45 Using fetal ultrasound to predict developmental outcomes in children born with cardiac defects in Manitoba: A historical cohort study

Abstract Background The cavum septum pellucidum (CSP) is a normal anatomical cavity found in the mid-brain of the fetus and is routinely imaged at the time of mid-pregnancy obstetrical ultrasound as part of anatomic survey. Presence of the CSP reflects intact intrahemispheric connections and an intact corpus callosum. The CSP space then closes by the time of birth or early infancy, however persistence of the CSP into adulthood has been described in those with the schizophrenic phenotype of DiGeorge syndrome. Preliminary work by our group in Manitoba has shown utility in measurement of CSP width to improve non-invasive prenatal diagnosis of DiGeorge syndrome among fetuses with and without conotruncal cardiac defects. However, there exists no information about the relationship between fetal CSP width and future neurodevelopmental outcomes, such as Bayley-III scores, in children with cardiac lesions in the absence of underlying genetic comorbidities. Objectives To determine the relationship between cavum septum pellucidum (CSP) width and neurodevelopment in children born with sever cardiac defects. Design/Methods This was a historical cohort study of children born with cardiac anomalies in Manitoba (2009-2019). Eligible children with developmental pediatric follow-up between ages 18-24 months corrected age were identified using a clinical repository. Post-processing measurement of fetal CSP at mid-gestation using stored obstetrical ultrasounds was performed by two blinded and independent reviewers in a standardized fashion. Ultrasound reports were also reviewed to collate data about routine fetal biometry and imaging data then linked to demographics, birth information, surgical details and neurodevelopmental outcomes (Bayley-III scales). Linear regression models were used to evaluate the relationship between CSP width and Bayley-III scores; inferential statistics were used to compare outcomes between groups (biventricular versus single-ventricle repair). Results 78 of 111 eligible cases had stored ultrasound images and included in the analysis. Most children were delivered at term following vaginal delivery (71%), with normal birthweight and Apgars. Median CSP width was 3.8mm (SD 1.2). There was a range conotruncal cardiac anomalies and 47% required single-ventricle repair. There was no obvious relationship between CSP width and Bayley-III motor (p=0.932) or cognitive domains (p0.686), however there was a trend towards larger CSP and higher Bayley-III language scores (p=0.049). Conclusion The influence of the deoxygenated blood on the developing fetal brain in modifying outcomes for children with severe cardiac defects needs future study to improve postnatal development in this high-risk group.