Authors :
Presenting Author: Mary Sagel, MS – Cleveland Clinic Foundation
Nicholas Swingle, MD – Cleveland Clinic
Juan C. Bulacio, MD – Cleveland Clinic, Cleveland, United States
Balu Krishnan, PhD – Cleveland Clinic
Vineet Punia, MD – Cleveland Clinic, Cleveland, OH, USA
Rationale: Hippocampal rhythms play an important role in the encoding and retrieval of episodic memory, a component of memory that shows substantial age-related decline. The purpose of this study is to examine the relationship between aging and hippocampal spectral features derived from intracranial SEEG while accounting for structural and clinical influences.
Methods: Resting-state awake and asleep SEEG recordings from adults with hippocampal sampling (2016–2020) were analyzed. Electrode placement within the hippocampal head was verified by epileptologists. Artifact-free 5-minute segments were used to compute power spectral density (PSD), from which aperiodic components were removed using the fitting oscillations and one over frequency (FOOOF) model to derive band-limited flat power and aperiodic parameters (offset and exponent). Features were averaged across time-of-day states. In the FOOOF model, the aperiodic offset reflects the power spectrum’s vertical shift, and the exponent captures its decay rate, independent of oscillatory peaks (Fig. 1a). Removing the aperiodic component yields a flattened spectrum used to calculate band-limited flat power (Fig. 1b). Linear mixed-effects models assessed associations with age, hippocampal volume, epilepsy duration, pathology, and time of day, with False Discovery Rate (FDR) correction applied to all results.
Results:
A total of 36 patients with epilepsy and 44 hippocampi were analyzed (Table 1). Figure 1c displays the results of the linear mixed effects model. The pathological hippocampus was associated with increased theta, alpha, beta, and total flat power, and decreased delta flat power and elevated aperiodic exponent, reflecting enhanced mid-frequency activity and steeper spectral decay. Greater hippocampal volume was linked to increased aperiodic offset (FDR p < 0.0001), suggestive of elevated broadband power. Longer epilepsy duration was associated with increased alpha power, decreased gamma power, and elevated aperiodic exponent. Although age alone did not yield significant effects after correction, trends included decreased delta power and exponent. A significant interaction between age and volume revealed that older individuals with larger hippocampi exhibited reduced aperiodic offset and exponent (FDR p = 0.013), indicating spectral flattening in this subgroup. SEEG signal during awake state was associated with significantly lower aperiodic offset and exponent consistent with a flattened spectral profile, which aligns with known neurophysiologic differences between sleep and wakefulness, offering a normative reference.