Abstract Background Gram-negative bacterial infections are a major cause of morbidity and mortality in intensive care units (ICU). This study aimed to evaluate clinical and microbiological predictors of in-hospital mortality in ICU patients with Gram-negative bacterial infections. Methods This retrospective cohort study included patients admitted to the ICU at Policlinico Umberto I, Rome, for at least 24 h, with microbiologically confirmed Gram-negative bacterial infections identified from blood cultures and/or respiratory cultures, including bronchoalveolar lavage (BAL) or tracheobronchial aspirate (TBA). Demographic characteristics, comorbidities, microbiological data, and treatment variables were collected from medical records. For the analysis, Gram-negative pathogens were classified as Enterobacterales, including Klebsiella pneumoniae, and non-fermenters, including Acinetobacter baumannii and Pseudomonas aeruginosa. Carbapenem resistance was assessed based on the driver pathogen, defined as the organism isolated from blood cultures when present, or otherwise the respiratory isolate, with priority given to non-fermenters in polymicrobial infections. Results 218 patients were included, with an overall in-hospital mortality of 41.3%. The median age was 56 years (IQR 47–69), and the majority were male (70.2%). Overall, 140 patients (64.2%) were colonized with multidrug-resistant organisms during hospitalization. 278 Gram-negative isolates were identified from blood, BAL, and TBA cultures, including 53 polymicrobial isolates. Carbapenem resistance was observed in 84.6% of Acinetobacter baumannii, 81.2% of Klebsiella pneumoniae, and 36.6% of Pseudomonas aeruginosa isolates. Adequate empiric antibiotic therapy within the first 24–48 h was administered in 29.4% of cases. In univariate analysis, multiorgan failure and increasing age were significantly associated with in-hospital mortality (P= 0.001 and P 0.001, respectively). Adequate empiric therapy showed a borderline association with mortality (P= 0.047). Microbiological factors, including site of isolation, pathogen group, polymicrobial infection, and carbapenem resistance, were not associated with mortality. These findings were confirmed in multivariable logistic regression analysis (Table 1), where multiorgan failure (OR 3.57; 95% CI 1.79–7.12; P 0.001) and increasing age (OR 1.05; 95% CI 1.03–1.07; P 0.001) were independently associated with mortality. Adequate empiric therapy was not associated with mortality after adjustment for confounders. Conclusions In ICU patients with Gram-negative infections, mortality appears to be primarily driven by clinical severity rather than microbiological factors. Further studies are needed to confirm these findings. Table 1.Multivariable logistic regression analysis of factors associated with in-hospital mortality.VariableOR95% CIP-valueAge (ppi)1.051.02–1.080.001Multiorgan failure3.811.77–8.210.001Adequate empiric therapy1.320.60–2.890.494MDR colonization0.550.26–1.170.121Pathogen groupa0.910.45–1.820.787CR organism0.960.47–1.960.910Polymicrobial infection0.720.32–1.590.413OR, odds ratio; CI, confidence interval; ppi, per point increase; MDR, multidrug-resistant; CR, carbapenem-resistantaPathogen group defined as non-fermenting Gram-negative versus Enterobacterales.
Bortolani et al. (2026) studied this question.