Intraoperative hippocampal electrocorticography frequently captures electrographic seizures and correlates with hippocampal pathology

Clinical Neurophysiology 129 (2018) 717–723

Contents lists available at ScienceDirect

Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph

Intraoperative hippocampal electrocorticography frequently captures electrographic seizures and correlates with hippocampal pathology Aashit K. Shah a,b, Darren Fuerst b, Sandeep Mittal b,c,d,e,⇑ a

Department of Neurology, Wayne State University, Detroit, MI, USA Comprehensive Epilepsy Center, Detroit Medical Center, Wayne State University, Detroit, MI, USA c Department of Neurosurgery, Wayne State University, Detroit, MI, USA d Department of Oncology, Wayne State University, Detroit, MI, USA e Karmanos Cancer Institute, Wayne State University, Detroit, MI, USA b

a r t i c l e

i n f o

Article history: Accepted 10 January 2018 Available online 2 February 2018 Keywords: Intraoperative electrocorticography Hippocampal seizures Temporal lobe epilepsy Hippocampus Hippocampal connections Epilepsy surgery

h i g h l i g h t s  Intraoperative hippocampal electrocorticography frequently captures spontaneous seizures.  Electrographic seizures are more likely in patients without disruption of hippocampal architecture.  Intraoperative hippocampal seizures may result from deafferentation from the temporal neocortex.

a b s t r a c t Objective: Relationship between electrographic seizures on hippocampal electrocorticography (IH-ECoG) and presence/type of hippocampal pathology remains unclear. Methods: IH-ECoG was recorded for 10–20 min from the ventricular surface of the hippocampus following removal of the temporal neocortex in 40 consecutive patients. Correlation between intraoperative hippocampal seizures and preoperative MRI, hippocampal histopathology, and EEG from invasive monitoring was determined. Results: IH-ECoG captured electrographic seizures in 15/40 patients (in 8/23 with abnormal hippocampal signal on MRI and 7/17 patients without MRI abnormality). Hippocampal neuronal loss was observed in 22/40 (Group 1), while 18/40 had no significant neuronal loss (Group 2). In Group 1, 4/22 had seizures on IH-ECoG, while 11/18 had electrographic seizures in Group 2. In 24/40 patients who underwent prolonged extraoperative intracranial EEG (IC-EEG) recording, hippocampal seizures were captured in 14. Of these, 7 also had seizures during IH-ECoG. In 10/24 IC-EEG patients without seizures, 3 had seizures on IH-ECoG. Conclusions: IH-ECoG frequently captures spontaneous electrographic seizures. These are more likely to occur in patients with pathologic processes that do not disrupt/infiltrate hippocampus compared to patients with intractable epilepsy associated with disrupted hippocampal architecture. Significance: Intraoperative hippocampal seizures may result from deafferentation from the temporal neocortex and disinhibition of the perforant pathway. Ó 2018 International Federation of Clinical Neurophysiology. Published by Elsevier B.V. All rights reserved.

1. Introduction In patients with pharmacoresistant temporal lobe epilepsy (TLE), the epileptogenic zone commonly involves the medial temporal lobe structures including the hippocampus. Removal of the

⇑ Corresponding author at: Department of Neurosurgery, Wayne State University, 4160 John R Street, Suite 930, Detroit, MI 48201, USA. Fax: +1 313 966 0368. E-mail address: [email protected] (S. Mittal).

hippocampus by resecting the medial temporal lobe either selectively or along with the temporal neocortex provides lasting seizure freedom (Datta et al., 2009; Kumar et al., 2013; Mittal et al., 2005; Olivier, 2000; Tanriverdi et al., 2008; Wiebe et al., 2001; Wieser et al., 2003; Wieser and Yasargil, 1982; Yasargil et al., 2010). Usefulness of electrocorticography (ECoG) as a tool during epilepsy surgery remains controversial (Chen et al., 2006; Mittal et al., 2016; San-juan et al., 2011; Schwartz et al., 1997; Stefan et al., 1991). Most investigators have directed their efforts at

https://doi.org/10.1016/j.clinph.2018.01.055 1388-2457/Ó 2018 International Federation of Clinical Neurophysiology. Published by Elsevier B.V. All rights reserved.

718

A.K. Shah et al. / Clinical Neurophysiology 129 (2018) 717–723

recording ECoG from the lateral and mediobasal temporal lobe surface during temporal lobe surgery (Fiol et al., 1991; Uda et al., 2015). There is very limited information in the literature describing intraoperative recording directly from the hippocampus prior to its removal. Interictal hippocampal spiking has been observed intraoperatively using either strip electrodes over the ventricular surface (McKhann et al., 2000; Oliveira et al., 2006; Polkey et al., 1989; Sugano et al., 2007; Tanaka et al., 1996) or using depth electrodes targeting the hippocampus (Kanazawa et al., 1996). Intraoperative ECoG recorded over the ventricular surface of the hippocampus has been used to guide the extent of hippocampal resection in patients with both lesional and non-lesional TLE (McKhann et al., 2000; Sugano et al., 2007). Rarely, spontaneous electrographic seizures have been captured by intraoperative hippocampal ECoG (Tanaka et al., 1996). In addition, spontaneous electrical activity resembling seizures have been demonstrated by several groups in vitro in hippocampal slice recordings (Borck and Jefferys, 1999; Karnup and Stelzer, 2001; Swartzwelder et al., 1988). The significance of electrographic seizures captured from the hippocampus remains unclear. We report our experience with systematic ECoG recording from the ventricular surface of the hippocampus during temporal lobe epilepsy surgery and discuss possible pathophysiological mechanisms underlying these spontaneous electrographic events. 2. Methods 2.1. Patients The study population consisted of 40 consecutive patients with medically intractable TLE who underwent tailored temporal lobe resection including medial temporal lobe structures by a single epilepsy surgeon (SM). A subset of this study group underwent staged epilepsy surgery with long-term extraoperative video-EEG recording with intracranial electrodes (IC-EEG). All patients were treated at the Wayne State University/Detroit Medical Center Comprehensive Epilepsy Center. The clinical, electrophysiological (including long-term video-telemetry and IC-EEG), radiographic, and histopathological characteristics were prospectively collected for each patient. Relationship between electrographic seizures [recorded during intraoperative hippocampal ECoG (IH-ECoG) and/or during long-term extraoperative IC-EEG monitoring in patients undergoing staged surgery] and clinical parameters, preoperative MRI, hippocampal histology, as well as, seizure outcome was analyzed. Patients with less than one year follow-up were excluded from the seizure outcome analysis. The study was approved by the local Institutional Review Board. 2.2. Surgical technique Patients were placed in supine position with the head turned away from the side of surgery. Most patients were under general anesthesia using a combination of midazolam, pancuronium, fentanyl, and propofol (n = 29); while propofol and remifentanil was used for patients that underwent surgery under local sedation (n = 11). Following craniotomy, patients underwent intraoperative ECoG (10–20 min) over the lateral surface of the temporal lobe to help define the posterior resection margin in cases of one-stage tailored temporal lobe resection. This was followed by en bloc microsurgical resection of the temporal neocortex with exposure of the ventricular surface of the hippocampus in all individuals. The hippocampus and parahippocampal gyrus along with their blood supply were preserved. After neocortical resection, patients underwent a brief (10–20 min) IH-ECoG using a 4-contact strip electrode placed along the long axis of the hippocampus (see

below). Subsequently, the entire head and body of the hippocampus (3.5–4 cm) was microsurgically removed en bloc up to the lateral mesencephalic sulcus (along with the parahippocampal gyrus) followed by subpial removal of the amygdala. The extent of hippocampal resection was not influenced by the results of IHECoG. Also, post-resection IH-ECoG of the residual tail of the hippocampus was not performed. Hemostasis was obtained after removal of the mesial temporal structures followed by routine closure. 2.3. Intraoperative hippocampal ECoG The depth of anesthesia sedation was adjusted during ECoG to obtain continuous EEG recording without suppression. Antiepileptic medication was not discontinued prior to surgery. Intraoperative ECoG was recorded over the exposed ventricular surface of the hippocampus following en bloc removal of the temporal lobe neocortex including the temporal pole while preserving the entorhinal cortex (Fig. 1). IH-ECoG was performed using a 4contact subdural strip electrode with platinum contact of 0.3 cm and interelectrode distance of 1 cm (PMT Corp., Chanhassen, MN or Ad-Tech Medical Instrument Corporation, Racine, WI) with electrode #1 placed over the tail of the hippocampus and electrode #4 covering the hippocampal head. In situ hippocampal EEG recording was performed over 10–20 min in a digitized monopolar referential format with a reference placed over the forehead at a sampling rate of 200 or 1000 Hz (Xltek EEG system, Natus Medical Corp., San Carlos, CA). EEG data were analyzed using a low frequency filter of 1 Hz and a high frequency filter of 70 Hz using referential and bipolar montages. EEG rhythms that exhibited temporal evolution in amplitude, frequency, and/or morphology were identified as electrographic hippocampal seizures. 2.4. Statistical analysis We assessed whether there was an association between presence of intraoperative hippocampal seizures with various clinical, radiographic, electrophysiological, and histopathological variables using chi-square analysis, one-way between groups ANOVA, and appropriate descriptive statistics using SAS 9.3 for Windows. A p value of less than 0.05 was considered significant. 3. Results Of the 40 patients included in this study, 15 (37.5%) had EEG rhythms on IH-ECoG that exhibited temporal evolution in amplitude or frequency usually recognized as electrographic seizures (Fig. 2). Hence, we refer to these intraoperative events as spontaneous electrographic hippocampal seizures. Table 1 outlines the relevant clinical, electrophysiological, neuroimaging, and histopathologic findings. 3.1. Patient demographics There were slightly more females than males (22 vs 18) in the study population, but this did not reach statistical significance (v2 = 0.40, n.s.). Average age for the sample was 39 years (SD = 39.6, range = 21–67) overall, 39.3 years (SD = 9.96) for females, and 38.7 years (SD = 13.7) for males. ANOVA indicated the difference in age was not significant. The difference in mean age for subjects with or without hippocampal seizures on IH-ECoG was not significant (F[1,38] = 0.11, n.s.). The mean seizure duration for patients with seizures recorded on IH-ECoG was 12 years, while it was 11 years in the patients without such seizures on IH-ECoG.

A.K. Shah et al. / Clinical Neurophysiology 129 (2018) 717–723

719

Fig. 1. A: Intraoperative photograph showing the ventricular surface of the exposed hippocampus during the second stage of surgery following resection of the lateral and inferior temporal neocortex. The arrows point to the entry of the orthogonally implanted hippocampal depth electrodes placed earlier for extraoperative intracranial EEG recording. B: Intraoperative photograph of the same patient following a typical placement of the 1  4 strip electrode over the hippocampus. Electrode #1 is over the posterior part of the hippocampal body while Electrode #4 is at the junction of the body and head of the hippocampus.

Fig. 2. Intraoperative hippocampal electrocorticography (IH-ECoG) recording utilizing 1  4 strip electrode as shown in Fig. 1B. The recording was made in a monopolar referential format with reference placed over the forehead at a sampling rate of 200 Hz. The EEG data is displayed here using a referential montage with a bandpass filter of 1– 70 Hz. The IH-ECoG shows an electrographic seizure with repetitive spike discharges that evolve in amplitude and frequency before an abrupt termination lasting 46 s.

720

A.K. Shah et al. / Clinical Neurophysiology 129 (2018) 717–723

Table 1 Patient data. No

Sex

Age

Hippo Sz in OR (IHECoG)

Hippo Sz in EMU (ICEEG)

Hippo abnormality on MRI

Sz duration (Years)

Sz outcome at 1 year

Sz outcome at 2 year

Sz outcome at 5 year

Group

Histopathology

1

F

42

Y

N

Y

1

1

1

1

2

2

F

39

Y

Y

Y

<1

1

1

1

3

F

37

Y

Y

N

8

1

1

No followup 1

4 5 6

M M M

21 50 45

N Y Y

N N Y

N Y Y

13 26 4

2 1 1

3 1 1

3 1 1

2 2 2

7 8 9

M M F

43 33 33

Y Y N

Y ND ND

N N N

17 31 1

4 1 1

4 1 1

4 1 1

2 2 1

10

M

30

N

ND

N

1

1

1

Died tumor recurrence

1

11

M

55

Y

ND

Y

<1

1

12

M

21

N

ND

Y

18

3

Died tumor recurrence 3

DNET without hippocampal involvement Oligodendroglioma Gr 2 with hippocampal involvement Meningioma grade 1 with normal hippocampus Temporal neocortical ulygyria Epidermoid Cortical dysplasia with hippocampal microheterotopia Gliosis Microheterotopia DNET with hippocampal involvement Oligodendroglioma Gr 2 with hippocampal involvement Infiltrating astrocytoma Gr 2 with hippocampal involvement Hippocampal sclerosis

13

M

40

N

N

Y

<1

1

14 15 16

F F M

42 22 67

N N N

N ND ND

N N Y

24 21 1

4 2 1

17

F

40

N

Y

Y

1

1

18

M

34

Y

N

N

24

3

Died tumor recurrence 2 2 Died tumor recurrence Died tumor recurrence 2

19 20

F M

51 57

N N

ND N

N Y

1 1

1 1

1 1

21 22 23 24 25 26 27

M F F F F F F

27 25 47 32 39 37 46

N N N N N Y N

Y ND Y ND ND Y Y

Y Y N N Y N Y

22 3 45 4 30 2 5

1 3 1 1 1 1 1

1 3 1 1 1 1 1

28 29 30 31

M M F M

27 39 45 22

Y Y Y N

Y ND Y N

N Y Y Y

5 <1 17 12

1 1 1 1

1 1 1 1

1 2 2 1

32 33 34

F F M

37 39 54

N N N

Y Y N

Y N Y

16 8 <1

1 1 1

1 1 1

2 2 1

35 36

M F

31 30

N Y

ND ND

Y Y

6 <1

1 4

1 4

2 1

37

F

62

N

ND

Y

1

1

1

1

38 39 40

F F F

22 56 41

N N Y

Y N ND

Y N N

22 7 22

1 1 1

1 1 1

1 1 2

2

1 Died SUDEP

1 1

1 1

2 1 1 1

3

2

1 Died tumor recurrence 1 3 1 1

2 1 1 1 1 1 1 2 2

Mixed oligoastrocytoma Gr 3, hippocampal sclerosis Gliosis Hippocampal sclerosis Anaplastic astrocytoma Gr 3 with hippocampal involvement Mixed oligoastrocytoma Gr 3 with hippocampal involvement Microheterotopia with gliosis in hippocampus Gliosis Mixed oligoastrocytoma Gr 2 with hippocampal involvement Hippocampal sclerosis Hippocampal sclerosis Hippocampal sclerosis Hippocampal sclerosis Hippocampal sclerosis Gliosis DNET without hippocampal involvement Hippocampal sclerosis Gliosis Gliosis Ganglioglioma with hippocampal involvement Gliosis Gliosis Mixed oligoastrocytoma Gr 2 with hippocampal involvement Gliosis Limbic encephalitis with hippocampal neuronal loss Mixed anaplastic oligoastrocytoma Gr 3 with hippocampal involvement Hippocampal sclerosis Hippocampal sclerosis Gliosis

Abbreviations: Hippo = hippocampal, Sz = seizure, OR = operating room, IH-ECoG = intraoperative hippocampal electrocorticography, EMU = epilepsy monitoring unit, ICEEG = intracranial EEG (long-term extraoperative recording), Gr = grade, DNET = dysembryoplastic neuroepithelial tumor, ND = not done, SUDEP: sudden unexplained death in epilepsy.

3.2. Electrophysiological analysis Of the 40 patients, 24 underwent staged epilepsy surgery with implantation of temporal depth and/or subdural grid/strip electrodes. Intracranial EEG was recorded from the mesial temporal structures (hippocampus, amygdala, parahippocampal gyrus) in all patients with invasive monitoring. Prolonged extraoperative

IC-EEG recording captured hippocampal seizures (alone or in conjunction with neocortical seizures) in 14/24 (58%) patients (v2 = 0. 67, n.s.). Of these 14 patients with epileptogenic zone involving the hippocampus; seven (50%) had electrographic hippocampal seizures during IH-ECoG performed during temporal resection. In the remaining 10/24 patients with exclusive extra-hippocampal seizure onset, three (30%) also had electrographic seizures during

721

A.K. Shah et al. / Clinical Neurophysiology 129 (2018) 717–723

IH-ECoG at time of second stage surgery. Of the remaining 16 patients undergoing one stage resection, 5 (31%) had seizures recorded on IH-ECoG (Table 2). In most of the 15 patients with intraoperative hippocampal seizures, the IH-ECoG session recorded 1 or 2 electrographic seizures lasting from 10 s to over 2 min. The electrographic seizures involved only part of the hippocampus, i.e. seizure isolated to one (n = 5) or two electrodes (n = 7), while the remaining 3 patients had seizures involving all 4 electrode contacts. Electrode #2 (located over the posterior half of the hippocampal body) was involved in the electrographic seizure in 9 patients; while electrode #3 (anterior half of hippocampal body) and electrode #1 (tail of the hippocampus) were each involved in 7 patients with electrographic seizures captured during IH-ECoG. The head of the hippocampus (covered by electrode #4) was least likely to participate in seizures (n = 3). The interictal epileptiform discharges (absent in several patients) were typically recorded from the same electrode(s) as the seizure onset. A typical seizure pattern consisted of repetitive slow (1–5 Hz) spikes or sharp waves gradually increasing in amplitude and at times spreading to adjacent electrodes (Fig. 2). This electrographic pattern of hippocampal seizures was different than those captured during long-term extraoperative IC-EEG. 3.3. Neuroimaging findings On preoperative MRI of brain, abnormalities involving hippocampus were detected in 23 patients. Of these 23 patients, 8 (35%) had seizures recorded on IH-ECoG. Of the 17 patients without hippocampal abnormality on MRI, IH-ECoG recorded seizures in 7 (41%) patients. Neuroimaging findings did not influence likelihood of intraoperative seizures on IH-ECoG (v2 = 0.90, n.s.).

(n = 1). Group 2 (n = 18) defined patients with grossly preserved hippocampal architecture including 14 cases of gliosis, 3 patients with microheterotopia, and 1 case without hippocampal abnormality. On IH-ECoG, 4 of the 22 patients (18%) in Group 1 had hippocampal seizures while 11/18 patients (61%) in Group 2 had electrographic seizures (v2 = 8.851, p < 0.01). Of these 22 patients in Group 1, 11 had two-staged surgery, and 6 of these 11 (54.5%) had hippocampal seizures in the epilepsy monitoring unit (EMU) during IC-EEG recording. Of 18 patients of Group 2, 13 underwent two-staged surgery and 8 of 13 (61.5%) had hippocampal onset of seizures in the EMU. Hence, there was no difference in likelihood of hippocampal seizures during IC-EEG recording between these two groups (v2 = 0.96, n.s.). From the neuroimaging perspective, preoperative MRI showed abnormality involving the hippocampus in 15 of 22 patients (68.2%) in Group 1; while 8 of 18 (44.4%) in Group 2 showed abnormality in the hippocampus. However, this difference did not reach statistical significance (v2 = 0.17, n.s.). Overall, the seizure outcome at one-year follow-up was similar between Groups 1 and 2. Nineteen of the 22 (86.4%) patients in Group 1 had excellent seizure outcome (Engel Class I or II), similarly 15 of 18 (83.3%) in Group 2 achieved the same outcome. Thus, postsurgical seizure control did not differ on the basis of integrity of the hippocampal architecture (v2 = 1.74, n.s.). At two-year follow-up, 14 of 17 (82.4%) patients in Group 1 again had excellent seizure outcome, and 16 of 18 (88.9%) had similar results in Group 2, and again seizure outcome was not dependent on hippocampal integrity (v2 = 0.31, n.s.). Although limited data was available for 5-year seizure outcome, it remained excellent (Engel Class I) for both Group 1 (5/6 patients; 83.3%) and Group 2 (7/10 patients, 70%); and, not surprisingly the difference between the groups was not significant (v2 = 0.34, n.s.).

3.4. Seizure outcome Seizure outcome at one year was Engel Class I in 32 (80%), Class II in 2 (5%), Class III in 3 (7.5%), and Class IV in 3 (7.5%). Seizure outcome did not differ significantly between 15 patients with electrographic seizures on IH-ECoG and 25 without such seizures (12 vs 22 with Engel Class I or II outcome). At the 2-year mark, 4 of 40 patients had died of tumor recurrence but nevertheless remained seizure-free following surgery. Of the remaining 36 patients, 28 (78%) achieved Class I outcome, with 3 had Class II (8%), 3 had Class III (8%) and 2 had Class IV (6%) outcome. At the 5-year mark, 3 additional patients had died (2 with tumor recurrence and 1 due to SUDEP) and one patient was lost to follow-up. Of the 16 patients who were evaluated 5 years after epilepsy surgery, 12 (75%) achieved Class I outcome, 3 had Class III (19%) and 1 had Class IV (6%) outcome. 3.5. Relationship between hippocampal histopathology and hippocampal seizures Patients were divided into two groups based on integrity of hippocampal architecture. Group 1 (n = 22) consisted of patients with histopathologically-proven neuronal loss from hippocampal sclerosis (n = 10), tumoral infiltration (n = 11), or limbic encephalitis

4. Discussion The extent of hippocampal resection does not always predict likelihood of seizure freedom. Nevertheless, insufficient removal of the hippocampus remains one of the most common causes of failed temporal lobe surgery (Germano et al., 1994; Jooma et al., 1995; Kanner et al., 1995; Stefan et al., 1996; Wyler et al., 1989). Usefulness of intraoperative ECoG for tailored temporal lobe resection remains controversial. Most investigators recorded ECoG from the lateral and mediobasal temporal lobe surface or performed limited recording with depth electrodes prior to resection during temporal lobe surgery (Burkholder et al., 2014; Kanazawa et al., 1996; MacDonald and Pillay, 2000; Schwartz et al., 1997). A study using a strip electrode inserted into the temporal horn via lateral approach and recording made prior to resection did not capture any seizures, but recorded spikes in most patients with temporal lobe epilepsy (Polkey et al., 1989). Important reciprocal connections exist between the hippocampus and the basal temporal neocortex and contribute to a complex neuronal network that can generate interictal spikes and seizures (Alarcón et al., 1997). The effects of resection of various parts of the temporal lobe has shown to have dynamic effects on the

Table 2 Hippocampal seizure data. Two-stage surgery with prolonged IC-EEG (n = 24) Hippocampal Sz during IC-EEG (n = 14)

One-stage surgery only (n = 16) No Hippocampal Sz during IC-EEG (n = 10)

IH-ECoG Sz

No IH-ECoG Sz

IH-ECoG Sz

No IH-ECoG Sz

IH-ECoG Sz

No IH-ECoG Sz

7

7

3

7

5

11

722

A.K. Shah et al. / Clinical Neurophysiology 129 (2018) 717–723

epileptiform discharges on ECoG. It is noted that excision of the temporal pole and hippocampus reduces neocortical spiking suggesting that these neocortical spikes depend on connections between the medial structures and the lateral temporal cortex (Oliveira et al., 2006). In contrast, the group in Montreal found that selective amygdalohippocampectomy through the middle temporal gyrus increases spiking over the temporal neocortical surface (Cendes et al., 1993). Although it is technically relatively straightforward to intraoperatively record ECoG from the hippocampus, such experience remains limited to a handful of centers. Resection of the anterior temporal neocortex (temporal pole) resulted in increased frequency or de novo appearance of hippocampal epileptiform activity during staged intraoperative ECoG (Oliveira et al., 2006). Similarly, interictal hippocampal spikes were observed following lateral temporal resection by the group from Seattle (McKhann et al., 2000). In their prospective study of 140 patients, the authors performed tailored hippocampal resection based on presence and extent of interictal hippocampal spiking. The length of hippocampal resection did not correlate with seizure outcome. However, presence of post-resection residual hippocampal spiking did correlate with worse seizure outcome (McKhann et al., 2000). Conversely, a randomized prospective study comparing partial (extending to the anterior cerebral peduncle) vs total hippocampal resection (extending to the collicular cistern) irrespective of electrophysiological findings found that patients undergoing a partial resection were less likely to become seizure-free (38% vs 69%). However, a large number of patients who were randomized to undergo partial hippocampal resection also became seizure-free (38%), thereby suggesting that the epileptogenic zone may be restricted to only to a focal segment of the hippocampus and that total hippocampal removal might not be necessary to achieve seizure freedom (Wyler et al., 1995). In this study, we demonstrate that the changes that fit the traditional criteria of electrographic seizures are frequently observed during a relatively short duration of IH-ECoG recording. The regularity at which electrographic seizures were captured was unexpected. We believe that this was partly related to our surgical technique whereby almost all of the lateral and inferior temporal neocortex was removed prior to the IH-ECoG while preserving the hippocampal-parahippocampal complex and their blood supply. The exact significance of this phenomenon is uncertain and the clinical relevance of electrographic hippocampal seizures remains unknown. We believe that the removal of the temporal neocortex enhances emergence of these electrographic ‘‘seizures.” This fact is supported by the group of patients in this study who underwent two-staged epilepsy surgery and had several days of extraoperative IC-EEG using hippocampal depth electrodes and did not show such high frequency of seizures even when the hippocampus was part of the epileptogenic zone. In addition, even in patients where the epileptogenic zone was outside the hippocampus, seizures were captured on IH-ECoG. While IH-ECoG has shown to have frequent epileptiform discharges (Polkey et al., 1989), electrographic seizures have not been previously reported. The current study indicates that age, gender and duration of epilepsy do not influence likelihood of recording seizures on IHECoG. Our results also suggest that although hippocampal sclerosis (with characteristic neuronal loss) is considered a prototype of hippocampal pathology leading to intractable epilepsy, acute seizures on IH-ECoG are less likely to be seen in these individuals. Similarly, in patients where the epileptogenic zone included the hippocampus (as shown by prolonged invasive extraoperative IC-EEG recording), chances of recording acute seizures during IH-ECoG was no better than 50%. The best predictor of the recording seizures on IH-ECoG was preservation of the hippocampal architecture.

4.1. Hippocampal connections The major afferents to the hippocampus are from the entorhinal cortex to the dentate gyrus through the perforant path. The entorhinal cortex receives input from perirhinal and posterior parahippocampal cortices, which in turn receive afferents from a variety of cortical areas including unimodal and multimodal cortices (Jones, 1993; Powell et al., 2004). Functional connectivity studies using electrical stimulation during ECoG showed strong connections from the orbitofrontal and other temporal lobe structures to the ipsilateral medial temporal structures (Alarcón et al., 1997; Lacruz et al., 2007). In contrast, limited input is received from the contralateral temporal lobe structures (Jiménez-Jiménez et al., 2015). During our IH-ECoG, both hippocampus and entorhinal cortex were preserved following en bloc resection of anterior, lateral and inferior temporal cortex; thereby resulting in deafferentation of these cortices from their normal inputs from various cortical structures. Normally, these inputs are inhibitory in nature and loss of this inhibition results in increase stimulation through the perforant pathway (Powell et al., 2004). Moreover, it is well known that stimulation of the perforant pathway leads to hippocampal seizures on hippocampal slice preparations (Bumanglag and Sloviter, 2008; Maru et al., 2002; Sutula et al., 1986). Hence, we postulate that removal of the temporal neocortex during surgery prior to recording IH-ECoG results in disinhibition of the perforant pathway leading to seizures in the hippocampus. Our observations may also provide additional insight into the potential underlying pathogenesis of seizures in patients with traumatic brain injury. It is known that the patients with temporal contusion/bleeds suffer from posttraumatic epilepsy at high rate (30%) (Diaz-Arrastia et al., 2009). Injury to the temporal neocortex involving the anterior and inferior surface may disrupt cortical inputs to the parahippocampal region (similar to our en bloc resection) that induces electrographic changes in the hippocampus acutely that are epileptogenic in nature and in due course may lead to spontaneous clinical seizures. 4.2. Limitations of study There are several limitations to this study. First, although the data was collected prospectively, it was analyzed retrospectively. We recorded hippocampal rhythmic discharges that resembled and were identified as electrographic seizures; however, it remains unclear whether this phenomenon represents normal physiological activity or truly a seizure with clinical significance. We also did not record ECoG from the ventricular surface of hippocampus prior to removal of the lateral and inferior temporal cortex limiting our observation to post-neocortical resection phase. Determination of the exact cause of EEG changes described here, as seizures on IHECoG on the basis of this study is speculative. A more definitive prospective study, such as staged resection of the temporal lobe while IH-ECoG is being recorded may provide additional understanding of intraoperative hippocampal seizures. 5. Disclosure None of the authors have potential conflicts of interest to be disclosed. References Alarcón G, Garcia Seoane JJ, Binnie CD, Martin Miguel MC, Juler J, Polkey CE, et al. Origin and propagation of interictal discharges in the acute electrocorticogram. Implications for pathophysiology and surgical treatment of temporal lobe epilepsy. Brain 1997;120(Pt 12):2259–82.

A.K. Shah et al. / Clinical Neurophysiology 129 (2018) 717–723 Borck C, Jefferys JG. Seizure-like events in disinhibited ventral slices of adult rat hippocampus. J Neurophysiol 1999;82:2130–42. Bumanglag AV, Sloviter RS. Minimal latency to hippocampal epileptogenesis and clinical epilepsy after perforant pathway stimulation-induced status epilepticus in awake rats. J Comp Neurol 2008;510:561–80. Burkholder DB, Sulc V, Hoffman EM, Cascino GD, Britton JW, So EL, et al. Interictal scalp electroencephalography and intraoperative electrocorticography in magnetic resonance imaging-negative temporal lobe epilepsy surgery. JAMA Neurol 2014;71:702–9. Cendes F, Dubeau F, Olivier A, Cukiert A, Andermann E, Quesney LF, et al. Increased neocortical spiking and surgical outcome after selective amygdalohippocampectomy. Epilepsy Res 1993;16:195–206. Chen X, Sure U, Haag A, Knake S, Fritsch B, Muller HH, et al. Predictive value of electrocorticography in epilepsy patients with unilateral hippocampal sclerosis undergoing selective amygdalohippocampectomy. Neurosurg Rev 2006;29:108–13. Datta A, Sinclair DB, Wheatley M, Jurasek L, Snyder T, Quigley D, et al. Selective amygdalohippocampectomy: surgical outcome in children versus adults. Can J Neurol Sci 2009;36:187–91. Diaz-Arrastia R, Agostini MA, Madden CJ, Van Ness PC. Posttraumatic epilepsy: the endophenotypes of a human model of epileptogenesis. Epilepsia 2009;50(Suppl 2):14–20. Fiol ME, Gates JR, Torres F, Maxwell RE. The prognostic value of residual spikes in the postexcision electrocorticogram after temporal lobectomy. Neurology 1991;41:512–6. Germano IM, Poulin N, Olivier A. Reoperation for recurrent temporal lobe epilepsy. J Neurosurg 1994;81:31–6. Jiménez-Jiménez D, Abete-Rivas M, Martín-Lopez D, Lacruz ME, Selway RP, Valentín A, et al. Incidence of functional bi-temporal connections in the human brain in vivo and their relevance to epilepsy surgery. Cortex 2015;65:208–18. Jones RS. Entorhinal-hippocampal connections: a speculative view of their function. Trends Neurosci 1993;16:58–64. Jooma R, Yeh HS, Privitera MD, Rigrish D, Gartner M. Seizure control and extent of mesial temporal resection. Acta Neurochir (Wien) 1995;133:44–9. Kanazawa O, Blume WT, Girvin JP. Significance of spikes at temporal lobe electrocorticography. Epilepsia 1996;37:50–5. Kanner AM, Kaydanova Y, deToledo-Morrell L, Morrell F, Smith MC, Bergen D, et al. Tailored anterior temporal lobectomy. Relation between extent of resection of mesial structures and postsurgical seizure outcome. Arch Neurol 1995;52:173–8. Karnup S, Stelzer A. Seizure-like activity in the disinhibited CA1 minislice of adult guinea-pigs. J Physiol 2001;532:713–30. Kumar A, Valentín A, Humayon D, Longbottom AL, Jiménez-Jiménez D, Mullatti N, et al. Preoperative estimation of seizure control after resective surgery for the treatment of epilepsy. Seizure 2013;22:818–26. Lacruz ME, Garcia Seoane JJ, Valentín A, Selway R, Alarcón G. Frontal and temporal functional connections of the living human brain. Eur J Neurosci 2007;26:1357–70. MacDonald DB, Pillay N. Intraoperative electrocorticography in temporal lobe epilepsy surgery. Can J Neurol Sci 2000;27(Suppl 1):S85–91. discussion S2-6. Maru E, Kanda M, Ashida H. Functional and morphological changes in the hippocampal neuronal circuits associated with epileptic seizures. Epilepsia 2002;43(Suppl 9):44–9. McKhann 2nd GM, Schoenfeld-McNeill J, Born DE, Haglund MM, Ojemann GA. Intraoperative hippocampal electrocorticography to predict the extent of hippocampal resection in temporal lobe epilepsy surgery. J Neurosurg 2000;93:44–52. Mittal S, Barkmeier D, Hua J, Pai DS, Fuerst D, Basha M, et al. Intracranial EEG analysis in tumor-related epilepsy: evidence of distant epileptic abnormalities. Clin Neurophysiol 2016;127:238–44.

723

Mittal S, Montes JL, Farmer JP, Rosenblatt B, Dubeau F, Andermann F, et al. Longterm outcome after surgical treatment of temporal lobe epilepsy in children. J Neurosurg 2005;103:401–12. Oliveira PA, Garzon E, Caboclo LO, Sousa PS, Carrete Jr H, Centeno RS, et al. Can intraoperative electrocorticography patterns predict surgical outcome in patients with temporal lobe epilepsy secondary to unilateral mesial temporal sclerosis? Seizure 2006;15:541–51. Olivier A. Transcortical selective amygdalohippocampectomy in temporal lobe epilepsy. Can J Neurol Sci 2000;27(Suppl 1):S68–76. discussion S92-6. Polkey CE, Binnie CD, Janota I. Acute hippocampal recording and pathology at temporal lobe resection and amygdalo-hippocampectomy for epilepsy. J Neurol Neurosurg Psychiatry 1989;52:1050–7. Powell HW, Guye M, Parker GJ, Symms MR, Boulby P, Koepp MJ, et al. Noninvasive in vivo demonstration of the connections of the human parahippocampal gyrus. Neuroimage 2004;22:740–7. San-juan D, Claudia AT, Maricarmen GA, Adriana MM, Richard JS, Mario AV. The prognostic role of electrocorticography in tailored temporal lobe surgery. Seizure 2011;20:564–9. Schwartz TH, Bazil CW, Walczak TS, Chan S, Pedley TA, Goodman RR. The predictive value of intraoperative electrocorticography in resections for limbic epilepsy associated with mesial temporal sclerosis. Neurosurgery 1997;40:302–9. discussion 9-11. Stefan H, Pauli E, Eberhard F, Ugrinovich R. Buchfelder M [‘‘Tailoring” resections in drug refractory temporal lobe epilepsy]. Nervenarzt 1996;67:306–10. Stefan H, Quesney LF, Abou-Khalil B, Olivier A. Electrocorticography in temporal lobe epilepsy surgery. Acta Neurol Scand 1991;83:65–72. Sugano H, Shimizu H, Sunaga S. Efficacy of intraoperative electrocorticography for assessing seizure outcomes in intractable epilepsy patients with temporal-lobemass lesions. Seizure 2007;16:120–7. Sutula T, Harrison C, Steward O. Chronic epileptogenesis induced by kindling of the entorhinal cortex: the role of the dentate gyrus. Brain Res 1986;385:291–9. Swartzwelder HS, Anderson WW, Wilson WA. Mechanism of electrographic seizure generation in the hippocampal slice in Mg2+-free medium: the role of GABAa inhibition. Epilepsy Res 1988;2:239–45. Tanaka T, Hashizume K, Kunimoto M, Yonemasu Y, Chiba S, Oki J. Intraoperative electrocorticography in children with medically intractable epilepsy. Neurol Med Chir (Tokyo) 1996;36:440–6. Tanriverdi T, Olivier A, Poulin N, Andermann F, Dubeau F. Long-term seizure outcome after mesial temporal lobe epilepsy surgery: corticalamygdalohippocampectomy versus selective amygdalohippocampectomy. J Neurosurg 2008;108:517–24. Uda T, Morino M, Minami N, Matsumoto T, Uchida T, Kamei T. Abnormal discharges from the temporal neocortex after selective amygdalohippocampectomy and seizure outcomes. J Clin Neurosci 2015;22:1797–801. Wiebe S, Blume WT, Girvin JP, Eliasziw M. Effectiveness, efficiency of surgery for temporal lobe epilepsy study G. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311–8. Wieser HG, Ortega M, Friedman A, Yonekawa Y. Long-term seizure outcomes following amygdalohippocampectomy. J Neurosurg 2003;98:751–63. Wieser HG, Yasargil MG. Selective amygdalohippocampectomy as a surgical treatment of mesiobasal limbic epilepsy. Surg Neurol 1982;17:445–57. Wyler AR, Hermann BP, Richey ET. Results of reoperation for failed epilepsy surgery. J Neurosurg 1989;71:815–9. Wyler AR, Hermann BP, Somes G. Extent of medial temporal resection on outcome from anterior temporal lobectomy: a randomized prospective study. Neurosurgery 1995;37:982–90. discussion 90-1. Yasargil MG, Krayenbuhl N, Roth P, Hsu SP, Yasargil DC. The selective amygdalohippocampectomy for intractable temporal limbic seizures. J Neurosurg 2010;112:168–85.