Mary A. Rutherford

Assessment of brain tissue injury after moderate hypothermia in neonates with hypoxic-ischaemic encephalopathy : a nested study of a randomised controlled trial

Summary Background Moderate hypothermia in neonates with hypoxic–ischaemic encephalopathy might improve survival and neurological outcomes at up to 18 months of age, although complete neurological assessment at this age is difficult. To ascertain more precisely the effect of therapeutic hypothermia on neonatal cerebral injury, we assessed cerebral lesions on MRI scans of infants who participated in the Total Body Hypothermia for Neonatal Encephalopathy (TOBY) trial. Methods In the TOBY trial hypoxic–ischaemic encephalopathy was graded clinically according to the changes seen on amplitude integrated EEG, and infants were randomly assigned to intensive care with or without cooling by central telephone randomisation. The relation between allocation to hypothermia or normothermia and cerebral lesions was assessed by logistic regression with perinatal factors as covariates, and adjusted odds ratios (ORs) were calculated. The TOBY trial is registered, number ISRCTN 89547571. Findings 325 infants were recruited in the TOBY trial between 2002 and 2006. Images were available for analysis from 131 infants. Therapeutic hypothermia was associated with a reduction in lesions in the basal ganglia or thalamus (OR 0·36, 95% CI 0·15–0·84; p=0·02), white matter (0·30, 0·12–0·77; p=0·01), and abnormal posterior limb of the internal capsule (0·38, 0·17–0·85; p=0·02). Compared with non-cooled infants, cooled infants had fewer scans that were predictive of later neuromotor abnormalities (0·41, 0·18–0·91; p=0·03) and were more likely to have normal scans (2·81, 1·13–6·93; p=0·03). The accuracy of prediction by MRI of death or disability to 18 months of age was 0·84 (0·74–0·94) in the cooled group and 0·81 (0·71–0·91) in the non-cooled group. Interpretation Therapeutic hypothermia decreases brain tissue injury in infants with hypoxic–ischaemic encephalopathy. The predictive value of MRI for subsequent neurological impairment is not affected by therapeutic hypothermia. Funding UK Medical Research Council; UK Department of Health.

43 Total Cerebral Volume Measurements Following Preterm Birth

Background. Preterm infants have reduced cerebral tissue volumes in adolescence, and relative loss of some brain regions in early childhood, but it is unclear if this is due to perinatal or more prolonged growth failure. We therefore compared total cerebral tissue volume in preterm infants at term equivalent age and term born controls.Methods. Subjects: We studied 89 preterm infants born at median 29.7 weeks GA without parenchymal lesions at median 40.57 weeks, and 20 term born controls (median 40.43 weeks GA). Image acquisition: A 1.5 Tesla MR system was used to acquire T1-weighted volume datasets with a voxel size 1x1x1.6mm, in addition to conventional and diffusion weighted imaging. Image processing: A non-rigid image registration algorithm transformed all images to a reference subject, and transformations were used to propagate region of interest (ROI) labels segmented in the reference anatomy to the corresponding structure in all subjects. The volume change for each ROI relative to the reference was computed, enabling cerebral tissue volume measurements (excluding CSF) to be calculated for each subject.Results. Mean cerebral tissue volume in preterm infants at term was 404.9cm3; and in term controls, 401.1cm3 (p=0.765). Diffuse white matter injury, oxygen requirement at 28 days, and intrauterine growth restriction were not associated with significant reductions in cerebral tissue volume.Conclusions. Despite evidence of regional and later growth failure, total brain tissue volume in preterm infants at term is similar to term born controls.

106 Cortical Growth and Neurocognitive Impairment in Preterm Infants

Background. Neurocognitive impairment after preterm birth is common and directly related to increasing prematurity. The neural substrate for this dysfunction is unknown, but cannot be wholly accounted for by major destructive brain lesions.Aims. To test the hypothesis that the timing of premature delivery is associated with a dose-dependent reduction in the rate of growth of the cerebral cortex which parallels the increase in neurocognitive impairment.Methods. MR images were obtained from a consecutive cohort of extremely preterm infants serially from soon after delivery until term corrected age and neurodevelopment was assessed using the Griffiths Developmental Scales at median 24 months corrected age. Total cerebral volume and cortical surface area were measured semi-interactively, and their rate of growth found to be related by a power law, the scaling exponent of which gives a direct measure of the growth rate of cortical surface area relative to cerebral volume. A generalized least squares random effects regression model was constructed to test the effect of gestational age at birth and potential confounding variables on the scaling exponent, and the relation to neurodevelopmental outcome was tested.Results. 113 infants born at 23–29 weeks gestation were imaged providing 274 images without major destructive lesions for analysis. Complete neurodevelopmental data are available on 63 infants. Increasing prematurity was associated with a reduced rate of cortical growth (p<0.0001) which was independent of intrauterine or postnatal somatic growth. There was a significant relation between reduced cortical growth and the Griffiths Developmental Quotient (p<0.05), seen with all subscales except the locomotor.Conclusions. Reduced rate of growth of the cerebral cortex parallels the dose-dependent effect of prematurity on neurodevelopmental function. Cortical growth failure may be a neural substrate for the high rate of non-locomotor neurocognitive impairment seen in preterm infants.