Outcomes of the Subsequent Periprosthetic Joint Infection Revisions after a Failed Debridement, Antibiotics and Implant Retention: A Multicentric Study of 197 Patients.

Background: The impact of prior unsuccessful debridement, antibiotics, and implant retention (DAIR) procedures on subsequent revisions is uncertain, with conflicting evidence. Despite 85% consensus against the second DAIR procedure following the 2018 International Consensus Meeting, a 2020 study reported high success rates for the aforementioned second DAIR procedure. Methods: We conducted a multicenter observational study reviewing data from patients with failed DAIR procedures between 2005 and 2021. Patients diagnosed with acute periprosthetic joint infection of the hip or knee were included, following ICM criteria. Failure was defined as uncontrolled infection leading to additional surgeries, prosthesis removal, infection-related mortality, or suppressive antibiotic therapy. Demographic, surgical, and microbiological variables were recorded. Results: Among 197 patients from 10 institutions with failed DAIR procedures were included: 88 (44.7%) received a second DAIR, 21 (10.7%) underwent one-stage revision, and 77 (39.1%) underwent two-stage revision. One-stage revision success rate was 76.2%, with no identified predictors of failure. Two-stage revision success rate was 79.3%; factors associated with failure included polymicrobial infections (p = 0.025) and revision procedures (p = 0.049). Second DAIR success rate was 54.5%; factors associated with failure included non-specialized surgical teams in the first DAIR (p = 0.034), non-exchange of mobile components (p = 0.0038), polymicrobial infections (p = 0.043), and antibiotic resistance (p = 0.035). Excluding patients with these risk factors increased the success rate to 83.3%. Conclusions: Second DAIR's overall success rate was 54.5%, significantly increasing to 83.3% when excluding patients with identified risk factors. These findings suggest considering second DAIR in carefully selected patients without these risk factors. Our study found success rates of 76.2% and 79.3% for one- and two-stage revisions, respectively, aligning closely with published data.