Abstracts

Rationale:
Selected children with medically refractory epileptic spasms have surgically remediable epilepsy. Seizure freedom rates ranging from 61% to 83% has been reported in three prior large series of children with epileptic spasms who underwent epilepsy surgery.  Invasive intracranial EEG monitoring has been used in 30-65% of children in these surgical series. We report our experience of selecting surgical candidates with epileptic spasms without invasive EEG monitoring.
Method:
Medical records of all consecutive children aged 5 years or under who had epilepsy surgery for intractable epileptic spasms at Cleveland Clinic between 2000 and 2018 were reviewed. Patients were classified as seizure free if they achieved an Engel class I at last follow-up. Predictors of prognosis of seizure outcome were analyzed.
Results:
Seventy patients with medically intractable epileptic spasms underwent surgical resections during the study period. All children had a minimum of 6 months of follow-up; mean follow-up was 4.7 years. Mean age at seizure onset was 6.8 months (+9.31) Mean age at surgery was 22.7 months (SD = 18.94). All children had active epileptic spasms, recorded on pre-operative video EEG. All 70 patients had an epileptogenic lesion identified on MRI brain; 17 patients (24%) had bilateral abnormalities. Etiologies included malformations of cortical development (58%), infarct/encephalomalacia (39%), and tumor (3%). Brain FDG-PET was performed in 54 (77%) of the patients. Surgical decisions were determined primarily by findings from video EEG, brain MRI, PET, and further modified based on pre-existing neurological deficits. None of the patients had intracranial EEG. Surgical procedures included hemispherectomy (44%), multilobar resections (23%), and lobectomy/lesionectomy (33%); of the lobar surgeries, 48% were temporal. At six months after surgery, 73% (51/70) had complete seizure freedom (Engel class IA). At a mean follow up of 4.7 years, 60% (42/70) were seizure free since surgery. Of 28 patients with recurrence, 17 patients (60%) reported improvement. Of 39 patients (56%) who had hypsarrhythmia on presurgical EEG, 35 (90%) had resolution of hypsarrhythmia on 6-month post-operative EEG. On multivariate analysis of prognostic predictors, shorter epilepsy duration (p=0.05) and smaller epileptogenic lesions (p=0.02) predicted favorable seizure outcome at 6 months after surgery. For long term seizure freedom, patients with unilateral abnormalities on MRI (p=0.001) and lobar or hemispheric epileptogenic lesions (p=0.02) (compared to multilobar) had higher chances to remain seizure free.
Conclusion:
Children with medically refractory epileptic spasms secondary to structural brain lesions can be selected for epilepsy surgery without the need for invasive monitoring. Seizure freedom rate comparable to prior published studies support this conclusion. A non-invasive approach allows for early surgery and potentially may improve outcome. Early surgery and shorter epilepsy duration was predictive of seizure freedom at 6-month follow-up. Absence of bilateral abnormalities on MRI, and lobar or hemispheric lesions were associated with favorable long term outcome.
Funding:
:Neurological Institute, institutional grant for statistical help