphates and for some carbohydrate metabolites. The main objective of the analyses on blood was to find out whether or not the increase in CBF (as evaluated from the arteriovenous difference in To,) was accompanied by signs of tissue acidosis due to CO, retention o r to accumulation of organic acids. In this series no control group was studied and the metabolite data were compared to a control series of animals, the brains of which were extracted at the same period, using identical techniques. Since the results of series A showed that the increase in CBF could not be explained by accumulation of carbon dioxide or of organic acids, a second series (series B) was made, with reduction of the hemoglobin content to about 3 g*(IOOmI)-l. In these animals, the tissue was frozen in situ either 2 or 30 min following the end of the bleeding-substitution procedure but no cerebral venous blood was sampled. However, in these groups cisternal CSF was sampled for measurements of glucose, lactate and pyruvate. A control group was obtained by withdrawing about 10 ml of blood and substituting an equal volume of fresh blood from donor rats, and by freezing the tissue about 15 min later. Since the purpose of these experiments was to estimate the degree of tissue hypoxia, if any, measurements were made of organic phosphates, glycolytic and citric acid cycle metabolites and some amino acids. Arterial and venous Po,, Pco,, pH, To, and hemoglobin content were measured as described previously (Borgstrom e t a/. 1974 a and b). Glucose, lactate and pyruvate in arterial blood and in CSF were analysed enzymatically (see below). Cortical tissue was extracted at - 22°C and enzymatic, fluorometric techniques (Lowry and Passonneau 1972) were used to measure glycogen, glucose, glucose-6-phosphate (G-6-P). fructose-1,6-diphosphate (FDP), dihydroxyacetone phosphate (DHAP), 3-phosphoglycerate (3-PG), pyruvate, lactate, citrate, a-ketoglutarate (a-KG), malate, aspartate, glutamate, glutamine, GABA, ammonia (NH:). phosphocreatine (PCr), creatine (Cr), ATP, ADP, and AMP. For references to individual methods, the reader is referred to a recent communication (Folbergrova e l a / . 1974). Statistical differences were evaluated with Student’s t-test. The following symbols are used: p - 0.05- *, p
Brain Energy Metabolism in Anesthetized Rats in Acute Anemia BY
HALLDOR JOHANNSSON and Bo. K. SIESJO
Received 28 October 1974
Abstract H. and B. K. SIESJO.Brain energy metabolism in anesthetized rats in acute anemia. Acta physiol. scand. 1975. 93. 515-524.
JOHANNSSON,
In order to evaluate if pronounced anemic hypoxia gives rise to signs of cerebral oxygen lack the blood hemoglobin content was reduced to 6 and 3 g.(IOO mI)-l. Cerebral blood flow increased in spite of the fact that there was a moderate reduction in blood pressure, and in mean tissue COa tension, and in the absence of signs of an increased glycolytic rate in the tissue. With a reduction in a hemglobin content to 3 g.(IOO mI)-l there was a moderate increase in tissue lactate content and associated changes in other carbohydrate metabolites and in amino acids of the type seen in hypoxic hypoxia, suggesting that tissue hypoxia was present. However, since the concentrations of phosphocreatine and adenine nucleotides remained constant this hypoxia must have been slight. It is concluded that there is cerebral vasodilation in the brain in pronounced anemic hypoxia, and that this vasodilatation, in combination with the reduced viscosity, creates favourable conditions for cerebral oxygenation.
In a preceding communication (Borgstrom et al. 1974 b) results were reported which showed that the cerebral blood flow (CBF) of lightly anaesthetized rats was progressively increased when the blood hemoglobin content was reduced from a normal value of 14-16 g.(lOO mI)-l to 12, 9, 6 and 3 gS(100 ml)-l. At hemoglobin contents of 12 and 9 g.(IOO ml)-I the increase in CBF could, at least theoretically, be attributed to reduction in blood viscosity but with more pronounced degrees of anemia the increase in CBF was far in excess of that which could have been due to viscosity changes. This suggests that the cerebral hyperemia occurring at marked degrees of normovolemic anemia is related to the fall in arterial oxygen content (Toy),i.e. to hypoxia. However, since neither the cerebral venous Po$ nor the venous oxygen saturation decreased significantly (Borgstrom et al. 1974 b) the data failed to give evidence of the presence of cerebral hypoxia. In the present communication we report on experiments in which the blood hemoglobin content was decreased to 6 and 3 g.(IOO ml)-* respectively, with subsequent measurements 515
516
H A L L D ~ RJ ~ H A N N S S O NAND BO K. SIESJO
of organic phosphates, glycolytic and citric acid cycle intermediates and of amino acids in the cerebral cortex. The objective of the study was to find out whether or not signs of tissue hypoxia develop at pronounced degrees of anemia. A priori, the presence of a normal venous Po*, and a normal venous oxygen saturation, would seem to exclude the possibility of cerebral hypoxia. However, experiments on arterial hypotension combined with moderate hypercapnia show that signs of energy failure in the brain may appear at normal, or supernormal venous Po, values (Eklof et al. 1973). Such paradoxical results suggest that under certain conditions cerebral blood flow may become grossly inhomogeneous and that an elevated venous Po, may reflect hyperemia in only part of the tissue (Siesjo et at. 1973). It is evident that provided such conditions arise only direct analyses of labile cerebral metabolites can reveal the presence of tissue hypoxia.
Methods As in the previous study (Borgstrom et al. 1974 b) the experiments were performed on male Wistar rats (310-410 g) that were initially anesthetized with 2-3 I% halothane and then maintained artificially ventilated on 704: N,O and 30% O,, with halothane administration discontinued. The body temperature was adjusted to 37°C and the arterial Pco, to 35-40 mm Hg. Both femoral arteries and one femoral vein were cannulated, allowing measurements of arterial Po,, Pco,, pH, oxygen content (To,), hemoglobin content, glucose, lactate and pyruvate, as well as bleeding from the artery and isovolemic substitution with fresh homologous plasma intravenously. There were two experimental series. I n one (series A) the posterior part of the superior sagittal sinus was exposed through a small burr hole, allowing sampling of cerebral venous blood. In these animals the blood hemoglobin content was reduced to about 6 g.(IOO ml)-*.Two and thirty min after that the procedure of bleeding and substitution had been completed arterial and cerebral venous blood were sampled whereafter the tissue was frozen in situ with liquid nitrogen (see Pontkn et al. 1973). Blood was analysed for Po., PCO, and pH and cortical tissue for organic phosphates and for some carbohydrate metabolites. The main objective of the analyses on blood was to find out whether or not the increase in CBF (as evaluated from the arteriovenous difference in To,) was accompanied by signs of tissue acidosis due to CO, retention o r to accumulation of organic acids. In this series no control group was studied and the metabolite data were compared to a control series of animals, the brains of which were extracted at the same period, using identical techniques. Since the results of series A showed that the increase in CBF could not be explained by accumulation of carbon dioxide or of organic acids, a second series (series B) was made, with reduction of the hemoglobin content to about 3 g*(IOOmI)-l. In these animals, the tissue was frozen in situ either 2 or 30 min following the end of the bleeding-substitution procedure but no cerebral venous blood was sampled. However, in these groups cisternal CSF was sampled for measurements of glucose, lactate and pyruvate. A control group was obtained by withdrawing about 10 ml of blood and substituting an equal volume of fresh blood from donor rats, and by freezing the tissue about 15 min later. Since the purpose of these experiments was to estimate the degree of tissue hypoxia, if any, measurements were made of organic phosphates, glycolytic and citric acid cycle metabolites and some amino acids. Arterial and venous Po,, Pco,, pH, To, and hemoglobin content were measured as described previously (Borgstrom e t a/. 1974 a and b). Glucose, lactate and pyruvate in arterial blood and in CSF were analysed enzymatically (see below). Cortical tissue was extracted at - 22°C and enzymatic, fluorometric techniques (Lowry and Passonneau 1972) were used to measure glycogen, glucose, glucose-6-phosphate (G-6-P). fructose-1,6-diphosphate (FDP), dihydroxyacetone phosphate (DHAP), 3-phosphoglycerate (3-PG), pyruvate, lactate, citrate, a-ketoglutarate (a-KG), malate, aspartate, glutamate, glutamine, GABA, ammonia (NH:). phosphocreatine (PCr), creatine (Cr), ATP, ADP, and AMP. For references to individual methods, the reader is referred to a recent communication (Folbergrova e l a / . 1974). Statistical differences were evaluated with Student’s t-test. The following symbols are used: p - 0.05- *, p
