Skip to main content
Patient Care

Journal Watch: Ketamine vs. Midazolam for Sedation

Reviewed This Month

Prehospital Sedation With Ketamine vs. Midazolam: Repeat Sedation, Intubation, and Hospital Outcomes

Authors: Holland D, Glober NK, Christopher S, Zahn E  

Published in: Amer J Emerg Med, 2020 May; 38(9): 1,748–53

Ketamine has gotten a lot of attention lately. The safety of prehospital ketamine use for sedation of agitated patients has been questioned in media reports and even by members of Congress. All this attention has caused medical directors around the country to reevaluate the use of ketamine in their systems. The best way to address these questions is with data and research. The authors of this month's very timely study evaluated the efficacy and associated adverse events following prehospital ketamine administration.   

The authors indicated that the most recent prehospital ketamine literature review only identified 10 studies that evaluated ketamine use for agitated patients. They further found that “no research exists on prehospital use of ketamine and associated hospital outcomes, leaving EMS leaders with little data to support or refute” its use. Conventional treatment of agitated patients in the prehospital setting, the authors indicated, includes administration of benzodiazepines for calming. 

This study compared ketamine to midazolam. To evaluate the efficacy of both medications, the authors examined “the need for repeat medical sedation by EMS or emergency department personnel.”  Adverse events were also compared, including the need for airway support or intubation, the need for physical restraints, length of hospital stay, and changes in GCS scores. 

This was a retrospective review of 163 consecutive patients chemically sedated for acute agitation by paramedics and transported to an urban Level 1 trauma center in Indianapolis. The study setting was a large third-service EMS agency, and the study period was from January 1, 2017 to April 1, 2018. During the study period, midazolam was the only benzodiazepine used in this agency, and ketamine was used solely for sedation. The agency had a chemical sedation protocol, and the decision to use ketamine or midazolam was based on the paramedic’s discretion. 

Patients were excluded if they were less than 18 years old. To only assess those with available outcomes, patients were also excluded if they were not transported to the Level 1 trauma center. 

Hospital records were reviewed to capture demographic information and suspicion of illicit drug use. Interestingly, the authors also used in-hospital records to capture prehospital medication dose and time, as well as airway interventions. It is unclear why this information was not obtained directly from prehospital care reports. However, the authors did capture prehospital GCS scores before and after sedation. They also utilized data available in both the in-hospital records and the prehospital care reports to collect data on repeat sedation dosing and timing, disposition details, and length of stay. 


There were 97 (60%) patients who received ketamine and 66 (40%) who received midazolam. In the ketamine group 78% were male, and 62% were male in the midazolam group (p=0.02). Patients in both groups were similar (p>0.05) with respect to age, race, insurance status, and weight. 

Almost three quarters of both ketamine and midazolam administrations were due to reported or suspected illicit drug ingestion. All ketamine administrations were intramuscular. Midazolam was administered intramuscularly (49%), intravenously (36%), and intranasally (15%). Most patients who received ketamine were administered 300 mg; however, four patients received 150 mg due to smaller estimated weights. Most patients who received midazolam were administered 5 mg; however, five patients received 2.5 mg for the same reason. 

GCS data was missing often. Only 50% of the ketamine patients and 66% of the midazolam patients had this information documented before and after sedation. There was no significant difference (p>0.05) noted when examining the change in average GCS achieved with ketamine (average  5, 95% CI: 3.6–6.4) to that of midazolam patients (average  4.5, 95% CI: 3.4–5.6). 

When evaluating all routes of administration, a significantly (p=0.01) greater proportion of patients received repeat sedation within 90 minutes in the ketamine group (47% vs. 27%). When only evaluating the IM route, no significant difference (p>0.05) in the proportion of patients who received repeat sedation was noted. There was no statistically significant difference noted (p>0.05) when evaluating total sedation doses by EMS or the emergency department (2.5 vs. 2.1), use of BVMs (7.2% vs. 4.5%), or use of physical restraints (17% vs. 16%). Intubation rates were similar among the groups. There were six (6.2%) patients intubated in the ketamine group, and five (7.6%) in the midazolam group. Almost three quarters of patients in both groups were discharged (p>0.05). Length of stay in the emergency department (9.7 hours vs. 9.4 hours), hospital (22.0 hours vs. 34.2 hours), and ICU (5.8 hours vs. 10.9 hours) were also similar between the ketamine and midazolam groups (p>0.05). The authors did not report deaths in either group. 


As with all studies, there were some limitations to this one, including a relatively small study population, only using data from one urban EMS agency, and only evaluating patients transported to a Level 1 trauma center. Nevertheless, this is an important study that adds to the available literature on the efficacy and safety of prehospital sedation for acutely agitated patients. The results of this study showed that ketamine and midazolam can be safely administrated in the prehospital environment. The majority of patients were discharged, and no deaths were reported. However, while similar in both groups, intubations were required in some patients. 

The authors rightly note more work needs to be done to prove the efficacy and safety of prehospital sedation for agitated patients. This should include evaluations that include a larger and more diverse patient population as well as prospective randomization.   

Antonio R. Fernandez, PhD, NRP, FAHA, is a research scientist at ESO and an assistant professor in the department of emergency medicine at the University of North Carolina–Chapel Hill. He is on the board of advisors of the Prehospital Care Research Forum at UCLA.

Back to Top