Empiric and targeted antibiotic therapy for bloodstream infections in internal medicine patients in Poland: a three-year analysis in a single centre using the AWaRe classification.

Background: Bloodstream infection (BSI), defined as bacteraemia accompanied by sepsis or septic shock, is a frequent cause of hospitalization in departments of internal medicine (DIM). This study aimed to characterize the epidemiology, microbiological profile, and antibiotic treatment patterns of BSIs in a DIM setting in 2021-2023 in Polish hospital.

Methods: A three-year, single-centre retrospective analysis was conducted at a DIM in Southern Poland. Medical records from 2021 to 2023 were reviewed for patients diagnosed with sepsis (ICD-10 codes A40-A41) with microbiological confirmation. Antibiotic use was assessed using Days of Therapy (DOT) and Length of Therapy (LOT). Empiric and targeted therapies were classified according to the WHO AWaRe (Access, Watch, Reserve) framework.

Results: A total of 124 BSI cases were identified, with an incidence rate of 5.8%. Community-acquired BSIs accounted for 84.3% of cases, while 15.7% were hospital-acquired. The predominant pathogens were Escherichia coli and Staphylococcus aureus. The median LOT for empiric therapy was 3 days (IQR 2-4), with third-generation cephalosporins being the most frequently used agents (78 patients; 43.3%, 228 DOTs; 43.9%). Targeted therapy had a median LOT of 8 days (IQR 5-10), most commonly involving penicillins (33 patients; 28.4%, 291 DOTs; 34.5%). Access group antibiotics were significantly more prevalent in targeted therapy compared to empiric therapy (p < 0.001).

Conclusions: Microbiological confirmation of BSI facilitates a safe de-escalation from broad-spectrum empiric antibiotics to narrow-spectrum targeted therapy, supporting antimicrobial stewardship in internal medicine settings.