The predictive value and reliability of ultrasound-guided synovial aspiration and biopsies for diagnosing periprosthetic shoulder infections

Introduction Preoperative diagnosis of periprosthetic shoulder infections (PSI) is difficult. Infections are mostly caused by low virulence bacteria and patients do not show typical signs of infection. The aim of this study was to determine the diagnostic value and reliability of ultrasound-guided biopsies for cultures alone and in combination with multiplex polymerase chain reaction (mPCR), serum markers, and/or synovial markers for the preoperative diagnosis of PSI in patients undergoing revision shoulder surgery. Materials and methods A prospective explorative diagnostic cohort study was performed including 55 patients undergoing revision shoulder replacement surgery. A shoulder puncture was performed preoperatively before incision to collect synovial fluid for mPCR analysis and for measurement of interleukin-6, calprotectin, white blood cell count (WBC), and polymorphonuclear cells. Also prior to revision surgery, six ultrasound-guided synovial tissue biopsies were collected for culture and two for mPCR analysis. A blood sample was obtained to determine serum C-reactive protein, WBC, and erythrocyte sedimentation rate. Six routine care tissue biopsies were taken during revision surgery and served as reference standard. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV; the primary outcome measure), and accuracy were calculated for ultrasound-guided biopsies, blood and synovial markers, mPCR, and combinations thereof. Results Routine tissue cultures were positive for infection in 24 patients. Cultures from ultrasound-guided biopsies diagnosed infection in 7 of these patients, yielding a sensitivity, specificity, PPV, NPV, and accuracy of 29.2%, 93.5%, 77.8%, 63.0%, and 65.6%, respectively. The best diagnostic value was found for the combination of ultrasound-guided biopsies for culture, synovial WBC, and calprotectin with a sensitivity of 69.2%, specificity of 80.0%, PPV of 69.2%, and NPV of 80.0%. Conclusion Ultrasound-guided biopsies for cultures alone and in combination with mPCR, and/or blood and/or synovial markers are not reliable enough to use in clinical practice for the preoperative diagnosis of PSI. Level of evidence Diagnostic study level II.