Prenatal Screening and Diagnosis for Down Syndrome Prevention

Overview
In 1979 the Canadian Task Force on the Periodic Health Examination reviewed the evidence for prenatal diagnosis of Down Syndrome (DS) and concluded that there was fair evidence to offer amniocentesis to pregnant women from high-risk groups, including parents with translocation of chromosome 2 1, a family history of DS, or maternal age over 35 years.< 1> This review will evaluate evidence which has been published recently on the use of the triple marker screen (multiple maternal serum markers), and chorionic villus sampling (CVS) for screening and diagnosis of DS pregnancies.

The logic underlying prenatal screening and diagnosis for DS involves the following assumptions: 1) Screening tests must be reasonably accurate in identifying DS fetuses. Confirmatory diagnostic tests must be highly accurate, with a good safety profile, and available during a period of gestation when pregnancy termination is safe and acceptable and 2) Early detection of DS pregnancies with provision for pregnancy termination provides informed reproductive choice for those who wish to use it.

Burden of Suffering
Down Syndrome, a congenital syndrome associated with chromosomal aneuploidy of all or part of chromosome 21, is the most common pattern of malformation seen in humans, with a median birth incidence rate of 1 per 1,000 births.<2> Clinically important problems include general hypotonia, mental retardation, growth retardation, and a significant risk for congenital malformations, of which a cardiac defect is the most common.<3> Though effective therapies are available for some of the specific malformations and problems associated with DS, there are no proven therapies available for the cognitive deficits. Literature on DS persons and their families has focused on dysfunctional outcomes. Up to 10% of families are unable or unprepared to cope with a DS child, and many others are affected by maternal depression, and difficulties with marital and sibling relationships.<4-6> Though family functioning may be adversely affected by a DS child, this is not a necessary outcome.<5>

Maneuver
Two approaches are available for prenatal diagnosis of DS: 1) Prenatal diagnosis with amniocentesis or chorionic villus sampling (CVS) offered to women at risk by virtue of their history (previous DS birth, advance maternal age, certain chromosomal rearrangements), and 2) prenatal diagnosis with amniocentesis offered to women identified as at increased risk through the use of screening maneuvers (maternal serum markers, ultrasonography).

Historical Risk Factors
Studies using a variety of methods have consistently demonstrated an increasing risk of DS birth with advancing maternal age (the risk of a DS birth at age 20 being approximately 1 in 1,500, rising to 1 in 30 at 45 years of age).<7-14> Parents carrying chromosome 21 rearrangements are also at an increased risk for DS pregnancies.<15-17> An increased rate of subsequent DS pregnancy has also been reported with parents of previous DS births in the absence of detectable chromosome rearrangements, independent of advancing age.<18-20> Other factors such as paternal age, parental consanguinity, second or third degree DS relationship, environmental radiation, etc. have not been consistently shown to significantly affect the incidence of the DS birth.

Maternal Serum Marker Screening
Maternal serum alpha fetoprotein (AFP), human chorionic gonadotropin (HCG), and unconjugated estriol (uE3) levels have all been shown to be associated with DS gestations. Individually these markers have relatively poor discriminatory power, but the simultaneous use of all three serum markers (Triple Screening) in the second trimester has received much attention recently. Probabilities derived from the individual markers and the maternal age-related risk are used to generate a post-screen probability of a DS fetus for each gestation.<21,22> Women with a post-screen probability that exceeds a certain cut off level (e.g. a 1 in 250 risk, similar to that of risk women age 35 years and over) can be offered amniocentesis.

Four cohort studies have compared the proportion of DS pregnancies identified through Triple Screening with the total number of DS pregnancies (the sum of DS fetuses detected prenatally and those DS pregnancies not detected, but observed as live births through surveillance of regional cytogenetics laboratory results).<23-26> Detection rates ranged from 48-91%, with false positive rates of 3.2-6% respectively. The different risk cutoffs (1 in 190 to 1 in 274) used in the studies account for some of this variation. A positive predictive value of 1.2-1.8% was achieved in the trials among low-risk women without advanced maternal age. Therefore, a woman with a pre-screen risk of 1 in 1,000 who tested positive upon screening would have a post-screen risk of 1 in 56 – 1 in 85 for a DS gestation. This post-screen risk is similar to that of women with advanced age who are currently offered prenatal diagnosis.

A recent descriptive study in women over 35 years of age, reported that using triple marker screening, 90% of DS pregnancies would have been detected with an amniocentesis rate of 25%.<27> The high negative predictive value of the screen in this age group may be helpful to women who wish to exclude the possibilities of a DS fetus but wish to avoid amniocentesis or CVS if possible. At this point there is still insufficient evidence to evaluate offering women over age 35 years only triple marker screening.

Ultrasound Screening
The abnormalities associated with DS (IUGR, hydrops, some cardiac anomalies, etc.) and differences in long bone length and nuchal fold thickness between DS and normal pregnancies that can be observed on ultrasound during the second trimester have recently been reviewed.<28> Only one prospective clinical trial of sonographic screening for DS fetuses has been reported. The sensitivity and positive predictive value of fetal nuchal fold thickening for DS was 75% and 25% respectively (12 out of 16 DS fetuses from a sample of 3,338 women were detected by sonography).<29> As the sample consisted mainly of high-risk women with advanced maternal age or other factors (DS rate of 1/209) the positive predictive value would be significantly lower in lower risk women. The trial did not address important measurement issues. Small differences in technique, equipment or operators may have a substantial impact on screening performance, and the results obtained from a small group of well trained and equipped operators in a research context will not necessarily generalize to widespread use.

Amniocentesis
Second trimester amniocentesis has been demonstrated as extremely accurate and reliable for prenatal diagnosis of DS with a controlled trial, and cohort studies.<30-33>

Chorionic Villus Sampling
Transcervical chorionic villus sampling (CVS) is a first trimester alternative to amniocentesis for diagnosis of chromosomal disorders. Accurate prenatal diagnosis has been obtained in over 99% of high-risk women with CVS with controlled trials vs. amniocentesis,<34-36> and cohort studies.<37-42> Amniocentesis is necessary to clarify the diagnosis, or to obtain a karyotype when CVS fails, as occurs for up to 5% of women. Transabdominal CVS is a new alternative to transcervical CVS and has comparable accuracy.<34,43,44> Placental position and operator experience may best determine the choice of these two procedures. Errors in diagnosis with CVS usually involve sex determination errors from maternal cell contamination, or mosaicism due to karyotypic abnormalities confined to placental tissue.

Pregnancy Termination
Pregnancy termination is sought by the majority of women who have undergone prenatal diagnosis revealing a fetus with DS. First trimester abortion following CVS represents the safest form of termination with serious complications being extremely rare.<45,46> Retained fetal products is common with second trimester abortions, but more serious complications are quite rare.<47,48> Women of unknown or D (Rh negative) bloodgroup undergoing induced abortion or amniocentesis should have repeat blood group antibody screening (see Chapter 11). Screening before CVS or other obstetrical procedures as well as in concert with complications is more controversial.

Effectiveness of Prevention and Treatment
No studies have been performed to prove that those utilizing screening and prenatal diagnosis for DS fetuses have better outcomes compared with those who do not. The goal of prenatal diagnosis is the provision of a safe and efficacious means of identifying pregnancies for those couples who wish to exercise reproductive choice.

Women at high risk for DS gestation, identified either by screening, advanced maternal age, or previous DS birth, can be offered accurate prenatal diagnosis in the second trimester with amniocentesis. Those identified at high risk prior to pregnancy, or during the first trimester can be offered first trimester CVS as an alternative to amniocentesis.

Adverse Effects
Both amniocentesis and CVS are associated with an increased risk of fetal loss. Some studies have suggested a procedure-related loss of up to 0.8% with amniocentesis,<29-32,49> and over 1% to 1.5% with CVS.<34-36> However, the results of the Canadian Multicentre Randomized Trial<36> have shown that the rate of procedure-related loss with amniocentesis can be as low as 0.04% in experienced hands.

The best evidence of risk from CVS has been obtained from randomized controlled trials of CVS versus amniocentesis.<33-36> These consistently show increased fetal loss when compared with amniocentesis, with an estimated procedure-related risk of fetal loss of at least 1% to 1.5%. Inexperience (i.e. the number performed by centre or operator), and the use of trans-cervical CVS rather than trans-abdominal approach appear related to a greater risk of fetal loss in these studies.

Existing trials of amniocentesis and CVS have inadequate sample sizes for detection and statistical testing of rare adverse effects. There have been suggestions of an increase in neonatal respiratory disease following amniocentesis,<30> and an increased risk of fetal limb reduction anomalies with CVS in case-control studies,<50,51> though the significance is unclear. The risk of limb reduction and other anomalies following CVS appears related to gestational age and CVS is discouraged prior to 10 weeks gestation.<52> Severe maternal complications with amniocentesis and CVS are rarely reported.<30-44>

Psychological distress associated with prenatal screening can include the fear of revealing an abnormal pregnancy, and facing a decision about pregnancy continuation, as well as anxiety over possible complications from diagnostic procedures and abortion. Women who have had a positive screening test may have greater distress than those women at the same risk from advanced age.<53,54> Distress is reduced following a diagnostic procedure, with confirmation of a normal pregnancy, but some anxiety related to the false positive test may persist despite reassurance.<55> No studies have addressed the potential for distress experienced by the significant number of women who test positive on the screen, but decline prenatal diagnosis and or pregnancy termination.

There is little available evidence on the long-term psychological implications of having used prenatal screening, diagnosis and termination of abnormal fetuses. Likewise, there is still little evidence about the burden of fear of giving birth to a DS child.

Decision Making in Prenatal Diagnosis
Critical ethical issues are raised by selective abortion for DS pregnancies. Full discussion of these issue is beyond the scope of this review. Society may interpret the offer of diagnosis and termination for DS fetuses as an implicit message that DS is by definition an undesirable state, and DS individuals worthless. Voluntary reproductive services may be promoted, seen, and evaluated by some in eugenic terms. Evolving societal pressures (including economic), may eventually serve to constrain choices, and create a stigma for the family with a DS individual. Diagnostic and preventive services must be aimed at increasing individuals’ control. They must be voluntary, not routine or expected, and offered in a value-sensitive fashion with emphasis on reliable information about DS and not just about the procedures.

Utilization of prenatal diagnosis is related both to views on the acceptability of pregnancy termination and the perceived risk of the fetus being abnormal.<56> Health care professionals play an important role in shaping these beliefs and many women feel that the potential for persuasion does exist.<57> The perception of harm or the nature of the disability may play a greater role in the decision than the actual probability of its occurrence.<58-61>

Given the low positive predictive value of a positive screen (1 to 2%) the effectiveness of this maneuver is usually predicated on an intention to proceed with amniocentesis in the event of a positive test, as well as on the couple’s understanding of the effect that a DS birth may have. The availability of support (DS societies, etc.) to assist in coping with a DS child, as well as the family’s own resources for coping should be considered when assessing the potential impact of a DS birth.

Informed consent for the screen must include details and risks of second trimester amniocentesis and pregnancy termination. A delay in obtaining test results, and in arranging procedures can mean that a women presenting for the triple screen at 16 weeks may not be informed of the result or be able to book an amniocentesis until 18 to 19 weeks gestation, with a further delay in obtaining results and arrangements resulting in termination of pregnancy at or beyond 21 weeks. Some women of advanced maternal age may prefer first trimester CVS for prenatal diagnosis rather than utilizing triple screening which would require second trimester amniocentesis if positive.

Recommendations of Others
The U.S. Preventive Services Task Force,<62> and The Society of Obstetricians and Gynecologists of Canada in association with the Canadian College of Medical Geneticists<63> have all recommended that amniocentesis or CVS be offered to high-risk women with a family history of DS, translocation or advanced maternal age. Other than recognizing their investigational status, no other organizations have addressed the use of multiple maternal serum markers or ultrasonography for DS screening, although the U.S. Preventive Services Task Force recommendation is currently under review. The Cochrane Pregnancy and Childbirth Group has recently reviewed genetic amniocentesis, ultrasound guidance during 2nd trimester amniocentesis, CVS compared with amniocentesis, transabdominal compared with transcervical CVS, and early amniocentesis.<64-68> They concluded that both amniocentesis and CVS are accurate, and that, although amniocentesis has a lower procedure-related fetal loss, the earlier diagnosis obtainable with CVS may make it more desirable for some individuals. Transabdominal CVS has been associated with less procedure-related loss than transcervical CVS.

Conclusions and Recommendations
There is fair evidence to support offering prenatal diagnosis, with CVS or amniocentesis to pregnant women with identified risk factors: advanced maternal age of 35 years or older, a history of previous DS pregnancy, or known carrier status for a chromosomal rearrangement associated with DS (B Recommendation). Fetal loss appears slightly increased with amniocentesis (approximately 0.8% procedure related fetal loss), and CVS (1.0% to 1.5%). The optimal choice of procedure may be influenced by a host of factors. An A Recommendation cannot be made because good evidence is lacking on personal and family outcomes, and the balance of risks and benefits for the group as a whole. However, it is clear that those women at very high risk (e.g. previous DS child, etc.) and who have significant anxiety over this possibility may benefit substantially from prenatal diagnosis.

Triple marker screening in the second trimester with AFP, HCG and uE3 when combined with maternal age-specific risks can offer an approximate 50% risk reduction for low-risk women (less than 35 years), with a false positive rate of approximately 5%. The evidence is derived from cohort studies using amniocentesis as the gold standard in those who screened positive, and follow-up through a regional cytogenetics laboratory in those who screened negative or declined amniocentesis. The results were consistent between trials when accounting for the differing cut-offs chosen. The positive predictive value of a positive screen is similar to, or higher than the age-related risk in women with advanced maternal age (>35 years) who are currently offered prenatal diagnosis, and the number of amniocenteses per DS fetus identified are fewer.

The studies of triple marker screening meet the Task Force criteria for type II evidence.<31> The Task Force concludes that there is fair evidence (B Recommendation) to support offering triple marker screening to women under 35 years of age when a comprehensive screening and prenatal diagnosis program is available (including education, interpretation and follow-up). It should be recognized that triple marker screening is not supported by the same strength of evidence or efficacy when compared with amniocentesis or CVS for advanced maternal age and other risk factors.

The Task Force is concerned with the limited sensitivity of the screening test, the number of women with false positive screens, and the sub-optimal rate of follow-through to amniocentesis noted in the trials. These limitations place a heavy burden on family physicians and obstetricians to provide full information to those couples interested in the screen. Even when a comprehensive screening program is available, many physicians may consider the burden of counselling low-risk women onerous. Screening of maternal serum markers outside of a fully coordinated program is undesirable. Triple marker screening followed by prenatal diagnosis may be offered to women over 35 years of age as an alternative to prenatal diagnosis alone. There is insufficient evidence for the offer of triple marker screening to replace exclusively the offer of prenatal diagnosis with CVS or amniocentesis to women with advanced maternal age.

Informed consent prior to triple marker screening must include: 1) the limitations in screening sensitivity and specificity; 2) the risks and harms associated with prenatal diagnosis and second trimester abortion; and 3) the psychological implications of screening and diagnosis, as well as the implications of having a DS child. Women consenting to the screen must also be aware of the delays inherent in the process and must understand the nature of a 20 week abortion (i.e. induction and delivery of a fetus).

There is a lack of sound evidence to support the use of individual maternal serum markers (such as AFP alone) for DS fetus screening. Maternal serum AFP measurement may be used to screen for neural tube defects (see Chapter 7). The optimal timing for AFP measurement, after 16 weeks, does not overlap with the optimal timing of DS serum marker screening, 15 weeks gestation. Ultrasonographic screening using long bone, and nuchal fold indices is not recommended for DS screening as part of the periodic health examination of pregnant women. Given insufficient evaluation for effectiveness and the concerns regarding measurement reliability and generalizability, there is insufficient evidence to support a recommendation for routinely offering ultrasonographic screening for early diagnosis of DS fetuses.

In these recommendations sole consideration has been given to the prenatal diagnosis of DS. Other chromosomal anomalies are frequently detected during prenatal diagnosis (Turner’s syndrome, Trisomy 13, etc.) and many may consider their detection important. Other chromosome anomalies have not been considered independently in these recommendations as the diagnostic issues with CVS and amniocentesis are similar to those with DS, there are few studies directly addressing these anomalies, and screening maneuvers have not been sufficiently evaluated to warrant review at this point. These recommendations do not apply to women who are at risk for non-chromosomal genetic abnormalities.

Unanswered Questions (Research Agenda)
There are a number of outstanding questions about DS prenatal screening and diagnosis. Continued evaluation of the triple screen is necessary. Research is in progress on screening with maternal HCG and AFP (double marker screening) and some experts suggest it should replace triple marker screening. Better age- and race-specific predictive values for triple marker screening are needed to provide confidence for individual estimates within subgroups such as women with significantly advanced maternal age. The value of routine ultrasonography for gestational age determination at the time of triple marker screening requires clarification. The development of a more accurate first-trimester DS screening technique would significantly improve the effectiveness and minimize harms associated with triple marker screening.

The potentially adverse effects of screening and diagnosis require further clarification. Information is needed on the outcomes of women with false positive screening tests, and those with positive screening tests who decline prenatal diagnosis and/or pregnancy termination. Comparison of screening tests followed by prenatal diagnosis vs. prenatal diagnosis alone for women of advanced maternal age or with other risk factors is indicated. The outcome of infants born following prenatal diagnosis requires continued monitoring with special attention to rare events such as limb reduction defects following CVS.

The impact of prenatal screening and diagnosis on societal perceptions of disabled individuals deserves attention. The consequences of DS births and alternatives to screening and diagnosis must be better understood, including the impact of coping skills and supportive interventions on the quality of life in families with handicapped children.

Evidence
The literature was identified with a MEDLINE search, from 1966 to March 1993, using the key words Down Syndrome, prenatal diagnosis, prevention, epidemiology and diagnosis subheadings. The review was initiated in February 1993 and recommendations finalized in March 1994 by the Task Force.

Acknowledgements
The principal author would like to acknowledge the assistance of Dr. William Feldman, and Dr. Elaine Wang in reviewing the evidence, and of Karen Huntley in preparing the manuscript.

Full Citation
Dick P. Prenatal screening and diagnosis for Down syndrome prevention. In: Canadian Task Force on the Periodic Health Examination. Canadian Guide to Clinical Preventive Health Care. Ottawa: Health Canada, 1994; 84-98.