Challenging the Strategy of Maternal Age–Based Prenatal Genetic Counseling

The decision to offer pregnant women prenatal diagnosis for genetic diseases has long had an age-based trigger.^{1 - 2} Early studies convincingly demonstrated an escalating rate of aneuploidy with advancing maternal age,³ but risks related to amniocentesis are independent of maternal age. When the flat risk-line from the invasive diagnostic procedure (iatrogenic pregnancy loss) crossed the age-related ascending risk-line of aneuploidy on a graph of adverse events, an intersection was determined that became the age standard for offering the procedure. Data from the 1970s appeared to suggest that at the age of 35 years a woman had a similar chance of delivering a child with aneuploidy³ as experiencing a miscarriage after an amniocentesis,⁴ and that age became “the” prompt for genetic counseling.² Much has been learned since then, but despite the substantially increased efficiency of diagnostic technologies and possibly reduced risks of procedure-related pregnancy loss,⁵ the recommendation remains unchanged. According to the current guidelines of the American College of Obstetricians and Gynecologists, “Women with singleton pregnancies who will be age 35 years or older at delivery should be offered prenatal diagnosis.”⁶ We suggest that advances in science and careful consideration of the ethics of testing should undermine any residual support for that standard.

Screening is performed to identify a population at increased risk for a particular problem so that scarce, expensive, or potentially dangerous diagnostic procedures can be offered to a group that would benefit most from the risk-benefit equation. If karyotype determination could be performed noninvasively, and at no risk to the fetus, the primary argument for not offering universal karyotype screening would be financial cost. However, the performance of amniocentesis carries a relatively small but real risk of pregnancy loss (0.5%-1%).^{4 ,7} Therefore, these procedures should be performed in as few unaffected pregnancies as possible, while at the same time maximizing the effectiveness of diagnosing affected fetuses. To make sense of studies on this topic, it is useful to compare the sensitivity of a screening test (detection rate of the disorder being sought) with its screen-positive rate (percentage of the population with a “positive” test result), because the latter will contain a significant number of patients who do not have the disorder (false-positive rate), as well as some who do.

Maternal age is a screening test for aneuploidy, but an extremely poor one. While the risk of having an infant with an abnormal number of chromosomes increases with maternal age, most older women deliver infants with normal karyotypes. Although older women are more likely than younger women to conceive aneuploid offspring,⁸ the majority of children with Down syndrome are born to younger mothers because they deliver a far higher proportion of all children. In a recent study of more than 38 000 women, Malone et al⁹ reported that the Down syndrome detection rate would have been only 31% if diagnostic testing had been limited to women who were 35 years or older. Furthermore, if the criterion for offering the test was age alone, by definition the screen-positive rate would be 100% for women 35 years or older and approximately 15% for the entire population.

The use of second-trimester maternal serum screening and combined ultrasound and maternal serum screening in the first trimester has substantially increased the sensitivity of screening for Down syndrome and other major karyotypic abnormalities. Maternal age is factored into the algorithms used in all of these more sophisticated forms of aneuploidy screening. In the study by Malone et al,⁹ the detection rate for Down syndrome was 81% using second-trimester maternal serum levels of α-fetoprotein, estriol, total human chorionic gonadotropin (hCG), and inhibin (the “quad” test), and 87% using first-trimester nuchal translucency measurements combined with maternal serum pregnancy-associated plasma protein A and free-β hCG with a 5% screen-positive rate. These detection rates can be improved and the screen-positive rates further reduced by combining first- and second-trimester screening studies in a variety of different ways.¹⁰ The question surrounding the current standard is why should potentially dangerous testing be offered to all women aged 35 years or older to detect slightly more than 30% of the affected cases, when the detection rate can be nearly tripled and invasive testing can be offered to fewer women who are actually at low risk for having a fetus with an aneuploidy.

The original basis for designating 35 years as the appropriate age for screening was the belief that at that age the risks of pregnancy loss from amniocentesis and the chance of detecting an aneuploid fetus reached numeric parity.^{3 - 4} Recent data suggest that the current rates of procedure-related fetal loss associated with amniocentesis may be much lower than those cited in earlier studies,⁵ but that is not the key ethical point in discussions of triggers for screening. There may be inherent appeal when beneficence (detecting an anomaly) and malfeasance (causing an iatrogenic pregnancy loss) are quantitatively matched. However, the more germane question is whether a 0.5% risk of losing a normal pregnancy⁴ is the same as a 0.5% chance that the fetus has Down syndrome.³

These rates may be the same, but the adverse outcomes are very different. For some patients who have struggled with infertility for years the thought of losing a karyotypically normal fetus may be devastating, whereas raising a child with Down syndrome is less concerning. For others, the notion of having a developmentally disabled child may be overwhelming, whereas fears of being able to conceive another pregnancy are not as relevant. These 2 data points are not equivalent in any meaningful sense, other than being numerically identical.

In 1979 a National Institutes of Health Consensus Development Conference concluded that “different women (couples) will interpret differently the burden of both a chromosomally abnormal child and of miscarriage. Therefore, their conclusions will differ as to whether or not an amniocentesis is advisable.”¹¹ In 1994, Nicolaides noted, “The perception of risk of miscarrying a wanted pregnancy, or the birth of a chromosomally abnormal baby is dependent on the expectations of the parents,”¹² and studies by Kuppermann et al confirm this to be true.^{13 - 14}

An argument in favor of offering invasive testing to all women aged 35 years or older is that while currently available screening tests have high sensitivities for detecting Down syndrome and some other aneuploidies,¹⁰ these tests are not effective in detecting fetuses with abnormal numbers of X chromosomes, which also increase with advancing maternal age. While these karyotypic abnormalities occur more frequently as women age, the actual number of affected fetuses is quite small. The risks of having a fetus with either 47,XXY or 47,XXX karyotypes are 1:1250, 1:833, and 1:303 at maternal ages 25, 35, and 40 years, respectively.¹⁵ Furthermore, the degree of disability associated with those disorders is considerably less than that associated with the other major aneuploidies.

There is another potential problem related to changing current recommendations, ie, no longer offering invasive testing to all women of “advanced maternal age.” If testing all women aged 35 years or older was no longer considered the standard of care, insurance companies might refuse to pay for the test in an older woman who simply wants the assurance that her fetus is karyotypically normal even though she had a “negative” screening study. This is not a reason to maintain the status quo. In our opinion every pregnant woman in the United States is entitled to have the karyotype of her fetus(es) determined if she is prepared to accept the risk to the pregnancy of obtaining that information. This is no more relevant for a 41-year-old woman with a screening risk for aneuploidy of 1 in 600 than it is for a 21-year-old with the same risk determined by comparable testing. The information obtained from karyotype determination has the potential to be of enormous importance to all obstetrical patients, and reimbursement policy should not mandate that a specific risk level must be superseded before approving diagnostic testing for women of any age.

Women who are offered amniocentesis for karyotype determination without prior serum testing, ultrasound screening, or both should be counseled about the risks and benefits of that option. While disclosure in the informed consent process previously was deemed adequate provided it met the common practice of the profession, more recent trends in the courts have moved beyond that standard to one requiring that the reasonable informational needs and expectations of ordinary individuals must be met.¹⁶ The older standard may have been met by offering invasive diagnostic tests to all women aged 35 years or older and the subset of younger women with calculated risks above a given threshold, but it is not as clear that the newer, more rigorous standard would be satisfied by that practice.

There is no a priori reason to believe that physicians and genetic counselors are fulfilling the reasonable needs and expectations of a woman when they assume that if her risks do not exceed a given trigger she has no interest in further information. That is substituting paternalism for individualized counseling by assuming the counselors know the level of risk a woman is willing to tolerate, or how she will balance potential procedure-induced miscarriage risks against those of detecting an aneuploidy. All women are entitled to know the results of their tests, regardless of how remote the estimated risk of aneuploidy. Appropriate counseling requires clear communication of adequate information to allow an individual to gauge the benefits and risks of a given choice, rather than merely foreclosing on choice by anticipating how an individual will weigh competing burdens.

The most rational approach for counseling any woman (or couple) interested in knowing the karyotype of a fetus is to allow her to choose a screening test that best meets her particular needs (the highest detection rate vs the earliest results vs the lowest screen-positive rate, etc) and then provide the actual risk figures derived from that study.¹⁰ At the same time, the patient should be informed about the risk of pregnancy loss from chorionic villus sampling or amniocentesis done by the physician who will actually perform that procedure in her case. She can then decide whether and how to proceed. With this approach, in a prenatal diagnostic universe of rapidly expanding potential and complexity, the core obstetrical ethic of respect for patient autonomy will continue to be honored. Seven years have passed since Kuppermann et al made this argument,¹⁴ but official recommendations and practice patterns in the United States have remained unchanged. Given the advances that have been made in prenatal screening, maternal age should no longer be used as a solitary indicator for offering women invasive testing for karyotype determination.