2022

An individual’s genes are inherited from their parents and are fixed. However, the expression of genes changes over time and in response to environmental cues. DNA methylation (DNAm) is one of the mechanisms that can regulate gene expression dynamically: the amount of methylation (a specific molecule) in a gene changes how much that gene is expressed. Since DNA methylation is important for brain development and is altered by early life adversity, we wanted to find out whether changes to DNAm might be responsible for linking low gestational age with altered brain development.

To do this, we used the TEBC to investigate whether children born preterm, in comparison to children born at term, have differences in DNA methylation in saliva and if so, whether those differences are associated with brain development using MRI. We analysed DNA methylation data and brain MRI from 258 babies in the TEBC.

We found that preterm birth is associated with a profound and widely distributed effect on DNA methylation across the genome. The sites on the genome that were affected provide new insights into the biological pathways that could mediate the effect of early life environmental exposures on brain growth. Finally, there were associations between DNA methylation and MRI scans which suggest that changes in DNA methylation may provide a link between preterm birth and atypical brain development in preterm infants.

Babies born preterm experience more stress than usual in their early life. Some people born preterm experience challenges regulating their emotions that could be linked to early life stress. In the TEBC, we investigated stress response and emotion regulation in infants born at term and infants born preterm.

Babies took part in a structured play session with their caregiver when they were 9 months old (corrected age). The caregiver was asked to ignore their baby for a short period during this play session. Before and after the play session small amounts of saliva were collected from the baby and cortisol (a stress hormone produced in response to stress that has a wide range of important roles in the body and brain) was measured from this. The infants’ behaviours and emotions during the play session were measured from video recordings.

Preterm infants showed fewer negative emotions compared to term born infants, and their stress hormone levels took longer to recover from stress. Term born babies who showed more negative emotions had higher stress hormone levels, but emotions and stress hormones were not related to one another in preterm infants.

In future studies we intend to explore whether early differences in stress response and emotion regulation may be useful for predicting childhood development and health.

TBC

Artificial intelligence (AI) is an integral part of our daily lives and has begun penetrating healthcare settings. This is especially so in complex healthcare settings where a large amount of healthcare data is generated. The intensive care provided for the most vulnerable newborn babies is becoming increasingly complex. However, the use of AI in newborn care is still in its infancy, lagging behind the adult healthcare settings.

AI can be used as a tool to analyse the large amount of data and complement other investigations, such as blood tests and imaging. This will support healthcare professionals in providing personalised efficient newborn care and shared clinical decision making with parents, besides reducing avoidable errors.

The study reviews the current potential application of artificial intelligence (AI) in the care of newborn babies. It also describes the challenges faced in integrating AI in newborn care. This is anticipated to raise awareness of the potential AI application among the neonatal community, accelerating newborn AI research and integration into clinical practice.

Artificial intelligence (AI) can be used to identify high-risk babies for timely targeted treatment and predict responses to treatment provided. It can also be used to support the interpretation of diagnostic investigations, such as those used in neurological disorders. AI application in image recognition will support the diagnosis of newborn conditions and the training of healthcare professionals. Key challenges in integrating AI in newborn care include the need for high-quality datasets to develop the AI tool and well-designed studies to robustly assess the developed AI tool.

The study findings will be of benefit to the neonatal community in understanding the potential and limitations of using artificial intelligence (AI) to improve the care we provide to the most vulnerable babies. Improved digital literacy and multi-disciplinary collaborations are needed to fully harness the potential of AI in newborn care.

Diffusion MRI is a brain imaging technique that is used to study the structure of brain connections – the ‘wiring’ that connects one region with another. Previously, using this technique we and others have shown that, on average, babies born preterm have less well-developed connections compared with babies born after a healthy pregnancy (full-term); this is linked to later learning problems. With diffusion MRI we can measure several different features of brain connections, which leads to a large and complicated set of information for each baby. It isn’t clear which particular feature or combination of features gives the most accurate overall picture of brain connectivity. Here, we used sophisticated statistical methods to summarise the data and find the feature/combination of features that provides the most accurate assessment of brain connections in preterm babies. The technique is based on calculating ‘general factors’ which summarise the extent to which connectivity features are similar across the brain.

We used brain images from 212 infants from the Theirworld Edinburgh Birth Cohort to derive these general factors. We found that the properties of brain connections are similar across the brain, meaning that if a baby has ‘weakness’ in one connection then s/he is quite likely to have weakness in other connections as well, and vice versa for strong connections. Further, general factors were associated with age at birth, meaning that it is easy to tell preterm babies apart from term-born peers by measuring general factors. These findings suggest that whole-brain estimates of white matter connections are an efficient tool to study atypical brain development and can be used to understand the causes and long term consequences of altered brain connectivity in early life.

Cortisol is a stress hormone that has a wide range of important roles in human health. These include regulating the immune system, metabolism and mood, and directing the body’s response to stressors. Based on studies showing that the how a person responds to stress is shaped by experiences in early life, we investigated whether cortisol secretion differed between infants born at term and infants born preterm through two studies. In a TEBC study, we measured saliva cortisol concentrations collected in the morning and evening (to assess how cortisol varies across the day) and before and after vaccination (to assess a stress response), at 4 months after birth. This study showed that infants born extremely preterm (before 28 weeks’ gestation) had less variation in cortisol across the day, and a blunted cortisol stress response. In this second study we analysed saliva cortisol concentrations that had previously been measured across infancy in a preterm birth cohort in Sweden. This analysis also showed that infants born extremely preterm had reduced variation in cortisol across the day, replicating what was seen in TEBC infants. Together these studies highlight that extremely preterm birth is associated with changes in cortisol secretion in infancy. In future studies, we intend to investigate whether these differences continue at older ages and/or predict childhood development and health.

Cortisol is a stress hormone that has a wide range of important roles in human health. These include regulating the immune system, metabolism and mood, and directing the body’s response to stressors. Based on studies showing that the how a person responds to stress is shaped by experiences in early life, we investigated whether cortisol secretion differed between infants born at term and infants born preterm through two studies. In TEBC, we measured saliva cortisol concentrations collected in the morning and evening (to assess how cortisol varies across the day) and before and after vaccination (to assess a stress response), at 4 months after birth. This study showed that infants born extremely preterm (before 28 weeks’ gestation) had less variation in cortisol across the day, and a blunted cortisol stress response. In a second study we analysed saliva cortisol concentrations that had previously been measured across infancy in a preterm birth cohort in Sweden. This analysis also showed that infants born extremely preterm had reduced variation in cortisol across the day, replicating what was seen in TEBC infants. Together these studies highlight that extremely preterm birth is associated with changes in cortisol secretion in infancy. In future studies, we intend to investigate whether these differences continue at older ages and/or predict childhood development and health.

One of the aims of Theirworld Edinburgh Birth Cohort (TEBC) is to find new ways of identifying during infancy those preterm children who are likely to develop problems later in childhood. This is so that interventions may be targeted to the right child during the pre-school years when they are likely to be most effective.

Cerebral lateralisation is the development of specialised processes for the left and right hemispheres of the brain and is an indication of typical brain development.  Lateralisation differences are linked to atypical social development in conditions such as autism. Based on the fact that preterm birth is closely linked with altered brain development and social difficulties in childhood, we aimed to find out whether we could identify early lateralisation differences in preterm infants using eye-tracking data. 

To do this, we used the TEBC to explore whether children born preterm, in comparison to children born at term, exhibited differences in the time they spent looking to faces versus objects in the left or right visual field. We analysed data from two eye-tracking tasks featuring social and non-social content with 51 preterm and 61 term-born infants aged 8-10 months. 

Infants born preterm showed a significantly reduced interest in social stimuli on the left versus the right compared to term children. These differences are indicative of atypical development of visual field biases, and therefore provide evidence of altered cerebral lateralisation in infancy in preterm infants. This reduction in left looking is the opposite pattern that we would expect to see in typical development. It is possible that this finding could be a physiological marker of early processing differences, and indicative of emerging lateralisation differences that are seen across the lifespan in people born preterm. The study takes us one step closer to being able to identify early children who may go on to need additional support, and they shed new light on the neural basis of learning.