Abstract Background Diabetic foot ulcers (DFUs) are common in individuals with diabetes mellitus (DM), frequently leading to infection, hospitalization, and amputation. Broad-spectrum empiric antibiotics targeting methicillin-resistant Staphylococcus aureus (MRSA) and Pseudomonas aeruginosa (PSA) are often overprescribed. In 2021, Ascension Seton implemented a local diabetic foot infection (DFI) guideline using a risk-stratified approach to optimize empiric antibiotic prescribing. This study evaluates its impact on antimicrobial stewardship and clinical outcomes. Methods This multi-site, retrospective cohort study included adults hospitalized with DFI or diabetes-related foot osteomyelitis (DFO) between January 2017 and September 2024. Patients were excluded if treated with antibiotics for 36 hours, had necrotizing fasciitis, or surgical site infections. The primary outcome was rate of empiric anti-MRSA and/or anti-PSA antibiotic use pre- vs. post-guideline implementation. Secondary outcomes included guideline concordance, escalation/de-escalation, treatment failure, 90-day readmission, and safety. Statistical comparisons used Chi-square and nonparametric tests. Results Of 170 patients (pre-guideline: n=102; post-guideline: n=68), empiric anti-MRSA use decreased from 66.7% to 54.4% (p=0.113) and anti-PSA use from 69.6% to 58.8% (p=0.156). Guideline concordance improved (46.1% vs. 55.9%, p=0.084). There were no significant differences in treatment failure (37.3% vs. 35.2%, p=0.813), 90-day readmission (28.4% vs. 29.4%, p=0.583), or adverse drug events (6.9% vs. 2.9%, p=0.288). No Clostridioides difficile infections or mortality occurred in either group. Conclusion A risk-stratified DFI guideline reduced empiric broad-spectrum antibiotic use and improved guideline adherence without compromising clinical outcomes. These findings support risk-based empiric therapy as a safe and effective antimicrobial stewardship strategy in DFI management. Disclosures All Authors: No reported disclosures
Doan et al. (2026) studied this question.