Kim Annink

257 General discussion / Conclusions & clinical implications / Future directions situations where MRI is not available or feasible, cerebral ultrasound (CUS) might be an alternative. The benefits of CUS are the lower costs, portability and the ability to monitor evolution of brain injury daily (29). In chapter four we describe the development and validation of a cerebral ultrasound scoring system for infants with HIE . We retrospectively included (near-)term neonates with HIE that received therapeutic hypothermia in two different hospitals. The final scoring system contains a deep grey matter score (composite score of hyperechogenicity of thalamus and of putamen, hypoechogenicity of the posterior limb of the internal capsule and the four column sign (defined as hyperechogenicity of the basal ganglia and thalami in the coronal view)) and a deep white matter score (including subcortical white matter involvement, periventricular white matter involvement and existence of edema). This score on day three to seven was well associated with adverse neurodevelopmental outcome (AUC=0.90), also in the validation cohort (AUC=0.89). The CUS on day one was not predictive for adverse outcome, but useful to detect antenatal pathology. Chapter five and six focus on neuroimaging and school-age outcome. It is important to assess neurodevelopment of infants with HIE into childhood, since some deficits can elucidate later in life and children can “grow into their deficits” (30–34). MRI at school- age enables us to investigate the consequences of the perinatal event and helps us to better understand which structures or connections are important for specific problems. This information might allow us to identify neonates at risk for problems at school-age, if the biomarkers at school-age can be related to neonatal injury. The aim of chapter five is to describe the effect of HIE on hippocampal volume in school-age children and to investigate the association between hippocampal volume and cognition and memory. Therefore nine- to ten-year-old children with mild or moderate HIE were included and compared to healthy controls, all born before therapeutic hypothermia became standard of care. Children with moderate HIE had significantly smaller hippocampal volumes compared to controls, also children with mild HIE seemed to have smaller volumes although this did not reach statistical significance. Furthermore the children with HIE had episodic memory problems. The smaller hippocampal volume (as percentage of total intracranial volume) was associated with poor long-term visuospatial memory. Afterwards, we started a new cohort study to investigate the ten year outcome in a group of children with HIE that were treated with hypothermia and a group that 12