Seizing the chance to improve babies' health

A research group at Cork University Maternity Hospital is developing and testing an EEG-based system to monitor high-risk babies…


A research group at Cork University Maternity Hospital is developing and testing an EEG-based system to monitor high-risk babies in real time and alert medical staff if there’s a problem

HOW CAN you tell by a baby’s outward appearance whether he or she is having a brain seizure? Often you can’t – which is why researchers are developing a new way to monitor high-risk babies for inward signs that help is needed.

“Up to three out of every 1,000 babies born at full-term experience brain seizures, but very premature babies may be more vulnerable, with some estimates of seizures happening in up to 50 per 1,000 births in this particular group,” explains clinical scientist Dr Geraldine Boylan, who heads the neonatal research group at Cork University Maternity Hospital.

“It’s something a lot of people don’t realise until it happens to them or a close relative,” she says. “Babies can get into distress, and usually that is because there’s a lack of oxygen or a lack of bloodflow to the baby’s brain. And if a baby has got a problem with its brain, one of the commonest things is that the baby will have a seizure. It’s like a reaction to something wrong in the brain.”

READ MORE

However, many seizures are subtle and are missed, she adds. “We have only realised recently that we don’t see babies having seizures, so we are missing a lot of them.”

One way to determine whether a high-risk baby is having a brain seizure is to monitor their brain non-invasively for electrical activity using an electro-encephalograph (EEG). But it’s not routinely done in neonatal intensive care, explains Boylan, and not spotting and treating seizures in vulnerable babies could allow longer term problems to develop.

“If a baby has developed brain injury around the time of birth, this could be a baby who could end up maybe in lifelong care, and if you have not recognised it early, you may not have treated it and the injury could get worse,” she says.

To help identify when a baby is having a seizure, Boylan’s team is developing and testing an EEG-based system to monitor high-risk babies in real time and alert medical staff if there’s a problem.

“If a baby is in difficulty, the problem usually manifests itself fairly soon, usually within the first 24 hours, so we monitor the babies for a few days after birth,” she explains.

The researchers monitored high-risk babies at the hospital’s neonatal intensive care unit and then picked through the EEG readings, identifying tell-tale signatures of brain seizures and developing an algorithm to detect them automatically.

“That took a lot of time because we had to give it every possible eventuality and then we had to test it and we tried to trick it,” says Boylan, a senior lecturer in paediatrics at University College Cork.

“Babies are moved around, they have X-rays taken, bloods done, their nappies are changed, their backs are patted and that will change the EEG, so our algorithm needs to be able to handle all these other things, these patterns.”

The labour paid off and the algorithm can now pick out seizures with 93 per cent accuracy, she notes. “We feel we have now absolutely nailed what we need to pull out of the EEG to tell us that a seizure is happening. And when you think that the alternative is that nobody knows about it, if you can detect 93 per cent without an expert sitting by the cot at four in the morning, that’s really good.”

The researchers are now testing the EEG system in real time in a trial based at Cork and in University College London. The monitoring approach, which is supported by medical device company CareFusion, could eventually be applied in cots in intensive care.

“We are hoping in a year from now the clinical trial will be completed and we can get it at the bedside,” says Boylan. “But already children are benefiting because their seizures are being diagnosed earlier.”

As well as picking out seizures, the EEG can also help identify babies with altered brain activity who could benefit from brain cooling treatment shortly after birth to help reduce injury.

But the available time window is short, so the team is developing a remote monitoring system that can relay EEG readouts to mobile devices, meaning that experts can assess them quickly.

The monitoring system is also part of an ongoing EU-wide clinical trial to better assess the use of anti-seizure medication in babies. And in the future, Boylan would like to see EEG monitoring routinely included along with measurements of heart rate and breathing at the cotside.

“To manage these babies the EEG on its own doesn’t give you the whole picture, but it adds to the management of the baby when it comes to seizures and [brain injury],” she says.

Moving the basic EEG research and engineering findings from the bench towards the patient has depended on a multi-disciplinary team of clinicians, nurses, scientists and computing experts working together, and Boylan stresses that initial State funding for the project through Science Foundation Ireland and the Health Research Board paved the way for securing funding from the Wellcome Trust and the EU later on.

If funds for basic research are cut in tomorrow’s Budget, it could mean a loss of traction for many groups here, she argues.

“My worry is if they cull the funding in research, then we have no leverage for funding anywhere else,” she says. “If they naively cut research funding, it would be a disaster for so many people.”

Meanwhile, the UCC-centred team is checking in with the children involved in the neonatal brain-monitoring study to assess the longer term impact of early detection and treatment of brain seizures up to age five.

“We don’t say thanks for being part of our study, now goodbye – we are trying to keep the children in the study as long as possible,” explains Boylan.

“The parents love it because they know their babies were high risk at birth and by getting involved in our research their children get neurodevelopmental assessments, then if there is a problem we can put them in touch with an intervention team. There’s no point in doing all this research if it doesn’t improve outcome. That’s the long-term goal, we want to improve the health of children.”