The basic principle underlying the pre-operative work-up is identification of the area of seizure onset (focus) and identification of the areas of function in the brain. Identification of the seizure focus is based on the seizure history and EEG monitoring. Patients may require hospital admission for continuous monitoring so that seizures can be captured on EEG with video. Imaging has a large role as well, especially high resolution MRI, which can show small lesions. The presence of a lesion in association with the seizure focus greatly increases the likelihood of achieving seizure control with surgery. Many patients who were previously thought to have no lesion are now found to have small areas of cortical dysplasia on MRI. These are malformed areas of cortex, which may generate seizures. SPECT (single photon emission CT) scanning is sometimes useful in identifying the seizure focus and occasionally PET (positron emission tomography) scanning and/or MEG (magneto encephalography) are required.

The other major preoperative issue is identifying areas of function. This helps us assess the risks of removing the seizure focus. The clinical and neuropsychology examinations are important and imaging plays a major role. Recent MRI techniques allow us to identify functional brain areas. Patients are given tasks while in the MRI unit, which may consist of finger or hand motion to identify those areas of the brain, or speech tasks to identify language areas of the brain. This is age dependent and may be difficult below 5 years of age. WADA testing is commonly used to determine which hemisphere is dominant for language and to assess the relative contribution of the two hemispheres to memory function. This requires a cerebral angiogram during which sodium amytal is injected into each internal carotid artery separately. After the injection, neuropsychology testing of language and memory is performed. This is also age dependant but is feasible in most children over 5 years of age.

In difficult situations, the process of localizing the seizure focus and/or localizing brain function may require surgery. Electrocorticography is an EEG performed on the surface of the brain. This allows us to assess the extent of the seizure focus and guides the extent of brain resection. Areas of function can be identified in the operating room by electrical stimulation applied to the surface of the brain, which will produce responses in the extremities, or by stimulation of nerves in the extremities which produces a response that can be recorded on the cortex.

Epilepsy surgery has been a part of neurosurgery for many years but its use in children has been increasing. In the past, there has been a belief that children might "grow out of" their seizures. As pediatric epilepsy has evolved, we have developed an improved knowledge of the natural history and are now able to predict which epilepsy syndromes will improve and which will not. Prior to MRI, some lesions were not well-recognized, especially cortical dysplasia. The high resolution MRI techniques that are now available allow us to identify these areas and direct our surgical efforts. There has also been recognition of the harmful effects of antiepileptic drugs and of frequent seizures on the developing brain.

Types of Pediatric Epilepsy Surgery

Temporal Lobe Epilepsy: This is the most common form of epilepsy requiring surgery in adults. It occurs in children as well and is often evident by adolescence. It is usually due to mesial temporal sclerosis in which the hippocampus is shrunken down and scarred and the seizures are coming from that area. Surgical resection of the hippocampus has about an 80% chance of rendering children seizure free and off medication. Other areas of the temporal lobe can be affected by lesions including low-grade tumors, particularly ganglioglioma and dysembryoplastic neuroepithelial tumor, and vascular lesions. Both of these are surgically curable lesions and their removal has a very high chance of stopping the seizures.

Extratemporal Disease: The seizure focus may be outside the temporal lobe with or without an identifiable lesion on imaging. The non-lesional extratemporal foci are more difficult to localize accurately and may require placement of subdural electrodes for a period of time (7-10 days). Placement requires burr holes and/or craniotomy. Patients then spend 7 - 10 days on the ward with the electrodes in place. This allows accurate identification of the seizure focus and mapping of areas of function. A second operation is then required during which children go back to the operating room and the seizure focus is removed, avoiding the functional areas. About two thirds of children with extratemporal, non-lesional epilepsy have a good result from surgery.

Hemispheric Disease: There are several pediatric disorders that affect all of one hemisphere and do not affect the other hemisphere. These include Rasmussen's encephalitis, Sturge-Weber syndrome, hemimegalencephaly, diffuse hemispheric cortical dysplasia and cerebral infarction. These disorders usually result in a contralateral hemiparesis and intractable epilepsy coming from that hemisphere. EEG's usually show that the seizures are coming from multiple parts of the hemisphere and that the opposite side is normal. In such situations, hemispherectomy is very effective in stopping the seizures (about 80% seizure free). Our technique of hemispherectomy is called peri-insular hemispherotomy, which disconnects the whole hemisphere without removing too much tissue. This decreases the long-term complications associated with hemispherectomy.

Vagal Nerve Stimulation: Placement of a subcutaneous stimulator (similar to a cardiac pacemaker) with the leads wrapped around the left vagus nerve results in decreased seizure activity. This was demonstrated in two randomized clinical trials in the 1990's. The exact mechanism of action is not well understood, but there are changes in the brain demonstrable on PET imaging during stimulation. Approximately half of appropriately selected patients get a good response (80-90% reduction in seizure frequency). Another quarter of patients decrease their seizure frequency by half. The stimulator is placed using a subclavicular and a neck incision during a one-hour procedure. The generator is then externally programmed to optimize the patient's seizure control. The battery life depends on the stimulation parameters but is predicted to last three to five years or more before a battery change is required. Patients are selected for vagal nerve stimulation if they have difficult to control epilepsy that is not suitable for surgical resection.

Summary

There are surgical options for children with epilepsy. The likelihood of responding to a new medication after failing two is quite low. Small lesions that can now be detected with high resolution MRI can produce very difficult to control epilepsy, which respond well to surgical resection. The potential effects of multiple medications and chronic seizures on the young developing brain should be kept in mind. Surgically treatable disorders should be recognized so that all treatment options can be considered.

	Home || Patient Education || Clinical Services || Faculty || Fellowship || Contact Information

copyright© 2000 Neurosurgery4kids.net. All Rights Reserved.