Search

Objective To perform a side-by-side comparison of two epileptogenicity biomarkers, high frequency oscillations (HFOs) and delayed responses (DRs), as a result of single-pulse electrical stimulation. Methods We have recorded stimulation-evoked HFOs and DRs in 16 epileptic patients undergoing presurgical evaluation using the stereoelectroencephalographic method. To evaluate converging and complementary information provided by the biomarkers, we analyzed them individually and for logical “and”/“or” combinations between them. 3D maps of the biomarkers’ distributions by recording location (inbound maps) and by stimulation location (outbound maps) were created to analyze their relationship with the epileptogenic structures. Results HFOs occur less frequently than DRs, by 18.7%, when counting by recording contacts, and more frequently, by 7.4%, when counting by stimulation contacts. 40.6% of the contacts exhibiting HFOs also exhibit DRs, and 44.1% of the contacts exhibiting DRs also exhibit HFOs. When combining biomarkers, there was a tradeoff between increased seizure onset zone (SOZ) sensitivity, from 21.3% to 73%, and decreased specificity, from 87.2% to 34.3%. Conclusions There is a moderate similarity in the information provided by the DRs and HFOs. Significance The biomarkers complement each other, but there is a tradeoff between different metrics for SOZ localization.

OBJECTIVE: To investigate functional coupling between brain networks using spectral changes induced by single-pulse electric stimulation (SPES).METHOD: We analyzed 20 patients with focal epilepsy, implanted with depth electrodes. SPES was applied to each pair of adjacent contacts, and responses were recorded from all other contacts. The mean response amplitude value was quantified in three time-periods after stimulation (10-60, 60-255, 255-500ms) for three frequency-ranges (Gamma, Ripples, Fast-Ripples), and compared to baseline. A total of 30,755 responses were analyzed, taking into consideration three dichotomous pairs: stimulating in primary sensory areas (S1-V1) vs. outside them, to test the interaction in physiologic networks; stimulating in seizure onset zone (SOZ) vs. non-SOZ, to test pathologic interactions; recording in default mode network (DMN) vs. non-DMN.RESULTS: Overall, we observed an early excitation (10-60ms) and a delayed inhibition (60-500ms). More specifically, in the delayed period, stimulation in S1-V1 produced a higher gamma-inhibition in the DMN, while stimulation in the SOZ induced a higher inhibition in the epilepsy-related higher frequencies (Ripples and Fast-Ripples).CONCLUSION: Physiologic and pathologic interactions can be assessed using spectral changes induced by SPES.SIGNIFICANCE: This is a promising method for connectivity studies in patients with drug-resistant focal epilepsy.

Background Temporal lobe epilepsy is frequently improved after selective resection. However, a small number of these patients have temporal-plus epilepsy and continue seizing. We aim to study the connectivity of the temporal lobe during wakefulness and sleep. Material and methods We included 25 patients with drug-resistant focal epilepsy who had depth electrodes exploration in 2016–2018. We implanted 9–18 electrodes per patient over a large network. We performed single pulse electrical stimulation protocol, as previously described, during wakefulness and sleep. We selected 40,678 statistically significant responses (rho > 0.4, p Results The difference between the mean responses in all temporal structures between sleep and wakefulness is 10.41 μV (p Conclusions Temporal lobe connectivity is modulated by sleep, EZ location, and pertainence of certain structures to the EZ, with statistically significant differences between left and right temporal structures.

Objectives Intracranial direct electrical stimulation (iDES) uses different parameters for mapping the epileptogenic and functional areas in patients with drug-resistant epilepsy. We aim at finding the common factor driving the electrographic responses to various iDES protocols reported in the literature. Methods We recorded early responses to single-pulse iDES in 11 subjects undergoing stereoelectroencephalographic presurgical evaluation. We systematically explored the role of several pulse parameters in evoking responses: monophasic versus biphasic pulses, current intensity, and pulse duration. We performed a correlation and regression analysis between responses to different protocols by amplitude, duration, and charge per phase. Results Regression analysis revealed that the responses were similar for the same charge per phase, regardless of their pulse duration and amplitude. Over eighty percent (82.8%) of the responses to variable pulse duration biphasic stimulation and between 58.6% and 81.9% of the responses to monophasic stimulation, depending on pulse polarity, were correlated to the responses evoked by the variable amplitude biphasic protocol, when expressing stimulus strength in terms of charge per phase. Conclusions Regardless of the combination of different stimulation currents, it is the underlying charge per phase parameter that determines the magnitude of the responses to single-pulse electrical stimulation. Significance Our results provide a unifying method for comparing iDES protocols.

Pre surgical evaluation in drug resistant epilepsy is based also on invasive EEG recordings in order to accurately establish the epileptogenic zone for surgical resection as well as the eloquent cortex that should be spared. During intracranial session we record ictal events and we apply different stimulation protocols to study functional cortex. In our centre we use the stereo-EEG method to identify the epileptogenic networks and mainly high frequency (50 Hz) stimulation for cortical mapping. By modulating the properties of the stimulation waveforms, in conjunction with the nonlinear response of the tissue to electrical stimulation, we were able to recover the physiological responses during the course of the stimulation train. The activation of the networks at stimulation levels evoking a clinical symptom was compared with the below-threshold stimulation, that did not elicit any clinical effects, allowing to evidence a selective recruitment of pathways associated with a particular symptom. The method is used for identification of physiologic networks associated with various somatosensory, motor or language-related symptoms, to investigate the impact of epileptogenic zone to functional cortex as well as to predict a possible post surgical deficit.

Objectives By analyzing the responses to intracranial electrical stimulation we aim at im-proving the localization of the seizure onset zone (SOZ) and the connectivity between different areas. Methods Our center uses a set of traditional and innovative advanced analysis methods for stereo-EEG signals recorded using depth electrodes. We analyze stimulation-evoked activity in time-frequency domain to detect inter-ictal epileptogenicity biomarkers like high-frequency oscillations (HFO, f > 80 Hz) and delayed responses (DR, t > 100 ms). Results of the different quantitative EEG analyses are represented as 3-D maps superimposed with patient’s anatomy, as visible on various imaging modalities, and their evolution in time is shown as movies. Cortico-cortical evoked potentials (CCEP) based on early responses (t = 10–110 ms) to single-pulse intracranial stimulation are used to calculate functional connectivity between cortical areas. Results Biomarker response maps and effective connectomes were calculated in a population of 24 patients undergoing presurgical evaluation for drug-resistant epilepsy. The sensitivity with respect to SOZ of HFOs was higher than the one of the DR by both response and stimulation location, whereas the specificity of the DRs was higher by the lo-cation of the stimulation. Discussion Despite variable sensitivity and specificity of the biomarkers, the positive pre-diction level of these biomarkers is less than 30%, taken separately or in combination. Conclusions Analysis of responses to intracranial electrical stimulation can contribute to the identification of the epileptogenic networks and for delineating the epileptogenic are-as. Significance We illustrate the value of using multiple advanced signal analysis techniques.

Objective We aim to study the connections of the posterior insula (pI) and anterior insula (aI) through cortico-cortical evoked potentials (CCEP) by single pulse electrical stimulation (SPES). Method We performed SPES stimulation protocol in a group of 8 refractory epilepsy patients presurgically explored with intracerebral depth electrodes, out of which 6 were operated on and are currently seizure-free. We selected the early responses (ER) obtained from contacts of 24 electrodes that were included in the anterior and posterior insula, when this structure was outside the seizure onset zone (SOZ). Responses were analyzed in terms of amplitude variations during wakefulness and sleep. Effective connectivity betweeen cortical was calculated based on the measured CCEPs. Results The ER amplitudes by SPES stimulation in the insula during wakefulness compared to sleep showed statistically significant ( p 0.05 ) differences in 6 patients. There are constant connections between aI and pI. We identified connectivity of the pI with language-related brain stuctures during wakefulness, and with the temporal mesial structures during sleep, possibly influenced by the SOZ. Right pI stimulation seems to elicit ampler reponses in the rolandic operculum and parietal structures during wakefulness, and in the primary motor and premotor cortex during sleep. Anterior insula has preferential connectivity with anterior cingulate gyrus during wakefulness and with the fusiform gyrus during sleep. Discussions Connectivity of the aI versus pI, as well as right versus left insula during wakefulness and sleep is different, partially supporting previous observations. Conclusion Sleep has been shown to alter the physiological connectivity of the insula. Further investigations are in progress to confirm these results. Significance These findings describe insular connectivity and variability during sleep, with possible implications in epilepsy surgery, sleep disorders understanding and language studies.

Objectives To assess the role of signal analysis methods for accurate delineation of the epileptogenic zone during surgery for drug resistant epilepsy. Methods We used surface video EEG as well as invasive studies by means of the stereo-EEG method to identify the epileptogenic networks. Signal analysis tools have been applied systematically during pre surgical evaluation to study ictal as well as interictal activity and responses to different stimulation protocols. Connectivity charts were computed based on analysis of cortico-cortical evoked potentials on single pulse electrical stimulation (SPES). The results were interpreted according with an individual patient hypothesis and integrated with clinical and anatomic data. Results We explored 63 patients at the University Hospital of Bucharest in the National Program for Epilepsy Surgery between 2012 and 2016 of which 57 received either open surgeries or undergone a procedure of radio frequency thermocoagulation (RFTC) applied on intracranial electrodes. In the group of patients ( n = 34) with more than 2 years follow up after surgery, 25 (74%) are seizure free. Signal post processing analysis was critical for surgical planning in 12 cases (48%). Discussion Systematic use of signal analysis approach improved the definition of surgical resection limits, made the recording sessions shorter without sacrificing information, and allowed a better definition of targets for RFTC protocols during SEEG recordings. Conclusion Signal analysis tools added complementary information where the traditional visual analysis was not helpful for surgical planning. Significance Post processing methods of the EEG signal integrated with the anatomo-clinical hypothesis offer unbiased valuable data to improve surgical outcome.

Objective To investigate default-mode network (DMN) deactivation in different types of focal epilepsy by single pulse electric stimulation (SPES). Method We analyzed 18 patients with focal epilepsy (8 frontal, 8 temporal, 2 posterior), implanted with depth electrodes. SPES was applied to each pair of adjacent contacts in seiwure onset zone (SOZ), and responses were recorded from the 10 different DMN-hubs; The responses to SPES in Gamma, Ripple (R) and Fast-Ripple (FR) bands were quantified in a 60–250 ms time-window following each stimulation pulse and compared to baseline. Time of propagation was determined based on N1 latency. These responses were compared across the three epilepsy types. Results Overall, we observed a significant deactivation of the DMN in the upper frequency ranges for all epilepsies (75% drop from baseline for gamma and 93% for R and FR). In the gamma band, the stimulation of frontal SOZs deactivated the ipsilateral posterior cingulate and precuneus significantly more than in the other epilepsy-types, with shorter general latencies of 22, 25 and 23 ms, respectively. Discussion Our results might explain why in prefrontal focal epilepsies and in generalized epilepsies, thought to have a frontal generator, the consciousness is preferentially and significantly impaired. Significance This is a promising method for quantifying the engagement of the consciousness processing areas by the epileptic focus. Detailed illustrations and tables for all 10 DMN-hubs, in all 3 high-frequency bands are provided.

Objectives We aim to map the function and connectivity of the cingulate gyrus (CG) using cortical direct electrical stimulation. Methods We selected patients explored by SEEG at the Emergency University Hospital Bucharest and at the Strasbourg University Hospital, that had minimum one electrode sampling the CG (anterior-ACG, middle-MCG or posterior-PCG), outside the epileptogenic zone. We systematically reviewed the functional stimulations (bipolar, 50 Hz, 0.25–3 mA). For 17 patients explored in Bucharest, cortico-cortical evoked potentials (CCEPs) elicited by single pulse electrical stimulations (SPES) were used to obtain the effective connectivity at a group level for each part of the CG. Results We included 429 bipolar stimulations from 108 patients. From 151 stimulations in the ACG, 24.5% elicited a clinical response like: emotions (19%), vestibular (13.5%), thoracic sensation (13.5%), smile (11%). There are 135 stimulations in the MCG of which 53.3% elicited clinical responses like: elementary motor (24.94%), proprioception (22.22%), elementary somatosensitive (13.2%), compulsive motor (2.77%). There were 143 stimulations in the PCG and only 18.88% were positive, for example: executive function (11.11%), vestibular (7.4%). CCEPs analysis revealed that ACG is connected with the prefrontal (frontobasal and frontopolar) regions, anterior insula, mesiotemporal regions. MCG is connected with the primary and supplementary motor areas, insulo-perisylvian region and parietal lobe. PCG is connected with the ACG, MCG, prefrontal, parietal and temporal regions. Discussion High frequency stimulations elicit clinical responses as a result of a network effect that we have mapped with CCEPs. Conclusions There is a regional organization of the cingulate gyrus for each part both in function and connectivity. Significance These findings help us understand the functional organization of the cingulate gyrus with possible implications in epilepsy surgery, depression or other mental disorders.

Objectives In deep comatose states, the EEG appears discontinuous, consisting of bursts of activity on a flat electrical background referred to as the burst-suppression (BS) pattern. We aimed to develop a quantitative EEG measure that can be applied to both BS and continuous EEG coma. Methods We recoded EEG changes in response to intermittent photic or electrical nerve stimulation epochs lasting 1 min. The multi-channel EEG was reduced to a binary signal of alternating stimulus processing and default classes. A default class ratio could then be defined for continuous EEG by analogy with the suppression ratio of BS. A global EEG reactivity index was derived to measure the relative reduction in default class ratio during (effects) and immediately following stimulation (after-effects). Results We found that in patients with BS coma, the EEG reactivity index correlated with clinical coma evaluation and prognostication scales. The same reactivity index measured during BS induced by general anesthesia, was found to reflect the degree of brain injury in experimental rodent models. Ongoing studies indicate that the EEG reactivity index also correlates with the clinical severity of coma with continuous EEG. Discussion Our data suggest that during BS, the bursts reflect stimulus-evoked activity whereas the suppression reflects the missing default activity. Conclusions We propose a novel EEG reactivity index that can be measured in BS and continuous EEG coma alike. Significance Our EEG reactivity index could provide a continuous measure for monitoring recovery from coma and the post-comatose disorders of consciousness. This work was supported by grants of the Romanian National Authority for Scientific Research, CNCS – UEFISCDI, project number PN-II-ID-PCE-2011-3-0847, PN-II-PT-PCCA-2011-3.2-1290 and PN-III-P2-2.1-PTE-2016-0114 to MM.

Purpose Design a protocol for probing the plastic nature of the epileptic networks. Methods We selected 3 patients with pharmaco-resistant epilepsy that were explored in the SEEG method. Single pulse electrical stimulations (SPES) (biphasic, 3 ms, 0.25–5 mA) were applied to adjacent contacts. We calculated the early responses in the 10–110 ms interval and considered only connections between contacts having a RMS value within the 3rd quartile of all the responses in an individual patient, correlated with the stimulation current (Spearman‘s rho >0.5, p Sublobar anatomical structures projecting to SOZ (inbound connections), as well as SOZ’s projections to other structures (outbound connections) were systematically assessed and named generically “epileptome”. SOZ‘s contacts were also stimulated in various relevant clinical situations (first, second patient-interictal awake vs. sleep vs. postictal, third-on medication vs. off) to understand the dynamics of the outbound analysis. Results Lack of medication or sleep activated SOZ’s network (mean increase in RMS 6%, 28% to baseline) and introduced new structures in the epileptome (hippocampus in one patient, contralateral basal temporal and precuneus in another). Post-seizure, SOZ lost some nearby connections and gained distant ones, not present interictally. Conclusions Intra-individually the epileptic focus engages in an evolving network configuration that can be accurately described.

a Stereoelectroencephalographic 3-D mapping of epileptogenicity using responses to single-pulse electrical stimulation Cristian Donos1 , Ioana Mindruta4,5, Jean Ciurea3, Alin Rasina3, Mihai Dragos Malaia5 ,Andrei Barborica1,2 1 Physics Department, University of Bucharest, Bucharest, Romania, 2 FHC Inc, Bowdoin ME, USA, 3 Neurosurgery Department, Bagdasar-Arseni Hospital, Bucharest, Romania, 4 Neurology Department, Carol Davila University of Medicine and Pharmacy, Bucharest, Romania, 5 Neurology Department, University Emergency Hospital, Bucharest, Romania