Previous Abstract | Next Abstract
Printable Version
October 26, 2015
1:00:00 PM - 3:00:00 PM
Room Hall B2-Area A
The Effect of Midazolam Presentation Dose on the Quantity of Midazolam Administered
Brent D. Ershoff, M.D., Sheila Ganjian, B.S., Joe C. Hong, M.D.
UCLA, Los Angeles, California, United States
Disclosures: B.D. Ershoff: None. S. Ganjian: None. J.C. Hong: None.

Midazolam is commonly administered for preoperative anxiolysis and sedation. According to established references, the premedication dose ranges from 0.02 mg/kg to 0.07 mg/kg. The wide dosing range may be due in part to variability in physician titration based on the patient’s age, medical comorbidities, anesthetic plan, as well as inter-physician variability. The dose in which midazolam was presented to anesthesiologists at UCLA was recently changed. In 2012, midazolam was available to anesthesiologists in 2mg vials. In 2013, due to a change in the supply chain, midazolam was available in 3mg syringes. We hypothesized that the dose in which midazolam is presented was associated with the quantity of midazolam administered.


This was a retrospective observational study examining the association between midazolam presentation dose and midazolam administration dose. We collected data on adult patients receiving general anesthesia at UCLA in 2012 (before the midazolam supply change) and patients in 2013 (after the supply change). The preoperative dose of midazolam administered as well as patient demographics were collected.  Comparisons between two groups were performed using fisher’s exact test for categorical outcome variables, and the independent samples t-test was used for continuous outcome variables.  


Data on 103 patients in the 2012 group and 134 patients in the 2013 group were included in the analysis.  Patient demographics including age, gender, ASA classification, and weight were not statistically different between the two groups.  The mean dose of midazolam administered in 2012 was 2.02 mg whereas the mean dose of midazolam administered in 2013 was 2.74mg (mean difference 0.71mg, 95 percent confidence interval: 0.44mg – 0.99mg, p<0.0001). In 2012, midazolam doses greater than or equal to 3mg were administered only 7.8 percent of the time as compared to 2013 when doses greater than or equal to 3mg were administered 57.5 percent of the time (odds ratio = 16.0, 95 percent confidence interval: 7.2 – 35.7, p<0.0001).


Our data show that the presentation dose of midazolam is associated with the administration dose.  This suggests that how midazolam is presented to physicians alters their administration behavior.  We speculate that the desire to give the full presentation dose, and thereby avoiding having to document a controlled substance’s waste, may explain part of this phenomenon. These results raise the question as to whether changes in midazolam presentation dose may potentially affect clinical outcomes such as postoperative Aldrete score and PACU length of stay. These data also raise the question as to whether similar phenomena occur with changes in the presentation of other drugs.

Copyright © 2015 American Society of Anesthesiologists