ORIGINAL ARTICLE

Transcutaneous Electrical Nerve

Stimulation for Phantom Pain and
Stump Pain in Adult Amputees

Matthew R. Mulvey, PhD*,†; Helen E. Radford, BSc†,‡; Helen J. Fawkner,

PhD*,†; Lynn Hirst, BSc‡; Vera Neumann, MD‡; Mark I. Johnson, PhD*,†
*Faculty of Health and Social Sciences, Leeds Metropolitan University, Leeds, U.K.;
†
Leeds Pallium Research Group; ‡Leeds Teaching Hospitals NHS Trust

n Abstract: Following amputation, 50% to 90% of indi- (P < 0.05) after 60 minutes of TENS. For five participants, it
viduals experience phantom and/or stump pain. Transcuta- was possible to project TENS sensation into the phantom
neous electrical nerve stimulation (TENS) may prove to be a limb by placing the electrodes over transected afferent
useful adjunct analgesic intervention, although a recent sys- nerves. Nonpainful phantom sensations and prosthesis
tematic review was unable to judge effectiveness owing to embodiment remained unchanged. This study has demon-
lack of quality evidence. The aim of this pilot study was to strated that TENS has potential for reducing phantom pain
gather data on the effect of TENS on phantom pain and and stump pain at rest and on movement. Projecting TENS
stump pain at rest and on movement. Ten individuals with sensation into the phantom limb might facilitate perceptual
a transtibial amputation and persistent moderate-to-severe embodiment of prosthetic limbs. The findings support the
phantom and/or stump pain were recruited. Inclusion crite- delivery of a feasibility trial. n
ria was a baseline pain score of ‡3 using 0 to 10 numerical
rating scale (NRS). TENS was applied for 60 minutes to gen- Key Words: transcutaneous electric nerve stimulation,
erate a strong but comfortable TENS sensation at the site of phantom limb pain, stump pain, analgesia, neuralgia,
stump pain or projected into the site of phantom pain. Out- stump, prosthesis embodiment
comes at rest and on movement before and during TENS at
30 minutes and 60 minutes were changes in the intensities
of pain, nonpainful phantom sensation, and prosthesis
embodiment. Mean (SD) pain intensity scores were reduced INTRODUCTION
by 1.8 (1.6) at rest (P < 0.05) and 3.9 (1.9) on movement
Over 77% of pain specialists use transcutaneous elec-
trical nerve stimulation (TENS) to manage chronic
Address correspondence and reprint requests to: Prof Mark I. John- pain, and over half of chronic pain patients given
son, PhD, Faculty of Health and Social Sciences, Leeds Metropolitan Uni-
versity, Civic Quarter, Leeds LS1 3HE, U.K. E-mail: m.johnson@leedsmet.
TENS report that they find it beneficial and wish to
ac.uk. continue using it.1 TENS is recommended as a non-
Place where work conducted: Seacroft Hospital, Leeds Teaching Hos-
pharmacological treatment for phantom pain and
pitals NHS Trust and The Centre for Pain Research, Leeds Metropolitan
University. stump pain,2–6 yet a recent Cochrane review conducted
Submitted: May 16, 2012; Accepted: June 27, 2012 by our team failed to find any randomised controlled
DOI. 10.1111/j.1533-2500.2012.00593.x
trails. We concluded that there was insufficient
evidence to judge TENS effectiveness in this
 2012 The Authors
Pain Practice  2012 World Institute of Pain, 1530-7085/12/$15.00
setting.7 Contralateral stimulation sites were used
Pain Practice, Volume 13, Issue 4, 2013 289–296 because of concerns that TENS may exacerbate pain in
290 • MULVEY ET AL.

individuals with tactile allodynia associated with Participants interested in taking part received an infor-
stump neuromas. In addition, there are concerns that mation pack and were contacted via telephone approx-
TENS may cause skin irritation which may affect imately 1 to 4 days later to be formally invited to take
stump integrity. Published case series have failed to part in the study.
report pain intensity data, and this has prevented the Inclusion criteria were average pain of at least 3 of
calculation of adequate sample sizes needed for a sub- 10 on a numerical rating scale (NRS) during the last
sequent clinical trial.8–10 They also failed to provide month and use of a prosthesis for at least 2 hours, 2
adequate information on TENS protocols, safety and times a week. Exclusion criteria were contraindications
tolerability, and satisfaction including potential barri- to TENS (noncomprehension of instructions, allergic
ers to use.7 Moreover, recent literature reviews have response to electrodes or conductive gel, dermatologi-
called for further investigations into efficacy and mech- cal lesions, epilepsy),20 and changes to opioid analgesic
anisms of nonpharmacological interventions for phan- medication within 3 days of study visit. Information
tom pain and stump pain, including TENS.11,12 regarding co-analgesics (such as anticonvulsants or an-
It is known that stimulating sensory afferents above tidepressants) was also collected but participants were
the stump can generate sensations of touch, joint move- not excluded if these had changed within 3 days prior
ment, and position in the phantom limbs of ampu- to study visit, as they have minimal immediate effects
tees.13–17 However, it is unclear whether it is possible on pain scores.
to target phantom pains by projecting TENS paresthe- The study visit lasted approximately 2 hours and
sia into the phantom limb, using electrodes placed over was facilitated by the Principal Investigator (MM).
peripheral nerves in the residual limb. If one can, it is Following written consent, the condition of the skin
uncertain whether this would reduce or exacerbate around the stump was noted for lesions. Participants
phantom limb pain. Moreover, we have previously selected a painful movement that occurred regularly
hypothesized that TENS could be used to facilitate per- during daily activity, for example, standing up from
ceptual embodiment of the prosthetic limb (the percep- sitting, standing from lying supine, walking, bending
tual awareness of the prosthesis in relation to the body) to pick up an object off the floor, or pulling on pros-
into the body schema,18 and initial experimental stud- thetic limb socket liner. The optimal TENS electrode
ies using able-bodied participants and a visuotactile placement positions were identified using a process of
illusion suggest that this is the case.19 Nevertheless, no trial and error which took no longer than 15 minutes.
clinical data on the use of TENS to aid the perceptual Two TENS electrodes (approximately 5 · 5 cm in size)
embodiment of prosthetic limbs are available to date. were placed at various sites above the distal end of the
The aim of this pilot study was to collect data on stump in an attempt to project TENS paresthesia to
pain intensity at rest and on movement (primary out- the most painful site, which was either in the stump,
come), perceptual embodiment of the prosthetic limb phantom limb, or both. Once the optimal location of
(at rest and on movement) and tolerability during a 1- TENS sensation was established, the TENS device was
hour in-clinic TENS intervention. In addition, data switched off until the investigator was ready to begin
were gathered on recruitment rates from a local NHS the TENS intervention. Participants were instructed
Amputee Rehabilitation outpatient Clinic and a Pain that they may need to adjust TENS amplitude to main-
Management outpatient clinic in West Yorkshire, U.K. tain a strong nonpainful TENS sensation during the
60 minutes intervention. TENS parameters were in line
with International Association for the Study of Pain
METHODS
(IASP) recommendations.21 Conventional TENS set-
This was a single-center study of 10 transtibial ampu- tings were chosen which were continuous pulse pat-
tee participants with phantom pain, stump pain, or tern, pulse duration of 80 ls, pulse frequency of
both. This study received ethical approval from the 100 Hz, strong but comfortable intensity.
National Research Ethics Service, Leeds (Central) During the TENS intervention, a Pro-TENS device
Research Ethics Committee (REC reference: 08/ (Nidd Valley Medical, Knaresbrough, U.K.) was
H1313/66). Participants were identified from local switched on and the amplitude of current increased
NHS Amputee Rehabilitation and Pain Management until strong nonpainful TENS sensations (electrical
Service databases (searched January to November paresthesia) were at the site of pain. TENS was
2009) and approached at their scheduled clinic review. switched off after 60 minutes, and a second stump
 TENS for Phantom Pain and Stump Pain • 291

observation was performed and any changes in skin the electrodes were placed on nonpainful skin either
condition noted. The following measures were taken at side of the affected area and TENS sensation was pro-
rest and on movement at baseline and at 30 and jected into the area of pain and allodynia. This method
60 minutes during the TENS intervention using an 11- of applying TENS did not exacerbate their pain.
point NRS: intensity of phantom pain and/or stump In 3 participants with phantom pain, it was possible
pain (0 = no pain, 10 = worst pain imaginable), inten- to elicit an evoked phantom sensation by palpating the
sity of nonpainful phantom limb sensation (0 = no sen- stump. By placing TENS electrodes at these sites, it
sations, 10 = strongest sensation imaginable), and was possible to project TENS sensation directly into
intensity of prosthetic limb awareness (0 = no aware- the phantom limb which participants reported to be
ness, 10 = strongest awareness imaginable). beneficial. It was not possible to elicit a referred phan-
A telephone follow-up 48 hours after the study visit tom sensation in the remaining 2 participants with
was used to gather information about adverse events phantom limb pain, although it was still possible to
and changes in concomitant medication. Intensity scores project TENS sensation directly into the phantom limb
for pain phantom sensation and prosthesis embodiment through a process of trial and error.
relative to baseline were calculated for each participant, All participants were able to wear their prosthesis
and a descriptive analysis was performed. and sit, stand, and walk with TENS electrodes in situ
during stimulation. During the 60-minute TENS inter-
vention, all participants remained seated, except when
RESULTS
undertaking the painful movement task.
Fifty-four individuals were invited to take part over an
8-month period. Ten of these individuals agreed to
Outcome Measurements at Rest and on Movement
participate (6 women; age range, 22 to 72 years; mean
age, 54; SD, 13.8) and all completed the study The mean baseline NRS for pain intensity at rest was
(Table 1). In the month prior to the study, all 10 par- below 3 of 10 for 4 participants (Table 2) despite these
ticipants experienced phantom limb pain, 9 experi- participants reporting an average pain of at least 3 of
enced phantom limb sensations, 9 experienced stump 10 on NRS during the last month. For one participant,
pain, and 2 had tactile allodynia on the stump. resting pain intensity was zero at baseline and
Transcutaneous electrical nerve stimulation was tol- remained at zero during TENS (Table 2). Resting pain
erated well by all participants. Participants reported no intensity decreased for the remaining 9 participants at
difficulty in using the TENS device or titrating TENS 30 minutes of TENS and continued to decrease for 8
amplitude to produce a strong nonpainful tingling sen- of these participants at 60 minutes of TENS
sation. No adverse events were reported during the (Figure 1A). Mean baseline NRS for pain intensity on
60-minute TENS session or at 48-hour telephone fol- movement was 3 or above for all participants
low-up. Post-TENS stump examination revealed slight (Table 2). Pain intensity on movement decreased from
reddening of the skin under the electrodes in two par- baseline for 9 participants at 30 minutes of TENS and
ticipants but this resolved within 5 minutes in both for all participants at 60 minutes (Figure 1B).
cases. There were no other reports of skin irritation as The mean (SD) change relative to baseline at
a result of having electrodes in situ. 60 minutes of )1.8 (1.6) at rest and )3.9 (1.9) on
movement was statistically significant (P < 0.05, Wil-
coxon signed-rank test). These inferential analyses
Optimal Electrode Placement
should not be used to establish the magnitude of treat-
For all participants, it was possible to apply TENS ment effects as the study was designed to allow estima-
electrodes to the residual limb and project TENS sen- tion of parameters for a feasibility study and was not
sation to the most painful site, which was either in the appropriately powered to establish intervention effec-
stump (n = 5) or in the phantom limb (n = 5). This tiveness. There was a marginal increase in mean non-
was achieved by applying electrodes below the knee painful phantom sensations after TENS at rest and on
onto the stump for 9 participants. One participant had movement; however, these changes did not reach sig-
pain which extended above the knee so electrodes were nificance. Scores for the intensity of prosthetic limb
placed above the knee at the leading edge of the pain. awareness were maximal for all participants at base-
For two participants who reported tactile allodynia, line and at 30 minutes and 60 minutes of TENS.
 292 •

Table 1. Participant Demographics and Pain Intensity Scores

Amputation Preamp Rest Pain Rest Pain Movement Movement Movement

Side, Time Since Pain and Current Electrode Location of Rest Pain @30 @60 Selected Pain Pain @30 Pain @60
MULVEY ET AL.

Age/Sex Etiology Amputation Duration Medication Positions* Pain Pre-TENS Minutes Minutes Movement Pre-TENS Minutes Minutes

60 F Left, vascular 7 year, Yes, Gabapentin Above stump Phantom 3 0 0 Reach—weight 4 2 0

10 months < 1 week anterior and bearing on
posterior stump
59 M Left, trauma 3 year, Yes, Paracetamol, Above knee Stump 4 0 0 Walking 3 1 0
7 months < 5 years tramadol, posterior and
gabapentin anterior
22 F Right, trauma 3 year, No, N/A Paracetamol, Above stump Stump 0 0 0 Walking 6 7 4
3 months ibuprofen anterior
(RLP);
gabapentin,
amitriptyline
(PLP)
58 F Right, 1 year, Yes, Pre-gabalin, Above stump Phantom 1 0 0 Walking 8 0 0
vascular 3 months < 6 months fentanyl posterior
patch
55 F Right, 7 year, Yes, Gabapentin, Above stump Phantom 4 3 2 Walking 5 3 3
vascular 7 months > 5 years MST anterior and
posterior
47 F Left, trauma Yes, Paracetamol On stump Phantom 7 4 3 Walking 8 6 5
3 months > 5 years (RLP), anterior
gabapentin
(PLP)
?M Right, 1 year, Yes, Gabapentin Above stump Stump with 3 0 0 Putting on 7 2 1
vascular 1 months < 6 months anterior tactile sock liner
allodynia
52 M Left, BKN re- 21 year, Yes, Cocodamol, On stump Stump with 3 2 3 Walking 5 3 3
amputation 2 months > 5 years ibuprofen, anterior tactile
paracetamol allodynia
61 F Right, trauma 18 year Yes, Cocodamol Above stump Stump 1 0 0 Walking 5 2 1
> 5 years posterior
72 M† Left, trauma 3 year, Yes, None Above stump Phantom 0 0 0 Walking 6 3 1
3 months < 6 months anterior

TENS, transcutaneous electrical nerve stimulation; Preamp, preamputation; RLP, residual limb pain; PLP, phantom limb pain; BKN, below knee.
*Below knee unless otherwise stated.
All participants reported both stump and phantom pain over the previous month except † who had stump pain without phantom pain.
 TENS for Phantom Pain and Stump Pain • 293

Table 2. Summary Data for Sample—Mean (SD)

Baseline 30 Minutes Change at 30 Minutes* 60 Minutes Change at 60 Minutes*

Intensity of pain
Rest 2.6 (2.2) 0.9 (1.5) )1.7 (1.4)** 0.8 (1.3) )1.8 (1.6)**
On movement 5.7 (1.6) 2 (2.1) )2.8 (2.3)** 1.8 (1.8) )3.9 (1.9)**
Intensity phantom sensation
Rest 7.8 (3.5) 9.3 (1.4) 1.5 (2.4) 9.7 (0.9) 1.9 (3.1)
On movement 8.0 (3.4) 9.7 (0.9) 1.7 (2.9) 9.7 (0.9) 1.7 (2.9)
Intensity prosthesis awareness
Rest 10 (0.0) 10 (0.0) 0.0 10 (00) 0.0
On movement 10 (0.0) 10 (0.0) 0.0 10 (00) 0.0

*A negative difference implies pain reduction at 1 hour.

**Significant reduction in pain intensity rating (P < 0.05) using Wilcoxon signed-rank analysis.

Figure 1. Individual response trajectories to transcutaneous electrical nerve stimulation (A) resting pain, (B) movement pain.
VAS, visual analog scale.
294 • MULVEY ET AL.

Unsolicited comments made by participants magnitude of effect against a placebo TENS control,
reflected the benefit of TENS: such as that described by Rakel et al.23 The low base-
‘‘TENS took my pain away in my phantom limb line pain scores at rest may have masked potential
and my stump… It didn’t just dull the pain, as I had reductions in resting pain during TENS (ie, a floor
expected it would, it actually took the pain away!’’ effect) as 7 participants were pain free after 60 minutes
(Participant 1). The five participants for whom TENS of TENS. The use of a pain relief scale, which stan-
paresthesia was projected into their phantom limb dardizes participants’ baseline scores (ie, zero pain relief),
reported that the sensation of TENS felt as it was aris- may prove more useful in monitoring TENS effects
ing from their prosthetic limb. One participant com- when baseline pain intensity scores at rest are low.
mented ‘‘it feels like the tingling is coming from my Transcutaneous electrical nerve stimulation did not
prosthetic leg’’. exacerbate pain or irritate the skin at the stump in the
10 participants. Two participants with tactile allodynia
who reported sharp stabbing and shooting stump pain
DISCUSSION
in response to light touch tolerated the application of
This study demonstrates that TENS is acceptable and TENS electrodes to nonallodynic skin proximal to the
well tolerated when placed on the amputation stump pain and also tolerated TENS sensations projected
in lower limb (transtibial) amputee patients. TENS within the allodynic area. This observation is interest-
reduced pain at rest and on movement when TENS ing because allodynia is generated in part by activation
sensation was projected into the main site of pain of low threshold mechanoreceptive peripheral afferents
which was either the phantom limb or stump. In some (Ab) whose input is amplified via sensitisation in the
but not all participants, it was possible to identify opti- central nervous system, resulting in pain.24 Paradoxi-
mal electrode positions by mechanically palpating the cally, TENS did not exacerbate pain in these cases
stump region to elicit a referred phantom sensation. despite activating low threshold mechanoreceptive
In addition, some participants reported that they expe- peripheral afferents (Ab). Human studies and animal
rienced TENS sensation arising from their prosthetic models have demonstrated that peripheral TENS
limb itself. Attributing TENS sensations to a prosthetic decreases ongoing central sensitisation by inhibiting
limb may help improve proprioceptive awareness of the activity of nociceptive responsive second order neu-
the prosthesis and facilitate perceptual embodiment.18 rones which may account for the reduction in symp-
The primary purpose of this study was to provide toms of mechanical hyperalgesia and allodynia seen
pilot data on treatment outcomes and tolerability, as here.25–28
well as recruitment strategies, to inform the design of This study demonstrated that it was possible to pro-
future investigations. The findings of this study are ject TENS sensation into a phantom limb by placing
limited by the small number of participants involved, the electrodes over the stump. Optimal placement was
which was in part due to recruitment difficulties. Fur- on areas that elicited referred phantom sensations
thermore, the lack of a control group prevents compar- when mechanically stimulated. Although TENS
isons of effectiveness being made. The short duration reduced phantom pain relative to baseline, it did not
of treatment per participant limits the extent to which change in the intensity of nonpainful phantom sensa-
the results can inform the outcome of long-term treat- tion scores although this may have been due in part to
ment use with TENS for amputee patients. high baseline intensity scores (ie, ceiling effect).
Although this study did not include a control group Prosthesis limb awareness scores were maximal for
and was not powered for a test of difference, it was all participants at baseline, suggesting that they had
noteworthy that the mean reduction in movement pain perceptually embodied their prosthetic limb, which is
intensity (relative to baseline) after 60 minutes of not unexpected as they were all established prosthetic
TENS was )3.9 (1.9) NRS points, which is almost limb users. Nevertheless, the study found that five par-
double the criteria for clinically meaningful change in ticipants reported experiencing paresthetic sensations
pain of two NRS points as described by the IMM- arising from their prosthetic limb itself when TENS
PACT guidelines.22 The reduction in pain at rest was was projected into their phantom limb. The synthesis
just below clinically meaningful at )1.8 (1.6). Taking of visual inputs from their prosthesis and the proprio-
into consideration the limitations of this pilot trial, ceptive experience of their phantom limb is likely to be
these data merit further exploration to establish driving this enigmatic experience of misattribution of
 TENS for Phantom Pain and Stump Pain • 295

TENS sensations to the prosthesis. We have previously 5. Jensen T, Nikolajsen L. Phantom pain and other phe-
published this hypothesis.18 nomena after amputation. In: Wall P, Melzack R. eds. Text-
Underlying mechanisms are likely to involve the book of Pain. London: Chirchill Livingstone; 2006: 799–814.
6. Nikolajsen L, Jensen TS. Phantom limb pain. Br
intermodal correlation of touch, visual, and proprio-
J Anaesth. 2001;87:107–116.
ceptive sensory information in the somatosensory areas 7. Mulvey MR, Bagnall AM, Johnson MI, Marchant
of the cortex, which are modulated by pre-existing PR. Transcutaneous electrical nerve stimulation (TENS) for
body representations.29–32 Integrating sensory function phantom pain and stump pain following amputation in
into a prosthesis and experiencing the prosthesis as a adults. Cochrane Database Syst Rev. 2010;5:CD007264.
corporeal structure (ie, a natural extension of one’s 8. Carabelli RA, Kellerman WC. Phantom limb pain:
own body) have been shown to be key elements of suc- relief by application of TENS to contralateral extremity.
cessful prosthetic rehabilitation.33,34 Incorporating sen- Arch Phys Med Rehabil. 1985;66:466–467.
9. Kawamura H, Ito K, Yamamoto M, et al. The trans-
sory function into prosthetic limbs involves highly
cutaneous electrical nerve stimulation applied to contralat-
invasive and costly surgical procedures.16,17,35–37 eral limbs of the phatom limb pain. J Phys Ther Sci.
TENS may prove to be an inexpensive, noninvasive 1997;9:71–76.
self-administered technique to aid the perceptual 10. Miles J, Lipton S. Phantom limb pain treated by elec-
embodiment of a new prosthetic limb. The findings trical stimulation. Pain. 1978;5:373–382.
support the delivery of a feasibility trial. 11. Moura VL, Faurot KR, Gaylord SA, et al. Mind-body
interventions for treatment of phantom limb pain in persons
with amputation. Am J Phys Med Rehabil. 2012;91:701–
CONCLUSION 704.
12. Subedi B, Grossberg GT. Phantom limb pain: mecha-
This study found that participants tolerated TENS nisms and treatment approaches. Pain Res Treat.
delivered to a painful phantom and/or stump and that 2011;2011:864605.
TENS may reduce pain both on movement and at rest. 13. Anani A, Korner L. Discrimination of phantom hand
The study also found that TENS sensation can be pro- sensations elicited by afferent electrical nerve stimulation in
jected into a phantom limb and that the TENS sensa- below-elbow amputees. Med Prog Technol. 1979;6:131–135.
tion feels as if it arises from a prosthetic limb. This 14. Anani AB, Ikeda K, Korner LM. Human ability to
may prove a useful aid for perceptual embodiment of discriminate various parameters in afferent electrical nerve
stimulation with particular reference to prostheses sensory
an artificial limb. A feasibility study that includes a
feedback. Med Biol Eng Comput. 1977;15:363–373.
placebo/sham TENS control as described by Rakel 15. Anani AB, Korner LM. Afferent electrical nerve
et al.23 is recommended to establish the potential for stimulation: human tracking performance relevant to pros-
undertaking a full randomised controlled trial. thesis sensory feedback. Med Biol Eng Comput. 1979;17:
425–434.
16. Clippinger FW, Seaber AV, McElhaney JH, Harrelson
ACKNOWLEDGEMENTS JM, Maxwell GM. Afferent sensory feedback for lower extrem-
This research was funded by a Leeds Metropolitan Uni- ity prosthesis. Clin Orthop Relat Res. 1982;169:202–206.
17. Riso RR. Strategies for providing upper extremity
versity Centenary Bursary. Participants for this research
amputees with tactile and hand position feedback—moving clo-
project were recruited from the Amputee Rehabilitation ser to the bionic arm. Technol Health Care. 1999;7:401–409.
Centre at Seacroft Hospital, Leeds NHS Trust. 18. Mulvey MR, Fawkner HJ, Radford H, Johnson MI.
The use of transcutaneous electrical nerve stimulation
(TENS) to aid perceptual embodiment of prosthetic limbs.
REFERENCES Med Hypotheses. 2009;72:140–142.
19. Mulvey MR, Fawkner HJ, Radford HE, Johnson MI.
1. Berman BM, Bausell RB. The use of non-pharmaco-
Perceptual embodiment of prosthetic limbs by transcutane-
logical therapies by pain specialists. Pain. 2000;85:313–315.
ous electrical nerve stimulation. Neuromodulation.
2. Black LM, Persons RK, Jamieson B. Clinical inqui-
2011;15:42–47.
ries. What is the best way to manage phantom limb pain?
20. Chartered Society of Physiotherapy. Guidance for the
J Fam Pract. 2009;58:155–158.
Clinical Use of Electrophysical Agents. London: Chartered
3. Flor H. Phantom-limb pain: characteristics, causes,
Society of Physiotherapy; 2006.
and treatment. Lancet Neurol. 2002;1:182–189.
21. Charlton J, ed. Stimulation-produced analgesia. In:
4. Halbert J, Crotty M, Cameron ID. Evidence for the
Core Curriculum for Professional Education in Pain. Seattle,
optimal management of acute and chronic phantom pain: a
WA: ISAP press; 2005: 93–96.
systematic review. Clin J Pain. 2002;18:84–92.
296 • MULVEY ET AL.

22. Dworkin RH, Turk DC, Wyrwich KW, et al. Inter- 30. Ehrsson HH, Spence C, Passingham RE. That’s my
preting the clinical importance of treatment outcomes in hand! Activity in premotor cortex reflects feeling of owner-
chronic pain clinical trials: IMMPACT recommendations. ship of a limb. Science. 2004;305:875–877.
J Pain. 2008;9:105–121. 31. Tsakiris M, Hesse MD, Boy C, Haggard P, Fink GR.
23. Rakel B, Cooper N, Adams HJ, et al. A new transient Neural signatures of body ownership: a sensory network
sham TENS device allows for investigator blinding while for bodily self-consciousness. Cereb Cortex. 2007;17:
delivering a true placebo treatment. J Pain. 2010;11:230–238. 2235–2244.
24. Doubell TP, Mannion RJ, Woolf CJ. The dorsal horn: 32. Tsakiris M, Longo MR, Haggard P. Having a body
state-dependant sensory processing, plasticity and the genera- versus moving your body: neural signatures of agency
tion of pain. In: Wall P, Melzack R, eds. Textbook of Pain. and body-ownership. Neuropsychologia. 2010;48:2740–
London: Chruchill Livingstone; 1999: 165–181. 2749.
25. Baba H, Doubell TP, Woolf CJ. Peripheral inflamma- 33. Montoya P, Larbig W, Grulke N, Flor H, Taub E,
tion facilitates Abeta fiber-mediated synaptic input to the Birbaumer N. The relationship of phantom limb pain to
substantia gelatinosa of the adult rat spinal cord. J Neurosci. other phantom limb phenomena in upper extremity ampu-
1999;19:859–867. tees. Pain. 1997;2:87–93.
26. Garrison DW, Foreman RD. Decreased activity of 34. Murray CD. An interpretative phenomenological
spontaneous and noxiously evoked dorsal horn cells during analysis of the embodiment of artificial limbs. Disabil Reha-
transcutaneous electrical nerve stimulation (TENS). Pain. bil. 2004;26:963–973.
1994;58:309–315. 35. Dhillon GS, Lawrence SM, Hutchinson DT, Horch
27. Garrison DW, Foreman RD. Effects of transcutane- KW. Residual function in peripheral nerve stumps of ampu-
ous electrical nerve stimulation (TENS) on spontaneous and tees: implications for neural control of artificial limbs.
noxiously evoked dorsal horn cell activity in cats with tran- J Hand Surg Am. 2004;29:605–615.
sected spinal cords. Neurosci Lett. 1996;216:125–128. 36. Dhillon GS, Horch KW. Direct neural sensory feed-
28. Ma YT, Sluka KA. Reduction in inflammation- back and control of a prosthetic arm. IEEE Trans Neural
induced sensitization of dorsal horn neurons by transcutane- Syst Rehabil Eng. 2005;13:468–472.
ous electrical nerve stimulation in anesthetized rats. Exp 37. Dhillon GS, Kruger TB, Sandhu JS, Horch KW.
Brain Res. 2001;137:94–102. Effects of short-term training on sensory and motor function
29. Botvinick M, Cohen J. Rubber hands ‘feel’ touch that in severed nerves of long-term human amputees. J Neuro-
eyes see. Nature. 1998;391:756. physiol. 2005;93:2625–2633.
Copyright of Pain Practice is the property of Wiley-Blackwell and its content may not be copied or emailed to
multiple sites or posted to a listserv without the copyright holder's express written permission. However, users
may print, download, or email articles for individual use.