Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Benzodiazepines are more effective than neuroleptics in reducing delirium and seizures in alcohol withdrawal

ACP J Club. 1998 Jan-Feb;128:8. doi:10.7326/ACPJC-1998-128-1-008

Related Content in the Archives
Symptom-triggered chlordiazepoxide was effective for alcohol withdrawal

Source Citation

Mayo-Smith MF, for the American Society of Addiction Medicine Working Group on Pharmacological Management of Alcohol Withdrawal. Pharmacological management of alcohol withdrawal. A meta-analysis and evidence-based practice guideline. JAMA. 1997 Jul 9;278:144-51.



To determine, using meta-analysis, the appropriate pharmacologic management of alcohol withdrawal.

Data sources

Studies were identified by searching MEDLINE (1966 to June 1995) using the terms substance withdrawal syndrome and ethyl alcohol. Bibliographies of studies and review articles were also examined.

Study selection

Controlled trials that studied humans and reported clinical end points.

Data extraction

Data were extracted on the interventions, study quality, and patient outcomes. Patient outcomes were severity and rates of the alcohol withdrawal syndrome, delirium, seizures, and completion of withdrawal; entry into rehabilitation; adverse events; and costs.

Main results

42 medications were identified in 143 trials. 65 controlled trials were included in the meta-analysis. All 6 placebo-controlled trials of benzodiazepines showed reduced signs and symptoms of alcohol withdrawal. Benzodiazepines were better than placebo at reducing delirium (risk difference [RD] 4.9 fewer cases of delirium/100 patients, P = 0.04) and seizures (RD 7.7 fewer cases/100 patients, P < 0.001) (Table). Long-acting benzodiazepines showed a trend toward fewer seizures (P = 0.07) compared with short-acting benzodiazepines. 2 studies showed that symptom-triggered dosing was as effective as fixed-dose therapy, but symptom-triggered dosing uses less medication and has a shorter duration of treatment. Neuroleptic agents, including phenothiazines and haloperidol, were less effective than benzodiazepines for reducing delirium (RD 6.6 more cases of delirium/100 patients, P = 0.002) and seizures (RD 11.4 more cases/100 patients, P < 0.001). β-Blockers and clonidine have been shown to reduce peripheral signs and symptoms of alcohol withdrawal, but evidence is insufficient to show reductions in delirium or seizures. Magnesium has not been shown to be effective in the management of alcohol withdrawal.


Benzodiazepines are more effective than phenothiazines in reducing delirium and seizures in patients with alcohol withdrawal and are recommended as first-line treatment for this condition.

Source of funding: No external funding.

For article reprint: Dr. M.F. Mayo-Smith, California Society of Addiction Medicine, 3803 Broadway, Oakland, CA 94611-5615, USA. FAX 510-653-7052.

Table. Benzodiazepines vs placebo for alcohol withdrawal*

Outcomes Benzodiazepines weighted EER Placebo weighted CER RRR (95% CI) Weighted ARR NNT (CI)
Delirium 2.1% 8.7% 83% (41 to 95) 6.6% 16 (10 to 39)
Seizures 0.9% 5.7% 80% (23 to 95) 4.8% 21 (12 to 123)

*Abbreviations defined in Glossary; RRR, ARR, NNT, and CI calculated from data in article.


Mayo-Smith and colleagues have produced a useful meta-analysis, and the resulting guidelines are applicable to psychiatry, primary care, internal medicine, and surgery.

Two medications used in Europe, chlormethiazole and meprobamate, were shown to be effective in reducing signs and symptoms of alcohol withdrawal but were not included because the reductions found in seizures and delirium were not shown to be statistically significant.

The meta-analysis also shows that, when fixed-dose regimens of benzodiazepines are compared with symptom-triggered dosing (symptoms rated according to a nurse-administered withdrawal rating scale), clinical features, including seizures, are well controlled and the total dose taken by the patient tends to be less. This treatment procedure, used in conjunction with the Clinical Institute Withdrawal Assessment-Alcohol (revised) (1) (which is designed to evaluate the severity of the alcohol withdrawal sundrome), is recommended in this guideline.

The meta-analysis does not review the relative benefits of inpatient compared with outpatient withdrawal from alcohol. If the withdrawal is to be assisted by medication, then the conclusions of the authors are helpful: Rapid-acting benzodiazepines are more likely to be abused and are therefore less applicable to outpatient treatment. It is also noteworthy that the more gradual action of longer-acting benzodiazepines may help explain why they have been shown to be more protective against seizures than shorter-acting benzodiazepines.

One point of great practical importance in managing alcohol withdrawal patients was outside the scope of this meta-analysis and guideline. Early treatment with benzodiazepines helps prevent serious complications. Too often in the general hospital, delirium develops before alcohol withdrawal is diagnosed, and management is then much more difficult. Guidelines are also needed on the use of interviews and blood tests to detect alcohol problems in the general hospital setting.

Jonathan Chick, MA, MPhil, MB, ChB
Royal Edinburgh HospitalEdinburgh, Scotland, UK


1. Sullivan JT, Sykora K, Schneiderman J, Naranjo CA, Sellers EM. Assessment of alcohol withdrawal: the Revised Clinical Institute Withdrawal Assessment for Alcohol scale (CIWA-Ar). Br J Addict. 1989;84:1353-7.