Die ventrale Schulterluxation : Assoziation zwischen Bankart-Läsion und Hill-Sachs-Delle

Background: The glenohumeral joint has the highest range of motion but also the highest rate of luxation in the human body. Shoulder luxations are typically anterior luxations which result from a fall onto the externally rotated and abducted arm. Separation of the humeral head and glenoid causes typical injuries on both elements of the glenohumeral joint: a cartilaginous or bony detachment of the glenoid rim (Bankart lesion) and a compression fracture of the humeral head (Hill-Sachs lesion). Bankart lesions are diagnosed in 47 - 80 % of all shoulders after luxation [Griffith et al. 2008, Tylor and Aciero 1997]. Hill-Sachs-lesions appear in 47 - 80 % of all shoulders after dislocation and in up to 100 % after recurrent dislocation [Bushnell et al. 2008b]. The aim of this study is to quantify the association between existence of both lesions after anterior shoulder dislocation and investigate an association between sizes of both lesions. Materials and Methods: The collective consisted of patients who were treated with shoulder luxation (S43.0) at the UKA between 2006 and 2013. Inclusion criteria were anterior luxation, presence of MRI-images of the affected shoulder and absence of other injuries or former operations of the shoulder. Bankart lesions were classified as labral or bony lesions. Hill-Sachs lesions were measured on axial images and classified with the help of a modified classification by Calandra et al. into grades I - III [Calandra et al. 1989]. The software SAS®, Version 9.2 was used for statistical analysis. Results: 105 patients with 110 affected shoulders were enclosed in this study (85 male, 20 female patients) with an age of 35,6 ± 16,5 (16 - 80) years. 77 shoulders had experienced primary dislocations, 33 recurrent dislocations. Bankart lesions were identified in 73 % of all shoulders, with 71 % being cartilaginous lesions and 29 % being bony lesions. Hill-Sachs lesions were diagnosed in 82 % of all shoulders with 27 % affecting only the articular surface of the humeral head (grade I), 66 % being small subchondral lesions (grade II) and 7 % being large subchondral defects. Hill-Sachs lesions had a size of 4,3 ± 3,2 (0 - 15,1) mm in depth and 13,1 ± 4,0 (5,6 - 29,7) mm in width. In 75 shoulders both lesions where found, while 13 showed no lesion. In 17 shoulders isolated Hill-Sachs lesions where diagnosed and in 5 shoulders isolated Bankart lesions. These findings resulted in an odds ratio of 11,47 (95% KI 3,60 - 36,52, p < 0,001) for the concomitance of both lesions. A correlation between the type of Bankart lesion (labral or bony) and the size of Hill-Sachs lesion (grade I - III) was proven (Spearman correlation coefficient 0,34, 95% KI 0,16 - 0,49, p < 0,001). Logistic regression showed an association between the existence of Bankart lesions and the depths of concomitant Hill-Sachs lesions (OR = 1,24, 95% KI 1,02 - 1,52, p = 0,033). The width of Hill-Sachs lesions was not associated with the existence of Bankart lesions. Hill-Sachs lesions in shoulders with bony Bankart lesions were significantly deeper and wider than in shoulders with labral lesions (p < 0,001). Conclusion: If either a Bankart lesion or a Hill-Sachs lesion is diagnosed in a shoulder after anterior luxation, it is approximately 11 times more likely to find the second type of lesion than an isolated lesion. There is a positive correlation between the sizes of Bankart and Hill-Sachs lesions. In the last years it has become apparent that Bankart as well as Hill-Sachs lesions have to be considered for the planning of operative treatment of anterior shoulder dislocation. It is of great importance to evaluate both lesions together as bipolar injuries. To this end exact diagnostics of both lesions are needed. The findings of this study can help to improve diagnostics of Bankart and Hill-Sachs lesions in MRI. We would like to draw attention on the association of Bankart lesion and Hill-Sachs lesion and to point out the importance of considering both lesions in treatment planning