An increased rate of fetal loss following amniocentesis has been reported by some studies; however, their findings have not been confirmed. This uncertainty complicates advising families at risk of having a child with chromosomal or developmental abnormalities about the risks and benefits of prenatal diagnosis. The issue is difficult to study because of ethical problems and the fact that patients are selected for amniocentesis based on complications of pregnancy or increased risk of fetal abnormality. Tongsong and colleagues conducted a controlled study to ascertain the true risk of fetal loss attributable to second-trimester amniocentesis.

More than 2,000 women who were at risk of having a child with chromosomal abnormalities but who were healthy, had confirmed gestational age, early antenatal care and no contraindication to amniocentesis were included in the study. The most common indication for amniocentesis was advanced maternal age (35 years or older), but 5 percent of the study participants had given birth previously to children with trisomy. Each study subject was matched with a control subject of the same age, parity and socioeconomic status who also met study entry criteria except for the risk of abnormality. The study subjects underwent amniocentesis between 15 and 24 weeks of gestation, and both study subjects and control subjects were monitored until delivery. All fetal losses were autopsied and tested for chromosomal abnormalities.

A total of 2,045 matched pairs completed the study. Pregnancy outcomes such as fetal loss, preterm rupture of the membranes, abruptio placentae, pregnancy-induced hypertension and cesarean delivery rates were comparable between the groups. The rates of spontaneous abortion before 28 weeks and of intrauterine fetal death were higher in the study group; the differences were not statistically significant and occurred in fewer than 2 percent of pregnancies.

The authors conclude that second-trimester amniocentesis is probably not associated with an increased rate of fetal loss in the absence of other conditions that could adversely influence pregnancy outcome.