PPuurrppoossee::

SSoouurrccee:: !"#$%&" ' (

))

)

) ) ) *)

PPrriinncciippaall ffiinnddiinnggss::

+

"

* *

,

-&'%#. / &!# ./0

CCoonncclluussiioonn::

+* *./ ! *

Objectif :

Sources : !" "#$%&$

'(

((

( ) ( *

((

Constatations principales : + +

" + ++

++

+

""

#,

++

-

++ "

.%&/ 0+

1213

. 0&2 "

Conclusion :

+

"++

EW sensations are as disturbing to the indi-vidual as that of pain. It stands alone amongmans sensations, because it is accompaniedby strong psychological and emotional com-

ponents. This is recognized by the InternationalAssociation for the Study of Pain (IASP), which

1000 REGIONAL ANESTHESIA AND PAIN

CAN J ANESTH 2001 / 48: 10 / pp 10001010

Preemptive analgesia I: physiological pathwaysand pharmacological modalities[Lanalgsie prventive I : mcanismes physiologiques et modalits

pharmacologiques]

Dermot J. Kelly MRCPI FFARCSI,* Mahmood Ahmad MD, Sorin J. Brull MD

From the Department of Anaesthesia,* Cork University Hospital, Wilton, Cork, Ireland, and the Department of Anesthesiology,University of Arkansas for Medical Sciences, Little Rock, Arkansas, USA.

Address correspondence to: Dr. Sorin J. Brull, Professor and Chairman, Department of Anesthesiology, University of Arkansas forMedical Sciences, 4301 W. Markham Street, Slot 515, Little Rock, AR 72205-7199, USA. Phone: 501-686-6119; Fax: 501-603-1421; E-mail: Sorin.Brull@uams.edu

Accepted for publication November 2, 2000.Revision accepted July 11, 2001.

F

Kelly et al.: PREEMPTIVE ANALGESIA (PART 1 OF 2) 1001

defines pain as an unpleasant sensory and emotionalexperience associated with actual or potential tissuedamage or described in terms of such damage.

At the beginning of the last century, Crile wasamong the first to introduce the concept of treatingpain prior to its onset: preemptive analgesia.1,2 Heobserved that if pain transmission was blocked prior tothe initial surgical incision, postoperative mortalitywas decreased. This analgesic technique was proposedinitially as a means for preventing postoperative shock;however, proponents of this technique, later termedpreemptive analgesia, also noted a marked decrease inthe intensity (and duration) of postoperative pain.

The last 20 years have seen significant scientificadvancement in our understanding of the physiology,pathophysiology and pharmacology of pain. In con-junction with this knowledge, there has been a resur-gence of the concept of preemptive analgesia andnumerous studies have addressed the purported bene-fits of this technique in the surgical patient. Thisreview summarizes current knowledge of the painpathways, mechanisms and neurotransmitters, anddescribes new aggressive approaches to acute painmanagement in the context of preemptive analgesia.

PPhhyyssiioollooggiicc ppaatthhwwaayyssSpecialized receptors provide information to the centralnervous system (CNS) about the state of the environ-ment in the vicinity of the organism. Each receptor isspecialized to detect a particular type of stimulus (e.g.,touch, temperature, pain, etc.) Those receptors in theskin and other tissues that sense pain are free nerve end-ings, while those for temperature detection can be freenerve endings, bulbs of Krouse or Ruffinis corpuscles.Receptors are distributed with varying densities in dif-ferent tissues. Pain receptors may be stimulated bymechanical damage, extremes of temperature, or by irri-tating chemical substances. While certain pain receptorsare responsive to only one of the above stimuli, mostcan be stimulated by two or more. When the painreceptors in peripheral tissues (such as skin) are stimu-lated, the nociceptive (pain) impulses are transmitted tothe CNS by two distinct types of neurons - the A-deltaand C nerve fibres. The A-delta fibres are large-diame-ter, fast conducting myelinated fibres, which transmitfirst pain - sharp, prickling, and injurious. The Cfibres are small-diameter, slower conducting unmyeli-nated fibres that are responsible for second pain -dull, aching and visceral type. The primary afferent sen-sory neurons from the periphery then enter the spinalcord and synapse with neurons in the dorsal horn. Thesecond-order neurons, arising from the dorsal horn,have long axons that decussate in the anterior commis-

sure and travel cephalad in the contralateral anterolater-al pathway (also known as spinothalamic tract). Some ofthe long axons that synapsed with type C neurons donot decussate, but pass cranially in the ipsilateral antero-lateral spinal pathway. The anterolateral spinal pathwayfibres terminate in the thalamus, from which neuronalrelays are sent to other CNS centres and the sensorycortex. These higher centres are responsible for the per-ception of pain and the emotional components thataccompany it.

There are four distinct processes in the sensorypathway: transduction, transmission, modulation andperception (Figure 1). Each of these processes pre-sents a potential target for analgesic therapy; thereforetheir physiology is described in some detail below.

TransductionNociceptors, the pain receptors, respond selectively tonoxious stimuli and convert chemical, mechanical, orthermal energy at the site of the stimulus into neuralimpulses, a process known as transduction (Figure 2).The primary afferent nociceptors are the terminalbranches of the A-delta and C fibres, whose cell bodiesare located in the dorsal root ganglia. Mendell3described a functional classification of nociceptive nervefibres. Wide-dynamic range (WDR) neurons are thosethat receive input from both noxious and non-noxious

FIGURE 1 Diagrammatic representation of the four processesinvolved in the sensory pathway: transduction, transmission, per-ception, and modulation. Primary afferent neurons transmit infor-mation from the periphery to the dorsal horn of the spinal cord.Afferent information is then transmitted via the spinothalamictracts by second-order neurons to the thalamus and to the sensorycortex. The descending inhibitory fibres (interrupted lines) modu-late the afferent input at the dorsal horn. Also represented are theagents that can modify the sensory input of each of the fourprocesses.

stimuli and that exhibit a graded response (i.e., they canincrease their firing rate in response to increasing levelsof stimulation); the high-threshold (HT) neurons arethose that are activated only by noxious (HT) stimula-tion; and low-threshold (LT) neurons are those activat-ed only by non-noxious (LT) stimulation.

When the A-delta and C fibres are activated by briefintense stimuli, with little accompanying tissue damage,the resulting transient pain serves as a physiologicalwarning. However, when nociceptors are activated bythe noxious stimuli that accompany tissue damage or

infection, a regional injury response occurs in theperiphery. Chemical substances and enzymes arereleased from the damaged tissues, increasing the trans-duction of painful stimuli. Prostanoids (prostaglandins,leukotrienes and hydroxyacids) are products of thearachidonic acid pathway and are major mediators ofthe hyperalgesia that accompanies inflammation.Prostaglandins (PGs) and leukotrienes cause sensitiza-tion of the peripheral receptors, reducing their activa-tion threshold and increasing their responsiveness toother stimuli.46 Kinins, such as bradykinin and kallidinhave numerous pro-inflammatory functions including:release of PGs, cytokines and free radicals from a varietyof cells; degranulation of mast cells and release of hista-mine; and stimulation of sympathetic neurons to alterblood vessel caliber.7 Kinins also contribute to plasmaextravasation by producing contraction of vascularendothelial cells.8 Bradykinin and PGs, particularlyPGE2, stimulate neurons directly, initiating the trans-mission of pain impulses along the nociceptive pathway.Peripheral vascular dilation and increased vascular per-meability are induced by the release of substance Pcaused by the axon reflex of the injured nerve.9 Thisincreased vascular permeability, accompanied by releaseof vasoactive mediators from mast cells, results in aninflammatory response (neurogenic edema).10 Theincreased vascular permeability also results in extravasa-tion of additional algogenic (pain producing) sub-stances, such as histamine and serotonin. Histamine canalso be released from mast cells during degranulation, aprocess promoted by substance P, kinins, interleukin-1and nerve growth factor. Histamine acts on sensoryneurons to produce pain and itching.11 Histamine stim-ulation of sensory neurons may evoke release of neu-ropeptides and PGs, leading to further inflammatoryeffects and hyperalgesia.12 Serotonin (5-HT) is a majorinflammatory mediator, especially in the initial phases ofthe inflammatory response.13,14 Released from mastcells and platelets during injury or inflammation, 5-HTcauses direct activation of sensory neurons via 5-HTtype 3 (5-HT3) receptor activation. Nociceptive impuls-es activating the sympathetic nervous system promotenorepinephrine release, which in turn accelerates sensi-tization of the nociceptors, creating another viciouscycle.8 Reactive oxygen species, such as hydrogen per-oxide, superoxide, and hydroxyl species, are producedby tissues during inflammation. These substances havebeen shown to enhance the effects of bradykinin, PGE2and other inflammatory mediators,15 similar to the syn-ergism between other algogenic substances, such asPGs, bradykinin, and 5-HT.8

In summary, the activity and sensitivity of sensoryneurons is profoundly altered by the mediators

1002 CANADIAN JOURNAL OF ANESTHESIA

FIGURE 2 Representation of the transduction process and themediators of inflammatory processes that lead to peripheral sensiti-zation of nociceptors.

FIGURE 3 Representation of the transmission process by prima-ry afferent (A-delta and C) fibres from periphery to the dorsalhorn of the spinal cord. The balance between excitatory andinhibitory transmitter release determines the intensity of afferentinformation and the state of sensitization that occurs followingperipheral injury.

released as a result of tissue injury and inflammation(Figure 2). These inflammatory mediators produce anincrease in nociceptor sensitivity, neurogenic edemaand hyperalgesia of tissues in the vicinity of the injury.These complex changes in peripheral signal processingresult in increased pain sensation, alteration in thequality and duration of pain, and may lead to alteredcentral pain processing and the development of chron-ic pain states.

Transmission in the dorsal hornWhen signal transduction has occurred, impulses aretransmitted via A-delta and C fibres to the dorsal hornof the spinal cord (Figure 3). The nerve fibres synapsein the superficial layers of Rexed laminae: the A-deltaneurons synapse in laminae I, II and V, and the Cfibres in laminae I and II. The borders between theselaminae are not distinct. In addition, there is consid-erable overlap of neuronal cell types between the lam-inae, with each lamina containing more than one typeof neuron.16 A variety of neurotransmitters arereleased by the incoming first order nociceptive neu-rons. One of these is substance P, a neurokinin, whichis released from HT fibres. The calcitonin gene-relat-ed peptide (CGRP) is released along with substanceP,17,18 and extends the spinal cord zone from whichsubstance P is released, thereby contributing toincreased excitability.19 In turn, substance P inducesthe release of excitatory amino acids (EAAs) such asaspartate and glutamate, which act on the AMPA (2-amino-3-hydroxy-5-methyl-4-isoxazole-propionicacid) and NMDA (N-methyl-D-aspartate) receptors.20Enhanced synaptic transmission due to release ofEAAs follows substance P release21 and the latter caninduce a prolonged enhancement of responses by dor-sal horn neurons to glutamate or NMDA.22,23 Thisenhanced depolarization causes calcium influx intopostsynaptic neurons, which induces persistentchanges in the excitability of the cell.8 The termwind-up has been used to describe the enhancedexcitability and sensitization of dorsal horn cellsinduced by the above mechanisms (Figure 3).

In addition to causing wind up, repeated noxiousstimulation of the dorsal horn may result in an increase inthe number of neurons in laminae I and II whose nucleiexpress C-fos protein, a protein thought to be involved inthe memory of pain.24,25 Pretreatment with morphine hasbeen shown to decrease the number of cells expressing C-fos protein. This suggests that preventing access of thetrigger signal to the CNS may attenuate the increasedsensitivity to painful stimuli, and reduce the hyperalgesiaand production of pain by non-painful stimuli (i.e., allo-dynia), which accompany tissue injury.24,25

PerceptionThe second order nociceptive afferent fibres have theircell bodies in the dorsal horn of the spinal cord, fromwhich they project axons to the higher CNS centresresponsible for processing of nociceptive information.As mentioned previously, most of the ascending fibresdecussate before travelling cranially in the spinothala-mic tract. The majority of the neurons comprising thespinothalamic tract are WDR or HT neurons;26 theycourse through the pons, medulla and mid-brain toterminate in specific portions of the thalamus. Fromthe thalamus, afferent information is carried to thesomatosensory cortex. The spinothalamic tract alsosends collateral branches to the reticular formation.The impulses transmitted via these tracts are responsi-ble for the sensory discrimination of pain and theemotional responses it evokes. The reticular formationis probably responsible for the increased arousal andaspects of the emotional-affective components of pain,as well as somatic and autonomic motor reflexes.2729The activation of supraspinal structures is mediated byEAAs,30 but the neurotransmitters involved in centralprocessing of nociceptive information have not yetbeen elucidated.

Analgesic therapy has traditionally targeted the painperception component of the analgesic pathway.Certain areas, such as the nucleus reticularis giganto-cellularis, one of the nuclei in which nociceptive sec-ond-order neurons terminate, are profoundly depressedby both general anesthetics and opioid analgesics.3134Despite this fact, traditional analgesic therapy using opi-oids has met with varying success due, in part, to thelack of binding specificity of parenteral and oral opioids.With the increasing understanding of the pain pathwaysand the processes involved therein, it is now recognizedthat pain is best controlled by using several analgesicagents, each of which acts on a specific site along thepain pathway. Such an approach also lessens the relianceon one particular agent or mechanism, and the result-ing synergism may avoid the side effects associated withhigh doses of individual agents.

Efferent pathways and pain modulationIn the early twentieth century, Sherrington35 empha-sized the importance of the interaction between excita-tory and inhibitory neuronal systems in the processingof incoming sensory information by the brain. It is nowknown that efferent pathways help to modify afferentnociceptive information. The efferent neuronal path-ways involved in pain modulation include: the corti-cospinal tracts, which commence in the motor cortexand synapse in Rexed laminae IIIIV; hypothalamicefferents, which arise in the hypothalamus and synapse

Kelly et al.: PREEMPTIVE ANALGESIA (PART 1 OF 2) 1003

in the mid-brain, pons, medulla and Rexed lamina I;and extensive efferent fibres from the periaqueductalgray matter in the mid-brain and the nucleus raphemagnus in the medulla, to the dorsal horn. Stimulationof these efferent (descending) pathways can modulatenociceptive transmission in the periphery, in the spinalcord by altering neurotransmitter release, orsupraspinally by activation of inhibitory pathways(Figure 1). It is well established that norepinephrine,serotonin and opiate-like substances (endorphins) areinvolved in the brainstem inhibitory pathways thatmodulate pain in the spinal cord.3638

Gamma-amino butyric acid (GABA) and glycineare two important inhibitory neurotransmitters thatact at the dorsal horn. Blockade of spinal GABA orglycine can result in allodynia, by removing inhibitorsthat control NMDA receptors.39 Failure of spinal inhi-bition may thus play a role in the etiology of neuro-pathic pain. Alternatively, when there is peripheralinflammation, the opposite effect can be seen: up-reg-ulation of spinal GABA receptors promotes inhibitionof afferent nociceptive impulses40 and decreased painsensation. Therefore, the sensitivity of spinal GABAreceptors can vary under different circumstances,resulting in modulation of nociceptive information.

Another neurotransmitter, somatostatin, is found incells of the dorsal root ganglion and in afferent terminalsof the dorsal horn of the spinal cord. It appears to bereleased in response to noxious stimuli, resulting inhyperpolarization and a reduced firing rate in dorsal hornneurons.41 However, although it appears to have benefi-cial analgesic properties, intrathecal administration ofsomatostatin can also result in motor dysfunction andparalysis at doses just above those which produce analge-sia. Thus, further studies are required to fully evaluatethe role of this peptide in antinociception.41

Galanin is found in a large percentage of primaryafferent nociceptive fibres, and is thought to be aninhibitory peptide. It is frequently co-localized withsubstance P and CGRP. Until development of anantagonist, however, the exact role of galanin in noci-ceptive transmission is unlikely to be elucidated.40

Mechanisms of hypersensitivityThe transmission of pain from peripheral tissues viathe spinal cord to the brain is not a simple process,with its own exclusive pathways; rather, it is dependenton the balance between excitatory and inhibitory neu-ronal systems.

Peripheral tissue injury can modify the responsivenessof the nervous system to stimuli at two sites: peripheralsensitization, which entails a reduction in the thresholdof nociceptive afferent peripheral terminals; and central

sensitization, an activity-dependent increase in theexcitability of spinal neurons.42 These two processes con-tribute to the postinjury state of hypersensitivity seenpostoperatively. This state manifests as an increase in theresponse to noxious stimuli and a decrease in the painthreshold. It is now apparent that the receptive field ofdorsal horn cells is not fixed, and can undergo a numberof changes. These changes include alterations in the sizeand location of the peripheral receptive field (spatialcomponent); changes in the sensitivity to different stim-uli (threshold component); changes in the selectivity ofthe receptor to mechanical, thermal or chemical stimuli(modality sensitivity component); or a change in activityof the receptor in relation to the timing of the stimulus(temporal component).43 Primary hyperalgesia refers toreceptor field changes within the area of injury, while sec-ondary hyperalgesia refers to changes in the undamagedtissue surrounding the area of injury. In addition, thesechanges are responsible for the misperception of pain inresponse to non-noxious stimuli - termed allodynia.42Primary hyperalgesia is explained by sensitization ofperipheral nociceptors,4446 while secondary hyperalgesiamay be caused by altered CNS processing of mechanore-ceptor impulses from peripheral tissues.45,4750

Clinical pain may be divided into two entities:inflammatory pain, which is the consequence of trau-ma to peripheral tissues (i.e., surgical incision, dissec-tion, burns, etc); and neuropathic pain, which is theresult of direct injury to nervous tissue (i.e., nervetransection). Both types of injury result in long-termchanges in the sensitivity of the nervous system, suchthat the intensity of subsequent stimuli necessary toinduce pain is reduced.

PPhhaarrmmaaccoollooggiiccaall mmooddaalliittiieessPreemptive analgesia is an antinociceptive therapywhose aim is to prevent both peripheral and centralsensitization, thereby attenuating (or, ideally, prevent-ing) the postoperative amplification of pain sensation.Treatment can be aimed at the periphery, at inputsalong sensory axons, or at CNS sites using single orcombinations of analgesics applied either continuous-ly or intermittently.

Regional anesthesiaAfferent blockade of all impulses, and in particular noci-ceptive impulses, prior to incision is central to the con-cept of preemptive analgesia. This can be achieved withlocal infiltration for superficial procedures, peripheralnerve or plexus anesthesia, or central neuraxial block-ade. Of importance is the fact that effective analgesiashould not only be well established prior to the surgicalincision but should be continued well into the postop-

1004 CANADIAN JOURNAL OF ANESTHESIA

erative period. The timing to stop delivery of the localanesthetic agent should be judged against the healing ofthe wound and degree of anticipated tenderness.Discontinuation of regional blockade when significantafferent input is still likely to be present would merelydelay the onset of surgical pain until after the pharma-cologic effects of the local anesthetic subside. Inpatients with preexisting central sensitization, theregional blockade should be continued for a longerduration in order to achieve a lower level of nociceptiveinput. Often, regional anesthesia is combined withadjuvant therapy in the postoperative period.

Non-steroidal anti-inflammatory drugs (NSAIDs)Peripheral sensitization, in which there is an increasein the sensitivity of HT peripheral sensory neurons,results from exposure of sensory nerve terminals toalgogenic substances and mediators released locally atthe site of injury. The aim of treatment should be toprevent the release (or to inactivate) the various neu-rotransmitters and inflammatory mediators, whichsensitize the peripheral nociceptors. By reducing PGsynthesis, cyclo-oxygenase inhibitors block the noci-ceptive response to endogenous mediators of inflam-mation, with the effect being greatest in tissues thathave been subjected to injury and trauma.51 NSAIDsrepresent diverse chemical entities, but their commonmechanism of action is inhibition of this PG-mediatedsensitization of nociceptors to chemical and mechani-cal irritants.52 Moreover, membrane stabilization byNSAIDs may account for the decrease in PG releaseseen at concentrations insufficient for effective inhibi-tion of cyclo-oxygenase.51 Glucocorticoids, tricyclicantidepressants, anti-arrhythmics and local anestheticsmay work by a similar mechanism, as all are consideredto possess membrane-stabilizing effects.

Central neuronal tissues also synthesize PGs, andspinally administered NSAIDs have been shown toreduce hyperalgesia. To what extent the central actionof NSAIDs contributes to the analgesic effect of sys-temically administered NSAIDs is unknown.53 It isapparent that in addition to their effects on PG syn-thesis, certain NSAIDs also affect the synthesis andactivity of other neuroactive substances, such as 5-HT,kynurenic acid and polyamines, thought to playimportant roles in the processing of nociceptiveimpulses in the dorsal horn.54

OpioidsAt the spinal cord, modulation of afferent input can beaccomplished by decreasing neurotransmitter release,by blocking the postsynaptic receptors (thereby block-ing the effects of the neurotransmitters), or by activat-

ing inhibitory pathways. Opioid receptors are a keysite of analgesia production and recent studies indicatethat spinal opioid systems can be enhanced or reducedunder different circumstances.40 The lamina I and thesubstantia gelatinosa in the dorsal horn, the zones inwhich C-fibres terminate, have the highest concentra-tions of opioid receptors in the spinal cord.Approximately 70% of the total opioid receptors in therat spinal cord are of the mu-subtype, approximately24% are delta, and approximately 6% are kappa recep-tors.55 Whether this situation mirrors that in humansis under investigation, but molecular biology hasdemonstrated close similarities between opioid recep-tors in humans and those in laboratory animals.56 Themajority of the mu-receptors in the spinal cord arefound presynaptically on the afferent nociceptive ter-minals.55 Opioids that are mu and/or delta agonistscause a reduced release from C-fibres of primary affer-ent neurotransmitters (substance P and gluta-mate).36,57,58 Opioids also inhibit the release ofCGRP.59 The predominance of presynaptic opioidreceptors on C fibres, as opposed to A-fibre terminals,accounts for the selective effect of spinal opioids onnoxious evoked activity.60 The deeper layers of thespinal cord contain relatively fewer opioid receptors;those present are believed to be situated on nocicep-tive circuitry such that they have selective inhibitoryeffects. When stimulated, these postsynaptic opioidreceptors hyperpolarize the membrane of dorsal hornneurons, thereby reducing activity in nociceptive path-ways.57,61 Studies with agonists of the delta opioidreceptors have highlighted their potential as analgesicspossessing fewer of the side effects usually associatedwith morphine. However, experience with kappareceptor agonists, to date, has been disappointing.62

Supraspinal injections of mu and delta agonists willproduce naloxone-sensitive analgesia,63 although theexact mechanism by which mu receptor activationproduces supraspinal analgesia is still unknown. Theexact role of the kappa receptors in the production ofsupraspinal analgesia is the subject of debate, withvarying opinions ventured in the literature.61,64,65

Opioids also act peripherally as analgesic agents.Mu receptor agonists prevent the nociceptor sensitiza-tion induced by inflammatory mediators, such asprostaglandin E2 (PGE2).

66 Delta and kappa receptorsare thought to be located on the sympathetic nervesand to mediate analgesia peripherally by blockingbradykinin-induced release of nociceptor sensitizingagents from nerve endings.67 Thus, opioids actsupraspinally, spinally and peripherally to produceanalgesia, thereby reducing sensitization both central-ly and peripherally.

Kelly et al.: PREEMPTIVE ANALGESIA (PART 1 OF 2) 1005

NMDA receptor antagonistsThere are large numbers of NMDA receptors in thehuman spinal cord;40 the conditions necessary for theirstimulation are complex and appear to only beachieved with repetitive C-fibre activity.68,69 When theC-fibre stimulus is maintained and its frequency orintensity is sufficient, the NMDA receptor becomesactivated and the resultant amplification or prolonga-tion of the response seems to underlie many forms ofcentral hyperalgesia.24,6973 Prolonged inflammatorypain, unlike its acute pain counterpart, is sensitive toNMDA antagonism.74 Since the NMDA receptorshave been implicated in pathological pain states,NMDA antagonists, such as ketamine or dex-tromethorphan, have been used to treat opioid insen-sitive neuropathic and cancer pains. NMDAantagonists have no effect on the afferent input ontodorsal horn, but they abolish the wind-up phenome-non, thereby converting the potentiated nociceptiveresponse to a normal response.75 Opioids, by contrast,reduce the release of neurotransmitters by presynapticC-fibres through binding to inhibitory receptors onthe C-fibres. While opioids will delay the onset ofwind-up by this mechanism, unless the dose is suffi-cient to stop all neurotransmitter release, wind-upmay still occur.60 Opioids and NMDA antagonists maybe used synergistically, and the combination hasshown marked inhibition of nociceptive responses.Ketamine is an NMDA receptor antagonist at sub-anesthetic doses and therefore has the capability toblock central hypersensitivity states at doses that arenot directly analgesic. Spinally administered localanesthetics also work synergistically with morphine inmodulating nociception by blocking afferent fibresand reducing neuronal excitability, thereby reducingNMDA-mediated activity.76

Alpha2 receptor antagonistsAnalgesia may also be produced by stimulation ofalpha2 adrenergic receptors in both the spinal cord andhigher centres.77 These receptors may be activated bydescending noradrenergic pathways or by exogenouscompounds, such as epinephrine, clonidine ordexmedetomidine. Studies indicate that alpha2 ago-nists exert a potent analgesic response,78,79 and thattheir potency is increased by concomitant opioid ther-apy.8082 While alpha2 agonists and opioids mediatetheir analgesic action through independent receptors,cross-tolerance may nonetheless develop between thetwo agents.83,84 Alpha2-agonists also have been shownto reduce the undesirable physiological and psycho-logical effects of opioid withdrawal.85 The exact mech-anism by which alpha2 agonists produce analgesia

remains unknown, though it is postulated that releaseof acetylcholine may play a role.86,87

Miscellaneous agentsNumerous other agents have been shown to haveanalgesic properties and may have a role in producingpreemptive analgesia, either singularly or in combina-tion therapy.

Cholecystokinin (CCK) can selectively reduce theanalgesic action of morphine at both spinal andsupraspinal sites. It appears that CCK acts as anendogenous control on the level of mu analgesia.Upregulation of its receptors results in decreased anal-gesia in some neuropathic models; decreased CCKconcentrations, in inflammatory models, results inenhancement of mu receptor effects.88 Alternatively,antagonists of the CCK receptor may enhance theanalgesic action of opioids.89,90

It has been shown that NMDA receptor activationresults in formation of nitric oxide (NO). Inhibitors ofNO synthetase, the enzyme responsible for synthesisof NO, are antinociceptive, suggesting a possiblefuture role in pain medicine.91 While it has beenshown that GABA agonists also have antinociceptiveactions,92 it remains to be determined whether theGABA receptor represents a viable target for futureanalgesic therapy.

Administration of antagonists to bradykinin, hista-mine and serotonin may be used to prevent or diminishthe formation of neurogenic edema and sensitization ofreceptors.93 However, the complexity of the peripheralneural mechanisms leading to hyperalgesia is not yetfully understood.70,94 The role of newly discovered andlonger established peripheral mediators such as thenerve growth factors, cytokines and catecholaminesfrom sympathetic nerve endings, as well as bradykinin,serotonin and prostanoids, awaits elucidation.70,94Accordingly, attention has focused on the CNS in thesearch for alternative processes that can be modifiedpharmacologically, in combination with peripherallyacting measures, to improve analgesia. Monoaminereuptake inhibitors can enhance the antinociceptionproduced by systemic opioids.95,96 Intrathecally admin-istered anticholinesterases, such as neostigmine, canproduce potent analgesia97 and can synergisticallyenhance the antinociception of both morphine andclonidine.98,99 Acetylcholinesterases appear to exerttheir analgesic action through a muscarinic, as opposedto a nicotinic, action (muscarinic agonists, but not nico-tinic agonists, are analgesic when injected intrathecal-ly).100,101 Amitryptiline, a tricyclic antidepressant, canenhance the antinociceptive effects of systemic opioids96and of intrathecal neostigmine.98 Potassium channel

1006 CANADIAN JOURNAL OF ANESTHESIA

openers, such as diazoxide and minoxidil, have alsobeen shown to produce antinociception when adminis-tered intrathecally.102 Intrathecal administration ofadenosine analogues produces antinociception; thiseffect can be blocked by the administration of adeno-sine receptor antagonists, such as methylxan-thines.103,104 What role, if any, these agents will play inthe treatment of pain requires further clinical investiga-tion and will probably form a large portion of futureresearch in this area.

RReeffeerreenncceess1 Crile GW, Lower WE. Anoci-Association.

Philadelphia: Saunders, 1914: 2235.2 Kehlet H. General vs. regional anesthesia. In: Rogers

MC, Tinker JH, Covino BG, Longnecker DE (Eds.)Principles and Practice of Anesthesiology. St. Louis:Mosby, 1993: 121834.

3 Mendell LM. Physiological properties of unmyelinatedfiber projection to the spinal cord. Exp Neurol 1966;16: 31632.

4 Birrell GJ, McQueen DS, Iggo A, Coleman RA, GrubbBD. PGI2-induced activation and sensitization ofarticular mechanonociceptors. Neurosci Lett 1991;124: 58.

5 Levine JD, Fields HL, Basbaum AI. Peptides and theprimary afferent nociceptor. J Neurosci 1993; 13:227386.

6 Cohen RH, Perl ER. Contributions of arachidonicacid derivatives and substance P to the sensitizationof cutaneous nociceptors. J Neurophysiol 1990; 64:45764.

7 Walker K, Perkins M, Dray A. Kinins and kinin recep-tors in the nervous system. Neurochem Int 1995; 26:1726.

8 Dray A. Inflammatory mediators of pain. Br JAnaesth 1995; 75: 12531.

9 Suzuki H. Recent topics in the management of pain:development of the concept of preemptive analgesia.Cell Transplant 1995; 4(Suppl. 1): S36.

10 Lynn B. Neurogenic inflammation. Skin Pharmacol1988; 1: 21724.

11 Simone DA, Alreja M, LaMotte RH. Psychophysicalstudies of the itch sensation and itchy skin (allokne-sis) produced by intracutaneous injection of hista-mine. Somatosens Mot Res 1991; 8: 2719.

12 Rang HP, Bevan S, Dray A. Nociceptive peripheralneurons: cellular properties. In: Wall PD, Melzack R(Eds.). Textbook of Pain, 3rd ed. Edinburgh:Churchill Livingstone, 1994: 5778.

13 Capasso F, Dunn CJ, Yamamoto S, Willoughby DA,Giroud JP. Further studies on carrageenan-inducedpleurisy in rats. J Pathol 1975; 116: 11724.

14 Garcia Leme J, Hamamura L, Leite MP, Rocha e SilvaM. Pharmacological analysis of the acute inflammato-ry process induced in the rats paw by local injectionof carrageenin and by heating. Br J Pharmacol 1973;48: 8896.

15 Holthusen H, Arndt JO. Nitric oxide evokes pain inhumans on intracutaneous injection. Neurosci Lett1994; 165: 714.

16 Wall PD. The dorsal horn. In: Wall PD, Melzack R(Eds). Textbook of Pain, 2nd ed. Edinburgh:Churchill Livingstone; 1994: 10212.

17 Murase K, Randic M. Actions of substance P on ratspinal dorsal horn neurones. J Physiol (Lond) 1984;346: 20317.

18 Skofitsch G, Jacobowitz DM. Calcitonin gene-relatedpeptide coexists with substance P in capsaicin sensi-tive neurons and sensory ganglia of the rat. Peptides1985; 6: 74754.

19 Schaible H-G, Freudenberger U, Neugebauer V, StillerRU. Intraspinal release of immunoreactive calcitoningene-related peptide during development of inflam-mation in the joint in vivo a study with antibodymicroprobes in cat and rat. Neuroscience 1994; 62:1293305.

20 Schneider SP, Perl ER. Selective excitation of neuronsin the mammalian spinal dorsal horn by aspartate andglutamate in vitro: correlation with location and exci-tatory input. Brain Res 1985; 360: 33943.

21 Kangrga I, Larew JSA, Randic M. The effects of sub-stance P and calcitonin gene-related peptide on theefflux of endogenous glutamate and aspartate fromthe rat spinal dorsal horn in vitro. Neurosci Lett1990; 108: 15560.

22 Willcockson WS, Chung JM, Hori Y, Lee KH, WillisWD. Effects of iontophoretically released amino acidsand amines on primate spinothalamic tract cells. JNeurosci 1984; 4: 73240.

23 Dougherty PM, Willis WD. Enhancement ofspinothalamic neuron responses to chemical andmechanical stimuli following combined micro-ion-tophoretic application of N-methyl-D-aspartic acidand substance P. Pain 1991; 47: 8593.

24 Coderre TJ, Katz J, Vaccarino AL, Melzack R.Contribution of central neuroplasticity to pathologi-cal pain: review of clinical and experimental evidence.Pain 1993; 52: 25985.

25 Tolle TR, Castro-Lopes JM, Evan G, ZieglgansbergerW. C-fos induction in the spinal cord following nox-ious stimulation: prevention by opiates but not byNMDA antagonists. In: Bond MR, Charlton JE,Woolf CJ (Eds). Proceedings of the VIth WorldCongress on Pain. Elsevier Science BV, 1991:299305.

Kelly et al.: PREEMPTIVE ANALGESIA (PART 1 OF 2) 1007

26 Willis WD Jr. Ascending somatosensory systems. In:Yaksh TL (Ed.). Spinal Afferent Processing. NewYork: Plenum Press, 1986: 398416.

27 Melzack R, Casey KL. Sensory, motivational, and cen-tral control determinants of pain. A new conceptualmodel. In: Kenshalo DR (Ed.). The Skin Senses.Illinois: Thomas, 1968: 42339.

28 Bowsher D. Role of the reticular formation in respons-es to noxious stimulation. Pain 1976; 2: 36178.

29 Price DD, Dubner R, Hu JW. Trigeminothalamicneurons in nucleus caudalis responsive to tactile, ther-mal, and nociceptive stimulation of monkeys face. JNeurophysiol 1976; 39: 93653.

30 Jensen TS, Yaksh TL. Brainstem excitatory amino acidreceptors in nociception: microinjection mapping andpharmacological characterization of glutamate-sensi-tive sites in the brainstem associated with algogenicbehavior. Neuroscience 1992; 46: 53547.

31 Kikuchi H, Kitahata LM, Collins JG, Kawahara M,Nio K. Halothane-induced changes in neuronal activ-ity of cells of the nucleus reticularis gigantocellularisof the cat. Anesth Analg 1980; 59: 897901.

32 Mosso JA, Kruger L. Spinal trigeminal neurons excitedby noxious and thermal stimuli. Brain Res 1972; 38:20610.

33 Ohtani M, Kikuchi H, Kitahata LM, et al. Effects ofketamine on nociceptive cells in the medial medullaryreticular formation of the cat. Anesthesiology 1979;51: 4147.

34 Yuge O, Kitahata LM, Collins JG, et al. Fentanyl andalfentanil suppress brainstem pain transmission.Anesth Analg 1985; 64: 597600.

35 Sherrington SC. The Integrative Action of the NervousSystem. New York: C. Scribner & Son, 1906.

36 Besson J-M, Chaouch A. Peripheral and spinal mecha-nisms of nociception. Physiol Rev 1987; 67: 67186.

37 Evans RH. The pharmacology of segmental transmis-sion in the spinal cord. Prog Neurobiol 1989; 33:25579.

38 Fields HL, Heinricher MM, Mason P.Neurotransmitters in nociceptive modulatory circuits.Ann Rev Neurosci 1991; 14: 21945.

39 Yaksh TL. Behavioral and autonomic correlates of thetactile evoked allodynia produced by spinal glycineinhibition: effects of modulatory receptor systems andexcitatory amino acid antagonists. Pain 1989; 37:11123.

40 Dickenson AH. Spinal cord pharmacology of pain. BrJ Anaesth 1995; 75: 193200.

41 Mollenholt P, Post C, Rawal N, Freedman J, Hokfelt T,Paulsson I. Antinociceptive and neurotoxic actionsof somatostatin in rat spinal cord after intrathecaladministration. Pain 1988; 32: 95105.

42 Woolf CJ, Chong M-S. Preemptive analgesiatreatingpostoperative pain by preventing the establishment ofcentral sensitization. Anesth Analg 1993; 77: 36279.

43 Willis WD Jr, Coggeshall RE. Sensory Mechanisms ofthe Spinal Cord, 2nd ed. New York: Plenum, 1991.

44 Hardy JD, Wolff HG, Goodell H. Experimental evi-dence on the nature of cutaneous hyperalgesia. J ClinInvest 1950; 29: 11540.

45 LaMotte RH, Shain CN, Simone DA, Tsai EFP.Neurogenic hyperalgesia: psychophysical studies ofunderlying mechanisms. J Neurophysiol 1991; 66:190211.

46 Raja SN, Campbell JN, Meyer RA. Evidence for dif-ferent mechanisms of primary and secondary hyperal-gesia following heat injury to the glabrous skin. Brain1984; 107(Pt 4): 117988.

47 Baumann TK, Simone DA, Shain CN, LaMotte RH.Neurogenic hyperalgesia: the search for the primarycutaneous afferent fibers that contribute to capsaicin-induced pain and hyperalgesia. J Neurophysiol 1991;66: 21227.

48 LaMotte RH, Lundberg LER, Torebjork HE. Pain,hyperalgesia and activity in nociceptive C units inhumans after intradermal injection of capsaicin. JPhysiol (Lond) 1992; 448: 74964.

49 Simone DA, Sorkin LS, Oh U, et al. Neurogenichyperalgesia: central neural correlates in responses ofspinothalamic tract neurons. J Neurophysiol 1991;66: 22846.

50 Torebjork HE, Lundberg LER, LaMotte RH. Centralchanges in processing of mechanoreceptive input incapsaicin-induced secondary hyperalgesia in humans.J Physiol (Lond) 1992; 448: 76580.

51 Kitahata LM. Pain pathways and transmission. Yale JBiol Med 1993; 66: 43742.

52 Dahl JB, Kehlet H. Non-steroidal anti-inflammatorydrugs: rationale for use in severe postoperative pain.Br J Anaesth 1991; 66: 70312.

53 Malmberg AB, Yaksh TL. Antinociceptive actions ofspinal nonsteroidal anti-inflammatory agents on theformalin test in the rat. J Pharmacol Exp Ther 1992;263: 13646.

54 McCormack K. The spinal actions of nonsteroidalanti-inflammatory drugs and the dissociation betweentheir anti-inflammatory and analgesic effects. Drugs1994; 47(Suppl 5): 2845; Discussion 467.

55 Besse D, Lombard M-C, Zajac J-M, Roques BP, BessonJ-M. Pre- and postsynaptic location of mu, delta andkappa opioid receptors in the superficial layers of thedorsal horn of the rat spinal cord. Prog Clin Biol Res1990; 328: 1836.

56 Uhl GR, Childers S, Pasternak G. An opiate-receptorgene family reunion. Trends Neurosci 1994; 17: 8993.

1008 CANADIAN JOURNAL OF ANESTHESIA

57 Dickenson AH, Sullivan AF. Electrophysiologicalstudies on the effects of intrathecal morphine onnociceptive neurones in the rat dorsal horn. Pain1986; 24: 21122.

58 Kangrga I, Randic M. Outflow of endogenous aspar-tate and glutamate from the rat spinal dorsal horn invitro by activation of low- and high-threshold prima-ry afferent fibers. Modulation by mu-opioids. BrainRes 1991; 553: 34752.

59 Pohl M, Lombard MC, Bourgoin S, et al. Opioid con-trol of the in vitro release of calcitonin gene-relatedpeptide from primary afferent fibres projecting in therat cervical cord. Neuropeptides 1989; 14: 1519.

60 Dickenson AH. A cure for wind up: NMDA receptorantagonists as potential analgesics. Trends PharmacolSci 1990; 11: 3079.

61 Dickenson AH. Mechanisms of the analgesic actions ofopiates and opioids. Br Med Bull 1991; 47: 690702.

62 Calderon SN, Rothman RB, Porreca F, et al. Probesfor narcotic receptor mediated phenomena. 19.Synthesis of (+)-4-[(alpha R)-alpha-((2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)-3- methoxybenzyl]-N,N-diethylbenzamide (SNC 80): a highly selective,nonpeptide delta opioid receptor agonist. J MedChem 1994; 37: 21258.

63 Pleuvry BJ. Opioid receptors and their relevance toanaesthesia. Br J Anaesth 1993; 71: 11926.

64 Leighton GE, Rodriguez RE, Hill RG, Hughes J. k-opioid agonists produce antinociception after i.v. andi.c.v. but not intrathecal administration in the rat. BrJ Pharmacol 1988; 93: 55360.

65 Millan MJ. k-opioid receptors and analgesia. TrendsPharmacol Sci 1990; 11: 706.

66 Levine JD, Taiwo YO. Involvement of the mu-opiatereceptor in peripheral analgesia. Neuroscience 1989;32: 5715.

67 Taiwo YO, Levine JD. k- and delta-opioids block sym-pathetically dependent hyperalgesia. J Neurosci 1991;11: 92832.

68 Dickenson AH, Sullivan AF. Evidence for a role ofthe NMDA receptor in the frequency dependentpotentiation of deep rat dorsal horn nociceptive neu-rones following C fibre stimulation.Neuropharmacology 1987; 26: 12358.

69 Dickenson AH, Sullivan AF. Differential effects ofexcitatory amino acid antagonists on dorsal hornnociceptive neurones in the rat. Brain Res 1990; 506:319.

70 Dray A, Urban L, Dickenson A. Pharmacology ofchronic pain. Trends Pharmacol Sci 1994; 15: 1907.

71 Dubner R, Ruda MA. Activity-dependent neuronalplasticity following tissue injury and inflammation.Trends Neurosci 1992; 15: 96103.

72 McMahon SB, Lewin GR, Wall PD. Central hyperex-citability triggered by noxious inputs. Curr OpinNeurobiol 1993; 3: 60210.

73 Price DD, Mao J, Mayer DJ. Central neural mecha-nisms of normal and abnormal pain states. In: FieldsHL, Liebeskind JC (Eds.). PharmacologicalApproaches to the Treatment of Chronic Pain.Seattle: IASP Press, 1994: 6184.

74 Haley JE, Sullivan AF, Dickenson AH. Evidence forspinal N-methyl-D-aspartate receptor involvement inprolonged chemical nociception in the rat. Brain Res1990; 518: 21826.

75 Davies SN, Lodge D. Evidence for involvement of N-methylaspartate receptors in wind-up of class 2 neu-rones in the dorsal horn of the rat. Brain Res 1987;424: 4026.

76 Fraser HM, Chapman V, Dickenson AH. Spinal localanaesthetic actions on afferent evoked responses andwind-up of nociceptive neurones in the rat spinalcord: combination with morphine produces markedpotentiation of antinociception. Pain 1992; 49:3341.

77 Wang Y-C, Su C-F, Lin M-T. The site and the modeof analgesic actions exerted by clonidine in monkeys.Exp Neurol 1985; 90: 47988.

78 Fielding S, Wilker J, Hynes M, Szewczak M, Novick WJJr, Lal H. A comparison of clonidine with morphinefor antinociceptive and antiwithdrawal actions. JPharmacol Exp Ther 1978; 207: 899905.

79 De Kock M, Crochet B, Morimont C, Scholtes J-L.Intravenous or epidural clonidine for intra- and post-operative analgesia. Anesthesiology 1993; 79:52531.

80 Omote K, Kitahata LM, Collins JG, Nakatani K,Nakagawa I. Interaction between opiate subtype andalpha-2 adrenergic agonists in suppression of nox-iously evoked activity of WDR neurons in the spinaldorsal horn. Anesthesiology 1991; 74: 73743.

81 Wilcox GL, Carlsson K-H, Jochim A, Jurna I. Mutualpotentiation of antinociceptive effects of morphineand clonidine on motor and sensory responses in ratspinal cord. Brain Res 1987; 405: 8493.

82 Goyagi T, Nishikawa T. The addition of epinephrineenhances postoperative analgesia by intrathecal mor-phine. Anesth Analg 1995; 81: 50813.

83 Paalzow G. Development of tolerance to the analgesiceffect of clonidine in rats. Cross-tolerance to mor-phine. Naunyn Schmiedebergs Arch Pharmacol 1978;304: 14.

84 Stevens CW, Monasky MS, Yaksh TL. Spinal infusion ofopiate and alpha-2 agonists in rats: tolerance andcross-tolerance studies. J Pharmacol Exp Ther 1988;244: 6370.

Kelly et al.: PREEMPTIVE ANALGESIA (PART 1 OF 2) 1009

85 Gold MS, Redmond DE Jr, Kleber HD. Clonidineblocks acute opiate-withdrawal symptoms. Lancet1978; 2: 599602.

86 Gordh T Jr, Jansson I, Hartvig P, Gillberg PG, Post C.Interactions between noradrenergic and cholinergicmechanisms involved in spinal nociceptive processing.Acta Anaesthesiol Scand 1989; 33: 3947.

87 Detweiler DJ, Eisenach JC, Tong C, Jackson C. Acholinergic interaction in alpha 2 adrenoceptor-medi-ated antinociception in sheep. J Pharmacol Exp Ther1993; 265: 53642.

88 Stanfa LC, Sullivan AF, Dickenson AH. Alterations inneuronal excitability and the potency of spinal mu,delta and kappa opioids after carrageenan-inducedinflammation. Pain 1992; 50: 34554.

89 Stanfa L, Dickenson A, Xu XJ, Wiesenfeld-Hallin Z.Cholecystokinin and morphine analgesia: variationson a theme. Trends Pharmacol Sci 1994; 15: 656.

90 Zhou Y, Sun Y-H, Zhang Z-W, Han J-S. Increasedrelease of immunoreactive cholecystokinin octapep-tide by morphine and potentiation of mu-opioid anal-gesia by CCKB receptor antagonist L-365,260 in ratspinal cord. Eur J Pharmacol 1993; 234: 14754.

91 Moore PK, Wallace P, Gaffen Z, Hart SL, BabbedgeRC. Characterization of the novel nitric oxide syn-thase inhibitor 7-nitro indazole and related indazoles:antinociceptive and cardiovascular effects. Br JPharmacol 1993; 110: 21924.

92 Aanonsen LM, Wilcox GL. Muscimol, -aminobutyricacidA receptors and excitatory amino acids in themouse spinal cord. J Pharmacol Exp Ther 1989; 248:10348.

93 Burch RM, Farmer SG, Steranka LR. Bradykininreceptor antagonists. Med Res Rev 1990; 10:23769.

94 Dray A, Perkins M. Bradykinin and inflammatorypain. Trends Neurosci 1993; 16: 99104.

95 Taiwo YO, Fabian A, Pazoles CJ, Fields HL.Potentiation of morphine antinociception bymonoamine reuptake inhibitors in the rat spinal cord.Pain 1985; 21: 32937.

96 Botney M, Fields HL. Amitriptyline potentiates mor-phine analgesia by a direct action on the central ner-vous system. Ann Neurol 1983; 13: 1604.

97 Bouaziz H, Tong C, Eisenach JC. Postoperative anal-gesia from intrathecal neostigmine in sheep. AnesthAnalg 1995; 80: 11404.

98 Eisenach JC, Gebhart GF. Intrathecal amitriptyline.Antinociceptive interactions with intravenous morphineand intrathecal clonidine, neostigmine, and carbamyl-choline in rats. Anesthesiology 1995; 83: 103645.

99 Abram SE, Winne RP. Intrathecal acetyl cholinesteraseinhibitors produce analgesia that is synergistic with

morphine and clonidine in rats. Anesth Analg 1995;81: 5017.

100 Svensson BA, Sottile A, Gordh T Jr. Studies on thedevelopment of tolerance and potential spinal neuro-toxicity after chronic intrathecal carbachol-antinoci-ception in the rat. Acta Anaesthesiol Scand 1991; 35:1417.

101 Gillberg PG, Hartvig P, Gordh T, et al. Behavioraleffects after intrathecal administration of cholinergicreceptor agonists in the rat. Psychopharmacology(Berl) 1990; 100: 4649.

102 Welch SP, Dunlow LD. Antinociceptive activity ofintrathecally administered potassium channel openersand opioid agonists: a common mechanism of action?J Pharmacol Exp Ther 1993; 267: 3909.

103 Sawynok J, Sweeney MI, White TD. Classification ofadenosine receptors mediating antinociception in therat spinal cord. Br J Pharmacol 1986; 88: 92330.

104 Sosnowski M, Stevens CW, Yaksh TL. Assessment of therole of A1/A2 adenosine receptors mediating thepurine antinociception, motor and autonomic func-tion in the rat spinal cord. J Pharmacol Exp Ther1989; 250: 91522.

1010 CANADIAN JOURNAL OF ANESTHESIA