A systematic literature search and review of the best evidence for intensive care treatment of refractory status epilepticus in children using continuous infusion of midazolam or anesthetic agents.
MEDLINE and EMBASE search before December 2013 using key words and/or Medical Subject Headings identified English-language citations that were screened for eligibility and used if 1) the study was about high-dose benzodiazepine or anesthetic agent for children; 2) the treatment protocol was described and used for refractory status epilepticus; 3) the outcomes included seizure control; and 4) the series included at least five children.
Sixteen studies (645 patients) were identified, including midazolam (nine studies), barbiturate (four studies), and other anesthetic approaches (three studies). When midazolam was used as the initial agent for refractory status epilepticus, the rate of clinical seizure control was 76%, which was achieved on average 41 minutes after starting the infusion. When midazolam was used in conjunction with continuous electroencephalography, the time to seizure control was much longer and the mean dose required for seizure control was 10.7 μg/kg/min compared with a lower dose (2.8 μg/kg/min) in the studies not using this form of monitoring, suggesting that continuous electroencephalography provided additional targets for treatment. Barbiturates were usually used after midazolam failed and treatment was started, on average, 66 hours after refractory status epilepticus onset with the goal of electroencephalography burst suppression, which was achieved, on average, 22.6 hours later. Among patients failing midazolam, barbiturate infusion was effective in 65%. Inhaled anesthetics, ketamine, and hypothermia were generally used after prior therapy with midazolam and barbiturates had failed, usually several days after seizure onset.
The data on intensive care treatment of pediatric refractory status epilepticus are of poor quality, yet they show a hierarchy in strategies: early midazolam, then barbiturates, and then trial of other anesthetic strategies. In addition, using a solely clinical endpoint for seizure control may be missing significant seizure burden in pediatric refractory status epilepticus.