Status Epilepticus Treatment: First-Line to Refractory Care

Status epilepticus treatment follows a staged, time-critical approach built around one central reality: the longer a seizure lasts, the harder it becomes to stop. The formal threshold for convulsive status epilepticus is five minutes of continuous seizure activity, and treatment typically begins with a benzodiazepine, escalating through a second antiseizure drug if needed, and potentially reaching continuous anesthetic infusions for the most resistant cases.¹ The biology underlying that urgency, the specific drugs used at each stage, and what happens when standard protocols fail are all more nuanced than most emergency-medicine summaries suggest.

Why the Clock Starts at Five Minutes

The current consensus defines status epilepticus as a seizure that has either outlasted the body’s normal mechanisms for shutting it down or has triggered processes that keep it going. An international task force set two key time thresholds for convulsive (tonic-clonic) seizures: the first, at five minutes, marks the point after which a seizure should be treated as continuous and unlikely to stop on its own; the second, at thirty minutes, marks the point after which there is a meaningful risk of lasting brain injury.² Those thresholds are grounded in both animal experiments and clinical data.³

The reason for urgency is not just that ongoing seizures are dangerous in general. Something specific changes at the cellular level during prolonged seizure activity. The brain’s primary “brake” system relies on receptors that respond to GABA, a chemical that dampens nerve firing. During sustained seizures, those GABA receptors get pulled inside nerve cells at an accelerated rate, effectively removing the brakes from the surface where they do their job.⁴ This internalization is driven by the seizure activity itself, creating a vicious cycle: the seizure reduces the brain’s ability to respond to the very class of drugs used to treat it.⁵ That is the biological basis for why benzodiazepines, which work through those GABA receptors, become less effective the longer a seizure continues. Every minute of delay narrows the treatment window.

First-Line Treatment With Benzodiazepines

The first drug given for status epilepticus is almost always a benzodiazepine. The question is which one and how it gets into the body. In a hospital, intravenous lorazepam has long been standard. But a landmark trial comparing intramuscular midazolam against intravenous lorazepam in the prehospital setting found that the intramuscular route was not only equivalent but actually produced better results by the time patients arrived at the emergency department. Seizures had stopped without rescue therapy in about 73% of patients who received intramuscular midazolam, versus about 63% of those who got intravenous lorazepam.⁶ The advantage came down to speed: the midazolam injection was administered roughly three and a half minutes faster than the lorazepam infusion, because starting an IV line on a seizing patient takes time. Lorazepam worked slightly faster once it was actually flowing, but the head start from quicker administration more than compensated.

In children, the picture is similar. A prehospital trial in pediatric patients found intramuscular midazolam was administered faster and appeared safe and effective, with seizure cessation rates comparable to intravenous lorazepam.⁷ A systematic review and meta-analysis of midazolam versus lorazepam in pediatric status epilepticus confirmed that seizure termination rates were essentially equivalent between the two drugs, with intranasal midazolam showing the practical advantage of shorter administration time and quicker seizure control upon hospital arrival.⁸

The development of non-intravenous delivery methods has been a quiet but significant advance. Intranasal diazepam and midazolam, along with intramuscular midazolam auto-injectors, have been shown to work as well as rectal or intravenous diazepam.⁹ These newer formulations matter because they allow treatment to begin before a paramedic or emergency physician is available, and because they spare patients the delay and difficulty of IV access during active convulsions.¹⁰

When Benzodiazepines Are Not Enough

Benzodiazepines fail to stop status epilepticus in a substantial minority of cases. The next question, which went unresolved for decades, is which second-line drug to reach for. The three main contenders are fosphenytoin (a form of phenytoin), levetiracetam, and valproate. A large randomized trial, known as ESETT, finally put them head to head. The results were striking in their uniformity: seizure cessation combined with improved consciousness at 60 minutes occurred in about 47% of patients given levetiracetam, 45% given fosphenytoin, and 46% given valproate.¹¹ No drug emerged as clearly superior.

An age-stratified analysis of the same trial explored whether children, adults, or older adults responded differently. No significant differences in efficacy or safety were detected by drug within any age group, though the overall success rates were somewhat lower in older adults across all three drugs.¹² In practice, this means the choice between these three drugs often comes down to the individual patient’s medical history, potential drug interactions, and the clinician’s comfort level rather than any firm advantage of one agent over another. A patient with liver disease might be better served by levetiracetam; someone with a cardiac conduction abnormality might do better avoiding fosphenytoin.

Refractory and Super-Refractory Status Epilepticus

When both a benzodiazepine and a second-line drug fail, the condition is classified as refractory status epilepticus. Treatment at this stage typically means transferring the patient to an intensive care unit and starting a continuous intravenous anesthetic, commonly propofol or midazolam at high-dose infusion, or a barbiturate such as pentobarbital. The goal is to suppress most brain electrical activity, producing a pattern on the EEG called burst suppression, where brief bursts of brain activity alternate with periods of near-silence. One study found that the amount of epileptic activity remaining within those bursts was a useful predictor of whether the anesthetic could eventually be weaned off successfully.¹³

If seizures return after the anesthetic is reduced, or if they persist for 24 hours or more despite adequate anesthetic infusion, the patient has crossed into super-refractory status epilepticus. This is among the most difficult scenarios in neurology, and the evidence base thins considerably. Ketamine, an NMDA receptor antagonist, has drawn interest because it works through a completely different mechanism than benzodiazepines and most standard antiseizure drugs. Instead of boosting the brain’s inhibitory GABA system, ketamine blocks excitatory glutamate receptors, which become increasingly important drivers of seizure activity as the condition progresses.¹⁴ This makes it a logical addition when GABA-targeting drugs have failed, though the evidence so far consists of case series and retrospective studies rather than large trials.

The Ketogenic Diet as a Rescue Strategy

One of the more unexpected tools in the super-refractory arsenal is the ketogenic diet, a high-fat, very-low-carbohydrate feeding regimen originally developed for childhood epilepsy decades ago. In an early study of critically ill adult ICU patients with super-refractory status epilepticus, 90% achieved ketosis, and seizures stopped in all of those patients within a median of three days.¹⁵ A more recent study of twelve patients reported resolution of super-refractory status epilepticus in 75% of cases, again with a median time to response of about three days after diet initiation. All responders were eventually weaned off both the anesthetic drugs and the ventilator.¹⁶

The diet is not without complications. Side effects in the ICU setting include gastrointestinal problems, malnutrition, metabolic abnormalities, electrolyte disturbances, and acute weight loss, though most of these can be managed with careful monitoring. No deaths have been attributed directly to the diet in published series. Still, implementing a ketogenic diet in a critically ill, often ventilated patient is logistically challenging, and the evidence remains limited to small case series.¹⁷

Non-Convulsive Status Epilepticus and Why It Gets Missed

Not all status epilepticus involves the dramatic convulsions people picture. Non-convulsive status epilepticus produces ongoing seizure activity in the brain without obvious physical shaking. A patient might appear confused, drowsy, or unresponsive, and the seizure can only be confirmed with EEG monitoring. This makes it easy to miss, particularly in intensive care patients who may be sedated or have other reasons for altered consciousness. In neurosurgical patients with reduced consciousness, one study found that diagnostic clarity was reached within the first 30 minutes of EEG recording in 96% of cases, suggesting that even a brief EEG is useful when the right clinical suspicion exists.¹⁸ However, the availability of continuous EEG monitoring and the expertise to interpret it remain limiting factors in many hospitals.¹⁹

The treatment principles for non-convulsive status epilepticus overlap with those for convulsive forms, but the urgency and aggressiveness of treatment are debated. Some non-convulsive presentations, such as absence status in a patient with known generalized epilepsy, carry relatively low risk and may respond quickly to a benzodiazepine. Others, particularly non-convulsive status epilepticus in a comatose patient, can signal serious ongoing brain injury and require escalation similar to refractory convulsive status epilepticus. The key challenge is recognizing it in the first place.

When the Immune System Is the Problem

A particularly devastating form of the condition is new-onset refractory status epilepticus, or NORSE, which strikes previously healthy people with no prior seizure history. A subset called FIRES (febrile infection-related epilepsy syndrome) typically follows a minor febrile illness. In many of these cases, no structural brain lesion or clear infection can be found, and standard antiseizure drugs fail. A growing body of evidence points to an immune or inflammatory process driving the seizures, which has prompted the use of immunotherapy.

First-line immune treatments typically include intravenous steroids, intravenous immunoglobulins, and plasma exchange. When those are insufficient, clinicians have turned to second-line agents that target specific inflammatory molecules. Anakinra, which blocks interleukin-1, and tocilizumab, which blocks interleukin-6, have shown the most promise in case reports and small series.²⁰ A review of 75 patients with NORSE treated with second-line immunotherapy found seizure control within the first two weeks in up to about 73% of patients given anakinra and about 70% of those given tocilizumab.²¹ In one detailed case report, a 21-year-old woman with FIRES had complete cessation of seizures within 24 hours of starting anakinra after over a month of refractory status epilepticus.²² The growing consensus is that early use of these immune-targeting drugs may be critical, though randomized trials are still lacking.

What Status Epilepticus Does to the Rest of the Body

The brain is the obvious target, but prolonged seizures inflict damage across multiple organ systems. Early in the episode, the body releases a surge of stress hormones that can affect the heart, lungs, and muscles. Cardiac arrhythmias and pulmonary edema are recognized complications of this hyperadrenergic state.²³ Sustained convulsions can also cause rhabdomyolysis, a breakdown of muscle tissue that releases proteins into the bloodstream and can lead to acute kidney injury.²⁴

At the level of the brain itself, prolonged seizure activity causes measurable neuronal injury. Neuron-specific enolase (NSE), a protein released when neurons are damaged, rises sharply in the blood within 24 to 48 hours after status epilepticus. In one study, NSE levels in patients after status epilepticus were roughly five times higher than in normal controls, and the levels correlated with both the duration of the episode and the patient’s outcome.²⁵ In children with convulsive seizures, elevated NSE was associated with a roughly fourfold increase in the odds of short-term brain injury, with an even stronger association in children whose seizures lasted ten minutes or longer.²⁶

Children With Febrile Status Epilepticus

Febrile status epilepticus, where a fever-related seizure extends beyond thirty minutes, is one of the most frightening experiences a parent can face. The short-term prognosis, fortunately, is more reassuring than it seems in the moment. A prospective study following 180 children who presented with febrile status epilepticus found no deaths and no new cases of cognitive or motor disability attributable to the episode.²⁷ Children who had febrile status epilepticus were more likely to have pre-existing neurological differences, suggesting the episode itself may not always be the primary culprit when problems are found.

Longer-term follow-up, however, paints a more cautious picture. A retrospective study of 57 previously normal children who had a first febrile status epilepticus found that about a quarter developed some neurological consequence, most commonly a speech deficit that was detectable an average of six months later. The strongest predictors of poor outcome were the number of different drugs needed to stop the seizure and how long the seizure lasted.²⁸ This underscores why treatment speed matters even in a condition with a generally favorable outlook: prolonged seizures, even febrile ones, are not guaranteed to be harmless.

Recovery and Long-Term Outcomes

The overall risk of poor long-term outcome after status epilepticus is high in both children and adults. The most important factor driving prognosis is the underlying cause, and separating the damage done by the seizures themselves from the damage caused by whatever triggered them is often difficult.²⁹ A study of patients who survived prolonged refractory status epilepticus found that most had significant functional decline at the time of hospital discharge, but the majority showed at least partial improvement over the following months. Cognitive testing in those patients found scores in the normal range for most survivors by the time of follow-up.³⁰

In patients whose status epilepticus was associated with acute encephalitis, a 12-month follow-up found that about two-thirds remained independent in daily activities, though roughly half developed ongoing epilepsy.³¹ The trajectory after status epilepticus is not a straight line: some patients improve steadily over weeks to months, while others plateau early. The period immediately after hospital discharge often looks worse than the eventual outcome, which is worth remembering for families in the early days of recovery.

Surgery as a Last Resort

When every drug and dietary intervention has failed and the seizures keep going, surgery enters the conversation. This is rare and reserved for super-refractory cases where a localizable brain region is clearly driving the seizures. A systematic review identified 114 patients who underwent acute neurosurgical intervention for super-refractory status epilepticus. Of those, 111 had resolution of the condition after surgery, and half were free of disabling seizures at last follow-up.³² Case reports illustrate what this looks like in practice: in two patients with identifiable structural brain lesions, surgical resection terminated super-refractory status epilepticus with no serious complications, and both remained free of disabling seizures at follow-up.³³

The results are selection-biased by nature: surgery is only performed when the seizure focus can be located and when the patient is a reasonable surgical candidate. But for those who meet the criteria, it offers a way out when nothing else has worked. The decision involves weighing the certain harm of ongoing seizures against the risks of brain surgery in a critically ill patient, a calculation that favors surgery when the alternative is indefinite anesthetic coma.

Older Adults and Treatment Complications

Status epilepticus in older adults carries particular challenges. The condition is more common in this age group, yet many of the standard drugs have higher complication rates in elderly patients. Phenytoin and fosphenytoin carry cardiac risks that are amplified in people who often already have heart disease. Benzodiazepines cause more profound sedation and respiratory depression. One pilot study explored using intravenous levetiracetam as the first-line drug in nine elderly patients with status epilepticus, bypassing the typical benzodiazepine-first approach because of safety concerns. It proved effective in all but one patient, with no adverse events or cardiac changes observed.³⁴ That is a very small study, and it has not reshaped standard guidelines, but it reflects a real clinical tension: the drugs that work fastest in younger patients may not be the safest choice in someone who is 80 years old with congestive heart failure.

Neurosteroids and the Next Generation of Drugs

The receptor internalization problem described earlier, where GABA receptors disappear from the cell surface during prolonged seizures, has spurred interest in drugs that can work through different parts of the GABA system. Neurosteroids are a class of compounds that modulate GABA receptors at a site distinct from where benzodiazepines bind. Ganaxolone, a synthetic neurosteroid, has drawn attention for its ability to control seizures through this mechanism, and it has already gained approval for certain genetic epilepsies.³⁵ Whether neurosteroids will ultimately prove effective for status epilepticus that has become resistant to benzodiazepines is an active area of research. The theoretical appeal is strong: if the receptors have been pulled away from the surface, a drug that works on a different binding site, or on receptors that are still present, could bypass the resistance mechanism entirely. Clinical trials are ongoing, and the field is watching closely.