Disability and Rehabilitation

ISSN: 0963-8288 (Print) 1464-5165 (Online) Journal homepage: http://www.tandfonline.com/loi/idre20

Effects of transcutaneous electrical nerve

stimulation alone or as additional therapy on
chronic post-stroke spasticity: systematic review
and meta-analysis of randomized controlled trials

Miriam Allein Zago Marcolino, Melina Hauck, Cinara Stein, Jociane

Schardong, Aline de Souza Pagnussat & Rodrigo Della Méa Plentz

To cite this article: Miriam Allein Zago Marcolino, Melina Hauck, Cinara Stein, Jociane Schardong,
Aline de Souza Pagnussat & Rodrigo Della Méa Plentz (2018): Effects of transcutaneous electrical
nerve stimulation alone or as additional therapy on chronic post-stroke spasticity: systematic
review and meta-analysis of randomized controlled trials, Disability and Rehabilitation, DOI:
10.1080/09638288.2018.1503736

To link to this article: https://doi.org/10.1080/09638288.2018.1503736

Published online: 16 Oct 2018.

Submit your article to this journal

Article views: 69

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=idre20
DISABILITY AND REHABILITATION
https://doi.org/10.1080/09638288.2018.1503736

REVIEW ARTICLE

Effects of transcutaneous electrical nerve stimulation alone or as additional

therapy on chronic post-stroke spasticity: systematic review and meta-analysis of
randomized controlled trials
Miriam Allein Zago Marcolinoa , Melina Hauckb , Cinara Steinb , Jociane Schardonga ,
Aline de Souza Pagnussata,b and Rodrigo Della M
ea Plentza,b
a
Graduate Program in Rehabilitation Sciences, Universidade Federal de Ci^encias da Sa
ude de Porto Alegre (UFCSPA), Porto Alegre, Brazil;
b
Graduate Program in Health Sciences, Universidade Federal de Ci^encias da Sa
ude de Porto Alegre (UFCSPA), Porto Alegre, Brazil

ABSTRACT ARTICLE HISTORY

Purpose: To evaluate the effects and to compare transcutaneous electrical nerve stimulation protocols, Received 11 December 2017
alone or as additional therapy in chronic post-stroke spasticity through a systematic review and meta-ana- Revised 5 June 2018
lysis of randomized clinical trials. Accepted 19 July 2018
Methods: Search was conducted in MEDLINE, Cochrane Library, EMBASE and Physiotherapy Evidence
KEYWORDS
Database through November 2017 (CRD42015020146). Two independent reviewers performed articles Muscle spasticity;
selection, data extraction and methodological quality assessment using the Cochrane Collaboration’s risk transcutaneous electric
of bias tool. The main outcome was spasticity assessed with Modified Ashworth Scale or other valid scale. nerve stimulation; stroke;
Meta-analysis was conducted using random effects method, and pooled-effect results are mean difference systematic review;
with 95% confidence interval. meta-analysis
Results: Of 6506 articles identified, 10 studies with 360 subjects were included in the review.
Transcutaneous electrical nerve stimulation alone or as additional therapy is superior to placebo TENS to
reduce post-stroke spasticity assessed with Modified Ashworth Scale (
0.52 [
0.74 to
0.30] p < 0.0001,
6 studies), especially in lower limbs (
0.58 [
0.82 to
0.34] p < 0.0001, 5 studies), which is in accordance
with the studies that used other scales. Low frequency TENS showed a slightly larger improvement than
high-frequency, but without significant difference between subgroups. Most studies present low or
unclear risk of bias.
Conclusion: Transcutaneous electrical nerve stimulation can provide additional reduction in chronic post-
stroke spasticity, mainly as additional therapy to physical interventions. Studies with better methodo-
logical quality and larger sample are needed to increase evidence power.

ä IMPLICATIONS FOR REHABILITATION

 Transcutaneous electrical nerve stimulation as additional treatment to physical interventions can lead
to additional reduction in chronic post-stroke spasticity.
 High and low frequency transcutaneous electrical nerve stimulation showed similar results, with
a smaller numerical superiority of low frequency TENS.
 More studies are needed to substantiate the best protocol of transcutaneous electrical nerve
stimulation to the treatment of spasticity.

Introduction Spasticity can reduce quality of life, functionality, and daily activ-
ities [6–9] because of articular deformities, disability, and pain [8].
Stroke is a cardiovascular non-communicable disease and the
However, spasticity may have positive effects by maintaining
main cause of acquired adult disability [1]. It is a global health-
muscle mass, and density bone by involuntary contraction, allow-
care problem. Fifteen million people worldwide suffer a stroke
ing orthostasis and gait [10,11]. Thus, patients’ treatment must
annually and five million are left permanently disabled [2]. This
focus on function improvement through combinations of thera-
disease is responsible for the greatest burden of disability-
peutic modalities considering spasticity risks and benefits [10].
adjusted-years, but more than 90% of these are attributable to
But, the effectiveness of each component of care falls short of evi-
modifiable risk factors [3]. Stroke incidence is declining as result
dence-based practice standards [5].
of better control of risk factors, but absolute numbers continue to
Transcutaneous electrical nerve stimulation (TENS) is a safe,
increase because of the ageing population [2].
low-cost, non-invasive treatment mainly known as a resource to
Deficits in motor control and limited mobility are the most rec-
treat pain conditions by selectively stimulating A-beta fibres, caus-
ognized impairment caused by stroke [1]. Chronic stroke survivors
ing pre-synaptic inhibition [12,13]. High-frequency TENS also
also may have spasticity (42%) [4] accompanied by hemiplegia [5].

CONTACT Rodrigo Della M
ea Plentz roplentz@yahoo.com.br, rodrigop@ufcspa.edu.br Department of Physical Therapy, Universidade Federal de Ci^encias da
Sa

ude de Porto Alegre (UFCSPA), 245 Sarmento Leite, 90050-170, Porto Alegre, Rio Grande do Sul, Brazil
ß 2018 Informa UK Limited, trading as Taylor & Francis Group
2 M. A. Z. MARCOLINO ET AL.

stimulates d-opioid receptors at the supra-spinal and spinal cord Methods

level and low-frequency stimulates l-opioid [12,14,15]. Both fre-
quencies in the brainstem may stimulate serotoninergic, noradre-
nergic, muscarinic and gabaergic systems [13,14] suggesting that
it can have effects other than analgesia. The TENS effect on the
spasticity of patients with upper motor neuron lesion was eval-
uated by several clinical trials [16–25]. It may act over spasticity
by stimulating Ia [16,26,27] and Ib afferent fibres [17] which acti-
vate local mechanisms in the spinal cord causing presynaptic
inhibition of efferent fibres [13,14].
Positive effects of TENS on spasticity, functionality, and gait
were shown by three recent systematic reviews [28–30]. Two
included different conditions without quantitative analysis and
reported a variety of protocols [28,29]. Mills and Dossa suggest
superiority in a combination of TENS with physical treatments
[29]. Meta-analysis of Lin et al., showed TENS can reduce spasticity
in stroke patients. However, only three studies, involving both
acute and chronic patients were included in spasticity analysis
[30]. Also, a commentary about this meta-analysis was published,
eliciting some major methodological failures and criticizing it as
ambiguous and misleading [31]. Thus, the effect of TENS on
chronic spasticity after stroke is still a matter for further analysis.
Clinical spasticity of chronic stroke survivors were generally
improved despite many different protocols with high (100 Hz)
[16,19–23,25] or low frequencies (5 Hz and 20 Hz) [17,18], and 30
[16,18–20,22] or 60 min [20,21,24] application time, besides fre-
quent association with another intervention [16,18–22,24,25].
However, functional improvement is not always followed by sig-
nificant spasticity reduction [16,24]. This way, in spite of the previ-
ous systematic reviews [28–30] effectiveness of TENS on chronic
post-stroke spasticity and the best application protocol remain
uncertain. Therefore, this study aims to systematically review
randomized clinical trials (RCT) which applied TENS alone or com-
bined with other therapies, compared to placebo or other inter-
ventions in chronic spasticity after stroke, to verify its general
efficacy and differences according to protocols characteristics by
meta-analysis.
Search strategy and selection criteria
This systematic review with meta-analysis was performed in
accordance with the Cochrane Collaboration [32] and Preferred
Reporting Items for Systematic Review and Meta-Analyses
Statement [33]. Electronic databases used for literature searches
(August 1968–November 2017) were MEDLINE (accessed by
PubMed), EMBASE, Cochrane Central Register of Controlled Trials
and Physiotherapy Evidence Database.
The search strategy used the individual or combined terms
“stroke”, “electric stimulation”, “transcutaneous electrical nerve
stimulation” and variations of these, and a previously proposed
sequence of words with high sensitivity in the search for RCTs
[34]. Words related to outcomes of interest were not included to
enhance the comprehensiveness of the search. Search terms were
adjusted to fit the requirements of each electronic database. No
restriction for date of publication and language was adopted.
Reference lists of selected articles were used as an additional
source of potential clinical trials. The complete search strategy
used for the PubMed database is shown in Table 1.
Independently, two reviewers (MAZM, MH) performed the
database searches and screened titles and abstracts of papers
according to a standard screening checklist based on eligibility
criteria. Inclusion criteria were RCT evaluating effects of TENS
alone or combined with other therapies in chronic spasticity of
stroke patients compared to placebo or other interventions. There
had to be more than one TENS application lasting at least 10 min
per intervention. Publication must be available in English, Spanish
or Portuguese languages. The primary outcome was spasticity
assessed by a validated clinical scale. The secondary outcome was
range of motion (ROM) assessed by goniometry. Full-text versions
of selected RCT (potentially eligible and uncertain) were inde-
pendently retrieved for complete review for two reviewers to
determine eligibility. Full-text articles should have presented
protocol intervention characteristics defined, such as frequency,
pulse duration, intensity, and application time. Exclusion criteria

Table 1. Search strategy on literature used for database medline accessed on PubMed.
(#1) Patient “Stroke”[Mesh] OR “Stroke” OR “Strokes” OR “Apoplexy” OR “CVA (Cerebrovascular Accident)” OR “CVAs (Cerebrovascular
Accident)” OR “Cerebrovascular Accident” OR “Cerebrovascular Accidents” OR “Cerebrovascular Apoplexy” OR “Apoplexy,
Cerebrovascular” OR “Cerebrovascular Stroke” OR “Cerebrovascular Strokes” OR “Stroke, Cerebrovascular” OR “Strokes,
Cerebrovascular” OR “Vascular Accident, Brain” OR “Brain Vascular Accident” OR “Brain Vascular Accidents” OR “Vascular
Accidents, Brain” OR “Cerebral Stroke” OR “Cerebral Strokes” OR “Stroke, Cerebral” OR “Strokes, Cerebral” OR “Stroke, Acute”
OR “Acute Stroke” OR “Acute Strokes” OR “Strokes, Acute” OR “Cerebrovascular Accident, Acute” OR “Acute Cerebrovascular
Accident” OR “Acute Cerebrovascular Accidents” OR “Cerebrovascular Accidents, Acute”
(#2) Intervention "Electric Stimulation"[Mesh] OR “eletric stimulation” OR “Electrical Stimulation” OR “Electrical Stimulations” OR “Stimulation,
Electrical” OR “Stimulations, Electrical” OR “Stimulation, Electric” OR “Electric Stimulations” OR “Stimulations, Electric” OR
"Electric Stimulation Therapy"[Mesh] OR "Electric Stimulation Therapy" OR “Stimulation Therapy, Electric” OR “Therapeutic
Electrical Stimulation” OR “Electrical Stimulation, Therapeutic” OR “Stimulation, Therapeutic Electrical” OR “Therapy, Electric
Stimulation” OR “Electrotherapy” OR “Therapeutic Electric Stimulation” OR “Electric Stimulation, Therapeutic” OR “Stimulation,
Therapeutic Electric” OR “Electrical Stimulation Therapy” OR “Stimulation Therapy, Electrical” OR “Therapy, Electrical
Stimulation” OR "Transcutaneous Electric Nerve Stimulation"[Mesh] OR "Transcutaneous Electric Nerve Stimulation" OR “Electrical
Stimulation, Transcutaneous” OR “Stimulation, Transcutaneous Electrical” OR “Transcutaneous Electrical Stimulation” OR
“Percutaneous Electric Nerve Stimulation” OR “Percutaneous Electrical Nerve Stimulation” OR “Transdermal Electrostimulation”
OR “Electrostimulation, Transdermal” OR “Transcutaneous Electrical Nerve Stimulation” OR “Transcutaneous Nerve Stimulation”
OR “Nerve Stimulation, Transcutaneous” OR “Stimulation, Transcutaneous Nerve” OR “Electric Stimulation, Transcutaneous” OR
“Stimulation, Transcutaneous Electric” OR “Transcutaneous Electric Stimulation” OR “TENS” OR “Electroanalgesia” OR “Analgesic
Cutaneous Electrostimulation” OR “Cutaneous Electrostimulation, Analgesic” OR “Electrostimulation, Analgesic Cutaneous”
(#3) Type of Study randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized controlled trials [mh] OR random allocation [mh]
OR double-blind method [mh] OR single-blind method [mh] OR clinical trial [pt] OR clinical trials [mh] OR ("clinical trial"[tw])
OR ((singl[tw] OR doubl[tw] OR trebl[tw] OR tripl[tw]) AND (mask[tw] OR blind[tw])) OR ("latin square"[tw]) OR placebos
[mh] OR placebo[tw] OR random[tw] OR research design [mh: noexp] OR comparative study [mh] OR evaluation studies [mh]
OR follow-up studies [mh] OR prospective studies [mh] OR crossover studies [mh] OR control[tw] OR prospectiv[tw] OR volun-
teer[tw]) NOT (animal [mh] NOT human [mh]
Search #1 AND #2 AND #3

were TENS application in non-spastic patients and other regions
than upper or lower limbs and intensities that produce strong
contractions with angular movement.
Meta-analysis inclusion criteria for spasticity were an assess-
ment using the Modified Ashworth Scale (MAS). We did not create
a unified scale for meta-analysis because other scales may con-
sider tendon jerk and clonus evaluation besides resistance to pas-
sive articular displacement which is the only evaluation
considered in MAS. All meta-analysis considers summary estimates
from included studies.
Disagreements were discussed between reviewers or arbitrated
by a third reviewer (SC) when a consensus was not reached. The
authors of the papers were contacted via e-mail to obtain explan-
ations of studies without sufficient information to evaluate eligi-
bility criteria. After contact, papers with insufficient information
were excluded. The study protocol was registered in the
International Prospective Register of Systematic Reviews
(PROSPERO: CRD42015020146) and can be fully assessed online at
http://www.crd.york.ac.uk/PROSPERO/display_record.
asp?ID=CRD42015020146.
Data analysis
Methodological design, number of subjects, comparison groups,
intervention protocol, and results of outcomes of interest were
extracted from selected full-text papers by two independent
reviewers (authors MAZM and MH). The main summary measures
were the end of intervention mean and standard deviation (SD).
Authors of studies with missing data were contacted by e-mail to
obtain results for synthesis or study quality assessment, or, when-
ever possible, techniques were used to estimate the summary
measures (when needed, data were extracted from figures or
graphs using the WebPlotDigitizer software version 4.0, and 95%
confidence intervals (95%CI) or standard errors were mathematic-
ally converted to SD). The study was excluded from meta-analysis
if data were still insufficient after this process.
The estimated pooled-effect was obtained by comparison of
the end of intervention scores of each group studied for meta-
analysis. The planned analysis in the protocol, was to compare
change of scores, however, considering that it may be less effi-
cient for outcomes which are unstable or difficult to measure pre-
cisely [32], and that this analysis showed to be more likely to
produce significant results [35], we decided to change the analysis
to end of intervention scores. Data from follow-up were not
included in the meta-analysis. Results were presented as weighted
mean difference (MD) with 95%CI. Analyses were performed using
the random effects method. The p-value 0.05 was considered
statistically significant. Statistical heterogeneity among studies
was assessed by chi-squared test (Chi2) and inconsistency test (I2).
I2 interpretation follows the thresholds described in Cochrane’s
Handbook (0%–40%: heterogeneity might not be important;
30%–60%: may represent moderate heterogeneity; 50%–90%: may
represent substantial heterogeneity; 75%–100%: considerable het-
erogeneity. I2 values in intersections of these thresholds were
judged in accordance with the strength of evidence for hetero-
geneity (p-value from the Chi2 test) [32].
For studies with more than one possible comparator to TENS,
we included data of only one comparison group in meta-analysis.
The criterion was the comparability between groups, which mean
that the only difference between the intervention group and the
comparator included in meta-analysis must be active TENS stimu-
lation in the intervention group. Sensitivity and subgroup analysis
was performed to explore differences in protocols considering the
TENS ON POST-STROKE SPASTICITY: A META-ANALYSIS 3
application site of TENS and protocol of intervention. All analyses
were done by Review Manager, version 5.3.5.
Risk of Bias Assessment
Two independent reviewers (authors MAZM and MH) assessed risk
of bias of included studies by considering items set by the
Cochrane Collaboration tool for assessing risk of bias (within and
across randomized trials): random sequence generation, allocation
concealment, blinding of participants and personnel, blinding of
outcome assessment, incomplete outcome data, selective report-
ing and other biases (considering the other possible sources sug-
gested by the Cochrane Collaboration). The judgment was
categorized as low, high or unclear risk of bias [32].
Results
Studies characteristics
Ten RCT [16–25] from 6506 records found in database searches
fulfilled eligibility criteria and were included in this systematic
review providing data from 360 patients. Figure 1 shows the flow
diagram of the selection process.
TENS was applied in continuous mode with high-frequency
(100 Hz) in seven articles [16,19–23,25], and low-frequency (20 Hz
[17] and 5 Hz [18]) in two. Pulse duration was adjusted between
0.1 and 0.3 ms in eight studies [16,17,19–23,25]. Intensity set
included subsensory threshold [22], sensory threshold [25], 50 mA
[23], maximum without muscle contraction [17], patient tolerance
[18], two to three times the sensory threshold [16,19–21] and vis-
ible muscle contractions [24]. A TENS burst with a frequency of
1.7 Hz in eight pulses with 0.14 ms was applied only in one study
[24]. Intervention time varies from 15 [23], 20 [17], 30
[16,18–20,22,25] to 60 min [20,21,24] applied once a day
[16,17,19–25] or twice a day [18], five [16,18–25] or six [17] days
per week for one [18], two [23], four [16,17,19,21], six [20,25] or
12 [24] weeks. The total time of TENS stimulation delivered was
150 [23], 300 [18], 480 [17], 600 [16,19], 900 [20,22,25], 1,200 [21],
1800 [20] and 3600 min [24]. Interventions were carried out in an
outpatient setting [16–23,25] and in the patient’s home [24].
TENS was applied alone in three studies [14,18,20]. Other stud-
ies used TENS combined with physical therapy [22,24,25], task-
related training (TRT) [16,20,21], Bobath inhibitory techniques [19]
or botulinum toxin A and stretching [18]. Combined interventions
were delivered at the same weekly frequency and weeks as TENS
[16,17,19–23,25], except for one study that applied TENS 5 days a
week and stretching for 7 days [18] and one study that performed
TENS 5 days a week and physical therapy twice a week [24].
Comparison groups included placebo TENS alone [17], placebo
TENS combined with interventions cited above [16,19–22,25],
physical therapy alone [24], therapeutic ultrasound and botulinum
toxin A alone [23] or botulinum toxin A combined with taping or
stretching [18] with the same weekly frequency and weeks as
intervention group, except for the botulinum toxin alone group
from one study that performed one application of botulinum
toxin A [18]. Studies characteristics and protocols are described in
Table 2. Assessment of spasticity was performed with MAS in
seven studies [16–19,22–24], and in other studies, it was per-
formed with the Composite Spasticity Scale [21,25] and Modified
Composite Spasticity Scale [20].
4 M. A. Z. MARCOLINO ET AL.

Identification
Records identified through database
Additional records identified
searching (n = 6,506): MEDLINE
through other sources
(3,294), EMBASE (2,817)
(n = 0)
COCHRANE (375) and PEDRO (20)

Duplicates removed (n = 935)

Screening

Records excluded in title

Records screened
and abstract
(n = 5,571)
(n =5,542)

Full-text articles assessed

Full-text articles excluded
for eligibility
because of (n = 19):
Eligibility

(n = 29)
11 articles another method
of intervention
2 articles verify only
immediate effect
3 articles do not assessed
Studies included in spasticity
qualitative synthesis 1 article with acute period
(n = 10) of disease
1 article with cerebral palsy
and stroke subjects
Included

1 article in other language

Studies included in quantitative

synthesis (meta-analysis) for stroke
(n = 6)

Figure 1. Flow diagram of study selection process.

Risk of bias
According to the risk of bias assessment, most information is from
studies with low or unclear risk of bias (Figure 2). Only two stud-
ies showed a high risk of bias in allocation concealment [23],
blinding of outcome assessment and selective reporting. The allo-
cation concealment in the study of Picelly et al. was considered
to have a high risk of bias because it refers that one of the inves-
tigators checked correct patient allocation according to the ran-
domization list, thus the allocation was not concealed to all
investigators [23]. High risk of bias for blinding of outcome assess-
ment was considered for the study of Sonde et al. because it
reports no blinding procedures and the description of the ran-
domization and assessment denotes that the baseline assess-
ments were made after randomizing the patient [24]. The same
study was considered to have a high risk of bias for selective
reporting because provided insufficient data for the outcomes
measured. Also, the study delivered the TENS equipment to the
intervention group use it at home, however, there was no
description of a monitoring process, with possible low adherence
or misuse, thus, we considered it as a possible bias source, with
uncertain risk [24]. Laddha et al. was considered to have unclear
risk for other bias because the authors do not reported any base-
line measurements the participants, just refers no statistical base-
line difference between the patients that ended the study, thus,
baseline imbalance between randomized participants could not
be assessed. None study with high risk of bias was included in
our meta-analysis [20].
Interventions effects
Spasticity – results from meta-analysis
Meta-analysis for spasticity includes five articles that assessed spas-
ticity with MAS and presented at least one comparable group
[16–19,22]. TENS, alone or combined with other therapies, is super-
ior to the same therapies with placebo TENS or with placebo TENS
alone to reduce spasticity with no heterogeneity with a MD of

0.57 point in MAS [95%CI ¼
0.82 to
0.33; I2 ¼ 0%] p < 0.0001;
n ¼ 128 – Figure 3(a)). A sensitivity analysis considering only studies
with the application to lower limbs was carried out with little
change in the result, with a MD of
0.62 point [95%CI ¼
0.90 to

0.34; I2 ¼ 0%] p < 0.0001; n ¼ 98 – Figure 3(b)) [17–19,22].
To explore differences in protocols, a subgroup analysis was
performed considering the frequency of TENS. Application of high-
frequency TENS added to physical interventions (physical therapy
[22], Bobath inhibitory techniques [19] or TRT [16]) was superior to
comparator group with placebo TENS added to physical interven-
tions with a MD of
0.49 points in MAS ([95%CI ¼
0.78 to
0.19;
I2 ¼ 0%] p ¼ 0.004; n ¼ 109 – Figure 4(a)) [16,19,22]. Application of
low-frequency TENS, 5 Hz added to stretching and botulinum toxin
A [18] or 20 Hz alone [17] was superior to comparator group with a
Table 2. Characteristics of the studies included in the systematic review.
Male Age ± SD
Author, Year Participants Intervention Comparators N gender (years-old) Protocol
Baricich et al., Chronic hemiple- 1) BTA þ TENS þ s- 2) BT-A þ Taping 1) 8 1) 5 1) 64.7 ± 7.3 All groups – BTA: injected into lateral and
2008 [18] gic adults after tretching 3) BT- 2) 8 2) 3 2) 60.3 ± 12 medial head of gastrocnemius on hemiplegic
stroke with spastic A þ Stretching 3) 7 3) 5 3) 61.2 ± 5.7 side (500 IU in 1.25 mL of 0.9% saline solution,
equinus foot adjusted in 150–250 IU range for each muscle
head according with spastic hypertonia grade
and muscle size)
1) TENS 1 Stretchinga:
Electrode Placement: Gastrocnemius
F: 5Hz
PD: not informed
Intensity: adjusted to patient’s tolerance
Volume: twice a day during 30min for 5 days
¼ 300min
2) Taping:
Adhesive taping at ankle and thigh during 5
days checked daily by physical therapist
3) Stretchinga:
Of calf muscles, twice a day during 30min for 7
days

Chen et al., Stroke patients 1) TENS 2) Placebo TENS 1) 12 1/2) 14 1/2) 57 (41-69) 1) TENSa:
2005 [17] with evident 2) 12 Electrode Placement: Junction of gastrocnemius
ankle spasticity muscle and Achilles tendon
F: 20 Hz
PD: 0.2 ms
Intensity: adjusted to maximum without induc-
ing muscle contraction
Volume: 20 min, 6 times a week, for 4 weeks
¼ 480 min
2) Placebo TENSa:
Same protocol as described in TENS group, with
intensity kept at zero.

Hussain et al., Stroke spas- 1) TENS þ Bobath 2) Bobath inhibi- 1) 15 1) 10 1) 53.6 ± 7.35 1) TENS 1 Bobath inhibitory techniquesa:
2013 [19] tic patients inhibitory tory techniques 2) 15 2) 8 2) 57.6 ± 6.53 Electrode Placement: in leg acupuncture points
techniques (ST 36, LV 3, GB 34, UB 60)
F: 100 Hz
PD: 0.2 ms
Intensity: 2 to 3 times the sensory threshold
Volume: 30 min, 6 times a week, for 4 weeks
¼ 600 min
2) Bobath inhibitory techniquesa:
Passive extension movement of big toe and
other four toes, ankle joint dorsiflexion, knee
joint extension, abduction and external rotation
of hip joint for 15 min, 6 times a week, for 4
weeks.

Jung et al., Chronic stroke 1. TENS þ Phys- 1. Placebo 1) 20 2) 20 1) 56.2 ± 10.4 All groups – Physiotherapy: Conventional ther-
2017 [25] patients with iotherapy TENS þ Phys- 2) 56.3± apy during 1 hour a day, and sit-to-stand train-
moderate to iotherapy 10.2 ing during 15min, after TENS or Placebo TENS, 5
TENS ON POST-STROKE SPASTICITY: A META-ANALYSIS

severe spasticity sessions a week over 6 weeks, under supervision

in the plantar of two licensed therapists.
(continued)
5
 6
Table 2. Continued.
Male Age ± SD
Author, Year Participants Intervention Comparators N gender (years-old) Protocol
flexors 1) TENS 1 Physiotherapy:
Electrode Placement: over the peroneal nerve
F: 100 Hz
PD: 0.2 ms
Intensity: at sensory threshold, defined as the
minimum tightly sensation felt by the patients.
Volume: 30 min, 5 times a week over 6 weeks
¼ 900 min
2) Placebo TENS 1 Physiotherapy:
M. A. Z. MARCOLINO ET AL.

Same protocol as TENS group, without real

stimulation.

Kim et al., Chronic stroke 1) TENS þ TRT 2) 1) 15 1) 9 1) 63.3 ± 8.3 All groups – TRT: arm or hand tasks related
2013 [16] patients with par- Placebo 2) 15 2) 8 2) 61.3 ± 9.9 to functional moments needed in daily life, dur-
etic arms TENS þ TRT ing 30min, 5 sessions a week over 4 weeks
before TENS or Placebo TENS
1) TENS 1 TRTa:
Electrode Placement: Muscle belly of triceps and
wrist extensors
F: 100 Hz
PD: 0.2 ms
Intensity: 2 to 3 times the sensory threshold
Volume: 30 min, 5 times a week over 4 weeks
¼ 600 min
2) Placebo TENS 1 TRTa:
Same protocol as TENS group, with intensity
kept at zero

Laddha et al., Chronic stroke 1) TENS (30 min) 2) TRT 1) 10 1) 8 1/2/3) 46.4 ± 6.9 All groups – TRT: general and individual task
2015 [20] patients þ TRT 3) TENS (60 min) 2) 10 2) 7 oriented exercises guided by physical therapists.
with spasticity þ TRT 3) 10 3) 4 All exercises were initially repeated 15 times
without compensation during 60min, 5 times a
week over 6 weeks, after application of TENS
1) TENS 30 min 1 TRTa
Electrode Placement: One on head of fibula over
common peroneal nerve, and another over the
belly of tibialis anterior lateral to proximal shin
of tibia
F: 100 Hz with short pulse duration around
50 ms
PD: 0.2 ms
Intensity: 2 to 3 times the sensory threshold
Volume: 30 min, 5 times a week over 6 weeks
¼ 900 min
2) TRTa
General exercises based on Ng and Hui-Chan
(2007) TRT protocol (describe next)
Individual exercises include balance exercises,
strength training and arm-training, with restraint
of the less impaired arm, training functional
tasks repetitively
3) TENS 60 min 1 TRT
Same electrode placement and TENS parameters
as TENS 30 min þ TRT group
(continued)
Table 2. Continued.
Male Age ± SD
Author, Year Participants Intervention Comparators N gender (years-old) Protocol
Volume: 60min, 5 times a week over 6 weeks
¼ 1800min

Ng and Hui- Chronic stroke 1) TENS þ TRT 2) Placebo 1) 21 1) 16 1) 58..4 ± 7.1 1) TENS 1 TRT:
Chan, patients TENS þ TRT 2) 20 2) 17 2) 57.1 ± 7.8 TENS:
2007 [21] with spasticity 3) TENS alone 3) 19 3) 17 3) 56.4 ± 9.1 Electrode Placement: Leg acupuncture points
4) Control 4) 20 4) 17 4) 57.3 ± 8.6 (ST 36, LV 3, GB 34, UB 60
F: 100Hz
PD: 0.2ms
Intensity: 2 to 3 times the sensory threshold
Volume: 60min, 5 times a week, for 4 weeks
¼ 1200min
TRT:
4 weight bearing and stepping exercises 2.5 or
5cm high wooden blocks: (1) loading exercise
on affected leg; (2) stepping up exercise with
affected leg; (3) stepping down exercise with
unaffected leg; (4) heel lifts from a dorsiflexed
position in standing; and 2 functional training:
(1) standing up from a chair, walking a short
distance, and returning to chair; and (2) walking
with rhythmic auditory cues generated by a
metronome during 60 min, 5 times a week, for
4 weeks after TENS or placebo TENS.
Standardized progression was made by physical
therapist
2) Placebo TENS 1 TRT:
Same TRT training and electrode placement as
TENS þ TRT group, delivered from identical-look-
ing TENS devices with electrical circuit discon-
nected inside
3) TENS alone
Same TENS protocol as TENS þ TRT group
4) Control:
No treatment

Park et al., Hemiplegic 1) 2) Placebo 1) 15 1) 12 1) 71.2 ± 3.4 All groups – Physical Therapy: ROM, func-
2014 [22] stroke patients TENS þ Physiothe- TENS þ Physiothe- 2) 14 2) 8 2) 71.1 ± 3.8 tional and gait training with a physical therapist
rapy rapy during 30min, 5 days a week for 6 weeks
1) TENSa:
Electrode Placement: Lateral and medial quadri-
ceps and gastrocnemius
F: 100Hz
PD: 0.2ms
Intensity: Sensory threshold was measured from
0.1mA and stimulated by 90% amplitude using
sub-sensory threshold. The participant perceives
no sensation
Volume: 30min, 5 times a week over 6 weeks
¼ 900min
2) Placebo TENSa:
Same protocol as described in TENS group, with
TENS ON POST-STROKE SPASTICITY: A META-ANALYSIS

intensity kept at zero

7

(continued)
 8
Table 2. Continued.
Male Age ± SD
Author, Year Participants Intervention Comparators N gender (years-old) Protocol
Picelli et al., Stroke patients 1) TENS 2) BT-A 1) 10 1) 5 1) 62.7 ± 12.9 1) TENSa:
2014 [23] (hemorrhagic or 3) 2) 10 2) 8 2) 65.2 ± 5.5 Electrode Placement: Origin and mid belly of
ischemic) with Therapeutic 3) 10 3) 6 3) 64.2 ± 8.7 gastrocnemius
spastic equinus ultrasound F: 100 Hz
foot PD: 0.3 ms
Intensity: 50 mA, without inducing muscle con-
traction
Volume: 15 min, 5 times a week over 2 weeks
¼ 150 min
M. A. Z. MARCOLINO ET AL.

2) BT-A:
Applied at 2 sites in each head of gastrocne-
mius muscle near the muscle origin, and at
mid-belly of muscle bulk (100 units for each
head of gastrocnemius muscle, 50U per injec-
tion site)
3) Therapeutic ultrasounda:
Over mid-belly of gastrocnemius muscle bulk,
perpendicular to its surface
Mode: Continuous with treatment head surface
of 5cm2
F: 1 MHz
Intensity: 1.5 cm/W2
Volume: 10 min, 5 times a week for 2 weeks
¼ 100 min

Sonde et al., Stroke patients 1) Self-applied 2) Physiotherapy 1) 26 1) 19 1) 71 ± 6 1) TENS 1 Physical Therapy:

1998 [24] with paretic arm Low-frequency 2) 18 2) 8 2) 73 ± 3.5 Electrode Placement: Elbow extensors or shoul-
TENS þ Physiothe- der abductors
rapy F: 1.7 Hz
PD: 8 pulses with interval of 0.014 ms
Intensity: that produces visible muscle contrac-
tion
Volume: 60 min, 5 times a week over 12 weeks
¼ 3600 min
ES self-applied at home
2) Physical Therapy:
At a day-care center, twice a week over
12 weeks
CG: Control group; TENS: transcutaneous electrical nerve stimulation; F: frequency; PD: pulse duration; ES: electrical stimulation; BTA: botulinum toxin type A; TRT: task-related training.
a
Group included on meta-analysis.

Figure 2. Review author’s judgement about Risk of bias by Cochrane
Collaboration’s tool for assessing risk of bias within and across randomized trials.
(a) Risk of bias within studies; (b) Risk of bias across studies. Other potential bias
evaluated were related to proper intervention administration and base-
line imbalance.
MD of
0.76 [95%CI ¼
1.20 to
0.32; I2 ¼ 0%] p ¼ 0.0008; n ¼ 39
– Figure 4(b)). Test for subgroup differences showed no statistical
difference between groups (p ¼ 0.32; I2 ¼ 0%).
Spasticity – results from studies not included in meta-analysis
Sonde et al. showed low-frequency TENS (burst, 60 min) combined
with physical therapy didn’t reduce spasticity of biceps brachialis
(p ¼ 0.07) but did not present summary measures [24]. Ng and
Hui-Chan demonstrated high-frequency TENS (60 min) alone or
combined with TRT improved spasticity of plantar flexors two
weeks earlier and greater than TRT combined with placebo TENS
or control, however, the amount of change was small, with the
TENS ON POST-STROKE SPASTICITY: A META-ANALYSIS 9
reduction of around 1 point of composite spasticity scores in
those groups (p < 0.01) [21]. Jung et al., showed TENS (30 min)
combined with physical therapy after stimulation, is superior to
placebo TENS combined with physical therapy over plantar
flexor spasticity, with a mean reduction in composite spasticity
score of
2.6 with TENS and
0.7 with placebo (p < 0.001) [25].
Accordingly, Laddha et al. showed high-frequency TENS com-
bined with TRT reduced spasticity of biceps brachialis in relation
to TRT (p < 0.05), and the results of 60 min TENS application
were significantly greater than 30 min, a MD of around 4 points
in comparison with a mean difference of approximately 2 points
in modified composite spasticity score, respectively (p < 0.05)
[20]. Picelli et al. compared TENS with therapeutic ultrasound or
botulinum toxin A. Only TENS group and botulinum toxin A
group showed significant reduction of spasticity at the end of
intervention assessment, but the comparison between groups
showed no significant difference. In the assessments 15 days
and 75 days after the end of the intervention, botulinum toxin
A showed better improvement than both TENS and therapeutic
ultrasound (p < 0.05) without significant difference between the
last two [23].
Range of motion
Meta-analysis for ROM included tree studies [18,19,21]. Addition
of active TENS to physical interventions was superior to placebo
TENS, with a MD of 1.46 degrees of ROM assessed with goniom-
etry ([95%CI ¼ 0.19–2.73; I2 ¼ 0%] p ¼ 0.02; n ¼ 86 – Figure 5).
Picelli et al. showed improvement in ankle ROM only in the
botulinum toxin group, with statistical superiority to both TENS
and therapeutic ultrasound groups at the end of treatment and
other follow-up times (p < 0.01) [23]. Laddha et al. did not provide
data of ROM assessment, but its results showed TENS combined
with TRT improved the ankle ROM (p < 0.05) [20].
Discussion
This systematic review with meta-analysis shows that the addition
of active TENS to the treatment of spasticity of chronic stroke pro-
vides additional benefits. Most studies used TENS in addition to
physical treatments, including physical therapy, Bobath inhibitory
techniques or TRT. Its effect as a unique therapy needs further
investigation. By subgroup analysis to investigate differences
among protocols, no significant difference was found between
high and low-frequency of stimulation. Besides, TENS can increase
ankle dorsiflexion ROM. To the best of our knowledge, this is the
first systematic review with meta-analysis to evaluate the efficacy
and to compare different TENS interventions in spasticity of
chronic stroke patients.
Three systematic reviews about the effect of TENS on spasticity
were published while we were conducting our study [28–30].
Clinical, neurophysiological and functional evaluations of TENS in
different diseases were presented by Fern
andez-Tenorio et al.
[28]. Besides positive effects, the results pointed to the need for
specific studies about application parameters, and how TENS acts
on spasticity [23]. Mils and Dossa also included TENS application
in different diseases in upper and lower limbs [29]. TENS
improved spasticity and gait with level one and two of evidence.
Furthermore, TENS combined with exercises or TRT achieved the
best results [29].
Lin et al. performed a meta-analysis about the effect of TENS
in stroke patients. For spasticity, only three studies were included,
with acute and chronic patients [30]. As their meta-analysis
compared only final spasticity values, it showed that TENS
10 M. A. Z. MARCOLINO ET AL.

Figure 3. Forest plot of the results of meta-analysis and sensitivity analysis for Transcutaneous Electrical Nerve Stimulation (TENS) on spasticity assessed by Modified
Asworth Scale. Values presented as mean difference and confidence interval of 95%. The mean effect was obtained with a model of random effects. (a) TENS applica-
tion to upper or lower limbs; (b) TENS application to lower limbs only.

Figure 4. Forest plot of the results of subgroup analysis by protocol for Transcutaneous Electrical Nerve Stimulation (TENS) on spasticity assessed by Modified
Asworth Scale. Values presented as mean difference and confidence interval of 95%. The mean effect was obtained with a model of random effects. (a) High-fre-
quency TENS; (b) Low-frequency TENS.

Figure 5. Forest plot of the results of meta-analysis and sensitivity analysis for Transcutaneous Electrical Nerve Stimulation (TENS) on range of motion (ROM) assessed
by manual Goniometry. Values presented as mean difference and confidence interval of 95%. The mean effect was obtained with a model of random effects.

supplementation had reduced spasticity in comparison with the stroke, we can assume that our search was more accurate to find
control groups [30], but it does not represent how much change the existent evidence about this topic.
the addition of TENS causes in spasticity. Considering that our There is still much controversy about parameters of TENS in
meta-analysis included five studies [16–19,22], only in chronic spasticity although its efficacy was recognized [14]. TENS may act

over spasticity by causing pre-synaptic inhibition of spastic
muscles [36,37]. TENS applied in the spastic muscle or its antag-
onist can interfere with pre-synaptic and reciprocal inhibition
mechanisms by the stimulation of Ia afferent fibres [11,16,26,27].
TENS on muscle-tendon junction may stimulate Ib afferent fibres,
responsible for the inhibitory pathway by Golgi tendon organ
[17]. Also, as before mentioned, high-frequency TENS and low-fre-
quency TENS stimulates d-opioid and l-opioid receptors, respect-
ively, at the supra-spinal and spinal cord level [12,14,15]. Both
frequencies in the brainstem may stimulate serotoninergic, nora-
drenergic, muscarinic and gabaergic systems [13,14] which can
interfere in the neuron excitability.
Subgroup analysis showed no significant difference between
low-frequency and high-frequency TENS. However, three studies
with high-frequency TENS and significant results in spasticity
weren’t included in the meta-analysis because MAS was not the
instrument to assess spasticity [20,21,25]. Laddha et al. showed
60 min TENS application was better than 30 min which suggests
that larger stimulation time of application may produce better
effect [20]. However, stimulation time of all studies included in
meta-analysis varied from 20 min to 30 min application in each
section, thus the effect of larger stimulation times was not ana-
lysed. Considering total stimulation time, low-frequency TENS
showed numerically grater effect with smaller total stimulation
time (300 min [18] and 480 min [17]) than high frequency
(600 min [16,19] and 900 min [22]). Only the study of Sonde et al.
does not find any improvement in spasticity using low-frequency
TENS. The burst mode applied in antagonist muscles was not able
to reduce spasticity as continuous mode applications present in
other studies, besides the largest total time of TENS application
(3600 min) [24]. No difference in results could be attributed to the
intensity set variation.
It is important to highlight that in all studies with high-fre-
quency TENS, it was added to physical interventions. Besides the
meta-analysed studies [13,16,19], the high-frequency TENS studies
assessing spasticity with other scales also found better results in
the combination of high-frequency TENS with TRT [20,21] and
physical therapy [25]. Combinations of therapies are usual and
recommended in clinical practice [10]. However, meta-analysed
studies with low-frequency TENS had combined it only with
stretching and botulinum toxin A [18], or as unique therapy [14].
None included study assessed the effect of continuous low-fre-
quency TENS as additional therapy to more complete physical
interventions such as physical therapy, TRT or Bobath inhibi-
tory techniques.
Our meta-analysis results showed that the addition of TENS
produces an additional reduction in spasticity of around 0.6
points in MAS. However, as there is no minimal clinical important
difference for spasticity assessed with MAS, we cannot determine
if our results are clinically important. Even though MAS is a widely
used measurement of spasticity in clinical practice [38], and fre-
quently used in clinical studies to assess the effect of a treatment
– as can be seen in the majority of the included studies of this
review [16–19,22–24] as well in previous systematic reviews with
meta-analysis including spasticity measurements [39–41], its valid-
ity and reliability has been criticized [38,42]. Various studies inves-
tigated the reliability of MAS in different populations and found
incompatible results for both same and different muscle groups
[43–46]. Regarding the muscle assessment included in our review,
MAS provided moderate to substantial test-retest reliability and
inter-rater reliability in spasticity measurement of hemiplegic
patients [42].
TENS ON POST-STROKE SPASTICITY: A META-ANALYSIS 11
Chronic spasticity can influence ROM through modifications of
muscle structure, and changes in articular connective tissue. It
reduces active and passive ROM, voluntary movements, gait sta-
bility, and daily life activities [47]. A combination of exercise and
TENS can effectively improve proprioception for each muscle
resulting in increased body orientation, and improvement of ankle
ROM may exert positive effects [22]. Our meta-analysis showed
TENS alone or combined with therapies improved ankle ROM by
1.46 degrees. As there is no established minimal clinically import-
ant difference for ankle dorsiflexion ROM [48], we cannot say
whether our results can represent a significant clinical improve-
ment. Interestingly, the study of Baricich et al. with TENS and
stretch versus stretch alone after botulinum toxin A found an
improvement of 8.75 degrees in the TENS group compared to
3.57 in the stretch group 15 days after end of intervention, sug-
gesting that changes in the rheological properties of spastic
muscles may need more time to be perceived [15].
As limitations of this systematic review, we can point to an
unclear risk of bias because of lack of information regarding
methodological design. In spite of trial report standards [49], stud-
ies don’t use checklists showing how the trial was designed, ana-
lysed, and interpreted. Inadequate allocation concealment leads
to selection bias and can impact significantly on the effect size of
continuous outcomes in physical therapy [50]. Most of the studies
included in this review showed no description of allocation
concealment and therefore the risk of bias for this criterion was
considered unclear. One study showed high risk of bias for this
criterion, but its results were not pooled on meta-analysis [23].
The evaluation of risk of bias about blinding of participants and
personnel presents a peculiarity because the outcome of spasti-
city cannot be influenced by the patient, thus blinding the patient
may not influence the results [51]. Non-pharmacological studies
with physical activities present a lot of limitations, such as learn-
ing curves, lack of blinding, low standardization of interventions
and co-intervention [39]. Besides the potential bias in the
included studies, the indirect measurement of spasticity using
clinical scales also represent a potential assessment bias in our
results, considering that it may not be sensitive enough to detect
minimal changes which render it difficult to perform an objective
quantification of the therapeutic effect of TENS in spasticity [17].
Also, lack of standardization of the TENS intervention could be
considered a limitation of this systematic review, because its vari-
ation can influence results, and make it difficult to extrapolate
synthesized data. Nevertheless, differences in TENS protocols
allow us to assess possible differences in their effectiveness.
Finally, as we pre-specified only two outcomes, spasticity [16–25]
and ROM [18–21,23], the effects of TENS in other outcomes that
are highly important to stroke patients such as walk speed
[17,19–22], balance [22,25], strength [25] and functionality [16,24]
still matter further evaluation.
Conclusion
This systematic review with meta-analysis showed TENS as add-
itional treatment to physical interventions can lead to an add-
itional reduction in chronic spasticity of stroke patients. Although,
we used only indirect measurement of spasticity by scales, which
may fail to provide an objective measurement of spasticity. The
evaluation of protocols showed numerically, but non-significant
difference between high and low-frequency TENS. However, few
studies were included with very different protocols, which may
influence results and reduce the extrapolation of synthesized
data. Therefore, it is suggested to conduct new large clinical trials
12 M. A. Z. MARCOLINO ET AL.
applying TENS in chronic spasticity, especially with low-frequency
associated with physical treatments, to verify the best protocol of
application and results.
Acknowledgements
We would like to thank Coordenaç~ao de Aperfeiçoamento de
Pessoal de N
ıvel Superior (CAPES).
Disclosure statement
No potential conflict of interest was reported by the authors.
ORCID
Miriam Allein Zago Marcolino http://orcid.org/0000-0001-
5333-3173
Melina Hauck http://orcid.org/0000-0001-9202-7574
Cinara Stein http://orcid.org/0000-0003-1833-8221
Jociane Schardong http://orcid.org/0000-0003-3635-8541
Aline de Souza Pagnussat http://orcid.org/0000-0001-
7837-5855
Rodrigo Della M
ea Plentz http://orcid.org/0000-0002-2662-8192
References
[1] Langhorne P, Coupar F, Pollock A. Motor recovery after
stroke: a systematic review. Lancet Neurol. 2009;8(8):741–754.
[2] WHO. Programmes. Cardiovascular disease. The atlas of
heart disease and stroke. Global burden of stroke
[Internet]. World Health Organization; 2004 [cited 2016 Aug
30] Available from: http://www.who.int/cardiovascular_dis-
eases/en/cvd_atlas_15_burden_stroke.pdf?ua¼.
[3] Hachinski V. Stroke is a burdensome but preventable brain
disorder. Lancet Neurol. 2016;15(9):892–893.
[4] Wissel J, Manack A, Brainin M. Toward an epidemiology of
poststroke spasticity. Neurology. 2013;80(3 Suppl 2):
S13–S19.
[5] Dobkin, B. Strategies for stroke rehabilitation. Lancet
Neurol. 2004;3(9):528–536.
[6] Welmer AK, von Arbin M, Wid
en Holmqvist L, et al.
Spasticity and its association with functioning and health-
related quality of life 18 months after stroke. Cerebrovasc
Dis. 2006;21(4):247–253.
[7] Hsu AL, Tang PF, Jan MH. Analysis of impairments influenc-
ing gait velocity and asymmetry of hemiplegic patients
after mild to moderate stroke. Arch Phys Med Rehabil.
2003;84(8):1185–1193.
[8] Gillard PJ, Sucharew H, Kleindorfer D, et al. The negative
impact of spasticity on the health-related quality of life of
stroke survivors: a longitudinal cohort study. Health Qual
Life Outcomes. 2015;13:159.
[9] Doan QV, Brashear A, Gillard PJ, et al. Relationship between
disability and health-related quality of life and caregiver
burden in patients with upper limb poststroke spasticity.
PM R. 2012;4(1):4–10.
[10] Bethoux F. Spasticity Management After Stroke. Phys Med
Rehabil Clin N Am. 2015;26(4):625–639.
[11] Stevenson V, Playford D. Neurological rehabilitation and
the management of spasticity. Medicine. 2012;40(9):
513–517.
[12] Tashani O, Johnson M. Transcutaneous electrical nerve
stimulation (tens) a possible aid for pain relief in develop-
ing countries? Libyan J Med. 2009;4(2):62–65.
[13] Machado AF, Santana EF, Tacani PM, Liebano RE. The
effects of transcutaneous electrical nerve stimulation on tis-
sue repair: a literature review. Can J Plast Surg.
2012;20(4):237–240.
[14] DeSantana JM, Walsh DM, Vance C, et al. Effectiveness of
transcutaneous electrical nerve stimulation for treatment of
hyperalgesia and pain. Curr Rheumatol Rep. 2008;10(6):
492–499.
[15] Maeda Y, Lisi TL, Vance CGT, et al. Release of GABA and
activation of GABA(A) in the spinal cord mediates the
effects of TENS in rats. Brain Res. 2007;1136(1):43–50.
[16] Kim TH, In TS, Cho HY. Task-related training combined with
transcutaneous electrical nerve stimulation promotes upper
limb functions in patients with chronic stroke. Tohoku J
Exp Med. 2013;231(2):93–100.
[17] Chen SC, Chen YL, Chen CJ, et al. Effects of surface elec-
trical stimulation on the muscle-tendon junction of spastic
gastrocnemius in stroke patients. Disabil Rehabil. 2005;
27(3):105–110.
[18] Baricich A, Carda S, Bertoni M, et al. A single-blinded,
randomized pilot study of botulinum toxin type A com-
bined with non-pharmacological treatment for spastic foot.
J Rehabil Med. 2008;40(10):870–872.
[19] Hussain T, Sosorburam T, Mohammad H, et al. The effect of
transcutaneous electrical nerve stimulation (TENS) com-
bined with bobath on post stroke spasticity. A randomized
controlled study. JUMDC. 2013;4(1):22–29.
[20] Laddha D, Ganesh GS, Pattnaik M, et al. Effect of transcuta-
neous electrical nerve stimulation on plantar flexor muscle
spasticity and walking speed in stroke patients. Physiother
Res Int. 2016;21(4): 247–256.
[21] Ng SS, Hui-Chan CW. Transcutaneous electrical nerve
stimulation combined with task-related training improves
lower limb functions in subjects with chronic stroke.
Stroke. 2007;38(11):2953–2959.
[22] Park J, Seo D, Choi W, et al. The effects of exercise with
TENS on spasticity, balance, and gait in patients with
chronic stroke: a randomized controlled trial. Med Sci
Monit. 2014;20:1890–1896.
[23] Picelli A, Dambruoso F, Bronzato M, et al. Efficacy of thera-
peutic ultrasound and transcutaneous electrical nerve
stimulation compared with botulinum toxin type A in the
treatment of spastic equinus in adults with chronic stroke:
a pilot randomized controlled trial. Top Stroke Rehabil.
2014;21 Suppl 1:S8–S16.
[24] Sonde L, Gip C, Fernaeus SE, et al. Stimulation with low fre-
quency (1.7 Hz) transcutaneous electric nerve stimulation
(low-tens) increases motor function of the post-stroke par-
etic arm. Scand J Rehabil Med. 1998;30(2):95–99.
[25] Jung KS, In TS, Cho HY. Effects of sit-to-stand training com-
bined with transcutaneous electrical stimulation on spasti-
city, muscle strength and balance ability in patients with
stroke: a randomized controlled study. Gait Posture.
2017;54:183–187.
[26] Decq P. Peripheral neurotomies for the treatment of focal
spasticity of the limbs. Neurochirurgie. 2003;49(2–3 Pt 2):
293–305.
[27] Decq P, Filipetti P, Lefaucheur J. Evaluation of spasticity in
adults. Oper Tech Neurosurg. 2004;7(3):100–108.

[28] Fern
andez-Tenorio E, Serrano-Mun ~oz D, Avendan ~o-Coy J,
Go
mez-Soriano J. Transcutaneous electrical nerve stimula-
tion for spasticity: a systematic review. Neurologia.
2016;pii:S0213-4853(16):30111–30116.
[29] Mills PB, Dossa F. Transcutaneous electrical nerve stimula-
tion for management of limb spasticity: a systematic
review. Am J Phys Med Rehabil. 2016;95(4):309–318.
[30] Lin S, Sun Q, Wang H, Xie G. Influence of transcutaneous
electrical nerve stimulation on spasticity, balance, and
walking speed in stroke patients: a systematic review and
meta-analysis. J Rehabil Med. 2018;50(1):3–7.
[31] Etoom M. Comments on: influence of transcutaneous elec-
trical nerve stimulation on spasticity, balance, and walking
speed in stroke patients: a systematic review and meta-
analysis. J Rehabil Med. 2018;50(1):94.
[32] Higgins JPT, Green S. Cochrane handbook for systematic
reviews of interventions version 5.1.0 [updated March
2011]. The Cochrane Collaboration; 2011. Available from
http://handbook.cochrane.org.
[33] Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting
items for systematic reviews and meta-analyses: the
PRISMA statement. PLoS Med. 2009;6(7):e1000097.
[34] Robinson KA, Dickersin K. Development of a highly sensi-
tive search strategy for the retrieval of reports of controlled
trials using PubMed. Int J Epidemiol. 2002;31(1):150–153.
[35] Fu R, Holmer HK. Change score or followup score? an
empirical evaluation of the impact of choice of mean dif-
ference estimates [Internet]. Rockville (MD): Agency for
Healthcare Research and Quality (US); 2015 Apr. Report
No.: 15-EHC016-EF.
[36] Detrembleur C, Lejeune TM, Renders A, Van Den Bergh PY.
Botulinum toxin and short-term electrical stimulation in the
treatment of equinus in cerebral palsy. Mov Disord.
2002;17(1):162–169.
[37] Aydin G, Tomruk S, Keleş I, et al. Transcutaneous electrical
nerve stimulation versus baclofen in spasticity: clinical and
electrophysiologic comparison. Am J Phys Med Rehabil.
2005;84(8):584–592.
[38] Thibaut A, Chatelle C, Ziegler E, et al. Spasticity after stroke:
physiology, assessment and treatment. Brain Inj. 2013;
27(10):1093–1105.
[39] Stein C, Fritsch CG, Robinson C, et al. Effects of electrical
stimulation in spastic muscles after stroke: systematic
review and meta-analysis of randomized controlled trials.
Stroke. 2015;46(8):2197–2205.
TENS ON POST-STROKE SPASTICITY: A META-ANALYSIS 13
[40] Park SW, Yi SH, Lee JA, et al. Acupuncture for the treat-
ment of spasticity after stroke: a meta-analysis of random-
ized controlled trials. J Altern Complement Med. 2014;
20(9):672–682.
[41] Guo P, Gao F, Zhao T, et al. Positive effects of extracorpor-
eal shock wave therapy on spasticity in poststroke patients:
a meta-analysis. J Stroke Cerebrovasc Dis. 2017;26(11):
2470–2476.
[42] Li F, Wu Y, Li X. Test-retest reliability and inter-rater reliabil-
ity of the Modified Tardieu Scale and the Modified
Ashworth Scale in hemiplegic patients with stroke. Eur J
Phys Rehabil Med. 2014;50(1):9–15.
[43] Blackburn M, van Vliet P, Mockett SP. Reliability of meas-
urements obtained with the modified Ashworth scale in
the lower extremities of people with stroke. Phys Ther.
2002;82(1):25–34.
[44] Ansari NN, Naghdi S, Moammeri H, et al. Ashworth Scales
are unreliable for the assessment of muscle spasticity.
Physiother Theory Pract. 2006;22(3):119–125.
[45] Mutlu A, Livanelioglu A, Gunel MK. Reliability of Ashworth
and Modified Ashworth scales in children with spastic cere-
bral palsy. BMC Musculoskelet Disord. 2008;9:44.
[46] Craven BC, Morris AR. Modified Ashworth scale reliability
for measurement of lower extremity spasticity among
patients with SCI. Spinal Cord. 2010;48(3):207–213.
[47] Miyara K, Matsumoto S, Uema T, et al. Feasibility of using
whole body vibration as a means for controlling spasticity
in post-stroke patients: a pilot study. Complement Ther
Clin Pract. 2014;20(1):70–73.
[48] Young R, Nix S, Wholohan A, et al. Interventions for
increasing ankle joint dorsiflexion: a systematic review and
meta-analysis. J Foot Ankle Res. 2013;6(1):46.
[49] Schulz KF, Altman DG, Moher D; CONSORT Group.
CONSORT 2010 statement: updated guidelines for report-
ing parallel group randomised trials. J Clin Epidemiol.
2010;63(8):834–840.
[50] Armijo-Olivo S, Saltaji H, et al. What is the influence of ran-
domisation sequence generation and allocation conceal-
ment on treatment effects of physical therapy trials? a
meta-epidemiological study. BMJ Open. 2015;5(9):e008562.
[51] Carvalho APV, Silva V, Grande AJ. Avaliaç~ao do risco de
vi
es de ensaios cl
ınicos randomizados pela ferramenta da
colaboraç~ao Cochrane [Evaluation of risk of bias from
randomized clinical trials by the Cochrane Collaboration’s
tool]. Diagn Tratamento. 2013;18(1):38–44.