| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 Department of Radiology, Duke University Medical Center, Box 3808, Duke
University Medical Center, Durham, NC 27710.
2 Department of Surgery, Division of Orthopedics, Duke University Medical
Center, Durham, NC.
Received March 14, 2004;
accepted after revision June 30, 2004.
Address correspondence to K. W. Harper.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. In our study population, we found that moderate to severe glenoid dysplasia is not a rare entity as seen on MRI, with an incidence of 14.3%. Furthermore, we found that there is a statistically significant increase in the incidence of posterior labral tears associated with shoulders with moderate or severe glenoid dysplasia compared with those shoulders with no dysplasia or mild dysplasia as evaluated on MRI. Even when the mild cases were included, the incidence of posterior labral tears was significantly increased versus healthy subjects, indicating that even mild cases of glenoid dysplasia may be clinically relevant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
Among our goals was to assess the incidence of the glenoid dysplasia as evaluated on MRI and to compare our findings with the established literature.
Findings of glenoid dysplasia have been described using radiography, CT, and more recently MRI [1, 2, 5, 6–9]. These abnormalities have been associated with shoulder instability [3, 4, 8–10] and are thought to predispose affected patients to accelerated degenerative joint disease [1, 6].
MRI has become the mainstay for the evaluation of most shoulder joint abnormalities. Using MRI for the evaluation of structural abnormalities of the shoulder, we have found that there is often an association between glenoid dysplasia, or deficiency of the inferoposterior glenoid rim, and posterior labral tears. The association between posterior glenoid deficiency and posterior labral tears has not, to our knowledge, been previously evaluated in the radiology literature. Our goal was to evaluate for glenoid dysplasia on shoulder MRI, to assess overall observed incidence, and to investigate its association with posterior labral tears.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Studies were grouped into four categories: normal, mild, moderate, and severe dysplasia (Fig. 2). Criteria for grading a case into one of the four categories include evaluation of the posteroinferior glenoid rim, excluding the most caudal section in which the glenoid cortex can be clearly seen; evaluation of the size and appearance of the posterior labral tissue; and evaluation of the presence or absence of posterior tilt to the glenoid articular surface. A case was defined as normal if there was preservation of the posterior glenoid rim, excluding the most caudal axial section; the posterior labrum was normal in size, without evidence of hypertrophy; and there was no significant posterior sloping of the articular surface of the glenoid (Fig. 3). Mild dysplasia was defined as rounding or truncation of the posteroinferior glenoid rim on one or two axial sections above the most caudal section (Fig. 4), but with reconstitution of the glenoid rim on more cephalad axial sections. The posterior labrum can look normal in size or mildly hypertrophied. Moderate dysplasia was defined as more significant rounding or truncation of the posteroinferior glenoid rim, extending several sections cephalad above the most caudal section, associated with hypertrophied low-signal posterior labral tissue (Fig. 5). Severe dysplasia was defined as marked rounding or truncation of the posteroinferior glenoid rim with either no reconstitution of the rim or reconstitution in only the most cephalad section. Severe cases should show markedly hypertrophied low- or mixed low- and intermediate-signal posterior labral tissue (Figs. 6A and 6B). In each category of dysplasia, the posterior slope of the glenoid articular surface when seen in conjunction with other findings of dysplasia was used as supporting evidence of the presence of posteroinferior glenoid dysplasia.
|
|
|
|
|
|
For purposes of statistical analysis, these groups were combined into two groups: a group of normal and mild cases, which we refer to as the "normal group," and a group of moderate and severe cases, which we refer to as the "dysplastic group." The observed incidence of glenoid dysplasia (both moderate and severe) was calculated. Observers experienced a perceived difficulty separating mild versus moderate cases, but felt a greater level of subjective comfort and consistency separating normal versus dysplastic glenoid. Therefore, we also included calculations that reclassify the mild cases into the dysplastic group with the moderate and severe cases. Studies were then evaluated for the presence of posterior labral tears. Criteria for the diagnosis of posterior labral tear necessitated visualization of joint fluid or intraarticular gadolinium extending either between the bony glenoid and posterior labrum or within the substance of the posterior labrum (Fig. 7), or the presence of a posterior paralabral cyst. The incidence of posterior labral tears was then calculated for both the normal group and the dysplastic group. The incidence of posterior labral tears was then calculated for the normal group and for the dysplastic group.
|
Fifteen cases of surgically proven posterior labral tears were then reviewed. Cases were all those with posterior labral tears diagnosed arthroscopically over a 5-year period by a single orthopedic surgeon specializing in sports medicine. MRI was analyzed by consensus of the three original observers. The incidence of moderate to severe glenoid dysplasia in patients with surgically proven posterior labral tears was then calculated.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Findings were also analyzed using normal cases versus all cases graded as having mild, moderate, or severe dysplasia. Thirty-nine (39.8%) of the 98 patients were thought to have mild, moderate, or severe dysplasia, whereas the remaining 59 (60.2%) were thought to have completely normal glenoids. In the normal category, five labral tears (8.5%) were identified, and in the dysplasia category, 13 labral tears (33.3%) were identified. This difference was also found to be statistically significant (p = 0.003).
Fifteen surgically proven cases of posterior labral tears were retrospectively analyzed for glenoid morphology by the original three observers. Of the 15 cases, 11 (73%) were graded as having moderate or severe posterior glenoid dysplasia.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Among the findings on imaging studies that have been used to define this entity is a dropping away of the normally flat plateau of the inferior glenoid articular surface [1]. This abnormality has been found to be identifiable on imaging studies such as CT and MRI. Weishaupt et al. [8] described both a delta (sharp posterior truncation of the angle of the inferior glenoid rim with a triangular or delta-shaped defect in the bony glenoid rim replaced by hypertrophied labrum or cartilage tissue) and a "lazy" J (rounded posteroinferior labrum with overlying hypertrophied labrum or cartilage) appearance of the bone deficiency as seen on CT. These findings correspond with the abnormal flat and convex glenoid rim shapes described by Inui et al. [10]. A similar appearance can be seen in healthy subjects at the most inferior aspect of the bony glenoid rim, possibly because of partial volume averaging with adjacent soft tissues, and must be distinguished from the spectrum of glenoid dysplasia. For this reason, in our study, the lower-most axial section of each pulse sequence was not used for evaluation.
Localized posteroinferior glenoid dysplasia has been linked to multidirectional instability of the shoulder [3, 4, 9, 10] and posterior instability [10]. Edelson [1] found skeletal changes of localized posteroinferior glenoid dysplasia in nine (75%) of 12 patients with symptomatic voluntary multidirectional shoulder instability, although he thought that coexisting ligamentous laxity also played a role in the shoulder instability in those patients. Weishaupt et al. [8] found that 93% of their patients with recurrent atraumatic posterior shoulder instability had a deficiency of the posteroinferior glenoid. These findings were seen in only 60% of patients with anterior instability, which was a statistically significant difference. They found, however, that it was the size of the defect—and not merely its presence—that was predictive of instability.
Several studies have attempted to quantify measurements of the bone deficiency or to measure glenoid version (abnormal slope or angle of the glenoid articular surface relative to the body of the scapula) as a predictor of instability. This has been used to successfully associate the degree of glenoid version with findings of instability [8–10]. Precise measurements and calculations may be less useful to the radiologist in everyday practice because the process of measuring and calculating angles is cumbersome. In addition, much of this quantitation was performed using CT, which better defines precise skeletal anatomy than MRI. Because MRI is considered the standard for evaluation of most shoulder abnormalities, we elected to use subjective criteria that could be used in day-to-day practice as our means of diagnosis.
In addition to the rounding or sloping of the posteroinferior glenoid rim seen in patients with glenoid dysplasia, a second imaging finding has been associated with this entity: cartilage and labral hypertrophy [1, 2, 5, 7]. This can be seen as a prominent triangular area of low or intermediate signal intensity bordering the posterior articular surface of the glenoid [5, 7].
A commonly accepted explanation is that this thickened material represents cartilage of the inferior glenoid that has failed to ossify [2, 4, 6, 7]. The glenoid rim normally develops from two ossification centers. The upper ossification center appears between ages of 9 and 15 years; the lower ossification center appears between ages of 12 and 16 years [4, 5, 7]. Most think that dysplasia of the posteroinferior glenoid results from failure of ossification of the inferior ossification center [2, 5, 7], leaving a bone defect and an abnormal, thickened cartilage cap in its place [7].
Glenoid dysplasia can be seen in patients across a wide range of ages. Children appear to be generally asymptomatic [6]. Affected individuals usually become symptomatic in the second or third decade of life [2, 4] and often after an increase in their level of activity [9]. A second peak in the fifth or sixth decade has been described [2]. Symptoms include pain, loss of range of motion, and instability [4, 5, 7].
Regarding posterior labral tears, it has been hypothesized that increased shear force during abnormal posterior movement of the humeral head may contribute to tears of the hypertrophied posterior cartilage and labrum [7]. In one of two cases of glenoid dysplasia reported by Munshi and Davidson [7], fissuring of the posterior soft tissues and extension of intraarticular contrast material into the abnormal soft tissues were noted on MR arthrography. Similarly, in another report, a patient with glenoid dysplasia was found to have contrast material extending into the substance of the hypertrophied posterior cartilage, consistent with ulceration [5]. We hypothesize that activities, such as bench-pressing heavy weights, that generate large shear forces on the posterior labrum and stabilizing soft tissues may predispose those patients affected by posterior glenoid deficiency to injury of the vulnerable posterior labrum.
In our study population, we found a high incidence (64.3%) of posterior labral tears in patients assessed by MRI to have moderate to severe posterior glenoid deficiency. This incidence is significantly higher than that of posterior labral tears in unaffected or mildly affected patients (10.7%). Even including the mild cases, the incidence of posterior labral tears relative to the healthy patient population remained much higher, with a 33.3% incidence in affected patients versus 8.5% in the unaffected population. Our data, therefore, indicate that affected patients are at higher risk of posterior labral tears.
Our limited observation of MRI arthrograms of 15 cases of surgically proven labral tears revealed a high incidence of moderate and severe cases of posterior glenoid dysplasia in that population (73%), lending further credence to the proposed correlation between the two entities. This study is only a preliminary look with a small sample size, but our findings further suggest the possible increased risk of posterior labral tears in patients with posterior glenoid dysplasia.
Limitations of this study include a fairly small sample size (n = 98). However, our results appear meaningful and warrant further study with a larger study population. Lack of surgical proof of posterior labral tears is also a limitation of our study, but complete data were not available in this population. Another limitation is the lack of surgical proof of glenoid dysplasia. During arthroscopy, surgeons neither look for nor note findings of glenoid dysplasia. Thus, surgical confirmation of glenoid dysplasia is not possible. The use of subjective criteria for evaluation of glenoid dysplasia also may limit reproducibility. This was thought a necessary evil because these findings will be subjectively evaluated in practice by most radiologists.
Bias may have been created in the observers by the heightened vigilance for the findings of glenoid dysplasia during the grading process. In an effort to avoid overdiagnosis, the original data were calculated to include mild cases with the normal group. Cases were graded for the purposes of statistical analysis, but because the findings of glenoid dysplasia can be seen in a spectrum, efforts to separate into different grades are somewhat difficult and manufactured. We found, however, that the most difficult distinction for the observers was in grading a case as mild versus moderate, and that decisions about grading a case as normal versus dysplastic were thought to be more consistent and certain. For this reason, data were also analyzed using the normal cases versus all cases with findings of dysplasia, although this analysis is also limited by artificial stratification. The incidence of dysplasia using this method appears too high when compared with the established literature. It is likely that there was observer bias and that the diagnostic threshold may perhaps need to be raised. However, even with these additional mild dysplasia cases included, the incidence of posterior labral tears remains significantly increased, indicating that perhaps even mild cases of dysplasia may have clinical significance.
Our retrospective review of surgically proven posterior labral tears is limited by a small sample size (n = 15). The observers reviewed the images with the knowledge of the presence of labral tears, and this could have created bias in the observers. It is only a preliminary observation that requires validation with a larger blinded study, but this observation further suggests the correlation between the spectrum of glenoid dysplasia and posterior labral tears.
In conclusion, authors of a number of previously published articles have described findings of radiography, CT, and MRI related to the various incarnations of glenoid dysplasia and have implicated this entity as associated with shoulder instability. However, to our knowledge, analysis of the relative incidence of posterior labral tears in affected patients to that seen in healthy subjects has not been reported.
We think that glenoid dysplasia is a clinically significant entity that can be recognized on routine shoulder MRI arthrograms. MRI grading of glenoid dysplasia is qualitative and subject to variation among observers; however, this variation appears to diminish as observers gain experience. The spectrum of glenoid dysplasia is more common than once thought. We hypothesize that thickened, hypertrophied labral or cartilaginous tissue is weaker and more likely to tear than a smaller, more normal labrum that is buttressed with bone. In our observations of our study population, patients with glenoid dysplasia as evaluated on MRI had a much higher incidence of posterior labral tears than those with normal glenoids.
Acknowledgments
We thank Lee Cothran for providing the illustrations.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  N. Shah and G. A. Tung Imaging Signs of Posterior Glenohumeral Instability Am. J. Roentgenol., March 1, 2009; 192(3): 730 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. B. Dewing, F. McCormick, S. J. Bell, D. J. Solomon, M. Stanley, T. B. Rooney, and M. T. Provencher An Analysis of Capsular Area in Patients With Anterior, Posterior, and Multidirectional Shoulder Instability Am. J. Sports Med., March 1, 2008; 36(3): 515 - 522. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Escobedo, M. L. Richardson, Y. B. E. Schulz, J. C. Hunter, J. R. Green III, and K. J. Messick Increased Risk of Posterior Glenoid Labrum Tears in Football Players Am. J. Roentgenol., January 1, 2007; 188(1): 193 - 197. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
