PHENOBARBITAL VS BENZODIAZEPINES FOR ALCOHOL WITHDRAWAL MANAGEMENT IN MEDICAL ICU: PRELIMINARY RESULTS OF PRE-POST PROTOCOL STUDY

Abstract

TOPIC: Critical Care TYPE: Original Investigations PURPOSE: Phenobarbital (PB) has been suggested as an alternative to benzodiazepines (BZD) for alcohol withdrawal (AW) management. Few studies have examined the efficacy of different PB regimens for the management of AW in the medical intensive care unit (ICU). METHODS: Retrospective, pre-post protocol, cohort study for patients with AW syndrome admitted to the medical ICU at Cleveland Clinic, Fairview Hospital, during the period from January 2019 through April 2021. Historically a symptom-triggered BZD based protocol utilizing CIWA-Ar has been employed. In September 2020, a PB-based protocol was implemented for AW management. The protocol included PB 260 mg intravenous (IV) followed by 130 mg IV repeated doses every 15 - 30 minutes as needed up to a total of 15 mg/kg (ideal body weight) to maintain CIWA-Ar score below 10. We included adult patients who were admitted to medical ICU primarily for AW management. We excluded patients intubated before ICU admission or the presence of a serious medical or surgical diagnosis that would have otherwise dictated admission to the ICU. The primary outcome is ICU length of stay (LOS). Secondary outcomes include hospital LOS, intubation rate, the number of adjunctive agents, medication side effects, requirement for sitter, restraining, ICU readmission and mortality rate. RESULTS: A total of 270 patients were screened, of which only 62 patients with AW were enrolled, 15 in the PB arm and 47 in the BZD arm. Overall, 32% of all patients presented as uncomplicated AW, 42% as delirium tremens, and 13% as AW seizure. No significant differences in basic characteristics between arms were observed. Compared to BZD arm, patients who received PB had significant shorter ICU LOS (median [IQR], 28 hours [24 – 37] vs 59 hours [34 – 115]; P=<.001) and hospital LOS (median [IQR] 4 days [3 – 5] vs 6 days [4 – 10]; P=.006). PB arm required less number of adjunctive agents to control AW (1 [0 – 1] vs 3 [2 – 4]; p=<.001) including less utilization of dexmedetomidine (1 [7%] vs 22 [47%]; p=<.001). None of the patients in the PB arm required restraint use (0 vs 27 [57%]; p=<.001). No significant difference between arms with regards to medication side effects, sitter requirement, ICU readmission, intubation rate, and mortality rate. CONCLUSIONS: The initial findings from our study are suggesting that our proposed PB-AW management protocol is associated with shorter ICU and hospital LOS. It may be considered as an effective and safe alternative to BZD. CLINICAL IMPLICATIONS: Our results will help in evolving AW management guidelines in ICU. DISCLOSURES: No relevant relationships by Saira Afzal, source=Web Response No relevant relationships by BASSEL AKBIK, source=Web Response No relevant relationships by Dina alayan, source=Web Response No relevant relationships by Sura Alqaisi, source=Web Response No relevant relationships by Mahmoud Alwakeel, source=Web Response No relevant relationships by Ellen Immler, source=Web Response No relevant relationships by Talha Saleem, source=Web Response No relevant relationships by Xiaofeng Wang, source=Web Response