Chromosomal Abnormalities

Nuchal Translucency and Trisomy 21

In 1866, Dr. Langdon Down reported a classification of individuals with developmental delay who had similar physical characteristics. He described the skin as “deficient in elasticity, giving the appearance of being too large of the body.” Fetuses with trisomy 21 as well as other aneuploidies often have excess fluid in the subcutaneous tissue behind the fetal neck ( Fig. 31.2A , Video 31.1 ). Sonographically, this appears as an echolucent fluid collection between the soft tissue over the cervical spine and an echogenic line representing the skin edge. This fluid space is called the nuchal translucency (NT) ( Fig. 31.2B ). The lucency is thought to represent mesenchymal edema and is often associated with distended jugular lymphatics ( Fig. 31. 2C ). The prevailing theory suggests an alteration in lymphangiogenesis and delayed lymphatic development. Other possible causes include cardiac failure and abnormal extracellular matrix, but these do not explain the localized and transient nature of the NT. Most likely, the cause is a complex interaction of factors. NT normally increases with advancing gestational age, and therefore the measurement is compared to crown-rump length (CRL, Fig. 31.2D ). An NT greater than 95% for CRL is considered thickened. An NT greater than 99% does not change significantly with CRL and is approximately 3.5 mm ( Fig. 31.2E ).

In 1992, Nicolaides et al. reported that an NT of 3 mm or greater in the first trimester was associated with a 35% risk of chromosomal abnormality. The association between chromosome anomalies and a thickened NT was subsequently confirmed in a large, prospective multicenter trial of 20,804 pregnancies. The risk of trisomy 21 can be calculated by multiplying the a priori (presumptive) risk by a likelihood ratio derived from the degree of deviation in NT from the expected NT. This methodology, in conjunction with a risk cutoff of 1 in 300, resulted in the identification of 80% of fetuses with trisomy 21, with a false-positive rate (FPR) of 5%. Snijders et al. evaluated the use of NT and maternal age to detect trisomy 21 in a multicenter trial that included 22 different sites and 306 trained sonographers. Using a threshold of 1 in 300, the sensitivity for the detection of trisomy 21 was 82% for an FPR of 8%. If the FPR was set at 5%, the sensitivity for the detection of trisomy 21 was 77%.

Serum Biochemical Markers

A variety of serum biochemical markers have different concentrations in pregnancies with trisomy 21 compared with euploid pregnancies. The maternal serum of women carrying fetuses with trisomy 21 has a higher concentration of free beta human chorionic gonadotropin (free β-hCG) and lower concentration of pregnancy-associated plasma protein A (PAPP-A) compared with the serum of women carrying euploid fetuses. The use of these serum markers, in conjunction with maternal age, results in identification of 62% of fetuses with trisomy 21 at an FPR of 5%.

Combined First-Trimester Screening

There is no association between NT measurement and serum levels of free β-hCG or PAPP-A in euploid fetuses or in those with trisomy 21. This independence allows the combination of NT screening and biochemical screening, resulting in a more effective method of risk assessment than either method individually. Wald et al. demonstrated that the combination of NT measurement with maternal serum PAPP-A and free β-hCG, known as the combined first-trimester screening test, results in the detection of 85% of trisomy 21 fetuses at an FPR of 5%.

First-trimester screening using NT measurement and serum biochemical measurement has been further validated in four large studies. The One-Stop Clinic to Assess Risk (OSCAR) screening trial in the United Kingdom studied 12,339 women with singleton pregnancies between 10 and 14 weeks' gestation. First-trimester screening was accepted by 97.5% of the women, and if they screened positive (risk ≥ 1 : 300), 77% underwent invasive diagnostic testing. There were 25 cases of trisomy 21, of which 23 (92%) were detected, with an FPR of 5%. The detection rates of both trisomy 13 and 18 were 100%.

A multicenter trial in North America (BUN study) similarly evaluated 8514 patients with singleton pregnancies between 74 and 97 days' gestation. A screening result was considered positive if the risk of trisomy 21 was 1 in 270 or higher or the risk of trisomy 18 was 1 in 150 or higher. There were 61 cases of trisomy 21; detection rate was 79% with a 5% FPR. The detection rate of trisomy 18 was 90% with a 2% FPR.

The Serum, Urine and Ultrasound Screening Study (SURUSS) evaluated the efficacy, safety, and cost-effectiveness of first- and second-trimester screening for trisomy 21. This prospective study was conducted primarily in the United Kingdom on 47,053 pregnancies between 9 and 13 weeks' gestation. In the first trimester the combined test had a sensitivity of 85% for the detection of trisomy 21 with an FPR of 6%.

The First and Second Trimester Evaluation of Risk (FASTER) was the largest trial based in the United States and was designed to determine how best to screen pregnant women for trisomy 21. This multicenter trial included 36,120 patients with complete first-trimester data, of whom 92 fetuses had trisomy 21. The trial included NT measurements as well as serum biochemistry in both the first and the second trimester, revealing the results to patients only in the second trimester, after both serum screens. The detection rate of trisomy 21 was 87%, 85%, and 82% at 11, 12, and 13 weeks' gestation, respectively, at a 5% FPR. For a detection rate of 85%, NT alone (without biochemical data) had an unacceptably high FPR of 20%. If the FPR was set at the more acceptable level of 5%, the sensitivity for detection of trisomy 21 fell to 68%.

Screening for aneuploidy in twin gestations is less robust than in singleton gestations and is hampered by nuances of placentation. Determination of amnionicity and chorionicity is crucial. Sebire et al. studied a series of 448 twin pregnancies and identified 88% of trisomy 21 fetuses (FPR 7%) using an NT less than 95%. The prevalence of a thickened NT is higher in euploid monochorionic twins than in euploid dichorionic twins, which may relate to placental structure or fetal anomalies. Discordance of NT measurements in monochorionic twins may predict twin-to-twin transfusion syndrome. Additionally, the contribution of biochemical markers increases the complexity of risk assessment in twin gestations because the concentration of markers relates to the pregnancy (as opposed to the individual fetus). Spencer and Nicolaides identified 75% of trisomy 21 fetuses (FPR of 7%) using biochemical markers and NT measurement. Wald and Rish reported that at a set FPR of 5%, the estimated detection rate of trisomy 21 was 84% in monochorionic twins and 70% in dichorionic twins. This compares to an 85% estimated detection rate in singletons. Recently, it has been reported that the risk of trisomy 21 in twins may not be as high as previously thought, especially in monozygotic twins; these data add to the complexity of risk assessment. In a large population study based in Europe examining the prevalence of Down syndrome among monozygotic and dizygotic twins, the adjusted risk ratio (RR) of Down syndrome was 0.58 (95% CI, 0.53-0.62) for twins compared with singletons. This was most pronounced in monozygotic twins where the adjusted RR was 0.34 (95% CI, 0.25-0.44). In dizygotic twins the adjusted RR was 1.34 (95% CI, 1.23-1.46).

A major question is whether an NT measurement exists above which there is no benefit to additional biochemical screening. Results of the FASTER trial were evaluated with specific attention to this question. An NT of 4 mm or greater was identified in 32 patients (0.09 %). In this group the lowest combined risk assessment for trisomy 21 in euploid fetuses was 1 in 8 and for aneuploid fetuses was 7 in 8. There were 128 patients with NT of 3 mm or greater. The lowest risk of trisomy 21 among euploid fetuses using combined screening was 1 in 1479, and the lowest risk among those with aneuploidy was 1 in 2. Only 10 patients (8%) had the risk lowered below 1 in 200, all of whom had normal outcome. These authors concluded that there is minimal benefit in waiting for combined screening results in fetuses with NT of 3 mm or larger and no benefit for those with NT of 4 mm or larger.

Integrated and Sequential Screening

Wald et al. introduced the concept of integrated screening, in which first- and second-trimester evaluation is used to provide a single risk estimate for trisomy 21.

The FASTER trial compared screening strategies across the first and second trimesters. There were 33,546 patients with complete first- and second-trimester data available, including 87 fetuses with trisomy 21. The fully integrated screen that included first-trimester NT measurement along with PAPP-A and a second-trimester quad screen resulted in the detection of 95% of fetuses with trisomy 21 at a 5% FPR or 87% if the FPR was set at 1%. A disadvantage of integrated screening is that the patient does not have any screening results in the first trimester, and fetuses at high risk of trisomy 21 are not identified until the midtrimester. Compounding this, as many as 20% of patients may not comply with the second-trimester screen.

Two alternative approaches have been proposed in response to the criticism of late notification of patients at “high risk” of aneuploidy. Each of these sequential protocols discloses first-trimester results to patients above a specific “high-risk” threshold. Women above the threshold and therefore considered at highest risk can be referred for genetic counseling and diagnostic testing early in gestation. In the step-wise protocol, those not identified in the high-risk group undergo second-trimester biochemical screen. In the contingent protocol, only those women at intermediate risk (≥1 : 50 and <1 : 1500) go on to the second-trimester biochemical screening while those at low risk (<1 : 1500) are not offered additional screening. In each protocol the results of first- and second-trimester screening are integrated and reported as a single number because if they are independently interpreted, the FPR is unacceptably high. Those with a risk of 1 in 270 or higher were offered genetic counseling and diagnostic testing.

Cuckle et al. retrospectively calculated the midtrimester risks of trisomy 21 from the FASTER trial data. In this study, patients were categorized as “high risk” (>1 : 30), “borderline risk” (1 : 30-1 : 1500), and “low risk” (<1 : 1500) based on the results of the first-trimester evaluation. Only patients in the borderline category underwent recalculation of risk based on second-trimester biochemical screening. The initial detection of trisomy 21 (>1 : 30) after first-trimester screening was 60% with an FPR of 1.2%. Of the remaining population, 23% were at borderline risk and calculated as having additional screening. Contingent screening identified 91% of fetuses with trisomy 21, with a 4.5% FPR. Stepwise screening, in which all women other than those at highest risk had calculated first- and second-trimester testing, had a detection rate of 92% with an FPR of 5.1%. Integrated screening, in which all women had both first-semester and second-trimester testing, had a detection rate of 88% with an FPR of 4.9%. This study showed that contingent screening, in which only 23% of the population went on to the second-trimester biochemical serum screen, had a similar detection rate of trisomy 21 as the protocols in which most if not all patients had recalculation of risk with second-trimester biochemistry.

Standardization of Nuchal Translucency Measurement Technique

For NT screening to be accurate, standardization of technique, training, and ongoing monitoring are crucial. This education, validation, and ongoing quality assurance has been supported by two organizations: the Nuchal Translucency Quality Review program based in the United States and the Fetal Medicine Foundation based in the United Kingdom. The criteria and caliper placement for measuring the NT are illustrated in Fig. 31.2 (see also Video 31.1 for real time scanning without measurement). The specific details for credentialing can be found at www.ntqr.org or www.fetalmedicine.org .

The fetal CRL must be between 45 and 84 mm although variations exist among laboratories that perform the biochemical component. The accuracy of the NT measurement and the CRL is critical because the NT measurement is converted into multiples of the median (MoM) based on the CRL. An adequate sonographic image to measure the CRL requires the fetus to occupy a majority of the image space and be in a neutral position. The longest straight line between the fetal crown and rump is measured at least three times and the average of three good measurements is used.

Haddow et al. showed that accuracy is critical when obtaining an NT measurement. They studied 4412 women who underwent first-trimester screening with biochemistry and NT in which no specific training in NT measurement was required although the method for measuring NT was a standard protocol. Measurement of the NT varied considerably between centers and could not be reliably incorporated into risk calculations. Furthermore, the center with the highest success rate in obtaining an NT measurement (100%) had the lowest sensitivity (0%) for identifying trisomy 21. Results from the BUN trial revealed that after training, measurements were initially smaller than expected compared with normative values developed by Fetal Medicine Foundation. With increasing experience, the measurements of the BUN trial were in concordance with published norms.