A Prospective Study of Intramuscular Droperidol or Olanzapine for Acute Agitation in the Emergency Department: A Natural Experiment Owing to Drug Shortages
|
Design
|
Single-center, prospective, open-label, observational study
N= 1,257
|
Objective
|
To compare intramuscular droperidol and olanzapine for treating agitation due to acute alcohol or drug intoxication
|
Study Groups
|
Droperidol (n= 538)
Olanzapine (n= 719)
|
Inclusion Criteria
|
Received either intramuscular droperidol or olanzapine to treat acute agitation due to acute alcohol or drug intoxication
|
Exclusion Criteria
|
Received other intramuscular medications (e.g., lorazepam, midazolam, haloperidol, ziprasidone) or intravenous droperidol or olanzapine
|
Methods
|
On arrival to the emergency department (ED), the treating physician determined the medication and dose. Data collection was performed beginning when patients received medication for agitation, continuing for 120 minutes after medication administration. The Altered Mental Status Scale (AMSS; ordinal scale from -4 [unresponsive] to 4 [most agitated]) score was used to evaluate sedation after medication administration.
|
Duration
|
July 2019 to March 2020
Data collection: 120 minutes after medication administration
|
Outcome Measures
|
Primary outcome: time to adequate sedation (elapsed time between administration of medication and reaching an AMSS score of zero or lower)
Secondary outcomes: need for additional (rescue) medications, total time in the ED, adverse events
|
Baseline Characteristics
|
|
Droperidol (n= 538)
|
Olanzapine (n= 719)
|
Median age (IQR), years
|
40 (31 to 54) |
43 (32 to 54) |
Male
|
377 (70%) |
536 (75%) |
Race/ethnicity
White
Black
Native American or Alaska Native
Hispanic
Asian
|
209 (39%)
202 (38%)
77 (14%)
32 (6%)
7 (1%)
|
287 (40%)
283 (33%)
114 (16%)
23 (3%)
10 (1%)
|
Alcohol concentration > 0
|
466 (86%) |
627 (87%) |
Median dose (IQR), mg
|
5 (5 to 5) |
10 (10 to 10) |
Presumed cause of agitation†
Alcohol intoxication
Illicit substance
Psychiatric illness
Medical
|
461 (86%)
79 (15%)
67 (12%)
9 (2%)
|
625 (87%)
96 (13%)
93 (13%)
6 (1%)
|
IQR, interquartile range
*There were 29 patients with no measured alcohol level (11 in the droperidol group and 18 in the olanzapine group) who were assumed to have a value of zero. Apart from 13 blood measurements, all alcohol concentrations were obtained with a breath alcohol device.
†Patients could have more than one cause for agitation.
|
Results
|
Endpoint
|
Droperidol (n= 538)
|
Olanzapine (n= 719)
|
Median time to adequate sedation (IQR), min*
|
16 (10 to 30)
|
17.5 (10 to 31)
|
Rescue medications
Entire encounter
Before adequate sedation achieved
After adequate sedation achieved
Median time until first rescue medication given (IQR), min
|
89 (17%)
44 (8%)
45 (8%)
36 (20 to 60)
|
173 (24%)
98 (14%)
75 (10%)
37 (22 to 64)
|
Median total time in ED (IQR), min
|
444 (349 to 543)
|
500 (391 to 631)
|
Adverse events
Extrapyramidal events
Hypotension
Hypoxemia
Oxygen supplementation
Intubation
|
6 (1%)
13 (2%)
20 (4%)
6 (1%)
4 (1%)
|
1 (< 1%)
19 (3%)
42 (6%)
30 (4%)
7 (1%)
|
*There were 35 patients who did not achieve adequate sedation within 120 minutes (12 with droperidol and 23 with olanzapine). There were 4 patients with an unknown time to sedation (2 in the droperidol group and 2 in the olanzapine group).
In a sensitivity analysis for the primary outcome which excluded 125 patients (115 droperidol; 10 olanzapine) who had diphenhydramine coadministered with the initial sedative, the estimated between-group difference for time to adequate sedation was -0.6 minutes (95% confidence interval [CI] -2.1 to 0.7 minutes). The hazard ratio for adequate sedation for droperidol compared with olanzapine was 1.11 (95% CI 0.98 to 1.26).
|
Adverse Events
|
See Results section
|
Study Author Conclusions
|
We found no difference in time to adequate sedation between intramuscular droperidol and olanzapine.
|
InpharmD Researcher Critique
|
This was an open-label, observational study, which may have allowed for certain biases due to physician and patient knowledge of treatment. Additionally, this trial was not randomized, and it was reported that drug shortages made either olanzapine (July to September 2019) or droperidol (November 2019 to March 2020) unavailable, making it uncertain whether the patients who were cared for when only droperidol was available were similar to patients cared for when only olanzapine was available; selection bias may have occurred when both drugs were available.
|