Targeted Antimicrobial Comparison for Treatment Individualization in Critically Ill Patients With CRAB Infection

The goal of this prospective, multicenter observational study is to learn which antibiotic treatment strategies are associated with better outcomes in critically ill adults with severe infections caused by carbapenem-resistant Acinetobacter baumannii (CRAB). The study will also evaluate the safety and costs of different treatments and develop prediction models to support personalized treatment decisions. The main questions it aims to answer are: Are there differences in 28-day survival between patients receiving polymyxin-based and non-polymyxin-based antibiotic treatments? Which treatment strategies are associated with better clinical recovery, bacterial clearance, and fewer adverse events, particularly acute kidney injury? Can clinical data and machine learning help predict treatment outcomes and support safer, more effective treatment decisions? Researchers will compare patients receiving polymyxin-based combination therapy with those receiving non-polymyxin-based treatments, including high-dose sulbactam-based, tigecycline-based, and newer antibiotic regimens. Treatment decisions will be made by the patients' treating physicians as part of routine clinical care, rather than assigned by the researchers. Participants will: Receive antibiotic treatments determined by their treating physicians. Have clinical information collected during hospitalization, including laboratory tests, bacterial cultures, organ function, antibiotic use, adverse events, and medical costs. Be followed for 28 days after starting targeted antibiotic treatment to assess survival and clinical outcomes. Be contacted by telephone if necessary to confirm their health status after hospital discharge.

Inclusion Criteria: * Age 18 years or older. * Admission to the intensive care unit (ICU) of a participating hospital. * Severe infection caused by carbapenem-resistant Acinetobacter baumannii (CRAB), with CRAB identified as the primary pathogen or a definite co-pathogen. Patients with concomitant infections caused by other pathogens may be included if CRAB remains the primary pathogen or a definite co-pathogen. * At least one eligible infection type: lower respiratory tract infection, bloodstream infection, or intra-abdominal infection. * Lower respiratory tract infection, if present, must meet the diagnostic criteria for ventilator-associated pneumonia, hospital-acquired pneumonia, or severe community-acquired pneumonia, with new or progressive pulmonary infiltrates and compatible clinical manifestations. * Bloodstream infection, if present, must be associated with systemic inflammatory response or sepsis and a positive peripheral blood culture for CRAB. Catheter-related bloodstream infections require removal or replacement of the central venous catheter within 48 hours after diagnosis. * Intra-abdominal infection, if present, must have imaging evidence of an infectious focus and compatible clinical manifestations, with adequate source control achieved before enrollment and no remaining obstruction. * Laboratory confirmation of CRAB by bacterial culture, identification, and antimicrobial susceptibility testing according to current CLSI or EUCAST criteria. Molecular testing alone is insufficient for confirmation. * Receipt of at least one dose of a study-defined intravenous anti-CRAB antibiotic for suspected or confirmed CRAB infection, with the initial treatment strategy classifiable into cohort A, B1, B2, or B3a. * Written informed consent provided by the participant or a legally authorized representative in accordance with ethics committee requirements. Exclusion Criteria: * CRAB identified as contamination, colonization, or of uncertain clinical significance, or not considered the primary pathogen or a definite co-pathogen responsible for the current infection. * Irreversible terminal condition documented before or at Day 0, together with a documented decision by the patient or family to discontinue active antimicrobial treatment intended to control the current infection, withdraw major life-sustaining support, or receive palliative care only. * A do-not-resuscitate order alone does not constitute an exclusion criterion if active antimicrobial treatment and necessary organ support are continued. Death within 24 hours after Day 0 alone must not be used retrospectively to justify exclusion. * Pregnant or breastfeeding women. * Participation within the 3 months before enrollment in another clinical trial that could interfere with the outcomes of this study.