Analysis of Cerebral Perfusion Using Head Ultrasound and Multisource Detector Near Infrared Spectroscopy (NIRS) Imaging

Purpose: There are 4 objectives for this study.

1. To determine if the use of cranial ultrasound and multi-source-detector near infrared spectroscopy can identify alterations in cerebral perfusion that are associated with long-term neurologic dysfunction.
2. Evaluate whether changes in resistive indices as determined by cranial ultrasound after cardiopulmonary bypass are associated with major neurologic dysfunction.
3. Describe regional changes in brain tissue oxygenation during selective cerebral perfusion using near-infrared spectroscopy and determine if there are critical thresholds that correlate with poor neurologic outcomes. To accomplish this, we will describe differences in cerebral blood flow and oxygenation as related to structural anatomy of the heart. Specifically, we will compare single ventricle and two ventricle hearts with regards to use of cardiopulmonary bypass with or without selective cerebral perfusion. We will also determine if a prolonged reduction in cerebral oxygenation intraoperatively is associated with ultrasound-diagnosed periventricular leukomalacia at discharge.
4. Prospectively collect patients' DNA to determine if differences exist in neurologic outcomes that are associated with genetic variation. Specifically, we will look at simple karyotyping, single nucleotide polymorphisms, and copy number variation differences between patients. This DNA will be collected and stored for future comparison with other children with congenital heart disease who undergo neurodevelopmental testing.

Background:

The incidence of congenital heart disease is at least 10 per 1000 live born children and children with congenital heart disease are at high risk for developing neurologic abnormalities1. Patients with a single ventricle have a mean full scale IQ between 89-97 and the incidence of severe mental retardation in hypoplastic left heart syndrome is as high as 18%2, 3. The factors influencing this risk are multiple and include preoperative, intraoperative, and postoperative events. One major risk factor is the use of deep hypothermic circulatory arrest (DHCA), which has historically been employed to obtain a bloodless operating field for aortic arch reconstruction. Since it is becoming increasingly evident that there is no safe duration of DHCA, newly developed strategies are aimed at maximizing cerebral blood flow and thereby improving neurologic outcomes4-7. One such strategy that has been widely employed is selective cerebral perfusion (SCP), which increases cerebral blood flow via a dedicated brain cannula advanced into the right common carotid artery during aortic arch reconstruction.8 This technique, which decreases the period of cerebral ischemia, has been shown in animal models to improve neurologic outcome, but similar studies have not been performed in humans9.

It is known that cerebral vascular resistance is increased after DHCA and it has been suggested that SCP may contribute to post-bypass cerebral vascular hypertension10, 11. Single channel near-infrared spectroscopy (NIRS) measurements have been shown to correlate with cerebral blood flow in animal models, but evaluation of SCP with multi-site-detector NIRS has not been performed, and it is unknown if there are regional oxygenation differences12, 13. This study would provide important information on the cerebral blood flow characteristics of regional tissue, particularly as it relates to unilateral perfusion as well as differences in patients with complex anatomy and compromised cerebral blood flow.

Cranial ultrasound with Doppler resistance measurements have been used to predict hypoxic ischemic encephalopathy in term and preterm infants14-16. It is known that infants with severe congenital heart disease have abnormal cerebral blood flow, particularly those with single ventricles, and that this puts them at increased risk for periventricular leukomalacia17-19. It is not clear, however, if these measurements are associated with adverse neurodevelopmental outcomes, and whether they can be used in combination with NIRS to provide bedside information on neurologic health and adequacy of current hemodynamic management.

While NIRS has been demonstrated to be associated with new or worsened ischemia on postoperative magnetic resonance imaging (MRI), it is often the case that these infants are not stable enough for the risks inherent in obtaining an MRI, including transport and sedation, and ultrasound may serve as an appropriate, safer, screening tool20, 21. We propose to use an experimental continuous diagnostic modality (NIRS), combined with a known screening tool (cranial ultrasound), to analyze and evaluate cerebral blood flow and oxygenation, and determine if abnormal neurodevelopmental outcomes can be predicted and potentially improved upon. The optimal timing of ultrasound performance is not known, and thus it is important to obtain pilot data to refine a diagnostic combination to best assess the risk of neurologic injury.

Another well known risk factor for poor neurologic outcome in children with congenital heart disease is genetics22. Single gene polymorphisms have been identified and associated with poor neurodevelopmental outcomes and more are being identified23. Further study needs to be done to better understand the genetic keys to congenital heart disease and the neurologic outcomes that result because of it.

This is a single center, prospective analysis of cerebral perfusion in infants up to 2 months of age who are undergoing repair of congenital heart disease. A total of 70 subjects will be consented on this protocol with the hopes that 60 will complete the study. Subjects will be expected to participate in the study for a total of 6 years, from enrollment to the final neurocognitive assessment. Subjects will receive $25 for the 12 month follow-up evaluation, $50 for the three year evaluation, and $75 for the 6 year evaluation as patient incentive.

Infants (up to 2 months) with congenital heart disease that undergo surgery at Children's Medical Center Dallas (CMCD) will be eligible. All such patients will receive standard pre-, intra-, and post-operative care.

Enrolled patients will undergo a pre-operative head ultrasound to determine resistive indices and Doppler flow of the middle cerebral artery. Patients will then have a multi-source-detector NIRS device placed on the scalp which will be held in place with Velcro. The probes are attached to wires that are connected to a data collection computer. This device will collect data intra-operatively about regional tissue oxygenation and blood flow and compare many different areas of the brain. A second head ultrasound will be obtained in the immediate post-operative period to determine resistive indices and Doppler flow (within 4 hours). Finally, a third head ultrasound will be performed within 48 hours of discharge to screen for periventricular leukomalacia, or injury to the brain's white matter. These examinations are not painful and produce no additional burden on the patient.

These diagnostic tests are part of a research process and are not standard of care. Single channel NIRS monitoring is standard of care and will be performed as it usually is.

Neurodevelopmental status will be assessed by a trained pediatric neuropsychologist at 12 months of age. The Bayley Scales of Infant and Toddler Development will be performed at the 12 month evaluation (2 hours). The Wechsler Preschool and Primary Scales of Intelligence will be performed at the 3 year evaluation (2 hours). The Children's Memory Scales, the Beery Visual-Motor Integration Tests, the Child Behavior Checklists, and the Woodcock Johnson Tests of Language Proficiency will be performed at the 6 year evaluation (3 hours). These tests are not considered standard of care for these patients, but can provide important information about learning disabilities and school performance.

Additionally, the subject will have the option to consent to providing a DNA sample. If the subject does consent, one sample of blood will be collected by a doctor, nurse, or licensed technician. We will take up to 30ml of blood to be collected on the day of the surgery from an already existing line. These samples will be drawn in accordance with CMCD's maximum blood sampling volume policies. The samples will be completely de-identified and assigned a unique two digit identifier prior to being sent to the Kernie lab at UTSW for extraction and analysis. This sample will be stored in a database to compare with future study subjects with congenital heart disease. This blood draw will occur through an already existing line (necessary for surgery) and will not add additional burden on the patient. Genetic testing may include analysis of single nucleotide polymorphisms, copy number variation, and simple karyotyping. Correlation with type of congenital heart disease, results of neurodevelopmental exams, and imaging results will be made with DNA results.