Association Between Subcutaneous Medication Administration and ICU Delirium

Introduction: Delirium is the most common manifestation of brain dysfunction in intensive care unit (ICU) patients and is a major contributor to morbidity, mortality, and prolonged length of stay. Pain has been identified as a significant contributor to delirium, but the role of mild, repetitive sources of pain (e.g., from subcutaneous drug administration) has not been investigated. The purpose of this study is to determine whether there is an association between ICU delirium and administration of subcutaneous medications like enoxaparin, heparin, and insulin in the ICU.


Methods: This retrospective chart review included critically ill adults admitted to the ICU for ≥72 hours from January to December 2020. Patients were excluded if their Richmond Agitation Sedation Scale score was -4 to -5 or if they were admitted with an acute neurological diagnosis, prior cognitive impairment, alcohol withdrawal, or dementia. The primary and secondary outcomes were delirium, defined by CAM-ICU, and mortality. Pearson correlation was used to explore the relationships between independent variables (i.e., number of different subcutaneous medications, individual medications, point of care glucose [POCG] tests) and outcomes with alpha < 0.05 indicated significance.


Results: There was a mean of 1.5±0.6 different subcutaneous medications administered in the ICU among 404 included patients. Number of subcutaneous medications was positively correlated with delirium (R=0.128; p=0.015) and mortality (R=0.112, p=0.032). Administration of subcutaneous insulin was the only individual medication associated with delirium (R=0.134; p=0.007) while enoxaparin was the only individual medication associated with mortality (R=0.130; p=0.032).The number of POCG tests in the ICU correlated with delirium (R=0.499, p< 0.001) and mortality (R=0.136, p=0.011).


Conclusions: A weakly positive correlation was found between number of subcutaneous medications in the ICU and delirium. Given the stronger correlation with POCG checks, this may be driven by frequency of the mild but repetitive noxious stimuli. Future research should clarify this association and seek to determine whether positive interventions to reduce mild noxious stimuli reduce the incidence of delirium.