IntroductionOsteoarticular infections (OAIs) of native joints lead to cartilage damage which may require subsequent arthroplasty. There is no consensus on systematic intraoperative microbiological sampling when performing an arthroplasty on a native joint with a history of OAI. We carried out a retrospective study to: (1) identify the frequency of the persistence of the microorganism(s) involved during the initial, presumed cured OAI, when performing an arthroplasty for sequelae of osteoarthritis, (2) to find an association between the length of time between the OAI and arthroplasty, and the recurrence of bacterial infection, (3) to assess the influence of the presence of hardware on the risk of infectious recurrence.HypothesisSystematic sampling is justified during a subsequent arthroplasty after an OAI, even after a prolonged period.Material and methodThis single-center, retrospective descriptive study included all patients whose indication for arthroplasty resulted from osteoarthritis, osteitis or bacterial osteomyelitis of a native joint, or in the aftermath of an infection post osteosynthesis. All patients were considered to have recovered from the initial infection at the time of the arthroplasty. Between 2008 and 2019, 92 patients were included in the study, with an average age of 56.5 years (range: 21–97 years). OAI occurred at a mean age of 35 years (range: 1–84 years). The average time from OAI to implantation was 15 years (range: 1–65 years). The bacteria most frequently found in the initial OAI was Staphylococcus aureus, involved in 35.8% of cases (n = 33/92).ResultsThe intraoperative samples came back positive in 17% of cases (n = 16/92), including 9 positive for the same bacteria as the OAI (56%, n = 9/16). For these 16 cases, the time between the OAI and the arthroplasty was 1 year for 5 patients, between 1 and 15 years for 5 patients and greater than 15 years for 6 patients. For 3 positive patients, the information on the initial microorganism was not known and 4 patients were positive for a bacterium different from the initial one. The time from the initial OAI to the arthroplasty was not associated with positive results (p = 0.38). There was no significant difference between a positive culture at the time of arthroplasty and the initial type of OAI [native joint versus presence of hardware and/or open fracture (p = 0.41)].ConclusionThe results of this work suggest there is value in microbiological sampling when performing an arthroplasty on a previously infected joint, regardless of the duration of the infection.