SEARCH

SEARCH BY CITATION

Keywords:

  • Epilepsy;
  • EEG;
  • Cortical stimulation;
  • Intracranial recording

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Purpose: The usefulness of single-pulse electrical stimulation (SPES) during intracranial recordings was evaluated in a pediatric population. This method is useful in identifying epileptogenic cortex in adult subjects.

Methods: We studied 35 children who were undergoing intracranial electroencephalography (EEG) recordings from two hospitals (King’s College Hospital and Great Ormond Street Hospital for Sick Children, London, United Kingdom). In each patient we studied all available contacts using a series of 10 or more single, brief (1 ms) electrical stimuli. The cortical responses were reviewed in detail. The data were examined for associations between response type, ictal onset zone, lesion boundary, and seizure outcome.

Results: We identified cortical responses to SPES that were similar to those reported in adults. In agreement with previous studies we found that two types of responses (“delayed” and “repetitive” responses) were associated with the ictal onset zone and the area of the presumed epileptogenic lesion. When these responses were present (54% of cases), the removal of the entire area responsible for the abnormal responses to SPES was associated with good outcome.

Conclusion: Cortical responses to SPES in children provide new and additional information in the investigation of epileptogenic cortex in children during assessment for epilepsy surgery. This may improve the outcome for this difficult but important group.

Single-pulse electrical stimulation (SPES) is a technique for identifying abnormal cortical excitability by observing the electrographic responses to brief (1 ms) electrical stimuli during subdural recordings in patients with epilepsy. This is interesting because seizure generation most likely involves an abnormal balance between excitation and inhibition (Arzimanoglou et al., 2003; Binnie et al., 2003), and, therefore, close inspection of the different types of cortical response to SPES may provide insight into the epileptic process.

Three types of responses to SPES have previously been reported in patients with epilepsy who were undergoing intracranial electroencephalography (EEG) studies (Valentin et al., 2002, 2005a). The most common response to SPES is a sharp wave, often followed by a slow wave, immediately after the stimulus. This response type, called the “early response” (ER), is ubiquitous, occurs after most stimuli at most electrode sites, and the amplitude of the response varies with stimulus intensity. The ER appears to reflect normal cortical excitability and is not related to the epileptogenicity of the underlying cortex. It does, however, provide information about cortical connectivity (Lacruz et al., 2007) because ERs that arise at increasing distances from the stimulating electrodes have appropriate increases in latencies. This probably indicates that stimulation activates cortico-cortical connections in the populations of underlying neurones. Furthermore, the ER appears to correspond to responses noted by other authors using similar paradigms (Brazier, 1964; Rutecki et al., 1989; Wilson et al., 1990; Buser et al., 1992; Matsumoto et al., 2004; Catenoix et al., 2005).

Two other types of responses to SPES appear to reflect abnormal cortical excitability. The “delayed response” (DR) was the first abnormal response to SPES discovered (Valentin et al., 2002) and has the form of an all-or-nothing sharp wave or spike, resembling an epileptiform discharge, occurring later than 100 ms after stimulus (although not necessarily after every stimulus). The sites where DRs arise have been demonstrated to correspond with the areas of seizure onset (Valentin et al., 2002, 2005a), and their removal is correlated with good surgical outcome (Valentin et al., 2005b). Like the ERs, the DRs may arise at some distance from the stimulating electrodes, and again this implies the activation of cortico-cortical connections. The long latency (100 ms–1 s) between stimulus and response may indicate multisynaptic connections. In animal studies, similar examples of long-latency epileptiform field potentials have been observed in response to single stimuli in slice preparations treated with agents that modify γ-aminobutyric acid (GABA) (Chagnac-Amitai & Connors, 1989a, 1989b; Empson et al., 1993).

The third response type to SPES is called the “repetitive response” (RR) and has the form of a successive repetition of the initial ER, often with a broad field and typically lasting for a second or longer. RRs are also markers of epileptogenic cortex because the stimulation site corresponds with the areas involved in seizure onset and their removal is associated with a good surgical outcome (Valentin et al., 2005a). RRs are different to DRs in both their waveform and their interpretation, and these differences may imply a different pathophysiologic mechanism. The fact that the RRs were often widespread (even bilateral) and yet were elicited by a focal stimulus implicates bigger “loops,” perhaps even thalamocortical loops. There is evidence from animal studies that some cases of “generalized” discharges (particularly those arising from secondary bilateral synchrony) originate from discrete cortical foci (Steriade & Contreras, 1998; Meeren et al., 2002, 2005), and we hypothesize that RRs represent that phenomenon.

Cortical responses to SPES, therefore, appear to provide useful information that is independent and different from the spontaneous epileptiform activity routinely acquired during intracranial monitoring. Unfortunately, although spontaneous epileptiform activity provides important information, there are some fundamental limitations with those data. Spontaneous interictal discharges provide information about the irritative zone, but this may not correspond well with the epileptogenic zone (Ojemann & Engel, 1987; Alarcon et al., 1994; Spencer et al., 2007). Furthermore, although spontaneous seizures provide direct information about the epileptogenic zone, in some patients seizures are infrequent, and there may be none during the monitoring period. SPES, on the other hand, does not rely on spontaneous activity and provides a method for interactively investigating abnormal cortical excitability.

To date, SPES studies have focused on adult subjects (most of whom have temporal lobe epilepsy); however, children with focal epilepsy present a more complicated problem with a wider variety of epileptogenic pathologies and where there may be a substantial conflict between patient safety and early surgical intervention. The behavior of some children (because of their age or other cognitive issues) may place them at higher risk of complications if they are subjected to prolonged periods of subdural monitoring. On the other hand, postponement of surgical intervention relegates them to an extended period of epilepsy and antiepileptic medication at a crucial stage of neurodevelopment. An early, tailored, surgical resection can circumvent these risks (Duchowny, 1989; Shields, 2004; Freitag & Tuxhorn, 2005; Sperli et al., 2006) and at the same time take advantage of the remarkable facility of children to recover significant function after early cerebral damage (Holloway et al., 2000; Hertz-Pannier et al., 2002; Liegeois et al., 2004; Gleissner et al., 2005; Kulak et al., 2006; Yuan et al., 2006). Therefore, SPES may be particularly useful in the investigation of some children with focal epilepsy.

In the current study we evaluate SPES in a pediatric population undergoing subdural EEG monitoring. These patients were recruited from the epilepsy surgery programs of two hospitals, and the results were acquired independently by neurophysiology staff from each hospital. This population of patients differs significantly from the predominantly adult populations previously reported. It differs with respect to its age distribution as well in the types of pathology and in the distribution of epileptic foci.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Subjects

Children who underwent intracranial recordings as part of their assessment in the epilepsy surgery programs at Great Ormond Street Hospital for Sick Children (London, United Kingdom) and King’s College Hospital (London, United Kingdom) were recruited for this study. We conducted studies on 35 children (18 males and 17 females). The median age was 14 years 2 months (with a range between 0 years 9 months and 17 years 7 months). Patient details are provided in Table S1. Patients were usually studied for about 7 days. In four patients the studies were extended for up to 10 days in order to capture infrequent seizures, and on two occasions the available study period was limited because of postoperative complications. SPES studies were conducted during interictal periods, although this could be difficult in those patients with frequent seizures. Studies were generally conducted while the patient was quiet and awake. If a seizure occurred, the study would be suspended until the patient recovered fully. Antiepileptic medications were sometimes modified according to clinical needs, and those decisions were not influenced by the patients’ inclusion in our study.

Lesion identification

Structural lesions were initially identified using magnetic resonance imaging (MRI) and classified by a consultant radiologist. Sixteen patients had lesions in the frontal lobe, eight in the parietal lobe, four in the occipital lobe, and 10 in the temporal lobe, and three had no clear lesions on neuroimaging. Twenty-six patients had lesions restricted to a single lobe and six patients had lesions involving two lobes: frontal and temporal in one, parietal and temporal in one, frontal and parietal in one, parietal and occipital in one, and occipital and temporal in two patients. In patients who underwent resective surgery, the presence and nature of lesions were confirmed with neuropathologic examination. Findings are summarized in Table S1. In patients with a single lesion, the presumed epileptogenic lesion in each patient was defined as the lesion identified by neuroimaging during presurgical investigations. In patients in whom there was more than one lesion, the presumed epileptogenic lesion was defined as the lesion closest to the ictal onset zone identified by intracranial recordings (see subsequent text).

Electrode placement

Subdural or intracerebral (depth) electrodes (described in detail elsewhere; Valentin et al., 2002) were placed under MRI guidance in accordance with the targets suggested by clinical history; seizure semiology; neuroimaging; scalp EEG; and sometimes neuropsychology, motor, and language functional MRI (fMRI), single photon emission computed tomography (SPECT), or positron emission tomography (PET). Subdural grids (used in 25 patients), subdural strips (30 patients), or depth electrodes (9 patients) covered a number of areas/lobes in each patient. Thirteen patients had electrodes implanted over the right hemisphere, 13 had electrodes implanted over the left hemisphere, and nine had electrodes implanted bilaterally. The frontal lobes were studied in 30 patients, the parietal lobes in 23, the occipital lobes in 9, and the temporal lobes in 27 patients. The number of electrodes implanted in each patient varied between 17 and 78 electrodes (the median number of electrodes implanted was 54).

Ictal onset zone

Ictal EEG data were reviewed by experienced clinical neurophysiologists, and reported without knowledge of the findings of SPES. The ictal onset zone was defined as the area that showed the first clear ictal electrographic change consistent with the beginning of a seizure. Such a change consisted of rhythmic sharp waves, sharp-and-slow wave complexes, spike-and-slow wave complexes, rhythmic delta or theta activities, sharpened delta or theta activities, regular spikes, or low-amplitude high frequency activity in the beta range. Ictal onset was arbitrarily classified as (1) focal if three or fewer adjacent electrodes were involved at onset, (2) regional if more than three electrodes in the same lobe were involved at onset, or (3) nonlocalized if EEG changes at ictal onset were bilateral, involved more than one lobe, arose from a diffuse electrodecremental event with no superimposed focal features, or if the clinical features of the seizure preceded the electrographic change. Focal and regional ictal onset zones were subclassified according to the lobe where the seizure began (i.e., frontal, parietal, occipital, or temporal).

Experimental protocol for SPES

SPES was performed in parallel with our standard video-EEG acquisition. SPES was conducted using a constant-current neurostimulator (Medelec ST10 Sensor, Oxford Instruments, Surrey, United Kindgom; or Digitimer Model DS2A, Digitimer Ltd., Welwyn Garden City, Hertfordshire, United Kingdom). The experimental protocol has been fully described elsewhere (Valentin et al., 2002). In summary, it involves repeated (one every 5–10 s) application of a brief duration (0.6 ms–1.0 ms) monophasic pulse of a current intensity between 4 and 8 mA. Typically, twenty individual pulses (10 of each polarity) were applied to every available pair of adjacent electrodes in contact with cerebral cortex (as well as the hypothalamic hamartoma in patient no. 26).

The on-line EEG was visually monitored, and normal and abnormal responses to SPES were easily identified during the recording session. In this way it was possible to get a fair indication of the responses to SPES in real time. If the results obtained after 10 stimuli were ambiguous, then repeated trials of 10 stimuli were applied until a clear result was obtained. After the data were acquired, the results were formally analyzed off-line using a one-tailed sign test (Valentin et al., 2002). The experimental procedure was approved by the ethics committees of both hospitals (Great Ormond Street Hospital for Sick Children R&D No. 05NR07 and King’s College Hospital ethics reference no. 99–017).

Identification of cortical responses

Cortical responses to SPES were reviewed and characterized for each stimulation in each patient. Briefly, the three types of responses to SPES that have been previously described were also identified in the present study: that is, early response (ER), delayed response (DR), and repetitive response (RR) to SPES (Valentin et al., 2002, 2005a).

Statistical analysis

The association between stimulation and DR was established by comparing the occurrence of spikes during 1 s before and 1 s after each pulse. It was assumed that spikes were related to stimulation if the number of stimuli showing spikes during the second following stimulation was greater than the number of stimuli showing spikes during the second prior to stimulation with p < 0.05 (one-tailed sign test). Data analysis was carried out with the Statistical Package for Social Sciences 10.0 (SPSS Inc. Chicago, IL, U.S.A; 1999).

Surgical outcome

Patients were routinely followed up as part of ongoing management after surgery. Surgical outcome was assessed according to the Engel classification system (Engel, 2001). We considered grades I and II as good outcomes and grades III and IV as poor.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Ictal onset zone

Seizures were captured in all patients (between two and >200 seizures captured in each patient). Ictal onset was focal in 4 patients, regional in 16, and nonlocalized in 15. Ictal onset involved the frontal lobe in 13 patients, the parietal lobe in 14, the occipital lobe in one, and the temporal lobe in 14 patients. Ictal onset was restricted to one lobe in 17 patients, to two lobes in 9 patients, to three lobes in 2 patients and there was a diffuse onset in 5 patients. Furthermore, in the patient with the hypothalamic hamartoma, the seizure onset was identified in the hypothalamus as well as diffusely in the frontal lobe, and in one patient (Patient 20) the ictal onset was contralateral to the side of the presumed lesion.

Relation between the ictal onset zone and the presumed epileptogenic lesion

Overall, ictal onset was often close to, or overlapped with, the lesion location (in patients for whom there was more than one lesion on imaging, this refers to the smallest distance from the ictal onset to the nearest lesion). Among the patients with an abnormality identified on preoperative imaging (n = 32), the ictal onset zone either overlapped with or came to within 1 cm of the lesion boundary in 24 patients (75%), and in one other patient the regional ictal onset was 2 cm from the lesion boundary. It should be noted, however, that 18 patients had nonlocalized ictal onset zones, and so in those cases an overlap between lesion and ictal onset zone is of uncertain usefulness.

Types of cortical responses elicited by SPES

We identified cortical responses similar to those reported previously (ERs, DRs, and RRs; Valentin et al., 2005a, 2002). In addition, we identified a new type of response: the “stable response” or SR (see description in subsequent text). As with all previous studies, SPES did not elicit any seizures in our patients. In 25 of our patients we identified either DRs, RRs, or SRs and the relationship between the ictal onset zone; the limits of the presumed epileptogenic lesion and responses to SPES (omitting ERs) in those patients are presented in Table 1.

Table 1.   Relationship between responses to SPES, seizure onset, and lesion boundary (in only those patients who showed DRs, RRs, or SRs)
Patient numberDelayed responses (DRs)Repetitive responses (RRs)Stable responses (SRs)
Relationship to ictal onsetRelationship to lesionRelationship to ictal onsetRelationship to lesionRelationship to ictal onsetRelationship to lesion
  1. contra, contralateral; OL, overlap with ictal onset zone or lesion; no DR, there was no delayed response; no RR, there was no repetitive response; no SR, there was no stable response; Distant, refers to responses arising more than 2 cm from the ictal onset zone or lesion; smaller, the area of the response is smaller than the area of ictal onset; bigger, the area of the response is bigger than the area of ictal onset.

  2. apatients had diffuse or nonlocalized seizure onsets.

  3. bpatient had DRs on the side of presumptive abnormality; however, the ictal onset was contralateral to the presumed area of interest.

  4. Distances are measured from the contact nearest the area of interest (either the seizure onset or the lesion).

 1Distant >5 cmno lesionno RRno lesionno SRno lesion
 2no DRno DROL/smaller<2 cmno SRno SR
 3OL/smaller<1 cmno RRno RRno SRno SR
 4OL/smaller<1 cmno RRno RROL/biggerOL
 5no DRno DRno RRno RROL bigger1 cm
 6Distant 3 cm<1 cmno RRno RRDistant >5 cm4 cm
 7OL/smaller<1 cmOL/smaller<1 cm1 cm1 cm
 8OL/smallerDistant 3 cmno RRno RRno SRno SR
 9no DRno DRno RRno RRDistant 3 cmDistant 3 cm
11OL/smaller2 cmOL/smaller1 cm2 cmDistant 3 cm
13no DRno DRno RRno RRDistant >5 cmDistant >5 cm
14aOL/smaller Distant 3 cmno RRno RROL/smallerDistant 5 cm
17Distant 3 cm1 cmno RRno RRno SRno SR
18no DRno DRno RRno RRDistant >5 cm2 cm
19no DRno DRno RRno RR1 cm<1 cm
20bDistant: contrano lesionno RRno lesionDistant contrano lesion
21ano DRno DRno RRno RROL2 cm
24OL/bilateral0 cmno RRno RRno SRno SR
25OLDistant >3 cmno RRno RRno SRno SR
28aOL/bigger0 cmno RRno RROL/bigger2 cm
29OL0 cmno RRno RRno SRno SR
30aOL/smaller<1 cmno RRno RRno SRno SR
31no DRno DRno RRno RRDistant >5 cmDistant 4 cm
34OLDistant 3 cmno RRno RRno SRno SR
35OL/smaller0 cmOL/smaller<2 cmno SRno SR
Early responses

ERs (Fig. 1A) were identified in all patients in most areas covered by electrodes. These responses are most commonly elicited around the stimulating electrodes, although occasionally they could arise distant from the stimulating electrodes.

image

Figure 1.   (A) An example of early responses (ERs) are demonstrated in this example, where stimulation of electrodes G015 (−) and G016 (+) results in ERs in adjacent contacts G012–14 (single arrow) with increasing latency at the more distant contacts. An example of a delayed response (DR) is also demonstrated, where this stimulation also elicited a DR that is maximal at contacts G023 and G032 (double arrow: these two contacts were close together on the anterior edge of the electrode array). This indicates that the abnormal cortex underlies the place where the response occurs (i.e., contacts G023 and G032). A representation of the electrodes on a brain map is presented at the bottom of the panel. (B) An example of a repetitive response (RR) is demonstrated in this example, where stimulation of contacts DB02 (−) and DB01(+) on a depth electrode elicits a diffuse rhythmic RR that is clearest in contacts G007, G008, G014, and G015, but has a field that includes surrounding contacts. This indicates that the area of abnormal cortex underlies the area of the stimulating electrodes (i.e., DB02 and DB01). A representation of the electrodes on a brain map is presented at the bottom of the panel, with the lesion represented as the shaded blob.

Download figure to PowerPoint

Delayed responses

DRs (Fig. 1A) were identified in 17 patients (49%), and were identified as focal in 9 patients, regional in 47, and nonlocalized in one patients. These responses were identified in the frontal lobe in eight patients, in the parietal lobe in six patients, in the occipital lobe in one patient, in the temporal lobe in four patients, and were bilateral in one patient. Most often these responses were restricted to a single lobe (13 patients).

Repetitive responses

RRs (Fig. 1B) were observed in four patients (11%). The locations of the stimulating electrodes where the RRs were elicited were identified as focal in two patients, regional in one, and nonlocalized (bilobar) in the other patient (although in that case the area where stimulation elicited RRs was located in adjacent electrodes around a lesion in the central sulcus).

Stable responses

This type of cortical response is described here for the first time (Fig. 2). SRs consist of a small spike or sharp wave, most often superimposed on the slow wave of an ER. SRs had a latency ≥100 ms. In contrast to DRs, however, SRs have a fixed latency (typically with a variation in latency of less than 20 ms). Furthermore, when present during the stimulation of a particular site, SRs arise after most or all stimuli. In this respect they resemble ERs, except that the latency is typically ≥100 ms.

image

Figure 2.   A “stable response” (SR) is demonstrated in this example, where stimulation of electrodes G007 (−) and G008 (+) consistently elicit an SR in adjacent contact G006 in the form of a slow wave and following blunt sharp wave. The latency of the sharp wave was 240 ms in the first panel, and then 250 ms in the following two panels in this example, demonstrating the relative consistency of the latency of this type of response. A representation of the electrodes on a brain map is presented at the bottom of the panel, with the lesion represented as the shaded blob.

Download figure to PowerPoint

SRs were identified in 14 patients (40%) and were identified as focal in 6 patients, regional in one, and nonlocalized in 7 patients. SRs were most common in the frontal and parietal lobes, especially around the area of the central sulcus (Fig. 3). SRs were identified in the frontal lobe in 10 of the 14 patients (71%), in the parietal lobe in 10 (71%), and in the temporal lobe in one (7%). SRs were identified in one lobe in 7 of the 14 patients (50%) and in two lobes in the other 7 (50%). For each patient showing SRs, the location of the SRs was consistent, and SRs could be elicited by stimulation of a variety of contacts. There was no clear relationship between the location of the SRs and the location of the stimulating electrodes.

image

Figure 3.   This figure presents the combined data for the distribution of “stable responses” (SRs) for all patients who demonstrated this class of response. The points marked with an “x” represent the distribution of all of the stimulating electrodes that elicited SRs. The points marked with an “o” represent the distribution of all of the electrodes that demonstrated SRs. The top panel represents combined data obtained from studies of the right hemisphere and the bottom panel represents combined data obtained from studies of the left hemisphere.

Download figure to PowerPoint

Relation between the cortical responses to SPES and the ictal onset zone

Early responses

ERs were identified in all lobes with electrode cover. They are most often seen around the stimulating electrodes when stimulating most sites. Therefore, there was no relationship between the location of the ERs and the location of the ictal onset zone.

Delayed responses

DRs were identified in areas that overlapped the area identified as the ictal onset zone in 13 of the 17 patients where DRs were observed (76%). The location of the DRs was similar to, or more localized than, the ictal onset zone in 11 (65%; although it should be noted that in three patients the ictal onset zone was nonlocalized). The area of the DRs was more extensive than the ictal onset zone in two patients. DRs were distant (>2 cm) from the ictal onset zone in four patients (24%).

Repetitive responses

Areas eliciting RRs overlapped the area identified as the ictal onset zone in all four patients (100%). In all four patients, the area eliciting the RRs was more localized than the ictal onset zone. In three of these patients, DRs were also observed. Interestingly in patient no. 2, the abnormal RRs were identified on stimulation of a depth electrode inserted through a 40-contact grid; however, we did not identify any other abnormal responses arising from stimulation of the overlying grid.

Stable responses

Areas showing SRs overlapped the ictal onset zone in 5 of the 14 patients who showed SRs (36%). The area of the SRs was more localized than the ictal onset zone in one patient (7%) and was less localized than the ictal onset zone in three patients (21%). In three patients, SRs arose within 2 cm from the ictal onset zone, and in six patients (43%) SRs were distant (>2 cm) from the ictal onset zone. Overall, there was no clear relationship between the location of SRs and the ictal onset zone.

Relation between the cortical responses to SPES and the location of the presumed epileptogenic lesion

Early responses

Because ERs were ubiquitous, there was no relationship between the areas showing ERs and the location of lesions.

Delayed responses

DRs were identified very near a lesion (either overlapping or within 1 cm) in 10 of the 15 studies in patients with DRs and lesions (67%) and within 2 cm of the lesion boundary in one patient (7%). DRs were distant (>2 cm) from the lesion in four patients (27%). There were two patients with DRs and no lesion on imaging.

Repetitive responses

Areas eliciting RRs were identified very near a lesion (either overlapping or within 1 cm) in two patients (50%) and within 2 cm in two patients (50%).

Stable responses

SRs were identified very near a lesion (either overlapping or within 1 cm) in four of the 13 patients showing SRs (31%). In three patients (23%) the location of the SRs was about 2 cm away from the lesion boundary, in six patients (46%) the location of the SR was distant (>2 cm) from the lesion, and in one patient there was no clear lesion. There was no significant relationship between SRs and lesion location.

Relation between the cortical responses to SPES and neuropathology

Postoperative neuropathologic examination revealed that DRs were identified in cases of dysembryoplastic neuroepithelial tumors (DNETs) in three patients, in cases with focal cortical dysplasia (FCD) in four patients, in cases of mesial temporal sclerosis (MTS) in two patients, in a single patient with meningioangiomatosis, and in two patients with normal neuropathologic evaluation. In five patients DRs were observed in the absence of neuropathologic data.

Postoperative neuropathologic evaluation revealed that RRs were identified in one patient with a DNET with associated calcification, in two patients with nonspecific findings, and in one patient with normal neuropathologic findings.

Postsurgical seizure outcome

Of the 35 patients studied, 29 progressed to have resective surgery in the area of the presumed epileptogenic zone and one had multiple subpial transection of epileptogenic cortex in the hand area of the motor strip. Three went on to have vagal nerve stimulators implanted, and one patient with a hypothalamic hamartoma had a partial resection and a deep brain stimulator implanted in the hypothalamus. Three patients left our care very soon after our study and were lost to follow-up. Therefore, we have post-resective surgical outcome data after 27 investigations (Table 2). The mean follow-up period is 2 years 11 months.

Table 2.   Surgical outcome
Patient numberTime of fup (years)Relationship of surgery to seizure onset zoneRelationship of surgery to lesionRelationship of surgery to areas with abnormal cortical responses (DR or RR)Engel score
  1. MST, multiple subpial transaction; No surgery, no resective surgery; fup, follow-up; NA, not applicable as patient did not show DR or RR.

  2. aVNS, vagus nerve stimulation.

  3. bplus deep brain stimulator.

 13 yearPartial removalNo lesionAll removedIV
 25 yearAll removedAll removedAll removedI
 3No surgeryNo surgeryNo surgeryNo surgeryNo surgery
 44 yearPartial removalAll removedAll removedI
 5Lost to fupMSTMSTMSTLost to fup
 61 yearPartial removalAll removedPartial removalIII
 74 yearPartial removalAll removedPartial removalIII
 84 yearAll removedAll removedNone removedIII
 94 yearAll removedAll removedNAIV
103 yearPartial removalAll removedNAII
112 yearPartial removalAll removedPartial removalI
121 yearNone removedAll removedNAIV
133 yearAll removedPartial removalNAII
142 yearPartial removalAll removedNone removedI
154 yearAll removedAll removedNAI
164 yearPartial removalAll removedNAII
173 yearPartial removalAll removedNone removedIII
182 yearAll removedPartial removalNAIV
191 yearPartial removalPartial removalNAI
20No surgeryNo surgeryNo surgeryNo surgeryNo surgery
213 yearPartial removalAll removedNAIV
221 yearNone removedAll removedNAIII
233 yearPartial removalAll removedNAI
24Lost to fupLost to fupLost to fupLost to fupLost to fup
251 yearNone removedAll removedPartial removalI
261 year bPartial removalbPartial removalbNAIV
276 yearAll removedPartial removalNAIV
281 yearPartial removalAll removedPartial removalI
292 yearNone removedAll removedAll removedII
30No surgeryaNo surgeryaNo surgeryaNo surgeryaNo surgerya
312 yearAll removedAll removedNAIV
32No surgeryaNo surgeryaNo surgeryaNo surgeryaNo surgerya
33No surgeryaNo surgeryaNo surgeryaNo surgeryaNo surgerya
34Lost to fupLost to fupLost to fupLost to fupLost to fup
353 yearPartial removalPartial removalPartial removalIV

Overall, 18 of the patients who underwent resective surgery (64%) had a “worthwhile improvement” or better (Engel class I–III), and 13 of those had a good outcome (Engel class I–II). In Table 3 we consider the relationship between postsurgical seizure outcome and the following parameters: (1) the removal of the presumed epileptogenic lesion, (2) the removal of the seizure onset zone, and (3) the removal of the areas responsible for DR and RR during SPES.

Table 3.   Summary of surgical outcome
 Surgical removalOutcome
Engel I–IIEngel III–IV
LesionAll removed119
Partial removal24
None removed00
Seizure onset zoneAll removed36
Partial removal87
None removed22
Abnormal cortical responses (DR & RR)All removed31
Partial removal33
None removed12
No abnormal cortical responses foundNot applicable59

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

We have confirmed that cortical responses to single-pulse electrical stimulation (SPES) in children are similar to the responses reported in adults (Valentin et al., 2002, 2005a). We have identified the early response (ER), delayed response (DR), and repetitive response (RR) as well as defining a new type of response that we have named the “stable response” (SR). In agreement with the earlier studies we found that the areas showing DRs and areas eliciting RRs were related to the ictal onset zone, although in our population RRs were infrequent. We found an overall correlation between ictal onset zone and the location of the DRs of 76%, which is similar to the 85% originally reported (Valentin et al., 2002). Both DRs and RRs were also correlated with the location of the presumed epileptogenic lesion. In our cohort, abnormal responses to SPES were elicited in patients with a wide variety of pathologies and the location of the abnormal cortex, as identified by SPES, followed the distribution of the lesions, regardless of their pathology. Therefore, these responses do not appear to be restricted to specific pathologies or locations, but rather reflect a general property of epileptogenic cortex. Nevertheless, abnormal responses were not elicited in all studies, even though all patients had their typical seizures during the monitoring period. We identified either DRs or RRs in 54% of our patients, and this is slightly less than that previously reported in adults (60%; Valentin et al., 2002).

SPES, therefore, provides useful information and has the advantage that it provides an interactive tool for investigating cortical excitability during interictal periods, and this provides important opportunities. SPES may identify epileptogenic cortex in patients where other data are discordant or limited (as is the case in patients with few seizures or reduced monitoring periods). Furthermore, SPES could be used in the operating theater during initial electrode placement in order to identify areas of cortex with abnormal responses, and thereby optimize electrode placement. This would be particularly useful in nonlesional cases or in cases where the identity of the epileptogenic lesion is uncertain, as can be the case with multiple lesions (e.g., tuberous sclerosis; Seri et al., 1998). A recent study of 37 patients used SPES intraoperatively (Valentin et al. in preparation) and found that abnormal responses to SPES can be seen under isoflurane or sevoflurane general anaesthesia.

If SPES is used to improve electrode placement and if it helps reduce monitoring times by providing additional information about epileptogenic cortex, then it will be particularly useful in the intracranial investigation of children with debilitating focal epilepsy. Some children will have difficulty tolerating prolonged periods of intracranial monitoring, and yet such invasive monitoring can identify surgical targets, and a successful surgical intervention will have a significant effect on their neurodevelopment potential (Duchowny, 1989; Shields, 2004; Freitag & Tuxhorn, 2005; Sperli et al., 2006).

Aside from the abnormal responses discussed earlier, we also identified two types of responses to SPES that had no relation to abnormal cortex. As reported in previous studies, the most common response to SPES was the ER (Valentin et al., 2002, 2005b). In addition to ERs, we have identified a new category of response that we have called the “stable response” (SR). Review of our previously acquired data (from adults and children) has revealed that these responses were also present in earlier studies but not reported because they were included as part of the early response due to their low amplitude and relatively blunt waveforms overlapping with the early response. SRs also appear to be a nonspecific response, with no correlation to the ictal onset zone or the presumed epileptogenic lesion. When present, SRs were usually a low amplitude sharp wave that was elicited by most stimulations and had a very consistent (although sometimes >100 ms) latency. SRs were identified in most cerebral lobes, but were more common in the frontal and parietal lobes, especially in the area of the central sulcus (Fig. 3). Identification of SRs is important because they may otherwise be misclassified as DRs, which have different clinical significance.

Finally, in the review of outcome in our cohort after resective surgery we found that among those patients who had abnormal SPES responses, 75% (three of four) had a good surgical outcome after total removal of all of the areas of cortex implicated by abnormal responses to SPES. The proportion of children showing good outcome after resection of all abnormal SPES areas is lower than that reported in a previous follow-up study carried out largely in adults (Valentin et al., 2005b). In that study, 95% of patients had a favorable surgical outcome if areas with abnormal responses to SPES were completely resected. Interestingly, when we consider the other common surgical targets (lesions and ictal onset zone) in our cohort, we find a good surgical outcome was observed in only 55% of patients (11 of 20) who had a surgical resection that included all of their presumed epileptogenic lesion, and a good surgical outcome was identified in only 33% of patients (3 of 9) who had complete removal of their ictal onset zone. Furthermore, when abnormal responses to SPES were not observed, then most patients (9 of 14; 64%) had a poor outcome.

In conclusion, we found cortical responses to SPES in children that have similar features to those identified in adults. Our study confirms that the abnormal responses to SPES (DRs and RRs) appear to correlate with abnormal cortex in the areas where seizures arise as well as with surgical outcome. SPES, therefore, provides a useful tool in the presurgical evaluation of patients being considered for epilepsy surgery. This technique has some potential. It may help delineate the epileptogenic zone and it may allow intraoperative planning to optimize positioning of subdural arrays. There is also much scope for further detailed research. SPES provides a tool for investigating in vivo cortical excitability that may shed more light on the epileptic process.

Acknowledgments

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

We are grateful to clinical and technical staff at Great Ormond Street Hospital for Children and King’s College Hospital.

We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Disclosure: The authors have no conflicts of interest to report.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information
  • Alarcon G, Guy CN, Binnie CD, Walker SR, Elwes RD, Polkey CE. (1994) Intracerebral propagation of interictal activity in partial epilepsy: implications for source localisation. J Neurol Neurosurg Psychiatry 57:435449.
  • Arzimanoglou A, Guerrini R, Aicardi J. (2003) Aicardi’s Epilepsy in Children. Lippincott Williams & Wilkins, Philadelphia, PA.
  • Binnie CD, Cooper R, Mauguiere F, Osselton JW, Prior PF, Tedman BM. (2003) Clinical neurophysiology volume 2: EEG, paediatric neurophysiology, special techniques and applications, Vol 2 Elsevier, Amsterdam.
  • Brazier MA. (1964) Evoked responses recorded from the depths of the human brain. Ann N Y Acad Sci 112:3359.
  • Buser P, Bancaud J, Chauvel P. (1992) Callosal transfer between mesial frontal areas in frontal lobe epilepsies. Adv Neurol 57:589604.
  • Catenoix H, Magnin M, Guenot M, Isnard J, Mauguiere F, Ryvlin P. (2005) Hippocampal-orbitofrontal connectivity in human: An electrical stimulation study. Clin Neurophysiol 116:17791784.
  • Chagnac-Amitai Y, Connors BW. (1989a) Horizontal spread of synchronized activity in neocortex and its control by GABA-mediated inhibition. J Neurophysiol 61:747758.
  • Chagnac-Amitai Y, Connors BW. (1989b) Synchronized excitation and inhibition driven by intrinsically bursting neurons in neocortex. J Neurophysiol 62:11491162.
  • Duchowny MS. (1989) Surgery for intractable epilepsy: Issues and outcome. Pediatrics 84:886894.
  • Empson RM, Amitai Y, Jefferys JGR, Gutnick MJ. (1993) Injection of tetanus toxin into the neocortex elicits persistent epileptiform activity but only transient impairment of GABA release. Neuroscience 57:235239.
  • Engel J Jr. (2001) A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: Report of the ILAE task force on classification and terminology. Epilepsia 42:796803.
  • Freitag H, Tuxhorn I. (2005) Cognitive function in preschool children after epilepsy surgery: Rationale for early intervention. Epilepsia 46:561567.
  • Gleissner U, Sassen R, Schramm J, Elger CE, Helmstaedter C. (2005) Greater functional recovery after temporal lobe epilepsy surgery in children. Brain 128:28222829.
  • Hertz-Pannier L, Chiron C, Jambaque I, Renaux-Kieffer V, Moortele P-FVd, Delalande O, Fohlen M, Brunelle F, Bihan DL. (2002) Late plasticity for language in a child’s non-dominant hemisphere: A pre- and post-surgery fMRI study. Brain 125:361372.
  • Holloway V, Gadian DG, Vargha-Khadem F, Porter DA, Boyd SG, Connelly A. (2000) The reorganization of sensorimotor function in children after hemispherectomy: A functional MRI and somatosensory evoked potential study. Brain 123:24322444.
  • Kulak W, Sobaniec W, Kuzia J-S, Bockowski L. (2006) Neurophysiologic and neuroimaging studies of brain plasticity in children with spastic cerebral palsy. Exp Neurol 198:411.
  • Lacruz ME, Garcia Seoane JJ, Valentin A, Selway R, Alarcon G. (2007) Frontal and temporal functional connections of the living human brain. Eur J Neurosci 26:13571370.
  • Liegeois F, Connelly A, Cross JH, Boyd SG, Gadian DG, Vargha-Khadem F, Baldeweg T. (2004) Language reorganization in children with early-onset lesions of the left hemisphere: An fMRI study. Brain 127:12291236.
  • Matsumoto R, Nair DR, LaPresto E, Najm I, Bingaman W, Shibasaki H, Luders HO. (2004) Functional connectivity in the human language system: A cortico-cortical evoked potential study. Brain 127:23162330.
  • Meeren HKM, Pijn JPM, Van Luijtelaar ELJM, Coenen AML, Lopes da Silva FH. (2002) Cortical focus drives widespread corticothalamic networks during spontaneous absence seizures in rats. J Neurosci 22:14801495.
  • Meeren H, Van Luijtelaar G, Lopes da Silva F, Coenen A. (2005) Evolving concepts on the pathophysiology of absence seizures: The cortical focus theory. Arch Neurol 62:371376.
  • Ojemann GA, Engel J. (1987) Acute and chronic intracranial recording and stimulation. In EngelJ (Ed) Surgical treatment of the epilepsies. Raven Press, New York, pp. 263288.
  • Rutecki PA, Grossman RG, Armstrong D, Irish-Loewen S. (1989) Electrophysiological connections between the hippocampus and entorhinal cortex in patients with complex partial seizures. J Neurosurg 70:667675.
  • Seri S, Cerquiglini A, Pisani F, Michel CM, Pascual Marqui RD, Curatolo P. (1998) Frontal lobe epilepsy associated with tuberous sclerosis: electroencephalographic-magnetic resonance image fusioning. J Child Neurol 13:3338.
  • Shields WD. (2004) Effects of epilepsy surgery on psychiatric and behavioral comorbidities in children and adolescents. Epilepsy Behav 5:S18S24.
  • Spencer SS, Sperling MR, Shewmon DA, Kahane P. (2007) Intracranial electrodes. In EngelJ, PedleyTA, AicardiJ, DichterMA, HeinemannU, MoshéSL, PorterRJ, TaylorDC (Eds) Epilepsy: A comprehensive textbook, Lippincott Williams & Wilkins, Philadelphia. pp. 17911832.
  • Sperli F, Spinelli L, Seeck M, Kurian M, Michel CM, Lantz G. (2006) EEG source imaging in pediatric epilepsy surgery: A new perspective in presurgical workup. Epilepsia 47:981990.
  • Steriade M, Contreras D. (1998) Spike-wave complexes and fast components of cortically generated seizures. I. role of neocortex and thalamus. J Neurophysiol 80:14391455.
  • Valentin A, Anderson M, Alarcon G, Seoane JJ, Selway R, Binnie CD, Polkey CE. (2002) Responses to single pulse electrical stimulation identify epileptogenesis in the human brain in vivo. Brain 125:17091718.
  • Valentin A, Alarcon G, Garcia-Seoane JJ, Lacruz ME, Nayak SD, Honavar M, Selway RP, Binnie CD, Polkey CE. (2005a) Single-pulse electrical stimulation identifies epileptogenic frontal cortex in the human brain. Neurology 65:426435.
  • Valentin A, Alarcon G, Honavar M, Garcia Seoane JJ, Selway RP, Polkey CE, Binnie CD. (2005b) Single pulse electrical stimulation for identification of structural abnormalities and prediction of seizure outcome after epilepsy surgery: A prospective study. Lancet Neurol 4:718726.
  • Wilson CL, Isokawa M, Babb TL, Crandall PH. (1990) Functional connections in the human temporal lobe. I. Analysis of limbic system pathways using neuronal responses evoked by electrical stimulation. Exp Brain Res 82:279292.
  • Yuan W, Szaflarski JP, Schmithorst VJ, Schapiro M, Byars AW, Strawsburg RH, Holland SK. (2006) fMRI shows atypical language lateralization in pediatric epilepsy patients. Epilepsia 47:593600.

Supporting Information

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Table S1. Patient details.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
EPI_2056_sm_Table1.doc100KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.