IntroductionShoulder instability is a common problem for military personnel due to the highly physical demands of work and training. This study assessed the pattern of glenoid labrum tears suffered by serving UK military personnel, the reliability of preoperative diagnostic methods (magnetic resonance arthrogram (MRA) vs clinical examination) and, finally, the outcomes of arthroscopic stabilisation in terms of satisfaction, pain, and return to sport and full deployment.MethodsRetrospective demographic and clinical data were collected for all patients within our unit who underwent arthroscopic shoulder stabilisation between September 2016 and January 2019. Patients underwent clinical examination for instability and subsequent imaging with MRA. For service evaluation, patient-reported outcome measure data and occupational outcome data were gathered preoperatively and postoperatively.Results41 military patients with shoulder instability were treated with arthroscopic stabilisation. 24.4% had an isolated anterior tear, and 41.5% had complex two-zone or pan-labral tears identified on arthroscopy. Clinical examination showed higher sensitivity, accuracy and negative predictive value for all labral tear patterns compared with MRA. Mean preoperative Oxford Shoulder Instability Score score was 18.58 (SE ±1.67) and mean postoperative score was 41.5 (SE ±1.13). 82.14% of the patients returned to full deployment during the study period and 85% had returned to sports.ConclusionComplex labral tear patterns are common in military personnel with shoulder instability, and clinical examination appears to be more effective than imaging at predicting injury pattern. Patients respond well to arthroscopic stabilisation with good rates of return to work and sport, regardless of chronicity of injury.