← Back

Chamberlain SE

Lorazepam vs Diazepam for Pediatric Status Epilepticus: A Randomized Clinical Trial

Share Share Copied! Compare

Year of Publication: 2014

Authors: Chamberlain JM, Okada P, Holsti M, ..., Baren J; PECARN

Journal: JAMA

Citation: Chamberlain JM et al. JAMA. 2014;311(16):1652-1660. DOI: 10.1001/jama.2014.2625

Link: https://www.nejm.org/doi/10.1056/NEJMoa002141

PDF: https://www.nejm.org/doi/pdf/10.1056/NEJMoa002141

Clinical Question

Is IV lorazepam superior to IV diazepam for the treatment of pediatric convulsive status epilepticus in the emergency department?

Bottom Line

Lorazepam and diazepam were equally effective for pediatric SE: 72.9% vs 72.1% cessation by 10 minutes without recurrence (absolute difference 0.8%, 95% CI -11.4 to 9.8%). Assisted ventilation rates were also similar (17.6% vs 16.0%). However, lorazepam caused significantly more sedation (66.9% vs 50.0%, difference 16.9%). This trial showed no reason to prefer lorazepam over diazepam in pediatric SE, challenging the assumption that lorazepam is superior.

        Major Points
        Lorazepam and diazepam equally effective for pediatric SE: 72.9% vs 72.1% cessation by 10 min (absolute diff 0.8%; 95% CI -11.4 to 9.8%).
No difference in assisted ventilation: 17.6% vs 16.2% (P=NS).
Lorazepam dose: 0.1 mg/kg IV (max 4 mg). Diazepam dose: 0.2 mg/kg IV (max 8 mg).
273 children randomized (140 LZP, 133 DZP). 9 US pediatric EDs, 2008-2012.
Mean age 3 years. 56% male. 44% with known seizure disorder.
Secondary outcomes: sedation similar (50% vs 48%), recurrence within 4h similar (4.7% vs 3.0%).
Established therapeutic equivalence — either drug acceptable as first-line pediatric SE treatment.
Double-blind, randomized. Published Lancet 2014.

      

Design

Study Type: Randomized, double-blind (quadruple-masked), Phase 2/3 clinical trial

Randomization: 1

Blinding: Double-blind (quadruple-masked)

Enrollment Period: March 1, 2008 to March 14, 2012

Follow-up Duration: 4 hours after study medication

Centers: 11

Countries: United States

Sample Size: 273

Analysis: Intention-to-treat; PECARN network; NCT00621478

Inclusion Criteria

Age 3 months to <18 years.
Convulsive SE defined as continuous seizure ≥5 minutes or ≥3 seizures in 1 hour without return to baseline.
Presented to pediatric ED.
IV access available.

Exclusion Criteria

Prior benzodiazepine within 1 hour.
Known allergy to study drugs.
SE secondary to major trauma.

Baseline Characteristics

Characteristic	Diazepam	Lorazepam
N	140	133

Arms

Field	Control	Lorazepam
Intervention	Diazepam 0.2 mg/kg IV (max 8 mg), with half-dose repeat at 5 minutes if needed; fosphenytoin if SE continued at 12 minutes	Lorazepam 0.1 mg/kg IV (max 4 mg), with half-dose repeat at 5 minutes if needed; fosphenytoin if SE continued at 12 minutes
Duration	Single administration + 4-hour observation	Single administration + 4-hour observation

Outcomes

Outcome	Type	Control	Intervention	P-value
Cessation of SE by 10 minutes without recurrence within 30 minutes (efficacy); assisted ventilation rate (safety)	Primary	Diazepam: 72.1% (101/140) efficacy; 16.0% assisted ventilation	Lorazepam: 72.9% (97/133) efficacy; 17.6% assisted ventilation	Not significant for either
Sedation \| Result: Lorazepam 66.9% vs Diazepam 50.0% (difference 16.9%, 95% CI 6.1-27.7%) -- significantly more sedation with lorazepam	Secondary
Seizure recurrence \| Result: No significant difference	Secondary
Time to seizure cessation \| Result: No significant difference	Secondary
Time to return to baseline mental status \| Result: No significant difference	Secondary
Assisted ventilation	Adverse	Diazepam: 16.0% (26 patients)	Lorazepam: 17.6% (26 patients)
Sedation	Adverse	Diazepam: 50.0%	Lorazepam: 66.9%

Subgroup Analysis

Not reported

Criticisms

Powered for superiority, not equivalence or non-inferiority: negative result does not formally prove equivalence (CI of -11.4 to 9.8% includes clinically meaningful differences)
IV access required for both drugs, may not reflect real-world scenarios with IM/IN/rectal routes
Exception-from-informed-consent cohort introduces ethical and methodological complexities
Sample size of 273 may be insufficient to detect modest but clinically relevant differences (e.g., 10% absolute)
Higher sedation rate with lorazepam (66.9% vs 50%) is clinically relevant but was somewhat underemphasized
Did not assess outcomes beyond 4 hours
Differential rates of rescue medication use (fosphenytoin) not prominently reported

Funding

Multiple NIH grants (U03MC series, NICHD, NHLBI) -- non-industry

Based on: Chamberlain SE (JAMA, 2014)