Abstract
Periprosthetic joint infection (PPJI) following shoulder arthroplasty is uncommon, with an overall rate of 0.98%. However, the rates following revision arthroplasty and reverse arthroplasty are much higher. Given the rapid increase in the prevalence of shoulder arthroplasty and the increasing revision burden, the cost of PPJI to society will likely increase substantially. The most common organisms found in PPJI following shoulder arthroplasty are Staphylococcus aureus, coagulase-negative Staphylococcus, and Propionibacterium acnes (P. acnes). P. acnes is especially common in males. Traditional testing for PPJI includes aspiration, white blood cell count (WBC), erythrocyte sedimentation rate (ESR), and c-reactive protein (CRP). Aspiration often yields a dry tap and when fluid is obtained for culture, a positive result is helpful but a negative result does not rule out PPJI. Although WBC, ESR, and CRP are often positive with PPJI in the lower extremity, they are most often negative in shoulder PPJI. Although bone scans and WBC labeled scans are used, they are expensive and have low sensitivity and specificity.
New testing and techniques have been reported in an attempt to improve sensitivity and specificity for PPJI. These techniques can be divided into tests on serum, synovial fluid, and tissue. Serum Interleukin-6 (IL-6) is highly specific (94%) for shoulder PPJI but has low sensitivity (14%). Synovial fluid can be tested for leukocyte esterase using a simple and cheap technique. In lower extremity PPJI it has shown to be helpful. It is not as helpful in shoulder PPJI with 30% sensitivity and 67% specificity. Alpha defensin has been reported to be more sensitive (63%) and as specific (95%) as traditional techniques but still lacks predictive value. Testing for specific cytokines (IL-2, IL-6, TNF- α) within synovial fluid is not widely used as yet but has shown promise with 80% sensitivity and 90% specificity. Obtaining tissue for culture and other testing is probably the most reliable way of confirming PPJI for the shoulder. Frozen sections taken at the time of revision can be helpful but is very pathologist dependent and institution specific. With a dedicated musculoskeletal pathologist, the finding of 10 or more WBCs per high powered field has been reported to be 72% sensitive and 100% specific for P. acnes and 63% sensitive and 100% specific for other organisms. Cultures from arthroscopic tissue biopsy have also been found to have high sensitivity (100%) and specificity (100%). Genetic testing of tissue biopsy specimens (PCR/NGS) has recently been reported and shows great promise. The significance of positive cultures and other tests, especially for P. acnes is unclear. There is a high rate of positive intra-operative cultures in primary cases of shoulder arthroplasty. In addition, intra-operative cultures taken at the time of revision, even in cases in which infection is not suspected, are frequently positive for P. acnes with weak correlation with rates of post-operative clinical infection.
In conclusion, shoulder PPJI is a difficult problem to deal with. The definition of shoulder PPJI is currently unclear and further study is needed. There is no ideal test to confirm it. A reasonable approach is to aspirate for culture, and perform serum tests for WBC, ESR, and C-reactive protein. If any of these is positive in the setting of a painful arthroplasty, PPJI should be assumed until proven otherwise. Operative tissue cultures are probably the most reliable test but the clinical significance is not always obvious. Synovial fluid cytokine profiles and tissue PCR/NGS show promise for the future.
