Mark B. Coventry Award: Synovial C-reactive Protein: A Prospective Evaluation of a Molecular Marker for Periprosthetic Knee Joint Infection

Abstract

C-reactive protein (CRP) serum assays are a standard element of the diagnostic workup for periprosthetic joint infection (PJI). However, because CRP is a marker for systemic inflammation, this test is not specific to PJI. Our purpose was to assess whether synovial fluid and serum assays alone could differentiate between infected and uninfected revision knee arthroplasties and to determine which of these methods had the greatest diagnostic accuracy. We collected synovial fluid specimens from 66 patients undergoing revision total knee arthroplasty. Patients were judged uninfected or infected by standardized criteria. Synovial CRP levels were measured using an individual CRP assay (15 samples; 10 infected, five uninfected) and a multiplex immunoassay platform (59 samples; 25 infected, 34 uninfected). Results from preoperative standard serum CRP assays conducted were also collected (55 samples; 25 infected, 30 uninfected). Sensitivity, specificity, and receiver operating characteristic curve analyses were performed for each assay with a diagnosis of infection based on previously established criteria. Synovial CRP concentrations differed between infected and uninfected joints in the multiplex and serum analyses. The area under the curve was 0.84 for the individual assay, 0.91 for the multiplex assay, and 0.88 for the serum CRP assay. Sensitivity and specificity were 70.0% and 100.0% for the individual enzyme-linked immunosorbent assay, 84.0% and 97.1% for the multiplex assay, and 76.0% and 93.3% for the serum CRP assay. An assay measuring CRP in synovial fluid may be more accurate in diagnosing PJI than the standard serum CRP assay. We believe such an assay holds promise as a new diagnostic marker for PJI. Level II, diagnostic study. See Guidelines for Authors for a complete description of levels of evidence.