Recent Advances in Epilepsy Surgery

Abstract

The modern practice of epilepsy surgery requires multiple modalities of presurgical investigations such as video-EEG, intracranial EEG, high-resolution imaging, advanced functional imaging, and clinical analysis. A multidisciplinary approach is essential, including close collaboration between neurosurgeons, neurologists, neurophysiologists, neuropsychologists, neuropsychiatrists, and neuroradiologists. Candidates for epilepsy surgery require a history of seizures that are refractory to appropriate medical therapy. A meticulous selection of candidates will lead to a better chance of freedom from seizure. Epilepsy surgery includes a variety of surgical procedures including resective surgery for focal refractory seizure, which offers a significant chance of seizure freedom in temporal and extratemporal lobe epilepsy. Palliative treatment for patients who are not candidates for resective surgery, such as vagal nerve stimulation, deep stimulation, and callosotomy, offers further options. We reviewed and analyzed the recent scientific literature and forthcoming advances that will impact on the future of epilepsy surgery. This chapter on recent advances in epilepsy surgery emphasizes improved methods of assessment, a better understanding of seizures, the development of new surgical techniques, and the outcome of epilepsy surgery.

1. Introduction

Epilepsy is a brain disorder characterized by brief disturbances in the normal electrical function of the brain resulting in seizures; 35% of adults with active epilepsy have seizures that are not seen by a neurologist or an epilepsy specialist [1].

Epilepsy is one of the most common chronic neurologic disorders with a worldwide prevalence of approximately 1.2% [2]. Epilepsy contributes to up to 25% of the global burden of neurological disease, and many neurological diseases are associated with seizures and epilepsy [3].

Recent data indicate that epilepsy mortality rates are rising significantly [4]. These data have generated significant concern from stakeholders and advocacy groups that the increase in epilepsy mortality may represent a failure to effectively treat epilepsy and prevent premature death [5].

Only 64% of patients with new-onset seizures are seizure-free by their third anti-epileptic drugs (AED) [6]. Thus, more than 35% of patients continue to have seizures and become recognized as refractory seizure patients.

The pre-surgical evaluation should result in a clear understanding of whether surgery can be undertaken and its potential benefit [7].

Neuroimaging developments with the introduction of magnetic resonance imaging (MRI), functional (fMRI), fluourodeoxyglucose F18 positron emission tomography (PET), single-photon emission computed tomography (SPECT), and magnetoencephalography (MEG) have facilitated the selection of patients for surgery, also reducing the number and severity of complications [8, 9]. Neurophysiological tests include invasive and noninvasive procedures and define the epileptogenic zone in a specialized center. Epilepsy surgery should be recommended through a multidisciplinary team [5].

In pharmacoresistant (PR) patients, epilepsy surgery must take into consideration the chance of seizure freedom and the adverse long-term effects of uncontrolled seizures [10]. Epilepsy surgery is underutilized even in developed countries because many physicians do not recognize that a treatable syndrome exists and in developing countries because of lack of resources or because many physicians do not recognize that a treatable syndrome exists [5].

Developments made in surgical techniques have significantly increased the effectiveness and safety of these techniques as such techniques have been demonstrated to improve seizure control/freedom outcomes [11] and increase patients’ life span [8] by reducing the number and severity of complications [8, 9].

2. Refractory seizure

One percent of the world’s population has active epilepsy, 30–40% of people with epilepsy have a seizure that is uncontrolled by medication [1], which accounts for 80% of the cost of epilepsy in the United States [12].

PR epilepsy is, therefore, a major health concern for patients, their families, and for the society. Treatment aim for epilepsy is seizure freedom with no side effects, as soon as possible. Full-service epilepsy centers are staffed by a multidisciplinary team consisting of neurologists, epileptologist, neurosurgeons, neuroradiologists, clinical neurophysiologists, neuropsychologists, psychiatrists, social workers, and nurses skilled in the management of epileptic seizures and their consequences [13]. These approaches permit recognition of true epileptic seizures and their causes, diagnosis of specific seizure types and epilepsy syndromes, and determination of which patients are truly FRs and might be candidates for surgical therapy.

Apparent pharmacotherapy failure does not necessarily mean that standard AEDs will not work. Alternative causes are seizures that are not epileptic, misdiagnosis of the seizure type or epilepsy syndrome, inappropriate use of AED such as inadequate doses or drug interactions and lifestyle issues, such as drug abuse, alcohol binging, stress, and sleep deprivation. Epilepsy centers have the ability to utilize specialized pharmacologic approaches, including enrollment in clinical trials of experimental anti-seizure drugs, to provide alternative treatments other than surgery,

The term “PR epilepsy” can no longer be taken literally, as there are now so many anti-seizure drugs that it would take a lifetime to try all of them alone and in combination in any given patient.

There are several reasons for PR, and research to clarify underlying mechanisms is important for the future development for more effective treatments [14].

Concerning the diagnosis of PR patients, the International League Against Epilepsy (ILAE) has proposed, as a verifiable hypothesis: “That PR is defined as failure of adequate trials of two tolerated, appropriately selected, and used antiepileptic drug schedules (monotherapies or in combination) to achieve sustained seizure freedom, which is defined as sustaining seizure freedom for a period 3 times the longest inter-seizure interval or 1 year whichever is longer” [15].

3. Epileptogenic zone

The epileptogenic zone is defined as the area of the cortex that is necessary and sufficient for initiating seizures and whose removal 0r disconnection is necessary for the complete abolition of seizures [16].

The pre-surgical evaluation to define the epileptogenic zone and its relationship with the eloquent area is essential for the ideal resection for a patient with drug-resistant focal epilepsy (Figure 1) [8].

Figure 1.

Epileptogenic complex cortical area. IOZ; ictal onset zone, IZ; Irritative zone, FZ; functional zone, SZ; symptomatogenic zone, EZ; epileptogenic zone, LS; lesion area.

Jehi [17] defined five cortical zones in the pre-surgical evaluation process, which include: the irrigative zone (IZ), which is the area of the cortex that generates electric spikes, with the best test to define that through EEG, MEG and EEG-fMRI; seizure onset zone (SOZ), which is that area of the cortex responsible of the clinical seizure tested by EEG, ictal SPECT, fMRI, and MEG; symptomatic zone (SZ), defined as the area of the cortex that, when activated produces initial ictal symptoms signs, observed by initial seizure symptomology; epileptogenic lesion (EL) includes macroscopic lesion that is causative of the epileptic seizure because the lesion itself is epileptogenic (cortical dysplasia) by secondary hyperexcitability of the adjacent cortex, tested by MRI; functional deficit zone (FDZ) defined as the area of the cortex that is not functioning normally in the inter-ictal period tested by neurological and psychological exams and by functional images (interictal SPECT and PET) (Table 1).

Critical zone	Diagnostic procedures
Functional deficit zone; area of the cortex with abnormal function in the intricately period	Clinical neurological and neuropsychological examination. PET and Inter ictal SPECT
Epileptogenic lesion area; were the lesion is responsible of the seizure, such as migratory disorders.	Brain MRI
Symptomatic zone; are of the cortex that that during seizure activation produce symptoms or signs.	Clinical initial seizure and Video EEG
Seizure onset zone; area of the cortex the initiate the clinical seizures	Video EEG, Ictal SPECT
Irritative zone: area of the cortex that generate inter ictal spikes	EEG, MEG, EEG-fMRI
Epileptogenic network	Micro dialysis; (Glutamate level at onset of seizure and in epileptogenic region)
	7TMR spectroscopy image(spectrometer); mainly in medical temporal lobe epilepsy) MTLE and MTS.
	Neurotransmitters; Central benzodiazepine receptors acts on ionotropic GABAA–regulation in seizure complex, which is reduced in epileptogenic foci and seizure onset.
	Dopamine receptor; Dopamine receptor D2/D3 decrease in epileptogenic area(PET) Opioid receptors, trough PET radiotracers, no changes in temporal lobe epilepsy.
Tractography	Difussion tensor imaging(DTI), can visualizeze Meyer Loop preoperativly.

Table 1.

Preoperative evaluation modalities for delimitation of epilepsy zone in refractory seizure patients.

It is helpful in the study of epileptogenic zone to use intracranial EEG (subdural electrode or SDE implantation via craniotomy) as a principal approach for intracranial EEG monitoring [18]. Nevertheless, there is no high-quality evidence indicating superiority of any one technique over the other intracranial EEG monitoring [19], specific brain MRI with sequence and voxel-based morphometric analysis, and the EEG-fMRI [20]. 3D multimodality images are a simultaneous display of different structural and functional data in each patient [17, 18, 19].

Unidentified epileptogenic source occurs in approximately 50% of patients coming to surgery [21]. According to advanced knowledge in EZ, defining the area of neuroconnectivity through the epileptogenic network is very important. This is achieved by using new tests for epileptogenic network detection: a) microanalysis; by detecting the extracellular glutamate level, through microanalysis device insertion; glutamate level is increased in the epileptogenic region and high at onset of seizure [22]. b) 7 T-MR spectroscopy, in the study of mesiotemporal lobe sclerosis (MTLS), the epileptogenic in the hippocampus is energetically developed, anterior more than posterior; a similar energetic loss is seen in the ipsilateral anterior thalamus and less significantly in the contralateral thalamus and hippocampus [23]. According to this, optogenetic stimulation of thalamic circuit can inhibit the parietal epileptogenic cortex.

4. Presurgical evaluation

The pre-surgical evaluation is the most crucial aspect for epilepsy surgery; all patients diagnosed with PR should be referred to an epilepsy center for pre-surgical evaluation to confirm the diagnosis of true epilepsy and type of seizure. At this stage of evaluation, around 50% of patients usually have idiopathic generalized seizure, which cannot be treated with resective epilepsy surgery; only patients with focal-onset seizure with loss of awareness (partial complex seizure) with or without secondary generalized seizure are surgical candidates and can complete the pre-surgical process for investigation. Therefore, the initial evaluation should include proper history, seizure semiology, and EEG. The final goal of pre-surgical evaluation is to define the EZ (region) defining the lateralization and localization of EZ (Table 2) [24].

Table 2.

Pre-surgical evaluation parameters of patients with refractory seizure.

The American Epilepsy Society conference had a consensus about the evaluation protocol through a multidisciplinary methodological approach involving physical exam, scalp video-EEG, telemetry, structural MRI (MRI epilepsy protocol), neuropsychological assessment, neuropsychiatric assessment, social worker, and nursing for patient support network, with realistic expectation of outcome [25]. Additionally parameters include fMRI, Wada test, PET, ictal SPECT, MEG, and intracranial EEG electrodes (Table 1) [25, 26].

4.1 Neurological and neuropsychological studies

Neurological and neuropsychological assessment is part of the preoperative evaluation of refractory seizure patients. The assessment should comprise baseline standardized measures of cognitive function in addition to wider measures of behavioral and psychosocial function [27]. The preoperative neuropsychological assessment contributes to seizure lateralization, localization, characterization, and provides predictions of cognitive risk associated with surgery [27].

For future outcome, it is important to include exploration of patient and family expectations from surgical treatment. Neuropsychological changes following surgery are a dynamic process, and it should be an integral part of the postoperative follow-up. The neuropsychologist plays an important role with postoperative rehabilitation and support of the patient and family members as a part of the multidisciplinary team of epilepsy surgery [28].

4.2 Electrophysiological studies

4.2.1 Noninvasive EEG monitoring

The standard procedure after confirming PR patient usually is initiated with video EEG-monitoring for detection of lateralization and localization of seizure focus. Patients are often admitted for several days, depending on their seizure frequency. Most often their AEDs are tapered off in order to capture about 3–5 typical and consistent seizures. There is no consensus on the tapering process, but in general very rapid tapering is avoided to minimize risk of status epilepticus or triggering aberrant seizure onset zones [24, 29].

Scalp-recorded EEG is usually performed for inter-ictal and ictal epileptiform patterns [29]. In the video-EEG monitoring, seizure semiology is recorded, which is important for surgical decision-making [29]. Occasionally when ictal/inter-ictal EEG epileptiform activity discharges are concordant with structural neuroimaging abnormalities, it could be sufficient for surgical localization and treatment [26]. High-density array of EEG electrodes (10–10 electrode placement) may be useful in patient with extra-temporal lobe focal seizure. Deeper seizure foci may not be detected via (scalp) electrodes. Further, scalp EEG recording is usually insufficient in extra-temporal epilepsy or even in non-lesional (normal MRI) temporal lobe epilepsy (TLE). Using extra electrodes placed based on the 10–10 international electrode system may add significant accuracy to scalp recording and in some cases avoid intracranial recording [24].

When seizure onset zone is poorly lateralized due to alternating seizure onset lateralization in bi-temporal asynchrony or frequent bilateral epileptiform discharges, it indicates a less favorable postsurgical seizure outcome [30]. Unilateral hippocampal atrophy on MRI and concordant unilateral interictal spikes are highly predictive of concordant ictal localization [31].

4.2.2 Invasive intracranial EEG

ECoG, implantation of SDE, is very helpful in detecting, seizure lateralization and localization. The duration of intra cranial-EEG monitoring is determined by seizure frequency, number of seizures needed to make a decision, and the time needed to perform mapping of the eloquent cortex. Interpretation of the data is based on EEG pattern recognition as well as clinical semiology.

It is essential to insert the intracranial EEG electrodes in proper locations at or in close vicinity to the epileptogenic zone although interpretation may be difficult in some cases since that frequency range may differ from one cerebral area to another due to neurophysiological properties of the anatomical structure and from one etiology to another [24, 32].

4.2.3 Stereoelectroencephalography (SEEG)

Recording EEG signals through surgically implanted depth electrodes provides the best coverage for deeper structures (such as hippocampus, amygdala, and insula) and deep sulci. Depth electrodes in various lengths and number of contacts are implanted using conventional stereotactic technique or by the assistance of stereotactic robotic devices through 2.5 mm diameter drill holes. Risk of infection and intracranial hemorrhage have been reported in 1% of patients. In other series small hemorrhages have been reported in 5.5%, of whom only 0.9% required surgery, and no mortality was reported [33, 34]. The planning of SEEG implantation requires formulating precise hypotheses about the possible epileptogenic zone, seizure onset, and propagation zones to be tested.

4.3 Neuroimaging studies

Advances in brain imaging technology have substantially improved seizure localization and surgical outcome [30]. All patients with clinical and evidence of focal-onset seizure should have brain imaging, including brain MRI [24].

4.3.1 Brain MRI

MRI is an important noninvasive tool for evaluation of patient with epilepsy that provides two critical data, a potential epileptogenic brain abnormality, and its surrounded anatomy. Whole brain coverage allows for the examination of the lesion location and its relationship to cortical eloquent areas [35]. The main role of brain MRI is to define structural abnormalities that may cause seizure. A high-resolution brain MRI, with epilepsy protocol, is recommended. Epileptogenic lesions in almost one half of those presenting with new-onset seizure are detected. Sensitivity of brain MRI is increased by using epilepsy protocol [36]. The common sequences used by most epilepsy centers include thin section of 1 mm coronal oblique T1 gradient echo, coronal oblique T2 series, high-resolution 3D sequences (sensitive to subtle cortical dysplasia or small tumors), and T2 flair (fluid attenuated inversion recovery) images performed on 3 Tesla or higher MRI systems [24].

4.3.2 Functional brain MRI (fMRI)

The functional MRI acquisitions are based on the blood oxygen level dependent (BOLD), the main target is to confirm language hemisphere dominance in patients that are candidates for possible temporal lobe resection. It has been widely used in the context of pre-surgical evaluation in refractory seizure, is a noninvasive procedure, and is able to provide more precise localization and more cost-effective than the invasive Wada Test (with intracarotid amobarbital injection [25]. According to the American Academy of Neurology, fMRI, it is part of the guideline for language and memory function and should be predicting postoperative verbal memory outcome [37]. It can be also used in mapping of primary motor, somatosensory cortex, or visual cortex, which is useful for certain tumors, gliosis, or focal cortical dysplasia, with close relationship to eloquent area [31]. EEG-fMRI may yield complementary information within the pre-surgical evaluation for patients with possible surgical treatment [36].

4.3.3 Magnetoencephalography (MEG)

MEG is a noninvasive functional neuroimaging method for investigating electrical neuronal activity of the living human brain by using sensors positioned around the head [38] that measure fluxes in the magnetic field caused by the same brain electrical activity, which is excellent for spatial and temporal resolution and is complementary to scalp EEG [38]. However, certain limitation for the wide use of MEG is due to signal disturbance caused by subject motion and the high maintenance cost [38].

4.3.4 Optic-magnetoencephalography (OP-MEG)

This method functions through optic-pumped magnetometers and high Tc SQUIDs and does not require a thermal isolation [39]. It allows to move the subject naturally while recording long-term OP-MEG recording akin to EEG telemetry, which is especially useful in pediatric epilepsy [39].

4.3.5 Single photon emission computerized tomography (SPECT)

This functional neuroimaging modality is based on a radioactive tracer, imaging hardware, and data analysis software. ^99mTc-ethyl cysteinate dimmer [ECD or Neurolyte^R] or ^99mTc-99 m HMPAO SPECT is used for measurement of regional cerebral blood flow in vivo [40]. Compared with ¹⁸F-FDG PET, brain SPECT has inferior spatial but superior temporal resolution, allowing identification of onset-zone and increased neuronal activity during the ictal phase, which is associated with increased metabolism and regional cerebral blood flow (RCBF) [40]. SPECT can also detect additional abnormalities in regions without structural abnormalities.

4.3.6 Fluorodeoxyglucose positron emission tomography (FDG-PET)

FDG-PET is an indirect marker of neuronal energy metabolism by measurement of glucose consumption. PET is obtained during the inter-ictal phase because cerebral uptake of FDG occurs over 30–40 minutes after injection and represents the imaging consumption of cellular metabolic process during the uptake period. The prolonged cerebral metabolic uptake makes FDG inappropriate for measuring rapid neural events considering the average seizure duration of 1–2 minutes; thus, ictal FDG PET is not usually feasible [40, 41]. Epileptogenic foci of inter-ictal TLE and extra TLE are associated with the area of reduced glucose metabolism that usually extends beyond the seizure-onset zone [30].

4.3.7 Hybrid PET/MRI

PET/MRI has shown improved diagnostic yields in detecting potential epileptogenic lesions in patients with refractory seizures presenting for possible epilepsy surgery [42]. The choice of tracer depends on the physiological process of interest such as oxygen consumption, glucose metabolism, or cerebral blood flow. The sensitivity of detecting unilateral temporal lobe hypometabolism by inter-ictal FDG-PET/MRI is 70–80% [42].

4.3.8 Non-FDG PET procedures (neurotransmitters)

4.3.8.1 Central benzodiazepine receptors

Carbon-11-labeled Flumazenil (FMZ) is a specific reversible bound antagonist to the central benzodiazepine site on the GABA_A receptor complex. GABA receptor binding and FMZ uptake are reduced in the epileptogenic foci, and the seizure onset zone has narrower distribution than the corresponding area of FDG hypometabolism. However, an area of focal decrease of FMZ uptake in the cortical regions remote from the primary focus may occur complicating localization of epileptogenic focus [43].

4.3.8.2 Dopamine receptors

Awareness of dopamine role in the pathophysiology of focal epilepsy is growing after the discovery of dopamine receptors (D1, D2, D3); D1 more pro-convulsant, and D2s have anticonvulsant effect [4, 44]. PET studies with the high affinity of D2/D3 receptors radioligand ¹⁸F-fallypride have shown that D2/D3 receptor levels are significantly decreased in the epileptogenic temporal lobe in all patients, including the temporal pole and lateral temporal region in patients with TLE and hippocampal sclerosis, which inhibit decreased FDG uptake. These findings suggest that dopaminergic system is part of the endogenous anticonvulsant mechanism that prevents generalization of the seizure [44].

4.3.8.3 Serotonin receptors

PET radio-tracer ¹¹C-α- methyl- tryptophan (AMT) is used for quantification of serotonin synthesis in the brain, which is increased in cortex epileptic area and reveals the epileptogenic focus in the inter-ictal state [44].

4.3.8.4 Opioid receptors

Functional imaging with PET radiotracers yields quantitative measurement of opioid binding mediated by μ, γ κ, opioid receptors. This has not demonstrated specific changes in TLE idiopathic epilepsy [4].

5. Surgical approaches and techniques

Surgery is an effective procedure for many patients with drug-resistant epilepsy (DRE). Unfortunately, surgery is underutilized; in the USA, <1% of patients with DRE are referred to epilepsy centers [45]. The common reasons for rejections include fear of complications, expenses, reservation about benefits, and misconceptions held by non-specialist physicians [6, 45]. The delay from onset of epilepsy to surgery averages >20 years resulting in impaired social and educational development [8]. Early surgery provides the best opportunity for seizure remission minimizing adverse social and psychological consequences and premature death [6].

Surgical therapy can eliminate disabling seizures completely in appropriately selected patients [46]. Multiple surgical procedures, including recent minimally invasive procedures, are now offered depending on the type of epileptic seizures and their presumed underling pathology (Table 3).

Table 3.

Epilepsy surgery, modalities according to the preoperative evaluation.

5.1 General anesthetic consideration

Perioperative anxiety should be managed with premedication in the preoperative area. Midazolam has been demonstrated to be effective in relieving anxiety. If intraoperative ECoG is planed the dosage of Midazolam and Benzodiazepines should be reduced to minimize their depression effect on ECoG intraoperatively [47].

Anesthesia can be induced with thiopental or Propofol; these drugs rapidly induce the unconsciousness nondepolarizing muscle relaxant in the administration after induction of general anesthesia [47].

5.2 Awake anesthesia in epilepsy surgery

The golden rule in epilepsy surgery is to resect the epileptogenic zone, with preservation of the neurological function [48]. The epileptogenic zone often shows no image abnormalities (FCD type I) in foci often located in functional areas, such as language or somatosensory areas. In such situations, awake surgery may be effective [48]. The main advantage of awake anesthesia is the clinical functional mapping intraoperatively under awake condition allowing the identification of eloquent area with close monitoring during resection. Positioning the patient is critical for the success of the technique “Asleep-awake-asleep.” Propofol does not interfere with ECoG monitoring when disconnected 20 minutes before [24]. Certain disadvantages have been reported, such as spatial limitation of craniotomy, limited intraoperative time, inability to have ictal recording, and limitation of the patient cooperation in young patients or patients with psychological or psychiatric disorders. However, in awake anesthesia, it is imperative to psychologically prepare the patient for the procedure. It has been reported that under awake anesthesia, patients with epileptogenic foci, close to functional area, may have improved seizure control and minimal neurological complications, through intraoperative mapping information and ECoG [48].

5.3 Implantation of strip, grid, and depth electrodes

Invasive monitoring tests are indicated in cases of nonconclusive noninvasive tests, with unclear lateralization or localization of the epileptogenic zone. Invasive monitoring is not an exploratory procedure, but it is a complementary test for lateralization and localization of the epileptogenic zone [49]. The most common electrodes are subdural strip and grids. Epidural electrodes are available but not widely used. The ideal electrode should be selected by epileptologist, epilepsy surgeon, and neurophysiologist together after review of the all patient data. Invasive electrodes are frequently placed bilaterally, if lateralization is unclear [50]. Strip and depth electrodes are useful for lateralization of seizure onset while grids are more helpful in localization.

Depth electrodes are more valuable for assessment of deep cortical structures, such as amygdala, hippocampus, insular, cingulum, and bifrontal cortex. They have multiple contact arrays, up to 12 nickel-chromium, or platinum contact and are commonly in use for bi-temporal mesial sclerosis, and it could be used in combination with subdural electrodes [51].

Subdural strip (Figure 2a–c) and grids (Figure 2d) are fine structures covered by silastic or Teflon sheets embedded in nickel-chromium or platinum. Each electrode is 2–4 mm contact in diameter with inter-electrode distance being generally 10 mm. Subdural grids are larger plates of rectangular arrays with several parallel rows up to 64 electrodes. It is an excellent choice for covering large cortical area to record inter ictal and ictal epileptogenic activity. Cortical stimulation and grid mapping are valuable to delineate the functional area and epileptogenic zone [52]. The most common complications are subdural hematoma, up to 16%, cerebral edema, (2–14%), CSF leakages (19–33%), brain edema, 2–14%, hemiparesis, 1.5% [52].

Figure 2.

Intracranial monitoring; a & b; skull x-ray images, showing temporal bilateral intracranial strip electrodes, c; brain CT, axial cut, bone window technique, showing bilateral temporal intracranial electrodes. And d; intraoperative view showing over the brain cortex, 2 grids (4x8 &4x5 electrodes).

5.4 Resective surgery for temporal lobe epilepsy

5.4.1 Temporal lobe epilepsy (TLE)

TLE is the most common focal seizure disorder in adults with mesial temporal sclerosis being the most common pathological entity (10). Several varieties of techniques have been described, including tailored temporal lobectomy, anteromesial temporal lobectomy, trans-Sylvian amygdalohippocampectomy, and temporal lobe lesionectomy [53]. The overall seizure freedom following temporal lobectomy for epilepsy has been reported to be between 74 and 82%, with best outcome for temporal lobe neoplasm (88–92%), followed by patients with MTS (70%), and the poorest control for cortical dysplasia [54].

Resection of the anterior temporal lobe, the amygdala, and part of the hippocampus is the most commonly performed resective epilepsy surgery [55]. The posterior margin of resection is 4.0–4.5 cm in the dominant side, and 5.0–5.5 cm in the non-dominant hemisphere, in order to minimize the speech and visual deficits and an en bloc resection of the amygdala, hippocampus, uncus, and fusiform gyrus [56]. Schram et al. [57], in a review study of 53 scientific papers, showed no differences between temporal lobectomy and selective amygdala hippocampectomy. Nevertheless, neurophysiological outcome was significantly better in selective amygdalohippocampectomy [57]. Approximately 25% of patients will develop some degree of memory impairment after temporal lobe lobectomy (Figure 3) [55].

Figure 3.

Brain MRI T1-weighted image, axial slide showing postsurgical temporal lobectomy and amygdalohypocampectomy.

5.4.2 Resective surgery in extra-temporal lobe epilepsy

Focal lesion, lobar or multilobar resections can be undertaken in the frontal, occipital, and parietal lobes with the expectation of curing or improving seizures. Lesions in eloquent areas of speech, language, and motor function may not be suitable for resection given the postoperative implications of this procedure. In areas adjacent to motor and somatosensory cortex, intraoperative neuro monitoring may be required. The outcomes of non-lesional or MRI-negative resections are less successful [55].

Most patients with extratemporal resections will have invasive electrode recordings because the epileptogenic zone is often not as well defined as in temporal-lobe epilepsy. The outcome for extra temporal-lobe resections is in the region of 60% [52]. Predictors of success include a greater extent of surgical resection, structural pathology on MRI, and concordant structural and electrophysiological imaging. For patients with cortical dysplasia, seizure freedom outcomes are reported to be in the region of 40–70% and are inversely related to the length of follow-up [58]. The best postoperative outcome is associated with type 2B focal cortical dysplasia [58].

Frontal lobe resections account for up to 30% of cases and carry a 1-year seizure remission rate of approximately 45% (range 21–61%) and less durable long-term outcomes [59]. The EZ frequently extends beyond MRI-defined lesions, and the resection may need to be tailored according to invasive EEG findings [52]. The best postoperative outcome is associated with type 2B focal cortical dysplasia, a focal seizure onset, and total resection of the EZ.

In insular resection, seizure remission after resection of insular tumor is in the range of 74–84% [60], and insular resection with non-lesional requires a meticulous analysis of the risk–benefit ratio.

Usually parietal seizures are associated with lesional areas; seizure freedom ranges between 45 and 78%, with the best being associated with a focal MRI lesion [61]. Occipital lobe resection seizure freedom averages 65% (range between 52 and 100%) [61]. While occipital lobe epilepsy surgery carries significant risk of postoperative visual dysfunction, seizure freedom is less than that of frontal and parietal lobe.

5.4.3 Functional hemispherectomy

When the EZ is extensive in one hemisphere, hemispherectomy, or functional hemispherectomy, may be considered. Generally, this is restricted to individuals who have a hemiparesis with loss of meaningful hand function [6]. Seizure freedom occurred in73% of patients. Most patients who are walking prior to surgery remain so afterward, whereas cognitive outcomes are usually stable, with language functions having developed in the contralateral hemisphere (Figure 4) [6].

Figure 4.

Intraoperative functional hemispherectomy view showing the middle cranial fossa after removal of the temporal lobe(a), falx cerebri (fc) and corpus callosum(cc) after removal of the parietal lobe, temporal lobectomy and disconnection of the frontal lobe(f), and occipital lobe(o).

6. Palliative treatment

6.1 Corpus callosotomy

This is a palliative interhemispheric surgical approach for PR patients, which consists of surgical disconnection of the corpus callosum to disrupt synchronization of epileptiform discharges between both brain hemispheres. It is indicated in multifocal ictal onset where seizures emanate from bilateral cortical foci at different frequency. Indication could also include failed or poor response with VNS implantation [62].

Concerning the outcome, several studies have suggested that complete corpus callosotomy is more effective, and its efficacy is sustainable with less relapse rates compared with anterior 2/3 corpus callosotomy (88% versus 58% in pediatric patients, [63]. Drop attacks improved from corpus callosotomy more than other generalized seizure types. Transient disconnection syndrome was significantly more likely in total corpus callosotomy than in anterior 2/3 corpus callosotomy [63, 64].

The decision whether to perform anterior 2/3 corpus callosotomy versus complete/total corpus callosotomy is based on the degree of cognitive impairment and developmental delay and is also guided intraoperatively in some subjects by the presence of EEG activity desynchronization or transformation of generalized epileptiform discharges to asynchronized (lateralized) epileptiform discharges during the surgical course of the corpus callosotomy procedure (Figure 5) [64].

Figure 5.

Corpus callosotomy; (a); brain MRI, coronal section T1 weighted, showing the disconnection of the corpus callosum (one row), (b); brain MRI sagittal section T2-weighted showing anterior 2/3 of the disconnection of the corpus callosum (2 rows).

6.2 Vagus nerve stimulation

This is a type of stimulation of the vagus nerve on a set schedule. It is a well-established palliative treatment although it seems that VNS is unlikely to offer a substantial advance in epilepsy surgery. Current evidence points toward a deactivation of the nucleus of the solitary tract with widespread projections to the dorsal raphe nucleus, locus coerules, thalamus, hypothalamus, amygdala, and hippocampus [18]. A metanalysis has demonstrated a 44.1% decrease in seizure frequency with a follow-up of more than 3 years (Figure 6) [65].

Figure 6.

Vagus nerve insertion, a: Vagus nerve dissection, b: Electrode insertion, around the VNS, c: The lead of the device fixed to the muscle, and introduced subcutaneous to the infrascapular area, d: Intraoperative cheek of the stimulator, before subcutaneous insertion.

6.3 Responsive neuro-stimulator (RNS)

RNS is a programmable neurostimulator, which is cranially implanted and connected to one or two depth and/or subdural cortical strip electrodes to detect the onset of seizure and stop it as it occurs. It is now gaining a major position in the USA as a new device with closed-loop system [66]. It is more effective in focal onset seizure, with 75% of seizure reduction [67].

6.4 Deep brain stimulation

This approach has attained approval as adjunctive therapy for refractory seizure in EU and USA [68]. Bilateral stimulation of the anterior nuclei of the thalamus (ANT) for epilepsy is indicated as an adjunctive therapy for reducing the frequency of seizure in adult patients with partial onset seizure with or without secondary generalized seizure [68]. The long-term data showed a median seizure reduction at 1 year of 41% increasing to 69% after 5 years [68]. The mechanism of action remains unknown. These modalities of adjunctive procedures are eventually available in most developed countries [69].

6.5 (t-VNS)

Transcutaneous vagus nerve stimulation has been proposed as an alternative method for the treatment of various psychiatric disorders. The application of this device in refractory seizures is still not conclusive concerning the efficacy, and future trials are needed [70].

6.6 Ablative procedure

There are multiple varieties of ablative procedures for refractory seizures, including:

6.6.1 SEEG-guide radiofrequency thermos coagulation

SEEG-guide RF-TC has been developed and expanded [71]. A single or multiple lesionotomy by coagulation should be performed between continuous electrode contacts, with progressive increasing in power till the impedance suddenly changes indicating that thermos coagulation has occurred. The best response was observed in patients with periventricular heterotopia, results showed pooled seizure free rate of 23% and response rate of 58% [71].

6.6.2 Laser interstitial thermal therapy (LITT)

LITT is also a stereotactic laser ablation (SLA). The laser applicator sheath is placed and lesionotomy performed with precise imaging technique using MRI. The ablation is carried out under continuous monitoring of MRI thermal image near real time. In MTLE, improvement of 58% was achieved according to Engle scale I outcome after 1 year [72].

6.6.3 Radiosurgery

Radiosurgery has been used for (MTLS), gelastic epilepsy associated with hypothalamic hamartomas and epilepsy with vascular malformation, and it has demonstrated a decrease in seizure frequency in MTLS, hamartomas of the hypothalamus, and AVM. Delayed therapeutic effect must be considered in treatment decision [73], and recently, it has been shown that radiosurgery of corpus callosum may in some cases result in seizure reduction [74].

7. Current surgical challenges and the future

Despite the advances in technology and surgical technique, control of seizure in the best centers ranges between 15 and 75% and has no changed for the last three decades [75]. Mechanism of epileptogenesis may involve epileptic network rather than a single focus. Using the depth electrodes, independent ictal onset with same semiology from the hippocampus, entorhinal cortex, and amygdala in the same patient with medial temporal lobe epilepsy can be detected. It is known that unidentified epileptogenic source happens in 50% of patients treated surgically [23].

It is known that 10–15 years after surgery, seizure control declines with only 15–50% remaining seizure free [8].

In cases of more than one pathology, such as MTLS, migratory disorder, or tumor, seizure is usually not controlled unless sources resection is addressed [76].

Neuropsychological tests demonstrated cognitive deficit in most patients without tumors or cavernomas [77]. Epileptogenic network is partially supported by Microanalysis (elevation of glutamate levels in epileptogenic region and greater increase after ictal onset, 7 T-MR spectroscopy detecting the mitochondrial function in cortico-subcortical area) and electrophysiological connectivity [24].

Open-loop device for constant stimulation of the anterior thalami (not approved by FDA) could control the epileptogenic site and the lesional area, reducing the seizure in less than 50% of patients in a clinical trial [78].

For better understanding of the epileptogenic network, we need bioelectric integrated telemetered intracranial monitoring. The next advances will need a molecular biosensor with wireless transmission of critical data. Medicine and surgery are not able to control seizure in all epileptic patients yet.

The new surgical roadmap comprises enhancing research, through collaboration, bioinformatics, information scientist biomedical informatics, and information technology.

8. Conclusions

Despite the recent advances in the technical aspects of surgical treatment and the diagnostic approaches in PR patients, outcome concerning seizure freedom remains within the same range as during the last decades. Resective and disconnecting surgery may have reached the ceiling of their possibilities, and alternative additional procedures are needed to achieve better outcome. Scientific knowledge about the epileptogenic zone is still evolving and will be the main key for the treatment of PR patients and seizure outcome.

Prospective multicentric studies are needed for the application of new diagnostic procedures, surgical techniques, including the minimally invasive ones and response neurostimulator for the state-of-the art epilepsy surgery.