Double-outlet right ventricle (DORV) is a conotruncal anomaly which occurs in one out of 10 000 live births. It is defined as a condition in which both great arteries (pulmonary artery and aorta) arise primarily from the morphologically right ventricle. Although conventional prenatal echocardiography provides a relatively high degree of accuracy in the diagnosis of congenital heart disease, it is difficult to depict the spatial relationship of the cardiovascular structures, especially in cases of DORV1, 2.
The inversion mode is a novel volume analysis approach to three-dimensional (3D) and four-dimensional (4D) ultrasonography for visualization of fluid-filled fetal structures. Similar to the postmortem pathologic casts, this modality has been proposed as a technique capable of providing ‘digital casts’ of the fetal cardiac chambers and great vessels3–7. The feasibility of the inversion mode in the antenatal detection of congenital heart disease has been described in several reports, but in only one of these was DORV involved5. We present here a case of DORV diagnosed prenatally by 3D and 4D volume-rendered imaging using the inversion mode.
A 25-year-old multiparous woman was referred to our unit for detailed fetal ultrasound examination because of suspected cardiac abnormalities at 27 weeks of gestation. Conventional two-dimensional (2D) echocardiography and color Doppler flow mapping revealed two parallel great vessels arising from the right ventricle in the outflow-tract view (Figure 1) and the diagnosis of DORV was made. In order to further clarify the anatomical correlation of this anomaly, the inversion mode was used. The fetal heart volume was acquired using 4D ultrasonography and the spatio-temporal image correlation (STIC) technique (Voluson 730 Expert, General Electric Medical Systems, Kretztechnik, Zipf, Austria). Using a curved array transducer (2–5-MHz), the sweep took 10 to 12.5 seconds at an angle of 20 to 25° at the level of the four-chamber view. 3D reconstruction was carried out using the ‘inversion’ rendering algorithm. Low threshold and transparency levels were adjusted until the structures of interest were visualized. The volume data were analyzed with 4DView 2000 version 2.1 computer software (General Electric Medical Systems, Kretztechnik). By transforming echolucent structures into solid voxels, the heart chambers and the lumen of the great vessels were converted into hyperechogenic 3D structures. Both the pulmonary artery and aorta were seen to arise in parallel from the right ventricle (Figure 2 and Videoclip S1), whereas these two great vessels cross in a normal heart (Figure 3, from a normal fetus at 25 weeks of gestation). Based on these findings, the diagnosis of DORV was confirmed. A neonatal echocardiographic examination after birth was consistent with the prenatal findings.
Many new developments in fetal echocardiography have led to greater understanding of the complex cardiac morphology and to more accurate prenatal diagnosis. The inversion mode, or ‘negative surface display’, is a novel volume analysis tool which displays fluid-filled structures as solid objects8, 9. With this rendering algorithm, a structure can be inspected from all directions. As shown in the present report, the spatial relationship between the great vessels and the right ventricle visualized in the inversion mode image was more readily discernible than that obtained from conventional ultrasonography. In addition to furnishing anatomic information, the inversion mode facilitates volumetric measurement of the cardiac cavities because of the more easily distinguishable boundaries between the cavities9.
Compared to other 3D and 4D techniques, the inversion mode has been shown to be superior to power Doppler in image quality, but it provides less information about neighboring tissue than does the glass body mode8. It can be used to display any fluid-containing structure, but it lacks information on velocity or direction of blood flow. Moreover, fetal movements may affect rendering and artifacts can be encountered. More studies on the comparative performance and more efforts to overcome the limitations of these new methods in prenatal diagnosis are needed.