A RETROSPECTIVE REVIEW AND ASSESSMENT OF BENZODIAZEPINES IN THE TREATMENT OF ALCOHOL-WITHDRAWAL IN HOSPITALIZED-PATIENTS

Little information has been published concerning differences among the benzodiazepines in treating hospitalized patients with severe symptom s of alcohol withdrawal. We attempted to determine the length and type of hospital stay, and the pattern and appropriateness of administrati on, dosage requirements, and costs associated with benzodiazepines in patients undergoing alcohol withdrawal. A 1-year retrospective analysi s was performed for 57 hospitalized patients. Appropriate therapy was defined as lorazepam for patients 60 years and older or those with hep atic dysfunction, and chlordiazepoxide or diazepam for all other patie nts. Drug costs were calculated based on acquisition costs. The mean n umber of days of benzodiazepine treatment and length of stay in the in tensive care unit (ICU) were 6.2 days (range 1-30 days) and 3.9 days ( range 0-12 days), respectively. Fifty-six patients were admitted to th e ICU for management or for monitoring continuous-infusion lorazepam; one patient received chlordiazepoxide on a general ward. Total mean lo razepam infusion required per patient was 324 mg (range 2-5956 mg). Th e total benzodiazepine acquisition cost was $56,489 (mean $1009, range $0.06-7157/patient). The total costs of benzodiazepine acquisition an d ICU charge were $404,346 (mean $7462/patient). Based on our criteria , 41 of 57 patients could have been treated appropriately with chlordi azepoxide, which would have resulted in an estimated drug-acquisition cost savings of at least $37,000. Mean benzodiazepine dosage requireme nts in patients hospitalized for alcohol withdrawal appear higher than previously reported. Approximately 70% of our patients were not of ad vanced age and had no evidence of organ dysufunction, and therefore, c ould have been treated with an oxidized benzodiazepine (i.e., chlordia zepoxide). Significant acquisition costs were associated with continuo us-infusion lorazepam.