How to Treat Refractory Seizures in Children

Welcome to PICU Doc On Call, A Podcast Dedicated to Current and Aspiring Intensivists.

I'm Pradip Kamat and I'm Rahul Damania, and we are coming to you from Children's Healthcare of Atlanta - Emory University School of Medicine.

Welcome to our Episode a 24-month-old girl with increased seizure frequency.

Here's the case:

A 24-month old girl presents to the ED with h/o shaking/jerking episodes in her sleep. The patient was in the care of her aunt when this acute episode occurred. When the father arrived from work, he saw his daughter having episodes of her body shaking alternating with heavy breathing. The patient would not wake up in between episodes. There was pertinently no history of trauma. 911 was called and when EMS arrived, she was starting to arouse and respond to stimuli. The patient was transported to the ED. In the ambulance, the patient continued to have similar shaking and jerking episodes and was given rectal diazepam. On arrival to ED, the patient had a fever of 38.5 Centigrade. Due to ongoing seizures, the patient was loaded with Fosphenytoin, after having been given a total of two doses of IV Lorazepam. The patient was subsequently intubated for airway protection and respiratory failure. A respiratory viral panel was negative for SARS-COV-2 but positive for Rhino-enterovirus. The patient was admitted to the PICU with cEEG monitoring and placed on mechanical ventilation with fentanyl + dexmedetomidine infusions with as needed Midazolam administrations

Her physical examination on arrival to the PICU was unremarkable. She wasn't interactive as she had just received sedation after intubation. On her neuro-examination, Pupils are equal and punctiform. The face is symmetric. The tongue is midline. Normal bulk and tone. No spontaneous movements were noted. No withdrawal to painful stimuli. Tendon reflexes were equal throughout. No clonus is noted.

Rahul, to summarize key elements from this case, this patient has:

• Fever
• Viral infection with Rhinoentero virus
• Generalized Tonic clonic seizure lasting > 5minutes
• Acute respiratory failure
• All of which brings up a concern for status epilepticus

Absolutely, we will get to this later on in the episode; however, remember that Status epilepticus is historically defined as single epileptic seizure of >30 minutes duration or a series of epileptic seizures during which function is not regained between ictal events in a 30-minute period

• Let's transition into some history and physical exam components of this case?

1. What are key history features in this child who presents with status epilepticus?

• Prolonged Seizures
• Fever with viral symptomatology which may act as a trigger
• A pertinent negative is that this patient had no history of trauma or co-morbid conditions such as a genetic syndrome.
• The patient also had no presumed ingestions as well.

1. Are there some red-flag symptoms or physical exam components which you could highlight?

• Important to look for rash (darkening of the skin = adrenoleukodystrophy), genetic facies, evidence of trauma —-all of which are absent in this girl
• To continue with our case, the patients labs were consistent with:
• Initial Labs: WBC 27K, with neutrophilic predominance, Hgb and platelets were normal. Initial CMP was normal except for a glucose of 233. Gas prior to intubation in the ED was 6.9/102/85/-9. (repeat after intubation 7.19/49/40/-9). Ionized ca 4.9mg/dl. A urine analysis was unremarkable.
• Head CT negative

OK to summarize, we have: 24-month-old girl who presented with prolonged seizures and acute respiratory failure

• All of which brings up the concern for status epilepticus the topic of our discussion today.
• Let's start with a short multiple-choice question:

A 14-year-old girl is brought to the PICU from the floor with new-onset status epilepticus. She was admitted to the floor on her second day after a posterior spinal fusion surgery and is still receiving intravenous fluids. Her seizure is described as generalized tonic-clonic. After initial stabilization and maintenance of her airway and hemodynamics, which of the following is most likely to reveal the cause of her seizures?

• A) Serum electrolytes
• B) Stat MRI brain
• C) Lumbar puncture
• D) cEEG

Rahul, the correct answer here is A) serum electrolytes. Patients especially after posterior spinal fusion surgery are at risk for hyponatremia secondary to SIADH or even hypotonic fluids used for maintenance. Correction of hyponatremia in a child with seizures requires 3% hypertonic saline. The seizure threshold is typically a serum Na of 125meQ/L. Serum electrolytes will also reveal the serum glucose which is especially important to check in infants who have seizures. A stat MRI is not warranted in this patient especially if she is alert and awake prior to the seizure. Additionally, it would be dangerous to send an unstable patient for an MRI. As the patient is afebrile, LP is less likely to be illuminating about the cause of her seizures. LP could be needed especially if there is a strong suspicion of infection such as meningitis but can be delayed if the patient is unstable and antibiotics initiated. While a CEEG may be needed especially if the patient is intubated or comatose and there is a risk of non-clinical seizures, it is not the first-line diagnostic tool.

Excellent explanation Pradip, it is of utmost importance to make sure you assess for electrolyte disturbances or glucose abnormalities in your rapid diagnostics when patients are seizing. Remember hyponatremia, hypoglycemia, and hypocalcemia. If you have a child with Seizures 

• As you think about our case, what would be your differential for rhythmic jerking movements that mimic or are associated with seizures?
• Movement disorders: Any abnormal involuntary movements such as Tics, tremor, chorea, athetosis, dystonia, myoclonus, ballismus, asterixis. Dyskinesia is a generalized term used for abnormal involuntary movements
• Migraine (its paroxysmal nature + association with neuro-deficits or altered consciousness) may lead to confusion with seizures.
• In infants paroxysmal non-epileptic disorders such as jitteriness, benign neonatal myoclonus may be confused with seizure
• Myoclonus from drugs such as etomidate or post drowning due to hypoxia reperfusion injury may be mistaken for seizures

Let’s transition and highlight key definitions of status epilepticus:

Previously defined as a seizure lasting > than 30minutes or recurrent seizures lasting > 30minutes without patient regaining consciousness between seizures. The new definition refers to SE as 5minutes or more of either continuous seizure or 2 or more discrete seizures between which there is incomplete recovery of consciousness.

Refractory SE = SE that persists despite the administration of first and second-line anti-seizure medications with different mechanisms of action.

Super refractory SE refers to SE that continues 24 hours or more after the onset of anesthetic therapy for SE and includes recurrence during reduction or withdrawal of anesthetic therapy.

Pradip what is the most common cause of seizures in the pediatric population?

The majority of pediatric SE (30-50%) involved febrile seizures. About 9-17% involved either acute metabolic derangement or a CNS infection. 12% of first seizures in children present with status epilepticus (Shinnar, Pediatrics 1996)

What is the pathophysiology of seizures and its progression to status epilepticus?

There is an imbalance between excitation and inhibition. Ineffective recruitment of GABA neurons coupled with excessive excitatory NMDA neuronal stimulation leads to initiation and propagation of the electrical disturbance in SE. Prolonged seizures lead to selective neuronal loss in the hippocampus, cortex, and thalamus.

There is neurotoxicity due to excitotoxicity (via excess stimulation from glutamate on NMDA and AMPA receptors) as well as hypoxic-ischemic injury (imbalance between increased metabolic demand and cerebral blood flow/oxygenation). Hypoxia, acidosis, hypotension, and hypercarbia add to the ongoing damage.

There are early (< 30minutes) and late (> 30minutes) time-related complications of status epilepticus which are nicely elucidated in the LearnPICU status epilepticus-pathophysiology. (http://www.learnpicu.com/neurology/status-epilepticus)

The risk of subsequent epilepsy after status epilepticus is 26-36% (Barnard, J child Neurol 1999 and Eriksson, Develop Med Child Neurol 1997).

Would you also mind highlighting the way seizures are classified?

Seizures are classified as Partial or generalized based on clinical presentation or EEG FINDINGS. Partial Seizures arise in specific areas of the brain and are further classified as simple, local, or focal. Generalized seizures arise from diffuse cortical areas at one time. They involve both cerebral hemispheres and consciousness is typically impaired. Generalized can present as motor movements or absence seizures during which no convulsions are seen.

• If you had to work up this patient with status epilepticus what would be your diagnostic approach?
• I would start with some basic labs such as glucose, serum electrolytes including magnesium and calcium. I also typically add a DIC panel and CPK for especially for prolonged seizures.
• If there is concern for infection then CBC with differential, Lumbar puncture, CRP, procal, appropriate cultures (urine, blood, and CSF) should be sent. Virals studies such as HSV PCR from blood/CSF as well as a respiratory viral panel.
• Another thing to look at is the drug levels of any previous anti-epileptic agents (as agent withdrawal or change can precipitate seizures).
• In selected cases where inflammation is suspected- ESR, CRP, vWF antigen may be required. additionally, oligoclonal bands, testing for antibodies including neuronal and ion channel antibodies may be required from blood as well as the CSF.
• Rarely evaluation for toxins, metabolic disease, ophthalmologic evaluation may be needed in selected cases.
• In patients with established epilepsy- imaging is typically not necessary. Otherwise, brain imaging (either a CT or MRI) is required especially for a new status epilepsy
• cEEG in the PICU is required especially if the patient is intubated or comatose as the patient could continue to have non-clinical status. The overall incidence of electrographic seizures in critically-ill patients was ~ 26%.
• Yes, Rahul - I would also like to highlight a “new-age technology” with regards to EEG.
• One study (Fung F. et al. Epilepsia 2020) devised a predictive model for capturing electrographic seizures in critically ill pediatric patients. The model had a sensitivity of 92% with a negative predictive value of 93%. Variables associated with increased capturing of seizures on this monitor included:
• age (<1 or >1 year of age)
• acute encephalopathy category
• clinical seizures prior to CEEG initiation
• EEG background (slow disorganized, discontinuous, or burst suppression background)
• epileptiform discharges during the initial 30minutes of the recording. We should be cognizant that equipment for cEEG, as well as staffing, may not be available at all centers.

To summarize, these are the common causes of seizures in the PICU — AED withdrawal or change, drug toxicity or withdrawal, electrolyte problems, hypertensive encephalopathy, tumor, TBI, vasculitis, renal/hepatic dysfunction, fever, hypoxia/ischemia, and postoperative conditions. Pre-existing epilepsy, genetic and central nervous system disorders can also present with seizures. Intensivists should be vigilant about non-convulsive status especially in children who have hypoxic injury s/p cardiac arrest, submersion injury, TBI, and stroke.

• If our history, physical, and diagnostic investigation led us to status epilepticus as our diagnosis what would be your approach to general management?
• In the initial phase (0-5minutes): I would focus on stabilization of the patient’s airway/breathing and hemodynamics. Establish IV/IO access and supplement patients’ oxygenation and focus on correcting any abnormal glucose or electrolytes.
• Medications: Benzodiazepines (BZDs) are the first-line agents for status epilepticus.
• The BZDs work by potentiating the neuro-inhibitory effects of Gamma-aminobutyric acid (GABA).
• Lorazepam, diazepam and midazolam are frequently used.
• Zhao ZY et al. (J Child Neurol. 2016) in a network meta-analysis of 16 RCTs including 1821 patients which compared the efficacy of midazolam, lorazepam, and diazepam in treating pediatric status epilepticus concluded that non-IV midazolam and IV lorazepam were superior to IV or non-IV diazepam, and IV lorazepam was at least as effective as non-IV midazolam.

Summary: IV Ativan and IV Midazolam if your patient has good access are equally effective

• Yes, All the aforementioned benzodiazepines are lipid-soluble entering the brain within 2 minutes of IV administration.
• Diazepam has the highest lipid solubility and is also highly protein-bound and thus has a large volume of distribution of the unbound drug. Thus the effective duration of action for diazepam in SE is 20-30minutes resulting in rapid redistribution compared to lorazepam which has a much smaller volume of distribution of unbound drug and thus has a longer duration of action in SE. Hence lorazepam is the preferred agent in the initial management of SE.
• Midazolam can be given intranasally or intramuscularly inpatient without IV access. In fact, one study (Silbergleit R et al. NEJM 2012) showed that IM midazolam was as effective and safe as IV lorazepam for prehospital seizure termination. Rectal Diazepam is an option if unable to get IV access.
• How many doses of benzodiazepines would you give Rahul, and what is the pharmacokinetics to keep in mind?
• More than two doses of benzodiazepines are associated with side effects without a substantial increase in efficacy. The potency of BZDs decreases 20 fold over 30 minutes of SE. Receptor trafficking of GABAa receptors resulting in movement of the receptors from the synaptic membrane into the cytoplasm where they become functionally inactive. This reduces the number of GABAa receptors available on the synaptic surface to bind BZD, and in turn, leads to a single seizure becoming self-sustaining a time-dependent resistance to BZD develops. Additionally > 2 doses increases risk of respiratory depression (43% risk compared to 13% with < 2 doses). Furthermore, only 13% of patients achieved seizure termination