Mode
Text Size
Log in / Sign up

Systematic Review Examines Cardiovascular Biomarkers for Pediatric Epilepsy Seizure AnticipationSeizures May Be Predicted Before They Start

AI-generated summary of the cited source, checked by automated accuracy review. How we work

Key Takeaway
Recognize that no devices predict seizures despite observed biomarker changes in pediatric epilepsy patients.

This systematic review synthesizes evidence regarding non-invasive physiologic and environmental biomarkers for pediatric epilepsy patients. The authors specifically examined cardiovascular biomarkers using electrocardiogram (ECG) measurement to assess seizure anticipation capabilities. Eleven observational cohort studies were included in the analysis, though the total patient sample size was not reported.

Regarding performance metrics, pre-ictal anticipation algorithm times ranged from 21.8 s to 32 min across the reviewed literature. Correlational studies observed cardiovascular biomarker changes 3.59 s to 40 min before seizures. Cardiovascular biomarkers using electrocardiogram (ECG) measurement were most commonly used within these studies.

The authors note significant limitations affecting clinical translation. Evidence in 9/11 of reviewed studies were rated as either low or very low certainty using the GRADE tool. Limitations include methodological flaws, risk of bias, inconsistent results, and indirect or sparse evidence. The review explicitly states there are no devices or systems on the market that predict seizures.

Practice relevance involves ongoing opportunities to build on findings. Further testing of cardiovascular biomarkers with other physiologic and environmental factors is needed. Larger sample size studies and a precision medicine approach to tailoring algorithms and biomarker measurements to individual patients are recommended. Clinicians should interpret these findings cautiously given the observational nature of the data.

  • Scientists find early warning signs in kids’ heart rhythms
  • Could help children with hard-to-control epilepsy
  • Not ready yet — still in early research stages

This could one day help families know a seizure is coming — before it happens.

Every parent of a child with epilepsy knows the fear. You watch your kid laugh, play, eat breakfast — and then, without warning, they freeze. Their eyes glaze over. They fall. A seizure takes over. There’s no signal. No heads-up. Just chaos.

But what if you could know it was coming?

A new review of research suggests that seizures in children might be predictable — not by brain waves alone, but by changes in the heart.

Epilepsy affects over 50 million people worldwide. In children, it’s one of the most common brain disorders. Most kids can control seizures with medicine. But about 1 in 3 still have seizures despite treatment. These are called refractory seizures — and they’re dangerous.

One of the biggest fears? SUDEP — sudden unexpected death in epilepsy. It’s rare, but real. And it often happens without warning.

Today’s devices can detect a seizure once it starts. Some wearables sense shaking or changes in movement. But none can predict one before it begins.

That’s what makes this research so important.

Families live with constant uncertainty. Will a seizure happen at school? During sleep? While swimming? A warning — even 30 seconds — could mean time to lie down, call for help, or stop driving.

The Hidden Signal

For years, scientists focused on the brain. EEGs track electrical activity. They look for storm-like spikes before a seizure.

But here’s the twist: the body may sound the alarm first.

This review found that changes in heart rhythm — measured by simple ECG patches — often shift before the brain shows any sign of a seizure.

Heart rate goes up. Rhythm becomes irregular. These changes can appear seconds — or even minutes — ahead.

It’s like your car’s check engine light coming on before the engine stalls.

Body Talks Before Brain Acts

Think of your nervous system like a city’s power grid. The brain is the main station. The heart is a major substation. They’re connected by wires — nerves — that send constant updates.

In epilepsy, the brain may be about to surge. But before it does, it sends out stress signals. These travel down the vagus nerve to the heart.

The heart responds — faster beats, uneven timing — like a warning siren.

That signal can be caught. With the right tools, we might read it like a weather forecast: “Seizure risk: high in 5 minutes.”

The team reviewed 11 studies involving children with epilepsy. All used non-invasive tools — stickers on the skin, wearable patches — nothing surgical.

They looked for patterns in heart rate, breathing, body temperature, and even sleep and weather.

The most consistent clue? Heart rhythm changes.

In some cases, algorithms spotted warning signs as early as 32 minutes before a seizure. Others saw shifts just 22 seconds prior.

One study found heart changes 40 minutes before brain activity shifted.

But results varied. Not every child showed the same pattern.

But there’s a catch.

This doesn’t mean this treatment is available yet.

While the signs are promising, most studies were small. Some had fewer than 10 kids. Others tracked just a few seizures.

And the quality of evidence? Low or very low in 9 out of 11 studies.

Why? Flaws in design. Inconsistent methods. Too few data points.

Also, every child’s body reacts differently. A rising heart rate might mean a seizure is coming — or just that the child is excited, scared, or running.

Telling the difference is hard.

Why This Could Be Different

Past attempts to predict seizures failed because they relied only on brain signals. But the brain is noisy. It’s hard to spot a coming storm in all that static.

By adding heart data, scientists get a second opinion from the body.

It’s like using both radar and barometric pressure to predict storms — not just one tool.

Experts say this multi-system approach could be more reliable. The body often reacts before the brain shows clear signs.

And since ECG tools are already safe, cheap, and wearable, they could be adapted quickly — if the science holds up.

If your child has epilepsy, especially hard-to-control seizures, this research offers hope — but not immediate help.

No device on the market today can reliably predict seizures.

Wearables that monitor heart rate (like smartwatches) are not accurate enough for medical use.

And no algorithm is approved for seizure forecasting in kids.

For now, the best step is to talk to your doctor about seizure tracking. Some families use journals or apps to spot patterns — sleep, stress, illness.

But don’t rush to buy a “prediction” device. Many sold online make big claims with no proof.

Still Early Days

The biggest limit? These studies weren’t designed to build a product. They were small, short, and inconsistent.

Most tracked kids for days or weeks — not months. Some used hospital monitors, not real-world wearables.

And none tested whether warnings actually reduced seizures or improved safety.

Also, most data came from children already in hospitals for monitoring. Their patterns may not reflect kids at home.

So while the signal is there, we’re not yet sure how to use it reliably.

Scientists now need larger, longer studies — in real homes, with real families.

The next step? Combine heart data with breathing, sleep, and even environmental triggers like weather or screen time.

Then build smart algorithms trained on individual kids — not averages.

Some research teams are already testing wearable patches that send alerts to parents’ phones.

But it could take years before a trusted, approved device is available.

Medical progress is slow. But for families living in fear of the next seizure, even a few minutes of warning could change everything.

Study Details

Study typeMeta analysis
EvidenceLevel 1
PublishedApr 2026
View Original Abstract ↓
IntroductionEpilepsy is one of the most common neurological disorders globally. While medications, surgical interventions, and dietary changes can be successful in controlling seizures, a subset of individuals experience refractory epilepsy and are at increased risk for sudden unexpected death in epilepsy (SUDEP). Efforts to provide a detection system using devices have been successful at identifying seizures once they start, but there are no devices or systems on the market that predict seizures. The purpose of this systematic review (PROSPERO ID: CRD42024444250) is to determine non-invasive physiologic and environmental biomarkers that can be used to forecast seizures in pediatric epilepsy patients.MethodsA systematic search of relevant literature was conducted in PubMed, Web of Science, CINAHL Ultimate, and EMBASE in August 2023. Articles were reviewed by two investigators in a two-step process. Data extraction occurred using two independent extractors to identify study characteristics, patient characteristics, and forecasting results. Evidence quality was evaluated by two investigators using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) tool.ResultsEleven observational cohort studies were included and cardiovascular biomarkers using electrocardiogram (ECG) measurement were most commonly used. Pre-ictal anticipation algorithm times ranged from 21.8 s to 32 min, while correlational studies observed cardiovascular biomarker changes 3.59 s to 40 min before seizures. This systematic review provides a comprehensive overview of the current evidence for seizure forecasting. However, the evidence in 9/11 of reviewed studies were rated as either low or very low certainty using the GRADE tool due to methodological flaws, risk of bias, inconsistent results, and indirect or sparse evidence.DiscussionThere are ongoing opportunities to build on our findings, including further testing of cardiovascular biomarkers with other physiologic and environmental factors, larger sample size studies, and a precision medicine approach to tailoring algorithms and biomarker measurements to individual patients.Systematic review registrationPROSPERO (ID: CRD42024444250).
Free Newsletter

Clinical research that matters. Delivered to your inbox.

Join thousands of clinicians and researchers. No spam, unsubscribe anytime.