This will be the final post this week about status epilepticus. I have long argued that our current status algorithms leave too many patients seizing for too long. I updated my suggested alternative algorithm this week and added a longer supplemental post explaining the reasons that I suggest early anesthetic agents (specifically propofol). As I was putting the final touches on those posts, a new RCT was published that seems incredibly relevant. Burman 2019 compared phenobarbital to phenytoin as the second line agent in status epilepticus.
The paper
Burman RJ, Ackermann S, Shapson-Coe A, Ndondo A, Buys H, Wilmshurst JM. A Comparison of Parenteral Phenobarbital vs. Parenteral Phenytoin as Second-Line Management for Pediatric Convulsive Status Epilepticus in a Resource-Limited Setting. Front Neurol. 2019; 10. [article]
The Methods
This is an open-label, single centre, randomized trial based in Cape Town, South Africa.
Patients
Children aged 1 month to 15 years in generalized convulsive status epilepticus (defined as any convulsive seizure lasting more than 5 minutes, or multiple discrete seizures with no extended period of recovery between the events).
Intervention
An algorithm using phenobarbital as the second line agent after 2 doses of a benzodiazepine. Phenobarbital was dosed as 20 mg/kg IV, with repeat doses of 10 mg/kg allowed if the seizure continued after 10 and 20 minutes.
Comparison
An algorithm using phenytoin as the second line agent after 2 doses of a benzodiazepine. Phenytoin was given as 20 mg/kg IV over 30 minutes. If there was no response to phenytoin, the patient was started on a midazolam infusion.
Outcome
The primary outcome was the number of seizures refractory to benzodiazepines that were aborted by the second line agent.
The Results
Over a 3 year period, they included 193 cases of status epilepticus, which after some exclusions and lost data resulted in 144 seizures randomized from 111 patients. The median age in this group was 28 months. About half the seizures were continuous and half intermittent. Half of the patients required a second line agent after the benzodiazepine.
For the primary outcome, 86% of patients had their seizure stopped with one dose of phenobarbital as compared to only 46% with the phenytoin (p=0.0003).
A second dose of phenobarbital stopped the seizures in 4 of the 5 children still seizing (80%), meaning that the phenobarbital algorithm only failed in 1 out of 72 patients (1.4%).
The median time from the administration of the second line agent to seizure cessation was 10 minutes with phenobarbital and 28 minutes with phenytoin (p<0.0001).
PICU admission was 13% with phenobarbital and 28% with phenytoin. (p=0.04)
Respiratory depression occurred 56% of the time with phenobarbital, but 70% of the time with phenytoin (p=0.68).
My thoughts
This is an interesting study. I am somewhat biased, because I have spent the last month or so preparing a post explaining why I think conventional anticonvulsants are failing in status epilepticus and suggesting earlier aggressive use of anesthetics agents. So I do need to say up front that a non-blinded single center trial will always have problems, and we should wait for replications before making any definitive conclusions.
This study was purposefully designed to look at seizure management in a resource limited setting. As a result, it may not extrapolate well to resource rich settings. For example, because of long delays to emergency department presentation, the median time to the first dose of benzodiazepine was 50 minutes. Furthermore, these children have illnesses I essentially never see in Canada or New Zealand. There was a 3-4% prevalence of HIV, and a 2-3% prevalence of previous tuberculosis meningitis (and the authors think that because of lost data these are actually underestimates). However, that was precisely the reason this trial was needed. Other trials, like CONSEPT and ECLIPSE, might not extrapolate well to this setting. (Looking at the numbers, I think there are more similarities in seizure management than there are differences.)
Phenytoin was given somewhat slower than in other algorithms (over 30 minutes instead of 20), which could explain the slightly low response rate (45% as compared to 60% in ConSEPT), but phenytoin never has great response rates. (Dalziel 2019)
One thing that needs to be said about all seizure studies is that we are only looking at surrogate outcomes. Long seizures result in really bad outcomes, so stopping more seizures and stopping seizures earlier sounds like a good idea. However, these outcomes are still surrogates for what we really care about: children walking out of hospital alive and with good neurologic status. It is possible that a drug could stop seizures earlier but still result in worse outcomes.
I think this study provides a strong argument for using anesthetics such as phenobarbital (or perhaps propofol) early in the management of status epilepticus. Despite presenting very late and having significant baseline health problems, the phenobarbital approach was successful in 98.6% of children. Furthermore, although phenytoin failed more than half the time, the subsequent midazolam infusion was 100% successful. In comparison, using phenytoin, levetiracetam, or the combination of both left 30-40% of children seizing after 2 hours in the recent CONSEPT and ECLIPSE trials. (Dalziel 2019; Lyttle 2019)
This open-label, single center trial from a resource limited environment is not practice changing on its own. We need large, blinded RCTs focused on survival with good neurologic status in order to be sure that anesthetic drugs are the best second line agents in status epilepticus. However, as I argued earlier this week, I think the evidence is already strong enough that I am using this algorithm while I wait for more research.
Bottom line
This trial has a number of limitations, but suggests that phenobarbital might be a better second line agent in status epilepticus than phenytoin.
Other FOAMed
My status epilepticus algorithm
My argument for a more aggressive status epilepticus approach
References
Burman RJ, Ackermann S, Shapson-Coe A, Ndondo A, Buys H, Wilmshurst JM. A Comparison of Parenteral Phenobarbital vs. Parenteral Phenytoin as Second-Line Management for Pediatric Convulsive Status Epilepticus in a Resource-Limited Setting. Front Neurol.. 2019; 10. [article]
Dalziel SR, Borland ML, Furyk J, et al. Levetiracetam versus phenytoin for second-line treatment of convulsive status epilepticus in children (ConSEPT): an open-label, multicentre, randomised controlled trial. The Lancet. 2019; http://dx.doi.org/10.1016/S0140-6736(19)30722-6
Lyttle MD, Rainford NEA, Gamble C, et al. Levetiracetam versus phenytoin for second-line treatment of paediatric convulsive status epilepticus (EcLiPSE): a multicentre, open-label, randomised trial. The Lancet. 2019; http://dx.doi.org/10.1016/S0140-6736(19)30724-X
Morgenstern, J. Phenobarbital second line in status epilepticus (Burman 2019), First10EM, June 6, 2019. Available at:
https://doi.org/10.51684/FIRS.8698
5 thoughts on “Phenobarbital second line in status epilepticus (Burman 2019)”
Hi Justin,
Thank you for your post and for reviewing our article.
As you have correctly said, our study was mainly focussed on a resource-limited setting where phenobarbital is widely used although becoming increasingly unavailable due to manufactures stopping production.
The large multi-centre studies (i.e. CONSEPT and ECLIPSE, ESETT) did not include phenobarbital as it is not used in the resource-rich world due to fear of long terms neurodevelopment consequences. The long-term neuro-morbidities caused by phenobarbital are controversial as the evidence is scarce and there is no evidence on whether there is any long-term consequence of using high doses of phenobarbital in the acute management of paediatric status epileptics. This is a follow-up study that is needed!
Best wishes,
Richard
Thanks for the comment! I think you are the first author who has found their paper on my website. Very honoured to have your insights here.
I agree the evidence of long term neuro problems are controversial. Your results are incredibly promising. It was a fantastic trial – congratulations!
I think the next big step forward in status research (aside from more studies like this that look at anesthetic agents as second line) is recognizing that we really need studies to look at longer term neuro outcomes rather than just time to seizure cessation. (Although that clearly makes the trials more complicated and more expensive.)
Look forward to reading more from you in the future.
Great post! Love this especially as a follow up to your other post earlier this week! As a new doctor (I’m only an intern) I would have never really put a paper like this in the context of resource rich or poor setting – I would’ve just assumed that it would be applicable in all healthcare settings (I know the bigger the n, the stronger the power) – What I mean to say is, thank you for making me understand that you also have to consider the setting and resources of where research is conducted and how this cannot always be extrapolated to other countries/healthcare settings. I hope I made sense!