Evaluation of the Use of Preventive Antibiotic Therapy in Patients Undergoing One-Step Prosthetic Revision Surgery with Low Preoperative Infectious Risk.

The prosthetic knee infection (PKI) rate in most centers ranges from 0.5 to 2% for knee replacements, depending on risk factors. Current PKI definitions may miss the identification of both early and late complications. There is no consensus on preventive or early empiric antibiotic therapy (EEAT) in the one-step exchange strategy for low-risk patients pending microbiology results. The aim of the study was to evaluate the potential role of EEAT in patients with comorbidities in preventing PKI and to evaluate differences in septic failure at 3, 6 and 9 months after prosthetic revision between patients undergoing EEAT and patients not undergoing EEAT. All adult patients undergoing one-step knee revision surgery at IRCCS Sacro-Cuore Don Calabria Negrar, from January 2018 to February 2021, were retrospectively included in a cohort observational study. Patients on antibiotic therapy before surgery or with preoperative ascertained PKI were excluded. Demographic characteristics, Charlson score, comorbidities, inflammatory markers, microbiological tests, imaging, infectious disease risk score and EEAT data were collected. Any postoperative complication or modification of antibiotic therapy at 14, 30, 90, 180 and 270 days after surgery was collected. A total of 227 patients were included: 114 comorbid low-risk patients received EEAT after surgery, pending microbiological results; while 113 non-comorbid low-risk patients did not receive any antibiotic therapy in the postoperative period. Among the EEAT group, 16 were diagnosed with PKI, compared with 10 in the untreated group. Regarding septic failure during the 9-month follow-up after revision surgery, we registered nine cases in the EEAT arm and four in the untreated arm. In three out of nine cases treated with EEAT who had a post-revision septic failure, the causative microorganism was not successfully empirically targeted by EEAT; in the untreated group, two out of four cases had a post-revision septic failure, despite the targeted treatment of intraoperatively identified causative microorganisms. According to our results, EEAT after revision surgery in patients with comorbidities, who are at higher risk of infection, did not prevent prosthetic knee infections. There was also no evidence of a reduction in subsequent septic failure within nine months of revision surgery between groups. More accurate risk-defining scores are needed to identify patients at risk of PKI complications.