Keywords. Shoulder; dislocation; instability; skeletal immaturity; arthroscopic Bankart repair; outcomes. Introduction. Non-operative management of traumatic shoulder instability in children has a recurrence rate of up to 100%. Short-term outcomes of surgery in adults results has a quoted recurrence rates of around 10%. The aim of this study was to examine the surgical outcomes of adolescent patients (aged 13 to 18 years) undergoing arthroscopic stabilisation for shoulder instability. Methods. All patients had a labral tear sustained whilst participating in contact sports (84% rugby). Atraumatic, primary joint hyperlaxity and dyskinetic causes were excluded. All patients had a primary arthroscopic stabilisation. Patients were reviewed in clinic or contacted by post with a standardised outcome proforma. Sixty one shoulders in 57 patients met the inclusion criteria and were included. Mean follow up was 22 months, mean age was 16.8 (range 13 to 18), and 98% were male. All were Stanmore type 1 pathology. 16% had a concomitant SLAP repair, and 16% had a posterior labral tear. Results. 31% (19) reported recurrent dislocation, and 11 patients required further surgery. This is significantly higher than published series for adults, despite the senior author being a tertiary specialist. Despite the high recurrence rate, the median improvement in shoulder function was 90% and the median VAS for pain was 0. The majority of patients (61%) had a full return to pre-injury sporting activities, while 23% returned to decreased sporting activity and 16% stopped. The mean post-operative Oxford instability score was 26.8 (SD 12.9 range 12 to 48). 90% of recurrent dislocations occurred whilst playing rugby, but other than this, we could not identify any significant risk factors for the 19 shoulders that had recurrent dislocations (gender, type of sport, hyper laxity, type of tear). Conclusions. This study demonstrates that adolescent patients have a high risk of recurrent dislocation following arthroscopic stabilisation. Patients should be appropriately counselled and alternative procedures such as the coracoid transfer should be considered. Level of evidence: IV