In 60 SEEG patients from OpenNeuro ds004100 and ds003876, interictal marker agreement with clinical SOZ was not higher after good surgical outcome

A pre-registered analysis tested whether interictal high-frequency activity (HFA, 80–150 Hz), spike rate, and line length agreed more with the clinical seizure-onset zone (SOZ) in patients with seizure-free outcomes. For each eligible patient, we measured each marker’s AUC for distinguishing SOZ from non-SOZ bipolar pairs, then compared outcomes in 60 patients from OpenNeuro ds004100 and ds003876. The expected higher AUC in successful cases was not observed: success-minus-failure differences were −0.060 for HFA (95% CI −0.150 to 0.026; permutation p=0.187), −0.007 for spike rate (95% CI −0.108 to 0.099; p=0.894), and −0.058 for line length (95% CI −0.140 to 0.024; p=0.181). No Holm-adjusted p value was below 0.5. Successful cases had a larger SOZ fraction (p=0.002), but that fraction was not correlated with HFA AUC. Clinical SOZ labels do not establish the true epileptogenic zone; clips were short, outcome definitions differed by site, and the HUP cohort included ablation. With 60 patients, the analysis can exclude only large effects and cannot rule out small effects.

stereoelectroencephalography · epilepsy · seizure-onset zone · interictal markers · high-frequency activity · spike rate · line length · surgical outcome · area under the curve · permutation test