Pauline Anderson

December 12, 2013

WASHINGTON, DC — Researchers have developed a small, light-weight seizure detector that patients can wear at home. The technology appears to match results with video electroencephalography (EEG), the "gold standard" for defining generalized tonic-clonic (GTC) seizures.

Worn on the upper arm, the device constantly monitors electrical activity of arm muscles with the aim of picking up specific electromyography (EMG) patterns typical of GTC seizures. The power generated by a GTC seizure at high frequencies is quite different from that generated during normal daily activities or in association with other seizure types, the researchers say.

GTC seizures are the most common type of seizure, with a 2012 estimate of 759,000 Americans being affected. This type of seizure is also the most severe, associated with trauma, fractures, and sudden unexplained death in epilepsy (SUDEP), said C. Akos Szabo, professor, neurology, University of Texas Health Science Center, San Antonio.

Dr. Szabo and colleagues at the South Texas Comprehensive Epilepsy Center carried out a study comparing the efficacy of a computer algorithm that assesses EMG signals with video EEG monitoring, which is used to record seizures when medications are being reduced or withdrawn.

Seizure detector device

Their results to date were presented here at the American Epilepsy Society (AES) 67th Annual Meeting.

Constant Monitoring

The fear of SUDEP is always a worry for patients with epilepsy; almost a quarter of deaths in people with epilepsy are thought to be attributable to SUDEP. Parents of children with epilepsy are often reluctant to leave them alone for fear they will have a seizure with no help readily available.

The current study included 33 consecutive patients admitted to the epilepsy unit with a history of GTC seizures not responsive to medications. Participants ranged in age from 14 to 64 years.

Although the patients were in the video EEG monitoring unit, electrodes were attached to their triceps and biceps. Researchers obtained a baseline measure of muscle activity and then constantly recorded the EMG signals. For each patient, researchers recorded an average of 42.4 hours, which provided over 1400 hours of surface EMG data.

They then used the seizure detection software algorithm that was developed by Brain Sentinel, a start-up company based in San Antonio, to analyze the EMG data. The algorithm continuously compares recorded surface EMG signals to the baseline sample of muscle activity. (GTC seizures are signalled with sustained activation of multiple frequency bands [30 to 40 Hz, 130 to 240 Hz, 300 to 400 Hz] of EMG activity.)

The team compared the time a seizure was detected with the computer program with the time that 3 independent board-certified neurophysiologists had identified the seizures occurring using the video EEG.

"We would run the data through the computer program, and it would tell us when the seizures were occurring in time, and this was compared to the video EEG recording," explained Michael Girouard, president and CEO of Brain Sentinel.

Dr. C. Akos Szabo and Michael Girouard

In this analysis, 197 seizures were recorded. Eleven patients had a total of 22 GTC seizures.

The EMG algorithm detected all but 1 of the GTC seizures. "We were pleased to detect 21 of the 22 generalized tonic-clonic seizures that occurred in the 33 subjects," said Dr. Szabo. "Even more importantly, these were detected within 30 seconds of the initial arm contraction, so it was very rapidly detected, even before the full seizure was completed."

Dr. Szabo noted that even though the patients were moving around somewhat, brushing their teeth and eating meals, there was only 1 false-positive detection with use of the algorithm. "This shows that it is a very specific algorithm for generalized tonic-clonic seizures," he said.

"Based on these results, this is a very sensitive and very specific algorithm, or computer program, that detects generalized tonic-clonic seizures, and it appears to be reliable."

The technology has been put into an actual device that is simple to use and easy to wear. When a patient has it on, the EMG signals are transmitted to a base station and stored remotely. When distinct patterns are detected, the device sets off an alarm, and appropriate personnel are alerted.

Quick Treatment

"By detecting the generalized tonic-clonic seizures, we hope that any caregiver that is in the vicinity of the patient will be able to go to patient and provide first aid, be able to deliver treatment, hopefully very quickly, and not only stop the seizure but prevent further seizures, even status epilepticus, from occurring," said Dr. Szabo.

Even if a patient goes into cardiac arrest, emergency medical personnel should be alerted in time for the patient to be resuscitated.

The technology needs to be tested outside of the video EEG monitoring setting "in situations where patients are doing other kinds of physical activities, which may involve more muscle movement or activation," said Dr. Szabo.

Such testing will take place at 9 centers across the United States, said Girouard. Researchers plan to enroll 100 patients into an open-label pivotal phase 3 trial of the device.

About 55,000 people die of complications of epilepsy every year. This, noted Girouard, is more than the number who die of colon and rectal cancer in the United States every year. If the device gets US Food and Drug Administration clearance, it will be the first of its kind to be approved in the United States, he said.

The technology is promising and has several advantages, including being easy to use and noninvasive, involving a single-modality algorithm, and potentially being relatively affordable, according to Elson L. So, MD, professor of neurology and director of the Section of Electroencephalography, Mayo Clinic College of Medicine, Rochester, Minnesota, and newly elected President of the American Epilepsy Society.

Asked to comment on this research, Dr. So said it appears to be accurate in determining the frequency of convulsive seizures in the epilepsy monitoring unit setting.

However, the current findings need to be validated in blinded and controlled studies that include more patients and are carried out "in the life environment," outside the controlled setting of an epilepsy monitoring unit.

Dr. So added that the value of the device in preventing SUDEP remains to be seen. "SUDEP prevention may also depend on other factors, such as the availability of intervention measures," he said. Age-specific data would also be useful, he added.

Home Monitoring Device for AEDs

Another home device, one that monitors levels of antiepileptic drugs (AEDs), is also under development.

Dr. Nick J. Christodoulides

In a separate presentation at the AES meeting here, researchers provided results of a pilot study of a saliva-based bio-nanochip assay system designed to detect phenytoin and phenobarbital, 2 commonly used AEDs.

The idea is that patients will eventually be able to take a mouth swab and, using a portable, hand-held device the size of a credit card, test drug levels at home, said Nick J. Christodoulides, PhD, senior scientist, McDevitt Laboratory, Rice University, Houston, Texas.

The study was funded by Brain Sentinel. Dr. Szabo has no financial interest in the company and reports no relevant financial relationships. Mr. Girouard is President and CEO of Brain Sentinel.

American Epilepsy Society (AES) 67th Annual Meeting. Abstract 2.037; Platform A.09. Presented December 9, 2013.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.