Chapter 11. Imaging in Obstetrics

Technology for imaging in obstetrics has advanced a great deal over the past decade, but the purpose remains the same: to evaluate the anatomy and well-being of the fetus and the mother. The primary imaging modality in obstetrics is the 2-dimensional (2D) ultrasound because it is safe and widely available. Three- and 4-dimensional (3D/4D) ultrasound and magnetic resonance imaging (MRI) are also being used in certain situations to enhance imaging techniques, but the cost of these modalities limits their widespread use. Finally, computed tomography (CT) scan has limited utility because of safety issues but may be necessary in evaluating certain maternal conditions.

Ultrasound evaluation uses sound waves at a frequency greater than that which the human ear can hear (>20,000 cycles per second or Hertz [Hz]) to obtain images. An ultrasound examination is performed in real time with images or video clips stored for review. Ultrasound probes contain a transducer that creates the ultrasound waves at different frequencies. Higher frequency transducers provide better resolution but have less tissue penetrations, whereas low-frequency transducers have lower resolution but better tissue penetration.

In obstetrics and gynecology, ultrasound imaging is generally performed 1 of 2 ways: either with a transvaginal probe or transabdominal probe. The choice of which probe to use generally depends on the structure of interest and its distance from the probe. For instance, imaging of the cervix or an early gestation is generally best achieved with a transvaginal probe, whereas evaluation of the fetus in the third trimester of pregnancy is best accomplished with transabdominal imaging.

Ultrasound during pregnancy when performed for medical indications is considered to be safe. There are no documented harmful effects to the fetus from diagnostic ultrasound. Ultrasound waves, however, are a form of energy and have been shown to raise tissue temperature with high energy output or prolonged exposure. To minimize this risk, it is recommended that energy output, as measured by the Mechanical Index, be kept less than 1.0 and that ultrasound be used for diagnostic purposes only (ie, not for entertainment purposes). Imaging of the fetus can be divided into ultrasound evaluations in the first trimester and evaluations in the second and third trimesters. In each trimester, the goals and the ability to evaluate the fetal anatomy differ.

First-Trimester Ultrasound Examination

There are a number of indications to perform first-trimester ultrasound. They include confirmation of an intrauterine pregnancy, assessment of pelvic pain and vaginal bleeding, estimation of gestational age, confirmation of viability, evaluation of number of gestations, genetic screening, evaluation of basic anatomy, and assessment of uterine and adnexal anomalies and pathology.

To begin with, a first-trimester ultrasound can be used to confirm an early pregnancy by documenting the location of a gestational sac and the presence or absence of a yolk sac and fetal pole. However, one should use caution in determining the location of a pregnancy based solely on the presence of a gestational sac because sometimes an intrauterine fluid collection could be a pseudogestational sac from an ectopic pregnancy and not a normal gestational sac. Please see Chapter 13 for a more thorough discussion of using ultrasound to evaluate for ectopic pregnancy.

An ultrasound in the first trimester can also be used to determine the gestational age and viability of a pregnancy by the presence or absence of cardiac activity. Gestational age should be determined by measuring the length of the fetal pole. If there is no fetal pole noted, the gestational age can be estimated by measuring the mean sac diameter of the gestational sac. To assess for viability, cardiac activity should be noted by transvaginal ultrasound when the fetal pole measures 4–5 mm, which corresponds to a gestational age of 6.0–6.5 weeks. Using transabdominal ultrasound, cardiac activity should be noted when the fetal pole corresponds with a gestational age of 8 weeks. Table 11–1 includes some guidelines for evaluation of an abnormal pregnancy. However, there are exceptions to the guidelines, so a follow-up ultrasound may be considered in order not to terminate a normal pregnancy.

The first trimester is also the best time to determine the number of gestations and to evaluate the chorionicity and amnionicity of multifetal pregnancies. An ultrasound will be able to show the number of gestational sacs, number of yolk sacs, and the number of fetal poles with cardiac activity. In cases of multifetal pregnancies, an ultrasound will also show the location of the placenta or placentas, the number of yolk sacs, and the presence or absence of a dividing membrane or membranes. In monochorionic-monoamniotic twins, there will be no dividing membrane between the fetuses, and there will be a single yolk sac (Fig. 11–1).

Monochorionic, diamniotic pregnancies in the early first trimester are characterized by a single chorion with 2 yolk sacs visualized. In many cases, the amnion may not be visible until approximately 8 weeks' gestation, at which time thin amniotic sacs may be seen surrounding each embryo (Fig. 11–2). By 10 weeks, monochorionic-diamniotic pregnancies are characterized by the T-sign on ultrasound, which represents a single placenta with 2 amnions. This is in contrast to dichorionic-diamniotic pregnancies, which will either show 2 separate placentas or show the classic lambda or twin-peak sign if there are fused placentas (Fig. 11–3). Conjoined twins can also be diagnosed during the first trimester based on ultrasound findings demonstrating a single amnion, chorion, and yolk sac and 2 fetal poles that are fused (Fig. 11–4).

In the last 20 years, extensive research has shown that first-trimester ultrasound can also be used as a screening test for Down syndrome, trisomy 18, and trisomy 13. This is performed by measuring the nuchal translucency, or area of fluid that accumulates behind the fetal neck (Fig. 11–5). The combination of this measurement with maternal β human chorionic gonadotropin and pregnancy-associated plasma protein-A levels provides a patient-specific risk that the patient can use to determine whether she wants definitive testing for fetal chromosomal abnormalities with either chorionic villus sampling or an amniocentesis. More recently, research has shown that by incorporating the presence or absence of the nasal bone on ultrasound into the algorithm, the detection rate for Down syndrome in the first trimester is 94%, with a false-positive rate of 5%. Some institutions are also looking at impedance to flow in the ductus venosus and tricuspid regurgitation as markers of genetic abnormalities early in pregnancy.

Finally, with improvements in ultrasound technology and implementation of first-trimester screening for aneuploidy, more attention has been paid to the fetal anatomical survey in the first trimester. Although there are no official guidelines regarding what constitutes a fetal anatomical survey in the first trimester, it is possible to evaluate the fetal brain, spine, stomach, bladder, kidneys, abdominal cord insertion, and extremities during this period. Anatomical survey in the first trimester can currently detect a number of fetal anomalies, and surely as experience continues to grow, the number and type of anomalies detected will also increase. However, there are many structures that cannot be adequately evaluated in the first trimester, and there are many fetal anomalies that do not manifest on ultrasound until later in pregnancy. Therefore, an additional ultrasound is recommended in the second trimester.

Second- and Third-Trimester Ultrasound Examination

In the second and third trimesters, transvaginal and transabdominal ultrasound can be used for screening for chromosomal and nonchromosomal fetal anomalies, fetal growth, fetal well-being, fetal lie and presentation, placental anomalies, and cervical insufficiency. Additionally, it can be used for evaluating for gestational age, number of gestations, and viability if a first-trimester ultrasound was not performed.

Anatomy

The fetal anatomical evaluation recommended in current guidelines can be adequately assessed transabdominally after approximately 18 weeks of gestation and detects approximately 70% of major anatomical anomalies. However, the majority of the anatomy can be seen as early as 16 weeks transabdominally and even as early as 14 weeks transvaginally. If a patient is going to have a single second-trimester ultrasound to evaluate the basic fetal anatomy, it should be performed after 18 weeks' gestation. This is with the understanding that she may need to have a follow-up ultrasound after 20 weeks if further imaging of the brain and/or the heart is recommended. Some structures that are not required in current guidelines but that can be assessed (eg, the corpus callosum in the brain) cannot be reliably viewed until approximately 20 weeks' gestation. Therefore, some experts recommend 2 anatomy ultrasounds during pregnancy; the first can be performed at approximately 14–16 weeks' gestation to allow for early diagnosis of major structural malformations and the second after 20 weeks' gestation to optimize evaluation of the heart and brain. Table 11–2 includes a guideline for fetal structures that should be assessed during a second-trimester anatomy ultrasound.

The second-trimester ultrasound can also be used as a genetic sonogram because major structural anomalies that can be seen are often associated with a chromosomal anomaly. Trisomy 13 and trisomy 18 fetuses have major structural anomalies that are identifiable by ultrasound in more than 80% of cases. However, fetuses with Down syndrome have major ultrasound-identifiable anomalies in only 25% of cases. These include certain cardiac anomalies, duodenal atresia, and ventriculomegaly. Ultrasound can also detect “soft” markers, which are variations in normal anatomy that are usually not clinically significant but can be associated with aneuploidy. Soft markers of Down syndrome include short femur or humerus length, renal pyelectasis, echogenic intracardiac foci, ventriculomegaly, or hyperechoic bowel. According to clinical studies, if the ultrasound does not reveal any soft markers, the risk of Down syndrome is reduced approximately 50–80%. A normal second-trimester genetic sonogram, however, does not eliminate the possibility of Down syndrome.

Growth

Ultrasound in the second and third trimester can also be used to evaluate fetal growth. Measurements of the biparietal diameter, head circumference, abdominal circumference or average abdominal diameter, and femoral diaphysis length can be calculated to determine an estimated fetal weight. This estimated fetal weight can be compared with the estimated fetal weights in published nomograms at each gestational age to evaluate the growth of the fetus. Indications for a fetal growth scan include measurement of fundal height less than expected based on gestational age, inability to measure fundal height because of fibroids or maternal obesity, multiple gestations, or maternal or fetal complications of pregnancy that are associated with fetal growth restriction.

The purpose of performing fetal growth scans is to identify fetal growth abnormalities (ie, fetal growth restriction and macrosomia). Intrauterine growth restriction is usually defined as an estimated gestational age less than the 10%. It can be associated with chromosomal and nonchromosomal anomalies, infection, and placental insufficiency. Identification of these fetuses is important because growth restriction is associated with fetal demise, and increased surveillance of the growth restricted fetuses may decrease this risk. Macrosomia can be defined as an estimated fetal weight greater than 4000 or 4500 g. Identification of these fetuses can be useful because of the association of macrosomia with postpartum hemorrhage, caesarean delivery, and shoulder dystocia. However, ultrasound is not a perfect estimate of fetal growth. The error rate can be as high as 15–20% depending on the gestational age and certain maternal characteristics such as body habitus and abdominal scar tissue. Also, if ultrasounds for growth are performed more often than every 2 weeks, the margin of error may be too great to determine whether appropriate growth has occurred. (See Chapter 16 on Disproportionate Fetal Growth for more detail.)

Evaluation of Fetal Well-Being

The purpose of using ultrasound for fetal surveillance is to identify fetuses that are at risk for intrauterine death or severe morbidity. Ideally, this will allow interventions such as early delivery in order to prevent these complications. Patients that may benefit from fetal surveillance include those with complaints such as decreased fetal movements, who carry fetuses with intrauterine growth restriction, or with medical or fetal complications that put them at risk for intrauterine death or severe morbidity. The 2 main methods of fetal surveillance are the biophysical profile or modified biophysical profile and Doppler ultrasound.

Biophysical Profile

The biophysical profile (BPP) was first introduced in 1980. It consists of using 4 ultrasound parameters and a nonstress test to assign a score that gives a fetus's risk of hypoxia or intrauterine death. Table 11–3 shows the scoring of the fetal BPP. A score of 8 or 10 (of a possible score of 10) is considered normal. A score of 6 is considered equivocal, and a score of 4 or less is considered abnormal. Oligohydramnios, regardless of the composite score, warrants further evaluation.

The modified BPP consists of performing a nonstress test with only an evaluation of the amniotic fluid volume as measured by the amniotic fluid index (AFI). The AFI is calculated by dividing the uterus into 4 quadrants and measuring the maximum vertical pocket of amniotic fluid in each quadrant. The theory for this is based on the idea that the nonstress test is a measure of short-term fetal status, and the AFI is a measure of long-term fetal status. The AFI can measure long-term status because placental dysfunction can cause decreased renal perfusion, which can lead to oligohydramnios. If either the nonstress test is not reactive or the amniotic fluid index is less than 5 cm, the test is considered nonreactive. The stillbirth rate within a week of testing for a normal BPP or modified BPP is approximately 0.6–0.8 per 1000, giving a negative predictive value of greater than 99.9% for both tests. Risk of adverse pregnancy outcome as correlated with BPP score is shown in Table 11–4.

Fetal Doppler Evaluation

Doppler ultrasound is emerging as a newer method to evaluate fetal well-being. Christian Doppler first described the Doppler effect in the 1800s as a way of describing the variation in the frequency of a light or sound wave as the source of that wave moves from a fixed point. In medicine, Doppler ultrasound is used as a measure of the speed at which blood is moving within a vessel. The 3 most common fetal arterial Dopplers are measured in the umbilical artery, middle cerebral artery, and uterine artery, whereas the most common fetal venous Doppler is measured in the ductus venosus. Umbilical artery Dopplers are a reflection of the placental circulation. As diseases begin to affect the placenta and increase resistance within the placenta, the end-diastolic flow in the umbilical artery begins to slow and eventually may become absent or even reversed.

Middle cerebral artery Dopplers work on a different principle (Fig. 11–6). In the presence of fetal hypoxemia, blood flow is redistributed to the brain, known as the brain-sparing effect. As a result, in worsening disease states, blood flow increases in the middle cerebral artery. This measurement can be used both for a general assessment of fetal well-being as well as an assessment for fetal anemia. Finally, ductus venosus Dopplers reflect cardiac compliance and cardiac afterload, which may increase with disease states that affect the placenta. Therefore, evaluation of the ductus venosus waveform can be used to assess fetal well-being.

Placental Location

In addition to the fetus, ultrasound in the second and third trimester can be used to evaluate the placenta for placental anomalies. It is standard to evaluate the placental location during the anatomy scan or third-trimester growth ultrasound, but if there is vaginal bleeding, one may also evaluate for placental abruption.

The placenta is usually described by its location, its relationship to the internal cervical os, and its appearance. Specifically, it is important to evaluate the placenta for any evidence of placenta previa. The distance of the lower edge of the placenta is measured in relation to the internal cervical os. Using ultrasound, the relationship of the placenta to the cervix can be described in 1 of 3 ways: complete previa, marginal previa, and no evidence of previa. If the placenta or placental edge covers the internal os, it is considered a complete placenta previa. A marginal placenta previa occurs when the placental edge is within 2 cm of the internal os but does not cover it. This is clinically significant because a placenta previa is associated with antenatal and intrapartum vaginal bleeding, and it is recommended that patients with a placenta previa have a caesarean section for delivery to decrease their risk of experiencing hemorrhage.

The placenta may also be evaluated for the presence of placenta accreta, which is most commonly associated with prior uterine surgery. A placenta accreta refers to trophoblastic villi that penetrate the decidua but not the myometrium and thus result in an abnormally adherent placenta. Other abnormalities of placental attachment include placenta increta, in which the trophoblastic villi penetrate the myometrium, and placenta percreta, in which the trophoblastic villi penetrate the myometrium and uterine serosa. On ultrasound, a placenta accreta can be suspected when there are placental lacunae, thinning of the myometrium over the placenta, and loss of the retroplacental hypoechoic space among other findings. Clinically, this is relevant because a placenta accreta can prevent separation of the placenta from the uterus after delivery of the fetus. Antenatal identification of placenta accreta can be used to ensure a planned delivery with the appropriate resources, which can reduce maternal morbidity and mortality.

Finally, ultrasound of the placenta can be used to evaluate for vaginal bleeding in the pregnant patient. The most concerning causes of third-trimester vaginal bleeding are placenta previa and placental abruption. Placenta previa may be suspected in the clinical scenario of painless vaginal bleeding and as noted before can be diagnosed by ultrasound. Placental abruption is often suspected when there is painful vaginal bleeding, but the sensitivity of ultrasound in diagnosing placental abruption is low. Ultrasound can only visualize hemorrhage in approximately 50% of cases of clinical placental abruption.

Maternal Evaluation

Ultrasound is also used in pregnancy to evaluate for maternal pathology, such as uterine fibroids and ovarian cysts or masses. However, imaging of certain maternal structures also has implications for obstetrical outcomes.

Cervical Evaluation

Ultrasound can be used to evaluate the cervix in the pregnant patient, as the shape and length of the cervix has been shown to correlate with preterm delivery. The best way to evaluate the cervix is by using transvaginal ultrasound. The most clinically applicable measurement is that of the cervical length. This is the distance of the closed cervix from the external os along the endocervical canal to the innermost closed portion of the cervix. It is also often standard to report the presence or absence of funneling or opening of the internal os. It is also important that approximately 3–5 minutes be spent evaluating the cervix because of the potential for dynamic changes. The shortest cervical length should be reported and used for clinical management.

The specific uses of cervical length measurements are rapidly evolving, but it is important to understand some basic principles for the use of these measurements. A normal cervical length measures between 25 and 50 mm from the mid-second trimester until the third trimester. A cervical length less than or equal to 25 mm at these gestational ages can be considered abnormal or “short.” In the third trimester, there is physiologic shortening of the cervix, making the distinction between normal and abnormal more difficult. The earlier in gestation and the shorter the cervical length, the higher is the risk of preterm delivery. In Table 11–5, the risk of preterm birth by cervical length at 24 weeks is shown.

Cervical length measurements can be used in the second trimester in both low- and high-risk women to evaluate the risk for preterm birth. There are now many studies evaluating different treatment options, such as progesterone or cerclage, to help prevent preterm delivery based on the length of the cervix. In the late second and third trimester, cervical length can also be assessed in the symptomatic patient to help guide the need for hospitalization or the administration of steroids.

Maternal Doppler Evaluation

Uterine artery Dopplers are based on the theory that the spiral arterioles of the uterine arteries are meant to be maximally dilated to ensure adequate blood flow to the uterus. Worsening disease states are associated with a waveform notch and low end-diastolic velocity from high impedance circulation in the uterine artery. There have been a number of studies that have evaluated first- and second-trimester uterine artery Dopplers as a predictor of preeclampsia and intrauterine growth restriction. The studies have shown that in combination with abnormal maternal serum analytes, uterine artery Dopplers have a high sensitivity for prediction of these outcomes. Additional research is still being conducted to identify more accurately which patients to screen and with which combination of uterine artery Dopplers and serum analytes. Currently, besides close monitoring, aspirin is the only intervention that has been shown to decrease the risk of developing these adverse outcomes. Again, however, additional research is needed to determine which patients would benefit from this therapy.

Up to this point, the discussion on imaging in obstetrics has been limited to 2D ultrasound. However, as technology is improving, the clinical uses of 3D and 4D ultrasound have started to be evaluated. The benefits of 3D ultrasound include improved evaluation of fetal anomalies, the ability to store information for reevaluation at a later time or by a different clinician, and improved maternal–fetal bonding. Specific fetal anomalies that have been studied with 3D include the fetal face, skeleton, neuroanatomy, and heart. 3D ultrasound not only helps to confirm a suspected anomaly seen in 2D ultrasound, but it also can help to explain findings to the parents. Figures 11–7 and 11–8 demonstrate 3D images of the fetal face. Figure 11–7 represents a normal fetal face, whereas Figure 11–8 demonstrates a fetus affected by unilateral cleft lip. Furthermore, 4D ultrasound can be used to evaluate fetal anomalies that may affect fetal movement.

Although ultrasound is the primary imaging modality used in obstetrics, MRI can be beneficial in certain cases. In the second trimester, obtaining or ascertaining a diagnosis is usually the purpose of the MRI, whereas in the third trimester, delivery planning is the goal. Abnormal fetal neuroanatomy, specifically ventriculomegaly, is the most common indication for a fetal MRI. MRI has been shown to alter the diagnosis or improve the accuracy of the diagnosis in up to 50% of studies performed for suspected fetal central nervous system anomalies. MRI can also be used to evaluate other fetal anomalies, such as genitourinary, gastrointestinal, cardiac, and musculoskeletal anomalies. For example, genitourinary anomalies are often associated with oligohydramnios, which can affect the imaging quality of ultrasound. MRI, on the other hand, is not dependent on the surrounding tissue, so it can be beneficial in these cases. In addition to fetal imaging, MRI has also been shown to be helpful in evaluating the placenta. In cases of suspected placenta accreta, increta, and percreta, where ultrasound findings are equivocal or a posterior placenta limits ultrasound evaluation, MRI can be used to more accurately diagnose the suspected abnormality.

Finally, although the safety of ultrasound in obstetrics has been well established, the safety of MRI is sometimes questioned. There have been no studies linking MRI to any teratogenic effects on the fetus. However, gadolinium use is not recommended during pregnancy. Although limited studies have not shown any adverse effect on the fetus, gadolinium has been shown to cross the placenta and may have a long half-life in the fetal compartment. For these reasons, it should not be used unless the potential benefit outweighs the potential risk of unknown fetal harm.