What's the real miscarriage risk from amniocentesis?

Question

If you look for information about the miscarriage risk from amniocentesis, the risk of 1 in 200 or around 0.5 percent gets repeated again and again. Examples:

Mayo Clinic

WebMd

However, they never seem to cite a source. I am looking for a semi-current study (let's say from the last 15 years, to account for current physician training and equipment) reporting the risks of amniocentesis.

What I am looking for would also need to have a control group - one of my problems with the above mentioned sources is that while spontaneous second trimester miscarriages are a lot less likely than first trimester ones, they do occur, even without amniocentesis.

Answer 1

I did end up finding some recent studies with a bit more searching.

Miscarriage risk from amniocentesis performed for abnormal maternal serum screening looked at 30.000 women with abnormal serum levels and divided them into two equally sized groups, one undergoing amniocentesis and one not. They found no significant difference in risk of miscarriage, the risk being around 0.5 percent for both groups. According to their abstract, the study size would have been large enough to detect a jump from that 0.5 percent to 0.75 percent, so an increase by 0.25, or 1 in 400.

No statistically significant difference was also found in Loss Rates After Midtrimester Amniocentesis, with a study group of 3.000 women and a control group of 30.000.

Revisiting the Fetal Loss Rate After Second-Trimester Genetic Amniocentesis: A Single Center’s 16-Year Experience looked at a single medical center's results from 1990 to 2006. That included 12.000 women undergoing the procedure and 50.000 who didn't. The control group has a miscarriage rate of 0.26%, the amniocentesis group one of 0.4%, the difference was statistically significant. That's an increase in 0.13%, or 1 in 769.

At least from those studies, the risk of 1 in 200 seems overstated.