Diagnostic parameters in native joint septic arthritis and development of a new evaluation score

Introduction: Septic arthritis (SA) is an orthopaedic emergency; delayed treatment endangers joint function and survival. Unlike periprosthetic infections, diagnostic criteria for native joints are poorly standardized. This study aimed to (1) evaluate diagnostic parameters including synovial white blood cell (sWBC) count, neutrophil percentage, serum C-reactive protein (CRP), and leukocyte count and (2) assess a new evaluation score combining these parameters. Methods: In a retrospective cohort study, cases of knee and shoulder SA treated at a German university hospital (2013-2022) were analysed. Parameters included synovial fluid analysis (sWBC, neutrophils), blood samples (CRP, leukocytes), and intraoperative cultures. Cut-offs for sWBC and neutrophils were determined using receiver-operating characteristic (ROC) analysis and the Youden index, comparing SA patients with non-infected controls. A new evaluation score for SA (Septic Arthritis Evaluation Score, SAES) was created: 2 points each for sWBC and neutrophils and 1 point each for CRP and leukocytes. Results: Complete data were available for 45 patients (71.4 % male, mean age 64.3 years). Knees were affected in 73.7 %, and shoulders were affected in 26.3 %. Median values were as follows: leukocytes - 11/nl, CRP - 158 mg L^-1, sWBC - 42 910/ $\mu$ L, and neutrophils - 93.6 %. ROC analysis identified an optimal sWBC cut-off of 35 650/ $\mu$ L (sensitivity of 64.4 %, specificity of 87.8 %). The SAES showed higher discriminatory performance; with a threshold $\ge$ 3 points, sensitivity was 95.6 %, and specificity was 70.7 %. Conclusions: In this retrospective cohort, commonly used laboratory parameters for native joint SA showed limited discriminatory ability when applied individually. A newly developed composite score combining synovial and serum markers demonstrated higher sensitivity within this dataset. Prospective validation in larger cohorts is required before clinical application.