Abstracts

Local and Network Functional Connectivity Changes After Stereotactic Thermocoagulation in Intracranial Recordings from Drug-resistant Epilepsy Patients

Abstract number : 1.178
Submission category : 3. Neurophysiology / 3E. Brain Stimulation
Year : 2022
Submission ID : 2204115
Source : www.aesnet.org
Presentation date : 12/3/2022 12:00:00 PM
Published date : Nov 22, 2022, 05:23 AM

Authors :
Sara Simula, MSc – Aix-Marseille Univ, INSERM, INS, Inst Neurosci Syst, Marseille, France; Elodie Garnier, Engr. – Aix-Marseille Univ, INSERM, INS, Inst Neurosci Syst, Marseille, France; Victor Lopez-Madrona, PhD – Aix-Marseille Univ, INSERM, INS, Inst Neurosci Syst, Marseille, France; Christian Bénar, PhD – Aix-Marseille Univ, INSERM, INS, Inst Neurosci Syst, Marseille, France; Fabrice Bartolomei, PhD, MD – Aix-Marseille Univ, INSERM, INS, Inst Neurosci Syst, Marseille, France

Rationale: About one third of epilepsy patients are resistant to treatment. Stereo-electroencephalography guided radiofrequency thermocoagulation (SEEG-guided RF-TC) aims at modifying epileptogenic networks by creating local thermocoagulative lesions in the interested areas to temporarily reduce seizure frequency in such patients. Although such lesions are hypothesized to reduce seizure frequency by functionally modifying the epileptogenic network, no account for functional connectivity (FC) changes (a measure of brain network function), nor local rhythms changes due to RF-TC has so far been presented. We evaluated, by means of SEEG recordings before and after RF-TC, whether a variation of the functional brain network properties after this procedure is related to the clinical outcome.

Methods: Interictal SEEG recordings from 34 drug-resistant epilepsy patients (17 responders and 17 non responders to RF-TC) were analyzed before and after RF-TC. To evaluate changes due to the therapy, three minutes-long segments of recording were chosen right before and 15 minutes after RF-TC. Bipolar montages were used; we performed a channel selection to include one channel per anatomical region and exclude contacts in the white matter. The FC changes were evaluated via linear (r2 index, for sub-bands) and non-linear (h2 index, for broadband 1 Hz-45 Hz) correlation coefficients between each couple of channels. We considered the network measure strength as a statistic of FC. The local changes were evaluated via the power spectral density (PSD) of each selected channel. Comparison of PSD and FC strength values before and after RF-TC were effectuated via z-score and mixed models. 

Results: In responders, thermocoagulated (TC) contacts showed a significant reduction of PSD after RF-TC, which was significantly different from non-TC contacts for all bands (delta: p = 0.002, theta: p = 0.006, alpha, beta, low gamma: p = 0.04). Nevertheless, no differences were found in PSD changes between TC and non-TC contacts in non-responders. At the network level, the strength value changed after RF-TC for the delta (p = 0.01), beta (p = 0.001) and broad (p = 0.02) bands between the responder and the non-responder groups. These differences were specific to the thermocoagulated areas only. Indeed, the strength variation of TC contacts in responders was significantly different from the one in non-responders (delta: p = 0.002, alpha: p = 0.009, beta: p = 0.0001, broad: p = 0.04).

Conclusions: Thermocoagulative lesions induce local and FC changes in brains of drug-resistant epilepsy patients lasting for at least 15 minutes. This study shows that thermocoagulation gives rise to different responses in brain network activity changes between responders and non-responders, and opens new perspectives for the investigation of longer-lasting functional connectivity changes after RF-TC.

Funding: European Research Council: Galvani project, ERC-SyG 2019, grant agreement No 855109
Neurophysiology