Impact of early cerebral oxygenation on neurodevelopment outcomes in high-risk preterm infants

Infants who are born preterm and/or at low birth weight (LBW) have an increased likelihood of subsequent neurodevelopmental delays and disability. Improving long-term neurodevelopmental outcomes for the increasing population of survivors of preterm birth is priority in new-born and paediatric care. Potential strategies to address this neurodevelopmental risk include understanding and mitigating the causes of preterm related brain injury within the neonatal period, developing robust developmental surveillance programmes for at risk infants and early identification, detection and intervention for developmental difficulties.

Firstly, we completed a retrospective study of 106 high-risk infants discharged from a tertiary neonatal unit over a 5 year period to characterise contemporaneous developmental follow up pathways and clinical outcomes.

Secondly, we developed and implemented a clinical pathway for the routine use of an early standardised developmental assessment (The General Movement Assessment, GMA). The feasibility of using this assessment to predict the likelihood of developmental disability within the context of our high-risk population and Developmental Paediatric service was then evaluated.

Thirdly, we performed a prospective observational cohort study of 60 infants born preterm (< 30 weeks) and with LBW (< 1000 grams) to explore the relationship between early neonatal physiological experiences and later development outcomes. Cerebral oxygenation was measured using Near-Infra Red Spectroscopy (NIRS) for the first 72 hours of life and we performed neurodevelopmental assessment at 3 months, 12 months and 24 months post term age (PTA). In this cohort of preterm infants, we examined the predictive values of the GMA on later developmental outcomes to explore whether this assessment can guide neurodevelopmental follow up and intervention.

We demonstrated that our existing neurodevelopmental surveillance has high coverage, with most infants (87%) in our region referred for neurodevelopmental surveillance followed up until 24 months PTA. For 68% of these infants, development was assessed as typical at 2 years PTA.

We established a framework and methodology for the implementation of GMA as a tool for routine neurodevelopmental assessment for at-risk infants in the Neonatal Intensive Care Unit (NICU). Additionally, in a group of high-risk preterm infants, we found that lower motor optimality scores on GMA at age 3 months PTA were associated with an increased probability of developmental delay at 12 and 24 months PTA. We demonstrated a trend towards an association between cerebral hypoxia burden in the first 3 days of life with sub-optimal GMA at 3 months PTA and lower language scores on developmental assessment at 12 months PTA.

Regular follow up for preterm LBW infants is essential to identify those at risk of neurodevelopmental difficulties and provide early targeted intervention to improve their long-term functional and wellbeing outcomes. Within our prospectively recruited cohort of high-risk infants we demonstrated that whilst rates of developmental difficulties are higher in this group compared to the general population, many infants had typical development at aged 2 years PTA. Conventional developmental surveillance is resource and time intensive for health services and whānau,with a low yield in terms of case identification. The use of robust early screening tools to identify those infants most at risk can help target longer-term follow up and minimise the ongoing medicalisation of infants following a normal developmental trajectory.

We have demonstrated that within our service, GMA can be successfully implemented as part of routine assessment for at risk infants. The GMA, and specifically the motor optimality score were associated with an increased probability of later developmental delay. We therefore propose that GMA is routinely used for this high-risk population to predict those infants most likely to follow an adverse neurodevelopmental trajectory and optimise opportunities for timely intervention.

In addition, early measurement of cerebral oxygenation using NIRS is feasible within the neonatal unit. This work highlights an association between cumulative cerebral hypoxia burden and later developmental outcomes, particularly language delay. Further work is needed to establish the relationship between cerebral hypoxia and short and longer term developmental outcomes in infants born preterm. However, strategies within the NICU to improve cerebral oxygenation are achievable and may offer another therapeutic strategy for neuroprotection in this vulnerable group of preterm infants.

In summary, although preterm birth is associated with a range of developmental difficulties, many children born extremely preterm have good developmental outcomes, but predicting which infants will follow typical developmental trajectory can be difficult. Early cerebral hypoxia is a potentially modifiable risk factor for preterm brain injury and we demonstrated an association between the duration of cerebral hypoxia and later developmental outcomes. Early developmental screening using GMA is accessible and acceptable and can help streamline follow up and direct resources to those most at risk of neurodevelopmental disability. More work is needed, but this thesis provides valuable information on early life exposures and on longer term outcomes, and provides an implementation strategy for early developmental screening.