Is my baby healthy? This is easily the number one question expecting moms want to know.
So when it comes to finding out if your unborn baby has Down Syndrome or another common chromosomal abnormality, which would you prefer: A faster, more accurate test or a slower, less accurate test?
Neither test increases the risk of miscarriage.
The answer seems pretty clear: More accurate is better.
I’m talking, of course, about Noninvasive Prenatal Testing (NIPT), which first became available to pregnant women in 2012.
So why are so many pregnant women not given the option of noninvasive prenatal testing (NIPT)?
What is NIPT?
NIPT relies on fragments of cell-free fetal DNA* that circulate in a mother’s blood to screen for the three most common trisomies (trisomy 21, commonly known as Down Syndrome, and trisomies 13 and 18) and common sex chromosome abnormalities.
NIPT is convenient – as easy as a simple blood draw – and can be done as early as 10 weeks. Plus, they can tell you your baby’s sex in the first trimester.
NIPT’s big selling point is accuracy. NIPT has a much higher detection rate than the older traditional screens, while being far less likely to raise a false alarm (a false positive).
Traditional screening tests, which rely on indirect signs of a genetic abnormality–hormonal markers and ultrasound measurements–cannot be performed until at least 11-13 weeks. The most accurate of these screens, the combined screen, is slow–taking until 15-17 weeks for a final risk estimate.
So, although the range of options and statistics can seem overwhelming at first, to my mind the choice is fairly simple: Are you comfortable with a very small increase of miscarriage**? If yes, then you should get amniocentesis or chorionic villus sampling. If not, skip the traditional screens in favor of NIPT.
This view differs from the one currently taken by the American College of Obstetricians and Gynecologists (ACOG). ACOG suggests that NIPT be offered to all pregnant women. However, they only recommend NIPT for “high risk” women.
“High risk” = women who have a family history of genetic disorders, women who are carrying twins or higher order multiples, women who have previously had a baby with a chromosomal abnormality, or who are over 35 –the cutoff for “advanced maternal age”***
Initially this recommendation made sense. Clinical studies had only assessed the performance of NIPT in high risk women.
Should NIPT be restricted to high risk women?
Probably not. The evidence is clear: several large studies (including one in the New England Journal of Medicine and another in PLOS) and a recent meta-analysis have now shown these tests are more accurate than the traditional screens, for all women, high or low risk.
A recent study, led by Mary Norton of UCSF, for example, looked at the performance of NIPT in over 15,000 women, most of whom were considered low risk.
- Of these 15,000 women, 38 were carrying a fetus with Down Syndrome. NIPT detected all 38 cases. The traditional first trimester screen missed 8.
- NIPT falsely identified 9 women as likely carrying a fetus with Down Syndrome, whereas the traditional first trimester screen falsely identified 854 women–almost 100 times as many!
- The results were similar for the other two trisomies: trisomy 13 and trisomy 18. While NIPT is not as good at detecting these two trisomies as they are at detecting Down Syndrome, it is still miles better than the traditional first trimester screen.
An obvious choice. Right?
Why isn’t NIPT recommended for all women?
Well, mainly because of cost. NIPT is more expensive than the traditional screens, so some insurers refuse to cover it for low risk women.
Some doctors may also shy away from recommending NIPT from fear that women will terminate healthy pregnancies after a false positive. Perhaps this fear comes from bad press coverage: several news articles have criticized NIPT (like this one), and cited cases of this exact scenario.
The companies often undeservedly get the blame in these cases, despite having been upfront and above board about NIPT’s accuracy.
Instead what probably went awry is that some doctors confused NIPT’s detection rate with the likelihood that a positive result on an NIPT is a true positive (in medical-speak, its positive predictive value).
NIPT can detect over 99% of fetuses with Down’s Syndrome, and over 91% of trisomies 18 and 13. But a positive result on an NIPT does NOT mean that you have a 99% chance of having a baby with Down’s Syndrome.
This is because the positive predictive value (the chance that a positive result is a true positive) depends not only on the test’s detection rate, but also on the test’s false positive rate and the base rate in the population.
A careful look at Norton’s study of NIPT in low risk women shows why. NIPT detected over 99% of fetuses with Down’s Syndrome (38 out of 38), but also gave 9 false positives.
If you do the math, that means that a positive test was a true positive only about 80% of the time. Far from perfect. But again better than the first trimester screen, where a positive was a true positive less than 5% of the time!
For uncommon conditions, even very accurate tests are wrong a lot of the time.
NIPT is not perfect. Of the prenatal testing options, only amniocentesis and chorionic villus sampling (CVS) are considered diagnostic.**** (And that is why further testing is often recommended after positive results.)
But without question, NIPT is better–by any reasonable standard–than traditional screening tests.
*[Most of these DNA fragments come from the placenta, not the fetus. Usually this does not matter, because he vast majority of the time, the placenta is genetically identical to the fetus. But in rare cases, the placenta contains a mix of chromosomally abnormal and chromosomally normal cells, even though the fetus is completely normal (a condition known as confined placental mosaicism.) Confined placental mosaicism can lead to inaccurate cell free DNA results. Chorionic villus sampling, which similarly relies on cells from the placenta, not the fetus, suffers from the same problem. A mosaic result on CVS should always be confirmed by amniocentesis, because about 80% of the time, the mosaicism is confined to the placenta, and the fetus is perfectly normal.]
**[The risk of miscarriage from invasive tests is often stated as about 1 in 100. But this estimate comes from older data, before doctors routinely used ultrasounds to guide these tests. The current risk of miscarriage from these tests is closer to 1 in 500 for chorionic villus sampling and 1 in a 1000 for amniocentesis, according to recent meta-analysis.]
***[This notion of “advanced maternal age” is a pet peeve of mine. So I’d like to pause for a moment to recall that there is nothing biologically significant about age 35. It’s an arbitrary cutoff point. It’s not like everything suddenly goes south egg-wise when you hit 35. Instead, the risk of carrying a fetus with a chromosomal abnormality rises throughout your 30s and 40s. In fact, because younger women have more babies than older women and are less likely to get tested for genetic disorders, the majority of children with Down’s Syndrome children in the U.S. are born to women under 35.]
****[Because Chorionic villus sampling relies on placental cells rather fetal cells, it is not 100% accurate. In rare cases, the placenta is genetically abnormal while the fetus is normal.]
Amy Kiefer holds a Ph.D. in Psychology and an M.A. in Statistics. She lives in the Bay Area with her husband and two children where she writes about fertility, pregnancy, and breastfeeding on her blog Expecting Science. Follow her @xpectingscience on Twitter or like Expecting Science on Facebook for more evidence-based parenting info.
(cover image via Visual hunt)