| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT |
Monday, May 1, 1:30 PM-3:30 PM
Abstracts 044-053 Moderators: Kenneth A. Buckwalter, MD and Patrick T. Liu, MD
1:30 PM
Keynote Address: MRI of the Shoulder with Arthroscopic Correlation
Patrick T. Liu, MD, Mayo Clinic, Scottsdale, AZ
1:50 PM
044. Rim-Rent Tears of the Rotator Cuff: A Common, and Easily Overlooked, Partial Rotator Cuff Tear
Vinson E.N.1*; Helms C.A.1; Higgins L.D.2; 1. Radiology, Duke University Medical Center, Durham, NC; 2. Orthopedic Surgery, Duke University Medical Center, Durham, NC.
Address correspondence to E.N. Vinson (envinson{at}hotmail.com)
Objective: Partial tears of the articular surface of the rotator cuff at its insertion (rim-rent tears), particularly those involving the anterior-most fibers of the supraspinatus tendon and those involving the infraspinatus tendon, are often overlooked on MRI. The purpose of this study is to determine the incidence of rim-rent tears and the sensitivity of MRI for their detection.
Materials and Methods: Two MSK-trained radiologists retrospectively reviewed 200 shoulder MR exams for the presence of cuff tears, and classified the tears as either full or partial thickness; of the partial tears, those of the rim-rent variety were noted and assessed for location. A separate cohort of 29 patients with surgically proven rim-rents were identified, and their preoperative MR exams and reports were reviewed.
Results: 117 cuff tears were identified in 101 (50.5%) of the 200 exams. 69 (59.0%) of the 117 tears were partial tears, and of these 49 (71.0%) were rim-rents. 24 (49.0%) of the rim-rents involved the anterior-most fibers of the supraspinatus, of which 16 (66.7%) were diagnosed prospectively. Of the 8 missed prospectively, 7 were in internal rotation. 24 (49.0%) of the rim-rents involved the infraspinatus, of which only 9 (37.5%) were diagnosed prospectively. Of the cohort of 29 patients with surgically proven rim-rents, 18 (62.1%) were prospectively diagnosed, and 28 (96.6%) were visible on MR in retrospect. 17 (60.7%) of these involved the anterior-most fibers of the supraspinatus, of which 12 (70.6%) were prospectively diagnosed. 9 (32.1%) of the 28 tears involved the infraspinatus, of which 4 (44.4%) were prospectively diagnosed.
Conclusion: Rim-rent tears are a common type of partial tear, comprising 71.0% of all partial tears in this series. Contrary to prior reports, the infraspinatus is a common site of rim-rents, with almost half of the rim-rents involving the infraspinatus in this series. These tears are often overlooked- only 44.4% were diagnosed prospectively in our cohort of surgically proven cases- possibly because they were previously felt to be uncommon. The anterior-most fibers of the supraspinatus tendon are also commonly involved, and these tears are also often overlooked on MRI, possibly due to internal rotation. Inspecting the articular surface of the distal infraspinatus tendon and the anterior fibers of the supraspinatus tendon, immediately lateral to the long head of the biceps tendon on coronal oblique images, should improve sensitivity for the detection of these common tears.
045. Magnetic Resonance Neurography in Patients with Upper Extremity Peripheral Neuropathy
Malfair D.*; Steinbach L.S.; Engstrom J.W.; Chin C.T.; Radiology, University of California, San Francisco, San Francisco, CA.
Address correspondence to D. Malfair (david.malfair{at}radiology.ucsf.edu)
Objective: Evaluation of peripheral nerve disease relies on clinical examination and electrodiagnostic studies and can be limited. MRN is a relatively new technique dedicated to imaging of peripheral nerves. This study demonstrates the utility of magnetic resonance neurography (MRN) in patients diagnosed with peripheral nerve dysfunction in the upper extremity and confirmed by electromyography (EMG).
Materials and Methods: MR neurograms of 14 patients were performed between July 2000 and October 2005 to investigate upper extremity peripheral nerve dysfunction. 1.5T axial T1, STIR, TSE T2 with fat saturation sequences were performed in all patients. Additional axial 3D DESS and post-gadolinum fat saturation sequences were also performed in 5 selected patients. The presence or absence of nerve signal abnormality and enlargement were recorded as well as the site and extent of signal abnormality. The findings were correlated with EMG data and clinical history.
Results: All patients were diagnosed with upper extremity peripheral nerve dysfunction based on physical exam and EMG testing. Thirteen of the fourteen patients demonstrated MRN abnormalities. Etiologies of the neuropathy were varied and included acute trauma (3), chronic compression (4), diabetic neuropathy (1) and idiopathic (5). Two cases demonstrated abnormalities in more than one nerve. Abnormalities were found in the ulnar, (8) radial (4) and median (3) nerves. MRN characteristics of the abnormal nerves included increased T2 weighted signal (15/15), increased size (5/15) and enhancement (5/5). The location of abnormality on MRN correlated with the abnormal EMG in 12 of 13 patients.
Conclusion: MRN is a sensitive tool for imaging of peripheral neuropathy. It compliments EMG by precisely localizing the area and extent of the abnormal nerve.
046. Overhead Throwing Athletes: Posterior Capsule Thickening on MR Arthrography
Petersen B.D.*; Tuite M.J.; Musculoskeletal Radiology, University of Wisconsin, Madison, WI.
Address correspondence to B.D. Petersen (bd.petersen{at}hosp.wisc.edu)
Objective: Overhead throwing athletes with shoulder pain may have internal impingement, which is associated with posterosuperior labral tears, posterior undersurface ro-tator cuff tears, and posterior humeral head cysts. They may also have a tight posterior capsule that predisposes to subtle anterior translation of the humeral head and contributes to the anterior instability seen in the extreme position of abduction and external rotation experienced by overhead athletes. This anterior subluxation is a possible cause of internal impingement. Our purpose was to determine if the posterior capsule in throwing athletes is thickened on MR arthrogram images.
Materials and Methods: We performed a retrospective review of all patients who had a shoulder MR arthrogram followed by shoulder arthroscopy over a 5-year period. 26 patients had a clinical diagnosis of shoulder pain while performing overhead athletic activity, with probable internal impingement. The control group was randomly selected from the 5-year study period and included 26 individuals who were not overhead athletes and whose MR arthrogram was prospectively interpreted as normal. The thickness of the most medial aspect of the capsule near the labral insertion was made on an axial T1 image through the 8 o'clock glenoid position. The mean capsular thickness was compared between the two groups using a student t-test.
Results: The mean capsular thickness of the throwing athletes was 2.9 mm (standard deviation 1.2 mm), while in the controls it measured 1.9 mm (standard deviation. 0.6 mm; p > 0.05).
Conclusion: There is a trend toward increased thickness of the posterior capsule in throwing athletes who go onto arthroscopy. A thickened capsule may be an MR arthrogram appearance of a tight posterior capsule as seen in some patients with internal impingement.
047. Error Classification in MR Imaging of the Rotator Cuff
Callaghan E.B.*; Bennett D.L.; Ohashi K.; El-Khoury G.Y.; Radiology, University of Iowa Carver College of Medicine, Iowa City, IA.
Address correspondence to E.B. Callaghan (eric-callaghan{at}uiowa.edu)
Objective: To classify error in interpretation of MR examinations of rotator cuff tears using a previously developed classification system of radiological errors. To our knowledge, errors in shoulder MRI interpretation have not been rigorously evaluated using a standard error classification scheme. Classification of error types may aid radiologists in identifying and correcting errors in interpreting shoulder MRI exams.
Materials and Methods: 162 patients with a preoperative MRI followed by shoulder surgery within 3 months were chronologically, retrospectively collected (08/2001 to 08/2003). Reports for both procedures were reviewed. The surgical results for each patient were simplified into three categories: full thickness tear, partial thickness tear, or no tear. Any discrepancies between the radiology and operative reports were divided into `major' and `minor' errors based upon disagreement between the presence of a tear (major) or extent of a tear (minor). Two experienced musculoskeletal radiologists, blind to the initial reports, independently reviewed the MRI exams of those subjects who had discrepancies with surgery. These reinterpretations were compared to the initial clinical reports to identify the sources of error. Sources of error were classified according to a previously reported system consisting of: complacency, faulty reasoning, lack of knowledge, underreading, poor communication (of abnormal findings), miscellaneous, or complications (Renfrew et al. Radiology, 1992).
Results: For full thickness tears, MRI sensitivity, specificity, pre-test probability, PPV, and NPV were 86%, 92%, 34%, 85%, 93%, respectively; partial thickness tears, 62%, 90%, 16%, 55%, 92%; and for any tear 88%, 91%, 52%, 91%, 88%. 28/162 subjects (17.2%) had discrepancies (17 major, 11 minor). The review of these 28 studies showed that the 17 `major' errors were attributed to complacency (23.5%), faulty reasoning (11.8%), underreading (35.3%), and miscellaneous (29.4%). Specific errors included: mistaking subacromial/subdeltoid fluid for a tear, mistakenly calling a tear when it is only seen on one sequence, metal artifact, and underreading.
Conclusion: The most common type of error was underreading. The majority of errors that we encountered were perceptual, which can be caught occasionally in the clinical setting if one focuses on common perceptual errors. No errors were caused by lack of knowledge, poor communication, or complications; this is different from previously reported error studies of other imaging modalities.
048. Changes in Pennation Angle Measurements with Supraspinatus Muscle Tendon Tears and Muscle Atrophy
Fillmore K.C.*; Schweitzer M.E.; Chang E.; Radiology, New York University Medical Center, New York, NY.
Address correspondence to K.C. Fillmore (fillmk01{at}med.nyu.edu)
Objective: Pennate muscles, such as the supraspinatus muscle (SS), are arranged so that the muscle fascicles form a common angle with the tendon known as the pennation angle (PA). We sought to study the PA of the SS by MR, in particular to see how it is affected by the different stages of impingement, as well as how it changes with two different models of atrophy.
Materials and Methods: At 1.5T the PA of the SS was measured in the coronal and axial planes in 137 patients; 28 with shoulder instability, 22 with subacromial impingement without cuff tear, 31 with SS tears, 42 with cuff tears and atrophy and 14 with neurogenic atrophy. PA measurements were obtained by two different readers on the same axial and coronal sequences.
Results: Patients without supraspinatus tears, interestingly either with instability or early impingement had similar mean PA 19.6/17.6 vs. 18.0/19.2 in the axial/coronal planes. Patients with supraspinatus tears had a mean PA of 18.8/14.9. Patients with tears and atrophy had angles 16.4/14.4 of which was similar to those with neurogenic atrophy (15.9/14.1).
Conclusion: PA measurements suggest that the muscle effects of impingement are related to stage only in so far as only whether atrophy is present or not. Patients with tear-related atrophy and neurogenic atrophy showed similar pennation angles suggesting that both models of atrophy result in similar changes in muscle.
049. Little Leaguer's Shoulder "Proximal Humeral Epiphyseolysis": MR Findings in Three Cases
Gaskin C.M.; Anderson M.W.; Taffoni M.J.*; Musculoskeletal Radiology, University of Virginia, Charlottesville, VA.
Address correspondence to M.J. Taffoni (taffonimd{at}gmail.com)
Objective: To describe the MRI findings in patients with the diagnosis of Little Leaguer's Shoulder. This entity is well described in the sports medicine literature and the plain film findings have been reported. To our knowledge, the MRI findings have never been described in the radiological literature.
Materials and Methods: Three patients with Little Leaguer's Shoulder, who underwent MRI, were reviewed. Specifically, this included history and physical examination, bilateral internal and external radiographs, and MRI of the involved shoulder. The MRI examinations were interpreted by two musculoskeletal radiologists.
Results: Two of the three patients were pitchers and the third was a tennis player. All subjects were male, with average age of 13 years. All complained of pain localized to the proximal humerus. All three patients demonstrated unilateral lateral physeal widening on plain radiographic examination. The MR exam in all three patients revealed abnormal high T2 and low T1 signal only along the lateral metaphyseal side of the physis.
Conclusion: MR examination is useful in the diagnosis of Little Leaguer's shoulder. It has a characteristic appearance that should be recognized. MRI may yield the appropriate diagnosis earlier than plain film findings, while excluding other causes of shoulder pain. This diagnosis has long been called "epiphysiolysis", although the MRI and plain film findings reflect only metaphyseal resorption along the physis.
050. MR Arthrographic Analysis of SLAP Lesions with an Emphasis on Accompanying Various Shoulder Abnormalities
Choi S.1*; Chun K.1; Kim K.; Kwon O.2; 1. Radiology, Uijeongbu Saint Mary's Hospital, Uijeongbu, South Korea; 2. Orthopedics, Uijeongbu Saint Mary's Hospital, Uijeongbu, South Korea.
Address correspondence to S. Choi (Medmath{at}hanmail.net)
Objective: To describe the pattern of various shoulder abnormalities with associated SLAP lesion using MR arthrography and to assess their clinical significance.
Materials and Methods: A retrospective review of SLAP lesions which was confirmed arthroscopically was performed from August 2003 to July 2005. Of 537 MR arthrography of the shoulder with arthroscopic procedure, 92 cases with SLAP lesion (only type 2 SLAP lesions are considered because type 1 is regarded as normal variant in degenerative shoulder, and type 3 and type 4 are only a small number of them) were identified. Imaging of MR arthrography, medical record and operative notes were reviewed and analyzed. MR arthrographic analysis included rotator cuff abnormality, acromioclavicular arthritis, adhesive capsulitis, glenohumeral arthritis, labral abnormality besides SLAP lesion, and paralabral cyst. We divided the patients with SLAP lesions into two age groups: those older than the age of forty and those forty years old or younger. Statistical analysis was performed to evaluate age influence for the various shoulder abnormalities with associated SLAP lesion.
Results: The prevalence of type 2 SLAP lesion was 17 % (92/537). Of the 92 type 2 SLAP lesions, isolated SLAP lesions with no associated any shoulder abnormality were 8 cases (8/92, 8%). Eighty four (91%) SLAP lesions were associated with various shoulder abnormalities including rotator cuff tendinosis (30/92, 32%), partial-thickness tear (36/92, 39 %), full-thickness tear (2/92, 2 %), acromioclavicular arthri-tis (46/92, 50 %), adhesive capsulitis (7/92, 7 %), glenohumeral arthritis (15/92, 16 %), labral abnormality (26/92, 28 %) and paralabral cyst (7/92, 7 %). A history of trauma was found in 19 patients. The SLAP lesions (60/92, 65%) in patients older than forty years of age significantly accompanied a rotator cuff abnormality (p < 0.001), glenohumeral osteoarthritis (p = 0.001), arcomioclavicular osteoarthritis (p < 0.001). In contrast, SLAP lesions (32/92, 35%) in patients with forty years old or younger significantly accompanied anterior or posterior labral lesions (p < 0.001).
Conclusion: Isolated SLAP lesions with no associated any shoulder abnormalities are uncommon. The patient age influences the prevalence of various shoulder abnormalities with associated SLAP lesions. MR arthrography is helpful for detection of accompanying shoulder abnormality with associated SLAP lesions.
051. Direct and Indirect MR Arthrography of the Glenohumeral Joint on an Open 0.3 Tesla System Using a Modified Three-point Dixon Sequence
Kim J.1*; Zoga A.C.1; Austin A.F.1; Parker L.1; Mamelak J.D.1,2; Morrison W.B.1*; 1. Radiology, Thomas Jefferson University Hospital, Philadelphia, PA; 2. Radiology, William Osler Health Center, Brampton, Ontario, Canada.
Address correspondence to J. Kim (jxk009{at}gmail.com)
Objective: Assess the use of MR arthrography of the glenohumeral joint for labral tear on an open 0.3T system using a modified 3-point Dixon sequence.
Materials and Methods: 52 patients with clinical suspicion for glenoid labral tear who were not candidates for imaging on a closed MR system had direct or indirect MR arthrography on an open 0.3T system. Standard MR arthrographic protocol was used in all patients with the addition of one modified 3-point Dixon water/fat suppression sequence, acquired in the plane felt most likely to be helpful based on clinical exam (23 coronal, 29 axial). MRIs were reviewed individually by 2 MSK radiologists at 2 sittings each, first using conventional arthrographic sequences only and the second using all sequences including the Dixon. At each sitting, labral tears were recorded and localized, and a confidence score was assigned to labral assessment as torn or not torn with 1 representing absolutely no tear and 5 representing definite tear. Readers also gave an overall score assessing the utility of the Dixon sequence on a 5-point scale with 0 as "did not help". Concordance of findings between readers was compared, and utility of the Dixon sequence in MR arthrography was explored on the basis of logged reader responses using the paired t-test and Wilcoxon signed rank test.
Results: Statistical methods revealed a mean confidence level increase for both readers using the Dixon sequences compared to standard protocol alone and that this change was significant, p < 0.005. For presence or absence of labral tear, mean improvements in confidence of 0.81 and 0.31 (reader 1 and 2) were recorded with the addition of the Dixon sequence. Overall, readers found the addition of the 3-point Dixon sequence helpful in establishing a diagnosis of labral tear in 46/52 and 32/52 cases, with an overall mean score of 2.8 on a 5-point scale. Addition of a Dixon sequence changed the diagnosis in 7 and 8 cases respectively. Changes in diagnosis with Dixon occurred 12 times with direct arthrography and only 4 on indirect, and occurred in 6 cases with an axial Dixon and 10 with coronal.
Conclusion: Adequate and reproducible demonstration of the glenoid labrum is possible using direct and indirect MR arthrography on open, low-field strength systems. When performing MR arthrography on an open, low-field strength unit, the 3-point Dixon sequence is useful for assessing the glenoid labrum. For practical purposes, if only one Dixon sequence can be acquired, coronal oblique images are recommended.
052. Sensitivity and Specificity in Detection of Labral Tears with 3.0 Tesla MR Imaging of the Shoulder
Magee T.H.*; Williams D.S.; 1. Radiology, Neuroskeletal Imaging, Merritt Island, FL.
Address correspondence to T.H. Magee (tmageerad{at}cfl.rr.com)
Objective: MR imaging of the shoulder has been found to be sensitive and specific for detection of labral tears at 1.5 Tesla or lower field strength as compared with arthroscopy. Three Tesla MR imaging of the shoulder has not been specifically assessed. This study assesses the sensitivity and specificity of MR imaging at 3.0 Tesla for labral tears as compared with arthroscopy.
Materials and Methods: Two experienced musculoskeletal radiologists retrospectively reviewed MR images of the shoulder in 100 consecutive patients. All patients had a proton density fat saturated axial, oblique coronal T1 and fat saturated intermediate weighted oblique coronal and oblique sagittal images performed. Any MR arthrograms performed in addition to conventional MR examinations were not included in the retrospective consensus review. The radiologists read the MR images by consensus without knowledge of arthroscopy results. Scans were read as to whether there were superior labral anterior to posterior (SLAP) tears, anterior or posterior labral tears. Sixty seven of the 100 patients went on to arthroscopy. After consensus review of MR images, arthroscopy results were compared with consensus MR interpretations.
Results: Of the sixty seven patients who went on to arthroscopy, twenty-one had SLAP tears. Eighteen of the sixty seven had anterior labral tears and seven of the sixty seven had posterior labral tears. Nineteen out of the twenty one patients with SLAP tears at arthroscopy were seen on prospective MR reading. Sixteen of the eighteen anterior labral tears and six of the seven posterior labral tears seen at arthroscopy were seen on prospective MR reading.
Conclusion: 3.0 Tesla MR imaging of the shoulder is very sensitive and specific as compared with arthroscopy in detection of superior, anterior and posterior labral tears.
053. Humeral Head Cysts: Location and Association with Pathology
Vinson E.N.*; Helms C.A.; Subhas N.; Radiology, Duke University Medical Center, Durham, NC.
Address correspondence to E.N. Vinson (envinson{at}hotmail.com)
Objective: Small cysts in the humeral head are often incidentally noted on shoulder MRI examinations. These cysts are commonly seen to occur in association with rotator cuff and/or labral pathology. The purpose of this study is to evaluate for a possible correlation between the location of humeral head cystic lesions and specific pathology of the rotator cuff and posterior labrum.
Materials and Methods: Two musculoskeletal-trained radiologists retrospectively reviewed 200 shoulder MRI examinations for the presence of humeral head cysts, and classified those cysts as to one of four locations in the humeral head on axial images: lesser tuberosity, anterior greater tuberosity, posterior greater tuberosity, or bare area. Associated pathology in the rotator cuff (tendinopathy, partial tear, or full thickness tear) and posterior labrum (irregularity, tear, or detachment) was noted in each case.
Results: 116 (58.0%) of the 200 shoulder MRI examinations demonstrated cysts in at least one of the described locations, and many had cysts in multiple locations. 6 (3.0%) examinations had cysts at the lesser tuberosity, and all 6 (100%) of these had associated subscapularis pathology. 26 (13.0%) of the studies demonstrated cysts at the anterior aspect of the greater tuberosity, and 23 (88.5%) of these had associated supraspinatus pathology. 63 (31.5%) of the studies demonstrated cysts at the posterior aspect of the greater tuberosity, and 43 (68.3%) of these had associated infraspinatus pathology. 58 (29.0%) of the studies demonstrated cysts at the bare area, and 14 (24.1%) of these had associated posterior labral pathology.
Conclusion: Humeral head cysts are commonly identified on shoulder MRI examinations, and the specific locations of the cysts often correlate with rotator cuff and/or labral pathology; this is particularly true of cysts located near the insertions of the subscapularis and supraspinatus tendons.
* Will present paper
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
