| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Diagnostic Imaging, Brown University School of Medicine, Rhode
Island Hospital, 593 Eddy St., Providence, RI 02903.
2
Department of Orthopedic Surgery, Brown University School of Medicine, Rhode
Island Hospital, Providence, RI 02903.
3
Bayside Orthopaedics, 300C Faunce Corner Rd., North Dartmouth, MA 02747.
Received September 21, 1999;
accepted after revision November 4, 1999.
Address correspondence to G. A. Tung.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. From 2211 MR imaging examinations, 51 (2.3%) cysts in 46 patients were identified. MR arthrography, (n = 5), cystography (n = 1), arthroscopy (n = 17), percutaneous needle aspiration (n = 4), and medical records were also reviewed (n = 46).
RESULTS. On MR imaging and arthrography, cysts were best viewed on T2-weighted images. Mean cyst diameter and volume were 2.2 cm and 2.8 cm3, respectively. Fifty-seven percent of cysts were located adjacent to the posterior labrum. On MR imaging and arthroscopy, a labral tear was identified in 27 (53%) and 15 (88%) patients, respectively. Eight cysts that caused compression neuropathy were large (mean size, 3.1 cm; p = 0.04) and located next to the posterior or inferior labrum. In four of five patients, MR arthrograms showed no intraarticular contrast material in the cyst. Cystograms showed no communication with the gleno-humeral joint space, and cyst aspiration resulted in temporary symptom relief; however, cysts recurred in three of four patients.
CONCLUSION. Most paralabral cysts are associated with labral tears. Paralabral cysts may be difficult to identify on MR arthrography unless a T2-weighted sequence is performed. Direct communication between a cyst and joint space rarely occurs. A posterior or inferior cyst may cause compression neuropathy of the suprascapular or axillary nerve, respectively. Cyst aspiration may result in temporary relief of symptoms, but an untreated labral tear should be suspected if cysts recur.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pathogenesis of paralabral cysts is unknown. A paraglenoid cyst could be a synovial cyst, ganglion cyst, or pseudocyst. A synovial cyst is lined by synovial cells and forms from evagination of the joint capsule or paraarticular bursa [10]. A ganglion cyst may arise from a joint capsule, bursa, ligament, tendon, or subchondral bone. A pseudocyst may result from the extrusion of joint fluid through a labrocapsular tear into adjacent soft tissues. This pathogenesis is similar to that of a meniscal cyst [11, 12].
Paralabral cysts are often incidentally discovered on MR imaging of the shoulder [1, 3]. To our knowledge, neither the findings of paralabral cysts on MR arthrography nor the characteristics of axillary compression neuropathy caused by paralabral cysts have been reported. We investigated the appearance of paraglenoid labral cysts and the frequency of concomitant labrocapsular disease on MR imaging and MR arthrography. We also assessed sonographically guided cyst aspiration as a primary treatment method.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Diagnostic Criteria
A paralabral cyst was noted when a focal well-defined collection of fluid
within 1 cm of the glenoid labrum appeared on MR imaging; subchondral cysts
were excluded. A unilocular or simple cyst does not reveal any septation or
loculation regardless of size (Fig.
1A,1B).
A cyst was defined as septated or multiloculated when at least one dividing
septum was visible inside the cyst. The maximal diameter of the cyst was
measured in three dimensions, and the volume was estimated according to the
prolate ellipse formula (cyst volume = d1 x d2 x
d3 x [
/3]).
|
|
On MR imaging, a labral tear was diagnosed when a simple or branched linear, stellate, or globular focus of increased signal intensity in the glenoid labrum extended to the articular or bursal labral surface or when the labrum was focally truncated or absent [13, 14]. Additional signs of a labral tear included labral displacement or an irregular margin, suggestive of focal fraying. On MR arthrography, labral tears were diagnosed when we noted abnormal focal extension of intraarticular contrast material into the fibrous portion of the labrum, labral fraying, displacement, or detachment [15,16,17]. The location of the tear and the paralabral cyst were described as superior, anterior, inferior, or posterior on the basis of a clockface description of the glenoid labrum [13, 14, 17]. Thus, a cyst adjacent to or a tear within the superior labrum was located between the 11- and 1-o' clock positions, around the supraglenoid tubercle and origin of the superior glenohumeral ligament. The anterior labrum was located between the 1- and 5-o' clock positions and included the origin of the middle glenohumeral ligament. The inferior labrum was located between the 5- and 8-o' clock positions and included the origin of the anterior and posterior bands of the inferior glenohumeral ligament. A tear of the posterior labrum was located between the 8- and 11-o' clock positions.
The diagnosis of muscle denervation resulting from compression neuropathy was made when we noted diffuse increased signal intensity in the muscle on a T2-weighted image, with or without muscle atrophy. With subacute muscular denervation, we noted abnormal signal intensity (increased signal intensity on T2-weighted images) with relatively preserved muscle mass [18,19,20] (Fig. 2A,2B,2C). Muscle atrophy was seen with chronic denervation [6] (Figs. 1A,1B and 3A,3B,3C).
|
|
|
|
|
|
MR Imaging Techniques
MR imaging of the shoulder was performed on a 1.5-T unit (Magnetom Vision;
Siemens Medical Systems, Erlangen, Germany) or a 0.2-T unit (Magnetom Open;
Siemens Medical Systems). Patients were placed in the supine position with
their shoulder in a neutral or slightly externally rotated position, and MR
imaging was performed with the receive-only flexible surface coil. After a
localization sequence, a four-sequence examination of the shoulder was
performed on the 1.5-T unit consisting of an axial T2*-weighted
two-dimensional (2D) fast low-angle shot sequence or an axial proton
density-weighted turbo spin-echo sequence, a coronal oblique proton
density-weighted turbo spin-echo sequence, a coronal oblique T2-weighted turbo
spin-echo sequence with fat saturation, and a sagittal oblique T2-weighted
turbo spin-echo sequence with fat saturation. The scan protocol for the axial
T2*-weighted 2D fast low-angle shot sequence included a TR/TE of
450/10, a flip angle of 25°, a 16-cm field of view, a 132 x 256
matrix, a 4-mm slice thickness, a 25% gap, and one acquisition. The scan
protocol for the axial proton density-weighted turbo spin-echo sequence
included a TR/effective TE of 2000/18, an echo train length of three, a 16-cm
field of view, a 252 x 256 matrix, a 3-mm slice thickness, no gap, and
one acquisition. The scan protocol for the coronal oblique proton
density-weighted turbo spin-echo sequence included a TR/effective TE of
2000/18, an echo train length of three, a 16-cm field of view, a 192 x
256 matrix, a 3-mm slice thickness, no gap, and two acquisitions. The scan
protocol for the coronal oblique T2-weighted turbo spin-echo sequence with fat
saturation included a TR range/effective TE of 3600-5000/96, an echo train
length of seven, a 16-cm field of view, a 256 x 256 matrix, a 4-mm slice
thickness, no gap, and two acquisitions. The scan protocol for the sagittal
oblique T2-weighted turbo spin-echo sequence with fat saturation included a TR
range/effective TE of 3500-5000/96, an echo train length of three, a 16-cm
field of view, a 256 x 256 matrix, a 3-mm slice thickness, a 10% gap,
and one acquisition.
On the 0.2-T magnet, after the localization sequence, a standard four-sequence examination was performed, consisting of an axial T2*-weighted 2D fast low-angle shot sequence or an axial proton density-weighted turbo spin-echo sequence, a coronal oblique T1-weighted conventional spin-echo sequence, a coronal oblique turbo spin-echo T2-weighted sequence, and a sagittal oblique T2-weighted turbo spin-echo sequence. The scan protocol for the axial T2*-weighted 2D fast low-angle shot sequence included a TR/TE of 560/17, a flip angle of 80°, an 18-cm field of view; a 192 x 256 matrix, a 5-mm slice thickness, a 20% gap, and two acquisitions. The scan protocol for the axial proton density-weighted turbo spin-echo sequence included a TR/effective TE of 2000/24, an echo train length of five, an 18-cm field of view, a 250 x 256 matrix, a 5-mm slice thickness, no gap, and two acquisitions. The scan protocol for the coronal oblique T1-weighted conventional spin-echo sequence included a TR range/TE range of 500-660/15-26, an 18-cm field of view, a 192 x 256 matrix, a 4-mm slice thickness, a 30% gap, and two acquisitions. The scan protocol for the coronal oblique turbo spin-echo T2-weighted sequence included a TR range/effective TE range of 2050-3000/80-100, an echo train length of seven, a 16-cm field of view, a 195 x 256 matrix, a 4-mm slice thickness, no gap, and two acquisitions. The scan protocol for the sagittal oblique T2-weighted turbo spin-echo sequence included a TR/effective TE of 3000/100, an echo train length of seven, an 18-cm field of view, a 196 x 256 matrix, a 4-mm slice thickness, a 30% gap, and two acquisitions.
Formal approval was obtained from the institutional review board to perform MR arthrography of the shoulder. A 2 mmol/l solution of gadopentate dimeglumine was prepared by diluting Magnevist (Berlex Laboratories, Wayne, NJ) or Omniscan (Nycomed, Amersham, Wayne, PA) in saline solution. To this diluted solution, we added 60% meglumine diatrizoate (Reno-M-60; Squibb, New Brunswick, NJ), 1% lidocaine, and 1:1000 epinephrine. Under fluoroscopic guidance, 20 ml of this solution was injected into the glenohumeral joint using standard techniques. MR imaging of the shoulder was then completed within 60 min of the intraarticular instillation of contrast solution. Using the 1.5-T unit and a circularly polarized flexible surface coil, fat-saturated T1-weighted conventional spin-echo images of the shoulder were obtained in the axial, coronal oblique, abducted externally rotated coronal oblique, and sagittal oblique planes (TR/TE, 780-800/12; field of view, 16 cm; matrix, 192 x 256; slice thickness, 3 mm; no interslice gap, acquisitions, two). Additionally, a T2-weighted turbo spin-echo sequence with fat saturation was performed in the coronal oblique plane (TR/effective TE, 3800/96; echo train length, seven; field of view, 16 cm; matrix, 252 x 256; slice thickness, 3-mm; interslice gap, 10%; acquisition, one).
Cyst Aspiration
Four paralabral cyst punctures were performed under sonographic guidance
using a standard sterile technique. In one patient, corticosteroid was
injected into the paralabral cyst after aspiration. In another patient, the
cyst was punctured under sonographic guidance and a cystogram was obtained
after instillation of iodinated contrast material (Fig.
4A,4B).
After cystography, contrast material was aspirated from the cyst.
|
|
Statistical Analysis
Tests of statistical significance were performed using the unpaired
Student's t test and the chi-square test with Yates correction. The
null hypothesis was rejected with a p value of less than 0.05.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Paralabral Cysts
The maximum diameter of cysts ranged from 0.5 to 6.1 cm (mean, 2.2 cm;
standard error of the mean [SEM], 0.2 cm). The average cyst volume was 2.8
cm3 (range, 0.1-23.9 cm3; SEM, 0.7 cm3).
Multilocular cysts were identified in 27 patients (59%); many of these were
tapered and pointed to the labrum (2-4, 6, and 7). Unilocular simple cysts
were often smaller than septated cysts. The location of cysts in proximity to
the glenoid labrum was posterior in 57% (n = 29), anterior in 21%
(n = 11), superior in 14% (n = 7), and inferior in 8%
(n = 4).
Labral Tears
On MR imaging, 27 labral tears were identified. With the exception of two
anterior tears, all the labral tears were located in the same anatomic
quadrant as the paralabral cysts. On arthroscopy, labral tears diagnosed on MR
imaging were verified in 15 patients, and in two patients, tears visible on MR
imaging were not confirmed. In three patients, arthroscopy showed additional
labral tears that were not visible on MR imaging. In five patients with two
labral cysts, four had both cysts in close proximity to one labral tear.
On MR arthrography, intraarticular contrast material dissected into a discrete labral tear in all patients but was not revealed inside the paralabral cyst in four of five patients (Fig. 5A,5B,5C). In one patient, intraarticular contrast material filled a small posterior paralabral cyst through a labral tear (Fig. 6A,6B). On MR arthrography, paralabral cysts were more difficult to identify on T1-weighted images and were best viewed on T2-weighted images (Fig. 5A,5B,5C); this was particularly true when cysts were small.
|
|
|
|
|
Compression Neuropathy Associated with Paralabral Cysts
MR imaging findings of infraspinatus (n = 6) (Fig.
1A,1B)
or teres minor muscle denervation (n = 2) (Figs.
2A,2B,2C
and
3A,3B,3C)
were found in eight patients; only two of these patients complained of
weakness and three had objective signs of shoulder external rotation weakness.
Electrodiagnostic studies were performed in six patients and were consistent
with compression neuropathy in two patients in whom MR imaging revealed signs
of muscle denervation. Electrodiagnostic studies revealed normal findings in
two patients in whom MR imaging of the muscle appeared normal. Two patients
had suprascapular nerve denervation on electrodiagnostic studies that were not
visible on MR imaging.
Signs of muscle denervation on MR imaging correlated with the size and location of paralabral cysts. The average volume of cysts associated with MR imaging signs of denervation was 6.0 cm3 (range, 0.7-23.8 cm3) compared with a mean volume of 2.2 cm3 (range, 0.1-18.7 cm3) for all other paralabral cysts (p = 0.04). The average maximum diameter of cysts associated with muscle denervation was 3.1 cm (range, 1.1-4.8 cm). Six cysts associated with infraspinatus muscle denervation were located in proximity to the posterior labrum (from 8- to 11-o'clock positions); two cysts associated with teres minor denervation were located in the 9- (Fig. 2A,2B,2C) and 7-o'clock positions (Fig. 3A,3B,3C), respectively.
Cyst Aspiration
In each of four cyst punctures, aspirate consisted of mucoid gelatinous
fluid that contained scattered histiocytes or foam cells. Follow-up shoulder
sonography at 4 months showed the recurrence of paralabral cysts in three of
four patients, including one patient in whom corticosteroid was injected into
the cyst after its contents were aspirated. Cystography revealed a
smooth-walled loculated cyst that tapered toward the glenoid labrum, but we
noted no contrast opacification of the glenohumeral joint or the paraglenoid
bursae (Fig.
4A,4B).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although often coincident, not all paralabral cysts are associated with labral tears revealed on MR imaging. In this study, labral tears were identified with 59% of paralabral cysts on nonarthrographic MR imaging. Four reasons explain this finding. First and most important, MR imaging may be insensitive to some labral tears [5, 13, 21]. Unlike MR imaging, MR arthrography revealed labral tears for all patients in this study. An important technical observation for MR arthrography is that all cysts were revealed on T2-weighted MR images but some were difficult to identify on fat-saturated T1-weighted MR arthrograms. Second, cyst-causing labral tears may spontaneously heal [1]. Third, paraarticular cysts may result from a labral disease other than a tear. For example, in the knee, meniscal cysts have been found in patients with osteoarthritis and calcium pyrophosphate deposition disease [12]. Fourth, paralabral cysts may be primary ganglion cysts that form spontaneously in a joint capsule, bursa, or tendon sheath in the absence of labral disease [22, 23].
Pathophysiology of Paralabral Cysts and Implications for
Treatment
That the principal cause of a paralabral cyst is an adjacent labral tear is
supported by coincidence, proximity, and the tapered shape of some cysts. In
this series, 93% of cysts were located in the same quadrant as the labral
tear. The prevailing theory is that these cysts may form after the
capsulolabral complex is torn or avulsed
[1,
3]. Through the tear, synovial
fluid can leak from the joint into paraarticular tissues causing the
development of cysts. In the knee, hip, and wrist, associations between
cartilaginous tears and ligament injuries have been linked to the presence of
paraarticular cysts [12].
Several surgical studies have reported that shoulder ganglia seem to originate in the fibers of the glenohumeral joint capsule, but that direct communication with the joint space rarely occurs [4, 7, 8, 24]. Direct communication between the joint space and the cyst was rarely observed in this study as well. Only one of five MR arthrograms showed intraarticular contrast material inside a paralabral cyst, and the glenohumeral joint space was unenhanced on cystograms we obtained. Four reasons explain this observation. First, a communication exists but was not revealed because of the type of imaging used. For instance, a study by Malghem et al. [25] revealed that imaging performed at least 1 hr after intraarticular contrast material administration was necessary to show contrast opacification of parageniculate cysts. At our institution, MR arthrography is performed within 1 hr of intraarticular contrast material administration; therefore, the contrast material may have appeared if delayed MR arthrography had been performed. Second, a communication exists but requires high intraarticular pressure or low intracystic pressure to fill the cyst from a joint injection. A study by Fritz et al. [6] concluded that chronic increased intraarticular pressure was necessary for tear-associated paraglenoid cysts to form. Third, incomplete healing of the labrocapsular lesion may lead to the formation of a one-way valve mechanism [3]. Fourth, a connection between the joint space and cyst may have closed because of the spontaneous healing of a labral tear [1].
Even though the communication between the cyst and glenohumeral joint space may be difficult to reveal, our experience with percutaneous cyst aspiration suggests that the cyst may recur if concomitant labral disease is not treated. In this study, percutaneous cyst aspiration provided only temporary benefit. Although a study by Fritz et al. [6] reported the successful treatment of paralabral cysts with CT-guided aspiration alone, cyst recurrence after open surgical cyst excision has been reported [2, 3]. Studies by Tirman et al. [1], Fehrman et al. [2], Iannotti and Ramsey [9], and Moore et al. [3] have reported a total of 16 paralabral cysts successfully treated by arthroscopic cyst aspiration alone. We believe that the important difference between arthroscopic treatment and that of percutaneous cyst aspiration is that the arthroscopic method decompresses the cyst through either a capsulotomy or a labral tear and then arthroscopic repair of the labrum is performed. Cyst recurrence after percutaneous aspiration suggests that the primary cause of the paralabral cyst has not been addressed and warrants a search for and treatment of an associated labral tear or other labrocapsular disease.
Compression Neuropathy Associated with Paraglenoid Cysts
Large paraglenoid cysts may compress the suprascapular nerve or axillary
nerve and cause shoulder weakness through denervation of external rotator
muscles
[1,2,3,
6]
(Fig. 7). Distal branches of
the suprascapular nerve, derived from the superior trunk of the brachial
plexus, supply the infraspinatus muscle after passing through the spinoglenoid
notch. As the nerve courses around the spinoglenoid notch to enter the
infraspinous fossa, it passes within 21 mm of the glenoid rim
[26]. In our study, posterior
and inferior paralabral cysts associated with muscle denervation had an
average diameter of 3.1 cm and were significantly larger than cysts
unassociated with muscle denervation. The axillary nerve is one of the
terminal branches of the posterior cord of the branchial plexus and innervates
the teres minor muscle. At the lower border of the subscapularis muscle, the
axillary nerve curves inferior to the glenohumeral joint capsule to traverse
the quadrangular space. In most patients, compression neuropathy of the
axillary nerve is caused by stenosis of the quadrangular space (quadrilateral
space syndrome) and a fibrous band or as a result of traumatic shoulder injury
[27,
28]. To our knowledge, we
present the first patient with axillary nerve compression caused by a
paraglenoid labral cyst diagnosed on MR imaging. In this patient, an inferior
paraglenoid cyst compressed the axillary nerve just proximal to the
quadrangular space. In another patient with teres minor muscle denervation, we
noted a posterior paralabral cyst in proximity to the spinoglenoid notch and
we postulate variant innervation of the teres minor muscle by a branch of the
suprascapular nerve. Although we could not find previous descriptions of this
anomalous innervation, variant anatomy of the terminal nerves of the brachial
plexus is common [29].
|
On MR imaging, muscle denervation is indicated by diffuse abnormal signal intensity in the affected muscle with or without muscle atrophy. With clinical recovery of motor function, signal intensity returns to normal [6, 18,19,20]. It has been postulated that the abnormal signal intensity in denervated muscle is caused by an increase in extracellular fluid without a significant change in total water content [30]. Using the short tau inversion-recovery sequence, West et al. [20] reported that signal intensity changes in muscle after peripheral nerve injury precede changes on electrodiagnostic studies. However, in two patients with infraspinatus denervation in this study, electrodiagnostic studies were more sensitive than MR imaging. Two explanations for this observation exist. First, it is possible that signal intensity changes in denervated muscle were not detected because fat-saturated echo T2-weighted imaging is less sensitive than the short tau inversion-recovery sequence. Second, cyst-related compression neuropathy may be predominantly a neuropraxic injury. In neuropraxic nerve injuries, myelin injury causes conduction delay or blockage but the axons are spared. In predominantly neuropraxic injuries, muscles have normal signal intensity on MR imaging. In contrast, in axonotometic nerve injuries, axonal loss distal to the site of nerve injury and abnormal signal intensity appear in affected muscles as a result of denervation [20].
Acknowledgments
We thank Tanya Entzian for her artistic contributions.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. P. Schroder, O. Skare, M. Stiris, E. Gjengedal, G. Uppheim, and J. I. Brox Treatment of Labral Tears with Associated Spinoglenoid Cysts without Cyst Decompression J. Bone Joint Surg. Am., March 1, 2008; 90(3): 523 - 530. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kassarjian, M. Torriani, H. Ouellette, and W. E. Palmer Intramuscular Rotator Cuff Cysts: Association with Tendon Tears on MRI and Arthroscopy Am. J. Roentgenol., July 1, 2005; 185(1): 160 - 165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P J O'Connor and A J Grainger Ultrasound imaging of joint disease Imaging, June 1, 2002; 14(3): 188 - 201. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
