The ability to detect seizure activity early may allow modulation of affected brain regions as a therapeutic option. Using electroencephalogram (EEG) technology, researchers have been able to detect pathological brain activity associated with seizures up to thirty minutes before a seizure occurs in patients with temporal lobe epilepsy.
Worldwide, more than 50 million people live with epilepsy, a brain disorder that makes them vulnerable to seizures. These seizures can cause physical damage, profoundly affect the quality of life, and increase the risk of death.
For those with epilepsy, the occurrence of seizures is often unpredictable, occurring seemingly randomly, as if they were “a bolt from the blue.” This unpredictability is a major limitation to patients’ daily lives.
In medical and scientific practice, episodes of brain activity in epilepsy are generally classified into three main stages: the interictal state (the period between seizures), the ictal state (when the seizure occurs), and the postictal state (immediately after the seizure).
However, recent studies using continuous electroencephalograms (EEG), which monitor the electrical activity of the brain, suggest that there is greater complexity to these states.
It was possible to identify periods in which the risk of a new seizure increases, subdividing the interictal state into specific states called pre-ictal and pro-ictal.
The pre-ictal state is what occurs a few seconds or minutes before the seizure, and is a type of immediate “warning” that the seizure is about to begin.
The pro-ictal state, on the other hand, represents a longer period, with subtle changes in brain activity that increase the likelihood of a seizure. These pro-ictal states, which can last many minutes or even hours, indicate a period in which the brain is more susceptible to having a seizure, even if it does not occur immediately.
It is still unclear how pro-ictal states differ from normal brain states, such as during sleep or while we are awake and alert. The lack of understanding of the exact mechanisms that create these pro-ictal states limits the development of effective treatments to prevent seizures.
Temporal lobe epilepsy is one of the most common and persistent types of focal epilepsy (when seizures begin in a specific area of the brain). This type of epilepsy is often treated with surgeries, such as microsurgery, laser ablation, or neuromodulation (electrical stimulation to control seizures).
Unfortunately, less than 50% of patients who undergo surgery achieve a lasting reduction in seizures. Epilepsy is increasingly being understood as a problem that affects networks of interconnected brain structures, rather than just one specific area of the brain.
Research suggests that abnormal changes in these networks, including areas such as the limbic thalamus, are key to the development of seizures.
The limbic thalamus is a region of the brain that helps control large networks of communication between temporal and extratemporal areas, regulates wakefulness and may be a promising target for neuromodulatory interventions in temporal lobe epilepsy.
Currently, attempts to predict seizures using EEG have focused primarily on activity in the cerebral cortex (the outer layer of the brain) and seek to identify seizures within minutes of onset.
However, researchers at UTHealth Houston have hypothesized that these pro-ictal states, characterized by abnormal brain activity and elevated seizure risk, may be identified long before seizure onset, possibly up to hours in advance.
These researchers suggested that by monitoring EEG that includes both the cortex and the limbic thalamus (known as thalamocortical EEG), it might be possible to detect these pro-ictal states with greater accuracy.
They conducted a study with 15 patients with temporal lobe epilepsy, who underwent EEG recordings of the limbic thalamus and cortex to localize seizures.
For each patient, EEG segments preceding seizures (up to 45 minutes before) and distant interictal moments (at least 4 hours away from any seizure) were analyzed. These segments were divided into 10-minute windows and randomly separated into two groups: one for training the analysis model and the other for validation.
A deep neural classifier (a machine learning tool) was used to distinguish brain activities associated with preictal from interictal states, in a personalized way for each patient. More than 1800 hours of continuous thalamocortical EEG were analyzed.
The results showed that each patient had identifiable pro-ictal states. The accuracy of the model was high, with a median area under the curve (AUC) of 0.92, indicating a strong ability to predict seizure risk.
In 13 of the 15 patients, pro-ictal states could be detected at least 45 minutes in advance, and in 2 patients, these states were detected 35 minutes in advance.
Fluctuations in Π and seizure events were observed during admission assessments. Plot of the classifier output (Π, blue lines) of the total electroencephalography (EEG) data obtained during the admission assessment for each participant. Seizure events are indicated by red lines. Missing EEG data are indicated by gray patches. Seizures appear to occur after increases in Π and cluster during periods of high Π values. Conversely, seizure occurrence is sparse during periods of low Π values. DOI: 10.1056/EVIDoa2200187
These findings suggest that, based on thalamocortical EEG, it is possible to identify brain activities that indicate a high risk of seizure in patients with temporal lobe epilepsy, and that these states can be detected more than half an hour before seizures.
The study concludes that the existence of these "non-physiological" brain states during periods of high seizure risk suggests that adaptive neuromodulation therapies, that is, interventions to modulate brain activity adaptively, could be applied at clinically significant times, with the potential to prevent seizures from developing.
READ MORE:
Pro-Ictal State in Human Temporal Lobe Epilepsy
Adeel Ilyas, Omar A. Alamoudi, Kristen O. Riley, and Sandipan Pati
NEJM Evid 2023; 2 (3)
DOI: 10.1056/EVIDoa2200187
Abstract:
Studies of continuous electroencephalography (EEG) suggest that seizures in individuals with focal-onset epilepsies preferentially occur during periods of heightened risk, typified by pathologic brain activities, termed pro-ictal states; however, the presence of (pathologic) pro-ictal states among a plethora of otherwise physiologic (e.g., sleep–wake cycle) states has not been established. We studied a prospective, consecutive series of 15 patients with temporal lobe epilepsy who underwent limbic thalamic recordings in addition to routine (cortical) intracranial EEG for seizure localization. For each participant, pro-ictal (45 minutes before seizure onset) and interictal (4 hours removed from all seizures) EEG segments were divided into 10-minute, nonoverlapping windows, which were randomly distributed into training and validation cohorts in a 1:1 ratio. A deep neural classifier was applied to distinguish pro-ictal from interictal brain activities in a patient-specific fashion. We analyzed 1800 patient-hours of continuous thalamocortical EEG. Distinct pro-ictal states were detected in each participant. The median area under the receiver-operating characteristic curve of the classifier was 0.92 (interquartile range, 0.90–0.96). Pro-ictal states were distinguished at least 45 minutes before seizure onset in 13 of 15 participants; in 2 of 15 participants, they were distinguished up to 35 minutes prior.
On the basis of thalamocortical EEG, pro-ictal states — pathologic brain activities during periods of heightened seizure risk — could be identified in patients with temporal lobe epilepsy and were detected, in our small sample, more than one half hour before seizure onset.
Comments