Kim Annink

255 General discussion / Conclusions & clinical implications / Future directions Part I: neuroimaging and follow-up Since many years there is a focus in neonatal research on the prediction of neurodevelopmental outcome in neonates at risk for brain injury. In neonates with HIE many biomarkers have been investigated (11). Discovery of (a combination of) biomarkers that accurately predict(s) neurodevelopmental outcome is valuable for clinical decision making ( e.g. redirection of care), to select patients at risk for adverse outcome for (future) interventions and early rehabilitation and it will be essential for the counseling of parents about the development of their child. At this moment there is no biomarker that perfectly predicts outcome, which is important to keep in mind while counseling parents. Neuroimaging is one of the three pillars to predict outcome in neonates with HIE, besides the clinical condition of the patient (e.g. abnormal neurological examination, multi-organ failure) and neurophysiology (12). Many studies have confirmed the important role of multiple MRI sequences in predicting outcome (12–16). Despite the availability of many research on neuroimaging in HIE, there are still gaps in our knowledge. Some of these gaps are addressed in Part 1. The focus in neonatal neuroimaging following HIE is traditionally on supratentorial brain injury. Based on postmortem histological studies it is known, however, that the cerebellum is also very vulnerable to hypoxic-ischemia (17–19). From extensive studies in preterm infants, we know that cerebellar injury is associated with not only motor problems, but also cognitive and behavioral problems (20). More research is needed on cerebellar injury in HIE and the consequence of cerebellar injury on the long-term neurodevelopmental outcome. Pathological examination can provide more information on the distribution and extent of cerebellar injury, which might be important for the effect of cerebellar injury on neurodevelopment. Purkinje cells are the main output neurons of the cerebellum and are very vulnerable to hypoxic-ischemia (21). In chapter two we describe the distribution of Purkinje cell injury in the cerebellar vermis. Term neonates with HIE were included if they had autopsy including hematoxylin and eosin (H&E) stained sections of the vermis. We found that the number of Purkinje cells was less in the bases of the sulci compared to the crowns of the sulci for both the lobules and folia. There were also more injured Purkinje cells in the bases of the sulci compared to the crowns. The pattern of Purkinje cell injury in the vermis is very 12