Randomized Controlled Trial of Intramuscular Droperidol Versus Midazolam for Violence and Acute Behavioral Disturbance: The DORM Study

Study objective

We determine whether droperidol, midazolam, or the combination is more effective for intramuscular sedation in violent and acute behavioral disturbance in the emergency department (ED).

Methods

We conducted a blinded randomized controlled trial of intramuscular sedation for violent and acute behavioral disturbance, comparing droperidol (10 mg), midazolam (10 mg), and droperidol (5 mg)/midazolam (5 mg). Inclusion criteria were patients requiring physical restraint and parenteral sedation. The primary outcome was the duration of the violent and acute behavioral disturbance, defined as the time security staff were required. Secondary outcomes included time until additional sedation was administered, staff and patient injuries, further episodes of violent and acute behavioral disturbance, and drug-related adverse effects.

Results

From 223 ED patients with violent and acute behavioral disturbance, 91 patients were included; 33 received droperidol, 29 received midazolam, and 29 received the combination. There was no difference in the median duration of the violent and acute behavioral disturbance: 20 minutes (interquartile range [IQR] 11 to 37 min) for droperidol, 24 minutes (IQR 13 to 35 minutes) for midazolam, and 25 minutes (IQR 15 to 38 minutes) for the combination. Additional sedation was required in 11 (33%; 95% confidence interval [CI] 19% to 52%) droperidol patients, 18 (62%; 95% CI 42% to 79%) midazolam patients, and 12 (41%; 95% CI 24% to 61%) in the combination group. The hazard ratio for additional sedation in the midazolam versus droperidol group was 2.31 (95% credible interval 1.01 to 4.71); for the combination versus droperidol, 1.18 (95% credible interval 0.46 to 2.50). Patient and staff injuries and number of further episodes of violent and acute behavioral disturbance did not differ between groups. There were two adverse effects for droperidol (6%; 95% CI 1% to 22%), 8 for midazolam (28%; 95% CI 13% to 47%), and 2 for the combination (7%; 95% CI 1% to 24%). An abnormal QT occurred in 2 of 31 (6%; 95% CI 1% to 23%) droperidol patients, which was not different from the other groups.

Conclusion

Intramuscular droperidol and midazolam resulted in a similar duration of violent and acute behavioral disturbance, but more additional sedation was required with midazolam. Midazolam caused more adverse effects because of oversedation, and there was no evidence of QT prolongation associated with droperidol compared with midazolam.

Introduction

Violence and aggression in the emergency department (ED) is a difficult and dangerous problem that can result in harm to the patient or staff. The majority of cases are due to acute delirium from alcohol intoxication, are due to psychostimulant toxicity, or are associated with deliberate self-harm and drug overdose.1, 2 Many of these patients will not respond to verbal de-escalation or accept oral medications, and some arrive in police custody already restrained. To allow the safe assessment, diagnosis, and treatment of these patients, parenteral sedation and physical restraint are usually required.

There is ongoing controversy about the safest and most effective medications for sedation of violence and acute behavioral disturbance in the ED.³ There is no type of medication that provides sedation in all patients with no adverse effects. Currently, the 2 major groups of medications used are benzodiazepines (midazolam, diazepam, and lorazepam) and antipsychotics (haloperidol, droperidol, and olanzapine). There is increasing evidence that the benzodiazepines fail to sedate a proportion of these patients because of benzodiazepine tolerance, and when larger doses are used a proportion are oversedated.⁴ Antipsychotics have been associated with cardiac dysrhythmias and QT prolongation,5, 6 and some are only mildly sedating, such as haloperidol.

Droperidol is a highly sedative antipsychotic that was widely used until the Food and Drug Administration in the United States issued a black box warning in 2001 because of concerns about QT prolongation and torsades des pointes. This was based on little evidence⁷ and an unusual number of spontaneous reports on 1 day, mainly from outside the United States.⁸ A systematic review of droperidol use failed to support a strong association with QT prolongation and torsades des pointes,⁷ and droperidol was not found to be commonly associated with torsades des pointes in a systematic review of torsades des pointes cases.⁹ However, the use of droperidol has fallen out of favor since the black box warning despite decades of its effective use in EDs and psychiatric units around the world.10, 11 Few controlled trials have been reported assessing its effectiveness for sedation and safety compared with that of other agents in the ED.

There is limited information about the intramuscular route for sedative drugs in violent and acute behavioral disturbance,² which is often the only possible route of administration for these patients. Intravenous sedation requires increased staffing; otherwise, it is effectively impossible and potentially dangerous to attempt to gain intravenous access. There is a real risk of needlestick injuries or other physical injuries to the staff, and most guidelines suggest intramuscular sedation in this setting. Antipsychotic medications such as droperidol are an option to benzodiazepines for intramuscular sedation, but there are few studies comparing intramuscular benzodiazepines with antipsychotics. To our knowledge, there is only 1 trial of intramuscular droperidol that compared it with midazolam and ziprasidone.² This showed that droperidol was at least as effective as midazolam, whereas midazolam required more rescue sedation. Studies of intravenous droperidol suggest it has a longer duration of action compared with that of benzodiazepines2, 12 but potentially a slower onset of action.¹³ All of these previous trials have used low doses, with many patients requiring further sedation. No previous trial has investigated the combination of intramuscular benzodiazepines and droperidol.

This study aimed to determine whether intramuscular droperidol, midazolam, or the combination results in a shorter duration of the violent and acute behavioral disturbance episode defined by the time security staff are required. In addition, the study aimed to determine for each drug treatment the requirement for additional sedation, staff and patient injuries, further episodes of violent and acute behavioral disturbance, and drug-related adverse effects, including the occurrence of QT prolongation with droperidol.