Pathogen characteristics and pre-index clinical factors associated with multidrug-resistant organism infections in intensive care unit patients.

BACKGROUND: Multidrug-resistant organism (MDRO) infections remain a major cause of healthcare-associated morbidity in intensive care units (ICUs). Local characterization of pathogen profiles and clinical factors associated with MDRO infection is needed to inform infection prevention and antimicrobial stewardship strategies. METHODS: This single-center retrospective observational study included consecutive adult ICU admissions from January to December 2025 with an ICU length of stay ≥48 h and available microbiological cultures. MDRO infection was defined according to clinical infection criteria and phenotypic non-susceptibility to at least one agent in ≥3 antimicrobial categories. A 1:2 propensity score matching approach was used to reduce baseline imbalance. Univariable and multivariable logistic regression analyses were performed to identify pre-index clinical factors associated with MDRO infection, and model discrimination was assessed using receiver operating characteristic (ROC) analysis. RESULTS: Among 986 eligible patients, 103 developed MDRO infection, yielding an incidence of 10.4%, with a median onset of 6 days after ICU admission. MDRO infections were mainly ventilator-associated/hospital-acquired pneumonia and bloodstream infection. Carbapenem-resistant Acinetobacter baumannii, carbapenem-resistant Enterobacterales, and carbapenem-resistant Pseudomonas aeruginosa were the predominant phenotypes. In the matched cohort (n = 309), pre-index clinical factors independently associated with MDRO infection included Acute Physiology and Chronic Health Evaluation II score (adjusted odds ratio [aOR] 1.04), pre-ICU antibiotic exposure (aOR 1.58), pre-index ICU carbapenem exposure (aOR 1.92), total pre-index systemic antibiotic duration in ICU (aOR 1.11 per day), pre-index mechanical ventilation (aOR 2.06), pre-index continuous renal replacement therapy (aOR 1.69), procalcitonin (aOR 1.39), and albumin (aOR 0.93). The model showed good discrimination, with an area under the ROC curve of 0.836 and a bootstrap-corrected value of 0.821. CONCLUSION: MDRO infections occurred in approximately one-tenth of ICU admissions and were primarily Gram-negative respiratory and bloodstream infections. Illness severity, pre-index antimicrobial exposure intensity and duration, pre-index organ support, and early inflammatory or nutritional biomarkers were independently associated with MDRO infection. These findings may support preliminary ICU risk stratification, but prospective validation and formal time-dependent exposure assessment are warranted.