Phenobarbital and symptom-triggered lorazepam versus lorazepam alone for severe alcohol withdrawal in the intensive...

A symptom-triggered lorazepam regimen is the standard for treating alcohol withdrawal syndrome (AWS) in an inpatient setting. However, in severe AWS, lorazepam requirements can reach significant amounts and lead to risk of delirium and propylene glycol toxicity. Phenobarbital has been shown to be an effective adjunctive therapy for AWS, reducing benzodiazepine use, in the emergency department. The purpose of this study is to determine the efficacy and safety of phenobarbital in adjunct to symptom-triggered lorazepam for severe AWS vs. lorazepam alone in the intensive care unit (ICU). A retrospective cohort was conducted at Cleveland Clinic hospitals from 2013 to 2018 of ICU patients with AWS receiving either phenobarbital adjunct to symptom-triggered lorazepam or lorazepam alone. The primary outcome was the total duration of treatment. Secondary outcomes include ICU length of stay, change in CIWA-Ar score at 24 h, incidence of hypotension, mechanical ventilation, and serum osmolar gap. A total of 72 ICU patients were included with 36 patients in each arm. The median duration of treatment in the phenobarbital adjunct arm was 2.7 days (IQR = 1.7-6.4), compared to 3.1 days (IQR = 1.6-4.8) in the lorazepam arm (p = 0.578). The median ICU length of stay was similar between both arms [4.1 days (IQR = 2.4-8.4) vs. 4.5 days (IQR = 2.8-6.1), p = 0.727]. The average change in CIWA-Ar from baseline at 24 h was significantly lower for those who received phenobarbital (1.8 ± 9.0 vs. 6.5 ± 8.5, p = 0.028). Three patients in the phenobarbital-adjunct group received mechanical ventilation after starting phenobarbital treatment. There were no new incidences of hypotension or increased osmol gap >10 mmol/L after starting treatment in both groups. In conclusion, phenobarbital is an effective adjunct to symptom-triggered lorazepam in severe alcohol withdrawal in the ICU with no significant difference in adverse events.