P/CHD
Alex J. Barker, PhD
Associate Professor
Children's Hospital Colorado, University of Colorado Anschutz Medical Campus
Alex J. Barker, PhD
Associate Professor
Children's Hospital Colorado, University of Colorado Anschutz Medical Campus
Richard M. Friesen, MD
Assistant Professor
Children's Hospital Colorado
Erin Englund, PhD
Assistant Research Professor
University of Colorado Anschutz Medical Campus
Takashi Fujiwara, PhD
Postdoctoral fellow
University of Colorado Anschutz Medical Campus, Children's Hospital Colorado
Sungho Park, PhD
Postdoctoral Fellow
Children's Hospital Colorado, University of Colorado Anschutz Medical Campus
Sarah A. Smith
MRI Research Technologist
Children's Hospital Colorado
Emily M. Bucholz, MD, PhD, MPH
Assistant Professor of Pediatrics
Children's Hospital Colorado
Michael V. Zaretsky, MD
Professor
Children's Hospital Colorado
Camila L. Londono Obregon, MD
Associate Professor
University of Colorado
Lorna P. Browne, MD
Professor
University of Colorado and Children's Hospital Colorado
Fetal echo markers of CoA have poor sensitivity and specificity. The aim of this study is to evaluate predictive capability of morphology and blood flow biomarkers in patients with prenatal suspicion of CoA using 4D flow MRI and slice to volume reconstruction (SVR) of T2-weighted black blood (T2w-BB) sequences obtained at 3T.
Methods: 28 patients with suspected CoA by fetal echo were referred for an IRB approved research CMR (34.4 ± 2.6 weeks). Normative reference data was collected in 31 healthy volunteers (33.9 ± 1.6 weeks) using an identical protocol. Cardiac gating was obtained with an MR-compatible Doppler ultrasound device at 3T (smartsync, Northh Medical). Fetal weight was measured from a 3D bFFE sequences to index vessel anatomy and flow measurements.1 4D flow MRI was oriented along a sagittal oblique plane with flows measured at the main pulmonary artery (MPA), ductus arteriosus (DA), ascending aorta (AAo), aortic arch (arch) and descending aorta (DAo). Combined cardiac output was computed (CCO = MPA + AAo). Multiplane T2w-BB sequences underwent SVR to create a 3D dataset.2 Transverse arch diameter (ARCH), distance from left carotid to left subclavian (DAI), and isthmus displacement distance from aortic isthmus to posterior wall of DAo (IDD) were measured (Fig. 1). Suspected CoA cases were classified into true positives and false positives according to postnatal diagnosis. Comparisons between CoA and non-CoA (false positives + volunteers) groups were performed using a Student's t-test with P< 0.05 deemed significant. Receiver operating characteristic (ROC) curves were computed.
Results: 14 patients were found to have true CoA postnatally (CoA+) and were compared to 31 control subjects and 14 false positive CoA (CoA-). Detailed results are shown in Fig. 2. The most significant morphology and flow differences, and a composite of the two were identified by p-values and area under the curve (AUC). Morphology: IDD indexed to fetal weight (IDD-i) was the most effective morphologic difference to distinguish the CoA+ group (P< 0.0001). Flow: AAo-i was the most effective flow feature (P< 0.0001). For both morphology and flow, weight indexing increased the statistical differences between the CoA+ group and the control and CoA- groups. Relative flow ratios: DA:CCO was the most effective flow ratio (P=0.0002). Composite scores of morphology and flow: IDD-i/AAo-i was the most effective feature (P< 0.0001). Predictive capability: The ROCs of the most significant biomarkers are shown in the right column of Fig. 2. Notably, the anatomy and flow composite IDD-i:AAo was the most sensitive and specific imaging biomarker (AUC = 0.996).
Conclusion: This study demonstrates that morphology and flow evaluation of fetal vasculature by 4D flow and 3D reconstruction has potential to predict postnatal diagnosis of CoA with a high degree of sensitivity and specificity.
Figure 1: A: Example of fetal volume segmentation from localizers. B: Morphologic arch measurements. IDD: Distance between the posterior wall of the aortic isthmus and the posterior wall of the descending aorta (isthmus displacement). DAI: Distance between the left common carotid artery and left subclavian artery (distal arch index/DAI). ARCH: Minimum transverse arch diameter. DAo: Diameter of descending aorta. C: 4D flow demonstrating planes of measurement for ascending aorta (AAo), main pulmonary artery (MPA), Arch, ductus arteriosus (DA) and descending aorta (DAo)
Figure 2: Comparison of Morphology, Flow and Composite Morphology/ Flow Measurements. Figure showing indexed morphological measurements, indexed flow measurements, relative flow ratio comparisons and composite morphology and flow scores for CoA+ patients vs Controls and CoA- patients. (CoA+= black triangles, CoA-= green triangles and healthy controls=black circles). The most significant results are outlined by a black box. Receiver operating characteristic (ROC) curves that show the best area under the curve (AUC) in each row are presented on the far right..png)
