Search   |  Advanced

Research Summary

Practice Guidelines for Managing Alcohol Withdrawal Delirium

Alcohol withdrawal delirium can cause serious morbidity and mortality if not treated appropriately. To inform clinical practice, researchers conducted a structured review and meta-analysis (including 9 prospective controlled trials) and developed evidence-based guidelines for managing alcohol withdrawal delirium.

Compared with neuroleptics, sedative-hypnotics were more effective at decreasing mortality (in 2 trials that had any deaths), and at shortening the duration of delirium (in 3 of 4 trials). In 2 studies reporting the time required to control agitation, intravenous diazepam was better than paraldehyde per rectum in 1, but intramuscular diazepam was no different from oral barbital in the other.

Based on these findings and review of other data, the researchers recommended the following:

  • providing comprehensive monitoring and supportive care
  • using parenteral, rapid-acting sedative-hypnotics (preferably benzodiazepines due to their more favorable therapeutic/toxic index) to achieve light sedation
  • considering pentobarbital or propofol if agitation is not controlled with initial large doses of benzodiazepines (based on case reports)
  • considering neuroleptics only when the patient has continued agitation, disturbed thinking, or perceptual disturbances despite sedative-hypnotic treatment


The practice guidelines outlined in this paper are very practical and reasonable. Although the studies examined are limited (the 9 trials were all published before 1979, 5 of the 9 included fewer than 20 subjects per treatment group, and conclusions about mortality were based on only 9 deaths), the evidence and years of clinical experience with these drugs support the use of sedative-hypnotics, primarily benzodiazepines, for alcohol withdrawal delirium.

Kevin L. Kraemer, MD, MSc


Mayo-Smith MF, Beecher LH, Fischer TL, et al. Management of alcohol withdrawal delirium: an evidence-based practice guideline. Arch Intern Med. 2004;164(13):1405–1412.
(view abstract)