FUNGAL PERIPROSTHETIC JOINT INFECTION AS A MARKER OF PERSISTENT POLYMICROBIAL BIOFILM: A MATCHED COHORT OF SURGICAL BURDEN AND CHRONIC OUTCOMES

Aim Fungal periprosthetic joint infection (PJI) is historically cited at fewer than 1% of all PJI cases, largely derived from centers without systematic fungal culture protocols. We hypothesized that fungal PJI represents a phenotype of persistent polymicrobial biofilm infection associated with disproportionate surgical burden rather than excess mortality, and that institutional prevalence is underestimated at centers without dedicated fungal surveillance. Methods A retrospective matched cohort study was conducted at a large academic referral center between 2015 and 2025. All intraoperative PJI specimens were processed with calcofluor white fluorescent smear and extended fungal culture incubation. Adults with confirmed fungal PJI of the hip or knee (n=64) were matched 1:3 to non-fungal PJI controls (n=181) on age (±8 years), sex, race, year of surgery, joint, and Elixhauser Comorbidity Index score. Binary outcomes were analyzed by logistic regression (adjusted odds ratio [aOR], 95% confidence interval [CI]; number needed to harm [NNH]); count outcomes by negative binomial regression (adjusted incidence rate ratio [IRR]). All models were adjusted for American Society of Anesthesiologists (ASA) physical status. Secondary analyses examined polymicrobial versus monomicrobial fungal PJI, and Candida albicans versus non-C. albicans species. Results Among 1,008 adults with PJI, 64 (6%) had confirmed fungal PJI. Candida species accounted for 94% of cases; 72% were polymicrobial. Baseline characteristics were well-matched (standardized mean difference <0.10 for all variables except ASA). Compared to matched non-fungal controls, fungal PJI was independently associated with chronic antibiotic suppression (93.8% vs. 46.4%; aOR 17.67, 95% CI 6.10–51.27; NNH 2), knee amputation (20.3% vs. 3.9%; aOR 6.08, 95% CI 2.25–16.42; NNH 6), flap use (aOR 5.32, 95% CI 1.92–14.73; NNH 8), and Girdlestone resection arthroplasty (aOR 4.13, 95% CI 1.26–13.58; NNH 10). Fungal PJI patients required more total surgeries (mean 5.30 vs. 2.35; aIRR 1.88, CI 1.57–2.25). Mortality was similar (17.2% vs. 15.1%; P=0.795). Polymicrobial fungal PJI showed consistently larger effect estimates than monomicrobial disease across all outcomes. Conclusions Fungal PJI prevalence substantially exceeds historical benchmarks under systematic fungal surveillance (6% vs. 1%). Rather than excess mortality, fungal PJI is characterized by a protracted cycle of surgical intervention and indefinite antimicrobial suppression. The 72% polymicrobial rate and Candida-dominant profile implicate mature polymicrobial biofilm biology as a likely mechanism of treatment refractoriness and suggest that current management strategies may be fundamentally insufficient.