- Email: [email protected]
The mechanism of acupuncture analgesia: a review
Complementary Therapies in Medicine (1997)5, 102-111
© PearsonProfessionalLtd 1997
The mechanism of acupuncture analgesia: a review J. Sims St George's Hospital Medical School, London, UK SUMMARY. In order to better understand the therapeutic effectiveness of acupuncture, questions about the means by which it operates need to be addressed. This article provides an overview of the neural, humoral and biomagnetic mechanisms that may contribute to the production of acupuncture analgesia. The concordance of some acupuncture points and trigger points is noted. Possible modulatory areas within the central nervous system are considered and parallels between the influence of acupuncture and the theory of diffuse noxious inhibitory control are explored. The concurrence of peripheral and central nervous activity with acupuncture stimulation suggests a functional if not an anatomical relationship between acupuncture points, meridians, and the nervous system. In addition, many studies demonstrate that acupuncture activates endogenous substances which inhibit nociceptive transmission. Two alternative theories are examined: that acupuncture analgesia occurs as part of a generalized stress response or as a result of an individual's suggestibility. At a practical level, the choice o f stimulation modality may influence clinical outcome, as different forms of stimulation, e.g. electroacupuncture and transcutaneous electrical nerve stimulation, can invoke subtly different pathways to produce analgesia. Future studies of the clinical effectiveness of acupuncture could be enhanced by using physiological measurement to assess both process and outcomes. This may also help in providing effective, individualized treatment.
INTRODUCTION
Mechanisms underlying acupuncture analgesia
As Lewith & Vincent1have recently stated, it is important that questions about whether acupuncture is therapeutically effective are raised alongside those about the means by which it may operate. Several review articles are available which consider the clinical efficacy of acupuncture analgesia (AA). 25 The clinical evidence remains somewhat equivocal. However, this is in part due to the weakness of research methodology in the area, a matter that is being addressed by current researchers. This paper will discuss the various mechanisms that are currently implicated in the production of the analgesic effect.
Before reviewing the information available on the anatomical and physiological pathways which may mediate an analgesic effect following acupuncture stimulation, a brief mention of the traditional Chinese medicine (TCM) viewpoint on the causes of pain will highlight the difference in underlying philosophies between TCM and Western orthodox medicine. Within the philosophy of TCM, pain may result from excesses (Shi) or deficiencies (Xu) in the circulation of blood and/or Chi (energy source). Shi is associated with stagnation of Chi, blood or stagnation due to cold. Insufficient nourishment of Chi, blood, or the organs and tissues in general, can create Xu pain. For example, Bensoussan7 cites the accumulation of cold or heat and the retention of phlegm as associated common causes of pain. In orthodox terms, the mechanisms whereby acupuncture operates can be categorized into three main areas: neural; humoral; and electromagnetic, although some overlap exists. Furthermore, these categories are not exhaustive: other mechanisms remain to be investigated and explained.
Acupuncture analgesia in Western medicine The most well known use of acupuncture in the West is for analgesic purposes. Meta-analyses of controlled trials of acupuncture usage in chronic pain have been published. 3,s Acupuncture has many other recorded uses, which have been discussed elsewhere? Reviews emphasize the need for a greater understanding of possible underlying mechanisms. Jane Sims, Division of General Practice & Primary Care, St George's Hospital Medical School, Cranmer Terrace, L o n d o n SW17 ORE, U K . 102
Mechanism of acupuncture analgesia
Cerebral cortex
~1
103
~
I I I ----.,~
.o_ E ..{:
g
'6(/3
Spinoreticular tract
D~;norphinA &B "', met-enkephalin Nora~trenalin So~atostati
Serotonin s*
""
I
omnioqa I l
~ica~";~d
nociceptors • principally via small fibr (As and C) acupuncture • impulses
~ular
principally via large fibn (A~ and As)
PERIPHERAL NERVOUS SYSTEM Main pain pathway
Neurological pathway inhibiting acupuncture analgesia (AA)
Neurological pathway inhibiting acupuncture analgesia (AA)
Figure Some of the neurological connections between the brainstem, the spinal cord and peripheral nerves that are activated by acupuncture. (Adapted, with permission, from Bensoussan A. The Vital Meridian: A Modern Exploration o f Acupuncture. Edinburgh. Churchill Livingstone, 1991 .)
NEURAL MECHANISMS Denervation experiments have served to support the neural model, i.e. that acupuncture effects operate via the nervous system? ,9 An early review 1° of the neurophysiology of AA proposed the involvement of sensory-neural integration. The Figure highlights the pathways involved. An intact nervous system is required for effective AA? 1 Local anaesthetic to a nerve blocks the effects of acupuncture in the area served by the nerve, suggesting neural involvement, z2
Blocking deep tissue receptors can abolish the characteristic needling sensation associated with acupuncture and the analgesic effect? ° However, whether the selective blocking required to test this mechanism can be achieved remains questionable.
Sensory (afferent) pathways From a T C M perspective, for acupuncture to be effective, the needling sensation, or de Qi, must be elicited. From a physiological viewpoint, a high-threshold
104
Complementary Therapies in Medicine
receptor must be stimulated to produce the acupuncture effect?3 Han & Terenius ~4proposed that needling stimulates somatic muscle afferents, e.g. A-8 (group III) and even the smaller C (group IV) nerve fibres, activating central opioid production. Han z5has implicated both A-5 and the larger A-~ fibres in transmission. Experimentation with rabbits ~ suggested that stimulating the smaller afferent fibres produced the greatest analgesic effect.
Spinal level Primary afferent A-J3 fibres end in the dorsal column nuclei in the spinal cord. High-frequency stimulation, as evoked by transcutaneous electrical nerve stimulation (TENS), operates by this route. Such fibres produce collaterals (branches) which stimulate inhibitory interneurones. One result is inhibition of the nociceptive input of C-fibres ~7operating at a segmental level, i.e. within the same spinal cord segment. A-8 afferent nerves also make connections at dorsal horn level, where they elicit the release of enkephalins (analgesic substances: see Humoral mechanisms below) from interneurones. Enkephalins act segmentally within the substantia gelatinosa of the spinal cord to inhibit the effects of nociceptive input. A-8 fibres also synapse with nerve cells that convey information via the spinothalamic tract to the periaqueductal grey matter (PAGM) of the midbrain. From there, descending inhibitory control upon the enkephalinergic interneurones occurs. In a cat model, Wu et al ~8observed the discharge of spinal cord dorsal horn neurons to heat stimulus to be temporarily attenuated by electroacupuncture (EA). This suggests modulation of the afferent transmission of pain stimuli. Experiments where the spinal cord was differentially lesioned suggest that the anterolateral as well as the dorsal column may mediate AA. 1° Correspondence of trigger points, motor points and acupuncture points Liu and colleagues~9conducted a double-blind study to investigate the correspondence between certain motor points, i.e. where the nerve passes through the fascia into the muscle, and acupuncture loci. They concluded that about 70% of acupuncture points correspond to motor points. Further, Melzack and associates2° noted similarities between trigger points and common acupuncture points for pain in terms of spatial distribution and pain patterning. Although they stem from different conceptual bases, the empirical approach involved in linking points and the effect of applying pressure to them could be expected to give rise to some concordance. Whilst acupuncture meridians are anatomically non-existent, the precise genesis of trigger points is also unclear. Pressure at trigger points evokes localized and referred pain in a classical pattern. At the
time their article was written, the underlying neural processes could only be tentatively explored. The impact of acupuncture stimulation can be observed at various levels throughout the central nervous system (CNS). Models that incorporate inhibition via higher centres, e.g. the thalamus and the brainstem, are required; 2I the intraspinal level of Melzack and WaU's original gate control theory22 of pain modulation has been superseded. The relatively delayed onset of AA and its prolonged effect recorded during many animal experiments implies additional supraspinal activity?° A widespread, intersegmentally mediated analgesic effect can occur which refutes any hypothesis23that the analgesic effect occurs only within areas served by the spinal segment that received the pain stimulus.
Brainstem level: midbrain and medulla The brainstem also mediates AA. TENS-conveying nerves (A-[~ afferents) end chiefly in the thalamus, but collaterals to the midbrain exist. From the midbrain pretectal region there are connections with the PAGM?4 Much of A-6 afferent input is conveyed to the reticular formation, where modulation occurs. Chang ~° reported changes in the electrical discharge of reticular neurons during manual needling of the Tsusanli acupuncture point. The pattern was not characteristic of 'pain'; rather, the reticular formation acts more as a relay station. Descending inhibition may also originate from this region?5 Animal studies provide increasing evidence of a spino-reticulo-spinal feedback loop for the detection and modulation of pain. Bing et al2~ compared the responses of subnucleus reticularis dorsalis (SRD) neurons in the rat medulla to various noxious stimuli and to acupuncture stimulation. Neurons responded to stimulation at different acupuncture points, particularly those contralateral to them; acupuncture stimuli use a similar afferent route. Similar responses were elicited by acupuncture at non-acupuncture points, which accords with a lack of topographical specificity for the analgesic effects of acupuncture?7
Diencephalon level The thalamus contains nociceptors as part of its neuronal relay system.28Thalamic nociceptive outflow has been inhibited during acupuncture in animals. Changes in electrical discharge from certain thalamic nuclei have been seen following EA. 29 A weakened aversive behavioural response, i.e. withdrawal from a noxious stimulus, was also observed. Similar effects were reported using CNS stimulation in particular areas and following the administration of the analgesic, fentanyl. In another series of experiments, thalamic nociceptive discharge was inhibited both by stimulating the midbrain raphe nucleus and by the
Mechanism of acupuncture analgesia application of EA. 3° EA thus appears to influence thalamic transmission of noxious information. An intact pituitary gland is required for an analgesic effect to occur. 28The hypothalamus also seems to modulate AA. The arcuate nucleus receives stimulation from the prefrontal cortex; this may be a channel for the modulation of pain by emotional and cognitive interpretation. Lesions to the arcuate nucleus can reduce AAY ,32
Descending pathways Higher centres may exert a descending inhibitory gating effect.33The PAGM is central to pain modulation by both ascending and descending inhibition. 34 Stimulation of the PAGM enhanced AA in rats. 35 Recently, the evidence for somatotopic organization within the PAGM has been suggested as a possible explanation of the need for using specific heterosegmental acupuncture points. 13 The midbrain, in particular the nucleus raphe magnus (NRM), acts as a relay station for descending inhibitory messages, g6 Liu et aP 7 proposed that EA may activate the NRM, thereby inducing analgesia. Medullary lesion involving the NRM has been shown to reduce EA effects. 38 EA analgesia in cats was markedly attenuated by the transection of the dorsolateral columns of the spinal cord, an area that contains descending fibres from the NRM. 39A somatotopic relationship between areas of the raphe nuclei and the spinal dorsal horns may be somehow connected to the relative specificity of acupuncture points in producing regional analgesia. 4°
Diffuse noxious inhibitory control Diffuse noxious inhibitory control (DNIC) theory proposes a counter-irritant stimulation which acts to inhibit pain messages. Counter-irritation has been tested in both human and animal studies. Research has highlighted the need for a noxious afferent input stimulus. 26,4IDNIC may act as a filter, extracting pain signals en route to higher centres?2 In addition, the signals could amplify the response to incoming noxious stimuli. Opioid receptors and other neurotransmitters are involved, 42,43as will be subsequently discussed. The impact of what may be DNIC due to sham needling, plus the overall placebo effect, may confound investigations of the clinical effectiveness of AA.2,5, 4
DNIC-associated activity does not tally directly with the acupuncture response. DNIC effects tend to be transient, whilst the effects of acupuncture can be long-lasting. Secondly, acupuncture does not appear to necessarily involve C-fibre activation. TENS and EA seem to utilize primarily A-a and -13 afferents, with C-fibres only being activated at higher threshold levels.12
105
HUMORAL MECHANISMS During the 1970s, it became clear that the delayed onset and duration of acupuncture analgesic response necessitated a hypothesis for a humoral influence. There is a need for more investigation in this area, given its complexity. The humoral model addresses the role of neurotransmitters, hormones and other chemical substances that circulate in the blood and cerebrospinal fluid (CSF). Many acupuncture studies have involved chemical blockade or enhancement to assess the impact of various neurotransmitters.
Opioid activity He 45 reviewed the CNS circuitry through which acupuncture appears to evoke analgesic activity via the release of endogenous opiate peptides. Various CNS sites are rich in opioids and their receptors. These will be briefly reviewed in relation to acupuncture studies. The research can be divided into three main categories: 1. The measurement of opioid levels, e.g. enkephalins, during/following acupuncture 2. The enhancement of opioid levels during/following acupuncture 3. The inhibition of AA using the opiate (morphine) antagonist naloxone.
Measurement of opioid levels Raised CSF endorphin levels following EA have been reported. 46 All studies demonstrated pain relief. Differential opiate release occurred in a study of lowfrequency EA. 47 CSF I]-endorphins were raised; metenkephalin levels remained unchanged. However, others48reported raised plasma met-enkephalins and stable ~3-endorphins in their patients. In an animal experiment, increased CSF leu-enkephalins were observed following EA. 49 The PAGM is a focal site for various endogenous opioids and their receptors. Zhang et aP° investigated the hypothesis that acupuncture elicits PAGM opioid release. Opiod assays from PAGM perfusate increased during EA, concordant with the degree of analgesic effect produced. In addition, both opioid and cholinergic mechanisms in the caudate nucleus (one of the basal nuclei) have been proposed to operate during AA. 51'52Xie et aP 3 reported an association between caudate levels of met-enkephalin and the degree of AA produced in rats. The hypothalamic arcuate nucleus has been shown experimentally to assist in [3-endorphin-mediated analgesia?4 Other mediation sites include the preoptic area 55and parts of the limbic system?6
106
Complementary Therapies in Medicine
Enhancement of opioid levels The administration of bacitracin which inhibits degradation of endogenous opioids enhanced AA in rabbits? 7 Enkephalin levels were enhanced in the intervention group. This finding can be supported by rat experiments. Cyclohexamide, which acts to reduce met-enkephalin production, attenuated both the EAinduced increase in met-enkephalins and the analgesic effect, as measured by a tail-flick test in rats? 8 Zhou et aP9 report similar findings from radioimmunoassay studies, albeit with very small sample sizes. They hypothesized that acupuncture may raise enkephalin levels by accelerating opioid biosynthesis. One type of amino acid, the D-amino acids, block the enzymes that would normally degrade endorphins: they can prolong AA. 6° In one experiment, tail-flick response to painful stimuli in rats was inhibited by EA, morphine and PAGM stimulation?~ Administration of d-phenylalanine enhanced all three analgesics, suggestive of a common underlying mechanism.
Inhibition of AA using naloxone The drug naloxone competes for opioid receptors, blocking opioid activity. In animal models, 5°,6~67opioid involvement was inferred, given the reversal of AA and behavioural responses following naloxone administration. Similar findings have been reported from other animal models, where the behavioural response to nociceptive stimulation was observed. Injection of naloxone into the septal area, 68 the nucleus accumbens69 and the amygdala69,7° partially reversed the effects of both morphine and EA in rabbits. This work supports clinical findings from human tooth pulp stimulation experiments where AA was seen to be naloxone-reversible. 7~ Acupuncture appears to influence the release of endogenous opioids onto receptor sites to elicit analgesia. Specific receptors, namely the g-1 type, seem responsible for the mediation of analgesic effects.72 CXBX-strain mice who are deficient in opioid receptors have a poor analgesic response to EA. 73 Not all the research favours an opioid explanation. Chapman and co-workersTM failed to reverse AA using naloxone in a tooth pulp stimulation experiment. In a small randomized double-blind trial of patients with chronic pain, no change in the pain relief afforded by manual acupuncture was encountered in those given naloxone.75 When comparing studies, however, in addition to noting the naloxone dosage used (and possibly the time of administration), the type of stimulation and the points chosen must be considered, since these could also modify outcome. Cheng & Pomerantz 62 reported that whilst analgesic effects could be produced by EA at both high (200 Hz) and low (4 Hz) frequencies, only the analgesia under the latter condition was reversed by naloxone.
Another set of experiments may also clarify reasons for the discrepant findings. Increased levels of both endogenous opioids and serotonin (5-HT) have been associated with acupuncture. A lowering of either one of these substrates resulted in a compensatory increase in the other.76 A reduced analgesic effect generally occurred with both naloxone and a 5-HT synthesis inhibitor. Pharmacological interference with both substances almost abolished analgesic effects. Other humoral activity
Over 20 neurotransmitters have been associated with the response to acupuncture. The key players are: serotonin (5-HT); dopamine (DA); noradrenaline (NA); and somatostatin, in particular CNS regions. (It should be recalled, however, that neurotransmitters tend to be found in some quantity in most areas of the brain.) Serotonin is thought to have a largely inhibitory effect. (However, in terms of specificity, it should be noted that, clinically speaking, it is the non-specific tricyclics that act as analgesics, rather than, say, SSRIs). Blood serotonin levels change with acupuncture intervention; the increase may be due to the release of a precursor or associated pituitary activity, since serotonin does not cross the blood-brain barrier. 77 Jin et alTM demonstrated enhanced 5-HIAA (indicative of catecholamine metabolism per se) and tryptophan (indicative of 5-HT activity) levels with EA and suggested that EA enhances 5-HT activity. From their experiments, Zhu & Shi79proposed that acupuncture activates the chain of events from opioid, and to a lesser extent 5-HT and NA release in the PAGM to descending 5-HT influence from the PAGM to the NRM. Han et aP° demonstrated this experimentally in a double-blind study using clomipramine, a 5-HT reuptake inhibitor. Several authors 78'8~have used 5, 6-DHT to deplete serotonin and have noted an attenuation of EA effect. Interestingly, the impact of acupuncture can be less in a depressed patient?2 This may accord with research that suggests that patients suffering from depression have lower serotonin levels?3 Painful impulses are also modulated, both positively and negatively, by catecholamines: DA, NA and somatostatin. In experiments, AA was enhanced following the introduction of exogenous cAMP and ATW4 These substances may activate intermediary hypothalamic-pituitary protein hormones. In a series of experiments, serotonin and catecholamine levels were measured during acupuncture. TM EA appeared to activate dopaminergic (DA) neurons. The analgesic effects of EA were enhanced using apomorphine, a DA agonist and reduced by 5,6-DHT, a dopamine receptor (D2) antagonist. Also, AA was accompanied by a reduction in brain NA levels. Unfortunately, interpretation of the findings is limited due to the small sample sizes.
Mechanism of acupuncture analgesia Cholinergic mechanisms may also be involved in AA, the caudate nucleus being a key areaY ,85 An association between AA and raised levels of acetylcholine (ACh) has been reported. 85 Ge et aP 6 conducted histochemical analysis of rat brain after acupuncture. They found raised levels of acetylcholinesterase (ACHE, the enzyme which aids ACh degradation) in the locus coeruleus and raphe nucleus, compared to a control group. Atropine (ACh antagonist) decreased the analgesic effect of acupuncture. However, given the widespread activity of ACh within the CNS, it is difficult to assess any specific role it might have. Analgesic antagonists include substance p,30cholecystokinin and cAME 8°'87Yonehara et aP8 suggested that EA inhibits substance P release evoked by tooth pulp stimulation in animals. They discussed the possibility that EA exerts an opioid inhibitory effect on substance P at the level of the trigeminal nucleus (brainstem region) and, via a relay system, may elicit monoamine release, with a consequent inhibitory, descending influence. Other experiments showed AA to be antagonized by gamma-amino butyric acid (GABA) release from the diencephalon region? 9 Antagonistic monoamine (DA, NA) involvement, chiefly via the locus coeruleus, may also occur. Dopaminergic influences may antagonize the opioid effect in the caudate nucleus. 90 Inhibition of adrenergic (sympathetic) activity by acupuncture may account for some of the analgesic pattern. Cao et a191 reported corroboratory physiological evidence following acupuncture. However, at spinal level, NA may have a facilitatory effect22 Catecholamines act differentially; AA was enhanced following direct microinjection of NA into the ventromedial hypothalamus. TM Ge et a186histochemically compared catecholamine levels in the locus coeruleus and raphe magnus using tissue from acupunctured animals and controls. Catecholamine levels appeared to be lower in the former group, perhaps as a result of increased monoamine oxidase activity (the enzyme which assists catecholamine degradation). The authors cite clinical evidence of a decrease in monoamines during acupuncture anaesthesia and reserpine (a monoamine antagonist) experiments, wherein analgesia has been enhanced. They conclude that the findings have been mixed, probably due to the various functions of the locus coeruleus, but favour a limited role for catecholamines in AA. Several experimenters 7° have sought to develop an understanding of the opposing roles of monoamines using antagonist drugs, i.e. cz- and [3-blockers. The evidence suggests that antagonistic (non-analgesic) functions operate via o~2 receptors, whilst analgesic effects operate via 13 receptors, the overall balance favouring the former activity.
107
Donor experiments Some interesting experiments have demonstrated the importance of circulating chemicals by giving acupuncture to one animal and then donating the body fluid to a recipient animal, with therapeutic outcome occurring in both? 2,93~6Unfortunately, methodological details appear to have been lost in the translation from Chinese.
BIOELECTRIC AND BIOMAGNETIC MEDIATION Bioelectric mediation supposes that the meridians are electrically distinct entities. Bioelectric changes associated with the meridians may precede any neurological or humoral reactivity to acupuncture. Acupuncture points correspond well with areas of low resistance (high permeability) around the body.97-99The resistance at a particular point seems to be lowered further in disease conditions. Many modern acupuncturists make use of this phenomenon in detecting the acupuncture points to use. Places with the greatest electromagnetic activity/bioluminescence correspond to TCM acupuncture points. 1°° Kirlian photography has been used to measure bioluminescence. This is thought to reflect the energy flow along the meridians, although evidence for its utility remains equivocal? °~Attempts to measure changes in the body's biomagnetic fields during acupuncture have been limited because the body's magnetic fields are very weak and are not readily detectable. 1°2It remains to be seen whether meridian flow is associated with charge movement during acupuncture. More recently, it has become possible to detect acupuncture meridians using radioisotopes. Darras et al ~°3 reported upon the use of technetium-99m to investigate acupuncture pathways. A particular migratory pattern was seen from various acupuncture points, which accorded with the traditional channels. Furthermore, these paths were distinct from the vascular and lymphatic circulation, in terms of rate and direction of flow. Concentrated areas of the tracer related not only to organs known to take up the tracer, but also to acupuncture points along a given meridian. The migratory pattern appeared to be altered in pathological conditions. In addition, changes in the tracer's migration rate from contralateral points of the body followed unilateral stimulation of an acupuncture point on the same meridian. The authors proposed that acupuncture stimulation evokes an effect via a neurochemical process. The pathways may be related to connective tissue diffusion following the neurovascular bundles along the extremities. This interstitial route needs further investigation.
108
Complementary Therapies in Medicine
Differential responses to various modes of acupuncture stimulation
Bioelectrical mediation may partially explain why acupuncture outcome may vary depending upon the type of stimulation employed, e.g. manual or electroacupuncture. 1°4 Furthermore, research evidence suggests that the intensities and frequencies used in EA may produce subtly different physiological outcomes. ~4,62,~°5-1°7In a therapeutic setting, therefore, one technique may be more appropriate than another.
effectiveness of acupuncture will be enhanced by the measurement of physiological parameters to assess both the process and outcomes. This will aid the comparison of acupuncture and alternative analgesics and may assist in the provision of effective, individualized treatment.
ACKNOWLEDGEMENTS The author would like to thank J. Tissier, R. Walker and J. Sizer for their valuable comments on this article.
Alternative hypotheses
Acupuncture analgesia as a generalized response to stress Some critics have suggested that AA may be purely a generalized response to trauma. However, researchers have described the distinctions between acupuncture and stressful stimuli, highlighting different processes a n d outcomes. 56'84'1°4'1°8111
Alternative hypotheses: hypnotic suggestibility Some researchers contest that acupuncture effect is due to a hypnotic-type suggestibility. Knox & Shum ~2 were unable to reverse hypnotic analgesia using naloxone, an indication that at least one different mechanism operates. There was some intra-individual commonality of responsiveness to acupuncture and hypnotism, however. Others have found this to be unrelated to analgesic outcome. 1~3
CONCLUSION Whilst both neural and humoral mechanisms are important, neither can fully explain such a fundamental phenomenon as differential point efficacy in certain painful conditions. TM There may be no distinct neural or circulatory difference between a real point and a nearby sham point, yet needling outcomes can be different. Needling changes the electrical activity at points and along channels; this may serve to correct a homeostatic imbalance. The clinical effectiveness of acupuncture per se is said to depend upon the specific placement of the needles at designated acupuncture points. For analgesic activity, point specificity appears to be less crucial than for other types of therapeutic intervention, e.g. for vomiting. Hence DNIC may be a valid model for partially explaining AA. The therapeutic benefits of AA may relate to activity other than neurological blockade, e.g. decreasing gastric acidity or eliciting vasodilation. H5 'Correct acupuncture, properly directed to the management of pain, will also increase oxygenation of cardiac muscle, reduce gastric acidity and control inflammatory responses. '7 Hopefully, future studies of the clinical
REFERENCES 1. Lewith GT, Vincent C. On the evaluation of the clinical effects of acupuncture: a problem reassessed and a framework for future research. J Alt Compl Med 1996; 2:79 90. 2. Lewith GT. How effective is acupuncture in the management of pain? J R Coll Gen Pract 1984; 34: 275-278. 3. Richardson PH, Vincent CA. Acupuncture for the treatment of pain: a review of evaluative research. Pain 1986; 24: 1540. 4. Patel M, Gutzwiller F, Paccaud F, Marazzi A. A metaanalysis of acupuncture for chronic pain. Int J Epidemiol 1989; 18:900 905. 5. ter Riet G, Kleijnen J, Knipschild R Acupuncture and chronic pain: a criteria-based meta-analysis. J Clin Epidemiol 1990; 43: 1191-1199. 6. Vincent CA, Richardson PH. Acupuncture for some common disorders: a review of evaluative research. J R Coll Gen Pract 1987; 37: 7%81. 7. Bensoussan A. Contemporary acupuncture research: the difficulties of research across scientific paradigms. Am J Acupunct 1991; 19: 357-365. 8. Bossy J. Morphological data concerning acupuncture points and channel network. Acupunct Electrother Res 1984; 9: 76-106. 9. Shen E. Participation of descending inhibition in acupuncture analgesia. In: Zhang XT ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 31-38. 10. Chang HT. Neurophysiological basis of acupuncture analgesia. Scientia Sinica 1978; 21: 829-846. 11. Levy B, Matsumoto T. Pathophysiology of acupuncture: nervous system transmission. Am Surg 1975; 41: 378-384. 12. Chiang CY, Chang CT, Chu H, Yang L. Peripheral afferent pathway for acupuncture analgesia. Scientia Sinica 1973; 16: 210-217. 13. Bowsher D. The physiology of stimulation-produced analgesia. Acupunct Med 1991; 9: 58-62. 14. Han JS, Terenius L. Neurochemical basis of acupuncture analgesia. Annu Rev Pharmacol Toxicol 1982; 22: 193-220. 15. Han JS. On the mechanism of acupuncture analgesia. Acupunct Res 1984; 3: 236-245. 16. Chen L, Tang J, Fan L e t al. An analytical study of the afferent fibres for impulses of acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 478489. 17. Duggan AW, Foong FW. Bicuculline and spinal inhibition produced by dorsal column stimulation in the cat. Pain 1985; 22: 249-259. 18. Wu J, Xing J, Xing Bet al. Inhibitory action of electroacupuncture on dorsal horn neurons of the spinal cord. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986: 3949. 19. Liu Y, Varela M, Oswald R. The correspondence between some motor points and acupuncture loci. Am J Chin Med 1975; 3: 347-358. 20. Melzack R, Stillwell DM, Fox EJ. Trigger points and acupuncture points for pain: correlations and implications. Pain 1977; 3: 3-27.
Mechanism of acupuncture analgesia 21. Melzack R, Jeans ME. Acupuncture analgesia: a psychophysiological explanation. Minn Med 1974; 57: 161-166. 22. Melzack R, Wall PD. Pain mechanisms. A a new theory. Science 1965; 150:971 979. 23. MacDonald AJR. Segmental acupuncture therapy. Acupunct Electrother Res 1983; 8: 267-282. 24. Rees H, Roberts MHT. Antinociceptive effects of doral column stimulation in the rat: involvement of the anterior pretectal nucleus. J Physiol 1989; 417: 375-388. 25. Oliveras JL, Besson JM, Guilbad G, Liebeskind JC. Behavioural and electrophysiological evidence of pain inhibition from midbrain stimulation in the cat. Exp Brain Res 1974; 20: 3244. 26. Bing Z, Villaneuva L, Le Bars. Acupuncture-evoked responses of subnucleus reticularis dorsalis neurons in the rat medulla. Neuroscience 1991; 44: 693-703. 27. Lewith GT, Machin D. On the evaluation of the clinical effects of acupuncture. Pain 1983; 16:111-127. 28. Takeshige C. Differentiation between acupuncture point and non-point by association with analgesia inhibitory system. Acupunct Electrother Res 1985; 10: 195-203. 29. Zhao ZQ, Shao DH, Yang ZQ. Inhibition of evoked discharges of posterior nuclear group of thalamus induced by stimulation of central grey matter, caudate nucleus and electroacupuncture in waking rabbits. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986. 30. Zhang G, Wei J, Zhou L e t al. Effect of stimulation of midbrain raphe nucleus on discharges of nociceptive neurons in the nucleus parafascicularis of the thalamus and its significance in acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986: 82-92. 31. Gao YS, Ku YH. Mechanism underlying the excitatory effect of nucleus arcuatus hypothalami on PAG-NRM system and its significance in electroacupuncture in rats. Acta Physiol Sinica 1983; 35: 409415. 32. Quo SY, Yin WP, Zhang HQ, Yin QZ. Role of hypothalamic arcuate region in lip acupuncture analgesia. Acta Physiol Sinica 1982; 34: 71-77. 33. Andersson SA, Ericson T, Holmgren E, Lindqvist G. Electroacupuncture effect on pain threshold measured with electrical stimulation of teeth. Brain Res 1973; 63: 393-396. 34. Fields HL. An endorphin-mediated analgesia system: experimental and clinical observations In: Martin JB et al, eds. Neurosecretion and brain peptides. New York: Raven Press, 1981: 199-212. 35.Huang C, Li XC, Zhang CC et al. Effect of electrical stimulation of midbrain periaqueductal gray on acupuncture analgesia in rats. Acta Physiol Sinica 1982; 34: 343-357. 36. Behbehani MM. Evidence that an excitatory connection between the PAG and nucleus raphe magnus mediates stimulation produced analgesia. Brain Res 1979; 170: 85-93. 37. Liu X, Zhu B, Zhang SX. Relationship between electroacupuncture analgesia and descending pain inhibitory mechanism of nucleus raphe magnus. Pain 1986; 24: 383-396. 38.Du HJ, Chao Y E Localization of central structures involved in descending inhibitory effect of acupuncture on viscero-somatic reflex discharge. Scientia Sinica 1976; 19: 137-148. 39. Shen E, Ma WH, Lan C. Involvement of descending inhibition in the effect of acupuncture on the splanchnically evoked potential in the orbital cortex of the cat. Scientia Sinica 1978; 21: 677-685. 40. Ju G, Shu S. Somatotopic projection of the raphe nuclei to spinal dorsal horn in the cat - a horseradish peroxidase study. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986:516 528. 41. Le Bars D, Villaneuva L, Willer JC, Bouhassira D. Diffuse noxious inhibitory controls (DNIC) in animals and in man. Acupunct Med 1991; 9: 47-56. 42. Le Bars D, Chitour D, Kraus E et al. Effect of naloxone upon diffuse noxious inhibitory controls (DNIC) in the rat. Brain Res 1981; 204: 387402. 43.Wilier JC, Broucker T, Le Bars D. Diffuse noxious inhibitory controls (DNIC) in man: involvement of an opioidergic link. Eur J Pharmacol 1989; 182: 347-355.
109
44. Needham J, Lu GD. Celestial lancets: a history and rationale of acupuncture and moxibustion. Cambridge: Cambridge University Press, 1980: 261. 45. He L. Involvement of endogenous opiod peptides in acupuncture analgesia. Pain 1987; 31: 99-121. 46. Sjolund B, Terenius L, Eriksson M. Increased CSF levels of endorphins after electroacupuncture. Acta Physiol Scand 1977; 100: 382-384. 47. Clement-Jones V, McLoughlin L, Tomlin Set al. Increased beta-endorphin but not met-enkephalin levels in human cerebrospinal fluid after acupuncture for recurrent pain. Lancet 1980; ii(8201): 946-948. 48. Kiser RS, Khatami M J, Gatchel RJ et al. Acupuncture relief of chronic pain syndrome correlates with increased plasma met-enkephalin concentrations. Lancet 1983; ii(8364): 1394-1396. 49. Huang Y, Wang QW, Wan FS et al. Increase of CSF enkephalin content by electroacupuncture in monkeys. Kexue Tongbao 1981; 26:631 632. 50.Zhang AZ, Pan XP, Xu SF et al. Endorphins and acupuncture analgesia. Chin Med J 1980; 93:673 680. 51. He LF, Xu SE Caudate nucleus and acupuncture analgesia. Acupunct Electrother Res 1981; 6:169-182. 52. He L, Xu S, Jiang C. Role of caudate nucleus in acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press 1986: 146-157. 53.Xie CW, Zhang WG, Hong XJ, Han JS. Relation between the content of central met-enkephalin and leu-enkephalin and analgesic effect of electroacupuncture in rats. Acta Physiol Sinica 1984; 36: 192-197. 54. Yin QZ, Duanma ZX, Guo SY, Yu XM, Zhang YJ. Role of the hypothalamic arcuate nucleus in acupuncture analgesia. Second National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing 1984: 398. 55.Wu GC, Jiang JW, Cao XD. Effect of microiontophoretic etorphine, noradrenaline and electroacupuncture on spontaneous discharge of neurons in rabbits preoptic area. Acta Physiol Sinica 1984; 36: 552-558. 56. Zhang AZ. Endorphin and acupuncture analgesia research in the People's Republic of China (1975-1979). Acupunct Electrother Res 1980; 5:131 136. 57. Tsou K, Zhao DD, Wu SH et al. Enhancement of acupuncture analgesia with concomitant increase of enkephalins in striatum and hypothalamus by intraventricular bacitracin. Acta Physiol Sinica 1979; 31:377-381. 58.Wu SH, Chang TS, Wan FS, Tsou K. Effect of intraventricular cyclohexamide on acupuncture analgesia and brain met-enkephalin. Acta Physiol Sinica 1980; 32:79-81. 59. Zhou G, Yi Q, Wu S et al. Enkephalin involvement in acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986: 267-278. 60. Cheng RSS, Pomerantz B. A combined treatment with Damino acids and electroacupuncture produced a greater analgesia than with either treatment alone: naloxone reverses these effects. Pain 1980; 8:231 236. 61. Takeshige C, Murai M, Tanka M, Hachisu M. Parallel individual variations in effectiveness of acupuncture, morphine analgesia and dorsal PGA-SPA and their abolition by d-phenylalanine. In: Bonica JJ et al, eds. Advances in pain research, vol 5. New York: Raven Press 1983: 563-569. 62. Cheng RSS, Pomerantz B. Electro-acupuncture analgesia could be mediated by at least two pain-relieving mechanisms; endorphin and non-endorphin systems. Life Sci 1979; 25: 1957-1962. 63. Lou AL, Zheng CQ, Bai HQ. Influence of microinjection of naloxone and procaine into nucleus raphe magnus on acupuncture effect of visceral traction response in dogs. Acta Physiol Sinica 1982; 34: 84-89. 64. Huang Y, Wang QW, Zheng JZ et al Analgesic effect of acupuncture and its reversal by naloxone as shown by method of conditioning responses in monkeys. Chung Hua I Hsenh Tsa Chih 1978; 58:1193 1195. 65. Ha H, Tan EC, Fukunaga H, Osamu A. Naloxone reversal of acupuncture analgesia in the monkey. Exp Neurol 1981; 73: 298-303.
110
Complementary Therapies in Medicine
66. Pomerantz B, Chiu D. Naloxone blockade of acupuncture analgesia: endorphin implicated. Life Sci 1979; 19: 1757-1762. 67. Fu TC, Halenda SP, Dewey WL. The effect of hypophysectomy on acupuncture analgesia in the mouse. Brain Res 1980; 202: 33-39. 68. Mo WY,, Chou YF, Wang GN. Analgesia by injection of morphine and antagonism of electroacupuncture analgesia by injection of naloxone into septal area of rabbits. Chung Kuo Yao Li Hseuh Pao 1982; 3: 150-154. 69.Zhou ZF, Du MN, Wu WY et al. Effect of intracerebral microinjection of naloxone on acupuncture- and morphine-analgesia in the rabbit. Scientia Sinica 1981; 24: 1166-1178. 70. Zhou ZF, Du MY, Wu WY, Han JS. The effect of injection of naloxone into habenula or amygdala on acupuncture analgesia in the rabbit. National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1979: 504. 71. Mayer D J, Price DD, Rafii A. Antagonism of acupuncture analgesia in man by narcotic antagonist naloxone. Brain Res 1977; 121: 368-372. 72. Pasternak GW. Central mechanism of opioid analgesia. Acupunct Electrother Res 1981; 6: 135-149. 73. Peets JM, Pomerantz B. CXBK mice deficient in opiate receptors show poor electroacupuncture analgesia. Nature 1978; 273: 675-676. 74. Chapman C, Benedetti C, Colpitts Y, Gerlach R. Naloxone fails to reverse pain thresholds elevated by acupuncture: acupuncture analgesia reconsidered. Pain 1983; 16: 13-31. 75. Kenyon JN, Knight C J, Wells C. Randomised double-blind trial on the immediate effects of naloxone on classical Chinese acupuncture therapy for chronic pain. Acupunct Electrother Res 1983; 8: 17-24. 76. Han JS, Tang J, Fan Set al. Central 5-hydroxytryptamine, opiate-like substances and acupuncture analgesia In: Zhang XT ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 309-316. 77. Malizia E, Andreucci G, Paolucci D et al. Electroacupuncture and peripheral beta-endorphin and ACTH (letter). Lancet 1979; 2: 535-536. 78. Jin G, Zhang Z, Yu L e t al. Role of brain serotonergic and catecholaminergic systems in acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 317-331. 79. Zhu LX, Shi QY. Activation of nucleus raphe magnus by acupuncture and enkephinergic mechanisms. J Tradit Chin Med 1984; 4: 111-113. 80. Han JS, Tang J, Ren MF et al. Central neurotransmitters and acupuncture analgesia. Am J Chin Med 1980; 8:331 348. 81.Du HJ, Shen E, Dong XW et al. Effect on acupuncture analgesia by (sic) injection of 5,6-dihydroxytryptamine in cat: a neurophysiological, neurochemical and fluorescence histochemical study. Acta Zool Sinica 1978; 24:11-20. 82. MacDonald AJR. Getting under the skin. Nurs Times 1986; 14: 47-48. 83.Van Praag HM. Depression. Lancet 1983; 2: 912. 84.Liu X, Yang Z, Li Q et al. The role of hypothalamuspituitary-adrenal system in acupuncture analgesia In: Zhang XT ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press 1986: 397407. 85.Wang CY, Yu B, Liu XC. The influence of electroacupuncture on the ACh levels in various regions of rat brain. National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1979: 453. 86.Ge Z, Huang W, Wang Z, Song W. The relation between acupuncture anaesthesia and neurotransmitters and enzymes of locus coeruleus and raphe nucleus. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 558-567. 87. Zhou Y, Sun YH, Shen JM, Han JS. Increased release of immunoreactive CCK-8 by electroacupuncture and enhancement of electroacupuncture analgesia by CCK-B antagonist in rat spinal cord. Neuropeptides 1993; 24: 139-144.
88. Yonehara N, Sawada T, Matsuura H, Inoki R. Influence of electroacupuncture on the release of substance P and the potential evoked by tooth pulp stimulation in the trigeminal nucleus caudalis of the rabbit. Neurosci Lett 1992; 142: 53-56. 89. Zhu JQ, Li CB, Xu F, Liang WJ. Relationship of the metabolism of GABA in mice brain to electroacupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986. 90.Xu SF, Lu WX, Zhou GZ, Sheng MP, Wang DL. Effects of cholinergic and dopaminergic systems of the caudate nucleus in acupuncture analgesia. Second National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1984: 471. 91. Cao XD, Xu SF, Lu WX. Inhibition of sympathetic nervous system by acupuncture. Acupunct Electrother Res 1983; 8: 25-35. 92. Cao XD, He LF, Xu XF, et al. The role of PAG in acupuncture analgesia. Second National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1984: 639. 93.Yang MMP, Kok SH. Further study of the neurohumoral factor endorphin in the mechanism of acupuncture analgesia. Am J Chin Med 1979; 7: 143-148. 94. Lung Ch, Sun AC, Tsao CJ et al An observation of the humoral factor in AA in rats. Am J Chin Med 1974; 2: 203-205. 95. Beijing Research Group of Acupuncture Anaesthesia. The role of some neurotransmitters of brain in finger acupuncture analgesia. Scientia Sinica 1974; 17:112-130. 96. Chen M, Zheng Y, Xue Z et al. Study of role of humoral factors in central action of acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 360-371. 97. Rabischong P, Niboyet JE, Terral C et al. Experimental basis of acupuncture analgesia. Nouv Presse Med 1975; 4: 2021-2026. 98. Reichmanis M, Marino AA, Becker RO. Electrical correlates of acupuncture points. IEEE Trans Biomed Eng 1975; Nov: 533-535. 99. Zhu ZX. Research advances in the electrical specificity of meridians and acupuncture points. Am J Acupunct 1981; 9: 203-216. 100. Luciani RJ. Direct observations and photography of electroconductive points on the human skin. Am J Acupunct 1978; 4: 311-317. 101. Van der Schaar W~. Organ pathology, diagnosis and acupuncture. Am J Acupunct 1976; 4: 252. 102. Jessel-Kenyon J, Cheng N, Blott B, Hopwood V. Studies with acupuncture using a SQUID biomagnetometer: a preliminary report. Compl Med Res 1992; 6: 142-151. 103. Darras J-C, de Vernejoul R Albarede R Nuclear medicine and acupuncture: a study on the migration of radioactive tracers after injection at acupoints. Am J Acupunct 1992; 20: 245-255. 104. Nappi G, Facchinetti F, Leguante G e t al. Different releasing effects of traditional manual acupuncture and electroacupuncture on propriocortin related peptides. Acupunct Electrother Res 1982; 7: 93-103. 105. Peets JM, Pomerantz B. Acupuncture-like transcutaneous electrical nerve stimulation analgesia is influenced by spinal cord endorphins but not serotonin. In: Advances in Pain Research and Therapy (9). New York: Raven Press, 1985: 519-525. 106.Han JS, Xie GX, Ding XZ, Fan SG. High and low frequency electroacupuncture analgesia are mediated by different opioid peptides. Pain 1984; (suppl) 2: 543. 107. Levine MF, Hui-Chan CWY. Conventional and acupuncturelike transcutaneous electrical nerve stimulation excite similar afferent fibers. Arch Phys Med Rehabil 1993; 74: 54-60. 108. Wei RS. Investigations on the mechanisms and therapeutic effects of acupuncture treatments. N E Med J Chin 1952; 2: 44-53. 109.Wen WL, Ho WKK, Wong HK et al. Changes in ACTH and cortisol levels in drug addicts treated by a new and rapid detoxification procedure using acupuncture and nalo×one. Chin Med E W 1978; 6: 241-246.
Mechanism of acupuncture analgesia
110. Pomerantz B. Relation of stress-induced analgesia to acupuncture analgesia. Ann N Y Acad Sci 1986; 467: 4~A A~7. 111. Kho HG, Kloppenborg PWC, van Egmond J. Effects of acupuncture and transcutaneous stimulation analgesia on plasma hormone levels during and after major abdominal surgery. Enr J Anaesthesiol 1993; 10: 197-208. 112. Knox VJ, Shum K. Reduction of cold pressor pain with acupuncture analgesia in high and low hypnotic subjects. J Psychol 1977; 86: 639-643. 113. Xu L, Fu Z, Xiang M e t al. The effects of acupuncture analgesia and its relation to blood endorphin, blood
111
histamine and suggestibility. National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1979. 114. Takeshige C. The central mechanism of acupuncture in acupuncture anaesthesia: differentiation of acupuncture point and non-point by the central analgesia producing system. Acupunct Electrother Res 1983; 8: 323-324. 115.National Health and Medical Research Council. Report of the working party on acupuncture. Canberra: Australian Government Publishing Services, 1988.
© PearsonProfessionalLtd 1997
The mechanism of acupuncture analgesia: a review J. Sims St George's Hospital Medical School, London, UK SUMMARY. In order to better understand the therapeutic effectiveness of acupuncture, questions about the means by which it operates need to be addressed. This article provides an overview of the neural, humoral and biomagnetic mechanisms that may contribute to the production of acupuncture analgesia. The concordance of some acupuncture points and trigger points is noted. Possible modulatory areas within the central nervous system are considered and parallels between the influence of acupuncture and the theory of diffuse noxious inhibitory control are explored. The concurrence of peripheral and central nervous activity with acupuncture stimulation suggests a functional if not an anatomical relationship between acupuncture points, meridians, and the nervous system. In addition, many studies demonstrate that acupuncture activates endogenous substances which inhibit nociceptive transmission. Two alternative theories are examined: that acupuncture analgesia occurs as part of a generalized stress response or as a result of an individual's suggestibility. At a practical level, the choice o f stimulation modality may influence clinical outcome, as different forms of stimulation, e.g. electroacupuncture and transcutaneous electrical nerve stimulation, can invoke subtly different pathways to produce analgesia. Future studies of the clinical effectiveness of acupuncture could be enhanced by using physiological measurement to assess both process and outcomes. This may also help in providing effective, individualized treatment.
INTRODUCTION
Mechanisms underlying acupuncture analgesia
As Lewith & Vincent1have recently stated, it is important that questions about whether acupuncture is therapeutically effective are raised alongside those about the means by which it may operate. Several review articles are available which consider the clinical efficacy of acupuncture analgesia (AA). 25 The clinical evidence remains somewhat equivocal. However, this is in part due to the weakness of research methodology in the area, a matter that is being addressed by current researchers. This paper will discuss the various mechanisms that are currently implicated in the production of the analgesic effect.
Before reviewing the information available on the anatomical and physiological pathways which may mediate an analgesic effect following acupuncture stimulation, a brief mention of the traditional Chinese medicine (TCM) viewpoint on the causes of pain will highlight the difference in underlying philosophies between TCM and Western orthodox medicine. Within the philosophy of TCM, pain may result from excesses (Shi) or deficiencies (Xu) in the circulation of blood and/or Chi (energy source). Shi is associated with stagnation of Chi, blood or stagnation due to cold. Insufficient nourishment of Chi, blood, or the organs and tissues in general, can create Xu pain. For example, Bensoussan7 cites the accumulation of cold or heat and the retention of phlegm as associated common causes of pain. In orthodox terms, the mechanisms whereby acupuncture operates can be categorized into three main areas: neural; humoral; and electromagnetic, although some overlap exists. Furthermore, these categories are not exhaustive: other mechanisms remain to be investigated and explained.
Acupuncture analgesia in Western medicine The most well known use of acupuncture in the West is for analgesic purposes. Meta-analyses of controlled trials of acupuncture usage in chronic pain have been published. 3,s Acupuncture has many other recorded uses, which have been discussed elsewhere? Reviews emphasize the need for a greater understanding of possible underlying mechanisms. Jane Sims, Division of General Practice & Primary Care, St George's Hospital Medical School, Cranmer Terrace, L o n d o n SW17 ORE, U K . 102
Mechanism of acupuncture analgesia
Cerebral cortex
~1
103
~
I I I ----.,~
.o_ E ..{:
g
'6(/3
Spinoreticular tract
D~;norphinA &B "', met-enkephalin Nora~trenalin So~atostati
Serotonin s*
""
I
omnioqa I l
~ica~";~d
nociceptors • principally via small fibr (As and C) acupuncture • impulses
~ular
principally via large fibn (A~ and As)
PERIPHERAL NERVOUS SYSTEM Main pain pathway
Neurological pathway inhibiting acupuncture analgesia (AA)
Neurological pathway inhibiting acupuncture analgesia (AA)
Figure Some of the neurological connections between the brainstem, the spinal cord and peripheral nerves that are activated by acupuncture. (Adapted, with permission, from Bensoussan A. The Vital Meridian: A Modern Exploration o f Acupuncture. Edinburgh. Churchill Livingstone, 1991 .)
NEURAL MECHANISMS Denervation experiments have served to support the neural model, i.e. that acupuncture effects operate via the nervous system? ,9 An early review 1° of the neurophysiology of AA proposed the involvement of sensory-neural integration. The Figure highlights the pathways involved. An intact nervous system is required for effective AA? 1 Local anaesthetic to a nerve blocks the effects of acupuncture in the area served by the nerve, suggesting neural involvement, z2
Blocking deep tissue receptors can abolish the characteristic needling sensation associated with acupuncture and the analgesic effect? ° However, whether the selective blocking required to test this mechanism can be achieved remains questionable.
Sensory (afferent) pathways From a T C M perspective, for acupuncture to be effective, the needling sensation, or de Qi, must be elicited. From a physiological viewpoint, a high-threshold
104
Complementary Therapies in Medicine
receptor must be stimulated to produce the acupuncture effect?3 Han & Terenius ~4proposed that needling stimulates somatic muscle afferents, e.g. A-8 (group III) and even the smaller C (group IV) nerve fibres, activating central opioid production. Han z5has implicated both A-5 and the larger A-~ fibres in transmission. Experimentation with rabbits ~ suggested that stimulating the smaller afferent fibres produced the greatest analgesic effect.
Spinal level Primary afferent A-J3 fibres end in the dorsal column nuclei in the spinal cord. High-frequency stimulation, as evoked by transcutaneous electrical nerve stimulation (TENS), operates by this route. Such fibres produce collaterals (branches) which stimulate inhibitory interneurones. One result is inhibition of the nociceptive input of C-fibres ~7operating at a segmental level, i.e. within the same spinal cord segment. A-8 afferent nerves also make connections at dorsal horn level, where they elicit the release of enkephalins (analgesic substances: see Humoral mechanisms below) from interneurones. Enkephalins act segmentally within the substantia gelatinosa of the spinal cord to inhibit the effects of nociceptive input. A-8 fibres also synapse with nerve cells that convey information via the spinothalamic tract to the periaqueductal grey matter (PAGM) of the midbrain. From there, descending inhibitory control upon the enkephalinergic interneurones occurs. In a cat model, Wu et al ~8observed the discharge of spinal cord dorsal horn neurons to heat stimulus to be temporarily attenuated by electroacupuncture (EA). This suggests modulation of the afferent transmission of pain stimuli. Experiments where the spinal cord was differentially lesioned suggest that the anterolateral as well as the dorsal column may mediate AA. 1° Correspondence of trigger points, motor points and acupuncture points Liu and colleagues~9conducted a double-blind study to investigate the correspondence between certain motor points, i.e. where the nerve passes through the fascia into the muscle, and acupuncture loci. They concluded that about 70% of acupuncture points correspond to motor points. Further, Melzack and associates2° noted similarities between trigger points and common acupuncture points for pain in terms of spatial distribution and pain patterning. Although they stem from different conceptual bases, the empirical approach involved in linking points and the effect of applying pressure to them could be expected to give rise to some concordance. Whilst acupuncture meridians are anatomically non-existent, the precise genesis of trigger points is also unclear. Pressure at trigger points evokes localized and referred pain in a classical pattern. At the
time their article was written, the underlying neural processes could only be tentatively explored. The impact of acupuncture stimulation can be observed at various levels throughout the central nervous system (CNS). Models that incorporate inhibition via higher centres, e.g. the thalamus and the brainstem, are required; 2I the intraspinal level of Melzack and WaU's original gate control theory22 of pain modulation has been superseded. The relatively delayed onset of AA and its prolonged effect recorded during many animal experiments implies additional supraspinal activity?° A widespread, intersegmentally mediated analgesic effect can occur which refutes any hypothesis23that the analgesic effect occurs only within areas served by the spinal segment that received the pain stimulus.
Brainstem level: midbrain and medulla The brainstem also mediates AA. TENS-conveying nerves (A-[~ afferents) end chiefly in the thalamus, but collaterals to the midbrain exist. From the midbrain pretectal region there are connections with the PAGM?4 Much of A-6 afferent input is conveyed to the reticular formation, where modulation occurs. Chang ~° reported changes in the electrical discharge of reticular neurons during manual needling of the Tsusanli acupuncture point. The pattern was not characteristic of 'pain'; rather, the reticular formation acts more as a relay station. Descending inhibition may also originate from this region?5 Animal studies provide increasing evidence of a spino-reticulo-spinal feedback loop for the detection and modulation of pain. Bing et al2~ compared the responses of subnucleus reticularis dorsalis (SRD) neurons in the rat medulla to various noxious stimuli and to acupuncture stimulation. Neurons responded to stimulation at different acupuncture points, particularly those contralateral to them; acupuncture stimuli use a similar afferent route. Similar responses were elicited by acupuncture at non-acupuncture points, which accords with a lack of topographical specificity for the analgesic effects of acupuncture?7
Diencephalon level The thalamus contains nociceptors as part of its neuronal relay system.28Thalamic nociceptive outflow has been inhibited during acupuncture in animals. Changes in electrical discharge from certain thalamic nuclei have been seen following EA. 29 A weakened aversive behavioural response, i.e. withdrawal from a noxious stimulus, was also observed. Similar effects were reported using CNS stimulation in particular areas and following the administration of the analgesic, fentanyl. In another series of experiments, thalamic nociceptive discharge was inhibited both by stimulating the midbrain raphe nucleus and by the
Mechanism of acupuncture analgesia application of EA. 3° EA thus appears to influence thalamic transmission of noxious information. An intact pituitary gland is required for an analgesic effect to occur. 28The hypothalamus also seems to modulate AA. The arcuate nucleus receives stimulation from the prefrontal cortex; this may be a channel for the modulation of pain by emotional and cognitive interpretation. Lesions to the arcuate nucleus can reduce AAY ,32
Descending pathways Higher centres may exert a descending inhibitory gating effect.33The PAGM is central to pain modulation by both ascending and descending inhibition. 34 Stimulation of the PAGM enhanced AA in rats. 35 Recently, the evidence for somatotopic organization within the PAGM has been suggested as a possible explanation of the need for using specific heterosegmental acupuncture points. 13 The midbrain, in particular the nucleus raphe magnus (NRM), acts as a relay station for descending inhibitory messages, g6 Liu et aP 7 proposed that EA may activate the NRM, thereby inducing analgesia. Medullary lesion involving the NRM has been shown to reduce EA effects. 38 EA analgesia in cats was markedly attenuated by the transection of the dorsolateral columns of the spinal cord, an area that contains descending fibres from the NRM. 39A somatotopic relationship between areas of the raphe nuclei and the spinal dorsal horns may be somehow connected to the relative specificity of acupuncture points in producing regional analgesia. 4°
Diffuse noxious inhibitory control Diffuse noxious inhibitory control (DNIC) theory proposes a counter-irritant stimulation which acts to inhibit pain messages. Counter-irritation has been tested in both human and animal studies. Research has highlighted the need for a noxious afferent input stimulus. 26,4IDNIC may act as a filter, extracting pain signals en route to higher centres?2 In addition, the signals could amplify the response to incoming noxious stimuli. Opioid receptors and other neurotransmitters are involved, 42,43as will be subsequently discussed. The impact of what may be DNIC due to sham needling, plus the overall placebo effect, may confound investigations of the clinical effectiveness of AA.2,5, 4
DNIC-associated activity does not tally directly with the acupuncture response. DNIC effects tend to be transient, whilst the effects of acupuncture can be long-lasting. Secondly, acupuncture does not appear to necessarily involve C-fibre activation. TENS and EA seem to utilize primarily A-a and -13 afferents, with C-fibres only being activated at higher threshold levels.12
105
HUMORAL MECHANISMS During the 1970s, it became clear that the delayed onset and duration of acupuncture analgesic response necessitated a hypothesis for a humoral influence. There is a need for more investigation in this area, given its complexity. The humoral model addresses the role of neurotransmitters, hormones and other chemical substances that circulate in the blood and cerebrospinal fluid (CSF). Many acupuncture studies have involved chemical blockade or enhancement to assess the impact of various neurotransmitters.
Opioid activity He 45 reviewed the CNS circuitry through which acupuncture appears to evoke analgesic activity via the release of endogenous opiate peptides. Various CNS sites are rich in opioids and their receptors. These will be briefly reviewed in relation to acupuncture studies. The research can be divided into three main categories: 1. The measurement of opioid levels, e.g. enkephalins, during/following acupuncture 2. The enhancement of opioid levels during/following acupuncture 3. The inhibition of AA using the opiate (morphine) antagonist naloxone.
Measurement of opioid levels Raised CSF endorphin levels following EA have been reported. 46 All studies demonstrated pain relief. Differential opiate release occurred in a study of lowfrequency EA. 47 CSF I]-endorphins were raised; metenkephalin levels remained unchanged. However, others48reported raised plasma met-enkephalins and stable ~3-endorphins in their patients. In an animal experiment, increased CSF leu-enkephalins were observed following EA. 49 The PAGM is a focal site for various endogenous opioids and their receptors. Zhang et aP° investigated the hypothesis that acupuncture elicits PAGM opioid release. Opiod assays from PAGM perfusate increased during EA, concordant with the degree of analgesic effect produced. In addition, both opioid and cholinergic mechanisms in the caudate nucleus (one of the basal nuclei) have been proposed to operate during AA. 51'52Xie et aP 3 reported an association between caudate levels of met-enkephalin and the degree of AA produced in rats. The hypothalamic arcuate nucleus has been shown experimentally to assist in [3-endorphin-mediated analgesia?4 Other mediation sites include the preoptic area 55and parts of the limbic system?6
106
Complementary Therapies in Medicine
Enhancement of opioid levels The administration of bacitracin which inhibits degradation of endogenous opioids enhanced AA in rabbits? 7 Enkephalin levels were enhanced in the intervention group. This finding can be supported by rat experiments. Cyclohexamide, which acts to reduce met-enkephalin production, attenuated both the EAinduced increase in met-enkephalins and the analgesic effect, as measured by a tail-flick test in rats? 8 Zhou et aP9 report similar findings from radioimmunoassay studies, albeit with very small sample sizes. They hypothesized that acupuncture may raise enkephalin levels by accelerating opioid biosynthesis. One type of amino acid, the D-amino acids, block the enzymes that would normally degrade endorphins: they can prolong AA. 6° In one experiment, tail-flick response to painful stimuli in rats was inhibited by EA, morphine and PAGM stimulation?~ Administration of d-phenylalanine enhanced all three analgesics, suggestive of a common underlying mechanism.
Inhibition of AA using naloxone The drug naloxone competes for opioid receptors, blocking opioid activity. In animal models, 5°,6~67opioid involvement was inferred, given the reversal of AA and behavioural responses following naloxone administration. Similar findings have been reported from other animal models, where the behavioural response to nociceptive stimulation was observed. Injection of naloxone into the septal area, 68 the nucleus accumbens69 and the amygdala69,7° partially reversed the effects of both morphine and EA in rabbits. This work supports clinical findings from human tooth pulp stimulation experiments where AA was seen to be naloxone-reversible. 7~ Acupuncture appears to influence the release of endogenous opioids onto receptor sites to elicit analgesia. Specific receptors, namely the g-1 type, seem responsible for the mediation of analgesic effects.72 CXBX-strain mice who are deficient in opioid receptors have a poor analgesic response to EA. 73 Not all the research favours an opioid explanation. Chapman and co-workersTM failed to reverse AA using naloxone in a tooth pulp stimulation experiment. In a small randomized double-blind trial of patients with chronic pain, no change in the pain relief afforded by manual acupuncture was encountered in those given naloxone.75 When comparing studies, however, in addition to noting the naloxone dosage used (and possibly the time of administration), the type of stimulation and the points chosen must be considered, since these could also modify outcome. Cheng & Pomerantz 62 reported that whilst analgesic effects could be produced by EA at both high (200 Hz) and low (4 Hz) frequencies, only the analgesia under the latter condition was reversed by naloxone.
Another set of experiments may also clarify reasons for the discrepant findings. Increased levels of both endogenous opioids and serotonin (5-HT) have been associated with acupuncture. A lowering of either one of these substrates resulted in a compensatory increase in the other.76 A reduced analgesic effect generally occurred with both naloxone and a 5-HT synthesis inhibitor. Pharmacological interference with both substances almost abolished analgesic effects. Other humoral activity
Over 20 neurotransmitters have been associated with the response to acupuncture. The key players are: serotonin (5-HT); dopamine (DA); noradrenaline (NA); and somatostatin, in particular CNS regions. (It should be recalled, however, that neurotransmitters tend to be found in some quantity in most areas of the brain.) Serotonin is thought to have a largely inhibitory effect. (However, in terms of specificity, it should be noted that, clinically speaking, it is the non-specific tricyclics that act as analgesics, rather than, say, SSRIs). Blood serotonin levels change with acupuncture intervention; the increase may be due to the release of a precursor or associated pituitary activity, since serotonin does not cross the blood-brain barrier. 77 Jin et alTM demonstrated enhanced 5-HIAA (indicative of catecholamine metabolism per se) and tryptophan (indicative of 5-HT activity) levels with EA and suggested that EA enhances 5-HT activity. From their experiments, Zhu & Shi79proposed that acupuncture activates the chain of events from opioid, and to a lesser extent 5-HT and NA release in the PAGM to descending 5-HT influence from the PAGM to the NRM. Han et aP° demonstrated this experimentally in a double-blind study using clomipramine, a 5-HT reuptake inhibitor. Several authors 78'8~have used 5, 6-DHT to deplete serotonin and have noted an attenuation of EA effect. Interestingly, the impact of acupuncture can be less in a depressed patient?2 This may accord with research that suggests that patients suffering from depression have lower serotonin levels?3 Painful impulses are also modulated, both positively and negatively, by catecholamines: DA, NA and somatostatin. In experiments, AA was enhanced following the introduction of exogenous cAMP and ATW4 These substances may activate intermediary hypothalamic-pituitary protein hormones. In a series of experiments, serotonin and catecholamine levels were measured during acupuncture. TM EA appeared to activate dopaminergic (DA) neurons. The analgesic effects of EA were enhanced using apomorphine, a DA agonist and reduced by 5,6-DHT, a dopamine receptor (D2) antagonist. Also, AA was accompanied by a reduction in brain NA levels. Unfortunately, interpretation of the findings is limited due to the small sample sizes.
Mechanism of acupuncture analgesia Cholinergic mechanisms may also be involved in AA, the caudate nucleus being a key areaY ,85 An association between AA and raised levels of acetylcholine (ACh) has been reported. 85 Ge et aP 6 conducted histochemical analysis of rat brain after acupuncture. They found raised levels of acetylcholinesterase (ACHE, the enzyme which aids ACh degradation) in the locus coeruleus and raphe nucleus, compared to a control group. Atropine (ACh antagonist) decreased the analgesic effect of acupuncture. However, given the widespread activity of ACh within the CNS, it is difficult to assess any specific role it might have. Analgesic antagonists include substance p,30cholecystokinin and cAME 8°'87Yonehara et aP8 suggested that EA inhibits substance P release evoked by tooth pulp stimulation in animals. They discussed the possibility that EA exerts an opioid inhibitory effect on substance P at the level of the trigeminal nucleus (brainstem region) and, via a relay system, may elicit monoamine release, with a consequent inhibitory, descending influence. Other experiments showed AA to be antagonized by gamma-amino butyric acid (GABA) release from the diencephalon region? 9 Antagonistic monoamine (DA, NA) involvement, chiefly via the locus coeruleus, may also occur. Dopaminergic influences may antagonize the opioid effect in the caudate nucleus. 90 Inhibition of adrenergic (sympathetic) activity by acupuncture may account for some of the analgesic pattern. Cao et a191 reported corroboratory physiological evidence following acupuncture. However, at spinal level, NA may have a facilitatory effect22 Catecholamines act differentially; AA was enhanced following direct microinjection of NA into the ventromedial hypothalamus. TM Ge et a186histochemically compared catecholamine levels in the locus coeruleus and raphe magnus using tissue from acupunctured animals and controls. Catecholamine levels appeared to be lower in the former group, perhaps as a result of increased monoamine oxidase activity (the enzyme which assists catecholamine degradation). The authors cite clinical evidence of a decrease in monoamines during acupuncture anaesthesia and reserpine (a monoamine antagonist) experiments, wherein analgesia has been enhanced. They conclude that the findings have been mixed, probably due to the various functions of the locus coeruleus, but favour a limited role for catecholamines in AA. Several experimenters 7° have sought to develop an understanding of the opposing roles of monoamines using antagonist drugs, i.e. cz- and [3-blockers. The evidence suggests that antagonistic (non-analgesic) functions operate via o~2 receptors, whilst analgesic effects operate via 13 receptors, the overall balance favouring the former activity.
107
Donor experiments Some interesting experiments have demonstrated the importance of circulating chemicals by giving acupuncture to one animal and then donating the body fluid to a recipient animal, with therapeutic outcome occurring in both? 2,93~6Unfortunately, methodological details appear to have been lost in the translation from Chinese.
BIOELECTRIC AND BIOMAGNETIC MEDIATION Bioelectric mediation supposes that the meridians are electrically distinct entities. Bioelectric changes associated with the meridians may precede any neurological or humoral reactivity to acupuncture. Acupuncture points correspond well with areas of low resistance (high permeability) around the body.97-99The resistance at a particular point seems to be lowered further in disease conditions. Many modern acupuncturists make use of this phenomenon in detecting the acupuncture points to use. Places with the greatest electromagnetic activity/bioluminescence correspond to TCM acupuncture points. 1°° Kirlian photography has been used to measure bioluminescence. This is thought to reflect the energy flow along the meridians, although evidence for its utility remains equivocal? °~Attempts to measure changes in the body's biomagnetic fields during acupuncture have been limited because the body's magnetic fields are very weak and are not readily detectable. 1°2It remains to be seen whether meridian flow is associated with charge movement during acupuncture. More recently, it has become possible to detect acupuncture meridians using radioisotopes. Darras et al ~°3 reported upon the use of technetium-99m to investigate acupuncture pathways. A particular migratory pattern was seen from various acupuncture points, which accorded with the traditional channels. Furthermore, these paths were distinct from the vascular and lymphatic circulation, in terms of rate and direction of flow. Concentrated areas of the tracer related not only to organs known to take up the tracer, but also to acupuncture points along a given meridian. The migratory pattern appeared to be altered in pathological conditions. In addition, changes in the tracer's migration rate from contralateral points of the body followed unilateral stimulation of an acupuncture point on the same meridian. The authors proposed that acupuncture stimulation evokes an effect via a neurochemical process. The pathways may be related to connective tissue diffusion following the neurovascular bundles along the extremities. This interstitial route needs further investigation.
108
Complementary Therapies in Medicine
Differential responses to various modes of acupuncture stimulation
Bioelectrical mediation may partially explain why acupuncture outcome may vary depending upon the type of stimulation employed, e.g. manual or electroacupuncture. 1°4 Furthermore, research evidence suggests that the intensities and frequencies used in EA may produce subtly different physiological outcomes. ~4,62,~°5-1°7In a therapeutic setting, therefore, one technique may be more appropriate than another.
effectiveness of acupuncture will be enhanced by the measurement of physiological parameters to assess both the process and outcomes. This will aid the comparison of acupuncture and alternative analgesics and may assist in the provision of effective, individualized treatment.
ACKNOWLEDGEMENTS The author would like to thank J. Tissier, R. Walker and J. Sizer for their valuable comments on this article.
Alternative hypotheses
Acupuncture analgesia as a generalized response to stress Some critics have suggested that AA may be purely a generalized response to trauma. However, researchers have described the distinctions between acupuncture and stressful stimuli, highlighting different processes a n d outcomes. 56'84'1°4'1°8111
Alternative hypotheses: hypnotic suggestibility Some researchers contest that acupuncture effect is due to a hypnotic-type suggestibility. Knox & Shum ~2 were unable to reverse hypnotic analgesia using naloxone, an indication that at least one different mechanism operates. There was some intra-individual commonality of responsiveness to acupuncture and hypnotism, however. Others have found this to be unrelated to analgesic outcome. 1~3
CONCLUSION Whilst both neural and humoral mechanisms are important, neither can fully explain such a fundamental phenomenon as differential point efficacy in certain painful conditions. TM There may be no distinct neural or circulatory difference between a real point and a nearby sham point, yet needling outcomes can be different. Needling changes the electrical activity at points and along channels; this may serve to correct a homeostatic imbalance. The clinical effectiveness of acupuncture per se is said to depend upon the specific placement of the needles at designated acupuncture points. For analgesic activity, point specificity appears to be less crucial than for other types of therapeutic intervention, e.g. for vomiting. Hence DNIC may be a valid model for partially explaining AA. The therapeutic benefits of AA may relate to activity other than neurological blockade, e.g. decreasing gastric acidity or eliciting vasodilation. H5 'Correct acupuncture, properly directed to the management of pain, will also increase oxygenation of cardiac muscle, reduce gastric acidity and control inflammatory responses. '7 Hopefully, future studies of the clinical
REFERENCES 1. Lewith GT, Vincent C. On the evaluation of the clinical effects of acupuncture: a problem reassessed and a framework for future research. J Alt Compl Med 1996; 2:79 90. 2. Lewith GT. How effective is acupuncture in the management of pain? J R Coll Gen Pract 1984; 34: 275-278. 3. Richardson PH, Vincent CA. Acupuncture for the treatment of pain: a review of evaluative research. Pain 1986; 24: 1540. 4. Patel M, Gutzwiller F, Paccaud F, Marazzi A. A metaanalysis of acupuncture for chronic pain. Int J Epidemiol 1989; 18:900 905. 5. ter Riet G, Kleijnen J, Knipschild R Acupuncture and chronic pain: a criteria-based meta-analysis. J Clin Epidemiol 1990; 43: 1191-1199. 6. Vincent CA, Richardson PH. Acupuncture for some common disorders: a review of evaluative research. J R Coll Gen Pract 1987; 37: 7%81. 7. Bensoussan A. Contemporary acupuncture research: the difficulties of research across scientific paradigms. Am J Acupunct 1991; 19: 357-365. 8. Bossy J. Morphological data concerning acupuncture points and channel network. Acupunct Electrother Res 1984; 9: 76-106. 9. Shen E. Participation of descending inhibition in acupuncture analgesia. In: Zhang XT ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 31-38. 10. Chang HT. Neurophysiological basis of acupuncture analgesia. Scientia Sinica 1978; 21: 829-846. 11. Levy B, Matsumoto T. Pathophysiology of acupuncture: nervous system transmission. Am Surg 1975; 41: 378-384. 12. Chiang CY, Chang CT, Chu H, Yang L. Peripheral afferent pathway for acupuncture analgesia. Scientia Sinica 1973; 16: 210-217. 13. Bowsher D. The physiology of stimulation-produced analgesia. Acupunct Med 1991; 9: 58-62. 14. Han JS, Terenius L. Neurochemical basis of acupuncture analgesia. Annu Rev Pharmacol Toxicol 1982; 22: 193-220. 15. Han JS. On the mechanism of acupuncture analgesia. Acupunct Res 1984; 3: 236-245. 16. Chen L, Tang J, Fan L e t al. An analytical study of the afferent fibres for impulses of acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 478489. 17. Duggan AW, Foong FW. Bicuculline and spinal inhibition produced by dorsal column stimulation in the cat. Pain 1985; 22: 249-259. 18. Wu J, Xing J, Xing Bet al. Inhibitory action of electroacupuncture on dorsal horn neurons of the spinal cord. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986: 3949. 19. Liu Y, Varela M, Oswald R. The correspondence between some motor points and acupuncture loci. Am J Chin Med 1975; 3: 347-358. 20. Melzack R, Stillwell DM, Fox EJ. Trigger points and acupuncture points for pain: correlations and implications. Pain 1977; 3: 3-27.
Mechanism of acupuncture analgesia 21. Melzack R, Jeans ME. Acupuncture analgesia: a psychophysiological explanation. Minn Med 1974; 57: 161-166. 22. Melzack R, Wall PD. Pain mechanisms. A a new theory. Science 1965; 150:971 979. 23. MacDonald AJR. Segmental acupuncture therapy. Acupunct Electrother Res 1983; 8: 267-282. 24. Rees H, Roberts MHT. Antinociceptive effects of doral column stimulation in the rat: involvement of the anterior pretectal nucleus. J Physiol 1989; 417: 375-388. 25. Oliveras JL, Besson JM, Guilbad G, Liebeskind JC. Behavioural and electrophysiological evidence of pain inhibition from midbrain stimulation in the cat. Exp Brain Res 1974; 20: 3244. 26. Bing Z, Villaneuva L, Le Bars. Acupuncture-evoked responses of subnucleus reticularis dorsalis neurons in the rat medulla. Neuroscience 1991; 44: 693-703. 27. Lewith GT, Machin D. On the evaluation of the clinical effects of acupuncture. Pain 1983; 16:111-127. 28. Takeshige C. Differentiation between acupuncture point and non-point by association with analgesia inhibitory system. Acupunct Electrother Res 1985; 10: 195-203. 29. Zhao ZQ, Shao DH, Yang ZQ. Inhibition of evoked discharges of posterior nuclear group of thalamus induced by stimulation of central grey matter, caudate nucleus and electroacupuncture in waking rabbits. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986. 30. Zhang G, Wei J, Zhou L e t al. Effect of stimulation of midbrain raphe nucleus on discharges of nociceptive neurons in the nucleus parafascicularis of the thalamus and its significance in acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986: 82-92. 31. Gao YS, Ku YH. Mechanism underlying the excitatory effect of nucleus arcuatus hypothalami on PAG-NRM system and its significance in electroacupuncture in rats. Acta Physiol Sinica 1983; 35: 409415. 32. Quo SY, Yin WP, Zhang HQ, Yin QZ. Role of hypothalamic arcuate region in lip acupuncture analgesia. Acta Physiol Sinica 1982; 34: 71-77. 33. Andersson SA, Ericson T, Holmgren E, Lindqvist G. Electroacupuncture effect on pain threshold measured with electrical stimulation of teeth. Brain Res 1973; 63: 393-396. 34. Fields HL. An endorphin-mediated analgesia system: experimental and clinical observations In: Martin JB et al, eds. Neurosecretion and brain peptides. New York: Raven Press, 1981: 199-212. 35.Huang C, Li XC, Zhang CC et al. Effect of electrical stimulation of midbrain periaqueductal gray on acupuncture analgesia in rats. Acta Physiol Sinica 1982; 34: 343-357. 36. Behbehani MM. Evidence that an excitatory connection between the PAG and nucleus raphe magnus mediates stimulation produced analgesia. Brain Res 1979; 170: 85-93. 37. Liu X, Zhu B, Zhang SX. Relationship between electroacupuncture analgesia and descending pain inhibitory mechanism of nucleus raphe magnus. Pain 1986; 24: 383-396. 38.Du HJ, Chao Y E Localization of central structures involved in descending inhibitory effect of acupuncture on viscero-somatic reflex discharge. Scientia Sinica 1976; 19: 137-148. 39. Shen E, Ma WH, Lan C. Involvement of descending inhibition in the effect of acupuncture on the splanchnically evoked potential in the orbital cortex of the cat. Scientia Sinica 1978; 21: 677-685. 40. Ju G, Shu S. Somatotopic projection of the raphe nuclei to spinal dorsal horn in the cat - a horseradish peroxidase study. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986:516 528. 41. Le Bars D, Villaneuva L, Willer JC, Bouhassira D. Diffuse noxious inhibitory controls (DNIC) in animals and in man. Acupunct Med 1991; 9: 47-56. 42. Le Bars D, Chitour D, Kraus E et al. Effect of naloxone upon diffuse noxious inhibitory controls (DNIC) in the rat. Brain Res 1981; 204: 387402. 43.Wilier JC, Broucker T, Le Bars D. Diffuse noxious inhibitory controls (DNIC) in man: involvement of an opioidergic link. Eur J Pharmacol 1989; 182: 347-355.
109
44. Needham J, Lu GD. Celestial lancets: a history and rationale of acupuncture and moxibustion. Cambridge: Cambridge University Press, 1980: 261. 45. He L. Involvement of endogenous opiod peptides in acupuncture analgesia. Pain 1987; 31: 99-121. 46. Sjolund B, Terenius L, Eriksson M. Increased CSF levels of endorphins after electroacupuncture. Acta Physiol Scand 1977; 100: 382-384. 47. Clement-Jones V, McLoughlin L, Tomlin Set al. Increased beta-endorphin but not met-enkephalin levels in human cerebrospinal fluid after acupuncture for recurrent pain. Lancet 1980; ii(8201): 946-948. 48. Kiser RS, Khatami M J, Gatchel RJ et al. Acupuncture relief of chronic pain syndrome correlates with increased plasma met-enkephalin concentrations. Lancet 1983; ii(8364): 1394-1396. 49. Huang Y, Wang QW, Wan FS et al. Increase of CSF enkephalin content by electroacupuncture in monkeys. Kexue Tongbao 1981; 26:631 632. 50.Zhang AZ, Pan XP, Xu SF et al. Endorphins and acupuncture analgesia. Chin Med J 1980; 93:673 680. 51. He LF, Xu SE Caudate nucleus and acupuncture analgesia. Acupunct Electrother Res 1981; 6:169-182. 52. He L, Xu S, Jiang C. Role of caudate nucleus in acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press 1986: 146-157. 53.Xie CW, Zhang WG, Hong XJ, Han JS. Relation between the content of central met-enkephalin and leu-enkephalin and analgesic effect of electroacupuncture in rats. Acta Physiol Sinica 1984; 36: 192-197. 54. Yin QZ, Duanma ZX, Guo SY, Yu XM, Zhang YJ. Role of the hypothalamic arcuate nucleus in acupuncture analgesia. Second National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing 1984: 398. 55.Wu GC, Jiang JW, Cao XD. Effect of microiontophoretic etorphine, noradrenaline and electroacupuncture on spontaneous discharge of neurons in rabbits preoptic area. Acta Physiol Sinica 1984; 36: 552-558. 56. Zhang AZ. Endorphin and acupuncture analgesia research in the People's Republic of China (1975-1979). Acupunct Electrother Res 1980; 5:131 136. 57. Tsou K, Zhao DD, Wu SH et al. Enhancement of acupuncture analgesia with concomitant increase of enkephalins in striatum and hypothalamus by intraventricular bacitracin. Acta Physiol Sinica 1979; 31:377-381. 58.Wu SH, Chang TS, Wan FS, Tsou K. Effect of intraventricular cyclohexamide on acupuncture analgesia and brain met-enkephalin. Acta Physiol Sinica 1980; 32:79-81. 59. Zhou G, Yi Q, Wu S et al. Enkephalin involvement in acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986: 267-278. 60. Cheng RSS, Pomerantz B. A combined treatment with Damino acids and electroacupuncture produced a greater analgesia than with either treatment alone: naloxone reverses these effects. Pain 1980; 8:231 236. 61. Takeshige C, Murai M, Tanka M, Hachisu M. Parallel individual variations in effectiveness of acupuncture, morphine analgesia and dorsal PGA-SPA and their abolition by d-phenylalanine. In: Bonica JJ et al, eds. Advances in pain research, vol 5. New York: Raven Press 1983: 563-569. 62. Cheng RSS, Pomerantz B. Electro-acupuncture analgesia could be mediated by at least two pain-relieving mechanisms; endorphin and non-endorphin systems. Life Sci 1979; 25: 1957-1962. 63. Lou AL, Zheng CQ, Bai HQ. Influence of microinjection of naloxone and procaine into nucleus raphe magnus on acupuncture effect of visceral traction response in dogs. Acta Physiol Sinica 1982; 34: 84-89. 64. Huang Y, Wang QW, Zheng JZ et al Analgesic effect of acupuncture and its reversal by naloxone as shown by method of conditioning responses in monkeys. Chung Hua I Hsenh Tsa Chih 1978; 58:1193 1195. 65. Ha H, Tan EC, Fukunaga H, Osamu A. Naloxone reversal of acupuncture analgesia in the monkey. Exp Neurol 1981; 73: 298-303.
110
Complementary Therapies in Medicine
66. Pomerantz B, Chiu D. Naloxone blockade of acupuncture analgesia: endorphin implicated. Life Sci 1979; 19: 1757-1762. 67. Fu TC, Halenda SP, Dewey WL. The effect of hypophysectomy on acupuncture analgesia in the mouse. Brain Res 1980; 202: 33-39. 68. Mo WY,, Chou YF, Wang GN. Analgesia by injection of morphine and antagonism of electroacupuncture analgesia by injection of naloxone into septal area of rabbits. Chung Kuo Yao Li Hseuh Pao 1982; 3: 150-154. 69.Zhou ZF, Du MN, Wu WY et al. Effect of intracerebral microinjection of naloxone on acupuncture- and morphine-analgesia in the rabbit. Scientia Sinica 1981; 24: 1166-1178. 70. Zhou ZF, Du MY, Wu WY, Han JS. The effect of injection of naloxone into habenula or amygdala on acupuncture analgesia in the rabbit. National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1979: 504. 71. Mayer D J, Price DD, Rafii A. Antagonism of acupuncture analgesia in man by narcotic antagonist naloxone. Brain Res 1977; 121: 368-372. 72. Pasternak GW. Central mechanism of opioid analgesia. Acupunct Electrother Res 1981; 6: 135-149. 73. Peets JM, Pomerantz B. CXBK mice deficient in opiate receptors show poor electroacupuncture analgesia. Nature 1978; 273: 675-676. 74. Chapman C, Benedetti C, Colpitts Y, Gerlach R. Naloxone fails to reverse pain thresholds elevated by acupuncture: acupuncture analgesia reconsidered. Pain 1983; 16: 13-31. 75. Kenyon JN, Knight C J, Wells C. Randomised double-blind trial on the immediate effects of naloxone on classical Chinese acupuncture therapy for chronic pain. Acupunct Electrother Res 1983; 8: 17-24. 76. Han JS, Tang J, Fan Set al. Central 5-hydroxytryptamine, opiate-like substances and acupuncture analgesia In: Zhang XT ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 309-316. 77. Malizia E, Andreucci G, Paolucci D et al. Electroacupuncture and peripheral beta-endorphin and ACTH (letter). Lancet 1979; 2: 535-536. 78. Jin G, Zhang Z, Yu L e t al. Role of brain serotonergic and catecholaminergic systems in acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 317-331. 79. Zhu LX, Shi QY. Activation of nucleus raphe magnus by acupuncture and enkephinergic mechanisms. J Tradit Chin Med 1984; 4: 111-113. 80. Han JS, Tang J, Ren MF et al. Central neurotransmitters and acupuncture analgesia. Am J Chin Med 1980; 8:331 348. 81.Du HJ, Shen E, Dong XW et al. Effect on acupuncture analgesia by (sic) injection of 5,6-dihydroxytryptamine in cat: a neurophysiological, neurochemical and fluorescence histochemical study. Acta Zool Sinica 1978; 24:11-20. 82. MacDonald AJR. Getting under the skin. Nurs Times 1986; 14: 47-48. 83.Van Praag HM. Depression. Lancet 1983; 2: 912. 84.Liu X, Yang Z, Li Q et al. The role of hypothalamuspituitary-adrenal system in acupuncture analgesia In: Zhang XT ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press 1986: 397407. 85.Wang CY, Yu B, Liu XC. The influence of electroacupuncture on the ACh levels in various regions of rat brain. National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1979: 453. 86.Ge Z, Huang W, Wang Z, Song W. The relation between acupuncture anaesthesia and neurotransmitters and enzymes of locus coeruleus and raphe nucleus. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 558-567. 87. Zhou Y, Sun YH, Shen JM, Han JS. Increased release of immunoreactive CCK-8 by electroacupuncture and enhancement of electroacupuncture analgesia by CCK-B antagonist in rat spinal cord. Neuropeptides 1993; 24: 139-144.
88. Yonehara N, Sawada T, Matsuura H, Inoki R. Influence of electroacupuncture on the release of substance P and the potential evoked by tooth pulp stimulation in the trigeminal nucleus caudalis of the rabbit. Neurosci Lett 1992; 142: 53-56. 89. Zhu JQ, Li CB, Xu F, Liang WJ. Relationship of the metabolism of GABA in mice brain to electroacupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anesthesia. Beijing: Science Press, 1986. 90.Xu SF, Lu WX, Zhou GZ, Sheng MP, Wang DL. Effects of cholinergic and dopaminergic systems of the caudate nucleus in acupuncture analgesia. Second National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1984: 471. 91. Cao XD, Xu SF, Lu WX. Inhibition of sympathetic nervous system by acupuncture. Acupunct Electrother Res 1983; 8: 25-35. 92. Cao XD, He LF, Xu XF, et al. The role of PAG in acupuncture analgesia. Second National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1984: 639. 93.Yang MMP, Kok SH. Further study of the neurohumoral factor endorphin in the mechanism of acupuncture analgesia. Am J Chin Med 1979; 7: 143-148. 94. Lung Ch, Sun AC, Tsao CJ et al An observation of the humoral factor in AA in rats. Am J Chin Med 1974; 2: 203-205. 95. Beijing Research Group of Acupuncture Anaesthesia. The role of some neurotransmitters of brain in finger acupuncture analgesia. Scientia Sinica 1974; 17:112-130. 96. Chen M, Zheng Y, Xue Z et al. Study of role of humoral factors in central action of acupuncture analgesia. In: Zhang XT, ed. Research on acupuncture, moxibustion and acupuncture anaesthesia. Beijing: Science Press, 1986: 360-371. 97. Rabischong P, Niboyet JE, Terral C et al. Experimental basis of acupuncture analgesia. Nouv Presse Med 1975; 4: 2021-2026. 98. Reichmanis M, Marino AA, Becker RO. Electrical correlates of acupuncture points. IEEE Trans Biomed Eng 1975; Nov: 533-535. 99. Zhu ZX. Research advances in the electrical specificity of meridians and acupuncture points. Am J Acupunct 1981; 9: 203-216. 100. Luciani RJ. Direct observations and photography of electroconductive points on the human skin. Am J Acupunct 1978; 4: 311-317. 101. Van der Schaar W~. Organ pathology, diagnosis and acupuncture. Am J Acupunct 1976; 4: 252. 102. Jessel-Kenyon J, Cheng N, Blott B, Hopwood V. Studies with acupuncture using a SQUID biomagnetometer: a preliminary report. Compl Med Res 1992; 6: 142-151. 103. Darras J-C, de Vernejoul R Albarede R Nuclear medicine and acupuncture: a study on the migration of radioactive tracers after injection at acupoints. Am J Acupunct 1992; 20: 245-255. 104. Nappi G, Facchinetti F, Leguante G e t al. Different releasing effects of traditional manual acupuncture and electroacupuncture on propriocortin related peptides. Acupunct Electrother Res 1982; 7: 93-103. 105. Peets JM, Pomerantz B. Acupuncture-like transcutaneous electrical nerve stimulation analgesia is influenced by spinal cord endorphins but not serotonin. In: Advances in Pain Research and Therapy (9). New York: Raven Press, 1985: 519-525. 106.Han JS, Xie GX, Ding XZ, Fan SG. High and low frequency electroacupuncture analgesia are mediated by different opioid peptides. Pain 1984; (suppl) 2: 543. 107. Levine MF, Hui-Chan CWY. Conventional and acupuncturelike transcutaneous electrical nerve stimulation excite similar afferent fibers. Arch Phys Med Rehabil 1993; 74: 54-60. 108. Wei RS. Investigations on the mechanisms and therapeutic effects of acupuncture treatments. N E Med J Chin 1952; 2: 44-53. 109.Wen WL, Ho WKK, Wong HK et al. Changes in ACTH and cortisol levels in drug addicts treated by a new and rapid detoxification procedure using acupuncture and nalo×one. Chin Med E W 1978; 6: 241-246.
Mechanism of acupuncture analgesia
110. Pomerantz B. Relation of stress-induced analgesia to acupuncture analgesia. Ann N Y Acad Sci 1986; 467: 4~A A~7. 111. Kho HG, Kloppenborg PWC, van Egmond J. Effects of acupuncture and transcutaneous stimulation analgesia on plasma hormone levels during and after major abdominal surgery. Enr J Anaesthesiol 1993; 10: 197-208. 112. Knox VJ, Shum K. Reduction of cold pressor pain with acupuncture analgesia in high and low hypnotic subjects. J Psychol 1977; 86: 639-643. 113. Xu L, Fu Z, Xiang M e t al. The effects of acupuncture analgesia and its relation to blood endorphin, blood
111
histamine and suggestibility. National Symposium of Acupuncture and Moxibustion and Acupuncture Anaesthesia, Beijing, 1979. 114. Takeshige C. The central mechanism of acupuncture in acupuncture anaesthesia: differentiation of acupuncture point and non-point by the central analgesia producing system. Acupunct Electrother Res 1983; 8: 323-324. 115.National Health and Medical Research Council. Report of the working party on acupuncture. Canberra: Australian Government Publishing Services, 1988.