Diagnosis of Hill-Sachs Lesion of the Shoulder

Size: px

Start display at page:

Download "Diagnosis of Hill-Sachs Lesion of the Shoulder"

Tyrone Payne
5 years ago
Views:

Acta Radiologica ISSN: 0284-1851 (Print) 1600-0455

com/loi/iard20 Diagnosis of Hill-Sachs Lesion of the

Mecozzi To cite this article: Luigi Pancione, G.

Mecozzi (1997) Diagnosis of Hill-Sachs Lesion of the

Submit your article to this journal Article views:

1 Acta Radiologica ISSN: (Print) (Online) Journal homepage: Diagnosis of Hill-Sachs Lesion of the Shoulder Luigi Pancione, G. Gatti & B. Mecozzi To cite this article: Luigi Pancione, G. Gatti & B. Mecozzi (1997) Diagnosis of Hill-Sachs Lesion of the Shoulder, Acta Radiologica, 38:4, To link to this article: Published online: 04 Jan Submit your article to this journal Article views: 581 Full Terms & Conditions of access and use can be found at

2 Acta Radiologica 38 (1997) Printed in Denmark. All rights reserved Copyright 0 Acta Radiologica 1997 A C TA RADIOLOGICA ISSN DIAGNOSIS OF HILL-SACHS LESION OF THE SHOULDER Comparison between ultrasonography and arthro-ct L. PANCIONE, G. GATTI and B. MECOZZI Department of Radiology, Ospedale Civile Maria Vittoria, Turin, Italy. Abstract Aim: The Hill-Sachs lesion is a compression fracture caused by impact on the trabeculae of the humeral head during anterior glenohumeral dislocation. The early and accurate identification of patients who risk recurrence of shoulder dislocation requires a suitable screening method. This should be characterized by high sensitivity and specificity, low cost, and repeatability. Material and Methods: The results of ultrasound examination as a screening method were evaluated, the indicator being the identificatiqn of the Hill-Sachs lesion. Using ultrasonography and conventional radiological techniques, and double contrast CT (arthro-ct), we studied 60 patients with posttraumatic instability of the shoulder. Results: Against arthro-ct as the true standard, ultrasonography showed a sensitivity of 95.6%, specificity of 92.8%, and diagnostic accuracy of 95%. Conclusion: Ultrasonography is an acceptable screening examination for recurrent scapulohumeral dislocation and should be applied prior to other techniques of investigation such as arthro-ct or MR imaging. Key words: Shoulder, dislocation; CT, ultrasonography. Correspondence: Luigi Pancione, Via dei Lerda 34, I Cuneo, Italy. FAX The first description of a fracture caused by the impact of the humeral head on the glenoid cavity dates back to MALGAIGNE in 1832 (8). In 1940 HILL & SACHS identified the connection of this fracture with articular instability (6). Today the lesion is known as the Hill-Sachs defect. Its reported incidence varies a great deal from one study to another. According to some authors, the lesion occurs in as many as 47% of patients already after the first episode of scapulohumeral dislocation (1, 7) and up to 100% in patients with recurrent shoulder dislocation (9, 12). With conventional radiographic examination it is not always possible to detect the lesion. Double-contrast CT (arthro-ct) and MR are the investigations of choice (3, 5, 12) but they involve high costs and are consequently ruled out as initial examinations. For this reason we evaluated ultrasound (US) examination as a screening method, using the identification of the Hill-Sachs lesion as the indicator. Material and Methods Using first US and subsequently plain radiography and arthro-ct, we studied 60 patients (47 men and 13 women aged 18-46, mean age 32 years) presenting clinical signs of posttraumatic instability of the shoulder. The US assessment was carried out as a comparison with the healthy side. We used a 7.5- MHz linear probe principally for transverse scans posterior to the humeral head, taking as reference the acromial spine and the plane passing through the coracoid process, with the arm in internal and external rotation in order to reveal alterations in the posterior profile of the humeral head. Each patient underwent radiography of the shoulder in the projections of Didiee, Hermodsson, Styker and Westpoint (2, 6, 9, 10). Immediately afterwards we injected 3 ml of a nonionic hydrosoluble iodated contrast medium (Iopamiro 300, Bracco) and 6 ml of air into the joint space via anterior access (11). This was fol- 523

3 L. PANCIONE ET AL. alteration. It becomes a second-degree lesion when there is an initial involvement of the subchondral bone, and a third-degree lesion in the case of gross subchondral bone defect (1). The radiographic criteria for positivity were: 1) irregularity of the contour of the lateral border of the humerus in frontal projection in external rotation; 2) a distinct grooved defect at the lateral border of the humeral head, more clearly visible in internal rotation; and 3) a vertical line of sclerotic bone resulting from fracture caused by impact and circumscribing the defect. The CT criteria of positivity were: 1) a grooved defect of variable width and extent at the posterolateral surface of the humeral head; and 2) a sclerotic border (this allowed differential diagnosis with respect to anatomical variants). The US criteria for positivity, compared with the healthy contralateral limb, were: 1) flattening of the convexity of the posterosuperior profile of the humeral head; and 2) a grooved defect in the humeral head itself (Fig. 1). The assessment was confined to the portion of the humeral head located above the coracoid process, the aim being to prevent the possibility of false-positives due to the posterior profile of the humeral neck. Patients suffering from bilateral dislocation were not included in the study. No attempt was made to assess the anterior glenoidal labrum or the relative position of the humeral head. Each criterion for positivity was considered valid even when it presented singly. Each patient was assessed by 3 observers using all the methods. The 3 techniques were used in the following sequence: US, radiography, arthro-ct. The results according to the criteria described above were extrapolated at the end of each examination. Fig. 1. a) US: modest and circumscribed grooved defect of osseus edge of humeral head (+). b) Arthro-CT Hill-Sachs lesion of slight degree (+). lowed by arthro-ct with the patient in the supine position and the arm being examined in the prone position; the contralateral arm was abducted and positioned above the head. From the subacromial level up to the surgical neck of the humerus, 3-mm-thick images at 3-mm intervals were taken, using the bone reconstruction algorithm (3). The Hill-Sachs lesion is defined as a first-degree lesion when the damage is limited to the osteochondral level, with involvement of the peripheral compact osseous layer only and without subchondral Results The presence of alteration in the humeral profile or the presence of a defect (with typical triangular appearance) provided the US diagnosis of a Hill-Sachs lesion in 45 patients. In the remaining 15 patients Table Reliability of US in diagnosing Hill-Sachs lesion (with arthro-ct as the true standard) CT+ CT- us us Sensitivity 95.6% Specificity 92.8% Diagnostic accuracy (A+D)/(A+B+C+D)=95% 524

Note the disappearance of the cortical thickness at the site of the lesion (+). the result was considered negative.

4 HILL-SACHS LESION OF THE SHOULDER Fig. 2. a) US: wide, deep and irregular grooved defect of osseus edge of humeral head (+). b) Radiography in axial projection: confirmation of a Hill-Sachs lesion with gross irregularities; concomitant forward pulsion of humeral head (+). c) Arthro-CT confirmation of the Hill-Sachs lesion. Note the disappearance of the cortical thickness at the site of the lesion (+). the result was considered negative. Conventional radiography confirmed the diagnosis in 39 cases, whilst arthrography identified the lesion in 44 cases (Fig. 2). Arthro-CT (Fig. 2), which we took as our gold standard, gave the diagnosis of the lesion in 46 patients, revealing a defect on the posterior profile of the humeral head that had been missed by US in 2 cases. The findings at arthro-ct were negative in the remaining 14 cases, one of which had been considered positive at US. As regards diagnosis of a Hill-Sachs lesion, US thus had a sensitivity of 95.6%, a specificity of 92.8%, and a diagnostic accuracy of 95% (Table). There were no cases of lesions confined exclusively to the cartilage. The McNemar test.did not reveal any significant differences in the findings between US and arthro-ct. Discussion Data in the literature would seem to imply that it is possible to diagnose a Hill-Sachs lesion in 100% of cases by means of conventional radiography and particular projections (12). The presence of a thin line of thickening on the humeral head posterior to the major tubercle with the limb in internal rotation corresponds to impact of the cortical layer on the spongiosa of the humeral head against the glenoid cavity. In more severe lesions, such as those that occur after recurrent dislocations, a defect is regularly seen. We believe that the figure of 100% is a little too optimistic, particularly in the less frequent first-degree injuries where the bone damage is very modest. An alteration due to arthrosis could easily be confused with a first-degree Hill-Sachs lesion. In these cases, methods using contrast media are certainly able to provide a better identification of the lesion as the intra-articular radio-opaque fluid or air occupies even the smallest cartilaginar defect. Furthermore, conventional radiography cannot always show associated lesions such as Bankart s fracture or lesions of the capsule and the rotator cuff, all of which are frequently encountered in traumatic dislocations. In these cases arthrography, and particularly arthro-ct, are able to characterize the lesion precisely and to establish the direction of instability (4). US provides a good visualization of the articular cartilage: it shows up as a thin band of tissue with low echogenicity surrounding the bone profile of the humeral head with its total US reflection. It is possible to evaluate almost the entire surface of the humeral convexity during the examination, by means of manoeuvers of internal and external rotation. Comparison with the contralateral limb further facilitates the examination. The typical Hill-Sachs defect is revealed wherever there is a loss of normal convexity in the osteochrondral profile of the humeral head. It sometimes assumes the appearance of a triangular defect (with the base of the triangle facing the cutaneous plane) or the appearance of a small concavity with a wide radius. The ease with which the signs can be interpreted enables the technique to be used even by operators with little experience of US. The dynamic examination makes it possible to judge whether the lesion reaches the glenoid cavity during movement and the extent to which the mo- 525

5 L. PANCIONE ET AL. tion of the limb is hindered. Furthermore, US shows associated alterations such as articular effusions and injury to the components of the rotator cuff. In our study the single false-positive result was due to misinterpretation of the anatomical neck as a lesion. The 2 false-negative results were found in patients suffering from gross superimposed deformation resulting from arthrosis. The Hill-Sachs lesion does not call for specific therapy as treatment is aimed at correcting the instability that has brought about the lesion. In the view of other reports, the presence of a Hill-Sachs lesion is pathognomonic of shoulder instability (1). For this reason, a considerable diagnostic value is attributed to such a finding. Conclusion: We believe that US examination should be the screening method of choice for identifying patients with scapulohumeral instability because of its low cost, noninvasiveness, reproducibility, and above all, its proven sensitivity. REFERENCES 1. CALANDRA J. J., BAKER C. L. & URIBE J.: The incidence of Hill-Sachs lesions in initial anterior shoulder dislocations. Arthroscopy 5 (1989), DANZING I. A,, GREENWAY G. & RESNICK D.: The Hill-Sachs lesion. An experimental study. Am. J. Sports Med. 8 (1980), DEUTSCH A. L., RESNICK D., MINCK J. H. et al.: Computed and conventional arthrotomography of the glenohumeral joint. Normal anatomy and clinical experience. Radiology 153 (1984), FINKE L. I., BERG D. J. & DAVIS J. L.: Double-contrast computed tomography of the shoulder. J. Am. Osteopath. Assoc. 89 (1989), HERMODSSON I.: Rontgenologische Studien bei die traumatischen und habituellen Schultergelenkungen nach vorn und nach unten. Acta Radiol. Suppl. 20 (1934). 6. HILL H. A. & SACHS M. D.: The grooved defect of the humeral head. A frequently unrecognized complication of dislocations of the shoulder. Radiology 35 (1940), HOVELIUS I.: Anterior dislocation of the shoulder in teenagers and young adults. J. Bone Joint Surg. 69 (1987), MALGAIGNE D. M. P.: Les luxations scapulo-humtrales. Nouveau moyen de les distinguer des fractures du col de I humcrus. Nouvelle mtthode de rkduction. ExpCriences faites B 1 HGtel-Dieu. Gaz. MCd. Paris 3 (1832), RESNICK D., GOERGEN T. G. & NIWAYAMA G.: Physical injury. In: Diagnosis of bone and joint disorders, 2nd edn., p Edited by D. Manke. Saunders, Philadelphia ROZING P. M., DE BAKKER H. M. & OBERMANN W. R.: Radiographic views in recurrent anterior shoulder dislocation. Acta Orthop. Scand. 57 (1986), TIJMES J., LLOYD H. M. & TULLOS H. S.: Arthrography in acute shoulder dislocation. South Med. J. 72 (1979), WORKMAN T. L., BURKHARD T. K., RESNICK D. et al.: Hill- Sachs lesion. Comparison of detection with MR imaging, radiography, and arthroscopy. Radiology 185 (1992),

US finding of the shoulder (with live demonstration) 인제의대상계백병원 안재기

US finding of the shoulder (with live demonstration) 인제의대상계백병원 안재기 Shoulder US Biceps tendon & Rotator Cuff Long Head of Biceps Tendon Subscapularis tendon Supraspinatus tendon Infraspinatus tendon Teres