MR Arthrography of the Labral Capsular Ligamentous Complex in the Shoulder: Imaging Variations and Pitfalls

Transcription

1 Downloaded from by on 12/31/17 from IP address Copyright RRS. For personal use only; all rights reserved Yang Hee Park 1 Ji Yeon Lee 2 Sung Hee Moon 1 Jong Hyun Mo 1 o Kyu Yang 3 Sung Ho Hahn 3 Donald Resnick 4 Received December 13, 1999; accepted after revision February 16, Presented at the annual meeting of the merican Roentgen Ray Society, New Orleans, May Department of Radiology, National Police Hospital, 58 Karakbon-dong, Songpa-gu, Seoul, , Korea. ddress correspondence to Y. H. Park. 2 Department of Radiology, Samsung Medical Center, 50 Irwon-dong, Kangnam-gu, Seoul, , Korea. 3 Department of Orthopedic Surgery, National Police Hospital, Songpa-gu, Seoul, , Korea. 4 Department of Radiology, University of California at San Diego Medical Center, 200 W. rbor Dr., San Diego, C JR 2000;175: X/00/ merican Roentgen Ray Society MR rthrography of the Labral Capsular Ligamentous Complex in the Shoulder: Imaging Variations and Pitfalls OJECTIVE. Using MR arthrography, we examined normal anatomy, anatomic variations, and pitfalls of imaging the labral capsular ligamentous complex in the asymptomatic shoulder. SUJECTS ND METHODS. We obtained 108 MR arthrograms of the glenohumeral joint in 95 asymptomatic volunteers with axial (108 shoulders) and oblique coronal (56 shoulders) images. We examined labral shape, patterns of capsular insertion, presence or absence of glenohumeral ligaments, and pitfalls of imaging. Our patients were men (auxiliary policemen) between 19 and 24 years old (mean age, 21 years). RESULTS. The shapes of labra were triangular (anterior part, 64%; posterior part, 47%), round (17%; 33%), flat (2%; 17%), cleaved (11%; 1%), notched (3%; 0%), or absent (2%; 2%). Using the system of Mosely and Overgaard, the anterior capsular insertion was type 1 in 63% of shoulders, type 2 in 20%, and type 3 in 17%; the posterior capsular insertion was type 1 in 60% of shoulders, type 2 in 31%, and type 3 in 9%. The superior and inferior glenohumeral ligaments were present in 99% of shoulders, but the middle glenohumeral ligament was present in only 79%. We also detected many pitfalls of MR imaging in identifying findings such as the undercutting of the labrum by cartilage (32% of shoulders), prominent axillary folds (46%), sublabral holes (7%) or recesses (33%), uford complexes (2%), and sulci between the biceps tendon and superior labrum (30%). CONCLUSION. Knowledge of normal anatomy and anatomic variations of the labral complex is important for the examination of the shoulder with MR arthrography. S houlder instability is a common condition that primarily affects young active people, especially athletes. To correctly diagnosis and properly treat shoulder instability, many methods have been used such as routine radiography, conventional arthrotomography, CT arthrography, and MR imaging [1 4]. MR arthrography is also an important imaging technique for assessing the labral capsular ligamentous complex of the shoulder joint [5 7]. The labral capsular ligamentous complex is an important component of shoulder stability [8, 9]. Many normal variations of this complex that are seen on MR imaging and that are confused with pathologic abnormalities have been reported in the literature, but their appearance on MR arthrography has received little attention [10 13]. We performed MR arthrography in asymptomatic shoulders to evaluate the appearance and frequency of anatomic variations and imaging pitfalls. Subjects and Methods We obtained 108 MR arthrograms of the glenohumeral joint in 95 asymptomatic volunteers with axial (108 shoulders) and oblique coronal (56 shoulders) images. Our patients were men (auxiliary policemen) between 19 and 24 years old (mean age, 21 years). Fifty right shoulders and 58 left shoulders were examined. MR arthrography was performed with fluoroscopic guidance, and intraarticular positioning of the needle was confirmed with a small amount of iodinated contrast material (Ultravist 370; Schering, erlin, Germany). Subsequently, ml of a mixed solution was injected, which was composed of 250 ml of saline, 0.1 mmol of gadolinium diethylene triamine pentaacetic acid (Magnevist; Schering), and 0.3 ml of epinephrine (1:1000). 0.5-T MR scanner (Gyroscan; Philips, Ienthoven, The Netherlands) was used, and T1-weighted spin-echo axial images (TR range/ TE range, /21 26), often supplemented with oblique coronal images, were obtained. The thickness and interval of the slices were 3.0 mm and 0.3 mm, respectively. During MR imaging, the volunteer was lying in the supine position with the arm in a neutral position. ll MR arthrographic images were indepen- JR:175, September

2 Park et al. Downloaded from by on 12/31/17 from IP address Copyright RRS. For personal use only; all rights reserved dently interpreted by three radiologists who were experienced in musculoskeletal MR imaging. Discrepancies among observers were resolved by consensus. The shape of the anterior and posterior parts of the labrum was evaluated, and descriptive terms such as triangular, round, flat, cleaved, notched, and absent were applied [11] (Fig. 1). t the inferior level of the glenoid cavity, the anterior and posterior areas of the capsular insertion were classified into one of three categories Note. Numbers are percentages. Some columns do not total 100% because of rounding. D according to the system of Mosely and Overgaard [14]: type 1 was a capsular insertion on the labrum, type 2 was an insertion occurring at the junction of the labrum and the glenoid, and type 3 was an insertion more medial to this junction on the cortical surface of the glenoid neck. The presence or absence of glenohumeral ligaments and additional findings such as the undercutting of the labrum by cartilage, sublabral holes, sublabral recesses, or uford complexes (the TLE 1 Shapes of nterior and Posterior Glenoid Labra on MR rthrography of 108 symptomatic Volunteers nterior Labrum Posterior Labrum Shape Superior Middle Inferior verage Superior Middle Inferior verage Triangular Round Flat Cleaved Notched bsent E combination of the absence of the anterior superior labrum and the presence of a cordlike middle glenohumeral ligament) [15] were noted. We also determined the presence or absence of a prominent axillary fold (simulating an intraarticular body) and any sulci between the biceps tendon and superior labrum on oblique coronal images. Results The morphology of the anterior and posterior parts of glenoid labrum was considerably variable, with several common dominant features (Table 1). The most common shape was triangular, found anteriorly in 64% of shoulders and posteriorly in 47% (Fig. 2). The second most common shape was round, found anteriorly in 17% of shoulders and posteriorly in 33% (Fig. 2). The flat labrum was found in anteriorly 2% of shoulders and posteriorly in 17% (Fig. 2). The cleaved and notched shapes were noted anteriorly in 11% and 3% of shoulders, respectively (Figs. 2C and 2D), especially at the C F Fig. 1. Labral shapes. F, Drawings show triangular labrum, anteriorly and posteriorly (); round labrum, anteriorly and posteriorly (); cleaved labrum, anteriorly (C); notched labrum, anteriorly (D); notch extends in labrum in more vertical fashion than horizontally oriented cleft; flat labrum, anteriorly and posteriorly (E); and absent labrum, anteriorly and posteriorly (F). Note glenoid rim has no visible labrum. 668 JR:175, September 2000

3 MR rthrography of the Shoulder Downloaded from by on 12/31/17 from IP address Copyright RRS. For personal use only; all rights reserved Fig. 2. Glenoid labral shape of various patients on T1-weighted axial MR arthrography. E, MR arthrograms show triangular variation, anteriorly and posteriorly (; note middle glenohumeral ligament [arrow]); round variation, anteriorly and flat, posteriorly (); cleaved variation, anteriorly (arrow, C); notched variation, anteriorly (arrow, D); and absent variation, anteriorly (E). Fig. 3. Three types of capsular insertion on T1-weighted MR arthrography., MR arthrogram shows type 1 (insertion on labrum, solid arrow) in posterior capsular insertion and type 3 (insertion more medial to junction of labrum and glenoid, open arrow) in anterior capsular insertion in 21-year-old man., MR arthrogram shows type 2 (insertion at junction of labrum and glenoid, solid arrow) in posterior capsular insertion and type 3 (open arrow) in anterior capsular insertion in 19-year-old man. superior and middle levels. Labral absence was revealed anteriorly in 2% of shoulders and posteriorly in 2% (Fig. 2E). Capsular insertion was considerably variable at the inferior glenoid level, with a type 1 capsule present anteriorly in 63% of shoulders, type 2 in 20%, and type 3 in 17%. The posterior capsular insertion was type 1 in 60% of shoulders, type 2 in 31%, and type 3 in 9% (Figs. 3 and 3). The superior and inferior glenohumeral ligaments were present in all cases except one (99%), but the presence of the middle glenohumeral ligament was not constant (present in 79% of cases) (Fig. 4). D We noted many appearances resembling labral defects, such as the undercutting of the labrum by hyaline cartilage (32% of shoulders) (Fig. 5), sublabral holes (7%) (Figs. 5 and 5C) or recesses (33%) (Fig. 5D), and uford complexes (2%) (Fig. 5E). Prominent axillary folds mimicked intraarticular bodies C E JR:175, September

4 Downloaded from by on 12/31/17 from IP address Copyright RRS. For personal use only; all rights reserved Fig year-old man with back pain. T1-weighted axial MR arthrogram fails to show middle glenohumeral ligament at superomedial level of glenohumeral joint, indicating absence of ligament. Note tendon of subscapularis muscle (arrow). G (46%) (Fig. 5F), Park and a et sulcus al. was seen between the biceps tendon and superior labrum (30% of shoulders) (Figs. 5C and 5G). D Discussion Several imaging methods have been used to evaluate the glenoid labrum, ligaments, and joint capsule. lthough MR imaging has been used in the examination of the glenoid labrum, there are some difficulties with the technique [13, 16]. Hajek et al. [17] were the first to perform MR arthrography with the injection of a mixture of saline solution and gadopentetate dimeglumine into the joint space; they found that many anatomic structures were better delineated because of capsular distention. The origin and course of the glenohumeral ligaments, tears of the glenoid labrum, and relationship of the labrum with the bicipital tendon and glenohumeral ligaments are better revealed on MR arthrography than on conventional MR imaging [5, 12]. It is essential to understand the normal anatomy and anatomic variations of the labral capsular ligamentous complex for the accurate interpretation of MR arthrograms in patients with clinically suspected shoulder instability. The glenoid labrum is composed of fibrocartilaginous tissue that has low signal intensity on T1- and T2-weighted MR images. The labrum deepens the glenoid cavity, from which the bicipital tendon and glenohumeral ligament originate. The size and shape of the glenoid labrum in asymptomatic subjects are variable. Neumann et al. [11] reported that the most common labral shape was triangular (anterior part, 45%; posterior part, 73%), and Fig. 5. T1-weighted MR arthrograms show findings resembling labral tears., Undercutting of labrum (arrow) is increased signal of hyaline cartilage underneath anterior labrum., Sublabral hole (arrow) is noted in anterosuperior portion of glenoid (high-signal-intensity effusion is interposed in this foramen). C, Oblique coronal view of same patient as in shows double Oreo cookie configuration with black (glenoid cortex, open arrow), white (sublabral hole, solid arrow), black (labrum, arrowhead), white (sulcus, open arrowhead), and black (biceps tendon, white arrow) areas, which are not superior labrum anteroposterior lesions. D, Sublabral recess (arrow) is visible underneath anterosuperior portion of glenoid labrum. E, Middle glenohumeral ligament (arrow) is thick and cordlike with absence of anterosuperior labrum, indicating uford complex. F, Synovial folds in axillary pouch simulate intraarticular bodies (arrows). G, Oblique coronal MR arthrogram shows small sulcus (solid arrow) between superior labrum (arrowhead) and biceps tendon (open arrow). E C F 670 JR:175, September 2000

5 MR rthrography of the Shoulder Downloaded from by on 12/31/17 from IP address Copyright RRS. For personal use only; all rights reserved the next most frequent appearance was round (19%; 12%). We found that, in the anterior part of the labrum, a round shape increased in frequency at inferior levels, and cleaved or notched shapes increased in frequency at the superior and middle levels, results comparable with those of Neumann et al. Liou et al. [10] suggested that the apparent cleft in the labrum was actually caused by the orgin of the inferior glenohumeral ligament at this location. In our study, in Figures 2C and 2D, we followed the clefts and notches inferiorly; however, we noted no relationship between the cleft and the orgin of the inferior glenohumeral ligament. Round and flat labra were more common posteriorly than anteriorly, unlike the results of Neumann et al. This discrepancy may be explained by capsular distention related to the intraarticular injection of contrast material. The attachment of the anterior part of the inferior capsule was most commonly type 1 (63% of shoulders), a finding that is comparable with that of a previous report [11]. In the posterior part of the capsule, type 1 (60%) and type 2 (31%) capsular attachments were commonly seen, findings that differ from those of Neumann et al. [11], who found type 1 in 100% of shoulders and type 2 in 0%. Types 2 and 3 insertions might be more commonly observed because of the capsular expansion seen on MR arthrography. The capsular attachments observed by Mosely and Overgaard [14] were at the level of the insertion site of the middle glenohumeral ligament. We observed the capsular attachment at the inferior level of the glenoid, but our results may have a considerable validity to verify the tendency of the capsular attachment. The glenohumeral ligaments are important for the stability of the shoulder joint [18]. The middle glenohumeral ligament may originate from the labrum or neck of the glenoid fossa and course posterior to the subscapularis tendon. The middle glenohumeral ligament is an important stabilizing structure at 40º of abduction and external rotation of the humerus [5, 12]. The middle glenohumeral ligament is most variable in size and frequency and may be attenuated or absent in up to 30% of normal shoulders [12]. One pitfall of MR imaging is that findings may mimic lesions suggestive of glenohumeral instability. In the superior portion of the labrum, a sublabral hole and recess have been recognized by other investigators [9, 19]. Cooper et al. [9] reported that the labrum was loosely attached by thin capsular tissue (sublabral recess) Fig. 6 Sublabral hole and sulcus between labrum and biceps tendon., Drawing shows glenolabral junction with normal sublabral hole (solid arrow) and sulcus (double arrow) between labrum (arrowhead) and biceps tendon (open arrow)., Drawing shows double Oreo cookie configuration with black (glenoid cortex), white (fluid in sublabral hole, single arrow), black (labrum), white (fluid in sulcus between labrum and biceps tendon, double arrow), and black (biceps tendon) areas. in five specimens and was not attached to the glenoid rim at all (sublabral hole) in four specimens among 23 newly frozen shoulders. De Palma et al. [19] reported that the superior labrum in fetuses and infants was firmly attached to the hyaline cartilage without evidence of a sublabral foramen, and alterations were seen in the superior labrum by the second decade of life. They concluded that the detachment of the superior labrum was an age-dependant degenerative phenomenon because its prevalence increased with the subject s age. In that study, nonattachment of the superior labrum was noted in only 17% of specimens from subjects in the second decade of life, more than 50% of subjects older than 20 years, and more than 95% of subjects in the seventh and eight decades of life. In our study, a sublabral hole was revealed in 7% of subjects, and a sublabral recess was revealed in 33% of subjects, findings probably related to the young age of our volunteers. In 30% of subjects imaged in the oblique coronal view (56 shoulders), a sulcus was revealed between the superior labrum and the biceps tendon. In some cases, a sublabral hole and a sulcus between the superior labrum and biceps tendon were revealed simultaneously (Figs. 5C and 6), a finding called the double Oreo cookie configuration. In our volunteers, this configuration was different from the description of Smith et al. [20]. They described glenoid cortex (black), sublabral recess (white), labrum (black), superior labrum anteroposterior tear (white), and labrum (black) as the double Oreo cookie configuration. The absence of the anterosuperior labrum and the presence of a cordlike middle glenohumeral ligament, known as the uford complex, was revealed in 2% of shoulders in this study. Williams et al. [21] reviewed 200 arthroscopic videotapes and reported that 1.5% of patients had findings suggestive of a uford complex, which should not be interpreted as detached labra. Some limitations may have affected our study. ll subjects were young men, so the effect of increasing age or sex cannot be evaluated. Normal shoulders were not verified with arthroscopy or surgery, although no history of shoulder instability or subjective shoulder symptoms were noted. In conclusion, knowledge of normal anatomy, anatomic variations, and pitfalls of labral capsular ligamentous complex is important for the examination of the shoulder with MR arthrography. References 1. McGlynn FJ, El-Khoury G, lbright JP. rthrotomography of the glenoid labrum in shoulder instability. J one Joint Surg m 1982;64-: Haynor DR, Shunman WP. Double contrast CT arthrography of the glenoid labrum and shoulder girdle. RadioGraphics 1984;4: McNiesh LM, Callaghan JJ. CT arthrography of the shoulder: variations of the glenoid labrum. JR 1987;149: Garneau R, Renfrew DL, Moore TE, El-Khoury GY, Nepola JV, Lemke JH. Glenoid labrum: evaluation with MR imaging. Radiology 1991;179: Palmer WE, rown JH, Rosenthal DI. Labralligamentous complex of the shoulder: evaluation with MR arthrography. Radiology 1994;190: eltran J, Rosenberg ZS, Chandnani VP, Cuomo JR:175, September

6 Park et al. Downloaded from by on 12/31/17 from IP address Copyright RRS. For personal use only; all rights reserved F, eltran S, Rokito. Glenohumeral instability: evaluation with MR arthrography. RadioGraphics 1997;17: eltran J, encardino J, Mellado J, Rosenberg ZS, Irish RD. MR arthrography of the shoulder: variants and pitfalls. RadioGraphics 1997;17: Turkel SJ, Panio MW, Marshall JL, Girgis FG. Stabilization mechanisms preventing anterior dislocation of the glenohumeral joint. J one Joint Surg m 1981;63-: Cooper DE, rnoczky SP, O'rien SJ, Warren RF, DiCarlo E, llen. natomy, histology and vascularity of the glenoid labrum. J one Joint Surg m 1992;74-: Liou JTS, Wilson J, Totty WG, rown JJ. The normal shoulder: common variations that simulate pathologic conditions at MR imaging. Radiology 1993;186: Neumann CH, Peterson S, Jahnke H. MR imaging of the labral-capsular complex: normal variations. JR 1991;157: Massengill D, Seeger LL, Yao L, et al. Labrocapsular ligamentous complex of the shoulder: normal anatomy, anatomic variation, and pitfalls of MR imaging and MR arthrography. Radio- Graphics 1994;14: Kaplan P, ryans KC, Davick JP, Otte M, Stinson WW, Dussault RG. MR imaging of the normal shoulder: variants and pitfalls. Radiology 1992;184: Mosely HF, Overgaard. The anterior capsular mechanism in recurrent anterior dislocation of the shoulder. J one Joint Surg r 1962;44-: Tirman PFJ, Feller JF, Palmer WE, Carroll KW, Steinbach LS, Cox I. The uford complex: a variation of normal shoulder anatomy MR arthrographic imaging features. JR 1996;166: Tsai JC, Zlatkin M. Magnetic resonance imaging of the shoulder. Radiol Clin North m 1990; 28: Hajek PC, Sartoris DJ, Neumann CH, Resnick D. Potential contrast agents for MR arthrography: in Notice to uthors vitro evaluation and practical observations. JR 1987;149: Resnick D. Internal derangement of joint. In: Resnick D, ed. Diagnosis of bone and joint disorders. Philadelphia: Saunders, 1995: De Palma F, Gallery G, ennett G. Variational anatomy and degenerative lesions of the shoulder joint. In: merican cademy of Orthopedic Surgery: instructional course lectures, vol. 6. St. Louis: Mosby, 1949: Smith DK, Chopp TM, ufdemorte T, Witkowski EG, Jones RC. Sublabral recess of the superior glenoid labrum: study of cadavers with conventional nonenhanced MR imaging, MR arthrography, anatomic dissection, and limited histologic examination. Radiology 1996;201: Williams MM, Snyder SJ, uford D. The uford s complex: the cordlike middle glenohumeral ligament and absent anterosuperior labrum complex a normal anatomic capsulolabral variant. rthroscopy 1994;10: In keeping with sound environmental and economic principles, the JR encourages all authors to submit manuscripts printed on both sides of the page. This practice not only will save paper but also will reduce the price of postage required to mail the manuscript. 672 JR:175, September 2000