Thompson & Thompson Genetics in Medicine, 8th Edition

CHAPTER 17. Prenatal Diagnosis and Screening

The goal of prenatal diagnosis and screening is to inform pregnant women and couples about the risk for birth defects or genetic disorders in their fetus and to provide them with informed choices on how to manage that risk. Some couples known to be at an elevated risk for having a child with a specific birth defect choose to forego having children. Prenatal diagnosis allows them to undertake a pregnancy with the knowledge that the presence or absence of the disorder in the fetus can be confirmed by testing. Many women or couples at risk for having a child with a severe genetic disorder decide to undertake a pregnancy and have been able to have healthy children because of the availability of prenatal diagnosis and the option of terminating an affected pregnancy if necessary. In some, prenatal testing can reassure and reduce anxiety, especially among high-risk groups. For still others, prenatal diagnosis allows physicians to plan prenatal treatment of a fetus with a genetic disorder or birth defect. If prenatal treatment is not possible, diagnosis during pregnancy can alert parents and physicians to arrange for appropriate management for the impending birth of an affected child in terms of psychological preparation of the family, pregnancy and delivery management, and postnatal care.

Prenatal diagnosis is the term traditionally applied to testing a fetus already known to be at an elevated risk for a genetic disorder to determine if the fetus is affected or not with the disorder in question. The elevated risk is usually recognized because of the birth of a previous child with the disease, a family history of the disorder, a positive parental carrier test, or when a prenatal screening test (discussed later in this chapter) indicates an increased risk. Prenatal diagnosis often, but not always, requires an invasive procedure such as chorionic villus sampling (CVS) or amniocentesis (both discussed later in this chapter) to acquire fetal cells or amniotic fluid for analysis. Prenatal diagnosis is meant to be as definitive as possible, giving a “yes or no” answer as to whether the fetus is affected with a particular disorder.

Prenatal screening, on the other hand, has traditionally referred to testing for certain common birth defects such as chromosomal aneuploidies, neural tube defects, and other structural anomalies in pregnancies not known to be at an increased risk for a birth defect or genetic disorder. Screening tests were developed because common birth defects most often occur in pregnancies not known to be at any increased risk and therefore the parents would not have been offered prenatal diagnosis. Screening tests are typically noninvasive, based on obtaining a maternal blood sample or on imaging, usually by ultrasonography or magnetic resonance imaging (MRI). Screening tests are typically designed to be inexpensive and sufficiently low risk to make them suitable for screening all pregnant women in a population regardless of their risk.

The ultimate goal of prenatal diagnosis is to inform couples about the risk for particular birth defects or genetic disorders in their offspring and to provide them with informed choices on how to manage that risk. In contrast, the goal of prenatal screening is to identify pregnancies for which prenatal diagnostic testing should be offered. Screening tests do not give a “yes or no” diagnostic answer about whether an abnormality is present. Rather, the risk for a birth defect derived from screening falls along a continuum relative to the background risk for an age-matched control group. The cut off for what is considered a positive screen is carefully set to balance sensitivity and specificity (i.e., false-negative and false-positive rates). Screening tests generally allow higher false-negative rates than would be acceptable for a diagnostic test to keep false-positive rates to a reasonable level, generally below 5%.

Traditionally therefore the distinction between prenatal diagnostic testing and prenatal screening has been made based on:

• Whether or not the pregnancy was known to be at risk for a particular disorder

• Whether the goal of the testing was a definitive diagnosis of a particular disorder or an assessment of risk relative to the background population risk

• Whether the test was invasive or noninvasive

Now, however, because of improvements in the safety of invasive procedures and advances in technology, the need to distinguish between diagnosis and screening is becoming much less clear. CVS or amniocentesis coupled with chromosomal microarray analysis (CMA) (see Chapter 5) is now being offered to every pregnant woman as a screening test, not only for the common chromosomal aneuploidies but also for other genomic imbalances, regardless of risk assessment based on personal or family history or noninvasive screening test results. Prenatal diagnosis is expanding beyond testing for specific disorders for which the fetus is known to be at risk to include any copy number abnormalities detectable by CMA and may, in the near future, include whole-genome sequence analysis of the fetus.

The purpose of this chapter is to discuss these various approaches to screening and diagnosis and to review the methodologies and indications as currently being used in this very rapidly changing field. The reader is cautioned, however, that because of technological advances in the methods available for assessing the fetus and the fetal genome, standards of care in prenatal screening and diagnosis are in flux.

Methods of Prenatal Diagnosis

Invasive Testing


Invasive testing utilizes CVS or amniocentesis to obtain fetal tissues. Amniocentesis refers to the procedure of inserting a needle into the amniotic sac and removing a sample of amniotic fluid transabdominally (Fig. 17-1A). The amniotic fluid contains cells of fetal origin that can be cultured for diagnostic tests. Before amniocentesis, ultrasonographic scanning is routinely used to assess fetal viability, gestational age (by determining various biometric parameters such as head circumference, abdominal circumference, and femur length), the number of fetuses, volume of amniotic fluid, normality of fetal anatomical structures, and position of the fetus and placenta to allow the optimal position for needle insertion. Amniocentesis is performed on an outpatient basis typically between the 16th and 20th week after the first day of the last menstrual period.


FIGURE 17-1 A, Amniocentesis. A needle is inserted transabdominally into the amniotic cavity, and a sample of amniotic fluid (usually approximately 20 mL) is withdrawn by syringe for diagnostic studies (e.g., chromosome studies, enzyme measurements, or DNA analysis). Ultrasonography is routinely performed before or during the procedure. B, Chorionic villus sampling (CVS). Two alternative approaches are drawn: transcervical (by means of a flexible cannula) and transabdominal (with a spinal needle). In both approaches, success and safety depend on use of ultrasound imaging (scanner). C, Preimplantation genetic diagnosis (PGD). Eggs are removed and used for in vitro fertilization. For blastomere biopsy, the fertilized embryos are incubated for 3 days, to the 8- to 16-cell stage, and a single blastomere is removed and undergoes genetic testing for a chromosomal abnormality or mendelian disorder. In this example, the embryo is affected (“X”) and after testing would not be implanted. In the blastocyst biopsy, approximately five trophectoderm cells (which will go to make the placenta and not the embryo proper) are removed and tested. Only those embryos that are unaffected will be implanted in the patient's uterus to establish a pregnancy.

In addition to fetal chromosome and genome analysis, the concentration of alpha-fetoprotein (AFP) can be assayed in amniotic fluid to detect open neural tube defects (NTDs) (see Chapters 8 and 14). AFP is a fetal glycoprotein produced mainly in the liver, secreted into the fetal circulation, and excreted through the kidneys into the amniotic fluid. AFP enters the maternal bloodstream through the placenta, amniotic membranes, and maternal-fetal circulation. It can therefore be assayed either in amniotic fluid (amniotic fluid AFP [AFAFP]) or in maternal serum (maternal serum AFP [MSAFP]). Both assays are extremely useful for assessing the risk for an open NTD but also for other reasons (see later discussion).

AFP concentration is measured by an immunoassay, a relatively simple and inexpensive method that can be applied to all amniotic fluid samples, regardless of the specific indication for the amniocentesis. To interpret an AFAFP, one compares the level to the normal range for each gestational age. If the AFAFP level is elevated (relative to the normal range for that particular gestational age), one must look for an open NTD as well as for causes other than an open NTD. Factors potentially leading to abnormally high concentrations of AFP in amniotic fluid are shown in Table 17-1. When the AFAFP assay is used in conjunction with ultrasonographic scanning at 18 to 19 weeks' gestation, approximately 99% of fetuses with open spina bifida and virtually all fetuses with anencephaly can be identified.

TABLE 17-1

Causes of Elevated Amniotic Fluid Alpha-Fetoprotein Other Than Neural Tube Defect


Note: Some of these causes of an elevated amniotic fluid AFP level can be confirmed or ruled out by ultrasonographic examination.

AFP, Alpha-fetoprotein.

If amniocentesis is performed for any reason, both the concentration of AFP in the amniotic fluid and a chromosome analysis of amniotic fluid cells are determined to screen for open NTDs and chromosomal and other genomic abnormalities, respectively. Other tests are performed only for specific indications.


The major complication associated with midtrimester amniocentesis at 16 to 20 weeks of gestation is a 1 in 300 to 1 in 500 risk for inducing miscarriage over the baseline risk of pregnancy loss of approximately 1% to 2% for any pregnancy at this stage of gestation. Other complications are rare, including leakage of amniotic fluid, infection, and injury to the fetus by needle puncture. Early amniocentesis performed between 10 and 14 weeks is no longer recommended because of an increased risk for amniotic fluid leakage, a threefold increased risk for spontaneous abortion, and an approximately sixfold to sevenfold increased risk for talipes equinovarus (clubfeet), over the 0.1% to 0.3% population risk. Early amniocentesis has now been replaced by chorionic villus sampling (see next section).

Chorionic Villus Sampling

CVS involves the biopsy of tissue from the villi of the chorion transcervically or transabdominally, generally between the 10th and 13th weeks of pregnancy (see Fig. 17-1B). Chorionic villi are derived from the trophoblast, the extraembryonic part of the blastocyst (Fig. 17-2), and are a ready source of fetal tissue for biopsy. As with amniocentesis, ultrasonographic scanning is used before CVS to determine the best approach for sampling.


FIGURE 17-2 Development of the tertiary chorionic villi and placenta. A, Cross section of an implanted embryo and placenta at approximately 21 days. B, Cross section of a tertiary villus showing establishment of circulation in mesenchymal core, cytotrophoblast, and syncytiotrophoblast. SeeSources & Acknowledgments.

The major advantage of CVS compared with midtrimester amniocentesis is that CVS allows the results to be available at an early stage of pregnancy, thus reducing the period of uncertainty and allowing termination, if it is elected, to be performed in the first trimester. However, unlike after amniocentesis, AFAFP cannot be assayed at this stage. Evaluation for a possible open NTD thus must be done by other methods, including MSAFP screening, amniocentesis for AFAFP, and ultrasonography.

The success of chromosome analysis by karyotype or CMA is the same as with amniocentesis (i.e., more than 99%). However, approximately 1% of CVS samplings yield ambiguous results because of chromosomal mosaicism (including true mosaicism and pseudomosaicism; see later); in these situations, follow-up with amniocentesis is recommended to establish whether the fetus has a chromosomal abnormality.


In prenatal diagnostic centers experienced in performing CVS, the rate of fetal loss is only slightly increased over the baseline risk of 2% to 5% in any pregnancy of 7 to 12 weeks of gestation and approximates the 1 in 300 to 1 in 500 risk seen with amniocentesis. Although there were initial reports of an increase in the frequency of birth defects, particularly limb reduction defects, after CVS, this increase has not been confirmed in large series of CVS procedures performed after 10 weeks of gestation by experienced physicians.

Preimplantation Genetic Diagnosis

Preimplantation genetic diagnosis (PGD) refers to testing during in vitro fertilization (IVF) to select embryos free of a specific genetic condition before transfer to the uterus (see Fig. 17-1C). This technology was developed in an effort to offer an alternative option to abortion for those couples at significant risk for a specific genetic disorder or aneuploidy in their offspring, allowing them to undertake a pregnancy even when opposed to pregnancy termination.

The two most common approaches are single blastomere biopsy and blastocyst biopsy. In blastomere biopsy, a single cell is removed from the embryo 3 days after IVF when there are 8 to 16 cells present. For blastocyst biopsy, the fertilized egg is cultured for 5 to 6 days until a blastocyst has developed (see Fig. 17-1C), and approximately five cells are moved from the trophectoderm (but not the inner cell mass, which will develop into the embryo itself; see Chapter 14). Diagnosis by polymerase chain reaction (PCR) has been undertaken for a number of single-gene disorders; chromosome abnormalities can also be detected using fluorescence in situ hybridization or CMA (see Chapters 4 and 5). Embryos that are found not to carry the genetic abnormality in question can then be transferred and allowed to implant, as is routinely done after IVF for assisted reproduction. Affected embryos are discarded. Data currently available on this technology suggest that there are no detrimental effects to embryos that have undergone biopsy.

Although PGD by blastomere biopsy has been performed many thousands of times worldwide, it is not without controversy. First, molecular analysis of a single cell is technically challenging; accuracy varies, with false-positive rates around 6% and false-negative rates around 1%, significantly higher than with analysis of specimens obtained by CVS or amniocentesis. The more recently developed blastocyst biopsy method provides more cellular material, with an apparently greater accuracy, but extensive studies are still ongoing. Second, although PGD was developed to avoid the ethical, religious, and psychological difficulties with pregnancy terminations, it still raises ethical concerns for those who consider the practice of discarding affected embryos as akin to abortion.

Noninvasive Prenatal Diagnosis

Prenatal Diagnosis of Anomalies by Ultrasonography

High-resolution, real-time scanning is widely used for general assessment of fetal age, multiple pregnancies, and fetal viability. Long-term follow-up assessments have failed to provide any evidence that ultrasonography is harmful to the fetus or the mother. The equipment and techniques used by ultrasonographers now allow the detection of many malformations by routine ultrasonography (Figs. 17-3 and 17-4). Once a malformation has been detected or is suspected on routine ultrasound examination, a detailed ultrasound study in three and even four dimensions (three dimensions over time, as with fetal echocardiography) may be indicated. With improvements in ultrasound resolution, an increasing number of structural fetal anomalies can be detected in the late first trimester (Table 17-2; see Fig. 17-3).


FIGURE 17-3 Ultrasonograms of spinal canal and neural tube. A, Normal fetus at 24 weeks of gestation; longitudinal midline view, with the sacrum to the left, thoracic spine to the right. Note the two parallel rows of white echoes that represent the neural arches. Also shown are echoes of the vertebral bodies and the overlying intact skin. B, Fetus with a neural tube defect, clearly showing the meningomyelocele sac protruding through the skin. SeeSources & Acknowledgments.


FIGURE 17-4 Ultrasonograms of hands (arrows). A, Normal fetus. B, Fetus with Holt-Oram syndrome, an autosomal dominant defect with congenital heart defects (often an atrial septal defect) and variable limb abnormalities caused by mutations in the TBX5 transcription factor gene. Note that there are only three obvious fingers and a thumb. The thumb is abnormal in shape (large and thick) and in position. SeeSources & Acknowledgments.

TABLE 17-2

Examples of Fetal Anomalies That Can Be Diagnosed or Ruled Out by Prenatal Diagnostic Ultrasonography


A number of fetal abnormalities detectable by ultrasound examination are associated with chromosomal aneuploidy, including trisomy 21, trisomy 18, trisomy 13, 45,X and many other abnormal karyotypes (Table 17-3). These abnormalities may also occur as isolated findings in a chromosomally normal fetus. Table 17-3 compares the prevalence of fetal chromosome defects in fetuses when one of these common ultrasound examination abnormalities is present as an isolated finding versus when it is one of multiple abnormalities. The likelihood of a chromosomally abnormal fetus increases dramatically when a fetal abnormality detected by ultrasound examination is only one of many abnormalities.

TABLE 17-3

Prevalence of Chromosome Defects in Fetuses with Selected Isolated and Multiple Sonographically Detected Abnormalities


Percent of Fetuses with Abnormal Karyotype


If Isolated Abnormality

If Multiple Abnormalities




Choroid plexus cysts



Cystic hygroma



Nuchal edema



Diaphragmatic hernia



Heart defects



Duodenal atresia






Renal abnormalities



Modified from Snijders RJM, Nicolaides KH: Ultrasound markers for fetal chromosomal defects, New York, 1996, Parthenon.

The finding of a normal fetus can be cautiously reassuring, whereas the identification of a fetus with an abnormality allows the couple the option of either appropriate pregnancy and delivery management or pregnancy termination. Consultation with a clinical genetics unit or perinatal unit should be initiated for counseling and further investigation should multiple congenital anomalies be found by ultrasonography or MRI.

Prenatal Ultrasonography for Diagnosis of Single-Gene Disorders

In some single-gene disorders for which DNA testing is possible but a blood or tissue sample is unavailable for DNA or biochemical studies, diagnostic ultrasonography can be useful for prenatal diagnosis. For example, Figure 17-4B shows an abnormal fetal hand detected by ultrasound examination in a pregnancy at 50% risk for Holt-Oram syndrome, an autosomal dominant disorder characterized by congenital heart disease in association with hand anomalies.

Ultrasonography can also be useful when the risk for a genetic disorder is uncertain and no definitive DNA-based testing is available.

Prenatal Ultrasonography for Diagnosis of Multifactorial Disorders

A number of isolated abnormalities that may recur in families and that are believed to have multifactorial inheritance can also be identified by ultrasonography (see Table 17-2), including neural tube malformations (see Fig. 17-3). Fetal echocardiography is also available at many centers for a detailed assessment of pregnancies at risk for a congenital heart defect (Table 17-4).

TABLE 17-4

Some Examples of Indications for Fetal Echocardiography*


*This list is not comprehensive, and indications vary between centers.

Determination of Fetal Sex

Ultrasound examination can be used to determine fetal sex as early as 13 weeks' gestation. This determination may be an important prelude or adjunct in the prenatal diagnosis of certain X-linked recessive disorders (e.g., hemophilia) for those women identified to be at increased risk. A couple may decide not to proceed with invasive testing if a female (and therefore likely unaffected) fetus is identified by ultrasound examination.

Indications for Prenatal Diagnosis by Invasive Testing

There are a number of well-accepted indications for prenatal testing by invasive procedures (see Box). Because of the increased incidence of certain trisomies with increasing age of the mother, the most common indication for invasive prenatal diagnosis is to test for Down syndrome (trisomy 21) and the two other, more severe autosomal trisomies, trisomy 13 and trisomy 18 (see Chapter 6). For this reason, prenatal diagnosis was most often used in the past in the setting of advanced maternal age. Current clinical guidelines, however, do not support using maternal age as the sole indicator for invasive testing for aneuploidies and, instead, recommend risk assessment be made by one or more of the noninvasive screening methods described later in this chapter.

In addition to fetal chromosome abnormalities, there are over 2000 genetic disorders for which genetic testing is available. Prenatal testing by amniocentesis or CVS can be offered with genetic counseling to couples known to be at risk for any of these disorders, but whether or not a couple considers the fetus to be at significant risk and the condition sufficiently burdensome to justify an invasive procedure and possible pregnancy termination is a personal, individual decision each couple must make for itself.

The traditional clinical approach to invasive prenatal diagnosis is to offer these procedures only in pregnancies for which the fetus has an increased risk for a specific condition, as indicated by family history, a positive screening test result, or other well-defined risk factors (but not maternal age alone). Reserving invasive testing for pregnancies with a documented increased risk for aneuploidy is supported by 2011 Practice Guidelines from the Society of Obstetricians and Gynaecologists of Canada and the International Society for Prenatal Diagnosis. In contrast, the American College of Obstetricians and Gynecologists (ACOG) has recommended that amniocentesis or CVS be made available to all women regardless of age and without a prior screening test indicating increased risk.

It is important to stress that invasive prenatal diagnosis cannot be used to rule out all possible fetal abnormalities. It is limited to determining whether the fetus has (or probably has) a specific condition detectable with the diagnostic testing method being used.

Principal Indications for Prenatal Diagnosis by Invasive Testing

• Previous child with de novo chromosomal aneuploidy or other genomic imbalance

Although the parents of a child with chromosomal aneuploidy may have normal chromosomes themselves, in some situations there may still be an increased risk for a chromosomal abnormality in a subsequent child. For example, if a woman at 30 years of age has a child with Down syndrome, her recurrence risk for any chromosomal abnormality is approximately 1 per 100, compared with the age-related population risk of approximately 1 per 390. Parental mosaicism is one possible explanation of the increased risk, but in the majority of cases, the mechanism of the increase in risk is unknown.

• Presence of structural chromosomal or genome abnormality in one of the parents

Here, the risk for a chromosome abnormality in a child varies according to the type of abnormality and sometimes the parent of origin. The greatest risk, 100% for Down syndrome, occurs only if either parent has a 21q21q Robertsonian translocation (see Chapter 6).

• Family history of a genetic disorder that may be diagnosed or ruled out by biochemical or DNA analysis

Most of the disorders in this group are caused by single-gene defects with 25% or 50% recurrence risks. Cases in which the parents have been diagnosed as carriers after a population screening test, rather than after the birth of an affected child, are also in this category. Mitochondrial disorders pose special challenges for prenatal diagnosis.

• Family history of an X-linked disorder for which there is no specific prenatal diagnostic test

When there is no alternative method, the parents of a boy affected with an X-linked disorder may use fetal sex determination to help them decide whether to continue or to terminate a subsequent pregnancy because the recurrence risk may be as high as 25%. For X-linked disorders, such as Duchenne muscular dystrophy and hemophilia A and B, however, for which prenatal diagnosis by DNA analysis is available, the fetal sex is first determined and DNA analysis is then performed if the fetus is male. In either of the situations mentioned, preimplantation genetic diagnosis (see text) may be an option for allowing the transfer to the uterus of only those embryos determined to be unaffected for the disorder in question.

• Risk for a neural tube defect (NTD)

First-degree relatives (and second-degree relatives at some centers) of patients with NTDs are eligible for amniocentesis because of an increased risk for having a child with an NTD; many open NTDs, however, can now be detected by other noninvasive tests, as described in this chapter.

• Increased risk as determined by maternal serum screening, ultrasound examination, and noninvasive prenatal screening test of cell-free DNA

Genetic assessment and further testing are recommended when fetal abnormalities are suspected on the basis of routine screening by maternal serum screening and fetal ultrasound examination.

• The pregnant woman or couple wishes invasive testing

Although limited at one time to a pregnant woman with no increased risk other than advanced maternal age, some current professional guidelines call for invasive testing to be offered to all couples.

Prenatal Screening

Prenatal screening has traditionally relied on both ultrasonography and measuring various proteins and hormones (referred to as analytes) whose levels in maternal serum are altered when a fetus is affected by a trisomy or an NTD. More recently, the field of prenatal screening and obstetrical genetics has taken a great leap forward with the discovery that maternal serum contains not only useful analytes but also cell-free DNA, of which a certain fraction is fetal in origin. Sequencing of this cell-free DNA using advanced technologies, as discussed later in this chapter, has made noninvasive screening for trisomies more sensitive and accurate compared to traditional analyte screening.

Screening for Neural Tube Defects

The AFAFP test described earlier is indicated for pregnancies that are undergoing amniocentesis due to a known high risk for an open NTD. However, because an estimated 95% of infants with NTDs are born into families with no known history of this malformation, a relatively simple screening test, such as the noninvasive MSAFP test, constitutes an important tool for prenatal diagnosis, prevention, and management.

When the fetus has an open NTD, the concentration of AFP in maternal serum is likely to be higher than normal, just as we saw previously in amniotic fluid. This observation is the basis for the use of MSAFP measurement at 16 weeks as a screening test for open NTDs. There is considerable overlap between the normal range of MSAFP and the range of concentrations found when the fetus has an open NTD (Fig. 17-5). Although an elevated MSAFP concentration is by no means specific to a pregnancy with an open NTD, many of the other causes of elevated MSAFP concentration can be distinguished from open NTDs by fetal ultrasonography (Table 17-5).


FIGURE 17-5 Maternal serum alpha-fetoprotein (AFP) concentration, expressed as multiples of the median, in normal fetuses, fetuses with open neural tube defects, and fetuses with Down syndrome. SeeSources & Acknowledgments.

TABLE 17-5

Causes of Elevated Maternal Serum Alpha-Fetoprotein Concentration


From Cunningham FG, MacDonald PC, Gant NF, et al: Williams obstetrics, ed 20, Stamford, CT, 1997, Appleton & Lange, p 972.

MSAFP is also not perfectly sensitive, because its assessment depends on statistically defined cutoff values. If an elevated concentration is defined as two multiples of the median value in pregnancies without any abnormality that could raise the AFP concentration, one can estimate that 20% of fetuses with open NTDs remain undetected. However, lowering the cutoff to improve sensitivity would be at the expense of reduced specificity, thereby increasing the false-positive rate.

The combined use of the MSAFP assay with detailed diagnostic ultrasonography (see later discussion) approaches the accuracy of AFAFP assay and ultrasonography for the detection of open NTDs. Thus first-degree, second-degree, or more remote relatives of patients with NTDs may have an MSAFP assay (at 16 weeks) followed by detailed ultrasound examination (at 18 weeks) rather than undergoing amniocentesis.

Screening for Down Syndrome and Other Aneuploidies

More than 70% of all children with major autosomal trisomies are born to women who lack known risk factors, including advanced maternal age (see Fig. 6-1). A solution to this problem was first suggested by the unexpected finding that MSAFP concentration (measured, as just discussed, during the second trimester as a screening test for NTD) was depressed in many pregnancies later discovered to have an autosomal trisomy, particularly trisomies 18 and 21. MSAFP concentration alone has far too much overlap between unaffected pregnancies and Down syndrome pregnancies to be a useful screening tool on its own (see Fig. 17-5). However, a battery of maternal serum protein analytes has now been developed that in combination with specific ultrasound measurements has the necessary sensitivity and specificity to be useful for screening. These batteries of tests are now recommended for noninvasive screening, although not for definitive diagnosis, during the first and second trimesters of all pregnancies regardless of maternal age.

First-Trimester Screening

First-trimester screening is ideally performed between 11 and 13 weeks of gestation and relies on measuring the level of certain analytes in maternal serum in combination with a highly targeted ultrasonographic examination. The analytes used are pregnancy-associated plasma protein A (PAPP-A) and the hormone human chorionic gonadotropin (hCG), either as total hCG or as its free β subunit. PAPP-A is depressed below the normal range in all trisomies; hCG (or free β-hCG) is elevated in trisomy 21 but depressed in the other trisomies (Table 17-6). Analyte measurements are combined with ultrasonographic measurement of nuchal translucency (NT), defined as the thickness of the echo-free space between the skin and the soft tissue overlying the dorsal aspect of the cervical spine caused by subcutaneous edema of the fetal neck. An increase in NT is commonly seen in trisomies 21, 13, and 18 and in 45,X fetuses (Fig. 17-6). NT varies with age of the fetus and thus must be determined with reference to gestational age.

TABLE 17-6

Elevation and Depression of Parameters Used in First- and Second-Trimester Screening Tests


AFP, Alpha-fetoprotein; β-hCG, human chorionic gonadotropin β subunit; PAPP-A, pregnancy-associated plasma protein A; uE3, unconjugated estriol.


FIGURE 17-6 Nuchal translucency measurements at 11 weeks of gestation. Nuchal translucency is a dark, echo-free zone beneath the skin in an ultrasonographic “sagittal section” through the fetus and is marked by two “+” signs connected by a yellow line. The average nuchal translucency is 1.2 mm at 11 weeks of gestation (95th percentile up to 2 mm) and 1.5 mm at 14 weeks of gestation (95th percentile up to 2.6 mm). A, Nuchal translucency of 1.2 mm in a normal 11-week fetus, the average for a normal fetus at this gestational age. B, Increased nuchal translucency of 5.9 mm, which is nearly 20 standard deviations above the mean and associated with a greatly increased risk for Down syndrome. SeeSources & Acknowledgments.

Second-Trimester Screening

Second-trimester screening is usually accomplished by measuring hCG in combination with three other analytes: MSAFP, unconjugated estriol, and inhibin A. This battery of tests is referred to as a quadruple screen. All of these substances are depressed below the normal range in all trisomies with the exception of hCG, which is elevated in trisomy 21 but depressed in the other trisomies, and inhibin A, which is elevated in trisomy 21 but not significantly affected in the other trisomies (see Table 17-6). Levels of these analytes can be affected by a number of factors, including race, smoking, IVF pregnancy, and maternal diabetes, and laboratories generally adjust for these variables. Extremely low levels of unconjugated estriol may be indicative of rare genetic conditions such as steroid sulfatase deficiency or the Smith-Lemli-Opitz syndrome.

Noninvasive Prenatal Screening by Analysis of Cell-Free Fetal DNA

The field of prenatal screening and obstetrical genetics is being revolutionized by the joining together of two major advances in the field of genomics, one biological and the other technological, to produce a new prenatal screening technology known as noninvasive prenatal screening (NIPS) (also sometimes referred to as noninvasive prenatal testing, NIPT). The biological discovery is that after 7 weeks of gestation, the serum of a pregnant woman contains fetal DNA that is not contained in the nucleus of a cell but is floating freely in the maternal circulation. Approximately 2% to 10% of the cell-free DNA in maternal blood is derived from the placental trophoblasts and is therefore fetal in origin. This cell-free fetal DNA, although mixed with DNA of maternal origin, provides a sample of the fetal genome that is available for analysis without the need for an invasive procedure. The technological breakthrough is the development of high-throughput sequencing methods that allow the sequencing of millions of individual DNA molecules in a mixture.

NIPS makes highly accurate, noninvasive screening of pregnancies for the common autosomal and sex chromosome aneuploidies possible, with sensitivities and specificities approaching 99% for trisomy 21. Cell-free fetal DNA in maternal serum has also been used to genotype the fetus at the Rh locus (see Chapter 9) and to determine fetal sex. Further refinements in the analysis of cell-free DNA will make noninvasive testing for many other genetic disorders, including many single-gene disorders, available for clinical care in the future.

Sequencing cell-free DNA in maternal serum has been implemented for aneuploidy detection in a number of different ways by different providers; an example designed to illustrate the concept is given here. Total cell-free DNA is subjected to next-generation sequencing, and millions of molecules of DNA are each mapped to its particular chromosome of origin (Fig. 17-7). The number of molecules that map to each chromosome is determined, without knowing which of the fragments is fetal and which maternal. Because chromosome 21 constitutes approximately 1.5% of total DNA in the genome, approximately 1.5% of total fragments should be assigned to chromosome 21 if the fetus and mother have two normal copies of chromosome 21. If, however, the fetus has trisomy 21, more sequences than expected will map to chromosome 21—a small but significant increase relative to the number of sequences that map to an appropriate reference chromosome or to the full set of chromosomes not including chromosome 21. A similar calculation can be used for the other common autosomal trisomies and for sex chromosome aneuploidies as well.


FIGURE 17-7 Schematic diagram of noninvasive prenatal screening (NIPS) for trisomies by next-generation sequencing of cell-free DNA in maternal serum. Fetal component of maternal serum cell-free DNA shown in red, maternal contribution in blue. Millions of molecules of DNA are sequenced and assigned to each chromosome by computerized alignment against the human genome. Highly accurate measurements of small but significant increases in the fraction of molecules assigned to chromosome 13, 18, 21 or X compared to a reference indicate increased risk for trisomy of each of these chromosomes.

Although NIPS provides a substantial improvement in sensitivity and specificity for fetal trisomies, particularly trisomy 21, and sex chromosome aneuploidies, it remains a screening test, not a diagnostic test. NIPS can also be used to detect Y chromosome sequences in maternal serum for the purposes of determining fetal sex; the test has false-positive and false-negative rates in the 1% to 2% range.

Integrated Screening Strategies

For standard first-trimester and second-trimester screening by ultrasonography and maternal serum analytes, a cutoff, chosen to keep false positives at 5%, results in sensitivities of first- and second-trimester screening, as shown in Table 17-7. Based on these parameters, a strategy was developed for combining the results of first-trimester and second-trimester testing to increase the ability to detect pregnancies with autosomal trisomies, particularly trisomy 21 (Fig. 17-8). These strategies have the advantage of giving couples found to be at significantly increased risk on the basis of first-trimester testing alone the choice of early invasive testing by CVS, rather than having to wait for second-trimester screening and use amniocentesis. The most common strategy, however, is to combine the risk as determined from first- and second-trimester screening tests in a sequential manner (see Fig. 17-8). In this stepwise sequential strategy, couples are identified as “screen positive” for Down syndrome once an ultrasound examination has confirmed fetal age and the estimated risk is found to be elevated. A couple showing increased risk by serum analyte screening can then be offered either NIPS or prenatal chromosome analysis (see Fig. 17-8). Without NIPS, this strategy can detect up to 95% of all Down syndrome cases with an approximately 5% false-positive rate. If NIPS is added, the sensitivity for trisomy 21 rises to greater than 99% with a less than 1% false-positive rate. Sensitivity for other trisomies is in the 90% to 95% range, but still with a remarkably low false-positive rate of less than 1%. Although NIPS is relatively new and more data are needed, initial measurements of sensitivity and false-positive rate for NIPS appear to provide improved screening parameters compared to currently available standard serum analyte screening. If this remarkable sensitivity and specificity are borne out, it is anticipated that NIPS may replace serum analyte screening for aneuploidies; MSAFP screening would remain, however, for NTDs.

TABLE 17-7

Sensitivity and False-Positive Rates for Trisomy 21 for Various Prenatal Screening Methods

Screening Test


False Positive Rate (1 Specificity)

First-trimester triple screen



Second-trimester quadruple screen



Combined first and second trimester



Noninvasive prenatal screening



Modified from Malone FD, Canick JA, Ball RH, et al: First-trimester or second-trimester screening, or both, for Down's syndrome, N Engl J Med 353:2001-2011, 2005; and Bianchi DW, Parker RL, Wentworth J, et al: DNA sequencing versus standard prenatal aneuploidy screening, N Engl J Med 370:799-808, 2014.


FIGURE 17-8 Prenatal screening decision tree used by physicians and patients to decide on which screening and diagnostic modalities to use to decide whether invasive testing is indicated. A number of different options shown as different alternative pathways are available at various points in the tree.

As with any screening test in medicine, it is critical for couples to be informed that screening for birth defects with measurement of maternal serum analytes, ultrasound scanning, and NIPS is a screening tool and not a definitive diagnostic test. They must also be counseled that screening tests will not reliably detect chromosome abnormalities other than the common trisomies and sex chromosome aneuploidies, mosaicism, or single-gene defects. Furthermore, only the second-trimester quadruple screen, which includes MSAFP, is helpful for detecting an open NTD in the fetus. Finally, women whose screening test result is considered to be “negative” must also be counseled that their risk for having a child with Down syndrome or another aneuploidy or NTD, although greatly reduced, is not zero.

Laboratory Studies

Cytogenetics in Prenatal Diagnosis

Either amniocentesis or CVS can provide fetal cells for karyotyping as well as for biochemical or DNA analysis. PGD, being a single-cell technique, has been used only for a limited number of DNA analyses and cannot be used for biochemical studies. Preparation and analysis of chromosomes from cultured amniotic fluid cells or cultured chorionic villi require 7 to 10 days, although chorionic villi can also be used for karyotyping after short-term incubation. Although short-term incubation using rapid metaphase analysis of villous cytotrophoblast tissue provides a result more quickly, it yields relatively poorer quality preparations, in which the banding resolution is inadequate for detailed analysis, as well as a higher rate of mosaicism. Some laboratories use both techniques, but if only one is used, long-term culture of the cells of the mesenchymal core is the technique of choice at present.

Fluorescence in situ hybridization (see Chapters 4 and 5) makes it possible to screen interphase nuclei in fetal cells for the common aneuploidies of chromosomes 13, 18, 21, X, and Y immediately after amniocentesis or CVS. This approach for prenatal cytogenetic assessment requires 1 to 2 days and can be used when rapid aneuploidy testing is indicated.

CMA (see Chapter 5) is replacing karyotyping for prenatal diagnosis under certain circumstances. Copy number variants (CNVs), including chromosome mutations, such as duplications, triplication, and deletions (see Chapter 4), can be detected at much higher resolution by CMA than can be accomplished even with high-resolution karyotyping. Both ACOG and the Society of Obstetricians and Gynaecologists of Canada have advised that CMA, rather than karyotyping, should be the first-line test when a fetal abnormality is detected by ultrasonography. ACOG, however, goes further by recommending that all women having invasive testing be given the option to have CMA, whether or not a structural abnormality is detected by ultrasonography. The difference between CMA being the first-line test when an anomaly is present versus simply being offered as an option when invasive testing of any kind is done reflects the fact that high-resolution genome analysis by CMA detects many CNVs of currently uncertain clinical significance. The number of false-positive CMA tests will be lower in a fetus with an anomaly than in one without because the prior probability that a CNV will be of clinical significance is higher when an anomaly is present. As experience and knowledge of copy number variation in the human genome improves (see Chapter 4), the medical relevance of a greater and greater fraction of CNVs will become clearer and the incidence of variants of uncertain significance will fall to levels that will result in CMA replacing fetal karyotyping for nearly all indications.

Chromosome Analysis after Ultrasonography

Because some birth defects detectable by ultrasonography are associated with chromosome abnormalities, CMA of amniotic fluid cells, chorionic villus cells, or (much more rarely) fetal blood cells obtained by insertion of a needle into an umbilical vessel (cordocentesis) may be indicated after ultrasonographic detection of such an abnormality. Chromosome abnormalities are more frequently found when multiple rather than isolated malformations are detected (see Table 17-3). The karyotypes most often seen in fetuses ascertained by abnormal ultrasonographic findings are the common autosomal trisomies (21, 18, and 13), 45,X (Turner syndrome), and unbalanced structural abnormalities. The presence of a cystic hygroma can indicate a 45,X karyotype, but it can also occur in Down syndrome and trisomy 18, as well as in fetuses with normal karyotypes. Thus complete chromosome assessment is indicated.

Problems in Prenatal Chromosome Analysis


Mosaicism refers to the presence of two or more cell lines in an individual or tissue sample (see Chapter 7). Because invasive prenatal techniques, particularly CVS, sample extraembryonic tissues of the placenta, and not the fetus itself, mosaicism found in cultured fetal cells may be difficult to interpret. The prenatal geneticist must determine if the fetus itself is truly mosaic and understand the clinical significance of any apparent mosaicism.

Cytogeneticists distinguish three levels of mosaicism in amniotic fluid or CVS cell cultures:

1. Mosaicism detected in multiple colonies from several different primary cultures is considered true mosaicism. Postnatal studies have confirmed that true mosaicism in culture is associated with a high risk that mosaicism is truly present in the fetus. The probability varies with different situations, however; mosaicism for structural aberrations of chromosomes, for example, is hardly ever confirmed.

2. Mosaicism involving several cells or colonies of cells from a single primary culture is difficult to interpret, but it is generally thought to reflect pseudomosaicism that has arisen in culture.

3. When apparent mosaicism is restricted to only a single cell, it is considered to reflect pseudomosaicism and is typically disregarded.

Maternal cell contamination is a possible explanation of some cases of apparent mosaicism in which both XX and XY cell lines are present. This problem is more common in long-term CVS cultures than in amniotic fluid cell cultures, as a consequence of the intimate association between the chorionic villi and the maternal tissue (see Fig. 17-2). To minimize the risk for maternal cell contamination, any maternal decidua present in a chorionic villus sample must be carefully dissected and removed, although even the most careful dissection of chorionic villi does not eliminate every cell of maternal origin. When maternal cell contamination is suspected and cannot be disproved (e.g., by genotyping with use of polymorphisms), amniocentesis is recommended to allow a second chromosome analysis.

In CVS studies, discrepancies between the karyotypes found in the cytotrophoblast, villous stroma, and fetus have been reported in 1% to 2% of pregnancies studied at 10 to 11 weeks of gestation. Mosaicism is sometimes present in the placenta but absent in the fetus, a situation termed confined placental mosaicism (Fig. 17-9). On occasion, a liveborn infant or fetus with nonmosaic trisomy 13 or trisomy 18 has been reported in a pregnancy with placental mosaicism with both the trisomic cell line and a normal cell line. This finding suggests that when the zygote is trisomic, a normal placental cell lineage, established by postzygotic loss of the additional chromosome in a progenitor cell of the cytotrophoblast, can improve the probability of intrauterine survival of a trisomic fetus.


FIGURE 17-9 The different types of mosaicism that may be detected by prenatal diagnosis. A, Generalized mosaicism affecting both the fetus and placenta. B, Confined placental mosaicism with normal (red) and abnormal (green) cell lineages present. C, Confined placental mosaicism with only an abnormal cell lineage present. D, Mosaicism confined to the embryo. SeeSources & Acknowledgments.

Confined placental mosaicism for any chromosome (but particularly for trisomy 7, 11, 14, or 15) raises the additional concern that the fetal diploidy may have actually arisen by trisomy rescue. This term refers to the loss of an extra chromosome postzygotically, an event that presumably allows fetal viability. If the fetus has retained two copies of a chromosome from the same parent, however, the result is uniparental disomy (see Chapter 5). Because some genes on the chromosome mentioned are imprinted, uniparental disomy must be excluded; two maternal copies of chromosome 15, for example, cause Prader-Willi syndrome, and two paternal copies are associated with Angelman syndrome (see Chapter 5).

CMA can detect some, but not all, cases of mosaicism. Because CMA uses pooled DNA from tissues or cultured cells and does not examine individual cells the way karyotyping does, it is less sensitive for the detection of mosaicism. Mosaicism in which 10% of the cells are aneuploid is difficult to detect as a copy number change by CMA, whereas 10% mosaicism will be detected with greater than 99% probability when 50 cells are examined by karyotype, as is typically done for the assessment of possible mosaicism. CMA is even less sensitive for detecting mosaicism for a copy number variation of only a segment of a chromosome unless it constitutes more than 20% to 25% of the cells under study.

Confirmation and interpretation of apparent mosaicism are among the most difficult challenges in genetic counseling for prenatal diagnosis because, at present, clinical outcome information on the numerous possible types and extents of mosaicism is limited. Further studies (e.g., amniocentesis that follows CVS, cordocentesis that follows amniocentesis) as well as the medical literature may provide some guidance, but the interpretation sometimes still remains uncertain. Ultrasonographic scanning may provide some reassurance if normal growth is observed and if no congenital anomalies can be demonstrated.

Parents should be counseled in advance of the possibility that mosaicism may be found and that the interpretation of mosaicism may be uncertain. After birth, an effort should be made to verify any abnormal chromosome findings suspected on the basis of prenatal diagnosis. In the case of termination, verification should be sought by analysis of fetal tissues. Confirmation of mosaicism, or lack thereof, may prove helpful with respect to medical management as well as for genetic counseling of the specific couple and other family members.

Culture Failure.

If couples are to have an opportunity to consider termination of a pregnancy when an abnormality is found in the fetus, they should be provided with the information at the earliest possible time. Because prenatal diagnosis is always a race against time, the rate of culture failure can be a concern; fortunately, this rate is low. When a CVS culture fails to grow, there is time to repeat the chromosome study with amniocentesis. If an amniotic fluid cell culture fails, either repeated amniocentesis or cordocentesis can be offered, depending on fetal age.

Unexpected Adverse Findings.

On occasion, prenatal chromosome analysis performed primarily to rule out aneuploidy reveals some other unusual chromosome finding, for example, a rare chromosomal rearrangement or a marker chromosome (see Chapter 5). In such a case, because the significance of the finding in the fetus cannot be assessed until the parental karyotypes are known, both parents should be karyotyped to determine whether the finding seen in the fetus is de novo or inherited. Unbalanced or de novo structural rearrangements may cause serious fetal abnormalities (see Chapter 6). If one parent is found to be a carrier of a structural rearrangement seen in unbalanced form in the fetus, the consequences for the fetus may be serious. On the other hand, if the same finding is seen in a normal parent, it is likely to be a benign change without untoward consequences. Potential exceptions to this guideline include the possibility of uniparental disomy in a region of the genome that contains imprinted genes. In this situation, an inherited balanced rearrangement may cause serious fetal abnormalities. This possibility can be excluded if there has been a previous transmission of the same balanced rearrangement from a parent of origin of the same sex as the transmitting parent in the current pregnancy.

Biochemical Assays for Metabolic Diseases

Although any disorder for which the genetic basis is known and the responsible mutation(s) identified can be diagnosed prenatally by DNA analysis, more than 100 metabolic disorders can also be diagnosed by biochemical analysis of chorionic villus tissue or cultured amniotic fluid cells; a few rare conditions can even be identified directly by assay of a substance in amniotic fluid. Most metabolic disorders are rare in the general population but have a high recurrence risk (usually 25% within sibships, because most are recessive conditions). Because each condition is rare, the experience of the laboratory performing the prenatal diagnostic testing is important; thus referral to specialized centers is preferable. Whenever possible, biochemical assay on direct chorionic villus tissue (as opposed to cultured tissue) is preferred to avoid misinterpretation of results due to the expansion in culture of contaminating maternal cells. Access to a cultured cell line from a proband in the family is highly advisable so that the laboratory can confirm its ability to detect the biochemical abnormality in the proband before the assay is attempted in CVS or amniotic fluid cells from the pregnancy at risk.

Of course, many metabolic disorders cannot be diagnosed by enzyme assay of chorionic villus tissue or cultured amniotic fluid cells because the enzyme is not expressed in those tissues or a reliable in vitro biochemical assay has not been developed. For these disorders, DNA sequencing to look for the pathogenic mutations responsible can be performed. Nonetheless, biochemical tests have one significant advantage over DNA analysis in some cases: whereas DNA analysis by direct detection of a mutation is accurate only for that mutation and not for other alleles at the locus, biochemical testing can detect abnormalities caused by any mutant allele that has a significant effect on the function of the protein. This advantage is particularly significant for disorders characterized by a high degree of allelic heterogeneity, genes in which mutant alleles occur in regions of the gene that are not routinely sequenced, or by a high proportion of new mutations. In addition, biochemical testing may be the only option for prenatal diagnosis if the causative mutations in the family have not been identified or are unknown.

Fetal DNA and Fetal Genome Analysis

As the specific basis for an increasing number of inherited disorders is determined (see Chapter 12), many conditions (some of which were not previously detectable prenatally by other means) can now be diagnosed prenatally by analysis of fetal DNA. Any technique used for direct mutation screening can be used for prenatal diagnosis, from allele-specific or gene-specific tests to whole-exome or whole-genome sequencing. As of the beginning of 2015, genetic testing registries report the clinical availability of DNA-based prenatal testing for more than 5000 genetic disorders caused by mutations in over 3500 genes. The degree of certainty of the diagnosis approaches 100% when direct detection of a mutation is possible, but testing will fail if the disorder in the patient is due to a mutation different from the one that is being sought.

Numerous diseases cannot yet be diagnosed prenatally, but every month additional disorders are added to the list of conditions for which prenatal diagnosis is possible either by biochemical testing or by DNA analysis. One of the contributions of medical geneticists to medical practice in general is keeping up with these rapid changes and serving as a central source of information about the current status of prenatal testing.

Prenatal diagnosis by DNA analysis may not be predictive of the exact clinical presentation in an affected pregnancy in the case of disorders characterized by variable expressivity. For example, in neurofibromatosis type 1 (Case 34), a specific mutation may lead to a severe clinical manifestation in one family member and a mild manifestation in another. Mitochondrial disorders (see Chapters 7 and 12) that result from mutations in mitochondrial DNA are particularly challenging for prenatal counseling because the mutations are almost always heteroplasmic, and it is difficult to predict the fraction of defective mitochondrial genomes any one fetus will inherit. Although there is uncertainty concerning the degree of heteroplasmy that will be passed on from mother to fetus, DNA analysis of samples from the fetus obtained by CVS or amniocentesis is likely to reflect the overall degree of heteroplasmy in the fetus and therefore should be a reliable indicator of the burden of pathogenic mitochondrial mutations in the fetus.

Although whole-exome or whole-genome sequencing of fetal DNA is not yet part of routine care, it is technically feasible now, and discussions are ongoing concerning whether whole-genome analysis by exome or genome sequencing of fetal DNA could serve as a prenatal screening test (see earlier discussion of NIPS). The ethical concerns posed by whole-genome analysis of fetuses are substantial. These include presymptomatic diagnosis of adult disorders, particularly those for which no treatments are known, stigmatization, damage to the parent-child relationship, and the impact of having to provide counseling for massive amounts of currently uninterpretable information arising from the discovery of variants of uncertain significance. This will be an area that bears watching closely in the years ahead, with increasingly important ethical and policy implications for the practice of fetal medicine and prenatal genetics.

Genetic Counseling for Prenatal Diagnosis and Screening

The majority of genetic counselors practice in the setting of a prenatal diagnosis program. The professional staff of a prenatal diagnosis program (physician, nurse, and genetic counselor) must obtain an accurate family history and determine whether other previously unsuspected genetic problems should also be considered on the basis of family history or ethnic background.

Ethnic background, even in the absence of a positive family history, may indicate the need for carrier tests in the parents in advance of prenatal diagnostic testing. For example, in a couple referred for any reason, one must discuss carrier testing for autosomal recessive disorders with increased frequency in various ethnic groups. Such disorders include thalassemia in individuals of Mediterranean or Asian background, sickle cell anemia in Africans or African Americans, and various disorders in the fetus of an Ashkenazi Jewish couple. However, because it is becoming increasingly difficult to assign a single ethnicity to each patient, the use of universal carrier screening panels, in which patients are tested for a large array of genetic disorders irrespective of apparent or stated ethnicity, are becoming more and more common.

The complexities posed by the availability of different tests (including the distinction between screening tests and diagnostic tests), the many different and distinctive indications for testing in different families, the subtleties of interpretation of test results, and the personal, ethical, religious, and social issues that enter into reproductive decision making all make the provision of prenatal diagnosis services a challenging arena for counselors. Parents considering prenatal diagnosis for any reason need information that will allow them to understand their situation and to give or withhold consent for the procedure. Genetic counseling of candidates for prenatal diagnosis usually deals with the following:

• The risk that the fetus will be affected

• The nature and probable consequences of the specific problem

• The risks and limitations of the procedures to be used

• The time required before a report can be issued

• The possible need for a repeated procedure in the event of a failed attempt

In addition, the couple must be advised that a result may be difficult to interpret, further tests and consultation may be required, and even then the results may not necessarily be definitive.

Elective Pregnancy Termination

In most cases, the findings in prenatal diagnosis are normal, and parents to be are reassured that their baby will be unaffected by the condition in question. Unfortunately, in a small proportion of cases, the fetus is found to have a serious genetic defect. Because effective prenatal therapy is not available for most disorders (see Chapter 13), the parents may then choose to terminate the pregnancy. Few issues today are as hotly debated as elective abortion, but despite legal restrictions in some jurisdictions, elective abortion is widely used. Among all elective abortions, those performed because of prenatal diagnosis of an abnormality in a fetus account for only a very small proportion. Without a means of legal termination of pregnancy, prenatal diagnosis would not have developed into the accepted medical procedure that it has become.

Some pregnant women who would not consider termination nevertheless request prenatal diagnosis to reduce anxiety or to prepare for the birth of a child with a genetic disorder. This information may be used for psychological preparation as well as for management of the delivery and of the newborn infant.

At the level of public health, prenatal diagnosis combined with elective termination has led to a major decline in the incidence in certain population groups of a few serious disorders, such as β-thalassemia (see Chapter 11) and Tay-Sachs disease (see Chapter 12). Similar data for the effects of prenatal screening, diagnosis, and elective termination on the birth incidence of Down syndrome in the United States are conflicting, however. Estimates range from a 24% increase to a 15% decrease in the numbers of babies born with Down syndrome over the 15- to 20-year time period ending in 2005. These data must be viewed against an estimated 34% increase in affected pregnancies that was expected due to a rise in average maternal age. The frequency with which couples carrying a Down syndrome pregnancy terminate a pregnancy also varies tremendously between different societies. For example, whereas approximately two thirds of couples in the United States choose to terminate a Down syndrome pregnancy, nearly 90% of couples in the United Kingdom choose to terminate such pregnancies.

Impact of Prenatal Diagnosis

It must be stressed, however, that the principal advantage of prenatal diagnosis is not to the population but to the immediate family. Parents at risk for having a child with a serious abnormality can undertake pregnancies that they may otherwise not have risked, with the knowledge that they can learn early in a pregnancy whether the fetus has the abnormality and can make an informed decision about whether or not to continue the pregnancy.

Although the great majority of prenatal diagnoses end in reassurance, options available to parents in the event of an abnormality—of which termination of pregnancy is only one—should be discussed. Above all, the parents must understand that in undertaking prenatal diagnosis, they are under no implied obligation to terminate a pregnancy in the event that an abnormality is detected. The objective of prenatal diagnosis is to determine whether the fetus is affected or unaffected with the disorder in question. Diagnosis of an affected fetus may, at the least, allow the parents to prepare emotionally and medically for the management of a newborn with a disorder.

In closing, the reader is again cautioned that because of technological advances in the methods available for assessing the fetus and the fetal genome, and because of ongoing discussions of social and ethical norms and governmental policies concerning prenatal diagnosis in different cultures and countries around the globe, standards of care in prenatal screening and diagnosis will continue to be subject to modification and refinement.

General References

Gardner RJM, Sutherland GR, Shaffer LG. Chromosome abnormalities and genetic counseling. ed 4. Oxford University Press: New York; 2011.

Milunsky A, Milunsky J. Genetic disorders and the fetus: diagnosis, prevention, and treatment. ed 6. Wiley-Blackwell: Chichester, West Sussex, England; 2010.

Specific References

American College of Obstetricians and Gynecologists Committee on Genetics. The use of chromosomal microarray analysis in prenatal diagnosis. Obstet Gynecol. 2009;122:1374–1377.

American College of Obstetricians and Gynecologists Committee on Genetics. Noninvasive prenatal testing for fetal aneuploidy. Obstet Gynecol. 2012;120:1532–1534.

Bianchi D. From prenatal genomic diagnosis to fetal personalized medicine: progress and challenges. Nat Med. 2012;18:1041–1051.

Bianchi DW, Parker RL, Wentworth J, et al. DNA sequencing versus standard prenatal aneuploidy screening. N Engl J Med. 2014;370(9):799–808.

Bodurtha J, Strauss JF. Genomics and perinatal care. N Engl J Med. 2012;366:64–73.

Chitayat D, Langlois S, Wilson RD, et al. Prenatal screening for fetal aneuploidy in singleton pregnancies. J Obstet Gynaecol Can. 2011;33:736–750.

Dugoff L. Application of genomic technology in prenatal diagnosis. N Engl J Med. 2012;367:2249–2251.

Duncan A, Langlois S, SOGC Genetics Committee, et al. Use of array genomic hybridization technology in prenatal diagnosis in Canada. J Obstet Gynaecol Can. 2011;33:1256–1259.

Fan HC, Gu W, Wang J, et al. Non-invasive prenatal measurement of the fetal genome. Nature. 2012;487:320–324.

Gregg A, Gross SJ, Best RG, et al. ACMG statement on noninvasive prenatal screening for fetal aneuploidy. Genet Med. 2013;15:395–398.

Malone FD, Canick JA, Ball RH, et al. First-trimester and second-trimester screening, or both, for Down's Syndrome. N Engl J Med. 2005;353:2001–2011.

McArthur SJ, Leigh D, Marshall JT, et al. Blastocyst trophectoderm biopsy and preimplantation genetic diagnosis for familial monogenic disorders and chromosomal translocations. Prenat Diagn. 2008;28:434–442.

Norwitz ER, Levy B. Noninvasive prenatal testing: the future is now. Rev Obstet Gynecol. 2013;6:48–62.

Talkowski ME, Ordulu Z, Pillalamarri V, et al. Clinical diagnosis by whole-genome sequencing of a prenatal sample. N Engl J Med. 2012;367:2226–2232.

Wapner RJ, Martin CL, Levy B, et al. Chromosomal microarray versus karyotyping for prenatal diagnosis. N Engl J Med. 2012;367:2175–2184.

Yurkiewicz IR, Korf BR, Lehmann LS. Prenatal whole-genome sequencing – is the quest to know a fetus’ future ethical? N Engl J Med. 2014;370:195–197.


1. Match the term in the top section with the appropriate comment in the bottom section.

a. Rh immune globulin

b. 10th week of pregnancy

c. Cordocentesis

d. Mosaicism

e. 16th week of pregnancy

f. Alpha-fetoprotein in maternal serum

g. Aneuploidy

h. Cystic hygroma

i. Chorionic villi

j. Amniotic fluid

________ method of obtaining fetal blood for karyotyping

________ usual time at which amniocentesis is performed

________ increased level when fetus has neural tube defect

________ contains fetal cells viable in culture

________ major cytogenetic problem in prenatal diagnosis

________ ultrasonographic diagnosis indicates possible Turner syndrome

________ risk increases with maternal age

________ usual time at which chorionic villus sampling (CVS) is performed

________ derived from extraembryonic tissue

________ used to prevent immunization of Rh-negative women

2. A couple has a child with Down syndrome, who has a 21q21q translocation inherited from the mother. Could prenatal diagnosis be helpful in the couple's next pregnancy? Explain.

3. Cultured cells from a chorionic villus sample show two cell lines, 46,XX and 46,XY. Does this necessarily mean the fetus is abnormal? Explain.

4. What two main types of information about a fetus can be indicated (although not proved) by assay of alpha-fetoprotein, human chorionic gonadotropin, and unconjugated estriol in maternal serum during the second trimester?

5. A couple has had a first-trimester spontaneous abortion in their first pregnancy and requests counseling.

a. What proportion of all pregnancies abort in the first trimester?

b. What is the most common genetic abnormality found in such cases?

c. Assuming that there are no other indications, should this couple be offered prenatal diagnosis for their next pregnancy?

6. A young woman consults a geneticist during her first pregnancy. Her brother was previously diagnosed with Duchenne muscular dystrophy and had since died. He was the only affected person in her family. The woman had been tested biochemically and found to have elevated creatine kinase levels, indicating she is a carrier of the disease.

Unfortunately, no DNA analysis had been conducted on the woman's brother to determine what type of mutation in the DMD gene he had. The woman was investigated by molecular analysis and found to be heterozygous (A1/A2) for a microsatellite marker closely linked to the DMD gene. No relatives except the parents of the woman were available for analysis.

a. Can the phase of the mutation in the woman be determined from analysis of the available individuals?

b. Can this information be used to diagnose her pregnancy?

c. What other molecular analysis could be performed on the fetus?

7. Discuss the relative advantages and disadvantages of the following diagnostic procedures, and cite types of disorders for which they are indicated or not indicated: amniocentesis, CVS, first-semester maternal serum screening, second trimester screening, noninvasive screening of cell-free fetal DNA (noninvasive prenatal screening [NIPS]).

8. Suppose the frequency of Down syndrome is 1 in 600 in pregnancies in women younger than 35 years. Consider the following two strategies for prenatal detection of the disorder:

• All pregnant women younger than 35 years are offered CVS or amniocentesis.

• All pregnant women undergo a sequential screening strategy, as follows: All participate in first-trimester screening with pregnancy-associated plasma protein A (PAPP-A), human chorionic gonadotropin (hCG), and nuchal translucency. Sensitivity is 84% with a false-positive rate of 5%. Those who score positive are offered CVS, and all use it. Those who score negative are screened during the second trimester with a quadruple maternal serum screening, which has 81% sensitivity and a 5% false-positive rate. Those who score positive are offered amniocentesis, and all use it.

Assuming that a population of 600,000 women younger than 35 years are pregnant:

a. How many CVS procedures or amniocenteses are done overall, given these two strategies?

b. What fraction of the total expected number of affected fetuses is detected under the two strategies? What fraction is missed?

c. How many CVS or amniocentesis procedures would need to be done to detect one fetus with Down syndrome under these two strategies?