The goal of the study was to compare early clinical and radiological results after open and arthroscopic Latarjet stabilisation in anterior shoulder instability.

Fourty eight out of 55 shoulders (87%) operated on between 2006 and 2011 using an open technique composed the OPEN group and 62 out of 64 shoulders (97%) operated on arthroscopically between 2011 and 2013 composed the ARTHRO group. The mean follow-up was 54.2months in OPEN and 23.4months in ARTHRO. Intraoperative data, clinical and radiological (X-ray and CT) results were assessed.

Average time of surgery was significantly shorter in ARTHRO than OPEN: respectively 110 and 120minutes. Six intraoperative complications (12.5%) in OPEN and 5 (8.1%) in ARTHRO were reported. Clinical results in OPEN and ARTHRO were respectively: satisfaction rate 96.8 and 91.9%, Walch-Duplay score 83.9 and 76.7 and Rowe score, 87.8 and 78.9 (P>0.05). Loss of external rotation was significantly higher in ARTHRO than in OPEN: 14° versus 7° (P>0.05). Recurrence was reported in 3 cases in both groups: 6.2% in OPEN and 4.8% in ARTHRO. The rate of pseudoarthrosis was insignificantly higher in OPEN: 3 (7.14%) versus 1 in ARTHRO (1.67%). Total graft osteolysis was significantly higher in OPEN, 0 in ARTHRO and 5 in OPEN (11.63%). The partial graft osteolysis around the superior screw was reported in 32 patients (53.3%) in ARTHRO versus 10 patients (23.81%) in OPEN. The bone block was placed flush with the glenoid rim in 27 shoulders (65.85%) in OPEN and in 25 (41.67%) in ARTHRO (P>0.05). Overhanging graft position was insignificantly more frequent in OPEN, 8 shoulders (19.51%) comparing to 10 (16.67%) in ARTHRO. The screws were placed more parallel to the glenoid surface in ARTHRO, the angles were 12.3° for the inferior screw and 12.6° for the superior one. In OPEN, they were respectively 15° and 17° and the difference was significant for the superior ones. The superior screw in ARTHRO were significantly more often loosened in 8 cases (13.33%) comparing to 0 cases in OPEN.

Both techniques showed satisfactory and comparable results. The clinical results were slightly better for open procedure with significant difference in Rowe score and loss of external rotation. Radiologically better positionning of the screws and graft healing were reported in arthroscopic technique, however osteolysis around the superior screw was significantly more often than in open procedure.