Abstract 4126358: Fetal cardiac physiology but not cerebrovascular resistance predicts neurodevelopmental outcomes in congenital heart disease

Abstract

Introduction: Fetal “brain sparing” autoregulatory mechanisms respond to reduced cardiac output or substrate delivery by decreasing cerebrovascular resistance to maintain cerebral blood flow. In patients with congenital heart disease (CHD), data are conflicting on the association between fetal cerebral Doppler measures and postnatal neurodevelopment, possibly due to the heterogeneity of fetal cardiac physiology. Aim: To evaluate whether fetal cerebrovascular resistance estimated by middle cerebral artery pulsatility index (MCA PI) and cerebroplacental ratio (CPR; MCA PI divided by umbilical artery PI) predict neurodevelopment in patients with CHD. Methods: This retrospective analysis of a prospectively-enrolled cohort included cases with isolated CHD and healthy controls with a family history of CHD. Cases were classified by fetal cardiac physiology: single vs. biventricular circulation, outflow obstruction (none, systemic, pulmonary), and cerebral substrate delivery (normal, mixed, abnormal). Subjects underwent fetal echocardiography with cerebral and umbilical doppler measures (normalized for gestational age) as well as neurodevelopmental evaluation at two years of age with the Bayley Scales of Infant and Toddler Development III and IV. Results: Fetal echocardiograms and neurodevelopmental testing were obtained in 82 cases and 48 controls. There was no difference in MCA PI or CPR between cases and controls; however cases with mixed vs. normal substrate delivery had lower z-scores for MCA PI (-1.08 + 1.16 vs. -0.43 + 1.38, p=0.05) and CPR (-0.74 + 0.62 vs. -0.37 + 0.59, p=0.02). Cases had lower Bayley scores across all composite domains: cognitive (97.8 + 11.8 vs. 103.4 + 14.7, p=0.03), language (97.6 + 15.6 vs. 105.6 + 21.3, p=0.03), and motor (94.0 + 10.7 vs. 100.4 + 10.9, p=0.003). MCA PI and CPR were not associated with Bayley scores in cases or in sub-group analysis by fetal cardiac physiology. In multivariable analyses of Bayley scores adjusting for medical and demographic data, lower motor composite score was predicted by mixed vs. normal substrate delivery (partial R2=12.4%, p<0.001) and single ventricle physiology (partial R2=6.5%, p=0.008). Conclusions: In patients with CHD, fetal single ventricle physiology with mixed substrate delivery was associated with worse motor development. The lack of association of fetal cerebral dopplers with neurodevelopment may suggest that decreased cerebrovascular resistance is not necessarily an effective compensatory mechanism in CHD.