Soterios Gyftopoulos MD, Abstract Co-Author: Nothing to Disclose

Avner Yoez Yemin MD, Presenter: Nothing to Disclose

James S. Babb PhD, Abstract Co-Author: Nothing to Disclose

Luis Saura Beltran MD, Abstract Co-Author: Nothing to Disclose

Jenny T. Bencardino MD, Abstract Co-Author: Nothing to Disclose

The purpose of this study was to assess the association between an engaging Hill-Sachs lesion (HSL) and its size and location, and the amount of glenoid bone loss utilizing MRI data.

33 consecutive patients (32 male, 1 female) with a history of anterior shoulder dislocation who underwent arthroscopy by 3 surgeons and had preoperative MRI (1.5T, 3T) at our institution were collected over a 9-month period. 11 patients had evidence of an engaging HSL on physical examination (EHL), while 22 did not (NHL). A single blinded reader reviewed each imaging study and documented the presence and size of glenoid bone loss (GBL; % of bone loss along the articular surface) and HSL, and presence of a labral tear (LT). The reader also measured a revised biceps angle (BA) defined as the angle between the biceps groove and medial margin of the HSL; an indicator of the HSL location/medial extent. Statistical analysis included Mann-Whitney, logistic regression, and Pearson correlation tests.

There was a statistically significant (SS) difference in the amount of GBL in the EHL group compared to the NHL group (mean 20.7% vs. 8.1%, p=0.002). There was no SS difference between any of the dimensions or overall surface area (SA) of the HSL in the EHL and NHL groups (SA: 3.60 vs. 3.23 cm3, p=0.272). There was no difference in the presence of labral tears. The BA was also larger in the EHL group, but not SS (mean 154.5° vs. 143.9°, p=0.069). When looking at an individual's data, there was SS utility in predicting an EHL when using GBL (AUC=0.808, p=0.007), but no SS for BA (AUC=0.698, p= 0.110) or SA (AUC= 0.570, p= 0.708). No SS association was found between any two of the imaging findings. Diagnostic accuracy was maximized for a specific GBL threshold of > 20%. Utilizing this threshold achieved a 63.6% (7/11) sensitivity and 95.5% (21/22) specificity.

There is a statistically significant association between the amount of glenoid bone loss and the presence of an engaging Hill-Sachs lesion with no significant associations with the location and size of the HSL. GBL of >20% appears to be a moderately sensitive, highly specific threshold for predicting EHL.

The amount of GBL plays an important role in the pathogenesis of an EHL; augmentation of significant GBL should be considered in the treatment, possibly first line, of this subset of patients.

Gyftopoulos, S, Yemin, A, Babb, J, Beltran, L, Bencardino, J, Engaging Hill-Sachs Lesion: Can It Be Predicted on MRI?.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12024353.html