Predicting seizures in children during pre-surgical monitoring

A recent study shows how seizures are predicted in children during pre-surgical monitoring.

Long–Term–Monitoring (LTM) is a valuable tool for seizure localization/lateralization among children with refractory–epilepsy undergoing pre–surgical–monitoring. The aim of this study was to examine the factors predicting occurrence of single/multiple seizures in children undergoing pre–surgical monitoring in the LTM unit. Majority of the admissions (92%) admitted to the LTM–unit for pre–surgical workup had at–least one seizure during a mean length of stay of 5.24 days. Home seizure–frequency was the only predictor influencing occurrence of single/multiple seizures in the LTM unit. Patients with low seizure–frequency are at risk for completing the monitoring with less than the optimum number (<3) of seizures captured.

Methods

• Chart review was done on 95 consecutive admissions on 92 children (40 females) admitted to the LTM-unit for pre-surgical workup.
• Relationship between occurrence of multiple (≥3) seizures and factors such as home seizure-frequency, demographics, MRI-lesions/seizure-type and localization/AED usage/neurological-exam/epilepsy-duration was evaluated by logistic-regression and survival-analysis.
• Home seizure-frequency was further categorized into low (up-to 1/month), medium (up-to 1/week) and high (>1/week) and relationship of these categories to the occurrence of multiple seizures was evaluated.
• Mean length of stay was 5.24 days in all 3 groups.

Results

• Home seizure frequency was the only factor predicting the occurrence of single/multiple seizures in children undergoing presurgical workup.
• Other factors (age/sex/MRI-lesions/seizure-type and localization/AED-usage/neurological-exam/epilepsy-duration) did not affect occurrence of single/multiple seizures or time-to-occurrence of first/second seizure.
• Analysis of the home-seizure frequency categories revealed that 98% admissions in high-frequency, 94% in the medium, and 77% in low-frequency group had at-least 1 seizure recorded during the monitoring.
• Odds of first-seizure increased in high vs. low-frequency group (p=0.01).
• Eighty-nine percent admissions in high-frequency, 78% in medium frequency, versus 50% in low-frequency group had ≥3 seizures.
• The odds of having ≥3 seizures increased in high-frequency (p=0.0005) and in medium-frequency (p=0.007), compared to low-frequency group.
• Mean time-to-first-seizure was 2.7 days in low-frequency, 2.1 days in medium, and 2 days in high-frequency group.
• Time-to-first-seizure in high and medium-frequency was less than in low-frequency group (p<0.0014 and p=0.038).

Read more here

Epileptic seizures can be predicted by device implanted in brain

A small implant in the brain correctly predicts when an epileptic seizure will occur.

A small device implanted in the brain has accurately predicted epilepsy seizures in humans in  a world-first study led by Professor Mark Cook, Chair of Medicine at the University of Melbourne and Director of Neurology at St Vincent’s Hospital.

“Knowing when a seizure might happen could dramatically improve the quality of life and independence of people with epilepsy,” said Professor Cook, whose research was today published in the international medical journal, Lancet Neurology.

Professor Cook and his team, with Professors Terry O’Brien and Sam Berkovic, worked with researchers at Seattle-based company, NeuroVista, who developed a device which could be implanted between the skull and brain surface to monitor long-term electrical signals in the brain (EEG data). 

They worked together to develop a second device implanted under the chest, which transmitted electrodes recorded in the brain to a hand-held device, providing a series of lights warning patients of the high (red), moderate (white), or low (blue), likelihood of having a seizure in the hours ahead.

The two year study included 15 people with epilepsy aged between 20 and 62 years, who experienced between two and 12 seizures per month and had not had their seizures controlled with existing treatments.

For the first month of the trial the system was set purely to record EEG data, which allowed Professor Cook and his team to construct individual algorithms of seizure prediction for each patient.

The system correctly predicted seizures with a high warning, 65 percent of the time, and worked to a level better than 50 percent in 11 of the 15 patients. Eight of the 11 patients had their seizures accurately predicted between 56 and 100 percent of the time.

Epilepsy is the second most common neurological disease after stroke, affecting over 60 million people worldwide. Up to 40 percent of people are unable to control their seizures with existing treatments.

“One to two percent of the population have chronic epilepsy and up to 10 percent of people will have a seizure at some point in their lives, so it’s very common. It’s debilitating because it affects young people predominantly and it affects them often across their entire lifespan,” Professor Cook said.

“The problem is that people with epilepsy are, for the most part, otherwise extremely well. So their activities are limited entirely by this condition, which might affect only a few minutes of every year of their life, and yet have catastrophic consequences like falls, burns and drowning.”

Professor Cook hopes to replicate the findings of the study in larger clinical trials, and is optimistic the technology will lead to improved management strategies for epilepsy in the future.

Read more here