RAMPART

Published

Silbergleit R, et al. "Intramuscular versus Intravenous Therapy for Prehospital Status Epilepticus". The New England Journal of Medicine. 2012. 366(7):591-600.
PubMed • Full text • PDF

Clinical Question

For prehospitalized patients in status epilepticus, how effective is intramuscular midazolam compared with intravenous lorazepam for seizure termination?

Bottom Line

Intramuscular midazolam is superior to intravenous lorazepam for prehospital seizure cessation.

Major Points

The current definition of status epilepticus is more than 5 minutes of continuous seizure activity or recurrent seizure activity within this period. This definition is based on the knowledge that seizures extending past 5 minutes are much less likely to spontaneously terminate and animal studies suggested that permanent neuronal injury and resistance to treatment may set in between 5-30 minutes. Rapid termination of seizure activity is therefore paramount, and with multiple antiepileptic agents available, randomized studies were needed to determine the most efficacious approach. The Veterans Affairs Status Epilepticus Cooperative Study Group showed better seizure control with lorazepam compared to phenytoin as an initial agent in the context of overt generalized convulsive status epilepticus. The subsequent PHTSE trial comparing diazepam, lorazepam, and placebo failed to find a significant difference between the two benzodiazepines, although there was a trend toward better termination with lorazepam. When intramuscular (IM) formulations of midazolam became readily available, these gained significant popularity with paramedics due to their convenience and ease of use. Also, unlike lorazepam, midazolam has favorable stability and shelf-life parameters when not refrigerated. A randomized study comparing midazolam and lorazepam was needed to demonstrate comparative efficacy between the two approaches.

Published in 2012, the Rapid Anticonvulsant Medication Prior to Arrival Trial (RAMPART) trial enrolled over 1,000 pediatric and adult patients with out-of-hospital status epilepticus requiring benzodiazepines. In a protocol that waived informed consent requirements due to the emergency circumstances, patients were randomized to either IM midazolam or intravenous (IV) lorazepam, which was given by emergency medical services (EMS) personnel. The primary outcome was cessation of seizure activity prior to arrival in the emergency department without requiring rescue therapy, estimating a 70% rate of seizure termination and assigning a noninferiority margin of 10%. In the intention-to-treat analysis of 893 patients, the primary outcome occurred in 73.4% of patients receiving midazolam compared to 63.4% of patients receiving lorazepam (P<0.001 for superiority). Treatment in the midazolam arm was associated with fewer hospitalizations, fewer ICU admissions, shorter ICU and hospital lengths of stay. There were no differences in the rates of recurrent seizure after ED arrival (10-11%), or need for endotracheal intubation within 30 minutes of ED arrival (14% in both arms). The authors conclude that prehospital midazolam is noninferior and perhaps slightly superior to lorazepam for the prehospital treatment of status epilepticus. Their analyses of temporal data suggest that the advantage is in part attributed to the ability to more rapidly administer an intramuscular medication than one by the intravenous route.

Guidelines

NCS Guidelines for the Evaluation and Management of Status Epilepticus (2012, adapted)

• Lorazepam and midazolam are class IA agents for emergent management of status epilepticus

Design

• Multicenter, randomized, double-blind, phase 3, noninferiority clinical trial.
• N=893 patients with out-of-hospital status epilepticus
  • Midazolam IM (n=448)
  • Lorazepam IV (n=445)
• Setting: 33 EMS agencies and 79 receiving US hospitals
• Enrollment: 2009-2011
• Analysis: Intention-to-treat
• Primary outcome: Termination of seizure without need for rescue therapy by arrival in ED

Population

Inclusion Criteria

• Paramedics or reliable witnesses verify 5 minutes of either continuous seizure activity or of repeated convulsive seizure activity where the patient does not regain consciousness (operationally defined as meaningful speech or obeying commands) between seizures.
• Patient is still seizing at the time of paramedic treatment with study medications
• Estimated weight ≥13 kg
• Subject to be transported to a RAMPART participating hospital

Exclusion Criteria

• Major trauma as the precipitant of the seizure
• Hypoglycemia (as defined by local EMS protocol or a glucose <60 mg/dL)
• Known allergy to midazolam or lorazepam
• Cardiac arrest or heart rate (HR) <40 beats per minute
• Sensitivity to benzodiazepines
• Medical alert tag marked with "RAMPART declined"
• Prior treatment of this seizure with diazepam autoinjector as part of another study
• Known pregnancy
• Prisoners

Baseline Characteristics

From the midazolam group.

• Mean age: 43 years
• Female sex: 44%
• Race: Black 51%, White 37%, Other 12%
• Dose tier: Low 14%, high 86%
• History of epilepsy: Yes 65%, no 25%, unknown 10%
• Final diagnosis: status epilepticus 90%, nonepileptic spell 7%, undetermined 3%
• Precipitating cause:
  • Non-receipt of anticonvulsant therapy 31%
  • Breakthrough seizure 27%
  • Coexisting condition that lowered seizure threshold 7%

Interventions

• Randomization to
  • Midazolam IM (5 mg for patients 13-40 kg, and 10 mg for patients >40 kg)
  • Lorazepam 4 mg IV
• All patients received a matching IM or IV placebo
• Attempts at IV access were made for up to 10 mins.
• Intraosseous access was permitted at any time and was considered equivalent to IV.
• Rescue therapy was recommended if the patient was still convulsing 10 mins after the last study medication.

Outcomes

Comparisons are midazolam vs. lorazepam unless otherwise specified.

Primary Outcome

Cessation of seizure without rescue therapy

73.4% vs. 63.4% (difference 10%; 95% CI 4.0-16.1; P<0.001 for noninferiority and P<0.001 for superiority)

Secondary Outcomes

Endotracheal intubation within 30 mins of ED arrival

14.1% vs. 14.4% (RR 0.98; 95% CI 0.70–1.34)

Hospitalization

57.6% vs. 65.6% (RR 0.88; 95% CI 0.79–0.98)

ICU admission

28.6% vs. 36.2% (RR 0.79; 95% CI 0.65–0.95)

Recurrent seizure within 12 hours of ED arrival

11.4% vs. 10.6% (RR 1.08; 95% CI 0.74–1.56)

Hypotension requiring vasopressor infusion

2.7% vs. 2.9% (RR 0.92; 95% CI 0.42–1.98)

IM injection site complications

4 vs. 2 (RR 1.99; 95% CI 0.30–10.70)

IV injection site complications

0 vs. 3

ICU length of stay

5.7 vs. 4.1 days

Hospital length of stay

6.7 vs. 5.5 days

Median time to administration

1.2 vs. 4.8 mins

Onset of action (termination of convulsions)

3.3 vs. 1.6 mins

Criticisms

• Due to study design, unable to conclude if superior seizure termination was due to IM route of administration vs. IV route or due to superiority of midazolam as an agent to lorazepam.
• An autoinjector was used in this study and results may be not fully translatable to conventional intramuscular injections

Funding

• The National Institute of Neurological Disorders and Stroke
• IM injectors were provided by the U.S. Department of Defense