Status Epilepticus: Clinical Analysis of a Treatment Protocol Based on Midazolam and Phenytoin

Discussion

This study demonstrates the effective use of midazolam and phenytoin in the treatment of generalized convulsive status epilepticus in children. There are definite benefits to using a uniform treatment protocol. Standardization increases the speed of administration of drugs in this vulnerable patient group. It also facilitates the judgment of the relative contribution of drugs used in multifactorial clinical entities, such as generalized convulsive status epilepticus. Many modern treatment protocols are based on a combination of a benzodiazepine and phenytoin.^{[1,9,15,16,33]} Although there is ample information on the efficacy of individual antiepileptic drugs, little is known about the efficacy of these treatment protocols. In adults, four treatment modalities for generalized convulsive status epilepticus were compared.^[34] In that study, initial treatment with lorazepam was found to be superior to phenytoin alone but was not better than either diazepam or phenobarbital or a combination of diazepam and phenytoin.^[34] However, no data on the relative effect of subsequent medication steps were given. There are no comparative studies on the treatment protocols of generalized convulsive status epilepticus in children. The present study is only the second to evaluate the relative contribution of stepwise administration of antiepileptic drugs in generalized convulsive status epilepticus in children. One earlier retrospective study in 81 children with acute convulsions showed that 69 of 81 patients responded to one or two doses of diazepam.^[35] In 5 of 10 nonresponders, cessation of seizures was obtained with the simultaneous administration of rectal paraldehyde with intravenous phenytoin. The lower benzodiazepine success rate in the present study is possibly due to the fact that only patients with generalized convulsive status epilepticus were included.

Initial treatment of seizures in children still largely consists of rectal diazepam. Although this drug has several disadvantages, it is historically the practical, readily available way of starting treatment by emergency services. Eighty-two of the 122 children in this study thus received initial diazepam rectally. Since no significant relationship between pretreatment with diazepam and the level of antiepileptic therapy was found, diazepam exposure did not seem to influence the effect of further treatment in the study population. These patients did not differ significantly from the rest of the study group in age, cause of status epilepticus, and Pediatric Index of Mortality score (data not shown). It is possible, however, that these 82 patients had more refractory generalized convulsive status epilepticus because convulsions did not cease with initial diazepam therapy. As the initial benzodiazepine in the clinical setting, we chose midazolam because of its better pharmacokinetic properties. It is unclear why midazolam can successfully terminate status epilepticus when diazepam has failed to do so. The effect is possibly dose dependent or related to the benzodiazepine-binding site on the γ-aminobutyric acid (GABA) receptor.^[2,36] In the present study, 48% of the cases had cessation of seizure activity on midazolam bolus alone.

Phenytoin is generally accepted as the second therapeutic agent following a benzodiazepine for children of 1 month and older.^{[1,12,15,16,37]} An additional 15% of patients had cessation of seizure activity after phenytoin was added. This response rate is in concordance with a previous observation that generalized convulsive status epilepticus response to subsequent steps with first-line agents, after the first step has failed, has limited success.^[12]

Fosphenytoin, a new antiepileptic drug that rapidly converts to phenytoin after intravenous or intramuscular administration, is an alternative. It has fewer side effects. Data in children are scarce, and more studies are needed to determine its place in treatment.^{[1,5,12,38,39]} Phenobarbital is advocated by some authors as a second therapeutic drug.^[33] It might be less suitable because of the later onset of action and prolonged sedative effect, which complicates neurologic evaluation. It might also increase the risk of respiratory depression when given immediately after a benzodiazepine.^[15] Besides, the parenteral form of phenobarbital is no longer available in many areas.

Over the last decade, the use of continuous midazolam infusion for refractory generalized convulsive status epilepticus in children has been advocated.^[5,18,30] In the present study, an additional 26% of patients had cessation of seizure activity with continuous midazolam. The mean infusion rate in patients in whom generalized convulsive status epilepticus was controlled was 0.24 mg/kg/hour, comparable to studies in adults^[40] and children,^[5,26,30] but considerably lower than the mean of 0.84 mg/kg/hour found in a pediatric study.^[18] In that study, continuous electroencephalographic (EEG) monitoring was used, and some of the 8 patients studied did not have generalized convulsive status epilepticus. In the present study, a loading dose of midazolam was given before increasing the infusion rate with the assumption that increasing the peak serum concentration would terminate seizures earlier.

There are several alternatives to continuous midazolam for refractory generalized convulsive status epilepticus. Use of high doses of phenobarbital or pentobarbital can be effective.^[15,41] In adults, propofol was as effective as midazolam in treating refractory generalized convulsive status epilepticus.^[42] On the basis of safety concerns, however, several regulatory authorities have banned the use of continuous propofol in children. Although valproate is now available in an intravenous form, its use in generalized convulsive status epilepticus in children has not been studied.

Pentobarbital has emerged as one of the standard choices for refractory generalized convulsive status epilepticus. In this study, in 11% of the cases (13 patients), continuous midazolam failed to stop status epilepticus and barbiturates were used. In our clinical experience, it is unlikely that higher infusion rates of midazolam will terminate seizure activity if repeated loading doses do not at least give temporary cessation of seizure activity. Therefore, barbiturates were sometimes given before the maximum infusion rate of 1 mg/kg/hour was reached. In six patients, the midazolam infusion rate before barbiturate use was 0.9 to 1.0 mg/kg/hour. Five patients had a lower midazolam infusion rate but needed to be intubated, for which a barbiturate was used, and in two patients, repeated midazolam bolus did not have any effect.

We expected to see a relationship between the level of antiepileptic therapy used and the cause of generalized convulsive status epilepticus. Although there was a significantly higher percentage of patients with acute symptomatic generalized convulsive status epilepticus in the barbiturate group, no other association between the cause of generalized convulsive status epilepticus and the level of treatment was found in this study, indicating that underlying illness is not a major factor in the success of antiepileptic therapy.

Fifty-two (43%) patients had to be artificially ventilated, of whom 11 might be related to the administration of continuous midazolam. Most of the patients were intubated electively. This is in contrast to two other studies in which no patient had to be intubated.^[5,30] A larger study in 147 pediatric intensive care unit patients, however, also found a high percentage (58%) of patients requiring endotracheal intubation.^[4] These differences might be partly explained by strategies used for treatment of generalized convulsive status epilepticus. In this patient group, the airway should be protected if the Glasgow Coma Scale score is beneath 8, as suggested by Advanced Pediatric Life Support protocols. Furthermore, owing to the high cerebral metabolism and oxygen consumption during generalized convulsive status epilepticus, adequate oxygenation and ventilation should be guaranteed.^[33] In our view, the threshold for intubation and mechanical ventilation is therefore low.

The mortality rate in our study group was 6%, which is comparable with the 3% to 11% mentioned in the literature for pediatric patients treated for status epilepticus.^{[1,2,3,4,5,6,8,33]} As in other studies, mortality was not attributable to generalized convulsive status epilepticus itself but was dependent on the underlying cause.^[15,43,44]

There are some limitations to this study. First, the exact duration of generalized convulsive status epilepticus, which is a major factor in determining outcome, could not be determined given the retrospective nature of the study. Second, because of the retrospective nature and the lack of a control group, midazolam-related side effects could not be precisely quantified. Third, although all patients on pentobarbital received EEG monitoring, this was not routinely used in all patients to determine cessation of seizures. Since generalized convulsive status epilepticus in adults converts to nonconvulsive status epilepticus in 20% of cases, it is possible that some of our patients had persisting nonconvulsive status epilepticus.^[34] This disturbing finding warrants further EEG controlled trials in the pediatric patient group. In conclusion, this study shows that a stepwise approach combining midazolam and phenytoin for children admitted to a pediatric intensive care unit with generalized convulsive status epilepticus is clinically effective in 89% of patients.

CLICK HERE for subscription information about this journal.

J Child Neurol. 2005;20(6):476-481. © 2005 BC Decker, Inc.

Cite this: Status Epilepticus: Clinical Analysis of a Treatment Protocol Based on Midazolam and Phenytoin - Medscape - Jun 01, 2005.