It seems the obstetric triple screen is one of the most confusing and anxiety-provoking tests that our pregnant patients go through. While the test itself requires only a single tube of blood from the mother, the psychological impact of testing for a severe abnormality in the fetus is immense. That tube of blood forces parents to consider some very difficult moral and ethical issues regarding how they would handle a diagnosis of a severe spinal defect or Down’s syndrome in their fetus.

In the next few paragraphs I’ve attempted to answer some of the most common questions that arise regarding the use of these screening tests.

The obstetric triple screen measures three markers in maternal blood. Alpha-fetoprotein (AFP) is produced in the fetal liver. It is normally found in amniotic fluid as a result of fetal urination and in the maternal blood as the result of passage across the placenta or membranes surrounding the fetus. AFP levels may be increased in the presence of certain developmental defects in the fetus. Abnormally low levels have been noted in some cases of Down’s syndrome. The second marker, Estriol is an estrogen produced by the placenta which has been found to be decreased in cases of Down’s syndrome. And finally, Human Chorionic Gonadotropin (HCG) is the "pregnancy hormone" which is the basis for most pregnancy tests. It is produced by the placenta and is generally increased in cases of Down’s syndrome.

What exactly is being tested for?

The neural tube defects (often referred to as Spina Bifida) are a group of congenital abnormalities of development of the central nervous system. Specifically, there is an abnormal opening in the head or spinal column which can be associated with negligible to severe neurologic impairment of the fetus. The prevalence of neural tube defects varies widely throughout the world. It is most common in Ireland, Scotland and Wales where about 1% of fetuses are affected. In the United States only 1-2 per 1000 fetuses are affected. Large scale screening for neural tube defects was first begun in the United Kingdom in the early 1980’s. Routine voluntary screening by measuring AFP in maternal blood came to the Unites States in 1986. The screen is felt to be able to detect 80-90% of significant neural tube defects.

Down’s syndrome, or Mongolism, is the most common chromosomal disorder found in newborns. An association between Down’s syndrome and abnormally low maternal AFP level was first noted in the early 1980s. Researchers found that 45-50% of all Down’s syndrome fetuses could be detected by offering AFP testing to all women. When HCG and Estriol are measured in conjunction with AFP (the Obstetric Triple Screen), the sensitivity for Down’s syndrome increases to approximately 60%.

The use of AFP for the detection of neural tube defects and the Obstetric Triple Screen for the detection of Down’s syndrome are both examples of screening tests. The ideal screening test would be 100% sensitive (meaning that it always detects the problem you’re looking for) and 100% specific (meaning that it only shows positive when a problem is truly present). Unfortunately, we do not live in an ideal world. Medical screening tests are designed to be as sensitive as possible so that problems won’t go undiagnosed. However, as a test becomes more and more sensitive, greater numbers of people who do not have the problem in question will screen positive because of the decreased specificity of the screen.

A quick example may shed some light on the concerns that many have voiced about the inaccuracies of these tests. If 1000 women are screened for neural tube defects, we would expect one of the fetuses to be identified as having a neural tube defect (prevalence of 1 per 1000 in the US). However, because the specificity of the test is only 95%, 50 patients (the remaining 5% of 1000) will have positive screens. This means that 49 out of 50 mothers with positive test results will still have normal fetuses. In other words, the "false positive" rate is 98%.

AFP, Estriol, and HCG blood levels are usually drawn from the mother between the 15th and 20th weeks of pregnancy. The laboratory analysis of the three markers takes approximately 7 days and the results are reported back to the doctors’ office. In our office, as with other testing done during the pregnancy, the patient is contacted by phone if there is an abnormal result. Otherwise the test results are reviewed with the patient at her next regularly scheduled visit.

If the test is abnormal, the subsequent evaluation depends on whether the test is pointing to an increased risk of neural tube defects or Down’s syndrome. A test is deemed positive for an increased risk of neural tube defects when the AFP level is greater than 2.5 times the median (midpoint of levels for a group of patients at the same gestational age), or 2 .5 MOM (multiples of the median). The first step in evaluation of an elevated AFP is an ultrasound examination. Although the sonogram is primarily used to confirm the dating of the pregnancy, other causes of elevated AFP levels including the presence of twins or congenital abnormalities can sometimes be seen.

If the dating of the pregnancy is correct, the test can simply be repeated in two weeks as long as the patient is still less than 20 weeks pregnant. If an abnormality is present in the fetus the AFP level will become more abnormal, whereas in the case of a false positive the AFP level will move toward that of the normal population. An initial level greater than 3.0 MOM is not expected to decrease into the normal range when repeated. In the rare case where the AFP level is above 3.0 MOM or remains abnormal after repeating, an amniocentesis may be considered. Amniotic fluid obtained through amniocentesis can be tested for AFP to see if the level is truly elevated or if the elevation in the maternal blood might simply be due to a "leaky" placenta. If the amniotic fluid AFP is not elevated, there is little further concern about neural tube defects, but the pregnancy may be at increased risk for poor fetal growth, fetal distress, preterm labor, and other related problems which, at least in theory, are caused by the poorly functioning, "leaky" placenta. If the amniotic fluid AFP is elevated, then another protein, acetylcholinesterace, can be measured to help distinguish between the presence of a neural tube defect and other possible congenital abnormalities which might also liberate elevated levels of AFP.

An increased risk for Down’s syndrome is reported when the laboratory evaluation of the three markers taken together result in a risk for Down’s syndrome which is equal or greater than 1 in 270. This cutoff is equal to the risk that a 35 year old woman will have a baby with Down’s syndrome. Age 35 has traditionally been the time at which women were offered amniocentesis for the diagnosis of chromosomal abnormalities. In reality, nothing magical happens to increase one’s risk at age 35. This is simply the age at which the risk of the amniocentesis (risk of miscarriage is usually quoted at about 1 in 250) is about equal to the risk of having a baby with a chromosomal abnormality. As such, it seems reasonable to offer an amniocentesis to these patients. Whether or not a patient proceeds with an amniocentesis is a highly personal decision which may depend greatly on how she would use this information.

The evaluation of an abnormal obstetric triple screen also begins with an ultrasound examination. Again this is done primarily to confirm the dating of the pregnancy, however, the radiologist will give special attention to aspects of the fetal anatomy which may be altered in a fetus with Down’s syndrome. Although a normal sonogram is reassuring, it has been shown that only a minority of fetuses with Down’s syndrome will be detected because of sonographic abnormalities. If the dating of the pregnancy is confirmed, amniocentesis should be offered to the patient. There is no place for repeat testing of obstetric triple screens if the dating is confirmed. Especially for AFP, the normal distribution of values for normal and Down’s syndrome pregnancies have a large amount of overlap. For this reason, an initial low value will tend to improve upon repeat testing in both normal and Down’s syndrome pregnancies. Repeating the test, therefore, can only decrease the detection rate for Down’s syndrome.

I’ m already over 35. Should I have this test?

Since the cutoff for an abnormal obstetric triple screen is based on the risk of a 35 year old woman, a few words should be said on the usefulness of this screen in women who are 35 years or older. First, the normal values which are used by laboratories to evaluate these tests are extrapolated from women under the age of 35. There simply is not enough data on older women to know whether this is a valid means of interpretation. Second, the test can not be used to decrease one’s age-related risk of Down’s syndrome. If a 38 year old patient’s obstetric triple screen is reported as 1 in 500 risk of Down’s syndrome, her actual risk is still equal to the age-related risk of 1 in 100. On the other hand, if the obstetric triple screen risk is reported as 1 in 20, then it is reasonable to assume that her risk of having a baby with Down’s syndrome is increased compared to other women her age. An amniocentesis is the only way to know for certain whether the fetus is affected.

Suffice it to say, these tests are far from perfect. The high false positive rate has caused many a sleepless night for anxious parents, and certainly many hours of explanation and consoling from obstetricians like myself. Despite this, I still believe the tests may provide useful information about a pregnancy, and are generally reassuring. The trick is to understand a test’s limitations. Screening tests are developed to tell us if a patient is at risk for a certain disease, not to diagnose that disease. Further evaluation is always required to make a diagnosis, and as noted above, that diagnosis is usually "NORMAL!"

Walter J. Hodges Jr., M.D., F.A.C.O.G.