Original Research
Obstetrics
Reexamining the optimal nuchal translucency cutoff for diagnostic testing in the cell-free DNA and microarray era: results from the Victorian Perinatal Record Linkage study

The findings of this study were presented in an e-poster format at the Human Genetics Society of Australasia Virtual Conference 2020, November 24–25, 2020.
https://doi.org/10.1016/j.ajog.2021.03.050Get rights and content

Background

The American College of Obstetricians and Gynecologists and the Society for Maternal-Fetal Medicine recently recommended offering genetic counseling and diagnostic testing for enlarged nuchal translucency at ≥3.0 mm, regardless of previous negative screening with noninvasive prenatal testing.

Objective

This study aimed to perform a population-based, individual record linkage study to determine the optimal definition of an enlarged nuchal translucency for the detection of atypical chromosome abnormalities.

Study Design

This was a retrospective study of women resident in Victoria, Australia, undergoing combined first-trimester screening during the 24-month period from January 2015 to December 2016. Linkages between statewide results for combined first-trimester screening, prenatal diagnostic procedures, and postnatal cytogenetic results from products of conception and infants up to 12 months of age were used to ascertain the frequency and type of chromosome abnormality by gestation and nuchal translucency measurement. An atypical chromosome abnormality was defined as any major chromosome abnormality other than whole chromosome aneuploidy involving chromosomes 21, 18, 13, X, and Y.

Results

Of the 81,244 singleton pregnancies undergoing combined first-trimester screening, 491 (0.60%) had a nuchal translucency of ≥3.5 mm, 534 (0.66%) had a nuchal translucency of 3.0 to 3.4 mm, and 80,219 (98.74%) had a nuchal translucency of < 3.0 mm. When grouped by nuchal translucency multiples of the median (MoM), 192 (0.24%) had a nuchal translucency of ≥3.0 MoM, 513 (0.63%) had a nuchal translucency of 1.9 to 2.9 MoM, and 80,539 (99.13%) had a nuchal translucency of <1.9 MoM. A total of 1779 pregnancies underwent prenatal or postnatal diagnostic testing, of which 89.60% were performed by whole-genome single-nucleotide polymorphism chromosomal microarray. The frequency of total major chromosome abnormalities was significantly higher in the group with a nuchal translucency of ≥3.5 mm (147 of 491, 29.94%) than the group with a nuchal translucency of 3.0 to 3.4 mm (21 of 534, 3.93%) or a nuchal translucency of <3.0 mm (71 of 80,219, 0.09%) (P<.001). There were 93 atypical chromosome abnormalities in the total screened cohort. The frequency of an atypical chromosome abnormality was 4.07% (95% confidence interval, 2.51–6.22), 0.37% (95% confidence interval, 0.05–1.35), and 0.09% (95% confidence interval, 0.07–0.11) in the groups with a nuchal translucency of ≥3.5 mm, 3.0 to 3.4 mm, and <3.0 mm, respectively. The frequency of atypical chromosome abnormalities was 4.69% (95% confidence interval, 2.17–8.71), 2.53% (95% confidence interval, 1.36–4.29), and 0.09% (95% confidence interval, 0.07–0.11) in the groups with a nuchal translucency of ≥3.0 MoM, 1.9 to 2.9 MoM, and <1.9 MoM, respectively. When defining thresholds for offering diagnosis with chromosomal microarray at 11 to 13 weeks, both a nuchal translucency threshold of 1.9 MoM and a fixed threshold of 3.0 mm captured 22 of 93 fetuses (23.7%) with an atypical chromosome abnormality. Of these, 50.0% had a coexisting fetal abnormality on ultrasound. However, the gestation-specific threshold of 1.9 MoM had a better specificity than 3.0 mm. The positive predictive value of an enlarged nuchal translucency for any atypical chromosome abnormality was 1 in 47 for nuchal translucency of >3.0 mm and 1 in 32 for nuchal translucency of >1.9 MoM. Our nuchal translucency threshold of 1.9 MoM captured 0.87% of fetuses, thus approximating the 99th centile.

Conclusion

A gestational age–adjusted nuchal translucency threshold of 1.9 MoM or 99th centile is superior to the fixed cutoff of 3.0 mm for the identification of atypical chromosome abnormalities. The risk of an atypical chromosome abnormality in a fetus with an enlarged nuchal translucency is more than tripled in the presence of an additional ultrasound abnormality.

Introduction

In the era of noninvasive prenatal testing (NIPT) with maternal plasma cell-free DNA, the combined first-trimester screening (CFTS) based on the nuchal translucency (NT) measurement and serum markers is waning in importance as a primary aneuploidy screening test, with many countries reporting NIPT uptake rates of 20% or higher.1,2 Despite the superior performance of NIPT for the detection of the common autosomal aneuploidies (trisomies 21, 18, and 13), professional societies still recommend the retention of the 11- to 13-week NT ultrasound. The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) recommends that a woman with a negative NIPT test result should still have fetal NT thickness measured and reported but without computing first-trimester risk estimates for trisomies 21, 18, and 13.3 However, no specific ISUOG guidance is given on the management of increased NT with a previous negative NIPT result; members are simply directed to “local guidelines.” Nevertheless, an NT of ≥3.5 mm is well accepted as a strong predictor of aneuploidy or other fetal abnormalities, and diagnostic testing with chromosomal microarray (CMA) is routinely offered in this situation, with an incremental yield of 4% to 7% over G-banded karyotype.4

AJOG at a Glance

The nuchal translucency (NT) measurement continues to have importance in the era of noninvasive prenatal testing (NIPT) and chromosomal microarrays (CMAs) for the detection of atypical chromosome abnormalities. The use of a fixed measurement of 3.0 mm to define an enlarged NT rather than a gestational age–adjusted metric is poorly substantiated in the literature. We used a large population-based cohort of women undergoing first-trimester screening to determine the optimal definition of an enlarged NT for detecting atypical chromosome abnormalities.

A gestational age–adjusted NT threshold of 1.9 multiple of the median (MoM) is superior to a threshold of 3.0 mm for the detection of atypical chromosome abnormalities, providing a similar sensitivity rate for 31% fewer “screen positives.” In the era of NIPT and CMAs, NT of >1.9 MoM or NT at the >99th centile seems to be a reasonable NT threshold at which to offer diagnostic testing (1 in 270 risk of an atypical chromosome abnormality). The risk of an atypical chromosome abnormality in a fetus with an enlarged NT was 11% in the presence of a coexisting fetal abnormality compared with 3% for fetuses with isolated enlarged NT.

We examined the diagnostic yield of various NT thresholds using high quality data from a large population-based cohort, overcoming the limitations of the previous studies such as ascertainment bias, small sample size, and low microarray utilization rates. We showed that a gestation-specific NT threshold such as 1.9 MoM or 99th centile is superior to a fixed 3.0 mm millimeter cutoff for offering microarray testing for atypical chromosome abnormalities. The positive predictive value (PPV) of an enlarged NT for any atypical chromosome abnormality is 1 in 32 for an NT of ≥1.9 MoM compared with 1 in 47 for an NT of ≥3.0 mm. The PPV rises to 1 in 9 in the presence of a fetal structural abnormality. A detailed first-trimester ultrasound including NT measurement continues to have importance in the NIPT era.

In the current era of NIPT and CMA, there is little contemporary evidence to define the optimal NT cutoff to recommend invasive prenatal diagnosis. Some recent reports have suggested that the NT threshold should be lowered from a fixed cutoff of 3.5 mm to 3.0 mm.5, 6, 7, 8 The American College of Obstetricians and Gynecologists and the Society for Maternal-Fetal Medicine (SMFM) recently recommended offering genetic counseling and diagnostic testing for enlarged NT at ≥3.0 mm or above the 99th centile.9 Others have argued that pathogenic copy number variants (CNVs) are rare in fetuses with an NT of <3.5 mm in the absence of additional structural abnormalities10,11 and not more frequent than the 1 in 250 to 270 risk estimated for fetuses without structural abnormalities.12,13 Most of the published literature are limited by small sample sizes, selective use of CMA, exclusion of pregnancies that did not undergo prenatal diagnosis, and single-center design. Furthermore, the existing data are limited to prenatal diagnosis cohorts and have not included postnatal samples from live infants, miscarriages, or stillbirths.

Another question that has not been examined in the CMA era is whether NT metrics that account for gestation are superior to fixed millimeter cutoffs for defining an enlarged NT. Previous studies investigating the use of multiples of the median (MoMs) and centile cutoffs14,15 did not find a difference in diagnostic yield compared with millimeter cutoffs, but these studies predated the use of CMA so the benefit for the detection of a full range of atypical abnormalities, including submicroscopic CNVs, is unknown.

Given the limitations of existing literature, we performed a population-based study timed during a period of high uptake of CFTS and CMA in our state. We performed individual record linkage of women undergoing CFTS and prenatal or postnatal diagnosis to calculate the rate of typical and atypical chromosome abnormalities by NT millimeter and MoM categories. In doing so, we aimed to derive an optimal definition of an enlarged NT for the detection of atypical chromosome abnormalities.

Section snippets

Population characteristics and data sources

This was a retrospective study of women resident in Victoria, Australia, undergoing CFTS in a singleton pregnancy during the 24-month period from January 1, 2015, to December 31, 2016. This study period was selected because it was the last 2 years of a single central provider of serum screening in our state. It also postdates the widespread adoption of CMA as the standard method of prenatal diagnosis analysis since 2012. Victoria has approximately 79,000 births per year and a median maternal

Results

During the 24-month study period, there were 81,379 CFTS results reported for women resident in Victoria, representing 52.98% of the total 153,601 registered births. The median maternal age of women having CFTS was 31 years (range, 14–53), with 77.71% aged <35 years. We excluded 135 pregnancies owing to multifetal gestation. Of the 81,244 singleton pregnancies with CFTS, 491 (0.60%) had an NT of ≥3.5 mm, 534 (0.66%) had an NT of 3.0 to 3.4 mm, and 80,219 (98.74%) had an NT of <3.0 mm (Figure 2).

Principal findings

If the definition of enlarged NT is to be revised to a gestational age–adjusted measurement, our results show that a threshold using a MoM threshold of 1.9 potentially results in better detection of atypical chromosome abnormalities than a fixed NT threshold of ≥3.0 mm, capturing the same number of atypical abnormalities for 31% fewer “screen positives.” Overall, there were 6 fewer common autosomal aneuploidies in the group with an NT of >1.9 MoM than those with an NT of >3.0 mm. However, these

Acknowledgments

We acknowledge the members of the Victorian Perinatal Record Linkage (PeRL) Collaboration and the support of the following organizations for their participation in the PeRL study: the Victorian Clinical Genetics Services (VCGS), Monash Pathology, Australian Clinical Labs, and Melbourne Pathology contributed the prenatal and postnatal diagnosis data. The serum screening laboratory at VCGS provided the NT and CFTS data. We thank the Victorian Infant Hearing Screening Program for assisting with

References (23)

  • L. Su et al.

    Clinical application of chromosomal microarray analysis in fetuses with increased nuchal translucency and normal karyotype

    Mol Genet Genomic Med

    (2019)
  • Cited by (15)

    • Increased nuchal translucency can be ascertained using transverse planes

      2022, American Journal of Obstetrics and Gynecology
      Citation Excerpt :

      There is a consensus that the detection of an NT >3.5 mm should prompt the offer of an invasive procedure to determine karyotype with chromosomal microarray because, in these cases, there is an increased risk of anomalies that would be otherwise undetectable.25 Moreover, some investigators have proposed that this should be considered in fetuses with an NT between 3.0 and 3.4 mm.26–28 In brief, accurate measurement is required for a combined test; however, in pregnancies undergoing NIPT, the relevant clinical information is whether the measurement is >3.5 or 3.0 mm, depending on the threshold that is chosen.

    View all citing articles on Scopus

    L.H. was funded by a National Health and Medical Research Council Early Career Fellowship (1105603, 2016–2019), a University of Melbourne Research Fellowship from the Faculty of Medicine, Dentistry and Health Sciences (2020), and a Medical Research Future Fund Emerging Leader Fellowship (2021–2025). J.H. was funded by a National Health and Medical Research Council Senior Research Fellowship (10121252). The funding bodies had no role in the conduct of the research or the writing of the manuscript.

    R.P.D. reports a commercial relationship with Roche Diagnostics and Natera and personal fees from Philips Ultrasound, outside the submitted work. D.N. reports a commercial relationship with Roche Diagnostics and Natera, outside the submitted work. The remaining authors report no conflict of interest.

    Cite this article as: Hui L, Pynaker C, Bonacquisto L, et al. Reexamining the optimal nuchal translucency cutoff for diagnostic testing in the cell-free DNA and microarray era: results from the Victorian Perinatal Record Linkage study. Am J Obstet Gynecol 2021;225:527.e1-12.

    View full text