Hypoxic-ischaemic encephalopathy results from an acute period of reduced oxygen (hypoxia) and blood flow (ischaemia) at the time of birth, and typically comes as the devastating end to an otherwise normal pregnancy. The most severely affected infants will die within days, whereas the majority who survive have a high chance of sustaining significant lifelong disabilities including cerebral palsy. This therefore has profound implications for the affected infant, their family, and wider society.
Magnetic resonance spectroscopy allows the non-invasive investigation of cerebral metabolism, and provides quantitative measures of brain injury. Since hypoxic-ischaemic encephalopathy is associated with acute metabolic changes in the perinatal period, magnetic resonance spectroscopy has the potential to give insight into the ongoing injury, and potentially predict the neurodevelopmental prognosis of affected infants at an early stage.
This thesis tests the hypothesis that magnetic resonance spectroscopy, acquired within two weeks of birth, is effective at predicting neurodevelopmental outcomes years after the initial hypoxic-ischaemic injury. This is achieved via a prospective multi-centre observational study across several institutions across the UK and USA, the results of which indicate exceptional prognostic value.
The following work tests the hypothesis that accurate magnetic resonance spectroscopy measurements can be performed in the context of routine clinical settings to assist prognostication. A novel acquisition and analysis strategy is developed and validated, demonstrating precise and accurate results in a clinical setting within a modest acquisition time.