Technical factors in early amniocentesis predict adverse outcome. Results of the Canadian early (EA) versus mid-trimester (MA) amniocentesis trial

The purpose of this study was to identify risk factors for fetal loss and other pregnancy complications associated with genetic amniocentesis. Data were acquired in the Canadian Early Amniocentesis Trial (CEMAT), a multicentred (12) prospective, randomized trial comparing continuous ultrasound-guided early amniocentesis (EA) and mid-trimester amniocentesis (MA) (CEMAT Group, 1998). Details of the procedure were recorded and analysed by allocation (EA versus MA), operator and centre, and correlated with pregnancy outcome. A total of 62 spontaneous pregnancy losses occurred between the procedure and 20 weeks' gestation among the 3691 patients who received their procedures within the allocated window (EA = 53/1916, MA = 9/1775). Technical factors correlating with these losses included procedures 'judged to be difficult' by the operator, and post-procedure amniotic fluid leakage or bleeding. Maternal risk factors included maternal hypertension (fetal loss 11.1 per cent, compared with non-hypertensive women, 2.6 per cent) increased body mass index (BMI) and gravidity of three or greater. Allocation to EA was predictive of fetal loss, as well as failed procedure, multiple needle insertions, amniotic fluid leakage, failed culture and talipes equinovarus, in excess compared with MA. In conclusion, in this large prospective randomized trial evaluating amniocentesis, specific maternal, fetal and procedural variables were found to be predictive of fetal loss and adverse pregnancy outcome. Performing amniocentesis before 13 weeks' gestation (EA) was the major predictive factor for adverse outcome. These data suggest that first-trimester chorionic villus sampling (CVS) and MA will likely remain the invasive procedures of choice for evaluation of fetal karyotype.