One Versus 2-Stage Revision for Shoulder Arthroplasty Infections: A Systematic Review and Analysis of Treatment Selection Bias

JBJS Rev. 2021 Sep 9;9(9). doi: 10.2106/JBJS.RVW.20.00219.


Background: There is controversy regarding the optimal treatment for infection following shoulder arthroplasty. The purpose of this systematic review is to analyze the bias in treatment selection, infection clearance rates, and functional outcomes after 1 versus 2-stage revision surgery for periprosthetic shoulder infections.

Methods: A systematic search strategy following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines was conducted in 4 phases. Articles were identified using MEDLINE (PubMed), Embase (Elsevier), and Cochrane Library databases with Boolean search terms related to infection after shoulder arthroplasty. Included articles were analyzed for quality, and data were extracted for use. Preoperative treatment selection bias was analyzed and postoperative infection clearance rates and functional outcome scores were compared between 1 and 2-stage revision surgery for periprosthetic shoulder infection.

Results: Overall, 163 1-stage shoulder procedures and 289 2-stage shoulder procedures were included in the analysis. Cutibacterium acnes was the organism most frequently grown on culture (37%) followed by coagulase-negative Staphylococcus (19%). The overall infection clearance rate was 95.6% for 1-stage and 85.2% for 2-stage procedures. In a comparison of the change in outcome scores from preoperatively to postoperatively between 1-stage and 2-stage revision, the Constant-Murley Score (CMS) improved 21.0 points (1-stage) versus 22.8 points (2-stage), the American Shoulder and Elbow Surgeons (ASES) score improved 26.2 points versus 33.6 points, and the Simple Shoulder Test (SST) score improved 3.5 points versus 6.4 points, respectively. Overall, 23 of 26 studies cited a reason for selection of a 1 versus 2-stage procedure, which was due to standard treatment protocol in 10 studies, based on the timing of the infection (acute versus subacute versus chronic) in 5, due to a combination of factors (age, comorbidities, intraoperative appearance, adequacy of debridement, bone loss) in 6, and due to preoperative identification of a specific organism in 2.

Conclusions: One-stage revisions resulted in higher infection clearance rates; however, 2-stage revisions resulted in greater functional improvement as measured with ASES and SST scores from the preoperative assessment to the final postoperative follow-up. The decision between 1 and 2-stage revisions is due to a combination of factors including pathogen type, timing of infection, findings on the preoperative clinical examination, the patient's own decision, the surgeon's preference, and the intraoperative soft-tissue/osseous appearance, which may have biased the overall results. There is no consensus in the literature on the decision between 1 and 2-stage treatment for periprosthetic shoulder infection, which is based on a combination of factors. However, both treatment strategies are effective in treating periprosthetic shoulder infection.

Level of evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Publication types

  • Systematic Review

MeSH terms

  • Arthroplasty, Replacement, Shoulder* / adverse effects
  • Arthroplasty, Replacement, Shoulder* / methods
  • Humans
  • Prosthesis-Related Infections* / diagnosis
  • Selection Bias
  • Shoulder Joint* / surgery
  • Treatment Outcome