It was really a fun event and the vibe was very relaxed. There was a lot of “soft stuff” with patients, families and advocates talking to the clinicians. However, there was also a good measure of hardcore science. I had the good fortune to sit in the front row and watch the ConSEPT trial be released… whilst sitting alongside Dr Damian Roland (my favourite Paeds researcher and fellow nerd!) Then in the tea break I also met Dr Simon Craig – one of the investigators and had a chance to see how the data looked from their perspective.
If you want to know about the current practice in CSE (convulsive status epilepticus) then you can read this paper by Drs Lawton, Goldstein, Tagg and Davis… I just know that I have seen their names somewhere before!
So what was this trial? Let us break it down in the usual way…
ConSEPT (Convulsive Status Epilepticus: Paediatric Trial) is an open, RCT comparing levetiracetam (LEV) to Phenytoin (PHE) in children with ongoing convulsive epilepsy after 2 doses of an IV benzodiazepine. The basic protocol is here:
POPULATION : the trial’s final cohorts included 233 children [114 in the PHE arm & 119 in the LEV side.] The duration of the seizures in this trial was quite long – many continuing for an hour.. so sick and difficult seizures to control. Children were excluded if they were already receiving either study drug.
INTERVENTION: This was an open-label trial. This was impossible to blind as the study drugs were delivered over different time frames. The pharmacology would be too fiddly to keep this part blinded.
The basic difference between the arms was which agent was used after the 2nd dose of benzos (mostly midazolam). Participants either got 40mg/kg of LEVETIRACETAM over 5 minutes OR 20mg/kg of PHENYTOIN over 20 minutes.
If seizures continued they then received the other agent at the same doses – i.e. they crossed over. If seizures were still ongoing, then they were treated “as per the clinicians” which may have required a rapid sequence intubation.
A NOTE on CONSENT: This was also a really interesting trial as it employed “deferred consent”. This means that children were randomised and treated prior to consent being obtained. They were subsequently counselled and consented or declined to be involved in the research. The process was really quite acceptable to the families and there was a very high rate of successful consents being obtained. This is great to see as it is very tough to get consent in these hyperacute clinical scenarios.
CONTROL: this is a head-to-head study design and I guess you would say that the PHE arm is the current “gold standard.
OUTCOMES: although the actual drugs were not blinded. The outcome was. A non-clinical researcher used video of the seizures to determine when they had actually ceased without knowing to which arm the child had been allocated.
The primary outcome was “clinical cessation of seizure activity five minutes following the completion of the infusion of the study medication”
So what did they find? Well by the numbers this is a negative trial. Or you could say that neither drug was superior.
The astute reader will notice that this means that kids in the LEV group were assessed at 10 minutes (5-minute infusion + 5 minutes wait) whereas the PHE group were assessed at 25 minutes (20-minute infusion + 5 minutes)… so this is a bit of an unfair advantage to phenytoin as there is a “natural cessation rate” over time.
THOUGHTS: This is a small trial of a rare condition. There are also 2 other similar trials attempting to ask the same question going on currently – one in the UK, one in the USA. So in a while we will have a good data pool to make more firm conclusions.
Does this change practice – well, yes maybe. Although this trial is too small to detect a difference in adverse events we all know about the problems with phenytoin. Chatting with a few friends who run simulations – there is a high rate of errors in giving phenytoin in a stressful scenario like CSE. In my career I have witnessed 2 disastrous outcomes as a result of medication errors with IV phenytoin. Levetiracetam has a much better safety profile and ease of delivery. So for me I think if it is:
can be given faster (even the fear of “getting it wrong” adds minutes to the delivery of phenytoin)
less side effects / adverse outcomes
… then I will use it as the next agent after benzos have failed. Of course, I will keep an open mind and look forward to seeing more data from our northern hemisphere colleagues in time.
Thoughts? Will try and add the actual article once it is published.
I am a GP working in Broome, NW of Western Australia. I work as a hospital DMO (District Med Officer) doing Emergency, Anaesthestics, some Obstetrics and a lot of miscellaneous primary care. Also on the web as @broomedocs | + Casey Parker | Contact