Parenteral Antipsychotic Choice and Its Association With Emergency Department Length of Stay for Acute Agitation Secondary to Alcohol Intoxication

Randomized Double-blind Trial of Intramuscular Droperidol, Ziprasidone, and Lorazepam for Acute Undifferentiated Agitation in the Emergency Department

BACKGROUND

The optimal agent to treat acute agitation in the emergency department (ED) has not been determined. The objective of this study was to compare the effectiveness and safety of intramuscular droperidol, ziprasidone, and lorazepam for acute agitation in the ED.

METHODS

This was a randomized, double-blind trial of ED patients with acute agitation requiring parenteral sedation. The study was conducted under exception from informed consent (21 CFR 50.24) from July 2004 to March 2005. Patients were randomized to receive 5 mg of droperidol, 10 mg of ziprasidone, 20 mg of ziprasidone, or 2 mg of lorazepam intramuscularly. We recorded Altered Mental Status Scale (AMSS) scores, nasal end-tidal carbon dioxide (ETCO₂ ), and pulse oximetry (SpO₂ ) at 0, 15, 30, 45, 60, 90, and 120 minutes as well as QTc durations and dysrhythmias. Respiratory depression was defined as a change in ETCO₂ consistent with respiratory depression or SpO₂  < 90%. The primary outcome was the proportion of patients adequately sedated (AMSS ≤ 0) at 15 minutes.

RESULTS

We enrolled 115 patients. Baseline AMSS scores were similar between groups. For the primary outcome, adequate sedation at 15 minutes, droperidol administration was effective in 16 of 25 (64%) patients, compared to seven of 28 (25%) for 10 mg of ziprasidone, 11 of 31 (35%) for 20 mg of ziprasidone, and nine of 31 (29%) for lorazepam. Pairwise comparisons revealed that droperidol was more effective that the other medications, with 39% (95% confidence interval [CI] = 3% to 54%) more compared to 20 mg of ziprasidone and 33% (95% CI = 8% to 58%) more compared to lorazepam. There was no significant difference between groups in need of additional rescue sedation. Numerically, respiratory depression was lower with droperidol (3/25 [12%]) compared to 10 mg of ziprasidone (10/28 [36%]), 20 mg of ziprasidone (12/31 [39%]), or lorazepam (15/31 [48%]). One patient receiving 20 mg of ziprasidone required intubation to manage an acute subdural hematoma. No patients had ventricular dysrhythmias. QTc durations were similar in all groups.

CONCLUSIONS

Droperidol was more effective than lorazepam or either dose of ziprasidone for the treatment of acute agitation in the ED and caused fewer episodes of respiratory depression.

The Incidence of QT Prolongation and Torsades des Pointes in Patients Receiving Droperidol in an Urban Emergency Department

INTRODUCTION

Droperidol carries a boxed warning from the United States Food and Drug Administration for QT prolongation and torsades des pointes (TdP). After a six-year hiatus, droperidol again became widely available in the US in early 2019. With its return, clinicians must again make decisions regarding the boxed warning. Thus, the objective of this study was to report the incidence of QT prolongation or TdP in patients receiving droperidol in the ED.

METHODS

Patients receiving droperidol at an urban Level I trauma center from 1997-2001 were identified via electronic health record query. All patients were reviewed for cardiac arrest. We reviewed electrocardiogram (ECG) data for both critically-ill and noncritical patients and recorded Bazett's corrected QT intervals (QTc). ECGs from critically-ill patients undergoing resuscitation were further risk-stratified using the QT nomogram.

RESULTS

Of noncritical patients, 15,374 received 18,020 doses of droperidol; 2,431 had an ECG. In patients with ECGs before and after droperidol, the mean QTc was 424.3 milliseconds (ms) (95% confidence interval [CI], 419.7-428.9) before and 427.6 ms (95% CI, 424.3-430.9), after droperidol (n = 170). Regarding critically-ill patients, 1,172 received droperidol and 396 had an ECG. In the critically-ill group with ECGs before and after droperidol mean QTc was 435.7 ms (95% CI, 426.7-444.7) before and 435.8 ms (95% CI, 427.5-444.1) after droperidol (n = 114). Of 337 ECGs suitable for plotting on the QT nomogram, 13 (3.8%) were above the "at-risk" line; 3/136 (2.2%; 95% CI, 0.05-6.3%) in the before group, and 10/202 (4.9%; 95% CI, 2.4%-8.9%) in the after group. A single case of TdP occurred in a patient with multiple risk factors that did not reoccur after a droperidol rechallenge. Thus, the incidence of TdP was 1/16,546 (0.006%; 95% CI, 0.00015 - 0.03367%).

CONCLUSION

We found the incidence of QTc prolongation and TdP in ED patients receiving droperidol to be extremely rare. Our data suggest the FDA "black box warning" is overstated, and that close ECG monitoring is useful only in high-risk patients.

Study Enrollment When "Preconsent" Is Utilized for a Randomized Clinical Trial of Two Treatments for Acute Agitation in the Emergency Department

BACKGROUND

Acute agitation in the emergency department (ED) represents a danger to both patients and their caregivers. Medication is often needed, and few high-quality randomized trials have evaluated the optimal drugs for this vulnerable population. In the United States, as of 2017, randomized trials of drugs typically cannot be conducted under Waiver of Consent (46 CFR 45.116), and Exception From Informed Consent trials (21 CFR 50.24) are limited to life-threatening conditions, are onerous, and require filing an investigational new drug application with the FDA. We sought to conduct a randomized double-dummy trial of inhaled loxapine versus intramuscular haloperidol + lorazepam for acute agitation in the ED by obtaining consent in advance ("preconsent") in patients at risk of future agitation, allowing study drug administration up to 3 years later if the patient presented with acute agitation.

OBJECTIVE

We sought to report the successful enrollment rate of patients preconsented at an earlier ED visit for this trial.

METHODS

This was an analysis of patients age 18 to 64 with bipolar I disorder or schizophrenia preconsented for enrollment in the trial (clinicaltrials.gov, NCT02877108) conducted at a single urban academic center seeing approximately 60,000 patients per year. Eligible patients were assessed for capacity to consent by trained research associates, and informed consent was obtained at an ED visit for the possibility of administering drugs for agitation within the next 3 years. In the event the patient later presented to the ED and the attending physician deemed the patient required treatment for acute agitation, preconsent was confirmed and study drug would be administered.

RESULTS

Over 67 days, 1,461 patients were screened in the ED, 269 had bipolar I or schizophrenia, 194 of whom had a contraindication to inhaled loxapine leaving 75 eligible patients; preconsent was obtained in 43 patients. Four additional patients who had not preconsented were consented for the trial in real time (three by surrogate, one patient had capacity while agitated) resulting in a total of 47 consented patients. Of these 47, a total of 12 were later removed from the study: 10 patients had unrecognized exclusion criteria for inhaled loxapine, one preconsented patient contacted the investigators at a later date and asked to be removed, and one surrogate revoked consent immediately after providing it. Only two patients were successfully enrolled, neither by preconsent: one was enrolled via a surrogate the day of enrollment, and the other was mildly agitated and had capacity to consent. The remaining patient with a valid surrogate consent did not receive study medication.

CONCLUSIONS

Utilization of preconsent to enroll patients in a randomized trial of treatments for acute agitation in the ED requires substantial resources and may not be feasible.

An open-label randomized trial of intramuscular olanzapine versus oral clonidine for symptomatic treatment of opioid withdrawal in the emergency department

Background: Patients with opioid withdrawal often present to the Emergency Department (ED), but many EDs do not have the infrastructure in place to initiate treatment with opioid agonists (methadone or buprenorphine). Therefore, ED management often entails symptomatic control. The purpose of this study was to compare olanzapine to clonidine for the treatment of opioid withdrawal symptoms. Methods: This was a prospective, randomized clinical trial comparing 10 mg of IM olanzapine to 0.3 mg of oral clonidine for symptoms of opioid withdrawal. Adult (18 years and older) ED patients reporting a history of opioid use and symptoms consistent with withdrawal were eligible. Patients were excluded if they had already received treatment during the ED encounter, were pregnant, incarcerated, or unable to provide consent. Patients were randomized 1:1 to receive olanzapine or clonidine for their initial treatment. A baseline Clinical Opiate Withdrawal Scale (COWS) score was calculated. After 30 min, the patient could receive any additional treatment at the ED physician's discretion. The primary outcome was need for additional medication (rescue) within 1 h of study medication administration. Secondary outcomes included change in COWS score and adverse reactions. Results: We enrolled 63 patients (33 olanzapine, 30 clonidine). Demographic characteristics were similar for both groups (median age 45, range 21-67, 54% male) as well as baseline COWS score (median score 11). The median time since last opiate use was 48 h for both groups (range 4-116). Rescue was given within 1 h for olanzapine for 9 (27%) patients and for clonidine in 19 (63%) patients (difference 36%, 95% CI 13-59%). Decrease in COWS score at 1 h was 8.3 for olanzapine and 5.1 for clonidine (difference 3.2, 95% CI 0.3-6). Adverse events were uncommon: akathisia (1, olanzapine), hypotension (2, clonidine), respiratory depression (0). Conclusions: Treatment of opioid withdrawal symptoms with 10 mg of IM olanzapine results in a lower incidence of rescue medication administration and improved symptoms (COWS score) compared to 0.3 mg of oral clonidine.