摘要：Periprosthetic joint infection (PJI) is a catastrophic complication of shoulder arthroplasty. Commonly used surgical treatments include one- or two-stage revision, but their effectiveness in controlling infection is uncertain. We aimed to compare re-infection (recurrent and new infections) rates; clinical measures of function and pain; and noninfection complication rates of one- and two-stage revision surgery for shoulder PJI using a systematic review and meta-analysis. We searched MEDLINE, Embase, Web of Science, and The Cochrane Library to February 2018. Longitudinal studies conducted in patients with shoulder PJI treated exclusively by one- or two-stage revision were eligible. No clinical trials were identified. Re-infection rates were meta-analysed using random-effect models after arcsine transformation. The re-infection rate (95% CI) in pooled analysis of eight one-stage studies (147 participants) was 5.3% (1.4-10.6). The corresponding rate for 27 two-stage studies (351 participants) was 11.5% (6.0-18.1). Postoperative clinical measures of function and pain were not significantly different between the two revision strategies. The pooled noninfection complication rate (95% CI) for one-stage and two-stage revision was 12.1% (6.1-19.5) and 18.9% (8.4-31.9) respectively. New evidence suggests one-stage revision is at least equally as effective as the two-stage in controlling infection, maintaining joint function, and improving complications in shoulder PJI.
