Systemic administration of alpha2 agonists causes sedation [1] [2] [3,4] 2 receptors.
Adrenergic receptors are found on cerebral vessels of all sizes throughout the brain [5,6] 2 agonists. In addition, a central site of action of alpha2 agonists is suggested by the well defined adrenergic system originating in brain areas involved in control of cardiovascular function [7] 2 , as bilateral lesions of the locus coeruleus decrease in CBF response to hypercapnia in cats [8] 2 ) agonist [9] 2 agonists.
We tested the hypothesis that the previously demonstrated global decrease in CBF during normoxia and the global restraint in cerebral vasodilation during hypoxia, which is produced by a systemically administered alpha2 agonist [4] 2 agonist dexmedetomidine (DEX) via ventricular-cisternal perfusion and measured its effect on normoxic CBF, hypoxia-induced cerebral hyperemia, cerebral oxygen consumption (CMRO2 ) and cerebral oxygen delivery.
Methods
This study was approved by our animal care and use committee. Nineteen beagle dogs (9-11 kg) were anesthetized with thiopental 12 mg/kg intravenously (IV), endotracheally intubated, and the lungs mechanically ventilated (Harvard respirator pump; Harvard Apparatus Co., South Natick, MA). Oxygen was administered to maintain PaO2 greater than 90mm Hg and isoflurane was administered in the inspiratory limb of the ventilator circuit by an isoflurane vaporizer (Ohio Medical Products, Model TC-3, Madison, WI). End-tidal isoflurane was monitored with an infrared analyzer designed for halogenated drugs (Nelcor 2500; Nelcor Inc., Haywood, CA) and the end-tidal isoflurane concentration was maintained at 1.4%. A single dose of pancuronium bromide (0.1 mg/kg, IV bolus) was administered to minimize muscle contraction from electrocautery in all animals. A catheter was placed retrograde via a femoral artery into the left ventricle for radiolabeled microsphere injection. Bilateral omocervical artery catheters were inserted and advanced into the aortic arch for measurement of mean arterial blood pressure (MABP), withdrawal of microsphere reference blood, and arterial blood gas analysis. Bilateral femoral vein catheters were placed for rapid removal of blood to prevent increases in cerebral perfusion pressure (CPP=MABP-sagittal sinus pressure) during administration of hypoxic gas mixtures.
The animal was turned prone and the head fixed in a head holder elevated 10 cm above midthoracic level. A catheter was placed in the sagittal sinus for withdrawal of cerebral venous blood and measurement of sagittal sinus pressure. A 3-mm craniectomy was performed 3-4 mm lateral to the midline and immediately anterior to the coronal suture over the left hemisphere. A 2.5-mm silicone elastic ventricular drain (Cordis, Miami, FL) was advanced 1-2 cm into the lateral ventricle after a dural incision. Location of the drain tip in the lateral ventricle was indicated by return of clear cerebrospinal fluid (CSF). Drain location was further verified by a rapid decrease of pressure in the lateral ventricle when CSF was allowed to escape from a catheter placed in the cisterna magna. Arterial and sagittal sinus pressure transducers were referenced to the right atrium. The lateral ventricular pressure and sagittal sinus pressure transducers were referenced to the level of the external auditory meatus. Pressures were measured (Isotec[TM] Transducer, Quest Medical Inc., Allen, TX) and recorded on a Gould-Brush polygraph (Cleveland, OH). Brain temperature was measured by a Mon-a-therm thermistor (Mallinckrodt Medical, St. Louis, MO) placed in the right parietal lobe and maintained at 38 +/- 1 degrees C with heating lamps throughout each protocol. Arterial blood glucose was measured with a glucose analyzer (YSI Model 2300; Yellow Springs Instrument, Co., Yellow Springs, OH).
Arterial and sagittal sinus blood PO2 , and pH were measured at 37 degrees C immediately after withdrawal using an ABL3 analyzer (Radiometer, Copenhagen, Denmark). Oxygen saturation and hemoglobin were measured spectrophotometrically with a Radiometer Hemoximeter OSM3. Arterial and cerebral venous O2 content were calculated from the measured O2 saturation and hemoglobin concentration and corrected for dissolved oxygen. CMRO2 was calculated by multiplying the arterial to cerebral venous O2 content difference by hemispheric CBF. Cerebrovascular resistance (CVR) was calculated by dividing CPP by blood flow to cerebral hemispheres. Cerebral O2 transport was calculated by multiplying blood flow to cerebral hemispheres by arterial O2 content. Cerebral oxygen extraction was calculated by dividing the arterial to sagittal sinus oxygen content difference by the arterial oxygen content.
CBF was measured with radiolabeled microspheres (15 +/- 0.5 micro m diameter; Dupont-NEN Products; Boston, MA) using the reference withdrawal method. Six radiolabels were used and injected in random sequence (153 ) Gd,114 In,113 Sn,103 Ru,95 Nb,46 Sc). Regional flow was determined in cerebral hemispheres, caudate nucleus, brainstem, cerebellum, and cerebral cortex (1 x 1 x 10-mm superficial slice from parietal lobe).
The central distribution of a drug administered according to the techniques of this study were assessed by determining the distribution of radioactivity produced after ventricular-cisternal administration of3 H-clonidine, also an alpha2 agonist, but less lipid soluble. Tritiated DEX was unavailable at the time of this study and a less soluble tracer was chosen to maximize diffusion through the CSF space. Tissue samples for3 H-clonidine measurement were stored at -70 degrees C and homogenized in 0.32 M sucrose prior to analyses. An aliquot of 100 micro L of tissue homogenate was used to quantify3 H-clonidine by liquid scintillation spectroscopy (Beckman Instruments, Model LS 1800, Fullerton, CA). Protein content of the homogenates was assayed by the method of Bradford [10]
Surgical preparation required approximately 45 min. During 15 min of stabilization the intraven-tricular drain was used for administration of 2 mL mock CSF [11]
In three animals, the effect of DEX concentration and specificity of effect by reversal with a highly specific alpha2 antagonist, atipamezole, were determined. After determination of baseline data, all variables were recorded after the ventriculocisternal perfusion was sequentially switched to DEX in concentrations of 1, 10, and 100 micro g/mL. After the last infusion of DEX, an intraventricular infusion containing atipamezole 1000 micro g/mL was administered. As with DEX, atipamezole was administered as 2 mL over 5 min and then an infusion at 0.2 mL/min was continued for 10 min. All variables were recorded near the end of the 10-min infusion period to assess the effect on CBF changes produced by the previously administered DEX or atipamezole.
In the hypoxic studies (n = 11) baseline measurements were obtained after 1 h of anesthesia at normoxia (15 min of mock CSF infusion). During lateral ventricular mock CSF infusion, isocapnic hypoxia was then produced by introducing nitrogen into the breathing circuit. All measurements were repeated 10 min after obtaining an inspired oxygen concentration of approximately 7%. The nitrogen was then discontinued and a third set of measurements obtained 15 min after reestablishing normoxia.
The intraventricular infusion was then changed to either mock CSF (same as initial infusion; n = 5) or DEX (100 micro g/mL dissolved in mock CSF; n = 6) with the initial 2 mL administered over 5 min. DEX and atipamezole were provided by Farmos Corp. (Turku, Finland). A fourth measurement was made 10 min after the initiation of DEX (DEX group) or mock CSF administration (CSF group). During lateral ventricular mock CSF or DEX infusion, isocapnic hypoxia was then produced by introducing nitrogen into the breathing circuit. All measurements were repeated 10 min after obtaining an inspired oxygen concentration of approximately 7%. The nitrogen was then discontinued and a sixth set of measurements obtained 15 min after reestablishing normoxia. In both groups, increases in CPP due to hypoxia were prevented by removal of blood into heparin-treated syringes. At the end of each experiment, the animal was killed with IV KCl and the brain harvested for subsequent analysis of regional CBF.
In an additional group of animals (CLONIDINE; n = 5), distribution of intraventricular infusion of a tritiated labeled tracer was assessed by infusion of3 H-clonidine (Sigma Chemical Co., St. Louis, MO).3 H-clonidine (0.32 micro g/mL; 20 micro Ci/mL) was dissolved in mock CSF. Two milliliters of the solution was infused over 5 min into the lateral ventricular catheter followed by an infusion of 0.2 mL/min for 30 min to mimic the total infusion time in the other two groups (i.e., from the start of the infusion to the end of hypoxia). During the infusion, the animals were maintained normothermic, normoxic, normocapnic, and normotensive. At the end of the infusion, the animal was killed with IV KCl and the brain rapidly removed and samples of regions similar to those in which CBF was measured were frozen for later analysis. In particular, samples were taken from regions adjacent to the CSF space and regions distant from the CSF space (cerebral hemisphere contralateral to the infusion site) as well as brainstem and cerebellum. Radioactivity for each sample was normalized to activity in the ipsilateral caudate nucleus, which was the region expected to have the highest number of counts per milligram of protein.
Two-way analysis of variance for between (group) and within (time) subjects design was used to compare total and regional CBF, blood gas, and hemodynamic variables during the measurement periods. P < 0.05 was considered significant. If a significant group x time interaction was demonstrated, one-way analysis of variance was performed for between group effects at individual time points and for within group effects (repeated measures). The Student-Newman-Keul's test was used to correct for multiple comparisons. Because SD increased with the mean value during hypoxia, a logarithmic transformation was performed on regional CBF data, prior to data analysis. Data in text, tables and figures are presented as mean +/- SE.
Results
(Figure 1 ) shows regional CBF in cerebellum, brainstem, and cerebrum after ventriculocisternal perfusion of DEX 1, 10, and 100 micro g/mL and atipamezole 1000 micro g/mL. In all three regions, there was a progressive decrease in flow with increasing concentrations of DEX with return to flow to baseline level with atipamezole infusion. In all animals, CPP and arterial blood gases were maintained at baseline level throughout the experiment (data not shown).
Figure 1: Flow in cerebellum, brainstem, and cerebrum in three animals receiving sequential ventriculocisternal perfusion of dexmedetomidine 1, 10, and 100 micro g/mL and atipamezole 1000 micro g/mL. Each infusion consisted of 2 mL administered into the ventricular catheter over 5 min followed by an infusion at 0.2 mL/min for a total of 10 min of drug exposure.
(Table 1 ) shows hemodynamic values in CSF and DEX groups during both episodes of hypoxia. During hypoxia a similar amount of blood was withdrawn (approximately 20 mL/kg) to prevent changes in CPP in each group. Therefore, CPP was similar between groups at each measurement point. Intraventricular infusion of DEX did not require withdrawal of blood and did not result in a change in CPP.
Table 1: Hemodynamic Values in Animals Receiving Mock Cerebrospinal Fluid (CSF) or Dexmedetomidine (DEX) Ventriculocisternal Perfusion
(Table 2 ) shows blood gas values in CSF and DEX groups throughout the protocol. PaO2 , PaCO2 and arterial oxygen content were similar in the two groups at each stage of the study. Within each group, the values were also similar during the two hypoxic challenges. Baseline O2 extraction was 0.24 +/- 0.04 and 0.23 +/- 0.02 in CSF and DEX groups, respectively. In both groups, O2 extraction increased with the initial hypoxia (0.36 +/- 0.03 and 0.44 +/- 0.06, respectively). Oxygen extraction after DEX (prior to the second exposure to hypoxia) was 0.36 +/- 0.02, which was greater than in CSF group after administration of mock CSF (prior to the second episode of hypoxia). Although extraction increased in both groups during the second episode of hypoxia, the extraction was higher in the DEX group (0.50 +/- 0.04 vs 0.31 +/- 0.03; P < 0.05).
Table 2: Blood Gas Values in Animals Receiving Mock Cerebrospinal Fluid (CSF) or Dexmedetomidine (DEX) Ventriculocisternal Perfusion
Baseline CMRO2 in CSF and DEX groups was 3.5 +/- 0.3 and 3.6 +/- 0.2 ml [center dot] min-1 [center dot] 100 g-1 , respectively. CMRO (2 ) was similar in the two groups at all periods of study except for a lower value during the second hypoxic challenge in the DEX group (3.9 +/- 0.4 vs 2.5 +/- 0.4 ml [center dot] min-1 [center dot] 100 g-1 ; P < 0.05).
(Table 3 ) shows flow to cerebellum, brainstem, caudate nucleus, cerebral hemispheres, and cerebral cortex in CSF and DEX groups. During baseline conditions, hypoxia increased flow to each of these regions. After recovery from hypoxia, DEX decreased normoxic flow in each region except the cerebellum. Although CBF increased in response to hypoxia in each region during DEX infusion, CBF values obtained during the second hypoxia challenge were significantly lower than those measured during the first hypoxic challenge.
Table 3: Regional Blood Flow Values (mL [center dot] min-1 [center dot] 100 g-1 ) in Animals Receiving Intraventricular Mock Cerebrospinal Fluid (CSF group, n = 5) or Dexmedetomidine (DEX group, n = 6)
(Figure 2 ) shows CVR in CSF and DEX groups. Baseline CVR was similar in the two groups (1.5 +/- 0.2 and 1.2 +/- 0.2 mL [center dot] min-1 [centered dot] mm Hg [center dot] 100 g, respectively). The absolute change in CVR during the first hypoxia challenge was not statistically different between groups (CSF group 1.0 +/- 0.2; DEX group 0.6 +/- 0.2 mL [center dot] min-1 [center dot] mm Hg [center dot] 100 g). DEX increased CVR from 1.2 +/- 0.1 to 2.0 +/- 0.2mL [centered dot] min-1 [center dot] mm Hg [center dot] 100 g. In both groups, the second hypoxia challenge decreased CVR, but the CVR was higher in the DEX group during hypoxia. The absolute change in CVR during hypoxia was not affected by DEX administration (first hypoxia challenge for DEX group, 0.6 +/- 0.2; second hypoxia challenge for DEX group, 0.9 +/- 0.2 mL [center dot] min-1 [center dot] mm Hg [centered dot] 100 g-1 decrease).
Figure 2: Cerebrovascular resistance (CVR) in mock cerebrospinal fluid group (CSF, n = 5) and dexmedetomidine group (DEX, n = 6) groups during both episodes of hypoxia. *P < 0.05 versus normoxia value; dagger P < 0.05 versus Recovery 1 value; double dagger P < 0.05 versus CSF group at same experimental condition.
(Figure 3 ) shows cerebral oxygen transport in CSF and DEX groups. In the CSF group, baseline cerebral oxygen transport was 14.0 +/- 2.1 mL [center dot] min-1 [center dot] 100 g-1 and was unchanged as a result of hypoxia or the second administration of mock CSF. In the DEX group, the initial hypoxia slightly decreased cerebral oxygen transport from 14.4 +/- 1.0 to 10.3 +/- 2.0 mL [centered dot] min-1 [center dot] 100 g-1 (P < 0.05). However, DEX administration resulted in a substantial downward and parallel shift in the response to hypoxia.
Figure 3: Cerebral oxygen transport in mock cerebrospinal fluid group (CSF, n = 5) and dexmedetomidine group (DEX, n = 5) groups. *P < 0.05 versus value prior to DEX administration.
(Figure 4 ) shows brain distribution of3 H-clonidine after its infusion into the left ventricular catheter for 30 min. Activity in the left caudate nucleus was used as a standard. As compared to the left caudate nucleus, activity was the highest in the choroid plexus. The tissue concentration in the cerebral cortex was approximately 1% of that found in the left caudate nucleus. In two animals, the concentration in the infused solution and in CSF collected from the cisterna magna catheter was assessed. The ratio of CSF to infused solution was 19% and 33%, respectively.
Figure 4: Regional distribution of3 H-clonidine administered by ventricular-cisternal perfusion. Two milliliters of solution containing 0.32 micro g (20 micro Ci) was administered over 5 min into the left lateral ventrical followed by an infusion of this solution at 0.2 mL/min for 30 min. Values are expressed in relationship to activity per milligram of protein in the left caudate nucleus for each individual animal.
(Figure 5 ) shows CBF, expressed as a percentage of the normoxic flow, in regions found to have high penetration of alpha2 agonist (e.g., caudate nucleus and brainstem) and a region found to have low penetration of alpha2 agonist (e.g., cerebral cortex) as depicted in Figure 4 . DEX did not influence the percentage increase in flow during hypoxia, regardless of the degree of alpha2-agonist penetration in any region. The hyperemic response to hypoxia was most robust in brainstem as compared to the other regions.
Figure 5: Regional flow during hypoxia expressed as percentage of prior normoxic value. *P < 0.05 versus H1 value; dagger P versus prior hypoxia in same group and region. H1 = first hypoxic exposure; H2 = second hypoxic exposure; CSF = cerebrospinal fluid; DEX = dexmedetomidine.
Discussion
During normoxia, ventricular-cisternal perfusion of DEX decreased blood flow throughout the brain without evidence of reduced CMRO2 . The lack of blood pressure response to ventricular-cisternal perfusion of DEX suggests that the DEX does not have its effect as a result of systemic absorption of this alpha2 agonist. Concentration effect and alpha2-receptor specificity were demonstrated in a separate group of animals. To minimize time for diffusion through the CSF space and bulk absorption via the brain capillaries we used, a short exposure time to ventriculocisternal perfusion. To determine the likely cerebral distribution of DEX during ventricular-cisternal perfusion we measured the distribution of another alpha2 agonist3 H-clonidine using an identical mode of administration. We found the highest3 H-activity in regions ipsilateral to the ventricular injection and along the pathway to the cisternal outflow. Cerebral cortex demonstrated only approximately 1% of the3 H-activity in the ipsilateral caudate nucleus, but both of these regions had similar reduction in blood flow during ventricular-cisternal perfusion of DEX. Therefore, it is unlikely that DEX mediates a reduction in CBF by activating local vascular alpha2 receptors. Ventricular-cisternal perfusion of DEX was associated with a higher CVR during hypoxia than in control animals. However, the percent increase in CBF and the absolute change in CVR during hypoxia were not decreased in the DEX as compared to the control group. Treatment with DEX reduced cerebral oxygen delivery, both during normoxia and hypoxia. During hypoxia, DEX-induced reduction in cerebral oxygen delivery was severe enough to decrease CMRO2 .
Systemic administration of the alpha2 agonist DEX results in sedation [1] [12] 2 [2,13] 2 agonist [14] 2 but it is widely used in experimental animal models as a baseline anesthetic (usually in combination with alpha-chloralose). Binding sites for alpha2 agonists are most highly concentrated in areas of brain involved in control of cardiovascular function [7] 2 adrenoceptors [15] [16] 2 adrenoceptors appears to involve inhibitory G proteins [17] [18] 2 agonist. However, in our study, the blood flow effect was similar in regions (e.g., ipsilateral caudate) found to have a very high concentration of alpha2 agonist (3 ) H-clonidine in our study) compared to regions having very low concentration of alpha2 agonist (e.g., that contralateral cortex) during lateral ventricular infusion. In addition, others have found that the effect on blood flow is not mediated by a change in pial vessel diameter [19]
The mechanism for the widespread reduction in normoxic CBF during ventricular-cisternal perfusion of DEX appears to be related to its effect on either a subcortical or brainstem structure, as distribution of3 H-clonidine was greatest in these areas. Since DEX is more lipid soluble than clonidine [20] 3 H-clonidine probably overestimates the spread through the CSF space during the time assessed. In addition, we believe that it is unlikely that the ventricular-cisternal perfusion of DEX resulted in significant absorption of DEX into the systemic circulation because DEX administration in our study lacked the increase in MABP which is normally present during its systemic administration [3,4]
Pharmacodynamic variables relevant to DEX loss from CSF [20] [20] [21] [20] [21]
In unanesthetized rabbits, stimulation of peripheral sympathetic nerves limits hypoxia-induced cerebral hyperemia in cerebrum by 13% but has no effects during normoxia [22] [22]
There are at least two candidate areas in brainstem that may mediate the effect of DEX on the CBF response to hypoxia: locus coeruleus and rostral ventrolateral medulla. Others have found that intraventricular injection of alpha2 agonist alters neuronal response of the locus coeruleus [23] [24] [25] [25] [3] 2 [26]
An important consideration is whether the lower flows during hypoxia in the present study is somehow metabolically mediated. For example, large-dose barbiturate anesthesia, which reduces CMRO2 , is associated with attenuated hypoxia-induced cerebral hyperemia [27] 2 under baseline conditions. On the contrary, CMRO2 was only reduced by DEX during hypoxia. Therefore, we believe that the reduction in CMRO2 during hypoxia from DEX administration is due to a drug-induced reduction of oxygen delivery (reduced oxygen content of blood during a period of nonmetabolically induced CBF reduction) to the point of tissue hypoxia.
Reduction of CBF during hypoxia, to the point of causing a reduction in CMRO2 during hypoxia after treatment with DEX, may have important clinical implications. DEX is presently being evaluated as an anesthetic adjunct in clinical anesthesiology [1,18]
Although DEX in combination with hypoxic hypoxia results in tissue hypoxia in the current study, the literature suggests that DEX may be neuroprotective in the setting of ischemia. The literature concerning the effect of alpha2-receptor agonists and antagonists in the setting of ischemia is conflicting. For example, DEX also improves neurologic outcome from transient incomplete and focal ischemia [28-30] [31] 2-receptor antagonist, ameliorated brain injury in rats exposed to transient forebrain ischemia [32,33] [32,34] 2-receptor agonist and an antagonist have therapeutic effects in the setting of cerebral ischemia has not been resolved.
In conclusion, we have found that central administration of the alpha (2 ) agonist DEX increases cerebrovascular tone during normoxia and severe hypoxia, without evidence of systemic absorption of the drug. DEX-induced vasoconstriction is sufficiently robust to limit cerebral oxygen delivery during hypoxia and result in reduced CMRO2 . Although the amount of cerebral vascular tone during hypoxia is greater in dogs pretreated with DEX, the absolute change in tone resulting from hypoxia is unaltered by DEX. Therefore, we do not believe that DEX alters the underlying mechanism for the cerebrovascular response to hypoxia.
We wish to thank Jane Paradise for assistance in preparation of this manuscript. The assistance of technicians in the Anesthesiology/Critical Care Medicine laboratory is also gratefully acknowledged.
Considerable sadness accompanies publication of the foregoing original article. Dr. Robert McPherson, a frequent contributor to Anesthesia & Analgesia and a valued reviewer of manuscripts submitted to the Neurosurgical Anesthesia Section, died suddenly on July 4, 1996. Bob, a nationally renowned neuroanesthesiologist recently selected for promotion to Professor of Anesthesiology at Johns Hopkins Medical Institutions, had been a member of the faculty at Johns Hopkins since completing his military obligation at Walter Reed Army Medical Center in 1980. His investigative work on neurophysiologic monitoring and on cerebrovascular effects of isoflurane, funded by the National Institutes of Health, ensures that his contributions to our specialty will be recognized for many years to come. His colleagues at Johns Hopkins and around the country will greatly miss his keen insight and warm friendship.
Donald S. Prough, MD
Section Editor
Neurosurgical Anesthesia
REFERENCES
1. Belleville JP, Ward DS, Bloor BC, Maze M. Effects of intravenous dexmedetomidine in humans. 1. Sedation, ventilation, and metabolic rater. Anesthesiology 1992;77:1125-33.
2. Zornow MH, Fleishcer JE, Scheller MS, et al. Dexmedetomidine, an alpha 2-adrenergic agonist, decreases cerebral blood flow in the isoflurane-anesthetized dog. Anesth Analg 1990;70:624-30.
3. Fale A, Kirsch JR, McPherson RW. alpha
2-Adrenergic agonist effects on normocapnic and hypercapnic cerebral blood flow in the dog are anesthetic dependent. Anesth Analg 1994;79:892-8.
4. McPherson RW, Koehler RC, Traystman RJ. Hypoxia, alpha-2-adrenergic, and nitric oxide-dependent interactions on canine cerebral blood flow. Am J Physiol 1994;266:H476-82.
5. Peroutka SJ, Moskowitz MA, Reinhard JF Jr, Snyder SH. Neurotransmitter receptor binding in bovine cerebral microvessels. Science 1980;208:610-2.
6. Skarby T. Pharmacologic properties of prejunctional alpha-adrenoceptors in isolated feline middle cerebral arteries; comparison with the postjunctional alpha-adrenoceptors. Acta Physiol Scand 1984;122:165-74.
7. Unnerstall JR, Kopajtic TA, Kuhar MJ. Distribution of alpha 2 agonist binding sites in the rat and human central nervous system: analysis of some functional, anatomic correlates of the pharmacologic effects of clonidine and related adrenergic agents. Brain Res 1984;319:69-101.
8. Bates D, Weinshilboum RM, Campbell RJ, Sundt TM Jr. The effect of lesions in the locus coeruleus on the physiological responses of the cerebral blood flow vessels in cats. Brain Res 1977;136:431-43.
9. Correa-Sales C, Rabin BC, Maze M. A hypnotic response to dexmedetomidine, an alpha-2 agonist, is mediated in the locus coeruleus in rats. Anesthesiology 1992;76:948-52.
10. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248-54.
11. Levasseur JE, Wei EP, Raper AJ, et al. Detailed description of a cranial window technique for acute and chronic experiments. Stroke 1975;6:308-317.
12. Zornow MH, Scheller MS, Sheehan PB, et al. Intracranial pressure effects of dexmedetomidine in rabbits. Anesth Analg 1992;75:232-7.
13. Karlsson BR, Forsman M, Roald OK, et al. Effect of dexmedetomidine, a sensitive and potent alpha 2-agonist, on cerebral blood flow and oxygen consumption during halothane anesthesia in dogs. Anesth Analg 1990;71:125-9.
14. Aisaka K, Kihara T, Koike M, et al. Effect of yohimbine on urethane-induced hyperglycemia in rats. Jpn J Pharmacol 1989;49:523-7.
15. Tsukahara T, Taniguchi T, Usui H, et al. Sympathetic denervation and alpha adrenoceptors in dog cerebral arteries. Naunyn Schmiedebergs Arch Pharmacol 1986;334:436-43.
16. Coughlan MG, Lee JG, Bosnjak ZJ, et al. Direct coronary and cerebral vascular responses to dexmedetomidine. Significance of endogenous nitric oxide synthesis. Anesthesiology 1992;77:998-1006.
17. Doze VA, Chen B, Tinkleberg JA, et al. Pertussis toxin and 4-aminopyridine differentially affect the hypnotic-anesthetic action of dexmedetomidine and phentobarbital. Anesthesiology 1990;73:304-7.
18. Maze M, Tranquilli W. Alpha-2 adrenoceptor agonists: defining the role in clinical anesthesia. Anesthesiology 1991;74:581-605.
19. Kanawati IS, Yaksh TL, Anderson RE, Marsh RW. Effects of clonidine on cerebral blood flow and the response to arterial CO
2 . J Cereb Blood Flow Metab 1986;6:358-65.
20. Eisenach JC, Shafer SL, Bucklin BA, et al. Pharmacokinetics and pharmacodynamics of intraspinal dexmedetomidine in sheep. Anesthesiology 1994;80:1349-59.
21. Levin VA, Landahl HD. Pharmacokinetic approaches to drug distribution in the cerebrospinal fluid based on ventricular administration in beagle dogs. J Pharmacokinet Biopharm 1985;13:387-403.
22. Busija DW. Sympathetic nerves reduce cerebral blood flow during hypoxia in awake rabbits. Am J Physiol 1984;247:H446-51.
23. Adams LM, Foote SL. Effects of locally infused pharmacological agents on spontaneous and sensory-evoked activity of locus coeruleus neurons. Brain Res Bull 1988;21:395-400.
24. Hong M, Milne B, Loomis C, Jhamandas K. Stereoselective effects of central alpha
2-adrenergic agonist meetomidine on in vivo catechol activity in the rat rostral ventrolateral medulla (RVLM). Brain Res 1992;592:163-9.
25. Underwood MD, Iadecola C, Reis DJ. Lesions of the rostral ventrolateral medulla reduce the cerebrovascular response to hypoxia. Brain Res 1994;635:217-23.
26. Buchweitz E, Edelman NH, Weiss HR. Effects of locus coeruleus stimulation on regional cerebral oxygen consumtion in the cat. Brain Res 1985;325:107-14.
27. Donegan JH, Traystman RJ, Koehler RC, et al. Cerebrovascular hypoxic and autoregulatory responses during reduced brain metabolism. Am J Physiol 1985;249:H421-9.
28. Hoffman WE, Kochs E, Werner C, et al. Dexmedetomidine improves neurologic outcome from incomplete ischemia in rat. Reversal by the alpha
2-adrenergic antagonist atipamezole. Anesthesiology 1991;75:328-32.
29. Maier C, Steinberg GK, Sun GH, et al. Neuroprotection by the alpha (
2-adrenoreceptor ) agonist dexmedetomidine in a focal model of cerebral ischemia. Anesthesiology 1993;79:306-12.
30. Hoffman WE, Baughman VL, Albrecht RF. Interaction of catecholamines and nitrous oxide ventilation during incomplete brain ischemia in rats. Anesth Analg 1993;77:908-12.
31. Matsumoto M, Zornow MH, Rabin BC, Maze M. The alpha
2 adrenergic agonist, dexmedetomidine, selectively attenuates ischemia-induced increases in striatal norepinephrine concentrations. Brain Res 1993;627:325-9.
32. Gustafson I, Westerberg E, Wieloch T. Protection against ischemia-induced neuronal damage by the alpha 2-adrenoceptor antagonist idazoxan: influence of time of administration and possible mechanisms of action. J Cereb Blood Flow Metab 1990;10:885-94.
33. Gustafson I, Miyauchi Y, Wielock TW. Postischemic administration of idazoxan, an alpha-2 adrenergic receptor antagonist, decreases neuronal damage in the rat brain. J Cereb Blood Flow Metab 1989;9:171-4.
34. Gustafson I, Westerberg EJ, Wieloch T. Extracellular brain cor-tical levels of noradrenaline in ischemia: effects of desipramine and postischemic administration of idazoxan. Exp Brain Res 1991;86:555-61.
