Coregistration of multimodal imaging is associated with favourable two-year seizure outcome after paediatric epilepsy surgery.

Abstract

Multimodal coregistration uses multiple image datasets coregistered to an anatomical reference (i.e. MRI), allowing multiple studies to be viewed together. Commonly used in intractable epilepsy evaluation and generally accepted to improve localization of the epileptogenic zone, data showing that coregistration improves outcome is lacking. We compared seizure freedom following epilepsy surgery in paediatric patients, evaluated before and after the use of coregistration protocols at our centre, to determine whether this correlated with a change in outcome. We included paediatric epilepsy surgery patients with at least one anatomical and one functional neuroimaging study as part of their presurgical evaluation. Preoperatively designated palliative procedures and repeat surgeries were excluded. Multiple pre-, peri-, and postoperative variables were compared between groups with the primary outcome of seizure freedom. In total, 115 were included with an average age of 10.63 years (0.12-20.7). All evaluations included video-EEG (VEEG) and MRI. Seven (6%) had subtraction single-photon emission CT (SPECT), 46 (40%) had positron emission tomography (PET), and 62 (54%) had both as part of their evaluation. Sixty (52%) had extratemporal epilepsy and 25 (22%) were MRI-negative. Sixty-eight (59%) had coregistration. Coregistered patients were less likely to undergo invasive EEG monitoring (p=0.045) and were more likely to have seizure freedom at one (p=0.034) and two years (p<0.001) post-operatively. A logistic regression accounting for multiple covariates supported an association between the use of coregistration and favourable post-surgical outcome. Coregistered imaging contributes to favourable postoperative seizure reduction compared to visual analysis of individual modalities. Imaging coregistration is associated with improved outcome, independent of other variables after surgery. Coregistered imaging may reduce the need for invasive EEG monitoring, likely due to improved confidence in presurgical localization. These findings support the use of multimodal coregistered imaging as part of the presurgical assessment in patients evaluated for surgical treatment of intractable epilepsy.