Epidemiology of ICU-Onset Bloodstream Infection: Prevalence, Pathogens, and Risk Factors Among 150,948 ICU Patients at 85 U.S. Hospitals

Abstract

Objectives: Bloodstream infections (BSIs) acquired in the ICU represent a detrimental yet potentially preventable condition. We determined the prevalence of BSI acquired in the ICU (ICU-onset BSI), pathogen profile, and associated risk factors.

Design: Retrospective cohort study.

Data sources: Eighty-five U.S. hospitals in the Cerner Healthfacts Database.

Patient selection: Adult hospitalizations between January 2009 and December 2015 including a (≥ 3 d) ICU stay.

Data extraction and data synthesis: Prevalence of ICU-onset BSI (between ICU Day 3 and ICU discharge) and associated pathogen and antibiotic resistance distributions were compared with BSI present on (ICU) admission (ICU-BSI POA ); and BSI present on ICU admission day or Day 2. Cox models identified risk factors for ICU-onset BSI among host, care setting, and treatment-related factors. Among 150,948 ICU patients, 5,600 (3.7%) had ICU-BSI POA and 1,306 (0.9%) had ICU-onset BSI. Of those with ICU-BSI POA , 4,359 (77.8%) were admitted to ICU at hospital admission day. Patients with ICU-onset BSI (vs ICU-BSI POA ) displayed higher crude mortality of 37.9% (vs 20.4%) ( p < 0.001) and longer median (interquartile range) length of stay of 13 days (8-23 d) (vs 5 d [3-8 d]) ( p < 0.001) (considering all ICU stay). Compared with ICU-BSI POA , ICU-onset BSI displayed more Pseudomonas , Acinetobacter , Enterococcus, Candida , and Coagulase-negative Staphylococcus species, and more methicillin-resistant staphylococci, vancomycin-resistant enterococci, ceftriaxone-resistant Enterobacter , and carbapenem-resistant Enterobacterales and Acinetobacter species, respectively. Being younger, male, Black, Hispanic, having greater comorbidity burden, sepsis, trauma, acute pulmonary or gastrointestinal presentations, and pre-ICU exposure to antibacterial and antifungal agents was associated with greater ICU-onset BSI risk after adjusted analysis. Mixed ICUs (vs medical or surgical ICUs) and urban and small/medium rural hospitals were also associated with greater ICU-onset BSI risk. The associated risk of acquiring ICU-onset BSI manifested with any duration of mechanical ventilation and 7 days after insertion of central venous or arterial catheters.

Conclusions: ICU-onset BSI is a serious condition that displays a unique pathogen and resistance profile compared with ICU-BSI POA . Further scrutiny of modifiable risk factors for ICU-onset BSI may inform control strategies.

Figure 1.

Case selection flowchart. BSI present-on-(ICU)-admission (ICU-BSI_POA) represents BSI with onset on the day of admission to the ICU or the day after. BSI acquired in the ICU (ICU-onset BSI) represents BSIs with onset on the third day of ICU admission or thereafter. ^{^}Greater than or equal to 20 yr old; *limited to inpatient period of up to 14 d preceding index ICU admission. BSI = bloodstream infection.

Figure 2.

Distribution of bloodstream pathogens among ICU patients. A, The distribution of bloodstream pathogens across all ICU patients with bloodstream infection (BSI) (green bars). B, The percentage distributions of bloodstream pathogens separately for BSI present on (ICU) admission (ICU-BSI_POA) (blue bars) and BSI acquired in the ICU (ICU-onset BSI) (orange bars), respectively, and displayed as paired graphs to enable visual comparisons. C, The distribution of bloodstream pathogens according to resistance profile separately for ICU-BSI_POA (blue bars) and ICU-onset BSI (orange bars), respectively.