Musculoskeletal ultrasound of the shoulder in systemic sclerosis

Abstract

Objectives:

To explore shoulder findings by ultrasonography and to find factors associated with shoulder ultrasonographic abnormalities in systemic sclerosis patients.

Methods:

A series of systemic sclerosis patients who attended the scleroderma clinic, Srinagarind Hospital, Faculty of Medicine, Khon Kaen University, Thailand, were prospectively evaluated for baseline characteristics, physical examination, and ultrasonography of both shoulders.

Results:

Seventy-four systemic sclerosis patients were enrolled in this study. Diffuse cutaneous type of systemic sclerosis was the most common type (62.2%). The three common systemic sclerosis–associated symptoms were skin tightness (28.5%), salt–pepper appearance (20.9%), and telangiectasia (11.6%). The prevalence of shoulder pain in systemic sclerosis patients was 43.2% (32/74). Sixty-eight patients (92%) had abnormal ultrasonographic findings. The most common ultrasonographic abnormalities were unilateral calcification inside the glenohumeral joint (45.9%), bilateral calcification inside the glenohumeral joint (36.5%), and bilateral supraspinatus tendinosis (28.9%). Skin edematous was the only factor associated with abnormal shoulder ultrasonographic findings. No association between shoulder pain and abnormal shoulder ultrasonographic findings was detected.

Conclusion:

Ultrasonographic abnormalities in the shoulder were common in the systemic sclerosis patients. The most frequent ultrasonographic finding of shoulder joints in systemic sclerosis patients was calcification inside the glenohumeral joint. Moreover, asymptomatic shoulder ultrasonographic abnormalities were prevalent in individuals with systemic sclerosis.

Introduction

Systemic sclerosis (SSc) is one of the rare connective tissue rheumatologic diseases. The skin tightness, internal organ involved fibrosis, and vasculopathy are the specific characteristic of the disease. ¹ Although SSc is an uncommon disease, it has high morbidity and mortality rate. The prevalence varies with ethnicity, gender, and geographic area. ² In Northeastern Thailand, the incidence of SSc is 1 case per 100,000 person-years and women are at higher risk than men (ratio 2:1). ¹ The pathology of SSc demonstrated that a large number of fibrotic tissue infiltrates into the skin layer and various internal organs. These characteristics induce skin and internal organs become thicker and harder. ³ There are two primary forms of SSc: limited cutaneous and diffuse cutaneous. Clinical manifestations of SSc are skin thickening, Raynaud’s phenomenon, renal crisis, pulmonary, and vascular involvements. ⁴

Musculoskeletal (MSK) disorders such as arthritis, arthralgia, tenosynovitis, myalgia, and muscle weakness were found in 40%–80% of patients with SSc.^5,6 Most previous studies reported MSK disorders in SSc patients are at wrist and ankle regions.^4–6 However, shoulder joint is commonly utilized in the activities of daily living which is at risk of prolonged repetitive trauma in the shoulder region that cause shoulder problem. Shoulder pain would have a large impact on the patients’ daily activities, rehabilitation programs, and quality of life. The etiologies of shoulder pain in SSc patients are shoulder muscle weakness, tendon friction rubs, calcification, and shoulder stiffness.^6–10 Shoulder muscle weakness which causes muscle imbalance around the shoulder region is commonly found in SSc (90%). ⁷ The investigations of muscle weakness are manual muscle testing, serum creatine kinase level, muscle biopsy, MRI, and electrodiagnostic study.^5–7 Tendon friction rubs at the shoulder, which is a physical examination finding, is highly specific for SSc were noted about 9%.^8,9,11 It is linked to diffuse skin involvement and disability. ⁹ Therefore, tendon friction rubs are significantly associated with more severe skin thickening and joint contractures which are the risk factors of shoulder pain in SSc patients. Patient with CREST syndrome showed widespread calcinosis in the periarticular soft tissues and intra-articular infiltration on their shoulders which can be seen using shoulder radiograph and CT scan. ¹⁰ Calcinosis is often correlated with pain, repetitive local inflammation, and functional impairment. Inflammation and skin involvement in SSc patients are important factors for the development of shoulder contractures which is an unfavorable prognostic factor in SSc. ¹¹ Moreover, shoulder stiffness lead to shoulder pain and limitation of overhead activities which directly affect the quality of life.

The methods for diagnosis of shoulder pain are composed of history, physical examination, and various special tests. The sensitivity of shoulder special tests is 26%–72%. ¹² These tests are not as sensitive as ultrasonography. It was reported that ultrasonographic abnormalities can be found in normal result of special tests. ¹³ The radiographic imagings for shoulder investigation are plain film, CT scan, MRI, and musculoskeletal ultrasound (MSK US). Plain film is mainly to evaluate bony structures in shoulder but cannot be detected soft tissue lesions. ¹⁴ Musculoskeletal ultrasonography is a noninvasive imaging tool that can be mostly investigated for soft tissue lesion such as tendinopathy, synovitis, and bursitis.^15–18

To the best of our knowledge, there are no published studies which investigated MSK US of shoulders combined with physical examinations in SSc patients. The objective of the study is to explore shoulder ultrasonographic findings and to find factors associated with shoulder ultrasonographic abnormalities.

Material and methods

Research design

A cross-sectional descriptive study

Participants

This study examined SSc patients who visited outpatient scleroderma clinic in Srinagarind hospital, which is the largest super tertiary care hospital in Northeast region of Thailand. Inclusion criteria were age over 18 years, a diagnosis of SSc, and provision of written informed consent. Exclusion criteria were history of traumatic injury or shoulder surgery on either side of the shoulder, inability to cooperate during the examination, and inability to sit in a chair for more than 30 min.

This study was approved by the Khon Kaen University Ethics Committee for Human Research (HE621276).

Methods

All participants obtained clinical assessment and ultrasonographic examination of both shoulders. Demographic data included age, sex, weight, height, handedness, and underlying disease were assessed. Duration and type of SSc, scleroderma-associated symptoms, medications, and latest erythrocyte sedimentation rate (ESR) were evaluated.

In case of participants with shoulder pain, the involved pain site, duration of pain, and the severity of pain during activities by a numerical rating scale were evaluated. Information was performed through interviews by a physician who was not a musculoskeletal ultrasonographic viewer.

Shoulder examinations were to find out the area of maximal tenderness. Measurement of the shoulder passive range of motion in all directions by a goniometer was performed. The special tests of both shoulders which consisted of Yergason’s test, Resisted internal rotation, Resisted external rotation, Neer impingement sign, Empty can test, O’ Brien Active Compression Test for acromioclavicular joint were appraised.^12,19,20

Each participant received ultrasonographic evaluation of both shoulders by a physician who had more than 5 years of experience in musculoskeletal ultrasonography. The musculoskeletal ultrasonography viewer was blind to the patient’s clinical symptoms and examination results. Ultrasonography was conducted in a standard protocol by GE LOGIQ e Ultrasound System (GE Healthcare, USA) using a 9–14 Hz linear transducer to collect images. During the ultrasonographic examination, participants sat straight in a chair. The ultrasonographic procedure was evaluated according to previous standard protocol.^21,22 The examined shoulder structures were the biceps tendon, the subscapularis tendon, the acromioclavicular joint (ACJ), the supraspinatus tendon, the subdeltoid–subacromial bursa, the infraspinatus tendon, the glenohumeral joint (GHJ), the synovitis at posterior recess of GHJ, and the dynamic supraspinatus impingement test. The definitions of ultrasonographic pathology were established based on international guidelines of The Outcomes Measures in Rheumatology (OMERACT 7) and The European Society of Musculoskeletal Radiology (ESSR).^23,24

Statistical analysis

All statistical analysis was examined by SPSS 19.0 (IBM SPSS statistics Version 19). Continuous data were shown in mean and standard deviation (SD) if the data were normally distributed and median with interquartile range (IQR) if the data were not normally distributed. Ordinal and categorical data were presented in frequency and percentage. Association between factors and shoulder ultrasonography abnormalities were analyzed by Fisher’s exact test. Factors, which had p value less than 0.2, would be further analyzed by logistic regression analysis (significant p value < 0.05).

Results

Of the 74 participants recruited in this study, all of them had no history of shoulder inflammatory arthritis, 51 (68.9%) were female. The mean age was 58 ± 12 years. Most patients had normal body mass index. Median duration of SSc was 60 months. The most common SSc type was diffuse cutaneous (62.2%). The three common SSc-associated symptoms were skin tightness (28.5%), salt–pepper appearance (20.9%), and telangiectasia (11.6%). The average latest ESR level was 43.5 mm/h (Table 1).

Table 1.

Baseline characteristics.

Data	Number	Percentage
Age (years), mean ± SD	57.9 ± 12.2
Gender
Male	23	31.1
Female	51	68.9
BMI (kg/m2)
Mean ± SD	21.1 ± 4.1
Underweight	21	28.4
Normal	32	43.2
Overweight	5	6.8
Obesity	16	21.6
Handedness
Right	64	86.5
Left	10	13.5
Underlying disease
Diabetes mellitus	9	24.3
Hypertension	15	40.5
Dyslipidemia	10	27.0
Thyroid dysfunction	3	8.1
Type of systemic sclerosis
Limited cutaneous SSc	28	37.8
Diffused cutaneous SSc	46	62.2
Duration of scleroderma (months), median (IQR)	60 (36–120)
Scleroderma symptoms
Skin edematous	4	2.3
Skin tightness	49	28.5
Pitting scar	10	5.8
Raynaud’s phenomenon	19	11.1
Calcinosis cutis	7	4.1
Telangiectasia	20	11.6
Salt–pepper appearance	36	20.9
Dysphagia	16	9.3
Tendon friction rub	11	6.4
History of digital ulcers
Yes	14	18.9
No	60	81.1
Skin score median (IQR)	4 (0–8.25)
Medications
Calcium channel blockers	19	25.7
Prednisolone	2	2.7
Immunosuppression	53	71.6
Erythrocyte sedimentation rate (mm/h) mean ± SD	43.5 ± 24.6

BMI: body mass index; SSc: systemic sclerosis; IQR: interquartile range.

Regarding to shoulder pain characteristics, less than half of patients (43.2%) had shoulder complaints. Median duration of shoulder pain was 48 weeks. The most common area of maximal tenderness was right anterior region (59.2%). The percentage of patients with shoulder stiffness was 28.4% and median duration of stiffness was 48 weeks. A majority of the abnormal shoulder special tests on the right side were pain on resisted internal rotation test (30.7%), followed by pain on resisted external rotation test (25.6%) and the empty can (15.4%) (Table 2).

Table 2.

Shoulder pain characteristics.

Data	Number	Percentage
Shoulder pain, number (%)	32	43.2
Shoulder pain sites, number (%)
Right	10	31.2
Left	8	25.0
Bilateral	14	43.8
Duration of shoulder pain (weeks), median (IQR)	48 (10–96)
Severity of shoulder pain during activity daily living (NRS), mean ± SD	5.5 ± 1.7
Severity of shoulder pain during rest (NRS), mean ± SD	2.5 ± 2.0
Area of maximal tenderness, number (%)
Right
Anterior	16	59.2
Posterior	4	14.8
Medial	2	7.4
Lateral	5	18.6
Left
Anterior	14	58.3
Posterior	5	20.8
Medial	2	8.3
Lateral	3	12.5
Shoulder stiffness, number (%)	21	28.4
Duration of shoulder stiffness (weeks), median (IQR)	48 (30–96)
Passive range of motion (degree)
Right
Flexion, mean ± SD	150.6 ± 32.6
Extension, median (IQR)	60 (45–67.5)
Abduction, median (IQR)	60 (140–180)
Internal rotation, mean ± SD	54.4 ± 27.3
External rotation, median (IQR)	90 (80–90)
Left
Flexion, mean ± SD	140 ± 35.4
Extension, mean ± SD	53.1 ± 16.4
Abduction, mean ± SD	150 ± 31.6
Internal rotation, median (IQR)	60 (20–87.5)
External rotation, mean ± SD	77.5 ± 16.0
Positive shoulder special tests, number (%)
Right
Yergason’s test	3	7.7
Resisted internal rotation	12	30.7
Resisted external rotation	10	25.6
Neer impingement sign	4	10.3
Empty can test	6	15.4
O’ Brien active compression test for ACJ	4	10.3
Left
Yergason’s test	2	6.7
Resisted internal rotation	6	20.0
Resisted external rotation	4	13.3
Neer impingement sign	5	16.7
Empty can test	9	30.0
O’ Brien active compression test for ACJ	4	13.3

IQR: interquartile range; ACJ: acromioclavicular joint.

According to the ultrasonographic shoulder findings, the top three abnormalities were unilateral calcification inside the glenohumeral joint (45.9%; Figure 1), bilateral calcification inside the glenohumeral joint (36.5%) and bilateral supraspinatus tendinosis (28.9%), respectively. All of SSc patients had no effusion in glenohumeral joint. However, the synovitis at posterior recess of glenohumeral joint was found in 20 SSc patients (unilateral synovitis 12.4%, bilateral synovitis 10.1%) (Table 3). Of the 32 SSc patients with shoulder pain, 96.9% of them had at least one ultrasonographic abnormality. In addition, 88.1% of SSc patients with asymptomatic shoulder pain had abnormal ultrasonographic finding.

Figure 1.

Calcification in glenohumeral joint.

Table 3.

Ultrasound findings.

Data	Number	Percentage
The biceps tendon, number (%)
Normal bilateral echogenicity	47	62.7
Unilateral effusion	20	26.7
Bilateral effusion	6	8.0
Unilateral calcification	2	2.6
The subscapularis tendon, number (%)
Normal bilateral echogenicity	60	80.0
Unilateral tendinosis	2	2.7
Unilateral calcification	10	13.3
Unilateral full rupture	1	1.3
Bilateral calcification	2	2.7
The acromioclavicular joint, number (%)
Normal bilateral echogenicity	71	95.9
Unilateral calcification	3	4.1
ACJ length (mm), mean ± SD	0.61 ± 0.14
The supraspinatus tendon, number (%)
Normal bilateral echogenicity	32	38.6
Unilateral tendinosis	9	10.8
Bilateral tendinosis	24	28.9
Unilateral partial thickness rupture	2	2.4
Unilateral full thickness rupture	1	1.2
Unilateral calcification	14	16.9
Bilateral calcification	1	1.2
Tendon thickness (mm), median (IQR)	0.51 (0.43–0.60)
The subdeltoid–subacromial bursa, number (%)
Normal bilateral echogenicity	61	82.4
Unilateral bursitis	11	14.9
Bilateral bursitis	2	2.7
The subdeltoid–subacromial bursa thickness (mm) median (IQR)	0.095 (0.08–0.12)
The infraspinatus tendon, number (%)
Normal bilateral echogenicity	61	82.4
Unilateral tendinosis	1	1.4
Bilateral tendinosis	1	1.4
Unilateral calcification	9	12.1
Bilateral calcification	2	2.7
The glenohumeral joint, number (%)
Normal bilateral echogenicity	13	17.6
Unilateral calcification	34	45.9
Bilateral calcification	27	36.5
The synovitis of glenohumeral joint, number (%)
No synovitis	54	60.7
Unilateral synovitis	11	12.4
Bilateral synovitis	9	10.1
The dynamic supraspinatus impingement, number (%)
Bilateral negative	57	78.1
Unilateral positive	14	19.2
Bilateral positive	2	2.7

IQR: interquartile range; ACJ: acromioclavicular joint.

No association between shoulder pain and abnormal ultrasonographic findings was detected (Table 4). Skin edematous which was based on clinical assessment by physical examination was the only factor associated with abnormal ultrasonographic findings (Table 5).

Table 4.

Factors associated with abnormal ultrasonographic findings.

Factors	Normal ultrasonographic findings (N)	Abnormal ultrasonographic findings (N)	p Value
Age (years)
<60	3	39
>60	3	29	1.00
Gender
Male	4	19
Female	2	49	0.07*
BMI (kg/m2)
<23	4	49
>23	2	19	1.00
Handedness
Right	6	58
Left	0	10	0.59
Type of systemic sclerosis
Limited cutaneous SSc	2	26
Diffused cutaneous SSc	4	42	1.00
Duration of scleroderma (months)
<60	3	29
>60	4	38	0.98
Skin score
<4	4	30
>4	2	38	0.40
ESR
<25	3	13
>25	9	49	0.75
Scleroderma symptoms
Skin edematous	2	2	0.03*
Skin tightness	4	45	1.00
Pitting scar	0	10	0.59
Raynaud’s phenomenon	2	17	0.64
Calcinosis cutis	1	6	0.46
Telangiectasia	2	18	0.66
Salt–pepper appearance	3	33	1.00
Dysphagia	3	13	0.31
Tendon friction rub	1	10	1.00
History of digital ulcers
Yes	1	13
No	6	54	0.60
Shoulder pain
Yes	1	31
No	5	37	0.23
Shoulder stiffness
Yes	1	19
No	5	49	1.00

BMI: body mass index; ESR: erythrocyte sedimentation rate.

^*p Value ⩽ 0.2.

Table 5.

Multivariate analysis of factors associated with abnormal ultrasonographic findings.

Factors	Odds ratio	95% CI	p Value
Gender	3.6	0.5–23.9	0.187
Skin edematous	10.6	1.04–107.18	0.046*

CI: confidence interval. 95% CI = 95% confidence interval; p value from logistic regression analysis.

^*p < 0.05.

Discussion

In our study, we found a higher prevalence of shoulder pain in SSc patients compared with the previous studies (23%–30.4%).^18,25 This may be due to the difference in SSc subtypes because a majority of our participants were diffuse cutaneous whereas other studies mainly were limited cutaneous form. Diffuse cutaneous type has a more extensive cutaneous pathology to proximal limbs and torso while limited cutaneous type has cutaneous problems mainly below elbows and knees. Moreover, the mean age of this study was slightly higher than previous studies (54–56 years).^18,25 Older age also is associated with a progressive degenerative changes such as arthropathies. ²⁶

Compared with patients with other rheumatologic disease as rheumatoid arthritis, the prevalence of shoulder pain was 58% which was higher than our study. ²⁷ This may be explained by different pathology of disease. Regarding to rheumatoid arthritis, an autoimmune reaction directly targets to synovial lining of joints which causes significant pain and inflammation around joint line. Moreover, this study showed that the prevalence of shoulder pain is higher than general population (6.9–26%). ²⁸ It can be assumed that SSc patient has skin edematous and leakiness of blood vessels in shoulder joints which can restrict motion of joint for a long time which lead to shoulder pain. ²⁶ Furthermore, inflammation of the joints is another cause of shoulder pain and often accompanied by joint stiffness. As mentioned above, the common SSc-associated symptoms were skin tightness. Tightening and thickening of the skin around the shoulder joint can directly raise intra-articular pressure and cause a crush of the articulating structures which provokes shoulder pain. ²⁵

Ultrasonographic abnormalities in the shoulder were common in the SSc patients. The most frequent ultrasonographic finding was calcification inside the glenohumeral joint. The result of our study was consistent with the previous study which found calcifications in the peritendinous areas and soft tissue and thickening of the A1 pulley which may induce hand contractures by causing mechanical impingement of the finger flexion deformity in SSc patients. ²⁹ Among the rotator cuff tendons, supraspinatus was the most common involved, followed by subscapularis and infraspinatus tendon. The result of this study differed from previous published study which found calcinosis overlying acromion, involving subacromial-subdeltoid bursa and inferior glenohumeral ligament. ³⁰ Calcinosis is an essentially feature of SSc and is often associated with pain, recurrent episodes of local inflammation and functional impairment.^30,31 Participants usually prefer to use their dominant limb for repetitive daily routine work. It was in accordance with the hypothesis that the influence of handedness on the movement and repetitive muscle activities of the shoulder complex was evident in chronic shoulder pain due to recurrent microtrauma.^31,32 We suspected that calcification at shoulder region is associated with calcinosis cutis, but in our study, only a few incidents were reported and no significant association between calcinosis cutis and abnormal ultrasonographic shoulder findings. Moreover, our study found small amounts of synovitis at posterior recess of glenohumeral joint. It was hypothesized that, pathology of synovitis in the SSc disease, most affected joints were metacarpophalangeal, proximal interphalangeal, wrists, and ankle joints.^33,34 However, early detection of synovitis in glenohumeral joint which might be a predictive factor of the diffuse cutaneous subset and carrying a poor prognosis would potential enable earlier plan of treatment and close monitoring. ³³

We found that most of the SSc patients with shoulder pain had ultrasonographic abnormality in at least one of the structures evaluated. Therefore, the accurate definite diagnosis of shoulder pain can be detected by MSK US. The proper treatment can be directly prescribed. Although one of the main clinical symptoms of shoulder disorder is the presence of pain, asymptomatic shoulder pathologies have been detected on ultrasonography. According to this study, ultrasonographic abnormalities were also found in nonpainful shoulders. The absence of any clinical of shoulder pain cannot exclude the presence of local shoulder pathologies. Ultrasonography is a useful investigation for detecting the presymptomatic stages that may undergo shoulder symptomatic pathologies. ³⁵ The knowledge of possible asymptomatic pathologies is important for understanding shoulder pain and dysfunction that may occur in the future. ³⁶

This study revealed the most of positive shoulder special tests were resisted internal rotation which indicated for subscapularis tendon lesions whereas the common ultrasonographic abnormalities in SSc were calcification inside the glenohumeral joint. We found that the assessment of shoulder special tests was not associated with the ultrasonographic results. Therefore, shoulder special physical examination used for the diagnosis of shoulder problems may had low diagnostic performance for detecting abnormality in the SSc shoulder joint. It was consistent with the study of Naredo et al. ¹³ which found the occurrence of ultrasonographic abnormalities in patients with normal results of special tests. However, further studies should be explored to analyses reliability of shoulder special tests in SSc patients. Regarding to clinical practice, ultrasonographic findings should be interpreted together with clinical assessments to determine definite the pathology of shoulder pain.

There was association between skin edematous and abnormal ultrasonographic shoulder findings. Calcinosis is frequently associated with soft-tissue thickening, representing edema with inflammatory cells in the acute stage or fibrotic changes in the chronic sclerotic stage. Moreover, edema with calcinosis is often associated with pain and inflammation which usually present with high serum ESR level.^1,29 In our study, latest ESR level which is a usual indicator of inflammation was high in abnormal ultrasonographic findings group but no statistically significant. In accordance with our result, intra-articular calcification which is a result from inflammation process may associate with skin edematous.

There are several limitations of our study. First, a single center study limits its generalizability. The pattern of sampling in our study had been significantly concentrated potential for geographical selection bias because most Thai SSc patients were diffuse cutaneous type. ³⁷ Second, our study investigated synovitis at glenohumeral joint only in posterior recess. Further studies should assess synovitis at glenohumeral joint in different recesses included anterior and axilla parts for informative data. Third, the study was evaluated at one time point and no clinical follow-up. We could not determine whether the shoulders will remain either normal or abnormal. However, it is hoped that our study results encourage physicians to concern the ultrasonographic findings in SSc patients with shoulder pain. The results of our study may guide the shoulder pain investigation concepts for the physicians who deal with SSc patients. Future prospective longitudinal studies about the shoulder ultrasonographics findings in SSc patients should be considered.

Conclusion

Ultrasonographic abnormalities in the shoulder were common in the SSc patients. The most frequent ultrasonographic finding of shoulder joints in SSc patients was calcification inside the glenohumeral joint. Moreover, asymptomatic shoulder ultrasonographic abnormalities were prevalent in individuals with SSc.