High Spatial Resolution and Expanded Frequencies: The Search for Ictogenesis

Understanding the where and when of seizure initiation (ictogenesis) is as important to epilepsy research and patient care as discovering the why. A single neuron cannot have a seizure; a seizure requires populations of neurons firing in synchrony. The question asked by two Mayo Clinic neurologists is how many neurons does it take—what is the smallest anatomic unit that gives rise to a seizure? The smallest unit could translate into the earliest possible moment of detection, and that could be the key to seizure suppression.

High Spatial Resolution: Microelectrodes and Microseizures

The smallest functional unit of the cortex is the cortical column. Identified in the 1950s by Vernon B. Mountcastle, cortical columns are stacked arrangements of cells, each dedicated to a specific function. Each column comprises approximately 1,000 to 7,500 neurons that serve as their own network. It is cortical columns that Squire M. (Matt) Stead, MD, PhD and Gregory A. Worrell, MD, PhD neurologists at Mayo Clinic in Rochester, Minnesota chose as the focus of their research into ictogenesis.

Ictogenesis (seizure initiation)

Enlarge

Cortical columns measure just 300 microns across. To understand how small that is, it is worth noting that the standard clinical intracranial EEG electrode records from an area approximately 20 mm square. Each EEG electrode captures the activity of millions of neurons and hundreds of cortical columns. Using microelectrodes just 40 microns in diameter, thinner than human hair, Drs Stead and Worrell are able to record brain activity in humans from individual cortical columns.

Their eureka moment came when they found evidence of seizure activity in a given column, while, as Dr Stead explains, "a millimeter away in any direction, in all the microelectrodes surrounding that column, it was quiet. There was no seizure activity." These "microseizures" were the first evidence that seizure activity occurs in the smallest unit of functional brain organization.

Microseizures are subclinical, and Drs Stead and Worrell have recorded them in the brains of people with and without epilepsy. Their hypothesis predicts that as a clinical seizure approaches, individual columns have more frequent seizures and recruit other columns of neurons. "The occasional synchronous oscillation within a column may be a normal phenomenon," explains Dr Stead, "but we might be looking at the basis of the seizure threshold. In people with epilepsy, whose seizure thresholds are lower, the frequency and density of these microseizures may be greater."

Another critical factor is timing. When do seizures start? "What we've found is that hours before a clinical seizure occurs, there is increased columnar activity," says Dr Worrell. "Capturing this increased activity may translate eventually into a brain stimulation device that could identify microseizure activity, long before a clinical seizure begins." The earlier seizure activity is detected and the more precise the location, the better the chances are to prevent or abort it.

Expanded Frequencies

Sometimes old ways persist. When brain waves were first recorded for clinical purposes, the frequency range was set at 1 to 70 Hz because those frequencies represented the mechanical and spatial limits of the recording pens. Today, despite the advent of computerized technology, clinical EEG recording frequencies remain at 1 to 70 Hz.

Drs Stead and Worrell theorized that ultraslow or extremely fast oscillations up to 1,000 Hz might be relevant to ictogenesis and designed their recording equipment accordingly. They have now demonstrated that both high- and low-frequency oscillations are clinically relevant signatures or biomarkers of the epileptogenic zone, the region surrounding the area from which seizures arise.

Implantation and EEG Analysis

W. Richard Marsh, MD one of the neurosurgeons at Mayo Clinic in Minnesota who performs epilepsy surgery and stimulator implantation, explains that patients who are undergoing intracranial monitoring using depth electrodes are given the option of including microelectrodes for research purposes along with the standard leads. Recording from microelectrodes on the surface of the brain is under development. "The techniques for electrode implantation are well established," Dr Marsh explains. "There's nothing innovative about it. The innovative part is the recording—both the frequencies recorded and the narrowness of the sites. There is substantial computational power behind the analysis because of the tremendous volume of data—terabytes of data—beyond the human capacity to analyze. The research team has developed programs to make sense of it."

From Theory to Practice

The Next Generation of EEG Machines
Drs Stead, Worrell, and colleagues have been awarded a Mayo Discovery Translational grant. Combined with funding from the Epilepsy Foundation of America, it will be used to build the first prototype of a new multiscale EEG machine—the next-generation EEG recording system. The new system will improve sensitivity through increased spatial sampling (microelectrode arrays) and wide frequency bandwidth recording.

Pinpointing the Epileptogenic Zone
Microseizure recording might also aid in a more precise delineation of the epileptogenic zone. It is standard practice during epilepsy surgery to remove the epileptogenic zone. Identifying columnar seizure activity may more precisely establish and individualize its margins.

Implantable Devices
Dr Worrell notes, "This work is motivated by the fact that many patients with epilepsy do not have successful treatment and require surgery or an implantable stimulation device." Patients with intractable epilepsy might only have 4 or 5 seizures a month, each lasting a minute or so. But, as he explains, "those 4 minutes mean they can't drive, can't swim, and may be afraid to go out in public for fear of becoming unconscious or incontinent and could even end up socially isolated. The importance of those 4 minutes is their unpredictability."

Because microseizures may occur hours before a clinical seizure in patients with epilepsy, a possible application of this research is an implantable device that would detect subclinical microseizure activity and use continuous feedback stimulation to prevent microseizures from turning into large-scale clinical seizures. Another advantage would be the ability to alert patients not only of seizure onset but also of periods when they will be seizure free.

Microrecording sites and broader recording frequencies are improving understanding of the where and when of ictogenesis. "If the hypothesis that seizures in patients with epilepsy stem from cortical columns proves true, the practical applications will follow," says Dr Marsh.