Advertisement

Subacromial Corticosteroid Injection on Poststroke Hemiplegic Shoulder Pain: A Randomized, Triple-Blind, Placebo-Controlled Trial

Published:February 13, 2012DOI:https://doi.org/10.1016/j.apmr.2012.02.002

      Abstract

      Rah UW, Yoon S-H, Moon DJ, Kwack K-S, Hong JY, Lim YC, Joen B. Subacromial corticosteroid injection on poststroke hemiplegic shoulder pain: a randomized, triple-blind, placebo-controlled trial.

      Objective

      To evaluate the effect of subacromial corticosteroid injection on hemiplegic shoulder pain (HSP).

      Design

      Multicenter, randomized, triple-blind, placebo-controlled trial.

      Setting

      Three primary and 1 university-affiliated tertiary-care hospitals.

      Participants

      Poststroke HSP patients (N=58) with evidence of rotator cuff disorder.

      Interventions

      Participants were randomly assigned to receive ultrasound-guided subacromial injection with triamcinolone 40mg (treatment group, n=29), or lidocaine (placebo group, n=29). After a single injection, participants were followed up for 8 weeks.

      Main Outcome Measures

      Visual analog scale (VAS) of the average shoulder pain level at day and night (VAS-day/night, the primary outcome measures), Modified Barthel Index, Shoulder Disability Questionnaire (SDQ), and angles of shoulder active range of motion (flexion, abduction, external rotation, and internal rotation) at pretreatment and weeks 2, 4, and 8 posttreatment.

      Results

      There was no significant difference between the 2 groups in the main outcome measures at pretreatment. Compared with the placebo group, VAS-day/night, SDQ, flexion, external rotation, and internal rotation showed significant improvement in the treatment group.

      Conclusions

      To our knowledge, this is the first randomized, placebo-controlled study to assess the efficacy of subacromial injection in HSP patients with evidence of rotator cuff disorder. Subacromial corticosteroid injection showed improvement in pain, disability, and active range of motion, and the duration of its efficacy continued up to 8 weeks.

      Key Words

      List of Abbreviations:

      ANOVA (analysis of variance), AROM (active range of motion), HSP (hemiplegic shoulder pain), SDQ (Shoulder Disability Questionnaire), VAS (visual analog scale)
      HEMIPLEGIC SHOULDER PAIN (HSP) is a common problem after stroke; its occurrence varies from 16% to 84% in stroke patients.
      • Gamble G.E.
      • Barberan E.
      • Laasch H.U.
      • Bowsher D.
      • Tyrrell P.J.
      • Jones A.K.
      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.
      • Lindgren I.
      • Jonsson A.C.
      • Norrving B.
      • Lindgren A.
      Shoulder pain after stroke: a prospective population-based study.
      • Griffin J.W.
      Hemiplegic shoulder pain.
      • Neviaser A.S.
      • Hannafin J.A.
      Adhesive capsulitis: a review of current treatment.
      • Snels I.A.
      • Dekker J.H.
      • van der Lee J.H.
      • Lankhorst G.J.
      • Beckerman H.
      • Bouter L.M.
      Treating patients with hemiplegic shoulder pain.
      These patients remain hospitalized for a longer period of time, and their shoulder pain may interfere with the rehabilitation process.
      • Wanklyn P.
      • Forster A.
      • Young J.
      Hemiplegic shoulder pain (HSP): natural history and investigation of associated features.
      The pathogenesis of HSP is not well known; most investigators believe that multiple factors contribute to HSP. Postulated causes of HSP include adhesive capsulitis, rotator cuff disorder, bicipital tendonitis, glenohumeral subluxation, complex regional pain syndrome, spasticity, myofascial pain, central neuropathic pain, and brachial plexopathy.
      • Griffin J.W.
      Hemiplegic shoulder pain.
      • Snels I.A.
      • Dekker J.H.
      • van der Lee J.H.
      • Lankhorst G.J.
      • Beckerman H.
      • Bouter L.M.
      Treating patients with hemiplegic shoulder pain.
      • Bender L.
      • McKenna K.
      Hemiplegic shoulder pain: defining the problem and its management.
      Because of the various causes and multiple factors involved in HSP, it is not easy to treat the pain. There is also a burden on the time and cost of the treatment. And many times, the result is not satisfactory even after treatment.
      • Turner-Stokes L.
      • Jackson D.
      Shoulder pain after stroke: a review of the evidence base to inform the development of an integrated care pathway.
      Rotator cuff disorder (including rotator cuff impingement/tendinopathy/tear and/or subacromial bursitis) is commonly found in stroke patients. A study on stroke patients with HSP,
      • Shah R.R.
      • Haghpanah S.
      • Elovic E.P.
      • et al.
      MRI findings in the painful poststroke shoulder.
      which used magnetic resonance imaging, found 35% of the subjects exhibiting a tear of at least 1 rotator cuff, bicep, or deltoid muscle, and 53% exhibiting tendinopathy of at least 1 rotator cuff, bicep, or deltoid muscle. Normally, the rotator cuff muscle puts downward pressure on the humerus head during the shoulder joint abduction and flexion. But in patients with hemiplegia, the weakened rotator cuff makes it difficult to put a smooth downward pressure on the humerus head, and this leads to impingements. It is presumed that frequent impingements in turn lead to repeated microtraumas and inflammatory responses, resulting in tendinitis or tendon tear.
      • Lanzer W.L.
      Clinical aspects of shoulder injuries.
      Corticosteroid injection is frequently performed in patients with HSP; however, it is still controversial when it comes to its efficacy. One study reported improvements in the range of motion and pain of hemiplegic patients after intra-articular corticosteroid injection.
      • 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.
      On the other hand, a randomized trial found that there was no improvement in pain or function.
      • Snels I.A.
      • Beckerman H.
      • Twisk J.W.
      • et al.
      Effect of triamcinolone acetonide injections on hemiplegic shoulder pain: a randomized clinical trial.
      Another study using subacromial corticosteroid injection showed significant effect in pain reduction.
      • Chae J.
      • Jedlicka L.
      Subacromial corticosteroid injection for poststroke shoulder pain: an exploratory prospective case series.
      But there was no control group in the study and the reduction of pain may have been the result of either spontaneous recovery or placebo effect. There was also a recent study,
      • Lakse E.
      • Gunduz B.
      • Erhan B.
      • Celik E.C.
      The effect of local injections in hemiplegic shoulder pain: a prospective, randomized, controlled study.
      which categorized the cause of shoulder pain into 2 groups, the adhesive capsulitis or subacromial impingement group, and administered an intra-articular or subacromial injection, respectively. According to the study results, both groups showed an increased range of motion and reduced pain. But this study was not blinded and also lacked a placebo control.
      There is a paucity of study on the efficacy of subacromial corticosteroid injection, and no randomized controlled trial exists to draw any conclusions. We performed a subacromial injection on stroke survivors with hemiplegia who were dignosed with rotator cuff disorder as one of the major causes of shoulder pain. We hypothesized that subacromial corticosteroid injection would have a positive effect on pain, range of motion, disability, and activities of daily living.

      Methods

      Participants

      Between October 2009 and July 2010, eligible participants with shoulder pain were recruited in the rehabilitation and neurosurgery clinics of a university hospital in Suwon City, South Korea and 3 other local clinics in Suwon, Goyang, and Busan City. Demographics of the participants, including age, sex, etiology, time since stroke, affected side, Brunnstrom stage of upper limb, and presence of spasticity, were recorded.

      Inclusion and Exclusion Criteria

      Inclusion criteria were: (1) hemiplegia after stroke, (2) HSP of at least 1 month in duration, (3) aged 20 to 70 years old, (4) clinically and ultrasonographically diagnosed rotator cuff disorder; a minimum of 1 positive finding from the physical tests showing correlation with the ultrasonographic evaluation, (5) pain defined as a score of 3 points or more on a 10-cm visual analog scale (VAS) rated from 0 (no pain) to 10 (worst imaginable pain), (6) muscle power of deltoid grade 2 (poor) or greater on the manual muscle test by the Medical Research Council Scale, and (7) a minimum score of 20 points for the Mini-Mental State Examination to ensure the patient can make his/her own decision to participate in the research and express changes in pain.
      Exclusion criteria were: (1) current adhesive capsulitis (restriction of passive motion >30° in ≥2 planes of movement, measured to onset of pain with a gravity inclinometer),
      • Buchbinder R.
      • Green S.
      • Forbes A.
      • Hall S.
      • Lawler G.
      Arthrographic joint distension with saline and steroid improves function and reduces pain in patients with painful stiff shoulder: results of a randomised, double blind, placebo controlled trial.
      (2) complex regional pain syndrome type I diagnosed according to the International Association for the Study of Pain,
      • Sandroni P.
      • Low P.A.
      • Ferrer T.
      • Opfer-Gehrking T.L.
      • Willner C.L.
      • Wilson P.R.
      Complex regional pain syndrome I (CRPS I): prospective study and laboratory evaluation.
      (3) full thickness tear of the rotator cuff in ultrasonographic examination, (4) biceps tendon disorders (not accompanying rotator cuff disorder), (5) severe spasticity of the Modified Ashworth Scale
      • Bohannon R.W.
      • Smith M.B.
      Interrater reliability of a modified Ashworth scale of muscle spasticity.
      grade 3 (considerable increase in muscle tone, passive movement difficult) and 4 (rigid in flexion or extension) in affected upper limb, (6) shoulder subluxation (the width between the inferior aspect of the acromion and the superior aspect of the head of the humerus >1 finger at a sitting or standing position without supporting the affected upper limb),
      • Boyd E.A.
      • Torrance G.M.
      Clinical measures of shoulder subluxation: their reliability.
      (7) severe motor weakness (muscle power of deltoid less than grade 2 on the manual muscle test), (8) primary osteoarthritis of glenohumeral joint in a simple radiograph, (9) the presence of another obvious explanation for the pain (ie, fracture, radiculopathy, myofascial pain, central neuropathic pain), (10) the presence of an unstable medical condition or a known uncontrolled systemic disease, including cancer, rheumatoid arthritis, endocrine disease, major depression, or schizophrenia, (11) previous trauma history of the currently affected shoulder, (12) evidence of recent alcohol or drug abuse, (13) previous corticosteroid injection history of the affected shoulder, or (14) incapable of communication owing to severe aphasia. To prevent postinjection hemorrhage, we did not include participants who were currently taking medication such as an antiplatelet agent or anticoagulation, with the exception of those who agreed to stop for a minimum of 5 days before the injection.

      Sample Size

      A power analysis programa was used to calculate the number of participants.
      • Faul F.
      • Erdfelder E.
      • Lang A.
      • Buchner A.
      G* Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences.
      Sample-size requirements were based on a placebo-controlled, subacromial bursa injection study of the treatment of shoulder pain due to rotator cuff disorders.
      • Petri M.
      • Dobrow R.
      • Neiman R.
      • Whiting-O'Keefe Q.
      • Seaman W.E.
      Randomized, double-blind, placebo-controlled study of the treatment of the painful shoulder.
      Mean pain intensity scores (0–5 scale; 0=worst) of the subacromial injection (40mg triamcinolone and lidocaine) and placebo (lidocaine only) groups in this study were 4.04 and 3.00, respectively, and the SD was 1.55. For alpha of .05 (2-tailed test), 80% power, 2 groups, and 3 repetitions, 52 participants were required. A total of 60 participants would be necessary to allow for 15% loss to follow-up. At the final analysis, 58 participants were included, with an analysis power of 86%. The study was approved by the institutional review board at the university hospital, and all participants gave their written informed consent.

      Diagnosis of Rotator Cuff Disorder

      To determine rotator cuff disorders, we examined physical and ultrasonographic evaluations. We checked for muscle power of rotator cuff and tenderness in the affected shoulder. The Neer,
      • Neer 2nd, C.S.
      • Welsh R.P.
      The shoulder in sports.
      Hawkins,
      • Hawkins R.J.
      • Kennedy J.C.
      Impingement syndrome in athletes.
      and Yocum
      • Yocum L.A.
      Assessing the shoulder History, physical examination, differential diagnosis, and special tests used.
      tests were performed to determine the impingement syndrome. To examine the presence of tendinopathy, Jobe test for supraspinatus tendon
      • Jobe F.W.
      • Jobe C.M.
      Painful athletic injuries of the shoulder.
      and Patte test for infraspinatus tendon were completed.
      • Leroux J.L.
      • Thomas E.
      • Bonnel F.
      • Blotman F.
      Diagnostic value of clinical tests for shoulder impingement syndrome.
      Participants with no positive findings in these tests were excluded.
      After the physical tests, participants were examined with commercial, real-time ultrasound equipment (Logiq P6,b Accuvix XQ,c SonoAce X8,c HD 11XEd) using 5- to 13-MHz linear-array transducers. The examiners evaluated the participants for the presence of a rotator cuff impingement/tendinopathy/tear, calcification, subacromial bursa fluid, or abnormalities of the long head of the biceps tendon. Decisions regarding the imaging interpretations were based on the finding of previously published studies.
      • Bianchi S.
      • Martinoli C.
      • Baert A.
      • et al.
      Ultrasound of the musculoskeletal system.
      • Naredo E.
      • Aguado P.
      • De Miguel E.
      • et al.
      Painful shoulder: comparison of physical examination and ultrasonographic findings.
      The subacromial bursitis was defined as hypoechogenic fluid filled bursa greater than 2-mm thick. Dynamic examination by moving the arm from neutral position to abduction was performed to evaluate impingement syndrome.

      Design and Randomization

      This study was a multicenter, randomized, triple-blind (the participants, injection operators, and evaluators), placebo-controlled trial. A schematic of the study is shown in figure 1. A total of 181 participants were recruited. Of them, 121 were excluded. After screening evaluations, 60 participants were randomly assigned to 1 of 2 groups by a block randomization method, which ensured that approximately equal numbers were allocated to each group.
      • Altman D.G.
      • Bland J.M.
      How to randomise.
      A computerized random number generator and table were used to perform group allocations, which were managed by an assistant (other than the authors). Four of the authors who are board certified physiatrists (S.H.Y., J.Y.H., D.J.M., and B.J.) with 5 to 10 years of experience in ultrasonographic evaluation; evaluators and the musculoskeletal radiologist (K.S.K.) attended a 2-day training course on the study procedure, physical tests, home exercise program, and ultrasonography for diagnosis and injection procedure. The investigators adhered to the injection protocol even after the training course. One of the authors (J.Y.H.) in the university hospital visited the local clinics to facilitate communication between the medical personnel and minimize the differences in diagnoses due to the use of ultrasonography. Clinical investigator meetings were also held once a month, on average.
      Figure thumbnail gr1
      Fig 1Consolidated Standards of Reporting Trials flowchart of the study.

      Intervention

      Subacromial injection

      After physical and ultrasonographic examination, the participants were randomized to have either a single injection of 4mL of 40mg (10mg/mL) triamcinolone acetonide and 1mL of 1% lidocaine or 5mL of 1% lidocaine. The injection was prepared by a sterile technique using opaque syringes so that the participants, injection operators, and evaluators remained blinded to the group assignment. Participants sat in an upright position and the arms were positioned behind their backs with internal rotation and hyperextension of the shoulder and with the elbow bent for longitudinal supraspinatus view. A 23-gauge, 6-cm long needle was inserted parallel to the transducer in a semioblique plane from the posterior side of the shoulder (fig 2) . The needle was advanced with real-time ultrasound equipment until the needle tip entered the bursa.
      Figure thumbnail gr2
      Fig 2Ultrasound-guided subacromial injection. (A) A 23-gauge needle (arrow) is inserted within the subacromial bursa (arrowheads). (B) The subacromial bursa was filled with injection fluid. Abbreviations: DEL, deltoid muscle; H, humerus; SST, supraspinatus tendon.
      Participants had to stay at the hospitals for more than 20 minutes after the injection for any signs of acute adverse reactions. Late adverse reactions including postinjection flare, disturbance of menstrual pattern, facial flushing, subcutaneous tissue atrophy, skin depigmentation, infection, and tendon rupture were also checked at each follow-up. Patients on analgesics, if any, were told to stop administering from 5 days before the injection.

      Exercise program

      Because corticosteroid injection may lead to a temporary weakening of the tendon,
      • Kapetanos G.
      The effect of the local corticosteroids on the healing and biomechanical properties of the partially injured tendon.
      participants were given instructions to avoid any heavy work or exercises on the shoulders for the first 2 weeks after the injection. Although systematic reviews of interventions for rotator cuff disorder and shoulder pain suggest that exercise can be an effective treatment,
      • Grant H.J.
      • Arthur A.
      • Pichora D.R.
      Evaluation of interventions for rotator cuff pathology: a systematic review.
      there is no consensus on an ideal exercise program to treat patients with rotator cuff disorder. In this study, we adopted the exercise program of a systematic review,
      • Kuhn J.E.
      Exercise in the treatment of rotator cuff impingement: a systematic review and a synthesized evidence-based rehabilitation protocol.
      which synthesized features of exercise protocols from clinical studies with the highest levels of evidence. When participants revisited the hospital for a follow-up evaluation at week 2 postinjection, evaluators handed out picture leaflets and provided an education on home exercise programs based on the adopted protocol. Recommendations to keep exercising were also given at each follow-up examination. Participants were also encouraged to write an exercise diary to record the exercise frequency, duration, and any difficulties. These diaries were reviewed at each follow-up visit. Participants with low compliance were contacted by phone and encouraged to continue exercising. From week 2, participants were instructed to start gentle and active range of motion (AROM) exercise without weight. From week 3, they were told to perform progressive strengthening exercises for the scapular stabilizing muscles; from week 4, they were told to perform rotator cuff strengthening with closed chain exercises. Exercises were to be performed 3 times a day lasting 10 minutes each round, and participants were to continue until they felt pain or the pain was no longer bearable to continue exercising. They were also prohibited from taking any additional medication and physical modalities, such as heat, cold, and electricity.

      Outcome Measurements

      Treatment efficacy was evaluated at pretreatment and weeks 2, 4, and 8 after the injection by the 4 evaluators who were not informed of the injected medication. Primary outcome measure was 10-cm VAS for shoulder pain intensity during the day (VAS-day) and the night (VAS-night). The participant answered the question: “with respect to the worst pain you have experienced in your life, what was the average level of your shoulder pain in the last one week?” by placing a mark somewhere along the line. Secondary outcome measures were AROM of the shoulder, the Korean version of the Modified Barthel Index,
      • Jung H.Y.
      • Park B.K.
      • Shin H.S.
      • et al.
      Development of the Korean Version of Modified Barthel Index (K-MBI): multi-center study for subjects with stroke.
      and the Shoulder Disability Questionnaire (SDQ).
      • Croft P.
      • Pope D.
      • Zonca M.
      • O'Neill T.
      • Silman A.
      Measurement of shoulder related disability: results of a validation study.
      AROM was measured by a goniometer for forward flexion, abduction, and external rotation of shoulder. The measurement was taken in a position not affected by gravity (flexion measured lying on the side of the unaffected limb, abduction in a supine position, and external rotation in a sitting position). Participants were asked to move their shoulders slowly and the angle on onset of pain was measured 3 times to record the average value. Scapular rotation was prevented by the evaluators' hand. External rotation was measured in 0° abduction of shoulder and 90° flexion of elbow position. Internal rotation was measured from the vertebra prominens at C7 to the tip of the fully extended thumb. The distance (cm) from the spinous process of C7 vertebra to the tip of the actively extended thumb on the back was measured using a measuring tape.
      • Edwards T.B.
      • Bostick R.D.
      • Greene C.C.
      • Baratta R.V.
      • Drez D.
      Interobserver and intraobserver reliability of the measurement of shoulder internal rotation by vertebral level.
      This methodology shows high intrarater reliability
      • Hoving J.L.
      • Buchbinder R.
      • Green S.
      • et al.
      How reliably do rheumatologists measure shoulder movement?.
      and is a useful method that is also relevant with the activities of daily living, such as dressing, bathing, and toileting. The functional level was evaluated with the Modified Barthel Index, composed of 10 criteria (personal hygiene, bathing, feeding, toileting, stair up/down, dressing, defecation, voiding, ambulation, and chair/bed transfer) on a scale of 0 to 100. An SDQ score is commonly used in the evaluation of shoulder disability in clinical trials, has proven levels of validity, and consists of 22 self-reporting items that participants respond to with either yes or no. The SDQ has demonstrated strong associations with quality of life measures and compares favorably with other published shoulder disability questionnaires in terms of overall validity and patient acceptability. The score ranges between 0 (no disability) and 22 (maximal disability).

      Statistical Analysis

      Comparisons of baseline characteristics of participants were based on the chi-square and Mann-Whitney U test. The effect of subacromial injection (treatment or placebo) was evaluated with repeated-measures analysis of variance (ANOVA). If the repeated-measures ANOVA for group by time interaction was significant, post hoc tests with Bonferroni correction was conducted to compare the 2 groups at each time point. Significance was accepted for P values of less than .05. Statistical analyses were performed by SPSS statistical software.e

      Results

      A total of 60 patients were recruited and of them, 2 were excluded from further follow-ups: 1 from the placebo group who claimed no effect from the injection and the other from the treatment group who cited difficulties of transportation as the reason for no longer visiting the hospital (see fig 1). Table 1 shows the baseline characteristics of the study participants and the ultrasonographic findings of affected shoulder. With the exceptions of hemiplegic side and the ultrasonographic finding of increased fluid in the biceps tendon sheath, there was no significant difference between the treatment group and placebo group in age, sex, stroke type, duration of stroke, duration of shoulder pain, Brunnstrom stages, and Modified Ashworth Scale. There was no significant difference between the treatment group and placebo group in VAS-day, VAS-night, SDQ, Modified Barthel Index, and AROM at pretreatment (P>.05). Changes in outcome measurements after subacromial injection are shown in table 2. Repeated-measures ANOVA showed significant effect of time in all outcomes measures (VAS-day, F2.01,112.35=30.93, P<.001; VAS-night: F2.04,114.49=33.42, P<.001; SDQ, F1.73,97.11=37.41, P<.001; Modified Barthel Index, F1.11,61.91=6.50, P=.011; flexion, F1.71,95.98=28.31, P<.001; abduction, F1.33,74.33=12.66, P<.001; external rotation, F1.72,96.30=14.63, P<.001; internal rotation, F1.56,87.56=8.99, P=.001), and significant group by time interactions for VAS-day (F2.02,112.35=7.67, P=.001), VAS-night (F2.04,114.49=8.91, P<.001), SDQ (F1.73,97.11=15.20, P<.001), flexion (F1.71,95.98=28.10, P<.001), external rotation (F1.72,96.30=8.67, P=.001), and internal rotation (F1.564,87.564=11.15, P<.001). This means that compared with pretreatment, all outcome measures improved significantly in both groups by time. Compared with the placebo group, VAS-day/VAS-night, SDQ, flexion, external rotation, and internal rotation showed significant improvement in the treatment group. Post hoc tests for between-group comparison revealed that of VAS-day at weeks 4 and 8, VAS-night at weeks 2, 4, and 8, SDQ at weeks 4 and 8, flexion at weeks 2, 4, and 8, and external rotation at weeks 2 and 4 showed significant improved in the treatment group.
      Table 1Baseline Characteristics of Participants
      CharacteristicsTreatment Group (n=29)Placebo Group (n=29)P
      Age (y)56.6±12.554.9±10.6.709
      Mann-Whitney U test for between-group comparison.
      Sex, n
       Men:women21:818:11.401
      Chi-square test for between-group comparison.
      Stroke type, n
       Ischemic:hemorrhagic15:1417:12.597
      Chi-square test for between-group comparison.
      Duration of stroke (mo)23.6±16.918.8±10.7.969
      Mann-Whitney U test for between-group comparison.
      Duration of shoulder pain (mo)6.6±4.37.0±3.4.502
      Mann-Whitney U test for between-group comparison.
      Hemiplegic side, n
       Dominant:nondominant21:812:17.017
      Chi-square test for between-group comparison.
      Brunnstrom stages.896
      Chi-square test for between-group comparison.
       Stage 1–200
       Stage 364
       Stage 4911
       Stage 51211
       Stage 623
      Modified Ashworth Scale.549
      Chi-square test for between-group comparison.
       Grade 021
       Grade 1712
       Grade 1+1211
       Grade 285
      Ultrasonographic findings, n (%)
       Increased fluid, BT sheath17 (58.6)9 (31.0).036
      Chi-square test for between-group comparison.
       BT partial tear1 (3.5)0 (0).317
      Chi-square test for between-group comparison.
       Increased fluid, SA bursa18 (62.1)14 (48.3).295
      Chi-square test for between-group comparison.
       RC tendinosis9 (31.0)14 (48.3).183
      Chi-square test for between-group comparison.
       RC impingement2 (6.9)3 (10.3).643
      Chi-square test for between-group comparison.
       RC partial-thickness tear6 (20.7)9 (31.0).372
      Chi-square test for between-group comparison.
       RC calcification4 (13.8)3 (10.3).689
      Chi-square test for between-group comparison.
      NOTE. Values expressed as mean ± SD or as otherwise indicated. Duplicated counts for multiple findings in ultrasonographic findings.
      Abbreviations: BT, biceps tendon; RC, rotator cuff; SA, subacromial.
      low asterisk Mann-Whitney U test for between-group comparison.
      Chi-square test for between-group comparison.
      Table 2Mean Changes of Outcome Measurements After Subacromial Injection
      Outcome MeasurementsTreatment Group (n=29)Placebo Group (n=29)Repeated-Measures ANOVA
      Repeated-measures ANOVA for group by time interaction (P<.05).
      Between-Group Comparison
      If the repeated-measures ANOVA for group by time interaction was significant, post hoc tests with Bonferroni correction were conducted (P<.05).
      FPP
      VAS-day score7.67.001
       Week 05.5±1.75.7±1.71.000
       Week 23.9±2.25.1±1.70.088
       Week 43.6±2.04.9±1.90.048
       Week 83.0±1.84.9±2.30.004
      VAS-night score8.91<.001
       Week 05.5±1.55.9±2.01.000
       Week 23.7±1.95.4±2.10.004
       Week 43.4±1.65.1±2.40.004
       Week 82.7±1.75.0±2.6<0.001
      SDQ score15.20<.001
       Week 016.9±3.816.6±2.71.000
       Week 212.8±5.615.9±3.60.064
       Week 412.3±5.815.8±3.70.044
       Week 811.1±5.715.2±3.90.012
      MBI score0.14.737
       Week 075.7±17.871.0±26.3
       Week 277.5±17.372.2±25.4
       Week 477.5±17.372.6±25.4
       Week 877.5±17.272.7±25.6
      Flexion28.10<.001
       Week 0113.9±18.7119.5±28.01.000
       Week 2135.9±21.5114.8±29.20.012
       Week 4140.8±19.9118.2±29.20.004
       Week 8143.7±20.4121.6±29.80.008
      Abduction1.57.217
       Week 084.9±26.790.5±29.8
       Week 299.0±30.596.1±41.5
       Week 4101.0±31.998.5±40.3
       Week 8107.0±34.6101.2±41.4
      External rotation8.67.001
       Week 034.8±10.936.4±20.31.000
       Week 249.6±15.335.3±20.10.016
       Week 450.7±16.338.1±22.60.042
       Week 852.6±18.040.8±22.50.124
      Internal rotation, cm11.15<.001
       Week 053.1±12.852.6±12.61.000
       Week 249.3±13.153.3±11.30.848
       Week 449.0±13.253.2±11.80.844
       Week 847.7±13.152.7±12.30.548
      NOTE. Values are expressed as mean ± SD degree unless otherwise indicated.
      Abbreviation: MBI, Modified Barthel Index.
      low asterisk Repeated-measures ANOVA for group by time interaction (P<.05).
      If the repeated-measures ANOVA for group by time interaction was significant, post hoc tests with Bonferroni correction were conducted (P<.05).
      There were no important complications, such as skin pigmentation or infection, other than facial flushing on days 1 to 5 after injection (2 participants in treatment group) and dizziness due to vasovagal reaction during injection (1 participant in placebo group).

      Discussion

      Subacromial injection with corticosteroid is known to improve pain and function in nonstroke patients with rotator cuff disorder.
      • Petri M.
      • Dobrow R.
      • Neiman R.
      • Whiting-O'Keefe Q.
      • Seaman W.E.
      Randomized, double-blind, placebo-controlled study of the treatment of the painful shoulder.
      • Buchbinder R.
      • Green S.
      • Youd J.M.
      Corticosteroid injections for shoulder pain.
      However, there is a paucity of studies on the efficacy of subacromial injection in stroke patients with rotator cuff disorder. Although there are a few studies that report improvements in the range of motion and pain after subacromial injection,
      • Chae J.
      • Jedlicka L.
      Subacromial corticosteroid injection for poststroke shoulder pain: an exploratory prospective case series.
      • Lakse E.
      • Gunduz B.
      • Erhan B.
      • Celik E.C.
      The effect of local injections in hemiplegic shoulder pain: a prospective, randomized, controlled study.
      it is still difficult to assess the efficacy because they either lack the number of participants, placebo control, or have the problem of an unblinded study design. This study was designed to complement previous studies; to our knowledge, this is the first randomized, placebo-controlled study to assess the efficacy of subacromial injection in poststroke HSP.
      Intra-articular and subacromial injections with corticosteroid are frequently used procedures in poststroke HSP but not enough evidence exists to prove its efficacy. A randomized trial
      • Snels I.A.
      • Beckerman H.
      • Twisk J.W.
      • et al.
      Effect of triamcinolone acetonide injections on hemiplegic shoulder pain: a randomized clinical trial.
      on the efficacy of intra-articular injection in HSP reported no improvement in pain and therefore did not recommend intra-articular injection. However, we believe the one-size-fits-all use of intra-articular injection on all HSP, without taking into consideration their various causes such as adhesive capsulitis, rotator cuff disorder, glenohumeral subluxation, complex regional pain syndrome, spasticity, and neuropathic pain, may have led to such result. Subsequent studies
      • Chae J.
      • Jedlicka L.
      Subacromial corticosteroid injection for poststroke shoulder pain: an exploratory prospective case series.
      • Lakse E.
      • Gunduz B.
      • Erhan B.
      • Celik E.C.
      The effect of local injections in hemiplegic shoulder pain: a prospective, randomized, controlled study.
      suggested the possibility of subacromial injection to HSP patients with evidence of rotator cuff disorder having a positive effect on pain and function. These suggestions were even more clearly confirmed by our study.
      The minimum clinically significant change of VAS pain severity is known to range from 1.3 to 3.0.
      • Todd K.H.
      • Funk K.G.
      • Funk J.P.
      • Bonacci R.
      Clinical significance of reported changes in pain severity.
      • Lee J.S.
      • Hobden E.
      • Stiell I.G.
      • Wells G.A.
      Clinically important change in the visual analog scale after adequate pain control.
      In several studies that used a subacromial injection of triamcinolone 40mg (the same dose as our study) to nonstroke subjects with rotator cuff disorders, the change from baseline VAS score was −3.6±2.9 and −3.5±2.9 at weeks 4 and 8, respectively,
      • Hong J.Y.
      • Yoon S.H.
      • Moon D.J.
      • Kwack K.S.
      • Joen B.
      • Lee H.Y.
      Comparison of high- and low-dose corticosteroid in subacromial injection for periarticular shoulder disorder: a randomized, triple-blind, placebo-controlled trial.
      and −4.0±2.6 at week 6.
      • Ucuncu F.
      • Capkin E.
      • Karkucak M.
      • et al.
      A comparison of the effectiveness of landmark-guided injections and ultrasonography guided injections for shoulder pain.
      However, in a previous study with stroke survivors, the change from baseline VAS score in the treatment group was −1.6±1.2 at week 4,
      • Lakse E.
      • Gunduz B.
      • Erhan B.
      • Celik E.C.
      The effect of local injections in hemiplegic shoulder pain: a prospective, randomized, controlled study.
      and in our study, −1.9±1.6 and −2.5±1.4 at weeks 4 and 8, respectively. This suggests that the subacromial injection may be less efficacious for stroke survivors than nonstroke patients in relieving pain. It is generally known that the pain is relieved due to the corticosteroid that reduces the inflammation of the rotator cuff.
      • Coombes B.K.
      • Bisset L.
      • Vicenzino B.
      Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials.
      But as mentioned earlier, various causes are involved in patients with HSP. It can be assumed, then, that anti-inflammatory property alone does not suffice for relieving pain in HSP patients.
      We used the Modified Barthel Index to find out whether the improvement in shoulder pain and AROM after the subacromial injection had an effect on the activities of daily living. Despite improvements in pain and AROM for treatment groups at week 8, there was no significant difference in the Modified Barthel Index between the 2 groups. This is probably because the Modified Barthel Index is not a test that is designed specifically for shoulders. For instance, it includes many items that have more to do with the functions of distal than proximal upper limbs, and functions that have little to do with upper limbs, such as toileting, stair up/down, defecation, voiding, and ambulation.

      Study Limitations

      First, in the case of stroke patients, it is difficult to diagnose the rotator cuff disorder accurately using a physical test due to the motor weakness of their hemiplegic limb. Studies on the correlation between a physical test and the ultrasonographic finding in diagnosing rotator cuff disorder in nonstroke subjects found the accuracy to be lower when a physical test alone was used for diagnosis, and therefore emphasized the added use of ultrasonographic evaluation for enhanced accuracy.
      • Naredo E.
      • Aguado P.
      • De Miguel E.
      • et al.
      Painful shoulder: comparison of physical examination and ultrasonographic findings.
      • Kim H.A.
      • Kim S.H.
      • Seo Y.I.
      Ultrasonographic findings of painful shoulders and correlation between physical examination and ultrasonographic rotator cuff tear.
      Based on these studies, we included participants only if a positive finding of rotator cuff disorder using the physical test correlated with those of the ultrasonographic evaluations. Still, there is (1) no standard criteria for diagnosing the rotator cuff disorder of poststroke HSP patients, (2) no evidence that the correlation of physical and ultrasonographic examination will improve the diagnostic accuracy of rotator cuff disorder in poststroke HSP patients, and (3) no way to figure out whether the discovered rotator cuff disorder is the cause of the current HSP because there is no knowledge about the prior shoulder anatomy. All these points may lead to a selection bias of this study.
      Second, although we tried to include HSP patients diagnosed only with rotator cuff disorder, those with other causes may have been included. For instance, clinical diagnosis criterion of the International Association for the Study of Pain
      • Sandroni P.
      • Low P.A.
      • Ferrer T.
      • Opfer-Gehrking T.L.
      • Willner C.L.
      • Wilson P.R.
      Complex regional pain syndrome I (CRPS I): prospective study and laboratory evaluation.
      was used to diagnose the complex regional pain syndrome, without 3-phase bone scan. This criterion is not always clear and may lead to inaccurate diagnosis. Also, while excluding the severe spasticity patients of grade 3 and beyond on the Modified Ashworth Scale from the study, we did not exclude less severe patients because most stroke patients have spasticity and cannot be entirely excluded.
      Third, there were 2 systematic reviews
      • Buchbinder R.
      • Green S.
      • Youd J.M.
      Corticosteroid injections for shoulder pain.
      • Coombes B.K.
      • Bisset L.
      • Vicenzino B.
      Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials.
      that claimed the corticosteroid injection may be of short-term benefit for shoulder pain. Also, the duration of action when using triamcinolone for periarticular injection is known to be 2 to 3 weeks.
      • Caldwell J.R.
      Intra-articular corticosteroids Guide to selection and indications for use.
      This study therefore set 8 weeks for a follow-up period. However, because a few articles report potential adverse effects of corticosteroid on tendons,
      • Coombes B.K.
      • Bisset L.
      • Vicenzino B.
      Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials.
      • Gaujoux-Viala C.
      • Dougados M.
      • Gossec L.
      Efficacy and safety of steroid injections for shoulder and elbow tendonitis: a meta-analysis of randomised controlled trials.
      longer-term effect of subacromial injection of about 6 or 12 months should be studied.
      Lastly, it is difficult to generalize because we excluded patients with severe motor weakness, or muscle power of deltoid less than grade 2, and those with subluxation. These are stroke patients with the highest risk and prevalence of HSP. And because these conditions are common in patients with HSP, excluding such patients would pose as a certain limitation.

      Conclusions

      To our knowledge, this is the first randomized, placebo-controlled study to assess the efficacy of subacromial corticosteroid injection in HSP patients with evidence of rotator cuff disorder. Subacromial injection showed improvement in shoulder pain, disability, and AROM; the duration of its efficacy continued up to 8 weeks after the single injection without serious complications. This study shows how subacromial injection can be an efficacious treatment for patients with HSP.
      • a
        G*Power3; Heinrich-Heine-University, 40225 Düsseldorf, Germany; available at: http://www.psycho.uni-duesseldorf.de/abteilungen/aap/gpower3.
      • b
        GE Healthcare Ultrasound Ltd, 65-1 Sangdaewon-dong, Sungnam, Republic of Korea.
      • c
        Samsung Medison Ltd, 995-7 Daechi-dong, Seoul, Republic of Korea.
      • d
        Philips Ultrasound Inc, 22100 Bothell Everett Hwy, Bothell, WA 98021.
      • e
        Version 17; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.

      Acknowledgment

      We thank Aeree Park, MA, for the translation of the manuscript.

      References

        • Gamble G.E.
        • Barberan E.
        • Laasch H.U.
        • Bowsher D.
        • Tyrrell P.J.
        • Jones A.K.
        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
        • Lindgren I.
        • Jonsson A.C.
        • Norrving B.
        • Lindgren A.
        Shoulder pain after stroke: a prospective population-based study.
        Stroke. 2007; 38: 343-348
        • Griffin J.W.
        Hemiplegic shoulder pain.
        Phys Ther. 1986; 66: 1884-1893
        • Neviaser A.S.
        • Hannafin J.A.
        Adhesive capsulitis: a review of current treatment.
        Am J Sports Med. 2010; 38: 2346-2356
        • Snels I.A.
        • Dekker J.H.
        • van der Lee J.H.
        • Lankhorst G.J.
        • Beckerman H.
        • Bouter L.M.
        Treating patients with hemiplegic shoulder pain.
        Am J Phys Med Rehabil. 2002; 81: 150-160
        • Wanklyn P.
        • Forster A.
        • Young J.
        Hemiplegic shoulder pain (HSP): natural history and investigation of associated features.
        Disabil Rehabil. 1996; 18: 497-501
        • Bender L.
        • McKenna K.
        Hemiplegic shoulder pain: defining the problem and its management.
        Disabil Rehabil. 2001; 23: 698-705
        • 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
        • Shah R.R.
        • Haghpanah S.
        • Elovic E.P.
        • et al.
        MRI findings in the painful poststroke shoulder.
        Stroke. 2008; 39: 1808-1813
        • Lanzer W.L.
        Clinical aspects of shoulder injuries.
        Radiol Clin North Am. 1988; 26: 157-160
        • 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?.
        Clin Rehabil. 2011; 25: 60-68
        • Snels I.A.
        • Beckerman H.
        • Twisk J.W.
        • et al.
        Effect of triamcinolone acetonide injections on hemiplegic shoulder pain: a randomized clinical trial.
        Stroke. 2000; 31: 2396-2401
        • Chae J.
        • Jedlicka L.
        Subacromial corticosteroid injection for poststroke shoulder pain: an exploratory prospective case series.
        Arch Phys Med Rehab. 2009; 90: 501-506
        • Lakse E.
        • Gunduz B.
        • Erhan B.
        • Celik E.C.
        The effect of local injections in hemiplegic shoulder pain: a prospective, randomized, controlled study.
        Am J Phys Med Rehabil. 2009; 88: 805-811
        • Buchbinder R.
        • Green S.
        • Forbes A.
        • Hall S.
        • Lawler G.
        Arthrographic joint distension with saline and steroid improves function and reduces pain in patients with painful stiff shoulder: results of a randomised, double blind, placebo controlled trial.
        Ann Rheum Dis. 2004; 63: 302-309
        • Sandroni P.
        • Low P.A.
        • Ferrer T.
        • Opfer-Gehrking T.L.
        • Willner C.L.
        • Wilson P.R.
        Complex regional pain syndrome I (CRPS I): prospective study and laboratory evaluation.
        Clin J Pain. 1998; 14: 282-289
        • Bohannon R.W.
        • Smith M.B.
        Interrater reliability of a modified Ashworth scale of muscle spasticity.
        Phys Ther. 1987; 67: 206-207
        • Boyd E.A.
        • Torrance G.M.
        Clinical measures of shoulder subluxation: their reliability.
        Can J Public Health. 1992; 83: S24-S28
        • Faul F.
        • Erdfelder E.
        • Lang A.
        • Buchner A.
        G* Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences.
        Behav Res Methods. 2007; 39: 175-191
        • Petri M.
        • Dobrow R.
        • Neiman R.
        • Whiting-O'Keefe Q.
        • Seaman W.E.
        Randomized, double-blind, placebo-controlled study of the treatment of the painful shoulder.
        Arthritis Rheum. 1987; 30: 1040-1045
        • Neer 2nd, C.S.
        • Welsh R.P.
        The shoulder in sports.
        Orthop Clin North Am. 1977; 8: 583-591
        • Hawkins R.J.
        • Kennedy J.C.
        Impingement syndrome in athletes.
        Am J Sports Med. 1980; 8: 151-158
        • Yocum L.A.
        Assessing the shoulder.
        Clin Sports Med. 1983; 2: 281-289
        • Jobe F.W.
        • Jobe C.M.
        Painful athletic injuries of the shoulder.
        Clin Orthop Relat Res. 1983; : 117-124
        • Leroux J.L.
        • Thomas E.
        • Bonnel F.
        • Blotman F.
        Diagnostic value of clinical tests for shoulder impingement syndrome.
        Rev Rhum Engl Ed. 1995; 62: 423-428
        • Bianchi S.
        • Martinoli C.
        • Baert A.
        • et al.
        Ultrasound of the musculoskeletal system.
        Springer Verlag, Berlin2007
        • Naredo E.
        • Aguado P.
        • De Miguel E.
        • et al.
        Painful shoulder: comparison of physical examination and ultrasonographic findings.
        Ann Rheum Dis. 2002; 61: 132-136
        • Altman D.G.
        • Bland J.M.
        How to randomise.
        BMJ. 1999; 319: 703-704
        • Kapetanos G.
        The effect of the local corticosteroids on the healing and biomechanical properties of the partially injured tendon.
        Clin Orthop Relat Res. 1982; : 170-179
        • Grant H.J.
        • Arthur A.
        • Pichora D.R.
        Evaluation of interventions for rotator cuff pathology: a systematic review.
        J Hand Ther. 2004; 17: 274-299
        • Kuhn J.E.
        Exercise in the treatment of rotator cuff impingement: a systematic review and a synthesized evidence-based rehabilitation protocol.
        J Shoulder Elbow Surg. 2009; 18: 138-160
        • Jung H.Y.
        • Park B.K.
        • Shin H.S.
        • et al.
        Development of the Korean Version of Modified Barthel Index (K-MBI): multi-center study for subjects with stroke.
        J Korean Acad Rehabil Med. 2007; 31: 283-297
        • Croft P.
        • Pope D.
        • Zonca M.
        • O'Neill T.
        • Silman A.
        Measurement of shoulder related disability: results of a validation study.
        Ann Rheum Dis. 1994; 53: 525-528
        • Edwards T.B.
        • Bostick R.D.
        • Greene C.C.
        • Baratta R.V.
        • Drez D.
        Interobserver and intraobserver reliability of the measurement of shoulder internal rotation by vertebral level.
        J Shoulder Elbow Surg. 2002; 11: 40-42
        • Hoving J.L.
        • Buchbinder R.
        • Green S.
        • et al.
        How reliably do rheumatologists measure shoulder movement?.
        Ann Rheum Dis. 2002; 61: 612-616
        • Buchbinder R.
        • Green S.
        • Youd J.M.
        Corticosteroid injections for shoulder pain.
        Cochrane Database Syst Rev. 2003; (CD004016)
        • Todd K.H.
        • Funk K.G.
        • Funk J.P.
        • Bonacci R.
        Clinical significance of reported changes in pain severity.
        Ann Emerg Med. 1996; 27: 485-489
        • Lee J.S.
        • Hobden E.
        • Stiell I.G.
        • Wells G.A.
        Clinically important change in the visual analog scale after adequate pain control.
        Acad Emerg Med. 2003; 10: 1128-1130
        • Hong J.Y.
        • Yoon S.H.
        • Moon D.J.
        • Kwack K.S.
        • Joen B.
        • Lee H.Y.
        Comparison of high- and low-dose corticosteroid in subacromial injection for periarticular shoulder disorder: a randomized, triple-blind, placebo-controlled trial.
        Arch Phys Med Rehabil. 2011; 92: 1951-1960
        • Ucuncu F.
        • Capkin E.
        • Karkucak M.
        • et al.
        A comparison of the effectiveness of landmark-guided injections and ultrasonography guided injections for shoulder pain.
        Clin J Pain. 2009; 25: 786-789
        • Coombes B.K.
        • Bisset L.
        • Vicenzino B.
        Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials.
        Lancet. 2010; 376: 1751-1767
        • Kim H.A.
        • Kim S.H.
        • Seo Y.I.
        Ultrasonographic findings of painful shoulders and correlation between physical examination and ultrasonographic rotator cuff tear.
        Mod Rheumatol. 2007; 17: 213-219
        • Caldwell J.R.
        Intra-articular corticosteroids.
        Drugs. 1996; 52: 507-514
        • Gaujoux-Viala C.
        • Dougados M.
        • Gossec L.
        Efficacy and safety of steroid injections for shoulder and elbow tendonitis: a meta-analysis of randomised controlled trials.
        Ann Rheum Dis. 2009; 68: 1843-1849