Hyperglycemia increases cerebral intracellular acidosis during circulatory arrest

Hyperglycemia Increases Cerebral Intracellular Acidosis During Circulatorv Arrest U

J

Richard V. Anderson, MD, Michael G. Siegman, MD, Robert S. Balaban, PhD, Toni L. Ceckler, PhD, and Julie A. Swain, MD Surgery Branch and Laboratory of Cardiac Energetics, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland, and Department of Surgery, University of Nevada School of Medicine, Las Vegas, Nevada

Phosphorus 31 nuclear magnetic resonance spectroscopy was used to assess cerebral high-energy phosphate metabolism and intracellular pH in normoglycemic and hyperglycemic sheep during hypothermic circulatory arrest. Two groups of sheep (n = 8 per group) were placed in a 4.7-T magnet and cooled to 15°C using cardiopulmonary bypass. Spectra were acquired before and during circulatory arrest and during reperfusion and rewarming. Intracellular pH and adenosine triphosphate levels decreased during circulatory arrest. Compared with the normoglycemic animals, the hyperglycemic group was significantly more acidotic with the greatest difference observed during the first 20 minutes of reperfusion (6.40

H

yperglycemia may develop in patients subjected to the stress of cardiac operation and hypothermic cardiopulmonary bypass (CPB). Disturbances of insulin and glucose metabolism stem from impaired insulin production, decreased glucose utilization, and increased levels of gluconeogenic hormones, such as epinephrine, cortisol, glucagon, and growth hormone [l, 21. In addition, the use of dextrose-containing solutions for peripheral fluid administration, CPB prime, and cardioplegia delivers a substantial exogenous glucose load. In nondiabetic patients this condition is self-limiting and often goes untreated. In diabetic patients, a more profound hyperglycemia may develop, resulting in serious complications and even death [3].

See also page 1131. During ischemia, the metabolism of glucose by anaerobic glycolysis results in lactate production and intracellular acidosis. Hyperglycemia will potentiate this reaction, resulting in increased lactate production and a further decrease in intracellular pH. During normothermic ischemia, severe acidosis increases cell damage and impairs energy metabolism after ischemia [4]. The effects of hyperglycemia during hypothermic cereAccepted for publication March 23, 1992 Presented at the Clinical Congress Surgical Forum of the American College of Surgeons, Chicago, IL, Oct 20-25, 1991. Address reprint requests to Dr Swain, Division of Cardiovascular Surgery, University of Nevada School of Medicine, 2040 W Charleston Blvd, Suite 601, Las Vegas, NV 89102.

0 1992 by The

Society of Thoracic Surgeons

f 0.08 versus 6.08 2 0.06; p < 0.001). Intracellular pH returned to baseline after 30 minutes of reperfusion in the normoglycemic group but did not reach baseline until 1hour of reperfusion in the hyperglycemic animals. Adenosine triphosphate levels were significantly higher in the hyperglycemic group during circulatory arrest. Repletion of adenosine triphosphate during reperfusion was similar for both groups. These results support the hypothesis that hyperglycemia during cerebral ischemia drives anaerobic glycolysis and thus leads to increased lactate production and a decrease in the intracellular acidosis normally associated with ischemia. (Ann Thorac Surg 1992;54:1126-30)

bra1 ischemia and reperfusion have not been defined. In this study, deep hypothermic circulatory arrest was used as a model of global cerebral ischemia. We employed in vivo phosphorus 31 nuclear magnetic resonance spectroscopy (NMR) to examine the effects of hyperglycemia on intracellular pH and high-energy phosphate metabolism during ischemia and reperfusion in the sheep brain.

Material and Methods

Animal Preparation All animal experiments were conducted with the approval of the Animal Care Use Committee of the National Heart, Lung, and Blood Institute in accordance with the “Guide for the Care and Use of Laboratory Animals” (NIH publication No. 85-23, revised 1985). Twenty sheep weighing between 18 and 34 kg were divided into two groups (normoglycemia and hyperglycemia). Hornless sheep were used because of the relatively thin skull bone in the frontoparietal region, which reduces interference with the NMR signal. Juvenile sheep (greater than 8 weeks of age) were used because their erythrocytes lack 2,3-diphosphoglycerate, which interferes with the accurate determination of the resonance position of inorganic phosphate. The inorganic phosphate resonance position is required for the calculation of intracellular pH. Anesthesia was induced with intravenous Telazol (1:l tiletamine hydrochloride and zolazepam hydrochloride), 0.1 mL/2.5 kg. Endotracheal intubation was performed, and inhalation anesthesia was maintained with a mixture of 1%halothane and 99% oxygen. Muscle paralysis was 0003-4975/92/$5.OO

ANDERSON ET AL HYI'ERGLYCEMIA DURING CIRCULATORY ARREST

Ann Thorac Surg 1992;54112&30

obtained with intravenous administration of pancuronium bromide, 0.1 mg/kg. Arterial pressure and venous infusion catheters were placed in the right femoral artery and vein, respectively. The left femoral artery was cannulated with a 14F or 16F CPB catheter. A 24F lighthouse-tip CPB cannula was placed into the right atrium through the right internal jugular vein. Heparin sodium (300 U/kg) was administered intravenously before arterial and venous cannulation. An insulated, single-turn, custommade radiofrequency coil 2 cm in diameter was secured over the exposed skull. The animal was then wrapped in an insulating blanket, placed in a Lucite cradle, and positioned in a 4.7-T, 30-cm clear-bore magnet. Arterial blood oxygen tension was maintained greater than 100 mm Hg; normothermic carbon dioxide tension was maintained between 35 and 45 mm Hg. Mean arterial pressure was maintained between 60 and 90 mm Hg at 37°C and between 40 and 60 mm Hg at 15°C. Nasopharyngeal temperature was monitored with a temperature probe unaffected by the magnetic field. Serum electrolyte and glucose levels were intermittently measured using a Chem Pro 1000 Clinical Chemistry Analyzer (Ortho Diagnostic Systems Inc, Raritan, NJ).

Cardiopulmonary Bypass The bypass circuit consisted of a centrifugal pump (model BP-80; Bio-Medicus, Eden Prairie, MN), a 33-pm arterial filter (model H-625; Bard Cardiopulmonary Division, Santa Ana, CA), and a hollow-fiber membrane oxygenator with integral heat exchanger (Maxima; Medtronic Blood Systems, Inc, Anaheim, CA). A cardiotomy reservoir (Gish Biomedical, Inc, Santa Ana, CA) was inserted parallel to the venous line for the administration of volume and as a reservoir for exsanguination of the animal during the arrest period. The pump was positioned 4.5 m from the magnet, and this necessitated corresponding lengths of extension tubing. The pump was primed with 2,500 mL of lactated Ringer's solution, 10,000 units of heparin, 50 mEq of sodium bicarbonate, 12.5 g of mannitol, 80 mg of 2% lidocaine hydrochloride, and 1 unit of sheep whole blood. Blood flow during CPB was maintained between 60 and 100 mL kg-' min-'. Flow was measured with an electromagnetic flowmeter (Bio-Medicus) that had previously been calibrated in the weak magnetic field (15 G) where the pump was positioned. The oxygenator was ventilated with a mixture of 1%halothane and 99% oxygen. After baseline NMR spectra were acquired at 37°C for 20 minutes, the animal was cooled to 15°C. Cooling took 29 to 53 minutes. During hypothermia, the arterial blood pH was managed using the alpha-stat scheme. After spectra acquisition at 15°C for 20 minutes, circulation was arrested for 1 hour, followed by 2 hours of reperfusion and rewarming. Before reperfusion, 25 g of mannitol and a second unit of whole blood were added to the cardiotomy reservoir. Inotropic support (dopamine hydrochloride, 5 pg * kg-' . min-l) was initiated with reperfusion in all animals. Rewarming was accomplished by maintaining a 10°C gradient between the blood and water bath temperatures. Sheep in which massive edema developed, manifested as

-

-

1127

pulmonary edema in the ventilator tubing and no return of cerebral adenosine triphosphate (ATP) levels, were unable to be weaned from bypass and were excluded from the study. Four of the 12 control animals had to be excluded for this reason. (These 4 animals required large volumes of fluid to maintain adequate flows while on CPB.) None of the hyperglycemic animals showed massive edema. Animals weaned from bypass were ventilated for the duration of the experiment. No attempt was made to wean from the ventilator. At the completion of 2 hours of reperfusion, animals were killed with intravenous administration of potassium chloride. Total time on CPB, including the arrest period, ranged from 4 hours 20 minutes to 4 hours 48 minutes.

Glucose Monitoring and Treatment All animals fasted 12 hours before the experiment. The initial glucose level was determined as soon as venous access was obtained but before the start of maintenance fluid administration. All animals were initially normoglycemic (glucose level < 8.3 mmoVL [150 mg/dL]). The control (normoglycemic) animals received lactated Ringer's as maintenance fluid. The animals in the hyperglycemic group were administered lactated Ringer's with 5% dextrose as maintenance fluid. By the time the animals were transferred to the magnet and preparations for CPB had been completed, the animals in the treatment group had become hyperglycemic (glucose level > 13.9 mmol/L [250 mg/dL]). Serum glucose levels were measured before initiation of CPB, during CPB at 37"C, during CPB at 15T, during early reperfusion, and at 1 hour of reperfusion when rewarming was complete. Animals in the treatment group with glucose levels that fell to less than 13.9 mmoVL received intravenous injections of 50% dextrose in 12.5mEq increments until the level was once again greater than 13.9 mmoVL. This was necessary on only three occasions.

N M R Spectroscopy The techniques of NMR spectroscopy used in this study have been described previously [5]. The animal was positioned so that the brain was centered in the bore of the magnet. The NMR data were collected with a General Electric CSI spectrometer (General Electric NMR Instruments, Fremont, CA). The data were acquired in 20minute blocks of 640 averaged acquisitions at 37°C before CPB, 37°C on CPB, and 15°C on CPB. Nine consecutive 20-minute blocks were then acquired, three during the arrest period and six during the reperfusion and rewarming period. Data were analyzed using the GEMCAP Lorentzian line-fitting program, a part of the GEMCSI software. Concentrations of ATP and phosphocreatine (PCr) are reported as percent change from baseline levels acquired at 37°C off CPB. Intracellular pH was determined by measuring the chemical shift of the inorganic phosphate peak relative to the position of the pH-independent PCr peak. The relation between pH and inorganic phosphate

1128

ANDERSON ET AL HYPERGLYCEMIA DURING CIRCULATORY ARREST

400 -

-

GLUC0SE lrngldll

200 -

1-

P

,,

,,/'

j:

-----___

,' "

Ann Thorac Surg 1992543 12630

8-

f

.---------t;_ _ _ _ _ _ _ _ _

NOHMOGLYCEMlA

NORMOGLYCEMIA

0 HYPERGLYCEMIA * D 0 05 YE NORMOGLYCEMIA

0 HYPERGLYCEMIA

*

licOOO1

YI

NORMOGLYCEM~A

-----t---i-i 0-

I

I

-9.

1

I

I

chemical shift at 37°C and 15°C has been determined previously [61.

Statistical Analysis

/'

(Fig 2). This is consistent with the observations of our group [5, 71 using a similar model. During the arrest period and through the first 20 minutes of reperfusion, the intracellular pH markedly decreased in both groups. Compared with the normoglycemic group, the hyperglycemic animals became significantly more acidotic. Intracellular pH returned to baseline after 30 minutes of reperfusion in the normoglycemic group but did not reach baseline until 1 hour of reperfusion in the hyperglycemic animals. The ATP and PCr concentrations followed a course similar to that of pH. Hypothermic bypass resulted in increased levels of ATP and PCr in both groups, which subsequently decreased with circulatory arrest. Compared with the normoglycemic animals, the hyperglycemic group demonstrated significantly higher concentrations of ATP during hypothermic bypass and arrest (Fig 3). Repletion of ATP during the reperfusion period was similar for both groups.

*

All results are reported as the mean the standard deviation. Statistical significance to the 95% confidence level was determined using the Excel Statmac macro statistical program. Unpaired Student's t tests were used for intergroup comparisons of means. A multivariate two-factor (group, period) analysis of variance was used as an omnibus screening test, followed by univariate two-factor analyses of variance and post hoc t tests.

Results Serum glucose levels were significantly greater in the hyperglycemic group throughout each acquisition period (Fig 1).Changes in ATP, PCr, and pH are shown in Table 1. Cardiopulmonary bypass at 15°C produced an increase in intracellular pH in control and hyperglycemic animals

Table 1 . Changes in High-Energu Phosvhates and lntracellular vH" PHi Time 37°C 37°C CPB 15°C CPB Arrest (min) 10 30 50 Reperfusion (min) 10 30 50 70 90 110

%A PCr

NG

HG

NG

7.01 t 0.03 7.01 ? 0.02 7.59 f 0.02

7.01 t 0.02 6.99 f 0.05 7.56 f 0.06

100 99 ? 3 123 f 5

6.98 f 0.26 6.58 f 0.06 6.48 t 0.05

7.17 f 0.03 6.48 f 0.05" 6.23 f 0.07'

6.40 f 0.08 6.97 ? 0.08 7.03 ? 0.08 7.07 C 0.05 7.11 f 0.08 7.07 f 0.03

6.08 ? 0.06' 6.46 2 0.23' 6.84 2 O.Mb 6.98 -+ 0.15 7.02 f 0.14 7.05 f 0.14

69 29 20

10 5 f4 ?

f

60 f 6 109 ? 6 110 ? 6 113 ? 9 112 f 8 112 ? 7

%A ATP HG 100 95 ? 4 125 f 6 78 30 20

NG

HG

100 99 -r- 5 109 f 5

98 f 3 115 f 5b

9 5 f6 f

f

54 f 7 87 f 13' 97 f lod 101 ? 15 103 ? 13 103 f 14

Data are shown as the mean t the standard deviation. Significance: p < 0.05 compared with normoglycemic group. Significance: p < 0.01 compared with normoglycemic group. compared with normoglycemic group.

a

ATP = adenosine triphosphate; pH, = intracellular pH.

CPB

=

cardiopulmonary bypass;

HG

=

hyperglycemia;

NG = normoglycemia;

86 44 34

?

67 92 93 90 89 90

f

100

107 ? 6' 62 ? 8' 40 f 2"

4

f4 f

5

5 4 ? 4 f7 ? 5 f6

74 96 93 89 89 89

?

f7

6 4 f6 ? 6 f8 ? ?

Significance: p < 0.001 PCr

=

phosphoaeatine;

ANDERSON ET AL HYFERGLYCEMIA DURING CIRCULATORY ARREST

Ann Thorac Surg 1992:54 112G30

I

0-1,

,

I

37OC 37'C 15OC c p ~c p

I

I

I

10

30

50

~- A R R E S T I m i n I l

*

,

10

I

30

-REPERFUSION

,

D ~005vsNORMOGLVCEMlA

50

,

70

I

90

I 110

lminl-

Fig 3. Percent change in cerebral adenosine triphosphate (ATP) levels compared with baseline of 100%. (CPB = cardiopulmonary bypass.)

Comment Experimental studies based on models of focal and global normothermic cerebral ischemia have demonstrated a relationship between hyperglycemia, lactic acidosis, and cellular damage. It has been shown that the normothermic brain will exhaust all free-energy stores after as little as 2 minutes of complete ischemia [8]. After ATP depletion, anaerobic glycolysis metabolizes each glucose molecule into two ATP molecules, two lactate molecules, and two hydrogen ions, resulting in intracellular acidosis. Hyperglycemia will potentiate this reaction and result in increased lactate production and a further decrease in intracellular pH [4, 91. Intracellular acidosis, in conjunction with other ischemia-induced alterations of cellular metabolism, produces dysfunction of ion-pumping enzyme systems, disruption of ion gradients, changes in membrane permeability, lipolysis, and proteolysis, and may lead to irreversible membrane damage [9]. Clinical data [lo, 111 suggest that diabetic and nondiabetic patients in whom hyperglycemia develops after the onset of ischemic stroke have a worse neurological outcome than nondiabetic, normoglycemic patients. The present study demonstrates that during hypothermic cerebral ischemia and early reperfusion, intracellular acidosis is exacerbated by hyperglycemia. Because the experimental design did not allow for the measurement of lactate concentrations, we are lacking direct evidence that increased anaerobic glycolysis and lactate accumulation is the primary mechanism of acidosis. Using 31P and 'H NMR spectroscopy, Corbett and colleagues I121 were able to demonstrate a linear relationship ( r = 0.94) between intracellular pH and lactate levels in neonatal piglet brains during normothermic hemispheric ischemia. The linear relationship extended to lactate concentrations of 40 pmol/g. Using a similar model in adult cats, Gyulai and co-workers [13] reported a strong correlation between intracellular pH and brain lactate concentrations up to 20 prnollg. In this study, both groups of animals exhibited increased levels of high-energy phosphates (ATP and PCr) during hypothermic CPB. As postulated by our group [5], this increase may be the result of a relative decrease in ATPase activity versus ATP synthesis (through glycolysis

1129

or oxidative phosphorylation) at low temperature. This would also explain the greater increase in ATP observed in the hyperglycemic animals (see Fig 3). During hypothermic bypass, increased levels of glucose might drive glycolysis and oxidative phosphorylation while ATPase activity is concurrently decreased, resulting in greater levels of ATP. With the onset of ischemia, hyperglycemia continues to drive glycolysis, resulting in a delayed decrease in ATP concentration and increased production of lactic acid. Studies using rats and mice have demonstrated that hyperglycemia induced before normothermic ischemia significantly impairs postischemic recovery of brain energy metabolism. Gardiner and associates [14] subjected normoglycemic and hyperglycemic rats to 15 minutes of normothermic hemispheric ischemia by clamping the right carotid artery. Both groups experienced a substantial but similar decrease in PCr and ATP. Lactate concentration increased significantly in both groups, with the lactate levels in the hyperglycemic animals being much higher than in the normoglycemic group. Recovery of energy metabolism was significantly slower in the hyperglycemic group, and this was associated with a marked, persistent lactic acidosis. Using the same model, Welsh and co-authors [15] obtained similar results with mice. Despite the greater severity and prolonged presence of acidosis in the hyperglycemic animals in this study, both groups exhibited nearly complete recovery of energy metabolism, a finding that leads us to conclude that major irreversible cell damage did not occur during ischemia and reperfusion. Hypothermia may have protected cellular integrity and preserved energy metabolism by slowing overall metabolism, thus making the cell less sensitive to acidosis and other ischemic insults. Although we have no evidence of cellular damage, it is conceivable that the areas of the sheep brain most susceptible to ischemia (which have not been defined) were not sampled by the NMR coil, which, in this model, was over the frontoparieta1 cortex. Moreover, it is possible that microinfarction occurred, but the extent was not great enough to significantly alter energy metabolism. The uniform recovery of energy metabolism in both groups of animals may be explained in terms of a lactate threshold. Previous work using rats has shown that cerebral ischemia resulting in tissue lactate levels less than 16 pmoVg produces only selective neuronal damage in predictably vulnerable areas. Astrocytes and endothelial cells are spared, thus providing a protective effect by preventing widespread cerebral edema. With lactate levels greater than 16 pmoVg, astrocytes and endothelium as well as neurons are damaged, resulting in infarction and progressive cerebral edema [4]. Tissue lactate levels achieved in this model may have remained below a threshold level, resulting in minimal or no cell damage. In summary, hyperglycemia resulted in increased cerebral intracellular acidosis during ischemia and reperfusion but did not adversely affect postischemic recovery of energy metabolism. Energy metabolism during the early ischemic period was enhanced by hyperglycemia. Further investigation is needed to define the relationship between

1130

ANDERSON ET AL

HYPERGLYCEMIA DURING CIRCULATORY ARREST

acid-base metabolism a n d energy metabolism d u r i n g hypothermia a n d hypothermic ischemia. We thank Stephanie Sellers and Daryl Despres for the animal care assistance they provided.

References 1. Black PR, Van Devanter S, Cohn LH. Effects of hypothermia on systemic and organ system metabolism and function. J Surg Res 1976;20:49-63. 2. Weiland AP, Walker WE. Physiologic principles and clinical sequelae of cardiopulmonary bypass. Heart Lung 1986;15: 369. 3. Mills NL, Beaudet RL, Isom OW, Spencer FC. Hyperglycemia during cardiopulmonary bypass. Ann Surg 1973;177: 20S5. 4. Plum F. What causes infarction in the ischemic brain? The Robert Wartenberg Lecture. Neurology 1983;33:222-33. 5. Swain JA, McDonald TJ Jr, Balaban RS, Robbins RC. Metabolism of the heart and brain during hypothermic cardiopulmonary bypass. Ann Thorac Surg 1991;51:105-9. 6. Kost GJ. pH standardization for phosphorous-31 magnetic resonance heart spectroscopy at different temperatures. Magn Reson Med 1990;14:49&506. 7. Swain JA, McDonald TJ Jr, Robbins RC, Balaban RS. Relationship of cerebral and myocardial intracellular pH to blood pH during hypothermia. Am J Physiol 1991;26O(Suppl): H1640-4.

Ann Thorac Surg 1992;54:1126-30

8. Lowery OH, Passonneau JV, Hasselberger FX, Schulz DW. Effect of ischemia on known substrates and cofactors of the glycolytic pathway in brain. J Biol Chem 1964;239:18-30. 9. Siesjo BK, Wieloch T. Cerebral metabolism in ischaemia: neurochemical basis for therapy. Br J Anaesth 1985;57:47-62. 10. Pulsinelli WA, Levy DE, Sigsbee B, Scherer P, Plum F. Increased damage after ischemic stroke in patients with hyperglycemia with or without established diabetes mellitus. Am J Med 1983;74:540-4. 11. Berger L, Hakim AM. The association of hyperglycemia with cerebral edema in stroke. Stroke 1986;17:86571. 12. Corbett RJT, Laptook AR, Nunnally RL, Hassan A, Jackson J. Intracellular pH, lactate, and energy metabolism in neonatal brain during partial ischemia measured in uiuo by 31Pand 'H nuclear magnetic resonance spectroscopy. J Neurochem 1988;51:1501-9. 13. Gyulai L, Schnall M, McLaughlin AC, Leigh JS, Chance B. Simultaneous 31Pand 'H nuclear magnetic resonance studies of hypoxia and ischaemia in the cat brain. J Cereb Blood Flow Metab 1987;7543-51. 14. Gardiner M, Smith M, Kagstrom E, Shohami E, Siesjo BK. Influence of blood glucose concentration on brain lactate accumulation during severe hypoxia and subsequent recovery of brain energy metabolism. J Cereb Blood Flow Metab 1982;2:429-38. 15. Welsh FA, Sims RE, McKee AE. Effect of glucose on recovery of energy metabolism following hypoxia-oligemia in mouse brain: dose-dependence and carbohydrate specificity. J Cereb Blood Flow Metab 1983;3:486-92.