Assessing Empiric Antimicrobial Therapy With the Modified Dundee Classification for Nonpurulent Skin and Soft Tissue Infections in a Community Hospital System

Background: The modified Dundee classification has recently been validated in various studies for nonpurulent skin and soft tissue infections. This has yet to be applied in the United States and within community hospital settings to optimize antimicrobial stewardship and ultimately patient care. Methods: A retrospective, descriptive analysis was performed of 120 adult patients admitted to St. Joseph's/Candler Health System for nonpurulent skin and soft tissue infections between January 2020 and September 2021. Patients were classified into their modified Dundee classes, and frequencies of concordance of their empiric antimicrobial regimens with this classification scheme in the emergency department and inpatient settings were compared, along with possible effect modifiers and possible exploratory measures associated with concordance. Results: Concordance with the modified Dundee classification for the emergency department and inpatient regimens was 10% and 15%, respectively, with broad-spectrum antibiotic use and concordance positively associated with illness severity. Due to substantial broad-spectrum antibiotic use, possible effect modifiers associated with concordance were unable to be validated, and overall no statistically significant differences among exploratory analyses across classification status were observed. Conclusions: The modified Dundee classification can help identify gaps in antimicrobial stewardship and excessive broad-spectrum antimicrobial usage toward optimizing patient care.