INTRODUCTION
Approximately 90% of all rheumatoid arthritis (RA) patients have painful shoulder symptoms (1). The painful shoulder was found to be affected by inflammatory changes in the first two years of the disease in about 50% of patients, and in 83% of patients in 14 years of the disease (2).
Chronic inflammation in RA affects the articular synovial membrane and spreads to other articulated structures, such as bursae, tendons, and tendon sheaths. When examining such patients, conventional ultrasound with high-resolution probes is a reliable method for assessing all pathological changes of the above-mentioned joint structures, including the shoulder joint (3).
Proliferative synovitis therefore affects not only the humeroscapular joint, but also the other joints and bursae of the shoulder complex, especially the subacromial-subdeltoid bursa, as well as the long head of the biceps tendon (4).
Significant bone changes in the joints are expected in RA patients. However, unlike the small joints of the hands and feet, periarticular shoulder changes affect the joint function much earlier than the bone structure (5-7). As a result, classical radiography reveals only late changes in advanced cases of the disease (8).
In contrast, diagnostic ultrasound can detect early changes in the soft tissues of the shoulder, subacromial-subdeltoid bursal effusion (9,10), as well as effusion of the long head of the biceps tendon sheath (11), changes in the humeroscapular joint (12), and erosion of the humeral head (13,14).
Painful shoulder which occurs due to a non-inflammatory etiology is often referred to as periarthritis humeroscapularis (PHS) in everyday practice. This diagnosis should be avoided, because it covers a large number of different clinical entities, which are all treated differently (15).
Therefore, we use the term painful shoulder of non-inflammatory cause, which means that it occurs without an underlying systemic inflammatory condition reported by patients that can be clinically determined. The cause of 60% cases of painful shoulder is subacromial impingement syndrome with consequential subacromial bursitis or supraspinatus muscle tendinitis. According to the research, 12% of these cases show adhesive capsulitis, 10% present with partial or complete rupture of the rotator cuff, 7% manifest with acromioclavicular joint osteoarthritis, 5% of cases are tendinitis of the long head of the biceps muscle, and 7% of cases are the result of some other causes (15).
Some of these causes occur on their own, and some occur in various combinations (16). Consequently, many morphological parameters occur together in both RA patients as well as in patients with non-inflammatory painful shoulder.
The purpose of this research was to investigate and possibly confirm the existence of a significant difference between the morphological parameters of painful shoulder in RA patients as compared to patients with painful shoulder of non-inflammatory etiology by using numerous radiographic and ultrasonographic diagnostic parameters.
MATERIALS AND METHODS
Description of the research
A cross-sectional study was conducted on patients diagnosed with RA presenting with a painful shoulder and patients with a painful shoulder of non-inflammatory etiology.
A conventional radiogram and ultrasound of both shoulders were performed in the RA patients, whereas in the patients with a non-inflammatory etiology the diagnostic procedures were done unilaterally, only on the painful shoulder. Additionally, the obtained parameters were analyzed and the findings of patients with painful shoulder in rheumatoid arthritis were compared with the parameters of patients with painful shoulder of non-inflammatory etiology.
Respondents
An ultrasound examination of 160 shoulders was performed by an experienced ultrasound radiologist who had been working in the musculoskeletal system diagnostics field for more than 20 years. The diagnostic devices used were a Shimadzu SDU 1200 (Kyoto, Japan) with a 10-MHz linear probe, a Toshiba Nemio (Tokyo, Japan) with a linear probe of 11 and 14 Mhz, and Logic 8 (General Electrics) with an 8-Mhz linear probe.
Clinical examinations were carried out in 40 RA patients, examining both shoulders (bilateral painful shoulder in 28 women and 12 men, mean age 59.4 ± 11.9 years, average disease duration 4.3 years). In the same way, 80 patients with painful shoulder of non-inflammatory etiology were examined (unilateral shoulder pain in 54 women and 26 men, mean age 53.2 ± 7.2 years).
Before the examination, all patients stated that the pain had been present for more than 6 weeks, with no record of trauma.
The patients in the first group had been diagnosed with RA, supported by data on elevated CRP values and with no prior record of painful shoulder, while those in the second group had no record of possible RA.
Methods
For all 160 shoulders, the thickness of the supraspinatus tendon was measured in the transverse and longitudinal sections, in the neutral shoulder position as well as in adduction and internal rotation of the shoulder. The mean values in both cross-sections and both positions were measured.
Transverse cross-section: the upper part of the humeral head, above the intertubercular sulcus, was measured in the thickest medial part.
Longitudinal cross-section: measured in the place where the tendon emerges under the shadow of the acromion. The tendon diameter of the long head biceps muscle was measured on the upper edge of the intertubercular sulcus, in both the transverse and longitudinal cross-sections.
The humeroscapular joint effusion (capsule-bone distance), measured transaxillary, approached the part of the humerus not covered by the rotator cuff.
The presence of the subacromial bursal effusion has also been detected in the long head biceps tendon.
Data on the rotator cuff echo structure was analyzed during the examination, particularly the inhomogeneity of the supraspinatus tendon, as well as small deposits of calcium salts and the partial or complete rupture of the rotator cuff. The supraspinatus tendon is defined as homogeneous if the regular fibrillar structure is preserved along with the border towards the subdeltoid bursa. A tendon is defined as inhomogeneous if the fibrillar structure of the tendon is disturbed without a clear border to the subdeltoid bursa.
Conventional radiography of the shoulder was mainly analyzed on the presence of calcifications in the tendons and bursae, cystic formations, bone sulci, and generalized osteopenia. Osteoarthritis of the acromioclavicular and humeroscapular joints, greater tubercle sclerosis, and subacromial osteophytes were noted.
Statistical data analysis
A T-test for independent samples was used to determine statistically significant differences between numerical parameters. The correlation between category variables was determined by the χ2 test.
The calculation of the sample size was carried out using an online program available onhttp://www.stat.ubc.ca/~rollin/stats/ssize/n2.html to calculate the number of respondents. The distance from the articulated sheath to the bone was taken as the main measure of the outcome. Preliminary measurements gave us a value of 3.1 mm in the painful shoulder of inflammatory cause group, and a value of 2.5 mm in the non-inflammatory painful shoulder group. The standard deviation was about 0.5 mm. With a significance level of 0.05 and a statistical power of 0.8, the sample size was calculated to be 11 participants for each group of independent data sets. The Statistica 6 software package (StatSoft Inc, Tulsa, USA) was used.
RESULTS
Paraarticular sulci were found on 24 shoulders in 12 RA patients, whilst in the group of patients with non-inflammatory painful shoulders only one sulcus was found in one patient’s shoulder.
Our study showed that there was no significant difference in the supraspinatus tendon thickness and long head biceps tendon in RA patients compared to the patients with a painful shoulder of non-inflammatory etiology.
The joint capsule-bone distance (an indicator of the intra-articular synovial fluid amount) was significantly higher in RA patients than in the group of patients with a painful shoulder of non-inflammatory etiology (Table 1).
*t-test
Considerably more RA patients had a long head biceps tendon effusion (χ2 = 16.78; P < 0.01) as well as subdeltoid bursal effusion ((χ2 = 33.63; P < 0.01) (Table 2).
There was no significant difference between the two groups of patients considering diffuse osteopenia (χ2 = 2.66; P = 0.10), rotator cuff calcification (χ2 = 1.51; P = 0.22), supraspinatus muscle tendon inhomogeneity (χ2 = 0.72; P = 0.39), partial rupture of the rotator cuff (χ2 = 0.28; P = 0.59), complete rupture of the rotator cuff (χ2 = 0.49; P = 0.48), sclerotic lesion of the greater tubercle of the humerus (χ2 = 0.11; P = 0.75), acromioclavicular joint osteoarthritis (χ2 = 0.44; P = 0.51), and humeroscapular joint osteoarthritis (χ2 = 0.04; P = 0.84) (Table 2).
In a significantly higher number of patients with non-inflammatory shoulder pain subacromial osteophytes were found compared to the RA patients (Table 2).
The RA shoulders were later divided into two groups:
– the first group with subdeltoid bursa and long head biceps tendon sheath effusion, which always appeared in conjunction, and
– the second group without an effusion.
There were notably fewer shoulders with subacromial osteophytes (χ 2= 23.11; P < 0.01) and acromioclavicular joint osteoarthritis (χ2 = 4.37; P < 0.05) in the first group compared to the painful shoulder of the non-inflammatory cause group (Table 3).
There was no significant difference between morphological parameters in the RA patients from the second group compared to the patients with non-inflammatory painful shoulder (Table 4).
DISCUSSION
The results of our study, in which we compared ultrasound and radiographic parameters of patients with painful shoulder in rheumatoid arthritis and patients with painful shoulder of non-inflammatory etiology, showed that there was a significant morphological difference between these two groups.
Moreover, our study has shown that the capsule-bone distance (an indicator of the intra-articular synovial fluid amount) was significantly higher in RA patients in comparison to the group of patients with non-inflammatory etiology of the painful shoulder.
In particular, more RA patients had an effusion of the long head biceps tendon sheath and a subdeltoid bursal effusion.
In a significantly higher number of patients with painful shoulder of non-inflammatory etiology subacromial osteophytes were found, in contrast to RA patients.
In a survey conducted in 2010, researchers Milutinović and Zlatković-Švenda showed by using ultrasound that RA patients are more likely to be associated with subdeltoid bursal and long biceps tendon sheath effusions, a bigger capsule-bone distance, cartilage reduction, and humerus head erosion in comparison to patients with a non-inflammatory etiology of painful shoulder (17). These results are similar to the results of our study, as expected.
It is known that chronic and progressive inflammatory diseases of the joint such as RA affect the synovial membrane and extend to extra-articular components (bursae, tendons, and tendon sheaths), causing damage to the joint cartilage (3,18).
The non-inflammatory painful shoulder symptoms most often originate from a subacromial bursa irritation in subacromial impingement syndrome, as well as from calcium salt deposits in calcifying tendinitis.
According to the available data, in RA patients it can be expected to find a greater diameter of the supraspinatus muscle tendon, as well as the long head biceps tendon, and a bigger capsule-bone distance (which represents a higher amount of intra-articular synovial fluid). However, our study only found a greater capsule-bone distance in RA patients, whilst the supraspinatus muscle tendon thickness and the long head biceps tendon thickness showed no statistically significant differences between the patient groups. The explanation of that result may be the shorter duration of the inflammatory rheumatic disease and the fact that the supraspinatus muscle and long head biceps tendon thickening is already recorded in the first stage of subacromial impingement syndrome, especially in younger patients (19–22). This fact diminishes the value of metric parameters in distinguishing changes in RA patients from those with painful shoulder of non-inflammatory etiology.
The long head biceps tendon and subdeltoid bursal effusions, as a common finding in RA patients, predominated in that group as expected. According to the experience and available literature, effusion can be also found in patients with a non-inflammatory etiology of the painful shoulder, especially in subacromial impingement syndrome, with consequential long biceps tendon bursitis and tendinitis, which may be found in asymptomatic patients as well. In cases of massive rupture of the rotator cuff, a passive effusion from the articulated area into the bursa can be detected (15,21,23,24).
Diffuse osteopenia of the shoulder bones was found in both the RA patients and patients with non-inflammatory shoulder pain, with no statistically significant difference. An inflammatory component and access to earlier local corticosteroid administration, as well as the lack of movement in the painful shoulders, can be an explanation.
Other parameters associated with painful shoulder of non-inflammatory etiology, such as calcifications, partial and complete ruptures of the rotator cuff, inhomogeneity of the supraspinatus tendon, humeroscapular and acromioclavicular joint osteoarthritis, sclerosis of the greater tuberosity of the humerus, and the presence of subacromial osteophytes, surprisingly showed no statistically significant difference between the two groups. Even the number of RA shoulders with soft tissue calcifications was higher in comparison with the other group (25,26).
Inhomogeneity of the supraspinatus tendon with an abnormal fibrillar structure and without a clear border towards the subdeltoid bursa can be also found in subacromial impingement syndrome (16,27).
Although partial and complete rupture of the rotator cuff are strongly associated with subacromial impingement syndrome in elderly patients (28,29), they can be found in advanced RA as well. Painful shoulder of non-inflammatory etiology is strongly associated with acromioclavicular osteoarthritis, large tuberous sclerosis, and the presence of subacromial osteophytes (30).
The RA patient shoulders were divided into two groups in the second phase of the survey. The first group was associated with subdeltoid bursa and long biceps tendon sheath effusions in conjunction, whilst the second group had no such affiliation.
Comparing the shoulders of both the RA and non-inflammatory groups, a statistically significantly higher number of shoulders with acromioclavicular joint osteoarthritis and subacromial osteophytes were observed in the non-inflammatory painful shoulder group.
A comparison of the shoulders of the second group with RA and the non-inflammatory painful shoulders showed no statistically significant difference between the parameters. The multifactorial characteristic of the painful shoulder in RA is additionally emphasized by this fact.
CONCLUSION
RA patients are not protected from the usual harmful effects, especially in the older population, and therefore the etiology of painful shoulder in the stated group can be multifactorial. Due to that fact, painful shoulder treatment should be modified and adapted to each patient individually, based on radiological (X-ray and ultrasound) diagnostics. The significance of the results obtained by this study emphasizes the importance of a constant analysis of the morphological shoulder parameters after a clinical assessment in determining the prognosis and therapy of patients with a painful shoulder.