Tomographic evaluation of Hill-Sachs lesions: is there a correlation between different methods of measurement?

Background Several methods are currently available to evaluate and quantify the glenoid or humeral bone loss; however, none is universally accepted, particularly in the case of Hill-Sachs (HS) lesions. Purpose To establish whether there is correlation among different methods of measuring HS lesions, and to investigate the correlation between glenoid bone loss and the various HS lesion measurements and to assess the inter-observer reliability of such measurements. Material and Methods We assessed computed tomography (CT) or arthro-CT scans taken from individuals with recurrent anterior glenohumeral dislocation. The scans were independently assessed by two examiners. The parameters assessed were as follows: HS lesion width and depth on the axial and coronal planes, articular arc loss on the axial plane, and percentage of glenoid bone loss on the sagittal plane. Results Scans from 50 shoulders were assessed. The percentage of articular arc loss and HS lesion width on the axial plane were the only measurements that exhibited strong correlation (r=0.83; P<0.001). The values of the correlation coefficient corresponding to HS lesion depth on the coronal plane were the lowest. Most of the measurements exhibited moderate correlation. The inter-examiner reliability was good relative to all measurements except for HS lesion width and depth on the coronal plane, for which it was moderate. Conclusion The measurements of articular arc loss and HS lesion width on the axial plane exhibited strong correlation. The inter-examiner reliability relative to articular arc loss, HS lesion width and depth on the axial plane, and glenoid bone loss was good.