Key points are not available for this paper at this time.
Abstract Background Early detection and prompt treatment are crucial for managing septic arthritis (SA). However, diagnosing SA at an early stage can be challenging since the symptoms are non-specific, and there is no definitive diagnostic test available. This study aimed to investigate the potential of biomarkers as a reliable diagnostic tool for SA. Methods A prospective study was conducted to collect residual blood and synovial fluid samples from patients who underwent blood sampling and arthrocentesis for differential diagnosis due to suspected SA. The collected samples were analyzed for C-reactive protein (CRP), procalcitonin, and presepsin. The data obtained from the analysis were subjected to statistical analysis. Results From January to November 2023, 58 patients were enrolled. They comprised 24 men and 34 women with a median age of 71.1 (interquartile range, IQR 64.0-79.3) years. Twenty-one patients were diagnosed with SA and 37 with non-SA, including osteoarthritis (n=2), osteonecrosis (n=1), meniscus tear (n=1), abscess (n=1), contusion (n=2), and mechanical complications of implants (n=30). There were no significant differences in age or sex distribution between the two groups. The median blood and synovial fluid CRP levels were 4.24 and 1.85 mg/dL, respectively, in the SA group and 0.24 and 0.11 in the non-SA group. The median blood and synovial fluid procalcitonin levels were 0.0383 and 0.0403 ng/mL in the SA group and 0.0358 and 0.0488 in the non-SA group. The median blood and synovial fluid presepsin levels were 307 and 1389 pg/mL in the SA group and 192 and 959 in the non-SA group. Except for blood procalcitonin, all showed statistically significant differences between the SA and non-SA groups. In the ROC curve analysis, blood CRP displayed the highest area under the curve (AUC) of 0.795 with a 0.47 mg/dL cut-off value, 90.5% sensitivity, and 64.9% specificity. Synovial fluid CRP displayed similar results of 0.786 AUC, 0.30 mg/dL cut-off value, 85.0% sensitivity, and 71.4% specificity. Blood and synovial fluid procalcitonin had low AUCs of 0.532 and 0.659, respectively, and were insufficient to differentiate between SA and NS groups. The AUC (0.696 vs. 0.704), sensitivity (76.2% vs. 76.2%), and specificity (67.6% vs. 62.2%) of blood and synovial fluid presepsin were similar; however, synovial fluid showed a significantly higher cut-off value, 4.5 times that of blood (237 pg/mL vs. 1087 pg/mL). Conclusions Procalcitonin was not a reliable marker for distinguishing between SA and non-SA, regardless of whether it was measured in blood or synovial fluids. However, CRP and presepsin proved useful markers when measured in either blood or synovial fluid. In particular, when measuring presepsin in synovial fluid, an optimal cut-off value is needed.
Choi et al. (Tue,) studied this question.