2015 ISAKOS Biennial Congress ePoster #2309

Evaluation of the Size of Hill-Sachs Lesions by 3D-CT: Influence of the Recurrences on Their Size

Ritsuro Ozaki, MD, Osaka, Osaka JAPAN
Shigeto Nakagawa, MD, Osaka, Osaka JAPAN
Take Yasuhiro, MD, PhD, Osaka, Osaka JAPAN
Ryo Iuchi, MD, Osaka, Osaka JAPAN
Naoko Mizuno, MD, Toyonaka, Osaka JAPAN
Tatsuo Mae, MD, PhD, Suita, Osaka JAPAN
Minoru Yoneda, MD, DMSc, Kashiwa-city, Chiba JAPAN

Yukioka hospital, Osaka city, Osaka, JAPAN

FDA Status Not Applicable

Summary: The accuracy of evaluating Hill-Sachs lesions on 3D-CT was analyzed, and the influence of the number of dislocations and subluxations on the prevalence and size of Hill-Sachs lesions was investigated. CT was highly accurate to evaluate osseous Hill-Sachs lesions, and episodes of complete dislocation were shown to be important for the occurrence and enlargement of Hill-Sachs lesions.

Rate:

Abstract:

Background

In shoulders with traumatic anterior instability, a large Hill-Sachs lesion is one of the most important risk factors for postoperative recurrence. While evaluation of Hill-Sachs lesions using computed tomography with 3-dimensional reconstruction (3D-CT) is increasing, there have been no previous reports measuring the size of Hill-Sachs lesions. In addition, there is no consensus regarding the occurrence and enlargement of Hill-Sachs lesions.

Purpose

The purposes of the present study were to analyze the accuracy of evaluating Hill-Sachs lesions on 3D-CT in comparison with arthroscopic findings and to investigate the influence of the number of dislocations and subluxations on the prevalence and size of these lesions.

Methods

The subjects were 142 shoulders (30 with primary instability and 112 with recurrent instability) who underwent arthroscopic Bankart repair for traumatic anterior shoulder instability, in whom Hill-Sachs lesions were evaluated preoperatively by CT scanning. To assess the size of Hill-Sachs lesions, we measured the length and width on the en face 3D-CT image that showed the Hill-Sachs lesion, while the depth was measured on the axial image obtained perpendicular to the longitudinal axis of the humeral shaft. First, the prevalence of Hill-Sachs lesions was compared with the arthroscopic findings. Then, the size of Hill-Sachs lesions confirmed by arthroscopy was re-measured using the previous CT data. In addition, the relation of Hill-Sachs lesions with the number of dislocations and subluxations was investigated.

Results

Hill-Sachs lesions were detected in 90 shoulders by initial CT evaluation, and were found in 118 shoulders at arthroscopy. The Hill-Sachs lesions missed by initial CT were 15 chondral lesions and 13 osseous lesions. However, all 103 osseous Hill-Sachs lesions were detected by reviewing the CT data. The sensitivity, specificity, and accuracy of the initial CT evaluation by comparison with arthroscopic findings as the gold standard were 76.3%, 100%, and 80.3%, respectively. In patients with primary subluxation, the prevalence of Hill-Sachs lesions was 26.7%, and the mean length, width, and depth of the lesions was 9.0%, 5.3%, and 2.1% respectively, while the corresponding numbers for primary dislocation were 73.3%, 27.7%, 14.8%, and 7.0%, showing statistically significant differences. Among all 142 shoulders the corresponding numbers were respectively 56.3%, 20.7%, 11.2%, and 4.8% in patients who had subluxations but never a dislocation; 83.3%, 33.4%, 19.1%, and 7.6% in patients with one episode of dislocation; and 87.5%, 46.8%, 22.2%, and 10.2% in patients with two or more episodes, showing statistically significant differences among these 3 groups. There were no differences in lesion measurements in relation to the number of subluxations.

Conclusion

CT was a useful imaging modality for evaluation of osseous Hill-Sachs lesions.
In shoulders with traumatic anterior instability, most Hill-Sachs lesions occurred and enlarged at the time of complete dislocation, and large lesions occurred due to one episode of dislocation. On the other hand, recurrent subluxation did not affect the prevalence and size of these lesions, so episodes of complete dislocation were shown to be important for the occurrence and enlargement of Hill-Sachs lesions.