Introduction: Cardiac MRI (CMR) is an excellent modality for cardiac anatomical and functional evaluation, but typically requires ECG gating. This presents a major problem in fetal CMR. The purpose of the study was to evaluate the technical feasibility, image quality and scan length of a method for fetal CMR and compare with matching fetal echocardiography data.
Patients and Methods: Fetal CMR was performed in 20 cases using 2 techniques: 1) Using a prototype sequence on a 3T Prisma Siemens MRI scanner and a retrospective self-gating technique (11 cases) 2) Using a device that can detect the fetal heart contraction using integrated simultaneous ultrasound on a 1.5 T Philips Scanner (9 cases).
Scans were performed between 29 and 34 weeks of pregnancy. All fetuses had a known echocardiographic diagnosis, serving as the standard of reference. Following the scan, reconstruction and analysis were performed, blinded to the echocardiographic data.
In the first group fetal cardiac pathology was present in 10/11 scans (including TGA, VSD, Pulmonic atresia, discrepancy in ventricular volumes). 1/11 scans was referred following CMV infection. All were diagnostic studies.
In the second group 8/9 case were diagnostic.
Results and Discussion: Average echocardiography and CMR scan length were 15 and 20 minutes, respectively. Images quality was high in 18/20 cases. The diagnosis achieved by fetal CMR was similar to echocardiographic diagnosis in 17/19 cases. Discrepancies were related to fetal position and maternal habitus. Both ventricles, the atria and the pulmonary veins are well demonstrated.
Conclusion: Although challenging, we demonstrate that fetal CMR was able to achieve comparable diagnostic results with fetal echocardiography. Yet, in order to obtain fully comprehensive fetal CMR scans, further improvements are necessary. Fetal CMR is a new, promising technique which may be incorporated in the future in the evaluation of complex fetal cardiac pathologies.