Epilepsy surgery is considered when seizures are incompletely controlled by medication and it is apparent that further medical therapy will be futile. Surgery is typically offered to patients who have uncontrolled seizures that are disabling, cause loss of awareness, pose risk of injury, impair quality of life, or are psychologically disruptive. When the question of surgery arises in such cases, a presurgical evaluation should be performed. The evaluation is carried out by multidisciplinary teams composed of epilepsy specialists in neurology, neurosurgery, neurophysiology, neuroradiology, psychiatry, and neuropsychology, with additional input from professionals in social work and nursing. This chapter reviews the basic objectives, methods, and principles of the presurgical evaluation. A brief overview of the testing generally included in presurgical evaluations is provided, with full details about each method referred to in subsequent chapters.

The goal of any treatment for patients with epilepsy is “no seizures, no side effects,” as advocated at the National Institutes of Health (NIH)-sponsored meeting Curing Epilepsy: Focus on the Future; A White House-Initiated Conference on Epilepsy (CURE).^48,94 For most patients, even a rare seizure can be an unacceptable level of control because of restrictions on driving and the impact on social and vocational status.^96,103 While the CURE group acknowledged that complete seizure control cannot be achieved in all patients, it recognized that complete absence of seizures made such a difference in a person’s quality of life that this objective should be as close to the standard of care as possible.

Physician experts in epilepsy care recommend that patients with refractory epilepsy, defined as failure of two to three appropriate antiepileptic drugs (AEDs) or disabling medication side effects, should be referred to a comprehensive epilepsy center for a further evaluation that includes surgery as one of the therapeutic options.^5,13,26 The purpose of the referral is to verify that patients have a correct epilepsy diagnosis (i.e., exclude nonepileptic seizures and improper seizure classification that may affect treatment recommendations) and to consider all potential further therapies, such as surgery, that might be employed in an attempt to stop seizures. Of course, not every patient referred to an epilepsy center will have surgery.⁹ Changes in medical management can often substantially improve seizure control or even eliminate seizures so that patients can be returned to their community physicians. Likewise, additional nonsurgical therapies, such as the ketogenic diet and vagus nerve stimulator (VNS), can be considered at the time of a presurgical evaluation.^37,39,99

There is now irrefutable evidence that uncontrolled seizures are associated with progressive anatomic and functional changes, neurologic compromise,^52,67,88,97 and higher than expected mortality rates.^74,78,90 Although there still continues to be some debate as to the definition of therapy resistant,^8,10,68 these considerations have led to the recommendation that therapy-resistant patients should be referred to an epilepsy center for presurgical evaluation once they have failed initial AED therapy.^12,14,33,35 This recommendation is especially important for infants and young children, who are at greatest risk of epilepsy-induced encephalopathy.^29,53,54 It has even been advised that children with uncontrolled seizures or infantile spasms under age 2 years should be referred to a specialty center regardless of magnetic resonance imaging (MRI) findings.²⁶ MRI-demonstrated lesions should also be referred to an epilepsy center for presurgical evaluation, regardless of seizure control, as these substrates may eventually lead to therapy-resistant seizures, or the lesion itself may prove to be progressive.^28,85 In addition, intractable focal epilepsy can result from low-incidence etiologies that require the special experience of a comprehensive epilepsy center. Examples of these pathologies include adolescents and adults with focal cortical dysplasia and hippocampal sclerosis, and children with hemimegalencephaly,⁸² Rasmussen syndrome,¹¹ Sturge-Weber Syndrome,³¹ tuberous sclerosis complex,³¹ Landau-Kleffner syndrome,⁷³ hypothalamic hamartomas,^45,66 and polymicrogyria.³ The upper limit at which surgery should be considered is not certain. We have had successful experience with surgery for patients in their 60s who are otherwise in good health.

The goal of a presurgical evaluation is to characterize a patient’s epilepsy thoroughly and completely. In doing so, the multidisciplinary epilepsy team determines what type of seizures and epilepsy syndrome the patient has and the likelihood that it can be treated successfully with surgery. At the completion of the process, the team should be able to provide the patient and his or her family with a comprehensive assessment of the risks and benefits of different treatments and the risks that may be associated with the natural history of the epilepsy syndrome. If surgery is an option, a detailed description of the proposed surgical procedure, surgical risks that may be encountered, and long-term seizure control that can be anticipated from medical therapy alone should be provided. This information must be
communicated effectively to patients and their families so they can make informed decisions about subsequent management. The epilepsy team cannot tell the patient what to do, but it can provide sufficient information and education, in a manner that patients can understand, so that they and their families are informed about all the options and can make decisions about their care. It should also be noted that there are no age restrictions in considering patients as possible candidates for epilepsy surgery.^43,69

The types of surgical procedures can be divided into two general categories.⁶⁹ Resective operations are those where the intent of the surgery is to remove the epileptogenic brain area. The goal of resective surgery is to eliminate seizures completely and, hopefully, improve cognitive and psychosocial outcomes (see Chapter 167). Most resective procedures involve removal of cerebral cortex, and they can be subclassified as focal (lesion and surrounding cortex, Chapters 176, 177, and 179), lobar or multilobar (Chapter 178), or hemispheric (hemispherectomy, Chapter 178). Resections can also involve removal of subcortical lesions, like hypothalamic hamartomas^35,36 (Chapter 250). Also included under this category are procedures like radiosurgery, where focused x-ray beams destroy defined epileptogenic regions of the brain, and multiple subpial transections (MSTs; Chapter 182), which are thought to disconnect epileptogenic zones selectively from the rest of the cortex.⁷² With palliative procedures, the intent of therapy is to stop the most disabling seizures and/or reduce seizure frequency, but not necessarily eliminate all seizures. The most common palliative operations are corpus callosotomy^40,65,77 (Chapter 180) and VNS^7,37 (Chapter 131). Most callosotomies involve sectioning of the anterior two thirds to four fifths of the callosum, although a complete callosotomy is occasionally done in individuals with a major unilateral deficit or after failure of initial anterior resection.

The ideal candidate for resective surgery is a patient with therapy-resistant epilepsy due to a clearly demarcated localized brain abnormality (symptomatic epilepsy) where removal of the pathologic lesion has a high probability of producing seizure control with acceptable or no side effects.^33,34 The precise identification and localization of the responsible epileptogenic brain region is the goal of the presurgical evaluation.⁴⁴ Thus, the evaluation aims to (a) identify discrete structural abnormalities using modern neuroimaging and, inferentially, from associated functional deficits; (b) provide convincing evidence that seizures arise at or near the structural lesion; and (c) establish whether eloquent brain regions that support speech and motor-sensory functions are at risk. If these objectives are satisfied, then there is a high likelihood of a good postoperative seizure control. However, these aims may be modified in some patients with well-defined seizure syndromes or in individuals at risk for seizure-induced cognitive deficits. For example, patients with MRI-demonstrated unilateral hippocampal sclerosis may have multiple areas of interictal or ictal abnormalities on electroencephalography (EEG) but still respond well to anterior lobe resection. Similarly, infants and children with symptomatic epilepsy may present with what appears clinically and electrographically to be generalized epilepsy, but they nonetheless sometimes respond well to surgical removal of an identified cortical abnormality such as cortical dysplasia.^23,24 Perhaps the ultimate localized cortical resection is cerebral hemispherectomy, an operation that has evolved considerably over the years.^25,30 The basic objective in evaluating patients for hemispherectomy is similar to that in focal resection, namely, to identify an area of brain that can be removed safely to eliminate seizures. The presurgical evaluation in patients who are candidates for hemispherectomy aims to establish the following: (a) a unilateral structural abnormality; (b) widely distributed abnormal excitability in that hemisphere; and, in older children and adults, (c) significant impairment of motor and sensory function (and perhaps language) in that hemisphere.

Corpus callosotomy and VNS can be useful when the realistic therapeutic goal is to reduce the burden of epilepsy without stopping all seizures. As a consequence, the clinical approach to such patients and the surgical risk–benefit ratio are different than for patients being considered for resective surgery. When epileptogenic areas of the brain are widespread and involve both hemispheres, localized cortical resection is not usually a viable option. However, interruption of the interhemispheric connections via corpus callosotomy or stimulation of the vagus nerve (VNS) can often palliate seizures in such patients. This is especially true for patients that suffer injuries from generalized tonic, atonic, and tonic–clonic seizures. Corpus callosotomy and VNS may also benefit patients with complex partial seizures who are not candidates for focal resection.^6,83 For palliative procedures, the evaluation should determine if there is (a) evidence of generalized epilepsy or nonlocalizable partial epilepsy with abnormal generalized or multifocal discharges on the EEG, (b) lack of suitability for a resective procedure, and (c) presence of tonic, atonic, tonic–clonic, or therapy-resistant complex partial seizures. Neuroimaging is less critical in planning palliative operations unless it strongly suggests a discrete lesion, in which case consideration should be given to a focal resection.⁹³

The overriding principle of the presurgical evaluation is to define abnormalities of structure and function that can be attributed to abnormal cortical areas and EEG-determined pathologic excitability, and then remove those cortical areas in which all relevant data are concordant. Historically, our concepts of what constitutes the cortical region(s) that should be removed to control seizures have been determined by the evolving tools available for clinical use.³⁴ Over time, a series of terms that embody these concepts have been developed.^32,81 They are recapitulated and described here based on the current state of our understanding and their application to the evaluation process. In general, the greater the degree of overlap among the different tests, the higher the likelihood of accurately defining the epileptogenic zone and achieving a good postsurgical result.