The study is a prospective, open randomized, non-inferiority trial with two parallel groups, comparing 6 weeks versus 12 weeks of antibiotic treatment following surgery procedure (debridement and retention, 1-stage or 2 stage exchange).

The duration of the treatment antibiotic of prosthetic joint infections is only based on experts' opinion ; this one varies from 6 weeks to several months according to the customs of the influencer.

The principal aim of this study is to explore the efficacy and safety of 6 weeks versus 12 weeks antibiotic therapy duration, both associated with surgical procedure (debridement and retention of implant, one-stage or two stages exchange), in PJI treatment.

The study concerns 410 men or women of more than 18 years include in 34 centres in France.

The duration of the study is of 4 years.

Although the infection risk following primary hip and knee arthroplasties is less than 2%, prosthetic joint infections (PJI) have become more frequent due to the increasing number of patients undergoing surgery. PJI remains one of the most serious complications of prosthetic joint implantation. Each PJI episode represents substantial morbidity with a 5.3-7.2-fold higher cost than for the initial arthroplasty. The management of PJI almost always necessitates the need for surgical intervention and prolonged courses of intravenous or oral antimicrobial therapy. Despite a significant amount of basic and clinical research in this field, many questions pertaining to the definition of infection as well as diagnosis and management of these infections remain unanswered. The optimal surgical and antibiotic treatment of PJI remains unclear. The infection of implant is difficult to treat. According to the Infectious Diseases Society America recommendations of Osmon et al. in 2012, antibiotic therapy duration varies: 3 months for hip arthroplasties (6 months for total knee arthroplasties) for PJI with arthroplasty retention, and 6 weeks for two-stage exchanges.

Sometimes, excessive antibiotic treatment durations are performed because recommendations are not evidence-based sensu strictu and only based on experts' opinion. Short therapy should be benefit to ecologic and economic impact.

While the usual treatment is by the parenteral route for the first 2-4 weeks, this attitude is not evidence-based either. Up to one-third of patients with PJI may experience antibiotic-related or catheter-related problems during parenteral treatment. The one-stage and two-stage exchange of the infected implant has rather comparable success rate in the surgical management of the PJI (about 90%). Recent no-randomized studies indicate that short antibiotic therapy (6 weeks) seems to be efficient.

Our proposal study is to explore the efficacy and safety of 6 weeks in the treatment of PJI, after a surgical procedure.

The principal aim of this study is to explore the efficacy and safety of 6 weeks versus 12 weeks antibiotic therapy duration, both associated with surgical procedure (debridement and retention of implant, one-stage or two stages exchange), in PJI treatment.

Treatments antibiotics are chosen according to the germ and according to the consensual recommendations. Used antibiotics already have the AMM in this indication.

Eligible patients have at least one symptom relating to the PJI (such pain, sinus tract, inflammatory surgical scar…) and microbiological documentation of PJI.The primary endpoint is the occurrence of clinical cure at month 24. Secondary endpoints are (i) adequate clinical and biological response at months 6 and 12, (ii) incidence of adverse events in the two groups, (iii) oral and/or intravenous antibiotic therapy efficacy.

Randomization is stratified by the anatomical location of the infected joint (hip or knee) and surgical procedure (debridement and retention, 1-stage or 2 stage exchange).

The follow-up consists on periodic clinical examinations, anamnesis and occasionally laboratory or radiological controls.

Inclusion Criteria:

• Men or women over 18, presenting at least one of the following clinical signs: pain, fever, dent or flow of scar
• Bacterial infection on osteo-articular prosthesis (hip or knee)
• Documented basterial infection
• Surgical care optimized with prothesis change or articular washing
• Infection requiring an antibiotic treatment by parenteral way Intravenous injection or per bone
• Treatment effective antibiotic begun in the most previously 21 days
• Signed consent

Exclusion Criteria:

• Patient with no proof of osteo-articular infection
• No surgical care
• Patient having more than one prothesis change because of sepsis
• Infection due to mycobactery, fungic infection or brucellienne infection
• Patient with life expectancy supposed lower than 2 years
• Guardianship patient
• Patient included in another study