Toxicology. 73 ( 1992) 169- 178 Elsevier Scientific Publishers Ireland
169 Ltd.
Blood lactate and catecholamine levels in the carbon monoxide-exposed rat: the response to elevated glucose Bharat
B. Sutariya”, David G. Penneyb, Joseph and Curtis J. Swansona
C. Dunbarb
‘Department of Biological Sciences and bDepartment of Physiology, Wayne State University School of Medicine, Deiroii. MI 48201 (USA) (Received
November
20th,
1991; accepted
March
6th, 1992)
Summary Previous studies have shown that elevated blood glucose is detrimental to the outcome in acute carbon monoxide (CO) poisoning. The present goals were to characterize the blood lactate and catecholamine changes and to determine whether elevated blood glucose results in increases in the levels of these substances. Two groups of adult Sprague-Dawley, Levine-prepared, female rats (n = 22 each) were exposed to 2400 ppm CO for 90 min: one group received nothing (CO alone), while the other group was infused with a 50% glucose solution (4 ml/kg) (CO + glucose). The usual hypothermia, hypotension, bradycardia and hemoconcentration associated with acute severe CO poisoning were observed. Survival rates were 68% and 54% in the CO alone and CO + glucose groups, respectively. Arterial blood pressure tended to decline more in rats that died; the difference was significant in the CO + glucose group. In the CO alone group, plasma glucose concentration was significantly lower after CO exposure in rats that died than in survivors (35 +Z 15 vs. 99 f 16 mgdl). In the CO + glucose group, glucose concentration was significantly higher after 45 min in rats that died (d) than in survivors (s) (447 f 29 vs. 324 f 31 mgidl). Elevated blood glucose in the CO + glucose group failed to significantly increase blood lactate; however, lactate tended to be higher in rats that died in both groups (CO alone group: 175 f 17 (d) vs. 138 f 9 (s); CO + glucose group: 154 * 10 (d) vs. 143 f 8 (s) 1.Plasma epinephrine and norepinephrine increased significantly 6-lo-fold and 2-6-fold in each of the two groups, respectively; however, catecholamine levels were not related to either the administration of glucose or survival. With regard to CO poisoning in this animal model, the results do not support the hypotheses that elevated blood glucose exacerbates the increase in blood lactate, that increased catecholamine increases glucose, or that greater CO-induced hypoglycemia results from increased lactate production. The results do show that death is related to abnormally high or low blood glucose, but that it is not due to higher blood lactate or catecholamine levels. Key words: Blood Pressure; Catecholamines; Carbon Hypothermia; Lactate; Neurologic index; Survival
monoxide;
Glucose;
Hematocrit;
Hypotension;
Introduction Carbon monoxide variety of immediate
(CO) is a poisonous and delayed morbid
gas which causes death, as well as a wide effects, brain damage being the most com-
Correspondence fo: David G. Penney, Department of Physiology and Occupational and Environmental Health, Wayne State University School of Medicine, 540 E. Canfield, Detroit, MI 48201, USA.
0300-483X’92/$05.00 0 1992 Elsevier Scientific Publishers Printed and Published in Ireland
Ireland
Ltd.
170
mon. In the rat, acute severe CO poisoning is characterized by hypotension, hypothermia, bradycardia, hemoconcentration and hyperglycemia [ 1,2]. A recent study using the Levine-prepared rat, demonstrated increased neurologic deficit and general morbidity following 90-min CO exposure in subjects experiencing elevated blood glucose [3]. Post CO hyperglycemia has also been linked to increased severity of brain dysfunction in humans acutely poisoned with CO [4]. Thus, the plasma glucose level appears to be an important determinant of cerebral viability and survival during the hypoxia and relative ischemia of CO poisoning. Whether elevated blood glucose permits an increased rate of glycolysis, generating more lactate, hence increased blood and tissue acidosis, is unknown. This was one issue which we wished to address. Several studies have reported elevated blood and urine catecholamine levels in acutely CO-poisoned animals [5,6]. Whether catecholamine, acting on hepatic /3adrenergic receptors, could be a major stimulus for increased glucose mobilization, thus hyperglycemia resulting from CO poisoning, is also unknown. This was a second issue we wished to address. Thus, it was deemed essential to characterize the blood lactate and catecholamine changes during CO exposure in a defined animal model and attempt to relate them to changes in blood glucose. Our studies have also shown that hypoglycemia is increasingly frequent and severe in rats exposed to CO at levels above 2400 ppm, with death occurring when plasma glucose levels fall below 40 mg/dl [7]. Extreme hypoglycemia also occurs in rats when CO-induced hypothermia is prevented, i.e., body temperature is clamped (unpublished data). We have suggested that in such cases, plasma glucose is being converted to lactate faster than it can be resupplied to the blood by the glucogenic organs. Consequently, such hypoglycemic subjects should have higher plasma lactate levels than the normo- or hyperglycemic subjects. This was a third issue we wished to address. This study was carried out in an animal model in which the metabolic, cardiovascular and neurologic responses to acute severe CO poisoning have been previously characterized [ l-3,7-9]. The goals were as follows: (i) to characterize the blood lactate and catecholamine changes which occur in response to acute CO poisoning, both in survivors and in those that succumb; (ii) to determine whether there is a relationship between catecholamine level and blood glucose; (iii) to determine whether elevated blood glucose increases lactate and catecholamine levels and (iv) to determine whether extreme hypoglycemia during CO exposure can be explained by higher lactate levels. Materials and methods Experimental design and CO exposure Prior (2 days) to the experiment, a modified Levine preparation [lo] was produced in female Sprague-Dawley rats (Charles River Breeding Labs., Wilmington, MA) 100-200 days of age (body weight: CO alone, 257 f 4 (n = 22); CO + glucose, 264 + 3 (n = 22). In all animals the left common carotid artery was catheterized with PE-50 polyethylene tubing under ketamine (80 mg/kg)/Rompun (4.84 mgkg) anesthesia, as previously described [l]. In it, catheters made of PE-50 polyethylene
171
tubing were inserted into both the jugular vein and common carotid artery toward the heart. The catheters were threaded under the skin to the nape of the neck and tied in place. The lo-cm external length of each catheter tube was plugged with an Amphenol pin (#220-PO2-100) and the catheter coiled up with masking tape when not in use. This procedure effectively occludes the major vessels to one side of the brain, placing it at increased hypoxic/ischemic risk, while providing ports for blood withdrawal, infusion of glucose and the monitoring of vital functions. The rats were housed individually in plastic shoebox cages and were fed ad libitum. On the day of the experiment, the rats were confined in plastic restrainers while providing easy access to both cannulas. The rats were allowed to recover under close observation; they were not behaviorally different from unoperated controls. Prior to confinement, all animals were assessed for neurologic damage using a standard behavioral scale [l-3] which might have been caused by the surgery. The rats were randomly assigned to one of the two treatment groups as follows: (i) those to inhale 2400 ppm CO (‘CO alone’) and (ii) those to inhale 2400 ppm CO plus receive a 50% glucose solution by jugular infusion (4 ml of 50% glucose (w/v in saline)/kg body wt) (‘CO + glucose’). The rats were confined in plastic restrainers and exposed to 2400 ppm CO for 90 min inside a large transparent plastic bag. CO exposure occurred in the absence of anesthesia. Two rats were exposed simultaneously. The glucose solution was infused through the jugular catheter using a Harvard Apparatus Co. infusion/withdrawal pump (model 940). Two-thirds of the volume was given over the first 20 min of CO exposure and one-third over the subsequent 70 min. Following CO exposure, the rats were immediately transferred to room air for recovery. The air-CO mixture was made up by mixing air from a small carbon vane compressor (40 Urnin) with CO from a commercial cylinder (Cryogenic Gases, Detroit, MI) as needed. CO concentration was continuously monitored with a Beckman model 870, solid state infrared analyzer (Beckman, Lombard, IL), calibrated with a standard CO (2900 ppm)-nitrogen gas mixture. Measurements
Blood was withdrawn from the carotid cannula at several time periods for the measurement of hematocrit, glucose, lactate and catecholamines. Blood glucose and lactate values were determined on a 20-~1 blood sample, using a Yellow Springs Instrument Co., 2300 STAT, Glucose/Lactate Analyzer. Catecholamine measurements were made using high pressure, liquid chromatography, with electrochemical detection, using a modification of the method of Mefford [ 111. Hematocrit was determined in duplicate by the microhematocrit method. Measurements were made 0, 45, 90, 210 and 330 min after the start of CO exposure. Rectal temperature was monitored with a Yellow Springs Instrument Co. thermometer model 43TD, using YSI 400 series probes. Blood pressure was measured with a Statham P23id pressure transducer and recorded on a Gould model 2400 chart recorder. Heart rate was derived from the blood pressure record. Deaths were counted if they occurred during or 12 h after CO exposure. Data analysis and graphic display were carried out on a Macintosh microcomputer, using the Excel and Cricket-Graph programs. Most values are means f SEM.
172
Student’s t-test was used for statistical analysis. of 0.05 or smaller were considered significant.
Differences
that resulted
in P-values
Results Significant decreases in body temperature (3-4”C), blood pressure (38-44 mmHg) and heart rate (60-80 beats/min) were observed during CO exposure both in rats infused with glucose and in those given no additional treatment (Fig. 1). Decrease in body temperature was significantly less (P < 0.01) in rats given glucose. Hematocrit increased significantly 2-4% in both groups; the CO alone group showed a significantly greater increase (P < 0.01) after 45 min of exposure. The survival rate was 68% of 22 rats in the CO alone group and 54% of 22 rats in the CO + glucose
-30
0
30
60
90
120
150
180
210
240
270
300
330
TIME (min.1 Fig. 1. Changes in various parameters in rats exposed to 2400 ppm CO for 90 min (circle, CO alone; square, CO + glucose). Vertical bars are 1 SEM. Animal numbers indicated in parentheses. Comparing values at same time: ++, P < 0.01.
173
group (not shown). This difference as calculated by the Chi-square method was not significant (P between 0.10 and 0.90). All parameters had returned to initial values by 4 h of recovery. Blood glucose concentration was significantly lower at the end of CO exposure in the CO alone rats that died than in the survivors (35 * 15 mg/dl vs. 99 f 16 mg/dl, respectively) (Fig. 2). In the CO + glucose group, glucose was significantly higher after 45 min of CO exposure in rats that died than in survivors (447 f 29 mg/dl vs. 324 * 31 mgdl). Arterial blood pressure declined more in the rats that died than in the survivors in both the CO alone and the CO + glucose groups (died, 49 mmHg; survived, 40 mmHg) (not shown). There was no significant difference in blood lactate concentration between rats that died and those that survived, whether or not they received exogenous glucose,
20
1 +++
Ol.',"~"~"~"""."'.'.“"'""" -30 0 30 60 90
120
150
160
210
240
270
300
330
600
500
6
,'"I
400 15)
5300
(12)
E 200
(12) (10)
(121
100 b -30
0
30
60
90
120
150
160
210
240
270
300
330
Time (min.) Fig. 2. Blood glucose concentrations of rats exposed to 2400 ppm CO for 90 min (circle, CO alone; square, CO + glucose; open symbols, survived; closed symbols, died (within 12 h of CO exposure)). Vertical bars are 1 SEM. Animal numbers indicated in parentheses. Comparing values at same time: +++, P < 0.001.
174 240
Ol",.'~"~"~"~'.~“~"""‘~'""" -30 0 30 60 90
J
120
150
160
210
240
270
300
330
300
330
160-
?. 0
120-
E 60-
01"~"~"'~'~"'~'~"~'~~"'"~"~"" -30 0 30 60 90
120
150
180
210
240
270
TIME (min.1 Fig. 3. Blood lactate concentrations of rats exposed to 2400 ppm CO for 90 min. Abbreviations and symbols the same as in Fig. 2. Fig. 4. Plasma catecholamine concentrations of rats exposed to 2400 ppm CO for 90 min. Abbreviations and symbols the same as earlier. Comparing values before and after CO exposure: +, P < 0.05; ++, P < 0.01: +++, P < 0.001.
although there was a tendency for higher lactate concentration in rats that died, especially in the CO alone group at 90 min (P < 0.06) (Fig. 3). The elevation of blood glucose in the CO + glucose group after 45 min of CO exposure (died + survivors, 380 A 25 mg/dl) failed to produce a significantly higher lactate concentration than in the CO alone group (150 + 7 mg/dl, in both groups). There was no correlation between blood glucose concentration and lactate concentration at either 45 or 90 min of CO exposure, in either group. Plasma epinephrine and norepinephrine increased significantly in both of the experimental groups following CO exposure, . however, the infusion of exogenous glucose failed to alter the rise in catecholamine concentration (Fig. 4). There were
175 1800 1600 El
1400
_ ,
E
GLUCOSE/ALIVE
1200 1000
EJ) Q
800 600 400 200 0
1200
Norepinephrine 1000
+
q GLUCOSE/ALIVE H
GLUCOSE/DEAD
800
+++
Time
(min.)
Fig. 4. Plasma catecholamine concentrations of rats exposed and symbols the same as earlier. Comparing values before
to 2400 ppm CO for 90 min. Abbreviations and after CO exposure; +, P < 0.05; ++.
P < 0.01; +++, P < 0.001.
no significant differences in catecholamine concentration in either group between rats that died and rats that survived. No correlation was seen between blood glucose concentration and catecholamine concentration, in either group, or between blood lactate concentration and catecholamine concentration.
Discussion The results of this study demonstrated the usual effects of acute severe CO poisoning in the rat, hypothermia, hemoconcentration, hypotension, bradycardia and altered blood glucose concentration [ 1,2]. The loss of normal body temperature results from decreased heat production through an inhibition of oxidative
176
metabolism and an increased heat loss due to peripheral vasodilation [8]. Decreased arterial blood pressure is a function of the peripheral vasodilation and depressed cardiac pump function involving both decreased heart rate and stroke volume. The rise in hematocrit is presumed to be due to the loss of plasma volume resulting from an increased endothelial permeability [ 121. Blood glucose concentration usually rises during CO exposure and then once again following CO exposure [7]. We have referred to these two glucose peaks, as a ‘camel-back’ pattern. This pattern was seen in the present study in the rats treated with CO alone. The magnitude of the initial peak and the depth of the valley following, are exacerbated by increased CO concentration (> 2400 ppm), saline infusion and the forced maintenance of normal body temperature. Rats not given exogenous glucose that die during CO exposure almost invariably have a blood glucose concentration below 40 mg/dl [7]. Recent studies have demonstrated an increased neurologic deficit and general morbidity following 90 min of CO exposure in Levine-prepared rats experiencing elevated blood glucose [2,3]. This was the case whether the elevated glucose resulted from CO exposure alone [3] or from CO exposure combined with the infusion of exogenous glucose [2]. The morphological correlate of neurologic deficit is cerebral cortical edema, which has been assessed both by measuring gross water content [1] and by using magnetic resonance imaging techniques (Jalukar et al., unpublished data). The elevations in blood glucose concentrations achieved during CO exposure in the present study in both the CO alone and CO + glucose rats were similar in magnitude to those reported previously [2]. Also like previous reports, rats experiencing either greatly depressed glucose levels during CO exposure, or greatly elevated glucose during CO exposure, showed the highest mortality. Thus, glucose level appears to be an important determinant of cerebral viability and survival during the hypoxia and relative ischemia (i.e., oligemia) of CO poisoning. It was hypothesized that an elevated blood glucose would permit an increased rate of glycolysis, thus generating more lactate. Our data indicate that this is not the case in the rats that received exogenous glucose and that blood lactate increased some lo-12-fold regardless of the blood glucose concentration. This observation has been confirmed in another study (Jalukar et al., unpublished data). It is possible of course that regional brain lactate concentration departed significantly from that of the carotid blood. The opposite condition was also not true, i.e., that severely hypoglycemic rats had depressed blood glucose because of an increased conversion of glucose to lactate. This suggests that the reason for the depressed blood glucose was a decrease in glucose release. Sokal [13] found that blood lactate increased from 9- 162 mg/dl in rats exposed to 4000 ppm CO for 40 min, an 1%fold increase. Considering that the CO concentration he used was nearly double that of the present study, the similarity in response was excellent. Sokal also noted a nearly 3-fold increase in blood glucose concentration, while pH fell to 6.85. Similar changes in blood lactate and glucose were reported by McGrath and Lee [ 141, in rats acutely inhaling CO. Using mice inhaling 3500 ppm CO, Moore et al. [ 151 found that blood lactate reached 90 mgdl after 11.5 min, while pH fell to only 7.25, the more modest acidosis probably due to the short exposure period. Measurements of regional brain lactate concentration have shown
177
it to increase to similar high levels in anesthetized rats ventilated with up to 20 000 ppm CO [16]. Although we did not measure blood pH, it is likely that levels of acidosis similar to those seen by Moore et al. were achieved in our study. Jn the past, a number of investigators have suggested that hypoxic/ischemic brain damage is the direct result of excessive lactate production and the attendant acidosis [17] and that brain lactate can be used to calculate brain pH. Recent nuclear magnetic resonance studies of brain lactate and pH, however, suggest that the two become dissociated during hypoxia/ischemia [18]. Although there were no differences in the blood lactate levels in our rats, it is still possible that that brain pH was different in the survivors versus the non-survivors. Evidence suggests that carboxyhemoglobinemia causes increased stimulation of the sympathetic nervous system. For example, CO-induced arterial hypotension initiates the carotid baroreflex, producing the tachycardia which is observed prior to hypoxic pacemaker depression [ 11. The peripheral chemoreceptors, especially the aortic bodies, are sensitive to CO [19] and increase their firing rate in its presence. Elevated plasma catecholamine levels were observed in anesthetized dogs ventilated with an air-CO mixture [6]. Catecholamine urinary excretion is also reportedly increased up to 5-fold in rats receiving CO by subcutaneous injection [5]. It has been suggested that the elevated catecholamine, acting on hepatic /3-adrenergic receptors, could be a major stimulus for increased glucose mobilization and hence blood levels. Our data failed to show any relationship between either the elevation of epinephrine or norepinephrine and blood glucose. Nevertheless, the inhibition of catecholamine release in vivo has been found to block CO-induced hyperglycemia [20]. On the other hand, the enhanced glycogenolysis and lactate production in the brain are reportedly not prevented in CO-exposed mice by pre-treatment with reserpine or a P-adrenergic blocking agent [21]. In summary, our data show that elevated blood glucose during acute CO poisoning does not further increase the blood lactate level above that present when glucose is normal or depressed, although mortality is increased at very high and very low glucose concentrations. The data also fail to show any relationship between the COinduced elevation of catecholamine concentrations and the rise in blood glucose. Thirdly, the results show that CO-induced hypoglycemia is not accompanied by a greater elevation of blood lactate. Finally, the data do not support the notion that mortality is related to the magnitude of the elevation in blood lactate or catecholamine concentrations. It remains unclear by what mechanism mortality is increased in acute CO poisoning. References D.G. Penney, K. Verma and J.A. Hull, Cardiovascular, metabolic and neurologic carbon monoxide poisoning in the rat. Toxicol. Letts., 45 (1989) 207.
effects of acute
D.G. Penney, CC. Helfman, J.C. Dunbar and L.E. McCoy, Acute severe carbon monoxide poisoning in the rat: Effects of hyperglycemia and hypoglycemia on mortality, recovery and neurologic deficit. Can. J. Physiol. Phannacol., 69 (1991) 1168. D.G. Penney, C.C. Helfman, J.A. Hull, J.C. Dunbar and K. Verma, Elevated blood glucose is associated with poor outcome in the carbon monoxide-poisoned rat. Toxicol. Letts., 54 (1990) 287. D.G. Penney, Hyperglycemia exacerbates brain damage in acute severe carbon monoxide poisoning. Med. Hypotheses, 27 (1988) 241.
178
5
6
7 8 9 10 11
12 13
14 15 16 17 18
19 20 21
D. Pankow
and W. Ponsold,
Effect of single and repeated
carbon
monoxide
intoxications
on urinary
catecholamine excretion in rats. Acta Biol. Med. Ger., 37 (1978) 1589. J.T. Sylvester, S.M. Scharf, R.D. Gilbert, R.S. Fitzgerald and R.J. Traystman, Hypoxic and CO hypoxia in dogs: Hemodynamics, carotid reflexes and catecholamines. Am. J. Physiol., 236 (1979) H22. D.G. Penney, P. Sharma, B.B. Sutariya and B.C. Nallamothu, Development of hypoglycemia is associated with death during carbon monoxide poisoning. J. Crit. Care, 5 (1990) 1. D.G. Penney, Acute carbon monoxide poisoning: Animal models. Toxicology, 62 (1990) 123. D.G. Penney, Modifying role of plasma glucose in acute carbon monoxide poisoning. Arch. Toxicol. (Recent developments in toxicology: Trends, methods and problems) Suppl., 14 (1991) 239. S. Levine, Anoxic ischemic encephalopathy in rats. Am. J. Pathol., 36 (1960) I. I.N. Mefford, Application of high performance liquid chromatography with electrochemical detection to neurochemical analysis: Measurement of catecholamines, serotonin and metabolites in rat brain. J. Neurosci. Methods, 3 (1981) 207. M. Ogawa, S. Shimazaki, N. Tanaka, T. Ukai and T. Sugimoto. Blood volume changes in experimental carbon monoxide poisoning. Ind. Health, 12 (1974) 121. J.A. Sokal, Lack of the correlation between biochemical effects on rats and blood carboxyhemoglobin concentrations in various conditions of single acute exposure to carbon monoxide. Arch. Toxicol., 34 (1975) 331. J.J. McGrath, K.C. Chen and P.S. Lee, Effects of intraperitoneal injection of carbon monoxide on circulating blood glucose and lactate. Fed. Proc.. 38 (1979) 1313. S.J. Moore, I.K. Ho and A.S. Hume, Severe hypoxia produced by concomitant intoxication with sublethal doses of carbon monoxide and cyanide. Toxicol. Appl. Pharmacol., 109 (1991) 412. V. MacMillan, Regional cerebral energy metabolism in acute carbon monoxide intoxication. Can. J. Physiol. Pharmacol., 55 (1977) 1I I. S. Rehncrona, I. Rosen and B.K. Siesjo. Excessive cellular acidosis: An important mechanism of neuronal damage in the brain. Acta Physiol. Stand., I IO (1980) 435. S. Lotito, P. Blondet, A. Francois, M. von Kienlin, C. Remy, J.P. Albrand. M. Decorps and A.L. Benabid, Correlation between intracellular pH and lactate levels in the rat brain during potassium cyanide induced metabolism blockade: A combined “P-‘H in vivo nuclear magnetic spectroscopy study. Neurosci. Lett., 97 (1989) 91. E. Mills and M.W. Edwards Jr., Stimulation of aortic and carotid chemoreceptors during carbon monoxide inhalation. J. Appl. Physiol., 25 (1968) 494. M. Gothert, P. bon monoxide C.-J. Estler, glycogenolysis
Hock, G. Malorney and H.-J. Seiler, Mechanism of hyperglycemia during acute carpoisoning. Naunyn-Schmiedeberg’s Arch. Pharmacol. (SuppI.), 274 (1972) R4l. F. Heim, H.P.T. Ammon and V. Zimmermann, Catecholamine-independent in brain during carbon monoxide poisoning. Pharmacology, 5 (197 I) 55.
169 Ltd.
Blood lactate and catecholamine levels in the carbon monoxide-exposed rat: the response to elevated glucose Bharat
B. Sutariya”, David G. Penneyb, Joseph and Curtis J. Swansona
C. Dunbarb
‘Department of Biological Sciences and bDepartment of Physiology, Wayne State University School of Medicine, Deiroii. MI 48201 (USA) (Received
November
20th,
1991; accepted
March
6th, 1992)
Summary Previous studies have shown that elevated blood glucose is detrimental to the outcome in acute carbon monoxide (CO) poisoning. The present goals were to characterize the blood lactate and catecholamine changes and to determine whether elevated blood glucose results in increases in the levels of these substances. Two groups of adult Sprague-Dawley, Levine-prepared, female rats (n = 22 each) were exposed to 2400 ppm CO for 90 min: one group received nothing (CO alone), while the other group was infused with a 50% glucose solution (4 ml/kg) (CO + glucose). The usual hypothermia, hypotension, bradycardia and hemoconcentration associated with acute severe CO poisoning were observed. Survival rates were 68% and 54% in the CO alone and CO + glucose groups, respectively. Arterial blood pressure tended to decline more in rats that died; the difference was significant in the CO + glucose group. In the CO alone group, plasma glucose concentration was significantly lower after CO exposure in rats that died than in survivors (35 +Z 15 vs. 99 f 16 mgdl). In the CO + glucose group, glucose concentration was significantly higher after 45 min in rats that died (d) than in survivors (s) (447 f 29 vs. 324 f 31 mgidl). Elevated blood glucose in the CO + glucose group failed to significantly increase blood lactate; however, lactate tended to be higher in rats that died in both groups (CO alone group: 175 f 17 (d) vs. 138 f 9 (s); CO + glucose group: 154 * 10 (d) vs. 143 f 8 (s) 1.Plasma epinephrine and norepinephrine increased significantly 6-lo-fold and 2-6-fold in each of the two groups, respectively; however, catecholamine levels were not related to either the administration of glucose or survival. With regard to CO poisoning in this animal model, the results do not support the hypotheses that elevated blood glucose exacerbates the increase in blood lactate, that increased catecholamine increases glucose, or that greater CO-induced hypoglycemia results from increased lactate production. The results do show that death is related to abnormally high or low blood glucose, but that it is not due to higher blood lactate or catecholamine levels. Key words: Blood Pressure; Catecholamines; Carbon Hypothermia; Lactate; Neurologic index; Survival
monoxide;
Glucose;
Hematocrit;
Hypotension;
Introduction Carbon monoxide variety of immediate
(CO) is a poisonous and delayed morbid
gas which causes death, as well as a wide effects, brain damage being the most com-
Correspondence fo: David G. Penney, Department of Physiology and Occupational and Environmental Health, Wayne State University School of Medicine, 540 E. Canfield, Detroit, MI 48201, USA.
0300-483X’92/$05.00 0 1992 Elsevier Scientific Publishers Printed and Published in Ireland
Ireland
Ltd.
170
mon. In the rat, acute severe CO poisoning is characterized by hypotension, hypothermia, bradycardia, hemoconcentration and hyperglycemia [ 1,2]. A recent study using the Levine-prepared rat, demonstrated increased neurologic deficit and general morbidity following 90-min CO exposure in subjects experiencing elevated blood glucose [3]. Post CO hyperglycemia has also been linked to increased severity of brain dysfunction in humans acutely poisoned with CO [4]. Thus, the plasma glucose level appears to be an important determinant of cerebral viability and survival during the hypoxia and relative ischemia of CO poisoning. Whether elevated blood glucose permits an increased rate of glycolysis, generating more lactate, hence increased blood and tissue acidosis, is unknown. This was one issue which we wished to address. Several studies have reported elevated blood and urine catecholamine levels in acutely CO-poisoned animals [5,6]. Whether catecholamine, acting on hepatic /3adrenergic receptors, could be a major stimulus for increased glucose mobilization, thus hyperglycemia resulting from CO poisoning, is also unknown. This was a second issue we wished to address. Thus, it was deemed essential to characterize the blood lactate and catecholamine changes during CO exposure in a defined animal model and attempt to relate them to changes in blood glucose. Our studies have also shown that hypoglycemia is increasingly frequent and severe in rats exposed to CO at levels above 2400 ppm, with death occurring when plasma glucose levels fall below 40 mg/dl [7]. Extreme hypoglycemia also occurs in rats when CO-induced hypothermia is prevented, i.e., body temperature is clamped (unpublished data). We have suggested that in such cases, plasma glucose is being converted to lactate faster than it can be resupplied to the blood by the glucogenic organs. Consequently, such hypoglycemic subjects should have higher plasma lactate levels than the normo- or hyperglycemic subjects. This was a third issue we wished to address. This study was carried out in an animal model in which the metabolic, cardiovascular and neurologic responses to acute severe CO poisoning have been previously characterized [ l-3,7-9]. The goals were as follows: (i) to characterize the blood lactate and catecholamine changes which occur in response to acute CO poisoning, both in survivors and in those that succumb; (ii) to determine whether there is a relationship between catecholamine level and blood glucose; (iii) to determine whether elevated blood glucose increases lactate and catecholamine levels and (iv) to determine whether extreme hypoglycemia during CO exposure can be explained by higher lactate levels. Materials and methods Experimental design and CO exposure Prior (2 days) to the experiment, a modified Levine preparation [lo] was produced in female Sprague-Dawley rats (Charles River Breeding Labs., Wilmington, MA) 100-200 days of age (body weight: CO alone, 257 f 4 (n = 22); CO + glucose, 264 + 3 (n = 22). In all animals the left common carotid artery was catheterized with PE-50 polyethylene tubing under ketamine (80 mg/kg)/Rompun (4.84 mgkg) anesthesia, as previously described [l]. In it, catheters made of PE-50 polyethylene
171
tubing were inserted into both the jugular vein and common carotid artery toward the heart. The catheters were threaded under the skin to the nape of the neck and tied in place. The lo-cm external length of each catheter tube was plugged with an Amphenol pin (#220-PO2-100) and the catheter coiled up with masking tape when not in use. This procedure effectively occludes the major vessels to one side of the brain, placing it at increased hypoxic/ischemic risk, while providing ports for blood withdrawal, infusion of glucose and the monitoring of vital functions. The rats were housed individually in plastic shoebox cages and were fed ad libitum. On the day of the experiment, the rats were confined in plastic restrainers while providing easy access to both cannulas. The rats were allowed to recover under close observation; they were not behaviorally different from unoperated controls. Prior to confinement, all animals were assessed for neurologic damage using a standard behavioral scale [l-3] which might have been caused by the surgery. The rats were randomly assigned to one of the two treatment groups as follows: (i) those to inhale 2400 ppm CO (‘CO alone’) and (ii) those to inhale 2400 ppm CO plus receive a 50% glucose solution by jugular infusion (4 ml of 50% glucose (w/v in saline)/kg body wt) (‘CO + glucose’). The rats were confined in plastic restrainers and exposed to 2400 ppm CO for 90 min inside a large transparent plastic bag. CO exposure occurred in the absence of anesthesia. Two rats were exposed simultaneously. The glucose solution was infused through the jugular catheter using a Harvard Apparatus Co. infusion/withdrawal pump (model 940). Two-thirds of the volume was given over the first 20 min of CO exposure and one-third over the subsequent 70 min. Following CO exposure, the rats were immediately transferred to room air for recovery. The air-CO mixture was made up by mixing air from a small carbon vane compressor (40 Urnin) with CO from a commercial cylinder (Cryogenic Gases, Detroit, MI) as needed. CO concentration was continuously monitored with a Beckman model 870, solid state infrared analyzer (Beckman, Lombard, IL), calibrated with a standard CO (2900 ppm)-nitrogen gas mixture. Measurements
Blood was withdrawn from the carotid cannula at several time periods for the measurement of hematocrit, glucose, lactate and catecholamines. Blood glucose and lactate values were determined on a 20-~1 blood sample, using a Yellow Springs Instrument Co., 2300 STAT, Glucose/Lactate Analyzer. Catecholamine measurements were made using high pressure, liquid chromatography, with electrochemical detection, using a modification of the method of Mefford [ 111. Hematocrit was determined in duplicate by the microhematocrit method. Measurements were made 0, 45, 90, 210 and 330 min after the start of CO exposure. Rectal temperature was monitored with a Yellow Springs Instrument Co. thermometer model 43TD, using YSI 400 series probes. Blood pressure was measured with a Statham P23id pressure transducer and recorded on a Gould model 2400 chart recorder. Heart rate was derived from the blood pressure record. Deaths were counted if they occurred during or 12 h after CO exposure. Data analysis and graphic display were carried out on a Macintosh microcomputer, using the Excel and Cricket-Graph programs. Most values are means f SEM.
172
Student’s t-test was used for statistical analysis. of 0.05 or smaller were considered significant.
Differences
that resulted
in P-values
Results Significant decreases in body temperature (3-4”C), blood pressure (38-44 mmHg) and heart rate (60-80 beats/min) were observed during CO exposure both in rats infused with glucose and in those given no additional treatment (Fig. 1). Decrease in body temperature was significantly less (P < 0.01) in rats given glucose. Hematocrit increased significantly 2-4% in both groups; the CO alone group showed a significantly greater increase (P < 0.01) after 45 min of exposure. The survival rate was 68% of 22 rats in the CO alone group and 54% of 22 rats in the CO + glucose
-30
0
30
60
90
120
150
180
210
240
270
300
330
TIME (min.1 Fig. 1. Changes in various parameters in rats exposed to 2400 ppm CO for 90 min (circle, CO alone; square, CO + glucose). Vertical bars are 1 SEM. Animal numbers indicated in parentheses. Comparing values at same time: ++, P < 0.01.
173
group (not shown). This difference as calculated by the Chi-square method was not significant (P between 0.10 and 0.90). All parameters had returned to initial values by 4 h of recovery. Blood glucose concentration was significantly lower at the end of CO exposure in the CO alone rats that died than in the survivors (35 * 15 mg/dl vs. 99 f 16 mg/dl, respectively) (Fig. 2). In the CO + glucose group, glucose was significantly higher after 45 min of CO exposure in rats that died than in survivors (447 f 29 mg/dl vs. 324 * 31 mgdl). Arterial blood pressure declined more in the rats that died than in the survivors in both the CO alone and the CO + glucose groups (died, 49 mmHg; survived, 40 mmHg) (not shown). There was no significant difference in blood lactate concentration between rats that died and those that survived, whether or not they received exogenous glucose,
20
1 +++
Ol.',"~"~"~"""."'.'.“"'""" -30 0 30 60 90
120
150
160
210
240
270
300
330
600
500
6
,'"I
400 15)
5300
(12)
E 200
(12) (10)
(121
100 b -30
0
30
60
90
120
150
160
210
240
270
300
330
Time (min.) Fig. 2. Blood glucose concentrations of rats exposed to 2400 ppm CO for 90 min (circle, CO alone; square, CO + glucose; open symbols, survived; closed symbols, died (within 12 h of CO exposure)). Vertical bars are 1 SEM. Animal numbers indicated in parentheses. Comparing values at same time: +++, P < 0.001.
174 240
Ol",.'~"~"~"~'.~“~"""‘~'""" -30 0 30 60 90
J
120
150
160
210
240
270
300
330
300
330
160-
?. 0
120-
E 60-
01"~"~"'~'~"'~'~"~'~~"'"~"~"" -30 0 30 60 90
120
150
180
210
240
270
TIME (min.1 Fig. 3. Blood lactate concentrations of rats exposed to 2400 ppm CO for 90 min. Abbreviations and symbols the same as in Fig. 2. Fig. 4. Plasma catecholamine concentrations of rats exposed to 2400 ppm CO for 90 min. Abbreviations and symbols the same as earlier. Comparing values before and after CO exposure: +, P < 0.05; ++, P < 0.01: +++, P < 0.001.
although there was a tendency for higher lactate concentration in rats that died, especially in the CO alone group at 90 min (P < 0.06) (Fig. 3). The elevation of blood glucose in the CO + glucose group after 45 min of CO exposure (died + survivors, 380 A 25 mg/dl) failed to produce a significantly higher lactate concentration than in the CO alone group (150 + 7 mg/dl, in both groups). There was no correlation between blood glucose concentration and lactate concentration at either 45 or 90 min of CO exposure, in either group. Plasma epinephrine and norepinephrine increased significantly in both of the experimental groups following CO exposure, . however, the infusion of exogenous glucose failed to alter the rise in catecholamine concentration (Fig. 4). There were
175 1800 1600 El
1400
_ ,
E
GLUCOSE/ALIVE
1200 1000
EJ) Q
800 600 400 200 0
1200
Norepinephrine 1000
+
q GLUCOSE/ALIVE H
GLUCOSE/DEAD
800
+++
Time
(min.)
Fig. 4. Plasma catecholamine concentrations of rats exposed and symbols the same as earlier. Comparing values before
to 2400 ppm CO for 90 min. Abbreviations and after CO exposure; +, P < 0.05; ++.
P < 0.01; +++, P < 0.001.
no significant differences in catecholamine concentration in either group between rats that died and rats that survived. No correlation was seen between blood glucose concentration and catecholamine concentration, in either group, or between blood lactate concentration and catecholamine concentration.
Discussion The results of this study demonstrated the usual effects of acute severe CO poisoning in the rat, hypothermia, hemoconcentration, hypotension, bradycardia and altered blood glucose concentration [ 1,2]. The loss of normal body temperature results from decreased heat production through an inhibition of oxidative
176
metabolism and an increased heat loss due to peripheral vasodilation [8]. Decreased arterial blood pressure is a function of the peripheral vasodilation and depressed cardiac pump function involving both decreased heart rate and stroke volume. The rise in hematocrit is presumed to be due to the loss of plasma volume resulting from an increased endothelial permeability [ 121. Blood glucose concentration usually rises during CO exposure and then once again following CO exposure [7]. We have referred to these two glucose peaks, as a ‘camel-back’ pattern. This pattern was seen in the present study in the rats treated with CO alone. The magnitude of the initial peak and the depth of the valley following, are exacerbated by increased CO concentration (> 2400 ppm), saline infusion and the forced maintenance of normal body temperature. Rats not given exogenous glucose that die during CO exposure almost invariably have a blood glucose concentration below 40 mg/dl [7]. Recent studies have demonstrated an increased neurologic deficit and general morbidity following 90 min of CO exposure in Levine-prepared rats experiencing elevated blood glucose [2,3]. This was the case whether the elevated glucose resulted from CO exposure alone [3] or from CO exposure combined with the infusion of exogenous glucose [2]. The morphological correlate of neurologic deficit is cerebral cortical edema, which has been assessed both by measuring gross water content [1] and by using magnetic resonance imaging techniques (Jalukar et al., unpublished data). The elevations in blood glucose concentrations achieved during CO exposure in the present study in both the CO alone and CO + glucose rats were similar in magnitude to those reported previously [2]. Also like previous reports, rats experiencing either greatly depressed glucose levels during CO exposure, or greatly elevated glucose during CO exposure, showed the highest mortality. Thus, glucose level appears to be an important determinant of cerebral viability and survival during the hypoxia and relative ischemia (i.e., oligemia) of CO poisoning. It was hypothesized that an elevated blood glucose would permit an increased rate of glycolysis, thus generating more lactate. Our data indicate that this is not the case in the rats that received exogenous glucose and that blood lactate increased some lo-12-fold regardless of the blood glucose concentration. This observation has been confirmed in another study (Jalukar et al., unpublished data). It is possible of course that regional brain lactate concentration departed significantly from that of the carotid blood. The opposite condition was also not true, i.e., that severely hypoglycemic rats had depressed blood glucose because of an increased conversion of glucose to lactate. This suggests that the reason for the depressed blood glucose was a decrease in glucose release. Sokal [13] found that blood lactate increased from 9- 162 mg/dl in rats exposed to 4000 ppm CO for 40 min, an 1%fold increase. Considering that the CO concentration he used was nearly double that of the present study, the similarity in response was excellent. Sokal also noted a nearly 3-fold increase in blood glucose concentration, while pH fell to 6.85. Similar changes in blood lactate and glucose were reported by McGrath and Lee [ 141, in rats acutely inhaling CO. Using mice inhaling 3500 ppm CO, Moore et al. [ 151 found that blood lactate reached 90 mgdl after 11.5 min, while pH fell to only 7.25, the more modest acidosis probably due to the short exposure period. Measurements of regional brain lactate concentration have shown
177
it to increase to similar high levels in anesthetized rats ventilated with up to 20 000 ppm CO [16]. Although we did not measure blood pH, it is likely that levels of acidosis similar to those seen by Moore et al. were achieved in our study. Jn the past, a number of investigators have suggested that hypoxic/ischemic brain damage is the direct result of excessive lactate production and the attendant acidosis [17] and that brain lactate can be used to calculate brain pH. Recent nuclear magnetic resonance studies of brain lactate and pH, however, suggest that the two become dissociated during hypoxia/ischemia [18]. Although there were no differences in the blood lactate levels in our rats, it is still possible that that brain pH was different in the survivors versus the non-survivors. Evidence suggests that carboxyhemoglobinemia causes increased stimulation of the sympathetic nervous system. For example, CO-induced arterial hypotension initiates the carotid baroreflex, producing the tachycardia which is observed prior to hypoxic pacemaker depression [ 11. The peripheral chemoreceptors, especially the aortic bodies, are sensitive to CO [19] and increase their firing rate in its presence. Elevated plasma catecholamine levels were observed in anesthetized dogs ventilated with an air-CO mixture [6]. Catecholamine urinary excretion is also reportedly increased up to 5-fold in rats receiving CO by subcutaneous injection [5]. It has been suggested that the elevated catecholamine, acting on hepatic /3-adrenergic receptors, could be a major stimulus for increased glucose mobilization and hence blood levels. Our data failed to show any relationship between either the elevation of epinephrine or norepinephrine and blood glucose. Nevertheless, the inhibition of catecholamine release in vivo has been found to block CO-induced hyperglycemia [20]. On the other hand, the enhanced glycogenolysis and lactate production in the brain are reportedly not prevented in CO-exposed mice by pre-treatment with reserpine or a P-adrenergic blocking agent [21]. In summary, our data show that elevated blood glucose during acute CO poisoning does not further increase the blood lactate level above that present when glucose is normal or depressed, although mortality is increased at very high and very low glucose concentrations. The data also fail to show any relationship between the COinduced elevation of catecholamine concentrations and the rise in blood glucose. Thirdly, the results show that CO-induced hypoglycemia is not accompanied by a greater elevation of blood lactate. Finally, the data do not support the notion that mortality is related to the magnitude of the elevation in blood lactate or catecholamine concentrations. It remains unclear by what mechanism mortality is increased in acute CO poisoning. References D.G. Penney, K. Verma and J.A. Hull, Cardiovascular, metabolic and neurologic carbon monoxide poisoning in the rat. Toxicol. Letts., 45 (1989) 207.
effects of acute
D.G. Penney, CC. Helfman, J.C. Dunbar and L.E. McCoy, Acute severe carbon monoxide poisoning in the rat: Effects of hyperglycemia and hypoglycemia on mortality, recovery and neurologic deficit. Can. J. Physiol. Phannacol., 69 (1991) 1168. D.G. Penney, C.C. Helfman, J.A. Hull, J.C. Dunbar and K. Verma, Elevated blood glucose is associated with poor outcome in the carbon monoxide-poisoned rat. Toxicol. Letts., 54 (1990) 287. D.G. Penney, Hyperglycemia exacerbates brain damage in acute severe carbon monoxide poisoning. Med. Hypotheses, 27 (1988) 241.
178
5
6
7 8 9 10 11
12 13
14 15 16 17 18
19 20 21
D. Pankow
and W. Ponsold,
Effect of single and repeated
carbon
monoxide
intoxications
on urinary
catecholamine excretion in rats. Acta Biol. Med. Ger., 37 (1978) 1589. J.T. Sylvester, S.M. Scharf, R.D. Gilbert, R.S. Fitzgerald and R.J. Traystman, Hypoxic and CO hypoxia in dogs: Hemodynamics, carotid reflexes and catecholamines. Am. J. Physiol., 236 (1979) H22. D.G. Penney, P. Sharma, B.B. Sutariya and B.C. Nallamothu, Development of hypoglycemia is associated with death during carbon monoxide poisoning. J. Crit. Care, 5 (1990) 1. D.G. Penney, Acute carbon monoxide poisoning: Animal models. Toxicology, 62 (1990) 123. D.G. Penney, Modifying role of plasma glucose in acute carbon monoxide poisoning. Arch. Toxicol. (Recent developments in toxicology: Trends, methods and problems) Suppl., 14 (1991) 239. S. Levine, Anoxic ischemic encephalopathy in rats. Am. J. Pathol., 36 (1960) I. I.N. Mefford, Application of high performance liquid chromatography with electrochemical detection to neurochemical analysis: Measurement of catecholamines, serotonin and metabolites in rat brain. J. Neurosci. Methods, 3 (1981) 207. M. Ogawa, S. Shimazaki, N. Tanaka, T. Ukai and T. Sugimoto. Blood volume changes in experimental carbon monoxide poisoning. Ind. Health, 12 (1974) 121. J.A. Sokal, Lack of the correlation between biochemical effects on rats and blood carboxyhemoglobin concentrations in various conditions of single acute exposure to carbon monoxide. Arch. Toxicol., 34 (1975) 331. J.J. McGrath, K.C. Chen and P.S. Lee, Effects of intraperitoneal injection of carbon monoxide on circulating blood glucose and lactate. Fed. Proc.. 38 (1979) 1313. S.J. Moore, I.K. Ho and A.S. Hume, Severe hypoxia produced by concomitant intoxication with sublethal doses of carbon monoxide and cyanide. Toxicol. Appl. Pharmacol., 109 (1991) 412. V. MacMillan, Regional cerebral energy metabolism in acute carbon monoxide intoxication. Can. J. Physiol. Pharmacol., 55 (1977) 1I I. S. Rehncrona, I. Rosen and B.K. Siesjo. Excessive cellular acidosis: An important mechanism of neuronal damage in the brain. Acta Physiol. Stand., I IO (1980) 435. S. Lotito, P. Blondet, A. Francois, M. von Kienlin, C. Remy, J.P. Albrand. M. Decorps and A.L. Benabid, Correlation between intracellular pH and lactate levels in the rat brain during potassium cyanide induced metabolism blockade: A combined “P-‘H in vivo nuclear magnetic spectroscopy study. Neurosci. Lett., 97 (1989) 91. E. Mills and M.W. Edwards Jr., Stimulation of aortic and carotid chemoreceptors during carbon monoxide inhalation. J. Appl. Physiol., 25 (1968) 494. M. Gothert, P. bon monoxide C.-J. Estler, glycogenolysis
Hock, G. Malorney and H.-J. Seiler, Mechanism of hyperglycemia during acute carpoisoning. Naunyn-Schmiedeberg’s Arch. Pharmacol. (SuppI.), 274 (1972) R4l. F. Heim, H.P.T. Ammon and V. Zimmermann, Catecholamine-independent in brain during carbon monoxide poisoning. Pharmacology, 5 (197 I) 55.
