Or, more accurately, is it reasonable to perform an intramuscular injection of midazolam rather than an intravenous injection of lorazepam for seizure-like activity in the prehospital setting?
In fact, some folks are taking this article and claiming that intramuscular midazolam is superior to intravenous lorazepam, that it's a "game changer."
Well, let's not go crazy here.
As with any piece of literature, the more vocal the giddiness I see perpetuated about the internet, the more cautious I am with rushing to judgement. It is, of course, a very well-designed, prospective, double-dummy, randomized, non-inferiority comparison between midazolam and lorazepam. The aim of the study is, essentially, to show that, even though midazolam is not typically as rapidly effective at terminating seizures, the time difference is made up by intramuscular route versus the time required for an IV start.
What's kind of odd that I see in this article is that nearly a third of the lorazepam group did not receive the benzodiazepine portion of the intervention - and they compare it to the midazolam group in which all but 5 patients received the intervention. When their primary outcome is the number of folks who arrived seizure-free in the Emergency Department - it seems as though the 7% absolute difference between the two groups could be easily explained by the fact that a third of the lorazepam group didn't receive an intervention. Most of the lorazepam group had the intervention withheld because they stopped seizing of their own accord at the time of enrollment, with a minority having the intervention withheld because IV access could not be obtained.
And, the differences favoring midazolam are hard to pin down whether it's actually medication superiority, or something different about the seizures. 42 patients in the lorazepam group failed to stop seizing after additional therapy, compared with only 22 in the midazolam group - is this a difference in efficacy, or a difference in the underlying disease process - which appears to be more resistant to any therapy, including rescue, in the lorazepam group?
But, in any event, this just nitpicking against the superiority argument, and not the non-inferiority argument. From a clinical standpoint, it is clearly safe and effective to use intramuscular midazolam for seizures in the prehospital setting. However, what I'd prefer to see is a similarly powered trial of intranasal midazolam, which takes all the injection risks for patient and provider out of the equation during the seizure. This is a good first step, but I think we can make effective treatment even safer if intranasal can be shown non-inferior as well.
"Intramuscular versus Intravenous Therapy for Prehospital Status Epilepticus"