5
Transcutaneous Electric Nerve Stimulation: The Effect of
Intensity on Local and Distal Cutaneous Blood Flow
and Skin Temperature in Healthy Subjects
Fiona L. Cramp, DPhil, Gillian R. McCullough, BSc, Andrea S. Lowe, DPhil, Deirdre M. Walsh, DPhil
ABSTRACT. Cramp FL, McCullough GR, Lowe AS, Walsh
DM. Transcutaneous electric nerve stimulation: the effect of
intensity on local and distal cutaneous blood flow and skin
temperature in healthy subjects. Arch Phys Med Rehabil 2002;
83:5-9.
Objective: To determine what effect transcutaneous electric
nerve stimulation (TENS) intensity has on local and distal
cutaneous blood flow and skin temperature.
Design: Double-blind conditions.
Setting: University research laboratory.
Participants: Forty subjects (20 men, 20 women) randomly
assigned to 1 of 4 groups (10 per group): control, above-motor-
threshold TENS, below-motor-threshold TENS, or perception-
threshold TENS.
Intervention: TENS (4Hz, 200␮s) was applied over the
median nerve of the right forearm for 15 minutes.
Main Outcome Measures: Blood flow measured by laser
Doppler flowmeter and skin temperature measured by skin ther-
mistor were recorded during TENS and for 15 minutes after it.
Results: Significant differences occurred between groups for
forearm (P ⬍ .0001; repeated-measures analysis of variance)
but not fingertip cutaneous blood flow. Post hoc Fisher tests
showed a significant increase in forearm blood flow during
TENS application in the above-motor-threshold TENS group
compared with the other 3 groups. No significant differences
between groups for skin temperature data were observed.
Conclusions: The effect of TENS on cutaneous blood flow
depends on whether muscle activity is induced. Low-frequency
TENS applied above the motor threshold significantly in-
creases local cutaneous blood flow. There were no significant
differences between groups for skin temperature.
Key Words: Blood flow velocity; Laser Doppler flowmetry;
Perfusion; Rehabilitation; Skin temperature; Transcutaneous
electric nerve stimulation.
© 2002 by the American Congress of Rehabilitation Medi-
cine and the American Academy of Physical Medicine and
Rehabilitation
N RECENT YEARS, attention has been focused on the
Ioneffects of transcutaneous electric nerve stimulation (TENS)
the autonomic nervous system, specifically the effects of
electric stimulation on peripheral circulation by stimulation of
From the Rehabilitation Sciences Research Group, School of Rehabilitation Sci-
ences, University of Ulster at Jordanstown, County Antrim, Northern Ireland.
Accepted in revised form March 8, 2001.
Supported by the Department of Education for Northern Ireland.
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit upon the author(s) or upon any
organization with which the author(s) is/are associated.
Reprint requests to Fiona L. Cramp, DPhil, Rehabilitation Sciences Research
Group, School of Rehabilitation Sciences, University of Ulster at Jordanstown,
Newtownabbey, Co. Antrim, BT37 0QB, N. Ireland, e-mail: afl.cramp@ulst.ac.uk.
0003-9993/02/8301-6537$35.00/0
doi:10.1053/apmr.2002.27478
the sympathetic nerves.1 Previous work2 performed by the
Rehabilitation Sciences Research Group in Northern Ireland
has shown that low-frequency TENS (4Hz), when applied at a
“strong but comfortable” intensity to a peripheral nerve, sig-
nificantly increased local cutaneous blood flow compared with
a high-frequency TENS application (P ⫽ .011) or a control
group. However, it has been suggested that a subjective “strong
but comfortable” sensation is produced at a lower intensity
when using a high-frequency TENS application compared with
a low-frequency TENS application.3 It is therefore conceivable
that the increase in blood flow reported with low-frequency
TENS was caused by a higher intensity being applied in this
group. Therefore, to extend our previous work, we investigated
the effect of TENS on blood flow when applied at intensities
above and below motor threshold as well as at perception
threshold in the current study.
Research to date2 on the effect of TENS on cutaneous blood
flow has produced equivocal results, perhaps because of the
positioning of blood flow probes and thermistors.4 The previ-
ously reported2 increase in cutaneous blood flow with low-
frequency TENS was obtained with the laser Doppler blood
flowmeter probe positioned midway between the TENS elec-
trodes. It has been suggested that increased blood flow attrib-
uted to electric stimulation may only occur local to the muscles
being stimulated.5 Jänig6 indicates that any changes to fingertip
blood flow during TENS procedures may provide further evi-
dence of its effects on the sympathetic nervous system because
fingertip blood flow is controlled by this division of the nervous
system. This hypothesis is supported by the fact that arterio-
venous anastomoses are present in the skin of the hands and
form the main blood supply to the skin of the digital pads.
These anastomoses connect the smaller arteries and corre-
sponding veins and are under the influence of the sympathetic
nervous system.7
The aim of the present investigation was to examine how
TENS affects local and distal cutaneous blood flow and skin
temperature when applied at different intensities. We chose the
median nerve as the site of a low-frequency (4Hz) TENS
application because stimulation applied there increases local
cutaneous blood flow.2 Laser Doppler flowmetry was used to
provide continuous, noninvasive, and real-time measurement
of blood flow.8 This method has previously been used at this
research center to study the effects of various modalities on
cutaneous blood flow.2,9-11
METHODS
General
Forty healthy subjects (20 men, 20 women) were recruited
from the staff and students of the University of Ulster after
approval was obtained from the University’s Research Ethical
Committee. Subjects were randomly assigned to 1 of 4 groups
(10 each, equal numbers of men and women): control, above-
motor-threshold TENS, below-motor-threshold TENS, and
perception-threshold TENS.
Arch Phys Med Rehabil Vol 83, January 2002
6 EFFECT OF TENS INTENSITY ON BLOOD FLOW, Cramp
Preparation of Subjects
For the duration of the experiment, subjects lay supine on a
plinth with their right forearm supinated. The course of the
right median nerve was mapped as detailed previously.2 The
reference point for positioning the TENS electrodes and blood
flow and thermistor probes was taken as a point on the course
of the median nerve halfway between the elbow crease and
proximal wrist crease. The TENS electrodes were placed 4.5cm
on either side of this point with the cathode positioned proxi-
mal. A skin thermistor probe was placed 2cm medial to the
reference point and a laser Doppler blood flowmeter probe 2cm
lateral (fig 1). An additional laser Doppler blood flowmeter
probe was placed on the distal phalanx of the index finger. Skin
thermistor probes were also positioned on the tip of the distal
phalanx of the middle finger and the middle phalanx of the
index finger. Additional electrodes were also attached to the
forearm to monitor muscle activity. Preparation of the forearm
(⬃15min) allowed the subject to rest and acclimatize to the
environment before recording started. After the experimental
preparation, subjects in the TENS groups were given a dem-
onstration of TENS on their left forearm. Throughout the
experimental procedure, subjects were required to remain as
still as possible.
Experimental Procedure
After subject preparation, blood perfusion was monitored in
the right forearm. Study protocol required subjects to be ex-
cluded from the investigation if their blood flow did not reach
a steady state after 20 minutes; however, no subjects were
excluded. After the settling-down period, the TENS unit was
Fig 1. Placement of the TENS electrodes, the skin thermistor
probes, laser Doppler flowmeter (LDF) probes, and Ag/AgCl record-
ing electrodes.
Arch Phys Med Rehabil Vol 83, January 2002
activated and subjects were asked to report as soon as they felt
a tingling sensation; the intensity was then adjusted according
to group allocation as detailed below. At this point, TENS was
then applied for 15 minutes and monitored at 1, 3, 5, 7, 10, and
13 minutes. After the TENS application, subjects remained at
rest, and recordings of blood flow and temperature continued
for another 15 minutes. In the control group, the TENS device
was not activated, but the electrodes were attached as described
for the treatment groups.
Temperature Recording Procedure
Skin and ambient temperature were recorded at 1-minute
intervals throughout the experiment; data were excluded if
ambient temperature changes exceeded ⫾ 0.5°C. The skin
thermistorsa were positioned as previously described (fig 1),
and the ambient thermistor was attached to the wall of the
laboratory. All thermistors were connected to a centralized data
logger (1200 series Squirrel meter).a
Blood Perfusion Measurement
The method for measuring blood perfusion was described in
a previous study.10 In common with the findings of others,9 the
recorded signal showed relatively pronounced fluctuations
caused by vasomotion. In all experiments, blood perfusion was
monitored at 1-second intervals, and the recorded signals were
subsequently smoothed (Acknowledge programb: smoothing
function at 100%) to facilitate measurement of mean values.
The smoothing function is a transformation that computes a
series of data points and replaces each value with a mean value.
Double-blind conditions applied in the present study because
the subject was unaware of the treatment group to which he/she
had been assigned, and a second investigator, who was un-
aware of group allocation, compiled blood flow data from the
stored recordings.
TENS Procedure
A 120Z TENS unitc was used with self-adhesive electrodesd
(2-in diameter) for electric stimulation. The TENS unit was
calibrated at 4-Hz frequency and 200-␮s pulse duration before
the experiment using a digital recording oscilloscope.e
Muscle activity resulting from TENS application was mon-
itored in the flexor carpi radialis muscle, which is innervated by
the median nerve. This muscle was chosen because pilot work
showed that activity in flexor carpi radialis occurred before
activity in muscles distal to the site of stimulation, such as
abductor pollicis brevis. A motor response (M response) was
always obtained in the flexor carpi radialis muscle before a
visible muscle twitch occurred. Two Ag/AgCl recording elec-
trodes were placed 2cm apart with the active electrode over the
muscle belly and the reference electrode 2cm proximal (fig 1).
A ground electrode was placed lateral to the recording elec-
trodes at a point equidistant from them. Both recording elec-
trodes and the ground electrode were connected to the Mystro⫹
monitorf to observe muscle response. An output from the
TENS unit, connected to the monitor, acted as an external
trigger. For M-response recording, the following standard set-
tings were used: 50-ms sweep duration, frequency bandwidth
10Hz to 10kHz, sensitivity 500␮V/division. After each pulse
delivered by the TENS unit, muscle activity was recorded, and
an average of the recorded M responses was displayed on the
screen.
In the below-motor-threshold TENS group, stimulation in-
tensity was increased until an M response first appeared; the
intensity was then decreased until the M response was absent.
If subjects reported a decrease in the TENS sensation, the
 EFFECT OF TENS INTENSITY ON BLOOD FLOW, Cramp
Fig 2. Summary of forearm
skin blood perfusion differ-
ences plotted against time for
each of the experimental
groups (mean ⴞ SEM; n ⴝ 10
for all groups). Negative val-
ues represent a decrease and
positive values an increase in
skin blood perfusion from
baseline. Significant differ-
ences between groups: *P <
.01; †P < .0001. Legend: Œ,
above motor threshold; 䢇, be-
low motor threshold; f, per-
ception threshold; 䊐, control.
intensity was increased until the subject either reported that the
initial subjective sensation had been reached or that an M
response was obtained, whichever occurred first. In the latter
case, the intensity was decreased again to below the setting that
produced an M response.
In the above-motor-threshold TENS group, stimulation in-
tensity was increased until a visible muscle contraction was
always obtained with a corresponding M response. The stim-
ulation was then increased until the subject reported a “strong
but comfortable” sensation. If subjects reported a decrease in
TENS sensation, the intensity was increased until the initial
sensation was regained.
For the perception-threshold TENS group, stimulation inten-
sity was increased until the subject’s first report of any TENS
sensation beneath the electrodes. Although muscle activity was
also monitored in this group, we never obtained a motor
response for these subjects. If subjects reported a loss in
sensation, the intensity was increased until the first sensation of
TENS was restored.
Data Analysis
For the purposes of analysis, baseline blood perfusion values
were obtained 20 seconds before activation of the TENS unit in
the TENS groups; further readings were taken at 3-minute
intervals thereafter up to the 30-minute point. In the control
group, readings were taken at equivalent times. Corresponding
skin and ambient temperatures were also obtained at these time
intervals. Statistical analysis on all data was performed on
calculated difference scores (ie, variation from baseline) to
allow for variation in resting values between subjects. Two-
factor repeated-measures and 1-factor analysis of variance
(ANOVA) and post hoc Fisher tests where appropriate were
performed by using StatView® 4 softwareg on a personal
computer.
7
RESULTS
Blood Flow
Figure 2 summarizes forearm skin blood perfusion values as
difference scores (from time 0) for each of the 4 experimental
groups. The below-motor-threshold TENS, perception-thresh-
old TENS, and control values remained relatively unchanged
over the 30-minute recording period. In the above-motor-
threshold TENS group, blood perfusion rose over the initial 3
minutes and continued to increase to a maximum of 2.85 ⫾ .58
perfusion units (mean ⫾ standard error of the mean [SEM]) at
minute 12; after the 15-minute application of TENS, perfusion
values returned to baseline. Repeated-measures ANOVA per-
formed on these data showed a significant effect both between
groups (P ⬍ .0001) and over time (P ⬍ .0001); a significant
interactive effect (P ⬍ .0001) was also obtained. One-way
ANOVA and post hoc Fisher tests showed a significant in-
crease in the above-motor-threshold TENS group at 3, 6, 9, 12
(P ⬍ .0001), and 15 minutes (P ⫽ .002) when compared with
the other 3 groups at each of these time points.
Results for cutaneous blood flow recorded from the index
fingertip are summarized in figure 3. The figure shows an
overall decrease in perfusion values for all groups over the
30-minute time period. Repeated-measures ANOVA per-
formed on these data showed no significant effect between
groups (P ⫽ .793) and no interactive effect (P ⫽ .559).
However, a significant effect was observed over time (P ⬍
.0001).
Skin Temperature
Figure 4 plots forearm skin temperature differences against
time for each group. The figure shows that little variation
existed between groups throughout the 30-minute recording
period. Repeated-measures ANOVA showed a significant (P ⬍
Arch Phys Med Rehabil Vol 83, January 2002
8 EFFECT OF TENS INTENSITY ON BLOOD FLOW, Cramp

Fig 3. Summary of index fin-

ger skin blood perfusion dif-
ferences plotted against time
for each of the experimental
groups (mean ⴞ SEM; n ⴝ 10
for all groups). Negative val-
ues represent a decrease in
and positive values an in-
crease in skin blood perfusion
from baseline. Legend: Œ,
above motor threshold; 䢇, be-
low motor threshold; f, per-
ception threshold; 䊐, control.

.0001) change in skin temperature over time but no significant
effect between groups (P ⫽ .238) and no interactive effect
(P ⫽ .358).
Index finger and middle finger skin temperature values dis-
played a steady decrease over time for all groups. Repeated-
measures ANOVA performed on index finger skin temperature
difference scores showed a significant change over time (P ⬍
.0001) but no significant effect between groups (P ⫽ .237) and
no interactive effect (P ⫽ .902). Similarly, middle finger dif-
ference scores displayed a significant change over time (P ⬍
.0001) but no significant effect between groups (P ⫽ .317) and
no interactive effect (P ⫽ .884).
Ambient temperature data were similarly analyzed as differ-
ence scores. Repeated-measures ANOVA revealed no signifi-

Fig 4. Summary of forearm

skin temperature differences
plotted against time for each
of the experimental groups
(mean ⴞ SEM; n ⴝ 10 for all
groups). Negative values rep-
resent a decrease in and posi-
tive values an increase in
temperature from baseline.
Legend: Œ, above motor thresh-
old; 䢇, below motor threshold;
f, perception threshold; 䊐,
control.

Arch Phys Med Rehabil Vol 83, January 2002

 EFFECT OF TENS INTENSITY ON BLOOD FLOW, Cramp
cant differences between groups (P ⫽ .209) and no interactive
effect (P ⫽ .916), although there was a significant change over
time (P ⬍ .0001). Ambient temperature ⫾ SEM increased from
23.38° ⫾ 0.16°C at baseline to 23.45° ⫾ 0.16°C at minute 30.
The largest variation in ambient temperature in any individual
experiment was 0.3°C.
DISCUSSION
Results of the present study showed that local cutaneous
blood flow increased significantly during the application of
low-frequency TENS over the median nerve at an intensity
above motor threshold, compared with TENS applied below
motor threshold, at perception threshold, or not at all (control
group). However, we found no significant differences between
groups for skin temperature or distal cutaneous blood flow. It
has previously been suggested that skin temperature does not
accurately reflect changes in skin blood flow as detected using
laser Doppler flowmetry,12 which may account for the lack of
consistency previously reported.2
Research to date suggests that TENS affects sympathetic
nervous system activity, although results remain equivocal,
with some authors claiming that TENS increases sympathetic
activity13,14 and others reporting a decrease in activity because
of TENS.1,15,16 Jänig6 suggested that fingertip blood flow was
controlled entirely by the sympathetic nervous system; in the
current study, low-frequency TENS applied over the median
nerve had no significant effect on fingertip blood flow, which
may suggest that it had no influence on the sympathetic ner-
vous system. This hypothesis is supported by the finding that
an intensity above motor threshold was required for TENS to
produce an increase in cutaneous blood flow. The increased
cutaneous blood flow that occurs during TENS is caused either
by a “muscle pump” action5 or by increased metabolic demand
of the contracting muscle.17 The rapid increase and decrease in
cutaneous blood flow corresponding to the onset and offset of
TENS in the present study would indicate that the muscle pump
action is the most likely mechanism. The present results also
support the suggestion by Mohr et al5 that increased blood flow
from electric stimulation is localized and may be limited to the
muscles being stimulated.
CONCLUSION
The effect of low-frequency TENS on cutaneous blood flow
is dependent on whether stimulation is applied at an intensity
above or below motor threshold. Any increase in cutaneous
blood flow produced by TENS appears to be local in nature,
when compared with control group, and does not appear to
occur because of depression of the sympathetic nervous sys-
tem.
References
1. Hollman JE, Morgan BJ. Effect of transcutaneous electrical nerve
stimulation on the pressor response to static handgrip exercise.
Phys Ther 1997;77:28-36.
9
2. Cramp AF, Gilsenan C, Lowe AS, Walsh DM. The effect of high-
and low-frequency transcutaneous electrical nerve stimulation
upon cutaneous blood flow and skin temperature in healthy sub-
jects. Clin Physiol 2000;2:150-7.
3. Burton C, Maurer DD. Pain suppression by transcutaneous elec-
tronic stimulation. IEEE Trans Biomed Eng 1974;21:81-8.
4. Scudds RJ, Helewa A, Scudds RA. The effects of transcutaneous
electrical nerve stimulation on skin temperature in asymptomatic
subjects. Phys Ther 1995;75:621-8.
5. Mohr T, Akers TK, Wessman HC. Effect of high voltage stimu-
lation on blood flow in the rat hind limb. Phys Ther 1987;67:
526-33.
6. Jänig W. The sympathetic nervous system in pain: physiology and
pathology. In: Stanton-Hicks M, editor. Pain and the sympathetic
nervous system. Boston: Kluwer Academic; 1990. p 17-89.
7. Davies DV, Davies F, editors. Gray’s anatomy: descriptive and
applied. 33rd ed. Glasgow: University Pr; 1962. p 726.
8. Obeid AN, Barnett NJ, Dougherty G, Ward G. A critical review of
laser Doppler flowmetry. J Med Eng Technol 1990;14:178-81.
9. Lowe AS, Walsh DM, Baxter GD, Allen JM. Low-intensity laser
irradiation (830 nm) reduces skin blood flow in humans. Lasers
Med Sci 1995;10:245-51.
10. McDowell BC, McElduff C, Lowe AS, Walsh DM, Baxter GD.
The effect of high and low frequency H-wave therapy upon skin
blood perfusion: evidence of frequency specific effects. Clin
Physiol 1999;19:450-7.
11. Noble JG, Henderson G, Cramp AF, Walsh DM, Lowe AS. The
effect of interferential therapy upon cutaneous blood flow in
humans. Clin Physiol 2000;20:2-7.
12. Kurvers HA, Tangeder G-J, Demey JG, et al. Skin blood flow
abnormalities in a rat model of neuropathic pain: result of de-
creased vasoconstrictor outflow? J Auton Nerv Syst 1997;63:
19-29.
13. Wong RA, Jette DU. Changes in sympathetic tone associated with
different forms of transcutaneous electrical nerve stimulation in
healthy subjects. Phys Ther 1984;64:478-82.
14. Casale R, Gibellini R, Bozzi M, Bonelli S. Changes in sympa-
thetic activity during high-frequency TENS. Acupunct Electrother
Res 1985;10:169-75.
15. Kaada B. Vasodilation induced by transcutaneous nerve stimula-
tion in peripheral ischemia (Raynaud’s phenomenon and diabetic
polyneuropathy). Eur Heart J 1982;3:303-14.
16. Sanderson JE, Tomlinson B, Lau MS, et al. The effect of trans-
cutaneous electrical nerve stimulation (TENS) on autonomic car-
diovascular reflexes. Clin Auton Res 1995;5:81-4.
17. Tracy JE, Currier DP, Threlkeld AJ. Comparison of selected pulse
frequencies from two different electrical stimulators on blood flow
in healthy subjects. Phys Ther 1988;68;1526-32.
Suppliers
a. Grant Instruments Ltd, Barrington, Cambridge, CB2 5QZ, UK.
b. Acknowledge 881, version 2.1; Biopac Systems Inc, 42 Aero
Camino, Santa Barbara, CA 93117.
c. ITO Co Ltd, 3-3, Toyotama-Minami, Nerima-ku, Tokyo 176-8605,
Japan.
d. PALS electrodes; Nidd Valley Medical Ltd, Conyngham Hall,
Knaresborough, North Yorkshire, HG5 9AY, UK.
e. Model 1602; Gould Electronics Ltd, Instrument Systems, Roebuck
Rd, Hainault, Ilford, Essex, IG6 3UE, UK.
f. Medelec Ltd, Manor Way, Old Woking, Surrey, GU22 9JU, UK.
g. Abacus Concepts Inc, 1918 Bonita Ave, Berkeley, CA 94704-1014.
Arch Phys Med Rehabil Vol 83, January 2002