Tjitske van Engelen

87 Host response in patients with ICU-acquired pneumonia surgery, trauma), S. aureus colonization status, Acute Physiology And Chronic Health Evaluation (APACHE)-IV score, and immunosuppressed status. Two additional analyses were performed: (1) biomarker trajectory in the subgroup of cases in whom also a sample was taken after the event of ICU-acquired pneumonia, and (2) the association between the time to develop an ICU-acquired pneumonia and baseline biomarker levels. For more details, see the statistical paragraph of Additional file 1. Results Patient characteristics and clinical outcomes ASPIRE-ICU enrolled 1997 patients, of whom 316 (15.8%) developed ICU-acquired pneumonia (cases) and 1681 (84.2%) did not (controls)(Figure 1). From all controls, we randomly selected a subset in a 2:1 ratio to the cases, resulting in 632 controls for host response biomarker analyses. Selected (n=632) and not selected (n=1049) controls did not differ regarding baseline characteristics and clinical outcomes (Additional file 1: Table S1). Table 1 shows baseline characteristics at ICU admission and clinical outcomes of the study population; missing clinical data is depicted in Table S2 in Additional file 1. At baseline, cases were similar to controls, with the exception that cases more often had immunosuppression as comorbidity (7.9% versus 4.4% in controls, P = 0.04) and less often mild liver disease (0.3% versus 3.5%, P = 0.01; Additional file 1: Table S3). In cases the median interval from ICU admission to the diagnosis of ICU-acquired pneumonia was 5 days (interquartile range [IQR] 3-9 days). The causative pathogen was S. aureus in 140 patients (44.3%) and Pseudomonas (P.) aeruginosa in 48 patients (15.2%); in 21 patients (6.6%) both pathogens were assigned as causative. Cases had a longer length of ICU stay, higher readmission rates, and a higher 90-day mortality. Figure 1. Flowchart of patient inclusion. ICU = Intensive Care Unit. 5

RkJQdWJsaXNoZXIy MTk4NDMw