Intramuscular Midazolam Best for Status Epilepticus

Susan Jeffrey

February 16, 2012

February 16, 2012 — A randomized trial confirms that prehospital administration of midazolam using an autoinjector stopped seizures more rapidly and reliably than intravenous (IV) treatment with lorazepam among patients in status epilepticus.

Upon arrival at the hospital, 73% of patients receiving intramuscular (IM) midazolam were seizure-free, compared with 63% of those treated with IV lorazepam.

"This double-blind, randomized trial showed that prehospital treatment with intramuscular midazolam was at least as effective as intravenous lorazepam in subjects in status epilepticus (P < .001 for noninferiority and for superiority)," the researchers, with first author Robert Silbergleit, MD, from the University of Michigan, Ann Arbor, conclude.

"From a clinical standpoint, our view is it doesn’t really matter whether it’s very, very noninferior or a little superior. Either way, this becomes a better way of treating status epilepticus," Dr. Silbergleit told Medscape Medical News. "This really I think provided the high-quality, randomized controlled data that we needed to say that this is not only acceptable, but probably a preferred therapy."

Dr. Robert Silbergleit

Establishing IV access in patients having seizures outside of the hospital can be challenging and time-consuming, the authors note. "Since intramuscular treatments can be given more quickly and reliably than intravenous treatments and have noninferior efficacy, our data support the use of the former route of administration by EMS [emergency medical services] personnel."

The findings, from the Rapid Anticonvulsant Medication Prior to Arrival Trial (RAMPART), are published in the February 16 issue of the New England Journal of Medicine.

Early Termination of Seizures Critical

Early termination of prolonged seizures in status epilepticus using IV administration of benzodiazepines improves outcomes, the researchers write. Increasingly, though, paramedics have begun to use IM midazolam rather than an intravenous agent, "largely because intramuscular administration is faster and is consistently achievable," the authors write.

"We know that the first couple of minutes of treating a seizure is your best opportunity for making a difference," Dr. Silbergleit said. "We’ve known from animal work and clinical work that the longer that status continues, the more and more refractory it becomes to therapy," he added, so the opportunity to treat status in the prehospital setting should be maximized.

However, there have been few clinical trial data supporting the efficacy and safety of IM midazolam. A previous clinical trial led by RAMPART co-principal investigator Daniel Lowenstein, MD, from the University of California, San Francisco, called the Prehospital Treatment of Status Epilepticus (PHTSE) trial, showed that IV lorazepam and IV diazepam were both more effective than placebo in the prehospital environment, with lorazepam appearing to be more effective than diazepam.

Despite that, Dr. Silbergleit noted, diazepam, which can be given rectally or intravenously, has been the staple agent for prehospital treatment because lorazepam is impractical for that environment; it requires refrigeration, is more expensive, and requires IV administration. Small pediatric trials suggested midazolam might be useful for administration without an IV — it is rapidly absorbed across mucous membranes or via muscle but hasn’t been tested in a randomized trial; it also doesn’t carry an indication for the treatment of status.

The RAMPART study was a randomized, double-blind noninferiority trial that compared IM midazolam with IV lorazepam on the termination of seizures before arrival at the hospital. Participating were 4314 paramedics from 33 EMS agencies and 79 receiving hospitals across the United States.

Patients included both children and adults in status epilepticus; patients whose convulsions had persisted for more than 5 minutes and who were still convulsing when paramedics arrived were randomly assigned to receive midazolam given via an investigational IM autoinjector (made by the same company that makes epinephrine autoinjectors for anaphylaxis [EpiPen], Meridian Medical Technologies) or IV lorazepam given via a prefilled syringe (Carpuject System, Hospira).

The RAMPART treatment kit

For eligible patients, paramedics used a kit that contained 1 autoinjector and 1 prefilled syringe, 1 of which contained study drug and the other placebo, and used both to achieve blinding. Adults and children weighing more than 40 kg receiving 10 mg of IM midazolam followed by IV placebo, or IM placebo followed by 4 mg of IV lorazepam. Children weighing 13 to 40 kg received 5 mg of midazolam or 2 mg of lorazepam.

The timing of various events during treatment was captured by using voice recordings.

The primary outcome was absence of seizures at the time of arrival at the emergency department without the need for rescue therapy. The trial tested the hypothesis that IM midazolam would be noninferior to IV lorazepam by a margin of 10 percentage points.

The absolute difference in the primary endpoint was in fact 10 percentage points favoring IM midazolam, with more patients arriving at the hospital with seizures absent than those receiving IV lorazepam.

Table. RAMPART: Primary Endpoint

Endpoint IM Midazolam (n = 448), n (%) IV Lorazepam (n = 445), n (%) Absolute Percentage-Point Difference (95% Confidence Interval) P Value for Noninferiority/Superiority
Seizures absent without rescue therapy 329 (73.4) 282 (63.4) 10 (4.0 - 16.1) < .001


Groups were similar in terms of secondary outcomes, including the need for endotracheal intubation (14.1% vs 14.4% for IM midazolam vs IV lorazepam, respectively) and recurrence of seizures (11.4% vs 10.6%, respectively).

Overall, time to treatment with use of the autoinjector was faster, although once IV treatment was started, time to cessation of seizures was faster than with IM administration. For patients whose seizures had stopped before hospital arrival, median times to active treatment were 1.2 minutes for IM midazolam vs 4.8 for IV lorazepam. Median times from active treatment to cessation of seizures were 3.3 minutes for the IM group vs 1.6 minutes for the IV group.

However, as the researchers anticipated, IM treatment was easier to achieve. Of patients who did not reach a primary endpoint, 31 never received IV treatment because of failure to achieve vascular access; in comparison, only 5 in the IM group didn’t receive study medication because of malfunction or misapplication of the autoinjector.

Adverse event rates were similar in the 2 groups, the researchers note.

The researchers estimate that about one quarter of EMS systems in the United States have already started carrying midazolam to use as a first- or second-line agent in status epilepticus, "so this validates their practice," Dr. Silbergleit noted.

NETT Investigators

The RAMPART study is the first of several studies now being conducted under the auspices of the Neurological Emergencies Treatment Trials (NETT) network, a multidisciplinary clinical trials infrastructure funded by the National Institutes of Neurological Diseases and Stroke (NINDS) to address neurologic emergencies (events such as trauma, stroke, or seizures), Dr. Silbergleit said.

In the 1990s, a network linking emergency medicine and neurology researchers was formed to complete the landmark NINDS rtPA Stroke Study, which established the efficacy of thrombolytic therapy for stroke, he noted. "After that trial, that whole infrastructure went away."

In ensuing years, John Marler, MD, head of clinical trials at NINDS, and William Barsan, MD, from the University of Michigan, Ann Arbor, and others explored the idea of launching an ongoing clinical trial infrastructure for neurologic emergencies. "So the NETT was created, a multidisciplinary collaboration between emergency physicians, neurologists, neurosurgeons, neurointensivists, trauma surgeons, to try to do clinical trials that try to move the intervention early, into the ambulance or the emergency department,  and then continue the intervention into the in-hospital setting," Dr. Silbergleit said.

RAMPART was the first of 5 trials now underway to report its findings. "We’re able to run these with all the same emergency response clinical trial teams, so we’ve got a real economy of scale, a real efficiency, and that’s self-sustaining and really successful," he said.

Interestingly, another collaboration brought the autoinjectors into this trial.

"At the same time that we approached NIH [National Institutes of Health] about doing a study of IM midazolam in status, the Department of Defense, through the NIH Countermeasures Against Chemical Threats (NIH CounterACT) had been developing midazolam in an autoinjector for IM delivery, for use in nerve gas-induced status epilepticus," Dr. Silbergleit said.

For the Departments of Defense (DoD) and Health and Human Services (HHS), then, the RAMPART study was an opportunity to confirm the effectiveness of the autoinjector under development in patients with seizures.

"There was great synergy when we realized that RAMPART was studying a similar problem that was of concern to the chemical defense community," David Jett, PhD, program director for NIH CounterACT and NINDS said in a statement. "This led to a perfect collaboration between HHS and DoD.

"The broader implication of RAMPART is that we now have critical information from studies in humans that a safe and effective tool may one day be available to enhance our public health preparedness," Dr. Jett noted. "Autoinjectors provide a highly practical way to treat hundreds of people quickly during an emergency."

Use of the autoinjector allowed the paramedics in RAMPART to give the IM shot in about 20 seconds and then move on to start the IV treatment, Dr. Silbergleit noted. For now, though, the autoinjectors are still investigational and not yet available for this use.

The investigational autoinjector

"I think the manufacturer of the autoinjectors is very likely to seek approval and sell these devices for EMS use, and I think that the US Food and Drug Administration has indicated that they think that would be a good idea," he said.

Bright Future

In an accompanying editorial, Lawrence J. Hirsch, MD, from the Division of Epilepsy and EEG, Comprehensive Epilepsy Center, Department of Neurology, Yale University School of Medicine, New Haven, Connecticut, points out that results of RAMPART showed that IM midazolam not only was noninferior but was superior to IV lorazepam.

"This difference was due to the more rapid administration of the intramuscular medication (1.2 minutes, vs 4.8 minutes with the intravenous route)," Dr. Hirsch writes. This more rapid administration outweighed the faster cessation of seizures seen with IV treatment once it was given, he noted.

Next steps in this field include home treatment with nasal or buccal benzodiazepines that will soon be widely available and may help prevent status epilepticus among patients at risk for prolonged or repetitive seizures, he writes. A comparison between IM and nasal or buccal administration of midazolam is needed, along with more research to determine the best next step after first-line therapy fails, including the possible use of combinations of medications and neuroprotective agents.

"Thus, the future of care for seizure emergencies is quite bright," Dr. Hirsch concludes. "The study reported by Silbergleit and colleagues is an important step in this direction. As soon as a practical intramuscular autoinjector for midazolam becomes widely available, the findings in this study should lead to a systematic change in the way patients in status epilepticus are treated en route to the hospital."

The study was funded by the National Institute of Neurological Disorders and Stroke, the National Institutes of Health Office of the Director CounterACT Program, and the Biomedical Advanced Research and Development Authority of the Assistant Secretary for Preparedness and Response. Disclosure forms for Dr. Silbergleit and colleagues and for Dr. Hirsch are found at

N Engl J Med. 2012;366:591-600. Abstract Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: