Objective: To assess the rate of fetal losses in twin pregnancies undergoing genetic mid-trimester amniocentesis.
Methods: In the first part of this investigation, a retrospective cohort study compared a group of women . 32 years old with twin pregnancies who underwent amniocentesis with a similar group unexposed to amniocentesis. Data were compiled from January 1990 to March 2004 for patients from a single institution. Pregnancies complicated by twin-to-twin transfusion syndrome, monoamniotic twins, or lethal fetal anomalies, and those treated by fetal reduction were excluded. The primary outcome was the loss of one or both fetuses prior to 24 weeks' gestation. In the second part of the investigation, a systematic review of the literature and a meta-analysis were performed.
Results: In the first part of the study, data were collected for 132 women exposed to amniocentesis and 248 women not exposed to amniocentesis. There was no significant difference in the rate of fetal losses between the two groups (3.0% vs. 0.8%, P = 0.10). No losses occurred within four weeks of the procedure. In the second part of the investigation, four studies, including ours, were considered for a meta-analysis of 2026 women with twin pregnancies. Compared with women unexposed to the procedure, amniocentesis in women with twin pregnancies increased the risk of fetal losses prior to 20 to 24 weeks' gestation (odds ratio 2.42; 95% confidence intervals 1.24-4.74, P = 0.01) with an additional risk of one adverse outcome (1 or 2 fetal losses) for every 64 amniocenteses.
Conclusion: Genetic mid-trimester amniocentesis in twin pregnancies is associated.