First trimester ultrasound screening for Down syndrome based on maternal age, fetal nuchal translucency and different combinations of the additional markers nasal bone, tricuspid and ductus venosus flow

Full text

ORIGINAL ARTICLE
First trimester ultrasound screening for Down syndrome based
on maternal age, fetal nuchal translucency and different
combinations of the additional markers nasal bone,
tricuspid and ductus venosus flow
Harald Abele
1
, Philipp Wagner
1
, Jiri Sonek
2,3
, Markus Hoopmann
1
, Sara Brucker
1
, Burcu Artunc-Ulkumen
4
and Karl Oliver Kagan
1
*
1
Department of Obstetrics and Gynaecology, University of Tuebingen, Tübingen, Germany
2
Fetal Medicine Foundation USA, Dayton, OH, USA
3
Division of Maternal Fetal Medicine, Wright State University, Dayton, OH, USA
4
Department of Obstetrics and Gynecology, Celal Bayar University, Turkey
*Correspondence to: Karl Oliver Kagan. E-mail: KOKagan@gmx.de
ABSTRACT
Objective To examine the performance of screening for Down syndrome based on maternal age, fetal nuchal
translucency (NT) and different combinations of the additional ultrasound parameters: nasal bone (NB), tricuspid
flow (TF) and ductus venosus (DV).
Methods Retrospective study at the University of Tuebingen, Germany including women who underwent chorionic
villous sampling between 2008 and 2014. Prior to invasive testing, the crown rump length, NT, NB, TF and DV were
measured. In each case, the added value of the additional markers NB, TF and DV were compared with screening
for trisomy 21 based on maternal age (MA) and NT thickness alone.
Results A total of 1916 pregnancies met the inclusion criteria, including 1823 fetuses with a normal karyotype and 93
with trisomy 21. Screening based on MA, fetal NT and one, two and three of the additional ultrasound markers
resulted in a detection rate of about 80%, 87% and 94%, respectively for a false positive rate of 3%.
Conclusion Detection rates for trisomy 21 in first trimester ultrasound screening are substantially higher if all three
additional markers rather than just one are assessed. © 2015 John Wiley & Sons, Ltd.
Funding sources: None
Conflicts of interest: None declared
INTRODUCTION
Although cell free fetal DNA (cffDNA) testing provides a
significant improvement in screening for trisomy 21, first
trimester combined screening based on maternal age (MA),
fetal nuchal translucency (NT) and the maternal serum
markers (free beta-hCG and PAPP-A) is still considered to be
the gold standard in most countries.
1,2
Moreover it plays an
important role for genetic screening in the presence of
multiple pregnancies. The detection rate for this traditional
approach is about 90% for a false positive rate of 3–5%.
3
The performance of first trimester combined screening can
be improved by using additional ultrasound markers. The
most promising ones are nasal bone (NB) and evaluation of
the tricuspid flow (TF) and the ductus venosus (DV) using
Doppler ultrasound. Each of these markers is abnormal
in 55% to 66% of the affected cases.
4–6
In a first trimester
screening study based on maternal age, fetal NT, maternal
serum biochemistry and one of these markers including
19 614 euploid pregnancies and 122 with trisomy 21, the
detection rates were 91–96% for a false positive rate of
about 2.5%.
4–7
As a matter of policy, the Fetal Medicine
Foundation (FMF) recommends the use of only one of
the additional markers in order to reduce inter-operator
variability and the time required for the examination.
8
Only few studies have examined the change in screening
performance of traditional combined screening if two or three
of these markers are added. These studies have indicated a
substantially increased detection rate of up to 100%.
9,10
Because the majority of affected fetuses have at least one
abnormal additional marker and because the prevalence of
these markers is low in the euploid population, the use of NT
and MA in combination with multiple additional ultrasound
markers may prove to be an appropriate alternative to classical
first trimester screening policies.
Prenatal Diagnosis 2015, 35,1–5© 2015 John Wiley & Sons, Ltd.
DOI: 10.1002/pd.4664

In this study we set out to examine the performance of
screening for trisomy 21 based on maternal age, fetal NT and
different combinations of the additional ultrasound parameters:
NB, TF and DV.
METHODS
This is a retrospective study involving patients that were seen at
the department of prenatal medicine at the University of
Tuebingen, Germany between 2007 and 2014. A search of our
database was performed to identify those pregnancies where a
CVS was carried out and where data on MA and weight, obstetric
history and all pertinent ultrasound parameters (crown-rump-
length [CRL], NT, NB, DV, and TR) were available.
At our department, a detailed ultrasound examination is
routinely carried out prior to chorionic villus sampling (CVS).
This includes CRL and NT measurements as well as a thorough
anatomical fetal assessment looking for the presence of major
anatomic defects.
11,12
In addition, the NB, TF and DV are
examined in each case.
The NB is considered to be abnormal if it is either not visible
or its echogenicity is the same or less than that of the skin.
Tricuspid regurgitation (abnormal TF) is diagnosed if it is
found during at least half of the systole and with a velocity of
more than 80 cm/s. The DV flow is assessed qualitatively, and
considered to be abnormal if the a-wave was reversed. All
operators are certified for the respective marker evaluations
by the FMF UK and measurements are performed according
to the published guidelines.
4–6
A CVS is offered to all women who consider their risk for trisomy
to be high enough to justify invasive testing.
13
The results of the
ultrasound examination, the karyotype and all relevant maternal
and pregnancy characteristics are entered into a digital database
(Viewpoint, GE Healthcare, Muenchen/Germany).
Pregnancies with a chromosomal abnormality other than
trisomy 21 were excluded from this study.
Because of the retrospective character of our study, an
approval of the local ethical committee was not necessary.
However, the ethical committee was informed about the study
(IEC number: 446/2015R).
Statistical analysis
In each case, the risk for trisomy 21 was calculated based on first,
MA, second, MA and NT thickness, third, MA, NT and one of the
additional markers NB, TF and DV, fourth two of these markers
and finally, all of these additional markers together.
The risks were computed based on the latest algorithms of
the FMF UK that incorporated DV flow as dichotomous
variable.
4–6,14–16
Detection and false-positive rates were
calculated by taking the proportions with risks above a given
risk threshold.
Figure 1 Distribution of fetal nuchal translucency measurements in
fetuses with trisomy 21
Table 1 Detection rate for fixed false positive rates in first trimester screening for trisomy 21 based on maternal age, fetal NT and
different combination combinations of additional sonographic markers
Fixed false positive rate, n(%)
1% 2% 3% 4% 5%
Screening policy
MA 3 (3.2) 3 (3.2) 4 (4.3) 6 (6.5) 8 (8.6)
MA +NT 36 (38.7) 52 (55.9) 63 (67.7) 67 (72.0) 67 (72.0)
MA +NT + NB 56 (60.2) 70 (75.3) 74 (79.6) 78 (83.9) 80 (86.0)
MA +NT + TF 61 (65.6) 70 (75.3) 74 (79.6) 76 (81.7) 79 (84.9)
MA +NT + DV 49 (52.7) 62 (66.7) 73 (78.5) 76 (81.7) 78 (83.9)
MA +NT + NB + DV 70 (75.3) 76 (81.7) 81 (87.1) 82 (88.2) 83 (89.2)
MA +NT + DV + TF 65 (69.9) 72 (77.4) 80 (86.0) 83 (89.2) 83 (89.2)
MA +NT + NB + TF 62 (66.7) 75 (80.6) 82 (88.2) 85 (91.4) 87 (93.5)
MA +NT + NB + TF + DV 69 (74.2) 82 (88.2) 87 (93.5) 88 (94.6) 89 (95.7)
MA, maternal age; NT, nuchal translucency; NB, nasal bone; TF, tricuspid flow; DV, ductus venosus flow.
H. Abele et al.
Prenatal Diagnosis 2015, 35,1–5 © 2015 John Wiley & Sons, Ltd.

RESULTS
The search of the database identified 1916 pregnancies that
met inclusion criteria. There were 1823 fetuses with a normal
karyotype and 93 fetuses with trisomy 21.
In 62.3% of the cases, invasive testing was carried out because
of an increased risk after first trimester risk assessment including
cell free fetal DNA testing and in 37.7% of the cases, karyotyping
was performed because of increased anxiety, respectively.
Median MA and weight was 34.9 (25
th
–75
th
IQR 30.9–38.0)
years and 65.2 (IQR 59.0–73.8) kg, respectively. Median CRL
was 66.7 (IQR 61.4–72.3) mm. Most patients were of Caucasian
origin (1880 [98.1%]).
The NT measurement was above the 99
th
percentile in 77
(4.4%) of euploid fetuses and in 66 (71%) of the fetuses with
trisomy 21. Figure 1 illustrates the NT distribution in the
aneuploid population.
In the normal population, the NB, TF and DV were abnormal
in 36 (2.0%), 31 (1.7%) and 64 (3.5%) cases, respectively. In
the trisomic population, these findings were abnormal in 57
(61.3%), 57 (61.3%) and 56 (60.2%) cases, respectively. In 108
(5.9%) of the normal and 89 (95.7%) of the trisomic fetuses, at
least one of the markers was abnormal.
Different screening policies and the respective detection
rates are shown in Table 1. Screening based on MA and fetal
NT resulted in a detection rate of 67.7% for a false positive rate
of 3%. By adding one of the ultrasound markers NB, TF or DV,
the detection rate increased to 78.5–79.6%. The examination of
two of these markers resulted in detection rates of 86.0–88.2%,
and the assessment of the NB, TF and DV in addition to the NT
and MA detected 93.5% of fetuses with trisomy 21 for the same
false positive rate.
In Table 2 detection and false positive rates are shown for
fixed risk cut-offs. For a risk cut-off of 1:50, detection and false
positive rates are 95.7% and 4.3% if all markers are assessed.
By assessing only one or two of the markers NB, TF or DV
in addition to MA and NT resulted in a substantially lower
screening performance.
DISCUSSION
Main findings of our study
In this study we have examined the detection rates for trisomy 21
in screening by MA, fetal NT and different combinations of
additional ultrasound markers. We have shown that the highest
detection rate can be achieved if in addition to MA and fetal NT,
all three other sonographic markers NB, TF and DV are assessed.
If only two or one of the additional sonographic markers is
examined, the detection rate is about 10 and 15% lower.
Comparison with previous studies
Previous studies have predominantly used first trimester
combined screening based on MA, fetal NT and maternal
serum biochemistry as gold standard with the addition of only
one of new markers, i.e. NB, TF and DV.
In a study performed by the FMF UK, 19614 euploid
pregnancies and 122 with trisomy 21 were examined. The
Table 2 Detection and false positive rates for fixed risk cut-offs in first trimester screening for trisomy 21 based on maternal age, fetal
NT and different combinations of additional sonographic markers
Fixed risk cut-offs, n(%)
Karyotype 1 in 10 1 in 50 1 in 100 1 in 1000 1 in 2500
Screening policy
MA Normal 5 (0.3) 74 (4.1) 272 (14.9) 1698 (93.1) 1823 (100.0)
T21 0 (0) 6 (6.5) 27 (29.0) 87 (93.5) 93 (100.0)
MA +NT Normal 58 (3.2) 159 (8.7) 214 (11.7) 975 (53.5) 1460 (80.1)
T21 63 (67.7) 80 (86.0) 86 (92.5) 93 (100.0) 93 (100.0)
MA +NT + NB Normal 48 (2.6) 114 (6.3) 160 (8.8) 595 (32.6) 1016 (55.7)
T21 72 (77.4) 82 (88.2) 85 (91.4) 93 (100.0) 93 (100.0)
MA +NT + TF Normal 40 (2.2) 119 (6.5) 159 (8.7) 578 (31.7) 1009 (55.3)
T21 72 (77.4) 79 (84.9) 86 (92.5) 92 (98.9) 93 (100.0)
MA +NT + DV Normal 47 (2.6) 128 (7.0) 178 (9.8) 652 (35.8) 1094 (60.0)
T21 67 (72.0) 80 (86.0) 83 (89.2) 93 (100.0) 93 (100.0)
MA +NT + NB + DV Normal 49 (2.7) 92 (5.0) 127 (7.0) 384 (21.1) 683 (37.5)
T21 80 (86.0) 83 (89.2) 85 (91.4) 93 (100.0) 93 (100.0)
MA +NT + DV + TF Normal 47 (2.6) 89 (4.9) 129 (7.1) 359 (19.7) 674 (37.0)
T21 78 (83.9) 83 (89.2) 85 (91.4) 93 (100.0) 93 (100.0)
MA +NT + NB + TF Normal 45 (2.5) 83 (4.6) 117 (6.4) 331 (18.2) 613 (33.6)
T21 80 (86.0) 86 (92.5) 87 (93.5) 93 (100.0) 93 (100.0)
MA +NT + NB + TF + DV Normal 41 (2.2) 79 (4.3) 100 (5.5) 239 (13.1) 398 (21.8)
T21 84 (90.3) 89 (95.7) 89 (95.7) 90 (96.8) 93 (100.0)
MA, maternal age; NT, nuchal translucency; NB, nasal bone; TF, tricuspid flow; DV, ductus venosus flow.
First trimester ultrasound screening
Prenatal Diagnosis 2015, 35,1–5 © 2015 John Wiley & Sons, Ltd.

detection and false positive rates using first trimester combined
screening alone was 90% and 5% respectively. These improved
to 90–95% and 2.5% respectively by including a single additional
marker.
4–6
In another more recent study by the FMF UK, the DV
flow was measured as a continuous instead of a dichotomous
variable. The addition of DV pulsatility index for veins to the
combined test resulted in a detection rate of 94% and a false
positive rate of 1.6%.
17
Only few studies have examined the screening performance of
combined screening with more than one additional marker.
Ghaffari et al. examined the additional markers in about 13 500
pregnancies including 33 cases with trisomy 21. Screening based
on combined screening resulted in a detection rate of 94% for a
false positive rate of 4.8%. The addition of the NB, TF and DV
increased the detection rate to 100% for a false positive rate of
3.4%.
9
Karadzov-Orlic et al. examined about 4000 euploid fetuses
and 97 fetuses with trisomy 21. They reported a 93% detection
rate with the additional examination of the NB, TF and DV,
which represents a 6% increase compared to combined
screening alone. The false positive rate was only slightly reduced
from 5.3 to 4.8%.
18
Hsiao et al. observed an increase in the
detection rate from 67% to 100% and a decrease of the false
positive rate from 4.1% to 2.4% if these additional three markers
were assessed after combined screening.
10
In contrast to these previous studies, we have focused on the
ultrasound examination only without including biochemical
screening. The detection rate increased from 68% using MA
and fetal NT only to 94% if MA, fetal NT and all three additional
markers were assessed. We acknowledge that the screening
performance based on MA and fetal NT was slightly lower than
expected.
19
This may be because of the referral character of our
study population: although we have noted a change in the
indications for invasive testing over the years, a major reason
for referral was still advanced maternal age followed by
increased NT. However, as none of our patients was referred
for abnormal results of the NB, TF or DV examination, our
results are still representative.
Limitations of our study
Our study has some limitations. First, we examined a high risk
population that was referred because of increased MA or
increased NT. For a fixed false positive rate of 3%, the detection
rate in screening based on MA alone was substantially lower than
the expected 20–30%. However, the detection rates for a fixed risk
cut-off are still representative as the MA and fetal NT distribution
and the prevalence of abnormal NB, TF and DV findings in our
cases with trisomy 21 is consistent with larger screening
studies.
4–6
In the general population, however, the false positive
rates for the respective risk cut-off should be lower than our study.
A second weakness lies in the fact that our risk calculation
did not take into account the fact that NB, TF and DV are not
entirely independent. The FMF software package allows the
use all different marker combinations. However, since the
studies of the FMF focused only on one of the additional
markers, it does not use correction factors if more than one
additional marker is examined. As a result, the precision of
individual risk assessment is diminished though the clinical
impact is probably minimal.
Finally, we acknowledge that the assessment of the additional
markersisapowerfultoolinthe hands of fetal medicine
experts. However, it may be too difficult to incorporate
this approach into general screening policies. Most of the
reproducibility studies focused on the assessment of the
DV. Maiz et al. reported on about 120 examinations that are
necessary to reach an acceptably high intra- and inter-
operator reproducibility.
8
Consistently, Borrell et al.reported
on an intra- and interclass correlation of 0.81 and 0.50 for
the examination of the DV which underlines the need for
careful training and quality assessment when using this
marker in screening for trisomy 21.
20
Implication for clinical practice
Our results indicate that in expert hands, first trimester
ultrasound screening based on MA, fetal NT and all three
additional markers results in detection rates of more than
95%. The respective false positive rate in our study was about
4% because of the referral character of our study population.
However, in a general population a lower false positive rate
can be expected. For a risk cut-off of 1:50, the false positive
ratebasedonMAandfetalNTandoneoftheadditional
ultrasound markers is 1.3–1.4%.
7
The examination of two or
three of these markers will lead to a further reduction in the
false positive rate.
Currently, a lot of the discussion focuses on how first
trimester screening can be combined with new but expensive
screening methods such as cffDNA analysis. We have
recently reported on a policy based on MA, fetal NT and DV
pulsatility index for veins for all patients followed by cffDNA
in patients with an intermediate risk. If the risk cut-offs 1 in
50 and 1 in 1000 are used, the detection and false positive
rates are 1.3 and 94%.
21
SuchapolicywouldlimitcffDNA
testing to about 9% of the population. Because of the
high negative likelihood ratio, it can be expected that the
additional examination of the NB and the TF would further
reduce the need of cffDNA testing.
CONCLUSION
In this study we have shown that in the hands of fetal medicine
experts first trimester ultrasound screening based on MA, fetal
NT and the additional examination of all three ultrasound
markers NB, TF and DV results in a higher detection rate than
if only one of these ultrasound markers is assessed.
WHAT’S ALREADY KNOWN ABOUT THIS TOPIC?
•First trimester screening for Down syndrome based on maternal age
and fetal NT detects about 70% of affected fetuses for a 3% false
positive rate.
WHAT DOES THIS STUDY ADD?
•The assessment of one, two and three of the additional ultrasound
markers nasal bone, tricuspid and ductus venosus flow increases
the detection rate to about 80%, 87% and 94%, respectively, for
the same false positive rate.
H. Abele et al.
Prenatal Diagnosis 2015, 35,1–5 © 2015 John Wiley & Sons, Ltd.

REFERENCES
1. Bianchi DW, Oepkes D, Ghidini A. Current controversies in prenatal
diagnosis 1: should noninvasive DNA testing be the standard screening
test for Down syndrome in all pregnant women? Prenat Diagn 2014;
34(1):6–11.
2. Chitty LS, Bianchi DW. Noninvasive prenatal testing: the paradigm is
shifting rapidly. Prenat Diagn 2013;33(6):511–3.
3. Kagan KO, Wright D, Baker A, et al. Screening for trisomy 21 by maternal
age, fetal nuchal translucency thickness, free beta-human chorionic
gonadotropin and pregnancy-associated plasma protein-A. Ultrasound
Obstet Gynecol 2008;31(6):618–24.
4. Kagan KO, Cicero S, Staboulidou I, et al. Fetal nasal bone in screening
for trisomies 21, 18 and 13 and Turner syndrome at 11–13 weeks of
gestation. Ultrasound Obstet Gynecol 2009;33(3):259–64.
5. Kagan KO, Valencia C, Livanos P, et al. Tricuspid regurgitation in
screening for trisomies 21, 18 and 13 and Turner syndrome at 11 + 0 to
13 + 6 weeks of gestation. Ultrasound Obstet Gynecol 2009;33(1):18 –22.
6. Maiz N, Valencia C, Kagan KO, et al. Ductus venosus Doppler in
screening for trisomies 21, 18 and 13 and Turner syndrome at
11–13 weeks of gestation. Ultrasound Obstet Gynecol 2009;33 (5):512–7.
7. Kagan KO, Staboulidou I, Cruz J, et al. Two-stage first-trimester
screening for trisomy 21 by ultrasound assessment and biochemical
testing. Ultrasound Obstet Gynecol 2010;36(5):542–7.
8. Maiz N, Kagan KO, Milovanovic Z, et al. Learning curve for Doppler
assessment of ductus venosus flow at 11 + 0 to 13 + 6 weeks’gestation.
Ultrasound Obstet Gynecol 2008;31(5):503–6.
9. Ghaffari SR, Tahmasebpour AR, Jamal A, et al. First-trimester screening
for chromosomal abnormalities by integrated application of nuchal
translucency, nasal bone, tricuspid regurgitation and ductus venosus
flow combined with maternal serum free β-hCG and PAPP-A: a 5-year
prospective study. Ultrasound Obstet Gynecol 2012;39(5):528–34.
10. Hsiao C-H, Cheng P-J, Shaw SWS, et al. Extended first-trimester
screening using multiple sonographic markers and maternal serum
biochemistry: a five-year prospective study. Fetal Diagn Ther
2014;35:296–301.
11. Snijders R, Noble P, Sebire N, et al. UK multicentre project on
assessment of risk of trisomy 21 by maternal age and fetal
nuchal-translucency thickness at 10–14 weeks of gestation. Lancet
1998;352(9125):343–6.
12. Kagan KO, Wright D, Valencia C, et al. Screening for trisomies 21, 18 and
13 by maternal age, fetal nuchal translucency, fetal heart rate, free-hCG
and pregnancy-associated plasma protein-A. Hum Reprod 2008;
23(9):1968–75.
13. Nicolaides KH, Chervenak FA, McCullough LB, et al. Evidence-based
obstetric ethics and informed decision-making by pregnant women
about invasive diagnosis after first-trimester assessment of risk for
trisomy 21. Am J Obstet Gynecol 2005;193(2):322–6.
14. Cuckle HS, Wald NJ, Thompson SG. Estimating a woman’s risk of having
a pregnancy associated with Down’s syndrome using her age and serum
alpha-fetoprotein level. Br J Obstet Gynaecol 1987;94(5):387–402.
15. Snijders RJ, Sundberg K, Holzgreve W, et al. Maternal age- and
gestation-specific risk for trisomy 21. Ultrasound Obstet Gynecol
1999;13(3):167–70.
16. Wright D, Kagan KO, Molina FS, et al. A mixture model of nuchal
translucency thickness in screening for chromosomal defects.
Ultrasound Obstet Gynecol 2008;31(4):376–83.
17. Maiz N, Wright D, Ferreira AFA, et al. A mixture model of ductus
venosus pulsatility index in screening for aneuploidies at 11–13 weeks’
gestation. Fetal Diagn Ther 2012;31(4):221–9.
18. Karadzov-Orlic N, Egic A, Milovanovic Z, et al. Improved diagnostic
accuracy by using secondary ultrasound markers in the first-trimester
screening for trisomies 21, 18 and 13 and Turner syndrome. Prenat
Diagn 2012;32(7):638–43.
19. Kagan KO, Etchegaray A, Zhou Y, et al. Prospective validation of
first-trimester combined screening for trisomy 21. Ultrasound Obstet
Gynecol 2009;34(1):14–8.
20. Borrell A, Perez M, Figueras F, et al. Reliability analysis on ductus
venosus assessment at 11–14 weeks’gestation in a high-risk population.
Prenat Diagn 2007;27(5):442–6.
21. Kagan KO, Wright D, Nicolaides KH. First-trimester contingent
screening for trisomies 21, 18 and 13 by fetal nuchal translucency and
ductus venosus flow and maternal blood cell-free DNA testing.
Ultrasound Obstet Gynecol 2015;45(1):42–7.
First trimester ultrasound screening
Prenatal Diagnosis 2015, 35,1–5 © 2015 John Wiley & Sons, Ltd.