Blood test or ultrasound: which gender result do you trust?
By the mybaby4D teamUpdated 14 min read
The short answer
Early in pregnancy, the blood test. Cell-free DNA screening (NIPT), ordered by your doctor or midwife, can report the sex from about 10 weeks, and the Australasian Sonographers Association calls it more accurate for the sex than a first-trimester scan. After 14 weeks, studies of medical scans find scan calls highly accurate too, though no test or scan is 100% sure. Mismatches are uncommon, but if yours disagree, trust neither yet: only your doctor or midwife can look into why.
Below: how each works, how sure each is, and what to do if they clash.
How NIPT and ultrasound each read the sex
A blood test and a scan answer the same question from opposite ends: one reads DNA, the other reads the baby’s body. Either way, what they detect is the baby’s sex.
The blood test reads DNA
Cell-free DNA screening, often called NIPT (noninvasive prenatal testing), is a blood test your doctor or midwife orders. It looks mainly for chromosomal conditions such as Down syndrome (MedlinePlus), and some labs also report the sex as a side result (Cleveland Clinic).
In pregnancy, a mother’s blood holds a mix of free-floating DNA, some from her own cells and some from the placenta. The placenta’s DNA is usually identical to the baby’s, MedlinePlus explains. The lab searches this mix for Y-chromosome DNA, which shouldn’t otherwise be there. Any it finds points to a boy. If it finds none, the baby “is presumed to be female,” a 2012 review explains (Wright and colleagues, BMC Research Notes).
The test needs enough DNA from the placenta, a share called the fetal fraction. MedlinePlus says it generally needs to be above 4%, which typically happens around the 10th week. Results can sometimes take up to two weeks (Cleveland Clinic). England’s NHS screening uses a version that “does not screen for other conditions or fetal sex” (NHS England). At-home kits are a separate product: see at-home gender blood tests.
The scan reads anatomy
An ultrasound reads the baby’s outside genitals. In the late first and early second trimester, the call depends mainly on the direction of a small bud between the legs, the genital tubercle, a 2009 review explains. Down suggests a girl, and up a boy. Later in the second trimester, the sonographer looks at the genitals directly (Odeh and colleagues, Obstetrical & Gynecological Survey).
How sure is the blood test?
Very sure, but not certain. “The result is highly accurate, but not 100% accurate,” Cleveland Clinic says of the sex reading.
A 2017 review by researchers in Birmingham, England, pooled 60 studies of cell-free DNA testing for sex, covering 11,179 tests in pregnancies with one baby, done at any week. Their search covered research from 1997 to April 2015. Pooled, the tests got each sex right about 99 times in 100. The review’s two measures, sensitivity and specificity, were 98.9% and 99.6% (Mackie and colleagues, BJOG).
The review has limits. One of its two measures varied with the lab method: 93.9% with conventional PCR, against 99.9% with real-time PCR. Only 11 of the 60 studies reported inconclusive results, and the authors called false and inconclusive results “poorly reported across all conditions.”
It’s less reliable with twins: MedlinePlus says cell-free DNA screening is “not as accurate if you are pregnant with more than one baby” (MedlinePlus).
How sure is the scan?
It depends on the week, and no scan is certain. In the NHS’s words, “it’s not possible for the sonographer to be 100% certain about your baby’s sex” (NHS). Here are two studies of medical scans:
11 to 14 weeks, Australia. At a specialist private obstetric practice in the Illawarra, New South Wales, 7 sonographers predicted the sex of 640 babies across all three trimesters. At 11 to 14 weeks, judging from a side view, 75% of babies got a correct call, counting those left without one; 91% of the calls made were right. Every call made after 14 weeks, into the third trimester, was right. The abstract gives no group sizes (Kearin and colleagues, 2014).
18 to 21 weeks, Denmark. At anomaly scans of 5,786 single pregnancies at two hospitals, with live births in 2011, the sex was called at 86.7% of scans. Of those calls, 99.2% were right, which “still leaves 8/1,000” babies with the wrong sex (Størup and colleagues, 2019, corrected 2020).
The blood test. Scan calls this early are “relatively inaccurate,” says the 2009 review, which advises against trying before 12 weeks.
From 12 to 14 weeks
Still the blood test. NIPT has “a greater accuracy for sex identification” than a first-trimester scan, says the Australasian Sonographers Association (ASA, 2022).
Later, with a clear view
Both did well in the studies above. Neither is 100% sure.
When they disagree
Neither, yet. Tell your doctor or midwife, and hold off on anything permanent until they’ve looked into it.
When a lab blood test and a later, clear scan agree, two different methods point the same way. For every way to find out, week by week, see When can you find out your baby’s gender?
Why they rarely disagree
Both tests trace back to conception. A baby gets an X chromosome from the mother and an X or a Y from the father, and “these X and Y chromosomes determine the child’s genetic sex.” The genitals then grow from tissue that starts out the same in boys and girls (MedlinePlus). The blood test reads the chromosomes, and the scan reads the body that develops.
In a few babies, the two don’t line up, because of a difference in sex development (DSD). The NHS calls DSD “a group of rare conditions involving genes, hormones and reproductive organs” (NHS).
In the studies above, the blood test and later scan calls were each right about 99 times in 100 or better, so a clash is the exception.
What can make them disagree
There are several possible explanations, a few of them medical, and only your doctor or midwife can find out which applies to you.
A label, a lab or a scan
Human error is one. A 2021 case report, mostly by researchers at the genetics lab Igenomix in Spain and the UK, names three kinds. They are “blood sample mislabelling, laboratory methodologic limitations,” and a poor view of the genitals on early scans (Balaguer and colleagues, Molecular Cytogenetics). In the Australian study above, early scan calls were the least reliable.
Not enough of the placenta’s DNA
A girl is read from the absence of Y-chromosome DNA, so the lab needs enough of the baby’s DNA to be sure none is there. A low share “can lead to an inability to perform the test or a false negative result,” MedlinePlus says. Reasons include testing too early, sampling errors and obesity in the mother. In the Birmingham review, a low share was among the reasons study authors gave most often for wrong sex results. Many labs don’t report results below 4%, doctors wrote in Contemporary OB/GYN in 2020 (Lassey and colleagues). If there isn’t enough, Cleveland Clinic says, “You can have the NIPT screening again.”
A twin who stopped growing
Sometimes a pregnancy starts as twins and one stops growing early: a vanishing twin. ISUOG, the International Society of Ultrasound in Obstetrics and Gynecology, warns that this “can render unreliable the results of NIPT.” The lost twin can keep releasing DNA into the mother’s blood, it explains (ISUOG, 2025). That DNA “can persist in the maternal circulation for a period of up to eight to 15 weeks,” a Quest Diagnostics team wrote in a 2020 case report (CAP Today).
If the lost twin was a boy, the test can report a boy while the baby who’s growing is a girl. In a case reported in 2016 from Penn State Milton S. Hershey Medical Center, the test showed XY, and a girl was born (Kelley and colleagues, JABFM). NHS guidance in England lists “a vanished twin pregnancy” among the reasons its test isn’t suitable. Can you see twins at 6 weeks? covers the vanishing twin in more detail.
Your own DNA, or a donor’s
The test reads your DNA along with the placenta’s, so it “can be affected by a variety of maternal conditions,” the Quest team notes. In England, NHS guidance says its version of the test isn’t suitable after a bone marrow or organ transplant or stem cell therapy, or within 4 months of a blood transfusion. Nor is it suitable for women who currently have cancer, unless it’s in remission.
A transplant from a male donor is a known cause of a mismatch. In the Igenomix case, a woman whose bone marrow came from her brother had so much Y-chromosome DNA in her blood that the test couldn’t report the sex. She had a girl. If any of this applies to you, tell your doctor or midwife before the test.
One cause the Birmingham review rules out: an earlier pregnancy with a boy. Some studies blamed wrong results on it, but the review says that reason “has since been disproven,” because the placenta’s free-floating DNA is cleared from the mother’s blood within hours of birth.
A difference in sex development
Sometimes both tests read their part correctly, but the baby’s body has developed differently from the usual pattern for its chromosomes. Only medical specialists can assess this. After birth, a team “with expertise in DSD” cares for the child and supports the family, MedlinePlus says (MedlinePlus).
What one lab’s reported mismatches came down to
A 2020 study from Natera, a U.S. testing company, covered 1,301,117 of its lab’s tests of one type (SNP-based) whose results showed the usual XX or XY. For 91, about 1 in 14,300, patients or providers reported a mismatch with a scan or other clinical findings (Dhamankar and colleagues, Obstetrics & Gynecology).
Outcome information was available for 83 of the 91. Human error explained 28, mainly a blood tube labeled wrongly when it was drawn, or a scan that misread the sex. Thirty turned out to be a difference in sex development, established by chromosome tests and clinical information. For the other 25, no diagnosis was possible.
Two limits matter. The lab could count only the mismatches it was told about, so the study can’t say how often results really differ. And figures from a group of reported cases can’t tell you what’s behind any one result, including yours. That’s for your doctor or midwife to look into.
If your results don’t match: what to do
Tell your doctor or midwife first. Before you buy, paint or announce anything. There are several possible explanations, a few of them medical, and only they can look into it.
Share the details. Bring the blood test report, and note the week of the blood draw and of the scan. Doctors writing in Contemporary OB/GYN say the result should be checked “including the gestational age at which the test was performed.” Mention any transplant, recent transfusion or lost twin.
Let them check with the lab. “Discussion with the reporting laboratory” can help with a low fetal fraction, they add.
You may be offered a detailed medical scan and genetic counseling. The same doctors advise a scan “by an experienced imager.” A genetic counselor “can help you understand your options,” Cleveland Clinic says.
Ask before any further test. A diagnostic test such as amniocentesis may be offered for medical reasons. It isn’t done just to find out the sex. The NHS puts its risk of miscarriage at 1 in 200, and your doctor or midwife can help you decide whether to have it (NHS).
Don’t lean on a keepsake scan. It doesn’t check your baby’s anatomy, so it can’t settle a medical question.
For families who’ve been told, we offer a Gender Confirmation: a 2D scan, “the classic black-and-white picture,” with 8 black-and-white prints, at our studio in Glendale, Los Angeles County. We book from 13 weeks, and we’re surest from 15.
If what we see is different from what you’ve been told, we tell you what we see, and how sure we are. We don’t check or explain blood tests: what a result means is a question for your doctor or midwife. When we can’t be sure, we say so, and you come back for a free re-check (our promise). No ultrasound can promise 100%.
Our scan is a keepsake, not a medical exam: it can’t check your baby’s anatomy or health, and it never stands in for the anatomy scan or any appointment with your doctor or midwife. The FDA and the AIUM discourage ultrasound for keepsakes alone (FDA, AIUM); their statements are on Safety, in plain words. Details are on the Gender Confirmation page, with prices, or call (818) 243-7010.
4D Bonding16 to 36 weeks. Faces look best from 28 to 34.Book 4D Bonding
Your date stays on this device. We never see it.
Know your due date? Key it into the planner to see which of our sessions you could book this week and which one opens next. It stays on your device and goes nowhere else.
Questions
Can NIPT be wrong about the gender?
Yes, though it’s uncommon. NIPT (cell-free DNA screening) is a blood test your doctor or midwife orders to screen for chromosomal conditions, and Cleveland Clinic calls its sex result highly accurate, but not 100% accurate. Causes of a wrong result include a labeling mix-up, too little of the baby’s DNA, and a twin who stopped growing early. If it disagrees with a scan, tell your doctor or midwife.
Which is more accurate for gender, NIPT or ultrasound?
In the first trimester, NIPT: the Australasian Sonographers Association says it has greater accuracy for sex identification than a scan. Later, scan calls are highly accurate too: at Danish medical anomaly scans of 5,786 pregnancies at 18 to 21 weeks, 86.7% got a sex call, and 99.2% of those calls were right. Neither is 100% sure.
My NIPT said boy, but the ultrasound says girl. Which is right?
Don’t pick one yet: tell your doctor or midwife, because there are several possible explanations, a few of them medical. They range from a labeling mix-up or an early scan call to a twin who stopped growing or a transplant from a male donor. Your doctor or midwife can check with the lab and may arrange a detailed medical scan or genetic counseling. A keepsake scan can’t settle it.
Can a vanishing twin affect NIPT gender results?
Yes. A twin who stops growing early can keep releasing DNA into the mother’s blood, which ISUOG, an international ultrasound society, warns can make NIPT results unreliable. If that twin was a boy, the test can report a boy even when the baby who’s growing is a girl. Tell your doctor or midwife if you know or think you lost a twin.
Can a previous pregnancy with a boy affect NIPT?
Not according to a 2017 review: it says that idea has since been disproven, because the placenta’s free-floating DNA is cleared from the mother’s blood within hours of birth. If your NIPT and a scan disagree, tell your doctor or midwife, because there are other possible explanations.
Can a gender scan confirm my NIPT result?
A later scan with a clear view can show the sex, and when it matches the blood test, two different methods point the same way. No scan is 100% sure, and a keepsake scan doesn’t check your baby’s anatomy, so it can’t settle a medical question. If the two disagree, tell your doctor or midwife.
No. Tested on 2,840,755 Swedish births, the Chinese gender chart did no better than a coin toss. How it’s read, where it came from, and what works instead.
At-home kits look for a boy’s DNA in your blood. Makers claim 99% or more from 6 or 7 weeks (2026); a 2011 review found lab tests before 7 weeks unreliable.