Neurodevelopmental outcome in perinatal asphyxia: prediction and measurement

Abstract

The aim of this thesis was to improve our ability to predict and measure neurodevelopmental outcome in early childhood with particular reference to high-risk infants with perinatal asphyxia. Methods: 1) Promising umbilical cord blood biomarkers were analysed for ability to predict performance in the Bayley Scales of Infant and Toddler Development (Edition 3) at three years. 2) A retrospective cohort was analysed for performance of a low-risk cohort on the Bayley-3 at two years and compared to standardised scores. 3) A survey asked parents to report prevalence and quality of touch-screen usage in their toddlers. 4) Pilot testing of a novel cognitive assessment tool, the“Babyscreen App”, was performed on a prospective low risk cohort. The Babyscreen App was administered alongside the Bayley-3 at age 18 months to 2 years. Results: 1) IL-16 predicted severe outcome with an area under the ROC curve of 0.83 (p= Levels ≥ 514 pg/mL predicted a severe outcome with a sensitivity of 83% and a specificity of 81%. 2 metabolite models were tested. Model A predicted abnormal outcome with an area under ROC curve of 0.77, p<0.01. Model B was robust to predict both severe outcome (area under ROC curve of 0.92, p<0.01) and intact survival (0.80, p=0.01). 2) 240 two year olds were analysed for performance on the Bayley-3. Language and fine motor scores were significantly higher compared to U.S standardised norms, 109 ± 13 v. 100 ± 15 , p<0.001, and 11.5 ± 2 v. 10 ± 3, p<0.001 respectively. 3) For the examination of touch-screen usage in toddlers, 82 questionnaires were completed by parents of typically developing children aged 12 to 36 months. 71% of toddlers included had access to touch-screen devices for a median (IQR) of 15 (9-26) minutes per day. By 24 months the majority of children were able to swipe, unlock and actively look for touch-screen features. 4) 95 children underwent administration of the Babyscreen App and the Bayley-3. Significant medium sized correlations occurred between various measures of app performance and cognitive composite scores on the Bayley-3. Combined measures of overall app performance could predict cognitive scores less than 90 (1SD below the mean of our cohort) with an area under the ROC curve of 0.69 (0.55-0.83), p=0.02. Conclusion: This thesis has shown that novel biomarkers measured in umbilical cord blood at birth can predict neuro developmental outcome and that a novel touch-screen application can assess cognition in toddlers.