🧠 neurology

Epileptogenicity alters intrahippocampal ripple propagation

This study demonstrates that intrahippocampal ripple propagation is proportionally greater in less-irritative non-seizure onset zone tissue compared to epileptogenic regions, suggesting that ripple signals exist on a continuous physiological-to-pathological spectrum and that propagation metrics could serve as a novel tool for quantifying hippocampal epileptogenicity.

Original authors: Chen, Y., Ye, H., Ye, L., Chen, C., Staba, R. J., Wang, S., Weiss, S. A.

Published 2026-06-16

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Chen, Y., Ye, H., Ye, L., Chen, C., Staba, R. J., Wang, S., Weiss, S. A.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The Big Picture: The Brain's "High-Frequency" Buzz

Imagine your brain is a massive, busy city. Most of the time, it hums along with a steady, low-frequency background noise (like traffic in the distance). But sometimes, specific neighborhoods generate very fast, high-pitched electrical "buzzes" called High-Frequency Oscillations (HFOs).

Scientists have known for a while that these buzzes come in two main flavors:

Ripples (RonO): Fast waves (80–250 Hz). These are like a busy but organized market. They happen in healthy brains too and are actually good for memory and learning.
Fast Ripples (FRonO): Even faster, chaotic waves (250–600 Hz). These are like a riot or a stampede. They are almost always a sign of trouble and are linked to where seizures start.

The big question this study asked is: If a part of the brain is "sick" (prone to seizures), do these buzzes travel differently than they do in a "healthy" part of the brain?

The Experiment: Mapping the Buzz

The researchers looked at 49 patients with drug-resistant epilepsy who had tiny wires (electrodes) placed deep inside their brains to record activity. They focused specifically on the hippocampus, a seahorse-shaped structure deep in the brain that handles memory.

They divided the hippocampi into three groups based on how "irritated" or "sick" they were:

Group 1 (The Seizure Zone): The part of the brain where the seizures actually started.
Group 2 (The "Highly Irritable" Zone): A part that wasn't the main seizure source, but was buzzing with abnormal electrical spikes very frequently.
Group 3 (The "Less Irritable" Zone): A part that wasn't the seizure source and only had a few abnormal spikes.

They then watched how the "ripples" (the 80–250 Hz waves) moved from one contact point on the wire to another, creating a map of how the signal traveled.

The Surprising Discovery

The researchers expected that the "sick" parts of the brain (Group 1) would be the ones where the signals traveled the furthest and fastest, acting like a spark spreading through dry grass.

Instead, they found the exact opposite for the "Ripples" (RonO):

The Healthy-Looking Zone (Group 3): The ripples here traveled the most within the hippocampus. The signal spread smoothly from one point to another, like a wave rolling through a calm, well-connected ocean.
The Sick Zones (Group 1 & 2): In the areas where seizures started or where the brain was highly irritated, the ripples didn't travel as far. They seemed to get stuck or die out quickly.

Think of it like this:
Imagine a group of people passing a secret message down a line.

In the healthy group (Group 3), everyone is listening and passing the message along smoothly to the next person. The message travels far.
In the sick groups (Group 1 & 2), the people are so chaotic or disconnected that they drop the message or can't hear the person next to them. The message stops almost immediately.

What About the "Fast Ripples"?

The study also looked at the "Fast Ripples" (the chaotic, riot-like waves).

Frequency: As expected, the "Fast Ripples" happened much more often in the seizure zones (Group 1) than in the healthy zones.
Travel: However, unlike the normal ripples, the researchers couldn't find a clear pattern in how these fast ripples traveled. They were too rare in the healthy groups to make a fair comparison.

The Main Takeaway

The paper suggests that the brain isn't just made of "healthy" and "sick" parts that act completely differently. Instead, it's more like a spectrum or a dimmer switch.

Healthy tissue allows these memory-related ripples to flow freely and travel long distances within the hippocampus.
Damaged or epileptic tissue disrupts this flow. The "roads" are broken, so the signals get stuck.

The authors conclude that we can't just look at a single electrical wave and say, "This is healthy" or "This is sick." Instead, the way the wave moves (its propagation) tells us about the health of the brain tissue underneath it. The more the ripple can travel, the more "intact" the brain's internal wiring likely is.

Summary in One Sentence

While we expected the "sick" parts of the brain to be the most active in spreading electrical signals, this study found that the healthiest parts of the hippocampus are actually the best at letting these memory-related signals travel smoothly, whereas the seizure-prone areas are too damaged to let the signals pass through.

Technical Summary: Epileptogenicity Alters Intrahippocampal Ripple Propagation

Problem Statement
High-frequency oscillations (HFOs), specifically ripples (80–250 Hz) and fast ripples (250–600 Hz), are established biomarkers for epileptogenic tissue. However, a critical challenge remains in distinguishing pathological HFOs, which indicate seizure-generating zones, from physiological HFOs, which are essential for memory consolidation and occur in healthy brain tissue. While previous studies have utilized HFO rates or source localization to identify the seizure onset zone (SOZ), the propagation properties of these signals across a spectrum of epileptogenicity are not fully understood. This study investigates whether the propagation characteristics of hippocampal HFOs differ based on the degree of local epileptogenicity, hypothesizing that the extent of epileptogenicity correlates with functional and morphological disturbances that alter signal propagation.

Methodology
The study analyzed non-REM sleep stereo-EEG (SEEG) data from 49 patients (68 hemispheres) with drug-resistant epilepsy who underwent pre-surgical evaluation at Zhejiang University.

Stratification: Hippocampi were categorized into three groups based on epileptogenicity:
1. SOZ Group: Hemispheres where the hippocampus was the seizure onset zone.
2. More-Irritative Non-SOZ Group: Non-SOZ hemispheres with high interictal epileptiform discharge (IED) rates (>6 IED/min).
3. Less-Irritative Non-SOZ Group: Non-SOZ hemispheres with low IED rates (<6 IED/min).
Signal Processing: Researchers detected "Ripples on Oscillation" (RonO, 80–250 Hz) and "Fast Ripples on Oscillation" (FRonO, 250–600 Hz) using a Hilbert-based detector, excluding events coinciding with spikes.
Network Analysis: To identify significant propagation pathways, the authors constructed empirical temporal networks with a maximum latency of 150 ms. These networks were validated against 1,000 permutation-generated surrogates to filter out chance connections.
Propagation Metrics: Significant paths were classified into three categories: (1) intrahippocampal propagation, (2) propagation from the hippocampus to other mesial-temporal lobe (MTL) structures, and (3) propagation from the hippocampus to the neocortex. The study compared the proportion of significant propagating HFOs originating from specific contacts across the three groups using mixed-effects ANOVA.

Key Results

HFO Rates: FRonO rates were significantly elevated in the SOZ group compared to non-SOZ groups ( $p < 1 \times 10^{-9}$ ). In contrast, RonO rates showed no significant differences across the three groups.
Intrahippocampal Propagation: A significant interaction was found between the epileptogenicity group and propagation category. Specifically, the proportion of intrahippocampal RonO propagation was highest in the less-irritative non-SOZ group. This proportion was significantly lower in the more-irritative non-SOZ group ( $p < 0.05$ ) and the SOZ group ( $p < 0.001$ ).
Long-Range Propagation: No significant differences were observed in the proportions of propagation from the hippocampus to other MTL structures or to the neocortex across the three groups.
Fast Ripples (FRonO): Due to the scarcity of statistically significant FRonO propagation paths (particularly in non-SOZ groups), no significant differences in FRonO propagation proportions were detected across groups.
Latency Sensitivity: The significant finding regarding intrahippocampal RonO propagation was observed with a 150 ms maximum latency but disappeared when a shorter 32.5 ms latency was used, suggesting the effect is driven by longer-latency, likely polysynaptic pathways.

Significance and Claims
The authors conclude that intrahippocampal RonO propagation is proportionally greater in less-irritative (likely more physiological) hippocampal tissue compared to epileptogenic tissue. This finding challenges the binary view of RonO as strictly physiological or pathological. Instead, the authors propose that RonO signals and their underlying substrates exist on a continuous physiological-to-pathological spectrum, where the degree of local structural and functional injury dictates propagation properties.

The study suggests that while HFO rates (specifically FRonO) remain a strong marker for the SOZ, the propagation proportion of RonO may serve as a novel metric to quantify hippocampal epileptogenicity. The reduction in propagation in the SOZ and more-irritative tissue is attributed to hippocampal injury and network disconnection, which disrupt the micro-architecture and homeostatic gating mechanisms necessary for ripple propagation. The authors emphasize that distinguishing physiological from pathological RonO based solely on signal features remains difficult, and their work supports a framework where these categories are not distinct but rather points on a continuum.