This study investigated whether hippocampal gamma oscillations recorded via stereo-electroencephalography (SEEG) can serve as sensitive neural biomarkers of memory impairment in mesial temporal lobe epilepsy (MTLE).

Gamma oscillations (30–100 Hz) within the hippocampus are critical for memory encoding, retrieval, and network synchronization. Disruptions in these rhythms have been linked to cognitive deficits in epilepsy and other neurological disorders, yet their predictive value for cognitive outcomes in MTLE remains unclear.

SEEG data from 34 MTLE patients with bilateral hippocampal coverage were retrospectively analyzed. Gamma-band activity during wakefulness and sleep was quantified using CURRY 9. Hippocampal gamma power was categorized as bilaterally preserved, decreased (>50% reduction), unilaterally absent, or bilaterally absent. Presurgical neuropsychological data from the Hopkins Verbal Learning Test–Revised (HVLT-R) and the Brief Visuospatial Memory Test–Revised (BVMT-R) were standardized (Normal, Average, High Average, Low Average, Borderline, Below Average, and Exceptionally Low) and aligned with individual gamma profiles to assess relationships between electrophysiological integrity and memory performance.

Patients with bilaterally preserved gamma (n=15) exhibited average to above-average memory scores. Bilateral gamma reduction (n=2) corresponded with borderline to low-average performance, and bilateral absence (n=3) with severe global deficits. Lateralized gamma loss predicted domain-specific impairment: left-sided loss (n=4) correlated with verbal deficits, while right-sided loss (n=6) aligned with visuospatial deficits. One patient with preserved gamma but impaired encoding showed frequent interictal discharges, indicating additional disruption mechanisms.

Our findings reveal a clear dose–response relationship between hippocampal gamma oscillations and memory function in MTLE. Robust gamma activity aligns with preserved cognitive performance, while escalating disruptions—especially bilateral loss—predict increasingly severe impairments. These electrophysiological patterns mirror material-specific, lateralized memory models and underscore the promise of hippocampal gamma as a biomarker. Incorporating gamma profiling into preoperative evaluation could improve individualized prediction of cognitive outcomes following mesial temporal resection or ablation, thereby guiding surgical decision-making to better preserve memory.