The Impact of Lesional Factors on Seizure Outcome Following Lesionectomy as a Surgical Management of Extratemporal Epilepsy, Case Series from a Developing Country and Review of Literature.

BACKGROUNDExtratemporal lesional epilepsy can be managed with various surgical techniques such as simple lesionectomy or more extensive resections, all of which aim at targeting the epileptogenic zone which is the key for achieving a favorable outcome. This study aimed at evaluating the effectiveness of lesionectomy in the treatment of extra-temporal epilepsy associated with a lesion on radiological imaging, and to show the effect of lesional factors on seizure outcome including the anatomical location, the relation to cerebral parenchyma, the extent of surgical excision and the histopathological nature of the lesion.METHODSA prospective study on 20 patients presenting with focal epilepsy, or focal epilepsy with secondary generalized seizures with evidence of focal lesion in an extratemporal location on MRI. Lesionectomy was done and we used the Engel’s classification for seizure outcome.RESULTSLesions were mostly tumors (85 %). The frontal lobe is the most frequent locations (60 %). Low-grade glioma represented 35 % while meningioma represented 45 % of all lesions (both intra-axial and extra-axial). Four patients were lost during follow up (mean 23.31 months). For the remaining 16 patients, 13 cases were tumors (81.25%). Lesionectomy achieved seizure freedom in 68.75 %.CONCLUSIONSIn a country with limited resources, lesionectomy is a valid technique for epilepsy surgery as long as the radiological data and the seizure semiology are concordant. Total lesionectomy provides good seizure control when the clinical and radiological data are concordant with seizure semiology, in particular with tumor-related epilepsy. A study comparing postoperative seizure outcome between intra-axial and extra-axial lesions on a larger scale and with a longer follow up period is recommended.

Machine Learning for Localizing Epileptogenic-Zone in the Temporal Lobe: Quantifying the Value of Multimodal Clinical-Semiology and Imaging Concordance

Background: Epilepsy affects 50 million people worldwide and a third are refractory to medication. If a discrete cerebral focus or network can be identified, neurosurgical resection can be curative. Most excisions are in the temporal-lobe, and are more likely to result in seizure-freedom than extra-temporal resections. However, less than half of patients undergoing surgery become entirely seizure-free. Localizing the epileptogenic-zone and individualized outcome predictions are difficult, requiring detailed evaluations at specialist centers.Methods: We used bespoke natural language processing to text-mine 3,800 electronic health records, from 309 epilepsy surgery patients, evaluated over a decade, of whom 126 remained entirely seizure-free. We investigated the diagnostic performances of machine learning models using set-of-semiology (SoS) with and without hippocampal sclerosis (HS) on MRI as features, using STARD criteria.Findings: Support Vector Classifiers (SVC) and Gradient Boosted (GB) decision trees were the best performing algorithms for temporal-lobe epileptogenic zone localization (cross-validated Matthews correlation coefficient (MCC) SVC 0.73 ± 0.25, balanced accuracy 0.81 ± 0.14, AUC 0.95 ± 0.05). Models that only used seizure semiology were not always better than internal benchmarks. The combination of multimodal features, however, enhanced performance metrics including MCC and normalized mutual information (NMI) compared to either alone (p < 0.0001). This combination of semiology and HS on MRI increased both cross-validated MCC and NMI by over 25% (NMI, SVC SoS: 0.35 ± 0.28 vs. SVC SoS+HS: 0.61 ± 0.27).Interpretation: Machine learning models using only the set of seizure semiology (SoS) cannot unequivocally perform better than benchmarks in temporal epileptogenic-zone localization. However, the combination of SoS with an imaging feature (HS) enhance epileptogenic lobe localization. We quantified this added NMI value to be 25% in absolute terms. Despite good performance in localization, no model was able to predict seizure-freedom better than benchmarks. The methods used are widely applicable, and the performance enhancements by combining other clinical, imaging and neurophysiological features could be similarly quantified. Multicenter studies are required to confirm generalizability.Funding: Wellcome/EPSRC Center for Interventional and Surgical Sciences (WEISS) (203145Z/16/Z).