Epilepsy surgery: Not a last resort

By Sean Lew, MD, and Mary Zupanc, MD

The problem

Epilepsy, or recurrent, unprovoked seizures, affects 0.5 to 1 percent of the population, with the most frequent incidence being in childhood. Every year approximately 25,000 to 40,000 children in the U.S. experience their first seizure that is not associated with a febrile illness or head trauma. Thirty to 65 percent of these children go on to have a second unprovoked seizure within the first year. Fortunately, in most patients the seizures can be controlled adequately with antiepileptic drugs (AEDs), and in some cases the epilepsy will resolve over time. However, roughly 30 percent of children with epilepsy cannot be controlled medically. In addition, even in children with well-controlled epilepsy, the morbidity associated with long-term AED use can be significant. Epilepsy surgery offers an alternative means of treating seizures in children and is an important weapon in the armamentarium of the epileptologist.

Early intervention is key

Infants and children have specific vulnerabilities to even brief, recurrent seizures. If in the midst of a developing brain there is an ongoing "epileptic storm," the brain undergoes aberrant synaptogenesis and apoptosis as well as changes in the development of inhibitory neurocircuitry. Therefore, immature brains quickly will become "programmed" to excitability with continued seizures. This results in the establishment of broad networks of epileptogenicity that can become progressively more difficult to manage. Frequent seizures result in a secondary encephalopathy with subsequent global impairment. The consequences of uncontrolled seizures on the immature, developing brain are severe: developmental delays, behavioral difficulties and irreversible cognitive declines are the norm. If the seizures and ongoing epileptic encephalopathy can be stopped with medication or surgery, windows of "developmental opportunity" can open. The result is significant developmental gains that would not have been seen if the ongoing epileptic encephalopathy had persisted. Fortunately, it does not take years to determine if a patient's epilepsy will be amenable to medical treatment. Sixty percent of children will be controlled with the initial AED prescribed. Of the 40 percent whose seizures persist, only an additional 10 percent will achieve good control with a second appropriately chosen AED. The remaining 30 percent constitute the intractable group and adequate control is unlikely to be achieved with subsequent medication trials. 

Optimal treatment equals the best quality of life

Even in children whose epilepsy is well controlled, the consequences of their epilepsy can be profound. They have a higher incidence of anxiety and depression, perform at a lower level academically and ultimately suffer from a discrepancy in employment and income when compared with peers. Adults with epilepsy are less likely to marry. Finally, side-effects from AEDs can be profound. Many older AEDs can cause sedation. Phenytoin causes osteoporosis with long-term use; it is suspected but unproven that the enzyme-inducing AEDs also cause osteoporosis. Carbamazepine and felbamate can produce aplastic anemia. Valproate increases the risk of polycystic ovarian syndrome and anovulatory cycles in women. Many AEDs are teratogenic, and some animal studies also have suggested that long-term AED use has oncogenic potential. For these reasons, epilepsy surgery should be considered not only for medically-refractory cases, but also in cases that suggest an opportunity for seizure freedom without medication (or with considerably less medication). 

Surgery is not a last resort

There is a tendency to view brain surgery as a high morbidity/mortality endeavor that should be reserved for the most dire cases. While there certainly are risks associated with epilepsy surgery, they are often far outweighed by the risks of continued medical treatment and poor seizure control, as well as the quality-of-life benefits associated with attaining seizure freedom or improved seizure control. Advances in diagnostic and surgical techniques have improved success rates and decreased complication rates.  Unfortunately, patients often are unsuccessfully managed for several years before being referred to a tertiary epilepsy center for surgical evaluation. Over that time, the patient often suffers irreversible cognitive losses, while the epilepsy becomes more difficult to manage or cure due to the spread of epileptogenic networks. It is likely that many incurable adult epilepsy patients would have been more easily treated surgically at an earlier age with greater preservation of cognition. The developing brain of a child also is much more plastic than that of an adult. It is therefore possible for eloquent functions such as motor control and language to re-establish in other areas of the brain in a young child – a feat not possible in a mature brain. Thus, it is important that surgical options be considered early in the course of treatment, rather than as a last resort.

Determinants of surgical candidacy

The decision of whether or not a child is a surgical candidate typically is made by a multidisciplinary team at a tertiary epilepsy center. Evaluations require expertise from a variety of fields including neurology, neurosurgery, neuropsychology and radiology.

Diagnostic modalities utilized to assist with decision-making include:

• Video electroencephalographic monitoring.
• Magnetic resonance imaging.
• Wada (selective angiography to anesthetize individual hemispheres of the brain to determine language/memory localization).
• Other neuroimaging:
  • Functional MRI.
  • Magnetoencephalography.
  • Positron emission tomography.
  • Single photon emission computed tomography.
• Neuropsychological testing.

Questions that are considered when determining surgical candidacy include:

Is there a structural lesion that appears to be generating the seizures?

Seizures associated with structural lesions, such as tumors, focal cortical dysplasias, vascular malformations, hamartomas, infarcts and mesial temporal sclerosis, are more often amenable to surgical treatments. In the majority of lesional cases, complete seizure freedom can be obtained surgically. Surgery can be considered even in cases with adequate seizure control with AEDs. Specifically, if removal of the lesion can be accomplished easily and with good success, surgery would be low risk but carry the potential for high benefit – seizure freedom, with either few or no AEDs.

What is the prognosis for continued nonsurgical therapy?

If a patient has failed with more than two AEDs, it is likely he or she will remain medically intractable. Some epilepsy syndromes, such as benign rolandic epilepsy, typically are self-limited and resolve over time. Other syndromes have a predictable, progressive course leading to medically intractable seizures that only can be cured with surgery. If one can identify the specific epilepsy syndrome or etiology for the epilepsy, the prognosis typically can be defined. Rasmussen's syndrome, epilepsy associated with mesial temporal sclerosis and large cortial malformations (such as hemimegalencephaly) almost always are associated with intractable seizures and a progressive course. These epilepsy syndromes should be evaluated for epilepsy surgery quite early.

Is it likely that an epileptogenic zone can be identified and safely treated?

Much of the preoperative assessment is geared toward determining the likelihood that a focal epileptogenic zone exists that can be resected or disconnected without generating unacceptable neurologic deficits. If the epileptogenic zone is suspected but not certain, the patient may require intracranial EEG monitoring via implanted electrodes to better define the source of the seizures and define eloquent cortex, areas of the brain where important motor or language functions reside. Some patients have seizures without any evidence of such localization. They are typically less optimal candidates for epilepsy surgery. However, in carefully selected cases, these patients can have successful resections of the epileptogenic focus. These cases do require invasive intracranial monitoring. Others have epileptogenic foci that include eloquent areas of the brain, making surgical resection or disconnection a less palatable option.

Is there a role for palliative surgery?

Some children have incurable epilepsy but still can benefit from surgery to decrease the frequency or severity of their seizures. For example, some patients with intractable generalized epilepsy gain substantial benefit from sectioning of the corpus callosum to eliminate drop attacks (tonic/atonic seizures), while they will continue to have other types of less-severe seizures. In other patients, they may have bihemispheric/multifocal epileptogenic zones. If one area is producing the predominant intractable seizures, resection of this area may provide a substantial reduction in seizure frequency and result in a significant improvement in quality of life.

Surgical techniques

A variety of surgical techniques are used to treat pediatric epilepsy. The appropriate surgery is determined by the extensive presurgical evaluation. The initial phase of surgery often is diagnostic – electrodes are temporarily implanted intracranially to better localize the epileptogenic zone and often to map areas of eloquent brain to be preserved. (See Figure 1.) There are three broad categories of therapeutic surgical techniques.

Figure 1. Intraoperative photograph demonstrating the use of implantable subdural electrodes.

Resection
Surgical resection of epileptogenic tissue is the mainstay of epilepsy surgery. Resections can be limited, as in the case of small lesions (lesionectomies) or areas of epileptogenic brain (topectomies). They also can include complete or partial resections of lobes of the brain (lobectomies). At the extreme end of the spectrum, the entire cerebral hemisphere can be removed (anatomic hemispherectomy) or disconnected (functional hemispherectomy) in the case of broad areas of pathology such as Rasmussen's syndrome, Sturge-Weber syndrome or perinatal infarctions.

Disconnection
If an epileptogenic zone is disconnected from the rest of the brain, the seizures can be cured (if the only connection is cut) or made significantly more manageable by slowing conduction (by eliminating the main conduit of transmission). A functional hemispherectomy involves both resection and disconnection as a means of eliminating the transmission of epileptogenic discharges to the normal brain. Disconnections often are performed to minimize the morbidity associated with a more extensive resection. Sectioning of the corpus callosum (corpus callosotomy) (See Figure 2.) often will slow conduction of seizures in the case of bilateral onset, thereby reducing the severity of seizures and often eliminating drop attacks. Multiple subpial transections are used to selectively disconnect eloquent cortex to allow for continued function while minimizing spread of epileptogenic discharges.

Figure 2. Postoperative MRI demonstrating a corpus callosotomy – a disconnection of the main white matter connections between the two hemispheres.

Stimulation
There is a trend in functional neurosurgery toward nondestructive techniques involving electrical stimulation of the nervous system to modulate activity. Currently, vagal nerve stimulation is the only mainstream stimulation technique used in pediatric epilepsy. It is typically a palliative procedure with minimal risk and side effects. Deep brain stimulation and cortical stimulation are techniques that show promise in adult human trials for managing some forms of epilepsy. Pediatric trials are forthcoming.

Summary

• The immature, developing brain is particularly susceptible to the ravages of poorly controlled seizures, leading to permanent developmental/cognitive losses and the development of broader epileptogenic networks.
• It is imperative that children with epilepsy achieve maximal control in a timely fashion to prevent these long-term sequelae.
• A significant subset of children with epilepsy can benefit from epilepsy surgery as a curative measure or for palliation.
• The earlier these patients are identified and treated, the better the neurological/developmental outcome.
• The decision-making process for determining surgical candidacy and the appropriate surgical approach is complex and requires a multidisciplinary team at a comprehensive epilepsy program. 

Sean Lew, MD, is a pediatric neurosurgeon at Children's Hospital of Wisconsin. He also is an assistant professor of Neurosurgery at the Medical College of Wisconsin and a member of Children's Specialty Group.

Mary Zupanc, MD, is a pediatric neurologist and medical director of Neurology at Children's Hospital of Wisconsin. She also is a professor of Neurology at the Medical College of Wisconsin and a member of Children's Specialty Group.