Abstract
BACKGROUND: Until today, a reliable diagnostic discrimination between periprosthetic joint infections (PJI) and aseptic failure (AF) after total joint arthroplasty (TJA) remains challenging. Nearly all recent research focused on synovial markers to be elevated in PJI rather than in AF patients. In this study, synovial bone sialoprotein (sBSP) was investigated in PJI and AF arthroplasty patients before revision surgery. METHODS: sBSP and C-reactive protein (CRP) were determined in synovial fluid samples of PJI (n = 13) patients fulfilling the MSIS criteria and AF (n = 25) patients. Beside descriptive analysis and comparison, computed statistics determined the area under the receiver operating characteristics curve (AUC) to evaluate the discrimination ability of the tested synovial markers. RESULTS: In patients with PJI according to the MSIS criteria, mean sBSP was significantly lower: 14.8 ng/ml (95% CI 5.5-24.1) vs. 38.2 ng/ml in the AF group (95% CI 31.1-45.3), p ≤ 0.001. Conversely, mean sCRP was significantly higher in PJI patients: 8.4 μg/ml (95% CI 0-17.2) vs. 1.8 μg/ml in the AF group (95% CI 0.9-2.8), p = 0.032. The AUC of sCRP in PJI patients was 0.71. The AUC of sBSP in AF revision arthroplasty patients was 0.83. The detection of osteolyses was not associated with higher sBSP concentrations. CONCLUSIONS: Considering the MSIS criteria, significantly higher sBSP concentrations were found in synovial fluid samples of AF compared to PJI patients. sCRP showed only fair, sBSP good discrimination potential. If it is not clear whether PJI is present or not, sBSP may be considered as an add-on synovial marker.