514

Top Stroke Rehabil 2012;19(5):514–522
© 2012 Thomas Land Publishers, Inc.
www.thomasland.com

doi: 10.1310/tscir1905-514

Top Stroke Rehabil 2012;19(6):514–522
© 2012 Thomas Land Publishers, Inc.
www.thomasland.com

doi: 10.1310/tsr1906-514

Evidence for Therapeutic Interventions

for Hemiplegic Shoulder Pain During

the Chronic Stage of Stroke: A Review

Ricardo Viana, MD,

1,2

Shelialah Pereira, PT, MSc,

1,3

Swati Mehta, MA,

1,3

Thomas Miller, MD, FRCPC,

1,2,3

and Robert Teasell MD, FRCPC

1,2,3

1

St. Joseph’s Health Care, Parkwood Hospital, London, Ontario;

2

Department of Physical Medicine and Rehabilitation, Schulich School

of Medicine and Dentistry, Western University, London, Ontario;

3

Aging, Rehabilitation and Geriatric Care Program, Lawson Health

Research Institute, London, Ontario, Canada

Objective: To determine the effectiveness of therapeutic interventions targeting hemiplegic shoulder pain (HSP) more
than 6 months post stroke. Methods: A literature search of multiple databases (PubMed, CINAHL, Ovid, and EMBASE)
was conducted to identify articles published in the English language from 1980 to April 2012. Studies were included if
(1) all participants were adults who had sustained a stroke; (2) research design was a randomized controlled trial (RCT)
that examined the effectiveness of any treatment for HSP; (3) all participants had experienced stroke at least 6 months
previously; and (4) an assessment of pain had been conducted before and after treatment using a standardized method.
The following data were extracted: patient characteristics (ie, age, gender, time since stroke), sample size, study design,
measurement of pain pre and post treatment, and adverse events. Results: Ten RCTs (PEDro scores 4–9) met inclusion
criteria and included a total sample size of 388 individuals with a mean age of 53.2 years (range, 43.6–73.2). Mean time
post stroke was 18.4 months. Three studies addressed the use of botulinum toxin type A (BTx-A); 2 studies examined
electrical stimulation; 3 studies focused on intraarticular glenohumeral corticosteroid injections; 1 studied subacromial
corticosteroid injections; and 1 study looked at massage therapy. Conclusions: Positive outcomes were noted with the use
of corticosteroid injections and electrical stimulation and confl icting results were seen regarding the use of BTx-A. Overall,
these targeted therapies provide benefi t in the treatment of HSP in individuals who are more than 6 months post stroke.
Key words: chronic stroke, hemiplegic shoulder pain

H

emiplegic shoulder pain (HSP) is a com-
mon complication post stroke, which
may reduce participation in rehabilita-

tion activities, contribute to activity avoidance,
and reduce quality of life.

1–5

It is estimated that

28% of individuals with HSP will develop symp-
toms within 2 weeks; by 4 months, up to 87%
will be affected.

6

Although the pain will resolve

by 6 months in the majority of cases, at least 20%
of patients experience persistent, often debili-
tating, symptoms.

6,7

It is not clear whether this

improvement is due to treatment or the natural
history of the condition. The etiology of HSP is
multifactorial. Mechanisms include structural
injury from glenohumeral subluxation, capsular
contractures, or rotator cuff pathology. Chronic
HSP may develop over time and is thought to
be due to treatment-resistant structural injury,
abnormal posture of the hemiplegic shoulder
that damages the surrounding tissues,

7

or peri-

articular muscle spasticity.

2,4

With such a variety of possible etiologies, it is

no surprise that interventions are equally varied.
Current management includes physiotherapy,
massage therapy, strapping, slings and other
supports to minimize glenohumeral subluxation,
intraarticular or subacromial corticosteroid
injections, suprascapular ner ve blocks,
percutaneous or superficial electrical muscle
stimulation, and botulinum toxin type A (BTx-A)
intramuscular injections.

8

Early management

focuses on prevention with proper positioning
and range of motion activities or on treatment
in the acute or subacute stages post stroke.

8

This

is expected given that most stroke survivors
present with symptoms during this time. It is
diffi cult to determine the treatment effect given

Therapeutic Interventions for Hemiplegic Shoulder Pain

515

the methodological quality of many of the studies;
however, the overall estimate of treatment effect
for the available treatments has been reported
at 30% to 50%

9

for the poststroke population

as a whole. This is likely due to the inconsistent
defi nition of HSP

7,10–12

or lack of an appropriate

clinical examination as a guide for treatment
modality selection. The objective of this review
is to determine the effectiveness of interventions
targeting HSP

≥

6 months post stroke and present

them by modality for specifi c, possible etiologies.

Methods

Search strategy

A literature search was conducted in multiple

databases (PubMed, Scopus, Ovid, and CINHAL)
to identify relevant articles published from 1980
to April 2012. Reference lists were also hand
searched for additional articles that may not have
been identifi ed during the original search.

Selection criteria

Studies were selected for inclusion if (1) all

participants were adults (

≥

18 years of age) who

had sustained a stroke; (2) research design was a
randomized controlled trial (RCT) that examined
the effectiveness of any treatment for HSP; (3) all
subjects included in the study had experienced
stroke at least 6 months previously; and (4) an
assessment of pain had been conducted before
and after treatment using a standardized method.
Control conditions could include a placebo,
active control, or no treatment depending on the
treatment under investigation.

Commentaries, letters, abstracts, reviews/

guidelines, imaging studies, case studies or case
series, non-human trials, and those articles that
were not in English were excluded.

Study selection and assessment of methodological
quality

The titles and abstracts of all the articles

identifi ed in the literature search were screened
for eligibility. Trials included for the review were
assessed for methodological quality using the

Physiotherapy Evidence Database (PEDro) scoring
system.

13

PEDro is a 10-item scale, which assesses

the internal validity of a study where each item
is awarded a score of yes (1) or no (0). The total
scores range from 0 to 10. The strength of evidence
was assessed using guidelines developed for the
Evidence-Based Review of Stroke Rehabilitation

8

organizing the PEDro scores into the following
categories: “excellent” quality, 9–10; “good” quality,
6–8; “fair” quality, 4–5; and “poor” quality <4.

Results

Search results

A total of 226 articles were identifi ed after

removing duplicates. After reviewing titles and
abstracts and hand screening reference lists,
10 articles were included in the review.

14–23

A

description of the search results is presented in
Figure 1.

Articles included studies evaluating the

effectiveness of BTx-A,

14–16

regarding superfi cial

and percutaneous electrical stimulation,

17,18

and focusing on intraarticular glenohumeral
corticosteroid injections,

19–21

subacromial steroid

injections,

22

and massage therapy.

23

Table 1

provides a detailed description of the studies
divided into their respective treatment categories.

The total sample size was 388 participants,

which was composed of 229 males and 159 females.
The mean age of this pooled sample was 53.2 years
(range, 43.6 –73.2), and the mean time post stroke
was 18.4 months.

In most of the studies,

14–20,22

participants received

“conventional” physical therapy consisting of
stretching and range of motion (ROM) exercises
and neuromuscular facilitation. Rah

22

was more

prescriptive, allowing participation in therapy to
start only 2 weeks post injection to minimize the
risk of tendon rupture. These participants received
a graduated return to therapy program starting
2 to 4 weeks post injection. In 2 studies,

21,23

no

details were reported regarding additional therapy.
One study

17

reported that a proportion of the

study participants received oral analgesia but did
not report the class, timing, frequency, or dose of
use. They did report that these participants were
equally represented in the treatment and control

516 T

OPICS

IN

S

TROKE

R

EHABILITATION

/N

OV

-D

EC

2012

Figure 1. Flow chart depicting results of the literature search.

Search Results (minus duplicates)= 225

(PubMed n=28, Ovid n=77, Scopus n=96, CINAHL n=24)

Articles not relevant to current
review = 216

Reviews (31)

Imaging study (18)

Case studies (20)

Case series (2)

Published abstracts (3)

<6 months post stroke (22)

Etiology studies (20)

Prevalence studies (36)

Treatment of arm function (30)

Guidelines (2)

Not English (3)

Pain assessment (22)

Time post stroke not indicated (1)

No outcomes reported (1)

Mixed time post stroke (4)

Study protocol (1)

Botulinum

toxin

(3)

Electrical

stimulation

(2)

Steroid

injection

(4)

Included in the review = 10

Articles identified by hand

searching references = 1

Active

therapies

(1)

groups, and no changes were made to the dose or
regimens during the study period.

Adverse events were reported by 4 studies

15–17,22

impacting a total of 25 participants (6.4% of total

sample; 13 in the intervention groups and 12 in
the control groups). Most were described as minor
including pain

15,16

or vasovagal syncope

16,22

at

the time of the intervention. The most signifi cant

Therapeutic Interventions for Hemiplegic Shoulder Pain

517

Treatment categories

Botulinum toxin

Three studies used BTx-A. The PEDro

scores ranged from 6 to 8 (good to excellent
methodological quality). Two studies

14,16

targeted

event was reported by Chae

17

where the tips of

the percutaneous leads remained intramuscular in
4 participants. These participants were followed
for 18 months, and there was no associated
infection or granulomatous reaction or associated
pain or functional limitation.

Table 1. Participant characteristics and study details

Study

PEDro

Country

No. of participants

(M/F)

Time post

stroke (months)

Intervention

Outcome

measure

de Boer et al,
2008

14

6

Netherlands

21

(12/9)

7.65

BTx-A (BOTOX) 100 U vs
placebo
Injection into subscapularis

VAS (-)

Kong et al, 2007

15

8

China

16

(11/5)

9.3

BTx-A (Dysport ) 500 U vs
placebo
Injection into pectoralis major
and biceps brachii

VAS (-)

Yelnik et al,
2007

16

7

France

20

(15/5)

16.75

BTx-A (Dysport ) 500 U vs
placebo
Injection into subscapularis

NPS (+)

Lakse et al,
2009

19

4

Turkey

38

(18/20)

12

Triamcinolone acetonide
(40 mg) + prilocain 9 mL +
TENS (20 min/d x 19 sessions) +
PT vs control [TENS (20 min/d x
19 sessions] + PT)

VAS (+)

Lim et al, 2008

20

9

South Korea

29

(15/14)

8.85

Triamcinolone acetonide (40 mg)
vs 100 U BTx-A (BOTOX)
•

Group 1: BTx-A into
infraspinatus, pectoralis
& supscapularis, saline
intraarticular

•

Group 2: Saline into the
muscles’ triamcinolone
intraarticular

NPS (+/−)

Yasar et al, 2011

21

5

Turkey

26

(17/9)

9.04

Triamcinolone acetonide
(40 mg) intraarticular
injection vs 10 mL 2% prilocain
suprascapular nerve block

VAS(+/−)

Rah et al, 2012

22

5

Republic of
Korea

58

(39/19)

21.2

Subacromial injection
triamcinolone acetonide
(40 mg) vs 5 mL 1% lidocaine

VAS (+)

Chae et al, 2005

17

7

USA

61

(36/25)

31.9

Intramuscular electrical
stimulation to supraspinatus,
posterior & middle deltoid, and
upper trapezius (6 h/d x 6 wks)
vs sling

NPS (+)

Kobayshi et al,
1999

18

5

Japan

17

(15/2)

28.77

Surface FES vs no FES
Target supraspinatus vs middle
deltoid vs none

VAS (+)

Mok et al, 2004

23

5

Hong Kong

102

(51/51)

Slow stroke back massage vs no
treatment
Massage provided for 10 min
per night for 7 days

VAS (+)

Note: BTx-A = Botulinum toxin type A; d = day; FES = functional electrical stimulation; h = hour; NPS = Numeric Pain Scale; PT = physical

therapy; TENS = transcutaneous electrical stimulation; VAS = Visual Analog Scale; W/C = wheelchair; wk = week.

518 T

OPICS

IN

S

TROKE

R

EHABILITATION

/N

OV

-D

EC

2012

40 mg intraarticular injection of triamcinolone
acetonide (TA). Follow-up was arranged for 2, 6,
and 12 weeks post treatment. Pain scores (NPS)
for both groups improved from baseline (BTx-A:
7.9 ± 0.3 to 3.2 ± 0.5; TA: 7.6 ± 0.5 to 5.2 ± 0.8,
at 12 weeks); however, there was no statistically
signifi cant difference between groups (P = .064).
Yaser et al

21

compared an intraarticular injection

of 40 mg TA to a suprascapular nerve block using
10 mL of 2% prilocain. Follow-up measures
were arranged for 1 hour, 1 week, and 1 month
post treatment. Pain score (VAS) was similar
between groups (P > .05). Although the authors
provided baseline scores, they did not provide
follow-up scores for each of the time points. One
study looked at subacromial injections in the
treatment of HSP

≥

6 months post stroke. Rah et

al

22

compared a subacromial injection of 40 mg TA

to 5 mL of 2% lidocaine. Follow-up was arranged
for 2, 4, and 8 weeks post treatment. Pain score
(VAS) improvement was signifi cantly greater in
the corticosteroid group for daytime pain at 4 and
8 weeks (P = .048 and P = .004, respectively) and
for night time pain at 2, 4, and 8 weeks (P = .004,
P = .004, P < .001, respectively).

Electrical stimulation

Two studies addressed the use of neuromuscular

electrical stimulation. One study examined
percutaneous stimulation and the other studied
surface stimulation. PEDro scores were 5 and 7.
Chae et al

17

compared intramuscular electrical

stimulation to the supraspinatus, posterior and
middle deltoid, and upper trapezius musculature
for 6 h/day for 6 weeks to use of a sling. Follow-up
was arranged for 3, 6, and 12 months post
treatment. Maximum pain score for the previous
week [Brief Pain Inventory question 12’s numeric
pain scale (NPS)] improved to a signifi cantly
greater extent in the treatment group [Stim:
baseline 7.59 ± 2.12 to

Δ

3 months 4.44 (3.68),

Δ

6 months 4.44 (3.56),

Δ

12 months 5.0 (3.3) vs

Control: baseline 6.52 ± 2.29 to

Δ

3 months 0.68

(1.85),

Δ

6 months 1.38 (2.81),

Δ

12 months 2.31

(3.21); P < .001, at all time points]. Kobayshi et
al

18

randomly assigned 17 participants to surface

stimulation of supraspinatus muscle (group 1)
versus surface stimulation of the middle deltoid

the subscapularis muscle. de Boer et al

14

injected a

total of 100 U (BOTOX) to 2 sites in the subscapularis
muscle at a 2:1 dilution. Pain, measured by the
visual analog scale (VAS), showed no signifi cant
difference with respect to control at 12 weeks
[BTx-A: 44.9 mm (15.2 mm) to 38.1 mm (18.2
mm) vs Control: 61.7 mm (23.2 mm) to 46.8 mm
(27.2 mm); P = .81]. Yelnik et al

16

injected a total

of 500 U (Dysport), of unknown dilution, targeting
the motor endplates using electrical stimulation.
The control group was injected with all the
constituents of Dysport solvent but no BTx-A, and
follow-up was performed at 1, 2, and 4 weeks. Pain
score, measured by the numeric pain scale (NPS),
showed a signifi cant improvement with respect to
control at 4 weeks post injection (BTx-A: 7.5 to 1.5
vs Control: 5.5 to 4; P = .025). One study targeted
the pectoralis major muscle. Kong et al

15

injected a

total of 500 U (Dysport) diluted in 2.5 mL of saline
with 250 U to pectoralis major and 250 U to biceps
brachii. The control group received intramuscular
injection of normal saline, and participants were
followed up at 4, 8, and 12 weeks. Pain score (VAS)
showed no signifi cant improvement with respect to
pain control at 1, 2, and 3 months (P = .21, P = .48,
and P = .5, respectively).

Corticosteroid injection therapy

Three studies investigated intra-articular

corticosteroid injections. PEDro scores ranged
from 4 to 9. Lakse et al

19

compared 40 mg

of intraarticular triamcinolone acetonide and
9 mL of prilocain with transcutaneous electrical
stimulation (TENS) for 20 minutes per day for
19 days and physical therapy (PT) versus TENS
and PT alone. Follow-up was arranged at 1 and
4 weeks. Pain scores (VAS) were signifi cantly
improved in the injection group at time points at
rest, during activity, and at night (week 1: P = .01,
P = .02, P = .00; week 4: P = .03, P = .03, P = .01)
when compared to the control group during these
activities. Lim et al

20

compared intra-articular

triamcinolone acetonide to intramuscular BTx-
A. One group received a total of 100 U of BTx-A
(BOTOX) at a 4:1 dilution into infraspinatus,
pectoralis major, and subscapularis muscles and
an intraarticular injection of saline; the other group
received intramuscular injections of saline and a

Therapeutic Interventions for Hemiplegic Shoulder Pain

519

and BTx-A into overactive, spastic, periarticular
muscles. There were no RCTs examining the use
of physical modalities, strapping or other shoulder
support systems, or analgesic medication,
specifi cally topical or oral agents including the
use of opioids, in chronic HSP. As a whole, the
treatments studied had a positive impact on HSP
management in this population. Because each
treatment targets a different, suspected, etiology,
they will be initially discussed separately.

BTx-A is thought to decrease the activity of

periarticular muscles leading to muscle relaxation
and improved ROM. There are no studies
to date demonstrating a correlation between
muscle activity or spasticity and HSP, but a
causal mechanism is inferred based on positive
treatment responses. In this review, among the
4 studies that examined the use of BTx-A, there
were differences in target muscle (subscapularis
vs pectoralis major), dose and type of BTX type A
(BOTOX vs Dysport), and use of neuromuscular
targeting (neuromuscular electrical stimulation vs
anatomical landmarking). Two of the studies were
considered to be positive and 2 were negative.

Even though all studies use a BTx-A product,

there is some variability between brands on dose
equivalence. de Boer

14

and Lim

20

use BOTOX,

whereas the others used Dysport. Wohlfarth et al

24

recently determined the dose equivalence between
Dysport and BOTOX was 2.3–2.5:1 respectively,
when mixed in saline. To compare dosages
between studies, we have used this conversion
and reported treatment dose in unit equivalents of
Dysport. The dosages used in the studies ranged
from 250 to 500 U.

The 2 negative studies

14,15

differed by target

muscle (subscapularis vs pectoralis major)
and were similar in the dose of BTx-A (250 U
equivalents of Dysport), the use of a low dilution
(2–2.5:1), and no neuromuscular guidance for
needle placement. The positive studies

16,20

differed

by target muscles (subscapularis muscle vs
subscapularis muscle + pectoralis major muscle +
infraspinatus muscle); they also differed from
each other and the negative studies in that one
used guidance (neuromuscular junction electrical
stimulation) and the other used a high dilution
of 4:1. Both of these studies used higher doses
(250 to 500 U equivalents of Dysport)

muscle (group 2) versus no surface stimulation
(control). The therapeutic electrical stimulation
(TES) protocol consisted of negative monophasic
rectangular pulses at a frequency of 20 Hz and
pulse duration of 0.3 ms for the stimulation.
Total stimulation time was 15 seconds including
3 seconds of rising time and 2 seconds of falling
time repeating in a pattern of 15 seconds on and
15 seconds off for 15 minutes twice a day for
6 weeks. Follow-up was arranged for 6 weeks
post treatment. The primary outcome in this
study was to determine the amount reduction in
shoulder subluxation post stimulation. Pain was
a secondary outcome and present in only 6 of the
17 participants (Group 1: 3; Group 2: 3; Control: 1).
A 15 cm VAS was used to measure pain. At the
conclusion of the study, there was a reduction
in mean VAS scores in the 2 intervention groups
(Group 1: 10.33 cm to 7.83 cm; Group 2: 8.93 cm
to 4.8 cm) compared to control (6.3 cm to 6.5 cm).

Hands-on therapy

Only one study addressed the use of hands-on

therapy. Mok et al

23

compared the use of slow

stroke back massage for 10 minutes per night
for 7 days to a no-treatment control. Follow-up
was arranged for 3 days after the last treatment.
The authors found that the treatment group had
a signifi cant reduction in pain during and after
the treatments (P < .05) compared to control, as
measured by the VAS.

Discussion

Ten RCTs representing 5 distinct treatment

modalities were identifi ed in this review that
examined the management of HSP in stroke
survivors

≥

6 months post stroke. Chronic HSP

is thought to differ from early onset HSP and
is associated with treatment-resistant structural
injury, abnormal posture of the hemiplegic
shoulder (including chronic subluxation), or
increased spasticity of periarticular muscles.

4

The

treatments studied in this population address these
presumed etiologies with the use of corticosteroid
injections (either intraarticular or subacromial),
electrical stimulation of weak periarticular muscle
to minimize subluxation in order to reduce pain,

520 T

OPICS

IN

S

TROKE

R

EHABILITATION

/N

OV

-D

EC

2012

volume injection. The targeted muscles were large,
so the negative studies may not have used a high
enough dose or dilution to get the desired effect
given they did not use neuromuscular targeting.

Intraarticular corticosteroid injections are

given primarily in the management of capsular
or glenohumeral pathology, whereas subacromial
injections are used to treat rotator cuff or bursal
pathologies. In this review, we identifi ed 4 studies
addressing the use of corticosteroid injections,
specifi cally TA, with confl icting results. Lakse
reported signifi cant improvement in pain control,
whereas the other groups showed no difference
between treatment groups. Intraarticular injections
were no better than BTx-A and the mean decrease
in pain score was in favor of BTx-A (P = .051),
whereas intraarticular injections were found
to be equal in effi cacy to suprascapular nerve
blocks. Unfortunately, Yaser et al

21

did not provide

follow-up pain scores but instead presented
the follow-up results as statistical differences in
repeated measurements, so we cannot comment
on whether there was a signifi cant change from
baseline in either the intraarticular or suprascapular
nerve block group. The single article addressing
subacromial injections

22

reported significantly

greater improvement in pain control. In conclusion,
corticosteroid injections show a trend toward
improving HSP in stroke survivors who are

≥

6 months post stroke, and intraarticular injections

appear as equally effective as BTx-A intramuscular
injections and suprascapular nerve blocks.

Two studies

17,18

addressed the use of

neuromuscular stimulation in the management
of HSP

≥

6 months post stroke. Device wear

times for both regimens were extensive, requiring
stimulation up to 6 hours per day for an average of
6 weeks. The study by Chae

17

was strongly in favor

of percutaneous neuromuscular stimulation for
the management of HSP as well as for treatment of
glenohumeral subluxation. Kobayshi

18

used surface

neuromuscular stimulation in the management
of shoulder subluxation and reported some
benefi t on pain in a small sample of participants.
In reviewing the available information, we can
conclude that the current state of the evidence in
the use of surface neuromuscular stimulation in
the management of HSP

≥

6 months post stroke

is modest at best.

All studies differed by the dilution of BTx-A used.

de Boer

14

used a 2:1 dilution, Kong

15

used 2.5:1,

and Lim

20

use 4:1. Yelnik’s study

16

did not report

the dilution used, so its impact on the positive
outcome is unknown. Dilution can impact the
effectiveness of BTx-A intramuscular injections. A
recent study by Gracies et al

25

explored the impact

of dilution in neuromuscular treatment of the
biceps brachii. In this study, one group received
100 U of BTx-A at 1:1 dilution targeted to the
neuromuscular junctions across 4 sites, a second
group received 100 U of BTx-A at 1:1 dilution over
4 sites away from the neuromuscular junctions,
and a third group received 100 U of BTx-A at 5:1
dilution over 4 sites away from the neuromuscular
junctions. At the conclusion of the study, it was
determined that low volume/low dilution targeted
injections were as effective as high volume/high
dilution nontargeted injections, and both are
superior to low volume/low dilution nontargeted
injections. This conclusion is supported by the
fi nding that BTx-A diffuses in tissues and across
tissue plains as far as 4.5 cm with dilutions of
3:1 to 5:1.

26

So at higher dilution, the BTx-A is

able to more easily diffuse to and act on the target
neuromuscular junctions, even if injected as far as
4 to 4.5 cm away. Dilution appears to be associated
with likelihood of pain relief in the studies included
in this analysis, as the de Boer

14

and Kong

15

studies used low dose, low dilution, nontargeted
injections with negative results; Yelnik

16

used high

dose, targeted injections with positive impact on
pain control; and Lim

20

used low dose and high

dilution and noted considerable change from
baseline. Adjusting dose and dilution is patient-
and clinician-specifi c, so no recommendation can
yet be made on standard formulations.

Both the pectoralis major and subscapularis

muscles were targeted in these studies. There are
a number of variables impacting on the outcome
of these studies, so a defi nitive recommendation
as to which muscle to target is not possible.
Also, no studies have identifi ed a correlation
between quantitative activity measured on
electromyography (EMG) in these muscles and the
incidence or intensity of HSP.

Only the Yelnik

16

study used neuromuscular

targeting. As we have learned from Gracies,

25

lack

of targeting may be overcome by high dilution, high

Therapeutic Interventions for Hemiplegic Shoulder Pain

521

spasticity and the presence or intensity of HSP, as well
as clearer recommendations regarding target muscle,
dose, and dilution of BTx-A and the importance of
neuromuscular targeting. Corticosteroid injections
into the joint or subacromial space appear to be
a good treatment option for HSP

≥

6 months post

stroke. Electrical stimulation can be considered
in individuals with HSP and glenohumeral
subluxation, as it will improve the degree of
subluxation and may help in pain control. There
is no information available regarding the impact
of shoulder support systems and/or the utility of
oral analgesia, including opioids. More research
is needed to more completely understand HSP
and determine which modality, or combination of
modalities, can best manage HSP

≥

6 months post

stroke.

Acknowledgments

We would like to acknowledge funding from the

Canadian Stroke Network.

Financial disclosure: We certify that no party

having a direct interest in the results of the research
supporting this article has or will confer a benefi t on us
or on any organization with which we are associated
and, if applicable, we certify that all fi nancial and
material support for this research (eg, NIH or NHS
grants) and work are clearly identifi ed.

Mok et al

23

studied the use of slow stroke back

massage. The study follow-up was limited to
3 days after the last treatment, so the authors were
unable to determine whether the improvement in
pain was lasting. Also, there was no blinding of the
intervention, which carries a signifi cant bias. The
available information points to limited evidence in
support of slow stroke back massage as an option
for pain control at the time of treatment, but
more research must be performed to determine
the persistence of the effect and whether there is
a number of treatments, or “dose,” that would be
needed to result in a lasting impact on pain control.

Conclusion

HSP is a prevalent and debilitating poststroke

complication contributing to impairment and
disability. The etiology of HSP is multifactorial;
chronic HSP is thought to be secondary to treatment-
resistant structural injury, abnormal posturing
or chronic shoulder subluxation, or increased
periarticular muscle spasticity. Treatments directed
toward these specifi c etiologies have demonstrated a
positive impact on pain control in this population.
More research is needed regarding the use of
BTx-A in order to improve its effi cacy. Specifi cally,
there needs to be a clearer understanding of the
relationship between quantitative muscle activity or

REFERENCES

1. Lindgren I, Jonsson AC, Norrving B, Lindgren

A. Shoulder pain after stroke, a prospective
population-based study. Stroke. 2007;38:
343–348.

2. Koog YH, Jin SS, Yoon K, BI Min BI. Interventions

for hemiplegic shoulder pain: systematic review
of randomized control trials. Disabil Rehabil.
2010;32:282–291.

3. Turner-Stokes L, Jackson D. Shoulder pain after

stroke: a review of the evidence base to inform the
development of an integrated care pathway. Clin
Rehabil. 2002;16:276–298.

4. Black-Schaffer RM, Kirsteins AE, Harvey RL.

Stroke rehabilitation. 2. Co-morbidities and
complications. Am Acad Phys Med Rehabil.
1999;80:S8–S16.

5. Gamble GE, Barberan E, Bowsher D, Tyrrell PJ, Jones

AK. Post stroke shoulder pain: more common than
previously realized. Eur J Pain. 2000;4:313–315.

6. Gamble GE, Barberan E, Laasch HU, Bowsher D, Tyrrell

PJ, Jones AK. Poststroke shoulder pain: a prospective
study of the association and risk factors in 152
patients from a consecutive cohort of 205 patients
presenting with stroke. Eur J Pain. 2002;6:467–474.

7. Kalichman L, Ratmansky M. Underlying pathology

and associated factors of hemiplegic shoulder
pain. Am J Phys Med Rehabil. 2011;90:768–780.

8. Teasell R, Foley N, Salter K, et al. Evidence-Based

Review of Stroke Rehabilitation. 14th ed. 2011.
wwwEBRSRcom.

9. Snels IAK, Dekker JHM, van der Lee JH, Lankhorst

GJ, Beckerman H, Bouter LM. Treating patients
with hemiplegic shoulder pain. Am J Phys Med
Rehabil. 2002;81:150–160.

10. Ratnasabapathy Y, Broad J, Baskett J, Pledger M,

Marshall J, Bonita R. Shoulder pain in people with
a stroke: a population-based study. Clin Rehabil.
2003;17:304–311.

522 T

OPICS

IN

S

TROKE

R

EHABILITATION

/N

OV

-D

EC

2012

19. Lakse E, Gunduz B, Erhan B, Celik EC. The

effect of local injections in hemiplegic shoulder
pain: a prospective, randomized, controlled study.
Am J Phys Med Rehabil. 2009;88:805-811; quiz
12–14, 51.

20. Lim JY, Koh JH, Paik NJ. Intramuscular botulinum

toxin-A reduces hemiplegic shoulder pain: a
randomized, double-blind, comparative study
versus intraarticular triamcinolone acetonide.
Stroke. 2008;39:126–131.

21. Yasar E, Vural D, Safaz I, et al. Which treatment

approach is better for hemiplegic shoulder
pain in stroke patients: intra-articular steroid
or suprascapular nerve block? A randomized
controlled trial. Clin Rehabil. 2011;25:60–68.

22. Rah UW, Yoon SH, Moon do J, et al. Subacromial

corticosteroid injection on poststroke hemiplegic
shoulder pain: a randomized, triple-blind,
placebo-controlled trial. Arch Phys Med Rehabil.
2012;93:949–956.

23. Mok E, Woo CP. The effects of slow-stroke back

massage on anxiety and shoulder pain in elderly
stroke patients. Complement Ther Nurs Midwifery.
2004;10:209–216.

24. Wohlfarth K, Sycha T, Ranoux D, Naver H,

Caird D. Dose equivalence of two commercial
preparations of botulinum neurotoxin type A:
Time for a reassessment? Curr Med Res Opin.
2009;25:1573–1584.

25. Gracies JM, Lugassy M, Weisz DJ, Vecchio M,

Flanagan S, Simpson DM. Botulinum toxin dilution
and endplate targeting in spasticity: a double-
blind controlled study. Arch Phys Med Rehabil.
2009;90:9–16 e2.

26. Lim EC, Seet RC. Botulinum toxin: description

of injection techniques and examination of
controversies surrounding toxin diffusion. Acta
Neurol Scand. 2008;117:73–84.

11. Zorowitz RD, Hughes MB, Idank D, Ikai T, Johnston

MV. Shoulder pain and subluxation after stroke:
Correlation or coincidence? Am J Occup Ther.
1996;50:194–201.

12. Bohannon RW, Larkin PA, Smith MB, Horton MG.

Shoulder pain in hemiplegia: statistical relationship
with five variables. Arch Phys Med Rehabil.
1986;67:514–516.

13. Moseley AM, Herbert RD, Sherrington C, Maher

CG. Evidence for physiotherapy practice: a survey
of the Physiotherapy Evidence Database (PEDro).
Aust J Physiother. 2002;48:43–49.

14. de Boer KS, Arwert HJ, de Groot JH, Meskers

CG, Mishre AD, Arendzen JH. Shoulder pain and
external rotation in spastic hemiplegia do not
improve by injection of botulinum toxin A into the
subscapular muscle. J Neurol Neurosurg Psychiatry.
2008;79:581–583.

15. Kong KH, Neo JJ, Chua KS. A randomized

controlled study of botulinum toxin A in the
treatment of hemiplegic shoulder pain associated
with spasticity. Clin Rehabil. 2007;21:28–35.

16. Yelnik AP, Colle FM, Bonan IV, Vicaut E. Treatment

of shoulder pain in spastic hemiplegia by
reducing spasticity of the subscapular muscle: a
randomised, double blind, placebo controlled
study of botulinum toxin A. J Neurol Neurosurg
Psychiatry. 2007;78:845–848.

17. Chae J, Yu DT, Walker ME, et al. Intramuscular

electrical stimulation for hemiplegic shoulder
pain: a 12-month follow-up of a multiple-center,
randomized clinical trial. Am J Phys Med Rehabil.
2005;84:832–842.

18. Kobayashi H, Onishi H, Ihashi K, Yagi R, Handa Y.

Reduction in subluxation and improved muscle
function of the hemiplegic shoulder joint after
therapeutic electrical stimulation. J Electromyogr
Kinesiol. 1999;9:327–336.

Copyright of Topics in Stroke Rehabilitation is the property of Thomas Land Publishers Incorporated and its

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.