Over the last two decades, there has been a considerable accumulation of experimental and clinical data relating to the pharmacology of α2-adrenoceptor agonists (α2 agonists), and their clinical use in anesthesia has increased. Among their many clinically significant properties, including sedative, anxiolytic, analgesic, and hemodynamic stabilizing effects (1,2), the antinociceptive actions of spinally or systemically administered α2 agonists are significant and varied. Indeed, the α2 agonists dexmedetomidine (DXM), clonidine (CL), and tizanidine (TZ), which are clinically available, seem to have a diversity of actions; thus, one drug may have a potent hypnotic action when given systemically, while another has a potent analgesic action when given spinally. Several clinical observations have suggested that oral CL has effects similar to those seen during its spinal administration (a potentiation of the neuronal block induced by spinal anesthesia) (3,4). Systemic (oral) administration of TZ produces sedative and hypnotic effects (5), and prolongs tetracaine spinal anesthesia (6).
However, whether there are differences in efficacy between systemically and epidurally administered α2 agonists is not known. Further, although α2 agonists are thought to produce antinociceptive effects via activation of spinal and supraspinal α2-adrenoceptors, no studies have examined the correlation between their analgesic potency and their relative binding affinity to α2-adrenoceptors in brain and spinal cord using the same experimental model. In addition, no study has yet compared the binding affinity of a given α2-adrenoceptor between the spinal cord and brain. Because it is possible that the different effects seen with different α2 agonists could be caused by differences in binding affinity between spinal and supraspinal neuronal structures, we undertook the present study in rats primarily to examine the analgesic action of three α2 agonists when they were given by systemic or epidural injection and to try to relate their analgesic effect to their binding affinity to α2-adrenoceptors in the spinal cord and brain.
All experimental procedures in the present study were approved by our animal research committee.
Adult male Sprague-Dawley rats, weighing 350–450 g, were anesthetized with pentobarbital (50 mg/kg, intraperitoneally). Each animal was prepared for the implantation of a permanent, indwelling epidural catheter. Briefly, after a midline incision and a laminectomy at T12, a polyethelene-10, 15-cm long catheter with an approximate volume of 15 μL (Natsume Co Ltd, Tokyo, Japan) was inserted into the epidural space under direct visual control, then advanced 2 cm toward the tail so that the tip was at approximately the L1-2 spinal level. The catheter was cemented to the bone, and its end was plugged with a stainless steel insert. The free end of the catheter was exteriorized through a second skin incision in the midline over the upper cervical area by tunneling it under the skin. The cervical portion of the catheter was fixed to the fascia of the neck muscles with nylon suture. After surgery, all animals were housed individually in a temperature- and light-controlled environment with free access to food and water. The patency of the catheter was checked periodically. Each animal received one or three of the drugs tested, with an interval of 2–4 days between administrations of the drugs tested. Any animals exhibiting signs of a neurological or motor deficit were eliminated from the study. Fifty-five rats were instrumented during the course of this investigation. Eight rats were excluded for neurological deficit. After the completion of drug testing, bromphenol blue was injected to aid postmortem verification of catheter positioning in each animal.
Animals were allowed 7 days to recover from surgery, and 2% lidocaine (150 μL) was injected through the epidural catheter to confirm correct positioning. The rats were habituated daily to the laboratory environment and to the analgesia-testing apparatus. An evaluation of the antinociceptive effect of each of the three α2 agonists, administered epidurally or IM, was performed by using the tail-flick test. The doses used were DXM (0.5, 1, 5, and 10 μg epidurally and 5 and 50 μg IM), CL (10, 25, 50, 100, 125, 250, and 500 μg epidurally and 10, 100, and 1000 μg IM), and TZ (5, 10, 50, 100, 250, and 500 μg epidurally and 100 and 1000 μg IM). A combination of yohimbine, an α2-adrenoceptor antagonist, and a given α2 agonist was administered epidurally to challenge the effect of the agonist. The numbers of rats used to test each drug or combination was six for each dose. A custom-made tail-flick apparatus was used to assess the ther- mal nociceptive response. It consisted of a variable-intensity 50-W quartz projector bulb, focused approximately 2 cm from the upper surface of the tail, and a photodetector-automatic timer. The latency to respond after the onset of the radiant heat stimulus was monitored, the first visible twitch or flick being taken as the response. Light intensity was adjusted to yield a 3- to 4-s mean baseline latency, and the automatic cutoff was set at 10 s to avoid damage to the tail. The compounds to be tested were dissolved in sterile saline and injected in a volume of 50 μL for epidural or 200 μL for IM administration. After each injection, the catheter was flushed with sterile saline (20 μL), with the flush completed within 5 min. Postdrug latency measurements were performed between 10 and 30 min after the injection. The difference between the baseline and postdrug latencies constituted the analgesic response at a given time point. These values were normalized further to give values of percentage maximal possible effect (%MPE) using the following formula : where TF is the postdrug tail-flick latency (in s), BL is the baseline latency (in s), and 10 s is the cutoff time. Thus, a %MPE value of 100 meant that the postdrug tail-flick latency had increased to 10 s. The ED50 value was taken as the half-maximal effective dose calculated from the dose-response curve for each animal.
For the first 30 min after the drug administration, systolic arterial blood pressure (AP) and heart rate (HR) were determined in conscious, warmed, restrained rats by a tail-cuff method, by using a programmed sphygmomanometer (MK-1000, Muromachi Kikai Co., Tokyo, Japan). The values given are the percentage changes (%Change in AP or HR) from baseline values, calculated by using the following formula :MATH The rats used for AP and HR measurements did not take part in the testing of tail-flick latencies.
To obtain whole brains and spinal cords, adult male Sprague-Dawley rats (200–250 g) were killed by decapitation without anesthesia (5 rats and 20 rats, respectively). The brain or spinal cord was rapidly removed, processed on ice, and homogenized in a 20 times volume of 0.32 M sucrose using a Biotron (Kimura Co, Tokyo, Japan). The homogenate was centrifuged (1000 g) for 10 min at 4°C. The supernatant was recentrifuged (20,000 g) for 20 min at 4°C; then, the pellet was washed twice with 50 mM Tris-HCl buffer (pH 7.4) by centrifugation (20,000 g) for 20 min at 4°C. The final pellet (crude synaptosomal fraction, P-2) was resuspended in Tris buffer and stored at −80°C until use.
A modification of α2 receptor binding assays of Loftus et al. (7) was used. The binding experiments were performed at 25°C for 40 min in duplicate by incubating the crude synaptosomal fraction with 2 nM of 3H-UK14304, a selective α2 agonist, in the presence or absence of various concentrations of tested agonists. Then, the reaction mixture was filtered rapidly through G-10 glass filters (Inotech, Dottikon, Switzerland) under reduced pressure (400 mbar) using a Cell Harvester System (Inotech) to trap the labeled membrane fragments, followed by a 3 × 3-mL washing with ice-cold Tris buffer. The residues on the filter were solubilized in a 4-mL liquid scintillation cocktail, and the radioactivity was measured by means of a liquid scintillation counter (Beckmann-LS9000, Fullerton, CA). The specific binding of 3H-UK14304 to α2-adrenoceptors was calculated from the difference between the counts obtained in the presence and absence of 2 mM clonidine. Then, the apparent dissociation constant (Kd) and the maximal binding capacity (Bmax) values were calculated from the saturation curves by using a concentration range of 0.1 to 5 nM of 3H-UK14304. Competition experiments of α2 agonists were performed with fixed concentration of 3H-UK14304. Ki values were calculated from competition data with Ki = IC50/(1 + L/Kd), in which IC50 was the concentration of unlabelled ligand that caused 50% inhibition of binding of 3H-UK14304, L was the concentration of radioligand, and Kd was the equilibrium dissociation of 3H-UK14304. Protein content was determined with the BCA® assay (Pierce, Rockford, IL) by using bovine serum albumin as protein standard. IC50 values were calculated from log regression-fitting of the dose-response data plotted. Data of saturation and displacement experiments were analyzed by using the nonlinear curve-fitting program LIGAND (8).
The mean values for ED50 or IC50 and Ki were evaluated for significant differences among the three α2 agonist groups using a one-way analysis of variance followed by post hoc testing by using the Scheffé method. The mean ED50 values for each α2 agonist were evaluated for significant differences between epidural and IM injection by using an unpaired t-test. The mean IC50 and Ki values for each α2 agonist were evaluated for significant differences between spinal cord and brain membranes by using an unpaired t-test. %Changes in HR and AP in each α2 agonist group were evaluated for significance by using analysis of variance followed by post hoc testing using the Scheffé method. P < 0.05 was considered to be significant.
Dexmedetomidine hydrochloride was provided by Abbott Laboratories, Abbott Park, IL; clonidine hydrochloride and yohimbine hydrochloride (MW 390.9) were obtained from Sigma Chemical Co, St. Louis, MO; tizanidine hydrochloride was provided by Sandoz Pharmaceuticals Ltd (Novartis Pharma K.K. at present), Tokyo, Japan; morphine hydrochloride (MW 321.8) was provided by Takeda Chemical Industries Ltd, Osaka, Japan; and 3H-UK14304 (76.5 Ci/mmol) was obtained from New England Nuclear, Boston, MA.
In preliminary experiments, epidurally or IM injected morphine induced antinociceptive effects in a dose-dependent manner (Figure 1A), the ED50 values being 13.4 ± 2.1 and 285 ± 23.3 μg/rat, respectively (mean ± SEM, n = 6).
DXM, CL, and TZ administered into the epidural space induced a dose-dependent increase in tail-flick latency (Figure 1B). The increased latencies with each agonist were constant for 30 min after its administration. Epidural saline induced slight hyperalgesia (%MPE = −11.6 ± 8.5, n = 6) in our experimental model. (Figure 1, C and D). Epidurally administered yohimbine did not itself significantly affect tail-flick latency at a dose of 250 μg (%MPE = −7.7 ± 2.8, n = 6), but its pretreatment significantly reduced the measured increase in %MPE induced by epidural α2 agonists (Figure 2). Likewise, on systemic administration, DXM, CL, and TZ each induced a dose-dependent increase in tail-flick latency (Figure 1B). For both epidural and systemic administration, the rank order of the antinociceptive potencies (compared by using the mean ED50 values) was DXM > CL > TZ (Table 1). DXM, CL, and TZ each exhibited a significantly greater antinociceptive potency on epidural than on systemic administration as assessed using the ED50 values.
In 50 mM Tris-HCl buffer at 25°C, specific binding of 3H-UK14304 (2 nM) reached equilibrium within 30 min and was maintained for at least 60 min. Nonspecific binding was <30% of total 3H-UK14304 binding. Saturation experiment data revealed, in both brain and spinal cord membranes, a single binding site with a Kd value of 1.63 ± 0.14 nM and 1.13 ± 0.17 nM, and Bmax value of 97.3 ± 9.3 fmol/mg protein and 79.1 ± 6.4 fmol/mg protein, respectively.
DXM, CL, and TZ each induced a concentration-dependent increase in the %inhibition of 3H-UK14304 stereospecific binding (Figure 3). The rank order of their binding affinities for α2-adrenoceptors (compared by using the mean IC50 and Ki values) was DXM > CL > TZ (Table 1) for both brain and spinal cord membranes. The binding affinity of CL in the spinal cord was not significantly different from that in the brain. In contrast, the binding of DXM and TZ was significantly greater in the spinal cord than in the brain (Table 1).
At doses large enough to produce near-maximal antinociceptive effects, epidural DXM (10 μg), CL (100 μg), and TZ (100 μg) significantly decreased HR (Figure 4). Conversely, at larger doses of epidural DXM and CL, there were trends toward relative increases in AP. Epidural TZ tended to decrease AP at either dose, but the effects didn’t significantly differ from those in the saline group. On IM administration, DXM (10 and 100 μg), CL (100 and 1000 μg), and TZ (100 and 1000 μg), each significantly decreased HR at both doses tested. On IM administration at the larger of those two doses (which produced maximal antinociceptive effects), DXM, CL, and TZ each significantly increased AP.
Our results indicate that, whether given systemically or epidurally, the α2 agonists DXM, CL, and TZ each produce a dose-related antinociceptive effect in chronically catheterized rats. The effects were approximately 5 times greater on epidural than on systemic administration, and the former was completely blocked by epidural yohimbine. In our model, epidural saline and yohimbine produce slight hyperalgesia and TZ’s potency was the smallest in the three α2 agonists. The rank order of antinociceptive potencies (based on changes in %MPE) for the three α2 agonists, regardless of whether they were given epidurally or systemically, was DXM > CL > TZ. This rank order for their antinociceptive effect was the same as that for their binding affinity to α2-adrenoceptors in the spinal cord, but not the same as that for the changes in AP and HR evoked by doses producing near-maximal antinociception. The binding affinity of DXM and TZ for α2-adrenoceptors was significantly greater in spinal cord membranes than in the brain, but the binding affinity of CL did not differ between spinal cord and brain. Potentially, these findings could help establish a background for the assessment of the spinal antinociception induced by α2 agonists directly injected in close proximity to neuronal structures.
Several studies have indicated that on intrathecal administration in rats, the α2 agonists DXM (9,10), CL (9,10,11), and TZ (11,12) [and others such as ST-91 (9,10)] produce an antinociceptive effect that can be inhibited or attenuated by α2-adrenoceptor antagonists such as yohimbine. In addition, many clinical studies have indicated that epidural α2 agonists such as CL, given alone or in combination with local anesthetics (13) or opioids (14), may have an analgesic effect or potentiate another drug’s analgesic effect (1). The epidural route is probably more applicable to clinical practice than the intrathecal one, and the present study is the first to provide comparative data on the antinociceptive effect, probably via activation of spinal α2-adrenoceptors, produced epidurally by the three α2 agonists clinically available at present. The spinal antinociception induced by α2 agonists is mediated by an inhibition of synaptic transmission within the dorsal horn of the rat’s spinal cord (15), principally (a) via a direct suppression of the activity of dorsal horn neurons (16), (b) via an activation of the descending noradrenergic inhibitory system (17), and (c) via an activation of spinal cholinergic neurons (18).
α2-Adrenoceptors are found at many sites in the spinal cord, including the dorsal horn and ventral horn (ventral > dorsal) (19) and in the cerebral cortex in humans (20), and three subtypes have been cloned, α2A, α2B, and α2C (21). In rats, the α2A-adrenoceptor subtype is involved in α2 agonist-induced antinociceptive (10,22) and hypnotic action (23) of α2 agonists. Using ligand binding, Lawhead et al. (24) and Smith et al. (19) found that α2-adrenoceptors are in their largest concentration in the sacral > lumbar = thoracic regions of the human spinal cord, and that, in any given region of the spinal cord, the α2A-adrenoceptor subtype accounts for at least 80%–90% of the α2-adrenoceptor population.
Until now, there has been neither a comparative study of the antinociceptive effects produced by systemic as opposed to spinal administration of α2 agonists nor a comparative study of their binding affinity for α2 adrenergic receptors in brain as opposed to spinal cord. Their rank order of potencies against nociceptive responses and the behaviors probably reflect differences in their agonist-receptor binding or selectivity (25) or in their heterogenicity (26) and may perhaps reflect differences in binding affinity between spinal cord and brain. Using systemic and intrathecal administration of α2 agonists, Yaksh et al. (27,28) concluded that α2 agonists modulate nociception through an action within the spinal cord, whereas supraspinal sites mediate their sedative effect. In the supraspinal region, their action on the locus ceruleus is considered to result in an increased activation of α2-adrenoceptors in the spinal cord (29). Although we used whole brain and whole spinal cord for our ligand binding study, leading to certain inherent limitations, the data may provide a framework for a more detailed evaluation of the clinical differences among α2 agonists.
Although we found that the relative analgesic actions of the three α2 agonists seemed to be consistent with their receptor binding affinity in the spinal cord, the changes in AP and HR associated with their administration, although dose-related for each agonist, were not consistent with their receptor binding. Previous studies have indicated that a variety of hemodynamic effects occur after the epidural (28,30,31) or intrathecal administration (11,32) of α2 agonists. Unfortunately, there are no comparative data as to the vascular adrenoceptor binding of the three agonists. On pial vessels, the effects of DXM are similar to those of CL when both are given into the intrathecal space, when using the intracranial window technique (33,34). α2 agonists given systemically can affect vascular tone (and thus AP) through central and peripheral effects and may also have a cardiac effect (for example, via a suppression of sympathetic nervous system activity). It is thought that the decreases in AP and HR evoked by systemic administration of α2 agonists could be predominantly caused by a central effect via their binding to imidazoline receptors (35,36). Thus, in addition to their peripheral action, it is possible that epidural α2 agonists could also cause cardiovascular changes via central effects. The variety of cardiovascular responses to α2 agonists seen by different authors could well be the result of variation among the different agonists in terms of their preferential site(s) of action (both inside and outside the central nervous system).
The epidural route of administration for α2 agonists was chosen to provide a comparison with their systemic effects in this study because of its common use in clinical practice. Because of its pharmacokinetic and pharmacodynamic behavior, CL given epidurally passes into the spinal cord cerebrospinal fluid (31,37). No data, however, are available as to whether epidurally administered DXM and TZ can penetrate the spinal meninges as easily as CL. Bernards and Hill (38) found that there is a biphasic relationship between the log octanol:buffer distribution coefficient and the meningeal permeability coefficient. The optimal octanol:buffer distribution coefficient for maximal meningeal permeability is between 129 and 560. On the basis of their properties, DXM and CL would be close to or within the optimal range for meningeal penetration, but TZ would be outside the optimal range (Table 2). Thus, TZ might exhibit poor penetration of the spinal meninges. However, as was the case with DXM and CL, the antinociceptive effect of TZ was significantly greater on epidural than on systemic administration in the present study. The rank order for the antinociceptive effects of spinally administered α2 agonists seems to be DXM > CL > TZ [DXM > CL (9), and CL > TZ (12)]; this order agrees with our findings for epidural administration, for which the dose needed is 8–10 fold higher than for spinal administration. Thus, TZ is the least potent by spinal as well as by epidural administration, and so we cannot say at present to what extent the penetration of the spinal meninges by a given α2 agonist affects its spinal antinociceptive effect.
In summary, the present results suggest that epidural administration of α2 agonists is approximately 5 times more effective than systemic administration at producing an antinociceptive effect. Our present results are consistent with the idea that differences in antinociceptive action among the clinically available α2 agonists DXM, CL, and TZ could be caused principally by differences in their receptor binding affinity within the spinal cord. However, we cannot eliminate the possibility that the differences between spinal administration and the systemic or epidural administration of centrally acting drugs could be a result of, to some extent, differences in lipid solubility and meningeal penetration.
1. Eisenach JC, De Kock M, Klimscha W. Alpha 2-adrenergic agonists for regional anesthesia: a clinical review of clonidine (1984–1995). Anesthesiology 1996; 85:655–74.
2. Maze M, Tranquilli W. Alpha-2 adrenoceptor agonists: defining the role in clinical anesthesia. Anesthesiology 1991; 74:581–605.
3. Ota K, Namiki A, Iwasaki H, Takahashi I. Dose-related prolongation of tetracaine spinal anesthesia by oral clonidine in humans. Anesth Analg 1994; 79:1121–5.
4. Liu S, Chiu AA, Neal JM, et al. Oral clonidine prolongs lidocaine spinal anesthesia in human volunteers. Anesthesiology 1995; 82:1353–9.
5. Takenaka M, Iida H, Kasamatsu M, et al. Tizanidine for preanesthetic medication. Masui 1996; 45:971–5.
6. Omote K, Satoh O, Sonoda H, et al. Effects of oral alpha 2 adrenergic agonists, clonidine and tizanidine, on tetracaine spinal anesthesia. Masui 1995; 44:816–23.
7. Loftus DJ, Stolk JM, U’Prichard DC. Binding of the imidazoline UK-14, 304, a putative full alpha 2-adrenoceptor agonist, to rat cerebral cortex membranes. Life Sci 1984; 35:61–9.
8. Munson PJ, Rodbard D. Ligand: a versatile computerized approach for characterization of ligand-binding systems. Anal Biochem 1980; 107:220–39.
9. Takano Y, Yaksh TL. Relative efficacy of spinal alpha-2 agonists, dexmedetomidine, clonidine and ST-91, determined in vivo by using N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline, an irreversible antagonist. J Pharmacol Exp Ther 1991; 258:438–46.
10. Takano Y, Takano M, Yaksh TL. The effect of intrathecally administered imiloxan and WB4101: possible role of alpha 2-adrenoceptor subtypes in the spinal cord. Eur J Pharmacol 1992; 219:465–8.
11. Kawamata T, Omote K, Kawamata M, et al. Antinociceptive interaction of intrathecal alpha2-adrenergic agonists, tizanidine and clonidine, with lidocaine in rats. Anesthesiology 1997; 87:436–48.
12. McCarthy RJ, Kroin JS, Lubenow TR, et al. Effect of intrathecal tizanidine on antinociception and blood pressure in the rat. Pain 1990; 40:333–8.
13. Glynn C, O’Sullivan K. A double-blind randomised comparison of the effects of epidural clonidine, lignocaine and the combination of clonidine and lignocaine in patients with chronic pain. Pain 1996; 64:337–43.
14. Rockemann MG, Seeling W, Brinkmann A, et al. Analgesic and hemodynamic effects of epidural clonidine, clonidine/morphine, and morphine after pancreatic surgery: a double-blind study. Anesth Analg 1995; 80:869–74.
15. Sullivan AF, Kalso EA, McQuay HJ, Dickenson AH. The antinociceptive actions of dexmedetomidine on dorsal horn neuronal responses in the anaesthetized rat. Eur J Pharmacol 1992; 215:127–33.
16. Murata K, Nakagawa I, Kumeta Y, et al. Intrathecal clonidine suppresses noxiously evoked activity of spinal wide dynamic range neurons in cats. Anesth Analg 1989; 69:185–91.
17. Sawynok J, Reid A. Role of ascending and descending noradrenergic pathways in the antinociceptive effect of baclofen and clonidine. Brain Res 1986; 386:341–50.
18. Detweiler DJ, Eisenach JC, Tong C, Jackson C. A cholinergic interaction in alpha 2 adrenoceptor-mediated antinociception in sheep. J Pharmacol Exp Ther 1993; 265:536–42.
19. Smith MS, Schambra UB, Wilson KH, et al. Alpha2-Adrenergic receptors in human spinal cord: specific localized expression of mRNA encoding alpha2-adrenergic receptor subtypes at four distinct levels. Brain Res Mol Brain Res 1995; 34:109–17.
20. Renouard A, Widdowson PS, Millan MJ. Multiple alpha 2 adrenergic receptor subtypes. I. Comparison of [3H]RX821002-labeled rat R alpha-2A adrenergic receptors in cerebral cortex to human H alpha2A adrenergic receptor and other populations of alpha-2 adrenergic subtypes. J Pharmacol Exp Ther 1994; 270:946–57.
21. Bylund DB, Blaxall HS, Iversen LJ, et al. Pharmacological characteristics of alpha 2-adrenergic receptors: comparison of pharmacologically defined subtypes with subtypes identified by molecular cloning. Mol Pharmacol 1992; 42:1–5.
22. Millan MJ, Bervoets K, Rivet JM, et al. Multiple alpha-2 adrenergic receptor subtypes. II. Evidence for a role of rat R alpha-2A adrenergic receptors in the control of nociception, motor behavior and hippocampal synthesis of noradrenaline. J Pharmacol Exp Ther 1994; 270:958–72.
23. Mizobe T, Maghsoudi K, Sitwala K, et al. Antisense technology reveals the alpha2A adrenoceptor to be the subtype mediating the hypnotic response to the highly selective agonist, dexmedetomidine, in the locus coeruleus of the rat. J Clin Invest 1996; 98:1076–80.
24. Lawhead RG, Blaxall HS, Bylund DB. Alpha-2A is the predominant alpha-2 adrenergic receptor subtype in human spinal cord. Anesthesiology 1992; 77:983–91.
25. Graham BA, Hammond DL, Proudfit HK. Differences in the antinociceptive effects of alpha-2 adrenoceptor agonists in two substrains of Sprague-Dawley rats. J Pharmacol Exp Ther 1997; 283:511–9.
26. Bylund DB. Heterogeneity of alpha-2 adrenergic receptors. Pharmacol Biochem Behav 1985; 22:835–843.
27. Yaksh TL. Pharmacology of spinal adrenergic systems which modulate spinal nociceptive processing. Pharmacol Biochem Behav 1985; 22:845–58.
28. Sabbe MB, Penning JP, Ozaki GT, Yaksh TL. Spinal and systemic action of the alpha 2 receptor agonist dexmedetomidine in dogs: antinociception and carbon dioxide response. Anesthesiology 1994; 80:1057–72.
29. Guo TZ, Jiang JY, Buttermann AE, Maze M. Dexmedetomidine injection into the locus ceruleus produces antinociception. Anesthesiology 1996; 84:873–81.
30. Eisenach JC, Shafer SL, Bucklin BA, et al. Pharmacokinetics and pharmacodynamics of intraspinal dexmedetomidine in sheep. Anesthesiology 1994; 80:1349–59.
31. Eisenach J, Detweiler D, Hood D. Hemodynamic and analgesic actions of epidurally administered clonidine. Anesthesiology 1993; 78:277–87.
32. Kroin JS, McCarthy RJ, Penn RD, et al. Intrathecal clonidine and tizanidine in conscious dogs: comparison of analgesic and hemodynamic effects. Anesth Analg 1996; 82:627–35.
33. Ishiyama T, Dohi S, Iida H, et al. Mechanisms of dexmedetomidine-induced cerebrovascular effects in canine in vivo experiments. Anesth Analg 1995; 81:1208–15.
34. Ishiyama T, Dohi S, Iida H. The vascular effects of topical and intravenous alpha-2 adrenoceptor agonist clonidine on canine pial microcirculation. Anesth Analg 1998; 86:766–72.
35. Kobinger W, Pichler L. Centrally induced reduction in sympathetic tone-a postsynaptic alpha-adrenoceptor-stimulating action of imidazolines. Eur J Pharmacol 1976; 40:311–20.
36. Mao L, Abdel-Rahman AA. Synergistic behavioral interaction between ethanol and clonidine in rats: role of alpha-2 adrenoceptors. J Pharmacol Exp Ther 1996; 279:443–9.
37. Castro MI, Eisenach JC. Pharmacokinetics and dynamics of intravenous, intrathecal, and epidural clonidine in sheep. Anesthesiology 1989; 71:418–25.
© 2000 International Anesthesia Research Society
38. Bernards CM, Hill HF. Physical and chemical properties of drug molecules governing their diffusion through the spinal meninges. Anesthesiology 1992; 77:750–6.