Outcomes of arthroscopic Hill-Sachs remplissage and anterior Bankart repair: a retrospective controlled study including ultrasound evaluation of posterior capsulotenodesis and infraspinatus strength assessment.

BACKGROUND: Hill-Sachs lesions are compression fractures that result from shoulder dislocation. They involve "engaging" the humeral head on the anterior glenoid rim when the arm is abducted and externally rotated. The defect grows as the number of dislocations increases.

HYPOTHESIS: Arthroscopic remplissage and anterior Bankart repair do not significantly affect infraspinatus strength while ensuring healing of the capsulotenodesis.

STUDY DESIGN: Cohort study; Level of evidence, 3.

METHODS: Sixty-one patients with traumatic anterior shoulder instability treated by arthroscopic Bankart repair and Hill-Sachs remplissage at least 24 months previously were compared with a control group of 40 healthy participants. Preoperative imaging included magnetic resonance imaging for Bankart lesion identification and computed tomography to quantify the humeral head defect. Active range of motion and clinical scores (Walch-Duplay, Constant-Murley, and Rowe) were assessed. External rotation (ER) and internal rotation (IR) were measured with arm at the side (ER1 and IR1) and abducted at 90° (ER2 and IR2). Infraspinatus strength was assessed with the scapula free (infraspinatus strength test [IST]) and retracted (infraspinatus scapula retraction test [ISRT]). Infraspinatus tenodesis and posterior capsulodesis healing were evaluated by ultrasound (US).

RESULTS: The follow-up median was 39.5 months (range, 24-56 months). One patient experienced a recurrence of instability at 34 months. In the remplissage patients, ER1 was significantly lower in the affected compared with the unaffected shoulder (P < .001). Mean IST and ISRT strength values did not show differences between sides. The mean Constant-Murley score rose from 62.9 ± 7.1 to 90 ± 5.2 (P < .0001). The Walch-Duplay and Rowe scores were excellent in 23 (78.6%), good in 6 (17.8%), and poor in 1 patient (both scores). The remplissage group had significantly lower ER1 (P < .001), ER2 (P < .001), and IR2 (P < .01) values compared with the control group. Differences in IST and ISRT between the groups were not significant. Capsulotenodesis healing and filling of the Hill-Sachs defect were confirmed by dynamic US in all subjects.

CONCLUSION: Arthroscopic remplissage is a reliable approach to Hill-Sachs lesions. The ER and IR restriction does not significantly affect quality of life. Infraspinatus strength recovery is satisfactory even compared with healthy subjects. Ultrasound examination allows accurate evaluation of capsulotenodesis healing.