Ventilation-induced brain injury in preterm neonates: causes, consequences and prevention

A large proportion of preterm infants will require some form of respiratory support at birth, due to their underdeveloped and immature lungs. Despite the necessity of respiratory support for survival, mechanical ventilation can be injurious to both the preterm lungs and brain. Recent studies have highlighted that more than 80% of preterm infants will inadvertently receive high and injurious tidal volumes (VT) in the delivery room. In preterm lambs, 15 min of high VT immediately following delivery is sufficient to initiate a cerebral inflammatory response and leads to cerebral haemodynamic disturbance. Similarly, in preterm babies, exposure to high VT in the delivery room was linked to increased incidence of intraventricular haemorrhage. To date, the investigation of respiratory support in the delivery room has focused only on otherwise healthy preterm neonates. This thesis aimed to determine, and characterise, the additional effects of common clinical scenarios with the initiation of ventilation. Chapter 3 of this thesis explored the combined effect of chorioamnionitis and the initiation of ventilation on the preterm brain. Chorioamnionitis is a common antecedent to preterm birth with its incidence increasing with decreasing gestational age; almost two-thirds of extremely preterm infants will be exposed to chorioamnionitis. Chorioamnionitis, similarly to ventilation, causes haemodynamic instability as well as instigating a fetal inflammatory response. It was found that the combination of two days of exposure to lipopolysaccharide (LPS) (to mimic chorioamnionitis) and 15 min of high VT ventilation (followed by conventional ventilation for 75 min) increased pro-inflammatory cytokine gene expression, densities of astrocytes and apoptotic cells within the brain, as well as causing haemodynamic instability. Interestingly, ventilating LPS-exposed lambs with a ‘protective’ ventilation strategy, shown previously to partially protect the preterm brain from ventilation-induced injury, did not reduce any markers of ventilation-induced brain injury. Chapter 4 of this thesis similarly investigated chorioamnionitis and the initiation of ventilation but additionally characterised the effect of antenatal glucocorticoid administration. This is also clinically relevant given a large proportion of preterm infants will have been exposed to glucocorticoids prior to birth. Unventilated lambs exposed to betamethasone had increased pro-inflammatory cytokine expression, astrocyte density and vascular extravasation within the brain. LPS administration prior to betamethasone made no additional difference. In contrast, betamethasone exposed ventilated lambs had stable haemodynamics and reduced cerebral inflammation and vascular extravasation. LPS administration prior to betamethasone again made no additional difference following ventilation. Thus, whilst our ‘protective’ ventilation strategy in Chapter 3 did not protect LPS-exposed lambs from ventilation-induced brain injury, betamethasone administration appeared to mitigate some markers of ventilation-induced brain injury, regardless of the presence or absence of intrauterine inflammation. Given the complex interactions between chorioamnionitis, glucocorticoids and the initiation of ventilation, three common clinical occurrences, we hypothesised that an adjunct therapy that could target the critical pathways involved in ventilation-induced brain injury may be of most benefit. Chapter 5 of this thesis investigated the use of Erythropoietin; studies in term and preterm babies have shown it has a neuroprotective capacity due to its anti-inflammatory and anti-apoptotic properties, as well as the ability to increase the integrity of the blood-brain barrier. Chapter 6 of this thesis investigated the use of human amnion epithelial cells, which are stem cell-like cells that have shown potential in pre-clinical models to ameliorate ventilation-induced lung injury; their effects in the brain are not as well characterised. To date, neither therapy has been characterised in combination with the initiation of ventilation. Erythropoietin showed more potential as a treatment for ventilation-induced brain injury as it successfully reduced cerebral vascular protein extravasation back to control levels, indicating it can act on the haemodynamic pathway involved in ventilation-induced brain injury. Human amnion epithelial cells also reduced protein extravasation, but not to the same degree. Both treatments had regional inflammatory effects in the white matter suggestive of incorrect dose or incorrect timing of administration. Further studies are required to elucidate their roles in relation to ventilation prior to clinical translation. Overall, this thesis has demonstrated that the intrauterine environment has the potential to alter brain inflammation and injury resultant from the initiation of ventilation. Given the complex interactions between chorioamnionitis, glucocorticoids and ventilation, three common clinical occurrences, a potential adjunct therapy is required. Erythropoietin, more so than human amnion epithelial cells, shows promise as a therapy for ventilation-induced brain injury; however, it requires follow-up in regards to dosing and timing as well as investigation into its interactions with chorioamnionitis and glucocorticoids prior to clinical translation.