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


Review: Sedative recovery rate with midazolam is no faster than with diazepam

ACP J Club. 1995 Mar-April;122:40. doi:10.7326/ACPJC-1995-122-2-040

Source Citation

Ariano RE, Kassum DA, Aronson KJ. Comparison of sedative recovery time after midazolam versus diazepam administration. Crit Care Med. 1994 Sep;22:1492-6.



To compare, through a qualitative review of the literature, the sedative recovery rates of intravenous midazolam and diazepam used for short-term sedation.

Data sources

Studies were identified using MEDLINE (1977 to 1993), InPharma (1981 to 1993), and bibliographies of retrieved articles.

Study selection

Selected studies compared intravenous midazolam and diazepam for short-term sedation (a single sedative event with either a titrated or fixed dose) in adults. Studies were excluded if they did not assess sedative effects. 28 studies involving 1901 patients were eligible for inclusion.

Data extraction

Dosing information and the differences in mean times to sedative recovery were noted. In 8 studies in which necessary data were available, observed differences were weighted by study sample size and then combined. 15 minutes was considered a clinically relevant difference in sedative recovery time.

Main results

16 of the 28 studies were of sedation for endoscopy; 6 involved dental procedures; and 6 involved drug administration in volunteers, prelocal anesthesia, plastic surgery, and cataract surgery. In 8 trials (29%), sedative recovery rates with diazepam were faster than those with midazolam (P < 0.05); in 19 trials (68%), no difference between the drugs was noted; and in 1 trial, midazolam showed a faster recovery rate than did diazepam (3%) (P < 0.05). For 8 trials (679 patients) with a commonly defined sedative recovery time, the mean weighted time difference was 4 minutes, 16 seconds, favoring diazepam for faster recovery. The median dosing ratio for these trials was 2.1:1 for diazepam over midazolam, with the midazolam dose varying from 0.07 to 0.16 mg/kg of body weight. Formal meta-analysis was not done because of the lack of homogeneity among studies.


Most studies comparing diazepam with midazolam showed similar sedative recovery rates. A faster recovery rate was achieved with diazepam in one third of the trials. The differences in elimination half-life for these 2 benzodiazepines do not correspond to differences in the sedative pharmacodynamic effects occurring with short-term administration.

Source of funding: Not stated.

For article reprint: Dr. R.E. Ariano, Department of Pharmacy, St. Boniface General Hospital, 409 Tach, Winnipeg, Manitoba R2H 2A6, Canada. FAX 204-237-2165.


Through a qualitative review of the literature, Ariano and colleagues concluded that no clinically important difference in recovery rates from sedation existed between diazepam and midazolam. Clinicians must not assume from this study, however, that diazepam and midazolam can be interchanged freely for short-term sedation. The lack of a validated sedation assessment tool makes it difficult to know if the degrees of recovery were similar in the compared studies (1). Also, the question of whether equipotent doses of drugs were used still remains. In most of the reviewed studies, midazolam was considered to be approximately twice as potent as diazepam; however, recent clinical evidence indicates that this ratio may be too low. According to 1 study, the diazepam-to-midazolam potency ratio may be closer to between 3:1 and 6:1 than to 2:1 (2).

This study does raise an important point about short-term sedation. Elimination half-lives of benzodiazepines cannot be used to predict recovery time. Because most benzodiazepines are lipid soluble, which allows rapid entry into the central nervous system and redistribution out of the central nervous system, the onset and recovery times of benzodiazepines with short- or long-elimination half-lives are relatively similar. Differences seen in clinical activity may depend more on receptor affinity (3).

Unfortunately, knowledge gained from studying the recovery rates from short-term sedation cannot easily be extrapolated to longer-term sedation needs, such as for patients in the intensive care unit. After multiple doses are administered, drug accumulation and tolerance, the presence of long-acting active metabolites, and renal and hepatic dysfunction assume more important roles in the duration of drug activity.

Eugene Y. Cheng, MD
Medical College of WisconsinMilwaukee, Wisconsin, USA.


1. Hansen-Flaschen J, Cowen J, Polomano RC. Beyond the Ramsay scale: need for a validated measure of sedating drug efficacy in the intensive care unit. Crit Care Med. 1994;22:732-3.

2. Nuotto EJ, Korttila KT, Lichtor JL, Ostman PL, Rupani G. Sedation and recovery of psychomotor function after intravenous administration of various doses of midazolam and diazepam. Anesth Analg. 1992;74:265-71.

3. Colburn WA, Jack MC. Relationships between CSF drug concentrations, receptor binding characteristics, and pharmacokinetic and pharmacodynmaic properties of selected 1,4-substituted benzodiazepines. Clin Pharmacokinet. 1987; 13:179-90.

Authors' Response

We disagree with Dr. Cheng's comments about whether equipotent doses were achieved. 19 of the 28 trials (68%), which included more than 1500 patients, involved titrating the 2 intravenous benzodiazepines to clinical sedative end points. The weighted (to sample size) dosing ratio from these 19 trials was 1.73:1. The remaining 8 trials in which responses were not titrated had a pooled weighted dosing ratio of 2.01:1.