Prenatal diagnostic tests have low risk of miscarriage

Aug 31 2006

Pregnant women who seek prenatal diagnostic testing to identify genetic or chromosomal abnormalities have a lower risk of miscarriage than previously believed, according to a UCSF study.

The findings are published in the September 2006 issue of the journal "Obstetrics and Gynecology."

Two standard tests--amniocentesis and chorionic villus sampling (CVS)--are common prenatal tests performed during the first and second trimester of pregnancy. Early testing using the CVS procedure has typically been thought to have a higher rate of miscarriage than amniocentesis. However, in a 20-year retrospective comparison study of the two procedures, researchers found that the loss rates for both procedures decreased over time.

"This is a significant finding for use as information in both patient counseling and in establishing widespread prenatal diagnostic and screening programs," said study co-author Mary E. Norton, MD, who is medical director of the Prenatal Diagnostic Center at UCSF Medical Center and associate clinical professor in the UCSF Department of Obstetrics, Gynecology and Reproductive Sciences.

Both amniocentesis and CVS are invasive diagnostic screening methods that carry a small risk of pregnancy loss. Amniocentesis requires insertion of a hollow needle through the abdominal wall and into the uterus to withdraw amniotic fluid. CVS is a biopsy procedure that involves removing a piece of tissue from the placenta. These samples are then cultured and chromosomes analyzed to determine abnormalities associated with Down syndrome and other genetic diseases such as cystic fibrosis, Tay-Sachs disease, and sickle cell disease.

CVS can be performed at an earlier gestation period (10-12 weeks) than amniocentesis (16-20 weeks), and therefore is preferred by some patients because of earlier detection of possible abnormalities, according to Norton.

The study looked at data from 9,886 CVS and 30,893 amniocentesis procedures performed at UCSF Medical Center from 1983 to 2003. Overall, the pregnancy loss rate for patients undergoing CVS was greater (3.12 percent) than that of amniocentesis (0.83 percent). However, when examining the data at five-year intervals, the difference between the two procedures coincided with the highest occurring from 1983 through 1987 and lowest from 1998 through 2003.

"We saw that the rate of miscarriage risk attributable to CVS had declined over time," said Norton. "One possible reason is that over time practitioners have become more proficient in performing CVS procedures."

When the researchers controlled for gestational age and maternal age, they found that there were no differences between the losses from CVS or amniocentesis.

Over the past several years, non-invasive screening tests for birth defects have seen marked improvement, according to Norton. While screening in the past was primarily available only in the second trimester, more recent studies have demonstrated that tests such as a specialized ultrasound of the fetal neck, known as "nuchal translucency " ultrasound, combined with early blood screening have a better detection rate and can be completed much earlier in pregnancy. Pre-screening using these techniques is often performed as the precursor for the more invasive diagnostic procedures.

"We have advanced significantly in our ability to screen for and detect birth defects," said Norton. "Where the primary criteria for determining a high risk pregnancy used to be maternal age, we are now able to perform blood tests or ultrasound in the first trimester, enabling us to determine much more accurately if a pregnancy is indeed at high risk."

In the past, women were faced with the decision to have CVS earlier in the pregnancy but with a higher risk of miscarriage, or to wait until their second trimester for amniocentesis, which was thought to be safer but doesn't provide results until several weeks later, according to Norton. With more and more women undergoing first trimester screening, an increasing number of them will be faced with the choice between amniocentesis and CVS, she added.

"If there is no difference in risk associated with CVS and amniocentesis, women are more likely to choose CVS because information can be provided much sooner," Norton said.

The authors emphasized that these results were obtained in a single center with a large volume of procedures. Because rates of miscarriage vary with experience of the providers, the results may not be applicable in all situations.