Chorionic villus sampling: experience with an initial 940 cases.

Of over 1000 patients referred to the Genetics & IVF Institute for chorionic villus sampling, 10% had an abnormal gestational sac at the time of initial presentation. Villus sampling could not be performed in 0.6% of cases, usually because of fibroids. Chorioamnionitis, probably related to the procedure, occurred in two patients (0.2%). The total fetal loss was 2.4%, which is not significantly different from the background fetal loss rate reported in normal pregnancies after eight weeks of gestation. The procedure-related loss was estimated as approximately 0.6%. The cytogenetic analysis routinely used was "direct" trophoblast karyotyping. There were no false-negative diagnoses. Trophoblastic mosaicism was observed in 1.3% of samples. Amniocentesis was performed in 3.5% of cases because of chorionic villus sampling failure, chorionic mosaicism, or elevated maternal serum alpha-fetoprotein (MSAFP) concentrations. Chorionic villus sampling was not associated with falsely elevated MSAFP levels at 15-18 weeks' gestation. Chorionic villus sampling has proved a safe and accurate prenatal diagnostic method associated with relatively low risk when performed by an experienced team, although additional amniocentesis will be necessary in a small percentage of pregnancies.