Abstracts

Rationale: Generalized tonic-clonic seizures (GTCS) are accompanied by cardiovascular and respiratory sequelae that threaten survival, with GTCS frequency a major risk factor for sudden unexpected death in epilepsy (SUDEP), a leading cause of untimely death in the condition. Widespread cortical thinning occurs in frontal cortex, while increased cortical thickness consistent with inflammatory processes appears in autonomic regulatory areas, including the cingulate, somatosensory, and insular cortices. However, the relationship between GTCS frequency and cortical thickness changes are unknown. Methods: We assessed regional cortical thickness using high-resolution T1-weighted images (3.0-Tesla) collected from 41 patients with GTCS (Mean age±SD: 36.5±12.8 years, 21 male,) at UCLA, Case Western Reserve University, and University College London, and 530 healthy controls (37.1±12.6 years; 220 male) from UCLA and two open access databases (OASIS and IXI). We used MATLAB-based SPM12 software to preprocess high-resolution T1-weighted scans, including intensity bias correction, and FreeSurfer 5.3 software to measure regional cortical thickness. Regions showing altered thickness vs controls were selected for correlation analysis between GTCS frequency and regional mean cortical thickness, scaled for total intracranial volume. Results: Seizure counts for the year preceding hospital admission ranged from 0-78. Collectively, all 41 patients showed mild negative correlations between cortical thickness and GTCS frequency bilaterally across multiple frontal regions, with the left medial orbitofrontal showing the highest correlation (r = -0.36). Dividing patients into low, medium, and high frequency seizure groups revealed that the medium frequency group (3-14 GTCS/yr) had strong positive correlations between seizure frequency and cortical thickness across numerous regions. These sites included the orbitofrontal and insular regions ranging from r = 0.42-0.55, midfrontal regions (r = 0.61-0.67), and the left cingulate, which had the strongest correlation (r=0.71). Conversely, the high frequency group (18-78 GTCS/yr) showed strong negative correlations across numerous areas, including the left insula (r=-0.57), medial orbitofrontal (r=-0.46), and right caudal midfrontal cortex (r=-0.51). Conclusions: In patients with 3-14 GTCS/year, higher seizure frequency correlates with thicker cortical values, both in areas of increased or reduced thickness relative to controls, consistent with acute changes due to inflammatory processes accompanying ictal events. In patients with >18 GTCS/yr, higher seizure frequency is associated with thinner cortex, presumably due to chronic changes overriding any inflammatory effects. Repeated overactivation from GTC seizure processes may also contribute to cortical thickness changes. These findings further characterize the nature of neocortical changes accompanying GTCS, and provide insights into how seizure frequency may affect SUDEP risk. Funding: This work was supported by U01 NS090407 and the Amgen Foundation.