February 2001

From University of California - San Francisco

Ultrasound in second trimester is not accurate measure to evaluate Down syndrome risk, UCSF study finds

Using ultrasound to detect Down syndrome in fetuses in the second trimester is not an accurate screening measure, though ultrasound is widely used for this purpose in the United States, according to a University of California, San Francisco study.

While ultrasound has been increasingly used to search for a number of markers for Down syndrome, the vast majority of fetuses that have one of the ultrasonographic markers will not be born with Down syndrome. Moreover, ultrasound results can prompt further, invasive testing that could harm the fetus, the report found.

The study, a systematic review of 56 articles, will be published in the February 28 issue of the Journal of the American Medical Association.

An ultrasound, also called a sonogram, is a test that uses sound waves to make a picture of a fetus. For many years, physicians have used ultrasound to detect certain markers for Down syndrome in fetuses during the second trimester of pregnancy. The seven markers include thickened nuchal fold, or thickening on the back of the neck; choroid plexus cyst, a small cyst in the fetal brain; and renal pyelectasis, a slightly higher level of fluid in the kidney.

These markers came to be used as a measure of risk for Down syndrome because studies initially found a higher incidence of these markers in women at increased risk of having a baby with Down syndrome. Elevated risk is based largely on the mother's age, a blood test that measures the levels of three substances in the mother's blood while pregnant, or a family history of chromosome abnormalities.

"In part because these markers are easy to see in a fetus, a search for these markers was extended to women who are not at high risk," said Rebecca Smith-Bindman, MD, study lead author and UCSF assistant professor of radiology, epidemiology and biostatistics. "If you just see one of these markers, then it's very unlikely the baby has Down syndrome. None the less, women in whom these markers are identified are often counseled that they are at increased risk."

The study found that most of the seven markers could not confirm or exclude the presence of Down syndrome. Only one marker, thickened nuchal fold, may be useful in distinguishing between a fetus that has Down syndrome and one that does not. However, even when this marker is seen, the risk of Down syndrome increases to only three percent, according to the study.

Using thickened nuchal fold and the other markers as a way to determine Down syndrome risk comes at a significant financial cost. For example, doctors would need to screen 15,893 women who are at average risk of carrying a fetus with Down syndrome for thickened nuchal fold to find one case of Down syndrome. They would need to screen 87,413 women at average risk for having a baby with Down syndrome for choroid plexus cyst to find one case of the defect.

The analysis also found many of these markers occur at the same rate in babies with and without Down syndrome.

Down syndrome is caused by a chromosome defect and results in mild to severe mental retardation and many other birth defects. It occurs in one out of 700 births in the United States. While using ultrasound to look for these seven markers to assess Down syndrome risk is not an effective screening measure, according to the study, ultrasound is useful in detecting certain categories of abnormalities that are considered strong indicators of Down syndrome, such as severe brain, gastrointestinal and heart anomalies, Smith-Bindman said.

"When a fetus has a severe abnormality, further invasive testing with amniocentesis is certainly warranted. However, this isn't the situation we are talking about in the study," she said. "These seven markers are harmless to the baby, most resolve in pregnancy and the vast majority of fetuses with these isolated markers are completely normal. When a physician sees one of these markers, I think it is wrong and potentially harmful to say there is something wrong with the baby. Most mothers get extremely anxious when they are informed that their baby is abnormal, when in fact their risk may not even be elevated when one of these markers is seen."

When one or more of these markers is found on an ultrasound, the patient is usually advised to undergo an amniocentesis to confirm whether or not her baby will have Down syndrome. This test involves inserting a needle in a woman's abdomen to remove a small sample of the amniotic fluid that surrounds the fetus. This fluid contains fetal cells. The chromosomes in these cells are examined to detect chromosomal abnormalities. Though this procedure can detect virtually all cases of Down syndrome, it also can result in fetal death. Up to one percent of fetuses are lost in mothers who undergo amniocentesis. "Use of these markers hugely increases invasive testing and only minimally increases the detection of an affected fetus," Smith-Bindman said

According to the study, four fetuses would be lost for every case of Down syndrome correctly diagnosed in women at average risk who are offered an amniocentesis because an ultrasound detected a choroid plexus cyst. Nearly three fetuses would be lost in women at average risk for having a fetus with Down syndrome who are offered an amniocentesis because an ultrasound detected renal pyelectasis.

"For mothers whose fetuses are found to have one of these markers, the risk of complication from amniocentesis is higher than the risk of having an affected baby," Smith-Bindman said. Moreover, ultrasound may miss abnormalities in fetuses who are affected by Down syndrome, and thus a normal ultrasound can give false reassurance to a woman who is at increased risk of having a baby with Down syndrome.

While ultrasound is useful in determining the gestational age a fetus and identifying growth disruptions and severe anomalies, Smith-Bindman said, it should not be used to determine Down syndrome risk. Instead, if women want to know if their fetus is affected by Down syndrome, they should have the blood test. The blood test finds about 70 percent of all Down syndrome cases. If the mother's risk is high based on that test, then she should undergo an amniocentesis even if the ultrasound is normal, as many fetuses affected by Down syndrome will have a normal ultrasound, Smith-Bindman said.

Other study authors are Wylie Hosmer, BS, staff member in the UCSF Department of Radiology; Vickie A. Feldstein, MD, UCSF associate professor of clinical radiology; Jonathan J. Deeks, MSc, Centre for Statistics in Medicine, Institute of Health Sciences, University of Oxford, Oxford, England; and James D. Goldberg, MD, California Pacific Medical Center in San Francisco.

The study was supported in part by the Mount Zion Health Care Systems, the University of California, San Francisco Academic Senate and the UCSF Research Education and Allocation Committee. Smith-Bindman was a Radiological Society of North America Nycomed Amersham Fellow while she completed this research.

This article comes from Science Blog. Copyright 2004

Archives 2001 D