SEEG pairs showed no clear SOZ-specific rise before marked seizure onset

This analysis asked whether clinically marked seizure-onset-zone (SOZ) bipolar pairs show increased line length or 30–100 Hz power during the 30 seconds before clinician-marked seizure onset, compared with non-SOZ pairs in the same seizure. It used 83 seizures from 24 patients with SEEG implants in the HUP iEEG dataset (OpenNeuro ds004100, version 1.1.3), covering 7,814 bipolar pairs, including 841 SOZ pairs. The primary patient-mean SOZ-minus-non-SOZ line-length change from 105–75 seconds to 30–0 seconds before onset was +0.0103 log10 (ratio 1.024; 95% CI −0.0036 to 0.027; 12/24 positive; two-sided sign-flip p = 0.23). The placebo contrast from 105–75 to 70–40 seconds was −0.0048 (p = 0.14). The 30–100 Hz power contrast was +0.0135 (95% CI −0.051 to 0.086; p = 0.71). SOZ-only line-length change was +0.0223 (p = 0.11); the non-SOZ change was +0.0121 (p = 0.31). The primary contrast did not differ between Engel 1 and Engel 2–4 outcomes (16 and 8 patients; p = 0.98). The results do not show a SOZ-specific change in the final 30 seconds before the marked onset. The confidence interval excludes a SOZ-specific line-length increase larger than about 6%. A small general rise may be present but was not significant. The analysis was limited to a partial sample after a 45-minute retrieval limit, one center, clinical SOZ labels, 30-second windows, clinician onset marks that may include earlier changes, and simple features.

stereo-EEG · epilepsy · seizure onset · seizure-onset zone · line length · high-frequency power · intracranial EEG