Diagnostic Accuracy of Synovial Calprotectin in Megaprosthetic Reconstructions: A Prospective Cohort Study from a Tertiary Sarcoma Center

Background/Objectives: Diagnosing periprosthetic joint infection (PJI) after megaprosthetic reconstruction may be difficult due to altered inflammatory responses, extensive prior surgery, and the limited performance of conventional criteria such as the 2018 ICM score. Synovial calprotectin is a rapid neutrophil-derived biomarker that may improve diagnostic accuracy in this challenging setting. The primary aim of this study was to evaluate the diagnostic performance of synovial calprotectin in detecting periprosthetic infection in patients treated with tumor megaprostheses; secondary aims included comparison with ICM classification, assessment in infection classification-inconclusive cases, and exploratory performance in patients with low CRP. Methods: This prospective study included 20 consecutive megaprosthesis patients evaluated for suspected PJI at ATTIKON University Hospital, Athens, with a minimum follow-up of 1 year after biomarker testing. Synovial calprotectin was measured using a lateral-flow assay (positive ≥ 50 mg/L) and compared with a predefined infection reference standard. ICM final status (0 = aseptic, 1 = inconclusive, 2 = infected) was recorded for all cases. Other synovial biomarkers (α-defensin, leukocyte esterase, synovial D-dimer) were not routinely available. The cohort had a mean age of 52.9 ± 22.5 years, 70% were male, and reconstructions involved the knee (80%), hip (15%), and humerus (5%). Preoperative cultures were positive in 40%, the median systemic WBC was 7100/μL, and the median time from last surgery to testing was 1.0 years (IQR 0.46–2.0). Among infected cases, the most common microorganisms were coagulase-negative staphylococci (61.5%) and Staphylococcus aureus (23.1%), with 30.8% demonstrating polymicrobial infection. Results: Thirteen of 20 patients (65%) were classified as infected. Using the ≥50 mg/L threshold, synovial calprotectin demonstrated high apparent diagnostic accuracy in this exploratory cohort, and no false positives, yielding a sensitivity of 92.3%, specificity of 100%, PPV of 100%, NPV of 87.5%, LR+ = ∞, and LR− = 0.08. The AUC for continuous values was 1.00. Agreement with the ICM final classification was substantial (κ = 0.76), with no directional discordance (McNemar p = 1.00). Among the three ICM-inconclusive cases, calprotectin correctly reclassified two (66.7%). In patients with low CRP (<10 mg/L), a clinically difficult subgroup, calprotectin maintained strong performance (sensitivity 75%, specificity 100%, NPV 85.7%). Conclusion: Synovial calprotectin demonstrated promising diagnostic performance for PJI in megaprosthesis patients, with high sensitivity and specificity, and substantial agreement with the 2018 ICM criteria. It successfully clarified most ICM-inconclusive cases and remained reliable even in patients with low CRP. These findings support calprotectin as a valuable adjunctive biomarker in the complex diagnostic environment of megaprosthetic reconstruction and justify further validation in larger cohorts.