Electroencephalography (EEG) and Deep Brain Stimulation (DBS) in Epilepsy

Study Purpose:

The purpose of this study is to assess EEG changes during electrical stimulation of the
thalamus to treat people with epilepsy.

Background on stimulation of the anterior nucleus of the thalamus for epilepsy:

Of the estimated 4 million people in the U.S. and Europe with epilepsy, approximately 1.2
million have recurrent seizures that do not respond to antiepileptic drugs. For these
patients, alternative treatments consist of neurosurgical removal of the seizure source from
the brain, or surgical implantation of a device to electrically stimulate the vagus nerve in
the neck. In spite of these options, a significant number of people suffer from seizures that
remain uncontrolled. For this group, deep brain stimulation (DBS) may be helpful.

The rationale for using deep brain stimulation (DBS) to treat epilepsy is that DBS disrupts
regulatory feedback loops in the brain that allow seizures to develop and spread. By
influencing these regulatory areas in the brain, electrical stimulation has the capacity to
reduce or eliminate seizures that originate in portions of the brain that cannot safely be
surgically removed. Vagus nerve stimulation (VNS) works in a similar way against seizures,
but VNS disrupts the regulatory loops indirectly, while DBS disrupts the regulatory loops
directly.

The thalamus is an appealing target for DBS to treat epilepsy because it has widespread
connections to, and influences on, the cerebral cortex, which is the outer layer of the brain
where seizures originate. Fourteen patients with various types of refractory epilepsy have
been treated with stimulation of the anterior nucleus (AN) of the thalamus during the past
several years. During the first 3 months of AN stimulation, the median seizure frequency
reduction, relative to baseline, was 64%. Eight of the 14 patients (57%) had a 50% or greater
decrease in seizure frequency (responders). Nine of the 14 patients had seizures presumed to
arise from the temporal or frontal lobes. During the first 3 months, these 9 patients had a
median 80% reduction in seizure frequency and 78% of them were responders. The AN was
selected for stimulation because of its many connections to the "limbic system", a cortical
system which is often a source of seizures.

Prior studies of thalamic stimulation for epilepsy (centromedian nucleus [CM]) The majority
of the research addressing stimulation of the thalamic centromedian nucleus (CM) for the
treatment of epilepsy has been reported by Velasco et al. (1-3). In 1987, the group published
a paper in which 5 patients with generalized tonic-clonic seizures were implanted
bilaterally. Stimulation was delivered 2 hours per day for 3 months. All patients experienced
an 80-100% reduction in generalized tonic-clonic seizures and a 60-100% reduction in partial
complex seizures. Much of the benefit observed continued beyond the 2-hour period of
stimulation, suggesting that the stimulation resulted in a long-term change in thalamic
activity.

In 1993, the same group published a report of 23 patients with externalized CM
stimulators(2,3) (9 with generalized tonic-clonic seizures, 3 with focal motor and secondary
generalized seizures, 5 with partial complex and secondary generalized seizures, and 6 with
generalized tonic seizures and atypical absence seizures). These investigators found that
stimulation resulted in a significant decrease in seizure frequency in the patients with
tonic-clonic and partial motor seizures. However, no significant changes were observed in
patients with complex partial seizures or generalized tonic seizures. Overall, 12 of the 23
patients had at least a 50% decrease in seizure frequency, and 1 patient was seizure free.

None of these studies from the Mexico City group (Velasco et al.) were blinded or controlled.
However, in 1992, Fisher et al. (4) from John Hopkins reported a double-blind, controlled
trial of CM stimulation. This treatment had no effects on the patients. Therefore, CM
stimulation was abandoned and AN stimulation was undertaken instead.

Electroencephalograms (EEG) and thalamic stimulation for epilepsy An electroencephalogram
(EEG) is a recording of brain wave electrical activity. The cerebral cortex nerve cells
generate the EEG signal. The principal features of the EEG studied in patients with epilepsy
are: 1) how rapidly and often the electrical signal changes amplitude ("frequency"), and 2)
any abnormal excessive very brief unexpected electrical discharge ("spike-waves").
Spike-waves are the type discharges characteristic of patients with epilepsy.

Electrical stimulation of the CM nucleus of the thalamus has resulted in less EEG spike-waves
(1-3). However, there have been no studies of EEG frequency in epilepsy patients receiving AN
stimulation therapy.

For this study the investigators will record EEGs on computerized monitors. These recording
machines are small portable computers (about the size of a "Walk-Man") which the patient
wears for one hour. The brain-wave sensing electrodes are attached to the patients' scalp
with a water-soluble temporary conducting glue. These monitors are placed on the patients in
the EEG laboratory at New York Hospital-Cornell. The recorded EEG signal is downloaded to our
EEG analysis computer. This is a routine test commonly performed to evaluate patients with
epilepsy, and the ambulatory EEG in and of itself is not experimental in any way.

EEGs will be recorded at the following times relative to the DBS treatment for epilepsy: 1
month before stimulator implant, 1 month after stimulator implant (but before stimulator
activation), after 3 months of DBS, after 6 months of DBS, and thereafter every 6 months or
after stimulation setting changes.

The recorded EEGs will be mathematically analyzed, according to the following line of
reasoning: Nerve cells in the thalamic AN connect to the cerebral cortex. AN activity (and AN
stimulation) therefore may change how rapidly the electrical potentials of the cortex nerve
cells change, which in turn may change the EEG frequency content. The investigators will
utilize computer programs to determine the EEG frequency content, using a tool called
"spectral analysis". The investigators will compare the spectral analysis between different
time periods during the DBS treatment regimen.

Another feature the investigators will study is "EEG coherence". This is a computer-generated
measure of how similar the brain wave frequencies are in different brain cortex regions.
Since AN nerve cell activity regulates cortex nerve cell activity, AN stimulation may cause
cortical activity in different regions of the brain to be more similar to each other. This
type of change in cortical activity might be part of the mechanism of action of DBS against
epilepsy. The investigators will compare the coherence analysis between different times
during the DBS treatment regimen.

Inclusion Criteria:

- To be scheduled to undergo deep brain stimulation therapy (DBS) in an attempt to treat
intractable epilepsy

Exclusion Criteria: