Prognosis of unexpected positive intraoperative cultures in arthroplasty revision: A large multicenter cohort.

<bold>Background: </bold>The positive-intraoperative-cultures-type prosthetic joint infection (PIOC-PJI) is considered when surgical cultures yield microorganisms in presumed aseptic arthroplasty revisions. Herein we assess the risk factors for failure in the largest cohort of PIOC-PJI patients reported to date.<bold>Methods: </bold>A retrospective, observational, multicenter study was performed during 2007-2017. Surgeries leading to diagnose PIOC-PJI included only one-stage procedures with either complete or partial prosthesis revision. Failure was defined as recurrence caused by the same microorganism.<bold>Results: </bold>203 cases were included (age 72 years, 52% females). Coagulase-negative staphylococci (n = 125, 62%) was the main etiology, but some episodes were caused by virulent bacteria (n = 51, 25%). Prosthesis complete and partial revision was performed in 93 (46%) and 110 (54%) cases, respectively. After a median of 3.4 years, failure occurred in 17 episodes (8.4%, 95%CI 5.3-13.1). Partial revision was an independent predictor of failure (HR 3.63; 95%CI 1.03-12.8), adjusted for gram-negative bacilli (GNB) infection (HR 2.68; 95%CI 0.91-7.89) and chronic renal impairment (HR 2.40; 95%CI 0.90-6.44). Treatment with biofilm-active antibiotics (rifampin/fluoroquinolones) had a favorable impact on infections caused by staphylococci and GNB.<bold>Conclusion: </bold>Overall prognosis of PIOC-PJI is good, but close follow-up is required in cases of partial revision and in infections caused by GNB. [ABSTRACT FROM AUTHOR]