Neutrophil Extracellular Trap-related Biomarkers Are Increased in the Synovial Fluid of Patients With Periprosthetic Joint Infections

Abstract

Abstract Background The diagnosis of periprosthetic joint infection (PJI) is a major challenge in clinical practice. The role of neutrophils in fighting infection has been increasingly understood, and one mechanism of action of these cells is neutrophil extracellular traps. However, little is known about this process in PJI. Questions/purposes (1) Are the biomarkers of neutrophil extracellular trap formation (citrullinated histone H3 [H3Cit], cell‐free DNA [cf-DNA], and myeloperoxidase [MPO]) increased in the synovial fluid of patients with PJI? (2) What is the diagnostic accuracy of biomarkers of neutrophil extracellular trap formation for PJI? Methods Between May 2020 and March 2021, 43 patients who underwent revision THA or TKA were enrolled in this study. Eleven patients were excluded and 32 patients were categorized into the PJI group (n = 16) or non-PJI group (n = 16) according to the 2018 Second International Consensus Meeting on Musculoskeletal Infection criteria. There were 15 men and 17 women in this study, with a median (range) age of 70 years (60 to 80 years). Twenty-seven patients had TKA and five had THA. We measured cf-DNA, MPO, and H3Cit in synovial fluid. The sensitivity, specificity, and receiver operating characteristic curve were calculated for each biomarker using the Musculoskeletal Infection Society criteria as the gold standard for diagnosis and considering a clinical surveillance of 2 years for patients in the non-PJI group. Results Patients with PJI had higher levels of synovial fluid cf-DNA (median [range] 130 ng/µL [18 to 179] versus 2 ng/µL [0 to 6]; p < 0.001), MPO (1436 ng/µL [55 to 3996] versus 0 ng/µL [0 to 393]; p < 0.001), and H3Cit (2115 ng/µL [5 to 2885] versus 3 ng/µL [0 to 87]; p < 0.001) than those in the non-PJI group. In receiver operating characteristic curve analyses, we observed near-perfect performance for all biomarkers evaluated, with an area under the curve of 1 (95% CI 0.9 to 1), 0.98 (95% CI 0.9 to 1), and 0.94 (95% CI 0.8 to 0.99) for cf-DNA, MPO, and H3Cit, respectively. The sensitivity for detecting PJI using synovial fluid was 100% for cf-DNA, 94% for MPO, and 88% for H3Cit. The specificity was 100% for cf-DNA and MPO, and 88% for H3Cit. Conclusion Our results show that neutrophils in the periprosthetic microenvironment release neutrophil extracellular traps as part of the bactericidal arsenal to fight infection. These results allow a better understanding of the cellular and molecular processes that occur in this microenvironment, enabling the design of more assertive strategies for identifying new biomarkers and improving the available ones. Novel studies are needed to define whether and how neutrophil extracellular trap-related biomarkers can be useful for diagnosing PJI. Level of Evidence Level II, diagnostic study.