Rethinking Seizure Care Blog

Web Based Model Could Help to Predict Epilepsy Risk

Posted by RSC Diagnostics on Feb 12, 2019

 Research - Wordcloud Concept. The Word in Red Color, Surrounded by a Cloud of Blue Words.

Kids who have suffered at least one seizure may benefit from this model, as it can help to determine the amount of risk the child has of being diagnosed with epilepsy. This study, published in Pediatrics, explains the testing for the model and how it was developed.

The Need for Research

In children, it is often common for epileptic seizures to be underestimated because of the variety of clinical symptoms  that children might present. The symptoms can vary widely, which means that it might not be immediately apparent that the child is suffering from epilepsy. In addition, the “initial interictal electroencephalogram (EEG) might have limited sensitivity.” More research was needed to help provide doctors with better and more accurate prediction methods for their young patients.

How Did the Study Work?

Researchers behind the study say earlier efforts to “identify prognostic clinical variables for seizure recurrence after first consultation” were not as effective, since the previous studies were only done on kids who had already been diagnosed with epilepsy. The prediction model was developed to help clinicians who are providing a workup for patients and who had access to an EEG.

The researchers created their model using retrospective data from 451 children. This information included clinical data, as well as the initial EEG findings. The children who were a part of the study were followed for a minimum of one year and they had either been diagnosed as having epilepsy or were unconfirmed. To validate the model, researchers then used case data from another 187 children.

The report claims that 45% of the kids “in the training cohort and 29% in the validation cohort had inconclusive epilepsy diagnoses at the start of the study.” After a year of follow-up with the patients, they were able to make definitive diagnoses in 94.2% of the children who were a part of the training cohort and 92% that were in the validation cohort.

They found that the modeling efforts they employed turned out to be quite successful and they would be able to be used for screening.

How Does the Model Work?

The model that was created was meant to provide a “rational approach to assist clinicians during the diagnostic process by combining routinely available clinical information in a multivariate way.” They believe that the model can make for a tool that will work well for independent screening. These findings should help clinicians to have a better idea of the likelihood of a seizure to be epileptic in its origin, which should help the professionals to make better and more informed decisions on the further treatment of the patient.

The team has even made their model available through an Internet application, which can help the clinicians to more easily integrate the model into their practice.

The model could end up being helpful to those children who have uncertain epilepsy diagnoses. It allows for high-risk cases to be identified in less time and with fewer misdiagnoses. Ultimately, this should help to improve the outcome for all patients, ensuring those who need it are more likely to get the specialist help they require.


Topics: Pediatric Epilepsy