Status epilepticus: faster termination of seizure with phenobarbital.

Clinical bottom line (level 1b)

More patients in status epilepticus who had phenobarbital with optional phenytoin had stopped fitting after 10 minutes than patients on diazepam and phenytoin (NNT = 3 at 10 minutes) . On average patients fitted for four minutes less.
The effect on subsequent intubations, general anaesthetics or side-effects was unclear.

Shaner et al: Neurology 1988; 38: 202-207

Expires September 2003

The study

Unblinded ?concealed randomised trial without intention-to-treat
Setting: emergency department, USA

35 patients (aged mean 50 years, 61% male) with 36 episodes of status epilepticus, defined as:

30 minutes of continuous seizures or recurrent seizures without regaining consciousness in between
uncertain history with three witnessed seizures in department within 5 minutes
3 or more seizures within one hour and obturated before arrival

Excluded if

<15 years old

anticonvulsants given before arrival

Control Group: (n = 18, 18 analysed): diazepam iv at 2 mg/min until patient stopped convulsing or 20 mg reached, and phenytoin simultaneously at 40 mg/min: loading dose of 18 mg/kg over 30 min if unknown levels of 0 to 4 mg/l; otherwise 75% of loading dose if 5 to 9 mg/l; 50% if 10 to 15 mg/l; 25% if 16 to 20 mg/l. If patients continued to fit, had continuous diazepam infusion at 8 mg/hr. If patients still continued to fit, they had general anaesthetic.
Experimental Group: (n = 18, 18 analysed): phenobarbital at 100 mg/min until 10mg/kg given. If patient continued to fit, received phenytoin infusion as above, and simultaneously had iv phenobarbital at 50 mg/min up to a total dose of 20 mg/kg until stopped convulsing. If patient continued to fit, intubated and had iv phenobarbital at 50 mg/min up to a total dose of 30 mg/kg. If patient still continued to fit, had general anaesthetic.
All patients had an oral airway, oxygen at 2l/min, thiamine 100 mg im and 50% glucose infusion. Patients had U&E, Cr, CK, AST, blood count, alcohol levels, anticonvulsants as indicated, arterial blood gases, ECG.
100% followed for 7 hours

The evidence

Outcome	Time to outcome	CER	EER	RRR (95% CI)	ARR (95% CI)	NNT (95% CI)
GA required	7 hours	1 (5.56%)	0 (0.00%)	100% (% to %)	5.56% (-5.03% to 16.1%)	18 (NNT = 6 to infinity; NNH = 20 to infinity)
intubated	7 hours	6 (33.3%)	6 (33.3%)	0% (-152% to 60%)	0.00% (-30.8% to 30.8%)	infinity (NNT = 3 to infinity; NNH = 3 to infinity)
convulsion time 10 minutes or more	10 minutes	8 (44.4%)	2 (11.1%)	75% (-2% to 94%)	33.3% (6.17% to 60.5%)	3 (2 to 16)
arrhythmia or hypotension	7 hours	3 (16.7%)	3 (16.7%)	0% (-331% to 77%)	0.00% (-24.4% to 24.4%)	infinity (NNT = 4 to infinity; NNH = 4 to infinity)

Outcome	Control Group (SD)	Experimental Group (SD)	Mean Difference (95% CI)
total of convulsion time/ minutes	9.0 ()	5.0 ()	-4.0 (-14 to 0)

Presumed cause % (95% CI):

subtherapeutic anticonvulsants: 50% (34% to 66%)
structural lesions: 33% (18% to 49%)
alcohol withdrawal: 28% (13% to 42%)
toxic/ metabolic: 11% (0.8% to 21%)
infections: 5.6% (0.0% to 13%)

Comments

Diazepam can be given rectally unlike phenobarbital - often a more convenient route of administration in an emergency situation.
Is 4 minutes worth the effort of using a less convenient drug?
Study not large enough to comment on effect on GA, arrhythmias or hypotension.

Citation

Shaner DM, McCurdy SA, Herring MO, et al: Treatment of status epilepticus: a prospective comparison of diazepam and phenytoin versus phenobarbital and optional phenytoin. Neurology 1988; 38: 202-207

Contributor: Chris Ball and Clare Wotton, November 2000
Reviewer:

_{Clinical Question.}

_Patient	_{status epilepticus}
_{Intervention or Exposure}	_{phenobarbital}
_Comparison	_{diazepam and phenytoin}
_Outcome	_{length of fit}