To compare our initial experiences with early amniocentesis and transabdominal chorionic villus sampling (CVS). We compared the diagnostic and pregnancy outcomes of our initial 250 patients undergoing early amniocentesis (at or before the 14th completed week of gestation) or transabdominal CVS (performed between 9.5-12.9 weeks' gestation). In both groups, the indication for prenatal diagnosis was advanced maternal age (35 years or older at estimated date of delivery). No diagnostic errors were made using either technique, and the culture failure rate for both methods was 0.8% (two of 250). Seven cytogenetic abnormalities in the early amniocentesis group and seven in the transabdominal CVS group were detected. Nine of the 250 women undergoing early amniocentesis reported spontaneous miscarriages following the procedure, compared to five in the transabdominal CVS group. The loss rates were 3.8% in the early amniocentesis group and 2.1% in the transabdominal CVS group among continuing pregnancies. Frequencies of premature delivery, small for gestational age infants, and associated structural defects in both groups were comparable. At this time, early amniocentesis cannot be assumed to be equal to conventional transabdominal CVS or amniocentesis with regard to safety or accuracy; only a large cohort randomized study will adequately determine the safety and efficacy of early amniocentesis.