Prenatal Diagnosis

PRENATAL DIAGNOSIS

Lorraine Dugoff in Danforth's Obstetrics and Gynecology, 10th Edition. Lippincott Williams & Wilkins. 2008

The ongoing advances in molecular genetics and discoveries in cytogenetics are increasing the potential indications for prenatal diagnosis. It is now possible to identify hundreds of genetic syndromes prenatally by using technology including florescence in situ hybridization (FISH), mutation analysis, and biochemical markers. There have been recent advances in prenatal screening for Down syndrome. The use of nuchal translucency and first-trimester maternal serum marker screening has made it possible to detect 85% of cases of Down syndrome as early as 10 weeks gestation. This chapter will review prenatal screening for aneuploidy and neural tube defects and the invasive techniques currently available in prenatal diagnosis, including chorionic villus sampling (CVS), amniocentesis, and percutaneous umbilical blood sampling.

Screening for Chromosomal Abnormalities

Down syndrome occurs in approximately 1 of 800 live births. In 95% of cases, it is a result of meiotic nondisjunction of the chromosome 21 pair, usually in the mother's gamete, resulting in a 47, +21 karyotype. The risk of a fetus with Down syndrome, as well as trisomy 13 and 18, increases with maternal age. Four percent of cases of Down syndrome result from a translocation, and approximately 1% result from mosaicism. These cases are not related to advanced maternal age.

A number of maternal serum markers have proven useful in screening for Down syndrome. Historically, a maternal age at delivery of 35 years was used as a cutoff to identify women at the highest risk for having a baby with Down syndrome. Various combinations of serum biochemical markers have been used to screen for Down syndrome since 1984, when it was found that low second-trimester maternal serum α-fetoprotein (MSAFP) levels were associated with Down syndrome. In the 1990s, it was reported that elevated human chorionic gonadotropin (hCG) levels and decreased unconjugated estriol (uE₃) levels were associated with Down syndrome. The combination of these three markers in combination with maternal age, the triple screen, or triple test yields a 69% detection rate for Down syndrome at a 5% positive screen rate. A fourth marker, inhibin A, which may be increased in the serum of women carrying a fetus with Down syndrome, further increases the detection rate for Down syndrome in the second trimester. When inhibin A is included in the second-trimester screening test, known as the quadruple or quad screen, the estimated detection rate increases to 81% with a 5% false positive rate. The triple and quadruple screens should ideally be offered between 15 and 18 weeks gestation, although they can be performed between 15 and 22 weeks. It is critical to know the precise gestational age, because the median values for the biochemical markers and the risk ratios are based on gestational age.

First-trimester screening for Down syndrome using fetal nuchal translucency, a measurement obtained by ultrasound, and maternal serum markers, pregnancy-associated plasma protein A (PAPP-A) and the free beta subunit of human chorionic gonadotropin (β-hCG) in conjunction with maternal age (combined screening), yields a detection rate of approximately 85% at a 5% false-positive rate. The optimal gestational age for first-trimester screening appears to be 11 weeks, as the detection rate may be the highest (87%) at this time. First-trimester screening may be performed between 10 weeks/3 days and 13 weeks/6 days. The first-trimester combined screen may also be used to screen for trisomy 18. Nuchal translucency alone is associated with detection rates of 75% rate for trisomy 18, 72% for trisomy 13, 87% for Turner syndrome, 59% for triploidy, and 55% for other significant chromosomal abnormalities.

Down syndrome screening strategies that involve a combination of first- and second-trimester markers yield the highest detection rates. There are a variety of possible approaches to combined first- and second-trimester screening. The various Down screening tests and their detection rates are listed in Table 1. The integrated screen determines a Down syndrome risk assessment based on a combination of maternal age, first-trimester nuchal translucency, and PAPP-A and the second-trimester quad screen markers. The patient is provided with a single risk for Down syndrome after the quad screen has been interpreted. The integrated screen yields a 94% to 96% detection rate at a 5% positive screen rate. A potential disadvantage of the integrated screen is that the patient does not receive any information regarding Down syndrome risk until the second trimester. The serum integrated screen is similar to the integrated screen except that the patient does not have a nuchal translucency measurement in the first trimester. The serum integrated screen is an effective screening option for patients who do not have access to a center that can perform nuchal translucency measurement This yields a detection rate of 88% at a 5% positive screen rate. The stepwise sequential screen consists of the measurement of nuchal translucency, PAPP-A, and β-hCG in the first trimester and the quad screen in the second trimester. The results are provided to the patient after each test. An advantage of the stepwise sequential screen is that Down syndrome risk assessment is provided after the first-trimester screen, which gives the patient the option of having CVS if her initial risk is high. The stepwise sequential screen has a 95% detection rate at a 5% positive screen rate. The contingent sequential screen determines an initial Down syndrome risk based on first-trimester nuchal translucency, PAPP-A, and β-hCG measurements. Women with the highest risk from first-trimester screening are offered invasive testing by CVS, and women with the lowest risks are told that second-trimester testing is not necessary. Women with intermediate risks after the first-trimester screen have their risks reassessed by integrating their first-trimester results with second-trimester quad screen results.

TABLE 1 Down Syndrome Screening Tests and Detection Rates (at a 5% positive screen rate)

Screening Test Detection Rate (%) First trimester NT measurement 64-70 NT, PAPP-A, free or total β-hCG (combined screen) 82-87 Second trimester MSAFP, hCG, unconjugated estriol (triple screen) 69 MSAFP, hCG, unconjugated estriol, inhibin A (quad screen) 81 First and second trimester Integrated (NT, PAPP-A, quad screen) 94-96 Serum integrated (PAPP-A, quad screen) 85-88 Stepwise sequential 95 First-trimester test result: Positive: CVS offered Negative: quad screen offered Final: risk assessment incorporates first and second results Contingent sequential 88 to 94 First-trimester test result: Positive: CVS offered Negative: no further testing Intermediate: quad screen offered Final: risk assessment incorporates first and second results NT, nuchal translucency; PAPP-A, pregnancy-associated plasma protein A; β-hCG, beta human chorionic gonadotropin; MSAFP, maternal serum alpha feto protein; CVS, chorionic villus sampling. American College of Obstetricians and Gynecologists. Clinical management guidelines for obstetrician-gynecologists. Screening for chromosomal abnormalities. ACOG Practice Bulletin No. 77, January 2007, with permission.

In cases of first-trimester screening where the fetal nuchal translucency is 3.5 mm or greater, patients should be offered a targeted ultrasound examination and fetal echocardiogram. In addition to the increased risk for aneuploidies, these fetuses are at increased risk for having structural abnormalities, including heart defects as well as genetic syndromes.

Women with an increased nuchal translucency measurement or abnormal first-trimester serum markers may be at increased risk for adverse obstetric outcomes, including preeclampsia, preterm birth, low birth weight, spontaneous fetal loss before 24 weeks gestation, and fetal demise later in gestation. Currently, there are no data to indicate whether or not fetal surveillance in the later pregnancy will be helpful in the care of these patients.

In addition to nuchal translucency, other ultrasonographic markers for Down syndrome have proven to be useful adjunctive noninvasive screening tools. Absence of the fetal nasal bone in the first trimester has been observed in fetuses with Down syndrome. The assessment of absence of the fetal nasal bone, increased resistance to flow in the ductus venosus, or the presence of tricuspid regurgitation may be used to further modify first-trimester Down syndrome risk assessment.

The American College of Obstetrics and Gynecology published an updated technical bulletin on Screening for Fetal Chromosomal Abnormalities in January, 2007. The previous bulletin had recommended that screening for aneuploidy should be offered to all women younger than 35 at their estimated date of delivery and that invasive prenatal diagnostic testing should be offered to all women who will be 35 years or older at the estimated date of their delivery and to women with risk factors for having a fetus with aneuploidy including a significant family history, a positive screening test or an abnormality noted on prenatal ultrasound. The updated bulletin recommends that screening and invasive testing should be available to all women who present for prenatal care before 20 weeks of gestation regardless of maternal age.

It is likely that new markers may be implemented in the future to improve the sensitivity and specificity of maternal serum screening. ADAM 12, a metalloprotease that binds insulin growth factor binding protein-3 (IGFBP-3), appears to be an effective early Down syndrome marker. Decreased levels of ADAM 12 may be detected in cases of trisomy 21 as early as 8 to 10 weeks gestation. Maternal serum ADAM 12 and PAPP-A levels at 8 to 9 weeks gestation in combination with maternal age yielded a 91% detection rate for Down syndrome at a 5% false-positive rate. When nuchal translucency data from approximately 12 weeks gestation was added, this increased the detection rate to 97%.

Cellfree fetal DNA was first detected in the maternal circulation over a decade ago. Fetal epigenetic markers such as DNA methylation or a placental epigenetic marker called maspin can be utilized to discriminate fetal DNA from maternal DNA. The detection of fetal DNA in the maternal circulation holds great promise for prenatal diagnosis of fetal disorders and pregnancy complications. To date, cellfree DNA has been used for fetal rhesus D blood typing and fetal gender determination for carriers of X-linked recessive disease and fetuses at risk for congenital adrenal hyperplasia.

Screening for Neural Tube Defects

Neural tube defects are an etiologically heterogeneous group of conditions characterized by failure of closure of the embryonic neural tube. These abnormalities of the brain and vertebral column can occur as an isolated defect or as part of a genetic syndrome. Isolated neural tube defects occur in approximately 1.4 to 2 per 1,000 pregnancies and are the second most common major congenital anomaly. They are thought to result from a combination of genetic predisposition and environmental influences. Approximately 90% to 95% of all infants with neural tube defects are born to women with no history of a child with a neural tube defect. Factors known to be associated with neural tube defects include low folic acid intake, geographic region, ethnicity, maternal valproic acid and carbamazepine exposure, high maternal core temperature, and maternal diabetes.

MSAFP screening for neural tube defects was introduced in the 1980s. MSAFP evaluation is an effective screening test for neural tube defects and should be offered to all pregnant women. This type of screening is most accurate from weeks 16 to 18, as there is the widest margin between abnormal and normal distributions at this period in gestation. Although screening for neural tube defects should optimally be performed between 16 and 18 weeks gestation, it can be done between 15 and 22 weeks.

In the United States, a screen-positive cutoff of 2.5 multiples of the median (MoM) is commonly used, yielding a screen-positive rate of approximately 5%. This results in the detection of more than 95% of anencephalic fetuses and 80% of fetuses with open spina bifida. It is important to adjust MSAFP values for diabetes, race, maternal weight, and multiple gestation. In insulin-dependent diabetics, the MSAFP level is approximately 60% of nondiabetic controls, and it is inversely correlated with the hemoglobin A_1C levels. Blacks have approximately 1.1 times the MSAFP level of whites, and Asians have an intermediate level between blacks and whites. The median twin MSAFP level from 16 to 20 weeks is approximately 2.5 multiples of the median for a singleton pregnancy. The detection rate for twins is approximately 80%.

An inaccurate gestational age determination is the most common reason for an abnormally elevated MSAFP. The false-positive rate can be lowered by performing an ultrasound examination before MSAFP screening to verify the gestational age and diagnose multiple gestations and cases of intrauterine fetal demise, which may also be associated with elevated MSAFP levels. Table 2 lists conditions that may be associated with an elevated MSAFP level.

Women with MSAFP levels higher than the predetermined cutoff (usually 2.0 to 2.5 MOM) and women with risk factors for carrying a fetus with a neural tube defect including a positive family history or previous affected

pregnancy, diabetes, or first-trimester valproic acid or carbamzepine use should be referred for genetic counseling and consideration of a diagnostic test. All women with a positive MSAFP screen should have a specialized ultrasound to further assess the risk of neural tube defects and rule out other fetal anomalies. According to the ACOG bulletin, genetic amniocentesis is the traditional diagnostic test offered to women with an elevated MSAFP. An elevated amniotic fluid AFP in association with the presence of acetylcholinesterase in the amniotic fluid is considered diagnostic for a fetal neural tube defect. If an amniocentesis is performed secondary to an elevated MSAFP, a sample of amniotic fluid should also be sent for cytogenetic analysis, as there are several studies that have reported an association between elevated MSAFP levels and fetal aneuploidy.

TABLE 2 Conditions Associated with an Elevated Maternal Serum Alpha-fetoprotein Level

Neural tube defects Fetal demise Multiple gestation Selective fetal reduction Ventral wall defects: Gastroschisis Omphalocele Esophageal or duodenal atresia Urinary tract disease: Renal agenesis Congenital nephrosis Polycystic kidney disease Obstructive lesion Integumental defects: Congenital ichthyosiform erythroderma Epidemolysis bullosa Hydrops or ascites Cystic hygroma Placental abnormalities: Placental lakes Retroplacental hemorrhage Hemangiomas of placenta and cord Maternal hepatoma

There are many centers in the United States that use ultrasound as a diagnostic tool in women with a high risk for neural tube defects. Studies have shown that in experienced centers, ultrasound can yield a 97% sensitivity and a 100% specificity in the diagnosis of neural tube defects. If the fetal anatomy is well visualized and no abnormalities are detected, the risks and benefits of both amniocentesis and specialized ultrasound examination can be discussed with the patient. Many high-risk patients decline to have amniocentesis performed after a reassuring specialized ultrasound examination. Alternatively, amniocentesis should be offered if ultrasound visualization of the fetus is suboptimal and in patients in whom a fetal defect is identified.

Indications for Prenatal Diagnosis

The most common indications for consideration of invasive prenatal diagnostic testing include an abnormal prenatal screen conferring an increased risk for fetal aneuploidy or spina bifida, a thickened nuchal translucency >3 mm, increased risk for a genetic condition based on parental carrier status, and identification of a fetal anomaly on prenatal ultrasound. Table 3 lists indications for consideration of invasive prenatal diagnostic testing.

TABLE 3 Indications for Invasive Prenatal Diagnostic Testing

Abnormal biochemical screening result Fetal anomaly diagnosed on ultrasound Thickened nuchal translucency (>3 mm) Cystic hygroma Patient request for invasive testing for fetal karyotype Parent or previous fetus/child with a chromosomal abnormality Parents are carriers for a monogenic disorder (i.e., Tay-Sachs disease, Huntington disease, myotonic dystrophy)

Invasive Prenatal Diagnosis Procedures

Amniocentesis

Amniocentesis was introduced in the 1950s for fetal sex determination. The first prenatal diagnosis case of trisomy 21 was reported in 1968. Since that time, the role of amniocentesis has dramatically expanded to include the diagnosis of various cytogenetic and biochemical abnormalities, fetal infections, and a multitude of mendelian disorders. Amniocentesis is the most extensively used fetal sampling technique.

Technique for Amniocentesis

Amniocentesis is typically performed at 15 to 16 weeks gestation. Prior to the amniocentesis, an ultrasound examination is performed to assess fetal viability and number, gestational age, and fetal anatomy. Next, an optimal pocket of amniotic fluid is identified, ideally avoiding the fetus, umbilical cord, and placenta. The abdomen is then prepped with an antiseptic solution. Then, using continuous ultrasound guidance, the physician introduces a 20- to 22-gauge spinal needle into the pocket of fluid (Fig. 1). Approximately 20 mL of fluid is aspirated. The first 1 to 2 mL of amniotic fluid is generally aspirated in a separate syringe and is discarded to minimize the chances of maternal cell contamination.

Multiple Gestation

Prior to the procedure, an ultrasound examination should be performed to determine the position of the fetuses, the placenta(s), and the membrane(s) separating the sacs.
Chorionicity should be determined. It is important to describe the location of the fetuses and their placentas at the time the procedure is performed, as it may be necessary to later identify a fetus with abnormal results. Although the fetuses can change their relative positions, their placentas cannot. Thus, it is valuable to trace the umbilical cords to their placentas and to describe any other ultrasound features, including phenotypic gender, that may help to identify the fetuses at a later date.

Figure 1. Amniocentesis

A number of techniques involving amniocentesis in twin gestations have been described. The most common technique used to ensure that the same sac is not sampled twice involves the injection of indigo carmine into the first sac after the amniotic fluid sample has been withdrawn. When higher-order multiples are sampled, indigo carmine can be injected into each successive sac. Methylene blue should not be used, as it has been reported to cause methemoglobinemia, hemolytic anemia, and intestinal obstruction.

A single needle puncture technique has been reported for twin gestations. After the first sac has been sampled, the same needle is advanced into the sac of the second twin. The first 1 to 2 mL of amniotic fluid is discarded to avoid contamination from the first twin. Another reported technique involves simultaneous visualization of two needles on each side of the separating membrane by using a curvilinear or linear transducer. Both of these techniques eliminate the need to inject dye into the amniotic cavity.

Potential Complications and Risks

Fetal Loss

Several large prospective trials have been undertaken to establish the safety of midtrimester amniocentesis. A multicenter study sponsored by the National Institute of Child Health and Human Development (NICHD) found no significant difference in fetal loss rates between 1,040 patients who underwent midtrimester amniocentesis (3.5%) and matched controls (3.2%). A Canadian trial also reported similar fetal loss rates in the amniocentesis group and a matched control group. A British study published in 1978 was not as reassuring as the two previous collaborative studies. The group of 2,428 women who underwent amniocentesis had a 1.0% to 1.5% increased fetal loss rate compared with the matched control group. This study has been criticized for ascertainment bias, as many of the matched controls were not selected until a later gestation than the subjects at the time of the amniocentesis, so some potentially eligible controls may have aborted before they had the opportunity to be selected. In addition, some of the patients in the amniocentesis group had elevated MSAFP levels, which may be associated with adverse pregnancy outcomes including fetal loss. None of these studies was randomized, and all were performed in the 1970s before ultrasound guidance was routinely used.

The only randomized trial addressing the safety of amniocentesis was performed in Denmark and published in 1986. In this trial, 4,606 low-risk women between 14 and 20 weeks gestation were randomized into two groups: amniocentesis under ultrasound guidance or no procedure. The total fetal loss rate was noted to be significantly higher in the amniocentesis group (1.7%) compared with the control group (0.7%) (P <0.01).>

A review of “contemporary” second-trimester amniocentesis publications that each included over 1,000 amniocentesis cases with sufficient detail and follow-up data up to 28 weeks gestation was published in 2004. Twenty-nine reports involving 68,119 amniocentesis procedures were examined. There was a procedure-related rate of excess pregnancy loss of 0.33% (95% confidence interval [CI], 0.09, 0.56) in a comparison of all studies to available control subjects. There were five controlled studies that used concurrent ultrasound needle guidance, including the randomized Danish trial. The procedure-related loss rate was 0.6% (95% CI, 0.31, 0.90) in the subset of the five controlled trials. The background loss rate among the control patients who did not undergo amniocentesis was 1.08%.

A procedure-related loss rate of 0.06% was reported in a study of the 3,096 women who were enrolled in the FASTER (First and Second Trimester Evaluation of Risk for Aneuploidy) trial who chose to have amniocentesis performed. This subgroup of women was compared with the 31,907 women in the FASTER trial who did not undergo amniocentesis. The spontaneous fetal loss rate at less than 24 weeks gestation in the study group was 1.0% and was not statistically different from the background 0.94% rate seen in the control group (P = 0.74; 95% CI, 0.026, 0.49). There have been a number of criticisms regarding the study design and statistical methodology of this trial. The authors acknowledged potential limitations of the study, including its nonrandomized study design and the insufficient sample size. More than 400,000 women would be needed in each arm of the study to have 80% power to detect a difference of 0.05% in spontaneous loss rates between the two groups.

Factors that have been reported to be associated with increased rates of fetal loss include a large number of needle insertions, using a needle greater than 18 gauge, perforation of the placenta, and discolored amniotic fluid. The NICHD study reported a 2.9% incidence of fetal loss with one needle insertion, 4.3% for two insertions, and an 8.1% loss rate with three or more insertions. Although brown fluid was related to adverse pregnancy outcome, the NICHD study did not find an association between bloody amniotic fluid and increased fetal loss rates. The Canadian study also reported a correlation between fetal losses and more than two needle insertions per procedure and the use of needles of 19 gauge or greater. The Danish study reported that withdrawal of discolored amniotic fluid, increased levels of MSAFP, and perforation of the placenta were associated with increased fetal loss rates.

The impact of placental penetration on the risk of pregnancy loss has been addressed in a review of nine reports in the literature involving a total of 5,203 transplacental amniocentesis procedures. In these cases, the loss rate of 1.4% was identical to the overall loss rate noted in the total group of 34,144 women who did not have placental perforation at the time of amniocentesis. Although there is no demonstrable increased risk, it still seems prudent to avoid the placenta where possible. In cases where this is not possible, the cord insertion should be identified and avoided and the thinnest portion of the placenta should be punctured.

Although a contemporary randomized trial would be the optimal approach to assess the fetal loss rate attributable to amniocentesis, it is unlikely that another randomized trial will be performed due to practical and ethical reasons. Although the exact risk associated with amniocentesis is controversial, it is not a completely innocuous procedure. Based on a review of all of the available literature, the risk associated with genetic amniocentesis is likely close to 1 in 300 in experienced hands. Factors associated with increased fetal loss should be avoided when possible.

Other Risks of Midtrimester Amniocentesis

Other potential procedure-related complications include leakage of amniotic fluid, amnionitis, vaginal bleeding, and needle puncture of the fetus. Leakage of amniotic fluid occurs in approximately 1% of women after undergoing amniocentesis. Fortunately, in most cases, this is minimal and resolves within several days. Pregnancies with ruptured membranes after amniocentesis result in better outcomes compared with pregnancies complicated by spontaneous rupture at a similar gestational age. Intra-amniotic infection after amniocentesis occurs in approximately 0.1% of cases. Vaginal bleeding may occur in 2% to 3% of cases and is self-limiting in most cases. With the use of continuous ultrasound guidance, needle puncture of the fetus should be avoidable in the great majority of cases.

Local Anesthesia

Two randomized trials compared local anesthesia using 1% lidocaine with no anesthesia in over 400 women undergoing midtrimester amniocentesis. There were no significant differences in pain perception between the two groups. Based on these data, local anesthesia should not be used for amniocentesis except in select cases.

Early Amniocentesis

Early amniocentesis, performed prior to 14 to 15 weeks gestation, potentially provides fetal karyotype results earlier than midtrimester amniocentesis. Studies have shown that the cytogenetic results from early amniocentesis cell cultures are as accurate as those obtained in the second trimester. In contrast to CVS, potential advantages associated with early amniocentesis include the use of a familiar technique that is widely available, reduction of maternal cell contamination and placental mosaicism, and the ability to assess amniotic fluid AFP.

The technique for early amniocentesis is similar to the technique used for midtrimester amniocentesis. Early amniocentesis may be more difficult to perform because the amnion and chorion are often still separated by the extraembryonic coelom until 14 weeks gestation. This may result in tenting and stretching of the amniotic membranes and prevent access to the amniotic cavity. This potential problem may be avoided by advancing the needle tip vigorously into a fetusfree pocket of fluid so that the tip “pops through” the membranes. In early gestation, access to an optimal pocket of fluid may require a transplacental approach. There does not appear to be any increased rate of complications specifically associated with transplacental early amniocentesis.

Complications and Risks

Based on the results of several studies, it appears that early amniocentesis at 11/0 and 12/6 weeks gestation is associated with an increased risk of fetal loss and talipes equinovarus. Currently, there are not enough data available to make conclusions regarding the safety of early amniocentesis between 13/0 and 14/6 weeks gestation. The Canadian Early and Mid-trimester Amniocentesis Trial (CEMAT) randomized 4,374 women to early amniocentesis (11/0to 12/6 weeks gestation) or midtrimester amniocentesis (15/0to 16/6 weeks gestation). All procedures were performed with a 22-gauge needle. There was a statistically significant increase in total fetal losses (7.7% vs. 5.9%) and talipes equinovarus (1.3% vs. 0.1%) in the early amniocentesis group. There was also a significantly increased rate of postprocedural amniotic fluid leakage in the early amniocentesis group (3.5% vs. 1.7% in the midtrimester amniocentesis group). Amniotic fluid leakage after early amniocentesis was associated with a 15% incidence of talipes equinovarus compared with a 1.1% incidence after procedures without leakage of fluid. There were no other significant congenital anomalies in either group, and there was no difference in the incidence of neonatal ventilation or prolonged oxygen use between the two study groups; however, a comparison with controls not exposed to amniocentesis was not performed.

Two studies randomized patients to early amniocentesis versus transabdominal CVS. Both studies reported a significantly increased incidence of talipes equinovarus in the early amniocentesis group. A study by Sundberg and colleagues was stopped prematurely secondary to the significant association between early amniocentesis and talipes equinovarus. There were nine cases of talipes equinovarus out of 548 cases in the early amniocentesis group and no affected cases out of 555 CVS procedures. The authors were unable to make a definitive conclusion regarding fetal loss, as although the total fetal losses were similar in both groups, the power of the study was limited since the study was terminated early. A study by Nicolaides and associates found significantly higher rates of talipes equinovarus and spontaneous loss in the early amniocentesis group (1.6% and 5.9%, respectively) compared with the CVS group (0.5% and 1.2%, respectively).

Summary

Amniocentesis performed at 15 weeks or later is safe and effective. The fetal loss rate associated with amniocentesis at this gestational age is likely approximately 1/300 in experienced hands. Amniocentesis should not be performed earlier than 12 weeks and 6 days due to the increased risk of fetal loss and talipes equinovarus. The safety of performing amniocentesis between 13 weeks, 0 days and 14 weeks, 6 days is unknown at this time. Amniocentesis procedures should generally not be performed at this gestational age except for in extenuating circumstances.

Chorionic Villus Sampling

CVS involves sampling of the developing trophoblast in the first trimester. The indications for CVS are similar to amniocentesis. CVS is usually performed between 10 and 12 weeks gestation. The chorion frondosum, which contains the most mitotically active cells in the developing placenta, is sampled by using either a transabdominal or transcervical approach, depending on the placental location and the preference of the patient and/or the physician performing the procedure (Figs. 2, 3). An ultrasound should be performed prior to the procedure to assess fetal viability, gestational age, and placental position.

Figure 2. Transcervical CVS

Figure 3. Transabdominal CVS

Transcervical Chorionic Villus Sampling

Fetal tissue sampling was first performed using a transcervical route in the late 1960s. Ultrasound guidance for CVS was first reported in 1979. Brambati and colleagues in Italy were the first to describe the technique that is used for transcervical CVS in most centers today. A polyethelene catheter with a malleable obturator is passed through the cervix into the thickest part of the placenta under ultrasound guidance. Placental trophoblast is then aspirated through the catheter into a 20-mL syringe containing tissue culture medium.

Transabdominal Chorionic Villus Sampling

This technique, which was first described in the medical literature in 1984, can be performed by using a freehand technique or a needle-aspirator transducer. Some operators prefer to use a double-needle technique, in which a smaller needle may be introduced into a larger needle. A 19- or 20-gauge needle is directed into the thickest part of the placenta that is readily accessible under ultrasound guidance. After the stylet is withdrawn from the needle, a syringe containing tissue culture medium is attached to the hub of the needle, and suction is applied as the needle is moved up and down through the placenta until an adequate amount of tissue is obtained.

The sample should be inspected to ensure that an adequate amount of chorionic villi have been obtained. The average sample from a CVS aspiration contains from 15- to 30-mg wet weight of villus material. Transabdominal samples tend to be slightly smaller. A 20-mg sample is ideal for cytogenetic analysis; however, results can be obtained after cell culture with significantly smaller samples. Larger samples of 20 to 40 mg of tissue may be required if direct molecular and biochemical studies are to be performed in addition to cytogenetic analysis.

Transcervical versus Transabdominal Approach?

In most cases, physician or patient preference will dictate which method is used, as in most cases, CVS can be performed by using either a transcervical or a transabdominal approach. However, in approximately 3% to 5% of cases, clinical circumstances will support one approach over the other. Posterior placentas in a retroverted uterus are especially amenable to a transcervical approach, whereas a transabdominal approach would be preferable in cases involving anterior placentas in an anteflexed uterus and fundal placentas. The transcervical and transabdominal approaches have been shown to be equally safe and effective provided that centers have expertise with both approaches.

Absolute contraindications to the transcervical approach include maternal blood group sensitization and active cervical or vaginal infections such as herpes, chlamydia, or gonorrhea. Relative contraindications include vaginal bleeding within 2 weeks prior to the procedure; uterine fibroids that prevent passage of the catheter; cervical polyps; and a markedly retroverted, retroflexed uterus.

Confined Placental Mosaicism

Chromosome mosaicism is defined as the presence of two or more karyotypically different cell lines in the fetoplacental unit arising from a single zygote. Frequently, in cases where a CVS result shows chromosome mosaicism, an amniocentesis will be performed to determine whether the fetus is affected. Confined placental mosaicism refers to cases in which the fetus does not carry the mosaic cell line that was identified in the placenta. It is detected in approximately 1% to 2% of CVS samples. Pregnancies with confined placental mosaicism may be at risk for spontaneous abortion, perinatal loss, or intrauterine growth restriction. A mosaic normal cell line in the placenta of fetuses with chromosomal abnormalities such as trisomy 13 or 18 may allow for prolonged survival of the fetus.

Pregnancies with confined placental mosaicism involving a trisomic cell line may be at risk for uniparental disomy. Uniparental disomy refers to the inheritance of two copies of a chromosome, or part of a chromosome, from the same parent. Isodisomy results when a chromosome or gene is present in two identical copies from the same parent, and heterodisomy results when nonidentical copies from the same parent are present. Trisomic rescue is thought to be the most common cause of uniparental disomy. Trisomic rescue occurs when a zygote is initially trisomic for a particular chromosome but subsequently loses the extra chromosome. In two thirds of the cases, the lost chromosome is of the parental origin that contributed the extra chromosome. In one third of the cases, the lost chromosome will result in uniparental disomy. Isodisomy results when nondisjunction occurs during meiosis I, and heterodisomy results from nondisjunction in meiosis II. Uniparental disomy has been described for chromosomes 6, 7, 11, 14, 15, and 16. Paternal isodisomy for chromosome 15 is associated with Angelman syndrome, and maternal isodisomy for chromosome 15 is associated with Prader Willi syndrome. Maternal isodisomy for chromosome 16 is associated with pregnancy loss and severe intrauterine growth restriction.

Karyotype analysis is usually insufficient to detect uniparental disomy, as specific parentally derived homologues cannot be distinguished at the microscopic level of resolution. Uniparental disomy can be detected with molecular genetic techniques. Thus, in cases of confined placental mosaicism involving trisomy, consideration must be given regarding uniparental disomy. Ideally, the patient should be referred for a prenatal genetics consultation so that she can receive appropriate counseling regarding the potential outcomes and be offered the opportunity to have molecular genetic studies to rule out uniparental disomy.

Safety of Chorionic Villus Sampling

Pregnancy Loss

Four prospective randomized controlled trials that provide data on the safety of CVS have been published to date. The three largest of these studies were performed in the 1980s, soon after the introduction of this procedure, when many providers were still gaining experience with the technique. Patients in these trials were randomized to CVS and midtrimester amniocentesis. An early prospective trial found CVS to have a 0.4% greater rate of loss before 20 weeks gestation and a 0.7% greater rate of loss up to 28 weeks gestation compared with midtrimester amniocentesis. Another early nonrandomized study found a 0.8% increase in fetal loss after CVS as compared with midtrimester amniocentesis. It has been shown that loss rates from CVS are inversely correlated with provider experience. It is unlikely that additional randomized trials will be performed secondary to problems with patient recruitment. A recent retrospective cohort study compared the loss rates before 24 weeks gestation in women who had amniocentesis and CVS with women who did not undergo an invasive procedure between 1983 and 2003. All providers performed a minimum of 50 amniocentesis or 100 CVS procedures under supervision. After adjusting for the background loss rate, the overall amniocentesis loss rate was 0.46%, and the rate from 1998 to 2003 decreased to 0.27%, or 1 in 370. The CVS loss rate was 2.35% overall and decreased to 1.16% in the interval between 1998 and 2003. The difference in pregnancy loss rates between CVS and amniocentesis procedures decreased over the 20-year interval. There was no clinically or statistically significant difference between CVS and amniocentesis loss rates in the most recent interval from 1998 to 2003.

Limb-Reduction Defects

In 1991, Firth and associates reported that 5 of 289 cases of CVS performed at 56 through 66 days gestation had severe limb abnormalities. Four of the infants had oromandibular-limb hypogenesis, and the fifth had an isolated transverse limb-reduction defect. Since this initial report, there have been many other publications both supporting and refuting this association. It appears as though there may be an association of transverse limb defects with CVS procedures performed very early in gestation. Brambati and colleagues observed a 1.6% incidence of severe limb-reduction defects in cases performed at 6 and 7 weeks, a 0.1% incidence at 8 to 9 weeks, and no increased incidence of limb defects in cases performed after 9 weeks. Eighteen of 19 cases of transverse limb reduction reported by Hsieh and associates, including four cases of oromandibular-limb hypogenesis, had CVS performed before 9 weeks. It also appears that limb-reduction defects may be associated with less-experienced operators. The CVS procedures in the Hsieh report were performed by relatively inexperienced operators in Taiwan.

The World Health Organization (WHO)-sponsored CVS registry in Philadelphia, which was started in 1983, reviewed outcome results from a total of 138,996 cases performed at 63 programs between 9 and 12 weeks gestation. There was no increased incidence of limb-reduction defects in the CVS population (5.2 to 5.7 per 10,000 cases) compared with the general population (4.80 to 5.97 per 10,000). The pattern distribution of limb defects, including transverse limb defects, in the CVS group was similar to the pattern of limb defects in the general population.

Summary

It is important that women are educated regarding the potential benefits and risks associated with CVS, including a potentially slightly higher loss rate compared with second-trimester amniocentesis. This increased risk may become negligible as operator experience increases. It appears as though CVS performed after 10 weeks by an experienced operator is not associated with an increased incidence of limb defects compared with the general population. CVS procedures should generally not be performed prior to 10 weeks gestation.

Percutaneous Umbilical Blood Sampling

First attempts to enter the fetal circulation were based on endoscopic techniques. In 1973, Hobbins and Mahoney published their ability to obtain fetal blood with an endoscopic technique by utilizing a 1.7-mm diameter endoscope combined with a 25-gauge sampling needle. Later, this team reported on the successful diagnosis of hemoglobinopathies by utilizing this technique.

Daffos was the first to utilize an ultrasound guided percutaneous technique to enter the umbilical cord. He initially used this technique to diagnose fetal toxoplasmosis in patients at high risk for this condition. Since that time, percutaneous umbilical blood sampling (PUBS) has been used in the diagnosis of fetal infection, hemoglobinopathies, and chromosomal abnormalities; to explore the etiology of fetal hydrops; to assess fetal acid- base status in growth restriction; and to diagnose and treat erythroblastosis fetalis. Presently, there are noninvasive approaches available that obviate the need to access the fetal circulation for the majority of the initial uses of percutaneous blood sampling. Percutaneous access to the fetal circulation is still used to perform intravascular transfusion to treat fetal anemia. This is most commonly performed in cases of Rh-sensitization, in which there is evidence of underlying significant fetal anemia based on noninvasive studies using ultrasound.

Technique

PUBS involves obtaining a sample of fetal blood by placing a needle into the umbilical vein, generally where it inserts into the placenta, as this is where the cord is least mobile. Fetal blood sampling has been obtained from the umbilical artery; however, the umbilical vein is preferred because it is larger and less likely to be associated with fetal bradycardia when punctured. In cases when it is not possible to access the umbilical cord vessels, alternative techniques including sampling from the intrahepatic vein and cardocentesis have been reported, but these techniques are associated with increased risks compared with umbilical cord blood sampling.

A variety of methods have been used for PUBS. The needle may be inserted by using a freehand technique or by using a needle-guiding device that is fixed to the transducer. A 20- or 22-gauge needle may be used. Once the needle has been guided into the umbilical vein, fetal blood is aspirated into a syringe (Fig. 4). In order to confirm that the sample is fetal in origin, the mean corpuscular volume (MCV) of the sample should be assessed. Fetal blood cells (140 fl) are larger than maternal cells (80 fl). The MCV of a sample of fetal blood should be above 100.

Risks

The most critical factor related to the safety of the procedure is operator experience. Ghidini and colleagues published a meta-analysis that was designed to assess the true risk of PUBS. Reports from centers that had performed more than 100 procedures were included. In an attempt to identify procedure-related loss rates, a subset of low-risk cases was analyzed that excluded patients with fetal pathologic conditions including chromosomal abnormalities, structural defects, intrauterine growth restriction, nonimmune hydrops, and fetal infection. The meta-analysis in this low-risk group of patients yielded a loss rate of 1.4% in patients less than 28 weeks gestation and 1.4% in patients greater than 28 weeks. The total fetal loss rates in the low-risk groups ranged from 1.2% to 4.9%. Almost half of the cases analyzed by Ghidini and his group were contributed by a single operator. His loss rate of 2 per 1,021 (0.2%) represented an extremely low loss rate. If one were to exclude these data from the meta-analysis, the resulting loss rates for PUBS in all cases was 7.2% (96/1328) and 3% in the low-risk group (20/660).

Figure 4. Percutaneous umbilical blood sampling

In addition to fetal loss, other complications associated with PUBS include bleeding from the puncture site in the umbilical cord, cord hematomas, transient fetal bradycardia, infection, and fetomaternal hemorrhage. Bleeding from the cord puncture site is the most common complication and may occur up to 40% of cases. In most cases, it is self-limited. Fetal bradycardia occurs in approximately 9% of procedures and lasts for a short time in most cases. The incidence of a clinically significant fetal bradycardia was noted to be 6.6% in one report. Both umbilical artery puncture and severe, early onset growth restriction were associated with increased rates of bradycardia.

Rh Prophylaxis

It is critical to obtain the blood type on all patients who undergo amniocentesis, CVS, or PUBS, as patients who are Rh-negative and unimmunized should be given a 300 mg intramuscular dose of Rh immune globulin to avoid Rh sensitization.

Future Trends in Prenatal Diagnosis

It is likely that future advances in ultrasound, molecular genetics, cytogenetics, and other technology will continue to increase the alternatives for less invasive and thus less risky prenatal diagnosis options. A trend in this direction has already been observed. Conditions that were first diagnosed using fetoscopy may now be diagnosed using ultrasound, in some cases as early as the first trimester. Identification of additional maternal serum and ultrasound markers may allow for earlier and improved identification of patients at risk for fetal aneuploidy as well as early identification of women at risk for obstetric complications. The discovery of fetal DNA in the maternal circulation holds great promise for noninvasive prenatal diagnosis of fetal disorders and pregnancy complications.

SUMMARY POINTS

• It is possible to detect approximately 85% of cases of Down syndrome in the first trimester by using a combination of maternal age, nuchal translucency measurement, and maternal serum PAPP-A and free or total β-hCG.
• Down syndrome screening strategies that involve a combination of first- and second-trimester markers yield the highest detection rates and may still provide the option for first-trimester invasive prenatal diagnosis.
• Accurate prenatal diagnosis is now possible for hundreds of genetic conditions through genetic testing.
• Amniocentesis performed at approximately 15 weeks gestation is safe and effective. The fetal loss rate is approximately 1/300 in experienced hands.
• As first-trimester screening continues to increase in popularity, the demand for invasive first-trimester prenatal diagnosis will likely increase. When performed after 10 weeks gestation by an experienced operator, the risk associated with CVS may be comparable to the risk associated with amniocentesis.
• All unimmunized Rh-negative women who undergo invasive prenatal diagnostic testing should be given a 300 mg intramuscular dose of Rh immune globulin to avoid Rh sensitization.