Secondary Logo

Journal Logo

General Articles

The Analgesic Potency of Dexmedetomidine Is Enhanced After Nerve Injury

A Possible Role for Peripheral alpha2-Adrenoceptors

Poree, Lawrence R. MD, PhD; Guo, Tian Z. MD; Kingery, Wade S. MD; Maze, Mervyn MB, ChB

Author Information
doi: 10.1213/00000539-199810000-00037
  • Free

Abstract

alpha2-Adrenoceptor (alpha2 AR)-mediated analgesia has been extensively investigated since the turn of the century and used clinically since 1984 for both acute and chronic pain states [1]. Placebo-controlled clinical trials have reported mixed results using transdermal, oral, and epidural clonidine (alpha (2) AR agonist) for neuropathic pain and for complex regional pain syndromes [2]. Similarly mixed results have been reported for clinical trials using IV morphine in neuropathic pain states, and it has been hypothesized that neuropathic pain is resistant to opioid analgesia [2].

Several investigators have examined the efficacy of intrathecal morphine in reducing radiant heat hyperalgesia in neuropathic rat models. There was a four- to sixfold right shift in the analgesic efficacy of intrathecal morphine for heat hyperalgesia in both the sciatic loose ligature and the L5-6 spinal nerve ligation models of neuropathic hyperalgesia [3,4]. One investigator found that the efficacy of intrathecal morphine was unchanged 1 wk after sciatic loose ligature injury, but by 5 wk postinjury, there was no morphine analgesic effect [5,6]. Another study also found that systemic morphine had no analgesic effect by 4 wk after sciatic loose ligature injury [7]. In one study, systemic morphine had similar analgesic effects for mechanical vocalization thresholds in sciatic loose ligature-injured rats and in control rats, but no statistical comparison was made between the neuropathic and control thresholds [8].

Morphine is also ineffective in neuropathic allodynia models. Intrathecal morphine had either no effect or a weak analgesic effect on von Frey fiber allodynia in the L5-6 ligation model [9-11], and systemic morphine had no effect on von Frey fiber allodynia in the loose sciatic ligature model [7]. Neither a very small dose of systemic morphine (30 [micro sign]g/kg intraperitoneally [IP]) nor a 10-[micro sign]g dose of intracerebral ventricular morphine had any effect on von Frey allodynia in the L5-6 ligation model [4]. Systemic morphine had no effect on von Frey allodynia in the sciatic loose ligature model [7]. Using the loose ligature model, thermal allodynia has been reported to be unresponsive to intrathecal morphine [12], but systemic morphine effectively reduced cold plate allodynia [7].

Based on the reduced effectiveness of morphine in neuropathic hyperalgesia and allodynia models, we postulated that spinal nerve ligation would induce a right shift in the analgesic dose-response to systemically administrated alpha2 AR agonists. To test this hypothesis, we investigated the dose-response effect of systemic dexmedetomidine on mechanical and heat nociceptive withdrawal thresholds in the hindpaws of normal rats and in the L5-6 ligation model. To confirm that these analgesic effects were mediated by the alpha2 AR, we attempted to block dexmedetomidine analgesia with the selective alpha2 AR antagonist atipamezole.

We also examined the dose-response effects of systemic dexmedetomidine on sedation because this is the major limiting side effect in the use of systemic alpha2 AR agonists for pain relief [1]. If the 50% effective dose (ED50) for sedation in neuropathic animals was larger than the ED50 for analgesia, then there may be a therapeutic dosage window with systemic dexmedetomidine for neuropathic pain.

An additional goal of this investigation was to determine the site of analgesic action for dexmedetomidine using the peripherally restricted alpha2 antagonist L-659,066. The alpha2 ARs are located diffusely in the nervous system; on the primary afferents, on sympathetic postganglionic neurons, in the dorsal laminae of the spinal cord, and within the brainstem [13]. There are also alpha (2) ARs on smooth muscle and endothelial cells in arteries and microcapillaries [14]. All these alpha2 AR sites provide an anatomical substrate for possible alpha2 analgesic activity at supraspinal, spinal, or peripheral levels. alpha (2) AR-mediated analgesia for acute nociception is mediated at a spinal and, possibly, a supraspinal level, whereas sedation is mediated at a supraspinal level [1].

Several clinical investigations have observed peripheral analgesic effects with clonidine in chronic pain states [15,16]. If neuropathic animals developed a novel peripheral site of alpha2 analgesic activity, this would raise the interesting possibility of developing peripherally restricted alpha2 agonists for analgesia in neuropathic pain states, thereby avoiding the sedative side effect ceiling encountered with current systemic alpha2 agonists.

In summary, the purpose of this study was to investigate the analgesic and sedative potency and the site of action of systemic dexmedetomidine in normal and neuropathic rats.

Methods

Male Sprague-Dawley rats were used for all experiments. Animals were housed in separate cages and maintained on a 12-h day/night cycle. All studies were conducted with approval from the Institutional Animal Care and Use Committee of the Veterans Administration Palo Alto Health Care System and Stanford University.

Neuropathic hyperalgesia was elicited by using the L5 and L6 spinal nerve ligation model [17]. The surgical procedure was performed in 200-220 g halothane-anesthetized rats. A dorsal midline incision was made at the lower lumbar level, and the left paraspinal muscle was dissected to expose the L6 transverse spinal process. The L6 transverse process was removed with a pair of small ronguer bone cutters, allowing for visualization of the left L4 and L5 spinal nerves. The left L6 spinal nerve was visualized by using a small curved glass probe to hook the nerve medial and ventral to the sacroiliac joint. The L5 and L6 spinal nerves were tightly ligated with 6-0 silk suture.

von Frey fibers were used to measure mechanical nociceptive thresholds. Four fibers provided 9, 23, 56, and 85 g of force and were applied to the lateral dorsal surface of the left hindpaw at a 90[degree sign] angle to the skin surface. Each fiber was presented three times in ascending order for a total of 12 stimuli. Stimulus-evoked withdrawal of the paw was recorded as a positive nociceptive response. A sigmoid Emax model was fit to the data using a maximal likelihood logistic regression technique [18,19]. The mechanical force that corresponded to a 50% probability of a nociceptive response was taken as the mechanical threshold. Postsurgical animals whose mechanical thresholds decreased by a minimum of 20 g by the 14th postoperative day were arbitrarily considered to be hyperalgesic to mechanical stimuli.

A thermoelectric Peltier device was used to measure thermal nociceptive thresholds, as previously described [20,21]. With the surface of the 4 x 4 cm Peltier plate maintained at a baseline temperature of 40[degree sign]C, the plantar surface of the paw was positioned on the surface of the Peltier and allowed to equilibrate for 10 s. After equilibration, the surface temperature of the Peltier was increased at a rate of 1[degree sign]C/s until the rat responded by withdrawing the paw from the plate or until the maximal cutoff temperature of 52[degree sign]C was reached. The average temperature that elicited a paw withdrawal response in three consecutive trials was taken as the thermal nociceptive threshold. Postsurgical animals whose individual thermal nociceptive thresholds decreased by a minimum of 1[degree sign]C by the 14th postoperative day were arbitrarily considered to be hyperalgesic to thermal stimuli.

After the two sensory tests, the spontaneous locomotor activity of the animals was recorded using an open-field activity meter. Testing was performed during the day in a dark quiet room. After 15 min of acclimatization to the room, the rats were individually placed in a 1660-cm2 round chamber with 16 infrared detectors surrounding the chamber to form a grid. As animals crossed the grid, the infrared detector triggered a counter. Grid crossings were recorded for 5 min. A statistically significant decrease in the number of grid crossings after drug administration was taken as an indication of sedation. With the larger doses of dexmedetomidine (>100 [micro sign]g/kg IP), there was a loss of the righting reflex.

The three behavioral tests described above were performed during each testing session starting with mechanical thresholds and ending with grid crossing activity. Dexmedetomidine was administered IP 20 min before the start of behavioral testing. The test sessions were scheduled for Monday and Thursday of each week for each of the experimental groups. The first two test sessions were used as training sessions with a third test session taken as the baseline. Animals that did not meet the criteria for mechanical or thermal hyperalgesia by the fourth postoperative testing session (14th postoperative day) were eliminated from the study and killed (39 of 120 animals). These animals typically did not have withdrawal responses and were assumed to have complete denervation of the paw. All animals that did meet the initial criteria for hyperalgesia remained on the biweekly behavioral testing protocol.

A group of neuropathic animals (n = 56) was used to establish the analgesic dose response relationship of dexmedetomidine in the hyperalgesic state. Each animal received one or two different doses of dexmedetomidine (5, 10, 20, 40, or 60 [micro sign]g/kg IP) in a randomized, blind fashion separated by 1 wk, with a saline control test session between each dexmedetomidine test session to confirm the maintenance of hyperalgesia. Another group of neuropathic animals (n = 12) was used to investigate the effects of two different alpha2 AR antagonists (atipamezole 1 mg/kg IP and L-659,066 1 mg/kg IP) on the analgesic actions of dexmedetomidine (20 [micro sign]g/kg IP). This group underwent antagonist control test sessions between each dexmedetomidine test session, but for these control sessions, the animals were treated with just the antagonist. Atipamezole and L-659,066 were administered 15 min before dexmedetomidine.

Control animals (n = 20) similar to the neuropathic animals in age and weight underwent the same biweekly testing paradigm to establish the dose-response relationship of dexmedetomidine in acute nociception. Each animal received two of the four doses of dexmedetomidine (20, 60, 120, or 300 [micro sign]g/kg) in a randomized, blind fashion separated by 1 wk. An additional group of control animals (n = 20) was treated with 5, 10, and 40 [micro sign]g/kg dexmedetomidine to used to establish the lower end of the grid crossing dose-response curve. Another group of control animals (n = 12) was used to investigate the effects of the antagonists (atipamezole 9 mg/kg IP and L-659,066 9 mg/kg IP) on dexmedetomidine (180 [micro sign]g/kg IP)-induced analgesia and sedation.

Dexmedetomidine (molecular weight [MW] 237), a selective alpha2 AR agonist, and atipamezole (MW 214), a selective alpha2 AR antagonist [22,23], were supplied by Orion Corporation (Turku, Finland). L-659,066 (MW 419), a peripherally restricted alpha2 AR2 antagonist [24], was supplied by Merck Sharp and Dohme Laboratories (NJ). All compounds were administered IP with 0.1-mL volume of 0.9% saline/100 g body weight. Dexmedetomidine was injected 20 min, and the antagonists 35 min, before behavioral testing.

Dose-response data were converted into percent maximal possible effect (%MPE): %MPE = ([postdrug response - predrug response]/[response cutoff - predrug response]) x 100. The response cutoff for von Frey fibers was 85 g; for Peltier testing, it was 52[degree sign]C; and for grid crossing behavior, it was 0 crossings/5 min. ED50 values and 95% confidence intervals were estimated by fitting a sigmoid Emax regression model to the data using a nonlinear regression technique [19,25]. Other data are presented as means +/- SEM. Groups were compared using repeated-measures analysis of variance and Dunnett's post hoc test.

Results

Two weeks after ligation of the L5 and L6 spinal nerves, 68% of the animals exhibited mechanical and/or thermal hyperalgesia. The hyperalgesia was unaffected by IP injections of saline and persisted for at least 5 wk. The mean mechanical nociceptive threshold decreased significantly from 62 +/- 3 to 28 +/- 3 g (Figure 1A). Similarly, the thermal nociceptive threshold in these animals decreased significantly from 49.5 +/- 0.2[degree sign]C to 47.8 +/- 0.2[degree sign]C. (Figure 1B) Spontaneous locomotor activity, as measured by open-field grid crossings, decreased during the first two presurgical test sessions but remained stable post-surgically relative to the third presurgical value of 73 +/- 4 crossings/5 min (Figure 1C). Control animals exhibited similar baseline mechanical and thermal thresholds at 59 +/- 2 g and 49.8 +/- 0.1[degree sign]C, respectively. Baseline grid crossing activity was also similar in control rats at 71 +/- 4 crossings/5 min.

Figure 1
Figure 1:
Persistent mechanical (A) and thermal (B) hyperalgesia gradually developed over a period of 2 wk after the ligation of L5 and L6 spinal nerves. Spontaneous locomotor activity, measured as grid crossings in 5 min, decreased during the first two presurgical test sessions but remained stable after the third presurgical test session. *P < 0.05 before versus after surgery. Repeated-measures analysis of variance with Dunnett's post hoc test (n = 20).

Dexmedetomidine administered IP to control animals increased mechanical and thermal thresholds in a dose-dependent manner, with ED50 values of 144 and 180 [micro sign]g/kg, respectively. By contrast, neuropathic animals responded to much smaller doses of dexmedetomidine, with mechanical and thermal ED50 values of 51 and 29 [micro sign]g/kg, respectively (Figure 2, A and Figure 2 B, Table 1). Both groups had similar responses to dexmedetomidine in the grid crossing assay, with the ED50 values for control and neuropathic animals of 12 and 9 [micro sign]g/kg, respectively (Figure 2C, Table 1). The 95% confidence intervals for dexmedetomidine analgesia in control and neuropathic animals did not overlap, but they did completely overlap for dexmedetomidine inhibition of grid crossing activity (Table 1). After nerve injury, there was a three- to sixfold left shift in the analgesic efficacy of dexmedetomidine, but no shift was observed for the sedative efficacy (Figure 2, Table 1).

Figure 2
Figure 2:
Dexmedetomidine increased mechanical (A) and thermal (B) thresholds and decreased locomotor activity (C) in both control (circles) and neuropathic (triangles) animals in a dose-dependent fashion. Measurements were made 2-5 wk after the ligation of L5 and L6 spinal nerves. Neuropathic animals demonstrated a three- to sixfold left shift for the analgesic efficacy of dexmedetomidine (A, B), but no shift was observed with the sedative effect (C). See Table 1 for 50% effective dose values, 95% confidence intervals, and number of animals. %MPE = percent maximal effect.
Table 1
Table 1:
Behavior Assays

In both control and neuropathic animals, the analgesic and sedative actions of dexmedetomidine were blocked by the alpha2 AR antagonist atipamezole administered 15 min before dexmedetomidine (Figure 3 and Figure 4). By contrast, pretreatment with the peripherally restricted alpha2 AR antagonist L-659,066 blocked only the analgesic actions of dexmedetomidine in neuropathic animals (Figure 4). Neither dexmedetomidine-evoked analgesia in control animals nor dexmedetomidine-evoked sedation in either group was affected by prior treatment with L-659,066 (Figure 3 and Figure 4). When given alone, large-dose atipamezole (9 mg/kg IP) tended to increase grid crossing behavior in normal rats (Figure 3C). This increased activity is probably attributable to alpha2 AR antagonist-induced norepinephrine release in the brain [26], with the resulting arousal and restlessness [23].

Figure 3
Figure 3:
Atipamezole (Ati; 9 mg/kg IP) but not L-659,066 (L659; 9 mg/kg IP) blocked the mechanical (A) and thermal (B) analgesic actions of dexmedetomidine (Dex; 180 [micro sign]g/kg IP) in control animals. Atipamezole, but not L-659,066, also blocked dexmedetomidine inhibition of locomotor activity (C) in control animals. Loss of righting reflex (LORR) occurred in all animals that received dexmedetomidine alone or in combination with L-659,066. *P < 0.05 versus saline (Sal). Repeated-measures analysis of variance with Dunnett's post hoc test (n = 12).
Figure 4
Figure 4:
Atipamezole (Ati; 1 mg/kg IP) and L-659,066 (L659; 1 mg/kg IP) blocked the mechanical (A) and thermal (B) analgesic actions of dexmedetomidine (Dex; 20 [micro sign]g/kg IP) in neuropathic animals. Atipamezole, but not L-659,099, also blocked the dexmedetomidine inhibition of locomotor activity (C) in neuropathic animals. Measurements were made 2-5 wk after the ligation of L5 and L6 spinal nerves. *P < 0.05 versus saline (Sal) alone. Repeated-measures analysis of variance with Dunnett's post hoc test (n = 25).

Discussion

The systemic administration of dexmedetomidine increased the mechanical and thermal paw withdrawal thresholds in neuropathic and control rats in a dose-dependent manner (Figure 2). This analgesic action is mediated by the alpha2 AR and is effectively blocked by the alpha2 AR selective antagonist atipamezole (Figure 3 and Figure 4). The dose-dependent analgesia observed in the control rats is consistent with previous studies demonstrating that systemic alpha2 agonists are potent analgesics in animal models of acute nociception [1].

There are little data on alpha2 AR-mediated analgesia in animal models of neuropathic hyperalgesia. The maximal analgesic effect of intrathecal ST-91 on hindpaw heat latencies was reduced after sciatic loose ligature compared with control [27]. Intrathecal tizanidine had no effect on hindpaw heat latencies after sciatic loose ligature injury, but it did have a slight analgesic effect on mechanical nociceptive thresholds [28]. After sciatic loose ligature injury, systemic clonidine had a moderate analgesic effect on mechanical nociceptive thresholds and a large analgesic effect on paw dip latencies to warm (44[degree sign]C) and cool (10[degree sign]C) water baths [29] and on cold plate (5[degree sign]C) paw lifting behavior [7]. Intrathecal dexmedetomidine and clonidine were effective analgesics for von Frey allodynia in the L5-6 ligation model [9,30], but continuous infusion of dexmedetomidine (30-60 [micro sign]g/d subcutaneously) had no effect on von Frey or cold allodynia in this model [31]. Systemic clonidine (<or=to250 [micro sign]g/kg) had no effect on von Frey allodynia in the loose ligature model [7].

None of these studies presented analgesic data evaluating the effect of nerve injury on the alpha2 analgesic dose-response curve in normal and neuropathic animals. Our results indicate that 2 wk after the L5-6 spinal nerve ligation, there was a three- to sixfold left shift in the analgesic efficacy of dexmedetomidine for mechanical and heat nociception in the neuropathic hindpaw (Table 1) and that this analgesia was mediated by the alpha2 AR. The finding of increased analgesic efficacy for alpha2 analgesia after nerve injury is in contrast to numerous reports of diminished analgesic efficacy for opioids in neuropathic models [3,4,6,7,9-12].

Studies using the L5-6 spinal nerve ligation model have shown that the systemic administration of the alpha2 AR antagonist atipamezole could increase or kindle cold allodynia, but not mechanical allodynia [31], and that systemic phentolamine (a nonselective alpha AR antagonist) reversed mechanical allodynia [32]. In contrast, our data showed that systemic atipamezole in neuropathic rats had no effect on mechanical or heat hyperalgesia in the neuropathic paw (Figure 4).

There was also a dose-dependent inhibition of locomotor activity in the control and neuropathic rats with systemic dexmedetomidine administration (Figure 2). There was no change in the efficacy of dexmedetomidine sedation (inhibition of grid crossing) in the neuropathic rats (Table 1), and the sedative effects were effectively blocked with atipamezole, which indicates alpha2 AR mediation (Figure 3 and Figure 4). There was no overlap of the 95% confidence intervals for the ED50 values for sedation and analgesia in either the control or the neuropathic rats (Table 1), and significant sedation occurred with dexmedetomidine doses below the threshold for effective analgesia in both groups (Figure 2). These data indicate that there is no therapeutic dosage window in this model of neuropathic pain allowing for analgesia without sedative side effects.

The ability of systemic dexmedetomidine to increase the nociceptive thresholds in control and neuropathic rats is due to analgesic rather than sedative effects. Unlike nociceptive assays, such as the hot plate or formalin tests, which require complex supraspinally mediated behavioral responses such as licking or jumping, the withdrawal reflexes are spinally mediated and are observed after spinal cord transection and in decerebrated rats [33,34]. Furthermore, even rats that were severely sedated always withdrew before the cutoff threshold was reached (Figure 2), and the large left shift observed for analgesia in the neuropathic rats was not observed for sedation, which indicates that changes in the analgesia ED50 did not correlate with changes in the sedation ED50.

Systemic administration of the peripherally restricted alpha2 AR antagonist L-659,066 had no effect on nociceptive thresholds in the control or neuropathic rats (Figure 3 and Figure 4). L-659,066 did not antagonize the analgesic effect of dexmedetomidine in control rats but was just as effective as the nonrestricted alpha2 AR antagonist atipamezole in antagonizing the analgesic effects of dexmedetomidine in neuropathic rats (Figure 4). Spinal nerve ligation may have induced the development of a novel site of alpha2 analgesia outside the blood-brain barrier.

Nerve injuries can cause dramatic changes in the coupling of adrenoceptors and calcium channels in dorsal root ganglion neurons, which may affect neurotransmitter release [35]. Immunohistochemistry with polyclonal antibodies demonstrate that the alpha2A and alpha2C AR subtypes are found in dorsal root ganglion neurons, and within 2 wk after a complete or partial sciatic section, there is a 10- to 20-fold increase in the number of neurons labeled with the alpha2A AR subtype, which may enhance the analgesic efficacy of alpha2 agonists [36]. One problem with this hypothesis is that, in neuropathic rats, the largest increase in alpha2A AR was on the medium-large diameter dorsal root ganglion neurons, which normally do not transmit nociceptive information but may play a role in neuropathic allodynia.

The hypothesis of a novel peripheral site of alpha2 analgesic activity in neuropathic animals conflicts with behavioral data demonstrating intrathecal alpha (2) analgesic activity in neuropathic animals [9,30] and with electrophysiological data demonstrating that nerve injury causes a left shift in the efficacy of intrathecal alpha2 agonist inhibition of the electromyographic flexor reflex [33,37,38]. Unlike atipamezole, L-659,066 did not antagonize the sedative effects of dexmedetomidine in neuropathic rats (Figure 4), which indicates that the integrity of the blood-brain barrier was intact at a supraspinal level. However, the spinal nerve ligation may have caused a local disruption of the blood-brain barrier at a segmental level in the spinal cord, allowing the L-659,066 to enter the dorsal horn. Further investigations are required to confirm whether nerve injury can generate a novel peripheral site of alpha2 analgesic activity.

In conclusion, we were surprised to observe a left shift in the analgesic dose-response curve with systemic dexmedetomidine in the neuropathic rats. No shift was observed for the sedative dose-response effect of dexmedetomidine in neuropathic rats, and although the nonrestricted alpha2 AR antagonist atipamezole could effectively block the analgesic effects of dexmedetomidine in both the control and neuropathic rats, the peripheral alpha2 AR antagonist L-659, 066 only blocked the analgesic effects of dexmedetomidine in the neuropathic animals. These data indicate that ligation of the L5-6 spinal nerves both caused sensitization to alpha2 analgesia and shifted the site of alpha2 analgesic action to outside the blood-brain barrier.

Special thanks to Dr. Patricia Cross for her support of this work and to Dr. Russell Wada and Dr. Martin Angst for their advice on regression techniques.

REFERENCES

1. Kingery WS, Davies MF, Maze M. Molecular mechanisms for the analgesic properties of alpha-2 adrenergic agonists. In: Borsook D, ed. Molecular neurobiology of pain. Seattle, WA: IASP Press, 1997:275-304.
2. Kingery WS. A critical review of controlled clinical trials for peripheral neuropathic pain and complex regional pain syndromes. Pain 1997;73:123-39.
3. Mao J, Price DD, Mayer DJ. Experimental mononeuropathy reduces the antinociceptive effects of morphine: implications for common intracellular mechanisms involved in morphine tolerance and neuropathic pain. Pain 1995;61:353-64.
4. Wegert S, Ossipov MH, Nichols ML, et al. Differential activities of intrathecal MK-801 or morphine to alter responses to thermal and mechanical stimuli in normal or nerve-injured rats. Pain 1997;71:57-64.
5. Yamamoto T, Yaksh TL. Stereospecific effects of a nonpeptidic NK1 selective antagonist, CP-96,345: antinociception in the absence of motor dysfunction. Life Sci 1991;49:1955-1963.
6. Yamamoto T, Shimoyama N, Asano H, Mizuguchi T. Time-dependent effect of morphine and time-independent effect of MK-801, an NMDA antagonist, on the thermal hyperesthesia induced by unilateral constriction injury to the sciatic nerve in the rat. Anesthesiology 1994;80:1311-9.
7. Jasmin L, Kohan L, Franssen M, et al. The cold plate as a test of nociceptive behaviors: description and application to the study of chronic neuropathic and inflammatory pain models. Pain 1998;75:367-82.
8. Attal N, Chen YL, Kayser V, Guilbaud G. Behavioral evidence that systemic morphine may modulate a phasic pain-related behaviour in a rat model of peripheral mononeuropathy. Pain 1991;47:65-70.
9. Yaksh TL, Pogrel JW, Lee YW, Chaplan SR. Reversal of nerve ligation-induced allodynia by spinal alpha-2 adrenoceptor agonists. J Pharmacol Exp Ther 1995;272:207-14.
10. Nichols ML, Bian D, Ossipov MH, et al. Regulation of morphine antiallodynic efficacy by cholecystokinin in a model of neuropathic pain in rats. J Pharmacol Exp Ther 1995;275:1339-45.
11. Bian D, Nichols ML, Ossipov MH, et al. Characterization of the antiallodynic efficacy of morphine in a model of neuropathic pain in rats. Neuroreport 1995;6:1981-4.
12. Lee SH, Kayser V, Desmeules J, Guilbaud G. Differential action of morphine and various opioid agonists on thermal allodynia and hyperalgesia in mononeuropathic rats. Pain 1994;57:233-40.
13. 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 Rev 1984;7:69-101.
14. Ruffolo RR, Nichols AJ, Stadel JM, Hieble JP. Pharmacologic and therapeutic applications of alpha 2-adrenoceptor subtypes. Annu Rev Pharmacol 1993;32:243-79.
15. Davis KD, Treede RD, Raja SN, et al. Topical application of clonidine relieves hyperalgesia in patients with sympathetically maintained pain. Pain 1991;47:309-17.
16. Gentili M, Juhel A, Bonnet F. Peripheral analgesic effect of intra-articular clonidine. Pain 1996;64:593-6.
17. Kim SH, Chung JM. An experimental model for peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 1992;50:355-63.
18. Hosmer DW, Lemeshow S. Introduction to the logistic regression model: applied logistic regression. New York: Wiley, 1989:1-24.
19. Holford NHG, Sheiner LB. Understanding the dose-effect relationship: clinical application of pharmacokinetic-pharmacodynamic models. Clin Pharmacokinet 1981;6:429-53.
20. Kingery WS, Castellote JM, Wang EE. A loose ligature-induced mononeuropathy produces hyperalgesias mediated by both the injured sciatic nerve and the adjacent saphenous nerve. Pain 1993;55:297-304.
21. Vallin JA, Kingery WS. Adjacent neuropathic hyperalgesia in rats: a model for sympathetic independent pain. Neurosci Lett 1991;133:241-4.
22. Virtanen R, Savola JM, Saano V. Highly selective and specific antagonism of central and peripheral alpha 2-adrenoceptors by atipamezole. Arch Int Pharcodyn Ther 1989;297:190-204.
23. Scheinin H, MacDonald E, Scheinin M. Behavioral and neurochemical effects of atipamezole, a novel alpha-2 adrenoceptor antagonist. Eur J Pharmacol 1988;157:35-42.
24. Clineschmidt BV, Pettibone DJ, Lotti VJ, et al. A peripherally acting alpha-2 adrenoceptor antagonist: L-659,066. J Pharmacol Exp Ther 1988;245:32-40.
25. Sheiner LB. Analysis of pharmacokinetic data using parametric models. II. Point estimates of an individual's parameters. J Pharmacokinet Biopharm 1985;13:515-40.
26. Laitien KS, Tuomisto L, McDonald E. Effects of a selective alpha 2-adrenoceptor antagonist, atipamezole, on hypothalamic histamine and noradrenaline release in vivo. Eur J Pharmacol 1995;285:255-60.
27. Yamamoto T, Yaksh TL. Spinal pharmacology of thermal hyperesthesia induced by incomplete ligation of sciatic nerve. Anesthesiology 1991;75:817-26.
28. Leiphart JW, Dills CV, Zikel OM, et al. A comparison of intrathecally administered narcotic and nonnarcotic analgesics for experimental chronic neuropathic pain. J Neurosurg 1995;82:595-9.
29. Kayser V, Desmeules J, Guilbaud G. Systemic clonidine differentially modulates the abnormal reactions to mechanical and thermal stimuli in rats with peripheral mononeuropathy. Pain 1995;60:275-85.
30. Lee Y-W, Yaksh TL. Analysis of drug interaction between intrathecal clonidine and MK-801 in peripheral neuropathic pain rat model. Anesthesiology 1995;82:741-8.
31. Kontinen VK, Paananen S, Kalso E. The effects of the alpha2-adrenergic agonist, dexmedetomidine, in the spinal nerve ligation model of neuropathic pain in rats. Anesth Analg 1998;86:355-60.
32. Kim SH, Na HS, Sheen K, Chung JM. Effects of sympathectomy on a rat model of peripheral neuropathy. Pain 1993;55:85-92.
33. Xu XJ, Dalsgaard CJ, Wiesenfeld-Hallin Z. Intrathecal CP-96345 blocks reflex facilitation induce in rats by substance P and C-fiber conditioning stimulation. Eur J Pharmacol 1992;216:337-44.
34. Pertovaara A, Kauppila T, Jyvasjarvi E, Kalso E. Involvement of supraspinal and spinal segmental alpha 2-adrenergic mechanisms in the medetomidine-induced antinociception. Neuroscience 1991;44:705-14.
35. Abdulla FA, Smith PA. Ectopic alpha 2-adrenoceptors couple to N-type calcium channels in axotomized rat sensory neurons. J Neurosci 1997;17:1633-41.
36. Birder LA, Perl ER. Upregulation of the alpha-2A adrenergic receptor subtype after peripheral nerve injury[abstract]. Soc Neurosci Abstr 1996;22:1803.
37. Xu XJ, Wikberg JES, Wiesenfeld-Hallin Z. The effect of intrathecal guanfacine and clonidine on the flexor reflex in rats with intact and sectioned sciatic nerves. Eur J Pharmacol 1993;235:161-4.
38. Puke MJC, Luo L, Xu X-J. The spinal analgesic role of alpha2-adrenoceptor subtypes in rats after peripheral nerve section. Eur J Pharmacol 1994;260:227-32.
© 1998 International Anesthesia Research Society