Pain, 54 (1993) 99-106

0 1993 Elsevier Science Publishers B.V. All rights reserved 0304-3959/93/$06.00

PAIN 02314

99

Is TENS purely a placebo effect? A controlled study on chronic low back pain *

Serge Marchand a, Jacques Charest a, Jinxue Li a, Jean-RenC Chenard ‘, Benoit Lavignolle b and Louis Laurencelle ’

a Universite’ du Qu&ec en Abitibi-Ttfmiscamingue, Dipartement des Sciences de la Sank, Rouyn-Noranda, Quebec (Canada), b Universite’ de

Bordeaux 14 Centre Hospitalier R&ional, Dipartements d’orthopddie et de Riducation Rhdaptation Fonctionneks, Bordeaux (France) and

’ Universite’ du Qu&c 2 Trois-RiviZres, De’partement des Sciences de l'activitk physique, Trois-RiviZres, Quebec (Canada)

(Received 6 October 1992, revision received 25 January 1993, accepted 1 February 1993)

Summary Although high-frequency low-intensity transcutaneous electric nerve stimulation (TENS) has been , extensively used to relieve low back pain, experimental studies of its effectiveness have yielded contradictory findings mainly due to methodological problems in pain evaluation and placebo control. In the present study, separate visual analog scales (VAS) were used to measure the sensory-discriminative and motivational-affective components of low back pain. Forty-two subjects were randomly assigned to 1 of 3 groups: TENS, placebo-TENS, and no treatment (control). In order to measure the short-term effect of TENS, VAS pain ratings were taken before and after each treatment session. Also, to measure long-term effects, patients rated their pain at home every 2 h throughout a 3-day period before and 1 week, 3 months and 6 months after the treatment sessions.

In comparing the pain evaluations made immediately before and after each treatment session, TENS and placebo-TENS significantly reduced both the intensity and unpleasantness of chronic low back pain. TENS was significantly more efficient than placebo-TENS in reducing pain intensity but not pain unpleasantness. TENS also produced a significant additive effect over repetitive treatment sessions for pain intensity and relative pain unpleasantness. This additive effect was not found for placebo-TENS. When evaluated at home, pain intensity was significantly reduced more by TENS than placebo-TENS 1 week after the end of treatment, but not 3 months and 6 months later. At home evaluation of pain unpleasantness in the TENS group was never different from the placebo-TENS group. These results suggest that TENS reduces both the sensory-discriminative and motivational-af- fective components of low back pain in the short term but that much of the reduction in the affective component may be a placebo effect. We conclude that TENS should be used as a short-term analgesic procedure in a multidisciplinary program for low back pain rather than as an exclusive or long-term treatment.

Key words: Transcutaneous electric nerve stimulation; Low back pain; Placebo; Chronic pain

Introduction

Transcutaneous electric nerve stimulation (TENS) has emerged as a popular method to treat various chronic pain conditions (Meyer and Fields 1972;

* Presented in part at the annual meeting of the American Pain

Society, New Orlean, Louisana, USA, 1991.

Correspondence to: Serge Marchand, Universite du Quebec en

Abitibi-Timiscamingue, C.P. 700, Rouyn-Noranda, Quebec J9X 5E4,

Canada.

Thorsteinsson et al. 1978; Francini et al. 1981; Fried et al. 1984; Lundeberg 1984; Brill and Whiffen 1985; Lehmann et al. 1986; Cheng and Pomeranz 1987; Lampe and Dunn 1987; Phero et al. 1987; Umeh 1987; Bremerich et al. 1988; Graff-Radford et al. 1989a,b; Leandri et al. 1990; Marchand et al. 1991~) including chronic low back pain (Fox and Melzack 1976; Mannheimer and Carlsson 1979; Dougherty 1982; Deyo 1983; Lehmann et al. 1983, 1986; Melzack et al. 1983; Deyo et al. 1990). However, the effectiveness of its use has been put in question by the contradictory findings of these various studies. A recent example is the Deyo

et al. (1990) study that found no apparent benefit of TENS for chronic low back pain. One explanation for these apparently contradictory results is the paucity of adequate experimental control over two fundamental types of methodological problems: placebo effect and pain measurement. Here, pain measurement problems include: (i) lack of sensitivity of pain measurement; (ii) inadequate measurement of the components of pain; (iii) measurement relying on the memory of pain and finally; (iv>. not accounting for pain modulation over time and c&text. These problems and how they can be addressed are briefly described as follows.

Sensititity 0; pain measurement Of importance in conducting studies on the effect of

an analgesic technique is the use of pain ratings that a$e sensitive enough to detect small changes in pain pkrception. Even in morphine analgesia, the clinical effectiveness of which has been clearly demonstrated, a clinically effective dose changes the perception of pain by only 1°C or 2°C in experimental thermal pain (Price et BI. 1985, 1986). A number of studies on TENS analgesia use a 4- or j-step verbal descriptor to mea- sure pain, (Thorsteinsson et al. 1978; Hansson and Ekblom 1983, 1984) which is less precise considering that the decision to pass to the next descriptor relies on the patient’s behavior. For example, a patient may rapidly shift from the first descriptor to the next even if only a small reduction in pain has been perceived while waiting for a relatively important change in pain per- ception to shift from the second to the third descriptor. Considering that changes in perception of pain due to physical treatments such as TENS or dorsal column stimulation (DCS) could be slight (less than the 20% or 25% change represented by the step verbal descriptors) (Bushnell et al. 1991; Marchand et al. 1991a,b), this degree of precision is clearly too low. This is supported by the study of Ekblom and Hansson (1988) where they concluded that a S-level verbal descriptor, as they had previously used, was not sensitive enough to measure the pain modulation produced by TENS. An alterna- tive to step verbal descriptors is the visual analog scale WAS) which has been successfully used to detect pain modulation by non-pharmacological analgesia (Chap- man et al. 1987; Talbot et al. 1987, 1989; Miron et al. 1989; Feine et al. 1990; Bushneel et al. 1991; Marchand et al. 1991a,b). These scales were sensitive enough to detect a change of less than 20% in the perception of experimental thermal heat pain and clinical pain dur- ing TENS or DCS treatment (Marchand et al. 1991a,b).

~ea.~~rement of the components of pain Pain can be differentiated into sensory-discrimina-

tive and motivational-affective components which vary in relative importance depending on the type of pain (Price et al. 1987). Studies show that certain treatments

affect 1 component more than the other (Gracely et al. 1978; Kupers et al. 1991; Price et al. 1985). Conse- quently, it is important to measure these 2 components of pain independently in order to determine if the analgesic procedure is preferentially affecting one as- pect. The VAS scale used in our laboratory for sepa- rately measuring the 2 components of pain has previ- ously been validated for measurement of experimental and chronic pain (Price et al. 1983).

Measurement relying on the memory of pain Pain is often evaluated by ‘pain relief or ‘self-rated

improvement’, that is by asking the patient to compare his post-treatment with his pre-treatment pain. Yet, studies have demonstrated that the memory for pain is not accurate (see Erskine et al. 1990 for review). If, as reports suggest, patients have a natural tendency to exaggerate their evaluation of past pain (Eich et al. 1985; Feine et al. 1989; Jamison et al. 1989; Linton and Gdtestam 1983; Linton and Melin 19821, then present pain of the same intensity could be evaluated as being less painful, leading to false-positive results. In order to avoid this problem, the patient should evaluate his present pain without comparing it to past pain.

Pain modulation ol’er time and context Even in patients suffering from chronic pain for

years, pain intensity changes between days and even during the same day (Gilpin et al. 1990; Behar et al. 1991; Bellamy et al. 1991; Solomon et al. 1991). Thus, if a pre-test is taken during a period in which the patient experiences a great deal of pain, any subse- quent evaluation of less pain might lead to the conclu- sion that pain was reduced by the treatment when in fact it could just be a normal variation. One explana- tion for modulation in pain has been a change in attention (Miron et al. 1989). which could be caused by the context or environment, such as being in an unfa- miliar laboratory or clinic. To take pain modulation over time and context into consideration, an evaluation should be made at home during several consecutive days at different times during the same day (Chenard et al. 1991).

Placebo effect Considering that 20-30% of clinical improvement

can be explained by placebo effects, it is important to control for placebo responses in evaluating the effec- tiveness of an analgesic procedure (Jospe 1978). To our knowledge, two studies have verified the effect of TENS on low back pain with a placebo control (Lehmann et al. 1986; Deyo et al. 19901, both concluding that TENS is not significantly different from a placebo. However, the placebo control in the study of Deyo et al. (1990) is problematic. That study used a heat treatment (e.g., moist-heat treatment called hot packs twice a week at

the laboratory and electric-heating-pad treatment 45 min (average) per day at home) on each group includ- ing the placebo group. However, the a&mption that the effect of such heat treatment was exactly the same in each group (sham TENS, exercise and/or TENS) is unverified. The conclusions regarding the placebo ef- fect of TENS in the Deyo et al. study can also be questioned on the grounds of pain evaluation. Indeed, Deyo et al. measured pain for short periods of time in the clinic without considering pain modulation over time and context. Finally, neither study (Lehmann et al. 1986; Deyo et al. 1990) measured independently the sensory-discriminative and motivational-affective com- ponents of pain, and both studies relied on the patient’s memory of pain.

ling for 4 variables: sex, weight, diagnosis and pain severity. Weight

was defined as normal or heavy (between 10 and 20% over the

Canadian Heart Association normal weight curve). Diagnosis was

artificially divided in 3 categories: (a) somatogenic pain (muscu-

loskeletal pain or facet syndrome with anterior branch irritation), (b)

neurogenic problems with root compression and, (c) more specific

pathology including a&losing spondylitis, rheumatoid arthritis, scol-

iosis of less than 15” and pyramidal syndrome. When a patient’s

diagnosis included multiple categories, he was included in the cate-

gory presenting the most clinical challenge. Pain severity was divided

into 3 categories based on the 3 days of home pain evaluation before

the treatment (see next section on pain measurement): mild ( < 40%

on the VAS measuring intensity of pain), moderate (40-60%) or

severe pain ( > than 60%).

The present study presents a detailed psychophysi- cal investigation of the effect of TENS on chronic low back pain using 2 distinct VASs for the sensory-dis- criminative and motivational-affective components. Pain (with no reference to past pain) was evaluated at the laboratory before and after each treatment session and at home every 2 h during a 3-day period prior to treatment and again at 1 week, 3 months and 6 months after the end of treatment. A control group was also included in order to measure the effect of the simple passage of time on pain perception,

In concordance with the requirements of the University Human

Subjects Review Committee, each subject was informed as to the

nature of the treatment they would receive, as well as their right to

withdraw from the experiment at any time without prejudice; each

signed a consent form. Patients in the TENS and placebo-TENS

groups were told that we were verifying different stimulation param-

eters to measure their efficacy in chronic low back pain management.

Treatment dekiption After a clinical evaluation to determine the appropriate electrode

position according to the dermatome to produce paraesthesia in the

area of pain, patients were asked to come to the laboratory twice a

week during 10 weeks to receive treatment. For the TENS group, the

stimulation was given at high frequency (100 Hz, using square-wave

125 ps pulses), low intensity (stimulus amplitude adjusted for each

subject to produce a clear, but unpainful, paraesthesia) using a

2-channel Medtronic adjustable stimulator (TENS 7720). After the

intensity was adjusted, it was unchanged for the 30-min treatment.

Methods

Subjects Recruited by medical referral and advertisements in the local

newspaper, 48 patients, between 18 and 60 years of age and with a history of chronic low back pain of more than 6 months, volunteered

to participate in this study. The clinical exclusion criteria of Nachem-

son (1980) were applied when subjects were excluded from the study:

scoliosis of more than 15” (no case), important lost of sensibility and

reflex indicating a root compression and a surgical indication (1

case), collapse of more than 3 lumbar level (no case), a spondylohs-

thesis of more than 1 cm (1 case) and obesity problems of more than

20% on the Canadian Heart Association scale (3 cases). Also ex-

cluded was 1 case with important psychological problems. Forty-two

patients (87.5% of the patients recruited) were finally accepted (see

Table I).

For the placebo-TENS procedure, a TENS apparatus with visual

(oscilloscope and voltage indicator) and sound feedback was short-

circuited so that no current passed to the electrodes. An indepen-

dent experimenter pretended to adjust the stimulator and asked the

patients to report when they first felt a small sensation. All patients

reported a sensation.

Finally, for the control group, an equivalent group of patients,

receiving no treatment, followed the same pain evaluation described

hereafter. They were informed that they would receive treatment 6

months later.

Pain evaluation

Treatment assignment

Home and laboratory pain evaluation were performed using 2

separate VASs, 1 for pain intensity and 1 for pain unpleasantness.

Each scale consists of a lo-cm horizontal line with descriptors at the

far right and far left. For the intensity scale the anchor point on the

right is ‘no pain at all’ and on the left is ‘the most intense I could

imagine’. For the unpleasantness scale, the anchor point on the right

is ‘not unpleasant at all’ and on the left is ‘the most unpleasant I

could imagine’. A pseudo-random assignment of the patients to the 3 groups, During a practice session, patients came to the laboratory to rate

TENS (14), placebo-TENS (12) and control (16), was done control- their present pain intensity (sensory-discriminative component) and

TABLE I

PATIENTS DESCRIPTION

101

TENS

Male

6

Female

8

Age (S.D.) (years)

35.46 (7.84)

Low back pain duration

(S.D.) (years)

10.73 (7.08)

Placebo-TENS

Control

6 6 35.08 i7.38j 8.83 (6.21)

8 8 37.25 (8.18) 6.51 (6.22)

total: 20 total: 22 mean age: 36 mean pain duration: 9

102

pain unpleasantness (motivational-affective component) with the as- sistance of an independent experimenter. The distinction between the 2 dimensions of pain was explained using the analogy of Price et al. (1983) that we have previously used successfully (Marchand et al. 1991b).

At home, throughout a 3-day period before treatment, subjects rated their pain intensity and unpleasantness on a VAS every 2 h. In order to insure that each subject was evaluating his pain as pre- scribed, a research assistant called the patients 3 times a day to verify their evaluation. The same rating was done througbout a 3-day period 1 week after the end of treatment and then throughout a 3-day period 3 months and 6 months later. In addition, during the 10 weeks of treatment, subjects rated in the laboratory their present pain intensity and pain unpleasantness on a VAS immediately before and immediately after each treatment session. The patients were informed that they would be contacted at pre-determined moments after the end of the treatment to evaluate their pain again. All treatment and pain measurement sessions with the oatients were done individually.

Statistical analysis Mean ratings were calculated for each laboratory and home pain

rating session. Pain rating data were analyzed statistically using repeated MANOVA (SuperAnova, from Abacus). For the compari- son between TENS and placebo-TENS sessions over time, post-hoc Dunnet t tests were calculated. A regression analysis was performed using a regression ANOVA comparing the session number with pain intensity and pain unpleasantness (Statview SE+ from Abacus).

Results

Prior to treatment, there was no statistical differ- ence between the 3 groups in home rating of pain intensity (F(2,37) = 0.006, P = 0.99) and pain unpleas- antness (F(2,37) = 0.04, P = 0.97) (Fig. 1).

111 TENS

Placebo-TENS

q Control

Fig. 1. Mean home pain VAS ratings of the low back pain prior to treatment for the 3 groups (TENS, placebo-TENS and control groups). All 3 groups are statistically equivalent for pain intensity

and pain unpleasantness.

Short-term results Comparing before and after each treatment session,

TENS significantly reduced pain intensity ratings (F(1,12) = 58.39, P = 0.0001) as well as pain unpleas- antness ratings (F(lJ2) = 35.08, P = O.OOOl), reduc-

a r’40

1

Pmxk) =0.008

i T

TENS PLACEBO

b

a BEFORE

q AFTER

TENS ’ PLACEBO ’

Fig. 2. Mean VAS pain ratings of low back pain intensity (a) and unpleasantness (b) immediately before and after each TENS and placebo-TENS treatment sessions. TENS as well as placebo-TENS significantly reduced both the intensity and the unpleasantness rat- ing. TENS reduced pain intensity significantly more than placebo-

TENS. This was not so with pain unpleasantness.

tions of 43% and 42%, respectively. Placebo-TENS produced similar reductions: intensity rating (F(1,12) = 10.00, P = 0.008) and unpleasantness (F(lJ2) = 8.14, P = 0.015>, 17% and 18%, respectively (see Fig. 2a,b).

When comparing the effect of TENS with pIacebo- TENS, the reduction in intensity rating is significantIy greater for TENS than placebo-TENS (F&24) = 4.22, P = 0.05) but not so for the unpleasantness rating (F(1,24) = 1.38, P = 0.252). In order to see if the signif- icant effect found for pain intensity would be masked by the non-significant results for pain unpleasantness in a general pain measurement, the intensity and un- pleasantness rating for TENS and pIacebo-TENS were combined. This resulted in no significant difference between TENS and placebo-TENS (F(1,24) = 2.82, P = 0.106).

To verify if there was a tendency for a cumulative effect of TENS over repeated treatment sessions, we calculated the linear regression for the mean intensity and unpleasantness rating, respectively, before each of the first 16 treatment sessions (Fig. 3). For the TENS group, we found a significant decrease over sessions in the intensity @X1,14) = 5.01, P = 0.042) as well as the unpleasantness (F(lJ4) = 6.01, P = 0.028) of pain. No

significant cumulative effect was found for the placebo-TFNS group for pain intensity (F(1,14) = 2.18, P = 0.162) nor for unpleasantness (F(1,14) = 0.65, P = 0.435).

a 40 0 TENS INT BEF -

. PLAC INTBEF - - -

151 0 2 4 6 8 10 12 14 16

Treatment sessions

b 35

1

0 TENSUNPLBEF -

8 PLAC UNPL BEF - - -

tct : 3 - 00

s ‘5: 0 f -

0 2 4 6 8 10 12 14 16 Ttwament sessbns

Fig. 3. Regression curves for pain intensity (a) and pain unpleasant- ness (b) ratings before each of the first 16 TENS or placebo-TENS treatment sessions. TENS significantly decreased pain intensity and pain unpleasantness rating in an cumulative manner over repeted treatment sessions. Placebo-TENS showed no cumulative effect for

intensity or unpleasantness.

Lmlg-&ml rt?sutts One week after the end of the 10 weeks of TENS

treatment, pain intensity was still evaluated as lower than before the treatment (F&27) = 16.22, P = 0.~4). The same effect was found for unpleasantness (F(1,27) = 14.34, P = O&008> (Fig. 4c, & TENS still had an effect 3 months and 6 months ifter the end of the treatment as compared to before treatment for pain intensity rating (3-month: F(1,27) = 0.9.65, P = 0.004; 6-month: F(1,27) = 24.55, P = 0.~1) as well as unpleasantness (3-month: = F(1,27) = 5.25, P = 0.03; 6-month: F(1,27) = 14.57, P = 0.0007).

-, ._ CONTROL -I

a2 I1 PM

5 ’ ’ BEF AFT 3-M S-M

Fig. 4. Short- and long-term effects measured by the home pain evaluation before (BEF), after (AFT), 3 months (3-M) and 6 months (6-M) for TENS, placebo-TENS and control groups. TENS as well as placebo-TENS produced a significant long-term effect, even 6 months

after the end of treatment.

The same results were found for the placebo-TENS group except immediately after the end of the treat- ment where only a non-significant tendency for pain reduction was found (F&36) = 1.94, P = 0.172) (Fig. 4b,,,). Placebo-TENS significantly reduced pain inten- sity rating 3 months (F(1,36) = 5.39, P = 0.026) and 6 months after the end of the treatment (F(1,36) = 7.09, P = 0.02). The same results were found for pain un- pleasantness rating (after: F&36) = 2.16, P = 0.15; 3- month: F(1,36) = 6.25, P = 0.017; 6-month: F(1,36) = 5.71, P = 0.039).

For the control group, pain intensity (F(3,39) = 1.37, P = 0.268) and pain unpleasantness (F(3,39) = 1.54, P = 0.218) were not significantly changed by the mere passage of time (Fig. 4a,,,).

When we compare TENS results with placebo-TENS (Fig. 5a, b), TENS is significantly more effective than the placebo in reducing pain intensity 1 week after the end of the treatment (t Dunnet (3, 72) = 2.50, P < 0.05). No significant difference was found 3 months (t Dunnet (3, 72) = 0.741, NS) nor 6 months (t Dunnet (3, 72) = 2.178, NS) after the end of the treatment. As for the unpleasantness rating, no significant difference

a 40- - z

go-

P .= e .$20-

2

.; IO-

5 z”

+ TENS

-. PLACEBO

- ’ BEFORE ’ AFTER ’ J-MONTH ’ B-MONTH ’

b g40- ui

BEFORE AFTER 3-MONTH B-MONTH

Fig. 5. Comparison of the mean home pain VAS ratings of TENS and placebo-TENS groups. There was a significant difference be- tween TENS and placebo groups immediately at the end of the treatment sessions for the intensity rating but no difference was

found for the unpleasantness rating,

between TENS and placebo-TENS was found at any moment after the end of the treatment.

Discussion

In contrast with prior studies (Lehmann et al. 1986; Deyo et al. 19901, the present results, indicating that TENS is more effective than placebo-TENS in reduc- ing the intensity of low back pain immediately after each treatment session as well as 1 week after the end of the treatment, confirm TENS as an efficient non- pharmacological short-term analgesia for low back pain.

Our finding, that in the short term, TENS reduced perceived pain intensity more than did placebo-TENS but that the 2 treatments reduced pain unpleasantness equally, raises two points. The first is the importance of measuring the 2 components of pain independently. This was demonstrated by the disappearance of any significant difference when the intensity and unpleas- antness ratings were combined. The second is an indi- cation that TENS might also include a placebo effect. The sensitivity of the unpleasantness component of

pain to a placebo is not surprising considering that the affective-emotional aspect of pain is easily modulated by psychological manipulations (Gracely 1979; Price et al. 1980). The placebo effect is a universal component found in al1 pha~acological or non-pha~acological analgesic procedures (Keefe et al. 1991) and, if ade- quately understood, can be used advantageously in the clinical setting and accounted for in research design by adding a control group.

Regarding long term effects, the present findings that TENS reduced low back pain as long as 6 months after the end of the treatment (Fig 4c,,,) may be explained by increased physical activity taking place as a result of decrease in pain. The value of an increase in physical activity has been found to be the most impor- tant aspect of low back pain treatment (Fordyce et al. 1973; Kraus et al. 1983; Waddell 1987; Deyo et al. 1990). The same long-term effect, afthough to a lesser degree, produced by the placebo treatment may be similarly explained (Fig. 4b ,,2).

The stability of pain in the control group over a period of 6 months indicates that the 3-day home rating procedure adequately accounts for the natural modulation of pain. This control for pain modulation increases the reliability of the results found in the treatment groups.

Finally, the fact that TENS reduced pain intensity and pain unpleasantness, despite a major piacebo ef- fect for the motivational-affective component, leads us to conclude that TENS is clinically useful as a short- term analgesic procedure for low back pain. As such, it constitutes a valuable component in a more complete multidisciplinary treatment program for low back pain which includes increased exercise. Due to the additive effect of TENS over time, we also recommend its repetitive use over a short period of time to potentiate its effect.

Acknowledgements

We wish to thank Paule Julien, Roxanne Duquette and Serge Daigle for technical assistance; and Sarah Shidler and Dr. M. Cathrine Bushnell for their pre- cious comments on earlier versions of this manuscript.

References

Behar, S., Reicherreiss, H., Goldbourt, U. and Kaplinsky, E., Circa- dian variation in pain onset in unstable angina peetoris, Am. J. Cardiol., 67 (1991) 91-93.

Bellamy, N., Sothern, R.B., Campbell, J. and Buchanan, W.W., Circadian rhythm in pain, stiffness, and manual dexterity in rheumatoid arthritis: relation between discomfort and disability, Ann. Rheum. Dis., 50 (1991) 243-248.

10.5

Bremerich, A., Wolfgang, W., Thein, T. and Dietze, T., Transcuta- neous electric nerve stimulation (TENS) in the therapy of chronic facial pain, J. Cranio-Max. Facial Surg., 16 (1988) 379-381.

Brill, MM. and Whiffen, J.R., Application of 24-hour burst TENS in a back school, Phys. Ther., 65 (1985) 135.5-1357.

Bushnell, M.C., Marchand, S., Tremblay, N. and Duncan, G.H., Electrical stimulation of peripheral and central pathways for the relief of musculoskeletal pain, Can. J. Physiol. Pharmacol., 69 (1991) 697-703.

Chapman, C.E., Bushnell, M.C., Miron, D., Duncan, G.H. and Lund, J.P., Sensory perception during movement in man, Exp. Brain Res., 68 (1987) l-9.

Chenard, J.R., Marchand, S., Charest, J., Li, J. and Lavignolle, B., Evaluation d:une approche ~m~rtementale de la lombalgie chronique: l’Ecole interactionnelle du dos, Sci. Comportement, 28 (1991) 225-239.

Cheng, R.S.S. and Pomeranz, B., Electrotherapy of chronic muscu- loskeletal pain: comparison of electroacupuncture and acupunc- ture-like transcutaneous electrical nerve stimulation, Clin. J. Pain, 2 (1987) 143-149.

Deyo, R.A., Conservative therapy for low back pain, J. Am. Med. Assoc., 250 (1983) 1057-1062.

Deyo, R.A., Walsh, N.E., Martin, D.C., Schoenfeld, L.S. and Rama- murthy, S., A controlled trial of transcutaneous electrical nerve stimulation (Tens) and exercise for chronic low back pain, N. Engl. J. Med., 322 (1990) 1627-1634.

Dougherty, R.J., Transcutaneous electrical nerve stimulation: an alternative to drugs in the treatment of acute and chronic pain, Ann”. Sci. Assemb., (1982) Abst.

Eich, E., Reeves, J.L., Jaeger, B. and Graff-Radford, S.B., Memory for pain: relation between past and present pain intensity, Pain, 23 (1985) 3X-379.

Ekblom, A. and Hansson, P., Pain intensity measurements in pa- tients with acute pain receiving afferent stimulation, J. Neurol. Neurosurg. Psychiat., 51 (1988) 481-486.

Erskine, A., Morley, S. and Pearce, S., Memory for pain: a review, Pain, 41 (1990) 255-265.

Feine, J.S., Dao, T.T.T., Lavigne, G.J. and Lund, J.P., Is treatment success due to faulty memory of chronic pain?, J. Dent. Res., abst. 68 (1989) 1018.

Feine, J.S., Chapman, C.E., Lund, J.P., Duncan, G.H. and Bushnell, MC., The perception of painful and nonpainful stimuli during voluntary motor activity in man, Somat. Mot. Res., 7 (1990) 113-124.

Fordyce, W.E., Fowler, R. Jr., Lehmann, J.F., Delateur, B.J., Sand, P.L. and Trieschmann, R.B., Operant conditionning in the treat- ment of chronic pain, Arch. Phys. Med. Rehabil., 54 (1973) 339-408.

Fox, E.J. and Melzack, R., Transcutaneous electrical stimulation and acupuncture: comparison of treatment for low-back pain, Pain, 2 (1976) 141-148.

Francini, F., Maresca, M., Procacci, P. and Zoppi, M., The effects of non-painful transcutaneous electrical nerve stimulation on cuta- neous pain threshold and muscular reflexes in normal men and in subjects with chronic pain, Pain, 11 (1981) 49-63.

Fried, T., Johnson, R. and McCracken, W., Transcutaneous electri- cal nerve stimulation: its role in the control of chronic pain, Arch. Phys. Med. Rehabil., 65 (1984) 228-231.

Gilpin, E.A., HjaIma~on, A. and Ross, J., Subgroups of patients with atypical circadian patterns of symptom onset in acute my- ocardial infarction, Am. J. Cardiol., 66 (1990) G7-Gll.

Gracely, R.H. Psychophysical assesment of human pain. In: J.J. Bonica, J.C. Liebeskind and D.G. Albe-Fessard (Eds.), Advances in Pain Research and Therapy, Raven Press, New York, 1979, pp. 805-824.

Gracely, R.H., McGrath, P. and Dubner, R., Validity and sensitivity

of ratio scales of sensory and affective verbal descriptors: manipu- lation of affect by diazepam, Pain, 5 (1978) 19-29.

Graff-Radford, S.B., Reeves, J.L., Baker, R.L. and Chiu, D., Effects of transcutaneous electrical nerve stimulation on myofascial pain and trigger point sensitivity, Pain, 37 (1989a) l-5.

Graff-Radford, S.B., Reeves, J.L., Baker, R.L. and Chiu, D., Effects of transcutaneous electrical nerve stimulation on myofascial pain and trigger point sensitivity, Pain, 37 (1989b) 1-5.

Hansson, P. and Ekblom, A., Trans~taneous electrical nerve stimu- lation (TENS) as compared to placebo tens for the relief of acute oro-facial pain, Pain, 15 (1983) 157-165.

Hansson, P. and Ekblom, A., Afferent stimulation induced pain relief in acute oro-facial pain and its failure to induce sufficient pain reduction in dental and oral surgery, Pain, 20 (1984) 273-278.

Jamison, R.N., Sbrocco, T. and Parris, W.C.V., The influence of physical and psychosocial factors on accuracy of memory for pain in chronic pain patients, Pain, 37 (1989) 289-294.

Jospe, M., The Placebo Effect in Healing, Lexington Books, Lexing- ton, MA, 1978, 170 pp.

Keefe, F.J., Beckham, J.C. and Fiilingim, R.B., The Psychology of Chronic Back Pain, Adult Spine: Principles and Practices, 1991, pp. 185-197.

Kraus, H., Nagler, W. and Melleby, A., Evaluation of an exercice program for back pain, Am. Fam. Phys., 28 (1983) 153-158.

Kupers, R.C., Konings, H., Andriaensen, H. and Gybels, J.A., Mor- phine differentially affects the sensory and affective pain ratings in neurogenic and idiopathic forms of pain, Pain, 47 (1991) 5-12.

Lampe, J. and Dunn, B., Symmetrical biphasic TENS waveform for treatment of back pain, Clin. J. Pain, 3 (1987) 145-151.

Leandri, M., Parodi, CL, Corrieri, N. and Rigardo, S., Comparison of TENS treatments in hemiplegic shoulder pain. Stand. J. Re- hab. Med., 22 (1990) 69-72.

Lehmann, T.R., Russel, D.W. and Spratt, K.F., The impact of patients with nonorganic physical findings on a controlled trial of transcutaneous electrical nerve stimulation and electroacupunc- ture, Spine, 8 (1983) 625-634.

Lehmann, T.R., Russell, D.W., Spratt, K.F., Colby, H., Liu, Y.K., Fairchild, M.L. and Christensen, S., Efficacity of electroacupunc- ture and TENS in the rehabilitation of chronic low back pain patients, Pain, 26 (1986) 277-290.

Linton, S.J. and Glitestam, G., A clinical comparison of two pain scales: correlation, remembering chronic pain, and a measure of compliance, Pain, 17 (1983) 57-65.

Linton, S.J. and Melin, L., The accuracy of remembering chronic pain, Pain, 13 (1982) 281-285.

Lundeberg, T., A comparative study of the pain alleviating effect vibratory stimulation, transcutaneous electrical nerve stimulation, electroacupuncture and placebo, Am. J. Chin. Med., 12 (1984) 72-79.

Mannheimer, C. and Carlsson, C.-A., The analgesic effect of transcu- taneous electrical nerve stimulation (TENS) in patients with rheumatoid arthritis. A comparative study of different pulse patterns, Pain, 6 (1979) 329-334.

Marchand, S., Bushnell, M.C. and Duncan, G.H., Modulation of heat pain perception by high frequency transcutaneous electrical nerve stimulation (TENS), Clin. J. Pain, 7 (1991a) 122-129.

Marchand, S., Bushnell, M.C., Duncan, G.H., Molina-Negro, P. and Martinez, S.N., The effects of dorsal column stimulation on measures of clinical and experimental pain in man, Pain, 45 (1991b) 249-257.

Marchand, S., Duquette, R., Chenard, J.R., Charest, J. and Li, .I., Short- and long-term effect of tens and placebo tens on low back pain, American Pain Society Abstracts, New Orleans, 1991c, Abst.

Melzack, R., Vetere, P. and Finch, L., Transcutaneous electrical nerve stimulation for low back pain comparison of TENS and

106

massage for pain and range of motion, Phys. Ther., 63 (1983) 489-493.

Meyer, G.A. and Fields, H.L., Causalgia treated by selective large fibre stimulation of peripheral nerve, Brain, 9.5 (1972) 163-168.

Miron, D., Duncan, G.H. and Bushnell, M.C., Effects of attention on the intensity and unpleasantness of thermal pain, Pain, 39 (1989) 345-352.

Nachemson, A., A critical look at conservative treatment for low back pain. In: J. Malcom (Ed.), The Lumbar Spine and Back Pain, Pitman Medical, New York, 1980, pp. 453-464.

Phero, J.C., Prithvi Raj, P. and McDonald, J.S., Transcutaneom electrical nerve stimulation and myoneural injection therapy for management of chronic myofascial pain, Dent. Clin. N. Am., 31 (1987) 703-723.

Price, D.D., Barrel], J.J. and Gracely, R.H., A psychophysical analy- sis of experimental factors that selectively influence the effective dimension of pain, Pain, 8 (1980) 137-149.

Price, D.D., McGrath, P., Rafii, A. and Buckingham, B., The valida- tion of visual analogue scales as ratio scale measures for chronic and experimental pain, Pain, 17 (1983) 45-56.

Price, D.D., Harkins, SW., Rafii, A. and Price, C., A simultaneous comparison of fentanyl’s analgesic effects on experimental and clinical pain, Pain, 24 (1986) 197-203.

Price, D.D., Harkins, SW. and Baker, C., Sensory-affective relation-

ships among different types of clinical and experimental pain, Pain, 28 (1987) 297-308.

Price, D.D., Von der Gruen, A., Miller, J., Rafii, A. and Price, C., A psychophysical analysis of morphine analgesia, Pain, 22 (1985) 261-270.

Solomon, B., Reicher-Reiss, H., Goldbourt, U. and Kaplinsky, E., Circadian variation in pain onset in unstable angina pectoris, Am. J. Cardiol., 67 (1991) 91-93.

Talbot, J.D., Duncan, G.H. and Bushnell, M.C., Effects of diffuse noxious inhibitory controls (DNICS) on the sensory-discriminative dimension of pain perception, Pain, 36 (1989) 231-238.

Talbot, J.D., Duncan, G.H., Bushnell, M.C. and Boyer, M., Diffuse noxious inhibitory controls (DNICs): psychophysical evidence in man for intersegmental suppression of noxious heat perception by cold pressor pain, Pain, 30 (1987) 221-232.

Thorsteinsson, G., Stonnington, H.H., Stillwell, G.K. and Elveback, L.R., The placebo effect of transcutaneous electrical stimulation, Pain, 5 (1978) 31-41.

Umeh, B.U.O., Initial experience with transcutaneous electrical nerve stimulation (TENS) for pain relief in Nigeria, Clin. J. Pain, 3 (1987) 142-144.

Waddell, G., A new clinical model for the treatment of low-back pain, Spine, 12 (1987) 632-644.