Status epilepticus (SE) is the most common neurologic emergency, with rates estimated between 1.3 and 81 cases per 100,000 people per year. Mortality estimates also range widely, between 3% and 50% in various studies. Rapid, effective treatment decreases risks of cardiac and respiratory complications, intensive care admission and death.
About two-thirds of patients respond to initial treatment with benzodiazepines, but the others need second-line therapy with an anti-seizure drug.
Clinical guidelines provide little information on efficacy and safety to guide the choice of second-line therapy. As a result, treatment choices for any given case may be driven as much by habit, training and tradition as by evidence.
Levetiracetam, valproate and phenytoin are the most commonly used second-line medications. Phenytoin was used as first-line treatment for SE for several decades, until a 1998 randomized, controlled trial compared four medications and found phenytoin the least effective. Its use then shifted to second-line treatment, at least in the United States, noted Eugen Trinka, chair of the Department of Neurology, Paracelsus Medical University, Salzburg.
"At that time, at least half of Europe, Russia, China and in large parts of Asia, doctors were using valproic acid or phenobarbital" as second-line treatment, he said.
About a decade later, the wheels were set in motion for another trial, this one comparing fosphenytoin, levetiracetam and valproic acid (valproate) to treat benzodiazepine-resistant SE. With US centers favoring phenytoin, and centers elsewhere generally preferring the other two medications, the trial was designed to test whether the three drugs were clinically equivalent.
Trial results: Clinical equivalence
The Established Status Epilepticus Treatment Trial (ESETT) included 384 patients seen at 57 emergency departments in the United States. It excluded patients whose seizures were primarily precipitated by major trauma, hypoglycemia, cardiac arrest or anoxia.
ESETT results-;published in 2019 in the New England Journal of Medicine-;showed that the three drugs were clinically equivalent, each stopping seizures in about 45% of cases.
There were no significant differences among the groups in terms of median time to seizure cessation, seizure recurrence 1 to 12 hours after drug infusion, or rates of life-threatening hypotension, arrhythmia or death.
The trial was stopped early at the recommendation of the data and safety monitoring board, which determined there was only a 1% chance of more data establishing a least or most effective treatment.
The trial results were a surprise in different countries, for different reasons, said Trinka. "In the United States, clinicians were surprised that levetiracetam and valproate were so effective," he said. "And in Europe, clinicians were surprised that phenytoin was still a viable option."
All things considered, valproate and levetiracetam are easier to administer, noted Trinka, and have fewer adverse effects. The pharmacokinetics of phenytoin (or fosphenytoin, a pro-drug of phenytoin that can be administered at a faster rate) require close monitoring, and the mechanism of action increases the risks of drug interactions.
Trinka did not expect fosphenytoin to be as well tolerated as it was, given these complexities.
But when you use phenytoin with clinicians who are experienced with it, you will get good results. When it's given in an environment where people are not familiar with the side effects, you will have worse results. So had this study been done elsewhere, the results may have been different."
Eugen Trinka, chair of the Department of Neurology, Paracelsus Medical University, Salzburg
Meta-analysis: Phenytoin least effective second-line drug
Six months before the ESETT results were published, a meta-analysis including 1,185 SE episodes was published in Neurology. The meta-analysis compared five second-line treatments. Phenobarbital topped the list, with an 80% probability of stopping seizures.
"Phenobarbital is long forgotten, but it's one of the most effective drugs you can give," said Trinka. "This meta-analysis confirmed that."
Next was valproate at 71%, then lacosamide at 66%, levetiracetam at 62% and phenytoin at 53%.
Only three studies included phenobarbital and only two included lacosamide; the authors suggested that future studies better quantify the effectiveness of these drugs as second-line SE treatment. They did conclude, however, that available evidence does not support the superiority of phenytoin in terms of effectiveness or cost-effectiveness.
Pediatric trials: Clinical equivalence
Two recent studies showed clinical equivalence for phenytoin and levetiracetam as second-line treatments for convulsive SE in children. Both were multicenter, open-label, randomized controlled trials:
- The EcLiPSE trial (Lancet 2019) included 286 children in the United Kingdom. The primary outcome was time from randomization to seizure cessation. Seizures were stopped in 70% of those given levetiracetam and 64% of those given phenytoin (not a significant difference); time to seizure cessation did not differ significantly (35 minutes in levetiracetam group; 45 minutes in phenytoin group).
- The ConSEPT trial (Lancet 2019) included 223 children in 13 hospitals in Australia and New Zealand. The doses and infusion rates were the same as in the EcLiPSE trial, but the primary outcome was clinical cessation of seizures at 5 minutes after drug infusion was complete. This outcome was reached in 60% of the phenytoin group and 50% of the levetiracetam group (not a significant difference).
Phenytoin is "dying a slow agonizing death"
Why continue to use phenytoin or fosphenytoin if it is at best clinically equivalent to other options, while also riskier and more costly? Pharmacologists and neurologists at the University of Tennessee Health Science Center recently called for the establishment of levetiracetam as the top choice for second-line SE treatment in the United States.
The authors stated there is insufficient evidence supporting phenytoin's superiority, and cited complexities with the drug's pharmacokinetics, administration and drug interactions, as well as its high cost.
James Wheless, senior author, provided a vivid analogy. "You need a car that gets you from place to place, of course-;but what if that car had no air conditioning or stereo or power windows?" asked Wheless, chief of pediatric neurology at the University of Tennessee Health Science Center. "If you could instead get a car with all of those features at a comparable price, and it also filled the need of providing transportation, wouldn't you prefer that car?"
Currently, the American Epilepsy Society recommends phenytoin, valproic acid or levetiracetam for benzodiazepine-resistant SE, but a 2018 survey – completed before the publication of the meta-analysis, ESETT, ConSEPT or EcLiPSE results-;found that 9 of 10 US pediatric epilepsy centers use phenytoin under these circumstances. (Pediatric Neurology, 2018)
Different nails, one hammer
The heterogeneity of SE-;its patient populations, etiologies, and outcomes-;is cited in many studies as being an obstacle to teasing out which medications are most effective. This heterogeneity is a clinical reality, however, and it makes the area a potentially rich source of knowledge and understanding about how the many forms of SE develop and how to stop them.
These recent publications are contributing to evidence about which medications are good choices for controlling SE as quickly as possible. But what's yet to come is research on whether different medications are better choices for certain causes of SE.
"It's well known that some SE etiologies are less responsive than others," said Trinka. "Some are extremely difficult to treat, such as autoinflammatory causes and some brain trauma. But at the moment, we treat them all the same."
Until more is understood about the mechanisms of SE and its various etiologies, Wheless said that clinicians will likely continue to use the medications they have always used, barring forceful guidelines that say otherwise.
"It's something like that old pair of jeans that maybe have holes in them and don't fit right anymore, but you continue wearing them," Wheless said. "It's true of anyone, not only physicians-;people are comfortable with what they know."
Medication availability also plays a role in treatment choice. After intravenous levetiracetam became available in Japan in late 2015, it quickly became a popular choice as second-line SE treatment. A March 2020 study found that in 2014, 28% of participants needing second-line SE treatment received phenytoin; by 2017, only 10% received phenytoin, with 35% receiving levetiracetam. Fosphenytoin use remained stable over the same time period, at about 30%. The study did not find measurable differences among mortality or length of hospital stay between patients treated with phenytoin and those treated with levetiracetam, however.
Other pieces of the SE puzzle
The true success of a treatment protocol begins with short-term seizure cessation, but rests on long-term patient outcomes. Such outcomes depend on much more than second-line therapy. In a 2017 meta-analysis, authors presented SE treatment delay as a major concern. The review of 17 studies found that between 17% and 64% of patients had received no treatment 30 minutes after seizure onset, with at best only half receiving any treatment before reaching the hospital. (Hospital transportation can take 30 minutes or more in some regions, and treatment was not provided during transport in any study.) Median delays to second-line treatment ranged from 69 minutes to 3 hours, and third-line treatment was sometimes delayed for days.
Also citing unreliable dosing of benzodiazepines, delays in diagnosis and other issues, the meta-analysis authors recommended an explicit, single, continuous set of procedures that bridges pre-hospital and in-hospital treatment and involves family members and caregivers, emergency responders, and hospital-based physicians.
They write, "More effective treatment of patients with status epilepticus may be achieved when care delivery is optimized through rigorously examining current practice, collaborating across disciplines, and creating pragmatic treatment protocols."
A shift in perspective may help move research forward, suggested Edward Bertram, epileptologist and professor of neurology at the University of Virginia.
"In all of these carefully performed studies, the results are surprisingly similar, with the seizures stopping approximately half of the time with each of the drugs," he said. "Perhaps we should no longer consider these treatments equally effective, but rather equally ineffective; we don't know, and will never know, what proportion of the seizures would have stopped without the addition of second-line treatment."
The medications used as second-line therapy for SE are some of the same medications that people with epilepsy take every day to prevent seizures, he noted.
"Many studies have shown that the pathophysiology of SE is quite different than the pathophysiology of epilepsy," Bertram said. "It may be time to focus on identifying treatments specifically for SE, and not on repurposing epilepsy drugs."