Non-invasively measured pressure drops are higher in neonates with coarctation of the aorta (CoA) compared to reference controls (RF_FR_418)

Coarctation of the aorta (CoA) remains difficult to diagnose even in the neonatal period. Advanced image processing methods including the virtual work-energy relative pressure (vWERP) method have recently been proposed and validated to quantify intravascular pressure drops [1], using non-invasively acquired 4D Flow cardiovascular magnetic resonance (CMR) as the only input.

The aim of this study was to quantify the cross-coarctation relative pressure using vWERP and 4D Flow CMR in neonates with antenatal suspicion of CoA.

Methods:

A non-sedated non-contrast neonatal feed and wrap CMR was performed on a 3T MRI scanner (MAGNETOM Prisma, Siemens Healthineers, Forchheim, Germany) in 30 neonates with an antenatal suspicion of CoA, using a compressed-sensing 4D Flow research sequence. The analysis was performed using a Matlab based research 4DFlow CMR tool [2]. The aorta was manually segmented and a tetrahedral mesh was generated to allow interpolation of 4D Flow CMR velocities. To derive cross-coarctation relative pressure, the inlet starting plane was placed at the point prior the first bifurcation in the ascending aorta, and the outlet plane at the descending aorta passed the coarctation (Figure 1). Implemented within the aforementioned image processing tool, vWERP was then used to quantify pressure drops over the cardiac cycle. Patients were divided into three groups; Group 1: patients who postnatally did not have CoA; Group 2: patients who postnatally had a hypoplastic aortic arch only, and Group 3: patients who had a duct dependent coarctation requiring neonatal surgery.   From the derived pressure drops, maximum and minimum pressure drops as well as the relative pulse pressure (rPP) was derived. ANOVA was used to compare the vWERP values between groups.

Results:

3/30 neonates were excluded: 1 due to other complex congenital heart diseases (CHD), 1 due to severe CoA and no flow in the descending aorta, and 1 due to insufficient image quality. From the included 27 neonates, 10 patients were in group 1, 8 patients in group 2, and 9 patients in group 3. (Median age at MRI scan was 3.00 days, IQR [1-6 days]); 55.6% Females (N=15) vs 44.4% Males (N=12). The maximum pressure drop showed statistically significant difference between group 1 (non-CoA) and 3 (CoA) (Mean maximum dP 4.6 ±  1.5 mmHg vs 13.3 ± 3.2 mmHg, p< 0.001) and between group 2 (hypoplastic arch) and 3 (CoA) (Mean maximum dP 6.8 ± 2.4 mmHg vs 13.3 ± 3.2 mmHg, p< 0.001, but no significant differences were seen between group 1 and 2. The rPP was significantly higher in group 3 vs group 1 (Mean rPP 17.62 ± 4.00 mmHg vs 7.23 ± 1.82 mmHg, p< 0.001), and in group 3 vs group 2 (Mean rPP 17.62 ± 4.00 mmHg vs Mean rPP 10.43 ± 2.73 mmHg), but group 1 and 2 were not significantly different.

Conclusion:

Using 4D Flow CMR and image processing tools such as vWERP, we quantified intravascular pressure drops in neonates with prenatal suspicion of CoA. Neonates with confirmed CoA showed significantly higher pressure drops compared with neonates who have no CoA or a hypo-plastic aortic arch only.

Figure (1): Shows the analysis of virtual work-energy relative pressure (vWERP) in a CoA patient using the Matlab based research 4DFlow CMR tool [2]. After manually segmenting the aorta, the mesh is generated (A), and then the velocity from the 4DFlow CMR data is interpolated in each node of the tetrahedral mesh (B), then the inlet plane was placed at the point prior the first bifurcation in the ascending aorta (C) which can be seen from the angiography image of the phase contrast in Matlab 4DFlow tool, and the outlet plane at the descending aorta passed the coarctation (D), then the vWERP is computed and analysed (vWERP maximum, vWERP minimum and the relative pulse pressure (rPP) (E)

Figure (2): box plots showing the Mean ± SD of maximum pressure drops using virtual work-energy relative pressure (vWERP) method in the three patient groups: Group 1 (non-CoA): patients who postnatally did not have CoA; Group 2 (hypoplastic arch): patients who postnatally had a hypoplastic aortic arch only, and group 3 (CoA): patients who had a duct dependent coarctation requiring neonatal surgery.

Figure (3): box plots showing the Mean ± SD of relative pulse pressures (rPP) using virtual work-energy relative pressure (vWERP) method in the three patient groups: group 1 (non-CoA): patients who postnatally did not have CoA; group 2 (hypoplastic arch): patients who postnatally had a hypoplastic aortic arch only, and group 3 (CoA): patients who had a duct dependent coarctation requiring neonatal surgery.