We fight to improve the lives
of everyone affected by epilepsy

Study helps predict children with difficult-to-treat epilepsy

18 Jun 2001

A new study may help neurologists predict which children with epilepsy will not respond to the most common medications, and thus may be candidates for treatment with more aggressive approaches.

Identifying these children would allow neurologists to determine earlier which children would be most appropriate candidates for treatment with aggressive therapies, such as surgery, recently approved drugs, vagal nerve stimulation, drugs that have higher risks of side effects or the ketogenic diet.

“This could in some cases prevent months and even years of poorly controlled seizures and their consequences,” said study author Anne T. Berg, PhD, of Northern Illinois University in DeKalb.

For the study, 613 children with newly diagnosed epilepsy seen by physicians in Connecticut were followed for at least 18 months and an average of nearly five years. Of those, 10 per cent met the criteria for difficult-to-treat, or intractable, epilepsy. Intractable epilepsy was defined as the failure of two or more first-line epilepsy drugs to control seizures; children had one or more seizures a month for 18 months.

The researchers identified three factors that increase the risk of having intractable epilepsy. Those with a type of epilepsy called cryptogenic/symptomatic generalized epilepsy have the greatest risk. Children with this type of epilepsy are usually young when they have their first seizure, often less than a year old. It can occur in children who were previously neurologically normal, called cryptogenic.

More often it occurs in children who were known to have neurological abnormalities, called symptomatic. In the study, 35 per cent of the children with this type of epilepsy met the criteria for intractable epilepsy, compared to eight per cent of those with other forms of epilepsy.

Also at high risk are those whose seizures occurred frequently when they first developed, or more than once a month. In children with less than one seizure a month, only one per cent met the criteria for intractability. The percentage increased with increasing seizure frequency and reached a maximum of nearly 27 per cent in those with 200 or more seizures per month, Berg said.

"This shouldn't necessarily mean that neurologists should take extraordinary measures with children who initially have a high seizure frequency, but these children should be observed and followed closely," Berg said.

The researchers also measured the electrical activity in the children's brains through electroencephalogram (EEG) tests. Children with slow activity in an area of the brain were also more likely to have intractable epilepsy than children with normal EEGs or with other EEG abnormalities common in epilepsy.

These last two factors – initial seizure frequency and EEG activity – are also important in children with a type of epilepsy called non-idiopathic partial epilepsy, who represented almost half of the study group. Partial epilepsy means that the seizure occurs in one area of the brain. “This is important because this is the type of epilepsy most frequently treated with surgery if medications fail,” Berg said.

Holmes states that large studies are needed to compare and evaluate several new epilepsy drugs and surgical techniques. "This study provides solid guidelines for researchers to evaluate the medical, dietary or surgical treatment for children who are destined to have a difficult course," he said. "Identifying the children is the first step. Now we need to figure out what the best treatments are for them."

The study was published in the journal Neurology.