Hippocampal deep brain stimulation in nonlesional refractory mesial temporal lobe epilepsy

Accepted Manuscript Title: Hippocampal deep brain stimulation in nonlesional refractory mesial temporal lobe epilepsy Author: Hongbo Jin Wenling Li Changzheng Dong Jiang Wu Wenqing Zhao Zengyi Zhao Li Ma Fa Ma Yao Chen Qianwei Liu PII: DOI: Reference:

S1059-1311(16)00040-6 http://dx.doi.org/doi:10.1016/j.seizure.2016.01.018 YSEIZ 2672

To appear in:

Seizure

Received date: Revised date: Accepted date:

5-10-2015 28-1-2016 29-1-2016

Please cite this article as: Jin H, Li W, Dong C, Wu J, Zhao W, Zhao Z, Ma L, Ma F, Chen Y, Liu Q, Hippocampal deep brain stimulation in nonlesional refractory mesial temporal lobe epilepsy, SEIZURE: European Journal of Epilepsy (2016), http://dx.doi.org/10.1016/j.seizure.2016.01.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1. Hippocampal deep brain stimulation showed an efficiency and safety in refractory mesial temporal lobe epilepsy. 2. Refractory mesial temporal lobe epilepsy patients with normal MRI

ip t

responded well to stimulation.

cr

3. The reasonable adjustment of stimulation parameters was an integral

Ac ce p

te

d

M

an

us

part of DBS.

Page 1 of 29

Hippocampal deep brain stimulation in nonlesional refractory mesial temporal lobe epilepsy

ip t

Hongbo Jina, Wenling Lib, Changzheng Dongb, Jiang Wub, Wenqing Zhaoa,b, Zengyi Zhaoc，Li Mac, Fa Mac, Yao Chenb, Qianwei Liub

Faculty of Graduate Studies, Hebei Medical University, Shijiazhuang,

cr

a

Department of Functional Neurosurgery, Hebei General Hospital,

Shijiazhuang, Hebei Province, China

Department of Neurosurgery, Second Hospital of Shijiazhuang,

M

c

an

b

us

Hebei Province, China

d

Shijiazhuang, Hebei Province, China

te

Corresponding author's name: Wenqing Zhao

Ac ce p

Affiliation: Faculty of Graduate Studies, Hebei Medical University Department of Functional Neurosurgery, Hebei General Hospital

Mailing Address: No. 348 Heping West Road, Shijiazhuang 050000, Hebei, P.R.China

Telephone number: +86031185988643/15933629181 E-mail address: [email protected]

Page 2 of 29

Abstract

ip t

Purpose

cr

To evaluate the efficacy of chronic continuous hippocampal deep brain stimulation (DBS) in nonlesional refractory mesial temporal lobe

us

epilepsy.

an

Methods

M

Three adult patients with medically intractable epilepsy treated with hippocampal DBS were studied. Two patients underwent invasive

d

recordings with depth stereo-electroencephalography (SEEG) electrodes

te

to localize ictal onset zone prior to implantation of DBS electrodes. All

Ac ce p

the patients with no lesion in brain magnetic resonance imaging (MRI) scan received bilateral implantation of DBS electrodes. Chronic continuous high-frequency hippocampal stimulation was applied during treatment .The number of seizures in each patient before and after stimulation was compared. Results Long-term hippocampal stimulation produced a median reduction in seizure frequency of 93%. Two out of these patients received unilateral activation of the electrodes and experienced a 95% and 92% reduction in

Page 3 of 29

seizure frequency after hippocampal DBS respectively. The last patient had bilateral electrode activation and had a seizure-frequency reduction of 91%. None of the patients had neuropsychological deterioration and

ip t

showed side effects. Generalised tonic-clonic seizures disappeared

cr

completely after hippocampal DBS.

us

Conclusions

Chronic continuous hippocampal DBS demonstrated a potential

an

efficiency and safety in nonlesional refractory mesial temporal lobe

M

epilepsy and might represent an effective therapeutic option for these

Ac ce p

te

d

patients.

1. Introduction

Epilepsy is one of the most common neurological diseases which

affects 0.5-1% of the general population [1]. Despite formal anti-epileptic drugs treatment, up to 30% of patients still have uncontrolled seizures [2, 3]. Mesial temporal lobe epilepsy is a particularly frequent common form of medically intractable epilepsy [4]. Nonlesional refractory mesial temporal lobe epilepsy characterized by epileptiform discharges in one side or both sides temporal lobes and normal brain MRI scan is not good candidate for ablative surgery of selective amygdalo-hippocampal complex alone or together with the anterior temporal lobe [5-8]. For these

Page 4 of 29

patients, hippocampal DBS has been proposed as a potential therapeutic option. We present the results obtained from these patients with nonlesional

ip t

refractory temporal lobe epilepsy who underwent hippocampal DBS. 2. Material and methods

cr

2.1 Patients

us

Three adult patients with intractable temporal lobe epilepsy were

treated with bilateral DBS electrodes implantation at Department of Functional Neurosurgery of Hebei General Hospital between June 2010

an

to December 2013. The presurgical evaluation included 3T brain MRI, video-EEG telemetry, interictal positron emission tomography (PET),

M

SEEG, magnetoencephalogram as well as neuropsychological and psychiatric assessments.

d

The following clinical symptom characteristics were considered

te

diagnostic for temporal lobe epilepsy: ascending epigastric aura or fear followed by complex partial seizures (CPS) characterised by staring and

Ac ce p

oropharyngeal automatisms (smacking lips, masticating) which may be accompanied by ipsilateral superior limb automatisms or contralateral superior limb dystonia, possibly progressing late to generalized tonic-clonic activity. The cause of epilepsy was not yet clear, which might be associated with birth injury, craniocerebral injury and febrile convulsion (Table 1).

Two patients underwent stereotactic depth electrodes implantation for

intracranial invasive monitoring to localize ictal onset zone before DBS electrodes implantation because of insufficient proof on scalp video-EEG recordings. Multiple contact intracerebral electrodes (diameter, 0.8 mm; 5-18 contacts, 1.5 mm in length and 2 mm apart [Dixi Medical, Besançon, France]) were implanted. A high resolution CT scan was performed after

Page 5 of 29

implantation of the SEEG electrodes to evaluate lesions and co-registered with the preoperative 3D T1-weighted MRI to allow precise localisation of the electrode contacts. It took ten days from removal of the depth recording electrodes to implantation of DBS electrodes.

ip t

These entry standards included: (1) seizure frequency of at least one complex partial seizure per month (2) epileptic discharges in temporal

cr

lobes due to scalp video-EEG recording (3) ictal onset zone located in

unilateral temporal lobe owing to invasive video-EEG monitoring (4) no

us

lesion in MRI scan.

All candidates understood the operational risks and curative effects

an

fully, signed their operation informed consents in this study conducted

M

by the Ethics Committee of Hebei Medical University 2.2 Implantation procedure of DBS electrodes and stimulation

d

paradigm

te

A computerized tomography scan acquired under stereotaxic head frame (CRW, Radionics, USA) was fused with the 3D T1-weighted MRI.

Ac ce p

A plan was made based on stereotactic CT/MRI fusion. The permanent quadripolar DBS electrodes (Medtronic 3487A) were inserted into the bilateral hippocampal heads perpendicular to the hippocampal longitudinal axis through frontal burr holes under general anesthesia. The third and fourth contacts were located in hippocampal head on each side as determined by the fused CT/MRI datasets, while the first and second contacts were located in parahippocampal gyrus (Figure 1). Intra-operative neuronavigation was used during electrode insertion. Each electrode has four cylindrical contacts. The electrodes were connected to the ipsilateral pulse generators (Medtronic Soletra 7426) implanted in subclavicular pockets through the subcutaneous wires (Figure 2). Chronic continuous high frequency stimulation was employed during

Page 6 of 29

treatment. The pulse width was set to 450 µs and remained unchanged in the whole course of treatment. The quadripolar configuration of DBS was employed in which the first contact was set as cathode and the fourth contact was set as the anode.

ip t

The stimulating frequencies of 130, 150 and 170 Hz were suggested in

cr

patients 1, 2, and 3 respectively.

The voltage was increased gradually by 0.1 V every 2 months if the

us

DBS failed to decrease seizures by≥ 90% , to a highest at 3.5 V, or until the patients became seizure-free or adverse reactions appeared.

an

2.3 AEDs

Therapeutic schemes were in accordance with pre-implantation to

M

assess the curative effect of DBS expediently. Subtle change of drug dosages was permitted but taking any new antiepileptic drug was not

te

practice.

d

allowed. The AED regimen was tapered according to best medical

2.4 neuropsychological evaluation

Ac ce p

All the cases were tested by Wechster Adult Intelligence Scale-Revised

China (WAIS-RC) and Wechsler memory scale-revised in China (WMS-RC) before and 1 year after DBS. 2.5 Follow-up and data analysis All the subjects were followed up on out-patient review every 2 weeks

during the first three months after hospital discharge, every three months afterward, or more frequently when necessary. Seizure frequency, seizure intensity, adverse events and pharmacotherapy were carefully recorded on a seizure diary. Simulation parameters were adjusted or unilateral hippocampal stimulation was switched to bilateral hippocampal stimulation due to long-term follow-up relevant assessments.

Page 7 of 29

Follow-up time ranged from 26 to 43 months (mean =34.7months). The mean number of seizures per month averaged over the last six months before implantation except the presurgical evaluation time was established as control and the mean number of monthly seizures averaged

ip t

over the maximal follow-up time after implantation was established as intervention.

cr

Because there were too few samples in this research for a valid

us

statistical interpretation, we have to describe the results. 3. Results

an

Bilateral temporal lobe spikes of interictal non-invasive EEG were showed in two patients (patients 2 and 3), and unilateral temporal lobe spikes in patient 1. Seizures originating from the left temporal lobe of

M

ictal non-invasive EEG recordings were showed in two patients (patients 1 and 2) and from the both temporal lobe in patient 3.

d

The ictal onset which generated from unilateral mesial temporal lobe

te

and conducted to the contralateral medial temporal structures within a few seconds was detected in patient 3 on the invasive video-EEG

Ac ce p

monitoring .

In patient 1, a focal ictal onset in left mesial temporal lobe and

ipsilateral interictal spikes were discovered from non-invasive EEG recordings, which was verified by the local cerebral glucose metabolism decrease in left mesial temporal lobe on PET without invasive EEG monitoring (Figure 3).

In patient 2, the left mesial temporal lobe ictal onset and bilateral mesial temporal lobe interictal spikes were showed on scalp video-EEG, which was confirmed by the focal left-sided hippocampal ictal onset zone on intracranial EEG (Figure 4). In patient 3, ictal nonsynchronous epileptiform discharges in both

Page 8 of 29

temporal lobes were found on scalp video-EEG recordings, whereas a right-sided medial temporal lobe regional ictal onset zone spreading to the contralateral temporal lobe was detected on medial temporal lobe and parahippocampal electrode contacts of invasive EEG monitoring

ip t

(Figure 5).

Postoperative seizure frequencies were compared with preoperative

cr

ones. High-frequency chronic Hip-DBS was able to reduce seizure frequency in this series. All the patients suffered from refractory complex

us

partial seizures with (patient 3) or without (patients 1 and 2) occasional secondary generalized tonic-clonic seizures. Overall seizure frequency

an

reduction ranged from 91-95% (mean=93%) in patients. Patient 1, 2 and 3 experienced respectively a 95%, 92% and 91% reduction in seizure

M

frequency after hippocampal DBS.

Two patients (patients 1 and 2) were activated unilaterally and one patient was activated (patient 3) bilaterally. The activated side was

d

determined by the ictal onset zone on initial EEG recordings or

te

predominant discharges. The second side was activated unless the first activated side failed to decrease seizure by ≥ 90%. Although the

Ac ce p

seizure-frequency reduction of patient 3 after activated bilaterally was relatively inadequate, generalized tonic-clonic seizures disappeared completely in patient 3. The output voltage ranged from 1.0V to 2.5 V (table 2).

Patient 1 responded well to the initial stimulation settings (output, 1.0V;

frequency, 130Hz; pulse width, 450 microseconds) and achieved a firm 95 % seizure frequency reduction. Patient 2 gained a stable 92% seizure frequency reduction using these stimulating parameters (output, 1.6V; frequency, 150Hz; pulse width, 450 microseconds). In patient 3, right-sided hippocampal DBS (output, 2.5V; frequency,

Page 9 of 29

170Hz; pulse width, 450 microseconds) failed to decrease seizure frequency by more than 90% after 30 months of follow-up, and then left-sided hippocampal DBS (output, 1.5V; frequency, 170Hz; pulse reduction was obtained after switching to bilateral DBS.

ip t

width, 450 microseconds) was added, a steady 91% seizure frequency

There was no postoperative neuropsychological deterioration presented

cr

and all the data were within the normal range (table 3).

us

The was no lesion in T1 and T2 weighted images of preoperative MRI (Figure 6).

an

4. Disscussion

In patients with nonlesional refractory temporal lobe epilepsy, no

M

underlying pathological abnormalities such as hippocampal sclerosis, focal cortical dysplasias, vascular or traumatic lesion can be demonstrated on MRI. Seizure outcome after resective surgery in these patients is less

d

advantageous than those patients with structural abnormality. In addition,

te

the excision of MRI-negative temporal lobe structure is associated with a higher risk for postoperative neuropsychological deficit [5, 9, 10]. The

Ac ce p

verbal memory decline and visual-spatial learning impairment was observed after resection of non-atrophic hippocampus in left and right temporal lobe epilepsy respectively [11-13]. Less than 15% of patients with long-term seizure freedom through temporal lobectomy may recur [14]. Therefore, temporal lobe resection may not suitable for all people with refractory epilepsy, especially for patients with normal MRI scan. For these patients, hippocampal DBS as an alternative procedure is available, partly with definite curative effects. Hippocampal DBS has been shown to be successful in the treatment of nonlesional refractory temporal lobe epilepsy, despite a relatively small number of patients in this study. The group of patients with normal MRI had comparable results to the same kind of cases reported by Velasco et al

Page 10 of 29

[15]. These authors observed consecutively the seizure outcome in hippocampal DBS of nine patients (Five had normal MRI scan with a seizure frequency reduction of >95%, while the rest four had hippocampal sclerosis with a seizure frequency reduction of 50-70%)

ip t

with refractory temporal lobe epilepsy ranging from 18 months to 7 years

and came to a conclusion that patients with normal MRI scan would

cr

respond better to hippocampal DBS, this was noted in this series. Vonck

et al also confirmed that hippocampal DBS in patients with normal MRI

us

scan brought about a significant decrease of seizures and interictal epileptiform discharges [16].

an

There are several possible reasons for the less effective DBS in cases with hippocampal sclerosis [15, 17, 18]. First, the incomplete neuronal network attributing to serious lack of neurons in sclerotic mesial temporal

M

tissue was insensitive to hippocampal DBS. Second, the impedance of sclerotic hippocampus was high and resistent to stimulative current,

d

although the patients received approximately the same voltage with or

te

without hippocampal sclerosis. Third, the sclerotic hippocampus was atrophic which might cause the DBS electrode to departure from

Ac ce p

stimulating target. Finally, epileptic patients with normal MRI were good responders to hippocampal DBS. The stimulating parameters should be taken into account when

assessing the efficacy of hippocampal DBS because seizure outcomes were

influenced

directly

by

stimulating

parameters.

Moreover,

stimulating parameters should be tailored individually for the most effective epileptic DBS parameters have not yet been reached a consensus. The seizure reduction in hippocampal stimulation with high-frequency (130-190 Hz) was shown on this research which was in agreement with several reports [19-21]. Böex and his colleagues carried out a comparative study of high (130 Hz) and low (5 Hz) frequency electrical

Page 11 of 29

stimulation in nonlesional temporal lobe epilepsy and drew a conclusion that high-frequency, but not low-frequency stimulation was associated with a seizure reduction[22], as we noted in this series. The EEG desynchronization induced by high-frequency stimulation might produce

ip t

an antiepileptic action.

Seizure outcome was likely to be ameliorated through increasing the

cr

output voltage gradually within the patient's tolerance. Stimulating voltage was set initially at 1.0 V and increased at 0.1 V to a maximum of

us

2.5V. Patient 1 and 2 accepted the electric stimulation at 1.0 V and 1.6 V respectively in case that the seizure frequency reduction of ≥ 90%.

an

Patient 3 experienced bilateral activation with left voltage of 1.5 V and right of 2.5 V.

M

The voltage increment of 0.1 V was safe and effective without adverse reactions in this study, as applied in some reports [16]. Seizure frequency might increase abnormally in the event that voltage increment during

d

hippocampal DBS exceed 0.2 V, which was confirmed by Cukiert et al:

te

seizure frequency had deteriorated in three patients for the augmenter of voltage was set at 0.5 V [18]. Voltage increment was suggested to be set

Ac ce p

at 0.1 V based on the results of this study. For the purpose of abolishing epilepsy or reducing seizure frequency, a

high intensity was applied in hippocampal deep brain stimulation. Boëx et al indicated that low voltage (<1.0 V) accompanying with bipolar configuration lead to increase of seizure frequency [23], therefor, the initial voltage was set to 1.0 V. Tellez-Zenteno and his colleagues reported that the seizure frequency increased surprisingly in one patient of his study who underwent hippocampal stimulation in a quadripolar configuration with a high voltage of 4.5 V, a pulse of 90 μs and at a

frequency of 190 Hz [24]. These findings suggested that high or low voltage might produce excitatory effects thus voltage should be within a suitable range in order to obtain a reduction in seizure frequency. The

Page 12 of 29

voltage between 1.0 V and 2.5 V is safe and effective for the patients with nonlesional refractory temporal lobe epilepsy. The mechanism of hippocampal stimulation in treatment of patients with mesial temporal lobe epilepsy is in the course of exploration.

ip t

Velasco et al have suggested that the perforant pathway activated by DBS

might lead to a polysynaptic inhibition of epileptogenic neurons in

cr

hippocampal CA1-CA4 regions in charge of the initiation and/or propagation of temporal lobe epilepsy [25]. A more likely hypothesis that

us

hippocampal structure as the hub in temporal lobe epileptogenic network is potentially involved in generation and/or propagation of epileptiform

an

activity and stimulative current of DBS could inhibit ictal focus and propagation paths has been taken [26-29]. We endorsed this thesis that hippocampal DBS exert an inhibitory effect not only on ictal focus but

M

also the epileptogenic network through local and remote modulation of network excitability.

d

Recently, several studies have highlighted that the hyperexcitable

te

subiculum might take part in the generation and propagation of temporal lobe epilepsy [30-33]. According to these reports, a satisfactory result was

Ac ce p

achieved in our study by means of high-frequency stimulating hippocampus and subiculum simultaneously to decrease the excitability of subiculum.

Because of too few cases in this research and lack of randomized

control experiment, the optimal parameters for DBS remain unclear. In conclusion, hippocampal DBS may represent both a less traumatic

and more efficient therapeutic regimen for patients with nonlesional refractory temporal lobe epilepsy via the individualized therapeutic regimen and optimum combination of various parameters. References [1] Hirtz, D., Thurman, D.J., Gwinn-Hardy, K., Mohamed, M., Chaudhuri, A.R. and Zalutsky, R. How

Page 13 of 29

common are the "common" neurologic disorders? Neurology 2007; 68:326-337. [2] Regesta, G. and Tanganelli, P. Clinical aspects and biological bases of drug-resistant epilepsies. Epilepsy Res 1999; 34:109-122. [3] Singhvi, J.P., Sawhney, I.M., Lal, V., Pathak, A. and Prabhakar, S. Profile of intractable epilepsy in a tertiary referral center. Neurol India 2000; 48:351-356.

ip t

[4] Engel, J., Jr. Mesial temporal lobe epilepsy: what have we learned? Neuroscientist 2001; 7:340-352.

cr

[5] Morino, M., Ichinose, T., Uda, T., Kondo, K., Ohfuji, S. and Ohata, K. Memory outcome following transsylvian selective amygdalohippocampectomy in 62 patients with hippocampal sclerosis. J Neurosurg 2009; 110:1164-1169.

us

[6] Alpherts, W.C., Vermeulen, J., van Rijen, P.C., da Silva, F.H., van Veelen, C.W. and Dutch Collaborative Epilepsy Surgery, P. Standard versus tailored left temporal lobe resections: differences in cognitive outcome? Neuropsychologia 2008; 46:455-460.

an

[7] Baxendale, S., Thompson, P., Harkness, W. and Duncan, J. Predicting memory decline following epilepsy surgery: a multivariate approach. Epilepsia 2006; 47:1887-1894. [8] Loring, D.W., Meador, K.J., Lee, G.P. and Smith, J.R. Structural versus functional prediction of memory change following anterior temporal lobectomy. Epilepsy Behav 2004; 5:264-268.

M

[9] Akama-Garren, E.H., Bianchi, M.T., Leveroni, C., Cole, A.J., Cash, S.S. and Westover, M.B. Weighing the value of memory loss in the surgical evaluation of left temporal lobe epilepsy: a decision analysis. Epilepsia 2014; 55:1844-1853.

d

[10] de Vanssay-Maigne, A., Boutin, M. and Baudoin-Chial, S. [Predictors of verbal memory decline following temporal lobe surgery]. Neurochirurgie 2008; 54:240-244.

te

[11] Stroup, E., Langfitt, J., Berg, M., McDermott, M., Pilcher, W. and Como, P. Predicting verbal memory decline following anterior temporal lobectomy (ATL). Neurology 2003; 60:1266-1273.

Ac ce p

[12] Kwan, P. and Brodie, M.J. Early identification of refractory epilepsy. N Engl J Med 2000; 342:314-319. [13] Martin, R.C., Kretzmer, T., Palmer, C., Sawrie, S., Knowlton, R., Faught, E., Morawetz, R. and Kuzniecky, R. Risk to verbal memory following anterior temporal lobectomy in patients with severe left-sided hippocampal sclerosis. Arch Neurol 2002; 59:1895-1901. [14] Kelemen, A., Barsi, P., Eross, L., Vajda, J., Czirjak, S., Borbely, C., Rasonyi, G. and Halasz, P. Long-term outcome after temporal lobe surgery--prediction of late worsening of seizure control. Seizure 2006; 15:49-55.

[15] Velasco, A.L., Velasco, F., Velasco, M., Trejo, D., Castro, G. and Carrillo-Ruiz, J.D. Electrical stimulation of the hippocampal epileptic foci for seizure control: a double-blind, long-term follow-up study. Epilepsia 2007; 48:1895-1903.

[16] Vonck, K., Boon, P., Achten, E., De Reuck, J. and Caemaert, J. Long-term amygdalohippocampal stimulation for refractory temporal lobe epilepsy. Ann Neurol 2002; 52:556-565. [17] Cuellar-Herrera, M., Velasco, M., Velasco, F., Velasco, A.L., Jimenez, F., Orozco, S., Briones, M. and Rocha, L. Evaluation of GABA system and cell damage in parahippocampus of patients with temporal lobe epilepsy showing antiepileptic effects after subacute electrical stimulation. Epilepsia 2004; 45:459-466. [18] Cukiert, A., Cukiert, C.M., Burattini, J.A. and Lima, A.M. Seizure outcome after hippocampal deep brain stimulation in a prospective cohort of patients with refractory temporal lobe epilepsy. Seizure

Page 14 of 29

2014; 23:6-9. [19] Velasco, A.L., Velasco, M., Velasco, F., Menes, D., Gordon, F., Rocha, L., Briones, M. and Marquez, I. Subacute and chronic electrical stimulation of the hippocampus on intractable temporal lobe seizures: preliminary report. Arch Med Res 2000; 31:316-328. [20] Vonck, K., Boon, P., Claeys, P., Dedeurwaerdere, S., Achten, R. and Van Roost, D. Long-term deep brain stimulation for refractory temporal lobe epilepsy. Epilepsia 2005; 46 Suppl 5:98-99.

ip t

[21] Fountas, K.N., Smith, J.R., Murro, A.M., Politsky, J., Park, Y.D. and Jenkins, P.D. Implantation of a closed-loop stimulation in the management of medically refractory focal epilepsy: a technical note. Stereotact Funct Neurosurg 2005; 83:153-158.

cr

[22] Boex, C., Vulliemoz, S., Spinelli, L., Pollo, C. and Seeck, M. High and low frequency electrical stimulation in non-lesional temporal lobe epilepsy. Seizure 2007; 16:664-669.

us

[23] Boex, C., Seeck, M., Vulliemoz, S., Rossetti, A.O., Staedler, C., Spinelli, L., Pegna, A.J., Pralong, E., Villemure, J.G., Foletti, G. and Pollo, C. Chronic deep brain stimulation in mesial temporal lobe epilepsy. Seizure 2011; 20:485-490.

an

[24] Tellez-Zenteno, J.F., McLachlan, R.S., Parrent, A., Kubu, C.S. and Wiebe, S. Hippocampal electrical stimulation in mesial temporal lobe epilepsy. Neurology 2006; 66:1490-1494.

M

[25] Velasco, M., Velasco, F., Velasco, A.L., Boleaga, B., Jimenez, F., Brito, F. and Marquez, I. Subacute electrical stimulation of the hippocampus blocks intractable temporal lobe seizures and paroxysmal EEG activities. Epilepsia 2000; 41:158-169.

d

[26] Vonck, K., Boon, P., Goossens, L., Dedeurwaerdere, S., Claeys, P., Gossiaux, F., Van Hese, P., De Smedt, T., Raedt, R., Achten, E., Deblaere, K., Thieleman, A., Vandemaele, P., Thiery, E., Vingerhoets, G., Miatton, M., Caemaert, J., Van Roost, D., Baert, E., Michielsen, G., Dewaele, F., Van Laere, K., Thadani, V., Robertson, D. and Williamson, P. Neurostimulation for refractory epilepsy. Acta Neurol Belg 2003; 103:213-217.

Ac ce p

te

[27] Bellistri, E., Sartori, I., Pelliccia, V., Francione, S., Cardinale, F., de Curtis, M. and Gnatkovsky, V. Fast Activity Evoked by Intracranial 50 Hz Electrical Stimulation as a Marker of the Epileptogenic Zone. Int J Neural Syst 2015; 25:1550022. [28] Toydemir, H.E., Ozkara, C., Uysal, O., Ozyurt, E. and Uzan, M. Complete seizure freedom is possible in patients with MTLE-HS after surgery in spite of extratemporal electro-clinical features. Epilepsy Res 2015; 113:104-112. [29] Chaibi, S., Lajnef, T., Ghrob, A., Samet, M. and Kachouri, A. A Robustness Comparison of Two Algorithms Used for EEG Spike Detection. Open Biomed Eng J 2015; 9:151-156.

[30] Cohen, I., Navarro, V., Clemenceau, S., Baulac, M. and Miles, R. On the origin of interictal activity in human temporal lobe epilepsy in vitro. Science 2002; 298:1418-1421.

[31] Wozny, C., Kivi, A., Lehmann, T.N., Dehnicke, C., Heinemann, U. and Behr, J. Comment on "On the origin of interictal activity in human temporal lobe epilepsy in vitro". Science 2003; 301:463; author reply 463. [32] Stafstrom, C.E. The role of the subiculum in epilepsy and epileptogenesis. Epilepsy Curr 2005; 5:121-129. [33] Huberfeld, G., Menendez de la Prida, L., Pallud, J., Cohen, I., Le Van Quyen, M., Adam, C., Clemenceau, S., Baulac, M. and Miles, R. Glutamatergic pre-ictal discharges emerge at the transition to seizure in human epilepsy. Nat Neurosci 2011; 14:627-634.

Page 15 of 29

Figure legends Figure 1 Position of the DBS electrode contacts in bilateral hippocampal heads

ip t

The images after fusion of postoperative CT and preoperative MRI revealed that the bilateral white round rings were intracranial DBS

cr

electrodes in bilateral hippocampal heads. The third and fourth contacts

were located in hippocampal head on each side, while the first and second

A: the fourth pair of DBS electrode contacts

M

Figure 2 Intracranial stimulus devices

an

B: the third pair of DBS electrode contacts

us

contacts were located in parahippocampal gyrus.

The photos showed the intracranial stimulus devices including the bilateral depth electrodes inserted into the hippocampal heads through

d

frontal burr holes and the bilateral subcutaneous wires connected the

te

electrodes and the pulse generators.

Ac ce p

Figure 3 Ictal onset of patient 1 on scalp video-EEG. A continuity of low or middle amplitude slow wave followed by middle or high spike wave rhythms was detected by left temporal leads (M1, F7, T3) during ictal scalp video-EEG. Figure 4 Ictal onset of patient 2 on intracranial EEG A low amplitude fast rhythm progressing to middle amplitude continuous spike wave was recorded on the first and second contacts of SEEG electrode (LB) implanted in left hippocampal head. Figure 5 Ictal onset of patient 3 on invasive EEG

Page 16 of 29

A spike slow wave rhythm was found on the first ,second and third contacts of SEEG electrode (RB) located in right hippocampal head, subsequently continuous spike wave rhythm was presented on the first

ip t

and second contacts of SEEG electrode (LB) located in contralateral

cr

temporal region.

us

Figure 6 Original preoperative MRI The was no lesion in T1 and T2 weighted images.

an

A,C,E:the T1 weighted image of patient 1,2,3

Ac ce p

te

d

M

B,D,F:the T2 weighted image of patient 1,2,3

Page 17 of 29

Table 1 Demographic characteristic of neuroimaging, invasive video-EEG recording, AED, and etiology Type

AED(md/d)

Ictal onset

Etiology

Pre-DBS CPS

OXC 900, VPA 1000

L focal MT

2 Normal

CPS

VPA 1000, CBZ600,LEV 1000

L focal MT

us

VPA 1000,LTG200

Craniocerebral

R regional MT with

injury

Febrile convulsion

an

3 NormalCPS+GTCS

Birth injury

cr

1 Normal

ip t

Pt MRI

early

M

Left-sided involvement

Pt: patient number; LGT: lamotrigine; LEV: levetiracetam; VPA:valproicacid; OXC:

d

oxcarbazepine; CBZ: carbamazepine; CPS: complex partial seizure; GTCS: Generalised

Ac ce p

te

tonic-clonic seizure; MT: mesial temporal.

Page 18 of 29

Table 2 Overview of changes in mean monthly seizure frequency, stimulation voltage during follow-up 6

St V

Ms

12

St V

Ms

18

St V

Ms

24

St V

Ms

30

St V

Ms

36

St V

Ms

65.0

4.3

L:1.0

5.2

L:1.0

4.7

L:1.0

3.9

L:1.0

3.5

L:1.0

2

38.0

7.3

L:1.2

5.7

L:1.5

3.7

L:1.6

3.4

L:1.6

3.0

L:1.6

3

19.3

15.3

R:1.2

13.6

R:1.5

7.1

R:1.8

6.7

R:2.1

2.6

R:2.4

3.3

2.2

St V

Ms L:1.0

R:2.5

2.0

us

1

42

L:1.1

48

St V

SFR

FU

95

35

92

26

91

43

Ms

ip t

PSF

cr

Pt

R:2.5

1.7

R:2.5

L:1.4

L:1.5

an

Note: 1. Pt: patient number; 2. PRE: preoperative seizure frequency; 3. Ms: mean monthly seizure frequency during months post operation. For example, 6 Ms indicated mean monthly seizure frequency during 6 months post operation; 4. St V: stimulation voltage at the end of the

Ac ce p

te

d

last; 7. FU: follow-up time in months.

M

previously mentioned period; 5. L: left side, R: right side; 6. SFR: seizure frequency reduction at

Page 19 of 29

Table 3 Results of Intellectual and memory level Patient 1

Patient 2

Patient 3

Post-DBS

Pre-DBS

Post-DBS

Pre-DBS

Post-DBS

Verbal IQ

100

96

98

96

102

97

Performance IQ

99

98

98

96

98

Full Scale IQ

100

95

99

95

102

MQ

102

97

101

97

ip t

Pre-DBS

96

cr

95

98

us

101

IQ: intelligence quotient; MQ: memory quotient

Ac ce p

te

d

M

an

Normal range is typically 90-110 for Verbal IQ, Performance IQ, Full Scale IQ and MQ.

Page 20 of 29

cr

ip t

Table

MRI

Type

AED(md/d) Pre-DBS

an

Pt

us

Table 1 Demographic characteristic of neuroimaging, invasive video-EEG recording, AED, and etiology

Normal

CPS

OXC 900, VPA 1000

2

Normal

CPS

VPA 1000, CBZ600,LEV 1000

CPS+GTCS

VPA 1000,LTG200

d

Normal

ep te

3

M

1

Ictal onset

Etiology

L focal MT

Birth injury

L focal MT

Craniocerebral injury

R regional MT with early

Febrile convulsion

Left-sided involvement

Pt: patient number; LGT: lamotrigine; LEV: levetiracetam; VPA:valproicacid; OXC: oxcarbazepine; CBZ: carbamazepine; CPS: complex partial seizure; GTCS:

Ac c

Generalised tonic-clonic seizure; MT: mesial temporal.

Page 21 of 29

cr

ip t

Table

6 Ms

St V

12 Ms

St V

18 Ms

St V

24 Ms

1

65.0

4.3

L:1.0

5.2

L:1.0

4.7

L:1.0

3.9

2

38.0

7.3

L:1.2

5.7

L:1.5

3.7

L:1.6

3.4

3

19.3

15.3

R:1.2

13.6

R:1.5

7.1

R:1.8

St V

6.7

30 Ms

St V

36 Ms

St V

3.3

L:1.0

an

PSF

L:1.0

3.5

L:1.0

L:1.6

3.0

L:1.6

2.6

R:2.4

M

Pt

us

Table 2 Overview of changes in mean monthly seizure frequency, stimulation voltage during follow-up

R:2.1

2.2

R:2.5 L:1.1

42 Ms

2.0

St V

R:2.5 L:1.4

48 Ms

1.7

St V

R:2.5 L:1.5

SFR

FU

95

35

92

26

91

43

Ac c

ep te

d

Note: 1. Pt: patient number; 2. PRE: preoperative seizure frequency; 3. Ms: mean monthly seizure frequency during months post operation. For example, 6 Ms indicated mean monthly seizure frequency during 6 months post operation; 4. St V: stimulation voltage at the end of the previously mentioned period; 5. L: left side, R: right side; 6. SFR: seizure frequency reduction at last; 7. FU: follow-up time in months.

Page 22 of 29

Table

Table 3 Results of Intellectual and memory level Patient 1 Verbal IQ Performance IQ Full Scale IQ MQ

Pre-DBS 100 99 100 102

Post-DBS 96 98 95 97

Patient 2 Pre-DBS 98 98 99 101

Post-DBS 96 96 95 97

Patient 3 Pre-DBS 102 98 102 101

Post-DBS 97 96 95 98

Ac

ce pt

ed

M

an

us

cr

ip t

IQ: intelligence quotient; MQ: memory quotient Normal range is typically 90-110 for Verbal IQ, Performance IQ, Full Scale IQ and MQ.

Page 23 of 29

c Ac

Page 24 of 29

e pt ce Ac Page 25 of 29

p ce Ac Page 26 of 29

ep t Ac c

Page 27 of 29

e pt ce

Ac Page 28 of 29

c Ac

Page 29 of 29