• BACKGROUND
    • The diagnosis of periprosthetic infection in shoulder arthroplasty remains a challenge. Conventional methods for evaluating periprosthetic joint infections are poor because of lower-virulence organisms affecting the shoulder. The aim of our systematic review was to evaluate the diagnostic accuracy of preoperative arthroscopic tissue cultures compared with tissue biopsy samples obtained at the time of revision surgery.
  • MATERIALS AND METHODS
    • We conducted a systematic search of the MEDLINE, Embase, and Cochrane Central databases. The inclusion criteria consisted of studies that used arthroscopy to obtain preoperative tissue cultures for the diagnosis of shoulder arthroplasty infections. Studies were excluded if they obtained non-arthroscopic tissue samples. We reported the sensitivity, specificity, positive predictive value, and negative predictive value. Culture findings from the arthroscopic biopsy samples were also compared with conventional tests of fluoroscopy-guided joint aspiration and serum inflammatory marker testing (positive erythrocyte sedimentation rate or C-reactive protein) within the included studies. A meta-analysis was performed to assess the overall diagnostic accuracy of the studies.
  • RESULTS
    • Our search strategy yielded 795 potentially relevant publications; 572 underwent title and abstract screening, and 14 studies underwent full-text review, of which 7 were included in our systematic review. The studies represented a balance of shoulder arthroplasty types, including anatomic total shoulder arthroplasty (n = 75, 38%), reverse total shoulder arthroplasty (n = 60, 30%), and hemiarthroplasty (n = 64, 32%). There were 56 of 120 arthroscopic procedures that returned positive tissue culture findings compared with 64 of 157 positive open biopsy culture findings obtained from revision surgery. The meta-analysis total for sensitivity and specificity for all studies combined indicated that arthroscopic tissue cultures (0.76 [95% confidence interval (CI), 0.57-0.88] and 0.91 [95% CI, 0.79-0.97], respectively) were superior to both aspiration (0.15 [95% CI, 0.03-0.48] and 0.93 [95% CI, 0.65-0.99], respectively) and a positive erythrocyte sedimentation rate or C-reactive protein level (0.14 [95% CI, 0.02-0.62] and 0.83 [95% CI, 0.56-0.95], respectively) in diagnosing periprosthetic shoulder infections.
  • CONCLUSIONS
    • Our systematic review demonstrated that preoperative arthroscopic tissue biopsy used for microbiology cultures accurately predicts intraoperative culture findings obtained during revision surgery with high sensitivity and specificity. Additionally, arthroscopy appears to be superior to conventional techniques of joint aspiration and inflammatory marker testing. Therefore, arthroscopic tissue cultures may be an emerging useful tool to help guide the management of periprosthetic infections in shoulder arthroplasty.