OBJECTIVE  To investigate the clinical characteristics of patients with carbapenem-resistant Enterobacteriaceae （CRE） Bloodstream Infections （BSI）, identify risk factors and construct a prognostic model, thereby providing a reference for early identification of high-risk patients and optimization of clinical intervention strategies.

METHODS  A retrospective analysis was conducted on the clinical data of 190 patients with CRE BSI admitted to Tianjin Third Central Hospital from Jan. 2016 to Jun. 2025. Demographic characteristics, underlying diseases, invasive procedures and other relevant indicators were collected. Patients were classified as 30-day survivors （n=68） and 30-day non-survivors （n=122） based on all-cause mortality within 30 days after diagnosis. A multivariate logistic regression model was used to summarize risk factors and construct a prognostic model. The model's discrimination, calibration and clinical utility were evaluated through the C-index, area under the curve （AUC）—receiver operating characteristic （ROC）, calibration curve, and decision curve analysis （DCA）. Internal validation was performed via the Bootstrap method.

RESULTS  The primary pathogenic bacterium detected in patients with CRE-BSI was Klebsiella pneumoniae （90.53%）. Urinary system diseases （OR=3.822, 95%CI: 1.601－9.476, P=0.003） and mechanical ventilation （OR=8.737, 95%CI: 2.789－30.356, P＜0.001） were identified as risk factors for poor prognosis in patients with CRE-BSI. The model's C-index was 0.819 （0.818 after validation）, and the AUC was 0.819. The calibration was satisfactory, and decision curve analysis demonstrated clinical net benefit.

CONCLUSIONS  K. pneumoniae is the primary pathogenic bacterium of CRE BSI, and urinary system diseases and mechanical ventilation are risk factors for poor prognosis, particularly in critical patients. The constructed prognostic model demonstrates good performance and high clinical utility. It provides an important basis for early risk stratification and has the potential to guide individualized clinical interventions, ultimately improving patient prognosis.