Popis: |
Background and purpose: The incidence rate of bloodstream infection (BSI) in patients with malignant tumor increases gradually with the progress of anti-tumor treatment. The treatment outcome is closely related to the infection of pathogen. At the same time, blood glucose also has a significant impact on the occurrence and development of the disease in this type of patients. This research aimed to retrospectively analyze the distribution characteristics of infectious pathogens isolated in patients with malignant tumors combined with bloodstream infection, the significance of serum glucose detection and prognostic value of its variation trend. Methods: Data of 434 malignant tumor patients with BSI and 409 patients without BSI treated in Fudan University Shanghai Cancer Center were retrospectively analyzed in this research. We utilized SPSS 26.0, Graphpad, Medcalc and Office software etc. to statistically analyze all the data covering clinical characteristics, infectious pathogens and biochemical parameters which were collected from Oct. 2019 to Dec. 2022. Results: The top three isolates in malignant tumor patients with BSI were Escherichia coli (29.4%), Klebsiella pneumoniae (13.8%) and Pseudomonas aeruginosa (4.8%). There were two or more mixed pathogenic bacteria in 8.5% patients. There were totally 100 deadly strains, among which the top three isolates were Escherichia coli, Klebsiella pneumoniae and Pseudomonas aeruginosa, accounting for 21.0%, 12.0% and 10.0% of the diseases, respectively. Survival analysis showed that mixed infection had a poorer prognosis compared to single infection (P=0.000). The fasting blood glucose level at the initial stage of symptoms was significantly higher in malignant tumor patients with BSI (median 7.39 mmol/L, interquartile range 5.95-9.88 mmol/L) than in tumor patients without BSI (median 5.97 mmol/L, interquartile range 5.25-7.06 mmol/L, P=0.000). The area under curve (AUC) of the receiver operating characteristic (ROC) curve of the patients with BSI determined by the fasting glucose level at the beginning of the disease was 0.718, which was higher compared with procalcitonin (PCT), the classic diagnostic marker (AUC=0.708). The combination of these two parameters could even improve diagnostic efficiency (AUC=0.761). Furthermore, survival analysis showed that the prognosis of patients with high level of fasting glucose at the beginning of BSI was poor (HR=3.067, 95% CI: 1.375-6.838, P=0.000). In addition, the greater the glycemic variability at the beginning of BSI, the higher the risk of death was shown (HR=2.150, 95% CI: 1.125-4.109, P |