Background Shoulder instability in athletics is a common occurrence. However, there is controversy as to whether non-operative versus operative management is the most effective treatment in youth athletes. We systematically reviewed recurrence and return to play (RTP) after shoulder instability events in youth athletes based on surgical versus non-surgical intervention.

Design The systematic review concerned studies published before August 2016. Statistical analysis was performed to compare rates of recurrence for each extracted risk factor. Pooled ORs were analysed using random-effects meta-analysis.

Results 17 studies comprising 654 total shoulder instability events met the criteria for inclusion (438 men and 158 women; 507 cases traumatic and 20 atraumatic). Patients were grouped by whether they received non-operative or operative treatment. The primary non-operative group was more likely to have recurrence compared to the primary operative group (OR=13.41; 99% CI 3.60 to 49.93, p<0.001). Patients <14 years old in the primary non-operative group were less likely to recur compared to those aged ≥14 years (OR=0.16; 99% CI 0.06 to 0.43, p<0.001). The rate of recurrence in patients aged <14 years was high (44.44%). For RTP, there is evidence that RTP rates were higher for primary operative patients (95.3%) versus primary non-operative (41.3%, Z=6.12, p<0.001) and secondary operative patients (77.6%, Z=2.66, p=0.008).

Conclusions This meta-analysis summarises a mix of 17 acceptable quality level II and III prospective and retrospective cohort studies. Given the superior rates of recurrence and RTP, primary operative treatment for shoulder instability should be considered in youth athletes aged ≥14 years. Additionally, the recurrence rate in athletes aged <14 years is significant.