Surgical approaches in nonlesional neocortical epilepsy

Yield of non-invasive imaging in MRI-negative focal epilepsy

OBJECTIVE

The absence of MRI-lesion reduces considerably the probability of having an excellent outcome (International League Against Epilepsies [ILAE] class I-II) after epilepsy surgery. Surgical success in magnetic-resonance imaging (MRI)-negative cases relies therefore mainly on non-invasive techniques such as positron-emission tomography (PET), subtraction ictal/inter-ictal single-photon-emission-computed-tomography co-registered to MRI (SISCOM), electric source imaging (ESI) and morphometric MRI analysis (MAP). We were interested in identifying the optimal imaging technique or combination to achieve post-operative class I-II in patients with MRI-negative focal epilepsy.

METHODS

We identified 168 epileptic patients without MRI lesion. Thirty-three (19.6%) were diagnosed with unifocal epilepsy, underwent surgical resection and follow-up ⩾ 2 years. Sensitivity, specificity, predictive values, and diagnostic odds ratio (OR) were calculated for each technique individually and in combination (after co-registration).

RESULTS

23/33 (70%) were free of disabling seizures (75.0% with temporal and 61.5% extratemporal lobe epilepsy). None of the individual modalities presented an OR > 1.5, except ESI if only patients with interictal epileptiform discharges (IEDs) were considered (OR 3.2). On a dual combination, SISCOM with ESI presented the highest outcome (OR = 6). MAP contributed to detecting indistinguishable focal cortical dysplasia in particular in extratemporal epilepsies with a sensitivity of 75%. Concordance of PET, ESI on interictal epileptic discharges, and SISCOM was associated with the highest chance for post-operative seizure control (OR = 11).

CONCLUSION

If MRI is negative, the chances to benefit from epilepsy surgery are almost as high as in lesional epilepsy, provided that multiple established non-invasive imaging tools are rigorously applied and co-registered together.

Efficacy of delayed 18F-FDG hybrid PET/MRI for epileptic focus identification: a prospective cohort study

OBJECTIVE

We sought to investigate the contribution of delayed ¹⁸F-FDG imaging data to epileptogenic zone (EZ) identification using a hybrid positron emission tomography/magnetic resonance imaging (PET/MRI) system.

METHODS

Forty-one patients with epilepsy underwent a brain dual time point ¹⁸F-FDG PET/MRI examination. All early imaging was acquired at approximately 40 min. Late imaging was classified as short delay (150.1 ± 20.2 min) or long delay (247.8 ± 24.6 min). Visual evaluation and scoring of ¹⁸F-FDG uptake at dual time points were performed. An SUV_mean asymmetry index (AI) was calculated representing the difference in uptake between the EZ and the contralateral side. The EZ location was defined by a multidisciplinary team based on findings on video electroencephalography, ¹⁸F-FDG, and MRI. EZ location was classified as extratemporal lobe epilepsy (extra-TLE) or temporal lobe epilepsy (TLE). MRI findings were classified as positive if there were signal/structural abnormalities, or negative. AI of dual time points was compared between MRI-positive and MRI-negative, between extra-TLE and TLE, and between short delay and long delay of the late imaging time point.

RESULTS

The AI at the delayed time points was increased by a mean of 3.7 over the early time point in all patients (P < 0.01). The biggest AIs were found in the MRI-positive group. The ΔAI between two imaging points were 3.71 ± 3.50 and 4.67 ± 7.94 for MRI-positive and MRI-negative; 4.52 ± 6.70 and 2.51 ± 2.42 for extra-TLE and TLE; and 4.24 ± 6.52 and 3.46 ± 2.90 for short delay and long delay groups, respectively. There were more patients with increased AI at the delayed time with MRI-positive (95.8%, 23/24), with extra-TLE (96.8%, 30/31), and with short delay time (93.7%, 30/32). Two observers who had no knowledge of the images chose 85.4% and 82.9% of the delay-time point images as the more obvious asymmetry from all images. The kappa value between the two observers was 0.66 with good agreement.

CONCLUSION

Delayed ¹⁸F-FDG PET imaging can be used to better identify EZs with relatively greater metabolic asymmetry between the EZ and contralateral regions.

Integrating artificial intelligence with real-time intracranial EEG monitoring to automate interictal identification of seizure onset zones in focal epilepsy

OBJECTIVE

An ability to map seizure-generating brain tissue, i.e. the seizure onset zone (SOZ), without recording actual seizures could reduce the duration of invasive EEG monitoring for patients with drug-resistant epilepsy. A widely-adopted practice in the literature is to compare the incidence (events/time) of putative pathological electrophysiological biomarkers associated with epileptic brain tissue with the SOZ determined from spontaneous seizures recorded with intracranial EEG, primarily using a single biomarker. Clinical translation of the previous efforts suffers from their inability to generalize across multiple patients because of (a) the inter-patient variability and (b) the temporal variability in the epileptogenic activity.

APPROACH

Here, we report an artificial intelligence-based approach for combining multiple interictal electrophysiological biomarkers and their temporal characteristics as a way of accounting for the above barriers and show that it can reliably identify seizure onset zones in a study cohort of 82 patients who underwent evaluation for drug-resistant epilepsy.

MAIN RESULTS

Our investigation provides evidence that utilizing the complementary information provided by multiple electrophysiological biomarkers and their temporal characteristics can significantly improve the localization potential compared to previously published single-biomarker incidence-based approaches, resulting in an average area under ROC curve (AUC) value of 0.73 in a cohort of 82 patients. Our results also suggest that recording durations between 90 min and 2 h are sufficient to localize SOZs with accuracies that may prove clinically relevant.

SIGNIFICANCE

The successful validation of our approach on a large cohort of 82 patients warrants future investigation on the feasibility of utilizing intra-operative EEG monitoring and artificial intelligence to localize epileptogenic brain tissue. Broadly, our study demonstrates the use of artificial intelligence coupled with careful feature engineering in augmenting clinical decision making.

Current and Emerging Surgical Therapies for Severe Pediatric Epilepsies

The use of epilepsy surgery in various medically resistant epilepsies is well established. For patients with intractable pediatric epilepsy, the role of intracranial electrodes, resective surgery, hemispherectomy, corpus callosotomy, neurostimulation, and multiple subpial transections continues to be very effective in select cases. Newer treatment and diagnostic methods include laser thermal ablation, minimally invasive surgeries, stereo electroencephalography, electrocorticography, and other emerging techniques. This article will review the established and emerging surgical therapies for severe pediatric epilepsies, their respective indications and overall efficacy.