Your comprehensive evidence-based neurology knowledge base.
347 topics across 9 neurology specialties
AES Guideline: Treatment of Convulsive Status Epilepticus (2016)
Evidence-based guideline from the American Epilepsy Society (AES), published in Epilepsy Currents (2016), reviewing 38 RCTs on pharmacologic treatment of convulsive status epilepticus in children and adults.
🔹 Bottom Line
Guideline: AES 2016 evidence-based guideline — 38 RCTs reviewed for CSE treatment in children and adults
First-line (Level A): BZDs are the initial therapy of choice — IM midazolam, IV lorazepam, or IV diazepam; all equivalently efficacious
No IV access (Level A): IM midazolam is superior to IV lorazepam (RAMPART: 73% vs 63%, p < 0.001)
Second-line (Level U): No evidence-based preferred agent — fosphenytoin, valproic acid, or levetiracetam as single dose
Efficacy drops sharply: VA Cooperative — 1st drug 55.5% → 2nd drug 7.0% → 3rd drug 2.3%
Fosphenytoin > phenytoin on tolerability (Level B); phenytoin is an acceptable alternative
Respiratory depression is LOWER with BZDs than with placebo — untreated SE is more dangerous (Level A)
Children: Non-IV routes (rectal diazepam, buccal/intranasal midazolam) are probably effective (Level B)
Guideline Overview
Source & Scope
Organization: American Epilepsy Society (AES) Guideline Committee
Evidence base: 38 RCTs identified — 4 class I, 2 class II, 32 class III
Scope: Pharmacologic treatment of convulsive SE in children and adults; seizures ≥5 min
NOT addressed: Refractory SE, neonatal SE, NCSE, etiology-specific therapy, surgery, ketogenic diet, neurostimulation
Key Definitions
Status epilepticus: Seizure lasting ≥5 min (operational definition for treatment initiation)
Traditional SE: ≥30 min continuous seizure activity → risk of permanent neuronal injury
Efficacy: Ability of drug to stop CSE
Tolerability: Incidence, severity, and impact of anticonvulsant-related adverse effects
Effectiveness: Encompasses both efficacy and tolerability
Safety: Life-threatening adverse events
Evidence Classification
Level
Evidence Required
Conclusion
Recommendation
A
≥1 class I study or ≥2 consistent class II
Established as effective/ineffective/harmful
Should be done / should not be done
B
≥1 class II study or ≥3 consistent class III
Probably effective/ineffective/harmful
Should be considered / should not be considered
C
≥2 consistent class III studies
Possibly effective/ineffective/harmful
May be considered / may not be considered
U
Insufficient evidence
Data inadequate or unproven
None
Q1: Which Anticonvulsants Are Efficacious?
Adults — Initial Therapy
Agent
Route
Evidence Level
Key Data
Midazolam
IM
Level A
RAMPART: 73% seizure cessation; superior to IV LZP without IV access
Lorazepam
IV
Level A
VA Cooperative: 59.1% vs placebo 21.1% (OR 4.8); may repeat ×1
Diazepam
IV
Level A
VA Cooperative: 42.6% vs placebo 21.1% (OR 2.3); may repeat ×1
Phenobarbital
IV
Level A
VA Cooperative: efficacious but slower administration rate
No difference in efficacy between IV lorazepam and IV diazepam (Level A)
IM midazolam has superior effectiveness compared with IV lorazepam in adults without established IV access (Level A)
IV lorazepam is more effective than IV phenytoin at stopping seizures ≥10 min (Level A)
Adults — Second-Line Therapy
Agent
Dose
Max
Evidence Level
Fosphenytoin
20 mg PE/kg
1500 mg PE
Level U
Valproic acid
40 mg/kg
3000 mg
Level B
Levetiracetam
60 mg/kg
4500 mg
Level U
Phenobarbital
15 mg/kg
—
Level B (alternative if above unavailable)
No evidence that any second-line option is better than the others
IV valproic acid has similar efficacy to IV phenytoin (Level B) with better tolerability
Insufficient data on levetiracetam as second therapy in adults (Level U)
All second-line agents should be given as a single full dose
Children — Initial Therapy
Agent
Route
Evidence Level
Notes
Lorazepam
IV
Level A
0.1 mg/kg/dose, max 4 mg; may repeat ×1
Diazepam
IV
Level A
0.15–0.2 mg/kg/dose, max 10 mg; may repeat ×1
Diazepam
Rectal
Level B
0.2–0.5 mg/kg, max 20 mg
Midazolam
IM
Level B
10 mg for >40 kg, 5 mg for 13–40 kg
Midazolam
Intranasal
Level B
Non-IV midazolam probably more effective than diazepam (IV/rectal)
Midazolam
Buccal
Level B
Faster time to seizure cessation vs rectal diazepam
No difference between IV lorazepam and IV diazepam in children (Level A)
Insufficient data on intranasal lorazepam, sublingual lorazepam, rectal lorazepam, valproic acid, levetiracetam, and phenytoin as initial therapy in children (Level U)
🔹 Clinical Pearl
The #1 treatment error in SE is BZD underdosing. Give the full guideline-recommended dose as a single administration. Initial therapy should NOT be broken into multiple smaller doses (except IV lorazepam and diazepam which can be repeated once).
Q2: Adverse Events
Adults (Level A)
Most common adverse events: Respiratory and cardiac symptoms with IV anticonvulsant administration
VA Cooperative data:
Agent
Hypoventilation
Hypotension
Cardiac Rhythm
Lorazepam
10.3%
25.8%
7.2%
Diazepam
16.8%
31.6%
2.1%
Phenobarbital
13.2%
34.1%
3.3%
Phenytoin
9.9%
27.0%
6.9%
Placebo (untreated)
22.5% treatment-emergent adverse events
Critical finding: Respiratory depression is LOWER in BZD-treated patients than placebo-treated patients (Level A)
This confirms that respiratory problems are a consequence of untreated SE, not of BZD treatment
No substantial difference in cardiorespiratory adverse events between BZDs and phenobarbital (Level A)
RAMPART Adverse Events (IM Midazolam vs IV Lorazepam)
Treatment-emergent AEs: IM midazolam 26.7% vs IV lorazepam 30.6%
Decreased consciousness: MDZ 9.5% vs LZP 8.8%
Respiratory depression: MDZ 6.4% vs LZP 10%
Hypotension: only 1.2% overall
No significant differences in adverse events between the two agents
Children
Respiratory depression is the most common clinically significant adverse event (Level B)
No substantial difference between midazolam, lorazepam, and diazepam by any route for respiratory depression rates (Level B)
Adverse events with BZDs reported less frequently in children than adults (Level B)
Respiratory depression after rectal diazepam in children: 1.2–6.4% across class III studies
Buccal midazolam: no respiratory depression reported in 2 class III pediatric studies
🔹 Clinical Pearl
Untreated SE causes MORE respiratory depression than BZD treatment. The VA Cooperative trial showed placebo-treated patients had a 22.5% adverse event rate vs ~10% with lorazepam. Never withhold BZDs due to fear of respiratory depression.
Q3: Most Effective Benzodiazepine
Adults (Level A)
No significant difference in effectiveness between IV lorazepam and IV diazepam
IM midazolam is established as more effective than IV lorazepam when IV access is not available (Level A)
Pharmacokinetic advantage: lorazepam has a longer duration of action (but not longer half-life) than diazepam
RAMPART Trial (2012) — Class I
IM Midazolam
IV Lorazepam
Difference
n
448
445
—
Seizure cessation
73.4%
63.4%
10% (95% CI: 4.0–16.1; p < 0.001)
Median time to cessation
1.6 min from active Tx
4.8 min from active Tx
IM faster due to quicker delivery
Result
IM MDZ met noninferiority AND demonstrated superiority
Children (Level A & B)
No significant difference between IV lorazepam and IV diazepam (Level A)
Non-IV midazolam (IM/intranasal/buccal) is probably more effective than diazepam (IV/rectal) in children (Level B)
Meta-analysis of 6 pediatric studies: non-IV midazolam vs IV/rectal diazepam → RR for seizure cessation = 1.52 (95% CI: 1.27–1.82) favoring midazolam
Similar respiratory complication rates (RR 1.49; 95% CI: 0.25–8.72)
Time to seizure cessation: shorter for intranasal midazolam vs IV diazepam in 2 studies
🔹 Clinical Pearl
IM midazolam is preferred first-line when IV access is not established (prehospital, seizing patient). Its superiority in RAMPART was driven by faster drug delivery, not greater pharmacologic potency. IV lorazepam and IV diazepam remain equivalent when IV access is available.
Q4: IV Fosphenytoin vs IV Phenytoin
Insufficient data to compare efficacy of phenytoin vs fosphenytoin (Level U)
Fosphenytoin is better tolerated than phenytoin (Level B)
Phenytoin is an acceptable alternative when fosphenytoin is unavailable (Level B)
Tolerability Comparison (3 Class III RCTs)
Adverse Effect
Fosphenytoin
Phenytoin
Pain at infusion site
17%
Greater (not quantified)
Pruritus at infusion site
48.6%
Lower
Cardiac arrhythmias
None reported
Present
BP changes
13.7 mmHg decrease
5.9 mmHg decrease
Infusion slowed/discontinued
Less frequent
63.6% experienced issues
No fosphenytoin-related cardiac arrhythmias, heart rate changes, respiratory changes, or BP changes in tolerability study
Fosphenytoin can be infused faster (150 mg PE/min vs 50 mg/min for phenytoin)
Q5: When Does Efficacy Drop? (Refractory SE)
VA Cooperative Trial (1998) — Class I
Only class I RCT to address sequential therapy efficacy
570 adults with overt CSE; 4-arm, double-blind, sequential design
If 1st drug failed → randomized to 2nd drug; if 2nd failed → 3rd drug
Therapy Phase
Success Rate
Interpretation
1st anticonvulsant
55.5%
Initial BZD therapy — best chance of success
2nd anticonvulsant
7.0%
Dramatic drop — highlights urgency of adequate 1st-line dosing
3rd anticonvulsant
2.3%
Virtually ineffective; consider anesthetics at this stage
However, these studies were not blinded and initial therapy was not part of the RCT
No clear evidence to guide third-therapy phase (Level U)
Third therapy options: repeat second-line agent, anesthetic doses of thiopental/midazolam/pentobarbital/propofol (all with cEEG)
🔹 Clinical Pearl
The VA Cooperative trial is the only class I study showing sequential efficacy decline: 55.5% → 7.0% → 2.3%. This dramatically demonstrates why adequate first-line BZD dosing is critical — each subsequent drug is substantially less effective. The study also showed that lorazepam was the only agent superior to phenytoin head-to-head (p = 0.001).
Oxygen via nasal cannula/mask; consider intubation if respiratory assistance needed
Initiate ECG monitoring
Finger-stick blood glucose: if <60 mg/dL →
Adults: thiamine 100 mg IV then D50W 50 mL
Children ≥2 years: D25W 2 mL/kg IV
Children <2 years: D12.5W 4 mL/kg
Attempt IV access; collect electrolytes, hematology, toxicology, AED levels
Phase 2: Initial Therapy (5–20 min)
A benzodiazepine is the initial therapy of choice (Level A). Choose ONE of 3 equivalent first-line options:
Agent
Route
Dose
Level
Repeat?
Midazolam
IM
10 mg (>40 kg); 5 mg (13–40 kg)
A
Single dose
Lorazepam
IV
0.1 mg/kg/dose, max 4 mg
A
May repeat ×1
Diazepam
IV
0.15–0.2 mg/kg/dose, max 10 mg
A
May repeat ×1
If the 3 above options are not available, choose one of:
Agent
Route
Dose
Level
Phenobarbital
IV
15 mg/kg, single dose
A
Diazepam
Rectal
0.2–0.5 mg/kg, max 20 mg, single dose
B
Midazolam
Intranasal
Level B
B
Midazolam
Buccal
Level B
B
Phase 3: Second Therapy (20–40 min)
No evidence-based preferred second therapy of choice (Level U). Choose ONE, give as a single dose:
Agent
Dose
Max
Level
Fosphenytoin
20 mg PE/kg
1500 mg PE
U
Valproic acid
40 mg/kg
3000 mg
B (one class II study)
Levetiracetam
60 mg/kg
4500 mg
U
If none of the above are available: IV phenobarbital 15 mg/kg (if not already given) — Level B
ESETT trial (published 2019, after this guideline) later confirmed all three are equivalent (~46–47% efficacy)
Phase 4: Third Therapy (40–60 min)
No clear evidence to guide therapy in this phase (Level U)
Options include:
Repeat second-line therapy
Anesthetic doses of thiopental, midazolam, pentobarbital, or propofol
All with continuous EEG monitoring
Consider escalating directly to anesthetics for severe/prolonged cases
🔹 Clinical Pearl
Key algorithm principles: (1) Give BZDs as a single adequate dose, not multiple smaller doses. (2) All second-line agents are dosed by weight with a maximum cap. (3) The ESETT trial (2019, post-guideline) validated the equivalent efficacy of fosphenytoin, VPA, and LEV as second-line agents at ~46–47% success rate each. (4) There is no evidence to support one third-line agent over another.
