Association between delay in intensive care unit admission and the host response in patients with community-acquired pneumonia

BACKGROUND: A delay in admission to the intensive care unit (ICU) of patients with community-acquired pneumonia (CAP) has been associated with an increased mortality. Decisions regarding interventions and eligibility for immune modulatory therapy are often made at the time of admission to the ICU. The primary aim of this study was to compare the host immune response measured upon ICU admission in CAP patients admitted immediately from the emergency department (direct ICU admission) with those who were transferred within 72 h after admission to the general ward (delayed ICU admission). METHODS: Sixteen host response biomarkers providing insight in pathophysiological mechanisms implicated in sepsis and blood leukocyte transcriptomes were analysed in patients with CAP upon ICU admission in two tertiary hospitals in the Netherlands. RESULTS: Of 530 ICU admissions with CAP, 387 (73.0%) were directly admitted and 143 (27.0%) had a delayed admission. Patients with a delayed ICU admission were more often immunocompromised (35.0 versus 21.2%, P = .002) and had more malignancies (23.1 versus 13.4%, P = .011). Shock was more present in patients who were admitted to the ICU directly (46.6 versus 33.6%, P = .010). Delayed ICU admission was not associated with an increased hospital mortality risk (hazard ratio 1.25, 95% CI 0.89–1.78, P = .20). The plasma levels of biomarkers (n = 297) reflecting systemic inflammation, endothelial cell activation and coagulation activation were largely similar between groups, with exception of C-reactive protein, soluble intercellular adhesion molecule-1 and angiopoietin-1, which were more aberrant in delayed admissions compared to direct ICU admissions. Blood leukocyte transcriptomes (n = 132) of patients with a delayed ICU admission showed blunted innate and adaptive immune response signalling when compared with direct ICU admissions, as well as decreased gene expression associated with tissue repair and extracellular matrix remodelling pathways. CONCLUSIONS: Blood leukocytes of CAP patients with delayed ICU admission show evidence of a more immune suppressive phenotype upon ICU admission when compared with blood leukocytes from patients directly transferred to the ICU. Trial registration: Molecular Diagnosis and Risk Stratification of Sepsis (MARS) project, ClinicalTrials.gov identifier NCT01905033. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s13613-021-00930-5.