Archived NREM SEEG features did not reliably distinguish resected from unresected non-SOZ pairs in 13 patients

We tested whether resected non-seizure-onset-zone (non-SOZ) bipolar SEEG pairs had higher archived interictal NREM features than non-resected non-SOZ pairs. In the primary analysis, 13 patients met the requirement of at least two pairs in each group. Mean patient-level AUCs were 0.578 for 80–300 Hz power, 0.574 for 0.5–2 Hz phase-to-amplitude coupling, and 0.558 for delta power; each had Holm-adjusted p = 0.91. Restricting to pairs with normal status at both contacts gave similar results in 12 patients (mean AUC 0.550, 0.579, and 0.563; Holm-adjusted p = 0.66). Each feature had AUC above 0.5 in about half of patients. The data do not show that resection beyond the clinical SOZ follows these features. The 95% confidence intervals do not exclude a moderate effect (upper bounds about 0.67–0.72); this is a low-power null result. Features and labels were taken from an archived publication and were not recomputed. Resection is a clinical decision with multiple inputs, and pairs on one shaft are not independent.

SEEG · stereo-EEG · epilepsy surgery · seizure onset zone · NREM sleep · high-frequency activity · phase-amplitude coupling · delta power · AUC · resection margin