High-Risk NIPT Results Confirmed in Only Half of 1,326 Pregnancies, Study Finds

When a non-invasive prenatal test (NIPT) flags a pregnancy as high risk for a chromosomal abnormality, the result is a screening signal, not a diagnosis. A new retrospective study from Hebei General Hospital in Shijiazhuang, China, has now quantified just how often those alarming signals hold up under confirmatory testing. Among 1,326 pregnant women whose high-risk NIPT results were followed by invasive prenatal diagnosis between January 2014 and December 2023, only 648 cases were confirmed as true positives, giving an overall positive predictive value (PPV) of 48.87%. In other words, for roughly half of the women in this cohort, the high-risk result did not correspond to the chromosomal abnormality suggested by the screening test.

The study, published in Frontiers in Genetics, analyzed women who attended the hospital’s Department of Reproductive Genetics after receiving high-risk NIPT results. Of 1,364 singleton pregnancies with high-risk results during the ten-year window, 1,326 women, or 97.21%, underwent invasive prenatal diagnosis and formed the final cohort. Confirmatory testing relied on conventional G-banded chromosomal karyotyping, supplemented where necessary by fluorescence in situ hybridization (FISH) or copy number variation sequencing (CNV-seq). Thirty-eight women were excluded: fifteen declined invasive testing after counseling, fourteen suffered miscarriage or intrauterine fetal death before sampling, and nine were lost to follow-up or transferred elsewhere.

The performance of NIPT varied dramatically by the type of abnormality flagged. The test performed best for trisomy 21, the cause of Down syndrome, with a PPV of 78.61% (294 of 374 high-risk cases confirmed). Trisomy 18 followed at 56.36% (62 of 110). But the picture deteriorated sharply for rarer conditions: the PPV for trisomy 13 was just 10.53% (8 of 76), and for sex chromosome abnormalities it was 47.11% (212 of 450). Microdeletion and microduplication syndromes showed a PPV of 37.65%, rare autosomal trisomies only 7.41%, and cases involving two chromosomal abnormalities just 5.26%. These figures underline that a high-risk call for certain karyotypes is far more likely to be a false alarm than for others.

Maternal age emerged as a meaningful modifier of predictive value for the common autosomal trisomies. The researchers divided the cohort into women under 35 and women aged 35 or older at the expected date of delivery. For trisomy 21, the PPV reached 88.05% in the advanced maternal age group, significantly higher than the 71.63% observed in younger women. Trisomy 18 showed a similar pattern, with a PPV of 70.21% among older women versus 46.03% among younger ones. No statistically significant age-related differences were seen for trisomy 13, sex chromosome abnormalities, microdeletion or microduplication syndromes, or rare autosomal trisomies. Overall, the PPV across all result types was 60.66% in the older group compared with 44.69% in the younger group.

A notable feature of the study period was the introduction, in July 2019, of a government-funded free NIPT screening program in the province, which allowed the researchers to compare testing conditions before and after the policy change. After implementation, all samples were centrally tested by BGI using standardized shallow whole-genome sequencing on the MGISEQ-2000 platform, whereas the earlier period involved three different sequencing platforms. The PPVs for trisomies 21, 18, and 13 rose numerically after the program began, from 77.61%, 55.56%, and 8.70% to 81.13%, 57.89%, and 13.33%, respectively. However, none of these differences reached statistical significance, and the authors attribute this partly to small subgroup sizes, the expansion of testing to a broader and potentially lower-risk population, and biological variability across time periods.

The study also examined how the numerical confidence score underlying an NIPT result relates to the likelihood that the finding is real. NIPT assigns a Z-score to each chromosome, with results classified as high risk at an absolute value of 3 or greater. For trisomy 21, the PPV climbed steadily with increasing Z-score: it was just 25.00% in the borderline range of 3 to less than 4, rose to 85.19% at Z-scores of 6 or higher, and reached 100.00% at Z-scores of 13 or above. Trisomy 18 showed a similar upward trend. Trisomy 13, by contrast, displayed no clear linear association between Z-score and PPV, a pattern the authors connect to the low GC content of chromosome 13, which can introduce sequencing bias, and to the placental dysplasia often seen in trisomy 13 fetuses, which may reduce the amount of cell-free fetal DNA in maternal circulation.

Ultrasound findings proved to be a powerful complementary tool for refining risk. For high-risk trisomy 21 and trisomy 18 cases, the PPV reached 100.00% when three or more ultrasound soft markers or fetal structural anomalies were present. Even for trisomy 13, whose overall PPV was low, the predictive value rose from 2.04% in cases with a normal ultrasound to 50.00% when two soft markers were seen. Sex chromosome abnormalities did not follow a simple gradient with the number of soft markers, but their PPV remained high at 100.00% with three or more soft markers and 75.00% with structural anomalies. The authors argue that this imaging-plus-molecular-genetics strategy substantially improves the efficiency of detecting chromosomal abnormalities and can help clinicians decide which women need prompt invasive testing.

Among the 450 pregnancies flagged for sex chromosome abnormalities, a striking asymmetry emerged: the PPV for sex chromosome gains was 73.01%, significantly higher than the 20.98% for sex chromosome losses. The authors attribute the high false-positive rate for losses, such as the 45,X karyotype associated with Turner syndrome, to biological and technical factors including GC-content fluctuations on the X chromosome and maternal sex chromosome mosaicism. Among the 212 confirmed sex chromosome abnormality cases, the overall pregnancy termination rate was 76.89%, but it varied widely by karyotype: 91.36% for 47,XXY, 79.17% for 47,XYY, 67.31% for 45,X or mosaicism, and 51.61% for 47,XXX. The authors emphasize that outcomes for these conditions vary and that some features may be improved through clinical intervention, supporting a multidisciplinary counseling approach that respects informed parental choice.

The study’s authors are candid about its limitations. It was a single-center retrospective analysis, which may limit generalizability. Placental tissue verification was not available for false-positive cases, so the contribution of confined placental mosaicism, a major known cause of discordant NIPT results, could not be directly confirmed. Fetal fraction data were unavailable for the cohort, and low fetal fraction can alter predictive values, particularly for trisomy 13 and rare autosomal trisomies. The ten-year span also involved multiple laboratories, varied sequencing platforms, and iterative bioinformatic tools, and although all labs followed national NIPT standards, the authors note that minor vendor-specific technical bias cannot be fully excluded.

The broader message for clinical practice is that a high-risk NIPT result should never be equated with a definitive diagnosis. Cell-free fetal DNA is mainly derived from apoptotic placental trophoblasts, so results can be affected by confined placental mosaicism, maternal chromosomal abnormalities, vanishing twins, and other factors. The authors recommend that clinical interpretation integrate maternal age, Z-score, ultrasound phenotype, and confirmatory diagnostic findings to enable stratified genetic counseling. For women with extremely high Z-scores for trisomy 21 or 18, timely invasive diagnosis is warranted; for those with borderline scores or low-PPV conditions such as trisomy 13 and sex chromosome aneuploidies, counselors should explain the elevated false-positive risk clearly. The authors also note that the waiting period between screening and confirmatory amniocentesis, which is typically available from around week 16, commonly induces anxiety, and that sustained counseling during this interval can help mitigate distress before final results arrive.

Subject of Research: Positive predictive values of high-risk non-invasive prenatal testing results in 1,326 pregnancies undergoing confirmatory invasive prenatal diagnosis

Article Title: Prenatal diagnostic outcomes and positive predictive values in 1,326 pregnancies with high-risk non-invasive prenatal testing results

Article References: Du, R., Zhao, B., Shi, X., Gao, Y., Li, Y., & Qiu, B. (2026). Prenatal diagnostic outcomes and positive predictive values in 1,326 pregnancies with high-risk non-invasive prenatal testing results. Frontiers in Genetics, 17, 1899719. https://doi.org/10.3389/fgene.2026.1899719

Image Credits: AI Generated

DOI: 10.3389/fgene.2026.1899719

Keywords: NIPT, positive predictive value, prenatal diagnosis, trisomy 21, trisomy 18, trisomy 13, sex chromosome abnormalities, Z-score, ultrasound soft markers, genetic counseling, cell-free fetal DNA, chromosomal aneuploidy