Physiotherapy 95 (2009) 185–191

Transcutaneous electrical nerve stimulation and transcutaneous

spinal electroanalgesia: A preliminary efficacy and
mechanisms-based investigation
Shea Palmer ∗ , Fiona Cramp, Kate Propert, Helen Godfrey
School of Health and Social Care, Faculty of Health and Life Sciences, University of the West of England,
Blackberry Hill, Bristol BS16 1DD, UK

Abstract
Objectives To determine the effects of transcutaneous electrical nerve stimulation (TENS) and transcutaneous spinal electroanalgesia (TSE)
on mechanical pain threshold (MPT) and vibration threshold (VT).
Design A prospective, single-blind, randomised, placebo-controlled trial.
Setting Laboratory based.
Participants Thirty-four healthy volunteers (12 men and 22 women; mean age ± standard deviation 30 ± 8 years). Exclusion criteria were
conditions affecting upper limb sensation and contraindications to electrical stimulation.
Interventions Participants were allocated at random to receive TENS (n = 8), TSE (n = 8), placebo (n = 9) or control (n = 9). Electrical
stimulation was applied for 30 minutes (from time 18 minutes to 48 minutes) via electrodes (5 cm × 5 cm) placed centrally above and below
the space between the C6 and C7 spinous processes, with 5 cm between electrodes.
Main outcome measures MPT (using an algometer) and VT (using a vibrameter) were recorded on seven occasions from the first dorsal
interosseous muscle of the right hand – at baseline (0 minutes) and then at 10-minute intervals until the end of the 60-minute testing period.
Results There were no statistically significant group differences in MPT (all p > 0.05). Significant group differences in VT were found at
20, 30 and 40 minutes (all p < 0.05). Post-hoc tests showed that the TENS group had significantly greater VT than both the placebo [median
difference 0.30 ␮m, 95% confidence interval (CI) −0.05 to 0.66] and control (0.51 ␮m, 95% CI 0.05 to 0.97) groups at 20 minutes, and
significantly greater VT than the control group (0.69 ␮m, 95% CI 0.20 to 1.17) at 30 minutes (all p < 0.008).
Conclusions Electrical stimulation did not alter MPT. The increase in VT during TENS may be due to distraction or antidromic block of
large-diameter nerve fibres. TSE failed to alter either outcome measure significantly.
© 2009 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved.

Keywords: TENS; TSE; Mechanical pain threshold; Vibration threshold; Healthy

Introduction of TSE (Bioinduction Limited, Bristol, UK) claim that it is

Different forms of electrical stimulation are commonly
used for the relief of a range of painful conditions [1],
although evidence to support their effectiveness is equivo-
cal. The most commonly used modality is transcutaneous
electrical nerve stimulation (TENS), but a newer form of
electrical stimulation, transcutaneous spinal electroanalgesia
(TSE), was developed in the UK in 1991. The manufacturers
∗ Corresponding author. Tel.: +44 117 3288919; fax: +44 117 3288437.
E-mail address: Shea.Palmer@uwe.ac.uk (S. Palmer).
more comfortable and therefore better tolerated than other
forms of electrical stimulation [2]. To date, only two pub-
lished research studies have directly compared the effects
of TENS and TSE [3,4], with conflicting findings. Macdon-
ald and Coates [3] carried out a small crossover study and
suggested that TSE was more effective than TENS for a het-
erogeneous group of patients with a range of chronic pain
conditions. Robb et al. [4] found that neither TENS nor TSE
was more effective than placebo stimulation for chronic pain
associated with breast cancer treatment. Further empirical
evidence of the relative effects of these modalities is clearly
required.

0031-9406/$ – see front matter © 2009 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.physio.2009.04.008
186 S. Palmer et al. / Physiotherapy 95 (2009) 185–191
There are numerous theories upon which electrical stim-
ulation is purported to relieve pain, but the main one is
the ‘pain-gate’ theory. This theory describes how activity
in large-diameter sensory nerve fibres (A␤ fibres) inhibits
the activity of cells in the spinal cord that normally trans-
mit nociceptive messages onwards to the brain [5]. Electrical
stimulation devices are specifically designed to stimulate
these large-diameter A␤ fibres. Any treatment that enhances
the sensitivity of A␤ fibres may therefore enhance the effec-
tiveness of the pain gate in reducing pain.
Specialist assessment known as quantitative sensory test-
ing can use touch, pressure or vibration stimuli of known
magnitudes to determine the sensitivity of A␤ fibres and
whether electrical stimulation (or other treatments) can alter
that sensitivity. A previous pilot study on healthy participants
[6] assessed the effects of interferential current on the per-
ception of activity within A␤ fibres by assessing the vibration
threshold (VT). Although the study failed to determine sig-
nificant differences, there was a very distinct trend towards
interferential current sensitising participants to vibration (i.e.
VT was reduced) compared with a control situation (n = 7 in
each group). It was theorised that electrical stimulation may
therefore act to enhance the efficiency of the pain gate by sen-
sitising large-diameter nerve pathways [6]. This observation,
however, requires further verification. No other published
studies have been found investigating the effects of electrical
stimulation on VT.
Quantitative sensory testing may also be used to deter-
mine sensitivity to a range of painful stimuli such as heat
or pressure. Such techniques can thus be used to investi-
gate whether electrical stimulation can desensitise people to
those painful stimuli. The mechanical pain threshold (MPT)
has often been used for this purpose, and there is relatively
consistent evidence that TENS is effective in raising MPT
[7–13]. McManus et al. [14] also found that interferential
current raised MPT significantly during stimulation. How-
ever, Alves-Geurreiro et al. [15] found no significant effects
of TENS, interferential current or action potential stimulation
on MPT.
To date, there has been little empirical investigation of
the analgesic efficacy of TSE. Simpson and Ward [16] found
no effects of TSE on the pain experienced by eight patients
with chronic critical limb ischaemia. Robb et al. [4] also
found no effect of TSE or TENS when compared with
placebo stimulation on pain associated with breast cancer
treatment. Thompson et al. [17] found no effects of active
TSE over placebo stimulation in low back pain. Some other
TSE studies have used MPT as an outcome measure. Mac-
donald and Coates [3] reported that the ratio of MPT in
tender areas relative to non-tender areas was reduced in a
subgroup of 16 patients with unilateral tenderness follow-
ing 60 minutes of stimulation with TSE. In another study
reported in the same paper [3], eight pain patients received
TSE or TENS in a double-blind randomised crossover design.
Six efficacy measures, including MPT, were all signifi-
cantly increased in the TSE group compared with the TENS
group. Towell et al. [18] found that TSE lowered mechan-
ical pain tolerance significantly (i.e. the opposite of what
would be expected with hypoalgesia) when high-intensity
TSE (250 V, n = 20) was applied across the spine in healthy
participants compared with low-intensity TSE (3–4 V, n = 20)
or sham stimulation (n = 20). When TSE was applied over
the shoulder joint at a high intensity (n = 10), there were
no changes in MPT or mood relative to sham stimulation
(n = 10), suggesting that the effects were specific to spinal
application.
To date, most experimental studies have focused either
on determining the specific physiological mechanisms acti-
vated by electrical stimulation or, more pragmatically, on its
efficacy (i.e. whether or not it reduces pain). Few studies
have attempted to ascertain both efficacy and mechanisms of
action within the same methodological design. This means
that the relative contribution of different analgesic mecha-
nisms is unknown. For example, although increased activity
in large-diameter sensory nerve fibres relative to pain fibres
in the spinal cord (the ‘pain-gate’ theory) predominates as
an explanation for pain relief when using electrical stimu-
lation, pain may also be relieved through other mechanisms
such as the role of the brain in suppressing pain perception
or the body’s own endogenous analgesics. For this reason,
it is important to investigate both efficacy and mechanisms
within the same experimental protocol.
This study therefore investigated the effects of TENS and
TSE on both MPT (as an indicator of efficacy) and VT (as an
indicator of the mechanism of action). Control and placebo
interventions were also incorporated into the experimental
design.
Methods
Ethical approval for this study was granted by the Uni-
versity of the West of England Bristol, School of Health and
Social Care Ethics Sub-Committee.
Recruitment
Thirty-four healthy volunteers (12 men and 22 women;
mean age ± standard deviation 30 ± 8 years) were recruited
from the student and staff population of the Health and Life
Sciences Faculty following advertisements. Those interested
in taking part contacted the lead author (SP) by e-mail, and an
information sheet and consent form were sent electronically.
Those replying to confirm their willingness to volunteer were
then given a convenient appointment time. As part of the
informed consent process, participants were told that they
might receive electrical stimulation which may or may not
be working and which they may or may not feel.
Inclusion criteria
Men and women aged 18–45 years.
 S. Palmer et al. / Physiotherapy 95 (2009) 185–191
Exclusion criteria
Any condition of the upper limb that would alter the sensi-
tivity to electrical stimulation, vibration or mechanical pain,
such as neurological impairment or pain. Contraindications to
electrical stimulation were as follows, based on the Chartered
Society of Physiotherapy (CSP) guidelines [19]: diagnosed
with cancer in the neck area; recent bleeding in the muscles
or other tissues in the neck area; and active tuberculosis in
the neck area. Participants were also asked to discuss any
of the following precautions to electrical stimulation with
the research assistant on an individual basis, based on the
CSP guidelines [19]: epilepsy or advanced cardiovascular
conditions, e.g. severe angina or cardiac arrhythmias; signifi-
cant impairment in the circulation or sensory loss in the neck
area; devitalised skin in the neck area, e.g. after recent radio-
therapy; and local acute skin condition in the neck area, e.g.
eczema, dermatitis. One potential participant was excluded
due to a benign meningioma at the base of the skull. None
of the other volunteers were excluded on the basis of these
criteria.
Randomisation
Following informed consent, participants were randomly
and exclusively assigned to one of four groups: (1) TENS –
125 ␮s, 100 Hz, ‘strong but comfortable’; (2) TSE – 1.5 ␮s
biphasic pulse, 2500 Hz, ‘comfortable’; (3) placebo – par-
ticipants were told that they would receive a form of TSE
that you cannot feel but the device was not switched on;
and (4) control – electrodes were applied but participants
were told that they would not receive any electrical stimula-
tion. Block randomisation using sealed envelopes was used
to ensure approximately equal numbers of participants in
each group. Separate envelopes were prepared for men and
women to ensure a balance of genders. The envelopes con-
tained the codes ‘W’, ‘X’, ‘Y’ or ‘Z’ randomised in blocks
of four and were prepared by the lead author (SP) using an
online randomisation tool (http://www.randomization.com,
accessed September 2007). Envelopes were opened by the
research associate employed on the project (KP), who
allocated a treatment to each of the codes on a ran-
dom basis and used this allocation throughout the trial.
All other members of the research team remained blinded
to the identity of groups until data analysis was com-
plete.
Participant positioning
Participants sat in a comfortable chair with their back sup-
ported. They were asked in advance to wear clothing with
a loose neckline so that the lower neck area could be easily
accessed (a suitable loose shirt was made available if they for-
got). The right hand was supported on a table placed to their
side. The hand and forearm were placed in a pronated position
for VT and MPT testing, but participants were free to place
their arm in any comfortable position at other times. A screen
187
was used to block the participants’ view of the Vibrameter
and electrical stimulator.
Electrical stimulation
All groups had two self-adhesive electrodes (5 cm × 5 cm)
positioned centrally (in the midline) above and below the
space between the C6 and C7 spinous processes, with 5 cm
between electrodes. This aimed to stimulate the spinal cord
(the main consideration for TSE) and the nerve roots sup-
plying the lateral aspect of the arm and thenar eminence
(a common way of applying TENS). It should be acknowl-
edged that the central electrode placement used in the present
investigation altered slightly from the paravertebral place-
ment recommended for arm pain by the manufacturers [2]. A
central electrode placement is recommended for TSE treat-
ment of other pain [2], however, implying effectiveness in
stimulating the spinal cord. The scientific rationale for recom-
mending a paravertebral placement for arm pain is therefore
unclear. Paravertebral arrangement of electrodes was found in
pilot work to cause uncomfortable muscle stimulation dur-
ing TENS, and therefore a central placement was used to
facilitate comparison between the two devices.
All forms of electrical stimulation were applied using
the Acticare IC device (Bioinduction Ltd., Bristol, UK).
Each type of stimulation was applied for 30 minutes as rec-
ommended for first-time TSE users [2], beginning at time
18 minutes and ending at time 48 minutes.
VT and MPT were recorded on seven occasions from a
point marked on the skin over the belly of the first dorsal
interosseous muscle of the right hand – at baseline (0 minutes)
and then at 10-minute intervals until the end of the 60-minute
testing period.
Vibration threshold
VT was measured using the Vibrameter IV (Somedic Pro-
duction AB, Sweden). The Vibrameter probe was placed in
contact with the skin and the reading was adjusted to account
for the weight of the probe. The vibration amplitude of the
probe was increased slowly until the participant stated that it
could be felt. It was then decreased slowly until the partici-
pant stated that the vibration disappeared. A flap was used to
cover the amplitude reading until the appearance and disap-
pearance thresholds were reached; at this point, the research
associate removed the flap and took a reading in micrometres
to two decimal places. This was repeated three times at each
time point. VT was later calculated as the mean of the appear-
ance and disappearance thresholds at each time point, and this
value was used for all data analyses. The manufacturer reports
measurement accuracy of ±5% over the full scale.
Mechanical pain threshold
MPT was measured using the Algometer II (Somedic Pro-
duction AB, Sweden). A 1 cm2 -probe was used and this was
188 S. Palmer et al. / Physiotherapy 95 (2009) 185–191
Table 1
Participant characteristics and baseline vibration threshold (VT) and mechanical pain threshold (MPT) scores. The 25th and 75th percentiles describe the lower
and upper limits of the range across which the middle 50% of values lie.
TENS
Male:female 3:5
Median age (years) (25th to 75th percentiles) 32.5 (27.5 to 37.8)
Median baseline VT (␮m) (25th to 75th percentiles) 0.88 (0.66 to 1.01)
Median baseline MPT (kPa) (25th to 75th percentiles) 204 (167 to 310)
TENS, transcutaneous electrical nerve stimulation; TSE, transcutaneous spinal electroanalgesia.
placed at right angles to the skin surface. The manual pres-
sure applied to the skin was increased slowly at a rate of
10 kPa/second using the built-in slope indicator. The pres-
sure was released as soon as the participant reported that the
pressure reached the ‘very first sensation of pain’. A flap was
used to cover the pressure reading on the device until MPT
was reached. At this point, the research associate removed
the flap and took a reading in kilopascals. This procedure
was repeated three times and the mean was taken as the MPT
at each time point. The manufacturer reports measurement
accuracy of ±3% over the full scale.
Success of blinding
As recommended by Deyo et al. [20], participants who
received TENS, TSE or sham electrical stimulation were
asked at the end of the trial whether they thought they had
used an active or inactive device.
Sample size
The sample size for this preliminary study was based on
previous studies where significant changes in MPT were
found using eight participants in each group [11,12]. This
was therefore used as a minimum recruitment target for the
present study, although recruitment continued throughout the
1-month period available for data collection. To account for
the possibility that such participant numbers were insuffi-
cient to detect significant changes, the correlations between
changes in MPT and VT were explored to inform the dis-
cussion. The study aimed to provide important preliminary
data to inform future studies, and a subsequent sample size
calculation was performed.
Data analysis
All data analyses were conducted by the lead author (SP)
who remained blind to the identity of specific groups until the
analyses were complete. Appropriate non-parametric anal-
yses were employed throughout the study. A significance
level of α = 0.05 was selected, and Bonferroni corrections
for multiple testing were applied where appropriate.
Age differences between groups were explored using a
Kruskal–Wallis test and separate Mann–Whitney U-tests.
VT and MPT data were expressed as change from base-
line, and all statistical analyses were conducted using these
TSE Placebo Control
3:5 3:6 3:6
20.0 (19.0 to 32.8) 27. 0 (20.0 to 35.5) 38.0 (29.5 to 43.0)
1.19 (0.42 to 1.92) 0.80 (0.66 to 1.12) 0.89 (0.44 to 1.18)
154 (118 to 175) 163 (146 to 243) 131 (120 to 191)
change data. Differences in VT and MPT were compared
between groups at each assessment point using separate
Kruskal–Wallis tests. Any significant results from this anal-
ysis were explored further using Mann–Whitney U-tests to
determine any significant differences between groups. The
relationships between change in VT and change in MPT were
explored using Pearson’s rank order correlation coefficients.
Results
Participant characteristics and baseline values for VT and
MPT are presented in Table 1. The median changes in VT
and MPT from baseline for each of the experimental groups
are presented in Figs. 1 and 2, respectively.
Overall, there was a significant difference in age between
groups (Kruskal–Wallis test, p = 0.029). Further analysis
using Mann–Whitney U-tests (with Bonferroni corrections
for multiple testing, α = 0.008) failed to identify signifi-
cant differences in age between individual group pairs (all
p ≥ 0.008), although the TSE and control group comparison
fell on the α cut-off level (p = 0.008). There were no signif-
icant differences in raw VT or MPT values between groups
at baseline (Kruskal–Wallis tests, both p > 0.05).
Kruskal–Wallis tests indicated that there were significant
differences in VT between experimental groups at 20, 30 and
40 minutes (all p < 0.05). There were no significant differ-
Fig. 1. Median change in vibration threshold (VT). All data are expressed
as a change from the baseline scores. The shaded area represents when elec-
trical stimulation was switched on (time 18 minutes to 48 minutes). TENS,
transcutaneous electrical nerve stimulation; TSE, transcutaneous spinal elec-
troanalgesia. *Statistically significant relative to control and placebo groups.
† Statistically significant relative to control group.
 S. Palmer et al. / Physiotherapy 95 (2009) 185–191
Fig. 2. Median change in mechanical pain threshold (MPT). All data are
expressed as a change from the baseline scores. The shaded area represents
when electrical stimulation was switched on (time 18 minutes to 48 minutes).
TENS, transcutaneous electrical nerve stimulation; TSE, transcutaneous
spinal electroanalgesia.
ences in VT at the other assessment points. Analysis of the
differences in VT between groups at 20, 30 and 40 minutes
was conducted using Mann–Whitney U-tests (with Bonfer-
roni corrections for multiple testing, α = 0.008). The TENS
group displayed significantly greater thresholds than both the
placebo [median difference 0.30 ␮m, 95% confidence inter-
val (CI) −0.05 to 0.66] and control (0.51 ␮m, 95% CI 0.05 to
0.97) groups at 20 minutes, and a significantly greater thresh-
old than the control group (0.69 ␮m, 95% CI 0.20 to 1.17)
at 30 minutes (all p < 0.008). There were no significant dif-
ferences between groups at the 40-minute assessment point
once the Bonferroni correction had been applied. There were
also no significant differences in VT between any other group
combinations.
There were no significant differences in MPT between
groups at any assessment point.
There were no significant relationships between changes
in VT and changes in MPT at any assessment point (Spear-
man’s rank order correlation coefficient range 0.036 to
−0.265, all p > 0.05).
Overall, there was a significant increase in median ambi-
ent temperature from 21.2 ◦ C before testing to 22.2 ◦ C after
testing (Wilcoxon signed ranks test, p < 0.001), although
there were no significant differences in ambient temperature
between groups before or after testing (Kruskal–Wallis tests,
both p > 0.05).
Four participants were unable to correctly identify whether
they had used an active or inactive device. Three of the four
guessed incorrectly and one was unable to guess, even when
pressed to do so. All four participants were in the placebo
group.
A sample size calculation was performed on the basis of
the mean ± standard deviation of the baseline TENS data
(234 ± 115 kPa), using 80% power, α = 0.05 and a 38%
increase in MPT [12]. This found that a minimum of 28 par-
ticipants would be required to detect a significant change of
that magnitude.
189
Discussion
Neither form of electrical stimulation altered MPT; this is
in contrast to previous laboratory studies on healthy partici-
pants which have demonstrated effects of TENS with similar
small participant numbers of 10 or less in each group [11–13].
Another trial with participant numbers of 10 in each group
failed to show significant changes in MPT with TENS [15].
Other studies have used in excess of 30 participants per group
[7,9,10], and all have demonstrated effects of TENS on MPT.
The sample size calculation suggested that the present study
had insufficient statistical power, although there were no clear
trends towards analgesic efficacy that might support such
a conclusion. Future reproduction of these findings with a
minimum participant number of 28 in each group would be
necessary to discount a type II error.
The parameters of the ‘high-frequency, low-intensity’
TENS groups in previous laboratory studies [7,9–13,15] are
very similar, with frequencies of 100–150 Hz, phase dura-
tions of 125–200 ␮s and an intensity described as ‘strong but
comfortable’. This compares very favourably with the present
study which used TENS settings of 100 Hz, 125 ␮s and a
‘strong but comfortable’ intensity. A number of studies have
found that such TENS settings were effective in increasing
MPT when electrodes were placed over the superficial radial
nerve in the forearm, and MPT measurements were taken
from the first dorsal web space [7,9,11–13]. Only two studies
[10,15] found that such TENS settings were ineffective with
a similar methodology.
TENS electrode placement represents an important differ-
ence between the current study and previous investigations.
MPT was measured in the same place in all investigations
but, rather than being placed over the superficial radial nerve,
electrodes were placed over the spinal cord in the present
experiment as this is recommended for TSE application [2].
Standardising electrode placement was considered necessary
in order to control a potentially important confounding vari-
able. Guidance for TSE electrode placement [2] is more
prescriptive than for TENS and therefore influenced choice
in this investigation. This may have reduced the effective-
ness of TENS. The complex interaction of different TENS
parameters, however, complicates attempts to assign causal-
ity to the effects of any single factor. Chesterton et al. [9], for
example, found that low-frequency, high-intensity, extraseg-
mental stimulation produced longer lasting effects on MPT
than high-frequency, low-intensity, segmental stimulation.
Chesterton et al. [10], however, found that high-intensity,
high-frequency, segmental stimulation (or a mixture of seg-
mental and extrasegmental) was most effective in altering
MPT. The effects of different combinations of parameters
and modes of application are therefore very complicated and
require further work.
TSE failed to alter either outcome measure significantly
in the current investigation. Previous literature also failed to
show beneficial effects of TSE on low back pain [17], pain
associated with breast cancer treatment [4], chronic critical
190 S. Palmer et al. / Physiotherapy 95 (2009) 185–191
limb ischaemia pain [16], or MPT or mechanical pain toler-
ance in healthy participants [18]. Only Macdonald and Coates
[3] have demonstrated positive effects of TSE in a series of
uncontrolled studies in a heterogeneous group of chronic pain
patients. On balance, the literature to date does not seem to
favour the analgesic effectiveness of TSE.
The increase in VT (indicating reduced perception of
activity within large-diameter nerve fibres) during TENS sug-
gests that, in the absence of analgesic effects on MPT, it
may have acted as a distraction. It was expected that VT
would be reduced during electrical stimulation, as suggested
in a previous pilot study using interferential current [6]. As
already discussed in relation to MPT, electrode placement
may have altered the effects of electrical stimulation on VT.
In contrast to the present study, Palmer and Macmillan [6]
placed electrodes on the forearm and measured VT between
the stimulating electrodes. There may also be a difference in
the effects of different modalities (TENS, TSE and inter-
ferential current). Antidromic activation of large-diameter
nerve fibres, extinguishing afferent A␤-fibre impulses [21],
is another possible mechanism explaining the reduction in
sensitivity to vibration observed during TENS. Such factors
require further exploration.
Walsh et al. [12] explored the relationship between
changes in negative peak latency and MPT, finding a sig-
nificant association (r = 0.90). A similar method of analysis
was used in the present investigation, but no significant cor-
relation was found between changes in VT and MPT. This
suggests that changes in the perception of A␤-fibre activity
were not associated with changes in pressure pain perception.
Future investigations of such mechanisms using alternative
TENS electrode placements (as discussed earlier) would be
interesting.
Partial success in blinding participants to treatment alloca-
tions was evident in this study. The only participants unable to
correctly identify whether they had used an active or inactive
device were in the placebo group. Five participants correctly
thought that they had used an inactive machine, three thought
that they had used an active machine and one was unable to
guess. All of those participants receiving active TENS or TSE
were convinced that they had used an active machine, pre-
sumably due to the accompanying sensations. The absence
of sensation in the placebo stimulation group introduced an
element of doubt, but it was far from a convincing placebo.
This supports previous observations on the problems inherent
in designing placebo-controlled trials of physical treatments
[20].
There was a small median increase of 1.0 ◦ C in the mean
ambient temperature during testing, but this was the same
for all participants. Any effect of temperature on nerve con-
duction is therefore unlikely to explain differences between
groups. Overall, there was a significant difference in age
between groups, although the practical significance of this
is likely to be negligible as neurophysiological differences
would not be expected within the relatively narrow age range
of the study participants.
These findings relate to healthy participants in a controlled
laboratory setting. Such experimental designs allow for care-
ful control of extraneous variables and thus are well suited
to investigations exploring specific mechanisms of action
or the effects of manipulating individual treatment param-
eters. However, appropriate care should be taken in direct
application of these results to the treatment of clinical pain
conditions.
Conclusion
TENS and TSE failed to alter MPT relative to a control
condition in the present investigation on healthy partici-
pants. This suggests a lack of analgesic efficacy, although
the effects of these modalities on clinical pain conditions
are likely to be much more complex than demonstrated in
this laboratory setting. The application of TENS increased
VT during stimulation, with distraction of participants or
antidromic activation of large-diameter nerve fibres being
possible explanatory mechanisms. TSE failed to alter either
outcome measure significantly, supporting the majority of
previous literature.
It is likely that the spinal placement of electrodes was
suboptimal for the application of TENS, as previous inves-
tigations have demonstrated analgesic effects on mechanical
pain when placed directly over the peripheral nerve. Repe-
tition of this investigation with a different TENS electrode
placement is therefore warranted.
Acknowledgements
The assistance of Helen French is gratefully acknowl-
edged in developing the study protocol.
Ethical approval: School of Health and Social Care Research
Ethics Committee, University of the West of England, Bristol
(Ref. No. HSC/07/09/62).
Funding: School of Health and Social Care, University of the
West of England, Bristol.
Conflict of interest: None.
References
[1] Pope GD, Mockett SP, Wright JP. A survey of electrotherapeutic
modalities: ownership and use in the NHS in England. Physiotherapy
1995;81:82–91.
[2] Bioinduction Limited. Acticare IC Interferential, TENS and TSE elec-
trotherapy system users manual. Bristol: Bioinduction Limited; 2005.
Available at: http://www.acticare.com [last accessed 26.02.2008].
[3] Macdonald AJR, Coates TW. The discovery of transcutaneous
spinal electroanalgesia and its relief of chronic pain. Physiotherapy
1995;81:653–61.
[4] Robb KA, Newham DJ, Williams JE. Transcutaneous electrical nerve
stimulation vs. transcutaneous spinal electroanalgesia for chronic pain
associated with breast cancer treatments. J Pain Symptom Manage
2007;33:410–9.
 S. Palmer et al. / Physiotherapy 95 (2009) 185–191
[5] Melzack R, Wall PD. Pain mechanisms: a new theory. Science
1965;150:971–9.
[6] Palmer S, Macmillan L. Effects of electrical stimulation on a mea-
sure of A-beta fibre sensitivity. In: 10th World Congress on Pain. 2002
[Abstract ID 426-P60].
[7] Aarskog R, Johnson MI, Demmink JH, Lofthus A, Iversen V, Lopes-
Martins R, et al. Is mechanical pain threshold after transcutaneous
electrical nerve stimulation (TENS) increased locally and unilaterally?
A randomized placebo controlled trial in healthy subjects. Physiother
Res Int 2007;12:251–63.
[8] Başkurt Z, Başkurt F, Özcan A, Yilmaz Ö. The immediate effects of
heat and TENS on pressure pain threshold and pain intensity in patients
with stage I shoulder impingement syndrome. Pain Clin 2006;18:
81–5.
[9] Chesterton LS, Barlas P, Foster NE, Lundeberg T, Wright CC, Baxter
GD. Sensory stimulation (TENS): effects of parameter manipula-
tion on mechanical pain thresholds in healthy human subjects. Pain
2002;99:253–62.
[10] Chesterton LS, Foster NE, Wright CC, Baxter GD, Barlas P. Effect
of TENS frequency, intensity and stimulation site parameter manipu-
lation on mechanical pain thresholds in healthy human subjects. Pain
2003;106:73–80.
[11] McDowell BC, McCormack K, Walsh DM, Baxter DG, Allen JM.
Comparative analgesic effects of H-wave therapy and transcutaneous
electrical nerve stimulation on pain threshold in humans. Arch Phys
Med Rehabil 1999;80:1001–4.
[12] Walsh DM, Foster NE, Baxter GD, Allen JM. Transcutaneous electri-
cal nerve stimulation (TENS): relevance of stimulation parameters to
neurophysiological and hypoalgesic effects. Am J Phys Med Rehabil
1995;74:199–206.
Available online at www.sciencedirect.com
191
[13] Walsh DM, Lowe AS, McCormack K, Willer J-C, Baxter GD, Allen
JM. Transcutaneous electrical nerve stimulation: effect on peripheral
nerve conduction, mechanical pain threshold, and tactile threshold in
humans. Arch Phys Med Rehabil 1998;79:1051–8.
[14] McManus FJ, Ward AR, Robertson VJ. The analgesic effects of interfer-
ential therapy on two experimental pain models: cold and mechanically
induced pain. Physiotherapy 2006;92:95–102.
[15] Alves-Geurreiro J, Noble JG, Lowe AS, Walsh DM. The effect of three
electrotherapeutic modalities upon peripheral nerve conduction and
mechanical pain threshold. Clin Physiol 2001;21:704–11.
[16] Simpson KH, Ward J. A randomized, double-blind, crossover study
of the use of transcutaneous spinal electroanalgesia in patients with
pain from chronic critical limb ischemia. J Pain Symptom Manage
2004;28:511–6.
[17] Thompson JW, Bower S, Tyrer SP. A double blind randomised
controlled clinical trial on the effect of transcutaneous spinal elec-
troanalgesia (TSE) on low back pain. Eur J Pain 2008;12:371–7.
[18] Towell A, Williams D, Boyd S. High-frequency non-invasive stimula-
tion over the spine: effects on mood and mechanical pain thresholds in
normal subjects. Behav Neurol 1997;10:61–5.
[19] Chartered Society of Physiotherapy. Guidance for the clinical use of
electrophysical agents. London: Chartered Society of Physiotherapy;
2006.
[20] Deyo RA, Walsh NE, Schoenfeld LS, Ramamurthy S. Can trials of
physical treatments be blinded? The example of transcutaneous elec-
trical nerve stimulation for chronic pain. Am J Phys Med Rehabil
1990;69:6–10.
[21] Johnson MI. Transcutaneous electrical nerve stimulation (TENS). In:
Watson T, editor. Electrotherapy: evidence-based practice. 12th ed.
Edinburgh: Churchill Livingstone Elsevier; 2008. p. 253–96.