Age Was Not Significantly Related to HFA Power or Slow-Wave Coupling at Clinically Normal SEEG Contacts

This analysis tested whether NREM 80–300 Hz power (high-frequency activity, HFA) and coupling of HFA amplitude to 0.5–2 Hz slow-wave phase (Tort modulation index, MI) at clinically normal stereo-EEG (SEEG) contacts change with age. It used archived feature tables from OpenNeuro ds007118, ds007119, and ds007120; no new features were extracted. The unit was the patient: median values over bipolar pairs with two normal contacts were analyzed for 38 patients (406 pairs; ages 0–45 years). Under a pre-registered plan, Spearman rho for HFA versus age was −0.11 (95% CI −0.44 to 0.22; Holm-adjusted p = 0.50), and for MI versus age was 0.28 (95% CI −0.10 to 0.60; adjusted p = 0.19). Thus, the data did not show significant age relations for either primary measure. The HFA interval excludes a strong monotonic relation (|rho| > 0.45), while the MI interval does not exclude a moderate positive relation (up to rho 0.60). An exploratory, uncorrected analysis found an age association for MI residual after HFA rank adjustment (rho = 0.40, p = 0.015), which requires testing in another cohort. Limitations include low power (about 0.45 to detect rho = 0.3 at alpha 0.05), few normal pairs per patient (median 6.5), only 90 seconds of NREM sleep per patient, uncontrolled hardware and contact location, and clinical normal labels that do not establish healthy tissue.

stereo-EEG · intracranial EEG · sleep · high-frequency activity · slow-wave coupling · age · epilepsy · Tort modulation index