Association Between Sedative Medication Administration and Delirium Development in a Medical Intensive Care Unit

Background: Delirium develops frequently in intensive care unit (ICU) patients. Societal guidelines have suggested that benzodiazepines may cause delirium. This study investigates if a change in sedation administration use over time is associated with changes in delirium incidence. Methods: This was a retrospective cohort study conducted over a 4 year time period in a medical ICU. All data was abstracted from a local data warehouse. The primary outcome of the study was the association between annual cumulative benzodiazepine use and incidence of delirium during the study period. Data was analyzed using descriptive characteristics and Spearman's correlation coefficient. Additionally, multivariate logistic regression was performed to identify independent risk factors for delirium development. Results: From 2015 to 2018, annual total benzodiazepine administration decreased from 62,215 mg to 18,105 mg lorazepam equivalents (p = <.01). The cumulative dose of dexmedetomidine increased, with 657,262 mcg administered in 2015 and 1,476,951 mcg in 2018 (p < .01). No differences in annual delirium incidence were found. Risk factors that were significantly correlated with delirium following multivariate logistic regression included acute respiratory distress syndrome, renal failure, hepatic failure, septic shock, severe alcohol withdrawal, vasopressor use, corticosteroid use, benzodiazepine use, antipsychotic use, opiate use, and propofol use. Conclusions: A profound change in sedation medication paradigm did not influence delirium rates in a medical ICU.