Neonatal seizures: Guidelines and consensus-based recommendations: Dr. Ronit Pressler

Seizures are common in neonates, but there is substantial management variability. The Neonatal Task Force of the International League Against Epilepsy (ILAE) developed evidence-based recommendations about anti-seizure medication management in neonates in accordance with ILAE standards.

Dr. Emma Carter speaks with first author Dr. Ronit Pressler about the guidelines and recommendations, and how they were established. 

Treatment of seizures in the neonate: Guidelines and consensus-based recommendations—Special report from the ILAE Task Force on Neonatal Seizures

Sharp Waves episodes are meant for informational purposes only, and not as clinical or medical advice.

Let us know how we're doing: podcast@ilae.org.

The International League Against Epilepsy is the world's preeminent association of health professionals and scientists, working toward a world where no person's life is limited by epilepsy. Visit us on Facebook, X (Twitter), Instagram, and LinkedIn.

Transcript

Dr. Emma Carter: Well, thank you so much for being here today, Dr. Pressler. We're here to talk about the recent publication in Epilepsia, the special report from the International League Against Epilepsy Task Force on Neonatal Seizures. The title of the article is Treatment of seizures in the neonate: Guidelines and consensus-based recommendations. I was going to open up the floor and let you introduce yourself.

Dr. Ronit Pressler: Thank you very much for asking me to talk about this paper. I'm Ronit Pressler. I'm a consultant at Great Ormond Street Hospital in London and have worked on neonatal seizures for nearly 30 years. And I was, together with Hans Hartman, co-chair of the Neonatal Task Force, working on the guidelines. 

The Neonatal Task Force consisted of a huge group, 26, 27 people, really representing all different areas of the International League Against Epilepsy. We had neonatologists, clinical neurophysiologists, neurologists, pediatric neurologists, methodologists Nathalie Jette and Francesco Brigo. We had a patient representative, so it was really a truly multidisciplinary team.

Dr. Emma Carter: Before we kind of jump into some of the questions we were going to talk about today, I was going to ask you a little bit about the background of the paper and why the group felt neonatal seizures required their own set of recommendations.

Dr. Ronit Pressler: So the Neonatal Seizure Task Force itself was created around 10 years ago and actually our first project was the classification of neonatal seizures, but parallel to that, there was a working group, which was from the WHO, the International League, and the International Bureau of Epilepsy, and they had published recommendations for the treatment of neonatal seizures in 2011.

And around 2017, they had decided that an update is needed. And so then they contacted the Neonatal Task Force, which had been created since. And so the new group was created basically out of that. So it's a combination of the old working group and the neonatal task force, which still exists.

So the reason why neonates need their own recommendations is really because neonates, immature brains, respond differently to antiseizure medications. So there is age-dependent neurotransmission with different receptor expression, which on one hand makes seizures look different. So seizure semiology is slightly different to that of older children and in adults, but also obviously they respond to anti-seizure medication differently. And therefore it was felt that neonates would benefit from their own recommendations. 

The other problem is that for older children and adults, or particularly for adults, there are like over 30 anti-seizure medications which have been developed and tested, many of them are licensed. But for neonates, that's not the case. There are very, very few randomized, controlled studies which look at neonatal seizures, and that's a huge problem because retrospective studies or prospective open studies, we know that they are not very reliable. Many of these don't use EEG to diagnose seizures, which is in itself a problem in neonates. So there are actually only a very few randomized controlled studies to base these guidelines on. 

Actually, in the last few years, there have been a number of publications, really important publications. And so it was felt that this would be important to take into account. Plus, in neonates, there is now cooling therapeutic hypothermia, which is now standard of care in high-income countries. And so it was felt it was important to take this into account for the guidelines. 

Dr. Emma Carter: Can you walk us through a little bit about each of the six recommendations and the paper, and maybe some of the highlights of these recommendations?

Dr. Ronit Pressler: Sure, of course. So, the first two questions are straightforward: What should you use first line and what should you use second line. I think one of the key issues of what is different in these recommendations to the 2011 guidelines is really that for the first time we have good evidence for the use of phenobarbitone (phenobarbital). Phenobarbitone obviously is a really old anti-seizure medication which has been around for over 100 years. But there is now good evidence from randomized controlled trials that in neonates, it's the most effective drug, as far as we know, to date. Particularly with a study, which was published in 2020, this study was a randomized controlled study comparing levetiracetam with phenobarbital. And what they found is that neonates respond to phenobarbitone, becoming seizure free 80 percent with phenobarbitone versus only 28 percent with levetiracetam. And so this study was done with EEG, a really very well conducted study, high standard. And so with this study, we can now recommend with good evidence that phenobarbitone should be used as a first line drug.

Much more difficult, the question about second line. So, for second line, we looked at several studies. The same study, the Sharpe study that I mentioned earlier; the Painter study, which compared phenytoin with phenobarbital; and the study by Geraldine Boylan, which compared different second-line medications.

The problem is all of these studies were on second line. The numbers were very small with very poor evidence, so we couldn't use, there was no evidence to support the use of any of these as second-line medication and then we did a Delphi and we could not agree. We needed 66 percent of people agreeing to make a recommendation using Delphi. We could not agree. We had first no agreement then we had lots of discussions. We reformulated the question, and again, no agreement. And then again, we tried to make the questions easier. We discussed how we could agree and after this third round we gave up and said this is just the way it is. This is what it is. There's no agreement, there's no evidence from the papers and it's no surprise that we can't agree either because we don't know. So basically second line, we recommend to use levetiracetam or phenytoin, and other options are lidocaine and midazolam, but we cannot say which one is better.

That was a bit disappointing. On the other hand, I think it's just, it really illustrates how little evidence we have and that we need more studies looking at second line treatment. 

The third question is a new question. When to stop? Actually, it was also in the 2011 paper. But it's a question which has had really lots of discussions more recently. So the question of when to stop antiseizure medication. Obviously, most of the seizures in neonates are acute symptomatic seizures and not part of an epilepsy syndrome. So the question is, when should you stop? 

Like 10, 20 years ago, babies were treated with whatever they were started on for 6 months, 12 months, or even a couple of years. And nowadays, we think this is not needed and there's no need to treat for such a long time. we looked at the evidence; there were no randomized controlled studies. We included three papers: One prospective and two retrospective studies. But obviously, because they were not randomized controlled, we couldn't do a grading. Between the systematic review and the Delphi, there was one paper published in 2021, which was a prospective study, which shows that continuing medication does not influence the risk of developing epilepsy or the neurodevelopmental outcome. Even though this study was published outside our timeframe, I think it influenced the Delphi. 

And so we did a Delphi and there was a high level of agreement. Everybody agreed that anti-seizure medication should be stopped in neonates if there was no evidence for neonatal onset epilepsy and the seizures were treated successfully within the neonatal period. So, if a baby responds to treatment and has no evidence for epilepsy, then anti-seizure medication should be stopped before the baby's discharged from the hospital. And the big difference from the 2011 guidelines is really that this is regardless of whether there are MRI abnormalities or EEG abnormalities. Because obviously these babies have brain injury. All of them have EEG abnormalities, all of them have MRI changes. But still continuing medication doesn't help and may actually harm them. 

The fourth one was about therapeutic hypothermia, and this was relatively straightforward. There were a number of studies, no randomized controlled studies, looking at the effect on seizures, but all of them showing that therapeutic hypothermia reduces seizure burden. It doesn't stop them completely, but it reduces seizure burden. Although the only caveat is really that the studies were not performed to look at their anti-seizure properties. But that's okay. 

Number five was should we treat electrographic seizures or should we only treat clinical seizures? There were three studies included in this. The problem is that within the timeframe we had two randomized controlled studies, six prospective studies, and three retrospective studies. The two randomized controlled studies were really badly underpowered, and then there was a third study, which was published outside the timeframe, also underpowered and addressing a different issue, or their primary outcome measure wasn't useful to answer the question.

So we were relying on a Delphi and here we agreed with 74%. So it was a good agreement that electrographic seizures should be treated. And I think that's really justifiable because in most of the cases, it's the treatment we give those babies which make them electrographic only. So the babies are on sedation, they're ventilated, they're on muscle relaxation. 

The last question, number six, the use of pyridoxine or pyridoxal 5’-phosphate (PLP) was relatively easy because there's so little evidence. So we had no randomized controlled studies, no prospective studies. But there were eight retrospective studies, most of them without EEG, so we did a Delphi asking which babies should be treated with pyridoxine or PLP, and we had really high agreement. So the recommendation is that a baby who had clinical or EEG characteristics suggestive of pyridoxine-dependent epilepsy should be treated with pyridoxine as add on, but also any baby who's not responding to second-line medication without a clear etiology should also have a trial of pyridoxine and then if that doesn't work, PLP.

Dr. Emma Carter: Thank you for walking us through the questions there. You may have already answered this next question talking about which question your group or maybe you thought would have the biggest impact on practicing physicians.

Dr. Ronit Pressler: I think the biggest impact is that we now can recommend phenobarbitone as first line anti-seizure medication with better evidence or there is now evidence, whereas in the previous recommendations, it was always said we recommend phenobarbital, but there's very little evidence. Now we can say there is, it's not high, but it's good enough evidence for the use of phenobarbitone as first line medication. And the second one, which I think is really important, is when to stop antiseizure medication. So this is something which we get asked all the time. When do you stop? And I can also see this, I have to admit this in my own hospital, sometimes babies get discharged even now with anti-seizure medication on board, even though they have clearly acute symptomatic seizures and they responded well to the first- or second-line drug, and they still get discharged with medication. And I think this is something we need to change.

Dr. Emma Carter: Can you also address how these recommendations may be applied in more low-income or middle-income countries that may not have some other resources that other countries have in regards to following these recommendations? 

Dr. Ronit Pressler: Yeah, I think it's a really important question. In principle, these guidelines are applicable to any setting, it shouldn't matter.

One point people always ask in the guidelines, we clearly say that we only include studies that use EEG for the diagnosis of seizures. And I think, obviously, EEG is the gold standard, and ideally, we all want to diagnose seizures with EEG or AEG. But of course, we know that in the clinical setting, that is not always the case.

Yes, in, in some countries, most babies get diagnosed with EEG, but even in the UK, this is not the case. Most babies get diagnosed with AEG or clinical only. And I think we need to differentiate between evaluating a new anti-seizure medication from clinical practice. If you want to test a new drug and want to say whether it's effective or not, then we need to use EEG as an outcome measure that is reliable. Whereas in clinical practice, if you don't have EEG, you have to use what you can. We still need to make sure that we don't treat events that are not seizures. But It doesn't mean that these guidelines are only applicable if you have EEG or AAG.

The only problem obviously is if those drugs are not available. So I know in some countries you can't get phenobarbital IV, you can only get it orally. Obviously then you have to adapt. 

Dr. Emma Carter: Thank you. I know at the very end of the article, there were a few additional considerations that were included in the paper focused on a standardized pathway for the management of neonatal seizures and hoping that that could be available in those neonatal units, as well as ensuring that we as providers are providing adequate, complete communication with guardians and parents on their neonate’s condition. We were wanting to see if you could share with us how the group collaborated and chose to include those considerations as well. 

Dr. Ronit Pressler: Okay again, very important question. So one thing was that when we started off with these guidelines, we thought it was really important to have something which is usable. I mean, user friendly. If you look at systematic reviews, these are so difficult to read and it's all about the methodology. There's nobody saying use this, then this, then this. It's all about “The evidence shows this.” And they are so complicated. Even I find them complicated, even though I was involved in them. And so we said, yes, we also want to provide treatment pathways. 

And obviously this is just a suggested treatment pathway. We are not saying use this one, but this is an example of what you can use. And so we had a couple of Delphi questions. We didn't do a systematic review on those, so they are not part of the guidelines themselves, but we did a Delphi and we said, okay, every hospital who treats neonates should have a pathway and we gave this as an example how this treatment pathway could look. And then we also provided the most important anti-seizure medication we would use with a dosage, with the side effects or adverse effects, just to make the whole guideline also a more useful guideline. 

And then the question about parents. So we had as part of our group a parent representative, Silke Mader, and for her, this was really important that this guideline also includes how to talk to parents or that we need to talk to parents. This is, of course, so important, and often we forget to put this in the guidelines because everybody says, “Oh, we do this anyway,” but it should be part of the guidelines. So we did a Delphi on that, and everybody agreed. 

Dr. Emma Carter: I know you mentioned the Delphi methodology. And just wanting to ask a little bit about what that process looked like for you and the group: how you had regular meetings or surveys, delegated the work within the group to complete this article. 

Dr. Ronit Pressler: So what was really most of the work for the group was the systematic review. You formulate the priority questions: that in itself is quite difficult. We had lots of discussions also with the methodologists, how to rephrase them. Is it possible to answer these questions? How do we phrase them? And we were really lucky that we had two fantastic methodologists as part of the group. Both of them are adult epileptologists, but they were fantastic. Without their help, we would have never been able to do this. 

We formulated the priority questions and then we searched the databases for the articles, and we found, like, over 2,000 papers. The abstract review was nearly 3,000 papers. And then we did the full article review and then we ended up with 218 papers, and of those papers, we did the data extraction. That was a lot of work. So everybody in the group was allocated a certain number of papers. They had to do the data extractions. Always two people independent of each other had to extract the data for each paper.

And then we did the grading. The way we worked was that most of the work we actually did via virtual meetings. We met once a year, either at the American Epilepsy Society meeting or at the International or European epilepsy meetings. But most of the meetings, as I said, were via Zoom. So the grading, the evaluation of the evidence, we had an extra meeting just with a few people, and we sat down for a weekend, face to face, and sort of without stopping, did the whole data extraction, the whole grading, and the whole analysis and that was one of the best things I think we did. Because often that drags on and on and on and we did it in two days. That was very useful.

And then the Delphi, that was more fun. I think most people enjoy the Delphi. We again had a different group to formulate the Delphi questions, a group of seven eight people and we met via Zoom. And then we had three rounds of Delphi questions or questionnaires going around. The ILAE helped there, which was fantastic. They sent them out, they sent the reminders, and then sent us the spreadsheets with all the responses in there. And then that was much easier to analyze. Even though we had three rounds, that didn't take that long. 

And between the three rounds, we had some discussions all done by Zoom. And then at the very end, we had another face-to-face meeting in a smaller group, for writing up. And we did most of the writing in this small group, which I think was a good step.

Dr. Emma Carter: Thank you for just going through the whole article, talking through the considerations there and each recommendation. In regard to the task force, what's maybe next on the list of things to do, or if you guys are already tackling something new at this point in time.

Dr. Ronit Pressler: So now the task force is led by Nick Abend and Magda Nunes. And the new topic they are looking at is status epilepticus in neonates. So neonatal status epilepticus. We're doing a systematic review. And then parallel, Hans Hartman and I are also looking at two priority questions, which were originally part of this.

When we started off, we had eight priority questions, and then we dropped two because we said they are not as high priority and they are two diagnostic priority questions. And we thought the most important ones were the therapeutic ones, so we wanted to get those out and published before we then tackle the diagnostic priority questions. And so we are looking at those at the moment as well. So they're basically two streams. One is the status epilepticus case and the other one is, are the two diagnostic questions, should we use EEG, AEG, or clinical diagnosis, or what are the differences in diagnosing neonatal seizures?

Dr. Emma Carter: That's very exciting. Looking forward to both of those. Just talking a little bit about you, Dr. Pressler, I know you mentioned you've been involved in neonatal neurology for some time now and just wanting to get your advice or lessons learned for our listeners, especially some of our younger colleagues who are listening and just starting out in the field.

Dr. Ronit Pressler: I mean, I've been doing neonatal seizures for so long, and when I started, this was really a much-neglected field. There was hardly anybody interested. Most people said, “Oh, well, we can't diagnose them. We can't treat them. Why are you interested in them?” But now people are more and more interested. 

People acknowledge how important it is that we look for, that we monitor the brain and look for neonatal seizures, treat seizures early. I mean, there is evidence that treatment of neonatal seizures is time sensitive. So the earlier we treat, the better they respond. And this is something we haven't looked at in these guidelines. And I think this is something which we need to look at the next time, the next round.