| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1
Department of Radiology, University of California at San Francisco, 505
Parnassus Ave., Ste. M392, San Francisco, CA 94143-0628.
2
Present address: Department of Radiology, Freeman Hospital, High Heaton,
Newcastle upon Tyne, NE7 7DN, United Kingdom.
3
San Francisco Magnetic Resonance Center, 3333 California St., San Francisco,
CA 94118.
Received April 21, 1999;
accepted after revision October 13, 1999.
Address correspondence to A. J. Grainger.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Fluid in the subcoracoid space, revealed on MR imaging of the shoulder, may lie in the subcoracoid bursa or the subscapularis recess (both structures can be distinguished with MR imaging). In our patients, subcoracoid effusions were often associated with anterior rotator cuff tears, including tears of the rotator interval.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
The superior subscapularis recess is also known as the subscapularis bursa. It represents a recess of the joint capsule projecting anteriorly between the superior and the middle glenohumeral ligaments. This recess lies between the subscapularis muscle and the anterior surface of the scapula and extends above the superior margin of the subscapularis tendon over which it may hang similar to a saddlebag (becoming adherent to the tendon's anterior surface) [1]. The size of the recess is related to the size of the middle glenohumeral ligament, becoming larger as the ligament becomes more deficient [2]. An inferior subscapularis recess of the glenohumeral joint is also described, projecting between the middle and the inferior glenohumeral ligaments [2]. Its more inferior position means that it is located deep in relation to the subscapularis muscle.
Because the superior subscapularis recess extends anterior to the subscapularis tendon, fluid in this recess may mimic an effusion in the subcoracoid bursa. The distinction between the two structures is of importance to radiologists because the causes of a subcoracoid bursa effusion may be different from the causes of a subscapularis recess effusion. Fluid in the subscapularis bursa may simply represent physiologic joint fluid, whereas fluid in the subcoracoid bursa represents a pathologic process, such as bursitis, or results from communication with another fluid-containing structure.
This study describes the MR anatomy of the subcoracoid bursa and identifies the characteristics that distinguish it from the superior subscapularis recess of the glenohumeral joint. Additionally, we examine associated abnormalities of the shoulder joint to better establish the significance of a subcoracoid bursa effusion.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All patients underwent routine MR imaging of the shoulder using a variety of sequences. All protocols included, in the neutral position, T1-weighted and T2-weighted MR imaging in the axial and oblique coronal planes and T2-weighted MR imaging in the oblique sagittal plane. All images were obtained on a 1.5-T Signa scanner (General Electric Medical Systems, Milwaukee, WI).
Using a workstation with MRVision software (MRVision, Menlo Park, CA), two radiologists reviewed the images and clinical histories of 16 patients. When available, the findings at arthroscopy or surgery were also recorded. While reviewing the MR images, the radiologists obtained measurements for the maximum dimensions of fluid collections. The size of subcoracoid bursa effusions was classified on the basis of the maximum diameter. Effusions were small (<1 cm), moderate (1-2 cm), or large (>2 cm) according to the system of Schraner and Major [3]. Radiologists also noted any communication between the subcoracoid effusion and the glenohumeral joint or subacromial bursa. Other MR imaging findings were recorded, including information regarding the rotator cuff and the rotator interval. The presence or absence of a subacromial bursa or glenohumeral joint effusion was also noted and subjectively graded as small or large. Consensus agreement was achieved.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
No statistically significant difference was observed between the axial dimensions of the two types of fluid collections. However, the mean caudal extent of the subcoracoid bursae (34.2 mm; SD, 4.5) was significantly greater than that of the subscapularis recesses (8.4 mm; SD, 4.1; p < 0.001). This difference in size was well depicted on sagittal MR images (Fig. 3). All patients with subcoracoid effusion had a maximum dimension greater than 2 cm, a large measurement according to the classification system of Schraner and Major [3].
|
Two patients with fluid in the subscapularis recess had evidence of tendinosis in the supraspinatus tendon; however, no rotator cuff tear was identified. All patients with subcoracoid bursa effusions had evidence of rotator cuff tears (confirmed surgically in six patients). Evidence of a rotator interval tear was also noted in 11 patients with fluid in the subcoracoid bursa (Fig. 4A,4B,4C). This evidence was highlighted by the presence of fluid in the rotator interval and appeared as irregularity and thickening of the rotator interval capsule and coracohumeral ligament. One patient also had a complete tear of the long head of the biceps, which may have been associated with a tear of the rotator interval.
|
|
|
All patients with rotator cuff tears had fluid in the subacromial bursa, and in seven of the patients, the quantity of fluid in the subacromial bursa was small compared with that in the subcoracoid bursa. Ten (77%) patients with subcoracoid bursa effusions had bursa that contained septa (Fig. 4A,4B,4C). No septa were noted in the patients with subscapularis recess. In three (23%) patients with subcoracoid bursa effusion, communication between the subcoracoid and subacromial bursae was noted (Fig. 5). No patients in our study had evidence of subscapularis muscle tears or signs of labral tear or cyst.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although MR imaging reports indicated that all patients had subcoracoid bursa effusions, we reviewed the images in the oblique sagittal plane and found three patients with communication between the subcoracoid fluid collection and the glenohumeral joint. No saddlebag appearance was seen with the subcoracoid bursa. Furthermore, the caudal extent of the subcoracoid bursa was greater than that of the subscapularis recess. A study by Horowitz and Tocantins [1] reported measurements of the mean caudal extension below the coracoid process (12.5-18.8 mm) less than those we observed (range, 12.0 -42.9 mm; mean, 34.2 mm). This difference may be caused by the fluid-filled distended bursae observed in our study.
Because we studied only three patients with subscapularis recess fluid, we cannot suggest concrete guidelines for distinguishing bursa from recess based on size. However, a distinction can be made by searching for the saddlebag appearance and communication with the glenohumeral joint as seen on oblique sagittal MR images.
The clinical significance of an effusion in the subcoracoid bursa is unclear. In a study by Schraner and Major [3], five of 11 patients had associated rotator cuff tears. This result deviates from our findings. Their patients were classified on the basis of the amount of bursal fluid assessed on sagittal MR images, and Schraner and Major found more patients (4/6) with rotator cuff tears among patients with moderate or large amounts of bursal fluid. Using the classification system from that paper, all the subcoracoid effusions in our study were large. Smaller effusions in our study were presumably not reported at the time of MR imaging. This may explain why our results contradict their conclusion: effusions of the subcoracoid bursa are frequently an incidental finding.
Our results suggest an association between large subcoracoid effusions and rotator cuff tears. In most of our patients (11/13), disruption of the rotator interval was also noted. Although it is difficult to diagnose tears of the rotator interval on MR imaging [4], we noted six patients with MR images that revealed tear extensions in the rotator interval.
Le Huec et al. [5] reported that in five of 10 patients with rotator interval tears, CT arthrography revealed contrast medium anterior to the subscapularis. It is possible that the contrast medium was located in the subcoracoid bursa; however, their report does not mention whether subcoracoid effusions were seen on MR imaging. Further studies are required to establish whether the presence of subcoracoid bursa effusions is an indication of rotator interval tears. Research shows the importance of the rotator interval and the detection of injury to its structures [6].
The presence of subcoracoid bursa fluid was uncritical to the diagnosis of rotator cuff tears in our patients. However, our results indicate that when an effusion is seen, a careful review of the cuff (including the rotator interval) is required. Other researchers have indicated that shoulder pain may result from an isolated subcoracoid bursitis [3, 7]. However, this diagnosis should be made only after other possible causes, including tears of the rotator cuff and interval, have been eliminated.
Horowitz and Tocantins [1] reported that the subcoracoid bursa communicated with the subacromial bursa in 11 of 100 shoulders. Another study, using subacromial bursography, revealed communication with the subcoracoid bursa in 10.7% of patients [8]. Schraner and Major [3] reported a higher frequency of communication between the two bursae (55%) than that seen in our study (23%) or the earlier studies. Given that this communication exists, patients with rotator cuff tears associated with subacromial bursa effusions should have fluid in the subcoracoid bursa. A direct continuity between the two bursae may make them indistinguishable, appearing as distended subacromial bursae extending anteroinferiorly (Fig. 5).
The association between rotator cuff tears, including tears of the rotator interval, and subcoracoid effusions in the absence of communication between the subcoracoid and subacromial bursae is harder to explain. Because the tears in our study were all anterior, the bursal effusion may relate to an inflammatory response produced after trauma. Alternatively, the effusion may represent a secondary response to impingement resulting from the cuff tear. Even though there was no qualitative evidence of subcoracoid impingement in our patients, formal measurements were not obtained.
Other researchers discuss the inadvertent injection of contrast media into the subcoracoid bursa during arthrography. If this occurs, contrast material may appear in the subacromial bursa as potential communication and, if the subcoracoid injection is overlooked, a false-positive diagnosis of a rotator cuff tear may be made [3, 9, 10].
Our study, similar to that of Schraner and Major [3], is limited by its retrospective nature. Only patients with shoulder symptoms were included. Additionally, because patients were included only if they had been reported with a subcoracoid bursa effusion, some patients with subcoracoid effusions or subscapularis recess fluid may have been excluded if their fluid was not reported. More studies are required to examine the incidence and significance of subcoracoid bursa effusions in asymptomatic patients.
In conclusion, our results suggest that the subcoracoid bursa and the subscapularis recess can be distinguished on MR imaging, particularly with a careful review of oblique sagittal MR images. On the basis of this study, subcoracoid bursa effusions do not appear to be an incidental finding but may be associated with the rotator cuff and rotator interval tears.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. Morag, J. A. Jacobson, G. Shields, R. Rajani, D. A. Jamadar, B. Miller, and C. W. Hayes MR Arthrography of Rotator Interval, Long Head of the Biceps Brachii, and Biceps Pulley of the Shoulder Radiology, April 1, 2005; 235(1): 21 - 30. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Martinoli, S. Bianchi, N. Prato, F. Pugliese, M. P. Zamorani, M. Valle, and L. E. Derchi US of the Shoulder: Non-Rotator Cuff Disorders RadioGraphics, March 1, 2003; 23(2): 381 - 401. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. M. Major and A. J. Grainger Imaging of the Subcoracoid Bursa Am. J. Roentgenol., March 1, 2001; 176(3): 812 - 813. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
