Empiric antibiotic use and de-escalation practices incommunity-acquired infections in adults: a retrospectiveobservational study at a tertiary care hospital in UAE

Abstract

Background: The increase in antimicrobial resistance (AMR) rates is a global burden, including the Middle East and Gulf Region. Gaining insight into antibiotic use patterns in the United Arab Emirates (UAE) is crucial for informing antimicrobial strategies, such as antibiotic de-escalation, which can play a key role in reducing AMR. Methods: This retrospective observational study collected data from a tertiary care hospital medical record. Eligible patients were adults admitted for management of Community Acquired Pneumonia (CAP), Urinary Tract Infections (UTI), or sepsis between January 2022 and December 2023. Patients’ characteristics such as demographics, comorbidities and severity of infection on admission, treatment received during hospital stay and clinical outcomes were recorded in a suitably designed data collection form. The pattern of antibiotic prescription was analyzed and the rate of empiric broad-spectrum use was evaluated. Eligibility for de-escalation was assessed based on microbiological results. Data analysis was conducted using Stata/SE 12.0 For windows and Microsoft Excel for windows. Results: Data from 207 patients (median age 41) showed that 45.4% had CAP, 40.1% UTI, 14.5% sepsis. Overall, 96 patients (46.4%) had culture specimens obtained pre-treatment. Among CAP and sepsis patients, positive cultures were observed in only 8 cases. Conversely, in UTI patients, 32 cultures were positive, with 28 identifying E. coli as the pathogen, 12 of which were ESBL-producing strains. Broad-spectrum antibiotics were initiated in 70.5% of patients, with meropenem being the most commonly prescribed (57.9%). Of 27 de-escalation- eligible patients, 25.9% were successfully de-escalated. Broad-spectrum prescribing was associated with recent intravenous antibiotic exposure, admission to the Intensive care unit (ICU) and higher C-reactive protein (CRP) levels. The odds of prescribing a broad-spectrum antibiotic increase by 1% for each unit increase in CRP level. The interquartile range (IQR) for duration of hospitalization was 3- 5 days. Successful therapy was achieved in 197 cases. Conclusion: Broad-spectrum antibiotics were frequently initiated empirically, with meropenem being the most common choice. While microbiological data supported de-escalation in 27 cases, it occurred in only a quarter of these. Opportunities exist to optimize empiric therapy and promote antibiotic stewardship strategies such as de-escalation.