Case Based Pediatrics For Medical Students and Residents
Department of Pediatrics, University of Hawaii John A. Burns School of Medicine
Chapter XVIII.5. Status Epilepticus
Loren G. Yamamoto, MD, MPH, MBA
January 2002

Return to Table of Contents

A 6 year old male is noted by his mother to have a generalized tonic clonic seizure. He is unresponsive and his eyes are rolled up. She tries to blow on his face and shake his shoulders, but the seizure continues. His face is turning blue so she calls 911. The sequence of events proceeds as below:

Time 0: Onset of seizure.

Time 3 minutes: Mom calls 911 (activating EMS).

Time 12 minutes: Ambulance arrives at home.

Time 14 minutes: Ambulance crew at the patient's side. Oxygen saturation 80%. Patient still seizing.

Time 20 minutes: Mask ventilation with oxygen, IV started. Oxygen saturation improves to 95%.

Time 25 minutes: Lorazepam 0.1 mg/kg IV given. Patient being transported, treatment en route. Oxygen saturation is now 100%.

Time 31 minutes: Patient arrives in the emergency department, still seizing.

Time 33 minutes: Lorazepam 0.1 mg/kg IV repeated.

Time 34 minutes: Phenytoin or fosphenytoin 10 minute infusion started (larger patients may require longer infusion times).

Time 44 minutes: Patient still seizing so another anticonvulsant such as phenobarbital is administered.

Time 54 minutes: Seizures continue.

Status epilepticus is defined as prolonged seizures which continue or occur in rapid succession with relatively brief intervals in between. A minimum time duration is not part of a universal definition. The most severe form of SE is generalized tonic clonic SE which results in respiratory compromise (hypoxemia and hypercapnia), with hypermetabolic cerebral activity (excessive cerebral oxygen demand). This results in respiratory and metabolic acidosis. Skeletal muscle contraction may result in rhabdomyolysis and hyperkalemia. Other forms of status epilepticus may not be as obvious. Partial complex SE and absence (petit mal) SE, are less common and may be difficult to recognize. The discussion in this chapter will focus on generalized tonic clinic status epilepticus.

Status epilepticus may be due to several causes. Patients with epilepsy or brain anomalies may present with SE, or they may have breakthrough seizures presenting with SE. SE can also result from periods of anticonvulsant non-compliance. SE may be due to encephalitis or other brain infections. SE may be due to drug overdose toxicity (e.g., isoniazid), traumatic brain injury, neoplastic conditions, metabolic derangements, encephalopathy or cerebrovascular accident.

The desired goal in the management of status epilepticus is to terminate the seizures and restore the patient to baseline as soon as possible while maintaining oxygenation, circulation and normoglycemia (1). Prognosis is determined by the etiology and the duration of SE (1). Complications of prolonged SE include hypoxic ischemic brain injury which may be due to the combined factors of prolonged hypoxia, hypermetabolic cerebral oxygen demand, and respiratory acidosis.

In addition to terminating the seizures, treatment for possible treatable causes of SE should be initiated. Acyclovir and ceftriaxone are commonly administered while treating SE, since the etiology of SE may not be determined until much later.

Since benzodiazepines have the most rapid anticonvulsant onset, it is well accepted that the initial drug to treat SE should be a benzodiazepine. Phenytoin and barbiturates can also be used IV, but their onset time is slower. Other IV drugs such as lidocaine and magnesium have been reported to have anticonvulsant activity, but their use in SE is not routine. Paraldehyde is difficult to use and is no longer pharmacologically available in the United States (2). Reports of propofol and inhaled anesthetics having efficacy in SE have been cited elsewhere (1,2). IV valproic acid has been suggested as an additional agent (3).

IV lorazepam (Ativan) is the most commonly recommended benzodiazepine for SE, because of its longer duration. Diazepam (Valium), which can be given IV or rectally, is also very effective. Midazolam (Versed) can also be used, with its main advantage being that it can be given IM, while the IM route is not recommended for lorazepam and diazepam. If the patient has an IV, lorazepam should be given (diazepam can be used if lorazepam is not available). If an IV is not available, then rectal diazepam or IM midazolam can be used. If rectal diazepam is used, its dose should be doubled or quadrupled in order to get the same effect as the IV dose. A commercially available unit dose rectal diazepam (Diastat) is available or one could administer the IV form of diazepam by: 1) For small doses, a 1cc syringe could be used with lubricant applied to the syringe itself. Then the barrel of the syringe can be directly inserted into the rectum. 2) For larger doses, the diazepam can be drawn into a larger syringe and a rubber catheter or feeding tube can be attached to the hub with the other end inserted into the rectum. The diazepam is pushed through the tube or catheter, followed by an air bubble to clear the tube.

SE is often treated in a stepwise approach with an anticonvulsant drug given every 5 to 10 minutes until SE terminates. A benzodiazepine is administered first (e.g., lorazepam 0.1 mg/kg). This can be repeated in 5 to 10 minutes, if SE continues. In most instances, in patients who were not previously on chronic anticonvulsants, IV loading with one of the phenytoins (phenytoin or fosphenytoin) should immediately follow the first benzodiazepine dose without waiting to see if the benzodiazepine will work. Infants, and especially neonates, may be an exception where phenobarbital may be preferred (1) . For epilepsy patients who are already on anticonvulsants, stat drug levels should be obtained to determine if these are within the therapeutic range. If they are not, an IV or oral dose (or via nasogastric tube) can be given to raise their level appropriately.

In many instances, this first or second benzodiazepine dose will terminate SE and restore the patient to their baseline state (1). These cases are relatively simple and are at less risk for complications. For the purpose of this discussion, this chapter will refer to this as simple status epilepticus. It should be expected that once the benzodiazepine wears off, seizures may recur unless a longer acting anticonvulsant has been administered or a subtherapeutic anticonvulsant level has been brought back up into the therapeutic range.

When the first one or two doses of a benzodiazepine fail to terminate SE, these patients are at higher risk for prolonged SE and complications secondary to this. Refractory status epilepticus can be loosely defined as SE which continues despite an initial anticonvulsant drug regimen (such as a benzodiazepine and phenytoin). If SE continues after two doses of a benzodiazepine, refractory SE should be anticipated. The treatment of refractory SE is more complex and controversial.

The case described in the beginning of the chapter indicates a typical sequence of drugs administered in an attempt to terminate status epilepticus. This sequence may seem fast or slow to you depending on your perspective. It is actually fast since these times are highly ideal and medications are given rapidly without hesitation or pause for prehospital communication. In most instances when conditions are less than ideal, it is common for one hour to elapse while initial anticonvulsant agents are still being administered. One published sequence which completes its second line of drugs within 30 minutes is deceivingly fast because "time 0" is when treatment (ABC+vascular access and glucose) is initiated and only 5 minutes is given to complete an IV benzodiazepine, IV glucose, IV phenytoin (published recommended infusion times for this drug alone are longer than 5 minutes), brief history and neuro exam, reassessment of vital signs, a check of anticonvulsant levels and a repeat phenytoin mini bolus (all between Time=10 minutes to Time=15 minutes) (1).

Another published sequence defines its "time 0" when the first benzodiazepine is administered IV (this equates to Time=25 minutes in the sequence described in the case). This sequence uses two doses of a phenytoin and two doses of phenobarbital to end at approximately 68 minutes from the time that the first benzodiazepine was administered (4) (add about 25 minutes to this for the total duration of SE).

Neither of these published sequences provides for a period of observation following the administration of an anticonvulsant to see if the drug is successful in terminating SE. This is a common practice and it adds to these time sequences which already approach or exceed one hour. One source states that "the treating physician should allow adequate time for the anticonvulsants to reach therapeutic levels in the brain" (5). A major problem with this approach is that the "wait and see" time periods add up rapidly if SE is not terminated. This time is frequently further prolonged by delays in obtaining IV access, longer prehospital transport times, medication administration delays and if smaller initial doses of benzodiazepines are initially used (e.g. 0.05 mg/kg of lorazepam). No agent or combination of agents is universally reliable in terminating SE (1). It is easy to see how SE can sometimes continue for 2 or 3 hours before it is finally terminated. This is not ideal.

Since it is best to minimize the time of SE, it is best to minimize the time in which anticonvulsants are administered (1). Because of this, it may be unwise to give small doses and wait between anticonvulsants to see if it works (1). Larger full loading doses of anticonvulsants should be preferred. The onset time pharmacology of these drugs may have to be ignored (onset times range from 2 to 30 minutes) in order to minimize the duration of the anticonvulsant sequence. This "wait and see" approach should be modified to a "give, look and give another" approach when managing refractory SE. Refractory SE can be suspected early (as opposed to simple SE) when the first two doses of a benzodiazepine fail to terminate SE.

In the pharmacologic management of refractory SE, there are three basic approaches: 1) the "wait and see" approach, 2) the "give, look and give" approach, 3) the early rapid sequence intubation approach. While the first approach ("wait and see") is commonly used, this practice should be discouraged in managing refractory SE; thus, it should be eliminated from consideration in order to minimize the duration of refractory SE as discussed above. This leaves only two reasonable approaches.

In the early rapid sequence intubation (RSI) approach, refractory SE is recognized early once failure to respond to the first two large doses of benzodiazepines and the phenytoin load is noted. RSI (intubation facilitated by pharmacologic paralysis and deep sedation) has been discussed elsewhere and is beyond the scope of this chapter (6). Early RSI will paralyze the patient to optimize oxygenation, restore ventilation (reversing respiratory acidosis) and halt skeletal muscle contractions (halting rhabdomyolysis and excessive peripheral oxygen consumption). Its commonly recognized disadvantage is that continued seizure activity cannot be determined without the use of continuous electroencephalography (EEG) which is not available in most centers (7). However, the treatment for continued seizures is more anticonvulsants. RSI is often accompanied by a potent anticonvulsant and cerebroprotective agent such as thiopental. Etomidate is often used as a sedative in RSI, but this agent may worsen a seizure condition or cause myoclonus (8). RSI facilitates the administration of maximal anticonvulsant therapy since there is no risk of respiratory depression (the patient is already paralyzed and intubated). Loading with maximal doses of IV phenytoin and phenobarbital can now take place. Although the patient is paralyzed and continued seizure activity cannot be recognized, maximal doses of a benzodiazepine, phenytoin and phenobarbital have been administered. What more could be done even if the clinician could still see the seizures? Adding barbiturate coma, a high dose benzodiazepine infusion, propofol or IV valproic acid could be considered (1,2,3). Short acting paralyzing agents allow reasonably prompt recovery to regain the ability to witness continued seizure activity (4). General anesthesia with barbiturate coma with paralysis is commonly recommended for SE exceeding 60 minutes (5), yet the decision to initiate RSI (a similar procedure) a few minutes earlier, is paradoxically frowned upon by some. Refractory SE durations approaching and exceeding an hour are not uncommon. Since neuronal damage may occur long before 60 minutes of seizures (5) and barbiturate coma may be time consuming to initiate, RSI performed earlier may provide its benefits early and facilitate the rapid initiation of barbiturate coma if it is required. RSI also facilitates expedient neuroimaging which is frequently necessary in SE patients. Several reports recommend endotracheal intubation in the management sequence of refractory SE (1,5,9). If this is so, then should intubation be performed early versus late?; without RSI versus with RSI? These preferences determine if one will be an advocate of an early RSI approach.

The "give, look and give" approach is also a reasonable approach for those who judge the disadvantages of early RSI to outweigh its advantages. Continued seizures after the first two large doses of a benzodiazepine and the phenytoin load should signify refractory SE and a high risk patient. There is a high risk of apnea when phenobarbital is given in combination with benzodiazepines, thus intubation may be necessary (1). Barbiturate coma, benzodiazepine infusion, propofol, IV valproic acid (1,2,3) and/or rapid sequence intubation (if not already done) can be considered after this point.

The table below describes a typical time sequence difference between the "Give look and give" approach compared to the "Early RSI" approach. The comparison starts at time=20 minutes when paramedics or medical personnel first arrive.

"Give look and give" approach
"Early RSI" approach
20 min.
Oxygen + p.r.n. mask ventilation.Oxygen + p.r.n. mask ventilation.
25 min.
Lorazepam 0.1mg/kg IV.Lorazepam 0.1mg/kg IV.
27 min.
Phenytoin 15-20mg/kg IV over 10 min.Phenytoin 15-20mg/kg IV over 10 min.
37 min.
Lorazepam 0.1mg/kg IV.Lorazepam 0.1mg/kg IV.
42 min.
Phenobarbital 15mg/kg IV slowly.RSI drug sequence begins.
46 min.
Visible seizures continue.Visible seizures stop. Patient intubated.
47 min.
Phenobarbital infusion continues.Phenobarbital 25 mg/kg IV.
50 min.
Visible seizures continue.Phenobarbital infusion complete.
52 min.
Phenobarbital infusion complete.
53 min.
Apnea develops.
55 min.
Intubation done under difficult conditions.
60 min.
Additional phenobarbital 10mg/kg IV.

In refractory SE, when maximal anticonvulsant administration may be required, the early RSI approach completes drug administration faster and intubation is facilitated. Additionally, the patient is paralyzed and intubated earlier facilitating oxygenation and halting skeletal muscle activity. Phenobarbital must be given slowly following a benzodiazepine to prevent apnea, but this often happens anyway. The time sequence can be shortened by using fosphenytoin which can be given IV faster than phenytoin, or interrupting the phenytoin infusion to give the second lorazepam dose.

In summary, to minimize the duration of SE, the "wait and see" approach should be discouraged in favor of the "give look and give" approach or the early rapid sequence intubation approach in order to minimize the duration of drug administration and the duration of SE. Refractory SE risk should be recognized earlier (when SE continues despite the first two benzodiazepine doses), so that anticonvulsant administration can be done more aggressively since time is critical.


1. After oxygen, the first drug that is administered to a patient in status epilepticus is from what drug class?

2. Name three other anticonvulsant drugs, not belonging to the class above, that can be given IV?

3. Name two ways that diazepam can be given in status epilepticus?

4. Which benzodiazepine has the longest duration?

5. In status epilepticus, what drug should be administered after a benzodiazepine in most instances (other than in neonates)?

6. What is the most serious complication of status epilepticus?

7. Name 5 causes of status epilepticus?


1. Haafiz A, Kissoon N. Status epilepticus: Current concepts. Pediatr Emerg Care 1999;15(2):119-129.

2. Mitchell WG. Status epilepticus and acute repetitive seizures in children, adolescents, and young adults: etiology, outcome, and treatment. Epilepsia 1996;37(supp 1):S74-S80.

3. Yamamoto LG, Yim GK. The Role of Intravenous Valproic Acid in Status Epilepticus. Pediatr Emerg Care 2000;16(4):296-298.

4. Lowenstein DH, Alldredge BK. Status Epilepticus. New Engl J Med 1998;338(14):970-976.

5. Status Epilepticus. In: Strange GR, et al (eds). APLS: The Pediatric Emergency Medicine Course, third edition. Dallas, TX: American College of Emergency Physicians and Elk Grove Village, IL: American Academy of Pediatrics, 1998, pp. 191-197.

6. Gerardi MJ, Sacchetti AD, Cantor RM, Santamaria JP, Gausche M, Lucid W, Foltin GL. Rapid-sequence intubation of the pediatric patient. Ann Emerg Med 1996;28(1):55-74.

7. Terndrup TE. Clinical issues in acute childhood seizure management in the emergency department. J Child Neurol 1998;13(Supp 1):S7-S10.

8. Bergen JM, Smith DC. A review of etomidate for rapid sequence intubation in the emergency department. J Emerg Med 1997;15(2):221-230.

9. Bebin M. The acute management of seizures. Pediatr Ann 1999;28(4):225-229.

Answers to questions

1. Benzodiazepines.

2. Phenytoin (or fosphenytoin), phenobarbital, valproic acid.

3. IV and rectal.

4. Lorazepam.

5. Phenytoin (or fosphenytoin).

6. Hypoxic ischemic brain injury, death.

7. Epilepsy, encephalitis, neoplasm, drug overdose, metabolic derangement, cerebrovascular accident, trauma, etc.

Return to Table of Contents

University of Hawaii Department of Pediatrics Home Page