| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Diagnostic Radiology and Organ Imaging, The Chinese University
of Hong Kong, Prince of Wales Hospital, 30-32 Ngan Shing St., Shatin, Hong
Kong SAR, China.
2 Department of Orthopaedics and Traumatology, The Chinese University of Hong
Kong, Prince of Wales Hospital, Shatin, Hong Kong SAR, China.
Received April 27, 2007;
accepted after revision June 1, 2007.
Address correspondence to J. F. Griffith.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Fifty patients (45 males, five females; mean age, 28.7 years; age range, 14–56 years) with anterior shoulder dislocation underwent shoulder CT examination before arthroscopy (mean time interval between CT and arthroscopy, 28.5 days; range, 9–73 days). Thirteen (26%) of the 50 patients had a single dislocation, whereas the remaining 37 patients (74%) had recurrent dislocation (mean, 8.2 dislocations; range, 2–50 dislocations).
RESULTS. Glenoid bone loss was evident in 41 (82%) of the 50 patients at arthroscopy. Compared with arthroscopy, CT had a sensitivity in detecting glenoid bone loss of 92.7%; specificity, 77.8%; positive predictive value, 95.0%; and negative predictive value, 70.0%. Three false-negative CT assessments had 5%, 10%, and 10% glenoid bone loss, respectively, at arthroscopy. Two false-positive CT assessments had 8.7% and 5.7% glenoid bone loss on CT, although no bone loss was apparent at arthroscopy. There was a strong correlation between CT and arthroscopy with respect to the severity of glenoid bone loss (r = 0.79, 95% CI = 0.659–0.877, p < 0.0001).
CONCLUSION. CT has both a high sensitivity and a high specificity for detecting glenoid bone loss, and agreement with arthroscopy regarding the severity of glenoid bone loss is good. CT can be used to assess glenoid bone loss and the need for bone augmentation surgery.
Keywords: arthroscopy • bone loss • CT • glenoid bone • shoulder • shoulder dislocation • sports medicine
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Arthroscopic assessment of glenoid bone loss is based on measuring the distance from the glenoid bare spot to the anterior and posterior glenoid rims (Figs. 1A, 1B, 1C, 2A, and 2B). CT assessment of glenoid bone loss is based on comparison of glenoid width on the affected shoulder with glenoid width on the contralateral normal shoulder in subjects with unilateral dislocation (Figs. 3A and 3B). No study, to our knowledge, has compared the accuracy of CT in predicting glenoid bone loss using arthroscopy as a comparative standard. This study compared the accuracy of CT in determining the presence and severity of glenoid bone loss in patients with unilateral shoulder dislocation.
|
|
|
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CT Examination Technique and Analysis
Each patient underwent simultaneous CT examination of both shoulders with
his or her arms positioned by the chest wall on an MDCT scanner (LightSpeed 16
Plus, GE Healthcare) that used 16 x 0.625 mm acquisitions with 400 mA,
120 kV, and a pitch of 1:1.75. The scanning plane extended from the acromion
to just below the glenoid. Double oblique reconstruction of each glenoid was
used to obtain oblique sagittal images en face to the glenoid
articular surface (Advantage Windows, version 4.2, GE Healthcare) (Figs.
3A and
3B). On this image, a line was
drawn along the long axis of the glenoid (Figs.
3A and
3B). The width of the glenoid
was measured at right angles to this long axis though the midportion of the
inferior glenoid (Figs. 3A and
3B). The presence of an
anterior straight line along the anterior glenoid rim was noted and measured
(Figs. 3A and
3B). A single investigator
performed all CT measurements and assessments unblinded to clinical
information.
Glenoid bone loss was diagnosed provided two criteria were present—namely, an anterior straight line to the glenoid rim on the image obtained en face to the glenoid articular margin and a relative reduction in glenoid width of the dislocating shoulder compared with the normal side [2]. The anterior margin of the normal glenoid has a curved contour [2]. Any straightening of that counter was deemed to represent an anterior straight line [2]. An anterior straight line or relative reduction in glenoid width in isolation was not deemed sufficient to constitute glenoid bone loss. The percentage of glenoid bone loss was calculated as the difference in glenoid width compared with the width of the normal nondislocating glenoid and was expressed to the nearest first decimal point.
Arthroscopic Assessment
General anesthetic was used for all arthroscopies. With the patient under
anesthesia and supine, the shoulder was examined for anterior, posterior,
inferior, and multidirectional instability. The patient was then turned to the
lateral decubitus position, with the trunk tilting backward 30°, the arm
was flexed to 20° and abducted to 45° under traction with a weight of
3–4 kg. Three arthroscopic portals
[7]—namely, the
posterior, anterosuperior, and anteroinferior and mid glenoid
portals—were made, and diagnostic arthroscopy was performed using a 4-mm
arthroscope. Via the posterior portal, an arthroscopic probe (calibrated in
1-mm increments) was used to determine and quantify the degree of anterior
glenoid bone loss with reference to the central bare spot of the glenoid
(Figs. 1A,
1B,
1C,
2A, and
2B). The tip of the probe was
first placed against the posterior glenoid margin, then against the bare spot
in the middle of the inferior glenoid, and finally against the anterior
glenoid margin. Arthroscopic measurements of glenoid bone loss were reported
in intervals of 5% (e.g., 5%, 10%, 15% bone loss).
Statistical Analysis
SPSS software (version 14.0, SPSS) for Windows (Microsoft) was used for
statistical analyses. Variables were expressed as mean and range or mean
± SD, as appropriate. The sensitivity, specificity, and positive and
negative predictive values of CT compared with arthroscopy in the detection of
glenoid bone loss were calculated. The paired Student's t test was
used to test for difference in mean glenoid bone loss measured by CT and
arthroscopy. Pearson's correlation coefficient was used to examine the
correlation between CT and arthroscopy in quantifying the percentage of
glenoid bone loss.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
There were three false-negative CT assessments. These patients did not meet the criteria for glenoid bone loss on CT examination, although arthroscopy showed 5%, 10%, and 10% glenoid bone loss, respectively. There were two false-positive CT assessments. In those cases, CT revealed 8.7% and 5.7% glenoid bone loss, although no bone loss was apparent at arthroscopy.
No significant difference was found between mean glenoid bone loss predicted by CT (11.0% ± 8.1%; range, 0–33.0%) or by arthroscopy (12.3% ± 8.8%; range, 0–40%) (p = 0.17). There was a strong correlation (r = 0.79) between CT and arthroscopy with respect to assessments of the severity of glenoid bone loss (r = 0.79, 95% CI = 0.659–0.877, p < 0.0001) (Fig. 4).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CT is a reliable means of quantifying glenoid bone loss [2, 3, 8]. CT allows both glenoids to be examined simultaneously. CT assessment of glenoid bone loss rests on comparing the width of the glenoid in the dislocating shoulder with that of the nondislocating shoulder in patients with unilateral dislocation. Because a slight side-to-side variation in glenoid width is normal and because the curved anterior glenoid margin may occasionally be straight in subjects without dislocation, two criteria should be fulfilled before glenoid bone loss is diagnosed on CT examination [2]. First, on a reconstructed image en face to the glenoid surface, an anterior straight line should be present, and, second, a relative reduction in glenoid width compared with the normal contralateral side should be present.
It follows that CT cannot be used to quantify glenoid bone loss in subjects with bilateral shoulder dislocation. In this setting, the shape of the glenoid will allow one to semiquantitatively categorize glenoid bone loss as absent, mild, moderate, or severe, although the percentage of bone loss cannot be accurately quantified because there is no normal glenoid for comparison [2]. As a result, we included only patients with unilateral dislocation in this study.
Sugaya et al. [3] described a method in which a best-fit circle is applied to the inferior glenoid on 3D reconstructed CT images. Those investigators used size of the bone fragment, if present, as a quantitative measure of bone loss and nonquantitative visual comparison with the contralateral side if no fragment was present [3]. No quantitative comparison between arthroscopic and CT evaluations was undertaken [3].
Itoi et al. [9] created osseous glenoid defects on cadaveric scapula, recording scapular length and width and the force required to induce dislocation. An osseous defect resulting in a glenoid width of less than 21% of the glenoid length produced instability. A limitation of that study may be that these defects were created on the anteroinferior margin of the glenoid, whereas CT-based studies have shown that bone loss occurs principally on the anterior aspect rather than on the anteroinferior aspect of the glenoid [2, 8].
Other than CT, the method of quantifying glenoid bone loss is arthroscopy [6]. Arthroscopic assessment involves identification of the bare spot, an area where the cartilage is thinned or absent in the center of the inferior glenoid [5]. Using a calibrated probe inserted through a posterior portal, the distance from the posterior glenoid margin to the bare spot should be equal to the distance from the bare spot to the anterior glenoid margin if no bone loss is present [6]. Reduction in the anterior margin-to-bare spot distance relative to the posterior margin-to-bare spot distance is an indication and measure of glenoid bone loss [6].
Glenoid bone loss is common in patients with anterior shoulder dislocation, particularly in those with recurrent dislocation [2]. Most of the patients in this study had recurrent shoulder dislocation. CT had a high sensitivity and specificity for the detection of glenoid bone loss. We found that there was a strong correlation (r = 0.79) between CT and arthroscopy in quantifying the severity of glenoid bone loss. The CT technique used has been shown to have good inter- and intraobserver variability [2] and is quick to evaluate with reconstruction and analysis taking 5–10 minutes.
The main limitations in using CT for assessment of glenoid bone loss are threefold. First, additional radiation exposure is incurred, but this exposure can be minimized by limiting the scanning plane to include both glenoids only. Second, CT assessment possesses an inherent error because of the known side-to-side variation in normal glenoid width. However, this error has been shown to be small [2], and as shown in this study, the correlation with arthroscopy is good. Third, extra time, cost, and effort are needed to include a CT examination in the imaging algorithm for shoulder instability, which typically comprises radiography and MRI [10]. Further study may enable selection of patients most likely to benefit from the additional CT examination based on appearances at MR examination.
Arthroscopic assessment of glenoid bone loss was used as the gold standard in this study. Other than CT, arthroscopy is the only known method of quantifying glenoid bone loss. Arthroscopy is not the ideal gold standard for several reasons. First, the bare spot may occasionally comprise a bare area rather than a discrete bare spot. Second, the calibrated probe inserted via the posterior portal may not always be aligned at right angles to the long axis of the glenoid. Alignment away from the true short axis of the glenoid tends to underestimate bone loss. Third, the millimeter-spaced lines along the calibration probe are counted visually via a second scope inserted through an anterosuperior portal. In practice, these finely spaced lines can be difficult to count arthroscopically, so quantifying the exact degree of glenoid loss can be difficult even with direct inspection [1]. As a result, arthroscopic measurements tend to be approximations, with descriptions as mild, moderate, or severe or, as in this study, with measurements reported in intervals of 5% (e.g., 5%, 10%, 15% bone loss) [1]. Fourth, a cadaveric study has shown that the bare area does not consistently lie in the center of the glenoid and tends to lie closer to the anterior rather than the posterior glenoid margin [11]. The mean difference between the anterior-to-posterior glenoid widths with the bare spot as a reference was 4.2 mm (range, 0.9–8.9 mm). This tends to lead to overestimation of bone loss and has prompted Kralinger et al. [11] to recommend preoperative CT rather than arthroscopy to assess glenoid bone loss in patients with shoulder dislocation.
There are two main limitations to our study. First, although all arthroscopic measurements were reported independently of the CT results, the surgeon performing arthroscopy was aware of the CT results at the time of arthroscopy. Preoperative quantitative assessment of glenoid bone loss is now standard practice in our hospital, so withholding that information from the surgeon during preoperative planning was not justifiable. Second, the average time lag between CT examination and arthroscopy was 28.5 days, with a maximum of 73 days. Whether any further dislocations occurred during this interval was not documented.
In conclusion, CT has a high sensitivity and specificity for detecting the presence of glenoid bone loss evident at arthroscopy. There is also good agreement (r = 0.79) between CT and arthroscopy regarding the severity of glenoid bone loss.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Diederichs, H. Seim, H. Meyer, A. S. Issever, T. M. Link, R. J. Schroder, and M. Scheibel CT-Based Patient-Specific Modeling of Glenoid Rim Defects: A Feasibility Study Am. J. Roentgenol., November 1, 2008; 191(5): 1406 - 1411. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Griffith, G. E. Antonio, P. S. H. Yung, E. M. C. Wong, A. B. Yu, A. T. Ahuja, and K. M. Chan Prevalence, Pattern, and Spectrum of Glenoid Bone Loss in Anterior Shoulder Dislocation: CT Analysis of 218 Patients Am. J. Roentgenol., May 1, 2008; 190(5): 1247 - 1254. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
