Clinical characteristics, risk factors and outcomes of Klebsiella pneumoniae pneumonia developing secondary Klebsiella pneumoniae bloodstream infection

Abstract

Abstract Purpose The clinical characteristics of Klebsiella pneumoniae (KP) pneumonia and KP bloodstream infection (KP-BSI) are often reported, while the risk factors for KP pneumonia developing into secondary KP-BSI (KP-pneumonia/KP-BSI) are largely unknown. Therefore, this study attempted to investigate the clinical characteristics, risk factors and outcomes of KP-pneumonia/KP-BSI. Methods A retrospective observational study was conducted at a tertiary hospital between January 1, 2018, and December 31, 2020. The patients were divided into groups of KP pneumonia alone and KP pneumonia/KP-BSI, and the clinical information were collected from medical records electronic system. Results A total of 409 patients were finally recruited. According to the multivariate logistic regression analysis, male sex (adjusted odds ratio [aOR] 3.7; 95% CI, 1.44–9.5), immunosuppression (aOR, 13.52; 95% CI, 2.53,72.22), APACHE II score higher than 21 (aOR, 3.39; 95% CI, 1.41–8.12), serum procalcitonin (PCT) levels above 1.8 ng/ml (aOR, 6.37; 95% CI, 2.67–15.27), ICU stay of more than 2.5 days before pneumonia onset (aOR, 1.09; 95% CI, 1.02,1.17), mechanical ventilation (aOR, 4.96; 95% CI, 1.2,20.5), Klebsiella pneumoniae isolates producing extended spectrum β-lactamase (ESBL-positive KP) (aOR, 12.93; 95% CI, 5.26–31.76), and inappropriate antibacterial therapy (aOR, 12.38; 95% CI, 5.36–28.58) were independent factors of KP pneumonia/KP BSI. In comparison with the patients with KP pneumonia alone, the patients with KP pneumonia/KP BSI showed an almost 3 times higher incidence of septic shock (64.4% vs. 20.1%, p < 0.01), a longer duration of mechanical ventilation, and longer lengths of ICU stay and total hospital stay (median days, 15 vs. 4,19 vs. 6, 34 vs. 17, respectively, both p < 0.01). Additionally, the overall in-hospital crude mortality rate in the patients with KP-pneumonia/KP-BSI was more than two times higher than that in those with KP pneumonia alone (61.5% vs. 27.4%, p < 0.01). Conclusion Factors including male sex, immunosuppression, APACHE II score higher than 21, serum PCT levels above 1.8 ng/ml, ICU stay of more than 2.5 days before pneumonia onset, mechanical ventilation, ESBL-positive KP, and inappropriate antibacterial therapy are independent risk factors for KP pneumonia/KP-BSI. Of note, the outcomes in patients with KP pneumonia worsen once they develop secondary KP-BSI, which merits more attention.