Multiplex Antibody Detection for Noninvasive Genus-Level Diagnosis of Prosthetic Joint Infection

Abstract

We developed and evaluated a multiplex antibody detection-based immunoassay for the diagnosis of prosthetic joint infections (PJIs). Sixteen protein antigens from three Staphylococcus species (Staphylococcus aureus, Staphylococcus epidermidis, and Staphylococcus lugdunensis) (8 antigens), Streptococcus agalactiae (4 antigens), and Propionibacterium acnes (4 antigens) were selected by comparative immunoproteomics using serum samples from PJI cases versus controls. A bead-based multiplex immunoassay that measured serum IgG against purified, recombinant forms of each of the 16 antigens was developed. We conducted a prospective study to evaluate the performance of the assay. A PJI was defined by the presence of a sinus tract and/or positive intraoperative sample cultures (at least one sample yielding a virulent organism or at least two samples yielding the same organism). A total of 455 consecutive patients undergoing revision or resection arthroplasty (hip, 66.3%; knee, 29.7%; shoulder, 4%) at two French reference centers for the management of PJI were included: 176 patients (38.7%) were infected and 279 (61.3%) were not. About 60% of the infections involved at least one of the species targeted by the assay. The sensitivity/specificity values were 72.3%/80.7% for targeted staphylococci, 75%/92.6% for S. agalactiae, and 38.5%/84.8% for P. acnes. The assay was more sensitive for infections occurring >3 months after arthroplasty and for patients with an elevated C-reactive protein (CRP) or erythrocyte sedimentation rate (ESR). However, it detected 64.3% and 58.3% of targeted staphylococcal infections associated with normal CRP and ESR values, respectively. This new multiplex immunoassay approach is a novel noninvasive tool to evaluate patients suspected of having PJIs and provides information complementary to that from inflammatory marker values.

Assessment of a Multiplex Serological Test for the Diagnosis of Prosthetic Joint Infection: a Prospective Multicentre Study

“…The sensitivity and specificity of the detection of anti-staphylococcal antibodies in our study are close to those reported by Marmor et al (72.3% and 80.7%, respectively) and lower than those reported by De Seynes et al (87.5% and 93.5%, respectively) [9,10]. The specificity is notably lower than that reported by De Seynes et al with the high number of positive serological reactions among uninfected patients (18.1%) in our study.…”

“…Patients were distributed evenly between PJI and mechanical problems. All PJIs were chronic, with the presence of a sinus tract in a third of cases and a significantly increased CRP level, which reinforces the importance of measuring CRP in the diagnosis of PJI [7][8][9][10].…”

“…Two recent studies have evaluated its contribution to the diagnosis of chronic PJIs [9,10]. The original study for the development and the evaluation of the test included 455 patients in 2 referral centres for complex osteoarticular infections, 176 PJIs and 279 uninfected patients [9].…”

“…Two recent studies have evaluated its contribution to the diagnosis of chronic PJIs [9,10]. The original study for the development and the evaluation of the test included 455 patients in 2 referral centres for complex osteoarticular infections, 176 PJIs and 279 uninfected patients [9]. The second prospective, single-centre study included 71 patients, 26 with aseptic failure and 45 with PJI in another referral centre for complex osteoarticular infections [10].…”

“…The second prospective, single-centre study included 71 patients, 26 with aseptic failure and 45 with PJI in another referral centre for complex osteoarticular infections [10]. The preliminary prospective study showed variable performance for detection, with a sensitivity and specificity of 72.3% and 80.7%, respectively for staphylococci, 75% and 92.6%, respectively, for S. agalactiae, and 38.5% and 84.8%, respectively for C. acnes [9]. The results reported in the second study for the sensitivity, specificity, positive and negative predictive values were 81.8%, 95.4%, 97.3% and 72.4%, respectively, for all patients and 87.5%, 93.5%, 94.6% and 85.3%, respectively, for staphylococcal infections [10].…”