Introduction

In this subchapter, treatments for neurogenic pain, head and facial pain, and neck pain are described. The focus is on electrical stimulation, but some information is included on laser acupuncture (LA) and low-intensity laser therapy (LILT), which have been used for many of the conditions covered.

Neurogenic pain

Neurogenic pain may be due to nerve injury, neuritis (nerve inflammation) or degenerative (non-inflammatory) neuropathy, peripheral or central in origin. Some neurogenic pain may accompany or follow infection.¹ It can take the form of neuralgia (paroxysmal or fulminating pain along the course of one or more nerves), or be less severe but prolonged, sometimes with a correspondingly increased or decreased excitability to electrical methods of stimulation.² Neurogenic paraesthesiae (distorted sensations) may include burning, ‘pins and needles’ or numbness.

Neurogenic pain may be associated with reduced variability of electrical activity in the thalamus (ventral caudal nucleus), but also increased activity (frequency) in other thalamic regions (see SubCh. 5.1). Interestingly, referred sensations on electrical stimulation of ear points are not uncommon with neurogenic pain.³

The neuralgias

Acupuncture has been used for neuralgia in the West since at least the 1850s.⁴

EA in particular is considered to be appropriate for treating neuralgic pain according to one respected acupuncture textbook,5 although Bruce Pomeranz at one time concluded that it was not superior to placebo for neuralgia (as opposed to many other sorts of pain).6

In a retrospective review of pTENS for chronic pain, it had positive effects in 63.2% of those with neuralgia.7

In one large retrospective Japanese review (N = 11648), 13.9% of respondents had used magnets as a treatment for neuralgia.8

Microwave acupuncture has been used for neuralgia, both in humans⁹ and in animals (see Ch. 10, ‘used for neuralgia’).

Ultrasound acupuncture has been used for neuralgic pain.¹⁰

Points used

Anton Jayasuriya has found that auricular LA can be helpful for neuralgia.¹¹

Parameters used

Voll, in his EAV system, proposed that 3.9 Hz acupoint stimulation should generally be used for neuralgia, although it is unclear why he adopted this particular frequency.¹²

In a review of TENS for pain, Thomas Lundeberg commented that neuralgia tends to respond well to CTENS.¹³

Postherpetic neuralgia (PHN)

Herpes zoster and its treatments are described elsewhere (see SubCh. 9.4). Following the acute stage, postherpetic neuralgia can develop if the virus has not been adequately contained.¹⁴ This is the most common and dreaded complication of herpes zoster.¹⁵ Although PHN resolves spontaneously within 3 months in around 50% of cases, it may persist in 14%–22%.^16,17 Like herpes zoster itself, PHN tends to be more of a problem in older people (it is uncommon in those under 60 years¹⁸), and once it has become chronic can be very difficult to treat whatever method is used.

Conventional treatment includes sympathetic blockade,¹⁹ antidepressant medication such as amitriptyline and the anticonvulsant carbamazepine,²⁰ although other drugs such as the NMDA blockers ketamine and simple magnesium have also been investigated.^21,22 As so often, the quality of many studies and reviews of conventional treatments for PHN has been criticised.²³ The conclusion of one systematic review of RCTs for PHN was that tricyclic antidepressants appear to be the only agents of proven benefit once PHN is established.²⁴ Preventive use of tricyclics at the onset of the acute phase has also been attempted,²⁵ and virustatics such as acyclovir may reduce time to pain relief²⁶ if not altogether prevent subsequent PHN.²⁷ Overmedication is a risk in the elderly.²⁸ Topical agents such as capsaicin and lidocaine may be helpful according to one informal review.²⁹

PHN most commonly affects the ophthalmic division of the trigeminal nerve (see SubCh. 9.4) and midthoracic dermatomes, with least probability of pain relief for PHN of the isolated ophthalmic nerve and of the brachial plexus, and greatest when it involves the jaw, neck and trunk.³⁰ The pain experienced may be both steady and paroxysmal. Sensory loss, allodynia and dysaesthesia (disturbances of sensation) are also common, whereas hyperalgesia (pain of abnormal severity following noxious stimulation³¹) is less so. There may be a better chance of recovery if allodynia or sensory deficit is absent.³²

In one observational study by an experienced practitioner,33 MA was not found to be useful for patients with PHN who were older than 65, who experienced severe pain or who had been suffering for longer than 6 months (indeed, any form of treatment after about 6 months is likely to help only a minority of patients34). However, MA was useful in other patients, more specifically if they were treated early in the course of the disease, and other clinicians have also reported that MA can be helpful for PHN,35 for instance using simple static needle retention.36 However, treatment may need to be continued for quite some time37 (the author of this report did not in fact specify whether MA or EA was used), and results are not always good.38 Indeed, in one RCT MA was found to be no more helpful than placebo for PHN.39 At least two reviews have concluded on the basis of this that acupuncture in general is ineffective for the condition.40,41

EA has been used in a number of studies for PHN,42,43,44,45,46,47 although in one retrospective review of low-frequency (LF) EA it was not considered particularly helpful.48 As with MA, more chronic PHN becomes less amenable to treatment.49

According to Noordenbos,50 the herpes zoster virus tends to destroy more of the larger myelinated afferent nerve fibres, leaving an excessive proportion of small A- δ fibres and non-myelinated fibres still functioning. Peter Nathan and Patrick Wall, on the basis of Noordenboos’s hypothesis, considered that CTENS would stimulate the remaining larger fibres and reduce pain. Indeed they found that, although short treatment did not have a great effect, 12 hours or more of CTENS did indeed alleviate PHN.51 Thus TENS52,53,54,55,56,57,58,59 and the independently developed Japanese variant, the ‘neuro-softer’,60 have been used for PHN. Indeed, in 1991 amitriptyline, topical capsaicin and TENS were considered the best therapy currently available for PHN.61 However, one Italian group found good results with TENS only if used within 3 months after onset62 (poor prognosis thereafter may be due to degenerative changes in the dorsal root ganglia63). Another Italian researcher observed no additional benefits with TENS at all when patients were already taking tricyclic and neuroleptic medication.64 In one report, facial PHN responded better to TENS than PHN on the trunk.65 Long-term results with CTENS, as with many conditions, are not as impressive as short-term benefits.66,67

LILT has been used for PHN,68,69,70,71 although some reviewers have inevitably questioned its effectiveness.72 LA together with focal irradiation was utilised by one author.73,74 LA reduced local hyperaesthesia in another study.75 ‘Colorpuncture’ (coloured light directed at acupoints, see Ch. 10) has also been used.76

Ultrasound has been used for PHN,77 although rarely.

Local electrical stimulation can sometimes aggravate PHN. If this occurs, ‘electroacupressure’ (see Ch. 12) may be a gentler and more appropriate approach.

As mentioned above, an intervention that is often used for this condition is nerve block, using different agents and injection sites according to the location of the pain. So, for example, Gasserian ganglion block has been successful in cases of ophthalmic PHN. Electrical stimulation of the ganglion via a needle electrode before the block gave temporary but complete relief in this case, although both MA and TENS had only been of slight benefit.78 Stimulation of the Gasserian ganglion has also been carried out via an implanted electrode, for PHN affecting the trigeminal nerve79 ( Trigeminal neuralgia). Spinal cord stimulation (SCS) is not necessarily helpful for PHN.80

Early treatment of acute herpes zoster is important in reducing the likelihood of subsequent PHN (see SubCh. 9.4, ‘24–72 hours’). Provided treatment is started early following the initial acute phase, EA and TENS may also be of benefit once PHN has started.

Given that intravenous magnesium can temporarily block the NMDA receptor and alleviate PHN, it would be interesting to see whether oral magnesium supplements could alleviate PHN and other types of neurogenic pain,⁸¹ particularly in conjunction with acupuncture, which may itself increase blood magnesium level.⁸²

Comparisons and combinations

As mentioned, although MA is probably not very effective for PHN, TENS, and particularly CTENS, has often been considered helpful.⁸³ However, in one small RCT, EA (5/60 Hz DD) for 6 weeks gave better results than CTENS.⁸⁴ Pain in the EA group worsened again after treatment was terminated, so some commentators have surmised that the increased attention received by these patients was a major factor in their improvement (also that the different stimulation parameters used contributed in some way). In another study in which both EA (‘muscle stimulation’ with needles) and TENS were used, EA was often helpful in patients with dermatomal lack of sensation (so CTENS was not appropriate). Again, though, continued treatment was necessary for extended pain relief.⁸⁵

EA has been combined with cupping for PHN.⁸⁶ The results of this study are fascinating if difficult to interpret: 100% of the 25 Chinese patients treated were reported as cured, but only 6.2% of German patients, and 20% of the latter did not respond at all!

LA was considered superior to MA in one small comparative study.⁸⁷ Some patients who did not respond to LA alone did better with a combination of both methods. Similarly, the combination of HeNe LA with simultaneous magnetic field treatment has been claimed to give better results than LA alone.⁸⁸

Points used

Keith Glennie-Smith has suggested that segmental acupuncture may not be effective because myelinated afferents have been preferentially destroyed by the herpes zoster virus, and that points rostral or contralateral to the affected dermatome or auricular points should be treated instead (although, curiously, he also advocates ‘surrounding the dragon’ locally as an effective method).⁸⁹ For the same reason, deep-muscle stimulation may be more effective than superficial dermatomal treatment.⁹⁰

Paravertebral (or huatuojiaji) points at the level of the lesion are recommended by George Ulett among others.^91,92

Points that appear several times in the studies in the database ( Postherpetic neuralgia (PHN)) include auricular, contralateral and huatuojiaji points, GB-34, LIV-3, LI-4, ST-36 and SP-6. The most common approach is to use some form of local treatment.

ST-36 has been described as the most frequently selected point for PHN in Chinese studies,⁹³ although I understand that in one ‘mixed’ TCM and Western approach taught in the UK distal points such as GB-34 and LIV-3 are recommended, with ‘local’ points above and below the affected dermatome.

Electrode placements in the TENS studies of PHN include trigger points, selected dermatomes, low electrical resistance acupuncture points, over the painful area and proximal to the pain.⁹⁴ John Thompson and Jacqueline Filshie suggest electrodes close to but not directly over the affected area in the same or adjacent dermatomes,⁹⁵ possibly straddling the affected dermatome.⁹⁶ Contralateral TENS may be an alternative.⁹⁷

In practice, Glennie-Smith uses LILT locally as well as over intervertebral points, directed at the posterior root ganglion and dorsal horn at the affected spinal level (as well as in dermatomes immediately above and below it).⁹⁸ Another approach is to irradiate the stellate ganglion.⁹⁹

When to use what – parameters for postherpetic neuralgia

PHN did not respond well to LF EA in one retrospective study,¹⁰⁰ although LF EA did give good results in Chinese (but not German) patients in another uncontrolled trial.¹⁰¹ However, 5/60 Hz DD EA did give better results than CTENS at least in the short term (see above for a brief discussion of this comparison).

In the PHN TENS studies, parameters varied widely: stimulation frequency from 10 to 180 Hz, intensities from ‘low’ to comfortable to the maximum tolerated, with duration from 20 minutes daily to ‘prolonged’. Even with this limited data, there does appear to be some evidence that high-intensity 100 Hz stimulation long term may be of benefit.¹⁰² As always, though, more research is necessary to confirm this.¹⁰³

Margareta Eriksson and colleagues, on the basis of a small uncontrolled study, considered that HF or burst TENS gave better results for PHN than ALTENS¹⁰⁴ ( Neuropathic pain (mixed and other studies). HF TEAS gave good results in one uncontrolled trial, for instance.¹⁰⁵ However, Mark Johnson suggests that when there is hyperaesthesia ALTENS – which he defines as LF currents in the form of single pulses or LF trains or bursts of pulses, or as motor level TEAS – should be used (perhaps contralaterally) rather than CTENS over the affected area.¹⁰⁶

Keith Glennie-Smith has used red LILT (660 nm) for local superficial irradiation (contact mode, 3 W/cm², 8–10 seconds/pt), with the same dose (24–30 J/cm²) of infrared LILT (820–840 nm, 50 mW, 60–70 Hz) for deeper penetration at intervertebral points¹⁰⁷ (904 nm may also be an appropriate wavelength¹⁰⁸). Higher output power (150 mW rather than 60 mW) may be more effective still.¹⁰⁹

Glennie-Smith suggests twice-weekly treatments initially, whereas Boris Sommer, writing more recently, recommends a more intensive approach, with five to six daily sessions initially, slowly reducing treatment frequency to once weekly.¹¹⁰

Acute herpetic neuralgia

As emphasised above, it is important to treat herpetic pain early if it is not to become chronic. MA¹¹¹ and EA^112,113 have been used for acute herpetic pain, with a reduced incidence of PHN with the latter in two studies.^114,115

TENS (over the stellate ganglion) is another treatment that has been investigated for acute shingles pain.116

LA117,118,119 and LILT120,121 are frequently used for acute zoster pain.

Ultrasound has occasionally been used for pain relief in the acute stage of herpes zoster.122

Comparisons and combinations

In one report on acupuncture for acute herpes zoster pain, EA gave better results than MA ‘in some cases’.¹²³

Points used

As with acute herpes zoster in general, the local lesions are often the focus of treatment, using the ‘surround the dragon’ technique, for example (see Figs. 9.6.1, 10.1c). Posterior nerve roots (or huatuojiaji points) in the same segment can be used, and in one report anterior segmental points were used as well.¹²⁴ A similar approach was used in the ‘PENS’ (EA) study by Hesham Ahmed and colleagues.¹²⁵ However, rather than treating the affected segment itself, stimulation was limited to the dermatomes above and below it. Of the traditional acupoints, GB-43, LIV-3 and LI-4 have been used in more than one study.

Parameters used

LF and HF and 10 Hz stimulation have been used for acute herpes zoster pain. In one interesting report, HF EA was used first (for analgesia), followed by LF EA (for its neurotrophic effect).¹²⁶ With LA/LILT, HeNe is common, although polarised non-laser irradiation has also been used.¹²⁷

(For craniofacial neuralgias other than PHN, and for occipital and cervicobrachial neuralgia, see below, ‘Facial pain’, and ‘Occipital neuralgia’ and ‘Neck pain’, respectively).

Intercostal neuralgia

EA has been used for intercostal neuralgia,128 as has ryodoraku.129

For intercostal neuralgia, David Ottoson and Thomas Lundeberg have recommended HF or burst TENS at just below pain threshold, with electrodes positioned above/below and anterior/posterior to the painful region.130 The Likon device (see Ch. 11) has been used for intercostal neuralgia.131

‘Microwave needling’ combining both forms of stimulation has been used for intercostal neuralgia.132

Magnets positioned at acupoints have been used for intercostal neuralgia.^133,134

Points used

George Ulett recommends using paravertebral points at the level of the lesion involved, as with PHN.¹³⁵ Apart from the huatuojiaji or back shu points (BL-17, BL-18), others such as GB-34, LIV-3 and LIV-14 have also been suggested,¹³⁶ or LIV-3 and LIV-14 with other distal points such as SJ-6 and SP-9 or ST-40 when the condition is considered as due to stagnation of Liver qi.¹³⁷ P-6 has been used for acute intercostal neuralgia.¹³⁸

Parameters used

In one EA device manual, HF or DD is recommended for intercostal neuralgia, at a moderate to strong intensity.¹³⁹ Another just favours DD.¹⁴⁰

Peripheral nerve injury and compression syndromes

Nerve damage may vary from total transection (neurotmesis), through disruption of the axon yet with the myelin sheath still intact (axonotmesis) to contusion of the nerve (neurapraxia) (see Ch. 4). It is sometimes difficult to differentiate between the effects of nerve damage and those due to reversible nerve compression, which may even follow prolonged muscle spasm (cramp), for example.^141,142 However, in chronic peripheral nerve injury there may well be somatotopic remapping in some regions of the brain and spinal cord^143,144,145 (as occurs too with phantom pain, see below, ‘somatosensory cortex’).

The application of acupuncture for peripheral nerve injury has been reviewed.¹⁴⁶

MA has been used successfully in a case of peroneal neuropathy probably due to peripheral nerve injury sustained during surgery.147 It has also been used for injury of the superior clunial (cluneal or clunical, buttock) nerve148 (see below, ‘Clunial nerve injury’).

The experimental studies that support a role for EA in nerve regeneration are discussed in SubChapter 6.3 (‘EA and nerve regeneration’). Strong stimulation may be counterproductive: in one study in rats of experimental neuralgia due to nerve compression, weaker EA (1–1.5 mA) was found to be more effective than stronger EA (3–4 mA) (see SubCh. 6.4, ‘alleviate experimental’). It is also clearly important that stimulation be started as early as possible after injury.

EA has been used clinically for peripheral nerve injury. The authors of one such study commented that EA of sutured nerve hastens resumption of nerve impulse conduction and reduces incidence of complications.¹⁴⁹

Donlin Long has claimed that peripheral nerve injury responds well to TENS in over 70% of cases,150,151 with similar results in TEAS studies by Ronald Melzack152 and Bruce Pomeranz’s group.153 Thus TENS (together with analgesics and infrared treatment) was helpful for burning pain and allodynia in a case of crush injury following the Kobe earthquake in Japan, although neither analgesics alone nor continuous epidural analgesia had been effective.154 However, Paolo Procacci’s Italian group found that TENS was only briefly helpful for peripheral nerve injury pain when the nerve had been extensively damaged.155 Also, Steven Wolf and colleagues consider that CTENS is less effective for peripheral nerve injury than for peripheral neuropathy. In contrast, others have found TENS (75–100 Hz, 200 µs, 20–30 minutes, 3–4 times daily) to be helpful for pain due to traumatic nerve lesions provided treatment is started early (within a year).156

TCET (i.e. stimulation applied cranially rather than locally) may also be beneficial for peripheral nerve regeneration (see SubCh. 6.3, ‘signs of regeneration’).

The role of LILT in regeneration of sciatic nerve injury has been explored in some depth in experimental studies by Semion Rochkind’s research group in Israel (see Ch. 4). LILT has been used in veterinary practice for neurapraxia,157 and in humans both during surgery for peripheral nerve disorders158 and for peripheral nerve injury pain following dentistry.159 However, as the authors of the last report cautiously comment, ‘the significance of applying a new treatment method that has had much public interest should not be overlooked’.160

Experiments on rats indicate that LF PEMF (3 G, 2 Hz, 20 ms pulse duration, 4 hours daily for 6 days) can improve axonal sprouting in crushed sciatic nerve. However, if the nerve is totally transected (and then immediately repaired) and PEMF treatment initiated without delay there appears to be no improvement, whereas if treatment is delayed for 5 days some clinical improvement (15–20%) does occur.161 This is in contrast to TENS, which appears to have more effect if applied immediately and less if application is delayed.162

Whatever method of stimulation is employed, it is vital that it be combined with sensory reeducation and other rehabilitation strategies.¹⁶³

Comparisons and combinations

In general, nerve damage may respond better to EA than to MA,¹⁶⁴ although Bruce Pomeranz has suggested on theoretical grounds that the opposite may be true (see SubCh. 6.3, ‘levels and densities’).

PEMF in combination with LIT has been used in the treatment of peripheral nerve lesions.¹⁶⁵

Points used

Peripheral nerve injury can result in pain if large-diameter myelinated afferents are damaged and no longer able to inhibit C-fibre signals via the gate mechanism in the spinal cord (see Ch. 7, ‘Spinal segmental inhibition’, for details of this). If this is the case, using CTENS peripherally is rather pointless, but positioning electrodes proximal to the site of injury can result in ‘dramatic successes’.¹⁶⁶ Thus when using TENS for traumatic and compressive mononeuropathies, electrodes should be positioned proximal to the lesion, avoiding regions of abnormal (low or increased) sensitivity.^167,168 Usually, the intention when using TENS or SCS (or, for that matter, acupuncture) is to elicit a sensation (paraesthesia, deqi) in the affected area. Clearly, if there is deafferentation and loss of sensation then this may not be possible. However, there are occasional reports of electroanalgesia without the development of paraesthesiae in the precise area of pain and deafferentation, and it may be important to attempt to select treatment locations from which paraesthesiae can be obtained that surround an area of deafferentation in order to obtain pain relief.¹⁶⁹ For acupuncture, points are usually selected from those near both ends of the injured nerve trunk, local points and ‘big’ points on the Large Intestine and Stomach (yangming) meridians.¹⁷⁰

Parameters used

Experimentally, both segmental HF TENS and extrasegmental low -intensity LF EA have been shown to relieve signs of allodynia following nerve injury in rats (see SubCh. 6.3, ‘thermal allodynia’). The former effect may be mediated by orphanin (OFQ) release within the spinal cord, although this is less likely with LF stimulation (see SubCh. 6.4, ‘CNS release of orphanin’).

Selective stimulation of large-diameter myelinated afferents using CTENS can sometimes give ‘striking’ relief from burning pain caused by peripheral nerve injury (indicating that large-fibre loss may well be responsible for some of the pain of traumatic neuropathy).¹⁷¹ Thus CTENS is quite often recommended for peripheral nerve injury pain.^172,173 Steven Wolf and colleagues observed that results improved with higher-intensity stimulation.¹⁷⁴

However, from the study data ( Neurogenic pain by region), it would seem that LF or interrupted EA and TENS is more commonly used than CTENS. There is less consistency when it comes to stimulation amplitude, although there may be a possible tendency to use lower-intensity stimulation.

The head and neck

Cervical pain following nerve injury during surgery has been treated with EA and MA.¹⁷⁵

Post-traumatic neuralgia of the infraorbital nerve unresponsive to local anaesthetic nerve block has been treated with TENS or tricyclic antidepressive drugs, or a combination of these. Results were less good when prior surgery had been attempted.176

LA has been used for sensory nerve damage following dental extraction.177 Further examples of treatment for facial pain following dental treatment can be found below, under ‘Facial pain’.

The shoulder and upper limb

Neurogenic arm (brachial) pain can arise from a variety of causes and become very distressing. Treatment can be problematic, especially if the condition is complicated by the presence of chronic regional pain disorder (CRPD) (see below, ‘Complex regional pain disorder’). A number of work-related conditions are sometimes considered under the umbrella terms cumulative trauma disorder (CTD) or repetitive strain injury (RSI), sometimes also known as repetitive motion disorder (RMD). These include thoracic outlet and carpal tunnel syndromes. Ergonomic aspects of the workplace need to be considered in any treatment programme for CTD, which may result from incorrect and static posture with resultant muscle tension, as well as overuse of particular parts of the body. Musicians may be prone to RSI, for example.¹⁷⁸ Exercise, exercises, moving around and relaxation can all be vital.

Thoracic outlet syndrome

This somewhat controversial diagnosis¹⁷⁹ was introduced in the section on peripheral vascular obstruction (see SubCh. 9.6). It is usefully described in Lü Shaojie’s book on nervous system disorders,¹⁸⁰ and also in an online article by Alejandro Katz.¹⁸¹ Symptoms include pain and paraesthesiae in the arm and scapula, with the ulnar nerve affected in 90% of cases. Thus ulnar somatosensory evoked potentials (SEPs) are usually abnormal,¹⁸² and the middle, fourth and little fingers tend to be most affected.

EA has been used for thoracic outlet syndrome. Although the authors of one study noted relatively long-term benefits, they also commented that treatment itself might need to be prolonged.183

Katz recommends that scalene stretching exercises should be carried out consistently three times daily.184 Massage and acupoint injection at jingbi (M-HN-41) have also been used for the condition.185

Points used

In a study from Pakistan, local points such as SJ-14 and LI-15 were combined with LI-4, LI-11 and ST-38 distally.¹⁸⁶ Joseph Wong suggests using points in the C3–C4 segment for the scalene muscle, HE-1, LU-2 or ST-12 for the brachial plexus and appropriate distal points on the arm.¹⁸⁷ Katz applies microcurrent to P-6 and/or SI-3 (negative) connected with anterior scalene muscle TrPs (positive), with LIV-4 and/or ST-41 as secondary points (usually contralateral). He uses surface electrodes positioned over inserted needles.

Parameters used

CTENS was found to be helpful for thoracic outlet syndrome in one uncontrolled study.¹⁸⁸

Compression syndrome of the lateral cutaneous forearm nerve

Meralgia paraesthetica (below) is a recognised disorder. Less well known is the corresponding condition in the arm. Compression of the lateral cutaneous nerve of the forearm, sometimes the result of vigorous exercise, can lead to pain but decreased tactile or pinprick sensation along the radial aspect of the forearm, tenderness to palpation over the nerve where it pierces the deep fascia of the arm (lateral to the bicipital tendon and proximal to the elbow crease) and decreased elbow extension when the arm is fully pronated. Electrical evoked response measurements will confirm the diagnosis.

Treatment for this condition has included TENS and ultrasound applied at the tender region of the upper arm.¹⁸⁹

Points used

Joseph Wong has suggested points such as GB-31, ST-31, ST-32 and SP-12.¹⁹⁰

Ulnar nerve entrapment

Ulnar entrapment at the elbow is not uncommon, with symptoms of paraesthesiae along the Small Intestine meridian from the elbow towards the wrist and hand. It is likely to be present if light tapping on the cubital groove or SI-8 (with the elbow flexed at 90 degrees) reproduces these sensations.¹⁹¹

EHF has been used for ulnar neuritis, with claims of full restoration of diseased nerve function in many patients.192

Katz recommends stretching exercises to assist recovery.

Points used

Katz applies microcurrent stimulation at SI-8 (positive) with SI-3 (negative), ipsilaterally (contralaterally, if there is too much local inflammation to treat locally). He uses surface electrodes positioned over inserted needles.

Carpal tunnel syndrome (CTS)

Pressure on the median nerve within the carpal tunnel can lead to paraesthesia in the wrist and hand, with muscle wasting and functional impairment if the condition is severe or longstanding. CTS is not uncommon in pregnancy, menopause or associated with rheumatoid arthritis.¹⁹³ Inflammation within the tunnel may be associated with repetitive hand movements. However, overuse may also lead to other conditions that may be mistaken for CTS, such as shoulder or elbow problems or ulnar entrapment.¹⁹⁴ A nerve conduction study will confirm the diagnosis.^195,196

Local steroid injections, ultrasound treatment and oral pyridoxine (vitamin B6) have all been used for CTS, with limited evidence that the first of these (and also oral steroids) may be effective, conflicting evidence on ultrasound and the likelihood that B6 is equivalent to placebo, all in the short term. NSAIDs and diuretics do not appear to give much benefit either. Use of a hand brace may be helpful, as may yoga.¹⁹⁷ Acute CTS may respond to ultrasound experimentally¹⁹⁸ and, although there is some evidence, if limited, that ultrasound may give some symptomatic relief in the long term¹⁹⁹ (if treatment is continued for 7 rather than just 2 weeks²⁰⁰), the majority of those given steroid injections are likely to need a repeat within 2–4 months.²⁰¹ Without treatment, long periods off work and lengthy recuperation may be necessary (half of all employed people with CTS have missed 30 days or more of work²⁰²). Given that only 40% of those undergoing surgery regain normal function (although up to 80% may experience considerable pain relief),²⁰³ that non-surgical methods do not seem terribly effective and that repeated steroid injections can actually contribute to neuronal degeneration,²⁰⁴ acupuncture may well be an appropriate intervention to try. Indeed, in 1997 the US National Institutes of Health Consensus Development Conference panel considered that there was at least some good evidence for acupuncture’s effectiveness for CTS,²⁰⁵ although some other reviewers do not agree.²⁰⁶

In one report of MA for CTS, median nerve conduction speed was assessed as a method of objectifying changes due to treatment.207

Good results have been reported in uncontrolled studies of EA for CTS.208,209

LILT and LA have been used for CTS ( Carpal/tarsal tunnel syndrome). Some benefits were reported, for example, in a RCT of LA for patients awaiting surgery for CTS (but not postoperatively).210,211 However, in one small placebo-controlled study no subjective or objective difference was found between LILT and placebo for median nerve compression (this report was wrongly titled when it was published212). Also, some reviewers consider that LA has not (yet) demonstrated useful effects when compared with placebo or control treatments.213

Some helpful prognostic indicators are outlined in the various LA/microcurrent studies by Margaret Naeser and her colleagues ( Carpal/tarsal tunnel syndrome). In particular, continuing smokers are not likely to achieve significant benefits from their protocol, nor are patients with persistent round-the-clock fingertip numbness for longer than 18 months (although they may experience some pain reduction). However, those with intermittent numbness are likely to experience improvements in both pain and numbness.²¹⁴ Whether these observations are valid for other forms of acupoint stimulation is not clear.

Permanent magnets taped over the carpal tunnel area have been used for wrist pain in employees at a turkey processing plant, and were found to alleviate pain without adverse effects.215 This approach is recommended by Roger Coghill.216 However, as yet rigorous studies do not support the use of magnet therapy for CTS.217

The Qi Gong machine (QGM) has been used to treat CTS, applied paraspinally at C7–T1, at the elbow (if painful) and locally, for 10–15 minutes at low intensity (if pain increases, the applicator is moved a short distance away from the area treated for about 5 minutes and then reapplied).²¹⁸

Stretching exercises may be an important part of treating this condition.²¹⁹ Chiropractic approaches have also been used.^220,221,222

Case study 9.11.1 by Ron Sharp details the treatment of bilateral carpal tunnel syndrome.

Case study 9.11.1 Bilateral carpal tunnel syndrome Ron Sharp

This was a referral from the patient’s general practitioner to the physiotherapy department acupuncturist. The diagnosis of carpal tunnel syndrome had been confirmed by nerve conduction tests, and the patient was on the waiting list for decompression surgery. She complained of bilateral arm and hand pain, with frequent pins and needles. The hands ached following the simplest of activities and became very hot and swollen, especially at night. These symptoms had been going on for over 18 months and had been getting worse over the last 6 months. There was no history of neck, shoulder, arm or wrist injury. Her heart was fine, but there was an alteration to circulation in the arms with some slight general swelling. Aggravating factors were holding and gripping, driving any long distance and repetitive lifting, which was inevitable at times in her work as a book dealer. Previous treatments included heat, neuromobilisations and manipulation, none of which had helped. Relief was gained only by rest and anti-inflammatory medication. On examination she had poor neck posture, with a poker neck. Gripping caused pain, reflexes were dull and equal and she had increased sensation on the left forearm. Both arms were cold to the touch, with tenderness to deep pressure. There was full range of movement at her shoulders, elbows and wrists, although there was some weakness on resisted wrist extension bilaterally, with pain. There was some slight restriction to neck rotation (down 10%), which was equal on both sides. Extension, side flexion and flexion were normal. The neural tension test was positive at the end of range on the median nerve complex bilaterally and tinel’s sign was positive bilaterally. The pulse was slow but even, and the tongue pale and dry, with a thin, white coating with good root. The treatment plan was to move the energy in the arms, mobilise the neck and address any posture problems at her work. Acupuncture seemed to be the best approach for the arms, with mobilisations of the neck and postural exercises. The first treatment was the needling of LI-10 and SJ-5, bilaterally with even method, agitating the needles every 5 minutes for a period of 15 minutes. The patient was also given wrist splints to use at night and during the day if a heavy workload was scheduled, supplemented with neck posture advice and exercise. There was a slight healing crisis with discomfort of the neck for only a few hours. The previous treatment was repeated, with needle retention for 20 minutes using the even method, followed by cervical manual traction in flexion for 3 × 30 seconds. Although she benefited from treatment and regained her full range of neck movement, the discomfort had not changed appreciably. The next few treatment sessions used the same points but with a strong drainage treatment at SJ-5 for 5 minutes, followed by even technique at SJ-5 and LI-4, with needle agitation every 5 minutes over a 20-minute period. This strategy seemed to work well; the patient had good pain relief for about 5 days, with slight occasional discomfort depending on her workload. The right side had responded more effectively than the left, and the discomfort on the left had become more focused in the hand. The strategy now had to address the left-hand discomfort, so the points for acupuncture were changed to right LI-10, left LI-4, and luozhen (extra point). These were needled with the even method, with needles left in place for 20 minutes. This worked quite well, but again the effect lasted for no more than 5 days. To try to lengthen the treatment effect I added electroacupuncture to the treatment regimen. The treatment consisted of right LI-10, left LI-4, and luozhen (extra point). Right LI-10 and left luozhen were connected to the electrodes and left LI-4 was needled with even method, all for 20 minutes. The treatment rate was 10 Hz, with an intensity that was just within the comfortable range, checked every 5 minutes to maintain the intensity and to counteract accommodation. This strategy succeeded and the patient was totally asymptomatic after another three sessions. When she had to be out of the area for several months, we decided she should use a piezoelectric wand stimulator while away to maintain her pain control. I prescribed two weekly treatments of three charges at LI-10 bilaterally. On review, her consultant decided to hold back on surgery for the time being, as acupuncture seemed to control her condition. The patient contacted me after being away for 5 months to say that she had no problems with her arms and hands. The consultant had discharged her and she would not require surgery. The piezoelectric wand had x been used only a couple of times and that was only after a heavy session of stock taking. Equipment inventory Needles: 1½" and 1" 30 gauge stainless steel Electroacupuncture unit: AWQ-100 Multi-purpose electronic acupunctoscope Piezoelectric wand stimulator (no further information available) Point repertory LI-10 Moves energy and increases circulation in the limb SJ-5 Specific for painful obstruction syndrome LI-4 Major local pain relief point Luozhen (M-UE-24) To stimulate the circulation Conclusion I feel that the body gets used to a particular treatment strategy, especially in chronic conditions. To counteract this, I change the strategy to try and ‘keep the body guessing’, so each new approach will have a maximum effect on qi (within the framework of the working diagnosis). I am therefore not afraid to change when response levels to treatment tail off. Electro acupuncture will always provoke a strong response when needed. It should not be used as a first line approach in chronic deficient conditions. (Another case by Ron Sharp is presented as Case study 9.4.2.)

Comparisons and combinations

EA has been combined with glucocorticoid acupoint injection for this condition (using triamcinolone-A or prednisolone)²²³ and also with Chinese herbal medicine.²²⁴

The combination of LA with microcurrent stimulation for carpal tunnel syndrome or repetitive strain injury has been championed by Margaret Naeser²²⁵ ( Carpal/tarsal tunnel syndrome). She was also one of those who presented to the NIH Consensus panel that adjudged acupuncture as having some role in the treatment of CTS.^226,227,228

Points used

In the single detailed EA study on CTS located, P-6 and P-7 were the points employed
( Carpal/tarsal tunnel syndrome).

Margaret Naeser uses LA at Heart meridian points near the wrist, P-7 and LU-9 (as well as jing Well and baxie points when treating peripheral neuropathy),²²⁹ whereas Alejandro Katz applies microcurrent to ~P-6 (positive) locally, with contralateral ST-41 and/or LIV-4 (negative). Whether or not symptoms are reproduced with palpation of ~P-6 after treatment will indicate which of the two contralateral points is likely to give better results. He uses surface electrodes, positioned over inserted needles if the patient is not too sensitive for this.²³⁰

Some authors have emphasised treating points in the neck region for CTS, in addition to limb points.^231,232,233

A caution
Median nerve injury following acupuncture is not unknown.234

Parameters used

When using microcurrent, Margaret Naeser starts treatment with 292 Hz, following this with 0.3 Hz²³⁵ LF EA, ‘to tolerance’, was used in the one detailed study located.²³⁶ It is surprising that there are not more EA studies on CTS.

With a low-power (5 mW) HeNe or 670 nm diode laser suitable for self-treatment at home, lengthy treatments are necessary. For instance, Margaret Naeser recommends a 7 J dose for P-7, which would take around 20 minutes (!), 2–4 J (6–12 minutes) for other painful areas and 1 J (3 minutes) for the other hand points in her protocol. If there is an improvement with these exposure times then shorter treatments can be attempted in later sessions,²³⁷ whereas with the higher-power (15 mW) laser used in her clinical study such long exposure times were not needed ( Carpal/tarsal tunnel syndrome). In children far shorter treatments can be effective,²³⁸ for instance when treating ‘Nintendonitis’. If Naeser’s recommendations are correct, the treatment times of 15 seconds per point in one German RCT in which 5 mW HeNe laser was used may have been insufficient to produce marked benefits, although some improvement in nocturnal pain was noted.^239,240

On the basis of Judith Walker’s finding of a sharp increase in 5-hydroxyindoleacetic acid (5HIAA) urinary excretion after 10 sessions of LILT,²⁴¹ and their own observation that pain level may decrease noticeably after nine sessions, Kenneth Branco and Margaret Naeser recommend at least 15 sessions for CTS with their LA/microcurrent protocol²⁴² or 4–8 weeks for the home treatment version.²⁴³

Plexus lesions

Traction lesions of the brachial plexus, often involving avulsion of nerve roots from the spinal cord (in motor cycle accidents, particularly), may result both in total and irreversible paralysis and in paroxysmal pain that continues indefinitely and responds poorly to medication.²⁴⁴ However, in a minority of cases early reconstructive surgery may be of benefit.²⁴⁵ Surgery has also been used for iatrogenic plexus lesions induced by radiotherapy.²⁴⁶

Brachial plexus injury and neuralgia have been treated with EA. Treatment was found to be useless in patients with neurotmesis (complete plexus lesion) or where nerves had been ligated, with better results in those whose injury was less severe, due to compression, contusion or traction247 and was relatively recent.248 The points used are generally selected according to where pain is experienced and the distribution of nerves or meridians passing through these areas.

TENS has been used for the severe pain of brachial plexus avulsion injury,249 for example with one electrode positioned over C5–C7 and the other just proximal to the site of pain in an area where there is afferent input250 ( Plexus lesions). TENS was also employed in one case of attempted suicide that resulted in a lesion of the brachial plexus, with beneficial analgesic and functional effects.251 Wynn Parry, an authority on plexus avulsion injury, has commented that the best method to reduce the pain is for the patient to engage in work that is totally absorbing, rather than dwelling on it.252

Clunial nerve injury

The clunial (cluneal, clunical) nerve, which innervates the skin of the buttock (clunis), has three divisions: the superior, middle and inferior. The former originates with the lateral cutaneous branches of the dorsal rami of L1–L3and the latter two from the dorsal and ventral rami of S1–S3 respectively.²⁵³ The medial branch of the superior clunial nerve (SCN) is contained in an osseofibrous tunnel as it passes over the iliac crest just lateral and superior to the posterior superior iliac spine.²⁵⁴ Superior clunial nerve entrapment is an often-overlooked cause of low back pain.²⁵⁵ Pain in the back and buttock, often extending into the leg,²⁵⁶ may begin following back surgery and the formation of adhesions. It can be severe enough to make sitting, walking or standing possible only for short periods.²⁵⁷ Conventionally, this entrapment neuropathy has been treated with steroid injection, thermal ablation or surgery.²⁵⁸

MA has been used for injury of the SCN,²⁵⁹ as have EA^260,261 and TEAS.²⁶² EA was found to be more effective than MA in one RCT.²⁶³

In one case report of SCS for SCN, TENS gave only partial, temporary relief, whereas the more invasive method was more effective, enabling the patient to return to her normal activities.²⁶⁴

Points used

Stimulation of trigger points lateral to L1–L2 may relieve clunial nerve pain.²⁶⁵ Thus huatuojiaji and ashi points have been used for SCN, in conjunction with points such as GB-34 or other distal points.

Parameters used

LF EA or TEAS appears to be used more commonly than HF stimulation, although too few studies are available to enable unequivocal recommendation of one rather than the other.

Piriformis syndrome

The peroneal division of the sciatic nerve passes through the piriformis muscle, rather than under it, in some 10–20% of the population (Fig. 9.11.1). Spasm, contracture or injury of the muscle then compresses the sciatic nerve, leading to symptoms that mimic those of discogenic sciatica, as well as buttock tenderness and pains that sometimes radiate to the lower abdomen and genitals (in some cases even with dyspareunia). Pain may be diffuse, and if the problem is chronic then atrophy of the lateral muscles of the calf may develop. The condition is more common in women than men, in lorry drivers and certain types of athlete.²⁶⁶ It may even be caused by prolonged sitting during extended neurosurgery. In complicated cases, lower lumbar radiculopathy may lead to secondary irritation of the piriformis muscle, while trochanteric bursitis in some cases is due to an underlying piriformis syndrome.²⁶⁷ Further descriptions of the condition can be found in the standard textbooks by David Legge²⁶⁸ and Lü Shaojie.²⁶⁹ Conventional treatment includes physical therapy (ultrasound, TrP injection, stretching and postural correction), steroid injection or surgery.²⁷⁰

Figure 9.11.1 Piriformis syndrome. This set of diagrams (rear and side views) illustrates the close proximity of the sciatic nerve and the piriformis muscle at the pelvic outlet. The lower pair of diagrams show a common anatomical variation in which the peroneal part of the sciatic nerve traverses the muscle fibres of piriformis – this occurs in approximately 10% of people. (Reproduced with permission of Dr Mike Cummings, Medical Director, British Medical Acupuncture Society.)

Piriformis syndrome has been treated using MA,²⁷¹ EA²⁷² and a ‘biofrequency spectrum’ device.²⁷³ LA has been used for piriformis muscle injury.²⁷⁴

Comparisons and combinations

For sciatic pain presumed to be due to piriformis muscle spasm (cramp), Willem Khoe has used ultrasound acupuncture (1 W for 60 seconds) at acupoints such as BL-27–BL-29 and LIV-12, at the greater trochanter and rectally, combined with 2 Hz EA at the L5 huatuojiaji points (15–20 minutes, to tolerance) and vitamin B12 point injection at auricular and hand points for the low back.²⁷⁵ It is unclear which were the most effective ingredients in this cocktail. Other authors have combined EA and acupoint injection for piriformis syndrome,²⁷⁶ for instance using glucocorticoid (triamcinolone-A or prednisolone) injections.²⁷⁷ EA has also been combined with TDP irradiation.²⁷⁸

Points used

For non-radicular compression lesions of the sciatic nerve, John Thompson has recommended that TENS electrodes are applied segmentally and proximal to the lesion.²⁷⁹

For sciatic pain presumed to be due to acute piriformis syndrome, Willem Khoe used KI-8, the xi -cleft point of the yinqiao mai (Yin Heel Vessel) extra meridian in his own variant of Yoshio Manaka’s diode cord treatment.²⁸⁰

Parameters used

There are too few studies available to enable generalisations on the most appropriate parameters to use. Both LF and ‘CW’ EA have been utilised ( Low back pain with nerve entrapment).

Meralgia paraesthetica

Nerve compression or injury in the groin can sometimes lead to neuritis in the lateral cutaneous nerve of the thigh, with symptoms in the lower two-thirds of the anterolateral thigh. It may occur during or following pregnancy, or in obese individuals or even as a result of wearing overly tight clothes²⁸¹ (‘jeans disease’²⁸²).

This condition has been treated with MA together with acupoint injection,²⁸³ with EA together with cupping,²⁸⁴ with electric plum blossom needling combined with moxibustion²⁸⁵ and also with TENS.²⁸⁶ Local stimulation was emphasised in the EA (HF) and plum blossom (LF) studies.

Amputation pain

Chronic postamputation pain is common, with both central and peripheral mechanisms responsible for its initiation and maintenance. Pain may be experienced in the stump or phantom pain, and may be burning or tingling, or cramping. The former tends to be associated with decreased blood flow in the residual limb and the latter with muscle spasm (cramp).²⁸⁷ To avoid the development of chronic pain, early treatment can be important.²⁸⁸

In one retrospective review of pTENS for chronic pain, it gave useful effects in 65.8% of cases of pain due to amputation.²⁸⁹ In contrast, some surgeons believe that TENS offers no lasting relief of pain following amputation surgery.²⁹⁰ It may be that this difference of opinion is because TENS is more likely to be helpful for milder postamputation pain.

Stump pain

Stump pain is sometimes caused by ‘ectopic’ sensory nerve discharges from a neuroma.^291,292,293

Donlin Long reported good initial results for stump pain with TENS, although effectiveness dropped off considerably over time.294,295 In one case, residual stump pain was completely relieved by TENS after severe long-term phantom pain had been effectively dealt with by surgery and phenol nerve blocks.296 There are other positive reports on TENS for stump pain.297 TENS may also result in more rapid stump healing if used postoperatively.298

MRT has been found to be beneficial for stump pain, muscle spasm (cramp) and oedema.

Ultrasound has been used for stump neuroma pain.²⁹⁹

Electric currents can be measured at regenerating finger stumps in children³⁰⁰ (or whole limb stumps in adult newts³⁰¹). Regeneration is highly unlikely in human adults, but EAV has been used for amputation pain both as a method of evaluating changes in acupoint electrical potential and for treatment.³⁰² Although such treatment may help pain, there are no reports of it encouraging human limb regeneration, even though theoretically this may be possible. The effect of amputation on the low electrical skin impedance found along meridians has been investigated.³⁰³

Comparisons and combinations

Both MA and TENS were found useful in one report on stump pain. PEMF, however, sometimes exacerbated it.³⁰⁴

Points used

For stump pain, John Thompson recommends that TENS electrodes are positioned paraspinally (in the same dermatome) and actually on the stump.³⁰⁵

TrP injection with a local anaesthetic agent has been used in the treatment of postamputation pain. Contralaterally symmetrical tender points with low electrical skin resistance to an applied 3 V DC signal were selected.³⁰⁶

Roger Coghill recommends magnets locally on the stump itself for stump pain (repeated 15-minute sessions).³⁰⁷

From the studies available, there are no clear patterns on which might be the most effective points to use. Both stump and contralateral stimulation have been tried.

Parameters used

From the few studies available on peripheral methods of stimulation for stump pain, it appears that HF stimulation is attempted more frequently than LF EA or TENS.

From studies³⁰⁸ and reviews³⁰⁹ of TENS, it does appear that stump pain can respond well to CTENS. However, if it is the case that different sorts of stump pain have different causes, as mentioned above, these should be taken into account. For burning pain and paraesthesiae, parameters likely to improve blood circulation should be used (see SubCh. 9.6), whereas for cramping pain, parameters for muscle spasm (cramp) would be preferable (see SubCh. 9.1, ‘Spasticity in cerebral palsy (CP)’ and SubCh. 9.2, ‘Spasticity’).

Phantom pain

Pain in a part of the body that is not there occurs in some 70–80% of patients after amputation, and can be very severe. Surprisingly, it can be congenital as well as acquired.^311,312,313 Although it is usually the pain of a phantom limb, phantom pain can also occur after removal of the bladder (or spinal cord injury with deafferentation of the bladder),³¹⁴ a breast,^315,316 or even tonsils.³¹⁷ Phantom pain, like stump pain, can be burning or tingling or, possibly more often, cramp-like,³¹⁸ and for the same reasons.³¹⁹ It should not be confused with non-painful phantom sensation, which may be tactile, kinetic (as if a leg is moving, for example) or even kinaesthetic (a limb may feel as if it is ‘telescoping’ over time). The sense of a phantom limb may be essential for an artificial limb to be usable.³²⁰ However, there may be a strong correlation between phantom pain and phantom sensation.³²¹

A phantom limb may be experienced in an awkward position if that is how it was during or just prior to traumatic amputation. It is more likely to be experienced as painful if there was pain at that time,^322,323,324 and if the circumstances of the loss were particularly unpleasant.³²⁵ However, if it is possible to prevent pain immediately prior to (elective) amputation, for instance with regional anaesthesia,^326,327 then phantom pain is less likely to develop afterwards. This may be the result of preventing changes in the primary somatosensory cortex that appear to accompany the development of phantom pain.³²⁸ For a similar reason, acupuncture has sometimes been recommended early on in rehabilitation.³²⁹ Interestingly, if phantom limb pain is relieved the phantom limb may seem to reposition itself ‘normally’.³³⁰ (This has been observed to occur in stages with repeated acupuncture treatment, for example.³³¹)

Whether general or regional anaesthesia is used during amputation appears to have no bearing on whether phantom pain develops or not.³³² Although from experimental studies it would appear that pain relief immediately postamputation is not likely to prevent the subsequent development of phantom pain, continuing early acupuncture treatment for some weeks has been recommended.³³³ Further surgery is unlikely to be helpful.³³⁴ As with many forms of chronic neurogenic pain, complete relief of pain may not be a realistic objective, although pain reduction may well be possible. The level of pain experienced may be related to stress, anxiety, exhaustion, and depression,³³⁵ and treatment regimens may need to take these and other contributory factors into account.

Acupuncture, both MA and EA, has been used for phantom limb pain^336,337 ( Phantom limb pain). In one EA study, results were better on the upper than the lower limb, and in women than in men.³³⁸ The author of a small study on phantom limb pain in which MA, EA and moxibustion were all used commented that stump pain is more difficult to treat.^339,340

TEAS341 and TENS342,343,344,345,346,347 can give useful results in phantom limb pain, particularly if the pain is mild.348 Although sham TENS had a considerable effect on phantom pain in one study,349 in another TEAS (Codetron) was helpful in only 25% of patients.350 As with many conditions, TENS tends to lose its effectiveness for phantom pain slowly over time.351 Auricular TEAS also affected non- painful phantom sensations in one crossover CT,352 although in an earlier single case report from the same group such sensations slightly in creased in response to the treatment.353

In addition to phantom limb pain, TENS has also been used for phantom bladder³⁵⁴ and phantom breast pain. In the latter case, when tolerance developed to treatment (that also incorporated auricular acupuncture and medication), metastases were subsequently detected.³⁵⁵

Electrical stimulation (‘faradisation’) of the brachial plexus was early on found by Weir Mitchell (see below, ‘1863’) to ‘resurrect’ a phantom hand that had been missing for 25 years.³⁵⁶ Both vigorous acupuncture³⁵⁷ and TENS³⁵⁸ have been reported as eliciting or exaggerating phantom limb pain (TENS applied to the stump could also revive a phantom that had been successfully suppressed).

MENS, in contrast, may relieve phantom limb pain.359

Gary Null has suggested that improved circulation in response to magnetic field application may also benefit phantom pain.³⁶⁰ Timo Töysä has found that magnets even well away from the affected limb may be helpful for phantom pain (he interprets this as supporting the meridian hypothesis).³⁶¹ Pulsed magnetic fields have also been applied at acupoints for phantom pain³⁶² ( Phantom limb pain).

SCS has been used for phantom limb pain, although with unpleasant side-effects in some cases.³⁶³

Focusing attention on phantom limb pain will affect it.³⁶⁴ Phantom limb pain may thus respond to EMG biofeedback,³⁶⁵ which offers the interesting prospect of combining such methods with treatments such as EA that are likely to affect muscle activity.

Comparisons and combinations

In one case report, the patient’s preference was for MA over both EA and TENS.³⁶⁶ In another, the combination of acupuncture and TENS was the only treatment that enabled the patient to wear his forearm prosthesis with a degree of comfort.³⁶⁷

In one uncontrolled study, EA gave better results than TENS; the best results were found with their combination.³⁶⁸ EA (at scalp points) has also been combined with MA and moxibustion for phantom pain.^369,370

Points used

From the acupuncture-based studies in the database ( Phantom limb pain), it would seem that auricular and contralateral treatment might be useful starting points. Scalp points and GB-20, Du-16, Du-24 and sishencong (M-HN-1) have also been used, together with major points on the limbs and huatuojiaji points.

Phantom limb pain is liable to be more severe if the stump itself is colder.³⁷¹ Thus Jacqueline Filshie has suggested that acupoints be used that are likely to block sympathetic activation.³⁷² In the case of phantom breast she has reported using both paraspinal (T1/T2) and suprascapular TrPs to good effect (the latter to reduce muscle spasm (cramp)). In one case report, MA was effective only at one very specific point.³⁷³

With conventional TENS, the peripheral nerve trunks are usually stimulated. However, if these are absent, as in the phantom limb, other strategies have to be adopted. David Bowsher, for example, suggested that the spinal cord itself should be stimulated at the segmental level involved.³⁷⁴ Other TENS electrode placements have included TrPs, acupoints, segmentally or contralaterally, over related peripheral nerves, or the stump itself. John Thompson, for instance, has recommended positioning electrodes as for stump pain (paraspinally and on the stump).³⁷⁵ Gessler and Struppler found that TENS over the nerve supplying extensor muscles of the phantom was particularly helpful, giving the impression of cramplike flexures ‘opening’ and relaxing.^376,377 Laitinen and others obtained good results with contralateral stimulation.^378,379

Given that local stimulation may exaggerate phantom pain, contralateral treatment should perhaps be attempted first, as mentioned. This approach has been tried using press needles at contralateral points, for example,³⁸⁰ as well as with local anaesthetic injections.³⁸¹ However, TrP anaesthetic block does not always give good results.³⁸²

When to use what – parameters for phantom pain

EA studies have predominantly employed LF stimulation, ‘to tolerance’, or at ‘bearable’ intensity.

In the TENS studies, a wide range of parameters has been used. CTENS is sometimes recommended,³⁸³ but is not always helpful.³⁸⁴ LF burst TENS is another possibility,³⁸⁵ although it was less effective than CTENS in one study where electrodes were positioned to stimulate phantom extensor muscle efferents.³⁸⁶ ALTENS has been used contralaterally.³⁸⁷ Given the likely involvement of an affective component and spinal segmental mechanisms in the development of phantom pain, MA and low-intensity HF EA or TENS have been suggested as appropriate parameters to use both preoperatively (for 3 days when possible) and following amputation.³⁸⁸ Rather as with CRPD, a gentle approach emphasising a sense of inner stillness has been recommended rather than treatment heroics that could reinforce the initial trauma.³⁸⁹ As with stump pain, parameters should be selected in accordance with whether the phantom pain is more burning or cramping (see above, ‘burning pain and paraesthesiae’).

Peripheral neuropathy

Peripheral sensory neuropathy may accompany endocrine conditions such as diabetes or hypothyroidism, or may result from infectious diseases such as leprosy, Lyme disease or HIV, some nutritional deficiencies (thiamine, folic acid, vitamin B12) or ingestion of toxic chemicals (including long-term use of alcohol or some medication). Unexplained sensory neuropathy may be associated with cancer that is only diagnosed many months later – the result of nerve compression from a growing tumour. Mixed sensory and motor neuropathies occur in uraemia and Guillain–Barré syndrome, as well as in most types of Charcot–Marie–Tooth syndrome,³⁹⁰ the commonest inherited peripheral polyneuropathy.³⁹¹ (In most variants of this condition as in multiple sclerosis it is the myelinated rather than unmyelinated fibres that are most affected,³⁹² although one type involves axonal degeneration alone.³⁹³) When acute, the symptoms of axonal polyneuropathy associated with excess alcohol consumption in conjunction with poor nutrition may resemble those of Guillain–Barré syndrome (see SubCh. 9.3), but without involvement of the immune system.³⁹⁴ In TCM terms, peripheral neuropathy is usually considered under the heading of wei bi, or wei syndrome (sometimes clumsily but expressively translated as ‘wilting-impediment syndrome’).³⁹⁵

Substance P (SP) may be low in the cerebrospinal fluid (CSF) of patients with peripheral neuropathy (but may be raised in those with arachnoiditis, see SubCh. 6.2, ‘peripheral neuropathy’).

Acupuncture has been suggested for the neuritic pain of leprosy,^396,397 although I do not know whether it has actually been used for this to any extent.

Encouraging results have been reported in uncontrolled studies of EA for both drug-induced neuropathy398,399 and Guillain–Barré syndrome400,401,402 ( Other/mixed conditions). Charcot–Marie–Tooth syndrome, in contrast, appears to have been treated rarely with EA. One older patient that I treated in whom the disease appeared to be progressing believed that TEAS aggravated her problems, whereas before when the disease had been stable both EA and TEAS had been helpful.

Although Donlin Long did not consider TENS (parameters selected according to patient preference) to be particularly useful for peripheral neuropathy,403 Steven Wolf and colleagues did consider CTENS as helpful (however, their definition of ‘peripheral neuropathy’ included some neuralgias).404 The Codetron ALTENS device was found helpful for a small cohort of peripheral neuropathy sufferers in one uncontrolled study.405 More specifically, TENS has been used for painful neuropathy in Guillain–Barré syndrome.406

Microcurrent TEAS (‘MEAS’?) was found helpful for peripheral neuropathy induced by antiretroviral medication in HIV patients.407,408

The neurological complications of chronic alcoholism have been treated with LA.409

MRT (58.57–69.6 GHz, 3 mW, 4–7 minutes daily at acupoints) has been used for alcohol-induced polyneuropathy, radiation sickness and infection, with better results in non-medicated patients.410 EHF acupoint stimulation has been found helpful for ‘peripheral neuritis’ (neuralgias, optic and auditory neuritis) in other reports.411

Comparisons and combinations

EA, Chinese herbal medicine and acupoint injection have been used together for drug-induced multiple radicular neuritis.⁴¹²

Methylcobalamin (mecobalamin), the neurologically active form of vitamin B12, may be beneficial for both peripheral and autonomic neuropathy.^413,414 A combination of peripheral electrostimulation with this approach should be investigated to see whether results could be improved further. A case of ulnar neuropathy following carpal tunnel surgery responded well to a combination of MA with high potency B-complex vitamins in my own practice.

Points used

For drug-induced neuropathy, the most commonly used points appear to be ST-36 for the lower limb and LI-11 for the upper limb, although of course usually not treated in isolation.

Electrode placements in the TENS/TEAS studies of peripheral neuropathy include acupoints, and proximal to or over painful areas.

With a two-output TENS machine, Gordon Gadsby and Michael Flowerdew suggest that one pair of electrodes can be positioned at the site of neuropathic pain and the other pair bilaterally at ST-36.⁴¹⁵

One interesting approach tried for peripheral neuropathy combined acupuncture with home use of cathodal TEAS at ST-36 for the leg, or LI-11 for the arm, the return electrode (anode) being placed in a waterbath in which the whole of the affected foot or hand was also immersed.⁴¹⁶ The Liss Pain Suppressor device was used in this case report, in which a soothing sensation was experienced rather than the tingling felt with conventional TENS ( Drug-induced neuropathy). A similar approach with conventional EA has also been suggested for polyneuropathy: LI-11 in conjunction with the interdigital baxie points for the upper limb, or ST-36 and SP-6 with the corresponding bafeng points for the leg.⁴¹⁷

Corresponding to her protocol for CTS and neuropathic hand problems ( Carpal/tarsal tunnel syndrome), for peripheral neuropathy in the lower extremities (feet) Margaret Naeser has suggested a home treatment protocol for LA, with local points such as BL-60, KI-3, KI-6, GB-40, LIV-4, ST-41 and SP-5, jing Well and bafeng points.⁴¹⁸

When to use what – parameters for peripheral neuropathy

ALTENS has been found useful for experimental neuropathy in rats (see SubCh. 6.4, ‘WDR activity’).

CTENS or TLEA, especially if not charge balanced, may not be suited to peripheral neuropathy with hyperaesthesia or serious sensory loss (see Ch. 12 , ‘TLEA, especially’). In the first instance local treatment may serve only to exacerbate pain, whereas in the second skin damage could occur without the patient’s awareness if treatment is continued for long periods.

It has been suggested, on the one hand, that stimulation at > 50 Hz may not be appropriate for peripheral neuropathy if denervation is a factor (see too Ch. 12). On the other, experimental research indicates that HF may more effectively reduce neuronal hyperexcitability than LF stimulation (see SubCh. 6.4, ‘way: more’). Furthermore, increased blood supply to the (sciatic) nerve, which presumably would be beneficial in cases of neuropathy, may occur only in response to HF and not LF EA.⁴¹⁹ However, evidence from clinical studies on the effectiveness of CTENS for peripheral neuropathy, for example, is limited.^420,421,422

Clinically, moderate intensity DD or LF EA has been recommended for polyneuropathy,⁴²³ although, as mentioned (see SubCh. 9.10, ‘parameters for nociceptive and neurogenic pain’), some practitioners do prefer to use CTENS/TLEA. From the studies in the database ( Peripheral neuropathy) it would seem that for peripheral neuropathy in general acupuncture-like (AL), TENS-like (TL), DD and intermittent stimulation are all used, but again possibly with a with a slight emphasis on LF (high-intensity) treatment.

Metabolic neuropathy may result in part from oxidative and nitrative stress (involving the vasodilator nitric oxide, NO). Adjunctive treatment with antioxidants such as alpha lipoic acid may therefore be beneficial.⁴²⁴

Diabetic neuropathy

Peripheral neuropathy is the commonest complication of diabetes mellitus.⁴²⁵ It may be evident in gait and balance abnormalities (postural dizziness) even before diabetes is diagnosed,⁴²⁶ and may already be evident in children and adolescents with poor metabolic control of their condition.⁴²⁷ In older diabetics, peripheral neuropathy can also be associated with peripheral ischaemia due to vascular disease. In fact, chronic peripheral arterial occlusive disease may itself cause axonal polyneuropathy,⁴²⁸ and it can sometimes be difficult to tell whether pain is due more to diabetes or to poor circulation (as in some other conditions where both may be involved such as erythromelalgia ( Peripheral circulation: other studies). In what is termed the ‘diabetic foot’, the combination of neuropathy and ischaemia can have serious consequences: a trivial lesion, unnoticed because of the neuropathy, can lead to rapid necrosis even of deep tissue, with the need for extensive surgery sometimes of the whole limb.⁴²⁹ Almost 1% of all diabetics require amputation, although this could very often be prevented with proper care.⁴³⁰ Indeed, nerve regeneration is clearly possible if the condition is not too advanced.⁴³¹ Most drugs used for diabetic neuropathy are associated with significant side-effects.⁴³²

It is important to remember that diabetic autonomic neuropathy can cause dysfunction or actual damage in many internal organ systems – cardiovascular (e.g. tachycardia), gastrointestinal (e.g. gastroparesis, diarrhoea), genitourinary (e.g. neurogenic bladder), adrenal and ocular (pupillary abnormalities).⁴³³ Also, with reduced afferent input from limbs affected by peripheral neuropathy, neuropathic changes can occur centrally as well. Thus afferent stimulation can be an essential part of treatment.⁴³⁴

Acupuncture has been used for both autonomic and peripheral diabetic neuropathy.

Experimentally, EA has been found to compensate for neurochemical changes in peripheral nerves that may be associated with the development of neuropathy (levels of cAMP and cGMP),435 and also to protect nerve function (as measured by motor nerve conduction velocity and muscle activity436). EA (at BL-23 and ST-36) appeared to be more protective against sensory neuropathy than TENS at the same points.437

For peripheral diabetic neuropathy, results with EA were superior to those of MA (mock EA) in one crossover RCT (see Ch. 8, ‘diabetic neuropathy’). EA was also superior to medication in other controlled studies.^438,439 Improvement was found to peak after about 6–8 sessions by Romanian researchers.⁴⁴⁰

Kaada’s TEAS protocol (see SubCh. 9.6) was found to improve symptoms of diabetic neuropathy even more than those of Raynaud’s disease in one very small, uncontrolled study.441

There are several other uncontrolled TENS studies on diabetic neuropathy, usually with positive outcomes. In one RCT, the combination of TENS with amitriptyline resulted in significantly greater pain relief than use of sham TENS with the medication.442 Treatment may be less effective if there is gross sensory loss and poor distal vascular perfusion.443 However, as well as for ‘pure’ neuropathy, TENS has been used in treatment of the diabetic foot (see SubCh. 9.6, ‘painful diabetic foot’).

Strong (painful) TENS may well elicit axon reflex vasodilatation when applied to the top (dorsum) of the foot. However, this response may be absent or reduced in those with diabetic neuropathy (this in itself may be useful diagnostically,⁴⁴⁴ (see SubCh. 6.3, ‘axon reflex’). Various forms of TENS have been used to screen for peripheral neuropathy.⁴⁴⁵

Norman Shealy, in an informal report, noted that pain relief as well as some neurological improvement occurred in 80% of those with diabetic neuropathy treated using an experimental ‘GigaTENS’ (EHF) device.446

LILT has been used for diabetic neuropathy, although its effectiveness for the condition is by no means established.447

In some interesting research by Constantin Ionescu-Tîrgovişte and Romanian colleagues, electrical potentials measured through needles inserted ~1 cm into acupoints were reduced, and sensory threshold increased, both in diabetics compared with non-diabetics⁴⁴⁸ and in diabetics with neuropathy compared with diabetics without neuropathy.⁴⁴⁹

Comparisons and combinations

Norman Shealy reports good results for diabetic neuropathy by combining CES (1 mA for 1 hour daily, using the Liss device) with MRT using experimental GigaTENS. Each of the 12 acupoints in his ‘ring of fire’ protocol was irradiated for 3 minutes daily, using 52–78 GHz at 1 µW, for two 5-day courses separated by an interval of 2 days, and then 10 once-weekly treatments. Progesterone cream was also applied.⁴⁵⁰

Points used

As with other forms of neuropathy, the most frequently used point is probably ST-36. Other Stomach and Gall Bladder meridian points are often used in addition, perhaps less commonly with points on the Bladder, Liver, Large Intestine and Spleen meridians.

TENS studies usually focus on treating the leg and foot directly, perhaps with paraspinal lumbar electrodes in addition. An electrode sock, stimulating the whole lower leg non-specifically but particularly those regions where electrical skin resistance is low, is an interesting innovation⁴⁵¹ (glove electrodes are also available).

Parameters used

It appears that LF stimulation is used more frequently than HF in the treatment of diabetic neuropathy. DD EA and HF TENS have also been utilised. There is little information available on the current intensities employed.

HIV-related peripheral neuropathy

Neither MA nor amitriptyline was found to be superior to placebo in one study on HIV-related peripheral neuropathy.⁴⁵² However, both EA⁴⁵³ and pTENS⁴⁵⁴ were helpful in single case reports on this condition.

Joseph Odom, using Manaka-style ‘ion cords’ but with deeper needling than is usual with this method, claims good results with HIV-related peripheral neuropathy (see Ch. 10, ‘deeper needling’).

Points and parameters used

There is little consistency in point selection for neuropathy purely due to HIV/AIDS (without further iatrogenic neuropathy), and little information on appropriate stimulation parameters.

Restless legs syndrome

Unpleasant (not usually painful) sensations in the legs, often disturbing sleep and relieved by movement, may be associated with peripheral neuropathy⁴⁵⁵ as well as impaired peripheral circulation (see SubCh. 9.6). This ‘restless legs syndrome’ has been treated with MA,⁴⁵⁶ EA⁴⁵⁷ and TENS^458,459 ( Restless legs syndrome). The combination of TENS and vibration was more effective than either separately in one very painful case.⁴⁶⁰ Combined treatment with EA and a multifrequency lamp is another possible approach.⁴⁶¹

BL-56, ST-36 and SP-10 have been used, among other points, for restless legs syndrome, as well as interrupted, LF and HF stimulation. As yet there are too few studies to hazard a guess as to what approach is optimal for this problem, which like so many is multifactorial in origin.

A caution when treating peripheral neuropathy
Lack of sensation and poor tissue nutrition can lead to serious problems following even seemingly minor injuries. Needling should be carried out carefully. Moxibustion may result in severe burns.462 Charge-balanced stimulation should be employed, whether with EA or TENS.

Radiculopathy-related pain

Radiculopathy-related pain is due to a combination of mechanical and biochemical events. Compression of the nerve root creates oedema, leading to intraneural inflammation (neuritis) and hypersensitivity.⁴⁶³ Symptoms may be sensory (hyperalgesia, allodynia), autonomic (vasoconstriction, excessive sweating on movement or needling, ‘goose bumps’, peau d’orange, dermatomal hair loss), or motor (muscle shortening, reduced range of movement, thickening of tendons where they attach to bone).⁴⁶⁴

Thus radiculopathy may alter both motor and sensory function. Neurologically, pathology is confirmed by needle electromyography (EMG) and somatosensory evoked potentials (SEPs) respectively,⁴⁶⁵ although abnormal EMG findings may occur in those without symptoms and should not be considered in isolation.⁴⁶⁶ Motor tract stimulation with assessment of evoked potentials from several muscles simultaneously can also be useful for diagnosing motor radicular lesions.⁴⁶⁷ When needling paraspinal muscles for the purposes of EMG evaluation, patients may report reproduction of their symptoms (pain or paraesthesia),⁴⁶⁸ suggesting inadvertent TrP treatment while (or rather than?) confirming the level of the root lesion involved.

Indeed, for many medical acupuncturists, dry needling of myofascial TrPs is a key component of successful treatment for radiculopathic pain and related symptoms. Chan Gunn, for example, claims that all myofascial pain conditions are, in origin, radiculopathies that cause concurrent muscle shortening that in turn perpetuates the radiculopathy by increasing disc compression. For Gunn, even peripheral lesions (as in carpal tunnel syndrome) and osteoarthritis may be secondary to nerve root entrapment. Thus relaxation of paraspinal muscles is a key to successful treatment.⁴⁶⁹ Peter Baldry, in contrast, considers that only around 10% of those with myofascial TrPs require deep dry needling to release nerve root compression.⁴⁷⁰ Of course, there are causes to back and neck pain other than radiculopathy,⁴⁷¹ and differentiating whether these are ‘purely’ muscular or are due to nerve root compression or other nerve entrapment, let alone the aetiology of such entrapment, is by no means an easy task. As is well known, much chronic back pain remains a mystery, for all the hi tech scans and invasive interventions now available,^472,473 although careful evaluation obviously helps to reduce the proportion of those who fail to improve with standard care.^474,475 Here low back pain and sciatica will be considered. Neck conditions are covered later on in this chapter (see below, ‘Neck pain’).

The low back and sciatica

Back pain can be considered under various headings, based on whether or not nerve compression or entrapment is an important factor ( Thoracic and low back). Papers on ‘sciatica’ without emphasis on assessment of aetiology (herniated disc, lumbar osteochondrosis or spondylosis, ‘third lumbar transverse process syndrome’, etc.) have been listed separately in the database for this chapter ( Sciatica). Here, studies on sciatica and lumbar pain with radicular involvement are considered together.

Sciatica is usually due to pressure on a nerve root and ensuing inflammation (neuritis), although piriformis syndrome (see above, ‘Piriformis syndrome’) and some forms of neuropathy may give similar symptoms.⁴⁷⁶ Very occasionally, conjoined nerve roots may lead to low back pain and sciatica, even without impingement on the nerve.⁴⁷⁷ A nerve root problem is most likely if flexion is not problematic, but extension and side bending towards the problem side increases pain in the affected extremity.⁴⁷⁸ L4–L5 and S1–S3 nerve roots are those most commonly affected.⁴⁷⁹ Thus L4–L5 lumbar disc herniation is one possible cause of sciatic pain. As Daniel Traum has pointed out, if L3–L4 nerve roots are affected then pain is likely to radiate to the sacroiliac joints, hips, posterolateral thighs and fronts of legs, with anteromedial numbness, weakness of knee extension and a decreased knee reflex. If L4–L5 nerve roots are involved, pain will radiate in much the same way (but down the posterolateral rather than anterior lower leg). Numbness in this case will be lateral, perhaps extending into the dorsum of foot and big toe with weakened foot dorsiflexion but reflexes normal. L5–S1 and S1 nerve root involvement will result in a similar pattern of pain radiation with numbness in the lateral leg, calf, foot and heel, as well as in the sole. Plantarflexion may be weak and the ankle jerk reflex reduced or absent.⁴⁸⁰ As a rough guide, L3–L4 symptoms occur along the Stomach meridian, L4–L5 symptoms along the Gall Bladder meridian and L5–S2 symptoms along the Bladder meridian.⁴⁸¹

However, radicular symptoms usually take around 10 years, on average, to show up after an initial episode of acute traumatic low back pain,⁴⁸² and disc protrusion may be present in up to 50% of those aged over 40 who do not experience back or leg pain at all.⁴⁸³

Bilateral sciatica, with ‘saddle’ anaesthesia, lower-extremity weakness, loss of bladder control and impotence, is likely to indicate cauda equina syndrome rather than a radicular problem. Interestingly, patients with sciatica who do not recover well from surgery for herniated lumbar disc may well have a smaller-diameter spinal canal as well, and so a degree of stenosis⁴⁸⁴ (although not as severe as in cauda equina syndrome, when the nerve roots descending within the canal are themselves compressed).

Conventional treatments for nerve root problems include disc surgery, radiofrequency ablation of the dorsal root ganglion of the implicated nerve and epidural steroids.⁴⁸⁵ However, repeat surgery is often unhelpful,⁴⁸⁶ epidural steroids give only limited and unsustained relief,⁴⁸⁷ sciatica is generally considered amenable to electrostimulation⁴⁸⁸ and pain due to disc herniation is well documented as responding to non-invasive treatment in many cases.^489,490 For more resistant problems, SCS has been used – for chronic pain due to cervical radicular damage⁴⁹¹ or ‘battered root syndrome’ following spinal surgery.⁴⁹² In general, reversible radicular pain due to entrapment is more likely to respond to treatments such as acupuncture than established nerve damage (neuropathy).⁴⁹³ For this reason, as indeed with surgery,⁴⁹⁴ better results are likely if treatment is started within 3 months of sciatic pain onset. Sensory deficits are more likely to persist than motor problems.⁴⁹⁵

Chen Ke-zheng has written one of his useful brief reviews of various acupuncture methods on the treatment of sciatica.⁴⁹⁶

Sciatica is frequently treated with MA ( Sciatica).497,498 Even in the West, acupuncture has been used for treating sciatica since at least the 1850s,499,500 with EA in use since around 1920501 if not earlier. Non-invasive electrotherapy for sciatica has also been used for many years.502

In Russian studies, pain reduction was associated with improved alpha motoneuron reflexes, more on the healthy than on the affected side (using plum blossom needling together with EA),⁵⁰³ and also with improved local and systemic circulation (with subdermal needles retained for 2–3 weeks).⁵⁰⁴ MA (combined with electrotherapy) also enhanced electrical excitability of muscle, in patients with sensory and motor disturbance attributable to lumbosacral radiculitis.⁵⁰⁵ For an acute exacerbation of long-term sciatica, MA together with antidepressant medication both extended short-term pain relief and reduced the number of days that treatment was needed, compared with MA (and placebo medication).⁵⁰⁶

There are many studies on EA for sciatica ( Sciatica). Severe sciatica may require lengthy treatment with EA.507 The author of one study who preferred to use MA, reserving EA for chronic cases, noted that improvements in patients who had undergone spinal fusion were slower than in those with a laminectomy.508 And although sexual function in men with lumbar pain improved with treatment, it did so less in cases with severe radicular involvement.509

Case study 9.11.2 is an example of the use of electroacupuncture for acute back pain.

Case study 9.11.2 Electroacupuncture for acute back pain: a case study Juliette Lowe

Mrs B telephoned me with acute back pain: ‘I am stuck on the floor and can hardly move, please can you help?’ Mrs B is a 68-year-old woman who is normally very fit and healthy, regularly taking her dogs for walks and leading an exemplary moderate life. She was in training to walk the West Highland Way and had been walking a little more than usual. There was no obvious cause for the back pain. She had moved a slightly heavy piece of furniture 2 days previously, but had felt no pain in doing so. She did, however, have a history of back trouble, but no recurrence after an operation 7 years ago to remove calcium from the fifth lumbar vertebra. The pain was localised in the right sacrum and radiated down and around the right thigh into the medial knee (along the course of the dermatome associated with L2 and L3). The pain was of a deep ache in nature with sharp stabs on movement, and was slightly relieved by heat. The most comfortable position for treatment was prone with the right knee slightly bent and cushions under the stomach and right knee. On palpation, there was tenderness around the erector spinae muscles lateral to the right lumbar vertebrae, the right gluteus maximus and the right sacral foramen (all Bladder meridian points, from BL-23 to BL-34). The diagnosis was relatively straightforward, the type of pain indicating qi and Blood stagnation localised in the Bladder channel. This was probably due to a combination of the recent lifting and increased use on top of an old injury where there would be a tendency to stagnation from scar tissue formation. The treatment principle was simple: to move qi and relieve Blood stagnation in the localised section of the Bladder channel using local, adjacent and distal points. I needled the local and adjacent points BL-23, BL-25, BL-28 to BL-34 and distal point BL-40 (on the right), using 3.75 cm (1½") needles in the lumbar area and 5 cm (2") needles in the sacral points. I used electroacupuncture in this case, as the pain was acute and seemed to be coming from muscle spasm (cramp), having found this method useful in the past. I paired up BL-23 and BL-25, BL-28 and BL-29, BL-30 and BL-31, BL-32 and BL-33 and used a DD wave with frequency settings on my AWQ104 stimulator of f1 = 25 and f2 = 50 (these readings were reached after adjustments to produce a sensation that dispersed the pain). The intensity required to achieve adequate stimulation was moderate (2–3) on all points, and after 5 minutes I increased the intensity slightly (3–4) as Mrs B had got used to the level of stimulation. Initially she felt some relief from the pain and described a sensation as if the electroacupuncture were waving away the discomfort. However, after 15 minutes she began to find her position uncomfortable, so I turned off the machine and removed the needles. After she had moved I massaged her lower back, sacrum and buttock and gently stretched her legs. I then gave Mrs B a few simple exercises based on the McKenzie ‘Treat your own back’ series, involving lying on the floor and arching the back gently, slowly increasing as she was able to, followed by counterbalance stretching, lying on the back and pulling the knees into the chest. I advised her to do these for 10 minutes every 2 hours. After the treatment Mrs B was still in a lot of pain and I didn’t hold out much hope for my treatment. However, when I returned the following day, Mrs B was a transformed woman. She looked bright and alert, was walking with relative ease and even offered me a cup of tea! The pain was now only on certain movements and if she stayed in one position too long. It was localised in the right sacrum and no longer radiated down the leg. It was more of an ache, with only occasional stabs. I repeated the treatment, but used fewer points: BL-23 to BL-29, BL-32 and BL-40 (on the right) and paired up points BL-23 and BL-24, BL-25 and BL-26, and BL-28 and BL-29 for electroacupuncture. I used the same DD wave as before, but found Mrs B required a higher intensity (4–5) on all points, increasing up to 6–7 after 10 minutes. This time I left the needles in for 20 minutes and followed the acupuncture with a deeper massage and fuller range of stretches and gentle manipulations. Mrs B found no major discomfort during the treatment and felt much better afterwards. She telephoned me the following week to say that she was fully recovered. I was surprised how effective the treatment was after the apparently poor initial reaction but, having treated many similar cases before and gaining predominantly good results, I should have been more confident. On the whole, I use electroacupuncture for acute cases of pain and recent paralysis as I was taught to do at the London School of Acupuncture and the Railway Hospital, Nanjing. However, having worked in a hospital pain clinic for 5 years, I found that it could also be useful in many chronic conditions, often relieving pain rather than actually curing it. I therefore use it widely in my everyday practice, trying different methods (straight acupuncture, electroacupuncture, moxibustion, heat lamp and cupping, etc.) until I find a method that provides the best relief for the patient, regardless of the duration of the condition. I do, however, still find that in general electroacupuncture is more beneficial in acute conditions and moxibustion or heat lamp in chronic.

In one Japanese report on pTENS, younger patients (< 40 years old) responded better to this modality.510

TENS is often considered to be effective for low back pain.511,512 In particular, it has been used on board fishing vessels for radiculopathy exacerbated by cold and overstrenuous activity513 and, followed by exercise, in cases of recurrent ischialgia due to conjoined nerve roots. (Of 10 patients, 4 were almost pain free after 1–2 weeks of TENS, and the others had significant pain reduction.)514 TENS has also been used for pain associated with verified herniated lumbar discs, as part of a comprehensive treatment programme that enabled recovery in 90% of patients.515

Although paraesthesiae should be elicited by TENS in the affected area, the pain itself may be reproduced when TENS is applied; this may occur, for instance, with sciatic radicular pain.⁵¹⁶ In one mixed study, CTENS was found to be less effective for radicular than for purely musculoskeletal pain.⁵¹⁷ In keeping with this, TENS may be more useful when pain is relatively localised,⁵¹⁸ but less so for sciatic or radiating pain^519,520 or if several nerve roots are involved.⁵²¹ It will also be less effective if there are signs of denervation⁵²² (spinal nerve root compression, with ‘loss of function’^523,524).

MENS has been used for radicular problems.525

‘Microwave acupuncture’ has been used for sciatica,^526,527 sometimes combining microwave stimulation with needling,^528,529 or with other physical therapy methods.^530,531

LILT has been used for radicular pain,532 which may respond well to trigger point (TrP) LA.533 In one uncontrolled study in which both MA and LA were used, patients with sciatica alone obtained a similar outcome to those with low back and sciatic pain. Results were better in younger patients (< 64 years old), more prolonged in recent cases (< 6 months) and slightly worse in those who had already had surgery.534 However, results with LILT may be short lived,535 and severe spinal pathology is likely to be unresponsive to LILT,536 which may therefore be more helpful in less chronic cases. Nevertheless, reduced protrusion of herniated lumbar discs has been documented with both CAT scans and EMG following LILT (904 nm, 9 J/cm2, 3–5 times each week for 4 months). CAT scans showed no change following conservative therapy.537

In one uncontrolled study, ‘ultrasound acupuncture’ was found to be ‘excellent’ for sciatica.538

Iontophoresis at acupoints has been used for lumbar pain (including that due to disc prolapse), with results superior to those of conventional medication.⁵³⁹ However, in a study comparing MA and iontophoresis, neither was effective in cases of root compression associated with lumbar osteochondrosis.⁵⁴⁰

Acupoint/ashi point injection is not infrequently used for sciatica.^541,542,543 Simple saline injection (not necessarily at traditional acupoints) has also been utilised.⁵⁴⁴

Stretching and gentle movement to relax muscle paraspinal muscles can be important for sciatica sufferers.⁵⁴⁵ Inactivity may be counterproductive.

Although sciatica associated with lumbar disc herniation (protrusion) may respond well to acupuncture, if the protrusion is large or both spinal cord and nerve root are compressed then acupuncture is less likely to be beneficial and surgery may be essential.⁵⁴⁶ However, even if surgery is necessary to avoid injury to the nerves and muscle atrophy in the lower limb, acupuncture may still be helpful to aid recovery and reduce pain postoperatively.⁵⁴⁷

Comparisons and combinations

EA was better than MA for sciatic pain in one controlled study,⁵⁴⁸ and better than simple retained needling for intervertebral disc protrusion.⁵⁴⁹ EA was also preferable to MA for sciatica in other reports,⁵⁵⁰ for instance using periosteal stimulation (with both MA and EA).⁵⁵¹ However, in another CT, MA (based on TCM syndrome differentiation) gave better results than EA for sciatica.^552,553 Pekka Pöntinen has warned that exacerbation due to hyperstimulation is more likely with EA.⁵⁵⁴ Similarly, Sin Yoke Min has cautioned that some patients can become sensitised to EA, which should therefore not be used at every session.⁵⁵⁵ Perhaps for this reason, Wu Jinwei has recommended that only gentle EA be used in some patients.⁵⁵⁶

MA was superior to acupoint injection of Chinese herbal medication in one report on sciatica.⁵⁵⁷

EA and TEAS were both found effective for sciatica in one comparative study.⁵⁵⁸ However, in low back pain due to degenerative disc disease EA (‘PENS’) at non-acupoints gave better results than paraspinal TENS.⁵⁵⁹ Also, following surgery for spinal osteochondrosis, EA was more helpful for pain than TENS with the same sinusoidal waveform and frequency, although both methods were effective for associated motor deficit.⁵⁶⁰

MA was more effective than EA for sciatica in one comparative study in which syndrome differentiation was emphasised.⁵⁶¹

In one RCT, LA gave better results than EA and ultrasound acupoint stimulation in discogenic lumbosacral radiculopathies.⁵⁶²

Microwave acupuncture was superior to MA in its effects in one study.⁵⁶³

EA for sciatica has been combined with various treatments, including TENS,⁵⁶⁴ the shendeng ‘sacred lamp’⁵⁶⁵ or other multifrequency devices,⁵⁶⁶ massage⁵⁶⁷ and acupoint injection.^568,569,570 The latter combination was found to improve effectiveness,^571,572 as was the addition of both acupoint injection and warm needling⁵⁷³ or a multifrequency lamp⁵⁷⁴ to simple EA. EA has also been combined with both infrared heat and point injection.⁵⁷⁵

EA with warm needling gave better results than just EA.⁵⁷⁶ Similarly, TDP irradiation with either MA⁵⁷⁷ or EA⁵⁷⁸ gives better results than MA alone. In contrast, the addition of acupuncture to ‘biofrequency spectrum’ treatment did not improve results in another report.⁵⁷⁹

The combination of EA and moxibustion with massage gave significantly better results than their combination with traction in one study where ultrasound scanning of prolapsed intervertebral discs was used to confirm these improvements objectively.⁵⁸⁰ When combined with EA and herbal ‘steaming’ the results of traction were improved.⁵⁸¹ EA with medication applied to the skin, although perhaps more effective than EA alone, was not superior to combined massage and traction.⁵⁸²

As with cervicobrachialgia, combining MA with intradermally implanted needles improves results in radicular lumbar pain. Not only are benefits maintained for longer, but the combination has a greater effect on peripheral circulation as well.⁵⁸³ In acute episodes, combining MA with antidepressant medication (amitriptyline or pyrazidol) also increases the duration of analgesia after treatment and reduced the number of treatments required.⁵⁸⁴ MA is often combined with massage for sciatica,^585,586 or with cupping.⁵⁸⁷ In one report, it was combined with TENS, traction and HF electric field treatment.⁵⁸⁸

It is important to note that acupuncture (MA, EA and warm needling) gave better results in one study when combined with conventional conservative treatment.⁵⁸⁹ Also, MA together with Chinese herbal medicine (CHM) may give results little inferior to those of steroid nerve block and better than those obtained with local anaesthetic injection.⁵⁹⁰

Points used

Huatuojiaji⁵⁹¹ and ashi⁵⁹² points can be important when treating radiculopathic pain such as sciatica, but strong local stimulation at hypersensitive points, or at many points, should be avoided at least initially. Bladder meridian points (BL-11, BL-23, BL-36, BL-40, BL-54, BL-57, BL-60, BL-62), Kidney points (KI-3, KI-7), Gall Bladder points (GB-30, GB-34, GB-40) and points such as ST-36, ST-44, SP-6, Du-2, Du-3, Du-4, Du-15 and Du-20 may all be relevant.⁵⁹³ From the studies in the database ( Sciatica), the most used points for sciatica are those on the Bladder and Gall Bladder meridians: BL-23, BL-36, BL-37, BL-40, BL-54, BL-57 and BL-60, and GB-31, GB-34, GB-39 and GB-40. Of the adjunctive points, only ST-36 appears frequently, while dumai (Governor Vessel) points are well down the list behind huatuojiaji and ashi points. Auricular points are rarely used. For lumbar pain with a radicular component, of course the points are very similar: BL-23, BL-25, BL-36, BL-40, BL-54, BL-57, BL-60, GB-30, GB-31, GB-34 and GB-39. (Highlighted points are those that occur in one or other of these lists, but not both.) If anything, huatuojiaji and ashi points are even more emphasised, ST-36 appears somewhat less frequently than in the sciatica studies and auricular points are little used, but one of the most significant differences is the use of local points Du-3 and Du-4. Numerous methods of selecting appropriate treatment points have been devised over the years. A number of these may be worth investigating, such as the method of ‘corresponding’ points outlined by Sun Peilin,⁵⁹⁴ or the use of unusual points such as LI-10 for sciatica.⁵⁹⁵

With TENS, long electrodes can be positioned paraspinally to span several nerve roots (for instance, L1–L5 or L3–S2 for sciatica).⁵⁹⁶

For sciatic pain, one microcurrent advocate positions electrodes anterior and posterior to the affected nerve root,⁵⁹⁷ whereas a manufacturer of microcurrent devices recommends positive stimulation over the nerve root (at L4 or L5), the negative electrode being positioned at BL-40.⁵⁹⁸ With the Likon device, stimulation is suggested locally, or distally over the Bladder meridian.⁵⁹⁹

With LILT, Margaret Naeser and Xiu-Bing Wei suggest directing the beam at the nerve root site from three or four different directions, and then adding LA at points such as LI-4 (for cervical radiculopathy) or BL-60 or BL-67 (for lumbar radiculopathy).⁶⁰⁰

In one MA study of sciatica, the combination of a single body point with scalp stimulation gave better results than with body points alone.⁶⁰¹

When to use what – parameters for radicular pain

DD mode has been recommended for muscle spasm (cramp) in neurological conditions,⁶⁰² and so could be considered for radiculopathic pain. It has been used in a number of EA studies on sciatica and lumbar radicular pain, for example in database sections ( Sciatica) and ( Thoracic and low back).

In one report of EA for the sequelae of polio, stroke and sciatica, the strongest stimulation was required for the first of these and the least strong for sciatica.⁶⁰³ However, the author of one MA/point injection study noted better results for sciatica with stronger stimulation.⁶⁰⁴ In the main, although obtaining deqi is important and the usual phrase ‘to tolerance’ appears frequently in the EA sciatica and low back pain studies, my (biased!) impression is that high-intensity stimulation is not emphasised unduly for these conditions. Indeed, some authors have emphasised that treatment should be gentle, at least in some cases.⁶⁰⁵

The authors of one retrospective mixed pain CTENS study observed that, whereas results were comparatively poor for radicular pain, they did improve with higher intensity stimulation.⁶⁰⁶

An interesting comment was made by Tian Deming, author of one uncontrolled study on sciatica, who stated that HF EA can be used for patients in robust health, but that LF EA should be used if not. There are certainly more LF than HF EA studies in the database
( Sciatica and Low back pain with nerve entrapment), as well as a few in which intermittent rather than continuous stimulation is employed. Sometimes LF EA has been used at body points and HF EA at ear points. In an important Swedish study, although results with all acupuncture modalities were similar shortly after a course of treatment, 2 Hz EA resulted in longer-lasting improvements than HF EA for ‘nociceptive’ sciatica (due to disc prolapse or degeneration, or chronic lumbar sprain, but without excluding actual nerve damage).⁶⁰⁷ Adrian White has suggested that LF EA may have been more ‘convincing’ as it involved muscle contraction rather than just a tingling sensation.⁶⁰⁸

In an early trial comparing ALTENS and CTENS for chronic low back and leg pain, analgesia was short lived with both modes.⁶⁰⁹ More patients found CTENS to be helpful, possibly because ALTENS was too strong for comfort.

In a very small (N = 8) study comparing CTENS, burst and modulated TENS, three out of four sciatica patients responded best to burst TENS. At 2 months, two still benefited.⁶¹⁰ However, both motor level burst TENS and CTENS were found to give prolonged analgesia in one Russian study on lumbar and sciatic pain due to spinal osteochondritis.⁶¹¹

In one EA (‘PENS’) study on degenerative disc disease, 30 minutes of stimulation gave better results than 15 minutes.⁶¹²

When using microcurrent applied through pads, 20 minutes of stimulation has been recommended as a minimum treatment duration.⁶¹³

As for polarity, as mentioned (see SubCh. 9.10) ‘positive probe proximal’ may be appropriate, the distal probe being positioned at a point in the area of referred pain in the leg. As pointed out above, it is often considered that radiating pain is more difficult to treat than purely localised back (or neck) pain. Whether non-charge-balanced or DC stimulation would improve results for radiating pain compared with biphasic charge-balanced treatment has not been adequately researched.

Needle depth may need to be adjusted according to whether sciatica is acute or chronic. Apart from Gunn, a number of authors consider deep needling to be important.^614,615 In one MA RCT, however, there was little difference in outcome whether GB-30 was needled to a depth of 0.5–1 cun or 2–3 cun.⁶¹⁶

Case study 9.11.3 details the use of automated methods of intramuscular stimulation in a case of low-back pathology.

Case study 9.11.3 Automated twitch-obtaining intramuscular stimulation and electrical twitch-obtaining intramuscular stimulation Jennifer Chu

The use of traditional acupuncture and electrical acupuncture is common in the management of musculoskeletal pain. However, clinically noticeable twitches are not an occurrence with acupuncture methods. Twitch elicitation though, is inherent with electromyography (EMG) performed with the standard monopolar needle electrode.1,2 The presence of these twitches in EMG relates to the firmness and slightly larger size of the needle used in EMG and the intramuscular movements of this needle during EMG. The standard monopolar EMG needle electrode measures 16/1000 inch in (0.4 mm) diameter and is larger, stronger, and more efficient for penetration through thickened skin or deep muscle tissues than an acupuncture needle that measures approximately 10/1000 inch (0.25 mm) in diameter. The major advantage is that the monopolar needle has a Teflon coated shaft with exposure only at the conical tip. TheTeflon allows smooth oscillation of this needle within the muscle for twitch elicitation. Movements or oscillations of the monopolar needle during EMG at motor end plate zones (MEPZs) cause electrophysiological and clinical level twitches to occur.1,2,3 The motor end plate zones that can twitch readily are clinically identified by the presence of tenderness at taut intramuscular bands, ‘ropes’ or nodes. These zones have the same defining clinical features as those used for the identification of trigger points.4,5 Additionally, the intramuscular or intermuscular grooves define the motor lines and on needle stimulation along these lines, there is a linear yield of twitches. Of significant clinical importance is the fact that the traditional acupuncture points often lie along the intermuscular muscle grooves. Previous reports identify trigger points and acupuncture points to be the same points.6,7 The author’s clinical experience suggests defining the motor end plate zones (MEPZs) as the anatomic correlates of trigger points and acupuncture points along meridians crossing over muscles. The motor lines are the anatomic correlates of acupuncture meridians relating to the muscles that they traverse. Although manual insertion of the acupuncture needle does not readily elicit clinical level twitches, they can occur with intramuscular stimulation (IMS) as described by Gunn.8 In IMS, use of a plunger forces the acupuncture needle into the depths of the muscle and acts as a more definitive stimulus. However, the acupuncture needle is unsuitable for oscillation purposes since tissues have a tendency to adhere to the bare stainless steel shaft of this type of needle during oscillatory needle movements. Therefore, with repeated oscillations, the acupuncture needle’s original trajectory deflects away from that necessary for stimulation of targeted MEPZs. Also, the acupuncture needle will bend, twist or corkscrew when pushed into stiff and tight skin and muscle tissue. There is twitch occurrence with trigger point injections, a sign used for identification of trigger points.9 However, due to the local anaesthetics injected into the trigger point, there is suppression not only of twitch elicitation but also of the beneficial effects of twitches for pain relief. Therefore, the therapeutic significance of the twitch remained unrecognised. The hypodermic needle can penetrate through skin and muscle tissue easily allowing it to stimulate the MEPZs. However, the bevelled tip of such a needle produces a cutting edge, making it unsuitable for the repeated twitch elicitation in multiple myotomes essential in the management of chronic refractory myofascial pain syndrome. Gunn hypothesises that spondylotic radiculopathy related myofascial pain arises from the traction effect of shortened muscle fibres on tendons, nerves, blood vessels, bones, intervertebral discs and joints.8 He achieved immediate muscle fibre contraction and relaxation through oscillating, twisting and turning motions of the intramuscularly inserted acupuncture needle. However, these effects are very limited compared to the twitch-related muscle contraction and relaxation.10 Also twitches, especially those elicited from deeper regions, or those strong enough to mobilise large areas of the muscle, or even a joint, can cause more significant contraction, relaxation and intramuscular stretching effects. The exercise effects from the twitches also contribute to increase circulation to the areas treated11,12 and reduce thixotropy (stiffness) of muscle fibres.13 Long-term outcome studies using manually performed twitch obtaining intramuscular stimulation (TOIMS) for the relief of cervical and lumbosacral radiculopathic pain gave promising results.14,15 However, manual delivery of mechanical stimulation is relatively ineffective and can result in repetitive strain injury and cumulative trauma to the physician. The development of automation to help insert, oscillate and retract the needle from the muscle then became medically necessary. This method, termed the automated twitch-obtaining intramuscular stimulation (ATOIMS), utilises a handheld device. The device allows intensive quality treatments to multiple points in multiple myotomes of patients with diffuse nerve-related chronic muscle pain such as fibromyalgia. The device also removes the risk of cumulative and repetitive stress injuries to the treating physician. Since the ATOIMS device regulates and maintains only one smooth trajectory for the oscillating needle, it minimises tissue trauma, treatment pain, and post-treatment pain. There is potential for more tissue trauma with manual TOIMS. This is from the irregular trajectory of the manually oscillated needle and the multi-directional needle movements made during treatment. The ATOIMS device improves reliability, efficiency and accuracy of treatment and reduces the labour involved in searching for the major MEPZs that can produce strong twitches. It is also essential for treatment of areas with very stiff and tight muscle tissue. However, even the use of the ATOIMS device is not ideal for eliciting twitches, since mechanical stimulation is not an efficient method for depolarising nerve and muscle tissue. Electrical excitation of nerve is more powerful than mechanical stimulation for excitation of such tissues. The development of electrical twitch-obtaining intramuscular stimulation (ETOIMS) was necessary for increasing both the total number and the force of elicited twitches. The twitches evoked with ETOIMS are more forceful because submaximal micro-electrical current stimulation of an intramuscular nerve trunk has a bias towards stimulation of large motor neurons due to their lower electrical threshold.16,17 Also, intramuscular electrical stimulation at the motor points elicits axon reflexes that not only activate the stimulated axon antidromically but also invade other branches of the same axon to stimulate the entire motor unit.18 This is in contrast to focalised or fasciculation type twitches with little to no force, which arise from stimulation of individual axons of the most peripheral axonal tree.16 The more effective pain relief with ETOIMS underscores the importance of force magnitude and total number of intramuscular twitches elicited in the control of myofascial pain syndrome.19,20 To effectively stretch muscle fibres shortened from partial denervation or neurapraxia, the stretch should ideally occur in the immediate vicinity of the affected muscle fibres. Electrical fields have a steep radial decline with distance from a microstimulus source.16 Such selective stimulation of axons is even more restrictive for mechanical stimulation as with ATOIMS. Thus ETOIMS and ATOIMS, with the capacity to excite only the intramuscular peripheral axons and larger nerve branches closest to the stimulating electrode, can concentrate the twitch related stretch effects to shortened muscle fibres present in the region of the twitches. As a rule, the twitches occur immediately upon muscle penetration. The first in a series of twitches usually has the strongest force, or may be sustained as in a tetanic contraction. Not uncommonly, a single stimulus evokes multiple simultaneous twitches causing the muscle to flutter or go into a cramp. These twitch characteristics indicate a neurogenic component with re-excitation from presence of ectopic activity.21 The abnormal excitability is due to denervation supersensitivity and is also related to the presence of demyelinated or regenerating nerve endings.22 The pain-relieving effects of percutaneous electrical nerve stimulation (PENS) may be similar to those of ETOIMS. With PENS, electrical stimulation of very superficial motor points can occur with skin penetration. The stainless steel shaft of acupuncture needles used in PENS conduct unknown quantities of electrical current to tissues for 30 minutes.23,24 Tissues at the tip and sides of the shaft of the needle are unnecessarily stimulated for prolonged periods. With ETOIMS, the stimulation is 0.5 seconds/point and the maximum charge delivered to the tissues at each site is quantifiable and minute. The charge is only 0.4 microcoulombs on using the Teflon coated monopolar needle with an exposed conical tip of 0.3 mm2. The Teflon insulation prevents electrifying tissue adjacent to the needle’s shaft. In both ATOIMS and ETOIMS, the movement of the tissues due to the twitches allows the needle to take a slightly different route through the tissues. Even if there is re-insertion of the needle through the same skin insertion location, the needle will deliver the next mechanical or electrical stimulus to a new site, therefore minimising any prolonged or repeated injuries to the same sites. Denervation-related muscle fibre shortening from spondylotic radiculopathy can recur or become progressive with further acute or chronic repetitive trauma to the spinal nerve roots. ATOIMS and/or ETOIMS are suitable for such conditions since, in skilled hands, these treatments are without side-effects and are safe for repetitive use. A case history Chief complaint Back and left-sided groin pain Current condition A 36-year-old male came for management of left-sided back, buttock and groin pain dating back to his teens when he started developing pain over his low back at the age of 15 years. Since 1992 he also had groin pain associated with what was diagnosed as prostatitis. In 1997, he was evaluated by the author for an EMG. At that point, his left groin pain was attributed to spondylosis of the lumbosacral spine. It was recommended that he undergo ETOIMS/ATOIMS. He was lost to follow-up since then and his left leg and left buttock pain persisted. He returned to the author in January 2000, his pain level rated at 6/10 on average, down to 2 or 3/10 on a good day, but as high as 9/10 on a bad day. The pain radiates from the left side of the low back to the left buttock and down the left posterior thigh, as well as involving the left groin. Associated symptoms are weakness and some numbness down to the left toes. Pain is better with lying down and weekly massage, worse with lifting, bending, driving and sexual activity. It is severe at night, but rarely wakes him. He falls asleep easily, does not wake unusually early and can sleep for 8 hours without medication. He has tried ultrasound therapy and physical therapy, as well as his weekly massage. Past medical history In a car accident 5 years previously he sustained some left shoulder pain for which he received physical therapy for 3 months. His pre-existing low-back pain was aggravated. He lost his left index finger in a sawing accident. Review of systems The patient has no blurring of vision, headaches or loss of consciousness, nor any chest pain or shortness of breath. He has no liver or kidney problems, but he does urinate about 10 times daily, as well as twice nightly. He has no erectile dysfunction, but some difficulty with ejaculation during intercourse. Physical examination He has a full range of motion in neck flexion and extension. He has mild limitation of motion in neck lateral flexion, on neck rotation and on shoulder external rotation, with moderate limitation of motion in shoulder internal rotation. He has full range of motion on shoulder flexion, abduction and extension on the left, with mild limitation of motion on shoulder abduction on the right and with a full range of motion on shoulder flexion and extension on the right. Shoulder and pelvis levels are elevated by 1 cm on the right. Spine flexion has a full range of motion. He has a full range of motion on spine lateral flexion and mild limitation of motion on extension. The supine straight leg raise test is 80° bilaterally. The sitting root sign is positive over both hamstrings with a grade II pain. There is negative Trendelenburg and the patient is able to do a full squat with no tight heel cords, as well as a full heel lift and toe lift bilaterally with alternate unilateral stance. Examination of tender points shows mild tenderness over the L2 to the S1 dermatomes as well as the C2 to C8 myotomes. There is grade 2+ paraspinal spasm at the T10 to S1 levels, with the left greater than the right. There is grade 2+ cervical paraspinal spasm with the left greater than the right and grade 2+ trapezius spasm on the left, greater than the right. Autonomic changes include skin thickness of +1 over the cervical, thoracic and lumbar spine. The pore size is +1 over the cervical and +2 over the thoracic and lumbar areas. Trophic neurogenic edema is noted to be +1 over the C2 through C8 dermatomes. There are pilomotor changes of +1 over the C2 to S1 level, as well as vasomotor changes of +1 from C2 to C5 dermatomes. Neurological examination The gait has good heel, toe and tandem walk. Reflexes are brisk and symmetrical bilaterally, with no Babinski, clonus or Hoffmann’s sign elicited. Assessment The patient was found to have nerve-related chronic low back and groin pain, which was probably secondary to spondylotic radiculopathy over multiple levels. Recommendations The patient was recommended to undergo weekly electrical ETOIMS/ATOIMS treatment over the lumbosacral paraspinal muscles as well as the muscles that wrap around the hip girdle, including the L2 to S1 myotomes. Treatment goals These were to decrease the intensity of pain and frequency of painful exacerbations, increase the range of motion and maximise the overall level of performance as regards daily activities. Results After the first treatment, his pain level decreased to 3–4/10, with most pain relief in his low-back region, but also a decrease in the burning painful sensation in his left groin. During treatment, all muscles treated in the L3 to S1 myotomes and the C4 to C7 myotomes showed difficulty in eliciting twitches owing to their tightness from chronic underlying nerve root involvement. After the fourth treatment, his pain level was as low as 2/10. The patient has continued with weekly treatments, was able to play golf for the first time since his pain was exacerbated in 1997 and has experienced improvement in his quality of life. He continues with treatment every 6–8 weeks to maintain his quality of life over a total follow-up period of 4 years since January 2000. For description of ATOIMS and ETOIMS devices refer to: 1999 Intramuscular Stimulation Therapy Facilitation Device and Method. United States Patent 5,968,063 2002 Intramuscular stimulation apparatus and method. United States Patent 6,532,390

Complex regional pain disorder

Complex regional pain disorder (CRPD) has been described as a devastating disease.⁶¹⁷ It was first recognised as a clinical entity in 1863 by Silas Weir Mitchell, who observed the burning pain caused by gunshot or other nerve injuries sustained in the American Civil War. Today, in the US alone, some five to six million people are estimated as suffering from CRPD.⁶¹⁸ The pain, which may affect one or more extremities, can become intense and quite disproportionate to the underlying pathology.⁶¹⁹ CRPD may follow even slight injury, which means it is often underrecognised particularly in children^620,621 (however, one perhaps unsympathetic reviewer considers it is over diagnosed!⁶²²). Estimates vary widely on the proportion of sufferers likely to recover spontaneously.

CRPD may come in two guises,⁶²³ which are not always clearly distinct. Type I corresponds to what used to be called reflex sympathetic dystrophy (RSD), and is more likely to be diffuse, without following any dermatomal or other logical distribution.^624,625 CRPD type II tends to have a more clearly somatic origin, and corresponds to causalgia (pain and hyperalgesia, confined to the distribution of an injured peripheral nerve⁶²⁶).

CRPD can accompany repetitive strain injury (RSI), peripheral neuropathy or entrapment/compression syndromes, or follow burn or fracture injury, surgery (spinal⁶²⁷ or mastectomy,⁶²⁸ for example) or other insult to an organ (for instance, after myocardial infarction or as ‘shoulder–hand syndrome’ following stroke,⁶²⁹ see SubCh. 9.2). It has even been suggested that interstitial cystitis may be a form of reflex sympathetic dystrophy.⁶³⁰ Not only sensory but also sympathetic nerves can be involved (hence the earlier name ‘reflex sympathetic dystrophy’, RSD). Various central and peripheral mechanisms may be implicated.^631,632 Following hand surgery, for instance, it has been suggested that CRPD type II (sometimes termed ‘algodystrophy’) may correspond to an exaggeration of normal post-traumatic inflammatory responses rather than sympathetic damage, and can be prevented by minimising postoperative inflammation, increasing venous return (especially by avoiding constrictive dressings or casts), controlling post-traumatic pain and promoting early active mobilisation.^[633] If this is so, it may provide a model for treating some non-iatrogenic CRPD if the initial injury does not go ignored.

In the first stage of CRPD (type I), sympathetically maintained pain is accompanied by sensory changes, allodynia, hyperalgesia, non-pitting oedema and sudomotor and vasomotor changes (the affected extremity is likely to be warm, dry and red or slightly cyanotic). However, sympathetic symptoms such as local sweating or changes in skin temperature and colour may not be present,⁶³⁴ and the degree of hyperalgesia or allodynia experienced may not correlate with the intensity of spontaneous pain.⁶³⁵ Even at this early stage, there may be localised changes in hair growth and nail texture.⁶³⁶ Stage 1 usually lasts for 1–3 months, but may persist indefinitely.⁶³⁷

In the second stage, the symptom picture widens and may include inflammatory changes in the skin, neurodermatitis, bruising, tremor, joint swelling, insomnia due to night pain, and emotional disturbances. The skin is likely to become shiny, pale, cool and clammy.⁶³⁸ Motor function can be impaired.⁶³⁹ Hair growth now diminishes, and there may be changes in bone structure.⁶⁴⁰ In some stage 2 CRPD, serum noradrenaline may be lower on the painful side than on the healthy side.⁶⁴¹ In the third stage there may be spontaneous skin ulceration, infection, muscle atrophy, flexion deformities and osteoporosis, as well as the depression that may accompany any form of chronic pain.⁶⁴²

Treatment for CRPD has to be intensive and multimodal. Ted Priebe, on whose useful account of CRPD much of the above description was based, recommends that exercises should be given along with acupuncture on the same or alternate days. Other practitioners precede occupational or physical therapy with sympathetic blockade, for example: stellate ganglion blocks for upper limb problems and lumbar block for lower limb symptoms.^643,644 At least six sympathetic blocks may be needed before there is any response (as with acupuncture the effect can be cumulative). It is important that exercise should not be forced or painful (‘only gain with no pain’ rather than ‘no pain, no gain’⁶⁴⁵). Surgery is likely to exacerbate the problem.⁶⁴⁶

As with PHN, treatment has to be started in the early stages of the condition (certainly within 6 months⁶⁴⁷) for best results. Response may be slow, and once the condition has become chronic, it may become intractable.⁶⁴⁸ (One estimate is that this occurs in around 40% of CRPD patients,⁶⁴⁹ another in some 10%.⁶⁵⁰) If it does respond, treatment may need to be repeated at intervals indefinitely.⁶⁵¹ Spontaneous remission is rare.⁶⁵² However, in children full recovery is much more likely than in adults.^653,654

MA has been used clinically for CRPD, results being better than with sham stimulation (however, the traditional acupuncturist who started this study was unable to complete, so the sample is small and the outcome as a result only suggestive of benefit rather than clearly positive).655 In another very small trial, results with 20 MA (at bilateral HE-5, SI-3, P-6, SJ-5, LU-7, LI-4 and LI-19) appeared superior to those of sympathetic guanethidine block, with better effects on circulation in the affected extremity, as well as on pain in some cases.656 MA has been used for CRPD-like shoulder–arm pain following stroke (see SubCh. 9.2) and heart surgery.657,658

Theoretically acupuncture, which can reduce both sympathetic overactivity and pain, should be helpful for CRPD.⁶⁵⁹ Furthermore, in experimental studies both MA and EA have been shown to reduce some components of acute neurogenic inflammation,⁶⁶⁰ and thus may possibly be of benefit as a preventive measure in the very early stages of CRPD.

Perhaps for this reason, various types of acupuncture (such as magnets or moxibustion used at acupoints) have been found helpful by children with CRPD (their parents were less convinced).⁶⁶¹ In general, however, clinical results with both acupuncture and TENS for CRPD have been described as ‘not very impressive’.⁶⁶² Thus Joan Hester has commented that whereas acupuncture may be helpful in early stages of the disease it is not as effective as sympathetic blockade and should be used with caution in patients with hyperpathia (hyperalgesia or allodynia or both, possibly prolonged).⁶⁶³ In contrast, in some cases Dharma Singh Khalsa has found acupuncture useful in later stages of the condition when blocks are no longer effective.⁶⁶⁴ Also, EA was more effective than MA in one study of poststroke shoulder–arm syndrome (not CRPD proper?)⁶⁶⁵ ( Hemiplegic shoulder pain and subluxation).

CTENS was not found useful for CRPD in some early studies666 and reviews,667 and it has not infrequently been noted that TENS becomes less effective over time for CRPD, as for other conditions.668,669 In one small study comparing TENS and SCS, both gave better results than sympathetic block or sympathectomy. Although there were more ‘excellent’ outcomes with TENS, there were fewer total failures with SCS.670 Thus some authorities have considered TENS useful in general for the condition,671,672 especially in combination with gentle exercises and relaxation.673 For example, TENS may be particularly useful for CRPD in children,674,675 together with a graduated programme of physical therapy.676

The localised cooling that can sometimes accompany CRPD may be improved by TENS.⁶⁷⁷ In one quite large retrospective study, TENS (15–25 Hz, modulated) used for 22 hours each day made a considerable contribution to successful rehabilitation not only in terms of pain control, but also for vasomotor symptoms. It was noted that during the first 2 weeks of treatment symptoms returned rapidly if TENS was discontinued, but that after 2–4 months the unit could be dispensed with for several days before symptoms returned.⁶⁷⁸

Unusually, TENS has been used for CRPD following mastectomy.⁶⁷⁹

Ryodoraku (pTENS) treatment has been given in cases of CRPD.680

There is one case report of CES for ‘full body’ CRPD associated with brain injury. Quality of life and ability to work improved markedly.681

LA has been used for CRPD following surgery, with improvements in oedema, diffuse pain and extreme temperature fluctuations. MA had been unable to help this case.682 LA has also been combined with infrared in the treatment of CRPD.683

Static magnetic fields have been found helpful for CRPD in small uncontrolled studies.684,685 PEMF, with its effects on bone repair (see Ch. 4) and osteoporosis,686 may be worth considering as an adjunctive treatment.

Other electrotherapy methods that have been investigated for CRPD include high-voltage pulsed galvanic (HVPG) and athermal pulsed shortwave (Diapulse) ( Complex regional pain disorder (CRPD)).

Both skin temperature and conductance measurements have been used to monitor changes in CRPD with ryodoraku stimulation.⁶⁸⁷

Points used

From the studies in the database ( Complex regional pain disorder (CRPD)), it appears that some authors emphasise local or ashi points, or both, although these should probably be used with caution initially. The most commonly used traditional acupoints are P-6, LI-4 and LI-11.

For lower-limb CRPD, Alejandro Katz has used microcurrent at ST-36 and the bafeng points between the toes ipsilaterally, or contralaterally for a few treatments if the condition is too painful initially. He positions surface electrodes over the needles.⁶⁸⁸ In a useful overview of CRPD treatment from a TCM perspective, John Stebbins recommends MA, followed with microcurrent stimulation, at points such as SJ-5, GB-34, LI-11, ST-36, ST-40, SP-6, SP-9, baxie and bafeng, together with points on the affected meridians both proximal and distal to the affected area.⁶⁸⁹

However, when using TENS, Ronald Melzack and Patrick Wall found that stimulating proximal to an area of causalgia could markedly reduce hyperaesthesia, whereas stimulation distal to the area could aggravate it.⁶⁹⁰ Furthermore, TENS (or other stimulation) applied to the affected limb/area will probably feel stronger and be tolerated less easily than on the unaffected side, and may lead to uncomfortable spreading sensations that tend to outlast the stimulus.⁶⁹¹ Contralateral treatment may thus be advisable.⁶⁹² In contrast, ipsilateral paravertebral (C6–T3) and anterior neck electrode positioning to ‘sandwich’ the stellate ganglion was found helpful for upper-limb CRPD in one small Italian study.⁶⁹³

There is no reason why non-local auricular EA could not be tried with CRPD. Auricular EA was more effective than MA in one study of acute neurogenic oedema (see SubCh. 6.3, ‘reduced neurogenic’), and ear points have been stimulated in several studies on CRPD.

When to use what – parameters for CRPD

From the studies in the database ( CRPD and central pain), there seems to be little consistency in the parameters used to treat CRPD with EA and TENS. EA has been applied at both high and low frequencies, at ‘maximum tolerable’ intensity or ‘maximum stimulation without discomfort’. Both CTENS and ALTENS have been employed. There are also several MENS (microcurrent TENS) studies, with varying protocols.

In animal experiments, deep needling may be more effective than superficial needling for acute neurogenic inflammatory oedema (see SubCh. 6.1, ‘ST-36/GB-30 MA’). However, very strong EA (70 mA!) may aggravate it, although EA at a more clinically appropriate
5 mA contains it (see SubCh. 6.4, ‘EA at 5 mA’).

CRPD is a chronic neurogenic condition, and as such may well be exacerbated by strong treatment local or distal to the problem area (see Ch. 12, ‘CRPD’). Thus Christer Carlsson advises against deep needling for CRPD.⁶⁹⁴ Katz similarly recommends superficial needle insertion and subthreshold stimulation only.⁶⁹⁵

CTENS is sometimes recommended as appropriate for CRPD.⁶⁹⁶ 30 Hz TENS, ‘to tolerance’, was used in the Italian study on stellate ganglion stimulation mentioned above.⁶⁹⁷ In a single case report of TENS for CRPD in a young child, 50 Hz (3.5 mA) was eventually found to give better pain relief than 90 Hz (2.5 mA), used initially.⁶⁹⁸

Extreme temperatures (ice packs or radiant heat) should be avoided.⁶⁹⁹

The overall message would appear to be that gentle stimulation is most appropriate for CRPD, and that it should be started as early as possible. Treatment may need to be prolonged (see above, ‘Response may be slow’).

Central pain

Central pain is defined by the International Association for the Study of Pain as pain initiated or caused by a primary lesion or dysfunction in the central nervous system.⁷⁰⁰ There are several reviews of the various central pain syndromes and their treatment.^701,702

In an experimental study, EA (and MA) at SI-17 had an anti-inflammatory effect on neurogenic inflammation of the brain’s dura mater greater than that of prednisolone, although this was less than that of saline acupoint injection (see SubCh. 6.1, ‘dura mater’).

TENS has been used in various types of central pain. However, some authors consider that TENS is ineffective against pain involving ‘disordered regulation’ of the central mechanisms modulating nociception.703

5-hydroxytryptophan (5HTP) the serotonin precursor has been investigated for its effects on central and deafferentiation pain,⁷⁰⁴ with potential implications for its combination with electrostimulation methods.

Pain following stroke

The usual image we have of the stroke victim is of someone with hemiplegia or at least a degree of motor impairment. However, sensory pathways and areas in the brain can be affected as well in the so-called ‘thalamic syndrome’, leading to pain that is sometimes very severe⁷⁰⁵ but often begins only some months after the stroke itself. This central poststroke pain occurs in up to 8% of patients. It is usually burning and is accompanied by sensory deficit, as well as by allodynia (in around half of patients). As with so many neurogenic pains, it may be reduced by adrenergically active antidepressants such as amitriptyline, but treatment must be started early to take effect.⁷⁰⁶

Perhaps because motor impairment is so much more visible than central pain, there is a relative paucity of studies on electrotherapy for the latter. Indeed, some authorities consider it unlikely to respond to peripheral treatment methods,⁷⁰⁷ with stimulation of the brain itself being used instead as a last resort (deep, to the thalamus or central gyrus, or more recently motor cortex stimulation,⁷⁰⁸ sometimes via extradural electrodes⁷⁰⁹). Frequent technical and clinical monitoring is necessary in such cases.⁷¹⁰

In one controlled study of EA for thalamic pain following stroke, 100 Hz stimulation of huatuojiaji points (M-BW-35) daily for 30 days gave results statistically equivalent to those of oral carbamazepine711 ( Central pain).

TEAS (at LI-4 and ST-36),⁷¹² TENS^713,714 and SCS have all been used for central poststroke pain.⁷¹⁵ (Indeed, the mechanism of EA at paraspinal points is probably similar to that of the more invasive method of SCS.)

Both ( Central pain) ipsi- and contralateral stimulation were used in the TEAS/TENS studies. In one of these, results were similar whichever side was stimulated. In the TEAS and other TENS study, emphasis was on treating the healthy side. In part this was because CTENS or TLEA becomes ineffective if there is sensory loss in the painful area to be stimulated (see Ch. 12).

However, 100 Hz was effective in the EA study (albeit at paraspinal points), whereas in one small TENS trial (N = 15), of those patients that did respond (four, or some 27%), three (20%) appeared to respond similarly to both CTENS and ALTENS, with only two benefiting from contralateral treatment.

Pain following spinal cord injury (SCI)

About 50% of patients with spinal cord injury (SCI) suffer from persistent central neurogenic pain.⁷¹⁶ The combination of motor and sensory impairment with such chronic pain can be truly devastating, limiting an already seriously impaired quality of life.⁷¹⁷ Central pain associated with SCI is defined as dysaesthesia or deafferentiation pain, characterised by burning, tingling or aching below the level of the lesion. This is associated with abnormal electrical activity in the regions of the thalamus that have lost their normal somatosensory input.⁷¹⁸ Incomplete injury correlates with better prognosis, including motor improvement (whether the lesion is complete or incomplete can be determined by digital rectal exam)⁷¹⁹

In one survey of chronic pain associated with spinal cord injury, conventional treatments considered most helpful were opioid medications, physical therapy and diazepam, whereas those rated least helpful were SCS, counselling or psychotherapy and administration of acetaminophen (paracetamol) or amitriptyline. Alternative treatments reported as most helpful were massage therapy and use of marijuana, whereas acupuncture was tried by many but rated only moderately helpful⁷²⁰ (presumably both diazepam and marihuana were found helpful for pain associated with muscle spasticity). In another smaller survey, this time of CAM alone, although acupuncture was the most frequently used modality of the four considered; it was rated lowest for satisfaction with pain relief (massage was rated highest; and chiropractic and herbal medicine were also considered).⁷²¹ Acupuncture was not found helpful in another case report of pain following SCI.⁷²²

However, in a careful but uncontrolled study of MA for chronic SCI pain (N = 22), nearly half the patients treated showed improvements in pain intensity and other aspects of their condition. More than a quarter reported more pain, which was still present 3 months later. Results were poorer in those with central pain or complete transection of the spinal cord and for pain below the level of injury.723

EA was also found helpful for chronic SCI pain, although once treatment was terminated the pain started to return.724 In contrast to the results with MA, EA was found helpful for neuropathic pain below the level of the spinal lesion, particularly when this was bilateral, symmetrical and constant. However, this study (N = 36), though carried out by experienced practitioners, was retrospective and uncontrolled.725

Auricular MA and TEAS (at SI-3 and BL-62) have been used together for acute SCI pain. Early treatment gave encouraging results in this CT.726

TENS has been used for pain following SCI,727 although of course not always successfully.728 In particular, as with lumbar pain in general, results may be better for localised rather than radiating pain.729 Long-term TENS and TEAS treatment have been reported as being of some benefit for sensory deficit in the legs following spinal cord injury, even though started only weeks or even years after the initial trauma. No detail of the sensory improvement was given.730

CES (including SPES) has been used for chronic SCI pain, with very noticeable effects on patients’ mood in some reports^731,732,733 but not others.⁷³⁴ SCS has been used successfully for neuropathic pain subsequent to spinal surgery.⁷³⁵

Neurogenic pain and numbness following penetration of a broken acupuncture needle into the spinal cord at C1/C2 level has been reported,736 as has asymptomatic penetration at C2/C3 level.737 In both cases, the needle had to be removed surgically, with no adverse effects postoperatively. There are other reports too of spinal cord injury following acupuncture.738 These all occurred in Japan, where the needles used for MA are often thinner than those recommended for EA.

Cauda equina syndrome, spinal stenosis and arachnoiditis

Bilateral sciatica, with ‘saddle’ anaesthesia, neurogenic claudication, lower-extremity weakness (a shuffling gait), loss of bladder control and impotence is likely to indicate cauda equina syndrome due to compression of the nerve roots within the central spinal canal at lumbar level rather than (or in addition to) a radicular problem laterally, at or after their point of exit. Unless surgical decompression is carried out promptly, there may be irreversible neurological changes.739 Indeed, even invasive SCS may not be effective for painful cauda equina injury.740 Some spinal stenosis is likely to respond to conservative therapy, however.741 Thus EA has been used, with some success, for symptoms due to such spinal compression.742

Stenosis may be due to compression not just by the canal walls, but also from intervertebral discs, epidural fat or spinal ligaments.⁷⁴³ Pain due to spinal cord compression is not infrequent in cancer patients.⁷⁴⁴ As with recovery from spinal surgery (see above, ‘smaller-diameter spinal canal’), patients with stenosis in addition to disc pathology may not respond well to acupuncture interventions⁷⁴⁵ (MA, EA, LA⁷⁴⁶ or ultrasound acupoint stimulation).

Arachnoiditis (arachnitis), or inflammation of the arachnoid membrane of the spinal cord, can lead to severe pain as well as paralysis. It has been treated with EA in combination with moxibustion.^747,748 In one report, it did not respond well to EA alone.⁷⁴⁹ Nor were CTEAS or CTENS found particularly helpful.^750,751

Points used

Denis Castillo has made the interesting suggestion that auricular and other acupoint microsystems may bypass spinal pathways, and so be useful when the spinal cord itself is damaged.⁷⁵² However, this is probably not an idea that would meet with much approval from neurophysiologists.

When to use what – parameters for central pain

CTENS or TLEA is unlikely to benefit central pain, as it involves segmental rather than supraspinal mechanisms (see Ch. 12).

Ottoson and Lundeberg, in their textbook on TENS, appear to suggest that CTENS is most likely to benefit deafferentation pain,⁷⁵³ but it has also been suggested that burst TENS may in fact give better results for central pain due to spinal cord or brainstem injury.⁷⁵⁴

Scars and neurogenic pain

Scar tissue can be associated with neurogenic pain due to the presence of neuromata, or the development of hyperaesthesia, for instance following surgery.⁷⁵⁵ (Treatment of scars is covered in SubCh. 9.4, ‘Scar tissue and adhesions’)

Head and facial pain

Head and facial pain can arise from various causes, neurological, musculoskeletal, dental or stomatological (see SubCh. 9.4). Pain can also arise from the eyes, the ears, or the nose and sinuses (SubCh. 9.4). In this section, headache, facial and dental pain will be considered.

TENS has been used as a treatment for both head and neck pain.756 Headaches of various types, trigeminal neuralgia and dental or orofacial pain are all considered to be standard indications for TENS.757

Headache

Recent data show that in the UK there are some 170 000 and 390 000 GP consultations for headache and migraine respectively each year, with one in eight suffering impaired ability to work on more than 20 days per year, and 40% having at least 1 day off work per year.⁷⁵⁹ Headaches are not limited to adults. In Germany, idiopathic (primary) headache is probably the most common symptom in children of school age; a fifth may be affected.^760,761

There are various sorts of headache. According to the International Headache Society, they can be subdivided as follows:

• Tension-type headache: steady, non-throbbing bilateral pain of the head, back of the neck and face. It may last up to several hours, and occur several times within a week. Usually episodic (< 180 days per year), but can become chronic (> 180 days/year). Although the autonomic symptoms typical of migraine are absent or mild, both types of headache may occur in the same patient,⁷⁶² and one may possibly lead to the other.⁷⁶³ It has been suggested that, in tension-type headache, tender or tight scalp points may be found predominantly along the Bladder meridian.⁷⁶⁴ In Western acupuncture terms, cervical myofascial trigger points are commonly implicated as contributing to headache.⁷⁶⁵

• Cluster headache: penetrating, non-throbbing and severe pain behind the eyes or in the temples, lasting for less than 2 (possibly 4⁷⁶⁶) hours, and continuing for 2–3 months at a time, often occurring at night.

• Post-traumatic headache: due to head or neck injury; it may be experienced as dull, aching, stabbing, sharp, or excruciating at the site of injury, and can occur in various time patterns, sometimes starting years after the initial trauma.

• Migraine headache: unilateral and intense head pain that may last several days, typically accompanied by nausea or vomiting, visual disturbances and extreme sensitivity to light, with a variety of other accompanying symptoms (sometimes some of those of a cold, for instance). Some authors suggest that, in migraine headache, tender or tight scalp points may be found predominantly along the Gall Bladder meridian.⁷⁶⁷

Headache may also be secondary to many other conditions such as viral infections, dental, jaw or sinus problems⁷⁶⁸ (see below, ‘Dental pain’, and SubCh. 9.4, ‘Sinusitis’), temporomandibular joint (TMJ) disorders (see below, ‘Temporomandibular joint dysfunction’), hypertension⁷⁶⁹ (see SubCh. 9.6), stroke (see SubCh. 9.2), intracranial aneurysm (sometimes with subarachnoid haemorrhage) or tumour, functional hypoglycaemia⁷⁷⁰ and opiate withdrawal⁷⁷¹ (see SubCh. 9.15).⁷⁷² Acute, severe headache in some of these conditions may signal a medical emergency. In all of them, headache can become a long-term problem.

In addition to these general categories, chronic daily headache (CDH) refers to frequent headaches that occur on 15 or more days a month. CHD may be primary or secondary, and is usually tension-type headache as indicated above, although it may also represent a chronic development of episodic migraine. The transformation from episodic to chronic may be triggered by drug abuse,⁷⁷³ so clearly non-pharmacological methods of treatment may be important, both to prevent such transformation occurring and also for those who already suffer CDH.⁷⁷⁴ Habitual snoring in someone with chronic daily headache is likely to indicate sleep-disordered breathing as a contributory factor.⁷⁷⁵

A useful measure of headache severity is the Headache Impact Test (HIT), which is available at www.headachetest.com and www.amIhealthy.com.⁷⁷⁶

CAM is frequently used by headache and migraine sufferers,^777,778 and acupuncture is no exception,^{779,780,781,782,783,784,785,786,787,788,789} with positive results reported in a number of observational studies.^790,791,792 EA in particular has been used for headache,⁷⁹³ as has ryodoraku⁷⁹⁴ ( Migraine) and ( Uncategorised/mixed headache).

At one time, it was claimed that around a third of patients receiving acupuncture in the West suffered from some form of headache.⁷⁹⁵ In one very large, more recent German survey of some 40 000 acupuncture patients with previously diagnosed low back pain, migraine or tension-type headache, or knee or hip arthrosis, around 26% claiming relief from pain were headache sufferers.⁷⁹⁶ However, some reviewers have considered as equivocal the evidence that acupuncture has any effect beyond that of a placebo in the treatment of headache,^797,798 or as inconclusive when acupuncture for headache is compared with other forms of treatment.⁷⁹⁹ Others have suggested that, whereas acupuncture may appear to benefit non-migraine headaches, this may be just because tension headache sufferers tend to be more responsive to non-specific effects.⁸⁰⁰ It also appears from some MA studies that this form of acupuncture may reduce headache frequency and analgesic consumption, rather than headache duration or pain severity.⁸⁰¹

In contrast, the 1997 NIH Consensus Development Conference panel concluded that there is good evidence that acupuncture can benefit headaches, although this does require further substantiation.⁸⁰² Recent and larger systematic reviews have also been slightly more positive. A Cochrane review on acupuncture for idiopathic headache, for instance, concluded that ‘patients should not be discouraged from trying acupuncture from a risk/benefit standpoint’.⁸⁰³ Critics of this review predictably observed that no methodologically rigorous studies were included that assessed clinically relevant outcomes, and that the lower quality trials reviewed resulted in more positive results. They concluded that ‘recommendations cannot be made to implement the use of acupuncture in the treatment of idiopathic headache based on current evidence’.⁸⁰⁴ Nevertheless, some of the Cochrane authors went on to complete another systematic review of acupuncture including EA and LA, this time for chronic headache (24 RCTs, 35 non-randomised studies). Here again, acupuncture appeared to have a positive effect, although better quality studies produced lower response rates, as usual. Furthermore, in this review it was clear that neither the patients nor the interventions in the RCTs were representative for routine acupuncture practice.⁸⁰⁵ In this context, it would appear that some less formal reviews are perhaps rather overenthusiastic about the benefits of acupuncture for headache.⁸⁰⁶

A useful if short overview of headache treatment with different acupuncture methods has been written by Chen Ke-zheng.⁸⁰⁷ Jean Bossy has written a longer account of various acupuncture approaches to headache and other head pain,⁸⁰⁸ and Johannes Bischko several reports on his own methods, which have influenced many European practitioners.^809,810 Julian Scott has published a comprehensive TCM analysis of some 40 types of headaches.⁸¹¹

Adam Lewenberg has described how EA (details not provided) may combine well with amitriptyline (the standard treatment for some forms of chronic headache, see below, ‘Tension-type headache’), reducing both the number and frequency of required treatment sessions and the medication dose necessary for therapeutic effect and enabling more rapid benefits with few if any side-effects. He considered this combination to be effective for most headache patients, including those with CDH refractory to other forms of treatment, and noted that his associates obtained good results with MA provided treatment also involved antidepressant medication.⁸¹²

There are many heterogeneous studies on EA for different and mixed types of headache
( Uncategorised/mixed headache). Curiously, one author does not consider EA as appropriate for headache due to external pathogenic factors,⁸¹³ although it has in fact been used for recurrent headache following fever⁸¹⁴ ( Other headache). Another found MA results with headache to be better than those for arthritis.⁸¹⁵

As mentioned, headache has been considered a standard indication for TENS.⁸¹⁶ Both TENS^817,818 and pTENS^819,820,821 have been used in a number of studies on mixed types of headache, with many reporting both significant improvement in symptoms and usefulness of these methods prophylactically⁸²² ( Head and facial pain). The effects of pTENS on headache have been investigated thermometrically.^823,824

TCET has been used for headaches of various aetiologies,825 and so has the Liss CES device.826,827,828 In one study on an Italian method of cranial TENS (4 Hz, 1 ms, sufficiently intense at an average 4.9 mA to produce strong local tingling; anode at yintang, cathode occipital), stimulation benefited patients whose headaches were of organic origin, concurrently raising serum endorphin levels, but was without effect in those whose headaches were psychogenic or in normal controls829 (these results should be compared with those of acupuncture in migraineurs, see below, ‘(MA) in migraineurs’). Cerebral galvanism (application of DC currents to the head) certainly used to be employed for ‘nervous headaches’830 (and psychotherapy in conjunction with faradic current as a conditional stimulus, see Ch. 7, for frontal psychogenic headaches831). DC has also been applied, between forehead and upper back832 ( Uncategorised/mixed headache).

‘Microwave needling’ has been used to treat headaches,833 as has EHF acupoint stimulation834 ( Uncategorised/mixed headache).

HeNe LA has been used for headache835,836 as well as TrP LA.837 MA, EA and LA were used in different combinations for various forms of post-traumatic headache (migraine, tension-type and cluster) following an earthquake in Italy, with good results in a number of the patients, in some of whom symptoms only began years after the original trauma838 ( Stress and relaxation). Some of the benefits of LILT for acute headache (vascular or as the result of occipital neuralgia) have been attributed to its effects on 5HT839 ( Uncategorised/mixed headache; see too below, ‘(5HT) in migraine’).

)

Headache has sometimes been attributed to changes in the electromagnetic environment. For example, headaches are a common response to the 3–6 Hz electric fields that accompany the notorious Alpine wind in Austria, the Föhn (see Ch. 4). Headaches associated with neck and shoulder tension have also been considered as being due to reduced magnetic fields in the environment, a symptom of Kyochi Nakagawa‘s ‘magnetic field deficiency syndrome’.840 Thus ELF PEMF therapy has been used for headaches.841 It was found to be effective, for example, for chronic headache and migraine in one double-blind randomised study.842 In another controlled trial, cerebral magnetic stimulation (0.72 G for 1 hour, applied in the caudal–rostral plane normal to the geomagnetic field, and with the patient supine) was also found to improve chronic intermittent headache, particularly in patients who were able to relax during treatment. Improvement was associated with increased EEG α activity, and reduced δ and θ.843

Acupoint stimulation using static magnetic fields has also been used for headache^844,845 ( Uncategorised/mixed headache).

Given the natural DC polarity of the body, with the brain positive relative to the ends of the limbs (see Ch. 4), it would be interesting to know whether working in some electrical environments tends to enhance headaches, while reversing the electrical field direction reduces them (DC treatments to the head are mentioned above). In line with this, Träbert’s currents (rectangular monophasic 143 Hz pulses) are still recommended for headache, with the positive electrode over the spine at the back of the neck and the negative positioned caudally.⁸⁴⁶ At a more subtle level, May Loo advocates using ion-pumping cords for headaches to potentiate movement of qi away from a locally inserted needle to a distal one.⁸⁴⁷

Electrical measurements on the forehead (and occiput) during successful treatment for headache in controlled studies using both acupuncture⁸⁴⁸ and PEMF⁸⁴⁹ have shown that low-resistance readings found before treatment tend to normalise during or after it only when there is considerable improvement in symptoms or their disappearance, but not if improvement is only slight, or early on during a treatment session.

Bloodletting from the ear apex has been used for vascular and ‘functional’ headache,⁸⁵⁰ and Novocain acupoint injection for ‘neurotic headache’.⁸⁵¹ Simple saline injection (not necessarily at acupoints) was found helpful for migraine and vasomotor headache in one uncontrolled study.⁸⁵²

Massage together with relaxation therapy was found to improve pain ratings more than MA in one RCT of chronic headache patients.⁸⁵³ In an interesting observational study (N = 181), massage, tapping and both active and passive movement of the most painful area, together with qigong breathing, were considered to enhance the effects of scalp MA alone for headache.⁸⁵⁴ Self-administered acupressure has also been used,⁸⁵⁵ as well as whole body exposure to musical vibration.⁸⁵⁶

A completely different approach is the inhalation of a particular TCHM preparation. This has been claimed to help ‘neurofunctional’, ‘blood vascular’, neurovascular and tension-type headaches, with both short-term and long-term benefits.⁸⁵⁷

Comparisons and combinations (general)

The combination of EA with cupping was found to give better results than EA alone in one study on mixed headache types ⁸⁵⁸ ( Uncategorised/mixed headache).

LA and MA for headache gave similar results in one report.⁸⁵⁹ However, in a small but more recent RCT, LA was not found to be superior to placebo for a mixed group of sufferers from migraine and and tension-type headache⁸⁶⁰ ( Uncategorised/mixed headache).

In an unusual study comparing DC and more conventional TENS, 1 mA DC (between the forehead and interscapular area) gave more prolonged benefits for chronic headache, although with adverse skin effects in a considerable proportion of patients⁸⁶¹ ( Uncategorised/mixed headache).

Points used for headache – general considerations

Peter Deadman and colleagues have prepared a useful synopsis of the main acupoints used in TCM treatment of headaches (17 local points, 21 adjacent and distal ones).⁸⁶² Many other body points have also been used when treating headache with MA,^863,864,865 as have microsystems such as Korean hand acupuncture.⁸⁶⁶ In the EA and other clinical studies for mixed headache types located here, the most commonly used points are GB-20, LIV-3 and LI-4, with others such as BL-10, GB-8, ST-36 and taiyang (M-HN-9) trailing behind and a scattering of other points used even less frequently. A number of authors emphasise TCM differentiation; others stress that local treatment is best carried out at ashi points.

Indeed, when considering what points to use for headache, it should not be forgotten that in studies on experimental dental or other head pain local points have sometimes proved more effective than distal ones (such as LI-4, see SubCh. 6.1). One Chinese reviewer of a large number of headache studies noted that this seemed to be particularly the case for migraine, but less important when treating headache of neurogenic origin.⁸⁶⁷ In keeping with this approach, for instance, BL-2 and Du-23 were recommended for use with the Likon device,⁸⁶⁸ and points all around the crown of the head are stimulated in one LILT protocol.⁸⁶⁹ However, strong stimulation to head points can result in headaches,⁸⁷⁰ and in one uncontrolled CTENS study the use of tender local points alone (not acupoints) was of little benefit for cluster headache, migraine or post-traumatic headaches of vascular origin.⁸⁷¹

Most formula approaches to treating headache (as if it were a single condition!) make use of LI-4, either because it is traditionally indicated for headache,⁸⁷² because of the dense innervation at the point,⁸⁷³ or because it lies within the C5⁸⁷⁴ or C6⁸⁷⁵ dermatome. Thus one EA device manual recommends GB-20 and taiyang (M-HN-9) with LI-4,⁸⁷⁶ whereas Dean Richards recommends LI-4 with yintang (M-HN-3) for pTENS.⁸⁷⁷ For CES using the Liss device, application at LI-4 as well as over the area of pain is suggested.⁸⁷⁸ Other recommendations are more sophisticated. George Ulett gives a combination of LI-4 with GB-20, Du-15, Du-20 and taiyang, for example.⁸⁷⁹

In one Chinese EA device manual,⁸⁸⁰ points are grouped on the basis of headache location, although some of the points also have a TCM rationale:

In another such manual,⁸⁸¹ points are given on the basis of TCM syndrome differentiation:

LI-4 has frequently been recommended in magnet treatments for headaches.⁸⁸²

With his modified version of Yoshio Manaka’s diode cord treatment, Willem Khoe used BL-59 for headaches.⁸⁸³

Joseph Wong has a very clear and useful exposition of points to use for different sorts of headache, in particular his suggestion that distal yangming points are used for anterior head pain, shaoyang for lateral pain and taiyang for posterior pain.⁸⁸⁴

Parameters for headache – general considerations

In one Chinese review of a number of headache studies, it was observed that continuing to manipulate the needle once deqi is obtained gives better results than simple needle retention.⁸⁸⁵ This could be used to justify the stronger stimulation possible with EA.

In general, LF or DD EA was used more than HF EA in the clinical studies located ( Uncategorised/mixed headache).

However, whereas DD EA has been suggested as appropriate for all forms of headache by the manufacturers of two different EA stimulators,^886,887 one author recommended HF EA (15 minutes daily, on alternate days) for headaches.⁸⁸⁸

For those headaches that are triggered by stress or tension, treatment that fosters relaxation may be helpful; 0.5 Hz stimulation has been suggested as useful for this purpose when using photic/auditory stimulation,⁸⁸⁹ and could be considered for peripheral electrostimulation as well.

Reinhold Voll⁸⁹⁰ made the following suggestions, among others, but without explanation:

• viscerally produced headaches – 8.5 Hz for biliary headache

• cerebral headache – 1.2 Hz and 6.3 Hz for headaches associated with hypersensitivity or after concussion; 4.9 Hz for meningeal headache.

Tension-type headache

As many as 90% of adults have had a tension headache at some time,⁸⁹¹ while around 3% of the population suffer chronic tension-type headaches.⁸⁹² Like most varieties of headache, these can be aetiologically multifactorial, and so may require a correspondingly catholic approach to treatment. They are usually muscular in origin (with ensuing ischaemia contributing to the pain), often with a stress-related component, but may also result from degenerative arthritis of the neck or be drug-induced (headache patients not infrequently become dependent on medication⁸⁹³), or both. There may be considerable overlap between tension-type headaches and myofascial pain (see SubCh. 9.10) or TMJ syndrome (see below, ‘Temporomandibular joint dysfunction’). Although antidepressant drugs are frequently prescribed initially, it appears that none of the currently available treatments for tension-type headache, such as drugs, EMG or other biofeedback methods^{894,895,896,897,898} or psychotherapy, demonstrates clear superiority over the others.⁸⁹⁹ Additional methods such as acupuncture tailored to individual symptomatology may well be important,^900,901 although some reviewers consider neither acupuncture nor TENS of proven benefit.⁹⁰²

Acupuncture has indeed been used for tension-type headache,903,904,905,906 with some trials indicating MA as only equivalent to placebo in effect and others suggesting that MA may be more useful, or as effective as standard treatment ( Tension-type headache). Both MA and physiotherapy were found to be effective for myogenic headache in one Finnish study, for example907 ( Tension-type headache).

Given that electrotherapy appears to be helpful for tension-type and cervicogenic headache,908 there are surprisingly few studies on EA for tension-type headache. In one mixed retrospective study it gave good results,909 but in a series of controlled studies by Jane Carlsson and colleagues standardised symptomatic EA appeared to be less helpful than causally and individually adapted physiotherapy, for example ( Tension-type headache).

In one small crossover pTENS trial, a single pTENS treatment resulted in greater increases in pressure pain threshold at stimulated TrPs than sham stimulation. It is unclear how clinically significant this would be910 ( Tension-type headache).

TENS (both with and without physical therapy) was superior to biofeedback-enhanced neuromuscular reeducation in one CT of tension-type headache.911 TEAS/TENS was found helpful in several uncontrolled studies as well.912 For instance, in one interesting report of over 5000 patients suffering from headaches of which 39.5% were cervicogenic in origin, manipulation of the cervical spine benefited 91% of those whose symptoms appeared to be due to vertebral displacement (67% of the cervicogenic group), whereas electrical nerve block using a monophasic variant of TENS benefited 80–90% of those whose symptoms were apparently caused by narrowing of the intervertebral foramina (33% of the cervicogenic group).913 TENS has also been found helpful for tension-type headache in children,914 alleviating symptoms in around 80% of sufferers.915

In one controlled study of CES for tension-type headache, active treatment was found more helpful than sham. However, this study suffered from quite a high dropout rate916 ( Tension-type headache).

PEMF (12 Hz, 50 G) has been used prophylactically for tension-type (cervical) and migraine headache.917 Magnets (sometimes combined with precious metal inserts) have also been applied at acupoints for .918

Anti-inflammatory medication injected at GB-20 has been used for tension-type headache in Israel⁹¹⁹ (compare the use of botulinum A toxin injections for migraine, described below, ‘Acupoint BTX-A injection’). In another point injection study, this time from Japan, patients with chronic intractable tension-type headache or migraine without aura were found to have significantly low levels of plasma β-endorphin prior to treatment, but this improved significantly, along with VAS levels of pain, after acupoint injection with mecobalamin (the active form of cyanobalamin, vitamin B12). Changes in cortisol and ACTH were not significant.⁹²⁰ Also, in a Chinese report, steroid (dexamethasone) injection at SI-17 was beneficial for vascular rather than tension-type headache. There appeared to be greater ‘vagus tension’ (parasympathetic overactivity) in the former, accompanied by higher levels of plasma pancreatic polypeptide. Treatment tended to normalise these in vascular headache sufferers, but left them unchanged in those with .⁹²¹

Massage, including acupressure⁹²² or strong TrP techniques, has been found helpful for neck pain and reduced movement in patients with .⁹²³ Additional methods such as postural education, isotonic home exercises and stretching to the cervical spine muscles may all contribute to improvement.⁹²⁴ Manipulation of the cervical spine is another possible approach, as mentioned above.

Comparisons and combinations (general)

In one small study on chronic migraine, tension and post-traumatic headache, EA was found to be superior to MA for all three types⁹²⁵ ( Uncategorised/mixed headache).

Points used

GB-20, GB-21 and LI-4 and TrPs are the most frequently used points in EA protocols for tension-type headache. Points such as SI-17 are less likely to be helpful.⁹²⁶

Joseph Wong clearly differentiates the points that will be useful for this type of headache, which is primarily muscular in origin, from those he suggests for migraine or cluster headache: SI-15, GB-14, GB-20, GB-21, LI-18, Du-16, Du-18 and taiyang (M-HN-9).⁹²⁷

Parameters used

For ‘vertebrally induced headache’, Voll recommended stimulation at 9.6 Hz, but without justification.⁹²⁸ LF EA, pTENS and TENS appear to have been used in the various clinical studies more than HF stimulation, but data are sparse in this respect.

Cluster headache

This is the most severe and debilitating of the primary headaches, and is sometimes classified as a form of vascular headache.⁹²⁹ It can become chronic and unresponsive to medication. Sufferers may have low CSF levels of met-enkephalin (ME)⁹³⁰ or substance P (SP),⁹³¹ and during attacks there also appears to be a deficient release of SP in the iris, salivary glands and nasal mucosa of the affected side.⁹³² (One result of this, for example, is that the ipsilateral pupil may become constricted, and so unresponsive to the usual constrictive effect of local TENS, as mentioned in SubCh. 9.4 ‘cluster headaches’). In some patients cluster headache can coexist with or may even develop from migraine⁹³³ (both may involve dysfunction in nitric oxide metabolism⁹³⁴).

It has been suggested that impaired pain transmission along the sensory pathways of the trigeminal nerve, associated with SP dysfunction, can lead to a rostral spread of neuronal (‘quasi epileptic’) irritability along them, and hence to the pain of the headache itself.⁹³⁵ An associated effect is that the inferior posterior hypothalamic grey matter ipsilateral to the pain is activated. HF stimulation of this region of the brain via implanted electrodes (180 Hz, 60 µs, 3 V) has been found in a small pilot study to relieve severe, intractable, chronic cluster headache⁹³⁶ (with possible implications for extracranial stimulation methods?). Perhaps more relevant to acupuncture practice might be the finding that daily nasal administration of capsaicin can result in a rapid decrease in the number and intensity of attacks, and even disappearance of symptoms.⁹³⁷ Would some forms of intranasal electrostimulation or LILT have similar effects?

MA has been attempted for cluster headaches, and although in one study it did significantly increase CSF ME it had little effect on CSF β-endorphin levels, which were within normal range in any case, or on the incidence of headaches.938 This result is in stark contrast to that in the Japanese point injection study on cluster headache mentioned above. I have not located EA studies on cluster headache.

Very-low-intensity PEMF has been used for cluster headache, the effect possibly mediated via the pineal gland.939

Points used

According to Joseph Wong, cluster headaches indicate a sympathetic dysfunction and are often associated with hypertension. He therefore recommends quite different points for treatment than for migraine: SI-17, BL-23, LIV-3, LI-4, Du-1, Du-14, Du-20 and yintang (M-HN-3).⁹⁴⁰ Other points that have been used for cluster headache include GB-20, LI-11, ST-4, ST-7 and taiyang (M-HN-9)⁹⁴¹ ( Cluster headache).

Other forms of headache

Acupuncture has been used for postlumbar puncture headache.⁹⁴² Nevertheless, this method was without success in one study (eventually all the patients responded to epidural injection of a simple plasma substitute, dextran 40).⁹⁴³

However, MA (at points such as BL-10, GB-11, LI-4, ST-8, Du-14 and Du-20) has been effective for postepidural headache.944 In another report, local Gall Bladder meridian points were selected according to headache location, together with LIV-3 and Felix Mann’s ‘cervical articular pillar’ points.945

Acupuncture has been used for headache following head trauma⁹⁴⁶ or pituitary surgery.⁹⁴⁷ However, May Loo has cautioned that acupuncture should be used for head trauma headache only if tests show no skull fracture or intracranial bleeding.⁹⁴⁸ As mentioned above, in one small CT, EA gave better results than MA for headache following trauma.⁹⁴⁹ There are EA and LA studies on this form of headache ( Trauma induced head pain).

EA was more effective than MA for persistent ‘neurotic’ headache in one small Chinese study.⁹⁵⁰ Acupoint injection (with Novocain) has been used for psychogenic headache.⁹⁵¹

Head and facial pains of vascular origin

Pain due to circulatory dysfunction can result from a restriction of blood flow, usually due to vascular constriction (ischaemia), or from excess blood flow to an area (hyperaemia). Both can be responsible for cerebrovascular disorders, and both can lead to headache.

Vascular headache – both migrainous and non-migrainous – has been treated with MA,^952,953 EA,⁹⁵⁴ pTENS,⁹⁵⁵ LA, magnetic stimulation and other methods ( Other vascular headache).

Cephalic hypertension is often associated with muscle tension around the head and headache, even if blood pressure readings are normal in the arms.⁹⁵⁶ Yoshiaki Omura has found that light to moderate headache is experienced with systolic pressures over about 160 mmHg, and severe headache if it exceeds 220 mmHg. Reduction of muscle tension with EA, TEAS or LA may reduce the raised blood pressure and accompanying headache.⁹⁵⁷ Ethyl chloride block of the carotid sinus region has also been used for this type of headache (associated with occipital muscle tension) and at the same time significantly reduced both regional and overall blood pressure⁹⁵⁸ (the same region has been stimulated using EHF as a treatment for hypertension⁹⁵⁹). This could be very important for many workers who use computers for long periods and develop considerable muscle tension problems as a result. However, there can also be significant neurochemical differences between vascular and tension-type headaches.⁹⁶⁰ Perhaps for this reason biofeedback methods are rather more successful for the former.⁹⁶¹

The severe headache that accompanies spontaneous subarachnoid haemorrhage has been treated with auricular acupuncture in addition to standard medication.⁹⁶²

Headaches following cerebrovascular lesions have been treated with magnetotherapy, LILT and their combination, in addition to intravenous LILT.⁹⁶³

Migraine

Migraine, as Oliver Sacks puts it, is not just a headache. Rather, it is a systemic condition, an expression of a major portion of the entire personality, a ‘strategy’ that may be employed to any emotional or biological end, a ‘visceral retreat’ inwards.⁹⁶⁴ However, there is disagreement on whether a migraine personality type (‘emotionally rigid’, or ‘tense, meticulous and obsessional in nature’) can be found.^965,966 Curiously, migraine sufferers may be at increased risk of eczema, asthma and other respiratory disorders⁹⁶⁷ (possibly because all these may be allergy related or, at least for migraine and asthma, because they share a functional abnormality of smooth muscle in blood vessels and airways). There is also some overlap between migraine and inflammatory bowel disease (IBD), constipation and diarrhoea⁹⁶⁸ (again perhaps because of smooth muscle dysfunction). In many cases, migraine runs in families.

At least some migraineurs may be more sensitive to painful stimuli (and itching) than the rest of us, but less aware of gentle tactile stimuli (at least, between attacks).⁹⁶⁹ However, there is even a headache-free type of migraine, in which other migraine symptoms are experienced without the headache,⁹⁷⁰ and some authors have proposed that in some instances having migraines may actually reduce susceptibility to tumours.⁹⁷¹

Migraine is usually classified as ‘classical’ (paroxysmal in quality with a preceding aura) or ‘common’ (without the aura, more reactive in nature and generally longer lasting). The latter is the more frequent type (occurring in roughly 10% of the general population as opposed to 2% for classical migraine, according to Sacks’s estimate). Sometimes a third type, ‘complicated’, is added, which is associated with neurological symptoms that persist after the migraine attack.⁹⁷²

Although Sacks has emphasised the polymorphism of migraine, and that there is no single ‘migraine process’, he himself considers it as a disorder of arousal sandwiched between prodromal and rebound stages, most symptoms representing an increase in parasympathetic tone, a diminution of sympathetic tone, or both. Conventionally it is generally understood as a form of vascular headache. Although opinions differ on the exact mechanisms involved, it seems that a constriction of the nerve-rich arteries at the base of the brain occurs initially, owing to neuropeptide release by the complex trigeminal system within the brain (not to be confused with the external pathways involved in trigeminal neuralgia) and possibly associated with a spreading depression of activity in the cortex.⁹⁷³ This constriction may be exacerbated by release of serotonin (5HT) from platelets in the blood, and is responsible for many of the non-headache symptoms of migraine. This constrictive phase is followed by a rebound vasodilation that in turn triggers release of prostaglandins and substance P, which lead to inflammation and swelling as well as increased sensitivity to pain.⁹⁷⁴ Calcitonin gene-related peptide (CGRP)⁹⁷⁵ and NO⁹⁷⁶ are probably also active in the dural inflammatory cascade that occurs with or causes these vascular changes. Pain occurs during the dilation phase.⁹⁷⁷ What Sacks calls this ‘sympathetic bloating’ is then resolved by ‘crisis’ (vomiting, urgent defecation) or ‘lysis’ (perspiration, urination, watering eyes), more usually with common migraine.

The vascular nature of migraine can be seen in those (the majority) who turn a deathly pale during an attack, shivering and cold (what Sacks calls ‘white’ migraine, or yang deficiency (xu) in TCM language), whereas others become flushed and hot (‘red’ migraine, perhaps yin xu or Liver yang rising). The pallor of ‘white’ migraine may be due to release of vasopressin, associated with nausea.^978,979 In those who experience increased urinary frequency during migraine, in contrast, vasopressin secretion may be reduced.⁹⁸⁰

In some migraine sufferers, headache may mutate into neurogenic angina or laryngospasm.⁹⁸¹ Others – but not all – may show similar symptoms and EEG abnormalities to those found in epilepsy (possibly because of ischaemic changes caused by repeated migraine auras in the territory of the basilar artery).⁹⁸² The EEG of migraineurs has been explored in a number of studies.^{983,984,985,986} A more consistent finding appears to be increased EEG driving in response to visual flicker at > 18–20 Hz.⁹⁸⁷ (Interestingly, the frequency of both the scintillating visual scotoma and the paraesthesiae that can precede or accompany migraine is around 8–12 Hz, in the EEG α range⁹⁸⁸.) Photic stimulation has itself been used to treat migraines,⁹⁸⁹ as well as other forms of headache.⁹⁹⁰ The migraine sufferers found relief more rapid with higher frequencies in the 0.5–50 Hz range and surprisingly preferred bright to low-level stimulation. Perhaps because it borders on so many other conditions, both in WM and in TCM terms,⁹⁹¹ migraine often goes unrecognised and is underdiagnosed.⁹⁹²

Women are more liable to suffer from migraine than men. Part of the explanation for the different incidence of migraines in men and women could well be that in men the white matter of the brain is more densely packed (supposedly giving them a better sense of direction), whereas in women there is already greater blood flow to the head (as a byproduct of which they are more able to multitask).⁹⁹³ Another reason for the greater incidence of migraine in women may lie in greater sensitivity to vasopressin (as a platelet aggregator).⁹⁹⁴ In many women migraine is associated with hormonal fluctuations, especially around the time of menstruation, possibly due to the effects of oestrogen on magnesium levels⁹⁹⁵ and vasodilation.⁹⁹⁶ After menopause, migraines often become less frequent, although migraines can actually begin or worsen perimenopausally.⁹⁹⁷ Older women also suffer less narrowing of the arteries in the brain than men, which may have other advantages.⁹⁹⁸ Menstrually related migraine is often associated with a disturbance of the basilar artery (at the base of the brain, anterior to the medulla oblongata and beneath the pons Varolii), in which case symptoms may include vertigo and poor muscular coordination.⁹⁹⁹

Migraine is usually episodic, but in some patients (as many as 2% of the US population) it becomes chronic, and almost daily. Such chronic migraine is associated with anxiety and depression, drug dependency¹⁰⁰⁰ and a generally reduced quality of life.¹⁰⁰¹ Within the brain, areas such as the red nucleus, substantia nigra and periaqueductal grey may become involved when episodic migraine is transformed into its chronic variant.¹⁰⁰²

Migraine can be an awful and terrifying experience, and some symptoms may overlap those of a transient ischaemic attack, even to the extent of a temporary hemiplegia with vertigo. (This occurs more in classical migraine, but is sometimes categorised as a separate type altogether, ‘hemiplegic migraine’.¹⁰⁰³) However, except when it becomes chronic its effects are temporary, and it does not necessarily predispose to stroke or vascular dementia even then.¹⁰⁰⁴ Conversely, some headaches can of course be symptomatic of hypertension or presage a cerebrovascular accident,¹⁰⁰⁵ headaches may also follow a stroke, and small-artery disease may result in eventual vascular dementia following years of classical migraine.¹⁰⁰⁶ Death from severe arterial spasm, itself rare, has been known to occur in a young migraine sufferer.

Although disturbances, even hallucinations, may occur in any sensory modality with migraine, in most people it is accompanied by cutaneous allodynia (possibly due to sensitisation of central trigeminovascular neurons). Once allodynia is established during an attack, neither the migraine nor the allodynia itself is likely to respond to tryptan medication such as sumatryptan (Imigran).¹⁰⁰⁷ Thus early treatment while the migraine is still mild would seem advisable.¹⁰⁰⁸ Comparable studies on acupuncture for migraine would be useful.

A variety of non-pharmacological methods, including acupuncture and TENS, have been used in migraine treatment.^{1009,1010,1011,1012} In one German hospital for TCM, for example, migraine was one of three conditions most frequently treated with acupuncture and TCHM.¹⁰¹³

The regulatory effects of acupuncture (MA) have often been used to advantage in disorders of cerebral circulation, either to enhance reduced circulation (see SubCh. 9.2) or to lessen hyperaemia, as with migraine.1014,1015,1016,1017,1018,1019 There is at least one review of the acupuncture studies on migraine,1020 and positive results have been reported in a number of uncontrolled MA studies.1021,1022,1023,1024,1025 Bob Flaws, in his very readable book on migraine from a TCM perspective, considers that acupuncture may help more than 50% of migraine sufferers.1026 In one Scandinavian study, long-term results after one course of treatment (mixed MA and EA) were indeed better for migraine than for tension headache, for example.1027 Nevertheless, it has not always been considered useful for migraine,1028 and evidence for acupuncture’s efficacy in this condition is still unclear in the minds of some reviewers.1029 In one migraine study, for example, MA was found useful prophylactically for the associated sickness and reduced migraine frequency, but not intensity.1030 MA has also been used for migraine-related vertigo as part of a comprehensive treatment programme1031 (possibly with better results in those with fewer unresolved emotional problems1032). MA (TrP needling) for migraine may also affect bowel activity1033 (not surprisingly, in view of the frequent overlap between migraine and bowel dysfunction). Others have found MA helpful for migraine pain,1034,1035,1036 emphasising its prophylactic role during pain-free interludes.1037

George Lewith has made the interesting observation that acupuncture can make patients more responsive to medication: ‘someone with migraine may find that while powerful analgesics did not help at all, after acupuncture, one aspirin does the trick’.¹⁰³⁸ Hu Jinsheng has commented that the causes of migraine in TCM terms may well differ in different countries, with treatment adjusted accordingly.¹⁰³⁹ As in Glenn Johnson’s study on migraine-related dizziness, Hu observed that headaches from psychological pressures seemed less responsive to treatment;¹⁰⁴⁰ others have concluded that those more extraverted patients with low scores for neuroticism will tend to report better results with MA for migraine.¹⁰⁴¹ (Although no psychological factors were found to predict response to either MA or massage/relaxation therapy in chronic headache patients in general, the latter do tend to worry about the causes of their pain, even if headache scores are reduced.¹⁰⁴²) A useful practical review of MA for migraine has been given by Agnes Chen.¹⁰⁴³ Other traditional methods include bleeding¹⁰⁴⁴ and ‘fire needling’.¹⁰⁴⁵

As with acupuncture in general, some authors consider that EA may be particularly helpful for migraine.1046 EA, like MA, has frequently been used for migraine ( Migraine). However, in one American study, although good results were reported initially, a second course of treatment was less effective.1047

In an open study of 166 migraineurs treated with MA, EA or pTENS, a positive effect was claimed in 95.9% of the patients, the results correlating with changes in electrophysiological and biochemical measures. MA had a less dramatic action than the electrostimulation interventions.¹⁰⁴⁸ Others too have found pTENS useful.^1049,1050

Alf Heydenreich was the great protagonist of pTENS both for headache and migraine, and his studies, a mixture of well-performed clinical trials and general reviews, are some of the most convincing of its benefits for the latter condition, prophylactically1051 as well as therapeutically1052,1053,1054,1055 ( Migraine).

Auricular EA for migraine has been described.¹⁰⁵⁶ Auricular MA and pTENS were used with apparent success in a case of complex migraine headache of sudden onset with aphasia in a 12-year-old girl, according to a complicated protocol involving the ‘auricular medicine’ methods developed by Paul Nogier and administration of homeopathic virus nosodes, but with insufficient detail in the published report to permit replication of the methods used. The migraines disappeared after 8 weeks of treatment, and did not recur during a 6-month follow-up period.¹⁰⁵⁷

George Ulett has used what he terms ‘neuroelectric acupuncture’ (TEAS) together with imagery conditioning in the treatment of a woman with schizoaffective disorder and migraine who was on a veritable cocktail of eight different medications. After only four sessions, once each month, she was able to reduce to two drugs (risperidone and lithium), with the migraines mostly controlled by her imagery work.1058 It is unclear from his report whether the migraines resulted from or preceded her use of so many different drugs. TEAS for migraine has been investigated by other authors too1059,1060 ( Migraine).

TENS does not appear to have been used that frequently for migraine, although there are some positive reports on this application.1061,1062,1063 It gave quite good results in one uncontrolled study, but was not found helpful if patients were also severely depressed,1064 as has been found for other conditions. It has been utilised for children with migraine.1065 The Skenar device, a variant of TENS, has been used for migraine.1066

CES (SPES) has been used for migraine,1067,1068 with various results depending on the type of migraine being treated.1069 Good results have been claimed in unpublished double-blind and uncontrolled studies for the NET-I device.1070 In one case report, CES was combined with MA.1071

Migraine has been considered an indication for LILT1072 or LA,1073,1074 at auricular points, for instance,1075 although not all agree that LA is effective for the condition.1076

Slowly rotating magnetic fields may trigger migraines.1077 However, PEMF has been used as a treatment for migraine1078 ( Migraine), as for tension-type headaches (see above, ‘12 Hz, 50 G’).

Roger Coghill, among others, has suggested using one or more magnets applied to the head for migraine (950 G, 10–15 minutes, 3–4 sessions).¹⁰⁷⁹ Magnetic press-pellets on auricular acupoints have also been proposed as a useful treatment between attacks.¹⁰⁸⁰

Magnets on distal (leg) acupoints have been used in the treatment of migraine.^1081,1082 Magnets have also been used on hand points for migraine in Koryo hand therapy,¹⁰⁸³ in conjunction with needling and ‘press-pellets’. Another approach to migraine treatment is to use magnets positioned over acupoints in combination with drinking ‘magnetised’ water.¹⁰⁸⁴

GHz (EHF) acupoint stimulation, combined with biofeedback and relaxation, was helpful for migraine in one controlled study.¹⁰⁸⁵

It is well known that compressing the common carotid artery eases migraine in around two-thirds of sufferers.¹⁰⁸⁶ (It is used experimentally in animals as a method of inducing cerebral ischaemia, and is also an old-fashioned remedy for travel sickness – as a child, I used to go on long car journeys with a crepe bandage wrapped tightly round my neck.) The positive correlation between radial and vertebral arterial flow and the negative correlation between these and carotid blood flow may explain the effects of Koryo hand therapy on migraine: treatment that enhances blood flow in the hands reduces that in the common carotid artery, which carries blood to the front of the brain (the vertebral artery takes blood to the rear of the brain).¹⁰⁸⁷ Migraine did not respond well to these methods, however, in some patients with additional problems such as diabetes or severe arthritis.

Following the serendipitous discovery in 1992 that cosmetic use of ‘botox’ (botulinum A toxin, BTX-A) to reduce wrinkles could also help migraines,¹⁰⁸⁸ intramuscular BTX-A injection has been used specifically for migraine. In one study the frontalis, temporalis, procerus and corrugator muscles were injected, as well as the cervical paraspinal muscles (if tender between attacks). Best results occurred in younger episodic migraine patients (< 40 years) who had not overused analgesics (muscle tenderness was unrelated to outcome).¹⁰⁸⁹ Acupoint BTX-A injection has been used for migraine too (25 i.u. in total, at GB-20), even for ‘status migrainosus’ or severe persistent migraine for which hospitalisation is often necessary, with immediate dramatic effects and no further attacks for 2 months.¹⁰⁹⁰ It would be useful to know how much the substance injected contributed to these results. Acupoint injection of mecobalamin, the neurologically active form of vitamin B12, which is sometimes used for Alzheimer-type dementia¹⁰⁹¹ or nerve regeneration (see above, ‘Methylcobalamin (mecobalamin)’), has also been found helpful for chronic intractable migraine, for example.¹⁰⁹² (Compare the injection of anti-inflammatory medication at GB-20 for tension-type headache mentioned above, ‘injected at GB-20’.) The combination of EA with acupoint injection of mecobalamine could well be worth exploring. Neural therapy (impletol injection) is another approach, certainly less costly than botox injections. ¹⁰⁹³

Many of the neurotransmitters discussed elsewhere in this publication are involved in the migraine process: the enkephalins, noradrenaline (NA, norepinephrine), dopamine, histamine, serotonin (5HT), acetylcholine (ACh) and GABA.¹⁰⁹⁴ Dopamine, for instance, may play a role in prodromal symptoms of nausea or drowsiness, as well as during the migraine attacks themselves.¹⁰⁹⁵ Serum levels of βEP,¹⁰⁹⁶ and lymphocyte levels of βEP in particular, may be low, possibly as a result of serotoninergic hypofunction,¹⁰⁹⁷ and may not increase in response to acupuncture (MA) in migraineurs, in contrast to non-sufferers. There have been similar findings in those with CDH.¹⁰⁹⁸ However, cranial TENS did raise βEP levels in patients with headache from organic causes, as described above (‘an Italian method’). Clearly, these differences require some clarification: are they due to dissimilarities in the techniques used, or the type of headache patients treated, for instance?

The role of serotonin (5HT) in migraine has been briefly reviewed.¹⁰⁹⁹ It is complex and not fully understood, in part because 5HT has both vasoconstrictor and vasodilator properties, depending which receptors are involved (see SubCh. 5.1), in part because of the hydra-headed nature of migraine itself. However, it does appear that low levels of 5HT may contribute to migraine in some sufferers (and SSRIs are therefore used to treat the condition), and a serotonergic link between migraine and multiple sclerosis has even been proposed.¹¹⁰⁰ Drugs that bind to 5HT _1c receptors may trigger migraines; those that inhibit 5HT _1c or bind to 5HT _1d receptors may prevent them. The serotonin precursor 5-hydroxytryptophan (5HTP) decreases the sensitivity of 5HT _1c receptors while increasing that of 5HT _1d receptors, and so may prevent migraines if taken for at least 2 months.¹¹⁰¹ As mentioned elsewhere in this publication (see SubCh. 9.10), 5HTP and acupuncture may well have synergistic effects.

Blood magnesium levels tend to be low in migraineurs, but may be increased even to above normal by MA (at SJ-5, GB-20, GB-34, GB-41, LIV-3, ST-36 and taiyang).¹¹⁰² Use of oral magnesium supplementation in conjunction with acupuncture could potentially be a useful approach to treating migraine. Other nutritional supplements may be helpful, such as vitamins B2 and B6.¹¹⁰³ A diet plentiful in omega-3 fatty acids may also help (these have an anti-inflammatory effect, just as they do for eczema and asthma¹¹⁰⁴).

Although caffeine is used as a treatment for headache, in part because of its vascular and ANS effects, it should not be forgotten that overuse of caffeine as well as withdrawal from it can contribute to headache/migraine.¹¹⁰⁵ Caffeine may also reduce the effectiveness of some forms of electrostimulation (see Ch. 12).

Bob Flaws, in his useful book on migraine, describes a number of TCM-based self-help approaches for migraine and its prevention.¹¹⁰⁶

Comparisons and combinations

Traditionally, MA has been combined with cupping in the treatment of migraine.¹¹⁰⁷

Cold extremities may accompany a migraine. Palmar thermal biofeedback has therefore been used to treat classical migraine, and CES has been found to enhance its efficacy.^1108,1109 Whereas qigong has similarly been combined with MA in order to influence circulation in the hands in migraineurs,¹¹¹⁰ I know of no studies in which biofeedback has been combined with EA or other forms of acupuncture for migraine, although this would be a logical combination given the cerebral blood flow effects of Koryo hand therapy, for instance. Paradoxically, cooling of the hands or feet to draw qi downwards and away from the head can help some migraineurs.¹¹¹¹

EA and TEAS may be similarly efficacious for migraine.¹¹¹²

In one RCT comparing MA and LA, results were comparable^1113,1114 (however, this study has been criticised on methodological grounds) ( Migraine).

DC treatment at 1 mA (between forehead and interscapular area) gave more prolonged benefits than more conventional TENS for migraine, although with frequent adverse skin effects, as already mentioned¹¹¹⁵ ( Migraine).

Magnetic treatment was found more effective than MA for some patients in one informal report.¹¹¹⁶

Points used

Traditionally, acupoints for migraine treatment are selected according to TCM syndrome differentiation (ben), along with local or meridian points (biao).¹¹¹⁸ Thus, in one Chinese MA study, selection of acupoints according to TCM syndrome differentiation improved results.¹¹¹⁹ Ipsilateral cervical and thoracic huatuojiaji points, together with GB-20, have also been used for migraine,¹¹²⁰ as well as thoracic huatuojiaji points (T4, T9, T10, T12) together with GB-20.¹¹²¹ However, in this case report it is unclear whether points were treated bilaterally or unilaterally, and with MA or EA. In another Chinese author’s report on treating migraine patients in Germany, huatuojiaji points were found useful in the ~10% of cases for whom an ‘improper’ sleeping position – prone, with the head turned to one side – was considered to be partially responsible for their condition.¹¹²² A different and somewhat heroic method used for migraine consisted of needling ipsilateral taiyang (M-HN-9) through to SI-18, with the needle travelling beneath the zygomatic arch, together with ipsilateral LIV-3 and contralateral LI-4, ipsilateral GB-8 and GB-20 being added in severe cases. Rather extraordinary results were claimed.¹¹²³ Of course, there are many other such variants.^{1124,1125,1126,1127}

Experimentally, MA at GB-20 has been shown to regulate blood flow in both vertebral and basilar arteries, with strong (reducing) stimulation decreasing blood flow more than mild (reinforcing) stimulation increased it.¹¹²⁸ And when neurogenically mediated plasma protein extravasation was induced in the dura mater of rats by electrical stimulation of the right trigeminal ganglion as a model for migraine in humans, EA (and MA) at SI-17 partially blocked this effect (more than LI-11 injection with prednisolone, for example, although less than saline acupoint injection at SI-17 itself, see SubCh. 6.1, ‘dura mater’). Rheoencephalography has been used in the diagnosis of migraine;¹¹²⁹ MA at SJ-17 has been shown to affect the rheoencephalogram of migraine patients.¹¹³⁰

In general, MA at distal points such as SJ-5, GB-34, LI-11 and ST-36, with local points on the head and shoulders and auricular points such as shenmen, Subcortex and Sympathetic, has been used for migraine together with additional auricular points depending on whether the patient is hypo- or hypertensive.¹¹³¹ Auricular EA has also been suggested for ‘every type of vasomotor headache’ together with EA at BL-17.¹¹³²

Because of parasympathetic involvement in migraine, Joseph Wong has suggested use of his system of ‘parasympathetic switches’ at acupoints such as HE-7, BL-1, P-6, SJ-17, SP-6, Ren-22 and Ren-23, claiming that normalising parasympathetic function in this way can rapidly abort migraine symptoms.¹¹³³

MA at scalp and head points has also been utilised for vascular headache.^{1134,1135,1136} Thus it is not surprising that GB-8 with GB-20 is one recommended EA combination for migraine,¹¹³⁷ while EA at SJ-17 was found to improve cerebral blood flow in migraine sufferers¹¹³⁸ and acupoint injection of the synthetic steroid dexamethasone at SI-17 has been used for vascular headache (although it was less effective for tension-type headache).¹¹³⁹ From the EA studies ( Migraine), it appears that the most commonly used points are BL-10, SJ-23, GB-14, LIV-3, ST-8, ST-36, Du-20, yintang (M-HN-3) and taiyang (M-HN-9).

In one auricular acupuncture textbook, suggested points for EA (or LA) include taiyang (not on the ear?), Occiput (towards Forehead), shenmen, Sympathetic, Subcortex and associated points such as Kidney, Gall Bladder, Liver, Stomach, Ear apex and Helix 6 (on the earlobe).¹¹⁴⁰

In the pTENS migraine studies, local points are selected according to the region affected (for instance, ‘vertebral artery’ or ‘carotid sinus’ points¹¹⁴¹), and stimulated in combination with distal/constitutional points such as HE-7, SI-3, SJ-5, LI-4 or ST-36. In one Russian report, once pain was stabilised using local stimulation distal points were selected according to whether the patient’s blood pressure was high (P-6, SP-6), low (GB-34, GB-41, Ren-14?) or indifferent (LI-11?, ST-36).¹¹⁴²

For acute migraine, Manik Hiranandani recommends TEAS at LI-4, Du-20 and taiyang (for unilateral pain, with positive polarity locally, negative at LI-4).¹¹⁴³ Local points such as GB-5, GB-6, GB-14 and taiyang, with distal points like SJ-5, GB-41, LI-4, LI-11, ST-44 or SP-6 have also been recommended.¹¹⁴⁴ Yoshiaki Omura positioned electrodes on the forehead in his TENS study on migraine.¹¹⁴⁵

For LA during an acute attack, Hiranandani suggests stimulating SJ-20, GB-21, LI-4, ST-8, ST-44, Du-20, taiyang and sishencong (M-HN-1) every 5–10 minutes till the attack abates (904 nm, 2 mW, pulsed at 10 Hz, 15–20 seconds/pt except for LI-4 and ST-44, on which 4 mW should be used for 20–30 seconds). Between attacks, courses of 10 daily sessions can be given at the same points (all for 15–20 seconds/pt), along with HE-7, BL-2, LU-9, LI-11, SP-6, auricular Internal secretion and shenmen.¹¹⁴⁶

Parameters used

Rhythmic stimulation to the head itself can affect cranial circulation (see SubCh. 6.5, ‘effects on cranial circulation’). Scalp MA, rotating the needles 100 times per minute (1.7 Hz) has thus been used for migraine.¹¹⁴⁷

There is little consistency in the parameters used in migraine studies ( Migraine). LF, HF and DD have all been employed. Information on stimulation intensity is sparse.

Reinhold Voll recommended 9.4 or 9.5 Hz for migraine with nausea.¹¹⁴⁸ Emad Tukmachi has suggested that LF EA should be used for acute migraine, and HF if it is chronic.¹¹⁴⁹ For ‘vasomotor’ headaches, 10 EA sessions daily or every other day have been recommended, for 15–20 minutes at auricular points or 30 minutes at points such as BL-17.¹¹⁵⁰

Hiranandani recommends TEAS at 12–30 Hz and high intensity during a migraine, at the points detailed above. However, given the increased photic driving to frequencies > 18–20 Hz in migraine sufferers (see above, ‘visual flicker’), at first sight it might seem that strong electrostimulation at such frequencies should be used with caution. Nevertheless, photic (flashing-light) stimulation to alternate eyes has surprisingly been found beneficial for migraines, with the frequency selected for patient comfort at around 30 Hz (the visual ‘critical fusion frequency’ (see Ch. 4),¹¹⁵¹ whereas stimulation at 8–12 Hz is more likely to provoke the scintillating visual scotoma and paraesthesia of migraine.¹¹⁵² Perhaps something can be learned from this for electrostimulatory treatments as well. Omura used 1 Hz TENS on the forehead in his study on migraine.¹¹⁵³

Both HF and LF CES have been used for migraine ( Migraine).

When using PEMF, different frequencies have been recommended by different manufacturers: 7–12 Hz¹¹⁵⁴ or 2 Hz,¹¹⁵⁵ for instance.

For LA, Hiranandani uses infrared (904 nm, 2 or 4 mW, pulsed at 10 Hz, for 15–20 or 20–30 seconds/pt), whether during or between attacks.

Non-migrainous vascular headache

Horton’s headache

In Horton’s syndrome (or ‘histamine headache’), one-sided headaches recur over the region of the external carotid artery, with locally increased temperature, rhinorrhoea and lacrimation.¹¹⁵⁶ It is sometimes described as migrainous neuralgia, can be confused with cluster headache, is more common in men and may leave permanent after-effects such as pupil contraction, ptosis or enophthalmos (a sunken eyeball, as opposed to the exophthalmos of hyperthyroidism).¹¹⁵⁷ Together, these are confusingly termed Horner’s syndrome (Bayard Taylor Horton was an American doctor, and Johann Friedrich Horner a completely unrelated Swiss ophthalmologist).

MA (at Du-14) has been used for histamine headache.¹¹⁵⁸ EA and acupoint injection have also been used for Horton’s syndrome.¹¹⁵⁹ In one study, MA only was used locally, with low-intensity EA reserved for distal ‘anti-inflammatory’ points. Local EA tended to aggravate the problem¹¹⁶⁰ ( Other vascular headache).

Charlin’s syndrome

In this syndrome, which may resemble migraine in its onesidedness and cluster headache in its brevity, there is facial pain, iritis and rhinorrhoea with tenderness along the side of the nose.¹¹⁶¹ It is due to dysfunction of the ciliary ganglion (at the back of the eye socket), is also termed nasociliary neuralgia and is sometimes of vascular (ischaemic?) origin.

It has been treated with EA and LA, surprisingly with better results in more chronic cases ( Other vascular headache).

Sluder’s syndrome

This is a neuralgia of the sphenopalatine ganglion in the sphenomaxillary fossa (deep beneath ST-7) and is experienced as a burning or boring pain focused in the area of the superior maxilla, but also radiating into the neck and shoulder.¹¹⁶² Like Charlin’s syndrome, it can be vascular in origin (with ipsilateral temporal vessels visibly engorged) and may be mistaken for trigeminal neuralgia, migraine or cluster headache (attacks are one sided and relatively brief). Direct stimulation of the sphenopalatine ganglion has in fact been used in treating primary trigeminal neuralgia¹¹⁶³ ( Trigeminal neuralgia).

Sluder’s syndrome has been treated with EA and LA, with better results in less chronic cases ( Other vascular headache). Intraoral TrP LA is one possible intervention.¹¹⁶⁴

Comparisons and combinations for non-migrainous vascular headache

EA gave better results than MA in one study on vascular headache (both migraine and non-migraine).¹¹⁶⁵ Treatment with a herbal paste ‘stuffed’ into the nose, together with acupoint injection, was also superior to MA for vascular headache (‘pale’ migraine)¹¹⁶⁶
( Other vascular headache).

Points used for non-migrainous vascular headache

Points such as P-6, GB-20, LI-4 and ST-36 are often used for non-migraine vascular headaches. However, given their variety, this is probably not a very useful generalisation. Studies on specific headache types tend to be designed quite idiosyncratically, sometimes using extrameridian or unusual auricular points. For the traditionalists, it may be reassuring to know that the author of one RCT on angioneurotic (vasomotor) headache found that points selected according to the method of the ‘eight techniques of the mysterious turtle’ (ling gui ba fa)¹¹⁶⁷ gave better results than points chosen according to their meridian pathways¹¹⁶⁸ ( Other vascular headache).

Parameters used for non-migrainous vascular headache

As with migraine, there is little consistent or useful information on stimulation parameters to use for these other sorts of headache. Voll¹¹⁶⁹ suggested that 4.0 Hz could be used for vasally induced headache and vertigo, but without explanation.

Facial pain

Facial pain can arise from various causes, and may require a correspondingly multidisciplinary approach to treatment.¹¹⁷⁰ When not idiopathic, the most frequent conditions leading to secondary facial pain include myofascial pain syndrome, sinusitis, cervical vertebral lesions, postherpetic neuralgia, malignant head and neck tumours and vascular lesions of the pain pathway within the skull.¹¹⁷¹ As with any pain, it may be chronic or acute, nociceptive or neurogenic in origin, and paroxysmal or persistent in nature. Chronic trigeminal neuralgia is the commonest example of paroxysmal neurogenic pain, while so-called atypical facial pain is an example of persistent neurogenic pain.¹¹⁷² Both forms of facial pain tend to occur more in women. Some of the headache types described above can also involve facial pain (just as some facial pain can lead to headache). Pain can also arise from temporomandibular joint (TMJ) dysfunction. In some complex cases, more than one type of pain can be present.¹¹⁷³

A variety of CAM modalities have been used in treating chronic facial pain, including acupuncture, relaxation and biofeedback (by 2002 there were three RCTs for each of the former and eight for biofeedback).¹¹⁷⁴ Various types of chronic facial pain (predominantly atypical facial pain and trigeminal neuralgia, but also glossopharyngeal pain) have been treated with acupuncture.^1175,1176 However, although some reviewers have concluded that it can be effective for a variety of chronic facial pains,¹¹⁷⁷ a number of authors suggest that it may be more effective when this is nociceptive (myogenic) in origin rather than neurogenic.^1178,1179 Some authors have had disappointing results with MA for facial pain syndromes in general.¹¹⁸⁰

LF EA was used in one complex case of chronic continuous facial pain diagnosed as psychogenic, together with anxiolytic and other medication and also kanpo herbal medicine. The pain gradually subsided only to be replaced by a new shooting pain, diagnosed as trigeminal neuralgia, that then responded to carbamazepine.1181

TENS (motor level, applied bilaterally, over the mandibular notch just anterior to the ears) has been used for facial pain (various neuralgias or neurovascular syndromes),1182 with some effect claimed even in supposedly hopeless cases.1183 It may also act as an aid to its differential diagnosis, applying different combinations of frequency and pulse duration and gauging the response.1184

Some authors advise caution when using TENS on the face, particularly if the patient experiences hyperaesthesia.¹¹⁸⁵ (Face points tend anyway to be more electrically reactive and sensitive to electrical stimulation than elsewhere on the body, see SubCh. 5.2, ‘face points’, and Ch. 4, ‘face and neck’). However, others recommend its use prior to medication or more invasive procedures.¹¹⁸⁶ Unfortunately, this is not always the order in which treatments are given for facial pain.

TrP LA has been used for facial pain.1187 However, there are a number of negative studies of LILT for chronic orofacial pain.1188 Thus, although helpful for trigeminal neuralgia, HeNe LA was of no benefit for neuralgia of the pterygopalatine ganglion, for example.1189

Magnetotherapy has been used for facial pain (sometimes even with subcutaneously implanted magnets).1190 In particular, local stimulation with the Magnetostom device (1.56 and 600 Hz, up to several times daily) has been recommended for facial neuralgia.1191

Comparisons and combinations – facial pain in general

When TENS gave insufficient relief from facial pain, results were markedly improved by combining it with local iontophoresis of NSAIDs.¹¹⁹²

Points used for facial pain – facial pain in general

The face is a well-innervated area, so perhaps it is not surprising that a microsystem of ‘face acupuncture’ has been developed,¹¹⁹³ as with other such systems in the head and neck regions.¹¹⁹⁴ Face acupuncture (MA) has been used experimentally for analgesia during surgery in dogs,¹¹⁹⁵ and for conditions such as stroke (again with MA)¹¹⁹⁶ rather than for facial pain itself. Sun Peilin has given a useful analysis of facial pain, listing points for MA according to TCM syndrome differentiation.¹¹⁹⁷

Parameters used for facial pain – facial pain in general

See below, under trigeminal neuralgia and atypical facial pain.

Trigeminal neuralgia, or tic douloureux

The trigeminal nerve, the largest cranial nerve, is similar to spinal nerves in that it has two roots, one motor and one sensory. The latter, the lower (or caudal/spinal) sensory nucleus, is continuous with the spinal cord (substantia gelatinosa) and joins with fibres from the locus coeruleus, or upper sensory nucleus, then passing via the Gasserian (semilunar) ganglion to the three main branches of the trigeminal nerve (Fig. 9.11.2). The first two of these are purely sensory; the third is a compound nerve and also supplies the muscles of mastication.¹¹⁹⁹

Figure 9.11.2 Superficial nerves of the head and neck, showing areas innervated by the three divisions of the C2–C5. Note carotid sinus and larynx, where electrical stimulation should be avoided. (Adapted from Anderson 2000.)

Trigeminal neuralgia (TN) is a chronic, severe, lancinating and burning pain on one side of the face. It is most commonly primary (idiopathic), although secondary TN (symptomatic of another condition) also occurs.¹²⁰⁰ Early on, spontaneous remission can occur, but this happens less frequently as the condition progresses. Attacks are initiated by sensory irritation at trigger areas within the region innervated by the nerve,¹²⁰¹ although some authors consider that secondary TN may not exhibit such trigger zones¹²⁰² (it is possible that they are confounding TN and a typical facial pain, described below).

The location of TN varies depending which is the main branch (division) involved, the first or ophthalmic (V1 or OB), second or maxillary (V2 or MxB) or third or mandibular (V3 or MnB, including the auriculotemporal nerve) (Fig. 9.11.2). V2 neuralgia is the most common, but there can be neuralgia of more than one division.¹²⁰³