•

TRANSCUTANEOUS ELECTRICAL NERVE STl:MULATl:ON (TENS): EFFECTS OF DURATl:ON OF STIMULATION ON ANTINOCICEPTION IN MAN

STFVEN R. CHIN

School of Physical and Occupational Therapy McGill University, Montrecil September, 1993

A thesis submi tted to the Facul ty of Graduate Studies and Research in partial fulf illment of the requirements for the degree of Master of Science in Rehabili tation Science

•

Cc) steven Chin 1993

•

i ABSTRACT

Transcutaneous

electrical

nerve

employed for over 2 decades in pain

stimulation lnanaqemf~nt,

(TENS)

ha:::;

beEm

however, definitive

analgesic parameters have yet ta be delermined. The objective of this study was to determine the influence of ID, 30 and 60 min of TENS on the time course and lTlagnitudei of modulation on flexion reflex (FR) and on subjeGtive ~atn E~stimat.es (visual analogue scale, VAS), as weIl as determining the relationship between FR responses and VAS scores following TENS of different duraticms. Ten (10)

normal subjects received electrical stimuli of maximal

tolerable intensity to the sole of their foot. 'l'he r€;sultant FR was recorded electromyographically from the ipsiliateral biceps femoris (BF) muscle. Participants rated the Jntensity of this stimulus on a VAS. In general our results indicated that TENS could suppress the FR

•

in 50% to 70% of subjects. However,

longer durations (30 and 60

min) of TENS have a small, but discernible difference in the number signif~cant

maximal

FR suppression when compared to 10 min of TENS. Moreover,

the 60

of subjects that show inhlbition of FR area and

min of TENS generated significÇlnt. depression of VAS scores in more subjects than 10 and 30 min of l'ENS and

ensurf~d

that any increase

in pain perception would not reach significance when compared to the shorter

~10

and 30 min) durations of TENS.

Suppressed VAS ratings arlll BF FR areas yielded moderate to high linear correlations (r=.40 to .90), while potentiated VAS ratings and BF FR areas produced low to moderate linear correlations (r=.05 to .51). Thirty (30) min of TENS

genE~rated

a parallel modulation of

onset of maximal VAS scores and FR area,

whi le 60 min of TENS

resul ted in a "temporal" dissociat ion between these two var iables. Interestingly, 60 min of placebo stimulation produced a parallel but negative linear correlation (r=-.74)

between potentiated VAS

scores and BF FR areas. Our results suggested that longer durations of TENS (up to 60 min) could be applied in acute clinical pain to produce antinociceptive

~

and analgesic effects in a greater number of subjects.

I~---ii RESUME La stimulation ~lectrique transcutan~e des nerfs (TENS) fut utilis~e pendant plus que deux decennies dans la gestion de la douleur cependant, les paramètres d'un analgésique définitif sont toujours a d~terminer. Le but de cette étude ~tait de d~terminer l'influence de la durée (10, 30 et 60 min) du TENS sur la dur~e et l'importance de modulation sur le r~flexe de flexion (FR) et sur l'estimation de la douleur subjective sur une ~chelle visuelle analogue (VAS), comme la détermination de la relation entre les r~action du FR et les point r~sultant du VAS, suite au stimulus de différent dur~es. , , , , -. Dix personnes en bonne sante ont sub~ une exper~ence ou d'un choc èlectrique d'une intensit~ maximale tolerable dans la plante , du fut ' tree ' tante FR enreg~s par de leur pied. La resul electromyographie au niveau du biceps fémoral ipsilatéral. Les participants sont ~valu~ l'intensitt de ce stimulus à l'~chelle visuelle analogue. , , , , , " En general nos resultats ont ~ndl.que que la stl.mulat~on ~lectrique transcutan~e des nerfs pourrait ~liminer le r~flexe de flexion chez 50% a 70% des sujets sournis a l'exp~rimentation. Cependant, une dur~e plus longue (30 a 60 min) de stimulation (TENS) r~sulte en une diff~rence assez minime mais bien distinte dans le nombre des subjets qui ont démontr~ une inhibition au niveau du réflexe de flexion, et une diminution maximale assez significante du FR en comparaison avec 10 min de stimulation (TENS). Enoutre, 60 min de stimulation ~lectrique transcutan~e des nerfs, ont gén6ré une depression importante à l'échelle visuelle analogue chez plus de sujets que dans le cas où TENS est administrée pour 10 ou 30 min. Cette durée de 60 mln

à

aidé la

prevention dans l'ensemble de toute augmentation potentielle de la douleur compan~e aux courtes dur'e de 10 ou de 30 min de stimulation (TENS). La mesure de l'inhibition

•

à l'èchelle visuelle analogue au

niveau du réflexe de flexion a montr~ une corr'lation lin'aire , graduellement importante (r=.40 a .90). Quand a l'evaluation a ,~

1 iii l'''chelle visuelle anall:lgue facilitatrice au niveau du r~flt~xe de flex.Lon ont prodi'lt une corrélation linéaire d'une importance mlnime et moyenne (r=.05 à . 5J") , 30 min de stimulat10n a g~nèré une modulation parallèle d'une attaque maximale de VN2 au niveau du r~flexe de flexion. Alors que 60 min de stimulation (TENS) ont fini par cr'er une dissociation entre ces 2 variables. Ce qui est intéressant, c'est que 60 min de stimulation sur le placebo a produit une corr~lation linéaire parallèle mais négative (r=-.74) entre les points a l"chelle visuelle analogue facilitatoire et au niveau du réflexe de flexion. , Pour conclurla, des durees plus longues de stimulation élec:trique transcu'tan&e des nerfs (jusqu'a 60 min ) peuvent ~tre appliqu~ dans le CilS d'une douleur aigue pour produire des effets analg~si.que et arttinociceptive chez un nombre beaucoup plus important de sujets.

•

•

iv DEDICATION

'ro rny loving

wif(~,

Tae, who l sa dearly cherish and adore;

Ta rny beautiful daughter, Brittany - yeu are the sunshine of rny life; To rny parents and rnether-in-law who have sacrificed sa rnuch and expected nothing in return;

l pledge rny undying love and gratitude •

•

•

•

1/

ACKNOWLEDGEMENTS

l would first like to express my most sincere gratitude to my thesis supervisor, Dr. Christina Hui-Chan. The endless support and demanding attention to

detail throughout this

project has only

served to deepen my respect for her. l have come to realize that excellence cornes in many forms, but rarely do they appear in one pers on

who

can

assume

so

many

roles.

For

your

g~idance,

understanding and patience - please accept my humble thanks. In

addition,

mentioned

as

a

my

colleagues

vital

link

to

and

classmates deserve

surviving

Grëtduate

to

be

School.

In

particular, Steven Mah, Anna Peruzzi and Jasmine Cooper, providcd •

a

small

but

loyal

network

sympathetic ear could

where

friendship,

always be found.

l

support

and

a

have appreciated the

opportunity to grow academically and spiritually with this fine group. l

am grateful to Dr. Georges Monette for

his expertise in

statistical analyses. His gentle and patient manner continues to impress me.

Mr.

S.

Lafontaine

must

also

be

acknowlegded

for

develaping the computer programme employed in this study. Lastly, but ma st importantly l would like ta thank my wife and our families for the constant encouragement and support throughout my education - without them l would never have been able to do this. The

•

author

received

Foundation of Canada .

a

bursary

from

the

Physiotherapy

•

vi Preface

According to section 7 of the Guidelines concerning Thesis PI eparation as

Research,

specif ied by the Faculty of Graduate Studies and

McGi11

University

(May,

1992*),

the student

has the

option of including as part of the thesis, the text of an original paper, or papers, suitable for submission to learned journals for publication. This opt ion has been taken for the present thesis wi th the

authorization

ot

the

Director,

School

of

Physical

and

Occupational Therapy. Chapter One is organized to provide a literature review and the problem formulation pertaining to the present study. •

Two

is a paper to be

Chapter

submitted for publication, describing the

influence of different durations of conventional TENS on the human flexion reflex. Chapter Three is also a paper to be submitteù for publication, which describes the influence of different durations of

conventional

TENS

on the subjective pain

estimates

and the

latter' s relationship with the human flexion reflex. Chapter Four is a summary of the results and provides conclusions with a section

* In accordanc(' with the stated requirement of the Faculty of Graduate Studies and Research, section 7 of Guidelines concerning Thesis Preparation is cited in full below: 7.

•

MANUSCRIPTS AND AUTHORSHIPS

The candidate has the option, subject to the approval of the Department, of including as part of the thesis the text, or duplicated publishad text (see below), of an original paper, or papers. In this case the thesis must still conform to aIl other requirements in Guidelines Concerning Thesis Preparation. Additional material (procedural and design data as weIl as description of equipment) must be provided in sufficient detail

•

vii (e.g. in appendices) to allow clear and precise judgement to be made of the importance and originality of the research re~orted. The thesis should be more than a collection of of manuscripts published or to be published. It must include a general abstracto...... a full introduction and literature reviewand a final over_all conclusion. Connecting texts which provide logical bridges between different manuscripts are usually desirable in the interest of cohesion. I t is acceptable for theses to include as chapters authentic copies of papers already pUblished, provided these are duplicated clearly on regulation thesis stationery and bound as an integral part of the thesis. Photographs or other materials which do not duplicate weIl must be included in their original forro. In such instances « connecting text are mandatory and supplementary explanatory material is almost always neces:;"'\ry.

•

The inclusion of manuscripts co-authored by the candida te and others is acceptable but the candidate is required to make an explici t statement on who contributed to such work and to what extent, and supervisors must attest to the accuracyof such claims 1 e.g. before the Oral Committee. Since the task of the candidate's interest to make the responsibilities of authors perfectly clear . Candidates following this option must inform the Department before i t submits the thesis for review.

LIST OF REFEERENCES

Chapter 2: Hui-Chan, C. w. Y. and Chin, S., Transcutaneous electrical nerve stimulation: effect of stimulation duration on the flexion reflex in man. In preparation for submission to Pain. Chapter 3: Chin, S. and Hui-Chan C.W.Y., Influence of different TENS durations on the relationship between subj ecti ve pain sensation and flexion reflex in man. In preparation for submission ta Pain .

•

viii TABLE OF CONTENTS

ABSTRACT •••••••••••••••••••••••••••

1

..

RESUME, •••••••••••••••••••••••••••• 1.1.

DEDICATION ...••••....•............. i v ACKNOWLEDGEMENTS •....•...••.•.••••• v

.

PREFACE •••••••••••••••••••••••••••• Vl.

TABLE OF conTENTS ..••••...••.•..••• viii

CHAPT ER ONE

GENERAL INTRODuc'rION ..•••.••..•••••••.•••••..• l

LITERATURE REVIEW ••••••••••••••••••••••••••••••••••••••• 2

Historical Development of TENS for Analgesia •••.••••.. 2 Applications and Efficacy of TENS ••..••••••••••••.•••• 3 TENS: Proposed Mechanisms of Action ..•••.•••••..••••.. 7 1. The Gate Control Theory of Pain •.••••.•••••••• 7 2. Conduct ion Block •.••••••..•••••..••••••••••••• 9

3. Endogenous Mechanisms of Pain Control •••.••••• 10 TENS Parameters .••••••••.•••••••••••••••••••••••.••.•• 15 1. Electrode Placement and Electrode size •••••••• 15 2. Waveform •••••••••••••••••••••••••••••••••••••• 16

3. Pulse width ••••••••••••••••••••••••••••••••••• 17

4. Frequency and Intensity of Stimulation: Conventional TENS and Acupuncture-like TENS ••• 17 Reviewof Conventional TENS and Acu-TENS •••••••••••••• 21 Flexion Reflex •••••••••••••••••••••••••••••••••••••••• 25

•

problem Formulation: Analgesie Effects of the Duration of stimulation ••••.• 29

CHAPTER TWO

TRANSCUTANEOUS ELECTRICAL NERVE STIMULATION:

EFFECT OF STIMULATION DURATION ON THE FLEXION REFLEX IN MA.N •••••••••••••••••••••••••••••••••••••••••••••• 3 :3

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '" ....... 3 4

Materials and Methods ................................. 39 Subj ects .•.•.•..•.•••.••••....•.•..•••.....••.... 39

Elicitation and Recording of Flexion Reflex •..... 40 Four Stimulating Conditions ...•.•...•.....••..... 42 Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Resul ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

Discuss ion ............................................ 47

FR Suppression by TENS of Different Duration ••••. 48 FR Potentiation by TENS and Placebo TENS .••••.... 49 Placebo Effects ...•••..••..•.•......••......•.... 50 Clinical Applications ..•••...•••••.••••...•••.... 52

CHAPT ER THREE

INFLUENCE OF DIFFERENT TENS DURATIONS ON

THE RELATIONSHIP BETWEEN SUBJECTIVE PAIN SENSATION AND THE FLEXION REFLEX IN MAN ..•••...••...•••....••....• 54 Summary ...•....................•...............•...... 55

Introduction .......................................... 56

Materials and Methods ..•••...•••..•••••....•...••....• 58 Subj ects ......................................... 58

Elicitation and Recording of Flexion Reflex •.•..• 59

•

Subjective Pain Estimates •••••••.•.•••.•..••....• 61 Four Stimulating Condltions .•••.•.•••••••••••...• 62

•

Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

.......................................... .64 Discussion. .......................................... .68 Results .•.•

VAS Modulation by TENS of Different Dur.ation or Placebo stimulation •••••••••••.••••••••••••••• 68 Relationship Between VAS Scores and FR Following TENS of Different Duration or Placebo stimula ticn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

CHAPTER FIVE

•

•

SUMMARY AND CONCLUSIONS ••

.76

significance of the Study •••••••••

................... .81

BIBLIOGRAPHY .•.••....•..•.••••••.•••.•..••••••••.••••••..•• 83

1

CBAPTER ONE

GENERAL XNTRODUCTION

•

2

LXTERATURE REVXEW HXSTORICAL DEVELOPHENT OF TENS FOR ANALGESXA Transcutaneous nerve stimulation (TENS) has been extensively documented in the treatment of acute and chronic pain conditions in humans (Melzack, 1975; Francinni et al.,1981; Hansson and Ekblom, 1983; Fried et al., 1984; Mannheimer and Lampe, 1984; Melzack and Wall, 1984; Gersh and Wolf, 1985; Phero et al., 1987). In fact, modern TENS has its roots dating as far back as 2500 B.C .. Carvings in Egyptian tombs depicted the use of electric eels to control pain

(sjolund and Eriksson,

1985). Records from ancient

Roman scripts describe how electric torpedo fish was used in the treatment of gout and headache (Sjolund and Eriksson, 1985). Animal forms

of

electrical

stimulation continued

through

to

the

mid

1700'5, until man was able to harness electricity. Rudimentary man made devices were constructed which were often offered as a medical panacea.

Unfortunately,

many

of

these machines

were

not

weIl

received by the medical community due to po or quality control and a

reliance on

patient testimonials

for their efficacy

(Lampe,

1978) . The 1800's saw marked increases in the development and use of electrotherapy Sarlandiere

for

pain

reported

electricity (Ersek,

relief

on

1981).

the

around

combination

In America,

machines" were developed and patented.

•

short-lived

analgesia

extractions

and

and

recommended

the

were for

world. of

In

France,

acupuncture

and

a number of "electricity These, however,

primarily chronic

employed back

and

provided

for joint

tooth pain

•

3

(Sjolund and Eriksson, 1985). In the mid 1800's, Oliver began using electrical

stimulation

analgesia

and

even

during

local

surgical

anaesthesia.

procedures However,

to

produce

the advent

of

anaesthetic gases replaced afferent stimulation as the method of choice for surgery (Sjolund and Ericksson, 1985). The

practice

of

using

electrical

stimulation

remained

relatively dormant until the the mid 1960's, when Melzack and Wall proposed a new theory on pain mechanisms, known as the Gate Control Theory. According to this theory, stimulation of large myelinated fibres (A alpha and A beta) close out pain

transmitted

unmyelinated C fibres •

could act as a gating mechanism to via

small

(Melzack

myelinated

and Wall,

1965).

(A delta)

and

This proposaI

spawned an enormous and renewed interest in electrical stimulation for pain management.

In 1967, based on the gate control theory,

Mortimer and Shealy began implanting dorsal column stimulators and peripheral

nerve

1976). However,

stimulators to

treat

pain

syndromes

(Warren,

prior to any invasive or surgical technique, aIl

patients were screened using TENS. These investigators soon found that TENS was very effective in controlling acute and chronic pain. Hence, the modern era of TENS was launched (Warren, 1976).

APPLZCATXONS AND EFFICACY OF TENS

The effects of TENS have been investigated in their various

•

clinical

applications.

subjects

with

These

include

multiple sclerosis

(Levin and Hui-Chan,

1992),

reducing

(Ersek,

1981)

spasticity

in

and hemiplegia

relieving itching and atopic eczema

4

(Bjorna and Kaada, 1987), assisting in the healing of non-united fractures (Kahn, 1982) and improving athletes' physical performance in running, cycling, and swimming (Kaada, 1984). Nevertheless, TENS has been predominantly studied with respect to its effects on pain. In acute pain,

such as that experienced by post-operative

patients, TENS has been found to have great value. Harvie (1979) determined that TENS could greatly reduce the patients' need for pain medication by as much as 75% to 100% in post-operative knee surgery. early

In addition,

ambulation

similarly,

and

recovery of range of motion at the knee, prompt

hospital

discharge

was

enhanced.

in post-operative cholecystectomy patients, TENS was

shown to have significant effects on pulmonary recovery, resulting in a decreased incidence of pulmonary complications (Ali et al., 1981). Acute non post-operative conditions have also been examined. Hansson and Ekblom (1983) reported that 38% of patients with acute oro-facial pain received greater than 50% of relief with TENS, compared with 10% in placebo TENS. Ordog (1987) reported that TENS was as effective as oral analgesics, but without their side effects in the treatment of patients wi th acute traumatic pain such as sprains, fractures and lacerations. In fact, transcutaneous cranial electrostimulation

with

high

frequency

intermittent

current

(Limoge's current) has been found to potentiate morphine analgesia by 3 fold as measured by the tail flick latency in rats (Auriacombe et ~

al.,

1990).

Addi tionally,

these

researchers

were

able

to

significantly reduce the symptoms of opiate withdrawal by 48% in

5

rats (as measured on Gellert' s scale). Marchand et al. demonstrated

that

TENS

could

significantly

increase

the

( 1991) pain

threshold of heat stimuli from 46.7 to 47.7 degrees celsius, as weIl as significantly reducing subjective pain ratings. The effectiveness of TENS on chronic pain conditions has shown greater variability, due in part to potential co-existing psychosocial problems. The general concensus is that TENS can be used as a component of a multifaceted approach to the management of chronic pain. Long (cited by Ersek, 1981) was able to demonstrate that 70% of chronic pain patients (spinal pain and cancer) obtained relief with TENS, exercise, drug withdrawal and psychotrophic medications, as opposed to 40% relief with normal hospital ward routine. Fried and colleagues (1984) reported that TENS was beneficial in 83.8% of cases with chronic post-traumatic pain. This was based on a 6 month follow-up questionnaire. Melzack

(1975),

using

the

McGill

Pain

Questionnaire,

determined that TENS was significantly more effective than placebo TENS in relieving severe clinical pain. The mean decrease of pain during stimulation was 75% for phantom limb pain, 62% for shoulder arm pain and 60% for low back pain. For many of these subjects, the study was the last hope, as other forms of treatment (including psychiatry,

neurosurgery and orthopaedics)

Melzack et al.

were not effective.

(1983) compared TENS with mechanical massage on

acute and chronic low back pain sufferers, and found that the TENS

•

treatment group produced significantly better pain relief. GraffRadford et al. (1989) reported a reduction of chronic myofascial

6

pain of up to 50% following 10 min TENS stimulation. However, local trigger

point

authors

to

sensitivity

suggest

that

remained long

unaffected,

term

management

prompting would

the

require

additional forms of therapy. The efficacy and duration of TENS analgesia, especially over long term usage, is of great importance for those afflicted with acute and chronic pain. Review of literature has shown that TENS efficacy generally decreases over time - from 30%-80% to 18%-60% of patients reporting satisfactory pain relief after 1 or 2 years of follow-up (Eriksson et al., 1979; Nathan and Bates, 1980; Sjolund and Eriksson, 1985). A

few

predicting

studies when

have

TENS

examined

would

be

patient

most

characteristics

effective.

in

Reynolds

and

colleagues (1983) determined that older, retired people with pain onset of less than one year, who had undergone minor or no surgery and had used non-opiate medication, responded better to TENS. Wolf et al. (1981) also reported that patients with a minimal history of surgery and pain medication received better pain relief from TENS. Neurogenic pain patients reacted more favourably to TENS analgesia than

psychogenic

visceral

origin

pain

patients

(Eriksson

et

intensity, on the other hand, (Eriksson et al.,

(Johansson,

al.,

1979).

1980) Age,

or sex

pain anü

of

pain

had little effect on the results

1979). Shealy (cited by Ersek, 1981)

reported

that focal pain responded more positively than diffuse pain, while Johansson (1980) determined that pain in the extremities responded ~

better than axial pain. Melzack and Wall (1984) reported that the

7

conditions MOSt amenable to to TENS treatment included areas in which there were muscle or skin

tendl~rness

associated wi th nerve

dysfunction such as post herpetic neuralgia.

TENS: PROPOSEO MECHANISMS OF ACTION 1. The Gate Control Theory of Pain In the Mid 1960's, Melzack and Wall (1965) proposed the gate control theory of pain based on their own research and consolidated this with perspectives on pain mechanisms at that time. The theory suggested that the substantia gelatinosa (SG) in the dorsal horn of the spinal cord, acted as a gating mechanism that could influence the

activity

of

transmission

(T)

cells

and

ultimately

the

perception of pain. Activation of large diameter myelinated fibres subserving touch, pressure and vibration could partially close this pre-synaptic gate and reduce the output of the T. cells. Conversely, activation of small diameter myelinated and unmyelinated fiures carrying noxious input, diminished this pre-synaptic inhibition. In addition,

a central control trigger, which monitored emotions,

attention and pa st experiences, was present to influence sensory input and the gate control system. Accordingly, the experience of pain was a result of temporal and spatial summation or integration of incoming stimuli. When a critical threshold of T cell firing had been surpassed, an action system would be initiated followed by a sequence of responses including a startle response, flexion reflex, postural adjustment, autonomie responses and avoidance behaviours ~

(Melzack and Wall, 1965).

8

Clinically, it suggested that selective stimulation of large diameter, fast conducting fibres could close the gate in the dorsal horn of the spinal cord and consequently reduce the sensation of pain. Conventional TENS, in which a tingling sensation or numbness is perceived, evolved as a result of this theory. High frequency pulsE'!S (50 Hz to 200 Hz) delivered at low intensity of stimulation (2 to 3 times sensory threshold)

preferentially activate large

diameter fibres. Ideally, stimulation is performed at segmental levels.

The the ory advanced a

mediation of pain control.

major

However,

spinal

component

in the

the gate control was not

without its critics (Schmidt, 1972; Nathan, 1976) Melzack and Wall (1982) had since updated and modified their original theory. They suggested that activity of the large diameter fibres

could

activate the

SG

and

simultaneously

inhibit

the

transmission from the nociceptive afferent fibres to the T cells, located in lamina V of the dorsal horn. This could occur by way of pre-synaptic and/or post-synaptic inhibition. Aiso incorporated into the new theory is a powerful negative feedback loop from the brainstem,

called

the

descending

inhibitory

control.

After

receiving ascending information from the T cells, it can modulate and inhibit the transmission of nociceptive signaIs at the level of the SG. Midbrain and brainstem nuclei, such as the periaqueductal gray

(PAG)

and nucleus raphe magnus

(NRM),

form part of this

descending mechanism.

•

The authors acknowledged that the the ory is far from complete and revis ions may be required in the future. However, the gate

9

control theory of pain has been a great catalyst in stimulating an immense amount of research on pain mechanisms and pain management.

2. Conduction Block

Antidromic conduction block is a peripheral mechanism that has been used to explain how high frequency TENS may induce analgesia. Its

proponents

purported

that

repetitive

stimulation fatigues and/or causes a

peripheral

conduction

nerve

block of small

nocicepti ve afferents to the central nervous system (Sweet and Law, 1983)

Pertovaara (1980)

demonstrated elevated pain thresholds to

thermal stimuli applied at sites distal to high frequency TENS, but not proximal to electrode placement. campbell and Taub (1973) were able to show that periodic bursts of 100 Hz and 50 V of electrical stimulation to digital nerves of a finger always produced a painful sensation. In contrast, continuous afferent stimulation of 100 Hz that

was

gradually

increased

to

50

V had

subjects

initially

reporting pain, but ultimately reporting a numbing sensation or analgesia. Concurrently, A delta fibre activity was reduced. These observations led to the belief that pain carrying affercnts were subject to antidromic conduction block or fatigue after

being

activated. However, Sweet and Law (1983) believed that analgesia resulting from peripheral stimulation did not require activation of A delta nor C fibres,

and thus could not result in fatigue or

blockade of the conduction of these axons. Rather, they proposed

•

that pain relief was due to stimulation of large diameter fibres

10

and central mechanisms. Callaghan et al. (1978), on the other hand, supported both a peripheral small fibre blockade and central gating mechanism via large fibre stimulation for TENS. They reported that painful limbs demonstrated sensory impairment as compared to normal limbs,

possibly

because

of

tonic

small

fibre

(pain)

activity

impairing large fibre afferent terminaIs via central inhibitory pathways.

However,

when TENS was applied to painful limbs,

the

sens ory sensitivity approached normal and the resultant analgesia was due to blockade/fatigue of the small fibre activity, while TENS to normal limbs caused an impairment in sensation due ta large fibre stimulation.

3. Endoqenous Mechanisms of Pain control An endogenous mechanism of pain control has been one of the most

intensely studied

hypothesis

put forth

in

an

attempt to

explain pain modulation. Evidence for an internaI circuitry of pain regulation

was

initially

seen with

electrical

discrete areas of the brain known to generate analgesia

(SPA).

stimulatio~

SPA was first performed on animaIs

1969; Mayer et al., Hayes,

stimulation

1971; Mayer and Liebeskind,

of

produced

(Reynolds,

1974; Mayer and

1975), and effective sites were determined to lie within

mesencephalic and diencephalic areas of the brain.

Furthermore,

micro-injection of morphine into these sites and only these sites produced

analgesia,

suggesting

a

common

pathway

for

SPA

and

narcotic action (Mayer and Priee, 1976). The discovery of opiate ~

receptor binding sites in the central nervous system of rats (Pert

-----

11 et al.,

1974)

and the isolation of of endogenous morphine-like

compounds in the brain (Hughes 1975) provided further support for this relationship between SPA and opiate analgesia. Evidence for the existence of SPA in humans first came when Hosobuchi

and

colleagues

periventricular

and

(1977)

implanted electrodes

periaqueductal

gray

matter

of

into

the

patients

suffering from intractable pain. They observed a cross-tolerance between opiate analgesia and SPA, and noted that naloxone could reverse findings

the

effects

of both types

of

analgesia.

Again,

these

further suggested that electrical stimulation and the

action of narcotics could be mediated along a common pathway. The mechanisms involved in an edogenous opjoid pain control system are still not weIl understood. Anatomical structures within the CNS implicated in this system include the: PAG, nuclei within the rostral ventral medulla (RVM) - especially the nucleus raphe magnus (NRM), the arcuate nucleus,

h~ad

of the caudate, the limbic

nuclei, the pretectal nucleus, cerebellum, medial forebrain bundle, lateral

hypothalamic

region,

septal

area,

the

dorsal

lateral

funiculus (DLF) and the dorsal horn of the spinal cord (Mayer and Priee, 1976; Basbaum and Fields, 1978, 1984; Watkins and Mayer, 1982; He, 1987; Cheng and Wong, 1987; Terenzi et al., 1991). Basbaum and Fields peptides

(1984)

proposed that endogenous opioid

(met- oc leu- enkephalin, beta endorphin or dynorphin)

could disinhibi t

PAG

output neurons.

These

project to the rostral ventral medulla

•

neurons would then

(NRM).

Opiates,

at this

level, could activate the rostral ventral nuclei by inhibiting the

-

-------

12

action of

inhibitory interneurons. Neurons from the RVM nuclei

project ta the dorsal horn nuclei of the spinal cord via the OLF. A

serotonergic mediated inhibition of nociceptors is thought to

oceur at this level. Moreover, Yashpal and Henry (1990; 1992) have opi~id­

demonstrated a purine (adenosine) involvement in adrenal

indueed antinociceptive effects at the spinal cord level with rats. Sweeney et al. synaptosomes

(1988) felt that morphine could induce spinal cord

to

release

adenosine

which

would

then

bind

to

adenosine receptors resulting in spinal analgesia. Activation of this endogenous opiate pain control system can occur through many enviromental conditions such as stress (Willer and Able-Fessard, 1980), hypertension, hypoglycemia, restraint and noxious stimuli (Basbaurn and Fields, 1984). LeBars et al.

noted that noxious stimulation (heat,

(1979)

pinch and transcutaneous electrical stimulation) to distant parts ~~

the rat could profoundly inhibit the activity of dorsal horn

convergent

neurones.

This

effect

was

termed

"diffuse

noxious

inhibitory controls" (DNIe). DNIC was shown to involve supraspinal structures, have long lasting analgesic post-stimulation effects, be partially reversed by naloxone and release met-enkephalin like materials (MELM). These findings thus suggested an activation of an endogenous opioid pain control system (LeBars et al., 1979; 1981; 1987). It has been postulated that DNIC may be the neural basis

underlying the analgesia produced by counter-irritant stimulation

•

including sorne forms of acupuncture and TENS. In

fact,

Bing

et

al.

(1990;

1991)

demonstrated

that

13

acupuncture

and

noxious

thermal

stimuli

could

induce

DNIC

mechanisms, because they have similar magnitudes and time course in their analgesia,

evoked

release

of

MELM

with

antinociceptive

effects that were greatly antagonized by naloxone. The cumulative research of LeBars et al. (1989) has shown that DNIC is mediated by serotonin and opioid peptides and involves a complex loop ascending in the anterolateral quadrant of the spinal cord (especially the spinal

reticular

tract)

to

supraspinal

structures

(NMR)

and

descending a10ng the DLF. So far an endogenous opiate mechanism of analgesia has been discussed. In fact, a non-opiate system has a1so been described during some types of stress-induced ana1gesia (SIA) (Watkins and Mayer, 1982; Terman et al., 1984; Amit and Ga1ina, 1986).

For

example, footshock has been shown to evoke both opiate (naloxone reversible and cross tolerant with morphine), and non-opiate (nonnaloxone

reversible

and

not

cross

tolerant

with

morphine)

analgesia. The location, duration and number of footshock appear to induce

the

two

systems

differentially.

Specifically,

forepaw

stimulation in rats have been shawn to produce an opiate mediated form of hypogesia, while hindpaw stimulation generated a non-opiate mediated form of analgesia (Watkins and Mayer, 1982; Terman et al., 1984;

Amit et al. ,1986). Prolonged intermittent 20-30 min footshock

yielded an opiate forro of analgesia, while short durations (3 min) of continuous footshock generated a non-opiate mediated circuitry (Lewis et al., 1980). Furthermore, Watkins and colleagues (1992) ~

demonstrated that the number of shocks cou Id preferentially elicit

-----

---

14

opiate and non-opiate analgesic systems. They reported that 2 tailshock

(early)

and

80-100

tail-shock

(late)

induced

an opiate

mediated analgesic mechanism, while 5-40 tail-shock produced a nonopiate form of hypogesia. The exact mechanism of SIA

has yet to be elucidated, however,

in the rat, the nucleus raphe alatus and DLF are implicated in the opiate mediated system, because les ions to these areas abolished the antinociceptive effects. On the other hand, lesions to the DLF only partially reduced the antinociceptive effects of the opiate induced

analgesia.

In

humans,

Young

and

Chambi

(1987)

demonstrated that electrical stimulation to the PAG and

PVG

have may

involve non-opioid mechanisms, because it demonstrated a lack of cross tolerance with morphine and an insignificant reversa1 of analgesia

wi th

naloxone

administration

(when

compared

with

placebo) • Terman et al. (1984) have reported that non-opioid forms of SIA may i'lvolve serotonin, norepinephrine, dopamine antagonists and Hl antihistamines in i ts circuitry. A neural circui try rather than humora1 requlation, has been suggested for both types of analgesia, because hypophysectomy and adrenalectomy

did

not

diminish

However, continuous co Id of SIA,

the

water swims

antinociceptive

(CCWS), an alternate method

have been shown to induce hormonal non-opiate forms of

analgesia because hypophysectomy reduced (Watkins and Mayer, administration •

effects.

1982;

Amit and Galina,

its anaigesic effects 1986),

while naloxone

did not (Steinman et al., 1990). Speculation as to

the logic behind havinq t",'o or more analgesic systems have

led

15

researchers to pastulate that multiple systems allow the animal ta adapt to continued or additional

st~:esses,

no demonstration of cross tolerance

betw~'en

because there has been the analgesic systems.

Furhtermore, activation of one system can inhibit the other one (Ami t and Galina, 1986; steinman et al.,

1990).

TEllS PARAMETERS

As

health

care

professionals

and

researchers

began

to

familiarize themselves with electrical stimulation, they noted that altering

TENS

parameters could affect

the resultant analgesia.

These parameters include: 1)

electrode placement and electrode size

2) waveform 3) pulse width 4) frequency and intensi ty of stimulation 5) duration of stimulation (see Proble. P01'llulation)

1. Electrode Placement and El.ectrode Size In clinical practice,

electrode placement is important for

achieving therapeutic resul ts. Electrodes are often placed around painful

sites,

contralateral connection,

dermatomal body

parts

Tsang and Chan

stimulation to

areas,

spinal

(Mannheimer, (1984)

cord

1978) .

segments In

the

and

latter

have demonstrated that TENS

a remote contralateral

body

part could

produce

inhibition of lOlN'er limb flexion reflex. Moreaver, trigger zones, •

per ipheral nerve routes and acupuncture points have been advocated

16

to treat chronic pain conditions

(Melzack, 1975; Mannheimer and

Lampe, 1984). It should be noted that these three areas are hiqhly correlated as beinq similar points electrode

placement

diagnosis

af

is

painful

dependent condition

(Melzack, upon and

In general,

1975).

the

location of

individual

pain,

respanse

to

treatment. Electrode size should

vary according to the

size

of the

stimulated area (i. e. the larger the stimulated area the larger the electrode).

standard electrode sizes are usually 3 cm x 4cm. A

method of increasing the effectiveness is ta use larqer electrodes, especially Eriksson,

for

painful

1985).

conditions

of

the

trunk

(Sjolund and

One should take note that electrode size will.

alter the perception of stimulation, the smaller the electrode, the qreater

the

sensation

under

similar

stimulation

intensity

(Mannheimer and Lampe, 1984).

2 • • av.fora

There

are

many

waveforms

availabl.e

includinq monophasic rectangular waves,

on

the

TENS

units,

modified spike waves and

biphasic rectangular waves (Lampe, 1978). Unfortunately, there has been no definitive research to determine the waveform that is most effective (Howson, 1978; Mannheimer and Lampe 1984).

In fact, i t

has been demonstrated that various waveforms reaching the nervous tissue are different from those generated by the TENS unit (Howson,

•

1978) •

17

3. Pulse wictth Pulse width or pulse duration, in conjunction with stimulus intensity, can affect the type of nerve fibre activated. strengthduration curves in exposed cat saphenous nerves have demonstrated that C fibres are activated at pulse durations greater than 200 microsec, while A delta fibres are activated at pulse durations greater than 10 microsec and A beta fibres are excited at pulse durations greater than 2 microsec (Howson, 1978). Thus, wide pulse widths

(greater

than

1

msec)

and

high

intensities

should

simultaneously acti vate large and small myelinated nerve fibres (N.B.

Unmyelinated

intensi ties).

C fibres

However,

it

are recruited only at must

be

understood

that

very high clinical

application of electrodes on the skin, rather than on the nerve directly, will alter the characteristics of the electrical stimuli that eventually reach the nerve, due to electrical resistance of the interposed skin and tissue. Note that most TENS units used in clinical research generate pulses that vary from 4 microsec to 100 microsec in duration (0' Brien et al., 1984) •

... Frequency and Intensity of stimulation: Conventional anel Acupuncture-lite TENS Frequency investigated animal evoked

•

and

intensity

parameters

have

in afferent

studies, Chung and colleagues spinothalamic

tract

cell

been

the

electrical (1984)

activity

most

commonly

stimulation.

In

found that C fibre was

most

strongly

inhibited by peripheral nerve stimulation at high frequency, when

•

18 stimulus intensi ty was high enough to activate A delta fibres. sjolund (1985)

a Iso ebserved that intensities recruiting both A

beta and A delta fibres were more effective in depressing the C fibre evoked flexion

reflex in rats than recruitment of A beta

fibres alone. Whether these findings could be generalized to man are not known at the moment. prior to the se reports, two schools of thought have emerged wi th respect to

the

frequency and intensi ty of stimulation. As

mentioned previously, conventional TENS incorporates high frequency pulses

(50

Hz

to

200

Hz)

delivered

at

a

low

intensity

of

stimulation (2 te 3 times the sensory perception threshold of the subject). A sensation of numbness or tingling has been described •

during this type of stimulation. Acupuncture-like TENS (acu-TENS) consists of low frequency pulses intensities

(3

to

5

contraction is typically

times

the

(1 to

4 Hz)

sensory

deli vered at high threshold).

Muscle

elicited during acu-TENS. Many patients

were unable to tolerate the uncomfortable single pulses of 1 to 4 Hz, thus sjolund and Eriksson (1985) developed a technique whereby short pulse trains of 100 Hz internaI frequency were applied at slow repetition rates of 1 to 4 Hz. This stimulation technique was designed to reduce the unpleasant sensation by ensuring tetanic muscular contraction. Conventional TENS has been reported to produce an immediate analgesic effect

of

up to a maximum of 15 min of stimulation

(Eriksson and Sjolund, 1976), with a relatively rapid offset of •

about 30 min post-stimulation (Hughes et al., 1984). Sweet and Law

19 (1983) confirmed this when they found that patients sUffering from

chronic pain of distal origin, claimed immediate analgesia within seconds of turning on implanted nerve stimulators at a frequency of 33 Hz to 75 Hz. Eriksson and Sjolund (1976) found that most chronic

pain patients treated wi th conventional TENS reported induction times of pain relief immediately or within the first 10 to 15 min of

stimulation,

whereas

subjects

receiving

acu-TENS

reported

initial analgesia 30 min after onset of stimulation. Willer et al. (198?) used

conventional TENS to demonstrate a rapid and extremely

significant depres:sion of the R2 compone nt of the human blink reflex that did not outlast the stimulation periode On the other hand, electro-acupunture produced a progressive depression of the R2

component of the blink reflex, that continued for up to 17 min

after stimulation.

Ander~&on

and Holmgren (t976) demonstrated that

experimentally produced tooth pain thresholds were increased more rapidly with conventional TENS. Its post-stimulation analgesia was also shorter when compared to acu-TENS. (However, see controversial findings elucidated below). Acu-TENS is based on Chinese electro-acupuncture that uses high intensity eletrical impulses of low frequency (1 to 2 Hz) to replace manual twirling. Its analgesic effect is purported to have a prolonged onset of about 30 min, lasting

for

hours

after

and a

stimulation

graduaI offset even

(Hughes

et

al.,

1984) •

Objective evidence was provided by Andersson, Holmgren and Roos (1977) who demonstarated that experimentally produced tooth pain

•

thresholds were gradually inreased to a maximal level and then

20

slowly returned to a control level 3 hours after acu-TENS. Melzack (1975) also reported a graduaI r€lief of pain and a sustained analgesia (up to weeks) in treating chronic pain subjects with acuTENS over several treatment sessions. Thus, the two distinct analgesic onset and offset patterns produced by conventional TENS and acu-TENS have led researchers to propose

different

mechanisms

of

actions

for

each

type

of

stimulation. Conventional TENS is based on the gate control theory of pain and suggests primarily a spinally mediated system of pain control. Other contributing factors include conduction block or fatigue of peripheral nerves as previously discussed. The slow onset and gradual offset of analgesia produced by acupucture and acu-TENS are consistent wi th release of opioid substances su ch as endorphin, enkephalin and dynorphin (Hughes et al., 1984; He, 1987). This hypothesis has gained support because naloxone hydrochloride, a known narcotic/opiate antagonist, was demonstrated

to

reverse

the

low

frequency

electro-acpuncture

analgesic effects on heat to mice (Cheng and Pomeranz, 1979; Bing et al., 1990). On the other hand, naloxone did not reverse the analgesia produced by high frequency electro-acupuncture, which was partially blocked by para-chlorophenylalanine (pCPA), an inhibitor of serotonin synthesis result implicated a

(Cheng and Pomeranz,

1979).

The latter

serotonergic involvement in mediating the

analgesic action generated by high frequency electro-acupuncture. (See also the work of Han et al., (1984, 1991) reported later on). •

Further evidence for an opioid mediation of acu-TENS analgesia came

•

21

from Sjolund and Eriksson (1979). These investigators demonstrated that chronic pain patients receiving acu-TENS reported suppression of

analgesic

effects

after

naloxone

injection,

while

those

receiving conventional TENS communicated no change in analgesic effects after the introduction of naloxone.

REVIEW OF CONVENTIONAL TENS AND ACU-TENS The evidence provided so far suggested that acu-TENS, with its graduaI onset, slow offset and naloxone reversibility, is mediated by endogenous opiate peptides. This is in contra st to the rapid onset and offset of conventional TENS, based on the gate control theory. However, these observations have come under criticism from •

a number of investigators whose research findings demonstrated results to the contrary. Hansson and Ekblom (1983) noted that the majority of patients reported maximal reduction in acute oro-facial pain after 15 to 30 min of both conventional TENS and acu-TENS application. Thus, there was no signif icant difference in the induction time of maximal analgesia for both types of TENS. Additionally, Chan and Tsang (1987) demonstrated a graduaI rather than a rapid onset and offset of the inhibi tory effects of conventional TENS on the flexion reflex in humans. In fact, Mannheimer and Carlsson (1979) observed that, 1...Q.. min of 70 Hz TENS provided an average of 18 hours of pain relief, while that of 3 Hz TENS provided 4 hours and that of 3 Hz with an interna] frequency of 70 Hz provided 15 hours of pain

4It

relief in rheumatoid arthritic patients.

22 Hughes et al.

(1984)

were able to discern a

significant

increase in plasma endorphin levels in healthy subjects with both acu-TENS and conventional TENS,

but no change in

plasma beta

endorphin levels in a placebo TENS group. They also found a positve relationship between increases in pain threshold (as measured by evoked potentials over the course of what appears to be peripheral nerves) and in beta endorphin levels. Similarly, Facchinetti and his colleagues (1984) reported increased nocicepti ve flexion reflex threshold and increased plasma opioid levels (beta endorphin and beta lipotropin) in healthy subjects receiving conventional TENS. On the contrary, O'Brien et al.

(1984),

stimulatinq for 2

hours with both acu- and conventional TENS, found no change in experimental pain thresholds (induced by electrical stimuli to the median nerve over the wrist) and in plasma beta endorphin levels in healthy normal subjects. In addition, naloxone hydrochloride did not alter the pain threshold. While these contradictory results could be attributed to the factors that will be presented below, Woolf et al. (1980) were able to show that the tail flick response time of rats to noxious heat and stimulated with hiqh frequency, low

intensity TENS parameters could be significantly reduced,

though not completely reversed after the application of naloxone. Consequently, there has been sorne indication for an endoqenous opioid involvement in conventional TENS. The controversial resul ts surrounding the issue of naloxone reversibility of TENS may be due to severai factors including the ~

specificity of the antagonist such as naloxone hydrochioride, its

23

dose dependent action and the specificity of the radioimmunoassay. With regard to t.he first factor, Han et al.

(19B4)

proposed that

different types of endogenous opiates may be released by different frequencies of stimulation. They were able to demonstrate, in rats, that the analgesic effect of

high frequency

(100

Hz)

electro-

acupuncture was strongly antagonized by dynorphin A antisera, but only weakly antagonized by naloxone. Furthermore, they found that low

frequency

(2

Hz)

electro-acupuncture

analgesia

could

be

completely blocked by naloxone and met-enkephalin antisera. These findings

suggested

that

dynorphin

may

be

released

by

higher

frequencies of stimulation (e.g. 100 Hz) and that enkephalin may be released by lower frequencies of stimulation

(e.g.

recently,

similar

Han

et

al.

(1991)

demonstrated

a

2 Hz). More frequency

dependent release of various opioids in humans. This research group observed a significant increase of Met-enkephalin-Arg-Phe (367%) with acu-TENS and a significant increase of dynorphin A (49%) with conventional TENS in the CSF of human sUbjects. The second factor may be due to the dose dependent action of naloxone on different types of opiate receptors. Han and Xie (1984) concluded

that

naloxone

was

more

effective

in

reversing

the

analgesic action of morphine th an that of dynorphin on the tail flick

latency of

rats to

noxious

heat.

In

fact,

there

exist

different classes of opioid receptors (e.g. mu, delta and kappa), and

•

naloxone

could

have

a

greater

receptor as opposed to another

affinity

to

bind

with one

(Hollt, 1983; Millan, 1986). For

example, Terenius (1981) demonstrated that sorne receptors (mu) are

24

highly sensitive to naloxone while other receptors (delta, kappa) were not. Han et al. (1986) speculated that nalç·xone, which readily antagonized low frequency induced analgesia, i)robably acted on mu receptors.

In contrast,

frequency

induced

they found that the antagonism of high

antinociception

required

higher

doses

of

naloxone, which they thought to be likely due to the involvement of kappa receptors.

watkins et al. (1992, 1992) further demonstrated

that delta opiate receptors mediated SIA at supraspinal levels, while kappa opiate receptors were involved at the spinal level. The inconsistent doses of naloxone injected into animaIs and humans

across

the

different

studies,

could

surely affect

the

reversibility of possible opiate analCJesia produced by TENS and acupuncture. O'Brien et al. (1984) used only one injection of 0.4 mg of naloxone hydrochloride in their study, while Sjolund and Eriksson (1979) gave several (4-8) injections of 0.4 mg/ml solution of naloxone hydrochloride to their

subje~ts.

To further confuse the

issue, O'Brien et al. (1984) reported a paradoxical increase in the pain

threshold

administration.

of

aIl

subjects

Sawynok et al.

after (1979)

naloxone have

hydrochloride

similarly

reported

instances where naloxone in low doses (2 mg) has produced analgesia in humans classified as sensitive to experimentally induced pain, while higher doses (7.5 to 10 mg) produced hyperalgesia. Sawynok et al.

(1979)

suggested

usinq

hiqher

doses

investigation of opiate mediated analgesia,

•

beca~

of

naloxone

for

se of the multiple

forms of opiate receptors, sorne of which require higher doses of naloxone for blockade.

25

Additionally, one should be cautious when interpreting the reported

level

of

plasma

beta

endorphin

after

eiectrical

stimulation. It must be considered in conjunction wi th how weIl the radioinununoassay

detects

the opiate

(Le.

specificity

of

the

radioinununoassay) , and the normal diurnal release of beta endorphin from the pituitary gland (Hughes et al., 1984). Thus, the time of day

that

the experiment has taken place is

addition,

of

importance.

In

increased leveis of beta endorphin do not necessarily

cause an increased pain threshold (Cheng and Wong, the latter is

1987), unless

demonstrated concurrently in the study.

One must interpret results from animal research wi th care, especially if they are to be applied to human clinical practice. For example, animal studies often employ direct stimulation of the nerve and these may be performed at suprathreshold intensities, which could be too painful for human subjects to endure. Finally, subjective reports of pain relief do not always give a reliable indication of analgesia. Objective rneasures (e. g. evoked potentials or flexion reflex measurements) should be included with subjective measures to strengthen the results of the study. The flexion

reflex

objective

(FR)

measure

described of

pain

below,

and

pain

appears

to

relief

be

under

a

valid certain

circurnstances.

l'LEXJ:ON REFLEX

The flexion •

reflex

(FR)

has been described

as a

defense

mechanism against nociceptive stimulation. This polysynaptic reflex

26

manifests itself as a protective withdrawal of a

limb from an

offending stimulus (Fanagel, 1973). Meinck and associates (1981) defined the FR as comprising of an organized stereotypical pattern of

inhibition

and

facilitation

of

limb

muscles

that

ensures

adequate removal of body parts from noxious stimuli. At the spinal level, the FR incorporates an ipsilateral limb flexion response due to activation of the flexor motorneuronal pool and inhibition of the

motorn~uronal

extensor

contralateral

limb

pool.

extension

as

a

concomi tantly, result

of

there

facilitation

is of

extensor motorneurones and inhibition of flexor motorneurones. This is

known

as

coordinated

the

"crossed

reflex

extensor

withdrawal

of

reflex" the

and

allows

stimulated

limb

for

a

while

sustaining an erect posture on the contralateral leg (Kandel and Schwartz,

1985).

The

FR

is typically elicited

in response

noxious stimuli, but comprises a component that has been

to

obs~rved

ta occur with non-noxious stimuli (Shahani and Young, 1971; Willer, 1983; see below). The

lower

limb

FR

electromyographic (EMG) sural

nerve

with

has

been

studied

techniques. Hugon

surface

electrodes

and

and described (1973)

using

stjmulated the

recorded

the

reflex

activity in the short femoral biceps of normal subjects. This lower limb FR was found to consist of 2 EMG bursts separated by a silent period.

The first component,

termed RAIl,

was elicited by weak

intens i ty stimul i and thought to be mediated by lower threshold

•

group II afferents. The second burst, elicited by higher intensity stimuli, was believed to be mediated by the higher threshold group

27 III afferents and thus termed RAIII. The RAIl response, occurring between 40-60 msec, was considered tactile in nature and involved in postural positioning of the foot

(Hugon,

1973).

The RAIII

response, with lateneies of 85-120 msee, was considered nociceptive in nature and responsible for the actual withdrawal of the limb (Shahani and Young, 1971; Hugon,

1973).

Indeed,

Willer et al.

(1976) were able to show that the nociceptive component of the FR was probably mediated by the small myelinated group III afferents. Thus,

ischaemic

block

of

the

sural

nerve

produced

graduaI

disappearance of the RAIl reflex due to inactivation of the larger diameter group II afferents. The persistence of the RAIII reflex revealed that it was likely mediated by the smaller group III afferents. The onset latency, amplitude and duration of the FR offer interesting parameters to researchers. They are affected by many factors. Increasing stimulus intensities over a certain range cou Id short en the onset latency (Shahani and Young, 1971) and augment the response amplitude (Dimitrijevic, 1973). Thus, maximum tolerable intensities

(Meinck et al.,

1981)

or supramaximal

intensities

(Shahani and Young, 1971) have been employed to elicit stable FR responses. Faganel (1973) demonstrated that stimuli such as scratch and pinprick could modulate the FR response. Shahani and Young (1971) advocated the use of electrical stimulation to more accurately

•

measure the onset latency. Moreover, Meinck et al. (1983) suggested that a train of electrical pulses, 10-20 msec in duration, was more

28

appropriate than single pulses or shorter trains in eliciting the FR, because it provided temporal summation at the central synapse. Different stimulation sites could alter the onset latency of the FR and its facilitatory or inhibitory effects on the involved muscles (Hagbarth, 1960; Shahani and Young, 1971; Willer et al., 1976). The initial position of the limb and the posture of the subjects were found to be excitability.

Faganel

variables that could

(1973)

influence FR

was able to produce longer onset

latencies when the ankle was voluntarily plantarflexed as opposed to tonic dorsiflexed ankle positions. Baxendale and Farrell (1980) reported an enhanced FR response when the knee was maintained in a relatively

extended position.

Shahani and Young

(1971)

noted

different onset latencies and FR amplitudes, depending on whether subjects were seated, supine or standing. The FR is weIl known to habituate with repetitve stimulation (Dimitrijevic, 1973; Meinck et al., 1983). However, habituation can be kept to a minimum with random time intervals between stimuli (Dimitrijevic, 1973; Chan and Tsang, 1985), and tonie activation of the muscles under study (Meinck et al.,1983; Chan and Tsang, 1985). Hagbarth (1960) observed that responses of various lower extremity muscles were facilitated by voluntary background contraction. The FR and pain sensation are known to be under supraspinal influence.

It

has

been

demonstrated

that

a

task

requiring

concentration could inhibit pain sensation and the RAIl response (Willer et al., 1979); while anxiety, apprehension, cold and a full •

bladder could potentiate pain sensation and the RAIIl response

29

(Willer et al., 1979; Meinck et. al., 1983). Willer (1983) believed that the reticular activating system is involved in the inhibition of the FR, while the limbic system is involved in the facilitation of the FR. The

FR

has

been

used

in

several

human

studies

as

a

quantitative index to examine the effect of different analgesic manipulations (Willer et al., 1976; 1979; 1980; Faccinetti et al., 1984; Chan and Tsang, 1987). Willer (1977)

believed that the FR

provides an objective measure of pain in research,

because of a

strong correlation between pain sensation and the RAIlI compone11t in the lower extremity. Similarly, Dalliare and Chan (1987) showed a positive linear relationship between pain sensation and the FR area. Moreover, therapeutic doses of morphine were found to produce a rapid and siqnificant depression of the nociceptive reflexes (Willer,

1983).

Thus,

it appears that depression

component of the FR can be used as a antinociception

following

of the RAIII

quanti tati ve measure of

the application of a

pain relieving

procedure such as TENS.

PROBLEN FORMULATION

ADalq•• ic Effect. of th. Duration of Stimulation As the battle rages

on to determine what pain inhibitinq

mechanisms are activated by conventional TENS and acu-TENS, there has

been

little

characteristics.

•

research

One

stimulation period,

such

examininq

parameter,

the

other

stimulation

duration

of

the

is often overlooked and taken for qranted.

•

30 LeBars

et al.

(1979)

noted

that DNIC had

long

lasting post-

stimulation effects related to the duration of noxious stimulation. Lewis et al. analgesia,

(1980)

that

demonstrated, albeit with footshock induced

brief

and

prolonged

stimulation

may

activate

different pain inhibitory mechanisms. They found that the analgesic effects of rats subjected to 30 min of intermittent inescapable footshock were naloxone and dexamethasone reversible. On the other hand, the antinociceptive effects of those rats subjected to only 3 min of inescapable foot shocks were not antagonized by naloxone or dexamethasone. Indeed,

few authors have given scientific justification in

choosing the temporal characteristics of stimulation. One research •

group justified their treatment ot 30 to 45 min by suggesting that if improvement in pain were to occur, then changes would be noted within this time period (Wolf et al.,

1981). Generally, in human

and animal research, the duration of stimulation has been anywhere from

5

to

75

min.

In

clinical

studies

on

acute

(e.g.

post-

operative) and chronic pain, the duration of stimulation could be much longer,

from

30 min ta 48 hrs

of continuous stimulation

(Melzack, 1975; Gersh, 1978; Parry, 1979; Ali et al., 1981). Only

a

few

researchers

have

considered

the

duration

of

stimulation and its effects on analgesia. Woolf and his colleagues (1980) used conventional TENS to stimulate rats for periods of 5, 10, 15, 30 and 45 min. They determined that the analgesic effects, as measured by the latency of tail flick to heat, began to appear •

after 15 min of stimulation and reached significant leveis only

31 during 30 to 45 min of stimulation. This group also observed that the post-stimulation effects of 30 min conditioning wore off after 40 min.

Unfortunately,

they did

not

compare post-stimulation

results generated by other durations of conditioning. Facchinetti et al. (1984) demonstrated that maximal plasma opioid levels were reached 20 to 40 min after a 30 min period of conventional TENS stimulation

in

human

investigators was

subjects.

In

addition,

able to demonstrate

a

this

positive

group

of

correlati0n

between plasma opioid levels and an elevation in the RIII threshold of the flexion reflex, which was maximal at the 50th to 60th min post-stimulation. Salar and colleagues (1981) compared the increase in maximal levels of beta endorphin in the cerebrospinal fluid (CSF)

with temporal parameters of 20,

conventional TENS.

45,

60,

and

90 min of

Their resul ts indicated that maximal opioid

levels were reached at the 20 min and the 45 min durations of TENS stimulation, but this returned to baseline levels by the 90th min of stimulation. On the basis of their findings, they recommended that patients use 1 hour cycles of electrotherapy several times a day to obtain best therapeutic effects. The preceding observations have lent support to the ide a that the duration of stimulation may be important in maximizing the duration of analgesia and the level of endogenous opioids which are implicated in pain analgesia. Thus it appears that the duration of stimulation, a parameter which has often been taken for granted,

•

could be an important component in contributing to the time course and/or magnitude of TENS analgesia. With this goal in mind, and by

32 using the nociceptive component

of the

flexion reflex as the

quantitative measure of pain and the VAS score as the qualitative measure of pain, the objective of the present study can be stated as follows:

To aatermine vhether 10, 30 and &0 min of conventional TENS ati.ulation pro4uce aiqnificantly different ti.e cours. and allount of modulation of tbe flexion reflex and au):)jective pain •• timates in Dormal b.altby subjects.

1

•

33

CHAPT ER TWO

TRANSCUTANEOUS ELECTRZCAL NERVE STIMULATION: EFFECTS OF STIMULATION DURATION ON THE FLEXION REFLEX IN MAN

•

•

34

The purpose of this study was ta

SUMMARY

conventional

TENS

of

10,

30,

60

min

investigate wh ether

duration would

generate

different time course and magnitude of modulation on the lower limb flexion

reflex

(FR).

Ten

normal

subjects

received

electrical

stimuli of maximal tolerable intensity ta the sole of their foot. The

resultant

FR

was

recorded

electromyographically

from

the

ipsilateral biceps femoris (BF) muscle. Regardless of duration of stimulation, TENS suppressed the FR in 50% to 70% of subjects. Longer durations (30 an1 60 min) of TENS inhibited the FR in a slightly greater number of sUbjects (n=7 and n=6, respectively) than 10 min of TENS (n=5). In these subjects, 30 and 60 min of TENS produced a

•

maximal inhibition of the FR that reached statistical

significance

(group mean=68.35

± S.0.35.36%

and 68.35

± 3.49%

respectively of control values, p

50

«

50 25

S T ' MULATION

n=7

TENS STIKJLATION

oL--L-_..t:==t:==::t:::::::::L-....L....-----l- _.1----' -20 10 o 10 20 30 50 70 90 TINE (HIN)

2S

o -20

1

J

o

20

40 THo[ (NIN)

6080

1

120

•

66 ta 85.70% ± 10.90% of control during the 40th min of stimulation. Sixt Y

(60)

min

of

(pxpcriment ta obtain their maximal

tolerable intensity and this

enabled them to be "mentally set" for such stimulation. Third, sorne subjects demonstrated an increase of pain estimates

(Gee below)

rather than a graduaI decline of VAS scores. The mean reduced VAS scores, "pooled" during 60 min TENS and 60 min placebo st imulation were significantly different (p