Effective spectrum-based antibiotic resistance index for monitoring resistance in Gram-negative bacilli

Abstract Background: Antimicrobial resistance (AMR) is a growing public health threat, and we currently lack accurate measures to track and trend this resistance. We developed the antibiotic resistance index (ARI) that aggregates resistance of Gram-negative bacilli (GNB) into a single metric which can be tracked across healthcare settings and over time. Methods: Culture data were collected from adult patients who met the CDC adult sepsis event criteria across 10 Barnes-Jewish HealthCare (BJC) hospitals between January 2018 and December 2023. An antibiotic’s effective spectrum (AES) was calculated as the ratio of susceptible GNB to all identified GNB. The ARI was calculated as the sum of the AES to which the isolate was resistant. Using the 20 most common GNB and 15 most common anti-GNB antibiotics routinely tested in antibiograms, we calculated the ARI for each BJC hospital during the study years. Results: 18,854 GNB cultured from 12,803 patients meeting CDC adult sepsis event criteria were included. AES varied between 0.15 for ampicillin and 0.94 for amikacin. A. calcoaceticus-baumannii complex had the highest ARI of 6.64 (IQR 4.00–9.28). Median hospital-level ARI fluctuated between 2.12 (IQR 0.40–3.83) in 2018 to 2.20 (IQR 0.34–3.86) in 2023. The ARI trajectories over time varied by medical center. Conclusion: ARI aggregates AMR in GNB and may facilitate monitoring across locations and over time. ARI and antibiotic effective spectra redefine narrow and broad spectrum of activity and offer a starting point for antibiotic utilization metrics.