Perinatal Ultrasound

Ultrasound Equipment

In an ultrasound exam, a transducer (probe) is placed directly on the skin, inside the vagina or other body cavity, or directly on the organ of interest intraoperatively. A thin layer of water-based gel acts as a coupling agent to potentiate sound wave transmission.

Obstetric studies may be performed by either the transabdominal or transvaginal approach, using transducers of the appropriate frequency, usually between 1.5 and 10 MHz. Choice of transducer is a trade-off between penetration with lower frequencies (essential in the obese patient) and resolution at the higher end (required for the smallest fetal structures).

Fetal Imaging Techniques

Real-time ultrasound, in which image brightness varies with the intensity of returning signals (B-mode), is the standard method of fetal imaging ( Fig. 11.1 ). It is required for confirmation of cardiac activity and fetal movement in living gestations. Brief signal bursts are followed by relatively long receptive intervals (1 : 10 ratios or greater); lower signal frequencies encounter less interference but reflect from fewer informative interfaces. Image quality varies with distance to the target, structure size, movement relative to the signal, and tissue transmission characteristics. Ideally, the transducer is close to its target; the transvaginal approach is preferable in early gestation for cervical studies and for gynecologic investigations. Suboptimal images are common with obesity, limited fluid interfaces, intervening structures, gas-filled viscera, scar tissue, and poor positioning relative to the sound beam. Diagnostic difficulty arises when different materials have similar echo characteristics, as do blood, urine, ascites, and the contents of many cysts.

M-mode ultrasound is a direct representation of beam reflection by moving edges (e.g., in cardiovascular imaging). Interpretation requires standardized, often hard-to-achieve, stable views. M-mode is useful in assessing arrhythmias, myocardial contractility, and pericardial effusions. M-mode “snapshots” efficiently document cardiac activity and rate ( Fig. 11.2 ).

Doppler ultrasound uses the frequency shift that occurs when sound beams are reflected off moving objects to demonstrate the presence, velocity, and direction of blood flow ( Fig. 11.3 ). Direct calculations from narrow, tortuous fetal and uterine vessels lack accuracy; to compensate, prenatal Doppler findings ideally are obtained at beam angles less than 35 degrees relative to umbilical vessels and 15 degrees for middle cerebral arteries. The flow indices are generally expressed as ratios, relative to median values. Color Doppler semiquantitatively assigns direction to blood flow; by convention, warm colors (red) denote movement toward the transducer, and saturation is keyed to velocity. Color Doppler illuminates cardiac, arterial, and venous structures ( Figs. 11.4 and 11.5 ). Color Doppler energy (power Doppler) reflects signal intensity; amplitude corresponds to blood cell motion. Power Doppler is sensitive to very low flow and effective independent of angulation; it is helpful for mapping vascular beds and for quick visualization of any fetal vessel ( Figs. 11.6 and 11.7 ).

Three-dimensional ultrasound analyzes returning echoes along a third axis. Images are manipulated electronically to render surfaces and volumes from multiple perspectives, both as static and real-time (“four-dimensional”) views. Surface rendering of subtle organ details enhance detection of anomalies, partially overcoming positional limits of standard scans and permitting a comprehensive review of fetal organs and skeleton ( Fig. 11.8 ). Three-dimensional studies have improved volume calculations, facilitated analysis of complex spatial relationships, and have better explained abnormal findings. Spatiotemporal correlation (by a database managing both space and time inputs) of heart movement with color and power Doppler augments standard cardiac imaging ( Fig. 11.9 ).

In spite of challenges, the most clinically mature applications for volume ultrasound technology are within the realm of obstetrics. The ability to reorient the active view for optimal visualization permits rapid identification of normal and abnormal structures. Advantages to volume imaging also include presentation of recognizable fetal anatomy and anomalies to parents, aiding informed decision-making, and aid in maternal-infant bonding.

Magnetic resonance studies of pregnancies are usually performed after the first trimester, in spite of lack of identified biologic adverse effects. Maternal oblique positioning prevents undesirable inferior vena cava compression, and images are obtained with ultrafast sequences in less than 1 second; claustrophobia can be difficult for some patients, even after sedation or anxiolytic use. The wide field of view, extremely high resolution, excellent soft tissue contrast, and multiple potential planes of construction make MRI an appealing imaging modality, particularly useful when ultrasound studies are compromised by obesity or oligohydramnios.

MRI is often used to elucidate problems initially identified by ultrasound examinations. Images are acquired in the axial, coronal, and sagittal planes relative to the fetus or orthogonal to the maternal pelvis. Gadolinium is placentally transferred and is contraindicated during pregnancy because of its exceptional persistence in tissue and known potential for adult renal injury. Its use rarely may be justified for assessment of placenta accreta or serious maternal disease.

Bioeffects and Safety

As part of the Food and Drug Administration's (FDA's) initiative to reduce unnecessary radiation exposure from medical imaging, health care providers are advised to consider techniques with little or no ionizing radiation, for example, ultrasound or MRI, and only if medically appropriate.

Diagnostic ultrasound energy has the potential for affecting tissues (bioeffects); two recognized mechanisms are heating and cavitation. Data collected during routine studies show that “gray-scale” B-mode ultrasound is associated with a negligible rise in temperature. To date, there has been no convincing evidence of harm to the human fetus.

By definition, ultrasound is inaudible to humans; moreover, oncogenic effects have not been identified. At nonclinical levels, ultrasound energy causes cell lysis, intracellular shearing, streaming effects, altered membrane permeability, and abnormal chromosome function. Heat exposure triples with each change in modality: from 2- or 3D, B-mode to M-mode, then color flow, before peaking during pulsed Doppler. Harmful levels should not be attained routinely but might occur during focal cranial pulsed Doppler interrogation or in a febrile patient, without unusually long exposures. Mechanical disruption from cavitating gas bubbles is improbable in the fetus. Both temperature and disruption risks are now displayed on equipment ( Fig. 11.10 ); thermal index (TI), a ratio between transducer output and the energy needed to warm up tissue temperature by 1°C, with a desired value below 2, is also categorized by tissue type: TI soft tissues, TI cranial structures, and TI bone. Mechanical index (MI) references pulse amplitude effects of compression and decompression, ideally maintained below 0.4 in fetal studies. Publication standards now usually require that these indices be displayed on submitted images. For safety, only medically essential examinations should be performed; settings and duration should be the minimum required to achieve adequate views.

Strong magnetic fields and radiofrequency waves are used in MRI with no known harmful effects, but large longitudinal studies are lacking. As with ultrasound, heat delivery to the fetus is a recognized hazard; in MRI, however, the maternal surface receives the greater thermal exposure. Noise from the magnetic coils (up to 120 dB) is, in theory, capable of causing acute hearing damage; fortunately, maternal tissue attenuation decreases fetal intensities by 25% to relatively safe levels. Direct magnetic bioeffects remain unproven; the FDA states that safety to the fetus “has not been established.”

The FDA has recommended that health care providers attempt to minimize exposure while maintaining diagnostic quality when using ultrasound. The lowest possible ultrasound exposure settings that obtain adequate image quality and gain the necessary diagnostic information should be used, following the as-low-as-reasonably-achievable (ALARA) principle. Spectral or “flow” Doppler should not routinely be used to “auscultate” the fetal heart rate in the first trimester because of its higher energy delivery; instead, adequate documentation of viability can be obtained with use of M-mode or conventional two-dimensional real-time ultrasonography with video or cine archiving.

Ethical Considerations

The education of physician sonologists encompasses visual recognition and interpretative tasks common to diagnostic imaging but also demands mastery of specific mechanical skills (or, at least, an ability to assess the latter in sonographers). Both ultrasound and MRI have rapidly evolved; clinically relevant frontiers are often explored by collaborators with pooled data, prerelease technology, and funding for staff and statistical analysis. Nuchal translucency measurement, a deceptively simple sonographic method for aneuploidy screening, provides a cautionary example ( Fig. 11.11 ). Its orderly dissemination, including certification courses, ongoing audits, professional society, and laboratory coordination, stands in sharp contrast to the viral dissemination of most techniques, yet erosion of competency, once achieved, remains a concern.

Ethical practitioners should be candid when informing patients of their ability to provide a requested service and assiduous in improving their skills. Professional judgment remains paramount in deciding how and when to incorporate new developments into personal clinical practice. The ability to conduct a knowledgeable, balanced discussion regarding interventions both local and international is essential when counseling patients overwhelmed by online information.

Prenatal identification of fetal sex for the purpose of selective termination is available for serious X-linked disorders but has been more widely applied to abort normal female fetuses because of a lower perceived value ( Fig. 11.12 ). The addition of preimplantation genetic evaluations and non-invasive prenatal screening has redistributed the onus without fully resolving the problem.

Nonmedical fetal ultrasound (also known as “keepsake” ultrasound) uses ultrasound to provide nondiagnostic pictures or to determine the sex of a fetus (for a “reveal” party) without a practitioner referral, directly paid for by the patient. Notwithstanding governmental and professional guidelines and warnings regarding ultrasound safety, the popularity with prospective parents, evident profitability, and absence of proven harm have provided keepsake businesses with the impetus for rapid expansion. A number of ethical issues are raised, including conflicts of interest for the commercial enterprise, the fetus, and the parents with respect to long-term effects. Current epidemiologic evidence is not synchronous with advancing ultrasound technology; a lack of evidence of harm is not the same as lack of harm. Applying four major theories of ethics and principles (the precautionary principle, theories of consequentialism and impartiality, duty-based theory, and rights-based theories) leads to the conclusion that obstetric ultrasound practice is ethical only if the indication for use is based on medical evidence, rendering “keepsake studies” ethically unjustifiable.

Multiple gestations may result in a number of ethical dilemmas. Advances and regulation in assisted reproductive technology (ART) have decreased the incidence of multifetal pregnancies, but fetal reduction remains a painful choice for parents facing the prospect of extreme prematurity in higher order multiples. Management of twin–twin transfusion syndrome (TTTS), anomalous co-twins, discordant growth, or distress far from term also necessitates choosing among unsatisfactory alternatives. An excellent review of the psychosocial consequences and the ethical issues associated with selective termination of pregnancy has been published.

Non-diagnostic studies, varying prognoses for a given diagnosis, and the inherent limitations of ultrasound and MRI studies lead to ethical issues in management and counseling. Patients and physicians alike may share unrealistic expectations for the predictive accuracy of targeted diagnoses.

Anomalies and variants linked to Down syndrome and other serious conditions (sonographic “markers”) identified during routine studies present patients and caregivers with unanticipated, unwelcome options, particularly if patients had previously explicitly refused serum screening or direct genetic testing. Ideally, informed consent discussions addressing risks, benefits, consequences, alternative strategies, and limitations of ultrasound and MRI should be provided to all patients before performing such imaging. Given the irreversible nature of both birth and abortion, prospective parents must ultimately judge for themselves their tolerance for uncertainty in diagnosis and for imperfection in their offspring.

Classification of Fetal Sonographic Examinations

A

First-Trimester Examination

First trimester transvaginal scans exclude ectopic implantation by demonstrating an intrauterine asymmetric or “double sac” gestational sac with yolk sac or embryo, ideally with visible cardiac activity; identifiable embryos in the fallopian tubes are uncommon. Heterotopic pregnancy occurs in less than 0.1% of spontaneous conceptions, leading to the pragmatic conclusion that finding an intrauterine pregnancy excludes an ectopic one, except after assisted reproductive technic (ART). Nevertheless, careful study of the adnexa, ovaries, and cervix is generally prudent. A “double sac” is usually seen transvaginally at levels of 1000-1500 international units of human chorionic gonadotropin before 5 1/2 menstrual weeks (see Fig. 11.10 ); but serum human chorionic gonadotropin (hCG) and visualization thresholds are variable. Finding a yolk sac or embryo confirms an intrauterine site, with growth of the gestational sac diameter of about 1 mm daily in early pregnancy. The embryonic disc is usually visible transvaginally once sac diameters exceed 15 mm. Embryonic length also increases daily by 1 mm; transvaginally identified cardiac activity is customary by the end of the sixth week (4 mm embryonic length) and obligatory by the seventh. Current practice postpones diagnosing a failed pregnancy until well past the accepted normal thresholds for appearance and progressive development of these structures.

Embryonic heart rates, slower at initiation, increase to more than 160 beats per minute (bpm) by the 9th week, declining slightly through the 13th week. Persistent rates below 100 bpm have been linked to risk for missed abortion, aneuploidy, and anomalies. When an embryonic heart rate (below 80 bpm) is detected between 6 and 7 weeks, a first-trimester demise is anticipated in approximately 25%, even if the rate subsequently normalizes. In such pregnancies, a follow-up scan is warranted. Prior to 6 weeks, an embryonic heart rate below 100 bpm is not necessarily indicative of a poor prognosis. Likelihood of survival into the second trimester is also significantly higher when there is concordance between biometrically calculated gestational age and menstrual dating.

Embryonic anatomic surveys are limited and necessarily provisional; however, early fetal period scans diagnose a number of entities accurately ( Fig. 11.13 ). First-trimester screening for early detection of abnormalities decreases the need for invasive testing (previously offered to every woman over 35 years old, thereby missing abnormal pregnancies in younger women), the associated risk of miscarriages, and enabled first-trimester invasive testing and diagnosis in high-risk pregnancies using chorionic villi sampling rather than a second-trimester amniocentesis. It also allowed the safer and more private option of first-trimester termination of an abnormal pregnancy before the pregnancy was visible, decreased anxiety, and provided reassurance about pregnancy well-being in both high-risk and low-risk situations. Progress in the field of first-trimester sonography continues in two intertwined directions: One is finding additional sonographic markers that increased the accuracy and sensitivity of detecting chromosomal abnormalities, especially Down syndrome. The other is improving the early detection of congenital anomalies. Midtrimester confirmation continues to be prudent for the majority of first-trimester findings (see Fig. 11.13 ). Later studies retain advantages with respect to the natural history of many anomalies and for visualization of heart, spine, and other problematic structures.

B

Standard Second- or Third-Trimester Examination

A standard second- or third-trimester sonogram includes an evaluation of fetal number, presentation, cardiac activity, fetal biometry, amniotic fluid and placental characteristics, and fetal organ survey. Second trimester transvaginal cervical measurement is encouraged for patients at risk for prematurity; however, the role of universal screening remains controversial. Examination of the maternal pelvic structures is otherwise performed transabdominally at this time, if feasible.

C

Limited Examination

A limited examination is performed to investigate a specific concern, usually under exigent circumstances. For example, a limited examination might identify fetal cardiac activity in the presence of bleeding or confirm presentation in early labor. Abbreviated sonographic studies are more acceptable when there has been a prior complete study; pragmatically, a full examination should be documented once the acute situation has stabilized for ongoing pregnancies.

D

Specialized Examinations

A detailed “targeted” anatomic examination is performed when an anomaly is suspected on the basis of history, biochemical or genetic results, or findings on prior scans. Other specialized examinations include fetal Doppler ultrasound, 3D imaging, biophysical profile, fetal echocardiogram, detailed neurosonography, and additional biometry or evaluation of organs not usually imaged on routine studies.

Applications of Ultrasound Studies

Genetic Screening

Genetic screening combines ultrasound study and biochemical testing to enhance the detection of chromosomal abnormalities. This approach has resulted in greater scrutiny of younger patients and less frequent age-based invasive testing. Presently, noninvasive screening for fetal aneuploidy (trisomies 13, 18, 21) is encouraged for all low-risk patients. Common noninvasive screening options include: (1) first-trimester screening (nuchal translucency measurement and maternal serum biochemical marker algorithm), (2) first- and second-trimester cell-free fetal DNA fragment analysis from maternal blood, (3) second-trimester serum screening (maternal age and serum biochemical marker algorithm), or (4) two-step integrated screening, which includes first- and second-trimester serum screening with or without nuchal translucency (integrated prenatal screen, serum integrated prenatal screening only, contingent and sequential screening variations). Different algorithms noticeably affect sensitivity, specificity, and predictive values; cost or convenience may factor in choosing a strategy. Analysis of cell-free fetal DNA, found at concentrations almost 25 times higher than those from intact nucleated fetal blood cells extracted from similar volumes of maternal blood. Cell-free fetal DNA (cfDNA) noninvasive prenatal screening is now in wide use for specific targeted chromosomal abnormalities, especially trisomy (an extra copy of a chromosome) or monosomy (a missing chromosome), and numerous commercial products are currently marketed for this indication. Patient acceptance has been high, perhaps in part because of early identification of fetal sex. Insufficient fetal fractions and indeterminate results appear more frequently in obese patients but may also occur with chromosomal abnormalities. Direct prenatal diagnosis by chorionic villus biopsy or amniocentesis should be routinely recommended for at-risk individuals on the basis of age, family history, abnormal screening, or ultrasound findings.

One-third of fetuses with Down syndrome will have anomalies, characteristically endocardial cushion defects ( Fig. 11.14 ), duodenal atresia ( Fig. 11.15 ), and more subtly, small atrioseptal and ventriculoseptal cardiac defects ( Fig. 11.16 ). Two-thirds may have second-trimester sonographic markers: for example, increased nuchal lucency, hypoplastic or absent nasal bones, and abnormal cardiovascular Doppler patterns in the first trimester, thickened nuchal fold ( Fig. 11.17 ) and nasal hypoplasia, ventriculomegaly, choroid cysts ( Fig. 11.18 ), hypoplasia of the fifth digit, decreased long bone ratios, enhanced echogenicity of papillary muscles and bowel, and renal pyelectasis ( Fig. 11.19 ). The predictive value of screening components is affected by ethnicity, habitus, maternal diet, and fetal sex, as well as by device and operator-dependent detection rates. The permutations may confound counselors attempting to elucidate (and patients trying to grasp) the difference between screening and diagnosis, and basic descriptions of risks and benefits. Recently described first-trimester evaluation of the posterior brain (intracranial translucency) provides an additional screening tool for open neural tube defects and other intracranial abnormalities ( Fig. 11.20 A and B ). Second-trimester sonographic follow-up and genetic evaluation may add critical information to the assessment of these abnormal findings. First-trimester identification of tricuspid regurgitation and increased ductus venosus resistance seems a promising, albeit technically challenging, addition to early screening protocols.

Assisted Reproduction

Ultrasound is essential for timing and guiding oocyte retrieval and helpful in embryo transfer; its role in judging endometrial receptivity is less clear. Saline ultrasound studies prior to fertility treatments routinely complement or replace hystero-salpingogram in assessment of the uterus and adnexa. Use of MRI may play an expanded role in structural and functional evaluation in the future.

Assisted reproductive technology results in more twins and higher-order multiples; early ultrasound study of embryos, amnionicity, and chronicity is essential to subsequent management. With the increasing popularity of ART, obstetricians and radiologists are more likely to encounter associated complications, especially in an emergency setting. These complications include ovarian hyperstimulation or torsion, ectopic or heterotopic pregnancy, and pregnancies of unknown location or undetermined viability. Ovarian hyperstimulation syndrome may occur following ovulation induction or ovarian stimulation and is characterized by bilateral ovarian enlargement, by multiple cysts, and third-spacing of fluids ( Fig. 11.21 ). Additional clinical findings may range from gastrointestinal discomfort to life-threatening renal failure and coagulopathy. Ovarian torsion should be excluded in any woman undergoing ART who presents with severe abdominal pain. Ectopic pregnancy resulting from ART has a relatively increased frequency of rarer and more lethal forms, including interstitial and cervical locations. Heterotopic pregnancies, simultaneous intrauterine and ectopic implantations, are more common in ART patients.

Ultrasonography is the first-line choice for identifying ART complications, although lack of specific symptoms may first trigger other approaches. Familiarity with characteristic strengths and limitations of these and other techniques will facilitate accurate, timely diagnosis and avert potentially serious consequences.

Multiple Gestations

Multiple gestation accounts for about 3% of all pregnancies. With an increase in the number of fetuses, scans become more complex, time consuming, and error prone; additionally, determination of zygosity is essential. The management of anomalous, discordant, or moribund co-twins differs significantly based on chorionicity ( Fig. 11.22 ). Monochorionic twins occur with a relatively constant frequency (1 : 250 pregnancies), unlike dichorionic twinning that may be influenced by race, heredity, maternal age, parity, and ART. Ultrasound assignment of chorionicity is most accurate for different-sexed dizygotic twins, but by evaluating sac appearance in early gestation, approaches this accuracy in gender-concordant pairs. Successful identification may occur throughout gestation by examining the dividing membranes at their placental origin. Dichorionic diamniotic twins are usually (95%) dizygotic, with independent risks for anomalies and placental malfunction. Monochorionic pairs are predictably monozygotic; attrition rates exceed 30% from early abortion, anomalies, and prematurity. Matched and isolated anomalies are both more common in monochorionic gestations; because of shared vasculature, loss of a co-twin may kill its sibling outright or produce severe neurologic damage in up to one-third of survivors.

Monochorionic TTTS, more common in females, is characterized by unbalanced, shared perfusion that restricts growth and amniotic fluid production in the donor and causes volume overload, cardiac dysfunction, and polyhydramnios in the recipient. Ultrasound staging has been used to time and to guide a variety of vascular ablation strategies. Serial amniocentesis may be helpful in milder cases. Management by these approaches has been modestly effective in decreasing stillbirths and prematurity in TTTS. Monoamniotic twins rarely experience TTTS but routinely encounter cord entanglements, resulting in high lethality rate of both fetuses ( Fig. 11.23 ). Management usually consists of a scheduled preterm cesarean prior to onset of labor. For all multiple gestations, serial ultrasound monitoring of growth, well-being, placental performance, and cervical length is common practice.

Twin reversed arterial perfusion (TRAP) sequence is another complication of monochorionic twinning, complicating approximately 1% of monozygotic pregnancies. Placentation among TRAP cases has been predominantly reported to be monochorionic diamniotic and to a lesser degree monochorionic monoamniotic twins. The proposed pathogenesis is the association of paired artery-to-artery and vein-to-vein anastomoses through the placenta combined with delayed cardiac function of one of the twins early in pregnancy. This situation allows blood pumped from the healthy twin “pump twin” to perfuse retrogradely the heart of the other twin, also known as the “acardiac” twin or “parabiotic twin.” Thus, flow in the artery and vein are reversed in the umbilical cord of the acardiac twin, giving rise to the acronym TRAP. Retrograde perfusion interferes with normal cardiac development, which rarely goes beyond the stage of tubular heart. Common abnormal findings in the “acardiac twin” include impaired or absent development of the cephalic pole, rudimentary or absent heart, abnormal or absent upper limbs, relative preservation of the lower limbs although clubbing and abnormal toes are common, abnormal viscera, and single umbilical artery. A common finding is massive edema around the upper body including the neck of the acardiac twin ( Fig. 11.24 A - C ).