Stratified cumulative antibiograms and antimicrobial resistance trends among hospitalized patients in a Chinese tertiary hospital: a 7-year retrospective study.

Hospital-wide cumulative antibiograms guide empirical therapy but may conceal clinically relevant resistance heterogeneity across inpatient settings. We conducted a retrospective, single-center study of inpatient antimicrobial susceptibility testing (AST) data collected at a Chinese tertiary hospital from 2019 to 2025. For each patient, the first isolate of each organism in each calendar year was selected and its complete AST panel retained. The primary comparison was between adult intensive care unit (ICU) wards and all other inpatient wards. Temporal trends were assessed using the conventional Cochran-Armitage test with false discovery rate correction, and sensitivity analyses excluded hospital-based SARS-CoV-2 high-activity periods. The analysis included 45,645 index isolates from 26,297 patients. Meropenem resistance among Acinetobacter baumannii isolates was 90.7% in adult ICU wards versus 61.9% in non-ICU wards. Corresponding ICU versus non-ICU percentages were 35.5% versus 20.4% for imipenem-resistant Pseudomonas aeruginosa, 57.0% versus 47.0% for oxacillin-resistant Staphylococcus aureus, and 34.7% versus 17.2% for imipenem-resistant Klebsiella pneumoniae. Hospital-wide percentages were lower than the corresponding adult ICU percentages. Carbapenem-resistant A. baumannii and K. pneumoniae declined after false discovery rate correction, whereas methicillin-resistant S. aureus, carbapenem-resistant P. aeruginosa, and levofloxacin-resistant Escherichia coli showed no robust temporal trends. Excluding SARS-CoV-2 high-activity periods did not materially change the main findings. Care-setting-stratified cumulative antibiograms identified resistance heterogeneity not apparent from hospital-wide summaries and may complement hospital-wide reporting for local empirical therapy and stewardship guidance.