Neither exploratory SEEG sex effect replicated in an independent 48-patient interictal cohort

A pre-registered test in 48 independent interictal SEEG patients (22 women, 26 men; ages 16–59) did not replicate two exploratory sex effects reported in 256 RAM epilepsy patients: higher female 4–13 Hz peak frequency (+0.38 Hz) and flatter 2–40 Hz aperiodic slope (+0.146). In the independent cohort, adjusted female-minus-male differences were −0.15 Hz (95% CI −1.24 to +0.87; one-sided permutation p = 0.61) for peak frequency and −0.154 (95% CI −0.39 to +0.14; p = 0.87) for slope; Holm-adjusted p = 1.0 for both. The test had about 0.25 power for d = 0.30 and therefore does not exclude an effect of that size for peak frequency. Fixed-effect pooled estimates remained positive for peak frequency (+0.33 Hz, 95% CI +0.02 to +0.64, p = 0.036) and uncertain for slope (+0.098, 95% CI −0.01 to +0.20, p = 0.071); this pooling ignores cohort heterogeneity. The independent cohort used rest-like interictal segments rather than task segments and had different sites and montages, no MNI location adjustment, and two adolescents as specified in the plan. The negative overall slope estimate was driven by the 11-patient ds003876 subgroup and was fragile: in the 37 HUP patients alone, the slope difference was +0.038 (95% CI −0.25 to +0.31).

SEEG · intracranial EEG · sex differences · epilepsy · spectral analysis · replication · aperiodic slope · peak frequency