Diagnostic performance of perioperative inflammatory and coagulation biomarkers for periprosthetic joint infection

Distinguishing periprosthetic joint infection (PJI) from aseptic failure remains challenging, as both conditions may present with systemic inflammatory changes. Although C-reactive protein (CRP) and other conventional inflammatory markers are widely used, their diagnostic performance is not always sufficient. This study aimed to evaluate whether available preoperative biomarkers reflecting systemic inflammation and coagulation could improve the identification of PJI in patients undergoing revision total hip arthroplasty (THA) or total knee arthroplasty (TKA). We retrospectively reviewed 379 patients who underwent revision total hip or knee arthroplasty between December 2019 and December 2024, including 172 patients with PJI and 207 with aseptic revision group. The last available preoperative laboratory results before revision surgery were analyzed, including hematological inflammatory indices, including neutrophil–lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR), monocyte-lymphocyte ratio (MLR), systemic immune-inflammation index (SII), albumin-globulin ratio, interleukin-6 (IL-6), C-reactive protein (CRP), D-dimer, and fibrinogen (Fib). Receiver operating characteristic curve analysis was used to assess diagnostic performance. Univariate logistic regression analyses were used to assess the associations between individual biomarkers and PJI, followed by an exploratory multivariable logistic regression model adjusted for gender. A nomogram was developed from this exploratory model and evaluated using calibration analysis and decision curve analysis (DCA). Among the evaluated biomarkers, SII showed good overall discrimination for distinguishing PJI from aseptic revision, with an area under the curve of 0.830. At the Youden-derived cutoff of 768.125, SII had a sensitivity of 56.6% and a specificity of 92.8%. When dichotomized at the Youden-derived cutoffs, elevated SII (> 768.125; OR 12.526; 95% CI 6.599–25.962) and CRP (> 6.840 mg/L; OR 7.942; 95% CI 4.664–14.075) were both associated with PJI (both P < 0.001). In an exploratory model adjusted for gender, SII (aOR, 9.751; 95% CI 4.905–20.988; p < 0.001) and CRP (aOR, 6.205; 95% CI 3.449–11.556; p < 0.001) remained associated with PJI. The nomogram incorporating SII, CRP, and gender achieved an apparent AUC/C-index of 0.867 (95% CI 0.818–0.917) and an optimism-corrected C-index of 0.865 after 1000 bootstrap resamples. The optimism-corrected calibration slope was 0.954, with a Brier score of 0.151. SII showed good discriminative ability for distinguishing PJI revision from aseptic revision. However, its limited sensitivity at the selected cutoff precludes its use as a standalone screening or rule-out test. SII should therefore be interpreted as an adjunctive marker together with established diagnostic findings. The nomogram incorporating SII, CRP, and gender demonstrated promising apparent performance after internal validation. The underlying revision-arthroplasty biomarker cohort was retrospectively registered in the Chinese Clinical Trial Registry on April 26, 2022 (ChiCTR2200059185), after data collection had begun. The present analysis was not prospectively registered.