Volume 88, Issue 3, Pages 298-301 (March 2007)

8 of 37

ABSTRACT

FULL TEXT

FULL-TEXT PDF (84 KB)

Poststroke Shoulder Pain: Its Relationship to Motor Impairment, Activity Limitation, and Quality of Life

Presented in part to the Association of Academic Physiatrists, March 2006, Daytona, FL.

Abstract

Chae J, Mascarenhas D, Yu DT, Kirsteins A, Elovic EP, Flanagan SR, Harvey RL, Zorowitz RD, Fang Z-P. Poststroke shoulder pain: its relationship to motor impairment, activity limitation, and quality of life.

Objective

To assess the relationship between poststroke shoulder pain, upper-limb motor impairment, activity limitation, and pain-related quality of life (QOL).

Design

Cross-sectional, secondary analysis of baseline data from a multisite clinical trial.

Setting

Outpatient rehabilitation clinics of 7 academic medical centers.

Participants

Volunteer sample of 61 chronic stroke survivors with poststroke shoulder pain and glenohumeral subluxation.

Interventions

Not applicable.

Main Outcome Measures

We measured poststroke shoulder pain with the Brief Pain Inventory question 12 (BPI 12), a self-reported 11-point numeric rating scale (NRS) that assesses “worst pain” in the last 7 days. Motor impairment was measured with the Fugl-Meyer Assessment (FMA). Activity limitation was measured with the Arm Motor Ability Test (AMAT) and the FIM instrument. Pain-related QOL was measured with BPI question 23, a self-reported 11-point NRS that assesses pain interference with general activity, mood, walking ability, normal work, interpersonal relationships, sleep, and enjoyment of life.

Results

Stepwise regression analyses indicated that poststroke shoulder pain is associated with the BPI 23, but not with the FMA, FIM, or AMAT scores.

Conclusions

Poststroke shoulder pain is associated with reduced QOL, but not with motor impairment or activity limitation.

Key Words: Pain, Quality of life, Rehabilitation, Shoulder, Stroke

Article Outline

• Abstract

• Methods

• Participants

• Explanatory Measures

• Dependent Measures

• Analysis

• Results

• Discussion

• Conclusions

• References

• Copyright

SHOULDER PAIN IS A COMMON complication of hemiplegia; its reported prevalence ranges between 5% and 84%.1, 2 Numerous studies have reported a relationship between poststroke shoulder pain and limited shoulder external rotation range of motion (ROM),3 sensory impairment,4 adhesive capsulitis,5 impingement,6 subluxation,7 spasticity,8 and complex regional pain syndrome (CRPS).9 Its relationship to motor impairment and activity limitation is less clear. Roy et al10 showed that stroke survivors with shoulder pain have more severe motor impairment during recovery. Others, however, have reported no relationship between shoulder pain and motor impairment.6, 11 In a study of 108 stroke survivors discharged from a hospital, Wanklyn et al12 reported that patients with shoulder pain had significantly greater activity limitation than patients without pain, based on their Barthel Index of disability scores. Another study,4 however, showed no relationship between shoulder pain and the Barthel Index scores. Finally, other authors2, 13 have indicated a relationship between shoulder pain and pain-related quality of life (QOL). This relationship has not been quantitatively demonstrated, however. While the pain experience alone is a sufficient reason for treatment, the importance of treating poststroke shoulder pain is further emphasized if it can be shown that it has a relationship to motor impairment, activity limitation, and QOL.

Our objective in this cross-sectional study was to test the hypothesis that poststroke shoulder pain, motor impairment, activity limitation, and pain-related QOL are statistically related.

Methods

Participants

We analyzed the baseline data of stroke survivors enrolled in a multicenter randomized clinical trial of percutaneous electric stimulation for the treatment of poststroke shoulder pain.14, 15 The clinical trials protocol was approved by the institutional review boards at each participating center. Participants were more than 12 weeks poststroke (hemorrhagic or nonhemorrhagic) and were at least 18 years old. Participants had (1) shoulder pain graded as at least 2 on the 11-point numeric rating scale (NRS) of the Brief Pain Inventory16 question 12 (BPI 12), (2) at least one-half fingerbreadth of inferior glenohumeral separation by palpation with the affected limb in a dependent position without manual traction, and (3) cognitive ability to fulfill study requirements (able to recall 3 objects after 30min and use an NRS). Patients were excluded if they had a history of arrhythmia with hemodynamic instability, previous stroke with persistent neurologic deficit, prestroke shoulder pathology, CRPS, any implantable stimulator, or uncontrolled seizures (>1/mo).

Explanatory Measures

We defined the primary explanatory or independent variable as being the BPI 12. In studies that predict outcomes, the explanatory or independent variable is the variable that predicts a particular outcome or predicts the dependent variable. The BPI is a multiple-question self-reported metric that assesses both pain intensity (sensory dimension) and the interference (reactive dimension) of pain with daily activities and QOL. The BPI has shown both reliability and validity across cultures and languages.16, 17 Its developers have suggested that BPI 12, the “pain worst” rating, may be selected as the primary response variable. The question asks participants to rate their worst shoulder pain in the last week on an 11-point NRS of 0 (no pain) to 10 (pain as bad as you can imagine).

We identified secondary explanatory measures because factors other than BPI 12 might influence motor impairment, activity limitation, and QOL. Based on results of previous studies, we added pain-free external rotation ROM as a secondary pain–related explanatory measure.3, 18 Time from stroke onset to study entry, stroke type (hemorrhagic vs nonhemorrhagic), side of hemiplegia, radiographic inferior subluxation,19, 20 and the Modified Ashworth Scale (MAS) score of the elbow flexors21 were added as secondary stroke specific explanatory measures. We selected the MAS score of the elbow flexors to avoid the confounding effect of pain associated with shoulder manipulation. The MAS requires that the range be taken to its maximum, limited only by the tone or fixed joint contracture. Many stroke patients, however, have reduced range of shoulder abduction and external rotation because of pain with or without increased tone. Thus, assessment of the shoulder is unlikely to provide an accurate assessment of tone. We selected the elbow on the assumption that tone is generalized.

Dependent Measures

Motor impairment and activity limitations were assessed with laboratory-based measures. Motor impairment was assessed with the upper-limb component of the Fugl-Meyer Assessment (FMA). We used 2 measures of activity limitation. Upper-limb specific activity limitation was assessed with the self-care component of the FIM instrument and the functional domain of the Arm Motor Ability Test (AMAT). Pain-related QOL was assessed with BPI question 23 (BPI 23), a self-reported pain questionnaire. Trained therapists administered all measures.

The upper-limb component of the FMA considers evolving synergy patterns as well as isolated strength, coordination, and hypertonia. The FMA’s reliability and validity have been documented.22, 23, 24 After rigorously evaluating its measurement properties, Gladstone et al concluded, “Based on the available evidence, the Fugl-Meyer Motor scale is recommended highly as a clinical and research tool for evaluating changes in motor impairment following stroke.”25(p232)

The self-care component of the FIM instrument assesses upper-limb−dependent tasks such as feeding, dressing, bathing, and toileting. The measure permits compensatory strategies and participants may use the unaffected upper limb to perform the tasks. The FIM instrument’s validity, structure, and stability have been demonstrated.26

In contrast to the FIM instrument, the AMAT assesses hemiparetic upper-limb−specific functional tasks and does not permit compensatory strategies. Unilateral tasks are performed with the affected upper limb. Bilateral tasks are performed using (or attempting to use) the dominant extremities in the same roles as before the stroke. The correlation between the functional ability and quality of movement domains of the AMAT is very close to 1.27 Therefore, as recommended by Kopp et al,27 we used only the functional ability domain. The AMAT has been shown to be reliable, sensitive, valid, and internally consistent.27, 28

We assessed pain-related QOL with the BPI 23, which assesses the degree to which pain interferes with a combination of daily activities that includes general activity, mood, walking ability, normal work, interpersonal relationships, sleep, and enjoyment of life. Interference is assessed on an 11-point NRS, where 0 indicates “does not interfere” and 10 indicates “completely interferes.” The summary score is the mean of all 7 domains. Psychometrics was reported earlier in the context of BPI 12.

Analysis

To evaluate the relationship between the explanatory and dependent variables, we generated 4 stepwise linear regression models corresponding to the 4 dependent measures. Table 1 shows the specific explanatory and dependent variables for each model. All models included BPI 12 as the primary explanatory pain variable and pain-free external rotation ROM as a secondary explanatory pain variable. All models also included time from stroke onset to study entry, stroke type, and side of hemiplegia, radiographic inferior subluxation, and MAS score as secondary stroke specific explanatory variables. Because activity limitation may be influenced by motor impairment, we added the FMA as an additional explanatory variable for the second and third models. We added the FIM and AMAT as additional explanatory variables for the fourth model because activity limitation may influence QOL.

Table 1.

Explanatory and Dependent Variables for the 4 Linear Regression Models


Model	Primary Explanatory Variable	Secondary Explanatory Variables	Dependent Variable
1	BPI 12	Pain-free external ROM, MAS, stroke onset to study entry, stroke type, side of hemiplegia, subluxation	FMA
2	BPI 12	Pain-free external ROM, FMA, MAS, stroke onset to study entry, stroke type, side of hemiplegia, subluxation	FIM
3	BPI 12	Pain-free external ROM, FMA, MAS, stroke onset to study entry, stroke type, side of hemiplegia, subluxation	AMAT
4	BPI 12	Pain-free external ROM, FMA, MAS, stroke onset to study entry, stroke type, side of hemiplegia, subluxation	BPI23

Results

Data were available for all 61 stroke survivors enrolled in the clinical trial. The mean age ± standard deviation (SD) of participants was 58.6±12.1 years, 44% were women, 84% sustained nonhemorrhagic strokes, and 62% had left hemiplegia. Other characteristics of the participants are shown in table 2.

Table 2.

Participant Characteristics


	Variable	Mean ± SD
	Time from stroke onset to study entry (mo)	58.4±12.1
	BPI 12 score	7.1±2.3
	BPI 23 score	4.3±2.8
	Subluxation (mm)	7.3±8.5
	Pain-free external ROM (deg)	37.3±21.6
	FMA score	18.7±12.6
	MAS score	1.8±1.2
	AMAT function score	1.0±1.0
	FIM self-care score	30.4±7.8

Table 3 shows the results of the stepwise regression analyses. BPI 12 was not associated with the FMA, but pain-free external rotation ROM and the degree of inferior subluxation were directly and inversely related to the FMA, respectively (model 1). The model explained 15% of the variance in the FMA. None of the explanatory variables was associated with FIM self-care (model 2). BPI 12 was not associated with AMAT function (model 3). AMAT function was directly related to FMA, however, and lower AMAT function scores were associated with the nonhemorrhagic stroke subtype. The model explained 72% of the variance in AMAT function. BPI 12 was directly related to BPI 23, explaining 20% of the variance in BPI 23 (model 4). None of the other explanatory measures was associated with BPI 23.

Table 3.

Results of Linear Regression Analyses


Model	Dependent Variable	Significant Factor(s)	β	F	R2	P
1	FMA	Overall model	NA	4.9	.15	.01
		Inferior subluxation	−.28	NA	NA	.023
		External rotation ROM	.26	NA	NA	.037
2	FIMself-care	⁎	⁎	⁎	⁎	⁎
3	AMATfunction	Overall model	NA	74.6	.72	<.001
		FMA	.83	NA	NA	<.001
		Stroke type	−.20	NA	NA	.005
4	BPI23	BPI 12	.46	15.4	.20	<.001

Abbreviation: NA, not applicable.

⁎

No significant factor.

Discussion

Poststroke shoulder pain was associated with reduced QOL related to pain. Our study, however, failed to demonstrate a statistical relationship between poststroke shoulder pain and motor impairment and activity limitation.

In an earlier World Health Organization definition,29 health and QOL reflected the constructs of physical, mental, and social well-being and not merely the absence of disease. Today, however, QOL is generally referred to as a multidimensional construct involving the physical, emotional, functional, and social domains, which allows us to view the impact of disability, illness, or pain on a person as a whole.30 Consistent with this construct, in this study we used the BPI 23 to assess the degree to which pain interferes with a combination of activities that include general activity, walking ability, mood, vocation, relationships, sleep, and general enjoyment of life. To interfere with specific tasks does not necessarily mean that shoulder pain prevents the completion of the tasks. The pain may simply make the tasks more difficult by requiring greater investment of emotional and volitional effort. Accordingly, BPI 23 has been used as a QOL measure in cancer31 and in low back pain interventional32 studies.

BPI 12, a self-reported measure of “worst pain” experienced in the 7 days before the assessment, was associated with pain-related QOL, but pain-free external rotation ROM was not. This apparent incongruity may be the result of the distinctiveness of each measure. Pain-free external rotation ROM likely overestimates the severity of daily pain. The pain experience is increased when a shortened muscle is stretched or if soft tissue is impinged between the humeral head and the acromion process on movement. Thus, pain-free external rotation ROM introduces artificial constraints that may not be relevant to participants’ routine daily activities. In contrast, BPI 12 assesses worst pain during the past week without specifying a level of activity so that the pain score is most relevant to participants’ real-life activities. Thus, it is not surprising that the stroke survivors’ perception of pain-related QOL was associated with BPI 12, but not pain-free external rotation ROM.

Contrary to expectations, BPI 12 was not associated with the FMA, FIM, or AMAT scores, for which there are several possible explanations. First, this study only included stroke survivors with poststroke shoulder pain. Previous studies10, 12 that reported a relationship between poststroke shoulder pain, motor impairment, and activity limitation assessed stroke survivors with and without shoulder pain. Within this broader population of stroke survivors, those with shoulder pain exhibited greater motor impairment and activity limitations. Our subjects all had significant shoulder pain with severe motor impairments and a mean FMA score of 18.7 out of a maximum of 66. The requirement of subluxation, which we correlated with severity of motor impairment, likely presents a selection bias and may have influenced the relationship between poststroke shoulder pain and motor impairment. Similarly, participants had severely impaired hemiparetic upper-limb specific activity limitations, with a mean AMAT function score of 1.0 out of a maximum of 5. Thus, it is possible that the presence of pain is more important than its degree in predicting motor impairment and activity limitation. Second, with regard to FIM self-care scores, the lack of a relationship to BPI 12 may be because compensatory strategies were allowed. Because the FIM instrument permits the use of the unaffected limb, pain in the poststroke shoulder may be less relevant. Accordingly, participants’ FIM self-care scores were relatively high, with a mean score of 30.4 out of a maximum of 42. Third, because both the FIM instrument and the AMAT assess task completion, it is possible that participants experienced significant pain but were still able to complete at least some of the prescribed tasks. Finally, the FMA, FIM instrument, and AMAT assess performance in the laboratory whereas BPI 12 and 23 reflect real-life experiences and activities, respectively. Participant performance in the laboratory and in real-life activities may not be congruent. Specifically, the incentive to complete a task is higher in the laboratory than at home, where some factors may actually provide disincentives. A factor such as pain, which may inhibit but not preclude certain tasks, has a greater effect in lower incentive environments.

Conclusions

Poststroke shoulder pain is associated with pain-related QOL. This provides a further incentive to develop effective rehabilitation prevention and treatment strategies for poststroke shoulder pain. Our data did not identify a relationship between poststroke shoulder pain, motor impairment, and activity limitation. This, however, may have been an artifact of a study design that did not include stroke survivors free of poststroke shoulder pain, rather than a true lack of a relationship.

References

1. 1Bender L, McKenna K. Hemiplegic shoulder pain: defining the problem and its management. Disabil Rehabil. 2001;23:698–705. MEDLINE | CrossRef

2. 2Turner-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. MEDLINE | CrossRef

3. 3Bohannon RW, Larkin PA, Smith MB, Horton MG. Shoulder pain in hemiplegia: statistical relationship with five variables. Arch Phys Med Rehabil. 1986;67:514–516. MEDLINE

4. 4Gamble 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. Abstract | Full-Text PDF (50 KB) | CrossRef

5. 5Ikai T, Tei K, Yoshida K, Miyano S, Yonemoto K. Evaluation and treatment of shoulder subluxation in hemiplegia: relationship between subluxation and pain. Am J Phys Med Rehabil. 1998;77:421–426. MEDLINE | CrossRef

6. 6Joynt RL. The source of shoulder pain in hemiplegia. Arch Phys Med Rehabil. 1992;73:409–413. MEDLINE

7. 7Lo SF, Chen SY, Lin HC, Jim YF, Meng NH, Kao MJ. Arthrographic and clinical findings in patients with hemiplegic shoulder pain. Arch Phys Med Rehabil. 2003;84:1786–1791. Abstract | Full Text | Full-Text PDF (682 KB) | CrossRef

8. 8Van Ouwenaller C, Laplace PM, Chantraine A. Painful shoulder in hemiplegia. Arch Phys Med Rehabil. 1986;67:23–26. MEDLINE

9. 9Tepperman PS, Greyson ND, Hilbert L, Jimenez J, Williams JI. Reflex sympathetic dystrophy in hemiplegia. Arch Phys Med Rehabil. 1984;65:442–447. MEDLINE

10. 10Roy CW, Sands MR, Hill LD. Shoulder pain in acutely admitted hemiplegics. Clin Rehabil. 1994;8:334–340. CrossRef

11. 11Zorowitz RD. Recovery patterns of shoulder subluxation after stroke: a six-month follow-up study. Top Stroke Rehabil. 2001;8:1–9. MEDLINE | CrossRef

12. 12Wanklyn P, Forster A, Young J. Hemiplegic shoulder pain (HSP): natural history and investigation of associated features. Disabil Rehabil. 1996;18:497–501. MEDLINE | CrossRef

13. 13Vuagnat H, Chantraine A. Shoulder pain in hemiplegia revisited: contribution of functional electrical stimulation and other therapies. J Rehabil Med. 2003;35:49–54quiz 56. MEDLINE | CrossRef

14. 14Chae 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. MEDLINE | CrossRef

15. 15Yu DT, Chae J, Walker ME, et al. Intramuscular neuromuscular electric stimulation for poststroke shoulder pain: a multicenter randomized clinical trial. Arch Phys Med Rehabil. 2004;85:695–704. Abstract | Full Text | Full-Text PDF (160 KB) | CrossRef

16. 16Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain Inventory. Ann Acad Med Singapore. 1994;23:129–138. MEDLINE

17. 17Tan G, Jensen MP, Thornby JI, Shanti BF. Validation of the Brief Pain Inventory for chronic nonmalignant pain. J Pain. 2004;5:133–137. Abstract | Full Text | Full-Text PDF (93 KB) | CrossRef

18. 18Yu DT, Chae J, Walker ME, Fang ZP. Percutaneous intramuscular neuromuscular electric stimulation for the treatment of shoulder subluxation and pain in patients with chronic hemiplegia: a pilot study. Arch Phys Med Rehabil. 2001;82:20–25. Abstract | Full Text | Full-Text PDF (52 KB) | CrossRef

19. 19Boyd EA, Goudreau L, O’Riain MD, Grinnell DM, Torrance GM, Gaylard A. A radiological measure of shoulder subluxation in hemiplegia: its reliability and validity. Arch Phys Med Rehabil. 1993;74:188–193. MEDLINE

20. 20Prevost R, Arsenault AB, Dutil E, Drouin G. Shoulder subluxation in hemiplegia: a radiologic correlational study. Arch Phys Med Rehabil. 1987;68:782–785. MEDLINE

21. 21Brashear A, Zafonte R, Corcoran M, et al. Inter- and intrarater reliability of the Ashworth Scale and the Disability Assessment Scale in patients with upper-limb poststroke spasticity. Arch Phys Med Rehabil. 2002;83:1349–1354. Abstract | Full Text | Full-Text PDF (79 KB) | CrossRef

22. 22Berglund K, Fugl-Meyer AR. Upper extremity function in hemiplegia (A cross-validation study of two assessment methods). Scand J Rehabil Med. 1986;18:155–157. MEDLINE

23. 23Sanford J, Moreland J, Swanson LR, Stratford PW, Gowland C. Reliability of the Fugl-Meyer assessment for testing motor performance in patients following stroke. Phys Ther. 1993;73:447–454. MEDLINE

24. 24Duncan PW, Propst M, Nelson SG. Reliability of the Fugl-Meyer assessment of sensorimotor recovery following cerebrovascular accident. Phys Ther. 1983;63:1606–1610. MEDLINE

25. 25Gladstone DJ, Danells CJ, Black SE. The Fugl-Meyer assessment of motor recovery after stroke: a critical review of its measurement properties. Neurorehabil Neural Repair. 2002;16:232–240. MEDLINE

26. 26Stineman MG, Shea JA, Jette A, et al. The Functional Independence Measure: tests of scaling assumptions, structure, and reliability across 20 diverse impairment categories. Arch Phys Med Rehabil. 1996;77:1101–1108. Abstract | Full-Text PDF (948 KB) | CrossRef

27. 27Kopp B, Kunkel A, Flor H, et al. The Arm Motor Ability Test: reliability, validity, and sensitivity to change of an instrument for assessing disabilities in activities of daily living. Arch Phys Med Rehabil. 1997;78:615–620. Abstract | Full-Text PDF (843 KB) | CrossRef

28. 28Chae J, Yang G, Labatia I. Upper limb motor function in chronic hemiparesis: concurrent validity of the Arm Motor Ability Test. Am J Phys Med Rehabil. 2003;82:1–8. MEDLINE | CrossRef

29. 29World Health Organization. International classification of impairments, disabilities, and handicap: a manual of classification relating to consequences of disease. Geneva: WHO; 1980;.

30. 30Tulsky D, Chiaravalloti N. Measuring quality of life in rehabilitation medicine. In: DeLisa J, Gans B, Walsh NE editor. 4th ed.. Physical medicine and rehabilitation: principles and practices. Vol 2:Philadelphia: Lippincott Williams & Wilkins; 2005;p. 1193–1204.

31. 31Hwang SS, Chang VT, Kasimis B. Dynamic cancer pain management outcomes: the relationship between pain severity, pain relief, functional interference, satisfaction and global quality of life over time. J Pain Symptom Manage. 2002;23:190–200. Abstract | Full Text | Full-Text PDF (104 KB) | CrossRef

32. 32Gimbel J, Linn R, Hale M, Nicholson B. Lidocaine patch treatment in patients with low back pain: results of an open-label, nonrandomized pilot study. Am J Ther. 2005;12:311–319. MEDLINE | CrossRef

a Cleveland Functional Electrical Stimulation Center, Cleveland, OH

b Department of Physical Medicine and Rehabilitation, Case Western Reserve University, Cleveland, OH

c Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH

d Charlotte Institute for Rehabilitation, Charlotte, NC

e Kessler Medical Rehabilitation Research and Education Corp, West Orange, NJ

f Department of Rehabilitation Medicine, Mt. Sinai School of Medicine, New York, NY

g Rehabilitation Institute of Chicago, Chicago, IL

h Department of Rehabilitation Medicine, University of Pennsylvania, Philadelphia, PA

i NeuroControl Corp, North Ridgeville, OH.

Reprint requests to John Chae, MD, ME, Dept of Physical Medicine and Rehabilitation, Case Western Reserve University, MetroHealth Medical Center, 2500 MetroHealth Dr, Cleveland, OH, 44109

Supported in part by the National Institute for Child Health and Human Development (grant nos. R44HD34996, K12HD01097), the National Center for Research Resource (grant no. M01RR0080), and by the NeuroControl Corp, North Ridgeville, OH.

A commercial party having a direct financial interest in the results of the research supporting this article has conferred or will confer a financial benefit upon the author or 1 or more of the authors. NeuroControl Corp has a direct interest in the content of this article with respect to a device NeuroControl intends to commercialize. Chae is a consultant to NeuroControl. Fang is an employee of NeuroControl. At the time of the study, Yu was a consultant to NeuroControl, but is no longer affiliated with NeuroControl.

1 Yu is now affiliated with Virginia Mason Medical Center, Seattle, WA, and Bioness Inc, Santa Clarita, CA

2 Kirsteins is now affiliated with Moses Cones Health System, Greensboro, NC

3 Zorowitz is now affiliated with Johns Hopkins Bayview Medical Center, Baltimore, MD.

PII: S0003-9993(06)01562-0

doi:10.1016/j.apmr.2006.12.007

8 of 37