41 - Pre-Operative Brain Injury in Neonates with Critical Congenital Heart Defects

Publication Number: 1034.41

Molly McGetrick, The University of Texas Southwestern, Dallas, TX, United States; Irtiza Sakif Islam, University of Texas Southwestern Medical School, Dallas, TX, United States; Nicholas Andersen, University of Texas Southwestern Medical School, Dallas, TX, United States; Rachel L.. Leon, University of Texas Southwestern Medical School, Dallas, TX, United States

Background: Neonates with critical congenital heart disease (CHD) have increased risk for abnormal brain development and brain injury. Many institutions employ pre-operative cranial ultrasound (US) while a minority of centers obtain pre-operative brain MRI in these patients. Clinical factors associated with pre-operative brain injuries are not well described, and currently, there are no formal guidelines regarding neuroimaging in this high-risk population.

Objective: To determine the prevalence and clinical factors associated with abnormal neuroimaging in a cohort of pre-operative neonates with critical forms of CHD; secondarily, to assess the utility of cranial US in detection of pre-operative brain injury.

Design/Methods: This retrospective cohort study included neonates with critical forms of CHD born between 2017 and 2025, who underwent cardiac surgery with cardiopulmonary bypass and had pre-operative brain MRI obtained within the first 30 days of life. Brain MRI results were categorized as abnormal if any congenital or acquired injuries were present other than subdural hemorrhage related to birth trauma. Logistic regression was used to determine the odds ratio of abnormal MRI for patients with specific clinical features and for specific cardiac diagnoses. The utility of cranial ultrasound (US) as a screening tool was assessed with confusion matrix analysis for the subset of patients who received both cranial US and brain MRI.

Results: In our cohort of 339 neonates with critical CHD, brain MRI abnormalities occurred in 155 (46%), with white matter injury (25%) and intracranial hemorrhage (24%) most common (Figure 1). Logistic regression found an increased adjusted odds ratio (aOR) of abnormal brain MRI in neonates requiring pre-operative intubation (aOR 3.5; p< 0.0001]), but other clinical factors were not associated with increased risk of abnormal brain MRI (Figure 2). In subgroup analysis, neonates with transposition of the great arteries (TGA) who underwent balloon atrial septostomy (BAS) had the highest risk of abnormal brain MRI (70% vs. 37% in TGA without BAS; p=0.02; Figure 3). In patients who had cranial US prior to brain MRI (n=215), cranial US had poor sensitivity (19%) and accuracy (55%) in identifying brain injuries, but showed 90% specificity.

Conclusion(s): In one of the largest pre-operative cohorts reported on to date, we found that brain injury was present in almost half of neonates with critical CHD, but only the need for pre-operative intubation was associated with brain MRI abnormalities. While screening head US was commonly performed, it had poor sensitivity to detect brain injury.

Prevalence of Specific MRI Brain Abnormalities in Neonates with Critical Congenital Heart Diseases
Stacked bar graph of the prevalence of specific types of brain MRI findings in each diagnostic category of CHD. HLHS: hypoplastic left heart syndrome; SV: single ventricle physiology; TGA: transposition of the great arteries; DORV: double-outlet right ventricle; TOF: tetralogy of Fallot; TAPVC: total anomalous pulmonary venous connection; WMI: white matter injury; IVH: intraventricular hemorrhage; IPH: intraparenchymal hemorrhage; ICH: intracranial hemorrhage; GMI: gray matter injury; AIS: arterial ischemic stroke

Clinical Factors Associated with Abnormal Pre-Operative Brain MRI
Forest plot showing adjusted odds ratios (aORs) and 95% confidence intervals (CIs) from a multivariable logistic regression model evaluating associations between clinical characteristics and abnormal pre-operative brain MRI findings. Predictors used in the model included gestational age at birth (weeks), birth weight (kg),male sex, low Apgar score (defined as an Apgar score <6 at either 1 or 5 minutes), presence of a genetic diagnosis, prematurity (gestational age <37 weeks), pre-operative intubation, assigned STS CHSD Risk, and STAT surgical complexity score. The dashed vertical line represents an aOR of 1 indicating no association. Pre-operative intubation was the only variable significantly associated with abnormal brain MRI findings (aOR = 3.5, 95% CI 2.1-6.3, p<0.0001).

Estimated Prevalence of Abnormal Pre-operative Brain MRI by Diagnostic Subgroup of CHD
Firth-penalized logistic regression without an intercept (one estimate per group; no reference group) showing estimated prevalence of abnormal pre-operative brain MRI with 95% confidence intervals in each of the diagnostic subgroups of CHDs in the cohort. VSD: ventricular septal defect; AA: aortic atresia; AS: aortic stenosis; SV: single ventricle physiology; BAS: balloon atrial septostomy; DORV: double outlet right ventricle; TOF: tetralogy of Fallot; TAPVC: total anomalous pulmonary venous connection.