The management of Joint Infections (JIs) requires a medical-surgical team which includes rheumatologists, infectious disease specialists, orthopedists as well as microbiologists.

This collaboration makes it possible to optimize patient care both from a functional point of view and from an infection point of view.

The diagnosis of these infections is based on clinic, imaging and bacteriology (microbiological samples).

Treatment is based on surgical treatment and appropriate antibiotic therapy, which will be charged only after the sample has been taken. For non-complex native septic arthritis, identification of the bacteria is expected before starting antibiotic therapy adapted to the germ.

For patients requiring surgical treatment: patients with complex native joint infections (comorbidities, allergies, etc.) or joint infections on hardware, the identification of the incriminated germ(s) cannot be expected. Probabilistic antibiotic therapy is then started while awaiting the microbiological results. This antibiotic therapy generally lasts 48 hours, while the results of the antibiogram are known. If for (JIs) on native joints recent recommendations were published by the French society of rheumatology in 2020, it is not the same for (JIs) on hardware. As a result, no updated recommendations indicate the probabilistic antibiotic therapy to prescribe immediately post-operatively. Currently, the old recommendations are no longer applied regarding positive grams. Daptomycin is preferred over vancomycin. On the other hand, for gram-negative germs, broad-spectrum beta-lactams are always prescribed. The investigators are questioning the benefit of these prescriptions given the epidemiological context of joint infections and bacterial resistance phenomena.

The investigators hypothesize that osteoarticular infections are mainly due to gram-positive bacteria.

By affirming this hypothesis, investigators could avoid prescriptions for broad-spectrum beta-lactams. This would lead to a reduction in patient exposure to broad-spectrum antibiotics (fewer adverse effects, ecological and economic gain).

Inclusion Criteria:

• Adult patient (≥18 years old)
• Hospitalized in the ortho-traumatology department
• Operated between January 1, 2021 and December 31, 2021
• Probabilistic antibiotic therapy with cefepime or piperacillin-tazobactam combined with an anti-gram + antibiotic
• Operated for: Change or washing of THA, TKA, PTE, ablation or resumption of osteosynthesis lower limbs and upper limbs and native osteitis
• Absence of written opposition in the medical file of the subject (and/or their legal representative if applicable) to the reuse of their data for scientific research purposes. It is the responsibility of the investigator to verify the absence of opposition in the patient's medical file.

Exclusion Criteria:

• Patient having expressed his opposition to the reuse of his data for scientific research purposes.
• Minor patient
• Patient operated for: Amputation, plantar perforating pain, soft tissue infection
• Patient for whom there is documentation leading to targeted probabilistic antibiotic therapy with beta lactams.

Adult patient (≥18 years old) operated between January 1, 2021 and December 31, 2021