[Randomized comparison of transabdominal, transcervical chorionic villi sampling and amniocentesis].

The transabdominal chorionic villus sampling method was compared with the transcervical route and second trimester amniocentesis in a 3-winged randomised trial. Examination of 45 epidemiological variables showed the three procedure groups to be comparable at enrollment. In 3079 women at low genetic risk, we compared transabdominal with transcervical chorionic villus sampling and amniocentesis. The total fetal loss was 10.9%, 6.3% and 6.4% in the transcervical, transabdominal chorionic villus sampling groups and the amniocentesis group, respectively (p < 0.001). The two CVS procedures were compared in 2882 low and high genetic risk women with cytogenetically normal fetuses. Rates of unintended post-procedure loss were 3.7% and 7.7% for transabdominal CVS and transcervical CVS, respectively (p < 0.001), difference in rates 4.0%, 95% C.I. +2.3% to +5.8%. By a priori ultrasound scanning, more transcervical than transabdominal procedures (p < 0.001) were considered to be inaccessible for sampling. Our data indicate that transabdominal allows freer access to the placental site than transcervical sampling and is easier to adapt to than transcervical CVS. Women run comparable risks with transabdominal CVS and amniocentesis. Given the results of our study, transabdominal procedures remain the first choice for prenatal diagnosis and, since in our hands transcervical sampling entails an increased fetal risk, we have abandoned transcervical CVS in our two study centres.