Difficulties encountered in a randomization trial of CVS versus amniocentesis for prenatal diagnosis

In April 1985, having completed a study of the short-term complications of chorionic villus sampling (CVS), we began a randomized comparison of CVS versus amniocentesis. Our study continued over a 15-month period, and during that time we had difficulty recruiting patients, with only 10.6% of 1254 women referred for prenatal diagnosis fully participating in this study. However, 30.2% of those eligible by dates and indication chose to enter the study. CVS was available in our province only through this study, and the two most common reasons for such a low rate of recruitment were reporting too late in pregnancy and the concern about the potential risks of CVS. Patients continued to seek counselling too late for CVS despite direct and continuous contact with regional physicians. Our patients' concern about risk might well vary with the attitude of their physicians towards CVS, and with the information provided at the time of pre-test counselling. The small number of patients actually enrolled did not permit any meaningful comparison of amniocentesis to CVS. However, our experience with pregnancies lost post-CVS suggests that a pregnancy with an apparent low implantation at the time of sampling may be at a higher risk of loss.