Hippocampal Interictal Spikes during Sleep Impact Long-Term Memory Consolidation.

OBJECTIVE: Non-rapid eye movement (NREM) sleep is supposed to play a key role in long-term memory consolidation transferring information from hippocampus to neocortex. However, sleep also activates epileptic activities in medial temporal regions. This study investigated whether interictal hippocampal spikes during sleep would impair long-term memory consolidation. METHOD: We prospectively measured visual and verbal memory performance in 20 patients with epilepsy investigated with stereoelectroencephalography (SEEG) at immediate, 30-minute, and 1-week delays, and studied the correlations between interictal hippocampal spike frequency during waking and the first cycle of NREM sleep and memory performance, taking into account the number of seizures occurring during the consolidation period and other possible confounding factors, such as age and epilepsy duration. RESULTS: Retention of verbal memory over 1 week was negatively correlated with hippocampal spike frequency during sleep, whereas no significant correlation was found with hippocampal interictal spikes during waking. No significant result was found for visual memory. Regression tree analysis showed that the number of seizures was the first factor that impaired the verbal memory retention between 30 minutes and 1 week. When the number of seizures was below 5, spike frequency during sleep higher than 13 minutes was associated with impaired memory retention over 1 week. INTERPRETATION: Our results show that activation of interictal spikes in the hippocampus during sleep and seizures specifically impair long-term memory consolidation. We hypothesize that hippocampal interictal spikes during sleep interrupt hippocampal-neocortical transfer of information. ANN NEUROL 2020;87:976-987.

Accelerated long-term forgetting in focal epilepsy: Do interictal spikes during sleep matter?

Accelerated long-term forgetting (ALF) is a particular form of amnesia mostly encountered in focal epilepsy, particularly in temporal lobe epilepsy. This type of memory loss is characterized by an impairment of long-term consolidation of declarative memory, and its mechanisms remain poorly understood. In particular, the respective contribution of lesion, seizures, interictal epileptic discharges, and sleep is still debated. Here, we provide an overview of the relationships intertwining epilepsy, sleep, and memory consolidation and, based on recent findings from intracranial electroencephalographic recordings, we propose a model of ALF pathophysiology that integrates the differential role of interictal spikes during wakefulness and sleep. This model provides a framework to account for the different timescales at which ALF may occur.

Rapid detection of generalized anxiety disorder and major depression in epilepsy: Validation of the GAD-7 as a complementary tool to the NDDI-E in a French sample.

OBJECTIVE: Generalized anxiety disorder (GAD) in people with epilepsy (PWE) is underdiagnosed and undertreated. The GAD-7 is a screening questionnaire to detect GAD. However, the usefulness of the GAD-7 as a screening tool in PWE remains to be validated. Thus, we aimed to: (1) validate the GAD-7 in French PWE and (2) assess its complementarity with regard to the previously validated screening tool for depression, the Neurological Disorders Depression Inventory for Epilepsy (NDDI-E). METHODS: This study was performed under the auspices of the ILAE Commission on Neuropsychiatry. People with epilepsy >18 years of age were recruited from the specialist epilepsy unit in Marseille, France. The Mini-International Neuropsychiatric Interview (MINI) was performed as gold standard, and the Penn State Worry Questionnaire (PSWQ) and the NDDI-E were performed for external validity. Data were compared between PWE with/without GAD using Chi(2) test and Student's t-test. Internal structural validity, external validity, and receiver operator characteristics were analyzed. A principal component factor analysis with Varimax rotation was performed on the 13 items of the GAD-7 (7 items) plus the NDDI-E (6 items). RESULTS: Testing was performed on 145 PWE: mean age = 39.38 years old (SD=14.01, range: 18-75); 63.4% (92) women; 75.9% with focal epilepsy. Using the MINI, 49 (33.8%) patients had current GAD. Cronbach's alpha coefficient was 0.898, indicating satisfactory internal consistency. Correlation between GAD-7 and the PSQW scores was high (r (145)=.549, P<.0001), indicating good external validity. Factor analysis shows that the anxiety investigated with the GAD-7 and depression investigated with the NDDI-E reflect distinct factors. Receiver operator characteristic analysis showed area under the curve of 0.899 (95% CI 0.838-0.943, P < 0.0001) indicating good capacity of the GAD-7 to detect GAD (defined by MINI). Cutoff for maximal sensitivity and specificity was 7. Mean GAD-7 score in PWE with GAD was 13.22 (SD = 3.99), and that without GAD was 5.17 (SD = 4.66). SIGNIFICANCE: This study validates the French language version of the GAD-7 screening tool for generalized anxiety in PWE, with a cutoff score of 7/21 for GAD, and also confirms that the GAD-7 is a short and easily administered test. Factor analysis shows that the GAD-7 (screening for generalized anxiety disorder) and the NDDI-E (screening for major depression) provide complementary information. The routine use of both GAD-7 and NDDI-E should be considered in clinical evaluation of patients with epilepsy.

One step closer to a global tool for rapid screening of major depression in epilepsy: validation of the French NDDI-E.

OBJECTIVE: Depression in people with epilepsy (PWE) is underdiagnosed and undertreated. The Neurological Disorders Depression Inventory for Epilepsy (NDDI-E) is a screening questionnaire used for detecting major depressive episode (MDE) in PWE, and is already validated in 10 languages. However a version in French, one of the world's widely spoken languages, was, until now, lacking. We aimed to translate and validate the French NDDI-E. METHODS: This study was performed under the auspices of the ILAE. People with epilepsy >18years of age were recruited from 2 specialist epilepsy units in Marseille, France. Two native French speakers and 2 native English speakers performed a forward-backward translation. The Mini International Neuropsychiatric Interview (MINI) was performed as the gold standard, and the Center for Epidemiological Studies Depression symptoms index (CES-D) was performed for external validity. Data were compared between PWE with MDE and PWE without MDE using the chi-square test and Student's t-test. Internal structural validity, external validity, and receiver operator characteristics were analyzed. RESULTS: Testing was performed on 116 PWE: mean age=40.39years (SD=13.83, range: [18-81]years old); 58.6% (68) were women; 87.1% had focal epilepsy. Using the MINI, we found that 33 (28.4%) patients had current MDE and that 15 (12.9%) patients had dysthymia; also, we found that 37 (31.9%) patients presented suicidal ideation and/or behavior. Cronbach's alpha coefficient was 0.838, indicating satisfactory internal consistency. Correlation between the NDDI-E and the CES-D scores was high (r(116)=0.817, p<0.0001), indicating good external validity. Receiver operator characteristic analysis showed an area under the curve of 0.958 (95% CI=0.904-0.986), (p<0.0001), indicating good capacity of the NDDI-E to detect MDE (defined by MINI). The cutoff for maximal sensitivity and specificity was 15. The mean NDDI-E score in PWE with MDE was 18.27 (SD=2.28), and the mean NDDI-E score in PWE without MDE was 10.61 (SD=3.63). SIGNIFICANCE: This study validated the French NDDI-E, with a cutoff score of 15/24 for MDE, similar to previous studies, and reinforces the NDDI-E as a global tool for detection of MDE.

Simultaneous recording of MEG, EEG and intracerebral EEG during visual stimulation: from feasibility to single-trial analysis.

Electroencephalography (EEG), magnetoencephalography (MEG), and intracerebral stereotaxic EEG (SEEG) are the three neurophysiological recording techniques, which are thought to capture the same type of brain activity. Still, the relationships between non-invasive (EEG, MEG) and invasive (SEEG) signals remain to be further investigated. In early attempts at comparing SEEG with either EEG or MEG, the recordings were performed separately for each modality. However such an approach presents substantial limitations in terms of signal analysis. The goal of this technical note is to investigate the feasibility of simultaneously recording these three signal modalities (EEG, MEG and SEEG), and to provide strategies for analyzing this new kind of data. Intracerebral electrodes were implanted in a patient with intractable epilepsy for presurgical evaluation purposes. This patient was presented with a visual stimulation paradigm while the three types of signals were simultaneously recorded. The analysis started with a characterization of the MEG artifact caused by the SEEG equipment. Next, the average evoked activities were computed at the sensor level, and cortical source activations were estimated for both the EEG and MEG recordings; these were shown to be compatible with the spatiotemporal dynamics of the SEEG signals. In the average time-frequency domain, concordant patterns between the MEG/EEG and SEEG recordings were found below the 40 Hz level. Finally, a fine-grained coupling between the amplitudes of the three recording modalities was detected in the time domain, at the level of single evoked responses. Importantly, these correlations have shown a high level of spatial and temporal specificity. These findings provide a case for the ability of trimodal recordings (EEG, MEG, and SEEG) to reach a greater level of specificity in the investigation of brain signals and functions.

A comparison of methods for assessing alpha phase resetting in electrophysiology, with application to intracerebral EEG in visual areas.

There are two competing views on the mechanisms underlying the generation of visual evoked potentials/fields in EEG/MEG. The classical hypothesis assumes an additive wave on top of background noise. Another hypothesis states that the evoked activity can totally or partially arise from a phase resetting of the ongoing alpha rhythm. There is no consensus however, on the best tools for distinguishing between these two hypotheses. In this study, we have tested different measures on a large series of simulations under a variety of scenarios, involving in particular trial-to-trial variability and different dynamics of ongoing alpha rhythm. No single measure or set of measures was found to be necessary or sufficient for defining phase resetting in the context of our simulations. Still, simulations permitted to define criteria that were the most reliable in practice for distinguishing additive and phase resetting hypotheses. We have then applied these criteria on intracerebral EEG data recordings in the visual areas during a visual discrimination task. We investigated the intracerebral channels that presented both ERP and ongoing alpha oscillations (n=37). Within these channels, a total of 30% fulfilled phase resetting criteria during the generation of the visual evoked potential, based on criteria derived from simulations. Moreover, 19% of the 37 channels presented dependence of the ERP on the level of pre-stimulus alpha. Only 5% of channels fulfilled both the simulation-related criteria and dependence on baseline alpha level. Our simulation study points out to the difficulty of clearly assessing phase resetting based on observed macroscopic electrophysiological signals. Still, some channels presented an indication of phase resetting in the context of our simulations. This needs to be confirmed by further work, in particular at a smaller recording scale.

Reading epilepsy from the dominant temporo-occipital region.

BACKGROUND: Reading epilepsy is a rare form of epilepsy, classified among idiopathic, age- and localisation-related (partial) epilepsies as a reflex epilepsy syndrome. Seizures usually consist of myoclonic jerks restricted to the jaw. However, distinct ictal features including visual symptoms and paroxysmal a- or dyslexia are described in some patients. The anatomical substrate of ictogenesis in reading epilepsy remains poorly understood. METHODS: The authors report here the case of a primary reading epilepsy for which ictal semiology was characterised by visual symptoms and dyslexia, investigated by MRI, interictal high-resolution EEG and PET, ictal video-EEG and SPECT. Brain MRI was normal. Interictal high-resolution EEG was performed with 64 scalp channels, a realistic head model and different algorithms to solve the inverse problem. RESULTS: Interictal source localisations highlighted the left occipito-temporal junction. Interictal PET demonstrated bilateral occipito-temporal hypometabolism with left-sided predominance. Ictal EEG showed a rhythmic discharge in left temporo-parieto-occipital junction channels, with left occipito-temporal predominance. MRI fusion of the coregistered subtraction between ictal and interictal SPECT individualised relative hyperperfusion affecting (a) the left occipito-parietal junction area, (b) the left lateral middle and inferior temporal gyri and (c) the left inferior frontal area. CONCLUSION: Besides reading-induced myoclonic jerks of the jaw, a second variant of reading epilepsy exists with clearly partial seizures manifested by visual symptoms and a- or dyslexia. These seizures originate from the occipito-temporal region of the dominant hemisphere, corresponding to the posterior part of the neural network that underlies the function of reading.

Brain regions underlying word finding difficulties in temporal lobe epilepsy.

Word finding difficulties are often reported by epileptic patients with seizures originating from the language dominant cerebral hemisphere, for example, in temporal lobe epilepsy. Evidence regarding the brain regions underlying this deficit comes from studies of peri-operative electro-cortical stimulation, as well as post-surgical performance. This evidence has highlighted a role for the anterior part of the dominant temporal lobe in oral word production. These conclusions contrast with findings from activation studies involving healthy speakers or acute ischaemic stroke patients, where the region most directly related to word retrieval appears to be the posterior part of the left temporal lobe. To clarify the neural basis of word retrieval in temporal lobe epilepsy, we tested forty-three drug-resistant temporal lobe epilepsy patients (28 left, 15 right). Comprehensive neuropsychological and language assessments were performed. Single spoken word production was elicited with picture or definition stimuli. Detailed analysis allowed the distinction of impaired word retrieval from other possible causes of naming failure. Finally, the neural substrate of the deficit was assessed by correlating word retrieval performance and resting-state brain metabolism in 18 fluoro-2-deoxy-d-glucose-Positron Emission Tomography. Naming difficulties often resulted from genuine word retrieval failures (anomic states), both in picture and in definition tasks. Left temporal lobe epilepsy patients showed considerably worse performance than right temporal lobe epilepsy patients. Performance was poorer in the definition than in the picture task. Across patients and the left temporal lobe epilepsy subgroup, frequency of anomic state was negatively correlated with resting-state brain metabolism in left posterior and basal temporal regions (Brodmann's area 20-37-39). These results show the involvement of posterior temporal regions, within a larger antero-posterior-basal temporal network, in the specific process of word retrieval in temporal lobe epilepsy. A tentative explanation for these findings is that epilepsy induces functional deafferentation between anterior temporal structures devoted to semantic processing and neocortical posterior temporal structures devoted to lexical processing.

Stereoelectroencephalography in presurgical assessment of MRI-negative epilepsy.

According to most existing literature, the absence of an MRI lesion is generally associated with poorer prognosis in resective epilepsy surgery. Delineation of the epileptogenic zone (EZ) by intracranial recording is usually required but is perceived to be more difficult in 'MRI negative' cases. Most previous studies have used subdural recording and there is relatively less published data on stereoelectroencephalography (SEEG). The objective of this study was to report the experience of our group in using SEEG in presurgical evaluation, comparing its effectiveness in normal and lesional MRI cases. One hundred consecutive patients undergoing SEEG for presurgical assessment were studied. Forty-three patients out of one hundred (43%) had normal MRI and 57 (57%) had lesional MRI. Successful localization was achieved with no difference between these two groups, in 41/43 (95%) normal MRI and in 55/57 (96%) lesional MRI cases (P = 1.00). Surgery was proposed in 84/100 patients and contraindicated in 16/100 with no significant difference between lesional and MRI-negative groups (P > 0.05). At 1 year follow-up, 11/20 (55%) of those having undergone cortectomy in the MRI-negative group and 21/40 (53%) in the lesional MRI group were entirely seizure free (P > 0.05) and these proportions were maintained at 2 years follow-up. Significant improvement in seizure control (ILAE outcome groups 1-4) was achieved in >90% cases with no difference between groups (P > 0.05). Of MRI-negative cases that underwent surgery, 10/23 (43%) had focal cortical dysplasia. This series showed that SEEG was equally effective in the presurgical evaluation of MRI-negative and lesional epilepsies.

High frequency gamma activity in the left hippocampus predicts visual object naming performance.

Access to an object's name requires the retrieval of an arbitrary association between it's identity and a word-label. The hippocampus is essential in retrieving arbitrary associations, and thus could be involved in retrieving the link between an object and its name. To test this hypothesis we recorded the iEEG signal from epileptic patients, directly implanted in the hippocampus, while they performed a picture naming task. High-frequency broadband gamma (50-150 Hz) responses were computed as an index of population-level spiking activity. Our results show, for the first time, single-trial hippocampal dynamics between visual confrontation and naming. Remarkably, the latency of the hippocampal response predicts naming latency, while inefficient hippocampal activation is associated with "tip-of-the-tongue" states (a failure to retrieve the name of a recognized object) suggesting that the hippocampus is an active component of the naming network and that its dynamics are closely related to efficient word production.

Electrophysiological study of the basal temporal language area: a convergence zone between language perception and production networks.

OBJECTIVE: Regions involved in language processing have been observed in the inferior part of the left temporal lobe. Although collectively labelled 'the Basal Temporal Language Area' (BTLA), these territories are functionally heterogeneous and are involved in language perception (i.e. reading or semantic task) or language production (speech arrest after stimulation). The objective of this study was to clarify the role of BTLA in the language network in an epileptic patient who displayed jargonaphasia. METHODS: Intracerebral evoked related potentials to verbal and non-verbal stimuli in auditory and visual modalities were recorded from BTLA. Time-frequency analysis was performed during ictal events. RESULTS: Evoked potentials and induced gamma-band activity provided direct evidence that BTLA is sensitive to language stimuli in both modalities, 350 ms after stimulation. In addition, spontaneous gamma-band discharges were recorded from this region during which we observed phonological jargon. CONCLUSION: The findings emphasize the multimodal nature of this region in speech perception. In the context of transient dysfunction, the patient's lexical semantic processing network is disrupted, reducing spoken output to meaningless phoneme combinations. SIGNIFICANCE: This rare opportunity to study the BTLA "in vivo" demonstrates its pivotal role in lexico-semantic processing for speech production and its multimodal nature in speech perception.

Hemispheric lateralization of voice onset time (VOT) comparison between depth and scalp EEG recordings.

Auditory-evoked potential (AEP)s elicited to French-language voiced stop consonant (/ba/) and voiceless stop consonant (/pa/) were studied in non-language-impaired epileptic patients and non-epileptic volunteers. First, depth AEPs recorded from the primary auditory cortex during pre-surgical exploration and scalp AEPs recordings using high resolution EEG (HR EEG-64 channels scalp EEG) were compared in the same patients. Both methods indicated that the processing of voiced and voiceless consonants was based on a temporal auditory coding. /Ba/ elicited a first complex (N1) at the onset of voicing and a second component [release component (RC)] time-locked to release. This processing took place specifically in the left primary auditory cortex. Source modeling of the RC showed that a left-greater-than-right amplitude of source probes (SP) both in epileptic patients with left-hemispheric language dominance [established by means of invasive tests (WADA test) and/or clinical data] and right-handed non-epileptic subjects. Our data suggest that the processing of VOT is related to hemispheric dominance for language and that scalp-recorded AEPs may represent an effective, non-invasive method to establish hemispheric dominance for language in clinical settings. This procedure could complement existing methods and could help to detect the dissociation between receptive and expressive language sometimes observed in patients with epilepsy.