The purpose of this study was to report the magnetic resonance arthrography (MRA) findings of patients considered to have stage-2 idiopathic adhesive capsulitis and to identify differences in rotator cuff pathologies between 2 age groups (ie, < and >60 years). Eighty-one patients, considered to have stage-2 idiopathic adhesive capsulitis based on medical history and a physical examination underwent MRA. Magnetic resonance evaluations focused on combined pathologies other than adhesive capsulitis, especially on the supraspinatus tendon. The pathologies of supraspinatus tendon were divided into 4 categories: normal, tendinosis, partial-tears, and full-thickness tears. Mean patient age was 66 years and mean symptom duration was 8.9 months. There were 38 men and 43 women. Overall, 50 patients (61.7%) showed some form of supraspinatus tendon pathology [small full-thickness tears 6 (7.4%), partial-thickness tears 25 (30.9%), and supraspinatus tendon tendinosis 19 (23.5%)]. The most common MRA finding, other than rotator cuff pathology, was the impression of adhesive capsulitis in 58 patients (71.6%). Group 2 (>60 years) showed a higher prevalence of full and partial-thickness tears (P<.05), however the overall prevalence of supraspinatus tendon pathologies were similar in those younger than and older than 60 years (P=.1795). Approximately 1/3 of stage-2 adhesive capsulitis patients showed some form of supraspinatus tendon tear by MRA, and <10% of patients who were considered to have idiopathic adhesive capsulitis had a full-thickness tear. Adhesive capsulitis patients older than 60 years appeared to be more likely to have a supraspinatus tendon tear.
Idiopathic adhesive capsulitis in the shoulder is considered a self-limited disease that can usually be treated by physical therapy.1 In Korea and Japan, idiopathic adhesive capsulitis is sometimes called the ‘50s shoulder’ or ‘Ohsibgyun,’ which describes a self-relieving shoulder pain that occurs mainly in the 50s. It is known that most patients are satisfied with the end result in terms of performing activities of daily living,2 although some reports have concluded that residual symptoms and limitations of motion can exist.3-6 Most physicians and surgeons treat this disease conservatively without further evaluation or study. However, idiopathic adhesive capsulitis can be encountered in many other age groups, and in some older patients (especially >60 years), one cannot rule out the possibility of combined disease (especially rotator cuff pathology) requiring surgical intervention.
Furthermore, the etiology of adhesive capsulitis remains elusive. It may be confusing to classify adhesive capsulitis as a primary idiopathic or secondary pathology. Full-thickness rotator cuff tears combined with a history of a stiff shoulder could be classified as secondary adhesive capsulitis. However, some patients who appear to have stage-2-idiopathic adhesive capsulitis based on signs and symptoms sometimes show rotator cuff pathologies, especially a small full-thickness rotator cuff tear or partial-thickness rotator cuff tear, on magnetic resonance arthrography (MRA) or magnetic resonance imaging (MRI). Furthermore, without MRI or ultrasonography, it is difficult to diagnose these patients with stage-2-idiopathic adhesive capsulitis (primary). Moreover, to our knowledge, few studies have addressed concomitant rotator cuff pathology and its influence on adhesive capsulitis.
This article reports the MRA findings of patients considered to have stage-2 idiopathic adhesive capsulitis based on medical history and physical examination, and characterizes the incidental prevalence of rotator cuff pathologies and identifies differences between patients older and younger than 60 years in terms of their MRA findings.
Materials and Methods
From June 2005 to January 2007, 81 nonconsecutive patients with stage-2 adhesive capsulitis were prospectively evaluated by MRA. Stage-2 adhesive capsulitis was defined as pain, and a global limitation of motion (global limitation was considered present when active and passive forward flexion was less than 90° to 100°, external rotation at side was -107 to 107, and internal rotation below the thoracic vertebral level).5,7-9 All included patients were selected according to strict criteria,10 and had normal physical examinations (except for the above range of motion limitation) and normal plain radiographic findings.
Because our clinic is a tertiary referral hospital, patients generally see >2 doctors prior to referral, and most patients request special examination in addition to plain radiographs. Therefore, we performed MRA on all patients, except those who were claustrophobic or those who could not afford it. All patients underwent anteroposterior and axillary lateral radiography prior to further radiologic examinations, and patients showing radiologic abnormalities, such as, osteoarthritis or calcific material at the glenohumeral joint were excluded. Patients with a stiff shoulder due to severe post-traumatic injury, a post-surgical state, or any other known combined shoulder pathology were also excluded. To maintain the level of MRA image quality, patients who had undergone MRA at another institute were excluded.
A 1.5-Tesla magnetic resonance imager (Signa; General Electric Medical Systems, Milwaukee, Wisconsin) was used throughout. Just prior to imaging, patients were injected with 2-3 mL of Iomeron 300 (Bracco Imaging SpA, Milano, Italy) to verify a proper joint space by fluoroscopy. Afterwards, 40 mL of saline solution mixed with Magnevist (250:1 Schering AG, Berlin, Germany) was injected into the shoulder joint for arthrography. Magnetic resonance images were obtained using 3-mm thick slices with 1-mm inter-slice gap distance, a 16-cm field of view, and a matrix size of 256×192.
Two musculoskeletal radiologists unaware of the study purpose evaluated MRA findings. Radiologic reports issued by these two radiologists were reviewed and any discordant findings were re-evaluated until consensus was reached. For rotator cuff pathologies, MRA findings were graded as; normal, tendinopathy or tendinosis, partial-thickness tear, and full-thickness tear of the supraspinatus tendon. To compare the prevalences of associated rotator cuff pathology according to age, we divided patients into 2 groups; group 1 (<60 years), group 2 (>60 years). Statistical analysis was performed to identify differences between these two groups.
The x2-test and Wilcoxon’s two-sample test were used to compare demographic factors in the 2 groups. For supraspinatus tendon pathologies, Fisher’s exact test was used to compare prevalences. All statistical analyses were performed using SAS software version 9.13 (SAS Institute Inc, Cary, North Carolina). All analyses were performed using a 95% confidence interval to determine statistical significance.
Mean symptom duration prior to examination was 8.9 months (range, 2-36 months; SD 6.1), and mean patient age was 66 years (range, 36-73 years; SD, 3.1). There were 38 men and 43 women. Involved shoulders were left 35 and right 46. Seven patients had a minor trauma history prior to the development of shoulder symptoms. Eleven patients had diabetes mellitus and 21 patients had more severe night pain.
Prevalence of Concomitant Pathology in Overall Group
Fifty of the 81 patients (61.7%) had a supraspinatus tendon pathology on MRA. A small (<1 cm) full-thickness supraspinatus tendon tear was observed in 6 (7.4%) and a partial-thickness supraspinatus tendon tear in 25 (30.9%). In addition, supraspinatus tendinopathy were observed in 19 (23.5%) (Figure 1). The most common MRA finding, other than rotator cuff pathology, was the impression of adhesive capsulitis in 58 patients (71.6%).
Among these 58 patients with adhesive capsulitis and an MRA diagnosis of adhesive capsulitis, 98.2% (57 of 58) had a reduced axillary pouch volume,9 93.1% (54 of 58) had thickening of the capsule and synovium,11,12 81.0% (47 of 58) had enhancement of the rotator interval synovium and capsule.13
Figure 1: Supraspinatus tendon pathology in the MRA of patients with idiopathic adhesive capsulitis. Abbreviation: SSP, supraspinatus tendon.
Figure 2: Other intra-articular pathologies. Abbreviations: AC, adhesive capsulitis, SSC, subscapularis tendon.
Subscapularis tendon partial tears and SLAP II were reported in 14 of the 81 (17.3%) and 24 study participants (29.6%) respectively (Figure 2).
Difference in Cuff Pathology Between Two Age Groups
No significant differences between the 2 age groups were found in terms of symptom duration, sex, involved shoulder, or previous history (trauma, diabetes mellitus, or night pain). Thus the overall prevalences of supraspinatus pathologies were similar in the 2 groups (P=.1795). However, the 2 groups were significantly different in terms of the prevalences of small full thickness and partial thickness tears of the supraspinatus tendon (P=.0312 and 0.0242, respectively), ie, there were 13 supraspinatus tendon partial tears and 5 supraspinatus tendon full tears in group 2, but 12 supraspinatus tendon partial tears and only 1 supraspinatus tendon full tear in group 1 (Figure 3).
| || |
Figure 3: Supraspinatus tendon pathology in group 1 (<60 years) and in group 2 (>60 years).
This study shows that about 60% of patients considered to have idiopathic phase-II adhesive capsulitis had some form of rotator cuff pathology on MRA. In particular, 40% of all study participants had some form of rotator cuff tear. These MRA findings confuse the diagnosis of adhesive capsulitis by raising the question as to whether this disease is “primary or secondary” in nature. If it is considered that all study participants had idiopathic adhesive capsulitis, since this was our initial observation, we would have overlooked the 40% of patients (34 of 81) with a frozen shoulder and combined rotator cuff tears. Nonetheless, in many patients, conservative treatment improved stiffness and pain, as would be expected of primary adhesive capsulitis patients.
More studies are required to determine whether such cuff pathologies detected by MRA affect final clinical outcomes. Recent studies have reported good results after arthroscopic capsulotomy, even though some patients with adhesive capsulitis had either partial or small full-thickness rotator cuff tears,2,10,14-17 which suggests that these patient might do well even with a combined rotator cuff pathology.
Adhesive capsulitis or a frozen shoulder can be classified as primary (or idiopathic) or secondary. Primary adhesive capsulitis is characterized by idiopathic, progressive, painful loss of active and passive shoulder motion, whereas secondary adhesive capsulitis has a similar presentation and progression, but results from a known intrinsic or extrinsic cause. Intrinsic factors implicated in secondary adhesive capsulitis include rotator cuff tears, bursitis, and tendonitis, whereas extrinsic factors are generally related to trauma. Our study focus was on the fact that patients that would normally be considered to have primary adhesive capsulitis might have actually secondary adhesive capsulitis by definition. Although, MRA can exaggerate rotator cuff pathologic findings, our study shows that more than one third of patients had some form of tear in the supraspinatus tendon. However, we were unable to conclude whether these patients had primary or secondary adhesive capsulitis. Stiffness of the shoulder in the present study may have been primary or secondary due to a supraspinatus tendon pathology, subscapularis tear, or SLAP II lesion by MRA.
Neviaser and Nevaser9 described 4 stages or phases of adhesive capsulitis, but it is important to note that these stages represent a disease continuum rather than discrete well-defined stages. In stage 1, patients report pain during active and passive range of motion. The pain is described as achy at rest and sharp during motion with symptoms that persist for <3 months. In stage 2, known as the “freezing stage”, symptoms present for 3 to 9 months with chronic pain and a progressive loss of range of motion. Most patients are in this disease stage at presentation, because they are alarmed by continuous pain and progressive loss of motion affecting daily life. Our patients had mean symptom duration of 8.9 months at presentation, which also indicates stage 2. Stage 3 is the “frozen stage” where patients experience minimal pain at night or rest (except at the ends of motion ranges) but have significant shoulder stiffness, and symptoms that have been present for approximately 9 to 15 months. Finally, stage 4 the “thawing phase” is associated with minimal pain and a progressive improvement in range of motion due to capsular remodeling.18
Approximately 70% of the 81 patients enrolled in the present study had an MRA diagnosis of adhesive capsulitis. Several studies have presented characteristic MRI findings that might help diagnose adhesive capsulitis,13,19-21 although these MRI findings have not been proven to be essential for diagnosis in patients with a stiff shoulder.21 Our MRA study also revealed findings of joint capsule thickening, synovium of >4 mm, and loss of the inferior axillary pouch. Indications for MRI or MRA in patients with stiff shoulder include that there is a concern about underlying rotator cuff integrity or possibility of a soft tissue or osseous pathologic process.12 Furthermore, a medium sized or larger full-thickness rotator cuff tear combined with adhesive capsulitis is considered a disease entity that differs from idiopathic adhesive capsulitis, and is commonly categorized as secondary adhesive capsulitis. The patients enrolled in the present study, seemed less stiff than true stage-2 idiopathic patients, and were slightly older. We strongly recommend that MRI or MRA be conducted in this group of patients.
The patients in this study with combined supraspinatus tendon tears would have been considered to have idiopathic adhesive capsulitis without a specific imaging study. This raises the question as to whether this shoulder stiffness occurs secondary to subacromial impingement or impingement (including rotator cuff pathologies) results from bursal changes caused by altered glenohumeral mechanics in stiff shoulder.10,22 Valtonen et al23,24 suggested that the primary disease is a subacromial pathology and that frozen shoulder is a secondary phenomenon, but inflammatory changes in the subacromial bursa with a supraspinatus tear to be associated with the source of the pathologic process in adhesive capsulitis.
In other studies, few comments have been made concerning the effects of these combined lesions, especially with respect to rotator cuff pathology.2,10,14-16,25-29 This might indicate that rotator cuff pathology combined with adhesive capsulitis has no definite influence on final outcome. Watson et al17 reported that 6 patients had rotator cuff tears (2 articular partial thickness rotator cuff tear, 4 small full thickness rotator cuff tear) in the study of 73 patients with frozen shoulder syndrome who were treated with an arthroscopic capsulotomy. In the study by Griggs et al2 of 75 patients with stage-2 adhesive capsulitis, 4 combined rotator cuff tears (3 partial-thickness rotator cuff tear and 1 small full-thickness rotator cuff tear) were detected using MRI.
Harryman III et al15 reported 9 patients with partial thickness rotator cuff tear of approximately 2 to 3 mm in depth (8 articular and 1 bursal surface) in the study of 30 adhesive capsulitis patients who failed nonoperative management. However, adhesive capsulitis patients in general population could produce different results since the above-mentioned studies focused on patients with refractory shoulder stiffness patients.
Ogilvie-Harris et al16 reported that 50% of their study patients had an associated minor intra-articular disease, such as a partial rotator cuff tear or a glenoid labrum lesion, and concluded that minor intra-articular pathologies are clinically irrelevant in terms of final outcome.
Figure 4: MRA image of a full-thickness rotator cuff tear with combined adhesive capsulitis.
Other authors have advocated that rotator cuff partial tears are not substantial enough to cause symptoms, and partial tear of rotator cuff in patients with adhesive capsulitis attributed to age-related degenerative findings, and furthermore, were not considered to be an etiologic factor.22,28 Despite overemphasizing tendency of MRI, roughly 10% prevalence of full thickness rotator cuff tears (Figure 4) in our study population might have some clinical impact on the final outcome of combined adhesive capsulitis. Nonetheless, the rotator cuff pathologies (ie, a partial tear or tendinosis) in the remaining population might be attributable to normal degenerative aging of the rotator cuff.
The present study shows that adhesive capsulitis in and advanced age (>60 years) have higher prevalence of full or partial thickness tears of the supraspinatus tendon. However, it is unclear whether rotator cuff tear with adhesive capsulitis is attributable to age-related degenerative findings or whether it is the primary source of frozen shoulder. Although the prevalences of rotator cuff tear differed significantly in our 2 age groups, we believe that the difference between the 2 groups have little meaning in terms of management, and we should focus on identifying those with adhesive capsulitis combined with medium or large sized rotator cuff tears, because these patients would benefit most from early surgical intervention.
Despite successful results of nonoperative treatment, it is generally accepted that manipulation and arthroscopic release under anesthesia should be considered in patients resistant to nonoperative measures.8 However, several key questions remain to be answered. How can we predict who will be resistant to treatment? Can we detect at an early stage those patients who are likely to have a poor outcome to nonsurgical treatments? Do the 60% of patients (especially the 40% with tears) in the present study, who had some form of rotator cuff pathology, have a fate that differs from those without associated rotator cuff pathology? Should we be aggressive when treating partial-thickness rotator cuff tear or a small full-thickness rotator cuff tear combined with adhesive capsulitis? Accordingly, we will follow-up and compare the outcomes of patients with and without rotator cuff pathology.
The present study has several limitations. First, this study is based on the MRA findings of patients with a clinically impression of stage-2 idiopathic adhesive capsulitis. We might have overdiagnosed the degenerations of rotator cuffs and partial tears. Furthermore, there were other intra-articular pathologies, which might need treatment, other than rotator cuff pathologies. However, none of the patients were confirmed to have this pathology on arthroscopy. Second, this is a nonconsecutive retrospective study, and thus, is subject to selection bias. The mean age of our study participants was 66 years, which is older than of typical idiopathic adhesive capsulitis patients. Ultrasonography was performed instead of MRA in those who were relatively young or who refused because of the cost, and these individuals were not included in the present study. In addition, those not prepared to join the long waiting list for MRA or who objected to the injection of contrast media opted for MRI and were also excluded. Third, although a radiologic assessment was performed by 2 musculoskeletal radiologists who had >5 years of experience, our MRA findings are still vulnerable to interobserver bias.
Approximately one third of stage-2 adhesive capsulitis patients were found to have some form of supraspinatus tendon tear on MRA. Furthermore, <10% of patients who were considered to have idiopathic adhesive capsulitis had a full-thickness tear, which suggests that the cause of painful stiffness in these patients might be a secondary condition. Finally, the prevalence of supraspinatus tendon tear appears to be higher in adhesive capsulitis patients older than 60 years.
- Krishnan SG, Hawkins RJ, Boker DJ. Clinical evaluation of shoulder problems. In: Rockwood CA, Matsen III FA, Wirth MA, Lippitt SB, eds. The Shoulder. Philadelphia, PA: W.B. Saunders; 2004:145-181.
- Griggs SM, Ahn A, Green A. Idiopathic adhesive capsulitis. A prospective functional outcome study of nonoperative treatment. J Bone Joint Surg Am. 2000; 82(10):1398-1407.
- Binder AI, Bulgen DY, Hazleman BL, Roberts S. Frozen shoulder: a long-term prospective study. Ann Rheum Dis. 1984; 43(3):361-364.
- Grey RG. The natural history of “idiopathic” frozen shoulder. J Bone Joint Surg Am. 1978; 60(4):564.
- Reeves B. The natural history of the frozen shoulder syndrome. Scand J Rheumatol. 1975; 4(4):193-196.
- Shaffer B, Tibone JE, Kerlan RK. Frozen shoulder. A long-term follow-up. J Bone Joint Surg Am. 1992; 74(5):738-746.
- Nobuhara K, Sugiyama D, Ikeda H, Makiura M. Contracture of the shoulder. Clin Orthop Relat Res. 1990; (254):105-110.
- Hannafin JA, Chiaia TA. Adhesive capsulitis. A treatment approach. Clin Orthop Relat Res. 2000; (372):95-109.
- Neviaser RJ, Neviaser TJ. The frozen shoulder. Diagnosis and management. Clin Orthop Relat Res. 1987; (223):59-64.
- Pollock RG, Duralde XA, Flatow EL, Bigliani LU. The use of arthroscopy in the treatment of resistant frozen shoulder. Clin Orthop Relat Res. 1994; (304):30-36.
- Emig EW, Schweitzer ME, Karasick D, Lubowitz J. Adhesive capsulitis of the shoulder: MR diagnosis. AJR Am J Roentgenol. 1995; 164(6):1457-1459.
- Harryman II DT, Lazarus MD. The Stiff Shoulder. In: Rockwood J, C.A., Matsen III FA, Wirth MA, Lippitt SB, eds. The Shoulder. Vol 2. 3rd ed. Philadelphia, PA: W.B. Saunders; 2004:1121-1172.
- Connell D, Padmanabhan R, Buchbinder R. Adhesive capsulitis: role of MR imaging in differential diagnosis. Eur Radiol. 2002; 12(8):2100-2106.
- Andersen NH, Sojbjerg JO, Johannsen HV, Sneppen O. Frozen shoulder: arthroscopy and manipulation under general anesthesia and early passive motion. J Shoulder Elbow Surg. 1998; 7(3):218-222.
- Harryman DT II, Matsen FA III, Sidles JA. Arthroscopic management of refractory shoulder stiffness. Arthroscopy. 1997; 13(2):133-147.
- Ogilvie-Harris DJ, Biggs DJ, Fitsialos DP, MacKay M. The resistant frozen shoulder. Manipulation versus arthroscopic release. Clin Orthop Relat Res. 1995; (319):238-248.
- Watson L, Dalziel R, Story I. Frozen shoulder: a 12-month clinical outcome trial. J Shoulder Elbow Surg. 2000; 9(1):16-22.
- Sheridan MA, Hannafin JA. Upper extremity: emphasis on frozen shoulder. Orthop Clin North Am. 2006; 37(4):531-539.
- Jung JY, Jee WH, Chun HJ, Kim YS, Chung YG, Kim JM. Adhesive capsulitis of the shoulder: evaluation with MR arthrography. Eur Radiol. 2006; 16(4):791-796.
- Lee JC, Sykes C, Saifuddin A, Connell D. Adhesive capsulitis: sonographic changes in the rotator cuff interval with arthroscopic correlation. Skeletal Radiol. 2005; 34(9):522-527.
- Manton GL, Schweitzer ME, Weishaupt D, Karasick D. Utility of MR arthrography in the diagnosis of adhesive capsulitis. Skeletal Radiol. 2001; 30(6):326-330.
- Wiley AM. Arthroscopic appearance of frozen shoulder. Arthroscopy. 1991; 7(2):138-143.
- Valtonen EJ. Subacromial betamethasone therapy. II. Subacromial betamethasone and methylprednisolone injections in treatment of frozen shoulder and supra spinam tendinitis. A comparative trial. Ann Chir Gynaecol Fenn Suppl. 1974; (188):9-16.
- Valtonen EJ. Subacromial betamethasone therapy, I: The effect of subacromial injection of betamethasone in cases of painful shoulder resistant to physical therapy. Ann Chir Gynaecol Fenn Suppl. 1974; (188):5-8.
- Bennett WF. Addressing glenohumeral stiffness while treating the painful and stiff shoulder arthroscopically. Arthroscopy. 2000; 16(2):142-150.
- Jerosch J. 360 degrees arthroscopic capsular release in patients with adhesive capsulitis of the glenohumeral joint--indication, surgical technique, results. Knee Surg Sports Traumatol Arthrosc. 2001; 9(3):178-186.
- Massoud SN, Pearse EO, Levy O, Copeland SA. Operative management of the frozen shoulder in patients with diabetes. J Shoulder Elbow Surg. 2002; 11(6):609-613.
- Ogilvie-Harris DJ, Wiley AM. Arthroscopic surgery of the shoulder. A general appraisal. J Bone Joint Surg Br. 1986; 68(2):201-207.
- Warner JJ. Frozen Shoulder: Diagnosis and Management. J Am Acad Orthop Surg. 1997; 5(3):130-140.
- Binder AI, Bulgen DY, Hazleman BL, Tudor J, Wraight P. Frozen shoulder: an arthrographic and radionuclear scan assessment. Ann Rheum Dis. 1984; 43(3):365-369.
Drs Yoo, Ahn, and Koh are from the Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, and Dr Lee is from the Department of Orthopedic Surgery, Korea University School of Medicine, Korea University Medical Center Ansan Hospital, Kyonggido, Korea.
Drs Yoo, Ahn, Lee, and Koh have no relevant financial relationships to disclose.
Correspondence should be addressed to: Yong Seuk Lee, MD, Department of Orthopedic Surgery, Korea University Ansan Hospital, 516 Gozan-dong, Danwon-gu, Ansan 425-707, Korea.