Abstract
Background:
Data comparing benzodiazepine (BZD) to nonbenzodiazepine (non-BZD) sedatives in patients on mechanical ventilation (MV) demonstrate non-BZD sedation is associated with reduced duration of MV and lower mortality and delirium rates in medical and cardiac intensive care unit (ICU) populations. Limited data exist in surgical/trauma ICU (STICU) populations.
Objective:
Evaluate BZD versus non-BZD sedation on days alive and free of MV at 28 days in STICU patients.
Methods:
This single-center, institutional review board–approved, retrospective cohort study evaluated adult patients admitted to the STICU, intubated for ≥24 hours, and required continuous non-BZD versus BZD sedation in a level I trauma academic hospital. Primary outcome was days alive and free of MV at 28 days. Secondary outcomes included percent of time patients were oversedated and rates of delirium and ventilator-associated pneumonia (VAP). A multivariable model was constructed to determine independent predictors of being alive and free of MV at 28 days.
Results:
After screening, 209 patients were included (BZD, n = 88; non-BZD, n = 121). BZD patients had higher Charlson comorbidity index (CCI) and APACHE II scores. Non-BZD patients had higher rates of traumatic brain injury. Patients receiving non-BZD sedation experienced more days alive and free of MV at 28 days compared with BZD sedation (24 [interquartile range, IQR = 22-25] vs 24 [IQR = 20-25]; P = 0.002). In a multivariable model, including APACHE II score and CCI, non-BZD sedation remained independently associated with increased odds of being alive and free of MV at 28 days (adjusted odds ratio: 6.98; P = 0.002). In addition, non-BZD sedation was associated with less time being oversedated (P < 0.0001), less delirium (P = 0.003), and lower rates of VAP (P = 0.0007).
Conclusions and Relevance:
In STICU patients requiring MV, non-BZD sedation was associated with more ventilator-free days, less time oversedated, reduced delirium, and reduced VAP. These findings extend prior ICU sedation data into the STICU population supporting the use of non-BZD strategies when clinically appropriate to optimize patient outcomes.
Get full access to this article
View all access options for this article.
