Reinfection rates after debridement, antibiotics, and implant retention, one-stage and two-stage revision for fungal periprosthetic joint infection following total hip and knee arthroplasty: a meta-analysis and systematic review

This systematic review and meta-analysis aimed to compare reinfection rates after debridement, antibiotics, and implant retention (DAIR), one-stage revision, and two-stage revision for fungal periprosthetic joint infection (PJI) following total hip and knee arthroplasty. A systematic search of PubMed, Embase, the Cochrane Library, and Web of Science was conducted from database inception to November 1, 2025. Studies reporting reinfection outcomes after DAIR, one-stage revision, or two-stage revision for fungal PJI following total hip or knee arthroplasty were included. Two reviewers independently screened the studies and extracted data, with disagreements resolved by a third reviewer. Pooled reinfection rates with 95% confidence intervals (CIs) were calculated using a random-effects model after Freeman–Tukey double arcsine transformation. Twenty-one studies involving 324 unique patients and 332 procedure-specific treatment events were included. The pooled reinfection rates after DAIR, one-stage revision, and two-stage revision were 93% (95% CI: 80%–100%), 28% (95% CI: 12%–48%), and 25% (95% CI: 13%–39%), respectively. In subgroup analyses based on extractable joint-specific data, reinfection rates did not differ significantly between hip and knee fungal PJI. However, the evidence for one-stage revision was limited by the small number of included patients, wide confidence intervals, and potential selection bias. Given the rarity of fungal PJI and the predominance of small observational studies, the pooled estimates should be interpreted with caution. DAIR was associated with a high reinfection rate when used as the definitive implant-retention strategy for fungal PJI, whereas revision surgery showed lower pooled reinfection rates. However, evidence for one-stage revision remains limited, and current data are insufficient to establish its equivalence or superiority to two-stage revision.