Resuscitation, 22 (1991) 65-74 Elsevier Scientific Publishers Ireland
65 Ltd.
The effects of continuous glucose infusion on blood, plasma, and brain glucose in anesthetized rats Margaret The Anesthesia
(Received
R. Weglinski,
Roger
E. Hofer and William
L. Lanier
Research Laboratories. Department of Anesthesiology, Mayo Clinic and Mayo Medical School, Rochester, MN 55905 (U.S. A.)
September
18th, 1990; revision’received
February
2nd. 1991; accepted
March
17th, 1991)
In models of cerebral ischemia, it is important to rigidly control brain glucose in the pcri-ischemic period because alterations in brain glucose can affect the severity of the postischemic injury. The following study evaluated the effect of a continuous glucose infusion as a means of producing stable increases in brain glucose that could be monitored by measuring either blood or plasma glucose. Fifty-four halothaneanesthetized rats were studied. Rats received either no treatment (control group; N = 6). saline 2 ml/h (N = 24). or glucose 1 g/kg per h in saline 2 ml/h (N = 24). In the latter two groups. samples of blood, plasma, and brain glucose were obtained at either 30,60, 120, or 180 min of the infusion (N = 6 per group per sample period). Saline infusion had no effect on either blood, plasma, or brain glucose. In contrast, glucose infusion produced a significant increase in all three variables, achieving plateau increases during the 6Gl80 min measurement periods [blood glucose = 197 f 20 mg/dl (mean f S.D.) at 60 min. 220 f 34 mg/dl at 120 min. and 217 f 22 mg/dl at 180 min versus control blood glucose = 89 f IO mgdl]. Regardless of the treatment group, there was excellent correlation between blood and plasma glucose (r = 0.99; P 4 0.001). blood and brain glucose (r = 0.96; P * 0.001). and plasma and brain glucose (r = 0.97; P 4 0.001). The authors conclude that continuous glucose infusions are an effective method to produce stable increases in brain glucose in experimental models; and, in contrast to other methods for achieving brain glucose increases, the brain glucose increases can be accurately assessed by measuring blood or plasma
Brain -
Glucose
glucose.
-
lschemia
-
Metabolism
-
Protection
-
Diabetes
-
Hyperglycemia
INTRODUCTION
It is a well appreciated clinical and laboratory phenomenon that hyperglycemia and increases in brain glucose will worsen the degree of injury the brain sustains. during and following a period of ischemia [l-8]. Enhancement of ischemic neurologic injury has been reported with even small, clinically relevant, acute increases in blood glucose (i.e. an increase of < 50 mg/dl) [4]. The proposed mechanism responsible for this glucose-induced exacerbation of ischemic injury is the enhanced cerebral acidosis that results from the glycolysis of glucose to lactic acid in the presence of ischemia [8--l 11. Address all correspondence and reprint requests to: William Anesthesiology, Mayo Clinic, Rochester, MN 55905, U.S.A.
0300-9572/91/$03.50 0 1991 Elsevier Printed and Published in Ireland
Scientific
Publishers
Ireland
L.
Lanier,
Ltd.
M.D..
Department
of
66
When designing models of experimental cerebral ischemia to be used in the evaluation of pathomechanisms of - and therapies for - cerebral ischemia, it is important to provide a control group of subjects with as little variability in the degree of postischemic injury as possible. This is especially important in models in which injury is assessed by neurologic functional recovery, because there is 2 large amount of variability inherent to the postischemic neurologic function scores [ 12, 131. One possible manner in which postischemic variability could be attenuated would be to rigidly control the brain glucose concentrations before and during ischemia. Several approaches to this problem have been evaluated; however, all have limitations, resulting in either an excessive amount of variability in glucose values in the control animals, or an inability to predict brain glucose by measuring blood or plasma glucose [4,12,14,15]. The present study in rats examined the effect of an alternative method of controlling glucose, i.e. a constant rate glucose infusion, to determine: (a) if plateau increases in blood, plasma, and brain glucose could be obtained, and (b) if measurement of blood or plasma glucose would accurately predict brain glucose concentrations. SUBJECTS AND METHODS
This protocol was reviewed and approved by the Institutional Animal Care and Use Committee of Mayo Clinic. Fifty-four male Sprague-Dawley rats of a consistent age weighing 290 f 3 g (mean f S.D.) were studied. Rats were fasted approximately 12 h prior to the study, but had free access to water until the experiment began. The animals were anesthetized in a plexiglass chamber with halothane 3% inspired in 0,. Tracheal intubation was performed via a tracheostomy and the lungs were ventilated using a rodent ventilator (Ugo Basile, Model 7025, Varese, Italy). Anesthesia was maintained with 1.3% inspired halothane in 50% 0, and NZ. The rats were initially paralyzed with pancuronium 0.5 mg i.m., and additional doses of pancuronium 0.3 mg i.m. were given as needed. All rats received 1.3% inspired halothane for more than 1 h before the brain biopsies were taken. Polyethylene catheters (PE-50) were placed into a femoral artery for blood sampling and pressure monitoring, and a femoral vein for fluid infusion. Rectal temperature was measured by a thermistor (YSI Model 73A, Yellow Springs Instruments Co., Yellow Springs, OH), and body temperature was maintained near 37°C using a heating lamp. Inspired concentration of halothane was monitored using the Rascal@ raman scattering multiple gas analyzer (Albion Instruments, Salt Lake City, UT). Arterial blood gases were determined by electrodes at 37°C (Instrumentation Laboratories, Inc., Lexington, MA), and blood and plasma glucose were measured using a YSI Model 23A glucose analyzer. This device has a detection range of O-28 pmol/ml (l-500 mg/dl) and a sensitivity of 0.1 pmol/ml (1 mg/dl)*.
*Operator’s manual, YSI Model 23A glucose analyzer, Yellow Springs Instruments Co., Yellow Springs, OH.
67
Rats were then positioned prone and secured using a stereotactic headframe. Following a midline incision over the cranium, the skin and muscle of the scalp were reflected, and the underlying calvarium was exposed. A plastic funnel was sutured into place over the calvarium, and the exposed tissues were covered with a sheet of paraffin to prevent heat and moisture loss. After the surgical preparation was complete, and following an additional 20 min stabilization period, arterial blood gases, acid-base status, blood glucose, mean arterial pressure (MAP), temperature, and inspired halothane concentration were measured. If the pre-established protocol criteria were not met, appropriate adjustments were made, and additional 10 min stabilization periods were allowed until the following criteria were met: Paoz, 125-175 mmHg; Pat@, 3640 mmHg; MAP > 70 mmHg; temperature, 36.5-37.5”C; inspired halothane concentration, 1.2-l .4X; and blood glucose < 130 mg/dl. If MAP was less than 70 mmHg, a single treatment of l-2 ml 0.9% saline i.v. was given. If MAP remained below 70 mmHg following treatment, the animal was excluded from the study. Rats were randomly divided into groups according to fluid infusion and the time interval until brain freezing. Control rats (N = 6) received no fluid infusion and their brains were frozen in situ after the 20 min stabilization period. Saline-treated rats (N = 24) received 0.9% preservative-free saline i.v. at a rate of 2 ml/h. Their brains were then frozen at either 30,60, 120, or 180 min (N = 6 for all groups). Of the control and saline-treated rats, 18 were shared with another simultaneously performed study from which the results have been published [ 151. Glucose-treated rats (N = 24) received glucose at a rate of 1 g/kg per h infused in 0.9”%preservative saline for a total fluid rate of 2 ml/h. Their brains also were frozen at either 30, 60, 120, or 180 min (N = 6 for all groups). Brain harvesting for subsequent glucose analysis was performed using a modification of methods previously described by Ponten et al. [16]. Briefly, the paraffin overlying the calvarium was removed, and the brain was frozen in situ by pouring liquid N2 into the funnel overlying the calvarium. At the same time, an arterial blood sample was obtained for determination of blood gases, acid-base status, glucose, and hematocrit. Temperature, MAP, and inspired halothane were also recorded. Mechanical ventilation was continued throughout the period of freezing. After the freezing was complete, the brain was removed while being bathed with liquid NZ. The frozen brain, stored at -72°C was prepared for analysis in a refrigerated box (-20°C). The venous sinuses and meninges were dissected away and the hemispheres were separated from each other. The cortex was then dissected from the remainder of the cerebrum. This yielded 100-200 mg of cortex per hemisphere for subsequent analysis of brain glucose. Glucose was chemically extracted from the cortex [17] and measured using the enzymatic fluorometric technique described by Lowry et al. [18]. The sensitivity of the enzymatic fluorometric method of glucose measurement was 0.2 pmol per g [ 141. The correlation coefficient between measured values and samples with known glucose values was 0.99 [14]. In the present study, the reported value for each animal represented the mean of the two hemispheres. Data obtained during glucose or saline infusions were compared to the control group using a completely randomized analysis of variance and F-tests. These tests also were used to test for differences between glucose and saline groups at a given duration of infusion. The Pearson product-moment correlation coefficient was used
to test for associations between blood, plasma, and brain glucose values. P < 0.05 was considered significant. All data were reported as mean f SD. RESULTS
All groups were well matched for weight and systemic physiologic variables during the control period and at the time of brain freezing. Physiologic variables for the groups at the time of brain freezing are listed in Table I. There was good agreement between right and left hemisphere brain glucose values. When all groups were combined, right hemisphere brain glucose was 3.14 f 1.47 pmol/g and left hemisphere brain glucose was 3.12 f 1.51 pmollg (P > 0.6 by paired ?-test). Additionally, there was a significant correlation between brain glucose in the two hemispheres (r = 0.98; P < 0.0001). In the control group, blood glucose was 4.9 f 0.6 Fmol/ml (89 f 10 mg/dl), plasma glucose was 8.2 f 1.0 ~mol/ml, and brain glucose was 2.07 f 0.29 pmol/g. Brain glucose values in this group were similar to those reported for anesthetized rats in other studies [ 19,201. Saline infusion in the saline-treated group had no significant effect on blood, plasma, or brain glucose (Figs. l-3). Glucose infusion in the glucose-treated group resulted in blood, plasma, and brain glucose values that were significantly greater than control group values and values following saline (Figs. l-3). Blood glucose increased to a plateau of 10.9 f 1.1 pmol/ml (197 f 20 mg/dl) 12.2 f 1.9 kmol/ml (220 f 34 mg/dl), and 12.0 f 1.2 ~moVml(217 f 22 mg/dl) at the 60, 120, and 180 min measurement periods, respectively. Likewise, plasma glucose increased to 17.2 f 1.8 lmol/ml, 19.6 f 3.5
Table 1.
Physiologic
variables
at the time of blood and brain sampling
PaOz
PaCO,
(mmHg)
(mmHg)
pH
in halothane-anesthetized
Hct (‘%I)
rats.
Rectal temperature
MAP (mmHg)
(“C) Control
160 f
24
40 f
5
7.36 zt 0.04
42 f
3
84 zt 13
37.1 f
0.3
153 f 15 149 f 16 139 ?? 9 138 f 14*
37 38 38 42
f zt f f
4 3 5 4
7.37 ?? 0.03 7.33 f 0.03 7.36 f 0.03 7.35 ?? 0.02*
41 41 40 41
f f zt f
4 3 3 2
82* 81* 82zt 91 *
37.1 37.3 37.0 37.1
0.3 0.2 0.3 0.3
162 152 158 162
36 39 39 42
f I * 2 * 3 zt 4
7.38 7.35 7.35 7.32
38 40 40 43
iz * f f
2 1 3 3
83 ?? 21 89, 8 89zt 4 94 f 15
Sulk-fretted
Time of sampling 30 min 60 min 120 min I80 min
7 7 9 9
f * k f
Glucose-treated
Time of sampling 30 min 60 min 120 min 180 min
f zt f f
20 24 28 18
f f f zt
0.03 0.02 0.02 0.02+
37.1 ?? 0.2 37.3 f 0.3 37.1 ?? 0.3 37.3 f 0.3
MAP = mean arterial blood pressure; Hct = hematocrit. All values are presented as mean f SD. (N =6 for all groups). *Significant difference between glucose- and saline-treated subjects (P < 0.05) at the same time period.
+Significantly
different
from control
(P < 0.05).
69
:=: ?? ? -e ?
Control Glucoss-Trratod Salh-Trratod
t
01 -
30
60
90
120
150
160
Time(minutes) Fig. 1. Blood glucose following glucose or saline infusion in halothane-anesthetized rats: n = 6 for all, groups. All values are mean f SD. Vertical bars equal one SD. *P < 0.05 versus saline. +P < 0.05 versus control.
0-O -A f-*
20
-
15
-
10
-
5
-
-
Control Glucose-Tnatad SalineTreated
30
60
90
120
150
190
Time(minutes) Fig. 2. Plasma glucose following glucose or saline infusion in halothane-anesthetized rats: n = 6 for all groups. All values are mean + S.D. Vertical bars equal one SD. *P < 0.05 versus saline. +P < 0.05 versus control.
70 0-e A-A
?? -
Control Glucose-Treated Sailno-Treated
-*
6-
o-
cmt?Ql
30
60
90
120
100
150
Time (minutes) rats: n = 6 for all Fig. 3. Brain glucose following glucose or saline infusion in halothane-anesthetized groups. All values are mean f SD. Vertical bars equal one S.D. *P < 0.05versus saline. +P < 0.05 versus control.
pmol/ml, and 19.4 f 2.2 ,umollml, respectively. Accompanying these increases were increases in brain glucose to 4.39 f 0.44 pmol/g, 5.19 f 0.95 pmol/g, and 5.08 f 0.43 pmol/g, respectively. The ratio of brain to plasma glucose did not differ from control with either saline 0-O -A f - -e
E E
0.35
-
0.30
-
0.25
-
Control Glucoro-Tnated Sallno-Treated
3 G
--.
--
I
I
i
I --+---___
t
--
f -_ 1
it 8 3 G
.-s
,o m
0.10 ConhI
30
60
90
120
150
160
Time (minutes) glucose in halothane-anesthetized rats: n = 6 for all groups. All Fig. 4. The ratio of brain-to-plasma values are mean f SD. Vertical bars equal one SD. There were no significant differences between glucose- and saline-treated groups, nor were there significant differences versus control.
71
or glucose infusion (Fig. 4). Irrespective of the treatment, there was significant correlation between blood and plasma glucose (r = 0.99; P e O.OOl), blood and brain glucose (r = 0.96; P --s O.OOl), and plasma and brain glucose (r = 0.97; P 4 0.001). DISCUSSION
Hyperglycemia and increases in brain glucose will worsen the degree of injury the brain sustains following a period of global ischemia [l-8]. The proposed mechanism by which hyperglycemia augments cerebral ischemic injury is as follows: during periods of hyperglycemia, glucose crosses the blood-brain barrier by facilitated diffusion and other secondary pathways, producing increased brain glucose concentrations [4,9,10]. In the event of ischemia, the brain anaerobically metabolizes glucose to lactic acid. If the brain is exposed to an ischemic insult during a period of increased brain glucose, lactic acid formation will be increased [8, lo]. It is the lactic acid that appears to be the metabolic toxin responsible for the augmentation of ischemic neurologic injury [9,11]. The osmotic effects of increased blood and brain glucose appear to have little influence on the augmentation of ischemic cerebral injury [7]. Theoretically, during ischemia both free glucose as well as glucose liberated from the metabolism of glycogen will contribute to the magnitude of the lactic acidosis [8]. In support of this concept are in vivo data demonstrating a correlation between preischemic brain glucose increases and postischemic lactate accumulation [lo] or cellular acidosis [21], and in vitro data demonstrating the depletion of neuronal glycogen during anoxia [22]. In in vivo studies of experimental ischemia, investigators often exploit these metabolic considerations to vary the magnitude of postischemic injury. In order to avoid the confounding influence of metabolism of brain glycogen (a compound that is difficult to quantify acutely), investigators often select study groups with similar ages, weights, phenotypes, and nutritional histories. In this setting, investigator-controlled acute alterations in brain glucose, per se, independent of demonstrated or anticipated alterations in brain glycogen, have been reported to alter outcome following ischemia [4,10,23,24]. Based on the above-mentioned considerations, when designing models of experimental cerebral ischemia, it is important to control free brain glucose as rigidly as possible in the peri-ischemic period. Depending on the issue to be studied, it may be desirable for the brain glucose values to rigidly remain within the normal range, while at other times, it may be desirable to produce an equally stable but increased brain glucose. Regardless of the techniques and goals, it is ideal if blood, plasma, and brain glucose quantities are managed by a method that will allow a rapid and accurate assessment of brain glucose in the peri-ischemic period. Since it is technically impossible to obtain rapid, accurate direct measurements of brain glucose, it then becomes important to alter brain glucose using a technique that will permit accurate assessment using an extracerebral monitor. The present study assessed the use of continuous infusions of saline or glucose to produce stable and predictable alterations in brain glucose. Using the techniques described, brain glucose values ranging from 1.07 to 6.84 pmol/g could be accurately predicted by measuring either plasma glucose or the more rapidly obtainable blood glucose. In addition to the technique we studied, there are several other alternative meth-
12
ods for manipulating brain glucose concentrations in study animals. One option is to make all animals hyperglycemic with a glucose bolus or infusion, and after the glucose load, wait until blood glucose levels return to a predefined level before inducing &hernia. Indeed, this technique has been reported from several laboratories, including our own [4,25]. The problem with this methodology is that the production of ischemia must be performed precisely at the desired time, or the blood glucose will decline to a value less than the target value. Furthermore, a recent study from our laboratory demonstrated that - in this scenario - during the period of decreasing blood glucose there may be a hysteresis between rapidly changing blood and brain glucose values. If hysteresis occurs, the blood glucose measurements may underestimate the glucose concentrations in the brain [ 141. Thus, brain glucose concentrations may be greater [14], and postischemic neurologic injury may be worse [4], than would be predicted by examining blood or plasma glucose values. A second method to alter brain glucose in a graded fashion would be to begin with diabetic rats, and treat the rats with an insulin infusion until blood glucose values reach a given range. Recent studies in our laboratory examining this technique found that insulin infusions, per se, had no effect on the relationship between blood and plasma glucose, blood and brain glucose, or plasma and brain glucose [15]. Unfortunately, for a given plasma glucose concentration, diabetic rats had an approximately 25-50% greater brain glucose concentration than non-diabetic rats, because the diabetic brain tends to store glucose [ 15,261. Thus, for a given blood glucose concentration, it is difficult to make comparisons between brain glucose concentration estimates in non-diabetic versus insulin-treated diabetic rats. Furthermore, when using diabetic rats, one must contend with the special needs of the diabetic rat in the pre- and postischemic periods (i.e. increased fluid needs, increased risk of infection, possible need for continued insulin therapy, poor wound healing), particularly if chronic studies are planned. A third method of rigidly controlling brain glucose was evaluated by the present study: i.e. the use of a constant rate saline or glucose infusion to produce plateau concentrations of both blood and brain glucose. These data indicated that, using this technique, blood, plasma, and brain glucose could be effectively manipulated over a wide range of values. Within the range of blood glucose values studied [i.e. values from 3.6 to 15.0 pmol/ml (65-271 mg/dl)], there was excellent agreement between blood, plasma, and brain glucose. The technique is simple, and by adjusting the glucose infusion rate, should provide stable brain glucose values that are assessable using blood or plasma glucose measurements. Furthermore, our data suggest that the use of small glucose infusions to produce slight increases in brain glucose may result in a slightly better correlation between plasma and brain glucose, or alternatively blood and brain glucose, than occurs in untreated or saline-treated animals (Fig. 4). If the technique has a fault, it is as follows: previous studies from our laboratory suggest that if too large a glucose infusion is given too rapidly, there may be saturation of portions of the mechanisms responsible for transporting glucose into the brain, resulting in a hysteresis between blood and brain glucose. However, when lesser glucose infusions are given (1 g/kg per h in the present study versus 6 g/kg per h in our previous study [ 14]), there is no hysteresis between brain and blood glucose. In summary, the present study in anesthetized rats evaluated the use of continuous
73
i.v. infusions of glucose to produce stable increases in blood, plasma, and brain glucose. The method was effective in producing stable glucose alterations, and in preserving the correlation between blood, plasma, and brain glucose concentrations found in untreated or saline-treated animals. The technique is easy to use, and avoids many of the problems associated with other methods in which brain glucose may be altered during studies of experimental cerebral ischemia. ACKNOWLEDGEMENT
The authors wish to thank William Gallagher, Leslie Phelps, and Rebecca Wilson for their technical assistance; and John D. Michenfelder, M.D., for critically reviewing the manuscript. REFERENCES
1
W.T. Longstreth Jr., P. Diehr and T.S. Inui, Prediction arrest. N. Engl. J. Med., 308 (1983) 1378-1382.
2
W.T. Longstreth Jr. and T.S. Inui, High blood glucose level on hospital admission and poor neurological recovery after cardiac arrest, Ann. Neural., 15 (1984) 59-63. W. Pulsinelli, S. Waldman, B. Sigsbee, D. Rawlinson, P. Scherer and F. Plum, Experimental hyperglycemia and diabetes mellitus worsen stroke outcome, Trans. Am. Neural. Assoc., 195 (1980)
3
4
5 6 7 8 9
10
II
12
13
14
of awakening
after out-of-hospital
cardiac
21-24. W.L. Lanier, K.J. Stangland, B.W. Scheithauer, J.H. Milde and J.D. Michenfelder, The effects of dextrose infusion and head position on neurologic outcome after complete cerebral ischemia in primates: examination of a model, Anesthesiology, 66 (1987) 39-48. R.E. Myers and S. Yamaguchi, Nervous system effects of cardiac arrest in monkeys, Arch. Neurol., 34 (1977) 65-74. E. Siemkowicz and A.J. Hansen, Clinical restitution following cerebral ischemia in hypo-, normo-, and hyperglycemic rats, Acta Neurol. Stand., 58 (1978) 1-8. W.A. Pulsinelli, S. Waldman, D. Rawlinson and F. Plum, Moderate hyperglycemia augments ischemic brain damage: A neuropathologic study in the rat, Neurology, 32 (1982) 1239-1246. F. Plum, What causes infarction in ischemic brain?: The Robert Wartenberg Lecture, Neurology, 33 (1983) 222-233. E. Siemkowicz and A. Gjedde, Post-ischemic coma in rat: Effect of different preischemic blood I IO (1980) glucose levels on cerebral metabolic recovery after ischemia, Acta Physiol. Stand., 225-232. F.A. Welsh, M.D. Ginsberg, W. Rieder and W.W. Budd, Deleterious effect of glucose pretreatment on recovery from diffuse cerebral ischemia in the cat. II. Regional metabolite levels, Stroke, I1 (1980) 355-363. R.E. Myers, Lactic acid accumulation as cause of brain edema and cerebral necrosis resulting from oxygen deprivation. In: Advances in Perinatal Neurology. Editors: R. Korobkin and C. Guilleminault, Spectrum, New York, NY. 1979, pp. 85-l 14. W.L. Lanier, W.J. Perkins, B.R. Karlsson, J.H. Milde, B. W. Scheithauer. G.T. Shearman and J.D. Michenfelder, The effects of dizocilpine maleate (MK-801). an antagonist of the N-methyl-Daspartate receptor, on neurologic recovery and histopathology following complete cerebral ischemia in primates, J. Cereb. Blood Flow Metab., 10 (1990) 252-261. A. Tateishi, J.E. Fleischer, J.C. Drummond, MS. Scheller, M.H. Zornow, M.R. Grafe and H.M. Shapiro, Nimodipine does not improve neurologic outcome after 14 min of cardiac arrest in cats, Stroke, 20 (1989) 1044-1050. M.R. Weglinski and W.L. Lanier, The effects of transient hyperglycemia on brain glucose in rats anesthetized with halothane, Anesthesiology, 73 (1990) 93-98.
74 15 16 17
18 19 20
21 22 23
24
25
26
R.E. Hofer and W.L. Lamer, The effect of insulin infusion on brain and blood glucose in hyperglycemic diabetic rats (abstract), Anesthesiology, 73 (1990) A1272. U. Ponten, R.A Ratcheson, L.G. Salford and B.K Siesjo, Optimal freezing conditions for cerebral metabolites in rats, J. Neurochem., 21 (1973) 1127-l 138. J. Folbergrova, V. MacMillan and B.K. Siesjo, The effect of moderate and marked hypercapnia upon the energy state and upon the cytoplasmic NADH/NAD+ ratio of the rat brain, J. Neurochem., 19 (1972) 2497-2505. O.H. Lowry, J.V. Passonneau, F.X. Hasselberger and D.W. Schulz, Effect of &hernia on known substrates and cofactors of the glycolytic pathway in brain, J. Biol. Chem., 239 (1964) 18-30. W.A. Kotlce, R.A. Hawkins, D.W, Davis and J.F. Beibuyck, Comparison of the effects of volatile anesthetics on brain glucose metabolism in rats, Anesthesiology, 66 (1987) 810-813. D.F. Dedrick, Y.D. Scherer and J.F. Biebuyck, Use of a rapid brain-sampling technique in a physiologic preparation: Effects of morphine, ketamine, and halothane on tissue energy intermediates, Anesthesiology, 42 (1975) 651457. W.R. Marsh, R.E. Anderson and T.M. Sundt, Jr., Effect of hyperglycemia on brain pH levels in areas of focal incomplete cerebral &hernia in monkeys, J. Neurosurg., 65 (1986) 693496. P. Lipton, Regulation of glycogen in the dentate gyrus of the in vitro guinea pig hippocampus; effect of combined deprivation of glucose and oxygen, J. Neurosci. Methods, 28 (1989) 147-154. L.G. D’Alecy, E.F. Lundy, K.J. Barton and G.B. Zelenoch, Dextrose containing intravenous fluid impairs outcome and increases death after eight minutes of cardiac arrest and resuscitation in dogs, Surgery, 100 (1986) 505-51 I. K. Nakakimura, J.E. Fleischer, J.C. Drummond, M.S. Scheller, M.H. Zornow, M.R. Grafe and H.M. Shapiro, Glucose administration before cardiac arrest worsens neurologic outcome in cats, Anesthesiology, 72 (1990) lOO5-IO1 I. P.A. Steen, S.E. Gisvold, J.H. Milde, L.A Newberg, B.W. Scheithauer, W.L. Lanier and J.D. Michenfelder, Nimodipine improves outcome when given after complete cerebral ischemia in primates, Anesthesiology, 62 (1985) 40-14. R.E. Hofer, W.L. Lanier, Effects of insulin diabetic rats, Stroke, 22 (1991) 505-509.
on blood,
plasma
and brain glucose
in hyperglycemic
65 Ltd.
The effects of continuous glucose infusion on blood, plasma, and brain glucose in anesthetized rats Margaret The Anesthesia
(Received
R. Weglinski,
Roger
E. Hofer and William
L. Lanier
Research Laboratories. Department of Anesthesiology, Mayo Clinic and Mayo Medical School, Rochester, MN 55905 (U.S. A.)
September
18th, 1990; revision’received
February
2nd. 1991; accepted
March
17th, 1991)
In models of cerebral ischemia, it is important to rigidly control brain glucose in the pcri-ischemic period because alterations in brain glucose can affect the severity of the postischemic injury. The following study evaluated the effect of a continuous glucose infusion as a means of producing stable increases in brain glucose that could be monitored by measuring either blood or plasma glucose. Fifty-four halothaneanesthetized rats were studied. Rats received either no treatment (control group; N = 6). saline 2 ml/h (N = 24). or glucose 1 g/kg per h in saline 2 ml/h (N = 24). In the latter two groups. samples of blood, plasma, and brain glucose were obtained at either 30,60, 120, or 180 min of the infusion (N = 6 per group per sample period). Saline infusion had no effect on either blood, plasma, or brain glucose. In contrast, glucose infusion produced a significant increase in all three variables, achieving plateau increases during the 6Gl80 min measurement periods [blood glucose = 197 f 20 mg/dl (mean f S.D.) at 60 min. 220 f 34 mg/dl at 120 min. and 217 f 22 mg/dl at 180 min versus control blood glucose = 89 f IO mgdl]. Regardless of the treatment group, there was excellent correlation between blood and plasma glucose (r = 0.99; P 4 0.001). blood and brain glucose (r = 0.96; P * 0.001). and plasma and brain glucose (r = 0.97; P 4 0.001). The authors conclude that continuous glucose infusions are an effective method to produce stable increases in brain glucose in experimental models; and, in contrast to other methods for achieving brain glucose increases, the brain glucose increases can be accurately assessed by measuring blood or plasma
Brain -
Glucose
glucose.
-
lschemia
-
Metabolism
-
Protection
-
Diabetes
-
Hyperglycemia
INTRODUCTION
It is a well appreciated clinical and laboratory phenomenon that hyperglycemia and increases in brain glucose will worsen the degree of injury the brain sustains. during and following a period of ischemia [l-8]. Enhancement of ischemic neurologic injury has been reported with even small, clinically relevant, acute increases in blood glucose (i.e. an increase of < 50 mg/dl) [4]. The proposed mechanism responsible for this glucose-induced exacerbation of ischemic injury is the enhanced cerebral acidosis that results from the glycolysis of glucose to lactic acid in the presence of ischemia [8--l 11. Address all correspondence and reprint requests to: William Anesthesiology, Mayo Clinic, Rochester, MN 55905, U.S.A.
0300-9572/91/$03.50 0 1991 Elsevier Printed and Published in Ireland
Scientific
Publishers
Ireland
L.
Lanier,
Ltd.
M.D..
Department
of
66
When designing models of experimental cerebral ischemia to be used in the evaluation of pathomechanisms of - and therapies for - cerebral ischemia, it is important to provide a control group of subjects with as little variability in the degree of postischemic injury as possible. This is especially important in models in which injury is assessed by neurologic functional recovery, because there is 2 large amount of variability inherent to the postischemic neurologic function scores [ 12, 131. One possible manner in which postischemic variability could be attenuated would be to rigidly control the brain glucose concentrations before and during ischemia. Several approaches to this problem have been evaluated; however, all have limitations, resulting in either an excessive amount of variability in glucose values in the control animals, or an inability to predict brain glucose by measuring blood or plasma glucose [4,12,14,15]. The present study in rats examined the effect of an alternative method of controlling glucose, i.e. a constant rate glucose infusion, to determine: (a) if plateau increases in blood, plasma, and brain glucose could be obtained, and (b) if measurement of blood or plasma glucose would accurately predict brain glucose concentrations. SUBJECTS AND METHODS
This protocol was reviewed and approved by the Institutional Animal Care and Use Committee of Mayo Clinic. Fifty-four male Sprague-Dawley rats of a consistent age weighing 290 f 3 g (mean f S.D.) were studied. Rats were fasted approximately 12 h prior to the study, but had free access to water until the experiment began. The animals were anesthetized in a plexiglass chamber with halothane 3% inspired in 0,. Tracheal intubation was performed via a tracheostomy and the lungs were ventilated using a rodent ventilator (Ugo Basile, Model 7025, Varese, Italy). Anesthesia was maintained with 1.3% inspired halothane in 50% 0, and NZ. The rats were initially paralyzed with pancuronium 0.5 mg i.m., and additional doses of pancuronium 0.3 mg i.m. were given as needed. All rats received 1.3% inspired halothane for more than 1 h before the brain biopsies were taken. Polyethylene catheters (PE-50) were placed into a femoral artery for blood sampling and pressure monitoring, and a femoral vein for fluid infusion. Rectal temperature was measured by a thermistor (YSI Model 73A, Yellow Springs Instruments Co., Yellow Springs, OH), and body temperature was maintained near 37°C using a heating lamp. Inspired concentration of halothane was monitored using the Rascal@ raman scattering multiple gas analyzer (Albion Instruments, Salt Lake City, UT). Arterial blood gases were determined by electrodes at 37°C (Instrumentation Laboratories, Inc., Lexington, MA), and blood and plasma glucose were measured using a YSI Model 23A glucose analyzer. This device has a detection range of O-28 pmol/ml (l-500 mg/dl) and a sensitivity of 0.1 pmol/ml (1 mg/dl)*.
*Operator’s manual, YSI Model 23A glucose analyzer, Yellow Springs Instruments Co., Yellow Springs, OH.
67
Rats were then positioned prone and secured using a stereotactic headframe. Following a midline incision over the cranium, the skin and muscle of the scalp were reflected, and the underlying calvarium was exposed. A plastic funnel was sutured into place over the calvarium, and the exposed tissues were covered with a sheet of paraffin to prevent heat and moisture loss. After the surgical preparation was complete, and following an additional 20 min stabilization period, arterial blood gases, acid-base status, blood glucose, mean arterial pressure (MAP), temperature, and inspired halothane concentration were measured. If the pre-established protocol criteria were not met, appropriate adjustments were made, and additional 10 min stabilization periods were allowed until the following criteria were met: Paoz, 125-175 mmHg; Pat@, 3640 mmHg; MAP > 70 mmHg; temperature, 36.5-37.5”C; inspired halothane concentration, 1.2-l .4X; and blood glucose < 130 mg/dl. If MAP was less than 70 mmHg, a single treatment of l-2 ml 0.9% saline i.v. was given. If MAP remained below 70 mmHg following treatment, the animal was excluded from the study. Rats were randomly divided into groups according to fluid infusion and the time interval until brain freezing. Control rats (N = 6) received no fluid infusion and their brains were frozen in situ after the 20 min stabilization period. Saline-treated rats (N = 24) received 0.9% preservative-free saline i.v. at a rate of 2 ml/h. Their brains were then frozen at either 30,60, 120, or 180 min (N = 6 for all groups). Of the control and saline-treated rats, 18 were shared with another simultaneously performed study from which the results have been published [ 151. Glucose-treated rats (N = 24) received glucose at a rate of 1 g/kg per h infused in 0.9”%preservative saline for a total fluid rate of 2 ml/h. Their brains also were frozen at either 30, 60, 120, or 180 min (N = 6 for all groups). Brain harvesting for subsequent glucose analysis was performed using a modification of methods previously described by Ponten et al. [16]. Briefly, the paraffin overlying the calvarium was removed, and the brain was frozen in situ by pouring liquid N2 into the funnel overlying the calvarium. At the same time, an arterial blood sample was obtained for determination of blood gases, acid-base status, glucose, and hematocrit. Temperature, MAP, and inspired halothane were also recorded. Mechanical ventilation was continued throughout the period of freezing. After the freezing was complete, the brain was removed while being bathed with liquid NZ. The frozen brain, stored at -72°C was prepared for analysis in a refrigerated box (-20°C). The venous sinuses and meninges were dissected away and the hemispheres were separated from each other. The cortex was then dissected from the remainder of the cerebrum. This yielded 100-200 mg of cortex per hemisphere for subsequent analysis of brain glucose. Glucose was chemically extracted from the cortex [17] and measured using the enzymatic fluorometric technique described by Lowry et al. [18]. The sensitivity of the enzymatic fluorometric method of glucose measurement was 0.2 pmol per g [ 141. The correlation coefficient between measured values and samples with known glucose values was 0.99 [14]. In the present study, the reported value for each animal represented the mean of the two hemispheres. Data obtained during glucose or saline infusions were compared to the control group using a completely randomized analysis of variance and F-tests. These tests also were used to test for differences between glucose and saline groups at a given duration of infusion. The Pearson product-moment correlation coefficient was used
to test for associations between blood, plasma, and brain glucose values. P < 0.05 was considered significant. All data were reported as mean f SD. RESULTS
All groups were well matched for weight and systemic physiologic variables during the control period and at the time of brain freezing. Physiologic variables for the groups at the time of brain freezing are listed in Table I. There was good agreement between right and left hemisphere brain glucose values. When all groups were combined, right hemisphere brain glucose was 3.14 f 1.47 pmol/g and left hemisphere brain glucose was 3.12 f 1.51 pmollg (P > 0.6 by paired ?-test). Additionally, there was a significant correlation between brain glucose in the two hemispheres (r = 0.98; P < 0.0001). In the control group, blood glucose was 4.9 f 0.6 Fmol/ml (89 f 10 mg/dl), plasma glucose was 8.2 f 1.0 ~mol/ml, and brain glucose was 2.07 f 0.29 pmol/g. Brain glucose values in this group were similar to those reported for anesthetized rats in other studies [ 19,201. Saline infusion in the saline-treated group had no significant effect on blood, plasma, or brain glucose (Figs. l-3). Glucose infusion in the glucose-treated group resulted in blood, plasma, and brain glucose values that were significantly greater than control group values and values following saline (Figs. l-3). Blood glucose increased to a plateau of 10.9 f 1.1 pmol/ml (197 f 20 mg/dl) 12.2 f 1.9 kmol/ml (220 f 34 mg/dl), and 12.0 f 1.2 ~moVml(217 f 22 mg/dl) at the 60, 120, and 180 min measurement periods, respectively. Likewise, plasma glucose increased to 17.2 f 1.8 lmol/ml, 19.6 f 3.5
Table 1.
Physiologic
variables
at the time of blood and brain sampling
PaOz
PaCO,
(mmHg)
(mmHg)
pH
in halothane-anesthetized
Hct (‘%I)
rats.
Rectal temperature
MAP (mmHg)
(“C) Control
160 f
24
40 f
5
7.36 zt 0.04
42 f
3
84 zt 13
37.1 f
0.3
153 f 15 149 f 16 139 ?? 9 138 f 14*
37 38 38 42
f zt f f
4 3 5 4
7.37 ?? 0.03 7.33 f 0.03 7.36 f 0.03 7.35 ?? 0.02*
41 41 40 41
f f zt f
4 3 3 2
82* 81* 82zt 91 *
37.1 37.3 37.0 37.1
0.3 0.2 0.3 0.3
162 152 158 162
36 39 39 42
f I * 2 * 3 zt 4
7.38 7.35 7.35 7.32
38 40 40 43
iz * f f
2 1 3 3
83 ?? 21 89, 8 89zt 4 94 f 15
Sulk-fretted
Time of sampling 30 min 60 min 120 min I80 min
7 7 9 9
f * k f
Glucose-treated
Time of sampling 30 min 60 min 120 min 180 min
f zt f f
20 24 28 18
f f f zt
0.03 0.02 0.02 0.02+
37.1 ?? 0.2 37.3 f 0.3 37.1 ?? 0.3 37.3 f 0.3
MAP = mean arterial blood pressure; Hct = hematocrit. All values are presented as mean f SD. (N =6 for all groups). *Significant difference between glucose- and saline-treated subjects (P < 0.05) at the same time period.
+Significantly
different
from control
(P < 0.05).
69
:=: ?? ? -e ?
Control Glucoss-Trratod Salh-Trratod
t
01 -
30
60
90
120
150
160
Time(minutes) Fig. 1. Blood glucose following glucose or saline infusion in halothane-anesthetized rats: n = 6 for all, groups. All values are mean f SD. Vertical bars equal one SD. *P < 0.05 versus saline. +P < 0.05 versus control.
0-O -A f-*
20
-
15
-
10
-
5
-
-
Control Glucose-Tnatad SalineTreated
30
60
90
120
150
190
Time(minutes) Fig. 2. Plasma glucose following glucose or saline infusion in halothane-anesthetized rats: n = 6 for all groups. All values are mean + S.D. Vertical bars equal one SD. *P < 0.05 versus saline. +P < 0.05 versus control.
70 0-e A-A
?? -
Control Glucose-Treated Sailno-Treated
-*
6-
o-
cmt?Ql
30
60
90
120
100
150
Time (minutes) rats: n = 6 for all Fig. 3. Brain glucose following glucose or saline infusion in halothane-anesthetized groups. All values are mean f SD. Vertical bars equal one S.D. *P < 0.05versus saline. +P < 0.05 versus control.
pmol/ml, and 19.4 f 2.2 ,umollml, respectively. Accompanying these increases were increases in brain glucose to 4.39 f 0.44 pmol/g, 5.19 f 0.95 pmol/g, and 5.08 f 0.43 pmol/g, respectively. The ratio of brain to plasma glucose did not differ from control with either saline 0-O -A f - -e
E E
0.35
-
0.30
-
0.25
-
Control Glucoro-Tnated Sallno-Treated
3 G
--.
--
I
I
i
I --+---___
t
--
f -_ 1
it 8 3 G
.-s
,o m
0.10 ConhI
30
60
90
120
150
160
Time (minutes) glucose in halothane-anesthetized rats: n = 6 for all groups. All Fig. 4. The ratio of brain-to-plasma values are mean f SD. Vertical bars equal one SD. There were no significant differences between glucose- and saline-treated groups, nor were there significant differences versus control.
71
or glucose infusion (Fig. 4). Irrespective of the treatment, there was significant correlation between blood and plasma glucose (r = 0.99; P e O.OOl), blood and brain glucose (r = 0.96; P --s O.OOl), and plasma and brain glucose (r = 0.97; P 4 0.001). DISCUSSION
Hyperglycemia and increases in brain glucose will worsen the degree of injury the brain sustains following a period of global ischemia [l-8]. The proposed mechanism by which hyperglycemia augments cerebral ischemic injury is as follows: during periods of hyperglycemia, glucose crosses the blood-brain barrier by facilitated diffusion and other secondary pathways, producing increased brain glucose concentrations [4,9,10]. In the event of ischemia, the brain anaerobically metabolizes glucose to lactic acid. If the brain is exposed to an ischemic insult during a period of increased brain glucose, lactic acid formation will be increased [8, lo]. It is the lactic acid that appears to be the metabolic toxin responsible for the augmentation of ischemic neurologic injury [9,11]. The osmotic effects of increased blood and brain glucose appear to have little influence on the augmentation of ischemic cerebral injury [7]. Theoretically, during ischemia both free glucose as well as glucose liberated from the metabolism of glycogen will contribute to the magnitude of the lactic acidosis [8]. In support of this concept are in vivo data demonstrating a correlation between preischemic brain glucose increases and postischemic lactate accumulation [lo] or cellular acidosis [21], and in vitro data demonstrating the depletion of neuronal glycogen during anoxia [22]. In in vivo studies of experimental ischemia, investigators often exploit these metabolic considerations to vary the magnitude of postischemic injury. In order to avoid the confounding influence of metabolism of brain glycogen (a compound that is difficult to quantify acutely), investigators often select study groups with similar ages, weights, phenotypes, and nutritional histories. In this setting, investigator-controlled acute alterations in brain glucose, per se, independent of demonstrated or anticipated alterations in brain glycogen, have been reported to alter outcome following ischemia [4,10,23,24]. Based on the above-mentioned considerations, when designing models of experimental cerebral ischemia, it is important to control free brain glucose as rigidly as possible in the peri-ischemic period. Depending on the issue to be studied, it may be desirable for the brain glucose values to rigidly remain within the normal range, while at other times, it may be desirable to produce an equally stable but increased brain glucose. Regardless of the techniques and goals, it is ideal if blood, plasma, and brain glucose quantities are managed by a method that will allow a rapid and accurate assessment of brain glucose in the peri-ischemic period. Since it is technically impossible to obtain rapid, accurate direct measurements of brain glucose, it then becomes important to alter brain glucose using a technique that will permit accurate assessment using an extracerebral monitor. The present study assessed the use of continuous infusions of saline or glucose to produce stable and predictable alterations in brain glucose. Using the techniques described, brain glucose values ranging from 1.07 to 6.84 pmol/g could be accurately predicted by measuring either plasma glucose or the more rapidly obtainable blood glucose. In addition to the technique we studied, there are several other alternative meth-
12
ods for manipulating brain glucose concentrations in study animals. One option is to make all animals hyperglycemic with a glucose bolus or infusion, and after the glucose load, wait until blood glucose levels return to a predefined level before inducing &hernia. Indeed, this technique has been reported from several laboratories, including our own [4,25]. The problem with this methodology is that the production of ischemia must be performed precisely at the desired time, or the blood glucose will decline to a value less than the target value. Furthermore, a recent study from our laboratory demonstrated that - in this scenario - during the period of decreasing blood glucose there may be a hysteresis between rapidly changing blood and brain glucose values. If hysteresis occurs, the blood glucose measurements may underestimate the glucose concentrations in the brain [ 141. Thus, brain glucose concentrations may be greater [14], and postischemic neurologic injury may be worse [4], than would be predicted by examining blood or plasma glucose values. A second method to alter brain glucose in a graded fashion would be to begin with diabetic rats, and treat the rats with an insulin infusion until blood glucose values reach a given range. Recent studies in our laboratory examining this technique found that insulin infusions, per se, had no effect on the relationship between blood and plasma glucose, blood and brain glucose, or plasma and brain glucose [15]. Unfortunately, for a given plasma glucose concentration, diabetic rats had an approximately 25-50% greater brain glucose concentration than non-diabetic rats, because the diabetic brain tends to store glucose [ 15,261. Thus, for a given blood glucose concentration, it is difficult to make comparisons between brain glucose concentration estimates in non-diabetic versus insulin-treated diabetic rats. Furthermore, when using diabetic rats, one must contend with the special needs of the diabetic rat in the pre- and postischemic periods (i.e. increased fluid needs, increased risk of infection, possible need for continued insulin therapy, poor wound healing), particularly if chronic studies are planned. A third method of rigidly controlling brain glucose was evaluated by the present study: i.e. the use of a constant rate saline or glucose infusion to produce plateau concentrations of both blood and brain glucose. These data indicated that, using this technique, blood, plasma, and brain glucose could be effectively manipulated over a wide range of values. Within the range of blood glucose values studied [i.e. values from 3.6 to 15.0 pmol/ml (65-271 mg/dl)], there was excellent agreement between blood, plasma, and brain glucose. The technique is simple, and by adjusting the glucose infusion rate, should provide stable brain glucose values that are assessable using blood or plasma glucose measurements. Furthermore, our data suggest that the use of small glucose infusions to produce slight increases in brain glucose may result in a slightly better correlation between plasma and brain glucose, or alternatively blood and brain glucose, than occurs in untreated or saline-treated animals (Fig. 4). If the technique has a fault, it is as follows: previous studies from our laboratory suggest that if too large a glucose infusion is given too rapidly, there may be saturation of portions of the mechanisms responsible for transporting glucose into the brain, resulting in a hysteresis between blood and brain glucose. However, when lesser glucose infusions are given (1 g/kg per h in the present study versus 6 g/kg per h in our previous study [ 14]), there is no hysteresis between brain and blood glucose. In summary, the present study in anesthetized rats evaluated the use of continuous
73
i.v. infusions of glucose to produce stable increases in blood, plasma, and brain glucose. The method was effective in producing stable glucose alterations, and in preserving the correlation between blood, plasma, and brain glucose concentrations found in untreated or saline-treated animals. The technique is easy to use, and avoids many of the problems associated with other methods in which brain glucose may be altered during studies of experimental cerebral ischemia. ACKNOWLEDGEMENT
The authors wish to thank William Gallagher, Leslie Phelps, and Rebecca Wilson for their technical assistance; and John D. Michenfelder, M.D., for critically reviewing the manuscript. REFERENCES
1
W.T. Longstreth Jr., P. Diehr and T.S. Inui, Prediction arrest. N. Engl. J. Med., 308 (1983) 1378-1382.
2
W.T. Longstreth Jr. and T.S. Inui, High blood glucose level on hospital admission and poor neurological recovery after cardiac arrest, Ann. Neural., 15 (1984) 59-63. W. Pulsinelli, S. Waldman, B. Sigsbee, D. Rawlinson, P. Scherer and F. Plum, Experimental hyperglycemia and diabetes mellitus worsen stroke outcome, Trans. Am. Neural. Assoc., 195 (1980)
3
4
5 6 7 8 9
10
II
12
13
14
of awakening
after out-of-hospital
cardiac
21-24. W.L. Lanier, K.J. Stangland, B.W. Scheithauer, J.H. Milde and J.D. Michenfelder, The effects of dextrose infusion and head position on neurologic outcome after complete cerebral ischemia in primates: examination of a model, Anesthesiology, 66 (1987) 39-48. R.E. Myers and S. Yamaguchi, Nervous system effects of cardiac arrest in monkeys, Arch. Neurol., 34 (1977) 65-74. E. Siemkowicz and A.J. Hansen, Clinical restitution following cerebral ischemia in hypo-, normo-, and hyperglycemic rats, Acta Neurol. Stand., 58 (1978) 1-8. W.A. Pulsinelli, S. Waldman, D. Rawlinson and F. Plum, Moderate hyperglycemia augments ischemic brain damage: A neuropathologic study in the rat, Neurology, 32 (1982) 1239-1246. F. Plum, What causes infarction in ischemic brain?: The Robert Wartenberg Lecture, Neurology, 33 (1983) 222-233. E. Siemkowicz and A. Gjedde, Post-ischemic coma in rat: Effect of different preischemic blood I IO (1980) glucose levels on cerebral metabolic recovery after ischemia, Acta Physiol. Stand., 225-232. F.A. Welsh, M.D. Ginsberg, W. Rieder and W.W. Budd, Deleterious effect of glucose pretreatment on recovery from diffuse cerebral ischemia in the cat. II. Regional metabolite levels, Stroke, I1 (1980) 355-363. R.E. Myers, Lactic acid accumulation as cause of brain edema and cerebral necrosis resulting from oxygen deprivation. In: Advances in Perinatal Neurology. Editors: R. Korobkin and C. Guilleminault, Spectrum, New York, NY. 1979, pp. 85-l 14. W.L. Lanier, W.J. Perkins, B.R. Karlsson, J.H. Milde, B. W. Scheithauer. G.T. Shearman and J.D. Michenfelder, The effects of dizocilpine maleate (MK-801). an antagonist of the N-methyl-Daspartate receptor, on neurologic recovery and histopathology following complete cerebral ischemia in primates, J. Cereb. Blood Flow Metab., 10 (1990) 252-261. A. Tateishi, J.E. Fleischer, J.C. Drummond, MS. Scheller, M.H. Zornow, M.R. Grafe and H.M. Shapiro, Nimodipine does not improve neurologic outcome after 14 min of cardiac arrest in cats, Stroke, 20 (1989) 1044-1050. M.R. Weglinski and W.L. Lanier, The effects of transient hyperglycemia on brain glucose in rats anesthetized with halothane, Anesthesiology, 73 (1990) 93-98.
74 15 16 17
18 19 20
21 22 23
24
25
26
R.E. Hofer and W.L. Lamer, The effect of insulin infusion on brain and blood glucose in hyperglycemic diabetic rats (abstract), Anesthesiology, 73 (1990) A1272. U. Ponten, R.A Ratcheson, L.G. Salford and B.K Siesjo, Optimal freezing conditions for cerebral metabolites in rats, J. Neurochem., 21 (1973) 1127-l 138. J. Folbergrova, V. MacMillan and B.K. Siesjo, The effect of moderate and marked hypercapnia upon the energy state and upon the cytoplasmic NADH/NAD+ ratio of the rat brain, J. Neurochem., 19 (1972) 2497-2505. O.H. Lowry, J.V. Passonneau, F.X. Hasselberger and D.W. Schulz, Effect of &hernia on known substrates and cofactors of the glycolytic pathway in brain, J. Biol. Chem., 239 (1964) 18-30. W.A. Kotlce, R.A. Hawkins, D.W, Davis and J.F. Beibuyck, Comparison of the effects of volatile anesthetics on brain glucose metabolism in rats, Anesthesiology, 66 (1987) 810-813. D.F. Dedrick, Y.D. Scherer and J.F. Biebuyck, Use of a rapid brain-sampling technique in a physiologic preparation: Effects of morphine, ketamine, and halothane on tissue energy intermediates, Anesthesiology, 42 (1975) 651457. W.R. Marsh, R.E. Anderson and T.M. Sundt, Jr., Effect of hyperglycemia on brain pH levels in areas of focal incomplete cerebral &hernia in monkeys, J. Neurosurg., 65 (1986) 693496. P. Lipton, Regulation of glycogen in the dentate gyrus of the in vitro guinea pig hippocampus; effect of combined deprivation of glucose and oxygen, J. Neurosci. Methods, 28 (1989) 147-154. L.G. D’Alecy, E.F. Lundy, K.J. Barton and G.B. Zelenoch, Dextrose containing intravenous fluid impairs outcome and increases death after eight minutes of cardiac arrest and resuscitation in dogs, Surgery, 100 (1986) 505-51 I. K. Nakakimura, J.E. Fleischer, J.C. Drummond, M.S. Scheller, M.H. Zornow, M.R. Grafe and H.M. Shapiro, Glucose administration before cardiac arrest worsens neurologic outcome in cats, Anesthesiology, 72 (1990) lOO5-IO1 I. P.A. Steen, S.E. Gisvold, J.H. Milde, L.A Newberg, B.W. Scheithauer, W.L. Lanier and J.D. Michenfelder, Nimodipine improves outcome when given after complete cerebral ischemia in primates, Anesthesiology, 62 (1985) 40-14. R.E. Hofer, W.L. Lanier, Effects of insulin diabetic rats, Stroke, 22 (1991) 505-509.
on blood,
plasma
and brain glucose
in hyperglycemic
