Abstracts

Rationale:
Differences in pre-operative network characteristics may contribute to varying rates of seizure-freedom following epilepsy surgery. Though previous literature has identified correlations between network characteristics and rates of seizure freedom and recurrence, it remains unclear how this information may impact surgical decision-making. We sought to identify a relationship between seizure freedom, pre-operative network characteristics, and surgical approach to establish a role for network mapping in epilepsy surgery planning.
Method:
The hypothesized irritative zone and seizure network were mapped for 22 patients with medically refractory temporal lobe epilepsy treated surgically with either trans-cortical selective amygdalohippocampectomy (SAH) or anterior temporal lobectomy with minimal hippocampus resection (ATLsH). The positively correlated seizure network (PSN) was mapped pre-operatively and median distribution of the network was calculated. Post-operatively, the PSN was mapped again at three-month follow-up to calculate percent PSN resection. Patients were tracked for either seizure freedom (SF) or recurrence (SR).
Results:
For the entire cohort, surgery resulted in a significant reduction in PSN connectivity (p< .001) There was no significant difference in pre-operative PSN connectivity between SF and SR patients (p=.81). Post-operatively, SF patients had lower PSN connectivity that approached significance (p=.051). Subgroup analysis in SF patients to compare network characteristics between ATLsH and SAH patients identified smaller median spatial distribution of the PSN in ATLsH patients (p< 0.001). Additionally, ATLsH patients exhibited a greater percent resection of their PSN post-operatively (p=.006).
Conclusion:
Surgery resulted in decreased PSN connectivity post-operatively. Interestingly, in a subgroup of SF patients, patients receiving ATLsH exhibited smaller pre-operative PSN distribution as well as greater post-op percent disconnection when compared to SAH patients. Previously, we have shown decreased spread of the PSN and increased PSN disconnection to be correlated with improved outcomes. Decreased distribution of the PSN to the temporal lobes may predict improved surgical outcome with ATLsH versus SAH secondary to more extensive resection resulting in increased PSN disconnection. Selection bias is certainly a limitation of our study, and an evaluation of how patient pre-operative network and neuropsychological characteristics may contribute to a surgeon’s decision to pursue ATLsH versus SAH may be useful. In future studies, we hope to identify predictors of network disconnection for each surgery type and correlate this with post-operative outcome.
Funding:
:No funding resources to report.