Morphological Remodeling of Scalp High-Frequency Oscillations Across BASED-Stratified Groups in Infantile Epileptic Spasms Syndrome

Background/Objectives: High-frequency oscillations (HFOs)—transient electroencephalography (EEG) activity above 80 Hz—are emerging biomarkers of infantile epileptic spasms syndrome (IESS). However, the relationship between their multidimensional characteristics and clinical severity remains poorly understood. This study aimed to clarify the association of scalp HFO morphology with severity across levels defined by the Burden of Amplitudes and Epileptiform Discharges (BASED) score, an interictal EEG grading scale for IESS. Methods: We enrolled 53 children with epilepsy (30 with IESS and 23 non-IESS controls) and quantified HFO frequency, duration, amplitude, and cycle count from automatically detected scalp HFOs during interictal EEG. Results: Patient-level median analyses demonstrated significant monotonic associations with BASED severity: HFO frequency decreased (Spearman ρ = −0.46, p = 0.001) and duration increased (ρ = 0.32, p = 0.026). Event-level mixed-effects models confirmed these findings, showing that frequency decreased by 10.6 Hz per BASED step (p < 0.001) and duration increased 1.18-fold per step (p = 0.011), whereas amplitude and cycle count showed no consistent associations. Phenotype-level enrichment analysis revealed that specific morphological signatures significantly distinguished severity levels, with severe IESS showing a marked reduction in the high-frequency/high-amplitude/short-duration class (OR = 0.49, 95% CI 0.33–0.73) and a shift toward low-frequency/long-duration phenotypes. Conclusions: Scalp HFOs showed lower frequencies and longer durations in higher BASED-stratified groups, suggesting that HFO morphology may provide quantitative information complementary to visual EEG assessment in IESS. These findings support the potential utility of HFO phenotypic stratification for objective evaluation and longitudinal monitoring of disease burden.