Determinants of mortality in carbapenem-resistant Klebsiella pneumoniae ICU Infections: a multivariable clinical and predictive modeling analysis
摘要
This study primarily aimed to identify independent predictors of ICU mortality among adult patients with microbiologically confirmed carbapenem-resistant Klebsiella pneumoniae (CRKP) infection. Secondary objectives included characterization of infection types, antimicrobial resistance patterns, and ICU length of stay within this CRKP cohort.
MethodsThis retrospective study included 180 adult ICU patients with microbiologically confirmed CRKP infection between January 2020 and December 2024. Demographic and clinical variables—including comorbidities, vasopressor use, mechanical ventilation, septic shock, and ICU length of stay—were extracted from electronic hospital records. Antimicrobial regimens and susceptibility profiles were recorded. Univariate and multivariate logistic regression analyses were used to determine independent predictors of mortality.
ResultsThe overall ICU mortality rate was 75% (135/180). Non-survivors were significantly older than survivors (75 vs. 61 years, p = 0.003). Vasopressor use (83.7% vs. 24.4%, p = 0.001) and mechanical ventilation (99.3% vs. 75.6%, p = 0.001) were markedly more common among non-survivors. Multivariate analysis identified age (OR 1.04; p = 0.001), vasopressor use (OR 13.42; p = 0.0001), and mechanical ventilation (OR 14.19; p = 0.018) as independent predictors of mortality. ROC analysis showed good discriminatory performance for vasopressor use (AUC 0.810), mechanical ventilation (AUC 0.780), and age ≥ 68 years (AUC 0.720). No statistically significant association between specific antimicrobial combination regimens and survival was observed.
ConclusionsCRKP infections in the ICU are associated with extremely high mortality, primarily driven by advanced age and acute clinical severity indicators. In this cohort, antimicrobial regimen type was not independently associated with survival, although treatment subgroup sizes were limited. Early identification of high-risk patients and optimization of supportive care remain critical.
Clinical trial numberNot applicable.
Graphical Abstract