3.0T常规MRI与MR髋关节造影诊断髋臼唇撕裂的对比研究

目的 在3.0 T场强中,比较常规髋关节MRI及MR髋关节造影对髋臼唇撕裂的诊断价值.方法 回顾分析44例髋关节病变患者的患侧髋关节常规MRI及MR髋关节造影资料,将每例患者的臼唇划分为前、上、后3处区域(共计132处),确定有无撕裂,并进行分型.其中5例患者经髋关节镜检查.对于常规MRI及MR髋关节造影获得的臼唇撕裂及分型数据差异比较采用Wilcoxon秩和检验,一致性比较采用Kappa检验.结果 常规MRI与MR髋关节造影诊断完全相同者计116处臼唇,只有16处存在诊断差异.其中,前者诊断无撕裂而后者诊断撕裂者9处,前者诊断撕裂但后者诊断无撕裂者6处,其余1处两者均诊断为撕裂但分型不同.常规MRI与MR髋关节造影的评价差异没有统计学意义(Z=0.347,P＞0.05),且具有极好的一致性(K=0.781,P＜0.01).在接受关节镜的5例患者中,常规MRI、MR髋关节造影及关节镜结果均完全吻合.结论 对髋臼唇撕裂,3.0 T常规髋关节MRI可获得与MR髋关节造影基本相同的诊断效果.     

MR arthrographic appearance of the postoperative acetabular labrum in patients with suspected recurrent labral tears

OBJECTIVE: The objective of our study was to evaluate the MR arthrographic appearance of the acetabular labrum in patients with a suspected recurrent acetabular labral tear after previous arthroscopic resection of a labral tear. CONCLUSION: The labrum after excision will appear shortened on MR arthrography. A recurrent labral tear can be diagnosed by the identification on MR arthrography of a new line to the labral surface, an enlarged and distorted labrum, or a new paralabral cyst.     

Sensitivity of MR arthrography in the evaluation of acetabular labral tears

OBJECTIVE: MRI has historically provided suboptimal visualization of tears of the acetabular labrum. Degenerative fraying and underlying cartilage abnormalities can often mimic tears of the labrum on conventional MRI. Administration of intraarticular gadolinium enhances the MRI appearance of the labrum to improve detection of labral abnormalities. This study examined the improved diagnostic sensitivity of MR arthrography compared with conventional MRI and the importance of confining the study to a small field of view. MATERIALS AND METHODS: Fifty-one hips were imaged in 48 patients. Fourteen hips underwent conventional MRI with a large field of view (30-38 cm). Seven hips underwent conventional MRI with a small field of view (14-20 cm). Thirty hips underwent MR arthrography with a small field of view (14-20 cm). Labral tears were diagnosed when contrast material was identified within the labrum or between the labrum and the acetabulum, when a displaced fragment was noted, or when a paralabral cyst was identified. All study results were compared with findings at the time of hip arthroscopy. RESULTS: Conventional MRI with a large field of view was 8% sensitive in detecting labral tears compared with findings at the time of arthroscopy. Diagnostic sensitivity was improved to 25% with a small field of view. MR arthrography with a small field of view was 92% sensitive in detecting labral tears. CONCLUSION: A combination of MR arthrography and a small field of view is more sensitive in detecting labral abnormalities than is conventional MRI with either a large or a small field of view.     

Evaluation of radial-sequence imaging in detecting acetabular labral tears at hip MR arthrography

Objective In recent years, radial imaging has been advocated for improved visualization of the acetabular labrum in magnetic resonance arthrography of the hip. The purpose of this study was to investigate whether radial imaging demonstrates labral tears not visible on standard imaging planes. Methods Fifty-four consecutive magnetic resonance (MR) arthrograms of the hip that included radial imaging over 2 years were retrospectively analyzed by two radiologists. Standard imaging planes and radial imaging were reviewed for identification of labral tears in four specific areas of the labrum: anterosuperior, posterosuperior, anteroinferior, and posteroinferior. The standard imaging sequences include fat-saturated spin-echo T1-weighted images in the coronal and oblique axial planes, non-fat-saturated T1-weighted images in the coronal and sagittal planes, and T2-weighted sequence in the axial plane. Radial imaging was performed as previously described using fat-saturated T1-weighted sequences. Results Using standard imaging planes, 50 anterosuperior, 31 posterosuperior, 10 anteroinferior, and 9 posteroinferior labral tears were detected in 54 MR arthrograms of the hip. Using radial sequences alone, 44 anterosuperior, 25 posterosuperior, 9 anteroinferior, and 5 posteroinferior labral tears were detected. In all four areas of the labrum, the radial imaging did not show any labral tear not seen on standard imaging planes. Discussion In MR arthrography of the hip, radial imaging did not reveal any additional labral tears. Standard imaging planes sufficiently demonstrate all acetabular labral tears.     

Acetabular labral tears: contrast-enhanced MR imaging under continuous leg traction

 Objective. The objective of this study was to evaluate the effects of continuous leg traction on contrast-enhanced MR imaging of the hip joint and to determine whether MR imaging under these conditions is useful for demonstrating acetabular labral tears. Design and patients. Nineteen hips underwent MR imaging with a T1-weighted spin-echo sequence, followed by MR imaging under continuous leg traction after intravenous injection of gadolinium-DTPA. Joint fluid enhancement and labral contour detection were evaluated. Eleven hips had labral tears shown by conventional arthrography, arthroscopy and macroscopic surgical findings. Assessment of labral tears by MR imaging was correlated with the diagnosis based on these standard techniques. Results and conclusions. Joint fluid enhancement was obtained in all hips at 30 min after injection. Superior and inferior labral surfaces were completely delineated in 1 hip on the unenhanced MR images, and in 7 and 13 hips, respectively, on the enhanced images under traction. The enhanced images under traction depicted 9 of the 11 labral tears. Comparison between the unenhanced image and the enhanced image under traction avoided mistaking undercutting of the labrum for a tear in 4 hips. Contrast-enhanced MR imaging under traction was valuable for detecting labral tears non-invasively and without radiation. Follow-up examinations using this method in patients with acetabular dysplasia can help to clarify the natural course of labral disorders and enable better treatment planning.     

Sublabral sulcus at the posteroinferior acetabulum: a potential pitfall in MR arthrography diagnosis of acetabular labral tears

OBJECTIVE: Our study correlated findings on hip MRI and MR arthrography with hip arthroscopy to assess the location, prevalence, and potential pitfall of a normal acetabular sublabral sulcus. MATERIALS AND METHODS: We retrospectively collected 58 hip MRI studies along with surgical reports in 58 patients who underwent hip arthroscopy over a 5-year period. Intraoperative photography (n = 23), radiography (n = 56), unenhanced MRI (n = 13), and MR arthrography (n = 54) studies were available for review. Two radiologists described hip anatomy on radiology studies with agreement by consensus. RESULTS: A normal posteroinferior sublabral groove was confirmed on available arthroscopy photographs in four (17.4%) of 23 hips. In each of these four patients, the anatomic sublabral groove correlated with apparent partial labral detachment on MR arthrography. On review of all studies, 13 hips (22.4%) without a posterior labral tear at surgery had imaging findings of a sublabral sulcus. The sulcus was not associated with acetabular dysplasia, which was radiographically noted in 12 cases (21.4%). Preoperatively, the sulcus was misdiagnosed as a tear in two cases. Labral tears were anterior or anterosuperior in 51 patients. CONCLUSION: A posteroinferior sublabral groove is a relatively common normal anatomic hip variation. If not recognized as normal, the sulcus may serve as a diagnostic pitfall on MR arthrography. Its location is distinct from most labral tears. We did not discover a sublabral sulcus at the anterior or anterosuperior acetabulum, the most common sites of labral injury.     

Evaluation of the acetabular labrum at 3.0-T MR imaging compared with 1.5-T MR arthrography: preliminary experience

Institutional review board approval and informed consent were obtained for this HIPAA-compliant study. The purpose of this study was to prospectively compare imaging of the acetabular labrum with 3.0-T magnetic resonance (MR) imaging and 1.5-T MR arthrography. Eight patients (four male, four female; mean age, 38 years) with hip pain suspicious for labral disease were examined at both MR arthrography and MR imaging. Presence of labral lesions, paralabral cysts, articular cartilage lesions, subchondral cysts, osteophytes, and synovial herniation pits was recorded. There was arthroscopic correlation of findings in five patients. MR imaging depicted four surgically confirmed labral tears that were identified at MR arthrography, as well as one that was not visualized at MR arthrography. MR imaging helped identify all other pathologic conditions that were diagnosed at MR arthrography and helped identify one additional surgically confirmed focal articular cartilage lesion. These results provide encouraging support for evaluation with 3.0-T MR imaging over 1.5-T MR arthrography.     

MR arthrography of the adult acetabular capsular-labral complex: correlation with surgery and anatomy.

OBJECTIVE: Our purpose was to describe the appearance of the acetabular capsular-labral complex on MR arthrography and to correlate this appearance with surgical findings in adult patients and with gross anatomic findings in cadavers. SUBJECTS AND METHODS: MR arthrography of the hip joint was performed in 40 patients and six cadavers. All patients underwent subsequent arthrotomy of the hip. MR arthrography consisted of a T1-weighted three-dimensional gradient-echo sequence in both the coronal oblique and sagittal oblique planes after intraarticular injection of a 2 mmol/l solution of gadopentetate dimeglumine. The normal and pathologic appearance of the capsular-labral complex was assessed, and the labra were evaluated on the basis of morphology, signal intensity, presence of a tear, and attachment to the acetabulum. MR arthrography findings were correlated with the surgical results in all patients and with the anatomic sections of the cadaveric hip joint specimens. RESULTS: MR arthrography images of the T-weighted three-dimensional gradient-echo sequences allowed visualization of the anatomic structures. The normal labrum was triangular, without any sublabral sulcus, and of homogeneous low signal intensity. A recess between the labrum and the joint capsule could be identified in instances in which no thickened labrum was present. Labral lesions included labral degeneration, a tear, or a detached labrum either with or without thickening of the labrum. The sensitivity for detection and correct staging of labral lesions with MR arthrography in the patient study was 91%; the specificity, 71%; and the accuracy, 88%. CONCLUSION: MR arthrography with T1-weighted three-dimensional gradient-echo sequences allows excellent assessment of the normal and pathologic acetabular capsular-labral complex.     

MR arthrographic findings of SLAP lesion type V

PURPOSE: The purpose of this study was to determine the findings of MR arthrography of the shoulder and to assess the role of MR arthrography in the diagnosis of superior labrum anterior to posterior (SLAP) lesion type V. METHODS AND MATERIALS: Two radiologists retrospectively reviewed fat-suppressed T1-weighted MR arthrography images of six patients who were diagnosed with SLAP lesion type V by arthroscopy. Each imaging plane, including the transverse, oblique coronal, oblique sagittal, and oblique transverse in abductor external rotation (ABER) position were evaluated for the following three findings: tear of the superior labrum at biceps tendon insertion, Bankart lesion, and continuity of the two former findings. RESULTS: Tear of the superior labrum was shown in all patients on oblique coronal images. Bankart lesion was noted in five patients on the transverse images and in four on the oblique sagittal images. On the oblique transverse images in ABER position, Bankart lesion was shown in all patients. The continuity of the two former findings was noted in three patients on the ABER positioned images. Therefore, three patients could be diagnosed as having SLAP lesion type V by MR arthrography in our series. CONCLUSION: It is difficult to detect all three findings of SLAP lesion type V in one imaging plane, however, a combination of multi-directional images may increase the feasibility of MR arthrography in diagnosing SLAP lesion type V.     

Hypoplasia of the glenoid neck of the scapula: imaging findings and report of 16 patients

OBJECTIVE: Our purpose was to describe the primary and associated imaging findings in patients with glenoid hypoplasia required for an accurate diagnostic appraisal of this uncommon abnormality. METHODS: The medical records and imaging studies of 16 patients with glenoid hypoplasia were retrospectively reviewed. Plain films were available in all patients, arthrography in 2, computed tomography (CT) in 5, CT arthrotomography in 3, magnetic resonance (MR) imaging in 9, and MR arthrography in 3 patients. The imaging studies were reviewed with special attention to the primary and secondary findings associated with glenoid hypoplasia. RESULTS: On radiographs, the 21 abnormal shoulders documented in the 16 patients with glenoid hypoplasia had an irregular (n=15) or shallow and smooth (n=6) articular surface of the glenoid. Computed tomography findings showed subluxation of the humeral head in 5 cases. On MR images, the hypoplastic bony glenoid and scapular neck were replaced by abnormal tissue of inhomogeneous signal intensity, showing the signal intensity characteristics of fibrocartilage (n=6) or fat (n=5). Widening of the inferior glenohumeral joint space was seen in all cases. Magnetic resonance imaging or MR arthrography revealed variable abnormalities of the labrum, including enlargement (10 shoulders), detachment of labrum from the bony glenoid (6 shoulders), intrasubstance tear (4 shoulders), and perilabral cyst (2 shoulders). The posterior labrum was absent (n=1) or torn (n=1) in one patient with bilateral deformity. Other findings included partial tear of the subscapularis tendon (n=1) and rotator cuff atrophy (n=1) and tear (n=2). CONCLUSION: Glenoid hypoplasia is a developmental skeletal anomaly comprising a spectrum of associated osseous and cartilaginous changes of the articular cartilage and glenoid labrum. Cross-sectional imaging studies, with or without arthrography, may help further characterize associated derangements of the shoulder, some of which are indicative of shoulder instability.     

MR arthrography of the posterior labrocapsular complex: relationship with glenohumeral joint alignment and clinical posterior instability

OBJECTIVE: The purpose of our study was to investigate the relationship between tears of the posterior labrocapsular complex and glenohumeral alignment on MR arthrography and the presence and extent of posterior labrocapsular tears in patients with posterior instability. MATERIALS AND METHODS: Posterior labrocapsular tears identified on 24 MR arthrograms and surgically confirmed were evaluated for length of tear and labrocapsular avulsion. These examinations and a comparison cohort of 70 normal MR arthrograms with normal findings were also evaluated for humeral head position relative to the glenoid fossa. Medical records were reviewed for clinical diagnosis of posterior instability and history of shoulder trauma. RESULTS: The position of the humeral head relative to the glenoid was significantly more posterior in patients with posterior labral tear than in patients with a normal posterior labrum (4.9 mm versus 0.7 mm; p < 0.0001). The mean length (+/- SD) of posterior labral tear was 15.9 +/- 1.7 mm, and a direct correlation was found between tear length and posterior humeral translation (r = -0.65; p = 0.002). Posterior labral tears were significantly longer (18.6 vs 13.1 mm; p = 0.04), and posterior humeral translation was greater (6.4 vs 3.4 mm; p = 0.006) in patients with labrocapsular avulsion than in those without avulsion. Twelve (50%) of the patients with posterior labrocapsular tear had posterior instability, and 10 (83%) had a history of macrotrauma. On MR arthrography, the mean posterior humeral translation was greater (6.2 mm +/- 0.08; p = 0.019), posterior labral tears were longer (19.4 mm +/- 1.7; p = 0.0008), and labrocapsular avulsion was more common (83%; p = 0.0001) in patients with posterior instability than in patients who had a posterior labral tear but a clinically stable shoulder. CONCLUSION: Clinical posterior instability is associated with excessive posterior humeral translation, long posterior labral tears, and posterior labrocapsular avulsion.     

Labral-ligamentous complex of the shoulder. Evaluation with double oblique axial MR arthrography

Purpose: 
To assess the ability of double oblique axial (DOA) MR arthrography in evaluating labral-ligamentous complex compared with conventional axial (CA) MR arthrography. Material and Methods: 
MR arthrography of 51 shoulders, subsequently examined with arthroscopy, were retrospectively reviewed. DOA imaging was performed in all 51 shoulders and both DOA and CA imaging in 37 using a 1.5 T unit with gradient recalled-echo T2*-weighted sequences. DOA imaging was performed using perpendicular planes to the long axis of the glenoid fossa obtained by an oblique sagittal scout image. We compared the ability of DOA with that of CA MR arthrography to assess labral injuries and to demonstrate the whole length of the anterior band of the inferior glenohumeral ligament (AIGHL), which were shown to be intact by arthroscopy. Results: 
For anterior labral injuries, sensitivity and specificity were 87% and 93% with CA, and 94% and 100% with DOA imaging, respectively. For posterior labral injuries, sensitivity and specificity were 47% and 100% with CA, and 79% and 96% with DOA imaging, respectively. There were no statistically significant differences between CA and DOA images, except for the ability to diagnose posterior labral injuries, where DOA imaging had a significant superior sensitivity (p = 0.0327). DOA images also demonstrated the whole length of the intact AIGHL in 10 of 11 shoulders, while CA imaging showed this in only 3 of 11. Conclusion: 
DOA imaging was equal or better than CA imaging for evaluating the labral-ligamentous complex.     

肩袖撕裂的肩关节MR造影评价

目的 评价肩关节MR造影对肩袖撕裂的诊断价值。方法 分析32例病人的肩关节MR造影图像,评价内容包括肩袖肌腱、盂唇、肱二头肌长头腱,结果与肩关节镜及肩部开放手术比较。结果 以肩关节镜和手术结果为标准,32例病人包括14例全层撕裂,6例下表面部分撕裂,12例无撕裂。肩关节MR造影判定有无肩袖撕裂的敏感性、特异性和准确性都为100％;诊断全层撕裂的敏感性、特异性和准确性分别为100％、94％和97％。同时,肩关节MR造影还正确诊断了所有的多肌腱撕裂、合并的盂唇异常及肱二头肌长头腱异常。结论 肩关节MR造影可以准确、全面地评价肩袖撕裂。     

Gradient-recalled echo sequences in direct shoulder MR arthrography for evaluating the labrum

Objective The purpose of this study was to determine the utility of fat-suppressed gradient-recalled echo (GRE) compared with conventional spin echo T1-weighted (T1W) sequences in direct shoulder MR arthrography for evaluating labral tears. Materials and methods Three musculoskeletal radiologists retrospectively reviewed MR arthrograms performed over a 12-month period for which surgical correlation was available. Of 180 serial arthrograms, 31 patients had surgery with a mean of 48 days following imaging. Paired coronal oblique and axial T1W or GRE sequences were analyzed by consensus for labral tear (coronal oblique two-dimensional multi-echo data image combination, 2D MEDIC; and axial three-dimensional double-echo steady-state, 3D DESS; Siemens MAGNETOM Sonata 1.5-T MR system). Interpretations were correlated with operative reports. Results Of 31 shoulders, 25 had labral tears at surgery. The GRE sequences depicted labral tears in 22, while T1W images depicted tears in 16 (sensitivity 88% versus 64%; p < 0.05). Subdividing the labrum, GRE was significantly more sensitive for the posterior labrum (75% versus 25%; p < 0.05) with a trend toward greater sensitivity at the anterior labrum (78% versus 56%; p = 0.157) but not significantly different for the superior labrum (50% versus 57%; p > 0.7). Specificities were somewhat lower for GRE. Conclusion Thin section GRE sequences are more sensitive than T1W for the detection of anterior and posterior labral tears. As the specificity of GRE was lower, it should be considered as an adjunctive imaging sequence that may improve depiction of labral tears, particularly smaller tears, in routine MR arthrography protocols. Preliminary results of this study were presented at the Annual Meeting of the Society of Skeletal Radiology, Orlando, FL, USA, on 21 March 2007.     

MR arthrography of the shoulder: comparison with conventional MR imaging

Twenty-three patients underwent both conventional MR imaging of the shoulder and MR shoulder arthrography for clinically suspected labral or rotator cuff abnormalities. Images obtained before and after contrast administration were studied independently, and without knowledge of clinical findings, by two radiologists for the presence of abnormalities of the glenoid labrum or rotator cuff. Results were correlated with surgical findings in all patients. Nine patients had surgically proved labral tears. MR arthrography detected all nine labral abnormalities, while six of the nine were missed on conventional MR imaging. Fourteen patients had surgically proved rotator cuff tears. MR arthrography detected 11 of the 14 tears and missed three partial tears on the bursal surface. Conventional MR imaging detected nine cuff tears and missed five tears; two of these were complete rotator cuff tears and three were partial tears of the undersurface of the rotator cuff. Our results suggest that MR arthrography enhances the accuracy of MR in the evaluation of the glenoid labrum and rotator cuff tendon.     

Can isotropic fast gradient echo imaging be substituted for conventional T1 weighted sequences in shoulder MR arthrography at 3 Tesla?

PURPOSE: To assess the practical utility of isotropic shoulder imaging in patients undergoing MR arthrography. Isotropic shoulder imaging can be performed in less than three minutes with use of fast gradients. MATERIALS AND METHODS: Two experienced musculoskeletal radiologists retrospectively interpreted MR images of the shoulder in 100 consecutive patients undergoing MR arthrography of the shoulder. All patients underwent MRI of the shoulder in oblique coronal, oblique sagittal, and axial planes on a 3.0-Tesla MRI system. All patients had conventional fast spin-echo T1-weighted imaging. All patients also had thin section (0.4 mm) isotropic spoiled gradient echo images performed. A total of 67 of the 100 patients underwent subsequent arthroscopy and results were compared with MR interpretations. RESULTS: There were 41 full-thickness supraspinatus tendon tears, nine partial-thickness supraspinatus tendon tears (seven articular surface and two bursal surface), 21 superior labral, 18 anterior labral, and seven posterior labral tears demonstrated by consensus retrospective reading of the 100 shoulder MR exams. There was no difference in interpretation of the isotropic images as compared to the conventionally acquired images in the oblique coronal, oblique sagittal, and axial planes. Some patients had more than one finding on each exam. A total of 67 patients went on to arthroscopy. There were 41 full-thickness supraspinatus tendon tears, nine partial-thickness supraspinatus tendon tears, 21 superior labral, 18 anterior labral, and seven posterior labral tears demonstrated on arthroscopy. All full- and partial-thickness supraspinatus tendon tears seen at arthroscopy were seen on consensus MR reading. A total of 19 out of the 21 patients with superior labral anterior posterior (SLAP) tears at arthroscopy were seen on consensus MR reading. A total of 16 of the 18 anterior labral tears and six of the seven posterior labral tears seen at arthroscopy were seen on consensus MR reading. Some of the 67 patients had more than one finding on arthroscopy. CONCLUSION: Isotropic imaging of the shoulder is practical in clinical imaging when performed with use of fast gradients on a 3-Tesla system. Isotropic imaging provides the same clinical information as conventional imaging and can be acquired in less than three minutes.     

Paralabral cysts in the hip joint: findings at MR arthrography

Purpose

The purpose of this research was to retrospectively characterize paralabral cysts of the hip as seen at MR arthrography.

Materials and methods

After Institutional Review Board approval, 704 patients who had MR arthrography were identified over a 3-year period and 40 patients were identified as having a cyst or fluid collection at the hip by MR report. MR images from these 40 patients were retrospectively reviewed by three radiologists where 18 were found to have a paralabral cyst, which were characterized as follows: location, configuration, contrast filling, size of the cyst, extent, direction, and whether associated osseous changes were present. In addition, the acetabular labrum was assessed for tears and, if present, the location and pattern were characterized.

Results

Paralabral cysts were located anterosuperiorly in 56%, anteriorly in 22%, posterosuperiorly in 17%, and anteroinferiorly in 6% of cases. The vast majority (94%) were multilocular and filled with intra-articular contrast medium. The average dimensions were 8?×?7?×?11?mm. The paralabral cyst demonstrated extracapsular extension in 72% of cases, with 39% located between the ilium and gluteus minimus, and 22% between the ilium and iliopsoas. Remodeling of the ilium adjacent to the cyst was observed in 50% of these cases. A labral tear was at the base of the labrum adjacent to the cyst in 78% of cases, while the tear was isolated to the body of the labrum in 22%. Tears were most commonly anterosuperior (55%) or anterior (28%) in location.

Conclusion

Our results show that paralabral cysts of the hip are most commonly located anterosuperiorly, are multilocular, fill with intra-articular contrast medium, have average dimensions up to 11?mm, and often extend extracapsularly between muscle and bone where they may remodel the adjacent ilium.     

Injuries of the superior portion of the glenoid labrum involving the insertion of the biceps tendon: MR imaging findings in nine cases.

The goal of this investigation was to describe the MR appearance of traumatic fraying or detachment of the superior portion of the glenoid labrum including the insertion of the tendon of the long head of the biceps. This condition is caused either by an acute injury or by repeated overhead motion during participation in sports. In nine patients with such a lesion, the arthroscopic report and MR images were available for review. These patients were 22-64 years old (mean, 38). In four patients only fraying was noted during arthroscopy, in four patients the superior part of the labrum was detached together with the insertion of the biceps tendon, and in one case there was a bucket-handle tear of the superior portion of the labrum. The MR images were retrospectively evaluated by three osteoradiologists in conference. Signal changes within the labrum and detachment of the labrum were noted, and the findings were compared with the results of arthroscopy. MR imaging did not allow recognition of simple fraying. In two of the five cases with arthroscopic findings of detachment of the superior labrum from the glenoid rim, differentiation between complete and partial labral detachments was not possible even with MR arthrography. However, in these cases the patient's age and history led to the correct diagnosis. We conclude that early traumatic abnormalities of the superior portion of the labrum cannot be detected with MR imaging. Complete detachment, however, can be demonstrated if the patient's age and history are taken into consideration.     

上方盂唇前后向撕裂的肩关节MR造影评价

目的 评价肩关节MR造影对上方盂唇前后向撕裂(SLAP损伤)的诊断价值.方法 由2名影像诊断医师独立回顾分析137例经肩关节镜证实的肩关节造影MR图像,判定有无SLAP损伤,并进行分型.SLAP损伤分4型:Ⅰ型为上盂唇毛糙,Ⅱ型为肱二头肌腱盂唇复合体从关节孟撕裂,Ⅲ型为上盂唇桶柄状撕裂,Ⅳ型为上盂唇桶柄状撕裂并同时累及肱二头肌长头腱.计算肩关节MR造影对SLAP损伤的敏感度、特异度和准确度,以及与肩关节镜分型的符合率.采用Kappa统计计算2名影像诊断医师评价的一致性.结果 137例患者中,肩关节镜证实SLAP损伤59例,包括SLAP损伤Ⅰ型6例(10.2%)、Ⅱ型50例(84.7%)、Ⅲ型3例(5.1%).肩关节MR造影的敏感度、特异度和准确度医师甲分别为86.4%(51/59)、78.2%(61/78)和81.8%(112/137),医师乙分别为88.1%(52/59)、84.6%(66/78)和86.1%(118/137).2名影像诊断医师的评价一致性极好(K=0.796).肩关节MR造影与肩关节镜的分型符合率医师甲为83.1%(49/59),医师乙为79.7%(47/59).结论 肩关节MR造影是评价SLAP损伤比较可靠的影像方法.