A Retrospective Analysis of Dexmedetomidine Use for Alcohol Withdrawal Syndrome in Critical Care Patients

Abstract

PURPOSE: Benzodiazepines (BZP) are the cornerstone of treatment for alcohol withdrawal syndrome (AWS). Critically ill patients often require high doses of BZP. Dexmedetomidine (DEX) is a selective centrally acting alpha-2 adrenergic agonist, and the evidence supporting the use of DEX for AWS is evolving and currently limited to small studies. To the best of our knowledge, this is the largest retrospective cohort study examining the use of DEX for AWS in critically ill patients. METHODS: Records were reviewed for patients (greater-than or equal to)19 years hospitalized for AWS in a 25-bed mixed medical-surgical intensive care unit (ICU) at an academic hospital during 2010-2013. Patients were categorized by treatment with BZP alone or BZP with adjunct DEX (100 per group). Data were collected on patient demographics, chemistry panels, withdrawal assessment, utilization of DEX, lorazepam (LZP) dose requirement, and adverse events. The primary outcome was efficacy (duration of ICU stay) and safety (hypotension, bradycardia) of adjunct DEX compared with BZP alone for treatment of AWS in critically ill patients. Secondary analysis included BZP utilization. RESULTS: Study subjects were on average 49 years old, male (94%), and Caucasian (70.5%). Baseline characteristics of each group were similar (NS). Compared with controls, patients who received DEX had a reduced ICU length of stay for AWS (3.75(plus or minus)0.22 vs. 4.52(plus or minus)0.23 days, P=0.017) and had more complications (P<0.001). DEX was administered for an average of 2.75(plus or minus)0.18 days, and the dose range was 0.5 ml to 850 ml. Patients who received DEX demonstrated more severe withdrawal days, had more CIWA assessments performed (P=0.003), and increased FiO2 requirements (P=0.002). Longitudinal logistic mixed modeling showed that LZP use was not statistically significant after adjusting for severity of withdrawal. For each additional day of severe withdrawal, there was a 12.5% increase in the odds of requiring a dose of LZP (p<0.0001, 95% CI=8.8%-16.3%). CONCLUSIONS: This study suggests that DEX when added to standard BZP regimen can effectively reduce length of stay in the ICU for treatment of AWS. DEX appeared to not affect LZP utilization when the data was adjusted for severity of withdrawal. Rather, the number of days with severe withdrawal was the key predictor of LZP administration. CLINICAL IMPLICATIONS: DEX may be an effective adjunct therapy in reducing length of stay in the ICU for treatment of AWS. The relationship of DEX and BZP utilization needs to be further investigated.