Will My Baby Have Epilepsy Later? What New Research Says About Seizures After a Difficult Birth

How two 2026 studies, one from hospitals in Rome and Modena and one from Zurich, are helping doctors tell which babies need closer follow-up for epilepsy

Some babies have seizures in their first days of life, often after a birth in which their brain was short of oxygen. Two studies published in 2026 looked at which of these babies go on to have epilepsy, meaning seizures that keep coming back later in childhood. Most babies did not develop epilepsy, and doctors can already spot many of those at higher risk using brain scans and brain-wave tests, while a new computer-based brain-wave measure shows early promise but still needs checking.

Why families ask this question

When a newborn has seizures, the first days are about getting through the emergency. But once the baby is ready to go home, a new worry often appears: will the seizures come back? Parents want to know what to watch for, how long to keep seeing specialists, and whether their child's future will include epilepsy. Doctors want to give honest answers without causing needless fear. Being able to tell which babies are at higher risk, and which are at low risk, helps everyone plan sensibly.

How doctors have tried to answer it in the past

One common reason for newborn seizures is hypoxic–ischaemic encephalopathy, usually shortened to HIE. This means a brain injury caused by too little oxygen and blood flow around the time of birth. Back in 1976, two doctors named Sarnat described three stages of how unwell these babies look, from mild to severe, and noticed that babies whose brain-wave recordings did not return to normal tended to do worse [1]. A brain-wave recording is called an EEG (electroencephalogram): small sensors placed on the scalp pick up the brain's electrical activity. Later, researchers showed that an EEG taken in the first 6 hours of life was very good at predicting how a baby would be doing at age 2 [2]. MRI scans (magnetic resonance imaging, a detailed picture of the brain made with magnets rather than X-rays) added another clue. Injury to two deep brain areas, the basal ganglia and the thalamus, turned out to be the typical mark of a sudden lack of oxygen at birth [3], and an early scan came to be regarded as the best single guide to how a child would develop [4].

Since then, a treatment called therapeutic hypothermia, or "cooling", has become standard care. The baby's body is cooled to 33.5 °C for 72 hours, starting within 6 hours of birth, to help protect the brain [5]. Cooling also changed how EEGs should be read, because in cooled babies the recording became most informative only after the first day [6]. Studies of epilepsy itself were smaller and scattered. In one large American project following babies who had seizures caused by an illness or injury around birth, 37 of 282 children (about 1 in 8) developed epilepsy by age 2 [7]. In a study from Modena, Italy, epilepsy developed in 2 of 23 babies who were cooled compared with 9 of 26 who were not [8]. And in a group of cooled children from the Bristol region, epilepsy became more common as they grew older, rising from 8 children (about 6 in every 100) at age 2 to 14 of 103 (about 13 in every 100) by ages 4 to 8 [9]. That last finding raised an important worry: if some epilepsy starts after age 2, ending check-ups at that age might miss it.

Two studies, two different approaches

The first study, led by Dr Cicala, looked back at the records of 159 babies treated with cooling for moderate or severe HIE at Gemelli Hospital in Rome and the University Hospital of Modena, Italy, between 2013 and 2019 [5]. Every child was followed for at least 2 years, and the children who developed epilepsy were followed for a median of 4 years, meaning half of them were followed for longer than that. The researchers asked a practical question: which information that doctors already collect, such as how ill the baby was, whether the baby had seizures, early EEGs and MRI scans, best predicts later epilepsy?

The second study, led by Dr Karatza at the University Children's Hospital and the University Hospital in Zurich, Switzerland, took a more technical approach [10]. The team re-analysed routine EEGs from 47 newborns who had seizures for any reason, including HIE, injuries involving the brain's blood vessels, and genetic conditions, and from 8 healthy newborns. A computer program counted tiny, very fast bursts of brain activity, called high-frequency activity or HFA, during sleep. Each burst lasted, on average, less than a tenth of a second. The idea grew from earlier research suggesting that similar fast bursts can signal a brain that is prone to seizures in older children [11], and from the same Zurich team's smaller first study in newborns [12].

What the cooling study found

Out of 159 cooled babies, 15 developed epilepsy, or about 1 in every 11 [5]. Put the other way round, 144 children, more than 9 out of 10, did not. For 9 of the 15, epilepsy started before their second birthday, often in the first 6 months. But for the other 6, it started later, between about 2 and 4 years of age. Three children had a type of epilepsy called infantile spasms, which appeared at 5 to 6 months of age; in one of the three, the spasms were found only because a routine screening EEG was done. Most of the children's epilepsy was the kind that starts in one part of the brain, and the researchers describe it as generally well controlled, although 10 of the 15 children still had at least one seizure in the past year, usually about once a year.

The children who developed epilepsy usually stood out early. Almost all of them (10 of the 11 whose illness grade was recorded) had the severe form of HIE, and 14 of the 15 had seizures as newborns. Every one who had an EEG at 6 hours of life showed a markedly abnormal pattern. And 9 of the 15 had the most extensive pattern of injury on MRI, involving the deep brain areas, compared with only 9 of the 143 children without epilepsy who had a recorded scan. By contrast, things like the baby's Apgar scores (the quick check of a newborn's condition in the first minutes after birth) or the blood tests taken from the umbilical cord at birth did not predict epilepsy. Children whose epilepsy began earlier were also more likely to have cerebral palsy, a condition affecting movement, and slower development.

What the brain-wave study found

In the Zurich study, newborns with seizures had more of the fast bursts than healthy newborns did [10]. Of the 47 babies with seizures, 9 died as newborns and 3 could not be followed up; the remaining 35 were followed until their first birthday. Among these, the 11 who developed epilepsy had, on average, more than twice as many fast bursts on their first EEG as the 16 who developed normally.

This is an interesting early finding, but it comes with important limits. Nine of the 11 babies who developed epilepsy had a genetic cause for their seizures, and only one had HIE, so the result does not yet tell us much about babies with oxygen-related brain injury. The study was small, the cut-off point used to separate higher- and lower-risk babies was worked out in the same group it was tested on, and children were followed for only 1 year. The authors themselves say the test needs to be confirmed in larger, separate groups of babies before it can be used in care.

What this means for your baby and family

If your baby had HIE and was cooled, these findings are largely reassuring: in this group, most children did not develop epilepsy. The doctors' existing tools, including how ill your baby was, the EEGs and the MRI scan, already help to identify the smaller group who need closer watching. A similar study of cooled babies in Buenos Aires, Argentina, also found that deep brain injury on MRI and a brain-wave pattern that stayed very quiet at 72 hours were linked to epilepsy [13].

For babies with these higher-risk signs, the Italian researchers suggest three things [5]. First, an early EEG check in the first months of life, since infantile spasms can be subtle. Second, teaching parents what seizures and spasms look like, so they can seek help early. Third, continuing check-ups beyond age 2, because some epilepsy starts later. If you are unsure whether your baby is in a higher-risk group, it is reasonable to ask your baby's doctors which of these signs applied and what follow-up plan they recommend.

These numbers describe groups of children, not any single child. Both studies looked back at existing records, both were fairly small, and babies who died were not included in the Italian study. Your own medical team knows your baby's details and is best placed to explain what they mean. Families wanting to learn more about long-term outcomes after cooling may find the In[Neo]Sight article on the school-age follow-up of a major cooling trial helpful.

What researchers are working on next

Researchers want to follow cooled children well beyond age 2, to understand late-starting epilepsy better. They also hope to combine the clues they already have, such as early EEGs and MRI scans, into simple risk guides that can be tested in many hospitals. For the fast-burst measure, the key next step is to see whether it still predicts epilepsy in babies with HIE or stroke once genetic conditions are set aside, and whether a fixed cut-off works in new groups of babies.

The bottom line

Epilepsy after a difficult birth treated with cooling is uncommon, affecting about 1 in 11 children in the Italian study. The babies most at risk can usually be recognised early from how ill they were, their brain-wave recordings and their MRI scans, and for them follow-up should continue past the second birthday. A new computer-based way of reading brain waves may one day add to these tools, but it still needs to prove itself in larger studies.

References

  1. Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Arch Neurol. 1976;33(10):696-705. doi:10.1001/archneur.1976.00500100030012 ↩
  2. Murray DM, Boylan GB, Ryan CA, Connolly S. Early EEG findings in hypoxic-ischemic encephalopathy predict outcomes at 2 years. Pediatrics. 2009;124(3):e459-467. doi:10.1542/peds.2008-2190 ↩
  3. Okereafor A, Allsop J, Counsell SJ, et al. Patterns of brain injury in neonates exposed to perinatal sentinel events. Pediatrics. 2008;121(5):906-914. doi:10.1542/peds.2007-0770 ↩
  4. Martinez-Biarge M, Diez-Sebastian J, Rutherford MA, Cowan FM. Outcomes after central grey matter injury in term perinatal hypoxic-ischaemic encephalopathy. Early Hum Dev. 2010;86(11):675-682. doi:10.1016/j.earlhumdev.2010.08.013 ↩
  5. Cicala G, Ricca O, Picilli M, et al. Epilepsy and Neurodevelopment Outcomes 24 Months after Neonatal Hypoxic-Ischemic Encephalopathy and Predictive Factors of Post-neonatal Epilepsy. Neuropediatrics. 2026;57(2):107-115. doi:10.1055/a-2781-6844 ↩
  6. Nash KB, Bonifacio SL, Glass HC, et al. Video-EEG monitoring in newborns with hypoxic-ischemic encephalopathy treated with hypothermia. Neurology. 2011;76(6):556-562. doi:10.1212/WNL.0b013e31820af91a ↩
  7. Shellhaas RA, Wusthoff CJ, Numis AL, et al. Early-life epilepsy after acute symptomatic neonatal seizures: A prospective multicenter study. Epilepsia. 2021;62(8):1871-1882. doi:10.1111/epi.16978 ↩
  8. Lugli L, Balestri E, Berardi A, et al. Brain cooling reduces the risk of postneonatal epilepsy in newborns affected by moderate to severe hypoxic-ischemic encephalopathy. Minerva Pediatr (Torino). 2021;73(2):150-158. doi:10.23736/S2724-5276.18.05224-6 ↩
  9. Liu X, Jary S, Cowan F, Thoresen M. Reduced infancy and childhood epilepsy following hypothermia-treated neonatal encephalopathy. Epilepsia. 2017;58(11):1902-1911. doi:10.1111/epi.13914 ↩
  10. Karatza P, Cserpan D, Lo Biundo SP, et al. Scalp high-frequency activity differentiates neonates with seizures from healthy neonates and indicates postneonatal epilepsy risk. Epilepsia. 2026;67(3):1303-1316. doi:10.1111/epi.70025 ↩
  11. Klotz KA, Sag Y, Schönberger J, Jacobs J. Scalp Ripples Can Predict Development of Epilepsy After First Unprovoked Seizure in Childhood. Ann Neurol. 2021;89(1):134-142. doi:10.1002/ana.25939 ↩
  12. Cserpan D, Guidi G, Alessandri B, et al. Scalp high-frequency oscillations differentiate neonates with seizures from healthy neonates. Epilepsia Open. 2023;8(4):1491-1502. doi:10.1002/epi4.12827 ↩
  13. Calligaris SD, Jozefkowicz M, Cáceres MJM, et al. Early predictors for later epilepsy after therapeutic hypothermia for neonatal hypoxic-ischemic encephalopathy. Epilepsy Behav. 2026;182:111130. doi:10.1016/j.yebeh.2026.111130 ↩