Midazolam IM or Ativan IV?

Our regional guidelines say in status epilepticus, we can give Ativan IV or Midazolam IM when there is no IV access.

I have used both, and my anecdotal experience has been the Ativan resolves the seizure most of the time, and Midazalom only some of the time.

I have often given Midazolam to a seizing patient with poor access, and then once that dose was on board, looked for and found an IV, administered Ativan IV, and stopped the seizure.

So based on that anecdotal experience, if I thought I could get an IV, I went right for the IV, and once I got it, I pushed Ativan. I only tried Midazolam if I thought I was not going to get an IV or if I failed repeatedly to get an IV.

Now we have some evidence. A new study has just come out in The New England Journal of Medicine.

Citing both the need to get an IV in difficult patients to effectively give Ativan and Ativan’s limited shelf life when not kept refrigerated, the researchers asked the simple question, “Is IM Midazolam noninferior to IV Ativan in terminating seizures in the prehospital environment?”

To do this, they conducted a very interesting study. Medics were given a sealed kit containing either 10 mg Midazolam in an injectable syringe (think Epi-Pen) and an IV placebo or an injectable placebo and 4 mg Ativan for IV injection. When they opened the kit, a voice recorder was activated. They announced when they had given the IM dose, when they had obtained IV access, when they had pushed the IV dose and when the seizure stopped.

Included in the study were all patients over 13 kg, who were still seizing on EMS arrival and who received the study medication. Exclusions included patients with major trauma, bradycardia < 40, hypoglycemia, and pregnancy.

The study utilized over 4000 paramedics from 33 EMS agencies, and 79 hospitals.

Here’s what they found:

IM Midazolam stopped 329 of 448 (73.4%) seizures; IV Ativan stopped 282 of 445 (63.4%).

It took 1.2 minutes to administer the Midazolam and 4.8 minutes to administer the Ativan.

The IM Midazolam took 3.3 minutes to stop the seizure; the IV Ativan took 1.6 minutes.

For total time(includes getting access), it took Midazolam 4.5 minutes and IV Ativan 6.8 minutes to stop the seizure.

Between the two drugs, there was no difference in the need for intubation, no difference in repeat seizures or in hypotension.

Patients who received IM Midazolam were less likely to be admitted or go to the ICU.

The authors concluded that “the intramuscular administration of Midazolam by EMS is a practical, safe, and effective alternative to the intravenous route for treating prolonged convulsive seizures in the prehospital setting.”

In an accompanying editorial, Laurence Hirsch, M.D. wrote: “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.”

Shortly after this study came out, I responded to a call of a man down. On a third walkup apartment, I found a man face down in a pool of blood seizing. I do not know if the patient was beaten or if he seized and fell cutting himself on furniture before hitting the ground (I did find out later he had an intrcerebral hemorrhage). I was by myself, but I had my controlled substances in my pocket. The patient was fully clothed and the light in the apartment was dim. My thought was to try to get an IV and give him some Ativan. But getting an IV in that apartment was troublesome. My house bag was on the bed by the door, I needed to get his shirt and sweater off and lay out my IV kit. But the blood pool was quite large. Instead, I just opted to draw up some Midazolam and give him an IM injection. 5 mg stopped the seizure – at least temporarily. Enough to enable us to board and carry the patient out to the ambulance, where he started seizing again. Another 5 of Midazolam and he again stopped. On the way, we got IV access, and when the patient started seizing again, we gave Ativan. The patient did not seize again in our presence.

Had I not read the study I might have concluded Ativan was the drug that finally ended the patient's ceasing based on anecdote, but having read it, my anecdotal case also supports the ability of Midazolam to stop at least temporarily a fairly significant seizure. I was satisfied with the way Midazolam worked on this call, and now that there is solid evidence of Midazolam's effectiveness in relation to Ativan, I will definitely alter my practice to consider going right to the Midazolam IM to stop any seizure in the future. My take away from reading I have done on this is that the sooner you can gain control of a seizure, the better. With IM Midazolam you don’t have the risk of missing the IV, and further delaying treatment.

I wonder in how many of my anecdotal cases where I gave IM Midazolam in the past, the seizure would have stopped on its own anyway if I held off the Ativan IV, or in how many cases a larger dose of Midazolam would have done the trick.

The doses used in the study are worth noting: 10 mg of Midazolam and 4 mg of Ativan. Our current protocols call for 5 mg of Midazolam or 2 mg of Ativan, repeated in 5 minutes.

Maybe our doses need to be raised when we next review our guidelines.


  • Laura S. says:

    The RAMPART study! My agency was a part of that. Nice to see that the efforts paid off!

    • medicscribe says:

      I think that’s great you were in the study. I would love to participate in one, particuarly one as important as this one.

  • Christopher says:

    The other big takeaway from the study was that patients treated under the protocol were less likely to require intubation, even though they used double the common IM and IV benzo doses!

    This is contrary to what most paramedics are taught, but makes sense: if you don’t stop the seizures, they’re more likely to need intubation.

  • ANDREW says:

    My experience with versed/midazolam has proven to be effective over and over. The dose I use is 0.1 mg/kg IV or 0.2 mg/kg IM or IN (to a max of 5mg iv or 10mg all other routes). The intranasal route ir my preferred method because it eliminates having to get close to the seizing patient with a sharp therefore reducing the risk of needle stick injuries. With that being said, I have not had the opportunity to try ativan, but I’d argue that the time saved through the administration of midazolam via IN or IM trumps all.

  • Nathan says:

    Its very easy to give Versed IM on IN in a seizing pt. I enrolled several patients in this trial. I had no idea which drug I was using. That being said our guidelines already allowed for IM/IN Versed if we wanted to use it. I give it through the pants and dont pull back on the plunger before injecting. I can give a dose of Versed within less than a couple seconds and back off to avoid injury to myself or the pt.

    • medicscribe says:

      Thanks for the comments. Nice to hear from those who particpated in the study. Thanks for your good work!

  • Mary says:

    The RAMPART research trial that you are referring to above was published in the New England Journal of Medicine in February. The group did a lot of pre-trial work when it came to medication dosages and found that for an adult 10mg of Versed is preferred over smaller doses. Also IN has its drawbacks due to patients having an already altered breathing patterns and of course those with deviated septum’s, etc.

  • Peter says:

    Great comments. Thanks. We have intranasal Versed in our guidelines, but our pharmacy doen’t yet provide us with the proper concentration to administer it.

Leave a Reply

Your email address will not be published. Required fields are marked *