Background: It is unclear whether cutoff values for synovial C-reactive protein (syCRP) differ between acute and chronic infected as well as aseptic revision total hip arthroplasty (rTHA) and revision total knee arthroplasty (rTKA). We aimed to evaluate this and to compare the diagnostic accuracy of syCRP with serum C-reactive protein (seCRP), synovial white blood cell count, and polymorphonuclear leukocyte percentage (syPMN%).

Methods: A total of 387 revision arthroplasties (83 rTHA, 304 rTKA) were analyzed for seCRP, syCRP, synovial white blood cell, syPMN%, and culture results. Using the International Consensus Meeting 2018 periprosthetic joint infection (PJI) criteria, 189 (42 rTHA, 147 rTKA) were classified as infected, and 198 (41 rTHA, 157 rTKA) were aseptic revisions. Diagnostic accuracy was assessed using receiver operating characteristic curves and area under the curve (AUC).

Results: The optimal syCRP cutoff for PJI diagnosis among rTHA and rTKA < 90 days was 9.05 mg/L (AUC: 0.85) and 4.95 mg/L (AUC: 0.92), respectively. For > 90 days, the optimal syCRP cutoff was 3.00 mg/L (AUC: 0.90) for THA and 1.65 mg/L (AUC: 0.88) for TKA. In early and late acute PJI (n = 48), syCRP (45 of 48 = 94%) showed a higher positivity rate than seCRP (43 of 48 = 90%), WBC (38 of 48 = 79%), and PMN% (42 of 48 = 88%). A similar trend was observed in chronic PJI (n = 141): syCRP (110 of 141 = 78%), seCRP (104 of 141 = 74%), WBC (105 of 141 = 74%), and PMN% (101 of 141 = 72%). Among 189 confirmed PJIs, both WBC and PMN% were negative in 34 of 189 (18%), of which syCRP and seCRP were positive in 13 of 34 (38%) and 10 of 34 (29%), respectively.

Conclusions: The syCRP may be useful as a supplementary biomarker in cases with suspected PJI, but normal synovial cell count. Our data suggest that different cutoff values based on the type of joint and the chronicity of rTHA and rTKA are clinically useful.