Chorionic villus sampling as safe as amniocentesis

NEW YORK (Reuters Health) - For women undergoing prenatal genetic testing, chorionic villus sampling (CVS) is no more likely than amniocentesis to lead to pregnancy loss, investigators report.

Because CVS can be performed earlier in the pregnancy, women can now be assured on the basis of these findings about the safety of early diagnosis, they say.

During amniocentesis, a sample of amniotic fluid is extracted by needle, while CVS takes cells from the placenta.

Most of the studies comparing CVS and amniocentesis were conducted in the 1980s, soon after CVS became available. Those studies revealed higher miscarriage rates when CVS was conducted, most likely because doctors were not yet adept at performing the new procedure, Dr. Aaron B. Caughey at the University of California, San Francisco and colleagues note in the Obstetrics and Gynecology.

However, if experience in CVS has improved over time, it could become the preferred procedure, because results can be obtained during the first trimester, whereas amniocentesis is usually performed in the second trimester.

To see if miscarriage rates with CVS have improved over time, Caughey's group reviewed records of patients treated at the Prenatal Diagnosis Center at the University of California, San Francisco between 1983 and 2003. Their analysis included nearly 10,000 CVS and 32,000 amniocentesis procedures that yielded normal results.

Prior to 1993, there were significant differences between procedures, favoring amniocentesis, the team found. Over time, the loss rate decreased for both procedures, but the change was significant only for CVS.

In "multivariate analysis," prior to 1993, CVS was associated with a significantly greater pregnancy loss than amniocentesis. However, the loss rates were no longer significantly different from 1993 to 1997.

Moreover, differences between the two procedures were both statistically and clinically nonsignificant between 1998 to 2003.

The authors estimate that in the period from 1998 to 2003, the adjusted loss rate was 1 in 370 for amniocentesis, compared with 1 in 360 for CVS.

If replicated in other centers, these results are of "paramount importance" in this new era of first trimester genetic screening, Caughey's team writes, because termination of pregnancy is safer during the first trimester if that is the course suggested by screening results.

SOURCE: Obstetrics and Gynecology September 2006.