Anesthesiology 2004; 100:1258–62 © 2004 American Society of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc. Spinal Adenosine Receptor Activation Reduces Hypersensitivity after Surgery by a Different Mechanism Than after Nerve Injury Hideaki Obata, M.D.,* Xinhui Li, Ph.D.,† James C. Eisenach, M.D.‡

Background: Intrathecal adenosine has antinociceptive ef- itself does not produce analgesia in normal rats to acute fects under conditions of hypersensitivity. T62 (2-amino-3-(4- noxious stimulation.6 In contrast, adenosine does pro- chlorobenzoyl)-5,6,7,8-tetrahydrobenzothiophen) is an alloste- duce long-lasting analgesia in a model of neuropathic ric adenosine receptor modulator that enhances adenosine pain.7 A common problem with administration of direct binding to the A1 receptor. Intrathecal T62 reduces hypersensi- Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/100/5/1258/354837/0000542-200405000-00030.pdf by guest on 26 September 2021 tivity to mechanical stimuli in a rat model of neuropathic pain agonists, such as adenosine itself, is adverse effects ␣ by a circuit that totally relies on activation of 2 adrenoceptors. caused by stimulation of all receptors, not just those Here, the authors tested whether this same dependence was involved in the therapeutic effect. One approach to present in the acute setting of hypersensitivity after surgery. prevent this problem is the use of positive allosteric Methods: Intrathecal catheters were inserted in male Sprague- Dawley rats. An incision of the plantar aspect of the hind paw modulators. T62 (2-amino-3-(4-chlorobenzoyl)-5,6,7,8- resulted 24 h later in hypersensitivity, as measured by applying tetrahydrobenzothiophen) is an allosteric adenosine re- von Frey filaments to the paw. At this time, rats received intra- ceptor modulator that enhances adenosine binding to thecal T62, clonidine, or the combination in a blinded, isobo- the A receptor through a conformation change.8,9 In- lographic design. The effect of the ␣ -adrenoceptor antagonist 1 2 trathecal administration of T62 reduces mechanical allo- idazoxan on T62 was also tested. 10,11 Results: Intrathecal T62 produced a dose-dependent antihy- dynia in a rat model of neuropathic pain and thermal 12 persensitivity effect, with no effect on ambulation or activity hypersensitivity in carrageenin-inflamed rats. Surgery is level. Clonidine also produced a dose-dependent antihypersen- another common cause of persistent pain and hyperal- sitivity effect. The ED40 (95% confidence interval) for T62 was gesia, and these pain phenomena in humans can be ␮ 0.77 (0.63–0.91) g, and that for clonidine was 1.23 (0.56–1.9) mimicked by paw incision in rats.13 Hypersensitivity to ␮g. Isobolographic analysis indicated synergism between T62 and clonidine. Intrathecal pretreatment with idazoxan only mechanical stimuli after paw incision exhibits a different partially inhibited the antihypersensitivity effect of T62. analgesic pharmacology compared with that of nerve Conclusions: Intrathecal T62 is effective for postoperative hy- injury or inflammation. One purpose of the current study persensitivity. The synergy of T62 with clonidine and its only was to evaluate the potency and efficacy of intrathecally partial antagonism by idazoxan suggest that T62 does not rely ␣ administered T62 in this experimental model of postop- entirely on activation of 2 adrenoceptors. These results indi- cate that, after surgery, T62 acts via a mechanism different from erative pain in rats. that of spinal nerve ligation, a model of chronic neuropathic An obligatory interaction between adenosine or T62 pain. and spinal noradrenergic function has been postulated in a rat model of neuropathic pain.11,14,15 Thus, the effects ADENOSINE is recognized to play a role in modulation of of adenosine or T62 after nerve injury are totally abol- 1 nociceptive transmission in the spinal cord. Both A1 and ished by destruction of noradrenergic terminals or ad- ␣ A2 subtypes of adenosine receptors are present in the ministration of 2-adrenoceptor antagonists. We specu- 2 ␣ substantia gelationosa. Although selective A2 agonists lated that this adenosine– 2 adrenoceptor interaction were shown to have no antinociceptive effect, A1 recep- would also be present in the efficacy of T62 after sur- tors have been clearly identified to produce antinocicep- gery. Another purpose of the study was to test this ␣ tion in the spinal cord by using selective agonists and hypothesis, first by examining the effect of 2-adreno- antagonists.3–5 The spinal administration of adenosine ceptor antagonism on the action of T62, and second by defining whether T62 and clonidine interacted in an additive manner, as one would expect if T62 were totally ␣ * Research Fellow, Department of Anesthesiology, Gunma University Graduate dependent on 2-adrenoceptor activation. School of Medicine. † Assistant Professor of Anesthesiology, ‡ F.M. James, III Professor of Anesthesiology, Wake Forest University School of Medicine. Received from the Department of Anesthesiology and Center for the Study of Pharmacologic Plasticity in the Presence of Pain, Wake Forest University School Materials and Methods of Medicine, Winston-Salem, North Carolina, and the Department of Anesthesi- ology, Gunma University Graduate School of Medicine, Maebashi, Gunma, Japan. Surgical Preparation Submitted for publication September 1, 2003. Accepted for publication Decem- The study was approved by the Animal Care and Use ber 7, 2003. Supported by grant No. GM35523 from the National Institutes of Health, Bethesda, Maryland; a grant from King Pharmaceuticals, Cary, North Committee of the Wake Forest University School of Med- Carolina; and a grant for Japanese researchers overseas from the Ministry of icine (Winston-Salem, North Carolina). Male Sprague- Education, Culture, Sports, Science and Technology, Tokyo, Japan. Address reprint requests to Dr. Eisenach: Department of Anesthesiology, Wake Dawley rats (weight, 250 g) obtained from Harlan (Indi- Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, anapolis, IN) were used in all experiments. Animals were North Carolina 27157. Address electronic mail to: [email protected]. Indi- vidual article reprints may be purchased through the Journal Web site, housed under a 12-h light–dark cycle, with food and www.anesthesiology.org. water ad libitum. For intrathecal administration, a ster-

Anesthesiology, V 100, No 5, May 2004 1258 SPINAL T62 FOR POSTOPERATIVE ANALGESIA 1259

ilized 32-gauge polyethylene catheter (RecathCo, Allison responses, a fixed-ratio combination of T62 and Park, PA) connected to 8.5 cm Tygon external tubing clonidine in a ratio of 0.16:1 wt/wt was administered. (Saint-Gobain Performance Plastics, Akron, OH) was in- Because the time of peak effect for T62 was 30 min later serted under halothane anesthesia, as previously de- than clonidine, T62 was administered first, followed by scribed.16 The catheter was passed caudally 7.5 cm from clonidine injection 30 min later. the cisterna magnum to the lumbar enlargement. Only An antagonist study was performed to test whether the animals without evidence of neurologic dysfunction af- effect of T62 in the postoperative pain model involves ␣ ␣ ter catheter insertion were used for studies. Paw incision 2-adrenergic receptor stimulation. The selective 2-ad- as described by Brennan et al.13 was performed 5 days renoceptor antagonist idazoxan, 30 ␮g, or saline was after intrathecal catheter implantation. For this, rats administered intrathecally 15 min before T62 injection.

were anesthetized with halothane, and after sterile prep- The effect of idazoxan alone was tested in another group Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/100/5/1258/354837/0000542-200405000-00030.pdf by guest on 26 September 2021 aration with 70% ethanol, a 1-cm long incision was made of animals. in the plantar aspect of the left hind paw, starting 0.5 cm Drugs were administered intrathecally in a volume of from edge of the heel toward the toe. The plantaris 5 ␮l and followed by a 10-␮l saline injection to flush the muscles was elevated and incised longitudinally. The catheter. T62 was dissolved in 20% dimethyl sulfoxide, wound was closed with two mattress sutures of 5.0 silk. except for the highest dose, which was dissolved in 30% dimethyl sulfoxide because of its low solubility. Other Behavioral Testing drugs were dissolved in normal saline. T62 was obtained Rats were placed individually in a plastic cage with a from King Pharmaceuticals (Cary, NC). Clonidine and plastic mesh floor and were allowed to acclimate to the idazoxan were purchased from Sigma Chemical Co. (St. environment for 30 min. Withdrawal threshold was de- Louis, MO). termined using calibrated von Frey filaments (Stoelting, Wood Dale, IL), beginning with the 2.0-gauge filament. Statistics Filaments were applied vertically to an area adjacent to Data were normally distributed and are shown as mean the wound for 6 s while the filament was gently bent. In Ϯ SEM. Time course and dose–response effects of T62 or the absence of a response, the filament of next greater clonidine were analyzed using a two-way analysis of force was applied. In the presence of a response, the variance, followed by a Student-Newman-Keuls post hoc filament of next lower force was applied. The tactile test. ED40 was calculated by linear regression analysis. stimulus producing a 50% likelihood of withdrawal was The isobolograms were constructed as previously de- determined using the up–down method, as described by scribed.18 In isobolographic analysis, the theoretical ad- Chaplan et al.17 General behavior, including ambulation ditive points lies on a line connecting the ED40 values of and activity level, was assessed throughout the time of the each drug. Experimental values that lie on or near testing. The investigator was blinded to drug treatment that line are considered to have additive interactions. for all studies. Values that lie below and to the left of this additive line are considered to be synergistic, whereas values that lie Drugs and Their Administration above and to the right of that line demonstrate a less- Drug testing was performed 24 h after paw incision. than-additive interaction. The difference between the Rats received intrathecal T62 (0.3, 0.5, 1, and 3 ␮g) or theoretical additive point and the experimentally deter- clonidine (1, 3, and 10 ␮g). The withdrawal threshold mined value was compared by the Student t test. A P was determined before (prepaw incision threshold) and value of less then 0.05 was considered to indicate statis- 24 h after incision (baseline) and then every 30-min for tical significance. 4 h after intrathecal injection, using an up–down method with von Frey filaments. Dose–response curves were conducted from a peak effect at each dose after Results conversion of withdrawal thresholds to percentages of maximum possible effect (%MPE), where % ϭ 100 ϫ Overall, paw incision resulted in a reduced paw with- (Postdrug Response Ϫ Baseline)/(Prepaw Incision drawal threshold 24 h after surgery compared with pre- Threshold Ϫ Baseline). surgical values (4.7 Ϯ 0.19 g after, 31 Ϯ 1.2 g before; After determining the efficacy of each drug, isobolo- P Ͻ 0.05; n ϭ 102), with no differences across treatment graphic analysis was performed to determine the type of groups. Intrathecal administration of T62, 0.3–1 ␮g, pro- interaction between T62 and clonidine. The dose pro- duced a dose-dependent antihypersensitivity effect, with

ducing a 40% MPE (ED40) was calculated from dose– a peak 60 min after injection (fig. 1). T62 did not affect response curves for each drug in reducing postoperative ambulation or the level of general activity. The maxi- ␮ hypersensitivity. Using an ED40 was necessary because mum dose studied, 3 g, was less effective than the maximal effects of T62 in the dose range studied were effect after administration of 1 ␮g(fig. 2). Clonidine also not greater than 50% MPE. For drug combination dose produced a dose-dependent antihypersensitivity effect,

Anesthesiology, V 100, No 5, May 2004 1260 OBATA ET AL.

Fig. 1. Time course of the antihypersen- sitivity effects of intrathecally adminis- tered T62 (left) and clonidine (right)in rats with 1 day after paw incision sur- gery. Presurgery baseline values are indi- cated at the time before 0. Withdrawal ؎ thresholds are expressed as mean SEM Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/100/5/1258/354837/0000542-200405000-00030.pdf by guest on 26 September 2021 for seven or eight rats in each group. For T62, * P < 0.05 versus vehicle (20% di- methyl sulfoxide) group, # P < 0.05 ver- -sus 0.3 ␮g–treated group, ؉ P < 0.05 ver sus 0.5 ␮g–treated group. For clonidine, * P < 0.05 versus 1 ␮g–treated group, # P < 0.05 versus 3 ␮g–treated group.

with a peak effect 30 min after injection (fig. 1). Dose (0.35–0.37) ␮g, just 31% of the theoretical additive total responses at the time of peak antihypersensitivity for dose (1.16 ␮g; P Ͻ 0.05). Isobolographic analysis indi-

T62 and clonidine are shown in figure 2. The ED40 (95% cated a synergistic interaction between T62 and confidence interval) for T62, based on the 0.3- to 1-␮g clonidine (fig. 4). Intrathecal pretreatment with 30 ␮g doses, was 0.77 (0.63–0.91) ␮g, and that for clonidine idazoxan only partially inhibited (by approximately 55%) was 1.23 (0.56–1.9) ␮g. The peak antihypersensitivity the antihypersensitivity effect of 1 ␮g T62 (fig. 5). Intra- effects of intrathecal injection of the fixed-ratio combi- thecal administration of 30 ␮g idazoxan alone did not nation of two drugs were observed 30 min after injection alter the withdrawal threshold (the thresholds before of clonidine (fig. 3). ED40 (total dose) was only 0.36 and 15, 45, and 75 min after idazoxan injection were 5.0 Ϯ 0.46, 4.6 Ϯ 0.49, 4.2 Ϯ 0.30, and 4.4 Ϯ 0.27 g, respectively; n ϭ 6).

Fig. 2. Log dose–response curves of intrathecal administration Fig. 3. Time course of the antihypersensitivity effects of a fixed- of T62 and clonidine on paw incision–induced mechanical hy- ratio combination of T62 and clonidine (0.16:1 wt/wt) intrathe- persensitivity. Peak effects were used to calculate percentages cally in rats with paw incision. Withdrawal thresholds are ex- .of the maximum possible effect (%MPE). Data are expressed as pressed as mean ؎ SEM for seven or eight rats in each group .mean ؎ SEM for seven or eight rats in each group. * P < 0.05 versus 15% group, # P < 0.05 versus 5% group

Anesthesiology, V 100, No 5, May 2004 SPINAL T62 FOR POSTOPERATIVE ANALGESIA 1261

thecally administered adenosine in rats with spinal nerve ligation is completely abolished by intrathecal pretreat- ment with idazoxan, 30 ␮g, or by destruction of spinal noradrenergic nerve terminals using neurotoxins.11,14 Spinal perfusion of adenosine releases norepinephrine in rats with spinal nerve ligation but not in normal rats,15 suggesting that this interaction does not occur in the normal animal, in which adenosine is inactive. The mechanisms of norepinephrine release by activation of

spinal adenosine A1 receptors after nerve injury are not

understood. Such mechanisms must involve disinhibi- Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/100/5/1258/354837/0000542-200405000-00030.pdf by guest on 26 September 2021

tion because A1 adenosine receptors are themselves in- hibitory, and the circuitry involved in such disinhibition is under active investigation in our laboratory. A recent report shows that spinal noradrenergic innervation den- sity to the lumbar dorsal horn is increased after nerve injury.19 Adenosine receptors are present on intrinsic interneurons in the dorsal horn.2 It is conceivable that these adenosine receptor–containing neurons interact

Fig. 4. Isobologram at the 40% maximum effective dose (ED40) with novel noradrenergic sprouts after injury to produce level for intrathecal T62 and clonidine on mechanical hyper- antihypersensitivity. sensitivity after paw incision. The ED40 values and their 95% confidence intervals are shown for each drug alone on the axis. In the current study, the combination of T62 and

The theoretical additive line is drawn between the two ED40 clonidine produced a synergistic interaction in rats with values. The ED40 value and 95% confidence interval observed incisional pain. Synergy usually indicates that the two for the fixed-ratio combination (circles) was found to be signif- icantly below the theoretical additive line, indicating a syner- drugs have different final pathways to produce their gistic interaction. * P < 0.05 versus theoretical additive point. effect, although other levels of interaction, such as al- tered drug disposition, can also be responsible. We did Discussion not measure tissue concentrations of drugs, so we can- not exclude a pharmacokinetic mechanism of synergy in Intrathecal administration of T62 inhibits postopera- the current study. Nonetheless, the observation of syn- tive hypersensitivity in this animal model of postopera- ergy is somewhat surprising if, as indicated by studies tive pain without motor or sedative effects, as deter- with spinal nerve ligation, the effect of T62 relies en- mined by gross ambulatory and spontaneous activity. tirely on stimulating spinal norepinephrine release, T62 also exhibits activity in other settings of hypersen- sitivity, such as mechanical allodynia in rats with spinal nerve ligation10,11 and thermal hypersensitivity in carra- geenin-inflamed rats,12 but is inactive for acute thermal and mechanical pain tests in normal rats.12 These obser- vations are consistent with the results from previous studies with intrathecal injection of adenosine itself, which suppresses mechanical allodynia after nerve inju- ry,7,14 but shows minimal effect in acute thermal pain.6 In the current study, intrathecal administration of T62 produced only modest effects in this model of postop- erative pain, exhibiting a ceiling effect at 1 ␮g and an inverted U–shaped dose response. This dose–response characteristic is not unexpected because it is similar to the bell-shaped concentration response for allosteric adenosine receptor modulators on agonist binding to the to A1 receptor in vitro, thought to reflect receptor an- tagonism at high drug concentrations.8 Fig. 5. Effects of intrathecal pretreatment of idazoxan, a selec- Several previous observations have suggested that re- ␣ tive 2-adrenoceptor antagonist, on the antihypersensitivity ef- duction in hypersensitivity elicited by intrathecal aden- fect of 1 ␮g T62. Rats were pretreated with saline or alterna- osine is mediated by a spinal circuit that involves the tively pretreated with idazoxan 15 min before T62 ␣ administration. * P < 0.05, ** P < 0.01 versus the group treated percentage of the maximum ؍ release of norepinephrine and actions on 2 adrenocep- with saline plus T62. %MPE tors. Therefore, the antihypersensitivity effect of intra- possible effect.

Anesthesiology, V 100, No 5, May 2004 1262 OBATA ET AL.

␣ which acts on 2 adrenoceptors. One would in that case References expect an additive interaction, and intrathecal adenosine 1. Sawynok J, Liu XJ: Adenosine in the spinal cord and periphery: Release and and clonidine do interact additively in spinal-ligated regulation of pain. Prog Neurobiol 2003; 69:313–40 14 rats. In contrast, a synthetic adenosine agonist inter- 2. Choca JI, Green RD, Proudfit HK: Adenosine A1 and A2 receptors of the substantia gelatinosa are located predominantly on intrinsic neurons: An autora- acts synergistically with clonidine in acute thermal no- diography study. J Pharmacol Exp Ther 1988; 247:757–64 20 ciception tests in normal rats. In addition, idazoxan 3. Poon A, Sawynok J: Antinociception by adenosine analogs and an adenosine only partially reversed the antihypersensitivity effect of kinase inhibitor: Dependence on formalin concentration. Eur J Pharmacol 1995; 286:177–84 T62 in the current study. Taken together, these results 4. Nakamura I, Ohta Y, Kemmotsu O: Characterization of adenosine receptors indicate that, in the postoperative model, T62 acts via a mediating spinal sensory transmission related to nociceptive information in the rat. ANESTHESIOLOGY 1997; 87:577–84 mechanism different from that following the spinal 5. Poon A, Sawynok J: Antinociception by adenosine analogs and inhibitors of

nerve ligation model: Both spinal activation and other adenosine metabolism in an inflammatory thermal hyperalgesia model in the rat. Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/100/5/1258/354837/0000542-200405000-00030.pdf by guest on 26 September 2021 Pain 1998; 74:235–45 mechanisms are involved in the antihypersensitivity ef- 6. Chiari AI, Eisenach JC: Intrathecal adenosine: Interactions with spinal fect of intrathecally administered T62 after surgery. clonidine and neostigmine in rat models of acute nociception and postoperative Adenosine receptor activation in the spinal cord is pro- hypersensitivity. ANESTHESIOLOGY 1999; 90:1413–21 7. Lavand’homme PM, Eisenach JC: Exogenous and endogenous adenosine posed to produce antinociception via presynaptic inhi- enhance the spinal antiallodynic effects of morphine in a rat model of neuro- bition of excitatory neurotransmitter release21 and pathic pain. Pain 1999; 80:31–6 22 8. Bruns RF, Fergus JH: Allosteric enhancement of adenosine A1 receptor postsynaptic inhibition of excitatory transmission. binding and function by 2-amino-3-benzoylthiophenes. Mol Pharmacol 1990; These mechanisms, at least in part, may be involved in 38:939–49 9. Musser B, Mudumbi RV, Liu J, Olson RD, Vestal RE: Adenosine A1 receptor- the antinociceptive effects of T62 in the postoperative dependent and -independent effects of the allosteric enhancer PD 81,723. J Phar- pain model. In the current study, we tested T62 24 h macol Exp Ther 1999; 288:446–54 after paw incision, and it is likely that the potency and 10. Pan HL, Xu Z, Leung E, Eisenach JC: Allosteric adenosine modulation to reduce allodynia. ANESTHESIOLOGY 2001; 95:416–20 efficacy of T62 changes in a complex fashion over time 11. Li X, Conklin D, Ma W, Zhu X, Eisenach JC: Spinal noradrenergic activation after surgery. mediates allodynia reduction from an allosteric adenosine modulator in a rat model of neuropathic pain. Pain 2002; 97:117–25 These results add to an emerging literature to suggest 12. Li XH, Conklin D, Pan HL, Eisenach JC: Allosteric adenosine receptor that postoperative pain exhibits a unique pharmacology modulation reduces hypersensitivity following peripheral inflammation by a central mechanism. J Pharmacol Exp Ther 2003; 305:950–5 of analgesia compared with other sustained pain models. 13. Brennan TJ, Vandermeulen EP, Gebhart GF: Characterization of a rat For example, although spinal N-methyl-D-aspartate recep- model of incisional pain. Pain 1996; 64:493–501 tor antagonists attenuate hypersensitivity in most models 14. Gomes JA, Li XH, Pan HL, Eisenach JC: Intrathecal adenosine interacts with a spinal noradrenergic system to produce antinociception in nerve-injured rats. of persistent pain, these are not effective in this postop- ANESTHESIOLOGY 1999; 91:1072–9 erative pain model.23 Surprisingly, intrathecal adminis- 15. Bantel C, Li XH, Eisenach JC: Intraspinal adenosine induces spinal cord norepinephrine release in spinal nerve-ligated rats but not in normal or sham tration of non–N-methyl-D-aspartate receptor antago- controls. ANESTHESIOLOGY 2003; 98:1461–6 nists24 and neurokinin-1 receptor antagonists25 are 16. Yaksh TL, Rudy TA: Chronic catheterization of the spinal subarachnoid space. Physiol Behav 1976; 7:1032–6 effective in this model. Further, hypersensitivity after 17. Chaplan SR, Bach FW, Pogrel JW, Chung JM, Yaksh TL: Quantitative incisional surgery is suppressed by intrathecal cycloox- assessment of tactile allodynia in the rat paw. J Neurosci Meth 1994; 53:55–63 ygenase 1 but not by cyclooxygenase 2 inhibitors.26 18. Tallarida RJ, Porreca F, Cowan A: Statistical analysis of drug-drug and site-site interactions with isobolograms. Life Sci 1989; 45:947–61 These observations suggest that mechanisms of hyper- 19. Ma WY, Eisenach JC: Chronic constriction injury of sciatic nerve induces sensitivity after incision differ those after inflammatory the up-regulation of descending inhibitory noradrenergic innervation to the lumbar dorsal horn of mice. Brain Res 2003; 970:110–8 or peripheral nerve injury. 20. Aran S, Proudfit HK: Antinociceptive interactions between intrathecally Interactions between T62 and clonidine may also be administered ␣ noradrenergic agonists and 5'-N-ethylcarboxamide adenosine. Brain Res 1990; 519:287–93 important for practical reasons. Clonidine is effective in 21. Mauborgne A, Poliénor H, Hamon M, Cesselin F, Bourgoin S: Adenosine 27 nerve injury–induced hypersensitivity and is approved receptor-mediated control of in vitro release of pain-related neuropeptides from for treatment of chronic neuropathic pain. Because the rat spinal cord. Eur J Pharmacol 2002; 441:47–55 22. Li J, Perl ER: Adenosine inhibition of synaptic transmission in the substan- clonidine therapy is limited by hypotension in some tia gelatinosa. J Neurophysiol 1994; 72:1611–21 patients with chronic pain,28 the current observation of 23. Zahn PK, Brennan TJ: Lack of effect of intrathecally administered N- methyl-D-aspartate receptor antagonists in a rat model for postoperative pain. synergistic interactions between clonidine and T62 sug- ANESTHESIOLOGY 1998; 88:143–56 gests that the clonidine dose, and potentially its adverse 24. Zahn PK, Umali E, Brennan TJ: Intrathecal non-NMDA excitatory amino acid receptor antagonists inhibit pain behaviors in a rat model of postoperative effects, could be reduced by the addition of T62. pain. Pain 1998; 74:213–23 In summary, intrathecal administration of T62 pro- 25. Yamamoto T, Sakashita Y: The role of the spinal opioid receptor like1 duces antinociception in the rat postoperative pain receptor, the NK-1 receptor, and cyclooxygenase-2 in maintaining postoperative pain in the rat. Anesth Analg 1999; 89:1203–8 model. By isobolographic analysis, T62 interacts syner- 26. Zhu X, Conklin D, Eisenach JC: Cyclooxygenase-1 in the spinal cord plays gistically with intrathecal clonidine and is only partially an important role in postoperative pain. Pain 2003; 104:15–23 27. Yaksh TL, Pogrel JW, Lee YW, Chaplan SR: Reversal of nerve ligation- inhibited by idazoxan. These results indicate that the induced allodynia by spinal ␣-2 adrenoceptor agonists. J Pharmacol Exp Ther mechanisms of antinociception by T62 in the postoper- 1995; 272:207–14 28. Eisenach JC, DuPen S, Dubois M, Miguel R, Allin D, Epidural Clonidine ative pain model differ from those in the neuropathic Study Group: Epidural clonidine analgesia for intractable cancer pain. Pain 1995; pain model. 61:391–9

Anesthesiology, V 100, No 5, May 2004