Beyond the Cumulative Antibiogram: A Weighted-Incidence Syndromic Combination Antibiogram to Guide Empirical Therapy of Bloodstream Infections in the Emergency Department.

INTRODUCTION: Traditional cumulative antimicrobial susceptibility test report (cAST) have limitations in supporting empiric antibiotic selection decisions. We evaluated the complementary roles of weighted-incidence syndromic combination antibiogram (WISCA) and the traditional cumulative antimicrobial susceptibility test report (cAST) in guiding empirical therapy for bloodstream infections (BSIs) in the emergency department (ED). METHODS: We performed a retrospective cohort study including all consecutive adult BSI episodes attended in the ED of a tertiary-care hospital in Barcelona, Spain (January 2022-December 2024). A hospital cumulative antimicrobial susceptibility test report (cAST) was generated according to standard methodology for pathogens with ≥30 bloodstream isolates. WISCA coverage was calculated for eight monotherapy and six combination empirical regimens by weighting in vitro susceptibility according to pathogen frequency. Coverage was further stratified by infection acquisition (community-acquired vs. healthcare-associated) and source of infection. RESULTS: A total of 1716 BSI episodes were analyzed (median age 78 years; 55.2% male); 69.4% were healthcare-associated and the main source was urinary tract infection (42%). The 1913 isolates recovered were predominantly Gram-negative (70.3%); ESBL-producing Enterobacterales accounted for 24.9% of Enterobacterales. The cAST covered 87.3% of isolates after species grouping, whereas WISCA incorporated all pathogens. Highest WISCA coverage among monotherapies was observed with imipenem (94.8%), meropenem (91.3%), and piperacillin/tazobactam (89.6%). Ceftriaxone coverage increased from 65.1% to 88.7% when combined with amikacin, whereas carbapenem-based combinations provided only limited additional coverage overall (meropenem plus amikacin, 92.8%; meropenem plus vancomycin, 97.6%). Coverage was consistently lower in healthcare-associated than in community-acquired BSIs, particularly for ceftriaxone, fluoroquinolones, and amoxicillin/clavulanic acid, and was highest in respiratory-source BSIs. CONCLUSIONS: WISCA complements the cAST by providing regimen-level, syndrome- and subgroup-specific coverage estimates, capturing pathogen diversity missed by species-threshold cAST reporting. Integrating WISCA into local antimicrobial-stewardship programs may refine empirical treatment choices for ED bacteremia, especially according to the BSI source and patient-specific characteristics, while supporting judicious use of broad-spectrum agents.