Accepted Manuscript

Efficacy of Addition of Transcutaneous Electrical Nerve Stimulation to Standardized

Physical Therapy in Subacute Spinal Spasticity: A Randomized Control Trial

Win Min Oo, M.B,B.S; M.Med.Sc

PII: S0003-9993(14)00432-8
DOI: 10.1016/j.apmr.2014.06.001
Reference: YAPMR 55866

To appear in: ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION

Received Date: 7 May 2014

Revised Date: 10 June 2014
Accepted Date: 11 June 2014

Please cite this article as: Oo WM, Efficacy of Addition of Transcutaneous Electrical Nerve Stimulation to
Standardized Physical Therapy in Subacute Spinal Spasticity: A Randomized Control Trial, ARCHIVES
OF PHYSICAL MEDICINE AND REHABILITATION (2014), doi: 10.1016/j.apmr.2014.06.001.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
TENS and physical therapy in subacute spinal spasticity
ACCEPTED MANUSCRIPT

Efficacy of Addition of Transcutaneous Electrical Nerve Stimulation to Standardized

Physical Therapy in Subacute Spinal Spasticity: A Randomized Control Trial

Win Min Oo (M.B,B.S; M.Med.Sc)

Department of Physical Medicine and Rehabilitation, Mandalay University of Medicine,

PT
Myanmar.

RI
Acknowledgement: I would like to thank Professor Dr. Win Nyi Shein, PhD, for management

SC
support, Professor Dr. Win Myint OO, PhD, for statistical advices, Sr. Consultant Dr. Soe

U
Thein, Dr.Med.Sc, physiotherapists and the participating subjects from University Hospital of
AN
Mandalay for assisting in this study.
M

Disclaimers: I certify that the views expressed in the submitted article are my own and not an

official position of the institution. I confirm that there was no presentation of this material to
D

any journal.
TE

Source of support: There was no financial support for this study.

C EP

Conflict of interest: None

AC

Correspondence: Win Min OO, Physiatrist, Department of Physical Medicine and

Rehabilitation, Mandalay University of Medicine, Myanmar. Ph: 095991053012

E-mail: winminoo@ummdy.com; drwinminoopmr@gmail.com

Reprint can be obtained from this author.

TENS and physical therapy in subacute spinal spasticity
ACCEPTED MANUSCRIPT

Word count of abstract: 319

Word count of body of the text: 3163

Number of figure: 1

Number of tables: 4

PT
RI
U SC
AN
M
D
TE
C EP
AC
 1
ACCEPTED MANUSCRIPT
1 Efficacy of Addition of Transcutaneous Electrical Nerve Stimulation to Standardized

2 Physical Therapy in Subacute Spinal Spasticity: A Randomized Control Trial

5 Abstract

PT
6

RI
7

8 Objective: To study the immediate and short-term efficacy of adding transcutaneous

SC
9 electrical nerve stimulation (TENS) to standardized physical therapy on subacute spasticity

10 within 6 months post-spinal cord injury

11 Design: Randomized controlled trial for 3 weeks

U
AN
12 Setting: University hospital of Mandalay
M

13 Subjects and intervention: Sixteen subjects with clinically determined spasticity were

14 randomly assigned to experimental group (n=8, 60 minute sessions of TENS over the
D

15 bilateral common peroneal nerves before 30 minutes of physical therapy) or control group
TE

16 (n=8, 30 minutes of physical therapy alone). All patients in both groups had access to

17 standardized rehabilitation care.

EP

18 Outcome measures: Composite spasticity score as primary end point to assess plantar flexor
C

19 spasticity which included three sub-scores: ankle jerk, muscle tone and ankle clonus scores.
AC

20 These sub-scores were designated as secondary end points. Serial evaluations were made at

21 baseline before study entry, immediately after the first and last sessions in both groups.

22 Results: On analysis for immediate effects, there was significant reduction only in composite

23 spasticity score (mean difference 1.75 [99% confidence interval (CI) 0.47 to 3.03], p= 0.002)

24 in the experimental group but no significant reduction was observed in all outcome variables

25 in the control group. The significant difference of composite spasticity score (1.63[99% CI
 2
ACCEPTED MANUSCRIPT
1 0.14 to 3.11], p= 0.006) was observed between the two groups. After 15 sessions of

2 treatment, significant reduction was determined in composite spasticity score (2.75[99% CI

3 1.31 to 4.19], p<0.001), muscle tone score (1.75 [99% CI 0.16 to 3.34], p=0.006) and ankle

4 clonus score (0.75 [99% CI 0.18 to 1.32], p=0.003) in the experimental group while none of

5 outcome variables revealed significant reduction in control group. The between-group

PT
6 difference was significant only in composite spasticity score (2.13 [99% CI 0.59 to 3.66], p=

RI
7 0.001) and muscle tone score (1.50 [99% CI 0.15 to 2.85], p=0.005) after 15 intervention

8 sessions.

SC
9 Conclusion: Addition of TENS to standardized physical therapy had synergistically anti-

10 spastic action, providing more effective reduction of clinical spasticity.

11
U
AN
12
M

13 Key words

14 Spasticity; Transcutaneous Electrical Nerve Stimulation; Spinal Cord Injury; Physical

D

15 Therapy; Rehabilitation
TE

16

17
EP

18 List of abbreviations
C

19
AC

20

21 ASIA American Spinal Injury Association

22 GABA Gamma-aminobutyric acid

23 SCI Spinal cord injury

24 TENS Transcutaneous electrical nerve stimulation

25
 3
ACCEPTED MANUSCRIPT
1

2 In upper motor neurone lesions, spasticity is clinically diagnosed by eliciting

3 exaggerated tendon reflexes and velocity-dependent muscle hypertonia in resting state.1 It can

4 further be separated on clinical basis as follows: (1) muscle hypertonia due to intrinsic tonic

5 stretch reflex, (2) hyperreflexia and clonus due to intrinsic phasic stretch reflex (3) muscle

PT
6 spasms due to increased exteroceptive polysynaptic reflexes.2

RI
7

8 Spasticity is a common disorder which develops in the estimated 65–78% of patients

SC
9 with spinal cord injury (SCI).3-5 Of these patients, 40-60 % reports it as clinical impairment

10 and about 50% requires drug treatment.4,5

11
U
AN
12 Spasticity not only leads to incapacitating complications such as muscle contracture,
M

13 pain, difficulty in functional activities but also increases the economic and caregiver

14 burdern.4,5 Suboptimal treatment of severe spasticity contributes to significant obstacles in the

D

15 rehabilitation process.5-7 Therefore, the spasticity should be treated effectively and

TE

16 judiciously.

17
EP

18 Effective treatment of spasticity is still a therapeutic challenge that will not respond to
C

19 a single therapeutic option in all individuals, and necessitates a combination treatment in

AC

20 which priority is usually given to the most conservative measures with the fewest side

21 effects.5,8

22

23 Vibration of Achilles tendon during a complete block of the common peroneal nerve

24 which innervated the antagonistic dorsiflexor muscles resulted in the decreased inhibition of

25 soleus monosynaptic reflex9. Therefore, it was suggested that stimulation of Ia afferent fibres
 4
ACCEPTED MANUSCRIPT
1 in the common peroneal nerve with transcutaneous electrical nerve stimulation (TENS) could

2 increase the inhibition of this reflex and decrease the plantar flexor spasticity.

4 TENS sustains obvious advantages such as low-cost, non-invasiveness, portability,

5 easy applicability and simplicity for use at home. It has few side effects and the long-term

PT
6 cost is lower than drug treatment.10-13

RI
7

8 On literature review, efficacy of TENS was studied in different spasticity models.12-20

SC
9 Most authors14-19 reported the reduction of spasticity after the single session of TENS.

10 Therefore, there was a deficiency of randomized control study on the efficacy of repetitive

11
U
TENS application, especially in subacute spinal spasticity. Moreover, there was also a lack of
AN
12 robust evidence for anti-spastic effects of a combination of TENS and the physical therapy

during subacute SCI period when the most of motor recovery could be expected21 and the
M

13

14 least of secondary biomechanical muscle changes such as fibrosis might take place.7,22
D

15
TE

16 TENS coupled with standardized physical therapy would be practical during subacute

17 SCI rehabilitation and synergistic in mechanism of action because neural component of

EP

18 spasticity would respond to TENS and the biomechanical component to physical therapy22
C

19 The objective of this study was to determine the immediate and short-term efficacy of the
AC

20 combination of transcutaneous electrical nerve stimulation (TENS) with standardized

21 physical therapy in subacute spinal spasticity.

22

23

24 Material and methods

25
 5
ACCEPTED MANUSCRIPT
1

2 Design and sample size

5 The study was designed as a prospective, randomized, controlled, single-blind clinical

PT
6 trial which was ethically revised and approved by the university ethical committee. The

RI
7 sample size was calculated using the composite spasticity score as the primary outcome

8 measure. The effect sizes of TENS application for immediate and short-term efficacy were

SC
9 obtained from the previous studies.12,20 It was determined with computer software Stata 9.2a

10 that 8 subjects per group were needed to achieve 88% power for significant between-group

11 difference at an α level of 0.05.

U
AN
12
M

13

14 Randomization and blinding

D

15
TE

16

17 The block randomization method23 was used to assign the patients to either
EP

18 experimental or control group. The randomization sequence was generated by using a

C

19 permuted block sequence from a random number generator and placed into sequentially

numbered, opaque and sealed envelopes (SNOSE technique)24 by an independent staff, who
AC

20

21 then drew lots. The treatment allocation was concealed to the outcome assessor by blinding

22 the group assignment and to data analyst by not being given the codes for treatment group

23 status.25

24

25
 6
ACCEPTED MANUSCRIPT
1 Subjects

4 Patients with new traumatic SCI and clinically verified spasticity who reported

5 spasticity as pain or limitation of daily activities, or both4 were recruited from university

PT
6 hospital between September 2011 and August 2012. Subjects were selected based on the

RI
7 following criteria: (1) between 18 and 60 years of age; (2) post-injury duration ≤ 6 months; 3)

8 spasticity over lower limb(s); (4) having the return of ankle jerk denoting the recovery from

SC
9 spinal shock.21

10

11
U
Potential participants were excluded from this study if they sustained (1) exacerbating
AN
12 factors of spasticity such as urinary tract infection, pressure ulcer;5,22 (2) systemic diseases
M

13 with peripheral neuropathy; (3) lumbosacral radiculopathy; (4) impaired cognitive function;

14 (5) history of other neurological disorder; (6) an implanted pacemaker; (7) metal implants in
D

15 the affected leg; (8) broken skin under the placement of electrodes10,26
TE

16

17 Before study entry, the demographic characteristics of participant such as age, gender,
EP

18 post-injury duration, the level of spinal cord injury, American Spinal Injury Association

score27 were collected and the baseline evaluations of composite spasticity score and its sub-
C

19

scores16 were conducted. All patients gave their written informed consent.
AC

20

21

22

23 Intervention

24

25
 7
ACCEPTED MANUSCRIPT
1 TENS stimulator

4 A portable dual-channel TENS / ES combob (TENS/ ES-320, dual-channel, ITO,

5 Japan) and two pairs of standard disposable self-adhesive square electrodes (4×4cm) were

PT
6 used for TENS administration.

RI
7

SC
9 Experimental procedure

10

11
U
AN
12 Skin was cleaned with cotton wool soaked in 70% methylated spirit to reduce skin
M

13 resistance and then two electrodes from each channel were applied to each common peroneal

14 nerve (L4-S2) in such a way that the first anode electrode was placed posterior to the head of
D

15 fibula and the second cathode electrode was applied over deep peroneal nerve 2 cm lateral to
TE

16 tibial bone and 2 cm below the head of fibula.

17
EP

18 The digital intensity control, pulse-width control and pulse-frequency control knobs
C

19 were locked by using keyboard lock function to guarantee constant output parameters during
AC

20 TENS application. The stimulation was simultaneously delivered to the patient in supine

21 position using the symmetric biphasic rectangular waves at a frequency of 100 Hz, a pulse-

22 duration of 0.2 msec and the intensity of 15 mA, which was twice the average sensory

23 threshold of TENS application in healthy subjects.16 This TENS parameter would not cause

24 muscle contractions because the phase duration was 0.1 msec and the phase charge in this

25 study would be 1.5 microcoulomb (0.1msec×15mA) which was typical for sensory but not
 8
ACCEPTED MANUSCRIPT
1 motor stimulation. Treatment was organized as one 60-min session in the morning on every

2 weekday for 3 weeks, always before physical therapy. After every treatment session,

3 electrodes were removed from the patient and the skin was cleaned and checked for any skin

4 irritation.

PT
6 Patients in the control group were not given TENS application. Their care was

RI
7 otherwise the same as that of patients in the experimental group. Standardized physical

8 therapy included inpatient SCI physiotherapy attempting to minimize or reverse impairments

SC
9 such as poor strength, restricted joint mobility and reduced dexterity. It also included

10 occupational therapy for training of functional skills such as dressing, walking, transferring

11
U
and propelling a wheelchair. Besides, a strong focus was given to education about spinal
AN
12 spasticity, triggering factors, proper positioning and heel cord stretching exercises. All these
M

13 treatments were given by the assigned physiotherapist.

14
D

15 If the participant missed two treatment sessions, the case would be defined as drop-
TE

16 out. To standardize the drug therapy, the patients were instructed at study entry not to alter

17 their drug treatment within the study duration. If any change in medical regime occurred, the
EP

18 subject would be counted as drop-out.

C

19
AC

20 Treatment compliance rate in both groups was monitored using patient’s diary and

21 calculated as the number of treatment-receiving days divided by the number of expected days

22 of the study (15 days) and multiplied by 100 to yield a percentage.28

23

24

25 Outcome measures
 9
ACCEPTED MANUSCRIPT
1

3 Clinical assessments were conducted using composite spasticity score16 at baseline,

4 immediately after the first and the last interventions for both groups by a physiatrist who was

5 familiarized with these scores. Outcome measures were collected from both stimulated legs

PT
6 but statistical analysis was conducted only for the dominant legs of the participants.

RI
7

SC
9 Primary outcome measure

10

11
U
AN
12 Composite spasticity score was used as primary outcome measure to determine the
M

13 overall tonic and phasic aspects of clinical spasticity. It resulted from the summation of all

14 three sub-scores: (1) ankle jerk score (2) muscle tone score and (3) ankle clonus score.20
D

15 While muscle tone score evaluated the tonic component, the ankle jerk and ankle clonus gave
TE

16 the information regarding with the phasic component.12 The respective scores extending

17 from 1-5, 6-9, 10-12, 13-16 stand for normal, mild, moderate and severe spasticity.41
EP

18
C

19
AC

20 Secondary outcome measures

21

22

23 The three sub-scores of composite spasticity score were designated as secondary

24 outcome scores to determine how TENS application effected on individual sub-scores. The

25 ankle jerk score was measured on a scale ranging from 0 (no reflex) to 4 (maximally
 10
ACCEPTED MANUSCRIPT
1 hyperactive reflex).20 The muscle tone score was evaluated on a modified double-weighted 5-

2 point Ashworth scale29 where 0 indicated no increase in muscle tone and 8 corresponded to

3 maximally increased muscle tone. The ankle clonus score ranges between 1 (clonus not

4 elicited) and 4 (sustained clonus).20

PT
6 Under the relaxed condition, all clinical spasticity scores were graded with manual

RI
7 techniques in supine position Ankle jerk was scored by flexing the examined leg of the

8 patient at both hip and knee and rotating it externally so that the lower leg rests across the

SC
9 opposite shin. Then the foot was dorsiflexed at the ankle and the Achilles tendon was

10 tapped with reflex hammer with maximal tendon percussion.30 Muscle tone score was

11
U
rated by passively stretching the ankle joint from maximun possible plantarflexion to
AN
12 maximun possible dorsiflexion at a moderate speed for three consecutive times.31 For
M

13 grading ankle clonus score, the knee was supported in a partly flexed position with one hand.

14 With the other hand, the foot was sharply dorsiflexed and maintained in dorsiflexion,
D

15 looking for rhythmic oscillations between dorsiflexion and plantar flexion. 30

TE

16

17
EP

18 Data analysis
C

19
AC

20

21 Descriptive analyses were performed to present the demographic and clinical

22 characteristics of the 2 groups. To evaluate baseline differences between the two groups,

23 Student t-tests were conducted on age and time post-injury (month), composite spasticity

24 score and all sub-scores while chi-squared tests were computed on gender and level of injury.

25 Because all data met the criterion of normality (the Shapiro-Wilks test), statistical
 11
ACCEPTED MANUSCRIPT
1 significance of the outcome variables between the two groups was determined with the

2 independent t-tests. Within-group differences of outcome measures in both groups were

3 analyzed using the paired sample t-tests. A p-value ≤ 0.05 denoted a statistically significant

4 difference. To correct the family-wise error rate, p value was adjusted with Bonferroni

5 method and the error rate was determined as p≤0.01. Mean differences and 99% confidence

PT
6 intervals were described where appropriate to enable conclusion on clinical relevance of the

study findings. All the statistical procedures were carried out using SPSS version 19.c

RI
7

SC
9

10 Results

11
U
AN
12
M

13 A total of 16 patients (32 spastic plantarflexors) completed the study. Figure 1

14 demonstrates the flow of participants through the trial. All received interventions as allocated.
D

15 The experimental and control groups were well matched at baseline in terms of age, gender,
TE

16 post-injury duration (month), level of injury, ASIA score and any of the outcome variables

17 assessed in this study (Table 1). Results of within-group differences for experimental and
EP

18 control groups were presented in Table 2 and 3 respectively and results for between-group
C

19 difference in Table 4.
AC

20

21

22 Immediate effects on spasticity

23

24

25 After the first session, the reduction of clinical spasticity assessed by composite
 12
ACCEPTED MANUSCRIPT
1 spasticity score was statistically significant in experimental group (mean difference 1.75[99%

2 confidence interval (CI) 0.47 to 3.03], p= 0.002). No such reduction was found in control

3 group (0.63 [99% CI -0.51 to 1.76], p=0.095. The between-group difference of composite

4 spasticity score was significant (1.63[99% CI 0.14 to 3.11], p= 0.006). Analysis of within-

5 group and between-group differences in sub-scores revealed no significant improvement.

PT
6

RI
7

8 Short-term effects on spasticity

SC
9

10

U
AN
11 After the final session, the composite spasticity score significantly improved in

12 experimental group (2.75[99% CI 1.31 to 4.19], p<0.001) but was not significant in control
M

13 group (1.13 [99% CI -0.55 to 2.80], p=0.051). The between-group difference of composite

14 spasticity score was also significant (2.13 [99% CI 0.59 to 3.66], p= 0.001). On sub-score
D

15 analysis, significant decreases were observed in muscle tone score (1.75 [99% CI 0.16 to
TE

16 3.34], p=0.006) and ankle clonus score (0.75 [99% CI 0.18 to 1.32], p=0.003) in experimental
EP

17 group, while none of the sub-scores in control group revealed significant reduction. On

18 analysis of between-group difference, muscle tone score (1.50 [99% CI 0.15 to 2.85],
C

19 p=0.005) was significant after 15 TENS sessions.

AC

20

21 During the entire study period, no patient in either group changed their treatment

22 regime. All participants completed the study as allocated, presenting no adverse events, and

23 compliance rate in both groups being 100%.

24

25
 13
ACCEPTED MANUSCRIPT
1 Discussion

4 Spasticity develops when hyperexcitability of α motor neurons occurs consequent to

5 an imbalance in the excitatory and inhibitory inputs.32 The mechanism of TENS in spasticity

PT
6 reduction is hypothesized to be mediated (1) by modulating excessive α-motor neuron

activity through dynorphin release33 and (2) by inducing synaptic reorganization through

RI
7

8 increased afferent sensory inputs.16,19,34 Large fibre afferent stimulation in the form of TENS

SC
9 can modulate the interneuronic activities in several spinal segments through segmental and

10 propriospinal pathways20 enhancing the inhibitory mechanisms such as reciprocal inhibition35

11 and presynaptic inhibition20.

U
AN
12
M

13 In spastic SCI individuals, there existed a decrease in presynaptic inhibition evoked

14 by muscle or tendon vibration at rest, and this could contribute to the hyper-excitability of Ia–
D

15 motoneuron synapse.36 Moreover, after repetitive application of TENS over common

TE

16 peroneal nerve in spastic hemiparetic subjects, there was the significant increase in the

17 vibratory inhibition of the H reflex,20 which suggested an indirect evidence of increase in pre-
EP

18 synaptic inhibition.
C

19
AC

20 The presynaptic inhibition was associated with primary afferent depolarization. It

21 exerted via axo-axonal gamma-aminobutyric acid (GABA) synapses, and decreased the

22 amplitude of presynaptic impulse and excitatory neurotransmitter release from Ia afferent

23 terminal.37 GABA is a neurotransmitter which involves in various inhibitory spinal

24 mechanisms. It was reported that liberation of the GABA from the spinal cord was increased

25 by application of high frequency TENS to rats.38

 14
ACCEPTED MANUSCRIPT
1

2 Although TENS could directly be applied to the sensory nerve or over cutaneous skin

3 receptors, dermatomal stimulation produced no or less significant anti-spastic effects.39

4 While electrical stimulation of the ankle dorsiflexors revealed the reduction of the plantar

5 flexor stretch reflex35, Embrey et al40 found that spasticity remained unchanged on modified

PT
6 Ashworth scale when electrical stimulation was applied to dorsiflexors for 65% of gait cycle

RI
7 and to plantarflexors for 35% of gait cycle.

SC
9 Due to limitation of resources, spasticity was evaluated only with clinical parameters

10 (composite spasticity score) due to its advantage of giving information on the phasic

11
U
components of the stretch reflex2,6 compared to the modified Ashworth scale. Levin and
AN
12 Hui-Chan had verified the test-retest reliability of this composite spasticity score in

hemiplegic patients (intraclass correlation coefficient =0.87).20 Its applicability had also
M

13

14 been demonstrated in grading plantar flexor spasticity of spinal origin.41

D

15
TE

16 Effect sizes of TENS application were observed to have considerable discrepancy

17 across the studies partly due to difference in stimulation parameters, treatment procedures
EP

18 and outcome measures used, making the comparative analysis of studies difficult.42 Roughly,
C

19 spasticity reduction was revealed only in high-frequency TENS but not low-frequency

TENS.33 Spasticity decrease was reported in a variety of studies where TENS was
AC

20

21 delivered in various pulse-durations ranging from 0.1 msec to 0.3 msec. Therefore, it

22 could be hypothesized that pulse-duration parameter seem to be less critical than frequency of

23 TENS.43

24
 15
ACCEPTED MANUSCRIPT
1 Based on findings from previous studies,18,20 the 60-min session of TENS was

2 assumed to be optimal. Compared to study by Han et al33 who studied the effects of TENS

3 over the acupoints of ST 36 inside tibialis anterior muscles below the knee joint and BL57

4 below the gastronemius muscle in 4 spinal cord injury patients for 3 months, the duration of

5 present study was shorter and limited only to three weeks because improvement in spasticity

PT
6 tended to reach a plateau after 2 weeks of daily TENS application.20,33

RI
7

8 In the present study, significant improvement was detected only in composite

SC
9 spasticity score, muscle tone score and clonus score after first TENS session, consistent with

10 the report by Chung and Cheng.18 However, their study included both acute and chronic SCI

11
U
patients up to 364 weeks post-injury. These data suggest that TENS may have an immediate
AN
12 effect on these components of clinical spasticity long after the onset of SCI. However there
M

13 are limits to such proposal. Priebe et al. reported that muscle tone score was a reliable

14 measurement tool for rating the tonic component of spasticity.44 On the other hand, Nielsen
D

15 and Sinkjaer concluded that the muscle tone score could overestimate spasticity due to the
TE

16 non-reflex biomechanical properties of the spastic muscle.45

17
EP

18 Evaluation after final intervention in experimental group showed significant decline in

C

19 all clinical outcome scores except for ankle jerk score. The results were consistent with the

study of Levin and Hui-chan20 in hemiparetic patients. The inconsistent response between
AC

20

21 muscle tone score and ankle jerk score could suggest the different effects of TENS on the

22 various neurophysiological aspects of spasticity as muscle tone score represented tonic

23 component of spasticity while ankle jerk score measured the phasic component.2,6

24
 16
ACCEPTED MANUSCRIPT
1 The conflicting outcomes between ankle jerk score and clonus score could be

2 explained as follows: tendon jerk resulted from hyperexcitability of stretch reflex6 while the

3 ankle clonus was mediated through various peripheral inputs (i.e. recurrent activation of

4 stretch reflexes) and central inputs (i.e. rhythmic activity of central oscillator within the

5 spinal cord in response to peripheral inputs).46 Therefore, TENS could produce the different

PT
6 responses in the components of clinical composite score because spasticity itself does result

from imbalance of diverse neurophysiological mechanisms.22,32

RI
7

SC
9 Aydin et al.12 described significant reduction of muscle tone score and ankle jerk

10 score after repetitive TENS application over the tibial nerve. Compared to this study,

11
U
inconsistent findings in phasic components can be attributed to different TENS parameters
AN
12 and application sites (common peroneal nerve vs tibial nerve).
M

13

14 However, these results may not apply to other patients groups such as multiple
D

15 sclerosis (MS) as the contradictory findings had been reported in the studies on these
TE

16 patients.13,47 This may be attributed to wide fluctuations of disease symptoms, wide

17 variability in the response of individual subjects, delay between TENS application and
EP

18 clinical assessment, and presence of precipitating factors for spasticity in participating

subjects.47
C

19
AC

20

21 Sonde et al.48 also reported the significant reduction of modified Ashworth score after

22 TENS administration in stroke patients for 3 months. Similar report was published from the

23 study of Alabdulwahab and Al-Gabbani49 in cerebral palsy patients for 1 week.

24

25
 17
ACCEPTED MANUSCRIPT
1 Limitation of this study

4 According to the Jadad scale,50 this study was deemed to hold adequate

5 methodological quality. Although the number of participating subjects in this study was in

PT
6 line with other TENS studies in spinal spasticity, further study with larger sample size is still

RI
7 needed to fortify the findings of this study. There was no group with sham TENS, and so it

8 could not exclude placebo effect. However, placebo effect was not significant in previous

SC
9 studies.17,19 Moreover, the benefit of spasticity improvement for functional activities should

10 be investigated in future studies, using validated functional scores for SCI patients for longer

11 study duration.
U
AN
12
M

13

14 Conclusion
D

15
TE

16

17 Combination of TENS with standardized physical therapy was beneficial in

EP

18 synergistic reduction of clinical spasticity on both immediate and short-term basis in subacute
C

19 phase of SCI rehabilitation. However, further replication of this study with larger sample size
AC

20 is required for final evidence.

21

22

23 Reference

24

25
 18
ACCEPTED MANUSCRIPT
1 1. Lance JW. Symposium synopsis. In: RG Feldman, RR Young & WW Koella (eds),

2 Spasticity: Disordered Motor Control. Symposia Specialists, Miami (1980), pp. 485–500.

3 Year Book Medical Publishers, Chicago.

4 2. Decq P. Pathophysiology of spasticity. Neurochirurgie 2003; 49: 163-184.

5 3. Maynard FM, Karunas RS, Waring WP III. Epidemiology of spasticity following traumatic

PT
6 spinal cord injury. Arch Phys Med Rehabil. 1990;71(8):566–569.

RI
7 4. Sköld C, Levi R, Seiger A. Spasticity after traumatic spinal cord injury: nature, severity,

8 and location. Arch Phys Med Rehabil 1999; 80: 1548–1557.

SC
9 5. Strommen JA. Management of spasticity from spinal cord dysfunction. Neurol Clin. 2013

10 Feb;31(1):269-86.

11
U
6. Adams MM and Hicks AL. Spasticity after spinal cord injury. Spinal Cord 2005; 43,
AN
12 577–586
M

13 7. Becker W, Letts L. Intrathecal baclofen for adults with spinal spasticity, Can J Neurol Sci

14 22:122-129, 1995
D

15 8. Balakrishnan S and Ward AB. The diagnosis and management of adults with spasticity. In:
TE

16 Barnes MP, Good DC, editors. Handbook of Clinical Neurology, Vol. 110 3rd series.

17 Amsterdam: Elsevier B.V. Pr; 2013. p 145-60.

EP

18 9. Dindar, F. and Verrier, M. (1975) Studies on the receptor responsible for vibration-
C

19 induced inhibition of monosynaptic reflexes in man J Neurol Neurosurg Psychiat. 38: 155-
AC

20 160.

21 10. Johnson M: Transcutaneous electrical nerve stimulation (TENS), in Kitchen S (ed):

22 Electrotherapy: Evidence-Based Practice. Edinburgh, Churchill Livingstone, 2002, pp 259-86

23 11. Bennett MI, Hughes N, Johnson MI. Methodological quality in randomised controlled

24 trials of transcutaneous electric nerve stimulation for pain: low fidelity may explain negative

25 findings. Pain. 2011 Jun;152(6):1226-32.12.

 19
ACCEPTED MANUSCRIPT
1 12. Aydin G, Tomruk S, Keles I, Demir SO and Orkun S. Transcutaneous electrical nerve

2 stimulation versus baclofen in spasticity: clinical and electrophysiologic comparison. Am J

3 Phys Med Rehabil 2005; 84: 584-592.

4 13. Armutlu K, Meric A, Kirdi N, Yakut E and Karabudak R. The effect of transcutaneous

5 electrical nerve stimulation in multiple sclerosis. Neurorehabil Neural Repair 2003; 17: 79-

PT
6 82.

RI
7 14. Bajd T, Gregoric M, Vodovnik L, et al: Electrical stimulation in treating spasticity

8 resulting from spinal cord injury. Arch Phys Med Rehabil 1985;66:515–7

SC
9 15. Potisk KP, Gregoric M and Vodovnik L. Effects of transcutaneous electrical nerve

10 stimulation (TENS) on spasticity in patients with hemiplegia. Scand J Rehabil Med 1995;

11 27(3): 169-174.
U
AN
12 16. Goulet C, Arsenault AB, Bourbonnais D, Laramée MT and Lepage Y. Effects of
M

13 transcutaneous electrical nerve stimulation on H-reflex and spinal spasticity. Scand J

14 Rehabil Med 1996; 28(3): 169-176.

D

15 17. Gregoric M. Suppression of flexor reflex by transcutaneous electrical nerve

TE

16 stimulation in spinal cord injured patients. Muscle Nerve 1998; 21: 166-172.

17 18. Chung BPH and Cheng BKK. Immediate effect of transcutaneous electrical nerve
EP

18 stimulation on spasticity in patients with spinal cord injury. Clin Rehabil 2010; 24: 202-210.
C

19 19. Joodaki MR, Olyaei GR and Bagheri H. The effects of electrical nerve stimulation of
AC

20 the lower extremity on H-reflex and F-wave parameters. Electromyogr Clin

21 Neurophysiol 2001; 41(1): 23-28.

22 20. Levin MF, Hui-Chan CW. Relief of hemiparetic spasticity by TENS is associated

23 with improvement in reflex and voluntary motor functions. Electroencephalogr Clin

24 Neurophysiol 1992; 85(2): 131-142.

 20
ACCEPTED MANUSCRIPT
1 21. Ditunno JF, Little JW, Tessler A and Burns AS. Spinal shock revisited: a four-phase

2 model. Spinal Cord (2004) 42, 383-395.

3 22. Barnes MP (2008). An overview of the clinical management of spasticity. In: MP Barnes

4 and GR Johnson (ed), Upper Motor Neurone and Spasticity, Cambridge University Press, New

5 York, pp. 1-8.

PT
6 23. Matts JP, Lachin JM. Properties of permuted-block randomization in clinical trials.

RI
7 Control Clin Trials. 1988;9:327–344.

8 24. Doig GS, Simpson F. Randomization and allocation concealment: a practical guide for

SC
9 researchers. J Crit Care. 2005;20:187–193.

10 25. Polit DF. Blinding during the analysis of research data. Int J Nurs Stud. 2011 May; 48(5):

11 636-41.
U
AN
12 26. Walsh DM and Basford JR. Transcutaneous Electrical Nerve Stimulation. In: HS
M

13 Smith(ed), Current Therapy in Pain, Elsevier, pp. 541-546.

14 27. Maynard Jr FM, Bracken MB, Creasey G et al. International standards for
D

15 neurological and functional classification of spinal cord injury. American Spinal Injury
TE

16 Association. Spinal Cord 1997; 35: 266–74.

17 28. Peterson AM, Nau DP, Cramer JA, Benner J, Gwadry-Sridhar F, Nichol M. A
EP

18 Checklist for medication compliance and persistence studies using retrospective

C

19 databases. Value Health. 2007 Jan-Feb;10(1):3-12.

AC

20 29. Ashworth B. Preliminary trial of carisoprodol in multiple sclerosis. Practitioner 1964;

21 192: 540–542.

22 30. Bickley L S, Hoekelman R A (2009). Bates' Guide to Physical Examination and

23 History Taking, 10 th edn. Lippincott Williams & Wilkins, pp 595-670.

24 31. Bohannon RW and Smith MB (1987). Interrater reliability of a modified

25 Ashworth scale of muscle spasticity. Phys Ther 1987; 67: 206–207.

 21
ACCEPTED MANUSCRIPT
1 32. Elbasiouny S, Moroz D, Bakr M and. Mushahwar V. Management of spasticity after

2 spinal cord injury: current techniques and future directions. Neurorehabil Neural Repair

3 2010; 24: 23-33.

4 33. Han JS, Chen XH, Yuan Y and Yan SC. Transcutaneous electrical nerve stimulation

5 for treatment of spinal spasticity. Chin Med J (Engl) 1994; 107(1): 6-11.

PT
6 34. Smania N, Picelli A, Munari D, Geroin C, Ianes P, Waldner A, Gandolfi M.

RI
7 Rehabilitation procedures in the management of spasticity. Eur J Phys Rehabil Med. 2010

8 Sep;46(3):423-38.

SC
9 35. Crone C, Nielsen J, Petersen N, et al. Disynaptic reciprocal inhibition of ankle extensors

10 in spastic patients. Brain 1994;117: 1161-8.

11
U
36. Faist M, Mazevet D, Dietz V and Pierrot-Deseilligny E. A quantitative assessment of
AN
12 presynaptic inhibition of la afferents in spastics: differences in hemiplegics and
M

13 paraplegics. Brain 1994; 117(6): 1449-1455.

14 37. Rudomin P, Schmidt RF. Presynaptic inhibition in the vertebrate spinal cord revisited.
D

15 Exp Brain Res 1999;129:1–37.

TE

16 38. Sluka, K.A., Maedai, Y., Lisi, T.L. & Vance, C.G.T. (2007). Release of GABA and

17 activation of GABA(A) in the spinal cord mediates the effects of TENS in rats. Brain Res.,
EP

18 1136, 43-50.
C

19 39. Hsieh JTC, Wolfe DL, Connolly S, Townson AF, Curt A, Blackmer J, Sequeira K,
AC

20 Aubut J. Spasticity after spinal cord injury: an Evidence-Based Review of Current

21 Interventions. Top Spinal Cord Inj Rehabil 2007;13(1):81–97

22 40. Embrey DG, Holtz SL, Alon G, Brandsma BA, McCoy SW. Functional electrical

23 stimulation to dorsiflexors and plantar flexors during gait to improve walking in adults with

24 chronic hemiplegia. Arch Phys Med Rehabil. 2010 May;91(5):687-96.

25 41. Nadeau S, Goulet C, Arsenault AB, Gravel D, Bourbonnais D, Lepage Y.

 22
ACCEPTED MANUSCRIPT
1 Assessment of the plantarflexor muscle tone in chronic stroke and spinal cord injury

2 subjects using the spasticity index. Neurorehabil Neural Repair 1999; 13: 57.

3 42. Claydon LS, Chesterton LS, Barlas P, Sim J. Dose-specific effects of transcutaneous

4 electrical nerve stimulation (TENS) on experimental pain: a systematic review. Clin J Pain.

5 2011 Sep;27(7):635-47.

PT
6 43. Vodovnik L, Stefanovska A, Bajd T. Effects of stimulation parameters on modification of

RI
7 spinal spasticity. Med Biol Eng Comput 1987 Jul;25(4):439-42.

8 44. Priebe MM, Sherwood AM, Thornby JI, Kharas NF and Markowski J. Clinical

SC
9 assessment of spasticity in spinal cord injury: a multidimensional problem. Arch Phys Med

10 Rehabil 1996; 77: 713–716.

11
U
45. Nielsen JF and Sinkjaer T. A comparison of clinical and laboratory measures of
AN
12 spasticity. Mult Scler 1996; 1(5): 296-301.
M

13 46. Beres-Jones JA, Johnson TD, Harkem SJ. Clonus after human spinal cord injury

14 cannot be attributed solely to recurrent muscle-tendon stretch. Exp Brain Res. 2003 Mar
D

15 ;149(2):222-36.
TE

16 47. Miller L, Mattison P, Paul L and Wood L. The effects of transcutaneous electrical

17 nerve stimulation (TENS) on spasticity in multiple sclerosis. Mult scler 2007; 13:
EP

18 527-533.
C

19 48. Sonde L, Kalimo H and Viitanen M. Stimulation with High-frequency TENS

AC

20 effects on Lower Limb Spasticity after Stroke. Advances in Physiotherapy,

21 2000;2(4):183-187

22 49. Alabdulwahab SS, Al-Gabbani M. Transcutaneous electrical nerve stimulation of hip

23 adductors improves gait parameters of children with spastic diplegic cerebral palsy.

24 NeuroRehabilitation. 2010;26(2):115-22.
 23
ACCEPTED MANUSCRIPT
1 50. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, McQuay HJ.

2 Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control

3 Clin Trials. 1996 Feb;17(1):1-12.

PT
6 Supplier List

RI
7

SC
9 a. StataCorp LP, 4905 Lakeway Drive, College Station, Texas 77845-4512, USA

10 b. ITO Co. Ltd, 3-3-3 Toyotama-Minami, Nerima-ku,Tokyo 176-0014, Japan. E-mail:

11 itolator.co.jp
U
AN
12 c. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.
M

13

14
D

15 Figure Legend
TE

16

17
EP

18 Figure 1. Consort flow diagram

C

19
AC

20

21

22
 ACCEPTED MANUSCRIPT
Table 1. Baseline characteristics of patients included in the study.

Experimental group Control group p-value

(n=8) (n=8)

Age in years, mean±SD 33.38±13.606 40.50±13.774 0.248

PT
Gender, male, n 8 7 0.302

Time post-injury in months,

RI
mean±SD 3.25±1.753 3.13±1.959 0.789

Level of injury, paraplegia,n 4 4 1.00

SC
ASIA score A, n 3 3

U
ASIA score B, n 3 2
AN
ASIA score C, n 1 1

ASIA score D, n 1 2
M

Baseline CSS, mean(SD) 11.75±0.89 12.25±0.89 0.278

Baseline AJS, mean(SD) 2.88±0.64 3±0.76 0.727

D

Baseline MTS, mean(SD) 6.50±0.93 6.75±1.49 0.693

TE

Baseline ACS, mean(SD) 2.38±0.52 2.50±0.54 0.642

C EP

ASIA, American Spinal Injury Association; CSS, composite spasticity score; AJS, ankle jerk
AC

score; MTS, muscle tone score; ACS, ankle clonus score;

 ACCEPTED MANUSCRIPT
Table 2. Comparison of spasticity measurements in experimental group

IE IIE IIIE p-value p-value

(n=8) (n=8) (n=8) II-I III-I

CSS 11.75±0.89 10.25±1.49 8.75±0.89 0.002 <0.001

PT
AJS 2.88±0.64 2.75±0.71 2.63±0.52 0.08 0.170

RI
MTS 6.50±0.93 5.50±0.93 4.75±1.04 0.033 0.006

ACS 2.38±0.52 2.00±0.54 1.63±0.52 0.080 0.003

U SC
IE, baseline evaluation; IIE, evaluation after first intervention; IIIE, evaluation after final
AN
intervention.
M

CSS, composite spasticity score; AJS, ankle jerk score; MTS, muscle tone score; ACS, ankle

clonus score.
D

Data are expressed as mean±SD; Paired sample t-tests were used to determine significant
TE

differences over time. Level of significance set at P ≤ 0.01 for multiple comparisons, bold

type indicating significance.

C EP
AC
 ACCEPTED MANUSCRIPT
Table 3. Comparison of spasticity measurements in control group

IE IIE IIIE p-value p-value

(n=8) (n=8) (n=8) II-I III-I

CSS 11.50±0.76 10.88±0.84 10.38±1.19 0.95 0.051

PT
AJS 3±0.76 2.88±0.64 2.50±0.93 0.351 0.104

RI
MTS 6±1.07 5.75±0.71 5.50±0.93 0.598 0.351

ACS 2.50±0.54 2.25±0.46 2.38±0.74 0.170 0.351

U SC
IE, baseline evaluation; IIE, evaluation after first intervention; IIIE, evaluation after final
AN
intervention.
M

CSS, composite spasticity score; AJS, ankle jerk score; MTS, muscle tone score; ACS, ankle

clonus score
D

Data are expressed as mean±SD; Paired sample t-tests were used to determine significant
TE

differences over time. Level of significance set at P ≤ 0.01 for multiple comparisons, bold

type indicating significance.

C EP
AC
 ACCEPTED MANUSCRIPT
Table 4. Comparison of spasticity measurements in experimental and control groups

IE IIE IIIE

Mean±SD Mean±SD Mean±SD

Experimental n=8 n=8 n=8

PT
Control n=8 n=8 n=8

RI
CSS Experimental 11.75±0.89 10±1.07 9±1.20

Control 12.25±0.89 11.63±0.92 11.13±0.84

SC
p-value 0.278 0.006 0.001

mean diff 0.5 1.63 2.13

(99% CI) (0.82 to 1.82)

U (0.14 to 3.11) (0.59 to 3.66)
AN
M

AJS Experimental 2.88±0.64 2.50±0.76 2.63±0.52

Control 3±0.76 2.88±0.64 2.50±0.93

D

p-value 0.727 0.303 0.744

TE

mean diff 0.13 0.38 -0.13

(99% CI) (-0.92 to 1.17) (-0.67 to 1.42) (-1.24 to 0.99)

C EP

MTS Experimental 6.50±0.93 5.50±0.93 4.75±1.04

AC

Control 6.75±1.49 6.50±0.93 6.25±0.71

p-value 0.693 0.049 0.005

mean diff 0.25 1 1.5

(99% CI) (-1.60 to 2.10) (-0.38 to 2.38) (0.15 to 2.85)

ACS Experimental 2.38±0.52 2.0±0.54 1.63±0.52

 ACCEPTED MANUSCRIPT
Control 2.50±0.54 2.25±0.46 2.38±0.74

p-value 0.642 0.334 0.035

mean diff 0.13 0.25 0.75

(99% CI) (-0.66 to 0.91) (-0.49 to 0.99) (-0.20 to 1.70)

PT
RI
IE, baseline evaluation; IIE, evaluation after first intervention; IIIE, evaluation after final

intervention.

SC
CSS, composite spasticity score; AJS, ankle jerk score; MTS, muscle tone score; ACS, ankle

clonus score; CI, confidence interval

U
Data are expressed as mean±SD; Independent t-tests were used to determine significant
AN
differences between groups. Level of significance set at P ≤ 0.01 for multiple comparisons,
M

bold type indicating significance.

D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Enrollment Assessed for eligibility (n=24)

Excluded (n=8)

 Not meeting inclusion criteria (n=6)

PT
 Declined to participate (n=2)

16 patients recruited & randomized

RI
Baseline Assessment was done. (n=16)

SC
Allocation

U
Allocated to experimental intervention (n=8) Allocated to control intervention (n=8)
AN
 Received TENS application  Received usual physical therapy

Usual physical therapy

M

Assessment
D

Immediately after first intervention (n=8) Immediately after first intervention (n=0)
TE

Assessment
EP

Immediately after final intervention (n=8) Immediately after final intervention (n=8)
C

Analysis
AC

Analysed (n= 8) Analysed (n= 8)