|Year : 2019 | Volume
| Issue : 1 | Page : 35-41
A clinical, electrophysiological, and imaging study on the different causes of poststroke shoulder pain
Tarek Saad Shafshak1, Mowaffak Moustafa Abdelhamid1, Mohamed Ihab Samy Reda2, Marwa Abdullah Amer1
1 Department of Physical Medicine, Rheumatology and Rehabilitation, Faculty of Medicine, University of Alexandria, Alexandria, Egypt
2 Department of Radiodiagnosis, Faculty of Medicine, University of Alexandria, Alexandria, Egypt
|Date of Web Publication||22-May-2019|
Dr. Marwa Abdullah Amer
Department of Physical Medicine, Rheumatology and Rehabilitation, Faculty of Medicine, Alexandria University, Al-Khartoum Square, Alexandria, 21131
Source of Support: None, Conflict of Interest: None
Background: Poststroke shoulder pain (PSSP) is a complex phenomenon. Defining the exact etiology would facilitate proper management. Aim: The aim of the study is to identify the different causes of PSSP. Settings and Design: Observational cross-sectional study at institution, university hospital, tertiary level of clinical care. Methods: A total of 30 stroke survivors with PSSP were enrolled in this study. Patients were subjected to neuromuscular clinical examination of the shoulder, and evaluation (spasticity assessment using Modified Ashworth scale and Fugl-Meyer upper extremity scale to assess impairment). The shoulder pain was evaluated using the numerical rating scale and the neuropathic pain diagnostic questionnaire (DN4). Electrophysiological studies (nerve conduction and electromyography) and imaging (plain radiograph and magnetic resonance imaging [MRI]) were done to reveal neuromusculoskeletal causes for PSSP. Results: One or more musculoskeletal cause occurred in all patients. These were rotator cuff tear/tendinopathy (90%), impingement syndrome (90%), bursitis (66.7%), and adhesive capsulitis (50%). Concomitant proximal neuropathies including (upper trunk brachial plexopathy, axillary neuropathy, and suprascapular neuropathy) were found in 40% of PSSP patients. The DN4 questionnaire was positive in only two patients. Statistical Analysis Used: Statistical Package for Social Sciences (SPSS version 20 Chicago, IL, USA). Conclusion: The most common causes of PSSP were the musculoskeletal causes. MRI studies were helpful in revealing the common causes of shoulder pain such as rotator cuff tear and bursitis in those disabled population in whom clinical assessment could be difficult.
Keywords: Magnetic resonance imaging, shoulder pain, stroke, survivors
|How to cite this article:|
Shafshak TS, Abdelhamid MM, Reda MI, Amer MA. A clinical, electrophysiological, and imaging study on the different causes of poststroke shoulder pain. J Int Soc Phys Rehabil Med 2019;2:35-41
|How to cite this URL:|
Shafshak TS, Abdelhamid MM, Reda MI, Amer MA. A clinical, electrophysiological, and imaging study on the different causes of poststroke shoulder pain. J Int Soc Phys Rehabil Med [serial online] 2019 [cited 2022 Jan 20];2:35-41. Available from: https://www.jisprm.org/text.asp?2019/2/1/35/258766
| Introduction|| |
Although the experience of pain is a survival mechanism; poststroke pain (PSP) can lead to poor recovery, reduced quality of life, and adds further challenge to stroke recovery journey.
Poststroke shoulder pain (PSSP), the most common type of PSP, is a complex phenomenon, and the exact etiology is unknown. PSSP could be related to soft-tissue lesions, impaired motor control, and altered peripheral or central nervous system activity with each one of these factors could present as a separate phenomenon, coexist, or evolve to trigger each other.
The objective of this study was to identify the different causes of PSSP.
| Methods|| |
Design and participants
This observational cross-sectional study enrolled 30 stroke survivors with stroke duration >1 month (28 patients had ischemic and 2 patients had hemorrhagic stroke) with PSSP. This was the maximum number of patients collected in 18 months. Exclusion criteria: history of shoulder pain before stroke onset, history of shoulder trauma or surgery, chronic inflammatory arthritis, and severe cognitive impairment that impeded communication.
The clinical diagnosis of stroke was made by a neurologist on the basis of the criteria of the World Health Organization (i.e., signs of focal disturbance of cerebral function that lasted longer than 24 h and had no apparent origin other than vascular). The diagnosis of stroke was confirmed by imaging technique [computed tomography or magnetic resonance imaging (MRI)].
Patients enrolled in this study were recruited from those attending the outpatient clinic of physical medicine, rheumatology, and rehabilitation department in a university hospital in the area where the research was done. All participants were informed about the nature of the study, and an informed consent was taken from all patients and was approved by the local ethics committee.
The following data were collected from all participants: (1) demographic data collection (age, sex, duration of stroke, and history of previous stroke); and (2) analysis of PSSP including – onset of shoulder pain after stroke, duration of PSSP, character of shoulder pain (words used by patients to describe pain), course, progression; besides relieving and aggravating factors.
All the studied patients underwent clinical neuromusculoskeletal examination for the shoulder, electrophysiological studies (nerve conduction and electromyography [EMG]) and imaging studies. Imaging included plain X-ray and MRI.
- Pain intensity was assessed using the numerical rating scale in which the patient selects a whole number (0–10 integers) with 0 representing no pain and 10 representing the extreme pain he can imagine
- Neuropathic pain diagnostic questionnaire (DN4) scores. DN4 was used to detect neuropathic component to PSSP. DN4 scale ranges from 0 to 10. A DN4 score of 4 or more is considered diagnostic for definite neuropathic pain
- Shoulder muscle power testing using the Medical Research Council scoring system
- Hemiplegic shoulder passive range of motion (ROM)
- Spasticity around the shoulder was evaluated using the Modified Ashworth scale (MAS). Scoring range 0–4
- Shoulder subluxation was assessed by the palpation method of the space between the acromion and the head of the humerus. The sulcus sign was graded as <1.0 cm (Grade I), 1.0–2.0 cm (Grade II), or >2.0 cm (Grade III)
- Assessment of impairment level of the arm function using Fugl-Meyer for the upper extremity. It included (a) motor performance scale (33 items - maximal score = 66), each item is rated on a 3-point ordinal scale (0 = cannot perform, 1 = performs partially, and 2 = performs fully); (b) sensory assessment (6 items - maximal score = 12). A higher score indicates less impairment.
- Nerve conduction studies (NCSs) were done to detect possible proximal nerve lesions. These included axillary, musculocutaneous, and suprascapular motor NCSs, besides lateral antebrachial sensory conduction study (on the paretic and nonparetic sides). The procedure was done according to Buschbacher and Prahlow. Side to side comparison and designated cutoff points were done
- Needle EMG was done for the deltoid, infraspinatus, and biceps brachii muscles on the paretic side. The procedure was done according to Preston and Shapiro. The standard pathological interpretation of needle EMG and the other obtained nerve conduction parameters were used to describe an underlying lower motor neuropathology for each patient
- Pain X-ray for the shoulder was done to detect shoulder subluxation and/or any bone lesion
- MRI assessment: All patients underwent T1- and T2-weighted multiplanar, multisequence MRI of the affected shoulder. They underwent axial proton density (PD) fat suppression, coronal gradient T2, turbo T2, T1, and PD fat suppression sequence, sagittal PD fat suppression sequence, and three-dimensional coronal gradient T2. All scans were reviewed by one radiologist based on standard radiologic criteria for the following abnormalities: acromioclavicular joint osteoarthritis, rotator cuff pathologies (tendinopathy/tear), bicipital tendinitis, bicipital tendon tenosynovitis, bursitis (subacromial-subdeltoid, subscapularis, and subcoracoid), and labral-ligamentous complex abnormalities. Specific criteria were as follows: full thickness tear of the rotator cuff muscles was defined as a gap in the entire width of the muscle on both coronal and sagittal images. A partial thickness tear was defined as abnormal signal on T2-weighted images involving the articular or bursal side of the tendon on both coronal and sagittal images. Tendonopathy was defined as increased diffuse signal in the tendon, whether on T2 or PD sequences, with respected intact margins of the tendon. The bursal fluid was defined as increased intensity in the bursa on T2-weighted images. Labral ligamentous abnormality was defined as a tear of the anterior or posterior aspect of the labrum. Biceps tendon tear and tendinopathy were defined based on criteria noted above for the rotator cuff.
Determination of potential causes of poststroke shoulder pain
A diagnosis of central pain was made if the features were fitting criteria proposed for the diagnosis of central poststokes pain (CPSP). Noncentral PSSP (due to peripheral cause) included musculoskeletal causes (nociceptive) and/or associated lower motor nerve lesion (peripheral neuropathic cause).
The diagnosis of tendonitis, bursitis, and/or arthritis was made according to clinical evaluation and imaging techniques as well as the known diagnostic criteria. The diagnosis of shoulder impingement was based on both clinical and MRI findings. Poststroke shoulder-hand syndrome was diagnosed if there were pain and limited ROM of the shoulder, wrist, and hand with sparing of the elbow; edema, warmth, and redness of the wrist and hand; and tenderness to palpation of the metacarpophalangeal joints. Different underlying lower motor neuropathologies were categorized as illustrated in [Table 1].
Data were analyzed using the Statistical Package for Social Sciences (SPSS version 20 Chicago, IL, USA). The distributions of quantitative variables were tested for normality using Kolmogorov–Smirnov test, which revealed that the data were not normally distributed. Quantitative data were described using median and range. Qualitative data were described using number and percentage.
| Results|| |
The patients' age ranged from 29 to 81 years (median = 53). There were 21 men (70%) and 9 women (30%). Twenty-nine (96.7%) patients were right handed and only one patient was left handed. Eighteen patients (60%) had right hemiplegia, and 12 (40%) patients had left hemiplegia. The duration of stroke ranged from 1.5 to 60 months (median = 4 months) and 80% of the studied patients (n = 24) had stroke duration of <1 years. History of the previous stroke was present in only 5 (16.7%) patients.
Spasticity in shoulder muscles was present (MAS >1) in ten patients (33.3%). Eleven patients (36.7%) had shoulder muscle power of Grade (0–2), 6 (20%) patients had Grade 3, 9 (30%) patients had Grade 4; while 4 (13.3%) patients had normal muscle power. Passive shoulder ROM and sensorimotor impairment are illustrated in [Table 2].
|Table 2: Passive shoulder Range of Motion and sensorimotor impairment among the studied patients (n=30)|
Click here to view
Pain characteristics are illustrated in [Table 3]. It was observed that the onset of shoulder pain after stroke ranged from immediate PSSP up to 3 years after stroke onset, while the duration of PSSP ranged from 1 month to 33 months. The DN4 questionnaire was positive in only two patients.
Plain radiograph revealed shoulder subluxation in only 6 (20%) patients. Fracture proximal humerus (anatomical neck) was observed in only 1 (3.3%) MRI revealed different pathologies.
Musculoskeletal causes were found to occur in all (100%) patients. One patient had adhesive capsulitis, the remaining other 29 patients had combinations of different musculoskeletal pathologies in the same patient [Figure 1]. In addition, concomitant neuropathic causes were found in 12 (40%) patients [Table 4].
|Figure 1: Different combination patterns of musculoskeletal causes of PSSP among the studied patients. PSSP: Post-stroke shoulder pain; AC: Adhesive capsulitis; RCT: Rotator cuff tear, SIS: Shoulder impingement syndrome|
Click here to view
| Discussion|| |
The present study revealed that both musculoskeletal causes and possibly proximal nerve causes (e.g., brachial plexopathy, axillary nerve lesion, and suprascapular neuropathy) could be accused for the development of shoulder pain in stroke patients. However, the prevalence of musculoskeletal abnormalities was higher (100%) than neuropathic causes.
Although clinical examination can provisionally identify the possible musculoskeletal causes for shoulder pain, MRI imaging added a lot to reveal hidden causes among such patients in whom clinical evaluation could be difficult [Table 4].
This study revealed different musculoskeletal causes responsible for the development of shoulder pain among stroke survivors. These included rotator cuff pathology (tendinopathy/tear) (90%), impingement syndrome (90%), followed by shoulder bursitis (66.7%), adhesive capsulitis (50%), bicipital tendinitis/tenosynovitis (23.3%), shoulder subluxation (20%), poststroke shoulder-hand syndrome (10%), and myofascial pain syndrome (10%). Fracture proximal humerus was present in 1 patient (3.3%). It was also noted that more than one cause can occur in the same patient.
Electrophysiological study revealed abnormalities in 40% of the studied patients; isolated axillary nerve lesion (10%), upper trunk brachial plexus lesion (20%), and suprascapular nerve lesion (10%), which might have contributed to shoulder pain in those patients.
The origin of PSSP is controversy. Some authors suggested that the contribution of musculoskeletal factors to the development of shoulder pain has been disputed and they were in favor that there was a neuropathic origin. Their concept was based on the findings that the prevalence rate of tendinitis, subacromial bursal effusion, and rotator cuff tear is similar in stroke patients with or without shoulder pain and in normal controls. Zeilig et al. observed that PSSP patients exhibited high rates of hyperpathia, allodynia, and dysesthesia, and DN4 questionnaire in their study further corroborated this finding. Therefore, they suggested that stroke survivors with shoulder pain have a neuropathic component (of central or peripheral origin) to the pain.
No patient in this study fulfilled the diagnostic criteria for CPSP as there were definite nociceptive causes that explain PSSP. Furthermore, the current study supports the nociceptive origin of PSSP.
The present study illustrated that dull aching pain is the most common (100%) character or word used by stroke survivors to describe their shoulder pain, whereas numbness, tingling, and burning were used in only 14.3%, 10%, and 6.7%, respectively. This could implicate that nature of shoulder pain among stroke survivors is nociceptive rather than neuropathic. Moreover, movement was the most common pain aggravating factor, as reported by all the studied patients, and rest was the most common relieving factor. These data might suggest that PSSP could be mainly related to biomechanical factors such as muscle imbalance and ligament laxity and muscle length changes which occur after hemiplegia. This also consolidates that the primary initiating event is nociceptive in origin rather than neuropathic. In support of this, is the finding that DN4 was positive in only two patients.
In the current study, DN4 questionnaire (specifically designed to diagnose neuropathic pain, especially in a condition producing different types of pain, such as stroke) was positive in only two patients. One of them had shoulder subluxation with electrophysiological evidence of lower motor nerve lesion (upper trunk brachial plexopathy and partial axillary nerve injury) and shoulder-hand syndrome. The second case had electrophysiological evidence of suprascapular nerve entrapment. It should be noted that these two patients with positive DN4 related their pain to shoulder movement which is considered a sign suggestive of nociceptive origin. Therefore, this study suggests that there was no definite primary role for neuropathic pain in the development of PSSP.
All of the examined hemiplegic shoulders (100%) demonstrated at least one structural abnormality on MRI. It was difficult to identify a single pathology of shoulder pain in patients with hemiplegia; there were several combinations of these pathologies among patients. For example, some patients with adhesive capsulitis on clinical basis also had rotator cuff tears, and some patients with shoulder subluxation also had shoulder-hand syndrome. This also has been reported by Lo et al. who studied 32 patients with PSSP (clinically and arthrographically) and identified four definite musculoskeletal causes of hemiplegic shoulder pain, which were frozen shoulder, shoulder-hand syndrome, shoulder subluxation, and rotator cuff tear with various combinations of these causes in each patient.
The current study revealed that rotator cuff pathology (tendinopathy/tear) was the most common structural abnormality (90%) in stroke survivors with shoulder pain. Rotator cuff tears (partial and total) occurred in nearly 76.6% of the studied patients affecting mainly supraspinatus tendon. The prevalence was much higher than that previously reported prevalence of rotator cuff tear 50% and 40%, respectively. This high prevalence could be explained by the fact that MRI was used in this study, which has been reported to be more sensitive than ultrasound for rotator cuff tear. Ultrasonography (US) and MRI have been reported to be similarly accurate in detection of complete rotator cuff tears; however, for partial tear, the US appears less accurate than MRI.
Shah et al., whose study constitutes the first MRI-based description of the painful shoulder among chronic stroke survivors, reported a prevalence of rotator cuff tear of 35%. This could be explained by difference in their study design, which included only stroke survivors of at least 3 months' duration after stroke. There is evidence that subjects with more recent strokes were more likely to exhibit rotator cuff tears, and the current study included patients with ≥1 month.
Shoulder bursitis was found to occur in 66.7% of the studied stroke survivors in the present study. This has been previously reported by Lee et al. who observed that subacromial–subdeltoid bursal effusion was the most common abnormality seen on sonographic abnormality occurring (50%) of their patients.
The prevalence of shoulder subluxation in the current study was 20%. This prevalence was similar to that reported by Pompa et al. who studied 25 stroke patients with shoulder pain and 16 stroke patients without shoulder pain and revealed a prevalence of shoulder subluxation of 20% those with PSSP.
It should be noted that earlier studies reported shoulder subluxation in stroke survivors with higher frequencies. Najenson et al. and Van Langenberghe and Hogan have reported it to occur in 81% and 45%, respectively. This decline in this study, in the prevalence of shoulder subluxation could indicate the increased awareness of medical care providers and caregivers regarding the possibility of shoulder injuries among those disabled population.
Some researchers found an increased incidence of shoulder-hand syndrome of the upper limb associated with glenohumeral subluxation,, this was in agreement with the current study. The three patients who fulfilled the criteria of shoulder-hand syndrome had shoulder subluxation.
Among the 12 patients with plexopathy and/or peripheral nerve lesion (axillary nerve and/or suprascapular nerve); only five patients had shoulder subluxation. Furthermore, the other seven patients had no history for shoulder traction before the development of PSSP. These suggest that shoulder subluxation was not the only cause for proximal nerve involvement. Patients with electrophysiological features of the upper trunk brachial plexopathies, in whom evident subluxation was not present, may suggest the presence of subclinical proximal neuropathy (especially among those who were diagnosed mainly on latency difference). This mechanism was previously suggested.
In this study, it is unlikely that bias had played a role; as the cause of pain was confirmed by objective investigations.
Results of the current study may be generalized to all stroke patients who do not have chronic inflammatory arthritis, cognitive impairment, and previous shoulder pain or trauma. Each evaluation step (NCS, functional evaluation, clinical assessment, or radiological examination) was done by one and the same doctor.
| Conclusion|| |
The most common causes of PSSP were musculoskeletal (nociceptive) in origin. MRI studies were helpful in revealing common causes of PSSP such as rotator cuff tear and bursitis, especially in patients in whom the clinical assessment was difficult.
The relatively small size of the studied patients (case series) is a limitation of the current study. Moreover, MRI examination was not done for the contralateral shoulder (for side to side comparison) as most of the patients refused performing MRI examination once again, especially for the nonpainful joint. MRI was done in this research because it is superior to US examination of the shoulder.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Widar M, Ahlström G, Ek AC. Health-related quality of life in persons with long-term pain after a stroke. J Clin Nurs 2004;13:497-505.
Lindgren I, Jönsson AC, Norrving B, Lindgren A. Shoulder pain after stroke: A prospective population-based study. Stroke 2007;38:343-8.
Rajaratnam BS, Venketasubramanian N, Kumar PV, Goh JC, Chan YH. Predictability of simple clinical tests to identify shoulder pain after stroke. Arch Phys Med Rehabil 2007;88:1016-21.
Lindgren I, Gard G, Brogårdh C. Shoulder pain after stroke – Experiences, consequences in daily life and effects of interventions: A qualitative study. Disabil Rehabil 2018;40:1176-82.
World Health Organization. World Health Report-Mental Health: New Understanding, New Hope Statistical Annex. Geneva: World Health Organization; 2001. p. 151-5.
McRae R, editor. The shoulder. In: Clinical Orthopedic Examination. 6th
ed. Philadelphia: Elsevier Churchill Livingstone; 2010. p. 49-74.
Johnson C. Measuring pain. Visual analog scale versus numeric pain scale: What is the difference? J Chiropr Med 2005;4:43-4.
Van Seventer R, Vos C, Meerding W, Mear I, Le Gal M, Bouhassira D, et al.
Linguistic validation of the DN4 for use in international studies. Eur J Pain 2010;14:58-63.
Medical Research Council. AIDS to the Investigation of Peripheral Nerve Injuries. 2nd
ed. London: Her Majesty's Stationery Office; 1943.
Bohannon RW, Smith MB. Interrater reliability of a modified Ashworth scale of muscle spasticity. Phys Ther 1987;67:206-7.
Silliman JF, Hawkins RJ. Classifcation and physical diagnosis of instability of the shoulder. Clin Orthop Relat Res 1993;291:7-19.
Jia X, Ji JH, Petersen SA, Freehill MT, McFarland EG. An analysis of shoulder laxity in patients undergoing shoulder surgery. J Bone Joint Surg Am 2009;91:2144-50.
Fugl-Meyer AR, Jaasko L, Leyman I, Olsson S, Steglind S. The poststroke hemiplegic patient. A method for evaluation of physical performance. Scand J Rehabil Med 1975;7:13-31.
Buschbacher R, Prahlow ND. Manual of Nerve Conduction Studies. 2nd
ed. New York: Demos Medical Publishing, LLC; 2006.
Preston DC, Shapiro BE. Electromyography and Neuromuscular Disorders Clinical Electrophysiological Correlations. 2nd
ed. Philadelphia: Elsevier Butterworth Heinemann; 2007.
Resnick D, editor. Internal derangements of joints. In: Bone and Joint Imaging. Vol. 2. Philadelphia: W. B. Saunders Company; 1996. p. 819-83.
Klit H, Finnerup NB, Jensen TS. Central post-stroke pain: Clinical characteristics, pathophysiology, and management. Lancet Neurol 2009;8:857-68.
Martin SD, Upadhyaya S, Thornhill TS. Shoulder pain. In: Firestein GS, Budd RC, Gabriel SE, McInnes IB, O'Dell JR, editors. Kelley and Firestein's Text Book of Rheumatology. 10th
ed. Philadelphia: Elsevier; 2017. p. 754-85.
Chen L, Wu Q. Clinical observation on treatment of 83 cases of posthemiplegic omalgia. J Tradit Chin Med 1998;18:215-7.
Meredith J, Taft G, Kaplan P. Diagnosis and treatment of the hemiplegic patient with brachial plexus injury. Am J Occup Ther 1981;35:656-60.
Chokroverty S, Medina J. Electrophysiological study of hemiplegia. Motor nerve conduction velocity, brachial plexus latency, and electromyography. Arch Neurol 1978;35:360-3.
Lee KH, Khunadorn F. Painful shoulder in hemiplegic patients: A study of the suprascapular nerve. Arch Phys Med Rehabil 1986;67:818-20.
Zeilig G, Rivel M, Weingarden H, Gaidoukov E, Defrin R. Hemiplegic shoulder pain: Evidence of a neuropathic origin. Pain 2013;154:263-71.
Távora DG, Gama RL, Bomfim RC, Nakayama M, Silva CE. MRI findings in the painful hemiplegic shoulder. Clin Radiol 2010;65:789-94.
Moosmayer S, Smith HJ, Tariq R, Larmo A. Prevalence and characteristics of asymptomatic tears of the rotator cuff: An ultrasonographic and clinical study. J Bone Joint Surg Br 2009;91:196-200.
Lo 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;84:1786-91.
Reilly P, Macleod I, Macfarlane R, Windley J, Emery RJ. Dead men and radiologists don't lie: A review of cadaveric and radiological studies of rotator cuff tear prevalence. Ann R Coll Surg Engl 2006;88:116-21.
Yoon TS, Kim DH, Park JW, Kwon BS, Ryu KH, Lee HJ, et al
. Causes of the Hemiplegic Shoulder Pain. J Korean Acad Rehabil Med 2010;34:158-62.
Zaiton F, Tantawy HI, Elsharkawy KA, Sanad G. Painful poststroke shoulder: Comparison of magnetic resonance imaging and high frequency ultrasonography. Egypt J Radiol Nucl Med 2011;42:47-55.
Shah RR, Haghpanah S, Elovic EP, Flanagan SR, Behnegar A, Nguyen V, et al.
MRI findings in the painful poststroke shoulder. Stroke 2008;39:1808-13.
Lee CL, Chen TW, Weng MC, Wang YL, Cheng HS, Huang MH. Ultrasonographic findings in hemiplegic shoulders of stroke patients. Kaohsiung J Med Sci 2002;18:70-6.
Pompa A, Clemenzi A, Troisi E, Di Mario M, Tonini A, Pace L, et al.
Enhanced-MRI and ultrasound evaluation of painful shoulder in patients after stroke: A pilot study. Eur Neurol 2011;66:175-81.
Najenson T, Yacubovich E, Pikielni SS. Rotator cuff injury in shoulder joints of hemiplegic patients. Scand J Rehabil Med 1971;3:131-7.
Van Langenberghe HV, Hogan BM. Degree of pain and grade of subluxation in the painful hemiplegic shoulder. Scand J Rehabil Med 1988;20:161-6.
Chang JJ, Tsau JC, Lin YT. Predictors of shoulder subluxation in stroke patients. Kaohsiung J Med Sci 1995;11:250-6.
Roberts WJ. A hypothesis on the physiological basis for causalgia and related pains. Pain 1986;24:297-311.
[Table 1], [Table 2], [Table 3], [Table 4]