Ultrasound Evaluation of Normal Fetal Anatomy

Summary of Key Points

Continued advancement of ultrasound technology including increase in frequency and choice of focal position have improved visualization of fetal anatomy and therefore have also increased the required anatomic knowledge of these structures for those performing and interpreting fetal sonograms.
Superficial anatomy of the fetus can be delineated with either two-dimensional or three-dimensional sonography and can be useful in confirming normal formation of facial structures and genitalia.
Fetal skeletal structures are the earliest recognizable fetal features, and their sonographic evaluation is highly dependent on the position of the fetus.
Identification of a normal three-vessel umbilical cord and normal portal venous anatomy is useful in determining whether cardiovascular development is normal.
The liver is the largest gastrointestinal organ and is the largest contributor to the size of the abdominal circumference.
The lungs become more echogenic later in pregnancy; however, the change is not related to lung maturity.
Kidneys grow throughout pregnancy, with a ratio of kidney circumference to abdominal circumference of 0.27 to 0.30.
The calvaria can obscure the portion of the brain nearest the transducer. Bone-free windows through the fontanelles can be used to increase visualization, and when one hemisphere is not well seen, symmetry is assumed unless images prove otherwise.

Our understanding of normal fetal anatomy as seen on sonograms continues to evolve. Instrumentation has improved steadily, yielding both improved and more consistent image quality. Among the most significant advances in fetal imaging has been the ability to choose the depth of the zone of best focus of the ultrasonic beam and to select frequency without changing transducers. With these capabilities, the area of anatomy being observed can be consistently inspected with the focused portion of the beam at an optimal frequency, both highly significant advantages. Recent technologic advances allow imaging at higher frequencies than a mere decade ago.

In addition to these technologic advances, sonologists have gradually improved their understanding of the anatomy portrayed on in utero sonograms. Unquestionably, clearer images have led the way to our improved understanding, but other factors have been involved; not the least of these has been the surge in ultrasonic imaging of premature neonates. These tiny neonates are the equivalent of second trimester fetuses as early, at times, as 24 weeks' gestation. Visualization of their brains ( Fig. 8-1 ) and abdominal ( Fig. 8-2 ) anatomy in the ex utero environment, which permits the use of higher frequency transducers, a greater selection of planes of section, and comparison with other imaging modalities, has done much to improve our understanding of fetal anatomy. This information can be, to a large extent, extrapolated to younger fetuses. As well, as magnetic resonance imaging (MRI) continues to be a growth area in fetal imaging, our ability to compare sonographic anatomy of the fetus with that portrayed on MR imaging further enhances our understanding of fetal sonographic anatomy ( Fig. 8-3 ).

The ability of sonography to detect intrafetal structures is a balance between spatial resolution and contrast. This balance, however, strongly favors contrast as the more important aspect of perception. For example, a large white dot on a white wall is difficult or impossible to see because no contrast differential exists even though the eye can spatially resolve easily a tiny black dot (high contrast) on the same wall. Structures possessing high levels of subject contrast can be consistently detected at a smaller size (often equating to an earlier age) than those displaying poor contrast. Although sonographic contrast agents are available for imaging in adults and children, current agents do not cross the placenta in sufficient concentration to affect fetal organs. Therefore, sonologists possess no agents to alter contrast of fetal organs and thus are totally dependent on subject contrast (inherent contrast) for visualization of internal fetal morphologic details. Clearly, spatial resolution is also a critical feature in defining morphologic structure but has not been the limiting factor in demonstrating fetal anatomy. Fortunately, sophisticated modern sonographic imaging systems have the ability to choose varying contrast settings, which enhance inherent tissue differences. An important aspect of the technical advancements allowing ever higher frequency imaging is that at higher frequencies, contrast differences between organs change, which can have diagnostic impact.

Another technology that has improved fetal imaging, particularly for earlier gestations, is the intracavitary (transvaginal) transducer. Certain aspects of fetal and even embryonic, anatomy can be seen with amazing detail ( Fig. 8-4 ). Although this methodology is crucial to modern imaging, we intend to concentrate our discussion on anatomy visualized by transabdominal transducers in fetuses beyond the 14th week. When examining later pregnancies, in the context of fetal anatomic imaging, we use transvaginal transducers predominantly to visualize the presenting part when transabdominal transducers fail to adequately image this region of the fetal anatomy (e.g., distal sacrum in a breech fetus).

Other parameters, important in fetal imaging, also cannot be controlled. Sonography is a tomographic technique. Appropriate positioning for obtaining the best tomographic plane is always desirable. However, we are unable to control fetal position to attain this end. We also cannot control maternal body habitus or the amount of amniotic fluid, both of which may dramatically alter our ability to discern fetal anatomy. Despite these problems, a large number of fetal structures are consistently visible sonographically.

High-resolution, real-time scanners with their flexible approach to imaging are mandatory for modern fetal sonography. In the following sections, various aspects of fetal anatomy are detailed as seen on such instrumentation. An estimate is made of the ability of ultrasound instrumentation to consistently demonstrate the anatomic part under consideration as well as an attempt to estimate when the fetus has attained sufficient size such that the anatomic structure is large enough to be detected. It is important to recall that size and visualization may be relative at any given stage of development. For example, in a small fetus with a well-distended urinary bladder, identification of the bladder is relatively easy. Alternatively, identification of the bladder will be difficult in a term fetus that has recently voided. The urine, in this instance, provides the “contrast” that ordinarily makes the urinary bladder an easy structure to perceive. If this contrast agent drains away, the size of the fetus (and thus its bladder) will not rescue one from the loss of contrast. Also important is the concept that the human eye sees best in the “relative” rather than the “absolute” sense of size. Thus, in a young fetus, the cerebral ventricle is much more readily seen than in an older fetus because the relative size of the ventricle compared with overall brain size is larger early (even though the absolute size is larger later).

If the sonographer begins with a specific intent to image a particular fetal part, it is frequently possible to succeed. To accomplish this end, the sonographer must (1) assess the precise fetal position; (2) consider whether the anatomic part of interest is best visualized in planes perpendicular to the fetal long axis or parallel to the fetal long axis; and (3) adjust acoustic imaging parameters, particularly time-gain compensation, and transducer angulation to visualize the area to best advantage. Obviously, such rules are the same throughout all of sonography. The challenge of imaging intrafetal structures is to apply these rules when fetal position is changing such that the current scanning plane is no longer applicable for the part one wishes to visualize.

The flexibility offered by real-time sonographic systems enables one to survey the fetus quickly to determine precise position. Second, the sonographic tomograms, which are rapidly generated (virtually “real-time” imaging), enable one to view a large volume of the fetus with closely spaced sections. Such a rapid look at many contiguous tomograms eliminates one of the basic flaws of tomographic imaging of a moving target. Finally, fetal movements are viewed directly, which enables one to quickly reorient the transducer to the optimal plane of section to image the structure of interest. As digital image storage and viewing continues to expand, cine clip technology has greatly improved the capture and viewing of many fetal parts, especially those that are in motion, like the heart. Surprisingly, many imaging groups fail to take advantage of cine technology when recording fetal examinations. This immensely powerful technology should not be left on the sidelines by imaging specialists.

Currently, three-dimensional imaging systems have reached a clinically relevant stage of development. These instruments depend on volume imaging . That is, a volume of tissue within the fetus is insonated and the data are gathered in the processing computer. From this volume, three-dimensional images are generated. Of perhaps greater importance in the future, individual planes of section in virtually any orientation can be computed. When and if it becomes possible to generate equivalent planar images to those obtained with a hand-held transducer, the current method of “fluoroscopic” sonography, the entire nature of sonography will change. The technical challenges of sonography will be greatly reduced ( Fig. 8-5 ).

Fetal parts of interest to the sonologist fall into three major categories of subject contrast that subsequently determine the relative ease with which the structure is sonographically visible: (1) structures that generate high-amplitude reflections (e.g., ossified bones, submucosa of fetal bowel, leptomeninges); (2) structures that generate no internal echoes (e.g., fluid-containing viscera); and (3) structures that generate midrange gray echoes (e.g., the parenchymal organs such as the lungs, brain, spleen, liver, kidneys, and muscles). The categories are listed from most visible to least visible. Within the last category, one may anticipate seeing a spectrum of gray shades that will enable distinction between parenchymal organs and intraorgan components. For example, the medullary portions of the fetal renal parenchyma generate lower amplitude internal echoes than do the surrounding cortical tissues and Bertin septa, thus enabling recognition of this separate component of renal tissue ( Fig. 8-6 ). Modern imaging systems enable distinctions previously not possible. An example from a relatively common pathologic circumstance is the apposition of the herniated spleen against the lung in left congenital diaphragmatic hernia. Previously these organs “blended” together, but currently they are readily discriminated, resulting in better estimations of residual lung volume in affected fetuses ( Fig. 8-7 ).

A feature of critical importance for organ imaging is the fetal position. A prone fetus is in an optimal position for imaging the kidneys, ordinarily difficult to perceive, but in a poor position to demonstrate the urinary bladder, which is usually easy to image. Determination of fetal position should be accomplished in all obstetric sonographic examinations from the second trimester onward. The fetal position should be determined as precisely as possible before an interpretation of fetal anatomy is begun, because the position of a structure will often influence our interpretation. The general fetal orientation is first assessed (e.g., cephalic, breech, oblique, or transverse). Once this is determined, the location of the fetal spine is noted. If, for example, the fetal spine is on the maternal left side and the fetus is in a cephalic presentation, one can judge that the fetus is lying on its left side ( Fig. 8-8 ). Conversely, if the fetus is breech, then it must be lying on its right side. The reverse is the case for breech and cephalic fetuses when the fetal spine lies on the maternal right side. In the transverse or oblique fetal positions the same rules apply but with a different orientation.

Such an analysis of fetal position is vital for proper interpretation of abdominal and thoracic situs (now a requirement of the AIUM/ACR/ACOG [American Institute of Ultrasound in Medicine/American College of Radiology / American College of Obstetricians and Gynecologists] practice guidelines for second and third trimester obstetric sonograms) and for identification of abnormal fetal structures. For instance, a rounded, fluid-filled structure in the left posterior portion of the upper fetal abdomen may be assumed to represent the fundus of the fetal stomach. However, a structure of identical appearance but located on the right side of the upper fetal abdomen must be interpreted as a pathologic lesion or an abnormality of situs.

It is important to recall that pathologic structures are frequently more visible than their normal counterparts (e.g., dilated small bowel loops are easier to detect than normal small bowel loops). However, it is even more important to keep in mind that the most difficult pathologic observation is to recognize the absence of a structure that ordinarily could be visualized (i.e., a missing portion of an extremity or the inability to see the stomach when esophageal atresia without tracheoesophageal fistula is present).

Superficial Anatomy of the Fetus

Routine sonography for obstetric indications in many cases does not require a survey of superficial fetal structures. However, when an anomaly is suspected, a careful look at superficial features of the fetus becomes important or even mandatory. Superficial anatomy considered in this section includes the face, ears, hair, and external genitalia. Importantly, technologic advancements in three-dimensional renderings of fetal sonograms have had a dramatic impact on visualization of superficial fetal anatomy.

The fetal face can be viewed with considerable clarity with two-dimensional sonography. Expectant mothers are often surprised to see their fetus so clearly ( Figs. 8-9 to 8-11 ). The brow, cheeks, eyelids (and occasionally even eyelashes), nose, lips, and chin can be seen with consistency. The nose and lips are the more important to image in detail (to exclude clefting). The alae, column, and nares can be clearly depicted ( Figs. 8-12 and 8-13 ). The upper lip is more important diagnostically than the lower lip and fortunately easier to see. Visualization is usually good enough to identify the philtrum. The cheeks are prominent, as expected, and the subcutaneous tissues of the cheek, because of the presence of a large fat pad, are brightly echogenic ( Fig. 8-14 ). The profile is also readily demonstrated (see Fig. 8-11 ) and provides useful information about the brow (e.g., frontal bossing), chin (e.g., hypognathia), and nose (e.g., midface hypoplasia and Down syndrome).

FIG 8-9, Coronal section of the fetal face that demonstrates the palpebral fissures (PF) particularly well. NB, nasal bone.

FIG 8-10, Sonogram of the fetal face. Even though this image is a tomogram with relatively little depth, facial features are seen well. Amniotic fluid surrounding the face provides the “contrast” for visualization. 1, brow; 2, eyelid; 3, cheek; 4, ala of nose; 5, nostril; 6, philtrum; 7, upper lip; 8, lower lip.

FIG 8-12, “Coronal” sonogram of the nose and mouth. Nasal structure is particularly well seen: 1, ala; 2, column; 3, nostril. The mouth displays less detail, although consistent and characteristic layering of echoes is seen. These layers are (presumably) the subcutaneous fat (4) and muscular tissue (5), the orbicularis oris muscle, and (6) the mucosal tissue.

FIG 8-13, A, True coronal image of the nose and upper lip. B, Inclined coronal image demonstrates a slightly different perspective. 1, nostril; 2, ala; 3, column; 4, upper lip; 5, cheek; 6, philtrum.

FIG 8-14, A, Axial section of the fetal face demonstrating the tongue. B, Midsagittal section of the fetal face demonstrating the tongue. 1, lip; 2, tongue; 3, fat pad of cheek; 4, nose; 5, maxilla; 6, chin.

The ears can be visualized quite well and their progressive maturation noted. The external auditory canal, helix (and antihelix in older fetuses), lobule, and tragus can be depicted ( Fig. 8-15 ), but the relative position of the ear (e.g., as in low-set ears) is difficult to judge—a task more readily accomplished with three-dimensional imaging. The ear may be protuberant and can be mistaken for an abnormality, especially an encephalocele.

FIG 8-15, Parasagittal ( A ) and coronal ( B ) sonograms of the fetal ear. AH, antihelix; AT, antitragus; EAC, external auditory canal; FT, fossa triangularis; H, helix; L, lobule; T, tragus.

When present, scalp hair is readily perceived in late fetuses. The bright linear echoes protruding from or paralleling the scalp and neck are quite conspicuous. The only benefit of recognizing hair is not to be misled into mistaking long hair for a pathologic process—namely, an encephalocele or cystic hygroma—because longer hair, wet and matted by the amniotic fluid, may trap some of the fluid between it and the skin of the occiput or neck, creating the false impression of a cystic mass in this location ( Fig. 8-16 ).

FIG 8-16, A, Hair may be mistaken for the outer membrane of a cystic mass in an older fetus. B, Scan at 90 degrees to A. This image clarifies that the fetus is normal. C, Parasagittal plane through a different fetus demonstrates hair above the ear. 1, hair; 2, portion of ear; 3, occipital bone; 4, subcutaneous tissues and muscles in the occipital region; 5, “trapped” amniotic fluid; 6, umbilical artery; 7, umbilical vein; 8, cerebellum; 9, helix; 10, lobule.

The external genitalia can be appreciated from early second trimester onward. Fetal sex can be quite accurately assigned. Ordinarily, this is not of clinical consequence. However, in certain circumstances, fetal sex should always be determined. These situations include all living twins in which a single placental site is seen or when monozygotic twinning, other than for reasons of placentation, would be considered detrimental to pregnancy outcome. All fetuses with suspected lower urinary tract obstruction should have fetal sex determined because the differential diagnosis is different in males and females. Certain other circumstances would require determination of sex if karyotyping were refused or impractical to perform and cell-free DNA screening were not possible or inconclusive. These indications include, but are not limited to, risk for X-linked disorders or when Turner syndrome is suspected because of a dysmorphic feature (e.g., cystic hygroma).

Female sex should be assigned only by identification of the major and minor labia ( Fig. 8-17 ). Assigning female sex based solely on an inability to see a penis will result in many diagnostic errors. Male genitalia are readily seen ( Fig. 8-18 ). The penis and scrotum are most obvious. Testes may be seen in the scrotal sac, sometimes as early as the beginning of the third trimester (the testes are intra-abdominal during most of gestation but are only visualized in the presence of ascites). Care should be taken to image the entirety of the phallus to avoid false foreshortening. Details of the penis, including the glans, urethra, and corpora cavernosa, may be appreciated ( Fig. 8-19 ). Even the foreskin is visible in some cases.