LAUSR

BACKGROUND Shoulder periprosthetic joint infection (PJI) is a significant complication after arthroplasty with high morbidity. An evidence-based algorithm for treatment of shoulder PJI is lacking in current practice. The purpose of this systematic review and meta-analysis was to understand and compare the role of single and two-stage shoulder arthroplasty revision for PJI. METHODS A comprehensive literature review was performed to identify all studies related to shoulder arthroplasty for PJI in PubMed, Scopus, EMBASE. Inclusion criteria for this systematic review were studies that reported on single or two-stage revision with infection eradiation and a minimum follow-up of 12-months and minimum of five patients for analysis. A random effect meta-analysis was performed and heterogeneity was assessed with Cochrane Q and I2. RESULTS A total of 13 studies reporting on single-stage revision and 30 studies reporting on two-stage revision were included in final analysis. The majority of positive cultures from single-stage revision for PJI resulted with C. acnes with 113/232 (48.7%) compared to 190/566 (33.7%) for two-stage revision. However, there was a lower percentage of MRSA positive cultures with 2.5% for single-stage compared to 9.7% for two-stage revision. The overall pooled random effect reinfection incidence was 0.05 (95% CI 0.02-0.08), with moderate heterogeneity (I2 = 34%, p=0.02). The reinfection rate for single-stage was 6.3% and 10.1% for two-stage revision but this was not significant (Q=0.9 and p=0.40). CONCLUSION Based on a systematic review with meta-analysis, single-stage revision for shoulder PJI is an effective treatment. Indeed, our analysis showed single-stage to be more effective than two-stage but this is likely confounded by a treatment bias given the higher propensity of virulent and drug resistant bacteria treated with two-stage in the published literature. This implies that shoulder surgeons treating PJI can be reassured of a low recurrence rate (6.3%) when using single-stage treatment for C. acnes or other sensitive, low virulence organisms. LEVEL OF EVIDENCE Level IV; Systematic Review.