Acta physiol. scand. 1975. 93. 505-514 From the Brain Research Laboratory, E-Blocket, University Hospital, and the Department of Surgery, University of Lund, Sweden
The Influence of Acute Normovolemic Anemia on Cerebral Blood Flow and Oxygen Consumption of Anesthetized Rats BY
LARSBORGSTROM, HALLDORJOHANNSSON and Bo K. SIESJO Received 22 October 1974
Abstract BORGSTROM, L., H. JOHANNSSON and B. K.SIESJO.The influence of acute normouolemic anemia on cerebral blood flow and oxygen consumption of anesthetized rats. Acta physiol. scand. 1975. 93. 505-514. The influence of acute normovolemic anemia on cerebral blood flow (CBF) and cerebral metabolic rate for oxygen (CMRo,) was studied in normocapnic rats under nitrous oxide anaesthesia. The arterial hemoglobin content was reduced to values of about 12, 9, 6 and 3 g.(IOO mI)-l by arterial bleeding and substitution with equal volumes of homologous plasma. The CBF increased in proportion to the reduction in hemoglobin content t o reach values of 500-600 per cent of normal at extreme degrees of anemia, but CMRo, remained unchanged. Cerebral venous Po, and oxygen saturation did not decrease below normal values, indicating that tissue hypoxia did not develop. However, since the increase in CBF at hemoglobin concentrations of below 9 g.(lOO mI)-l was far in excess of that expected from the decrease in viscosity the results indicate that dilatation of cerebral resistance vessels occurred. This dilatation, which was obviously related to the fall in arterial oxygen content, cannot be explained by any of the current theories proposed to explain cerebral hyperemia in hypoxia.
Although there is much information on cerebral blood flow (CBF) and cerebral oxygen consumption (CMRo,) in acute and chronic anemias important problems remain unsolved. Chronic anemias in man (pernicious, sickle cell, and other types) were studied by Scheinberg (1951) and by Heyman et al. (1952), who found an increased CBF and decreased CMRo,. These authors concluded that cerebral hypoxia contributed to the increase in CBF and decrease in CMRo,. Hbwever, since administration of 85-100 per cent oxygen (Heyman et al. 1952) or specific therapy (Scheinberg 1951) did not bring CMR,, back to normal, it appeared doubtful if the fall in CMRo, was due to hypoxia alone. Subsequent experiments on acute anemia have been largely designed to study the factors responsible for the cerebral hyperemia. Haggendal and Norback (1966 b) found little variation in CBF in dogs over a hematocrit range of 70-30 per cent and concluded that the increase in CBF, occurring below a hematocrit of about 30 per cent, was due to cerebral hypoxia, the decrease in viscosity playing only a minor role. Data presented by this group of workers (Haggendal and Norback 1966 a) suggested that CMRo, was maintained at a 505
506
LARS BORGSTROM, H A L L D ~ RJ ~ H A N N S S O NAND BO K.
SIESJO
hematocrit of 20 per cent. Similar results were reported by Michenfelder and Theye (1969) who reduced hemoglobin concentration from 12.4 to 5.6 ga(1OO ml)-l in the same species. In their experiments CBF increased from 77 to 108 ml.(lOO g)-l. min-l but CMRo, remained constant. Since cerebral venous Po*was reduced from 37 to 30 mm Hg cerebral hypoxia may be assumed to have contributed to the increase in CBF. In a subsequent study on man (Paulson et al. 1973), the hematocrit was reduced from 38 to 28 per cent. This procedure gave an increase in CBF by 15-20 per cent but since cerebral venous Po, remained unchanged the authors concluded that the hyperemia was largely due to reduced blood viscosity, and that cerebral hypoxia should not have been present. The present experiments were designed to study circulation and oxygen supply to the rat brain in acute normovolemic anemia. The blood hemoglobin concentration was reduced to minimal values of 3 g.(lOO m l p , i.e. to about 20 per cent of normal, with subsequent measurements of CBF, CMRo, and cerebral venous Po,. The objectives of the experiments were to study the efficiency of the circulatory adaptation in moderate to severe anemia, and to obtain further information on the cause of the hyperemia. A preliminary cornmunication on part of the present material has been published (J6hannsson and Siesjo 1974 a).
Methods Since the procedures and methods used in the present study were similar to those reported i n two previous communications (Johannsson and Siesjo 1974 b, Borgstrom et al. 1974 a) only the main outline is given here, with an emphasis on differences in procedures. The experiments were performed on unstarved male rats (350400g) that were anesthetized with 2 per cent halothane, tracheotomized and maintained artificially ventilated on 70 per cent N,O and 30 per cent 0,. The respirator was set to give a Paco, of 35-40 mm Hg and body temperature was kept close to 37°C. Both femoral arteries and one femoral vein were cannulated and the posterior part of the superior sagittal sinus was exposed for sampling of cerebral venous blood. Thirty min after the end of the operative procedures, and when a respiratory steady state was at hand, the blood hemoglobin concentration was reduced by gradual bleeding from one arterial catheter and gradual replacement with homologous plasma via the venous catheter. Fresh plasma was obtained from donor rats that were anesthetized with halothane (2-2.5 X). Bleeding from the arterial catheter occurred at a rate of about I ml.min-l. Infusion of plasma was done at the same rate, and the amount of plasma given was equal to the loss of arterial blood. In order to obtain blood hemoglobin concentrations of 9 and 3 g.(IOO mI)-l about 5 and 22 ml, respectively, of whole blood had to be substituted with plasma. I n these groups the final hemoglobin concentrations were thus attained after about 5 and 20 min, respectively. There were two main series of experiments. In the first series (series A) the animals were allowed a steady state period of 30 min at blood hemoglobin concentrations of about 9, 6 and 3 g.(100 mI)-l before CBF and CMRo, were measured. A control group was obtained by maintaining animals at normal hemoglobin concentration for a comparable period. This control group was identical to that previously reported (Jbhannsson and Siesjo 1974 b). CBF was measured according t o the Kety and Schmidt principle (1948), using 133Xenonand repeated sampling of arterial and cerebral venous blood during desaturation (Norberg and Siesjo 1974, see also Johansson and Siesjo 1974 b). The partition coefficient (1)used was that given by Veal and Mallet (1965), corrected for hematocrit. CMRo, was calculated by multiplying CBF with the arteriovenous difference in oxygen content (AVDo,). In each animal the AVDop used was the mean of the values measured just before and after 2-3 min of desaturation. In the second main series (series B) CBF was calculated from AVDo,, assuming constant CMRo,. Blood hemoglobin concentration was reduced t o 12, 9, 6 and 3 g.(IOO mI)-l, respectively. A control group was obtained by withdrawal of 5-10 ml of blood and substitution with the same amount of donor blood. In these animals, arterial and cerebral venous total oxygen contents (To,) were measured 2, 5, 15 and 30 min following the end of the bleeding-substitution procedure. In order to allow calculation of CBF changes
507
CBF AND CMRo, IN ANEMIA
TABLE I. Effect of acute normovolemic anemia on physiological parameters. Number of animals
Hb content g.(IOO m0-l
Body temp. "C
Arterial pressure
Pao, mm Hg
Paco, mm Hg
PH
6 6 5 6
14.8f0.4 9.0f 0.3 5.7f0.3 3.3f0.6
36.8 f0.2 36.9fO.l 36.9f0.2 36.7 f0. I
143f2 139f4 134f3 119f4
140k6 139f5 129+5 142f5
37.3 f0.1 38.5 & 0.6 38.7f0.4 3S.Of0.5
7.382k0.009 7.403 f0.008 7.401 f0.013 7.373 f 0.001
in each individual animal, AVDo, was also measured prior to reduction in hemoglobin content. CBF in anemia (a) was then derived (in per cent of control) from AVDo, before anemia (c) as [CBF],
=
100. [AVDo,]; [AVDo,],'
At each sampling occasion venous blood was also collected for measurements of Po, and arterial blood for measurements of hemoglobin content, Po2, Pco, and pH. In both series repeated withdrawal of arterial and venous blood during the measurements of CBF and CMRo, necessitated replacement via the venous catheter. In order to maintain the hemoglobin content constant the hematocrit of the infused blood was properly adjusted by mixing plasma and whole blood. Arterial and venous Pol, Pco, and pH were measured with microelectrodes operated at 37°C with due corrections for any differences in temperature between animals and electrodes. Blood hemoglobin content was measured photometrically with a Vitatron Hb-meter. Arterial and venous To, was measured polarographically on 25 ,uI samples (Fabel and Lubbers 1964, see also Borgstrom er a / . 1974 b). Statistical differences were calculated with the student's t-test. The following symbols are used: p
The Influence of Acute Normovolemic Anemia on Cerebral Blood Flow and Oxygen Consumption of Anesthetized Rats BY
LARSBORGSTROM, HALLDORJOHANNSSON and Bo K. SIESJO Received 22 October 1974
Abstract BORGSTROM, L., H. JOHANNSSON and B. K.SIESJO.The influence of acute normouolemic anemia on cerebral blood flow and oxygen consumption of anesthetized rats. Acta physiol. scand. 1975. 93. 505-514. The influence of acute normovolemic anemia on cerebral blood flow (CBF) and cerebral metabolic rate for oxygen (CMRo,) was studied in normocapnic rats under nitrous oxide anaesthesia. The arterial hemoglobin content was reduced to values of about 12, 9, 6 and 3 g.(IOO mI)-l by arterial bleeding and substitution with equal volumes of homologous plasma. The CBF increased in proportion to the reduction in hemoglobin content t o reach values of 500-600 per cent of normal at extreme degrees of anemia, but CMRo, remained unchanged. Cerebral venous Po, and oxygen saturation did not decrease below normal values, indicating that tissue hypoxia did not develop. However, since the increase in CBF at hemoglobin concentrations of below 9 g.(lOO mI)-l was far in excess of that expected from the decrease in viscosity the results indicate that dilatation of cerebral resistance vessels occurred. This dilatation, which was obviously related to the fall in arterial oxygen content, cannot be explained by any of the current theories proposed to explain cerebral hyperemia in hypoxia.
Although there is much information on cerebral blood flow (CBF) and cerebral oxygen consumption (CMRo,) in acute and chronic anemias important problems remain unsolved. Chronic anemias in man (pernicious, sickle cell, and other types) were studied by Scheinberg (1951) and by Heyman et al. (1952), who found an increased CBF and decreased CMRo,. These authors concluded that cerebral hypoxia contributed to the increase in CBF and decrease in CMRo,. Hbwever, since administration of 85-100 per cent oxygen (Heyman et al. 1952) or specific therapy (Scheinberg 1951) did not bring CMR,, back to normal, it appeared doubtful if the fall in CMRo, was due to hypoxia alone. Subsequent experiments on acute anemia have been largely designed to study the factors responsible for the cerebral hyperemia. Haggendal and Norback (1966 b) found little variation in CBF in dogs over a hematocrit range of 70-30 per cent and concluded that the increase in CBF, occurring below a hematocrit of about 30 per cent, was due to cerebral hypoxia, the decrease in viscosity playing only a minor role. Data presented by this group of workers (Haggendal and Norback 1966 a) suggested that CMRo, was maintained at a 505
506
LARS BORGSTROM, H A L L D ~ RJ ~ H A N N S S O NAND BO K.
SIESJO
hematocrit of 20 per cent. Similar results were reported by Michenfelder and Theye (1969) who reduced hemoglobin concentration from 12.4 to 5.6 ga(1OO ml)-l in the same species. In their experiments CBF increased from 77 to 108 ml.(lOO g)-l. min-l but CMRo, remained constant. Since cerebral venous Po*was reduced from 37 to 30 mm Hg cerebral hypoxia may be assumed to have contributed to the increase in CBF. In a subsequent study on man (Paulson et al. 1973), the hematocrit was reduced from 38 to 28 per cent. This procedure gave an increase in CBF by 15-20 per cent but since cerebral venous Po, remained unchanged the authors concluded that the hyperemia was largely due to reduced blood viscosity, and that cerebral hypoxia should not have been present. The present experiments were designed to study circulation and oxygen supply to the rat brain in acute normovolemic anemia. The blood hemoglobin concentration was reduced to minimal values of 3 g.(lOO m l p , i.e. to about 20 per cent of normal, with subsequent measurements of CBF, CMRo, and cerebral venous Po,. The objectives of the experiments were to study the efficiency of the circulatory adaptation in moderate to severe anemia, and to obtain further information on the cause of the hyperemia. A preliminary cornmunication on part of the present material has been published (J6hannsson and Siesjo 1974 a).
Methods Since the procedures and methods used in the present study were similar to those reported i n two previous communications (Johannsson and Siesjo 1974 b, Borgstrom et al. 1974 a) only the main outline is given here, with an emphasis on differences in procedures. The experiments were performed on unstarved male rats (350400g) that were anesthetized with 2 per cent halothane, tracheotomized and maintained artificially ventilated on 70 per cent N,O and 30 per cent 0,. The respirator was set to give a Paco, of 35-40 mm Hg and body temperature was kept close to 37°C. Both femoral arteries and one femoral vein were cannulated and the posterior part of the superior sagittal sinus was exposed for sampling of cerebral venous blood. Thirty min after the end of the operative procedures, and when a respiratory steady state was at hand, the blood hemoglobin concentration was reduced by gradual bleeding from one arterial catheter and gradual replacement with homologous plasma via the venous catheter. Fresh plasma was obtained from donor rats that were anesthetized with halothane (2-2.5 X). Bleeding from the arterial catheter occurred at a rate of about I ml.min-l. Infusion of plasma was done at the same rate, and the amount of plasma given was equal to the loss of arterial blood. In order to obtain blood hemoglobin concentrations of 9 and 3 g.(IOO mI)-l about 5 and 22 ml, respectively, of whole blood had to be substituted with plasma. I n these groups the final hemoglobin concentrations were thus attained after about 5 and 20 min, respectively. There were two main series of experiments. In the first series (series A) the animals were allowed a steady state period of 30 min at blood hemoglobin concentrations of about 9, 6 and 3 g.(100 mI)-l before CBF and CMRo, were measured. A control group was obtained by maintaining animals at normal hemoglobin concentration for a comparable period. This control group was identical to that previously reported (Jbhannsson and Siesjo 1974 b). CBF was measured according t o the Kety and Schmidt principle (1948), using 133Xenonand repeated sampling of arterial and cerebral venous blood during desaturation (Norberg and Siesjo 1974, see also Johansson and Siesjo 1974 b). The partition coefficient (1)used was that given by Veal and Mallet (1965), corrected for hematocrit. CMRo, was calculated by multiplying CBF with the arteriovenous difference in oxygen content (AVDo,). In each animal the AVDop used was the mean of the values measured just before and after 2-3 min of desaturation. In the second main series (series B) CBF was calculated from AVDo,, assuming constant CMRo,. Blood hemoglobin concentration was reduced t o 12, 9, 6 and 3 g.(IOO mI)-l, respectively. A control group was obtained by withdrawal of 5-10 ml of blood and substitution with the same amount of donor blood. In these animals, arterial and cerebral venous total oxygen contents (To,) were measured 2, 5, 15 and 30 min following the end of the bleeding-substitution procedure. In order to allow calculation of CBF changes
507
CBF AND CMRo, IN ANEMIA
TABLE I. Effect of acute normovolemic anemia on physiological parameters. Number of animals
Hb content g.(IOO m0-l
Body temp. "C
Arterial pressure
Pao, mm Hg
Paco, mm Hg
PH
6 6 5 6
14.8f0.4 9.0f 0.3 5.7f0.3 3.3f0.6
36.8 f0.2 36.9fO.l 36.9f0.2 36.7 f0. I
143f2 139f4 134f3 119f4
140k6 139f5 129+5 142f5
37.3 f0.1 38.5 & 0.6 38.7f0.4 3S.Of0.5
7.382k0.009 7.403 f0.008 7.401 f0.013 7.373 f 0.001
in each individual animal, AVDo, was also measured prior to reduction in hemoglobin content. CBF in anemia (a) was then derived (in per cent of control) from AVDo, before anemia (c) as [CBF],
=
100. [AVDo,]; [AVDo,],'
At each sampling occasion venous blood was also collected for measurements of Po, and arterial blood for measurements of hemoglobin content, Po2, Pco, and pH. In both series repeated withdrawal of arterial and venous blood during the measurements of CBF and CMRo, necessitated replacement via the venous catheter. In order to maintain the hemoglobin content constant the hematocrit of the infused blood was properly adjusted by mixing plasma and whole blood. Arterial and venous Pol, Pco, and pH were measured with microelectrodes operated at 37°C with due corrections for any differences in temperature between animals and electrodes. Blood hemoglobin content was measured photometrically with a Vitatron Hb-meter. Arterial and venous To, was measured polarographically on 25 ,uI samples (Fabel and Lubbers 1964, see also Borgstrom er a / . 1974 b). Statistical differences were calculated with the student's t-test. The following symbols are used: p
