RISK FACTORS FOR HIP AND KNEE PROSTHETIC JOINT INFECTIONS CAUSED BY GRAM-NEGATIVE BACTERIA

Aim To investigate whether it is possible to identify a subgroup of patients with prosthetic joint infections (PJIs) at increased risk of having an infection caused by Gram negative bacteria (GNB). Methods Observational retrospective multicenter study on patients with hip and knee PJIs undergoing infection surgery between 2015 and 2019 at the participating centers. PJI was defined according to 2018 ICM criteria and was considered caused by GNB in presence of culture of at least one intraoperative peri-prosthetic tissue sample and/or sonication fluid yielding GNB. To assess risk factors for GN-PJI univariate and multivariate logistic regression analysis was carried out. Statistical significance was considered for p values < 0.05. Results Two-hundred and eighty-one patients were included, 137 (49%) were males, median age was 68 years (IQR 55-75), median Charlson comorbidity index was 3 (IQR 2-5); the hip was involved in 147/281 (52%) cases, osteoarthritis was the most frequent indication for prosthesis placement (64%) and 93/281 (33%) had already undergone a revision surgery. The most frequent surgical approach was a two-stage revision (61%), and the overall treatment failure rate was 19%. A Gram-negative etiology was found in 51 cases (18%); among these patients the most frequent etiological agents were Enterobacterales (n=34). A pre-operative synovial fluid culture was obtained in 93 patients (34%), and it was positive in 79; after surgery, 185/281 patients (66%) received a broad-spectrum empiric antibiotic therapy with GNB-coverage. Univariate and multivariate logistic regression analysis selected hip PJI (OR 2.88, 95% CI 1.44-5.77, p=0.003) and delayed infection (OR 0.45, 95% CI 0.21-0.94, p=0.03) as the factors independently associated to a GN-PJI Conclusion Currently, in the absence of a pre-operative etiological diagnosis, a broad-spectrum antimicrobial therapy is used for patients undergoing surgical treatment of PJIs, pending the results of intra-operative cultures. However, a widespread use of GNB-coverage may lead to more adverse reactions, resistance selections and costs. To identify risk factors for GNB-PJI may reduce the use of empirical GNB-coverage when treating PJIs without a pre-operative etiological diagnosis. According to our findings, an empirical Gram-negative coverage is justified in the case of hip PJIs and in early and acute late infections.