Ultrasonography is used routinely during pregnancy to screen and diagnose fetal anomalies. Two-dimensional ultrasound is usually adequate in women at low risk for malformations. When technical factors limit optimal evaluation or a malformation is suspected, further imaging with three-dimensional ultrasound and magnetic resonance imaging is becoming increasingly common. Three-dimensional ultrasound allows the manipulation of data acquired from two-dimensional ultrasound to recreate an infinite number of views, thereby enhancing the ability to evaluate the fetal anatomy. When three-dimensional ultrasound is either unavailable or inadequate, fetal magnetic resonance imaging permits detailed evaluation of the suspected anomaly and assesses the presence of associated anomalies. In this chapter, we review the techniques, advantages, limitations, and clinical applications of these two fetal imaging modalities.
Introduction
Undoubtedly, the most important breakthrough in the field of obstetrics has been the incorporation of two-dimensional ultrasound (2D-US) in clinical practice. Its cost-effectiveness, wide availability and portability, real-time capabilities, and safety for the mother and fetus, has made this technique the main imaging modality of the fetus and its surrounding environment. Indeed, it is difficult to consider the antenatal surveillance of any pregnancy without the use of 2D-US, which provides crucial information for pregnancy management in all three trimesters, including pregnancy diagnosis and location, fetal viability and number, fetal biometry for establishing gestational age and detecting abnormal growth, placental location, amniotic fluid volume, and fetal anatomy. In addition, with the adjunct use of Doppler ultrasound, it allows functional investigation of the utero–placental and intrafetal circulations. Although most of this important clinical information can be obtained by the general obstetrician or sonographer, the evaluation of fetal anatomy for detecting congenital anomalies has evolved significantly over the past 2 decades, and currently requires considerable knowledge, expertise, and training to carry out.
During the 1970s and 1980s, a plethora of studies provided robust evidence that many congenital malformations were able to be diagnosed prenatally using ultrasound. Most of these malformations, however, were detected incidentally, mainly among high-risk women. Subsequently, as a result of the improvement of ultrasound equipment and its image quality, more sophisticated ultrasound technique, and wider availability of dedicated operators, it soon became possible to use 2D-US in the first and second trimesters as a genetic screening test for low-risk pregnancies. It was not until the late 1990s, however, that it became widely accepted that all pregnant women should undergo one or two 2D-US examinations at specific gestational ages to detect congenital anomalies.
Despite the significant advancements that 2D-US has contributed to the area of prenatal care, it has inherent constrains in depicting the fetal anatomy. Specifically, visualisation can be limited by certain factors, including maternal habitus, abdominal scarring, the size and position of the fetus, the progressive ossification of the fetal skeleton producing acoustic shadowing, and oligohydramnios. In addition, the interpretation of certain ultrasound findings may be difficult even for the skilled operator, and the inability to carry out alternate analysis, such as volume reconstruction and obtaining additional views from already-captured 2D-US images, can considerably limit the potential of this technique in obstetrical practice. Although many of these factors may not have a significant effect when examining a structurally normal fetus, the need for more accurate, versatile, and high-definition technologies that can overcome some of these technical factors can be crucial when the maternal–fetal medicine specialist faces a pregnancy complicated with a fetal malformation.
The relatively recent incorporation of three-dimensional ultrasound (3D-US) and magnetic resonance imaging (MRI) in the field of obstetrics has overcome many of the limitations of 2D-US for the anatomic evaluation of the fetus, especially when a fetal malformation is suspected. In this chapter, we aim to review advantages, limitations, and clinical applications of these two fetal imaging modalities as adjunctive techniques to 2D-US to assess the normal and abnormal fetal anatomy. Their role in depicting selected structural congenital anomalies is described in chapter 7 in this issue of Best Practice and Research Clinical Obstetrics and Gynaecology .
Three-dimensional ultrasound
General considerations
Three-dimensional ultrasound is an expansion of 2D-US in that it consists of the acquisition of multiple bi-dimensional parallel sections by a specially designed volumetric transducer. Although a free-hand technique was used for most volumetric transducers during the initial stages of development of 3D-US, modern equipment is now equipped with mechanical transducers, which are able to generate high-resolution images because all sections are obtained in slices of similar width. Using sophisticated software programmes, the digital information acquired is integrated and stored as voxels (basic volumetric picture elements), instead of the conventional planar data, or pixels, used in 2D-US. Subsequently, the volume dataset can be further reformatted using different rendering algorithms, either at the bedside during the actual scan or offline after downloading the information onto a computer workstation to obtain further sections and to digitally navigate within the volume. Although most of the information generated by 3D-US is transformed into a solid, static block of data, rapid computer processors allow the examination of the target organ during fetal motion, which is called real-time 3D-US or four-dimensional ultrasound.
Although fetal imaging using 3D-US was first described more than 20 years ago, its use in the obstetric population became commonplace only after developments in computer-image processing. As its main components are obtained from bi-dimensional data, 3D-US has the same advantages and limitations as 2D-US, including the major advantage of being non-invasive and safe throughout pregnancy. In addition, as 3D-US allows the acquisition and storage of volumetric information, evaluation of any section within the acquired volume dataset, an infinite number of views, is possible. This digital information can be processed offline and even transmitted electronically for remote consultation. Main disadvantages of 3D-US include the greater expense of this technology compared with 2D-US. In addition, it is highly dependent on operator skill and the quality of the original volume as obtained by 2D-US. It is also dependent upon the absence of fetal movements at the time of dataset acquisition given that motion artifacts considerably reduce its performance in depicting the fetal anatomy. In addition, the entire targeted organ may not be able to be captured in the area of view and produce pseudo-amputations and filling artifacts that may cause concern to the parents.
In cases of suspected fetal malformations, 3D-US, where available, is a complementary imaging modality that can aid significantly in further evaluation and diagnosis. As the data used to carry out 3D-US are gathered during real-time 2D-US, a complete and well-executed 2D-US remains crucial. Nevertheless, the capability of examining fetal structures with 2D-US in a real-time fashion is a great asset that cannot be reproduced by 3D-US. The latter technique, on the other hand, allows reconstruction of high-resolution images of the surface of the fetus, particularly the face and limbs ( Figs. 1 and 2 ). Anecdotally, such images are highly desired and appreciated by parents, although the utility of 3D-US in improving the pregnancy experience and maternal–fetal bonding has not been clearly established.
Technique
The basic 3D-US protocol used to scan pregnant women is exactly the same as the 2D-US protocol, although the use of dedicated volumetric transabdominal or transvaginal probes is essential. The transvaginal route performs better than the transabdominal route in the first trimester ( Fig. 3 ) as well as in the second and third trimesters in specific circumstances, such as 3D-US imaging of the central nervous system. The technique includes identification of the target organ with 2D-US, the selection of the region of interest with a box on the screen, the capture of the volume dataset with a pre-determined sweep of variable angle and width, and the storage of such information for further analysis. To avoid contamination with motion artifacts, it is recommended to capture the volume while the fetus is still and the woman is holding her breath. Steadying the fetus with the sonographer’s free hand and the use of a pedal for triggering the acquisition is also recommended. Fetal motion artifacts, however, are becoming less of a significant problem owing to the use of mechanical probes with faster acquisition times, usually in the range of 2–4 s.
Once the dataset is captured and stored, the basic information is automatically displayed in a multiplanar mode, in which all three orthogonal views are seen simultaneously on the screen ( Figs. 4 and 5 ). The rendered image can also be displayed on the same screen or enlarged onto a separate full screen. This allows the visualisation of the organ or segment to be analysed from different perspectives that are all perpendicular to each other. Manipulation of the volume using a reference dot, which is moved through the volume by using the X-, Y-, and Z-buttons, allows pinpointing any desired area within the volume dataset on the screen and further digital navigation throughout the region of interest in a ‘tomographic fashion’ in the other two planes. The selected dataset can be further processed using different rendering algorithms, including surface, maximum, and inverse modes for enhancing the surface, the skeleton, and low-signal areas of the fetus, respectively. If the visualisation of the region of interest is obstructed by fetal parts, umbilical cord, placenta, or uterine wall, the redundant solid tissue can be removed by rotating the volume and carrying out an anatomic digital dissection with an image processor — the magic cut mode. Another useful tool is the multislice view, in which hundreds of consecutive parallel views are automatically displayed and navigated in successive pages at any desired thickness and angle to study the anatomical relation within any given structure ( Fig. 6 ). The application of this technique in the evaluation of the fetal face and palate is shown in Figs. 7 and 8 .
Safety issues
Current evidence and more than 40 years of widespread clinical use suggest that obstetric 2D-US is safe for the mother and the fetus. Theoretically, the collection of 2D-US images to form the 3D-US volume does not require additional energy that could pose harm to the fetus. Nevertheless, because of the possibility of potential risks, it is recommended that the use of multiple scans, particularly during the first trimester, should be restricted to those medically indicated, and prolonged time during examinations should be avoided. On the other hand, 3D-US has the potential to decrease the time of examination considerably, as most of the fetal organs can be captured with one sweep and further studied offline, but its role in shortening the scanning time remains to be determined. Future developments in 3D-US automation algorithms will likely lead to improved efficiency in visualising fetal structures, carrying out biometry, and calculating structural volumes, resulting in a reduction of the scanning time. The elective use of 3D-US, particularly for the sole purpose of obtaining a picture of the fetal face, has been strongly discouraged.
Three-dimensional ultrasound
General considerations
Three-dimensional ultrasound is an expansion of 2D-US in that it consists of the acquisition of multiple bi-dimensional parallel sections by a specially designed volumetric transducer. Although a free-hand technique was used for most volumetric transducers during the initial stages of development of 3D-US, modern equipment is now equipped with mechanical transducers, which are able to generate high-resolution images because all sections are obtained in slices of similar width. Using sophisticated software programmes, the digital information acquired is integrated and stored as voxels (basic volumetric picture elements), instead of the conventional planar data, or pixels, used in 2D-US. Subsequently, the volume dataset can be further reformatted using different rendering algorithms, either at the bedside during the actual scan or offline after downloading the information onto a computer workstation to obtain further sections and to digitally navigate within the volume. Although most of the information generated by 3D-US is transformed into a solid, static block of data, rapid computer processors allow the examination of the target organ during fetal motion, which is called real-time 3D-US or four-dimensional ultrasound.
Although fetal imaging using 3D-US was first described more than 20 years ago, its use in the obstetric population became commonplace only after developments in computer-image processing. As its main components are obtained from bi-dimensional data, 3D-US has the same advantages and limitations as 2D-US, including the major advantage of being non-invasive and safe throughout pregnancy. In addition, as 3D-US allows the acquisition and storage of volumetric information, evaluation of any section within the acquired volume dataset, an infinite number of views, is possible. This digital information can be processed offline and even transmitted electronically for remote consultation. Main disadvantages of 3D-US include the greater expense of this technology compared with 2D-US. In addition, it is highly dependent on operator skill and the quality of the original volume as obtained by 2D-US. It is also dependent upon the absence of fetal movements at the time of dataset acquisition given that motion artifacts considerably reduce its performance in depicting the fetal anatomy. In addition, the entire targeted organ may not be able to be captured in the area of view and produce pseudo-amputations and filling artifacts that may cause concern to the parents.
In cases of suspected fetal malformations, 3D-US, where available, is a complementary imaging modality that can aid significantly in further evaluation and diagnosis. As the data used to carry out 3D-US are gathered during real-time 2D-US, a complete and well-executed 2D-US remains crucial. Nevertheless, the capability of examining fetal structures with 2D-US in a real-time fashion is a great asset that cannot be reproduced by 3D-US. The latter technique, on the other hand, allows reconstruction of high-resolution images of the surface of the fetus, particularly the face and limbs ( Figs. 1 and 2 ). Anecdotally, such images are highly desired and appreciated by parents, although the utility of 3D-US in improving the pregnancy experience and maternal–fetal bonding has not been clearly established.
Technique
The basic 3D-US protocol used to scan pregnant women is exactly the same as the 2D-US protocol, although the use of dedicated volumetric transabdominal or transvaginal probes is essential. The transvaginal route performs better than the transabdominal route in the first trimester ( Fig. 3 ) as well as in the second and third trimesters in specific circumstances, such as 3D-US imaging of the central nervous system. The technique includes identification of the target organ with 2D-US, the selection of the region of interest with a box on the screen, the capture of the volume dataset with a pre-determined sweep of variable angle and width, and the storage of such information for further analysis. To avoid contamination with motion artifacts, it is recommended to capture the volume while the fetus is still and the woman is holding her breath. Steadying the fetus with the sonographer’s free hand and the use of a pedal for triggering the acquisition is also recommended. Fetal motion artifacts, however, are becoming less of a significant problem owing to the use of mechanical probes with faster acquisition times, usually in the range of 2–4 s.
Once the dataset is captured and stored, the basic information is automatically displayed in a multiplanar mode, in which all three orthogonal views are seen simultaneously on the screen ( Figs. 4 and 5 ). The rendered image can also be displayed on the same screen or enlarged onto a separate full screen. This allows the visualisation of the organ or segment to be analysed from different perspectives that are all perpendicular to each other. Manipulation of the volume using a reference dot, which is moved through the volume by using the X-, Y-, and Z-buttons, allows pinpointing any desired area within the volume dataset on the screen and further digital navigation throughout the region of interest in a ‘tomographic fashion’ in the other two planes. The selected dataset can be further processed using different rendering algorithms, including surface, maximum, and inverse modes for enhancing the surface, the skeleton, and low-signal areas of the fetus, respectively. If the visualisation of the region of interest is obstructed by fetal parts, umbilical cord, placenta, or uterine wall, the redundant solid tissue can be removed by rotating the volume and carrying out an anatomic digital dissection with an image processor — the magic cut mode. Another useful tool is the multislice view, in which hundreds of consecutive parallel views are automatically displayed and navigated in successive pages at any desired thickness and angle to study the anatomical relation within any given structure ( Fig. 6 ). The application of this technique in the evaluation of the fetal face and palate is shown in Figs. 7 and 8 .
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
