Episode 70- “Revisited”: Can we give levetiracetam as an IV push?

Hello and welcome to Episode 70 of ER-Rx. This week, I’m doing something I haven’t done before on this show. I’m going to revisit Episode Two of the podcast, which came out in April of 2020. It’s crazy to think that this podcast is almost 2 years old now, and for those of you who have stuck with me this whole time- thank you so much!

Episode Two was entitled “can we give levetiracetam as an IV push?”. And I know that there’s listeners out there who mispronounce it “leve-tira-cetam” (it’s kind of weird)- so I’ll just call it “Keppra” from here on out. Since the release of Episode Two, there’s been more studies published discussing this, along with numerous social media polls, discussions, and journal clubs asking people about their site’s Keppra administration process. My ER has been pushing mostly undiluted Keppra for about a year and a half now, so I also have some personal experiences to share towards the end of the episode.

As a little background, we should by now all be aware of the fact that seizing patients, especially those in status epilepticus, need to be given high-dose benzodiazepines followed by high-dose, second-line anti-epileptic drugs like Keppra, phenytoin, or valproic acid as quickly as humanly possible since time to dose has been shown to actually affect patient outcomes. Given its ease of administration and impressive safety profile, Keppra has become the most often used second-line agent for patients in status epilepticus- not to mention its routine use for other seizure disorders and as seizure prophylaxis not only in the ER, but throughout the entire hospital.

Back in Episode Two, we discussed a study by Burakgazi, et al that showed that giving doses up to 1,500 mg as an IV Push (IVP) was safe and effective. Two other prospective studies showed that doses of up to 3,000 mg, diluted 1:1 with normal saline and given over 5-6 minutes was well-tolerated in both adults and children. The ESET Trial used a 1:1 dilution for doses of up to 4,500 mg and gave that over 10 minutes.

Given these and other studies, a growing number of sites were starting to give Keppra as an IVP, not only in the ER for “STAT” doses, but also throughout the entire hospital for maintenance dosing. A couple of years ago, a poll sent out to the Neurocritical Care Society showed that most sites are either allowing IVP or are thinking about moving in that direction. Most sites had maximum IVP dose of 1,000 mg or 1,500 mg, whereas other sites had no dose maximums. More recently, a Neurocritical Care Journal Club poll on Twitter showed a wide variety of practices. To summarize, a good number of sites are allowing the max status epilepticus dose of 4,500 mg to be given IVP over 3-5 minutes- with some sites getting fancy and giving the dose as 3 separate 1,500 mg doses. But still many other sites only allow 1,000 or 1,500 mg as the max IVP dose.

That being said, we now have more data on the safety of pushing doses greater than 1,000 or 1,500 mg. One such study is the retrospective review by Alkazemi, et al out of Boston published last year in 2021. At this site, nurses were diluting the total volume of all Keppra doses 1:1 with normal saline and giving it over 5 minutes. This study included over 250 IVP administrations in 84 patients. About 20% of doses were given in the ER, about 20% were for new seizures or status epilepticus loading, and 50% of doses were given in a peripheral line (133 administrations).

The authors showed that IVP Keppra in doses up to 4,000 mg was associated with similar incidence of bradycardia (3.2% vs 1.5%, P= 0.34), hypotension (5.2% vs 3.5%, P=0.44), sedation (19.3% vs 27.9%, P=0.12), and peripheral IV site reactions (0% vs 0.6%, P=0.39) as giving it slower in a piggyback. The added benefit of pushing all doses was that the IVP group had a significantly shorter time between order verification and dose administration for all “STAT” orders and loading doses for patients in status (19 mins vs 40 min (P <0.001)). The only downside of this study was that only 6% of the IVP doses were greater than 2,000 mg with a maximum dose of only 4,000 mg given- so we really have no information for giving 4,500 mg as an IVP from this study alone.

A second retrospective review by Haller, et al regarding this issue was also published last year out of Phoenix. At this site, Keppra was given undiluted up to a max dose of 4,500 mg over 5 minutes. This larger study included over 900 patients with over 8,500 doses of undiluted Keppra given. Again, about 20% of doses were given in the ER, about 50% of doses were for new seizures, and about 80% of the doses were given in a peripheral line. A massive 3,674 (43%) of doses were greater than 1,000 mg. Again, only 15 doses were 4,000 mg or greater, and of those there were only 2 doses of 4,500 mg given. And just like in the previous study, adverse events were minimal with only 2 documented, but very questionable, injection sites reactions occurring.

In Episode Two, I recommended IVP for doses up to and including 1,500 mg. For doses greater than 1,500 mg up to 3,000 mg I recommend diluting 1:1 with normal saline and giving it over 5 minutes. For doses greater than 3,000 mg I recommended diluting in 100 mLs of normal saline and giving it over 10 minutes. But since that time and after actually implementing this process, my thoughts have changed. I’ve found that the 1:1 dilution can be confusing and difficult to actually carry out. So, to simplify the process we settled on pushing doses up to and including 1,500 mg over 5 minutes and putting doses greater than 1,500 mg in 50-100 mLs of normal saline and running it on a pump over 5 minutes or just telling the nurses to run it “wide-open” without the use of a pump. Whether giving it as an IVP or as a quick infusion, I’ve yet to hear about any adverse events.

Besides better workflow and shorter time to administration, the practice of pushing all doses is also associated with a cost savings of about $15 per dose, given the reduced use of materials and pharmacy and nursing time. I should mention that this is a very rough estimate from my own site done about 2 years ago, so this number may not be 100% accurate. This may seem like a small amount per dose, but over a large span of time with thousands of doses given, this really adds up, especially if this practice is implemented hospital-wide for every single Kcentra administration.

In conclusion, I do think that given what we know with these newer studies, it is reasonable to give up to 4,500 mg of Keppra undiluted over 3-5 minutes. This will reduce operational delays like preparation in the IV pharmacy, will mitigate the issue of potential drug shortages of premixed Keppra products, remove the confusion with diluting 1:1 at the bedside, and even side-step small volume fluid bag shortages. This will all in turn save your site time and money, reduce time to drug administration, and will hopefully lead to better patient outcomes if we can break the seizure faster. Personally, I’m ok with diluting larger doses in 50-100 mLs of normal saline (of course until a fluid bag shortage comes around), throwing a line on it, and just having the nurses run it over 5-10 minutes or “wide-open” without a pump. The end result is pretty much the same, and nurses don’t have to stand at the bedside for 5 minutes to give a dose.

As always, thank you so much for your time. Please remember to follow @errxpodcast on Instagram and to subscribe to the podcast on Spotify, Apple Podcasts, and even YouTube by searching “ER-Rx.” To make it easy to subscribe, I have direct links in the Show Notes. And while you’re subscribing, please also consider leaving me a review, especially on Apple Podcasts. This way, more of our colleagues can find the show. Finally, I’d like to thank Kellie for supporting the show on BuyMeACoffee.com. Every dollar donated helps keep the show free to access for everyone and also keeps it ad-free.