Monitoring the oxygenation of the preterm brain : What is there to gain?

Abstract

Despite advances in perinatal care, preterm birth is still associated with adverse neurodevelopmental outcome, often caused by cerebral injury. The most common forms of cerebral injury are all associated with disturbances in cerebral oxygen and blood supply. Traditionally, infant wellbeing is evaluated by monitoring, heart-rate, arterial oxygenation and blood pressure. Although blood pressure is certainly crucial for maintaining cellular homeostasis, it is by no means a direct representative for cerebral oxygen and blood supply.

In this thesis, non-invasive techniques such as Near-InfraRed Spectroscopy (NIRS), perfusion-index, and Magnetic Resonance Imaging (MRI) are used to assess (cerebral) haemodynamics. On the neonatal intensive care unit of the Wilhelmina Children’s Hospital in Utrecht, the Netherlands, monitoring of cerebral oxygenation during the first 72h of life in all preterm infants is part of routine clinical care. Given certain assumptions, the oxygen saturation presented by NIRS, can also be used as a surrogate of perfusion. NIRS-data was collected in 999 preterm infants (born <32 weeks of gestation). Furthermore, neonatal and adult populations are studied to develop a non-invasive MRI technique that provides cerebral oxygen saturation throughout the brain, including deeper brain structures.

The cerebral oxygenation is increased and more variable preceding a large intraventricular bleed and these infants also received more dopamine in support of their blood pressure. Low blood pressure on its own was not associated with decreased cerebral or renal oxygen saturation, which can be interpreted as not being associated with decreased cerebral and renal perfusion. In addition, low blood pressure was not related to adverse neurodevelopmental outcome, where low cerebral oxygenation does seem to be related to neurodevelopmental outcome. To more thoroughly investigate this observation, reference values were generated for various gestational ages during the first 72h after birth. Infants who spend more time outside, so either low or high cerebral oxygenation, these reference values had lower neurodevelopmental outcome scores at 2 years of age.

Where NIRS provides regional, but continuous, information, recent advances in MRI have provided new techniques to assess (neonatal) cerebral haemodynamics. It can be concluded that whole brain perfusion imaging is possible, providing more insight into brain development. However, neonatal heamodynamic imaging could benefit from a white paper. In addition, this thesis shows that imaging whole brain cerebral oxygen saturation is possible, both in adults and neonates. This could complement NIRS assessments with information on deeper brain structures and on the long run serve as a validation tool for NIRS measurements in daily neonatal care.

In conclusion, NIRS has merit in daily clinical care, even when absolute numbers are used as opposed to trend-monitoring. Cerebral oxygen saturation is associated with both short and longer term outcomes. To be done next: 1) to assess whether modifying cerebral oxygenation results in better neurodevelopmental outcome, 2) determine what is the true cerebral oxygenation in (preterm) neonates by MR-based validation, and 3) use cerebral oxygenation in conjunction with blood-pressure to provide an on-line measure of cerebral autoregulation.