EVALUATING THE ROLE OF NEXT-GENERATION SEQUENCING IN DIAGNOSING PERIPROSTHETIC JOINT INFECTIONS: INSIGHTS FROM A PROSPECTIVE STUDY

Aim To evaluate the diagnostic value of next-generation sequencing (NGS) in patients with suspected periprosthetic joint infection (PJI), using European Bone and Joint Infection Society (EBJIS) criteria for case definition. Method We are prospectively including patients undergoing revision surgery for suspected PJI. Diagnosis was established according to the EBJIS criteria. In addition to routine microbiological diagnostics (synovial fluid, periprosthetic tissue, and sonication fluid cultures), all available samples from each patient will be also processed for NGS. DNA extraction and human DNA depletion were performed using the Molzym Ultra-Deep Microbiome protocol, followed by library preparation, metagenomic sequencing and taxonomic profiling. Clinical, demographic, and microbiological data were collected and analysed, focusing on concordance between NGS findings and culture results and their relevance to the clinical diagnosis. Results A total of 75 patients are currently included in the cohort. Based on EBJIS criteria, 33 were classified and treated as PJI. NGS was performed on joint and sonication fluid from four random patients. The proportion of human DNA in all analysed samples was acceptably low, indicating effective depletion of host DNA during the extraction process. In two patients with confirmed Staphylococcus epidermidis PJI, both culture and NGS identified the same pathogen. The proportion of bacterial DNA was higher in the sonication fluid compared to joint fluid, and coverage depth for S. epidermidis was also greater in sonication-derived samples. In one patient not treated as infection, NGS detected Cutibacterium acnes DNA in both joint and sonication fluid. However, the patient had no clinical signs of infection and has remained asymptomatic during follow-up, indicating likely contamination. In another patient treated as PJI based on clinical and laboratory criteria, but with negative cultures, NGS revealed only a very low bacterial DNA signal in both sample types, with no identifiable pathogen. Conclusions NGS shows promise as an adjunct tool in the diagnosis of PJI, particularly in cases with inconclusive culture results. As further samples are processed, we look forward to expanding our dataset and identifying meaningful trends. Defining a diagnostic threshold for bacterial DNA percentage will be essential to improve interpretability. Given the complexity of NGS results, additional analyses such as metagenomic assembly and resistance gene identification are planned to complement taxonomic profiling and enhance pathogen characterization. Close collaboration with microbiologists or sequencing specialists will likely remain crucial for clinical decision-making.