Summary: Using EEG technology, researchers were able to detect pathological brain activity associated with seizures up to thirty minutes before a seizure occurred in patients with temporal lobe epilepsy. The ability to detect epileptic activity early could allow modulation of affected brain regions as a therapeutic option.
Source: UT Houston
Seizures can be predicted more than 30 minutes before onset in patients with temporal lobe epilepsy, opening the door to therapy using electrodes that could be activated to prevent seizures from occurring, new research from UT Health Houston.
The study, led by Sandipan Pati, MD, associate professor in the Department of Neurology at McGovern Medical School at UTHealth Houston, was recently published in NEJM recordings.
“The ability to predict seizures before they happen is a major advancement in the field of epilepsy research,” said Pati, lead study author and fellow at the Texas Institute for Restorative Neurotechnologies at UTHealth Houston. Neuroscience.
“These results are important because they suggest that we may be able to develop more effective therapies for epilepsy, which could greatly improve the quality of life of patients who suffer from this disease.”
Surgery is a common treatment for many epileptic patients. But when seizures affect larger areas of the brain, surgical removal of part of the brain is not an option. Neuromodulation therapy could offer an alternative solution for patients suffering from these seizures, Pati said.
Previous studies of continuous electroencephalography (EEG) – the measurement and recording of electrical activity in different parts of the brain – have suggested that seizures in people with focal epilepsy tend to occur during periods of increased risk, represented by pathological brain activities called “pro-ictal states”.
EEG-based detection of pro-ictal states is critical to the success of adaptive neuromodulation, with early seizure detection allowing electrodes to be therapeutically applied to the seizure onset area and thalamus of the brain.
To distinguish these pro-ictal states, Pati’s team studied a prospective, consecutive series of 15 patients with temporal lobe epilepsy who underwent limbic thalamic recordings in addition to routine intracranial EEG for localization of seizures. In total, they analyzed 1,800 patient-hours of continuous EEG.

The researchers were able to detect pro-ictal states in patients with temporal lobe epilepsy at least 35 minutes before the onset of the seizure. Pro-ictal states were distinguished at least 45 minutes before seizure onset in 13 of 15 participants. In two of the 15 participants, they were distinguished up to 35 minutes before.
While Pati believes that modulating these brain regions during pro-ictal periods may be an effective therapeutic approach for the treatment of temporal lobe epilepsy, his theory has yet to be tested in clinical trials. However, this information could lead to the development of electrical or drug therapies aimed at preventing seizures.
“This study was made possible through the collaboration of a team of experts in neurology, neurosurgery and neuroscience,” he said. “This underscores the importance of interdisciplinary research to advance our understanding of brain disorders.”
Temporal lobe epilepsy is the most common seizure disorder, affecting some 50 million people worldwide. There are two temporal lobes, one on each side of the head behind the temples.
Mesial temporal lobe epilepsy, which accounts for about 80% of all temporal lobe seizures, involves seizures beginning in or near a part of the brain called the hippocampus, which controls memory and learning. Neocortical or lateral temporal lobe epilepsy involves seizures beginning in the outer part of the temporal lobe.
Omar A. Alamoudi, Ph.D., a postdoctoral researcher in the Department of Neurology at McGovern Medical School at UTHealth Houston, contributed to the study. Other co-authors included Adeel Ilyas, MD, and Kristen O. Riley, MD, both of the University of Alabama at Birmingham.
See also

About this temporal lobe epilepsy research news
Author: Press office
Source: UT Houston
Contact: Press Office – UT Houston
Picture: Image is in public domain
Original research: Free access.
“Pro-ictal state in human temporal lobe epilepsy” by Adeel Ilyas et al. Files HIM
Abstract
Pro-ictal state in human temporal lobe epilepsy
BACKGROUND
Continuous electroencephalography (EEG) studies suggest that seizures in people with focal epilepsy occur preferentially during periods of increased risk, characterized by pathological brain activities, called proictal states; however, the presence of pro-critical (pathological) states among a plethora of otherwise physiological states (e.g., sleep-wake cycle) has not been established.
METHODS
We studied a prospective, consecutive series of 15 patients with temporal lobe epilepsy who underwent limbic thalamic recordings in addition to routine intracranial (cortical) EEG for seizure localization. For each participant, the pro-ictal (45 min before onset of seizure) and interictal (4 hours removed from all seizures) EEG segments were divided into non-overlapping 10-min windows, which were randomly assigned into cohorts training and validation in a 1:1 ratio. A deep neural classifier was applied to distinguish pro-ictal brain activities from interictal brain activities in a patient-specific manner.
RESULTS
We analyzed 1800 patient-hours of continuous thalamocortical EEG. Distinct proictal states were detected in each participant. The median area under the classifier receiver operating characteristic curve was 0.92 (interquartile range, 0.90–0.96). Proictal states were distinguished at least 45 minutes before onset of seizures in 13 of 15 participants; in 2 of the 15 participants, they were distinguished up to 35 minutes before.
CONCLUSION
Based on thalamocortical EEG, pro-ictal states – pathological brain activities during times of increased seizure risk – could be identified in patients with temporal lobe epilepsy and were detected, in our small sample , more than half an hour before the onset of the crisis.