Risk-free prenatal testing may increase risk of prenatal testing

SMFM logo imageIn 2014, I attended the Society for Maternal-Fetal Medicine (SMFM) annual conference. Turns out, research presented there showed that a prediction of mine is happening: non-invasive prenatal screening may increase the risk of diagnostic prenatal testing.

In 2013, I wrote about presenting a poster at the American College for Medical Genetics & Genomics (ACMG) annual conference. My poster was on the nonmaleficent justification for non-invasive prenatal screening. A quick summary:

  • Nonmaleficence embodies the Hippocratic Oath’s well-known statement “first, do no harm.”
  • Non-invasive prenatal screening (NIPS) poses no risk to the life of the mother or baby and has highly accurate screen results for mother’s already identified with an increased chance of having a child with Down syndrome.
  • Invasive testing, while the only way to know for certain if the child has Down syndrome, poses a risk of miscarriage.
  • Therefore, the principle of nonmaleficent would support offering NIPS to all women considering invasive testing, since it may reduce the total number exposed to the risk of miscarriage of invasive testing.
Fig. 1 Inverse relationship

Fig. 1 Inverse relationship

My poster also addressed the relationship between the number of invasive diagnostic tests performed and the risk of miscarriage. It should not be surprising that the more a practitioner performs a procedure, the better they get at it, and the risk of miscarriage decreases. Conversely, the less a Maternal-Fetal Medicine specialist performs invasive testing, the greater the chance of miscarriage. Therefore, if women who otherwise would undergo invasive testing instead choose to rely on the NIPS results and not proceed with an amnio or CVS, then the fewer of those procedures, and the greater the risk of miscarriage associated with amnio and CVS due to lack of practice.

This is what I mean by the headline: that risk-free prenatal testing, NIPS, may increase the risk of diagnostic prenatal testing. Put another way, while it is safer than ever to get a “maybe” with a NIPS result, that same testing technology may increase the risk of knowing for certain whether the child has Down syndrome.

Fig. 2 NIPT & Inverse relationship

Fig. 2 NIPT & Inverse relationship

My argument last year was based on the following logic: that if the risk of miscarriage is inversely related to the number of procedures performed, then a reduction in the number of procedures will increase the risk of invasive testing. With more than two years of NIPS being in clinical practice, real world experience is  bearing out my prediction.

I was at the SMFM conference exhibiting as part of the National Center for Prenatal & Postnatal Down Syndrome Resources along with the National Down Syndrome Adoption Network. It was a great example of collaboration between non-profits with aligned missions of improving the counseling expectant mothers receive about prenatal testing and Down syndrome. We enjoyed meeting with many attendees who stopped by our booth and spoke with us in between sessions.

Nearly every single practitioner who stopped by our booth said they were performing not just fewer invasive procedures at their practices, but far fewer procedures due to their patients relying on the NIPS results. I must say “nearly” because there was a single practitioner who said she had not seen a fall off in the number of procedures since NIPS was introduced–but this was because her practice had not been performing that many amnios and CVS’s before NIPS.

This anecdotal experience was bore out by the studies presented at the conference. In one study of a single center over nine-years, the researchers concluded that:

Our results suggest that he introduction of [first trimester screening] and NIPT screening resulted in a significant decrease in amniocentesis testing at our institution.

Another study done by the University of California, San Francisco, had the same experience after NIPS was introduced:

The rate of invasive procedures significantly declined following the introduction of NIPT. Women have more options, as significantly fewer women declined any further testing following a positive [first trimester combined screening] with the introduction of NIPT. However, this may impact CVS and amniocentesis training for future MFM fellows.

(emphasis added). And, that last line was corroborated by another study presented at the conference.

A study was done to test the hypothesis of whether “operator related factors affect procedure related loss and complication rates for CVS.” The study’s results were:

When procedurists did 10 or less CVS per year, the [procedure related loss] and [total complications] were 6.9 and 9.2% respectively which is significantly more than when [more than 10 per year] are preformed ([procedure related loss] 1.5% [ total complications] 3.0%].

Did you catch that? When a physician is experienced with CVS, only about 15 pregnancies out of 1,000 will be lost due to the procedure. But, when a physician is less experienced, 69 pregnancies out of 1,000 will be lost–more than four times the number of pregnancies lost with an experienced physician. Moreover, whereas only 30 would have complications with an experienced physician, 92 would have complications with a less experienced physician–a tripling of complications due to lack of experience.

While these results are significant, they should not be surprising. One of the main advantages promoted by the labs that offer NIPS is that their more accurate screening results will cause fewer women to have invasive testing. Again, this is a good thing if it means fewer women will be exposed to the risk of miscarriage associated with invasive testing. However, for those women who want to confirm a NIPS result through invasive testing, they need to be informed of the actual risk of miscarriage associated with an amnio or CVS. Based on the experience shared by the practitioners who visited our booth at the SMFM conference and the research presented from multiple centers, that risk is very likely increasing as more women accept and rely upon their NIPS results.


  1. This principle of nonmaleficence, it seems to have a false assumption that it is ethical to reduce harm to one group whilst continuing, or increasing, harm to another. How is that resolved?

    • Inferring from your comment, Mike, that while NIPS poses no risk to the life of the mother or the child, directly, it more often than not leads to termination of pregnancy following a positive result, your question is how is it nonmaleficent to the child given that likely outcome? In and of itself, NIPS can be justified with the principle of nonmaleficence since it minimizes the number of pregnancies exposed to invasive testing and its attendant risk of loss. Whether the termination of pregnancy for fetal anomaly (TOPFA) can be ethically justified, well, that’s a different ethical analysis. But of those mothers who have terminated, a significant percentage say they are making the more humane choice of sparing their child from a lifetime of suffering, which nonmaleficence could be one of the principles cited for that justification.

      • Isn’t the starting point, this practice causes harm? The response then, is how to reduce harm. I would argue that the ethical response is to reduce harm to both groups without discrimination. Your logic doesn’t address the discrimination, there is a false assumption that reducing harm to one group whilst continuing (or increasing) harm to another is ethical. I would argue that is not ethical, it assumes reducing harm to one at the expense of the other is neutral. Termination is not the ethical issue, it’s about a discriminatory response to a harmful practice.

        • Well, for the mother’s who choose prenatal testing so they can know beforehand, prepare their loved ones, and arrange appropriate prenatal and postnatal care, their experience is that prenatal testing actually reduces the harm experienced by families who have a postnatal diagnosis. And, NIPS allows for them to know with greater accuracy and no risk to their child. An increasing number of women are choosing to continue their pregnancies and are finding the support they need earlier than ever. NIPS, in and of itself, is providing information. You are correct that there are ethical issues when it is administered in a discriminatory fashion based on the inadequate training, persistent societal stigma, no off-setting public awareness campaigns, and choosing to invest in prenatal testing rather than fully-fund social support services. I’ve written about that here and elsewhere. The ethical analysis for this particular post concerns the situation where there is prenatal testing, and since there is, what testing protocol does the least harm. Given the precipitous drop in pregnancies that otherwise would have miscarried from an invasive test, but were spared because the mother chose to stop and rely on the NIPS result, that is a nonmaleficent result.

  2. You sir, explained it well but I have doubt. Can array CGH find all chromosome differences?

    • Speaking from memory, and as someone who is not trained in genetics, my understanding of array CGH is that it identifies gene differences, because that is what it is looking at, not the individual chromosomes. So, to identify structural conditions like translocation, a karyotype would still be needed.


  1. […] screening laboratories’ representatives, some of their highest officers. And, as at the Society for Maternal Fetal Medicine, and the American Society for Human Genetics, my colleague Stephanie Meredith and I will be the […]

Leave a Reply

%d bloggers like this: