Symptom-Triggered Benzodiazepines in Alcohol Withdrawal

From Wiki Journal Club
Jump to: navigation, search
Daeppen JB, et al. "Symptom-triggered vs fixed-schedule doses of benzodiazepine for alcohol withdrawal: a randomized treatment trial". Archives of Internal Medicine. 2002. 162(10):1117-21.
PubMedFull textPDF

Clinical Question

Among patients with alcohol withdrawal, is there a benefit to symptom-based therapy compared to fixed-schedule benzodiazepines?

Bottom Line

This trial demonstrated that the symptom-based approach (based on the CIWA-Ar scale) resulted in shorter hospitalizations and fewer benzodiazepines given compared to fixed-schedule benzodiazepines among patients with alcohol withdrawal.

Major Points

Strategies for the acute management of alcohol withdrawal range widely, with some favoring scheduled benzodiazepines, others favoring scheduled alpha-antagonists and antiepileptics, and still others preferring a symptom-based approach. This study compared scheduled benzodiazepines to as-needed benzodiazepines given according to the revised Clinical Institute Withdrawal Assessment for Alcohol (CIWA-Ar) scale[1], finding that the symptom-based approach resulted in shorter hospitalizations and fewer benzodiazepines given, and was not associated with an increase in adverse events. A major criticism was the small study size (117) which rendered adverse events such as seizures as very rare. Additionally, the study was conducted at acute inpatient psychiatric alcohol withdrawal units, and may not generalize to inpatient medicine wards.


  • Double-blinded, parallel-group, randomized, controlled trial
  • N=117
    • Symptom-triggered benzodiazepines (n=56)
    • Fixed-schedule benzodiazepines (n=61)
  • Setting: two inpatient alcohol withdrawal hospitals in Switzerland
  • Enrollment: August 1999 to October 2000
  • Follow-up: None (only while inpatient)


Inclusion Criteria

  • Adult patients
  • Alcohol dependence by DSM-IV criteria

Exclusion Criteria

  • Last alcoholic beverage >72h prior to admission
  • Daily use of medications to treat alcohol withdrawal for 30 days prior to admission (such as benzodiazepines, barbiturates, clomethiazole)
  • Major cognitive, psychiatric, or medical comorbidity
  • Opiate or stimulant dependence
  • Not fluent in French

Baseline Characteristics

From the symptom-triggered group, which was similar to the fixed-schedule group except where specified.

  • Age: 46.1 years
  • Male: 80.0%
  • White: 96.4%
  • Marital status:
    • Married: 30.4%
    • Separated/divorced/widowed: 50.0%
    • Never married: 19.6%
  • Currently employed: 55.4%
  • 7-day alcohol consumption:
    • Time since last drink: 13.8 hours (symptom-triggered) vs. 19.0 hours (fixed-schedule), P=0.04
    • Total number of drinks: 87.9 drinks
  • 30-day alcohol consumption:
    • Mean drinking days per week: 5.9
    • Mean number of drinks per occasion: 12.0
    • Maximum number of drinking days per week: 6.7
    • Maximum number of drinks per occasion: 21.2
  • Blood alcohol concentration: 0.5 g/L
  • Biological markers:
    • GGT: 210.4 u/L
    • CDT: 8.1%
    • MCV: 96.4 fL
  • Alcohol dependence:
    • DSM-IV criteria: 5.1
  • Withdrawal history:
    • DSM-IV criteria: 76.8%
    • Severity of Alcohol Dependence Questionnaire (SADQ) score (out of 60): 29.2
    • SADQ score ≥35: 32.1%
    • History of seizures and/or delirium: 19.6%


  • Patients randomized to fixed-schedule or symptom-triggered oxazepam
  • Fixed-schedule group received oxazepam 30mg q6h x4 doses, then 15mg x8 doses while awake.
  • Symptom-triggered group received placebo (mannitol) 30mg q6h x4 doses, then 15mg q8h.
  • In each group, CIWA-Ar scores were calculated q30mins and oxazepam was given as needed:
    • For scores 8-15, oxazepam 15mg was given
    • For scores >15, oxazepam 30mg was given
  • Subjects monitored for withdrawal symptoms for 48h after study completion


Results presented as symptom triggered vs. scheduled therapy.

Primary Outcomes

Treated with oxazepam
39% vs. 100% (P<0.001)
Cumulative oxazepam dose
38mg vs. 231mg (P<0.001)
Fixed-schedule oxazepam: 0mg vs. 225mg (P<0.001)
As-needed oxazepam: 38mg vs. 6.9mg (P<0.05)
Treatment duration
20 vs. 63 hours (P<0.001)
Highest CIWA-Ar score
Day 1: 8 vs. 6 (P<0.01)
Score ≥8, day 1: 45% vs. 28% (P not specified)
Day 2: 5 vs. 3 (P<0.01)
Score ≥8, day 2: 21% vs. 7% (P<0.05)
Day 3: 4 vs. 3 (P<0.01)
Score ≥8, day 3: 18% vs. 3% (P<0.01)

Well-being Schedule[2]
Patients were asked retrospectively on the 3rd day of treatment, 1=best, 6=worst

Health concerns last 3 days
4 in each
Anxiety last 3 days
4 in each
Depressed last 3 days
6 in each
Energy last 3 days
6 vs. 5 (P not specified)

Health-Related Quality of Life, adapted from MOS-SF-36[3]
Patients were asked retrospectively on the 3rd day of treatment

Physical functioning last 3 days
92 vs. 84 (P<0.01)
Vitality last 3 days
60 vs. 55 (P not specified)
Energy last 3 days
67 vs. 66 (P not specified)

Adverse events

1 vs. 0 (P not specified)
0 in each
Delirium tremens
0 in each


Funding provided by Wyeth AG, Switzerland.

Further Reading

  1. The CIWA scale at the Veteran's Affairs webpage
  2. Fazio, Anthony F. A concurrent validational study of the NCHS General Well-Being Schedule. US Department of Health, Education, and Welfare, Public Health Service, Health Resources Administration, National Center for Health Statistics, 1977.
  3. MOS-SF-36 Survey