IT administration of clonidine with local anaesthetic (LA) improves the quality and duration of the LA block but may cause greater hypotension (Bonnet et al. 1989). Similarly, epidural clonidine (1–4 μg kg−1) with LA improves the quality and duration of the block but does not increase hypotension. Caudal administration (0.75–3 μg kg−1) with LA increases anaesthesia duration two- to threefold (Lee and Rubin 1994). This may be of especial utility in paediatric anaesthesia, as children appear less susceptible to adverse haemodynamic
changes (as blood pressure is less dependent on sympathetic tone at this age). Clonidine is also efﬁcacious for labour analgesia (epidural doses of 1 μgkg−1) with the epidural combination of 75 μg clonidine and 50 μg fentanyl more than doubling the duration of analgesia produced by bupivacaine alone (Celleno et al. 1995). Furthermore, clonidine is a commonly employed adjuvant for brachial plexus blocks as it signiﬁcantly prolongs the duration of anaesthesia
provided (Bernard and Macaire 1997).
Clinical Application: Chronic Pain
Epidural administration of clonidine has been associated with signiﬁcant anal- gesia for deafferentation pain post spinal cord injury (Glynn et al. 1986), spas- ticity (Rémy-Néris et al. 1999), chronic arachnoiditis (Glynn et al. 1988) and for chronic sharp and shooting pains (Byas-Smith et al. 1995). Furthermore,
in the treatment of intractable cancer pain, epidural clonidine (100–900 μg)
produces a dose-dependent analgesia lasting up to 8 h (Eisenach et al. 1989).
Two weeks of analgesia is also produced by continuous epidural infusion of
30 μg h−1 during which time the initial sedative effect disappears (Eisenach et al. 1995). Clonidine is now regarded a second line pharmacotherapy for the
treatment of cancer pain as an adjunct or an alternative to opioids.
Conclusion: α2 Adrenergic Agonists
In both the acute and chronic pain settings, α2 adrenergic agonists remain rela- tively underused. With the expanding role of dexmedetomidine in the intensive care setting (Coursin et al. 2001) further information about the analgesic ac- tion of this class of agents will become available. A recent study of the systemic administration of dexmedetomidine in human volunteers found signiﬁcant analgesia against thermal-induced pain (Cortinez et al. 2004), although the modality of experimental pain inﬂuences the estimation of analgesic efﬁcacy in this type of study (Maze and Angst 2004). In addition, a recent randomised controlled trial of 34 post-operative patients showed that Dex provided more efﬁcacious analgesia than morphine (Arain et al. 2004). However, as described above (Sect. 2.3), regional rather than systemic approaches appear more efﬁ- cacious, although further clinical studies are required to quantify the systemic
analgesia afforded by α2 adrenoceptor agonists. Preclinical investigation into the downstream effectors beyond surface receptor signalling may facilitate the development of a new class of drugs to allow more effective pain relief.
As discussed in Sect. 2, the intrinsic antinociceptive system involves cholinergic signalling to modulate pain responses. Acetylcholine release induced by DINs or exogenous sources such as α2 adrenoceptor and opiate agonists induces an antinociceptive effect (Chen and Pan 2001). Recent interest in this class of
agents has been sparked by a series of interesting discoveries using genetic and pharmacological approaches.
Cholinesterase inhibitors have been used in anaesthesia for many years and their analgesic potential was highlighted over 70 years ago; recently their IT administration for analgesia has gained deserved interest (Hood et al. 1997). In human volunteers, neostigmine was shown to have an analgesic effect when given intrathecally and to potentiate systemic alfentanil analgesia. This effect was correlated with increased CSF acetylcholine.
Nicotinic acetylcholine receptor (nAChR) and muscarinic acetylcholine re- ceptor (mAChR) agonists are antinociceptive (Traynor 1998; Eisenach 1999). Cholinergic agonists have similar efﬁcacy to morphine; they lack the long-term addiction and withdrawal side-effects (Bannon et al. 1998; Swedberg et al. 1997) and thus may have utility in both acute and chronic pain settings. Furthermore, intranasal nicotine has recently shown potency as a post-operative analgesic (Flood and Daniel 2004).
Cholinergic Receptors and Substrates
Nicotinic cholinergic receptors are ligand-gated ion channels formed from pentamers of α subunits (containing the ACh binding site) with β, γ, δ or ε subunits. Neuronal nAChR are considered more diverse than their muscle equivalents, with α2–10 and β2–4 subunits cloned in neurons. α2–4 and 6 can form heteromeric channels with β2–4 subunits; α7–9 form homomeric channels. In the rat brain, messenger RNA (mRNA) for α4, α7 and β2 subunits are widely expressed, though the β2 subunit is less abundant in the brain of
primates (Tassonyi et al. 2002; Gotti et al. 2004). nAChR are excitatory in nature and likely inﬂuence learning, memory, arousal and analgesia.
mAChR are G protein-coupled channels which are either excitatory (M2 and 4) or inhibitory (M1, M3, M5) in nature. Even-numbered channels inhibit adenylyl cyclase via Gi/Go while odd-numbered channels, coupled to Gq/G11,
activate phospholipase C (Caulﬁeld and Birdsall 1998). Similar to α2 adreno-
ceptors, M2 receptor activation also activates GIRK channels (Fernandez-
Fernandez et al. 1999).
Investigation of the receptor subtypes involved has been furthered greatly by the use of genetic manipulation. Knock-out mice lacking the α4 or β2 nAChR subunit showed reduced nicotine-elicited antinociception in the hot plate test (Marubio et al. 1999). In addition, morphine antinociception is reduced in M4 and M4/M2 knock-out mice (Duttaroy et al. 2000). Similar
to α2 adrenoceptor agonists, M2 and M4 receptors are coupled to Gi proteins
and are inhibitory in nature, and dual knock-out of both receptors abolished
the antinociception induced by oxotremorine, a muscarinic agonist (Duttaroy
et al. 2002). Sole knock-out of the M2 receptor reduced antinociception (but to a lesser extent than dual knock-out) and also reduced the inhibitory action of muscarine on CGRP release from peripheral nerve endings (Bernardini et al.
2002). The antinociceptive action of oxotremorine, the muscarinic agonist, is reduced in mice with a GIRK-2-null mutation, indicating primarily a post- synaptic action of this drug via GIRK channels. As M2 channels are known to
activate GIRK channels and knock-out of either reduced muscarinic receptor antinociception, it is likely that M2 and GIRK channels mediate muscarinic antinociception, at least in mice. Therefore the likely cholinergic receptor subtypes modulating pain are the nAChR α4 or β2, and mAChR M2 and M4.
Potentiation of endogenous acetylcholine by the IT use of anti-cholinesterase inhibitors has recently gained renewed interest after recent clinical trials. Act- ing via both muscarinic and nicotinic receptors, acetylcholine is known to play a role in both endogenous and exogenous analgesia. Neostigmine (IT)
potentiates opioid and α2 adrenoceptor agonist-induced analgesia (Detweiler et al. 1993; Hood et al. 1997). Clinically the combination of neostigmine IT and opioid analgesia has been used successfully in gynaecological and obstet-
ric anaesthesia; particular utility for mobile epidurals has been noted (Lau- retti et al. 1998; Roelants and Lavand’homme 2004). However, nausea remains a limitation of this novel strategy and further investigation is required to fully
describe the extent of this side-effect.
Recent interest in nAChR agonists as analgesics was prompted by epibatidine, a compound isolated from the Ecuadorian tree frog, Epipedobates tricolor. Epibatidine is 200 times more potent than morphine but has a narrow ther- apeutic window with signiﬁcant toxicity problems, including cardiovascular and motor effects (Decker and Meyer 1999). Newer compounds such as ABT-
594 (Bannon et al. 1998) have an improved safety proﬁle but similar efﬁcacy. Importantly, ABT-594 also lacks the withdrawal side-effects and physical de- pendence of opioid analgesics, at least in rats. In a series of studies, Bannon and colleagues investigated the systemic efﬁcacy of ABT-594 in models of thermal, inﬂammatory and neuropathic pain; the compound proved superior to morphine. The neuroanatomical locus for this effect is still under discus- sion but activation of the LC and serotonergic neurons in the nucleus raphe magnus which in turn activates DINs is likely to mediate the systemic anal- gesia of nAChR agonists (Bannon et al. 1998; Bitner et al. 1998). Consistent
with this concept, intracerebroventricular pre-treatment with an α4 antisense
oligonucleotide attenuated systemic nAChR agonist analgesia. At the level of the spinal cord, serotonin release contributes to nAChR agonist analgesia likely via activation of non-α4β2 nAChR. Activation of these receptors likely acts via
’volume transmission’ whereby pre-synaptic nAChR increase serotonin re- lease rather than directly stimulate neuronal ﬁring (Cordero-Erausquin and Changeux 2001). nAChR agonists also activate noradrenergic and muscarinic
DINs (Rogers and Iwamoto 1993) when administered systemically. Signiﬁcant controversy still plagues the analgesic effects of nAChR agonists administered intrathecally with both pro- and antinociceptive actions reported (Rueter et al.
2000); therefore, further study is required before use of these compounds for
regional analgesia. Systemic administration of nAChR agonists likely induces analgesia predominantly through supraspinal mechanisms with subsequent activation of DINs.
Activation of muscarinic receptors induces antinociception in various pain paradigms including thermal, inﬂammatory and neuropathic pain (Wess et al.
2003; Kang and Eisenach 2003; Shannon et al. 2001). Both central and periph- eral mechanisms of mAChR agonist analgesia exist. Central effects are likely targeted to the dorsal horn of spinal cord where M2 receptors predominate.
Peripheral activation of M2 receptors likely contributes to analgesia via re- duced CGRP release. However, peripheral mechanisms lead to some concern about autonomic effects of these ligands which will require careful evaluation;
mAChR agonists may lend themselves to regional techniques rather than sys- temic approaches (potentially in contrast to nAChR agonists). Furthermore, as there is little evidence for supraspinal effects of muscarinic agonists there
may be little need to employ systemic administration.
As vedaclidine exhibits the typical antinociceptive proﬁle of a muscarinic agonist and activates M2 and M4 receptor subtypes but inhibits the odd num- bered receptors (Shannon et al. 2001), it is likely that the inhibitory action of the even-numbered channels is the predominant antinociceptive mediator. Fur- thermore, consistent with pharmacogenomic data, muscarinic antinociception is pertussis toxin sensitive, which indicates the role of inhibitory G protein sig- nalling, likely via M2 and M4 receptors. Further evidence that even numbered channels mediate muscarinic agonist antinociception is that M2 receptors are known to couple to GIRK receptors and GIRK-null mutants exhibit reduced muscarinic antinociception. This also indicates a post-synaptic mechanism of action; however, muscarinic antinociception is also reduced in the presence
of the γ-aminobutyric acid (GABA)B receptor antagonist CGP55845, which is thought to represent a pre-synaptic effect of augmented endogenous GABA re-
lease inhibiting neurotransmitter release (Li et al. 2002). Antagonists of the M2, M3 and M4 receptor subtypes inhibit this GABA release (Zhang et al. 2005). In
addition, as M4 receptors play a crucial role in the potency of α2 adrenoceptor agonists against neuropathic pain (Kang and Eisenach 2003), targeting this receptor subtype may prove useful for problematic pain management.
Conclusion: Cholinergic Agents
Cholinergic agents represent a novel and potent class of new analgesics. We await the results of clinical trials to investigate their role in clinical medicine, but we hope they will enjoy clinical utility as independent and adjunctive anal- gesic therapy. Further investigation into the neural substrates and the receptor subtypes involved will allow further development of cholinergic strategies to combat pain.