Kim Annink

129 The papez circuit and school-age outcome in HIE Multivariable logistic regression: Papez circuit and neurodevelopmental outcome Based on the results above, the association between volumes of the Papez circuit (MB, fornix, hippocampus) and intelligence and memory appeared to be most relevant. These brain volumes were strongly correlated with each other (Supplemental Figure 1). Because of multicollinearity it was therefore impossible to include the different brain volumes in one model. To correct for other brain volumes, we therefore made two different models. First, multivariable logistic regression was performed with the neuropsychological test results as dependent variable and MB atrophy, total white matter, total grey matter, total volume lateral ventricles and total cerebellar volume as independent variables. Secondly, we performed the same analyses but with total hippocampal volume instead of MB atrophy (Table 5). Table 5: Multivariable logistic regression models for different neuropsychological tests with MB and with total hippocampal volume. Outcome measure Model with MB Odds ratio (95%CI) Model with hippocampus Odds ratio (95%CI) Cognition TIQ MB atrophy 35.00 (5.53-221.39) Hippocampus 0.19 (0.07-0.55) VIQ Grey matter MB atrophy 0.98 (0.97-1.00) 7.26 (1.1-48.33) White matter 0.97 (0.95-1.00) PIQ MB atrophy 32.50 (5.12-206.16) Hippocampus 0.19 (0.06-0.53) Processing Speed MB atrophy 3.75 (0.90-15.66) Hippocampus 0.48 (0.23-0.99) Memory Verbal long- term memory – immediate recall MB atrophy 50.00 (5.40-463.20) Hippocampus Lateral ventricles 0.24 (0.09-0.65) 1.2 (0.99-1.40) Verbal long-term memory – delayed recall MB atrophy n/a a Hippocampus Grey matter 0.16 (0.05-0.47) 1.02 (1.00-1.03) Verbal Working Memory Grey matter Lateral ventricles 0.99 (0.98-1.00) 1.11 (0.99-1.26) Grey matter Lateral ventricles 0.99 (0.98-1.00) 1.11 (0.99-1.26) Visual-spatial Working Memory n.s. n.s. n.s. n.s. 6