Visualization of the corpus callosum by fetal ultrasound may be problematic. Due to its arched shape, it cannot be demonstrated using standard axial planes, while coronal planes enable visualization onscreen of only a small portion at a time. Therefore, sagittal midline views should be obtained using either the transvaginal or the transfundal approach1. In coronal and sagittal planes, the corpus callosum is anechogenic or hypoechogenic and the presence of hyperechogenicity signifies possible pathology, mainly callosal lipoma.
In a recent issue of the Journal, the use of three-dimensional (3D) reconstruction of sagittal planes was proposed as an alterative method for the study of brain anatomy and particularly of midline structures2, 3. Surprisingly, in the paper of Correa et al.2 and also in lectures presented by leading authorities in the field of 3D ultrasound, 3D reconstructed images depicted the corpus callosum as a hyperechogenic structure; Pilu et al.3 showed similar images but recognized that in 3D sagittal images the corpus callosum cannot been differentiated clearly from the cavum septi pellucidi.
Since lipomas of the corpus callosum are rare, the possibility of an artifact causing the hyperechogenicity should be considered. This type of artifact may be due to the simultaneous use of 3D pixel reconstruction in conjunction with SRI (speckle reduction image). The structure observed represents the interface between the cingulate gyrus, the cingulate sulcus, cerebrospinal fluid and the blood flow in the callosal arteries (Figure 1). A curvilinear lipoma shows similar characteristics on 3D imaging (Figure 2) but usually does not interfere with the 2D visualization of the corpus callosum (Figure 3).
This issue may also be important in the diagnosis of vermian anomalies due to similarities between the structure of the vermis and the cerebellar hemispheres.
Great caution is required before the widespread introduction of any new technique, and at present we suggest that critical decisions regarding the diagnosis of pathological midline anomalies should not be made based on 3D reconstructions.