Combining fetal CMR derived hemodynamic and anatomical risk stratification tools significantly improves prenatal prediction of coarctation of the aorta

Background: Prenatal diagnosis of coarctation of the aorta (CoA) remains challenging and widespread use of risk stratification tools is still lacking. This study aims to assess the added clinical value of fetal cardiovascular magnetic resonance (CMR) in improving prenatal prediction of CoA.

Methods: Based on a literature review conducted in August 2023 we prospectively characterised three risk groups: low, moderate and high risk for CoA(Figure 1a). CMR protocol: 10 blackBlood stacks in 5 different directions were acquired at 1.5Tesla(Siemens Magnetom Sola) and processed using slice to volume(SVR) reconstruction(Northh Medical). Standard viewing software was used for anatomical measurements. 2D flow imaging of the mitral valve(MV) and the ascending aorta(AA) were analysed using suiteHEART® software(NeoSoft). The prospectively designed reporting template is publicly available(5). The CMR reporter was blinded to echocardiographic results and CMR risk stratification was completed prior to delivery. BMI SPSS V30 was used for statistical analysis.

Results: 21 fetuses with prenatal suspicion of CoA on echocardiogram were prospectively enrolled between 11/2023 to 06/2025. 9/21 fetuses underwent surgical neonatal CoA repair, 1 fetus has continued surveillance for a hypoplastic aortic arch(HAA).


SVR and MV flow were diagnostic in all 21 fetuses. AA flow was only diagnostic in 24% (5/21). Therefore, MV inflow was used as surrogate for aortic outflow for the hemodynamic risk stratification score.


The CoA group had less MV inflow(p< 0.05), a smaller MV annulus(p< 0.05) and increased isthmal displacement(p< 0.05).


On echocardiographic assessment 5/6 fetuses in the high risk group had postnatal CoA, 4/10 in the moderate and 0/5 in the low risk group(chi quare p=0.046).


Anatomical CMR derived assessment did not differentiate well between the groups as individual markers or combined risk score(p=0.056).


On haemodynamic fetal CMR assessment 5/5 fetuses in the high risk group had postnatal CoA, 4/8 fetuses in the moderate and 0/8 in the low risk group(p=0.002).


Post-hoc combined hemodynamic and anatomical parameters further improved risk stratification with 9/10 fetuses in the high risk group having postnatal CoA (1/10 HAA surveillance), 0/3 in the moderate and 0/8 in the low risk group(p< 0.001).

Conclusion: MV inflow is easier to obtain in fetal CMR than AA flow and can be substituted for cardiac index calculation.


Both echocardiographic and hemodynamic CMR risk assessment correctly identified a low risk group without any missed CoA diagnoses. Echocardiographic assessment had a large moderate risk group with almost 40% risk of CoA. Hemodynamic CMR assessment increased the number of low risk fetuses but did not significantly alter size and risk of the moderate risk group. When combining anatomical and hemodynamic CMR assessment, high and low risk fetuses are correctly identified and additionally the moderate risk group is significantly decreased in size providing increased prenatal diagnostic certainty.

Fetal parameters and measurement compared between fetuses with postnatally confirmed CoA and those with postnatally normal aortic arch

	False positive cases		Coarctation cases
	mean	SD	mean	SD	p value
fetal weight in kg	2.12	0.23	2.17	0.28	0.67
Mitral valve inflow in mls	2.93	0.91	1.81	0.92	0.01
Tricuspid valve inflow mls	6.40	1.61	5.14	2.04	0.15
Systemic cardiac index in mls/kg/min	188	45	122	63	0.02
Mitral valve in mm	9.00	1.40	7.29	1.24	0.01
LV length in mm	22.61	3.79	19.64	3.52	0.08
Ascending aorta in mm	5.68	1.83	5.30	1.17	0.65
Transverse arch in mm	3.49	0.68	2.91	0.81	0.10
Isthmus in mm	3.05	0.78	2.87	0.98	0.66
Isthmal displacement in mm	1.88	0.93	4.88	2.13	0.01
Descending aorta in mm	5.45	1.27	4.92	0.86	0.27
Ductal arch in mm	5.60	1.29	5.21	0.84	0.41
Tricuspid valve in mm	10.38	1.84	9.19	1.42	0.11
RV length in mm	23.13	3.87	23.08	4.16	0.98
TV/MV ratio	1.17	0.19	1.29	0.15	0.11
Transverse arch/descending aorta ratio	0.67	0.14	0.61	0.18	0.39
Transverse arch/ductal arch ratio	0.67	0.21	0.55	0.11	0.12
MV inflow indexed to body weight in mls/kg	1.37	0.35	0.85	0.46	0.01
TV/MV inflow ratio	2.30	0.61	3.60	2.06	0.10
Transverse arch indexed to body weight in mm/kg	1.61	0.37	1.31	0.39	0.09

MV = mitral valve; TV = tricuspid valve; LV = left ventricle; RV = right ventricleFigure 1a depicts how the different risk scoring differentiates between false positive and true coarctation of the aorta (CoA) cases. Figure 2 B depicts the different risk scoring for low moderate and high risk group
Figure 2 depicts the 3D reconstruction of blackBlood imaging in a case with postnatal COA and one with postnatally normal aortic arch