Characteristics associated with culture-positivity versus antibiotic administration in patients with cirrhosis on ICU admission: a retrospective cohort study in 5 hospitals
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Objective
To test the hypothesis that selected SIRS and SOFA criteria measured at ICU admission have divergent associations with (1) the development of new culture-confirmed infections and (2) early empiric broad-spectrum antibiotic receipt in critically ill patients with cirrhosis.
Design
Retrospective, observational study using electronic health record (EHR) data.
Setting
Both community and academic ICUs within a five-hospital health system between July 2017 and December 2025.
Subjects
Adult patients >18 years of age diagnosed with cirrhosis admitted to the ICU within 96 hours of hospital admission. Patients transferred from outside facilities and those with positive cultures collected >24 hours before ICU admission were excluded.
Interventions
None.
Measurements
Multivariable logistic regression tested the association of the most aberrant values in nine SIRS/SOFA vital sign and laboratory parameters measured during the 24 hours prior to ICU admission with (1) the development a newly positive culture collected during the 24 hours before through the 96 hours after ICU admission and (2) receipt of broad-spectrum antibiotics during the 24 hours before through the 24 hours after ICU admission.
Main Results
The presence of hyperthermia, hypothermia, hypoxia, or hypotension at ICU admission was associated with both culture positivity and antibiotic receipt. The presence of leukocytosis was associated with antibiotic receipt but not culture positivity. Among patients in the 21.6% of encounters with newly positive cultures, normothermia and normal WBC were common (respectively present in 97.2% and 77.5% of encounters) and associated with antibiotic omission at ICU admission despite the subsequent development culture positivity.
Conclusions
Traditional physiologic markers of infection are often absent in critically ill patients with cirrhosis who develop new culture-confirmed infections at ICU admission, possibly contributing to delayed antibiotics. There may be opportunities to improve antibiotic delivery and stewardship at ICU admission in this high-risk population.
