The return of Why we do what we do focuses on the first drug class in the management of status epilepticus – benzodiazepines. What make them such a good choice? How are they best administered? When should we stop giving them? Read on and find the answers to these questions and more.

How is status epilepticus defined?

Status Epilepticus is defined as persistent seizure activity or intermittent activity without return to baseline between episodes that last for more than 5 minutes. Note that these recommendations and this discussion is for patients older than 4 weeks of age. Neonatal status epilepticus is a whole different enchilada – the first line is Phenobarbital, though levetiracetam is also being studies.

How do benzodiazepines stop seizures?

Originally developed in the 1930s and first utilized as anxiolytics. It wasn’t until two decades later that they gained favor as anti epileptics. Benzos are potentiate the effect of GABA at the GABA-A receptor and cause sedation, hypnosis, anxiolysis, muscle relaxation and of course, stop seizure activity.

Which benzos should we be familiar with?


  • Dose: 0.1 mg/kg IV, max 4 mg should be administered by slow IV push over two minutes
  • Assess for impact over the subsequent 2-5 minutes
  • Can suppress seizures for 4-6 hours – note that this is much shorter than the half-life which is 10 to 13 hours in children and up to 40! hours in newborns. It is not distributed into fat as much as other benzos – therefore it hangs around longer.


  • Dose: 0.2 mg/kg IV (max 8 mg)
  • IM: 10 mg for > 40 kg, 5 mg for 13-40 kg, or 0.2mg/kg for weight <12kg
  • Diazepam is very lipid soluble, and rapidly crosses the blood-brain barrier thus 10 to 20 seconds after administration it can stop a seizure
  • It readily redistributes into adipose tissue redistribution of the drug into adipose tissue and thus the duration of action is as short as <20 minute
  • Stable at room temperature – thus it is a good choice for home storage – therefore it is available as a rapidly absorbed rectal gel (Diastat)


  • Dose: 10 mg for > 40 kg, 5 mg for 13-40 kg, or 0.2mg/kg for weight <12kg
  • It terminates seizures in <1 minute and generally has a short half life in the CNS
  • Intramuscular, intranasal, oral, buccal, or rectal – a great choice when an IV is not available
  • Buccal administration of midazolam was found in one study of 177 children to be more effective than rectal diazepam in terminating seizures

Is one agent better than the rest?

In general the answer is not really – in my estimation, and that of many of my peers lorazepam and midazolam sit at the head of the pack for use in status epilepticus in the Peds ED. Let’s take a look at the evidence in more detail shall we? As you’ll see, our current understanding has matured since the 1990s.

The influence of diazepam or lorazepam on the frequency of endotracheal intubation in childhood status epilepticus
Chiulli DA, et al. J Emerg Med. 1991

  • Retrospective investigation of 142 children with seizures (27% had status epilepticus) with a focus on rats on intubation rates
  • Patients that got lorazepam had were intubated 27% of the time (4/15) versus 73% (8/11) of the diazepam group, p=0.026
  • No difference in groups with regard to age, weight, sex, seizure type, seizure duration, and appropriate anticonvulsant dosage
  • Bottom line: Really small study with a slant towards lorazepam from a safety perspective

Lorazepam versus diazepam in the acute treatment of epileptic seizures and status epilepticus
Appleton et al., Dev Med Child Neurol. 1995

  • Lorazepam vs diazepam were prospectively evaluated in 102 children
  • One agent was given on even days, another on odd days
  • 76% of lorazepam patients had their seizures controlled vs 51% diazepam
  • 3% of the lorazepam group had respiratory depression versus 15% diazepam
  • Bottom line: The methodology introduced some notable biases. Results suggested that lorazepam was more efficacious and safer

Comparative audit of intravenous lorazepam and diazepam in the emergency treatment of convulsive status epilepticus in children
Qureshi A,et al., Seizure. 2002

  • Bifurcated study design in which the investigators used diazepam for 6 months, then one year of lorazepam
  • Outcomes included latency to stopping of seizure (ceased within 15 minutes of IV drug) plus adverse events
  • IV diazepam was used in 17/26 patients 65% had seizure control in <15 minutes
  • IV lorazepam was used in 31/59 patients – also with 65% seizure stoppage rate

Drug management for acute tonic-clonic convulsions including convulsive status epilepticus in children: Systematic review of four trials
Appleton et al., Cochrane Database Syst Rev, 2008

Review of 4 studies – they concluded the following:

  • IV lorazepam is as effective as intravenous diazepam in the treatment of acute tonic clonic convulsions, 19/27 (70%) versus 22/34 (65%), RR 1.09 (95% CI 0.77 to 1.54)
  • Buccal midazolam controlled seizures in 61/109 (56%) compared with 30/110 (27%) of rectal diazepam treated episodes with acute tonic-clonic convulsions, RR 2.05 ( 95% CI 1.45 to 2.91)
  • Intranasal midazolam is as effective as intravenous diazepam in the treatment of prolonged febrile convulsions, 23/26 (88%) versus 24/26 (92%), RR 0.96 (95% CI 0.8 to 1.14)

Intramuscular versus Intravenous Therapy for Prehospital Status Epilepticus
Silbergleit et al NEJM, 2012

  • Double blind RCT of IM midazolam versus IV lorazepam in both adults and children administered by EMS personnel
  • Primary outcome was no longer seizing upon arrival to the ED
  • Hypothesis was no different fference with midazolam <10% different
  • The IM midazolam group saw 329/448 subjects (73.4%) seizure free versus 282/445 (63.4%) in the IV lorazepam group (absolute difference, 10 percentage points; 95% CI, 4.0 to 16.1; P<0.001 for both noninferiority and superiority)
  • Seizures stopped faster in the midazolam group overall 1.2 versus 4.8 minutes
  • Both treatment groups were similar
  • Almost ¾ arrived in the ED seizure free

Lorazepam vs Diazepam for Pediatric Status Epilepticus
Chamberlain et al., JAMA, 2014

  • Double blind RCT age 3 months to 18 years with status epilepticus across 11 Pediatric Emergency Departments
  • 273 patients; 140 randomized to diazepam and 133 to lorazepam (both IV)
  • If seizures continued at 5 minutes they got ½ the original se, followed by fosphenytoin 12 minutes henceforth
  • The primary outcome was cessation of status epilepticus by 10 minutes without recurrence within 30 minutes. they saw that diazepam worked in 101/140 (72.1%) and lorazepam in 97/133 (72.9%) – absolute efficacy difference of 0.8% (95% CI, −11.4% to 9.8%)
  • Additionally they looked at need for assisted ventilation 4 hours after drug administration – 26 in each group needed assistance, 16.0% of diazepam and 17.6% of lorazepam – absolute risk difference, 1.6%; (95% CI, −9.9% to 6.8%)
  • Lorazepam patients were a bit more likely to be sedated overall 66.9% vs 50%, absolute risk difference, 16.9% (95% CI, 6.1% to 27.7%)

How should I use benzos to stop a real life seizure?

I’ll keep it simple and answer based on the following scenarios:

If you have an IV/IO

Give lorazepam

If you don’t have an IV, and can give IM

Give midazolam

If you can’t give it IM and the patient has a rescue med use it

Generally rectal diazepam, but buccal or nasal midazolam 0.2 mg/kg, max 10 mg may be more effective

How many doses is too many?

Two, that’s it. In general you should consider ANY benzo given by pre-hospital personnel as a single dose – IM or IV. Home doses may not have been administered completely, or could have latency. It is up to you and your institution if the timing and administration of home meds counts.
Why two, well simply, the risk of respiratory depression increases if you give a third dose. This was noted by Chin et al in Lancet Neurol, 2008 in a prospective study where they noted that treatment with more than two doses of benzodiazepines was associated with respiratory depression (OR 2.9, 1.4-6.1).