Title

Author(s)

Rapid aneuploidy testing or traditional karyotyping, or both, in

prenatal diagnosis

Leung, Wing-cheong;

Citation

Issued Date

URL

Rights

2010

http://hdl.handle.net/10722/132149

The author retains all proprietary rights, (such as patent rights)

and the right to use in future works.

Rapid Aneuploidy Testing or Traditional Karyotyping, or Both,

in Prenatal Diagnosis
By
Wing Cheong LEUNG
MBBS, FHKAM(O&G), FHKCOG, FRCOG,
Cert RCOG(Maternal and Fetal Med)

A thesis submitted in fulfilment of the requirement for

The Degree of Doctor of Medicine
At the University of Hong Kong
Li Ka Shing Faculty of Medicine
September 2010

Abstract of thesis entitled

Rapid Aneuploidy Testing or Traditional Karyotyping, or Both, in
Prenatal Diagnosis
Submitted by
Wing Cheong LEUNG
For the Degree of Doctor of Medicine (MD)
At The University of Hong Kong
Li Ka Shing Faculty of Medicine
In September 2010

The major advantages of rapid aneuploidy testing (RAT), using either fluorescence in
situ hybridisation (FISH) with chromosome-specific DNA probes or quantitative
fluorescencepolymerase chain reaction (QF-PCR) with chromosome-specific small tandem
repeat markers, include fast reporting within 24 to 48 hours and earlier relief of anxiety.
Their accuracy in prenatal diagnosis of the common aneuploidies of chromosomes 21, 18, 13,
X and Y has been demonstrated (targeted testing). On the other hand, traditional karyotyping
examines all 23 pairs of chromosomes. Apart from the common aneuploidies, a wide range
of chromosomal abnormalities can be identified by this technique, including rearrangements,
such as translocations and inversions that may be balanced or unbalanced. But karyotyping is
labour-intensive and the results are usually not available for 2 weeks or more.
The challenge is to decide what would be the best approach - rapid aneuploidy testing or
traditional karyotyping, or both? The results of studies in this thesis provide information on
the pros and cons of each approach. The performance of a stand-alone RAT approach
depends on the indication for prenatal diagnosis. A high standard ultrasound examination is
essential. It is reasonable to use a stand-alone RAT approach when prenatal tests are
performed for indications such as positive Down screening with no ultrasound abnormalities.
It is important to note that if this new approach is used in prenatal diagnosis, for every 1000
invasive tests performed, up to 4 potentially clinical significant chromosomal abnormalities
could be missed. If both techniques are available, women should have the autonomy to
choose the approach (RAT, or traditional karyotyping, or both) after being fully informed of
the pros and cons.
(269 words)

To my wife, Connie & my son, Leslie

Declaration

I declare that the thesis and the research work thereof represents my own work, except where
due acknowledgement is made, and that it has not been previously included in a thesis,
dissertation or report submitted to this University or to any other institution for a degree,
diploma or other qualifications.

Signed

Acknowledgements
This research idea was generated and most of the studies in this thesis were performed
during my subspecialty training in Maternal Fetal Medicine (MFM) in Toronto (1999), Hong
Kong (2000-1) and London (2002-3). The studies would not have been possible without the
help and support from many people. I am also grateful to all the women and their partners
who joined the studies.
First of all, I want to thank Prof. Terence Lao and Dr. Mary Tang, MFM Subspecialty
Training Programme Directors, Department of Obstetrics & Gynaecology (O&G), HKU,
who have supported the research idea and studies from the very beginning and always given
excellent advice. I would also like to thank Prof. PC Ho, Chair Professor, Department of
O&G, HKU, who has supported my overseas training in Toronto and London. Special thanks
to Prof. Grace Tang and Prof. Lawrence Tang, HKU who have encouraged me to start
writing the thesis and reminded me from time to time not to give up; and Prof. Frank
Chervenak, Weill Medical College of Cornell University, New York, who has helped me to
arrive at the final conclusion of the thesis. I am particularly indebted to Prof. Greg Ryan &
Prof. Knox Ritchie, Perinatal Centre, University of Toronto, who have inspired me on this
research idea and allowed me to use the Toronto database; Prof. Lyn Chitty & Prof. Charles
Rodeck, Fetal Medicine Centre, University College London, who have also supported this
research idea and allowed me to use the London database; and Dr. Elizabeth Lau,
Department Manager, Prenatal Diagnosis & Conselling Department, Tsan Yuk Hospital, who
has supported the studies with a lot of time and effort. I also thank Dr. Robert Chin, Chief of
Service, Department of O&G, Kowloon West Cluster Hospitals, who has allowed me time to
complete the thesis on top of the clinical work.
I appreciate all the help, comments and suggestions from my co-authors: Dr. WL Lau, Dr.
CP Lee, Dr. YH Lam, Dr. KY Tse, Prof. Vivian Chan, Dr. Cora Ngai, Dr. Helena Lam, Dr. KY
Leung, Dr. Rebecca Tang, Dr. Shell Wong, Prof. TK Lau, Dr. KT Tse, Dr. SF Wong, Dr. WK To,
Ms. Lucy Ng, Dr. Grace Wong, Dr. Jonathan Waters, Prof. David Chitayat, Prof. Gareth
Seaward, Prof. Rory Windrim, Prof. Jon Barrett and Prof. Elizabeth Winsor. Special thanks to
Dr. Daniel Fong for statistical advice.
My appreciation also goes to the staff of Mrs. Wu Chung Prenatal Diagnostic
Laboratory, Tsan Yuk Hospital; Ms. YP Lee and Ms. Vivian Chan, Nurse Specialists, Prenatal
Diagnosis & Conselling Department, Tsan Yuk Hospital; Ms. Jane Chan, Ms. KY Wong,
Research Assistants; and Ms. Amanda Chung (clerical support). Without their support, the
studies and thesis would not have been completed.
ii

Finally, I want to express my sincere gratitude to my wife, Connie, my son, Leslie, and
my parents, for support and patience during these years of studies.

iii

Table of Contents

Declaration
Acknowledgements .
Table of Contents ..

i
ii
iv

Chapter 1

INTRODUCTION & OUTLINE OF THESIS .

Chapter 2

LITERATURE REVIEW ..

Chapter 3

BACKGROUND STUDY ..
16
Role of amniotic fluid interphase fluorescence in situ hybridization (FISH)
analysis in patient management

Chapter 4

APPLICATION OF RAPID ANEUPLOIDY TESTING IN PRENATAL

DIAGNOSIS

4.1
4.2
4.3
4.4

Positive Down screening ..

Advanced Maternal Age ...
Abnormal Ultrasound Findings
Other Indications Thalassaemia

Chapter 5

ANXIETY STUDIES

5.1

A randomized controlled trial on the effect of fast reporting by amnio-PCR on

anxiety levels in women with positive biochemical screening for Down
syndrome ..
65
A prospective study on the effect of rapid aneuploidy testing on anxiety levels
and quality of life measures in women and their partners with positive Down
screening ..
72

5.2

Chapter 6

DISCUSSION & CONCLUSION .

References ..
Curriculum vitae .
Publications (resulted from the work of this thesis)

iv

26
35
47
56

82
88
98
99

Chapter 1
Introduction & Outline of Thesis

TITLE OF THESIS
Rapid Aneuploidy Testing or Traditional Karyotyping, or Both, in Prenatal Diagnosis

INTRODUCTION
The most frequent fetal chromosomal abnormalities involve the autosomes 21, 18, 13, and
sex chromosomes X and Y. Aneuploidy or alterations in copy number of these chromosomes,
including trisomy 21 (Down syndrome), trisomy 18 (Edwards syndrome), trisomy 13
(Pataus syndrome), 45,X (Turners syndrome), 47,XXY (Klinefelters syndrome) and
triploidy (presence of three copies of each chromosome), account for 80% of clinically
significant chromosomal abnormalities diagnosed in the prenatal period. Down syndrome is a
well-recognized cause of mental retardation, cardiac, and other congenital abnormalities.
Edwards syndrome and Pataus syndrome lead to multiple congenital abnormalities and
early neonatal death. The phenotype of Turners syndrome is highly variable and includes
short stature, amenorrhoea, infertility, cardiac and renal malformations. Klinefelters
syndrome is associated with a relatively mild phenotypic abnormality. Fetuses with triploidy
are severely growth retarded and usually die in utero.

Figure 1. Traditional karyotyping showing a cultured amniotic fluid sample with Down
syndrome (47,XY,+21)
The traditional method for prenatal diagnosis of these common aneuploidies involves
2

analysis of banded metaphase chromosomes from cultured amniotic fluid cells

(amniocentesis) or chorionic villi (chorionic villus sampling). It is known as karyotyping, for
which all 23 pairs of chromosomes are examined (Figure 1). Apart from the common
aneuploidies, a wide range of chromosomal abnormalities can be identified by this technique,
including rearrangements, such as translocations and inversions that may be balanced or
unbalanced. Traditional karyotyping is labour-intensive and the results are usually not
available for 2 weeks or more. Advances in molecular diagnostics, using either fluorescence
in situ hybridisation (FISH) with chromosome-specific DNA probes or quantitative
fluorescencepolymerase chain reaction (QF-PCR) with chromosome-specific small tandem
repeat markers, can be applied to diagnose the common aneuploidies within 1 to 2 days. The
sensitivity and specificity of FISH and QF-PCR, collectively described as rapid aneuploidy
testing (RAT), have been demonstrated in a number of large-scale studies and compare
favourably with traditional karyotyping for the diagnosis of the common aneuploidies.
Unlike karyotyping, these technologies only allow the identification of the chromosomal
abnormalities that are specifically sought (targeted testing).
Currently, RAT (FISH or QF-PCR) is being used to give a rapid result for the common
aneuploidies as an adjunct to karyotyping. This combined approach clearly increases the cost
of prenatal diagnosis. One school of thought suggested that if the indication for prenatal
diagnosis is an increased risk of Down syndrome arising from a positive screening test result
or advanced maternal age, then karyotyping could be replaced by RAT. On the other hand,
this new approach is not supported by the other school because certain chromosomal
abnormalities, although infrequent and usually of debatable significance, would be missed.
This has resulted in much debate in the field of prenatal diagnosis.
The aim of this thesis is to answer the following central research question:
Rapid Aneuploidy Testing or Traditional Karyotyping, or Both, in Prenatal Diagnosis
This research idea was generated and most of the studies in this thesis were performed during
my subspecialty training in Maternal Fetal Medicine (MFM) in Toronto (1999), Hong Kong
(2000-1) and London (2002-3).

FISH
FISH (Ward et al., 1993; Eiben et al., 1999a; Estabrooks et al., 1999; Pergament et al., 2000;
Witters et al., 2002) involves hybridisation of selected chromosome specific DNA sequences
3

that have been labelled with fluorescent dyes to chromosome preparations. The fluorescently
labelled sequences stick to the corresponding DNA of the chromosomes and can be
visualised under the microscope (Figure 2). Normal samples show 2 dots per cell nucleus,
whereas trisomic samples show 3 dots. 50 to 100 cells are usually analysed to allow for
low-level background and signal overlay that occur during FISH procedures.

Figure 2. FISH assays on interphase nuclei of uncultured amniotic fluid cells probed with
LSI 21 (Vysis) for 21q22.13 to 21q22.2 region on the long arm of chromosome 21. Two
signals detected from a normal sample (A) and three signals detected from a sample with
trisomy 21 (B)

QF-PCR
QF-PCR (Pertl et al., 1994; Verma et al., 1998; Levett et al., 2001; Cirigliano et al.,
2001&2004; Mann et al., 2001) involves the amplification of chromosome-specific repeated
DNA sequences known as small tandem repeats (STRs). STRs are stable and polymorphic,
varying in length between subjects, depending on the number of times the tri-, tetra- or
penta-nucleotides are repeated. The sample DNA from amniotic fluid or chorionic villi is
amplified by PCR using fluorescent primers so that products can be visualised and quantified
as peak areas of the respective repeat lengths using an automated DNA sequencer with the
gene-scan software (Figure 3). DNA amplified from normal subjects who are heterozygous
(having alleles of different STR lengths) will show 2 peaks with the same area. DNA
amplified from trisomic subjects will show either an extra peak (triallelic) with the same area,
or only 2 peaks (diallelic), one of them twice as large as the other. The number and variety of
STR markers multiplexed together differ between assays and determine assay efficiency.

Figure 3. QF-PCR with STR markers showing Trisomy 21 from amniotic fluid. Arrows
showed diallelic pattern with ratio 2:1 (D21S1411) or triallelic patterns with ratio 1:1:1
(D21S1414, D21S1412). Normal patterns (1:1 ratio) are observed for chromosome 18
(D18S535, D18S51, D18S386) and chromosome 13 (D13S631, D13S258)

OTHER MOLECULAR METHODS FOR RAT

Other PCR-based approaches to RAT (Mann et al., 2004) include homologous gene
quantitative PCR (HGQ-PCR) (Lee et al., 1997) and real-time PCR (Zimmermann et al.,
2002). In addition, multiplex ligation-dependent probe amplification (MLPA) (Hochstenbach
et al., 2005) and microarray comparative genomic hybridisation (array CGH) (Choy et al.,
2008) can also be used for RAT. These other methods are less extensively studied when
compared to FISH or QF-PCR in RAT and studies in this thesis are referring to RAT using
FISH or QF-PCR only.

QF-PCR vs. FISH

Comparing QF-PCR with FISH (Hulten et al., 2003):
1. The risk for misdiagnosis of the common aneuploidies by either FISH or QF-PCR is
relatively small.
2. FISH is more labour intensive than QF-PCR.
3. Maternal cell contamination may constitute more of a problem with FISH than with
QF-PCR.
4. Fetal mosaicism remains a challenge by either method.
5

Maternal Cell Contamination

Maternal cell contamination of fetal material may arise during any of the invasive prenatal
sampling procedures. With FISH, mixture of maternal and fetal XY cells are readily
detectable but maternal and fetal XX cells are indistinguishable. Using QF-PCR, maternal
cell contamination is readily detected by the characteristic pattern with extra alleles or
skewed ratios between peaks for all target chromosomes (Stojilkovic-Mikic et al., 2005).
Fetal Mosaicism
It refers to the occurrence of more than one cell line containing different chromosomal results.
Using FISH, examination of a large number of interphase nuclei facilitates the diagnosis of
mosaicism. Low-grade mosaicism is likely to be missed. QF-PCR is also capable of
identifying autosomal mosaicism, when the trisomy is present in more than 15% (Donaghue
et al., 2005).

OUTLINE OF THESIS
In order to answer the central research question: Rapid Aneuploidy Testing or Traditional
Karyotyping, or Both, in Prenatal Diagnosis, the following studies were performed:
Chapter 2 is a literature review within this research theme and the pros and cons of RAT
stand-alone approach in prenatal diagnosis are discussed, including the suggestion of studies.
Chapter 3 described a background study on the role of RAT in patient management. The
research idea of this thesis is generated from this study. The feasibility of RAT, karyotyping,
or both is then studied in different indications for prenatal diagnosis - Chapter 4.1 (positive
Down screening), Chapter 4.2 (advanced maternal age), & Chapter 4.4 (other indications
using thalassaemia as an example). In Chapter 4.3, the role of ultrasound examination in the
RAT stand-alone approach is elaborated. The effect of RAT on anxiety levels in women and
their partners with positive Down screening result is also studied Chapter 5.1 & 5.2.
Chapter 6 is general discussion and conclusion.
Chapters 2, 3, 4.1, 4.2, 4.3, 4.4, 5.1 & 5.2 are reproduced from my published papers with
permission from the corresponding journals.

Chapter 2
Literature Review
Reproduced with permission from Expert Rev Mol Diagn:
Leung WC, Lau ET, Lao TT, Tang MH. Rapid aneuploidy screening
(FISH or QF-PCR): the changing scene in prenatal diagnosis? Expert Rev
Mol Diagn 2004;4:333-7.

ABSTRACT
The accuracy of new molecular diagnostics, fluorescence in situ hybridization or quantitative
fluorescence-polymerase chain reaction (collectively known as rapid aneuploidy testing), in
prenatal diagnosis has already been demonstrated in a number of large studies. The challenge
now is how to apply them clinically in the most cost-effective manner. It is now time to
appraise whether RAT can replace traditional karyotyping when amniocenteses are
performed for increased risk of Downs syndrome by maternal serum screening or advanced
maternal age in the absence of ultrasound abnormality. The ten most recent studies from the
literature within this research theme are reviewed and the pros and cons of this new approach
in prenatal diagnosis are discussed, including the suggestion of future studies.

INTRODUCTION
Amniocentesis allows prenatal diagnosis of fetal chromosomal abnormalities during
pregnancy. It is commonly restricted to pregnancies considered to be at high risk: as a result
of antenatal biochemical and ultrasound screening tests, due to advanced maternal age, or
due to a history of a parental balanced chromosomal rearrangement. The traditional gold
standard method for detection of fetal chromosomal abnormality involves analysis of banded
metaphase chromosomes from cultured cells (karyotyping). This technique identifies a wide
range of chromosomal abnormalities, including alterations in copy number (aneuploidy) as
well as chromosomal rearrangements, such as translocations and inversions that may be
balanced or unbalanced. However, the technique is labor-intensive and results are not usually
available for 2 weeks or more.
Advances in molecular diagnostics, using either fluorescence in situ hybridization (FISH)
with specific DNA probes (Ward et al., 1993) or quantitative fluorescence-polymerase chain
reaction (QF-PCR) with specific small tandem repeat markers (Pertl et al., 1994), can be
applied to diagnose the common aneuploidies of chromosomes 21, 18, 13, X and Y within 1
or 2 days. The sensitivity and specificity of these tests, collectively described in this review
as rapid aneuploidy testing (RAT), have been demonstrated in a number of large-scale
studies either using FISH (Eiben et al., 1999a; Estabrooks et al., 1999; Pergament et al., 2000;
Witters et al., 2002) or QF-PCR (Verma et al., 1998; Levett et al., 2001; Mann et al., 2001).
They compare favorably with traditional karyotyping for the diagnosis of the most frequent
and clinically important aneuploidies (trisomies 21, 18 and 13) as well as sex chromosome
aneuploidies. These technologies will, however, only allow the identification of the
chromosomal rearrangements that are specifically being sought.
8

If RAT (FISH or QF-PCR) is being used to give a preliminary rapid result for the major
aneuploidies in addition to traditional karyotyping, it would clearly increase the cost to the
service provider. One school of thought suggested that if the indication for amniocentesis is
an increased risk of Downs syndrome, such as positive screening test result or advanced
maternal age, traditional karyotyping could be effectively replaced by RAT (Thilaganathan et
al., 2000; Thein et al., 2000; Ryall et al., 2001; Leung et al., 2003). However, this new
approach for prenatal diagnosis is not supported by the other school (Witters et al., 2002;
Evans et al., 1999; Lewin et al., 2000; Leung et al., 2001; Homer et al., 2003) because certain
chromosomal abnormalities, although of a small number and might or might not be clinically
significant, would be missed. The aim of this review article is to summarize and analyze the
findings of these studies that gave rise to the two different schools of thoughts, focusing on
the pros and cons of this new approach in prenatal diagnosis (Table 1).

Table 1. Comparison of rapid aneuploidy screening (RAT) & traditional karyotyping results
[10] Thilaganathan et al., 2000; [14] Evans et al., 1999; [11] Thein et al., 2000; [15] Lewin
et al., 2000; [12] Ryall et al., 2001; [16] Leung et al., 2001; [6] Witters et al., 2002; [13]
Leung et al., 2003; [17] Homer et al., 2003; [18] Grimshaw et al., 2003
COMPARISON OF ANEUPLOIDY SCREENING & TRADITIONAL KARYPING
RESULTS
9

Table 1 showed a comparison of RAT and traditional karyotyping results from ten recent
studies in chronological order from 1999 to 2003 (Thilaganathan et al., 2000; Evans et al.,
1999; Thein et al., 2000; Lewin et al., 2000; Ryall et al., 2001; Leung et al., 2001; Witters et
al., 2002; Leung et al., 2003; Homer et al., 2003; Grimshaw et al., 2003). RAT (FISH or
QF-PCR) was actually performed in four of these studies (Witters et al., 2002; Thilaganathan
et al., 2000; Leung et al., 2001; Grimshaw et al., 2003). In the other six studies (Thein et al.,
2000; Ryall et al., 2001; Leung et al., 2003; Evans et al., 1999; Lewin et al., 2000; Homer et
al., 2003), RAT was only assumed to have been performed. The majority of samples were
amniotic fluid. The results of this analysis thus mainly referred to second-trimester prenatal
diagnosis by amniocentesis. All indications were included in order to have a global view and
to avoid missing those cases categorized under ultrasound abnormalities that may also have a
positive maternal serum screening test or advanced maternal age.
The results were divided into four groups: concordant normal, concordant abnormal, false
positive and false negative.

CONCORDANT NORMAL
As expected, concordant normal RAT with normal karyotyping results were the most
common scenario (96.8%). In other words, traditional karyotyping did not give additional
information to RAT in the great majority of women having amniocenteses. Most of these
amniocenteses were performed because of positive maternal serum Down screening test
result or advanced maternal age. The major advantage of RAT in this great majority group is
that the rapid normal result within 1 to 2 days can relieve the anxiety of the women and their
partners much earlier than when they have to wait for the traditional karyotyping result,
which can take up to 3 weeks. However, a recent randomized controlled trial has suggested
that this advantage of RAT may be lost if the women still have to wait for the traditional
karyotyping result (Leung et al., 2002). One possible explanation is that the woman, although
being told that the RAT result is normal, is still having a significant degree of anxiety as she
has to wait for the confirmation by the karyotyping result. This anxiety might be alleviated if
the RAT report is considered to be final. The other advantage of using RAT as a stand-alone
test is cost saving. Instead of adding the cost of RAT on top of that of traditional karyotyping,
the cost of the latter can be saved by the RAT alone approach. In the age of ever-escalating
cost in the provision of healthcare, especially in a government-funded public medical care
system, the savings can be redirected to enhance existing or fund new programs, thus
maximizing the effect of limited resources.
10

It is important to note that 0.14.4% of RAT results were noninformative mainly due to
maternal cell contamination (Witters et al., 2002; Thilaganathan et al., 2000; Grimshaw et al.,
2003). A back-up mechanism (e.g., the facility to culture cells until the RAT result is
available) has to be in place if RAT is used as a stand-alone test.

CONCORDANT ABNORMAL
Concordant abnormal RAT with the same abnormal karyotyping results accounted for 2.3%
of the results (Table 1). In this group of women, we can again see that traditional karyotyping
does not give additional information to RAT. At present, like many centers (Adinolfi &
Sherlock, 2001), our center is already offering the option of termination of pregnancy based
on abnormal RAT results without waiting for the traditional karyotyping confirmation
(Leung et al., 2003). In this regard, RAT can again be considered as a stand-alone test.

FALSE POSITIVE
This refers to the scenario when the RAT result is falsely abnormal in the presence of a
normal karyotype, which can potentially result in termination of a normal pregnancy. It had
occurred in one (Thilaganathan et al., 2000) out of the 214,031 cases in the ten recent studies
that have been reviewed (Table 1). This case was a 45,X/46,XX mosaic predicted by FISH
but 46,XX found on karyotyping (Thilaganathan et al., 2000). The author gave the
explanation that it was the result of extreme variation in size of the alpha satellite
centromeric region of the X chromosome (Thilaganathan et al., 2000). This was a rare
occurrence and manifested as very low signal strengths on fluorescent microscopy. When
very low signal strengths are encountered, FISH analysis is repeated with alternative probes
or reliance should be placed on karyotyping (Thilaganathan et al., 2000). Nevertheless, the
absence of a false-positive result is a basic prerequisite if RAT is to be used as a stand-alone
test.

FALSE NEGATIVE
This refers to the scenario when the RAT result is normal but the traditional karyotyping
result is abnormal. It accounted for 0.9% of the results in our review (Table 1). These
abnormal karyotypes could be divided into two groups: not clinically significant (0.6%) and
11

clinically significant (0.4%). The not clinically significant group referred to those balanced
translocations or other chromosomal rearrangements of known familial origin. The clinically
significant group included rare aneuploidies (other than chromosomes 21, 18, 13, X and Y),
unbalanced translocations or other chromosomal rearrangements, balanced de novo
translocations and marker chromosomes. Strictly speaking, they should not be considered as
false-negative RAT results because the FISH or QF-PCR that had been used could only
detect the common aneuploidies (chromosomes 21, 18, 13, X and Y). However, this is the
group of chromosomal abnormalities that would be missed if RAT is to replace traditional
karyotyping.
Some but not all of the clinically significant chromosomal abnormalities that cannot be
detected by RAT would have evidence of major structural abnormalities or soft markers of
aneuploidy on ultrasound examination (Thein et al., 2000; Leung et al., 2001). It has
therefore been suggested that RAT can be used as a stand-alone test in conjunction with
detailed ultrasound examination, while traditional karyotyping will only be performed for
those cases with ultrasound abnormalities (Leung et al., 2003). The cost effectiveness of this
approach depends not only on the standard of ultrasound examination but also on the
incidence of ultrasound abnormalities in the entire group of women undergoing
amniocenteses. A prospective study is necessary to address this issue.
Nevertheless, there would be some clinically significant false-negative RAT chromosomal
abnormalities that do not have ultrasound abnormalities. It is important to note that the
clinical significance of these chromosomal abnormalities, in particular balanced de novo
translocations and marker chromosomes, is very different from that of trisomy 21, 18 or 13
(Warburton, 1991). The risk of an adverse clinical outcome (including impaired intellectual
development, learning difficulties and physical abnormalities) for this cohort of
chromosomal abnormalities varies from 5 to 15% (Warburton, 1991). Identification of these
balanced de novo translocations and marker chromosomes in the absence of ultrasound
abnormalities often poses difficult counseling issues, may not be in the best interest of the
parents or the fetus and presents a difficult choice regarding the continuation of the
pregnancy.
In addition to the above clinically significant chromosomal abnormalities, false-negative
RAT results also included those balanced translocations or other chromosomal
rearrangements of known familial origin. Although they are not clinically significant, they
have the potential to result in unbalanced products in future pregnancies.

12

EXPERT OPINION
When RAT was compared with traditional karyotyping, the incidences of concordant normal,
concordant abnormal, false positive and false negative results were very consistent among
the various studies (Thilaganathan et al., 2000; Evans et al., 1999; Thein et al., 2000; Lewin
et al., 2000; Ryall et al., 2001; Witters et al., 2002; Leung et al., 2003; Homer et al., 2003;
Grimshaw et al., 2003) except one (Leung et al., 2001) that was limited by the small number
and the fact that more than a third of the amniocenteses were performed for major ultrasound
abnormalities (Table 1). However, regarding whether RAT could replace traditional
karyotyping as a stand-alone test, the opinions were less consistent. The authors of four
studies (Thilaganathan et al., 2000; Thein et al., 2000; Ryall et al., 2001; Leung et al., 2003)
supported this approach while those of five studies did not (Witters et al., 2002; Evans et al.,
1999; Lewin et al., 2000; Leung et al., 2001; Homer et al., 2003). The authors of one study
did not commit and stated in the article that further research in this area is needed (Grimshaw
et al., 2003). This inconsistency reflects the controversy of this new approach of RAT in
prenatal diagnosis, which may be influenced by the perspective of the investigators (Ogilvie,
2003).
The authors believe that RAT can replace traditional karyotyping if the indication for
amniocentesis is positive Down screening test or advanced maternal age and when there is no
ultrasound abnormality. But then one must accept the risk that for every 1000 amniocenteses
performed, up to four potentially clinical significant chromosomal abnormalities may be
missed. Some people may argue from an ethical point of view that since amniocentesis is an
invasive procedure that carries a small risk of miscarriage, we should maximize the
information that can be obtained by performing full karyotyping. However, we must also
realize that Down screening by maternal serum markers or advanced maternal age is a
program designed to detect primarily Downs syndrome and therefore in principle, follow-up
test with RAT alone would have realistically fulfilled the expectations of the couples and
obstetricians. Furthermore, even the performance of full karyotyping does not mean that the
information is maximized, for example, the presence of microdeletions and common
mutations are not tested.

FIVE-YEAR VIEW
It is predicted that this new approach of RAT replacing traditional karyotyping when
amniocentesis is performed for positive Down screening test or advanced maternal age in the
absence of ultrasound abnormality would be particularly applicable and cost effective when a
13

national screening program for Downs syndrome is to be launched in a country like China
with an enormous population and thus high throughput of samples and where traditional
karyotyping is not that deep seated or widely available as in the Western world. In that case,
it may be recommended limiting the RAT further to that of chromosome 21 only.
However, there are still a number of preparation studies that have to be performed before this
new approach can be put into practice:
a prospective study addressing the role of good quality ultrasound examination: the overall
proportion of cases with soft markers or major ultrasound abnormalities and particularly the
proportion of false negative cases without ultrasound abnormalities
a randomized controlled trial comparing the anxiety levels of women undergoing
amniocenteses for increased risk of Downs syndrome, one group with RAT alone and the
other group with traditional karyotyping
acceptability studies of this new approach from the point of view of obstetricians,
maternalfetal medicine subspecialists and most important, the women and their partners
a cost-effectiveness analysis comparing the setting up of a population screening program
for Downs syndrome using this new approach versus traditional karyotyping alone
an ethical study on the pros (avoid unnecessary termination of pregnancies because of
parental anxiety) and cons (miss the chance of prenatal diagnosis of those chromosomal
abnormalities not detectable by RAT) of this approach.

KEY ISSUES
The accuracy of rapid aneuploidy testing (RAT) (fluorescence in situ hybridization or
quantitative fluorescence-polymerase chain reaction) in prenatal diagnosis has already been
demonstrated.
Current and future research should focus on whether RAT could replace traditional
karyotyping when amniocentesis is performed for indications, such as positive Down
screening test or advanced maternal age.
Even under this new approach, RAT cannot completely replace traditional karyotyping
14

because the latter is required for those cases with ultrasound abnormalities.
A high-standard ultrasound examination is essential for this new approach.
The major advantages of this new approach include fast reporting within 24 to 48 h and
earlier relief of anxiety.
Using this approach in prenatal diagnosis, for every 1000 amniocenteses performed, up to
four potentially clinically significant chromosomal abnormalities may be missed (e.g.,
balanced de novo translocations and marker chromosomes).
This new approach is most applicable and cost-effective when a national screening program
for Downs syndrome is to be launched in a country with a large population and thus
high-throughput of samples and where traditional karyotyping is not very deep seated or
widely available.

15

Chapter 3
Background Study
Reproduced with permission from Prenat Diagn:
Leung WC, Chitayat D, Seaward G, et al. Role of amniotic fluid interphase
fluorescence in situ hybridization (FISH) analysis in patient management.
Prenat Diagn 2001;21:327-32.

16

ABSTRACT
We retrospectively reviewed 309 amniotic fluid interphase fluorescence in situ hybridization
(FISH) analyses performed from October 1995 to June 1999 to assess the role of interphase
FISH in the management of patients at increased risk for fetal aneuploidies. Gestational age
and indications for amniocentesis, clinical interventions after FISH results, as well as
interventions after final culture reports were analyzed. There were 244 (79%) normal, 50
(16%) abnormal and 15 (5%) inconclusive FISH results. There were no false-positive or
false-negative results, but there were nine (3%) clinically significant chromosomal
abnormalities not detectable by FISH. Of the 50 women with abnormal FISH results, 26
(52%) elected to terminate the pregnancy prior to the availability of the standard
chromosome analysis. In two of the fetuses with trisomy 21 no abnormalities were reported
by ultrasound examination. Our experience indicates that interphase FISH results played an
important role in decision making, especially for pregnancies close to 24 weeks gestation.
Standard karyotype analysis is still required for detection of chromosome abnormalities not
detectable by interphase FISH techniques and for clarification of unusual or inconclusive
FISH results.

INTRODUCTION
The traditional gold standard for prenatal diagnosis of chromosome abnormalities involves
analysis of banded metaphase chromosomes obtained from cultured amniotic fluid cells. This
is an expensive, labor-intensive laboratory procedure and results are not available for 13
weeks. The waiting period for results can be very stressful, especially when patients are
referred at gestations close to the time limit when therapeutic termination of pregnancy may
be an option. Fluorescence in situ hybridization (FISH) using probes specific for
chromosomes 13, 18, 21, X and Y has the potential to detect more than 80% of clinically
significant chromosome abnormalities within 24 h (DAlton et al., 1997; Eiben et al., 1999a;
Estabrooks et al., 1999; Evans et al., 1999). The availability of a test which is much faster,
but which provides less information than the traditional method, raises important questions
about the appropriate use of the FISH technique.
We present a retrospective review of 309 amniotic fluid interphase FISH analyses focusing
on the role of FISH in patient management. As use of the interphase FISH technique
becomes more widespread, it is important that patients be informed of its advantages and
limitations.

17

MATERIALS AND METHODS

The results of 309 amniotic fluid interphase FISH analyses performed from October 1995 to
June 1999 at the University of Toronto Perinatal Complex were retrospectively reviewed.
FISH was offered as an adjunct to the standard chromosome analysis at the discretion of the
obstetric or genetic staff. There was general agreement that interphase FISH testing would be
offered only if the results were likely to alter the options available to the patient, but rigid
selection criteria were not established prior to the beginning of the study.
The FISH analyses were performed on uncultured amniocytes using DNA probes specific for
chromosomes 13, 18, 21, X and Y. Due to changes in the availability of commercial probes
during the study, testing from October 1995 until August 1998 was performed using probes
obtained from Oncor (Gaithersburg, MD, USA) while from September 1998 to June 1999 the
probes were obtained from Vysis Inc. (AneuVysion, Downers Grove, IL, USA). In both
instances, alpha satellite probes were used for chromosomes 18, X and Y and locus specific
probes were used for chromosomes 13 and 21. Fifty nuclei were scored for each probe. The
result was classified as normal if normal signal patterns were observed in >80% of nuclei and
abnormal if >60% of the nuclei had abnormal signal patterns. Results were reported as
inconclusive if the above criteria were not met. The results of interphase FISH analyses were
compared to standard karyotype analyses from cultured cells.
The gestational age at amniocenteses and the indications for testing were recorded. The
indications were classified as ultrasound abnormalities, positive maternal serum screen
(MSS), advanced maternal age or other. The ultrasound abnormalities were classified as
minor abnormalities or soft markers (e.g. pyelectasis, choroid plexus cysts, short femur,
thick nuchal fold, intracardiac echogenic focus, mild ventriculomegaly, echogenic bowel,
clinodactyly, sandal gap toes) or major abnormalities (e.g. cleft lip, cardiac abnormality,
cystic hygroma, omphalocele, polyhydramnios, severe intrauterine growth restriction). For
the purpose of the present study the indications were classified in a hierarchical fashion. If
one or more ultrasound abnormalities were reported, the patient was classified as having
ultrasound abnormalities irrespective of the MSS result or maternal age. Maternal serum
screen was based on second trimester analysis of alpha fetoprotein, human chorionic
gonadotrophin and unconjugated estriol. Late maternal age was defined as age 35 years or
older at expected date of delivery.
Patient charts were reviewed to ascertain documented patient decisions and clinical
interventions following the availability of the interphase FISH results in addition to those
18

after the final karyotypes became available.

RESULTS
The indications for interphase FISH analyses on 309 amniotic fluids are shown in a
hierarchical fashion in Table 1. The identification of one or more ultrasound abnormalities
was the most common indication for amniotic FISH analysis (203/309=66%). Of the patients
with fetal ultrasound abnormalities, 28 also had a positive MMS. Although all of the patients
were eligible for traditional chromosome analysis, the decision to perform interphase FISH
analysis was strongly influenced by the gestation at the time of the amniocentesis (Figure 1).

Table 1. Indications for interphase FISH testing

Figure 1. Gestation at amniocentesis

The results of the amniotic fluid FISH analyses are compared with the final karyotype in
Table 2. There were 244 (79%) normal, 50 (16%) abnormal and 15 (5%) inconclusive FISH
19

results. Of the normal FISH results, 234 had a normal G-banded karyotype while ten had
abnormalities not detectable with the probes used. There were no false-positive or
false-negative results. Two unusual FISH results were classified as concordant with the final
karyotype. In the first case, only one signal was detected for the Oncor probe specific for
band 13q32-q33 and the karyotype revealed an unbalanced translocation [46,XX,der(13)
t(10;13)(q26;q32)mat]. The maternal translocation was not known prior to the amniocentesis.
The FISH result for the second unusual case was consistent with trisomy for both
chromosomes 18 and 21. The final karyotype was 47,XY,+der(18),i(21)(q10).

Table 2. Comparison of results from interphase FISH from uncultured amniocytes with
karyotypes from cultured amniocytes
There were 15 inconclusive FISH results (Table 2). Four were subsequently shown by
chromosome culture to be aneuploid. All four were considered suspicious by the laboratory
but, because the results did not meet the scoring criteria, the physician was advised to wait
for the culture result. One of the inconclusive results was due to an extremely rare familial
variation resulting in an X probe signal at the centromere of one of the number 19
chromosomes (Winsor et al., 1999). Inconclusive results were most commonly the result of
an insufficient number of nuclei or weak signals due to technical problems. Almost half of
the inconclusive results (7/15) were from amniotic fluids obtained between 25 and 40 weeks
while only 18% (56/309) of the total number of fluids tested were of this late gestation.
There were ten cases with chromosomal abnormalities that were not detectable by interphase
FISH analysis (Table 3). Nine were clinically significant chromosome abnormalities, while
one had mosaicism for a very small unidentified marker chromosome and no phenotypic
abnormalities were detected in the infant. Overall, 63 of the chromosomal abnormalities were
20

considered clinically significant (54 detectable by interphase FISH and nine not detectable by
interphase FISH). Of these, 50 (79%) chromosomal abnormalities were actually detected by
FISH. For the 13 clinically significant chromosomal abnormalities that were not known until
the culture results were available, four were due to inconclusive FISH results. The remaining
nine abnormalities were undetectable by FISH and provision of a normal FISH result could
be perceived as providing false reassurance.

Table 3. Chromosome abnormalities not detectable by interphase FISH analysis

One amniotic fluid obtained at 33 weeks failed to grow. The interphase FISH result indicated
trisomy 21 and the pregnancy was terminated at another centre. Ultrasound examination
detected short femurs, nuchal thickening and an atrial ventricular septal defect consistent
with the diagnosis of Down syndrome. This case is documented in Table 2 for completeness,
but technically cannot be classified as concordant because no culture result was available.
Although amniotic fluid culture failure is rare, this case illustrates a potential advantage of
testing uncultured nuclei.
The abnormalities detected by interphase FISH are shown in Table 4 according to gestation
and decision regarding pregnancy termination. Of the 50 patients for whom an abnormal
result was reported by FISH analysis, 26 chose to terminate the pregnancy prior to the
availability of the final culture result. Of these, 24/26 had at least one minor or major
ultrasound abnormality. The remaining two patients had amniocentesis at 21 and 23 weeks
because of a positive MSS and no ultrasound abnormalities were recorded. Both had FISH
results indicating trisomy 21. An additional 11/50 patients had pregnancy termination
following confirmation of the abnormality by cultured chromosomes. Three patients tested at
2022 weeks chose to continue their pregnancy to term despite karyotype confirmation of
trisomy 21 FISH results. One patient with a triploid fetus aborted spontaneously following
amniocentesis. The remaining 9/50 patients had amniocentesis at >24 weeks and the
abnormal FISH results allowed for patient management choices, such as not performing
21

Caesarian section for fetal distress.

Table 4. Gestational age at amniocentesis and relationship between interphase FISH results
and pregnancy termination

DISCUSSION
Following the publication of several large studies documenting the accuracy of interphase
FISH studies on uncultured amniotic fluid (Eiben et al., 1999a; Estabrooks et al., 1999; Ward
et al., 1993) the demand from both patients and physicians for this type of testing has
continued to increase. The obvious advantage of FISH is that results are usually available
within 24 h compared with conventional cytogenetic culture which requires 13 weeks.
However, the decision to offer a quicker result instead of the gold standard chromosome
analysis from cultured amniocytes depends on several factors. Since major decisions
regarding pregnancy management may be involved, the risk of false-positive results is of
primary concern. In the present study of 309 patients, there were no false-positive results.
Similarly, no false-positive amniotic fluid interphase FISH results were reported by Jalal et
al.(1998), Eiben et al.(1999a), Estabrooks et al.(1999) or Pergament et al.(2000) using similar
AneuVysion probes (Vysis Inc.) for chromosomes 13, 18, 21, X and Y. The five
false-positive results previously reviewed by Winsor et al.(1999) (Ward et al., 1993;
Verlinsky et al., 1995; Verlinsky et al., 1998; Mercier & Bresson, 1995; Tardy & Toth, 1997)
involved the use of other probes.
Additional considerations include the risk of false-negative or inconclusive results in addition
to chromosomal abnormalities that are not detectable with the probes commonly used for
interphase FISH testing. We found no false-negative results, but approximately 5% were
reported as inconclusive. Eiben et al.(1999a) reported no false-negative results in 3280 cases,
but subsequently reported a case with a normal female FISH result and 45,X in cultured
amniotic fluid cells (Eiben et al., 1999b). Estabrooks et al.(1999) reported three
false-negative results out of approximately 10,000 fluids. The proportion of inconclusive and
22

undetectable abnormalities varies with the criteria for specimen acceptance and reason for
testing (Evans et al., 1999; Morris et al., 1999; Feldman et al., 2000; Lewin et al., 2000).
Although the evaluation of laboratory performance of the interphase FISH testing was one of
our goals, the focus of the present study was to evaluate its role in clinical management and
patient decisions. The gestation of the pregnancy at the time of testing obviously influences
the options available. Of the 50 patients with abnormal FISH results, 26 (52%) chose to
terminate the pregnancy prior to availability of the culture result (Table 4). Eleven additional
patients with abnormal FISH results terminated the pregnancy after the availability of the
culture result. A retrospective review of termination of pregnancy does not fully measure the
impact of the information from interphase FISH testing. Some patients require a longer
period of time to decide whether or not to continue a pregnancy and the early availability of
information could allow this additional time. In some circumstances, delays may have been
due to the availability of consultation services and hospital facilities. One of the limitations in
assessing the impact of FISH results is that decisions are based on a combination of factors,
such as gestation of the pregnancy and abnormalities detected by ultrasound, and not just on
the abnormal FISH results.
Although the clinical decisions for patients with abnormal chromosome results are easier to
document, our informal experience indicates that relief of anxiety provided by normal results
is perhaps even more important than identification of abnormalities. This reduction in anxiety
based specifically on interphase FISH results was formally documented by Tucker
(Pergament et al., 2000; Tucker, 1997). Reduction of maternal anxiety is a major concern
following positive MSS results or reporting of minor abnormalities on ultrasound. Abuelo et
al.(1991) reported that patients with abnormal screening tests become more alarmed about
the possibility of carrying an abnormal fetus than women over 35 years (who had no
screening tests done), even though their statistical risks were similar. Despite the apparent
benefits of an early result based on FISH results, it is important to advise patients that FISH
testing may result in increased anxiety if the FISH assay is inconclusive or if the FISH result
is normal and a chromosomal abnormality not detectable by the assay is subsequently
found.
After the accuracy and limitations of interphase FISH testing of amniotic fluid are
established, the question is how this technique can appropriately be incorporated into clinical
practice. In 1993, the American College of Medical Genetics issued a policy statement
indicating that interphase FISH should be used as an adjunct test only, with standard
chromosome analysis serving as the primary diagnostic and confirmatory evaluation. It was
specifically stated that irreversible therapeutic action should not be initiated on the basis of
23

FISH analysis alone (ACMG, 1993). The ACMG is currently revising this policy (February
2000, www.faseb.org/genetics/acmg).a DAlton et al.(1997) suggested that therapeutic
decisions may be appropriate when FISH on uncultured amniocytes confirms the
interpretation of other findings, such as ultrasonography. Bryndorf et al.(2000) reported that
in their tertiary referral center in Copenhagen, 72% of the terminations of chromosomally
abnormal pregnancies were based on FISH and ultrasound results rather than on conventional
cytogenetic results. Morris et al.(1999) from the UK suggested that owing to the benefits of
interphase FISH and its popularity with both patients and clinicians, it should be offered
routinely, together with standard cytogenetic analysis. Thein et al.(2000), also from the UK,
suggested that FISH may be a satisfactory alternative test to cytogenetics in women
undergoing prenatal diagnosis because of advanced maternal age or abnormal maternal
serum screening when high quality ultrasound imaging is provided. They also noted that a
potential advantage of FISH is that it would not detect clinically insignificant minor
chromosome rearrangements and thus may reduce unnecessary anxiety and even avoid
unnecessary fetal loss. Pergament et al.(2000) reported a study in the USA where 2336/3969
(59%) of women elected to have FISH analysis in addition to routine chromosome analysis.
Costs were covered either by the patient or a third party (insurance scheme). In a publicly
funded system, such as Canada, the issue is largely a question of financial resources. Our
experience suggests that, if they were given the option, most patients and health care workers
would request FISH followed by cultured chromosomes. Because the established standard of
care is complete chromosome analysis, if the patient has an invasive procedure the
expectation is that all known chromosome anomalies would be detected. Thus, we do not
anticipate that a proposal to offer only FISH on amniotic fluid would be acceptable. As
demonstrated by two of the patients in the present study with trisomy 21 by interphase FISH
and confirmed by culture, the absence of abnormal findings by ultrasound does not rule out
the possibility of aneuploidy. In our centre, because of limited staff and the high cost of
commercially available probes, FISH has been offered only to patients with a high suspicion
of aneuploidy and to patients undergoing amniocentesis at a gestation close to the time limit
when pregnancy termination would usually be an option.
The present study indicates that amniotic fluid interphase FISH results can play an important
role in counselling and decision making, especially for those patients referred close to the
time limit when pregnancy termination would be an option. For patients with a high risk of
aneuploidy, the availability of interphase FISH testing has virtually replaced fetal blood
sampling which is technically more demanding and carries a higher risk of pregnancy loss
compared with amniocentesis. Given the extremely small risk of false-positive FISH results
and the fact that we were able to detect 79% of clinically significant chromosomal
abnormalities, we believe it is reasonable to use this information in decision making provided
24

that patients are informed of the benefits and limitations of testing. Our challenge is to find
an economical way to provide rapid results for a broader patient population.

ACKNOWLEDGEMENT
We thank Martha Wood-Burgess for technical expertise.

NOTE
a
ACMG Test and Technology Transfer Committee. 2000. Technical and clinical assessment
of fluorescence in situ hybridization: An ACMG/ASHG position statement. I. Technical
considerations. Genetics in Medicine 2(6): 356361.

25

Chapter 4.1
Application of Rapid Aneuploidy Testing in
Prenatal Diagnosis
Positive Down Screening
Reproduced with permission from Obstet Gynecol:
Leung WC, Lau ET, Lao TT, Tang MH. Can amnio-polymerase chain
reaction alone replace conventional cytogenetic study for women with
positive biochemical screening for fetal Down syndrome? Obstet Gynecol
2003;101:856-61.

26

ABSTRACT
Objective
To determine whether amniopolymerase chain reaction (amnio-PCR) can replace
conventional cytogenetic study for confirming the karyotype of fetuses in women with
positive biochemical screening for fetal Down syndrome.
Methods
To check the accuracy of this technique in our laboratory, we first compared the amnio-PCR
results with those of conventional cytogenetic study in 235 patients referred from June 1999
to December 2001 for prenatal diagnosis in a referral center in Hong Kong. We then
reviewed the results of 1526 amniotic fluid cultures performed for positive fetal Down
syndrome screening between January 1997 and December 2001 and classified them as
detectable or not detectable by amnio-PCR, using the assumption that we had replaced
conventional cytogenetic study with amnio-PCR.
Results
The 235 amnio-PCR results were all informative, without a false-positive or false-negative
result. Of the 1526 cases with positive fetal Down syndrome screening and no ultrasound
abnormalities, only two cases of sex chromosome abnormalities and two cases of marker
chromosomes would have been missed if conventional cytogenetic study had been replaced
by amnio-PCR.
Conclusions
Amnio-PCR can be an alternative to conventional cytogenetic study for women with positive
biochemical screening for fetal Down syndrome and no demonstrable fetal structural
abnormality.

INTRODUCTION
Second-trimester maternal serum screening for fetal Down syndrome has become an
established practice in many countries (Cuckle, 2000). The use and acceptability of screening
with human chorionic gonadotrophin, alpha-fetoprotein, and/or unconjugated estriol has been
demonstrated by numerous large prospective intervention studies (Cuckle, 2000). Serum
screening generates a ratio that characterizes the risk of Down syndrome in the fetus. Women
being assigned a risk above an arbitrary cutoff are designated screen positive and offered
amniocentesis for a karyotype. Conventional cytogenetic study of the cultured amniocytes is
27

performed, and a full karyotype report is available within 2 to 3 weeks. In addition to

chromosome 21, all the other chromosomes are also assessed.
Von Eggeling et al.(1993) devised a polymerase chain reaction (PCR) based test that
allowed rapid detection of trisomy 21 in less than 24 hours. This technique can be applied to
deoxyribonucleic acid (DNA) from uncultured amniotic fluid cells that have been amplified
with small tandem repeat markers using the PCR technique and fluorescence-labeled primers
(amnio-PCR) (Pertl et al., 1994). The small tandem repeat markers specific for chromosome
21 will only be able to detect aneuploidy of chromosome 21. In a masked prospective study
where 2167 pregnant women were offered amniocentesis, Verma et al.(1998) showed that the
rapid test was informative in 99.6% of cases and there were no false-positive or
false-negative diagnoses of Down syndrome.
We have performed amnio-PCR for various indications since 1999 in our Prenatal Diagnostic
and Counselling Department. In all cases, the diagnosis was confirmed by conventional
cytogenetic study. Positive Down syndrome screening is the second most common indication
for amnio-PCR, the first being ultrasound abnormalities.
Our research question is whether amnio-PCR can replace conventional cytogenetic study for
women with only positive biochemical screening for fetal Down syndrome. In other words,
we want to assess the likely outcome if we were to change to a policy of PCR analysis
without conventional cytogenetic study for all of the amniocentesis samples from women
with positive biochemical screening for fetal Down syndrome and no fetal abnormality
detected on ultrasound examination. The potential advantage of this approach is the relief of
womens anxiety during the 2- to 3-week waiting time for the conventional cytogenetic study
report (Leung et al., 2002). The potential disadvantage of this approach is that other
chromosomal abnormalities apart from aneuploidy in chromosome 21 would not be detected.
It would be important to know how many of these undetectable chromosomal abnormalities
are clinically significant if the corresponding ultrasound examination does not show any fetal
abnormality.

MATERIALS AND METHODS

The data in this study were obtained from the database of the Prenatal Diagnostic and
Counselling Department, Tsan Yuk Hospital, which is one of the referral centers for Hong
Kong. We first studied all cases of amnio-PCR performed as a rapid test for prenatal
diagnosis from June 1999 to December 2001. During this period, PCR was performed as an
28

adjunct to conventional cytogenetic study at the discretion of the individual obstetrician.

There was general agreement that amnio-PCR would be offered only if the results were likely
to alter the management of the individual case. For the purpose of the present study, the
indications for amniocentesis were categorized in a hierarchical fashion as follows: 1)
ultrasound abnormalities, 2) positive Down syndrome screening, 3) advanced maternal age, 4)
previous child with chromosomal abnormalities, and 5) others. If one or more ultrasound
abnormalities were reported, the patient was classified as having ultrasound abnormalities
irrespective of the Down syndrome screening result or maternal age. Ultrasound
abnormalities refer to one or more fetal structural abnormalities, soft ultrasound markers, or
evidence of fetal growth restriction (estimated fetal weight less than the tenth centile in local
growth chart). Positive Down syndrome screening refers to an assigned risk of one in 250 or
greater by assay of alphafetoprotein and human chorionic gonadotrophin between 15 and 20
weeks gestation. Advanced maternal age is defined as 35 years or older at the expected date
of delivery.
Amnio-PCR was performed on uncultured amniocytes using small tandem repeat markers for
chromosomes 13, 18, 21, X, and Y, with the choice of the chromosome to be tested
dependent on the indications for amniocentesis. Cells from 2 to 3 mL of amniotic fluid were
used for DNA preparation by a modified alkaline lysis method. Amniocytes were lysed in 20
uL of 0.2-mol/L potassium hydroxide, added with proteinase K (Roche Molecular
Biochemicals, Mannheim, Germany) to a final concentration of 200 ug/mL, and incubated at
65C for 10 minutes. The solution was then neutralized with an equal volume of buffer
containing 0.5-mol/L Tris (pH 8.5) and 0.2-mol/L hydrogen chloride and heated at 95C for
10 minutes. Deoxyribonucleic acid (2 uL) in the supernatant was then used for a single-tube
multiplexed PCR of 10-uL volume using primers from small tandem repeat markers
D21S1411, D21S1412, and D21S1414 (Pertl et al., 1999) labeled with IRD 700 or IRD 800
dye (LI-COR Inc., Lincoln, NE). Electrophoresis was performed on a LICOR 4200 automatic
DNA sequencer, and amplified products were analyzed with GeneImager software (LICOR).
All cases were followed by conventional cytogenetic studies. The cytogenetic results were
classified as those detected or not detected by amnio-PCR.
We reviewed retrospectively the results of all amniotic fluid cultures performed between
January 1997 and December 2001 with the indication of positive Down syndrome screening
from our database. The cytogenetic results were classified as being detectable or undetectable
by amnio-PCR, with the assumption that we were using a policy of PCR analysis not
followed by conventional cytogenetic study for all of the amniocentesis samples from women
with positive biochemical Down syndrome screening and no demonstrable fetal abnormality.

29

RESULTS
We had performed 235 amnio-PCR tests from June 1999 to December 2001. Amnio-PCR
was informative in all cases. Table 1 shows the various indications in a hierarchical fashion,
the number of abnormal karyotypes detected, and whether they were detected or not by
amnio-PCR. The presence of ultrasound abnormalities was the most common indication
(56.2%), followed by positive Down syndrome screening (28.9%), advanced maternal age
(6.8%), previous child with chromosomal abnormalities (all were trisomies) (2.6%), and
others (5.5%). There were altogether 63 abnormal karyotypes. Amnio-PCR could detect 57
of them (90.5%) (Table 2). There were no false-positive or false-negative amnio-PCR results
regarding the diagnosis of aneuploidy in chromosomes 13, 18, 21, X, and Y. Amniotic fluid
culture failure occurred in two cases. The two amniocenteses were performed at 33 and 38
weeks, respectively, both because of ultrasound abnormalities. Amnio-PCR detected trisomy
21 in one case and no evidence of trisomy 13/18/21 in the other.

Table 1. Indications for the 235 amniocenteses (with polymerase chain reaction performed),
number of abnormal karyotypes detected and whether they were detected by polymerase
chain reaction

Table 2. Abnormal karyotypes detected by polymerase chain reaction

There were six cases of abnormal karyotypes not detected by amnio-PCR. None of them
would be detectable by the small tandem repeat markers for chromosome 13, 18, 21, X, or Y
30

used in this study. Four of them had ultrasound abnormalities, and the clinical outcome is
shown in Table 3. Of the two cases of no ultrasound abnormalities (Table 4), one was
balanced Robertsonian translocation inherited from the father, with no clinical significance.
Amniocentesis was performed because the father was a known balanced Robertsonian
translocation carrier. The other case (47,XX,+mar) was clinically significant because the de
novo marker was found to be 15q, involving the PraderWilli/Angelman syndrome
chromosome region. The clinical outcome was a phenotypically normal live birth but with
the potential of having mental and developmental delay. Amniocentesis was performed
because of positive Down syndrome screening. Thus this was the only case that would have
been missed if we had adopted a policy of PCR analysis not followed by conventional
cytogenetic study for the amniotic fluid samples obtained for the indication of positive
biochemical Down syndrome screening.

Table 3. Abnormal karyotypes not detected by polymerase chain reaction, associated with
ultrasound abnormalities (n=4)

Table 4. Abnormal karyotypes not detected by polymerase chain reaction, and not associated
with ultrasound abnormalities (n=2)
The results of the retrospective review of all of the 1526 cases with amniotic fluid cultures
performed between January 1997 and December 2001 with the indication of positive Down
syndrome screening are shown in Table 5. There were 53 cases of trisomy 21 (3.5%), four
cases of trisomy 18 (0.3%), one case of trisomy 13 (0.1%), three cases of sex chromosome
abnormalities (0.2%), and four cases of other chromosomal abnormalities (0.3%).

31

Table 5. Amniotic fluid standard karyotype analysis for positive Down syndrome screening
from 1997 to 2001
If we had adopted the policy of PCR analysis not followed by conventional cytogenetic study
for all of the amniocentesis samples from women with positive biochemical Down syndrome
screening and no demonstrable fetal abnormalities, six cases of chromosomal abnormalities
(six of 65 [9.2%]) would have been missed. The nature of the chromosomal abnormalities
and the clinical outcome of these six cases are shown in Table 6. We had not included a case
of mosaic trisomy 21, 47,XX,+21/ 46,XX(10:24), because technically this could be detected
by amnio-PCR or reported as inconclusive, depending on the level of mosaicism. The six
cases that would have been completely missed by amnio-PCR included two of marker
chromosomes, two of balanced Robertsonian translocation, and two of sex chromosome
abnormalities. One case of marker chromosome (15q) was the one described previously
(Table 4). The other one was a de novo marker of unknown nature with a normal phenotype.
Regarding the two cases of balanced Robertsonian translocation, one was paternal in origin
and one was de novo; both were of no clinical significance. The two cases of sex
chromosome abnormalities had completely different outcomes. For the 47,XXX/45,X(31:10)
case, the parents accepted the diagnosis and continued with the pregnancy. A baby girl was
born at 41 weeks gestation with normal phenotype. For the 47,XXY case, the parents did not
accept their child having Klinefelter syndrome, and termination of pregnancy was performed.
There is another case of sex chromosome abnormality (45,X), but it would not be missed
under the proposed policy because it is associated with hydropic changes on ultrasound
examination.

Table 6. Abnormal karyotypes from amniocentesis for positive Down syndrome screening
that would be missed under the new approach
32

DISCUSSION
Amnio-PCR has been shown to diagnose chromosome aneuploidy accurately (Pertl et al.,
1994; Verma et al., 1998; Levett et al., 2001; Mann et al., 2001). Its major advantage is fast
reporting, within 2448 hours. Relative to the other fast reporting technique of amniotic fluid
interphase fluorescence in situ hybridization, amnio-PCR is less labor intensive. Many
centers are already terminating pregnancy based on abnormal amnio-PCR results without
waiting for the conventional cytogenetic confirmation (Adinolfi & Sherlock, 2001). The
potential use of amnio-PCR as a stand-alone test for indications such as positive serum Down
syndrome screening and advanced maternal age has been suggested (Mann et al., 2001). For
cases of no ultrasound abnormalities, one might worry that chromosomal abnormalities apart
from trisomy 21 would be missed by this approach. However, we must realize that Down
syndrome screening is a program designed to primarily detect Down syndrome. Therefore, in
principle and in practice, follow-up testing with amnio-PCR alone in patients with positive
Down syndrome screening would have realistically fulfilled the expectations of the patients
and obstetricians.
Active research is being done on how to achieve these realistic expectations of patients,
providers, and policy makers for prenatal testing (Marteau, 2002). In fact, all of the screening
methods (age, history, biochemical) are backed up by a high-level ultrasound examination to
look for fetal abnormalities. Therefore, amnio-PCR as a stand-alone test is not literally alone.
In our database of 1526 cases with positive Down syndrome screening, all four cases of
trisomy 18 and one case of trisomy 13 were associated with major ultrasound abnormalities.
Even under the proposed policy, these cases would have been identified because
conventional cytogenetic study would be performed in the presence of ultrasound
abnormalities. Furthermore, they would also be diagnosed by amnio-PCR using small
tandem markers for chromosomes 13 and 18.
The cases that are likely to have been missed by this approach are the sex chromosome
abnormalities, balanced translocations, and marker chromosomes that are not associated with
ultrasound abnormalities. Balanced translocations with normal phenotypes are basically
clinically insignificant, and there is no consensus for the management of the other conditions.
Counseling regarding sex chromosome abnormalities is often difficult. The presence of
marker chromosomes is an even more difficult area for counseling, particularly in the
absence of phenotypic abnormalities (Warburton, 1991). The revelation to the parents of
these chromosomal abnormalities found incidentally in a program designed to primarily
33

detect Down syndrome could lead to unnecessary anxiety and even unwarranted termination
of pregnancy, as parental reactions can vary greatly. The outcome can be unpredictable, with
some parents requesting termination of pregnancy and others accepting the diagnosis, as
illustrated by the two cases in our study.
The adoption of a policy of PCR analysis alone for the amniocentesis samples from women
with only positive biochemical Down syndrome screening has two advantages. The first and
obvious one is the fast reporting. There is a general belief that the fast reporting of normal
amnio-PCR results can relieve parental anxiety while they are awaiting the final report from
the cytogenetic study. However, a recent randomized controlled trial showed that amnio-PCR
did not alleviate anxiety in women who are screen positive for Down syndrome screening
(Leung et al., 2002). One possible explanation is that the woman, although told that the
amnio-PCR result is normal, is still having a significant degree of anxiety while waiting for
the confirmation by the full karyotype report. This anxiety might be alleviated if the
amnio-PCR report is considered to be final. Such an approach will also eliminate the
problems encountered with culture failure and the dilemma of repeat amniocentesis, fetal
blood sampling, or assuming the result to be normal. A prospective study on the anxiety
levels of these women, with one group randomized to having amnio-PCR as a stand alone
test and the other group to having amnio-PCR followed by conventional cytogenetic study, is
being planned in our hospital. The second advantage is cost savings. Instead of adding the
cost of amnio-PCR on top of that of conventional cytogenetic study, the cost of the latter can
be saved by the amnio-PCRalone approach. In this age of ever-escalating costs in the
provision of health care, especially in a government-funded public medical care system as in
Hong Kong, the savings can be redirected to enhance existing programs or fund new ones,
thus maximizing the effect of limited resources.
In conclusion, we think that amnio-PCR can replace conventional cytogenetic study for
women with positive biochemical screening for fetal Down syndrome if ultrasound
examination does not show any fetal abnormality, provided that the center has accumulated
enough experience with the technique of amnio-PCR and is confident of the standard of
ultrasound examination. The risk of missing clinically significant chromosomal abnormalities
is very small. Furthermore, this approach might be more effective in terms of anxiety relief
for women with false positive Down syndrome screening.

34

Chapter 4.2
Application of Rapid Aneuploidy Testing in
Prenatal Diagnosis
Advanced Maternal Age
Reproduced with permission from Hong Kong Med J:
Leung WC, Lau ET, Lau WL, Tang R, Wong SF, Lau TK, Tse KT, Wong
SF, To WK, Ng LK, Lao TT, Tang MH; Working Group on Prenatal
Diagnosis and Counselling, Hospital Authority. Rapid aneuploidy testing
(knowing less) versus traditional karyotyping (knowing more) for
advanced maternal age: what would be missed, who should decide? Hong
Kong Med J 2008;14:6-13.

35

ABSTRACT
Objectives
The application of rapid aneuploidy testing as a stand-alone approach in prenatal diagnosis is
much debated. The major criticism of this targeted approach is that it will not detect other
chromosomal abnormalities that will be picked up by traditional karyotyping. This study
aimed to study the nature of such chromosomal abnormalities and whether parents would
choose to terminate affected pregnancies.
Design
Retrospective study on a cytogenetic database.
Setting
Eight public hospitals in Hong Kong.
Participants
The karyotype results of 19 517 amniotic fluid cultures performed for advanced maternal age
( 35 years) from 1997 to 2002 were classified according to whether they were detectable by
rapid aneuploidy testing. The outcomes of pregnancies with abnormal karyotypes were
reviewed from patient records.
Results
In all, 333 (1.7%) amniotic fluid cultures yielded abnormal karyotypes; 175 (52.6%) of these
were detected by rapid aneuploidy testing, and included trisomy 21 (n=94, 28.2%), trisomy
18 or 13 (n=21, 6.3%), and sex chromosome abnormalities (n=60, 18.0%). The other 158
(47.4%) chromosomal abnormalities were not detectable by rapid aneuploidy testing, of
which 63 (18.9%) were regarded to be of potential clinical significance and 95 (28.5%) of no
clinical significance. Pregnancy outcomes in 327/333 (98.2%) of these patients were
retrieved. In total, 143 (42.9%) of these pregnancies were terminated: 93/94 (98.9%) for
trisomy 21, 20/21 (95.2%) for trisomy 18 or 13, 19/60 (31.7%) for sex chromosome
abnormalities, and 11/63 (17.5%) for other chromosomal abnormalities with potential clinical
significance. There were no terminations in the 95 pregnancies in which karyotyping results
were regarded to be of no clinical significance.
Conclusions
Knowing less by the rapid aneuploidy stand-alone testing could miss about half of all
chromosomal abnormalities detectable by amniocentesis performed for advanced maternal
age. Findings from two fifths of the latter were of potential clinical significance, and the
36

parents chose to terminate one out of six of the corresponding pregnancies. If both techniques
are available, parents could have enhanced autonomy to choose.

INTRODUCTION
The most frequent fetal chromosomal abnormalities involve the autosomes 21, 18, 13, and
sex chromosomes X and Y. Aneuploidy or alterations in copy number of these chromosomes,
including trisomy 21 (Down syndrome), trisomy 18 (Edwards syndrome), trisomy 13
(Pataus syndrome), 45,X (Turners syndrome), and 47,XXY (Klinefelters syndrome)
account for 80% of clinically significant chromosomal abnormalities diagnosed in the
prenatal period. Down syndrome is a well-recognised cause of mental retardation, cardiac,
and other congenital abnormalities. Edwards syndrome and Pataus syndrome lead to
multiple congenital abnormalities and early neonatal death. The phenotype of Turners
syndrome is highly variable and includes short stature, amenorrhoea, infertility, cardiac and
renal malformations. Klinefelters syndrome is associated with a relatively mild phenotype
abnormality.
The traditional method for prenatal diagnosis of these common aneuploidies involves
analysis of banded metaphase chromosomes from cultured amniotic fluid cells
(amniocentesis) or chorionic villi (chorionic villus sampling). It is known as karyotyping, for
which all 23 pairs of chromosomes are examined. Apart from the common aneuploidies, a
wide range of chromosomal abnormalities can thus be identified by this technique, including
rearrangements, such as translocations and inversions that may be balanced or unbalanced.
Traditional karyotyping is labour-intensive and the results are usually not available for 2
weeks or more. Advances in molecular diagnostics, using either fluorescence in situ
hybridisation (FISH) (Ward et al., 1993; Eiben et al., 1999a; Estabrooks et al., 1999;
Pergament et al., 2000; Witters et al., 2002) with chromosome-specific DNA probes or
quantitative fluorescencepolymerase chain reaction (QF-PCR) (Pertl et al., 1994; Verma et
al., 1998; Levett et al., 2001; Cirigliano et al., 2001&2004; Mann et al., 2001) with
chromosome-specific small tandem repeat markers, can be applied to diagnose the common
aneuploidies within 1 to 2 days. The sensitivity and specificity of FISH and QF-PCR,
collectively described as rapid aneuploidy testing (RAT), have been demonstrated in the
aforementioned studies and compare favourably with traditional karyotyping for the
diagnosis of the common aneuploidies. Unlike karyotyping, these technologies only allow
the identification of the chromosomal abnormalities that are specifically sought (targeted
testing).

37

Currently, RAT (FISH or QF-PCR) is being used to give a rapid result for the common
aneuploidies as an adjunct to karyotyping. This combined approach clearly increases the cost
of prenatal diagnosis. It has been suggested that if the indication for prenatal diagnosis is an
increased risk of Down syndrome arising from a positive screening test result or advanced
maternal age, then karyotyping could be replaced by RAT (Mann et al., 2001; Thilaganathan
et al., 2000; Thein et al., 2000; Ryall et al., 2001; Leung et al., 2003; Ogilvie, 2003; Nicolini
et al., 2004). This new approach has been challenged, because certain abnormalities,
although infrequent and usually of debatable significance, would be missed (Witters et al.,
2002; Evans et al., 1999; Lewin et al., 2000; Homer et al., 2003), which has resulted in much
debate in this area of prenatal diagnosis. For example, initiatives by the UK Government (UK
Department of Health, 2003) to replace traditional karyotyping with new screening
programmes involving RAT, FISH and QF-PCR, were objected by the UK Association of
Clinical Cytogeneticists (ACC). Caine et al. (2005) from the ACC undertook a retrospective
cytogenetic audit on 119,528 amniotic fluid and 23,077 chorionic villus samples from 1999
to 2004 to assess the probable clinical impact of these proposed policy changes. They
showed that about 1% of all prenatal samples would have a chromosomal abnormality
undetected by RAT and that a third of these might have a significant risk of serious
phenotypic consequences if RAT was used alone (Caine et al., 2005). However, these
workers had not addressed two important issues relevant to the debate on RAT versus
traditional karyotyping (Leung & Lao, 2005). The first was the role of ultrasound
examination for structural abnormalities in the fetus. A recent prospective study on 1589
amniocenteses samples obtained for various indications showed that 69% (9/13) of clinically
significant chromosomal abnormalities not detectable by RAT had fetal abnormalities
detected by ultrasound (Leung et al., 2004a). Kagan et al. (2007) had recently shown that
more than 98% of all chromosomal abnormalities can be detected if QF-PCR was performed
in all samples and karyotyping in about 16% of the samples selected on the basis of
ultrasound findings before amniocentesis. The second issue is the clinical outcome of
chromosomal abnormalities that are not detectable by the RAT stand-alone approach (Leung
& Lao, 2005). The clinical significance of the latter, particularly those without
ultrasound-detected fetal abnormalities, is very different from that of trisomy 21, 18, or 13
(Ogilvie et al., 2005). Prenatal identification of this group of chromosomal abnormalities
often poses difficult counseling issues, as termination of pregnancy (TOP) may be
unnecessary and not in the best interests of the parents or the fetus. Here we try to identify
the nature of the chromosomal abnormalities from a cohort of pregnancies with
amniocenteses performed for advanced maternal age and whether the parents would choose
TOP.

38

METHODS
This study utilised the database of the Prenatal Diagnostic Laboratory, at Tsan Yuk Hospital,
which receives all the prenatal samples from eight public hospitals in Hong Kong. The
results were sent to the referring hospitals for further action. If necessary, the parents were
counselled on the chromosomal abnormalities, mainly by obstetricians in the individual
hospital, and in some cases with the help of paediatricians and geneticists. In this database,
the indications for amniocentesis were categorized in a hierarchical fashion as follows: (1)
ultrasound detected fetal abnormalities, (2) positive Down syndrome screening, (3) advanced
maternal age ( 35 years old), and (4) other indications. If one or more ultrasound-detected
fetal abnormalities were reported, the case was classified as having ultrasound detected
abnormalities (1); this was irrespective of the Down syndrome screening result or maternal
age. If amniocentesis was performed for positive Down screening in a woman aged 35 years
or above, the case was classified as (2). We retrospectively reviewed the results of 19,517
amniotic fluid cultures performed specifically for advanced maternal age (3) from 1997 to
2002. Within the same period, there were 1061 amniotic fluid cultures performed for
ultrasound-detected fetal abnormalities, 1629 for Down syndrome screening and 2039
cultures for other indications.
The results were categorised into normal and abnormal karyotypes. Abnormal karyotypes
were divided into common aneuplodies of chromosomes 21, 18, 13, X and Y, and other
chromosomal abnormalities. The latter were further subdivided into a group with potential
clinical significance (de-novo balanced translocations and chromosomal rearrangements,
unbalanced translocations and chromosomal rearrangements, uncommon autosomal trisomies,
de-novo markers) and another group with no clinical significance (balanced
translocations/chromosomal rearrangements/markers of familial origin, de-novo balanced
Robertsonian translocations). A second classification was performed on whether the results
were detectable by RAT. The outcomes of these pregnancies with abnormal karyotypes were
reviewed from patient records in each hospital, through our Working Group on Prenatal
Diagnosis and Counselling.

RESULTS
The results of karyotyping of 19,517 amniotic fluid cultures performed for advanced
maternal age from 1997 to 2002 are shown in Figure 1. There were 333 (1.7%) abnormal
karyotypes. Among these, 175 (53%) were common aneuploidies detectable by RAT. The
other 158 (47%) were not detectable by RAT. The latter were further subdivided into 63
39

(19%) with potential clinical significance and 95 (29%) with no clinical significance.

Figure 1. Traditional karyotyping results of amniocenteses performed for advanced maternal

age from 1997 to 2002 (RAT denotes rapid aneuploidy testing)
We obtained the pregnancy outcomes in 327/333 (98.2%) of these patients (Tables 1-3). In
total, 143 (43%) pregnancies were terminated; 93/94 (99%) trisomy 21, 20/21 (95%) trisomy
18 or 13, 19/60 (32%) sex chromosome abnormalities, 11/63 (18%) other chromosomal
abnormalities with potential clinical significance, and 0/95 (0%) with no clinical
significance.

40

Table 1. Pregnancy outcomes of common aneuploidies detectable by rapid aneuploidy testing

(RAT)

Table 2. Pregnancy outcomes of chromosomal abnormalities with potential clinical

significance not detectable by rapid aneuploidy testing

Table 3. Pregnancy outcomes of chromosomal abnormalities with no clinical significance not

detectable by rapid aneuploidy testing
Terminations of pregnancy were separated into two groups. In group 1, they were performed
due to major chromosomal abnormalities or major ultrasound-detected structural fetal
abnormalities. Major chromosomal abnormalities referred to trisomy 21, 18 and 13 (Table 1),
41

and one case of 5p- or cri-du-chat syndrome (Table 2). Four others of potential clinical
significance were terminated because of structural fetal abnormalities detected subsequently
by ultrasound examination. They included cleft lip and palate, micrognathia, clinodactyly,
microcephaly, aortic stenosis (Table 2). In group 2, although the chromosomal abnormalities
were not major and no ultrasound-detected fetal abnormalities were present, the parents
could not accept the uncertainty in clinical outcome, which varied from normal to a degree of
mental impairment and/or physical abnormality that might not even be evident at birth.
Certain sex chromosome abnormalities were also considered under this category.
Table 4 shows the details of the 143 cases with chromosomal abnormalities among those
undergoing TOP, of which 25 (18%) belonged to group 2.

Table 4. Chromosomal abnormalities with termination of pregnancy (TOP)

DISCUSSION
The clinical application of RAT as a stand-alone approach in prenatal diagnosis is subject to
much debate. The pros and cons of this approach have been discussed in recent review
42

articles (Nicolini et al., 2004; Shaffer & Bui, 2007; Hulten et al., 2003; Mann et al., 2004;
Leung et al., 2004b; Dudarewicz et al., 2005). The major criticism is that such targeted
testing would miss the diagnosis of certain chromosomal abnormalities that will be picked up
by traditional karyotyping. The counter argument is that 60% of these abnormalities are not
clinically significant, and the other 40% are only of potential clinical significance (excluding
those with major ultrasound-detected structural abnormalities in the fetus) (Leung & Lao,
2005).
The feasibility of the RAT stand-alone approach depends on the indication for the invasive
prenatal test. With the presence of major ultrasound-detected fetal abnormalities, traditional
karyotyping should be performed to look for structural chromosomal abnormalities apart
from aneuploidies (Thein et al., 2000; Leung et al., 2004a; Leung et al., 2001). The RAT
stand-alone approach is best when the invasive prenatal test is performed for an identified
increased risk of Down syndrome from a positive Down screening test. We have previously
demonstrated the feasibility of this approach in 1526 cases, with amniocenteses performed
for positive biochemical Down screening (Leung et al., 2003). For the present study, we
chose the much larger dataset of amniocenteses (19,517 cases) performed specifically for
advanced maternal age, focusing on the pregnancy outcomes of the 333 cases with
chromosomal abnormalities.
As expected, the great majority (98.3%) of amniocenteses performed for advanced maternal
age showed normal results (Figure 1). For this large group of parents, RAT could exclude the
possibility of fetal Down syndrome and relieve anxiety within 1 to 2 days of amniocentesis
(Leung et al., 2008b). To supplement this fast report with the traditional karyotyping whose
results become available in 2 to 3 weeks seems unnecessary.
The pregnancy outcomes of the various common aneuploidies detectable by RAT could be
very different (Table 1). Essentially, all cases with non-mosaic trisomies 21, 18 and 13 were
terminated. If RAT was performed, the decision to terminate the pregnancies could be made
2 to 3 weeks earlier (Aldinolfi & Sherlock, 2001). On the other hand, sex chromosome
abnormalities pertained to another issue (Table 1). None of them had major
ultrasound-detected fetal abnormalities in this dataset of amniocenteses performed for
advanced maternal age. Two thirds of the parents decided to continue with the pregnancies,
which resulted in livebirths with no morphological abnormality at birth, while the other third
decided to terminate the pregnancies. Not surprisingly the prognosis of persons with sex
chromosome abnormalities is very different from that of trisomies 21, 18 and 13. Thus, some
obstetricians, clinical geneticists, and genetic counsellors are uneasy about testing and
reporting the sex chromosome status of all fetuses undergoing invasive prenatal tests
43

(Abramsky et al., 2001; Sagi et al., 2001; Donaghue et al., 2003). Identification of sex
chromosome abnormalities such as XXX, XYY and XXY (Klinefelters syndrome), that are
either asymptomatic or associated with a relatively mild phenotype, often poses difficult
counselling issues and may not be in the best interests of the parents. Such findings tend to
increase parental anxiety and present a difficult choice regarding the continuation of the
pregnancy. On the other hand, the Turners syndrome (45,X) phenotype is highly variable
with respect to short stature, amenorrhoea, infertility, cardiac malformations (coarctation of
aorta) and renal complications (Brock et al., 1992). Besides, up to 99% of fetuses with
Turners syndrome are miscarried during the first and second trimester of pregnancy (Hook,
1978), and those that do not miscarry usually have ultrasound abnormalities (Huang et al.,
2002). Donaghue et al. (2003) has proposed a selective policy for fetal sexing if RAT is to be
used as a stand-alone test. Aneuploidies of X and Y chromosomes will be determined by
RAT only in cases displaying ultrasound abnormalities consistent with Turners syndrome
and those at risk of inheriting a sex-linked disorder. The ultrasound findings in Turners
syndrome include: cystic hygroma, nuchal thickening of 5 mm or more, adjusted nuchal risk
of 1:5 or higher, hydrops, nuchal oedema or coarctation of the aorta (Donaghue et al., 2003).
This targeting policy for sex chromosome tests may avoid the unintentional finding of
conditions of borderline significance, such as XXX, XYY and XXY, during prenatal testing
for Down syndrome.
When amniocenteses were performed for advanced maternal age, chromosomal
abnormalities not detectable by RAT (Tables 2 and 3) were unexpected by the couples as
well as the obstetricians (van Zwieten et al., 2005). They would have been excluded from
prenatal testing, if RAT were applied as a stand-alone test in the absence of
ultrasound-detected structural fetal abnormalities. Some of them had potential clinical
significance (Table 2) while others had no clinical significance (Table 3). For those
chromosomal abnormalities with potential clinical significance (Table 2), 49 couples decided
to continue with the pregnancies. Eleven couples decided to terminate the pregnancies: four
owing to the presence of fetal abnormalities (cleft lip and palate, micrognathia, clinodactyly,
microcephaly, aortic stenosis), which were discovered during subsequent ultrasound
examinations after the chromosomal abnormalities were identified (i.e. group 1, Table 2).
One termination was for 5p- (cri-du-chat) syndrome and the other six because the parents
could not accept the uncertainty in clinical outcome varying from normal to a certain degree
of mental impairment and physical abnormalities, which might not be diagnosed even at birth
(i.e. group 2, Table 2) (Warburton, 1991). So, what might ensue if traditional karyotyping is
replaced by RAT? The 5p- syndrome would be missed. The other six chromosomal
abnormalities with potential clinical significance would not be diagnosed and the respective
parents would not have the chance to consider TOP.
44

For chromosomal abnormalities with no clinical significance (Table 3), karyotyping provided
no useful additional information (except that the finding of a familial translocation will have
implications for future pregnancies). Instead it led to additional counseling time to convey
information and to relieve parental anxiety. None of these 95 cases had TOP.
Overall, 143 pregnancies were terminated, 25 (17.5%) of them belonged to the group in
which the chromosomal abnormalities were not major (19 sex chromosome abnormalities
and six chromosomal abnormalities with potential clinical significance) and there were no
major ultrasound-detected structural fetal abnormalities (Table 4). If RAT (for trisomies 21,
18 and 13 only) had been used as a stand-alone approach, it could have been coupled with
RAT testing for sex chromosome aneuploidies only when ultrasound-detected abnormalities
were consistent with Turners syndrome (Donaghue et al., 2003). In which case, traditional
karyotyping would only be needed when there were major ultrasound-detected structural
fetal abnormalities (Thein et al., 2000; Leung et al., 2004a; Leung et al., 2001), and
consequently the 25 cases without major abnormalities would not be diagnosed and the
parents would not have the option to terminate these pregnancies. One might argue that this
is a beneficence-based approach (the physician making decisions that are best for the patient,
without regard to personal gain or the interests of others). Some of these 25 TOP were
probably unnecessary because the chromosomal abnormalities are only of potential clinical
significance, particularly in the absence of ultrasound-detected fetal abnormalities. On the
other hand, the ethics of prenatal diagnosis should be autonomy-based (the capacity of a
rational individual to make an informed, uncoerced decision; in medicine, respect for the
autonomy of patients is considered obligatory for doctors and other health care professionals).
If the parents choose to have TOP for minor chromosomal abnormalities, we have the
obligation to make sure that the counselling is thorough, to ensure that they have the
information about the outcomes in order to exercise their autonomy to continue or terminate
the pregnancy. If after that process, the parents still choose to terminate the pregnancy, then
their decision and autonomy must be respected. A medical attitude of we will not look for
sex chromosome abnormalities and other minor chromosomal abnormalities as we do not
consider you should have a termination could be considered paternalistic (a figurehead that
makes decisions on behalf of others for their own good, even if this is contrary to their
wishes). If both techniques are available, parents should have the autonomy to choose the
approach (RAT, or traditional karyotyping, or both) after being fully informed of the pros
and cons. Bui (2007) recently reported their experience in Stockholm, when Swedish women
were given this choice, 70% of them chose the RAT stand-alone approach (Bui, 2007). It
would be interesting to study the parental preference in our Chinese population, particularly
in Mainland China, which operates a one-child policy.
45

ACKNOWLEDGEMENT
We would like to thank Prof. Frank Chervenak, Weill Medical College of Cornell University,
New York, for his comments on the manuscript.

46

Chapter 4.3
Application of Rapid Aneuploidy Testing in
Prenatal Diagnosis
Abnormal Ultrasound Findings
Reproduced with permission from Prenat Diagn:

Leung WC, Waters JJ, Chitty L. Prenatal diagnosis by rapid aneuploidy

detection and karyotyping: a prospective study of the role of ultrasound in
1589 second-trimester amniocenteses. Prenat Diagn 2004;24:790-5.

47

ABSTRACT
Background
Reliable methods are available for rapid aneuploidy testing (RAT) for the prenatal diagnosis of
trisomies 21, 18 and 13. This study examines the potential advantages and limitations of using
RAT as a replacement rather than as an adjunct to traditional karyotyping.
Methods
One thousand five hundred and eighty-nine consecutive pregnancies referred for cytogenetic
assessment were offered RAT (FISH or QF-PCR) as an adjunct to traditional karyotyping. The
results of these two processes were compared, and the effects of three policies for cytogenetic
evaluation were determined: RAT alone, a combination of RAT for all and traditional
karyotyping for cases with ultrasound anomalies or a policy of RAT and traditional
karyotyping in all cases.
Results
RAT was uninformative because of maternal-cell contamination in 37 (2.3%) cases compared
to 4 (0.3%) cases of culture failure in traditional karyotyping. RAT and traditional karyotyping
results were concordant in 1526 of 1548 (98.6%) cases. All non-mosaic cases of trisomies 21,
18 and 13 and cases of triploidy were correctly identified by RAT, and there were no
false-positive diagnoses. The gold standard of a traditional karyotype in each case would have
detected all chromosomal abnormalities. A policy of RAT alone would have identified 60 of
73 (82%) clinically important chromosomal abnormalities. The addition of a full karyotype for
cases with evidence of ultrasound abnormalities would have improved detection to 95%.
Conclusions
A policy offering RAT to all patients, but restricting traditional karyotyping to cases with
ultrasound anomalies, would reduce the number of traditional karyotypes requested by 70%,
but maintain a 95% detection rate for all clinically important chromosomal abnormalities.
Further studies are required to determine whether similar results could be obtained in district
general hospital units and to determine whether this approach would be acceptable to health
professionals and patients.

INTRODUCTION
Amniocentesis allows prenatal diagnosis of fetal chromosomal abnormalities during
pregnancy. It is commonly restricted to pregnancies considered to be at high riskas a result
48

of prenatal biochemical and ultrasound screening tests, due to advanced maternal age or due to
a history of a parental balanced chromosomal rearrangement. The traditional gold standard
method for detection of fetal chromosomal abnormality involves analysis of banded
metaphase chromosomes from cultured cells (karyotyping). This technique identifies a wide
range of chromosomal abnormalities, including alterations in copy number (aneuploidy) as
well as chromosomal rearrangements such as translocations and inversions that may be
balanced or unbalanced. However, the technique is labour-intensive and results are not usually
available for two weeks or more.
Advances in molecular genetics, using either fluorescence in situ hybridization (FISH) (Ward
et al., 1993) or quantitative fluorescence-polymerase chain reaction (QF-PCR) (Pertl et al.,
1994), can be applied to give karyotype results within one or two days. The sensitivity and
specificity of these tests, collectively described in this article as rapid aneuploidy testing
(RAT), have been demonstrated in a number of studies either using FISH (Eiben et al., 1999a;
Estabrooks et al., 1999; Pergament et al., 2000; Witters et al., 2002) or QF-PCR (Verma et al.,
1998; Levett et al., 2001; Mann et al., 2001). They compare favourably with traditional
karyotyping for the diagnosis of the most frequent, clinically important aneuploidies (namely
trisomies 21, 13 and 18) (Table 1). These technologies will, however, only allow the
identification of the chromosomal rearrangements that are specifically being sought. If the data
from the published studies (Witters et al., 2002; Evans et al., 1999; Lewin et al., 2000; Homer
et al., 2003; Thein et al., 2000; Ryall et al., 2001; Leung et al., 2003; Leung et al., 2001;
Thilaganathan et al., 2000; Grimshaw et al., 2003) is pooled, then overall 0.9% of other
rearrangements are not detected (Table 1), although it is not possible to estimate the proportion
that carried a risk of adverse outcome. The views on whether RAT should be offered alone or
not were mixed, with several authors expressing the view that RAT would miss too many
significant rearrangements (Witters et al., 2002; Evans et al., 1999; Lewin et al., 2000;
Homer et al., 2003).

49

Table 1. Comparison of rapid aneuploidy testing (RAT) and traditional karyotyping results in
the literature
In the United Kingdom, RAT is increasingly being used to give a preliminary rapid result for
the major trisomies. This is in addition to traditional karyotyping, and clearly increases the
cost to the service provider. If the indication for cytogenetic analysis is an increased risk of
Down syndrome, then conventional karyotyping could be effectively replaced by RAD
(Mann et al., 2001; Thein et al, 2000; Ryall et al., 2001; Leung et al., 2003). This would,
however, miss a small number of chromosome abnormalities that are not detectable by these
techniques (Evans et al., 1999; Lewin et al., 2000). This prospective study examines the
hypothesis that karyotypic examination can reliably be limited to RAT in pregnancies with
normal ultrasound findings prior to amniocentesis. The relevance of ultrasound findings when
karyotyping is performed in the first trimester will be addressed in a separate study.

METHODS
All women undergoing amniocenteses in a tertiary fetal medicine unit between 1st April
2000 and 31st March 2003 were offered rapid aneuploidy testing (FISH or QF-PCR) as well
as traditional karyotyping. FISH, using DNA probes specific for chromosomes 13, 18, 21, X
and Y according to the manufacturers instructions (Vysis (UK) Ltd), was used in the first
year of the study. QFPCR, using small tandem repeat markers for chromosomes 13, 18 and
21 (see Levett et al. (2001) and Mann et al. (2001) for methodology), was used for the
remaining two years. Traditional karyotyping was performed using standard methods
(Rooney, 2001). As markers for sex chromosomes were not used in the later cohort, these
50

chromosomal abnormalities have been excluded from subsequent analysis. Gestational age,
indication for testing and details of ultrasound examinations were recorded. Indications
included a Down syndrome risk of 1:250 on maternal serum screening (MSS), advanced
maternal age (AMA) 35 years at the estimated date of delivery, parental translocation
carrier and the presence of anomalies on ultrasound examination. Positive ultrasound
findings included major structural abnormalities and markers of chromosomal abnormality
such as increased nuchal translucency (1114 weeks), nuchal oedema, echogenic bowel,
dilated renal pelves, choroid plexus cysts, intracardiac echogenic focus, clinodactyly and
sandal gap.
Cytogenetic results were classified as being normal or abnormal. Abnormal results were
subdivided into those that were clinically significant or insignificant (Table 2). The results of
RAT and traditional karyotyping were compared by indication. Statistical analysis was
performed using the SPSS/PC 10.0 software package. A chi-squared test was used to compare
the number of clinically significant chromosomal abnormalities that could not be detected by
RAT for each indication. A probability of p < 0.05 was considered statistically significant.

Table 2. Comparison of rapid aneuploidy testing and karyotyping results within each
indication group

RESULTS
Traditional karyotyping was performed on 1589 amniotic fluid samples. Of these, 521 samples
also underwent FISH and 1068 had QF-PCR for RAT. Rapid testing was uninformative in 37
(2.3%) samples due to presumptive maternal-cell contamination. FISH failed (22/521 = 4.2%)
more commonly than QF-PCR (15/1068 = 1.4%). Traditional karyotyping was normal in
35/37 (94.6%) of these, but was abnormal in the remaining two cases (trisomy 21 and
69,XXY). The traditional karyotype was not available in four samples (0.3%) due to culture

51

failure; all four cases had a normal RAT result. Rapid aneuploidy detection, and traditional
karyotyping results were available for comparison in 1548 samples (97.4%).

Table 3. Concordant abnormal results with rapid aneuploidy testing and karyotyping
Of the 1548 cases, 1526 (98.6%) had concordant results using RAT and traditional karyotype
techniques. There were no false-positive or false-negative RAT results for trisomies 21, 18 or
13 or for triploidy (Table 3). In the remaining 22 cases, traditional karyotyping revealed a
chromosomal rearrangement that was not found using the RAT technique (Tables 2 and 4).
RAT identified 60 of the 82 (73%) chromosomal abnormalities found, including 60 of the 73
(82%) abnormalities that had a risk of adverse clinical outcome (Table 2). Of the 22 discordant
results, 12 were karyotyped because of suspicious ultrasound findings, 5 because of positive
MSS, 4 on the basis of AMA and one while undergoing an invasive procedure for Rhesus
disease (Table 4). Of these 22 cases, 13 had clinically significant chromosomal abnormalities;
9 in the group karyotyped because of an ultrasound finding, 2 in the MSS group and 2 for
AMA, although one case in the MSS group was a balanced de novo rearrangement with a low
risk of adverse outcome. Ultrasound assessment was repeated in the four latter cases once the
abnormal karyotype had been reported, and was found to be normal on each occasion.
The prevalence of clinically important chromosomal rearrangements that were not detected by
RAT was significantly higher (p < 0.01) in fetuses karyotyped on the basis of abnormal
ultrasound findings (9 in 390) than for other indications (MSS, 2 in 629; AMA, 2 in 450). A
prenatal cytogenetic evaluation policy limited to RAT detects 60 of 73 (82%) of clinically
important chromosomal abnormalities. If cases with visible ultrasound anomalies are also
offered a traditional karyotype, the detection rate improves to 69 of 73 (95%), an improvement
in detection of 13%. A policy of RAT with a traditional karyotype for fetuses with visible
anomalies at the time of ultrasound examination will miss 5% of clinically significant
anomalies that would have been detected if a traditional karyotype had been offered in all
cases. In our cohort, a total of 485 fetuses (31.3%) had an ultrasound abnormality. In 390 cases,
ultrasound findings were the primary indication for karyotyping, and sonographic anomalies

52

were identified in 95 of the fetuses undergoing amniocentesis based on an increased risk on

MSS or AMA.

Table 4. Discordant results (normal rapid aneuploidy detection; abnormal karyotype)

53

DISCUSSION
This study confirms the accuracy of RAT for the prenatal diagnosis of non-mosaic trisomy 21,
18, 13 and triploidy, whether by FISH or QF-PCR. As a result, in common with other centres,
we advise parents of the results of RAT and offer further management as indicated, without
waiting for traditional karyotype confirmation (Adinolfi & Sherlock, 2001). This provides
the majority of patients with early reassurance, an advantage that is lost with a two-phase
reporting procedure (Leung et al., 2002). A quick result also gives parents more time to make
decisions, with an earlier option for termination of pregnancy, in the event of an abnormal
result. Limiting prenatal cytogenetics to these techniques would provide accurate
karyotyping information for the chromosomal abnormalities that are described to patients
undergoing screening and reduce the cost of prenatal diagnosis.
This study also shows that limiting karyotyping to a RAT technique would result in an 18%
reduction in the number of clinically significant chromosomal abnormalities detected.
Clinically significant abnormalities include unbalanced chromosomal rearrangements, de
novo balanced chromosomal rearrangements, marker chromosomes and rare aneuploidies
(Cases 1 to 13, Table 4). Of these 13 cases, 9 (69%) also had evidence of a major structural
defect or marker of aneuploidy on ultrasound examination, an association that has previously
been noted by other groups (Thein et al., 2000; Leung et al., 2001). These data suggest that,
for fetuses found to be at increased risk of aneuploidy, if full karyotyping is reserved for
those cases with ultrasound anomalies visible at the time of invasive testing, 95% of all
clinically important chromosomal abnormalities will be detected.
A potential problem of limiting cytogenetic evaluation to RAT is that 2.3% of samples were
uninformative (usually because of evidence of presumptive maternal cell contamination),
compared to 0.3% using traditional techniques. Any policy based on RAT alone should
include the facility to store amniotic fluid or to initiate cell culture, at least until the RAT
result is available. During the course of the study, the RAT technique employed was changed
from FISH (uninformative in 4.2% of cases) to QF-PCR (uninformative in 1.4% of cases).
As experience of QF-PCR has increased, there has been a further reduction in uninformative
results to 0.8% of cases (as shown in an audit of results from the last six months from our
Cytogenetics laboratory).
In addition to the chromosomal abnormalities described as being clinically important, there
was a group of nine cases with a familial, balanced translocation that would have been

54

missed with a policy of RAT alone. Although these are not significant in this pregnancy, they
have the potential to result in unbalanced products in future pregnancies (Homer et al., 2003).
The recent UK Government white paper on Genetics stated that the results of prenatal tests
should be available within three days (Our Inheritance, Our Future: Realising the potential
of genetics in the NHS White paper; HMSO, 2003). A policy of limiting prenatal
cytogenetics to RAT, and a full karyotype in the presence of ultrasound anomalies would
have achieved this in 70% of our cases while maintaining a 95% detection rate for clinically
important chromosomal abnormalities. These results should be interpreted with some caution,
as some fetuses with ultrasound anomalies were karyotyped after an anomaly scan at 20
weeks, and these anomalies may not always be as obvious if ultrasound examination is
performed at 16 weeks prior to amniocentesis for other indications such as serum screening.
The acceptability of this approach for both health professionals and their patients needs to be
determined before it can be employed in clinical practice.

ACKNOWLEDGEMENTS
We thank our colleagues at the Cytogenetics Laboratory, NE Thames Genetics Service, for
performing FISH and karyotype analyses and the staff at the Cytogenetics Centre, SE Thames
Genetics Service and Cytogenetic DNA Services Ltd., for performing QF-PCR analyses and
Jon Hyett for his helpful comments on the manuscript. Wing Cheong Leung was supported by
a Scholarship from the Ho Hung-Chiu Medical Education Foundation, Hong Kong SAR.

55

Chapter 4.4
Application of Rapid Aneuploidy Testing in
Prenatal Diagnosis
Other Indications Thalassaemia
Reproduced with permission from Mol Hum Reprod:
Tse KY, Leung WC, Leung KY, Lee CP, Ng LK, Lau ET, Vivian Chan,
MHY Tang. Full karyotyping, rapid aneuploidy diagnosis or both when
invasive prenatal testing is performed for diagnosis of thalassaemia? Mol
Hum Reprod 2006;12:55-9.

56

ABSTRACT
A retrospective study was performed to compare the detection rate of chromosomal
abnormalities by different approaches of full karyotyping, rapid aneuploidy testing (RAT) or
both when invasive prenatal testing is performed for diagnosis of thalassaemia. The
karyotype results of 1120 prenatal samples obtained from thalassaemia couples from January
1985 to December 2002 in a referral centre for prenatal diagnosis were studied. The detection
rate of chromosomal abnormalities by four different approaches were compared: (i)
karyotyping for all samples; (ii) RAT (21,18,13,X,Y) for all samples; (iii) RAT for all
samples + karyotyping for cases with ultrasound abnormalities; and (iv) RAT (21,18,13) for
all + RAT (X,Y) for cases with ultrasound abnormalities consistent with Turner syndrome +
karyotyping for cases with ultrasound abnormalities. Normal karyotypes were found in 1103
samples (98.5%). There were 17 cases (1.5%) of chromosomal abnormalities: four cases
(0.36%) were clinically significant, eight cases (0.7%) were of borderline clinical
significance and five cases (0.44%) were not confirmed by subsequent prenatal or postnatal
tests. The incidences of autosomal (7/1120 = 0.63%) and sex chromosomal (5/1120 = 0.45%)
abnormalities were not higher than those (0.41% and 0.22%, respectively) from newborn
surveys (Hook & Hamerton, 1977) (P = 0.398 and 0.216, respectively). Approach 1 would
detect all 17 chromosomal abnormalities. Approach 2 would detect three of four clinically
significant chromosomal abnormalities but not detect six of eight chromosomal abnormalities
of borderline clinical significance and three of five chromosomal abnormalities not
confirmed by subsequent prenatal or postnatal tests. Approach 3, in addition, would be able
to detect all four clinically significant chromosomal abnormalities. Approach 4 would detect
all four clinically significant chromosomal abnormalities but would not detect seven of eight
chromosomal abnormalities of borderline clinical significance and four of five chromosomal
abnormalities not confirmed by subsequent prenatal or postnatal tests. RAT (21,18,13) for all
+ RAT (X,Y) for cases with ultrasound abnormalities consistent with Turner syndrome +
karyotyping for cases with ultrasound abnormalities seemed to be the best approach for the
detection of chromosomal abnormalities when invasive prenatal testing is performed for
diagnosis of thalassaemia.

INTRODUCTION
Thalassaemia is common in Hong Kong. About 4.3% of the antenatal population have
-thalassaemia trait and 2.8% have -thalassaemia trait (Chan et al., 1997; Sin et al., 2000).
Couples with -thalassaemia or -thalassaemia trait carry 25% chance of having a fetus with
either homozygous 0-thalassaemia or -thalassaemia major, respectively. Fetuses with
57

homozygous 0-thalassaemia are not compatible with life and usually die in-utero. On the
other hand, children with -thalassaemia major require blood transfusion throughout life that
carries long-term morbidity. Prenatal screening and diagnosis have been used to prevent the
birth of infants with these severe forms of thalassaemia (Beris et al., 1995; Leung et al.,
2004). Prenatal diagnostic tests for thalassaemia couples include chorionic villus sampling,
amniocentesis and cordocentesis to obtain samples for fetal DNA analysis (chorionic villi,
amniotic fluid) or haemoglobin analysis (fetal blood). Although there is no evidence from the
literature that the prevalence of common aneuploidies or other chromosomal abnormalities is
higher in fetuses of thalassaemia couples, karyotyping is also performed in our practice in
addition to the DNA analysis for thalassaemia.
Karyotyping or conventional cytogenetic study refers to the analysis of banded metaphase
chromosomes from cultured cells. This technique identifies a wide range of chromosomal
abnormalities, including alterations in copy number (aneuploidy) as well as chromosomal
rearrangements, such as translocations and inversions that may be balanced or unbalanced.
This technique is labour intensive. For amniotic fluid and chorionic villi cultured cells,
results are not usually available in 2 weeks or more. Advances in molecular diagnostics,
using either fluorescence in-situ hybridization (FISH) with specific DNA probes (Ward et al.,
1993) or quantitative fluorescencePCR (QFPCR) with specific small tandem repeat
markers (Mansfield, 1993; Pertl et al., 1994), can be applied to diagnose the common
aneuploidies of chromosomes 21, 18, 13, X and Y within 12 days. The sensitivity and
specificity of these tests, collectively known as rapid aneuploidy testing (RAT), have been
demonstrated in a number of large-scale studies either using FISH (Eiben et al., 1999a;
Estabrooks et al., 1999; Pergament et al., 2000; Witters et al., 2002) or QFPCR (Verma et
al., 1998; Levett et al., 2001; Mann et al., 2001; Cirigliano et al., 2004). They compare
favourably with karyotyping for the diagnosis of the most frequent and clinically important
aneuploidies (trisomies 21, 18 and 13) as well as sex chromosome aneuploidies. These
technologies will miss chromosomal abnormalities such as balanced de novo translocations
and marker chromosomes (Leung et al., 2004b; Caine et al., 2005; Ogilvie et al., 2005). It has
been suggested that if the indication for karyotyping is an increased risk of Downs syndrome,
such as positive screening test result or advanced maternal age, karyotyping could be
effectively replaced by RAT (FISH or QF-PCR), provided there is no structural fetal
abnormality detected on ultrasound examination (Ogilvie et al., 2005; Thein et al., 2000;
Thilaganathan et al., 2000; Ryall et al., 2001; Leung et al., 2003; Leung & Lao, 2005;
Ogilvie, 2003). The approach of using RAT as a stand-alone test instead of karyotyping when
invasive prenatal testing is performed for diagnosis of thalassaemia has not been studied.
The aim of our study was to compare the detection rate of chromosomal abnormalities, when
58

invasive prenatal testing is performed for diagnosis of thalassaemia, by four different

approaches: (i) karyotyping for all samples; (ii) RAT (21,18,13,X,Y) for all samples; (iii)
RAT for all samples + karyotyping for cases with ultrasound abnormalities; and (iv) RAT
(21,18,13) for all + RAT (X,Y) for cases with ultrasound abnormalities consistent with
Turner syndrome + karyotyping for cases with ultrasound abnormalities.

METHODS
The data in this study were obtained from the database of the Prenatal Diagnostic and
Counselling Department, Tsan Yuk Hospital, which is one of the referral centres for prenatal
diagnosis in Hong Kong. The methods for prenatal screening and diagnosis for thalassaemia
in Hong Kong has been described in a recent paper on cost-effectiveness of the programme
(Leung et al., 2004). DNA analysis for thalassaemia was performed in the DNA diagnosis
laboratory of the Department of Medicine, University of Hong Kong. Karyotyping was
performed in the Prenatal Diagnostic Laboratory of Tsan Yuk Hospital.
We reviewed retrospectively the karyotype results of all prenatal samples obtained from
thalassaemia couples from January 1985 to December 2002. The maternal age and the type
of thalassaemia were recorded. The prevalence of common aneuploidies or other
chromosomal abnormalities was compared to those published in the literature (Hook &
Hamerton, 1977). The karyotype results were classified as detectable or not detectable by
RAT for the common aneuploidies of chromosomes 21, 18, 13, X and Y. The presence of
prenatal ultrasound fetal abnormalities and the clinical outcome of those cases with
chromosomal abnormalities were retrieved from hospital records either from our obstetric
unit or other referring obstetric units.
Statistical analysis was performed using the SPSS/PC 11.5 software package. The categorical
variables were compared using chi-square test or Fishers exact test as appropriate. A P-value
< 0.05 was considered statistically significant.

RESULTS
Over a period of 18 years from January 1985 to December 2002, 1187 invasive prenatal tests
were performed for the diagnosis of thalassaemia. Sixty-seven of them were excluded from
the study, because of incomplete data on pregnancy outcome. The mean age of the women at
sampling was 30.3 4.5 years (SD) (range = 1745 years). The type of thalassaemia couple
59

in the studied 1120 cases included 697 -thalassaemia couples (62.2%), 414 -thalassaemia
couples (37.0%), and 9 -thalassaemia couples (0.8%). The mean gestation when the
invasive prenatal test was performed was 15.6 4.0 weeks (SD) (range = 932 weeks). The
invasive tests included 641 amniocenteses, 461 chorionic villous samplings and 18
cordocenteses.
Normal karyotypes (46,XX or 46,XY) were found in 1103 samples (98.5%). There were 17
cases (1.5%) of chromosomal abnormalities (Tables 13) in these 1120 samples including
two cases (0.18%) of common aneuploidies (21, 18, 13), four cases (0.36%) of sex
chromosome aneuploidies, one case (0.09%) of 92,XXXX, one case (0.09%) of
46,XY/46,XX, three cases (0.27%) of other aneuploidies and six cases (0.54%) of balanced
translocations/inversions. The incidence of chromosomal abnormalities did not differ
between - and -thalassaemia (P = 0.452) or between maternal age < 35 and 35 years (P =
0.214) which might be limited by the small sample size.

Table 1. Clinically significant chromosomal abnormalities (case number: 1-4)

Table 2. Chromosomal abnormalities with borderline clinical significance (case number:

5-12)
Table 1 summarized the four cases (0.36%) of clinically significant chromosomal
abnormalities. Termination of pregnancy was chosen by the parents. Case 1 and 2 were
terminated because of trisomy 21 and 18, respectively. Case 3 (45,X) was terminated because
of the presence of major ultrasound abnormalities. Case 4 (47,XX,+16) had major ultrasound
fetal abnormalities. This was the only clinically significant case that would have been missed
60

if RAT had been used as a stand-alone test instead of karyotyping when invasive prenatal
testing was performed for diagnosis of thalassaemia. Table 2 summarizes the eight cases
(0.7%) of chromosomal abnormalities with borderline clinical significance. They included
sex chromosome abnormalities (case 5 and 6), inversions (case 7, 8 and 9) and balanced
translocations (case 10, 11 and 12) either inherited from one of the parents or de novo in
origin. The prognosis of sex chromosome abnormalities is very different from that of trisomy
21, 18 and 13. Some obstetricians, clinical geneticists and genetic counsellors are uneasy
about testing and reporting the sex chromosome status for all fetuses undergoing invasive
prenatal tests. The balanced translocations and inversions were unlikely to be clinically
significant for these pregnancies but may pose significant risk to future pregnancies. Table 3
summarizes the remaining five cases (0.44%) of chromosomal abnormalities that were not
confirmed on subsequent prenatal (case 13, 15 and 17) or postnatal (case 14, 16) samples.
They were probably a result of maternal cell contamination (case 13), confined placental
mosaicism (case 14, 16 and 17) or cell culture problem (case 15). Excluding these five cases,
there were 12 cases (1.1%) of chromosomal abnormalities: seven autosomal (case 1, 2, 4, 9,
10, 11, 12) and five sex chromosomal (case 3, 5, 6, 7 and 8) abnormalities. The overall
incidence of autosomal abnormalities (7/1120 = 0.63%) was not significantly higher than that
(0.41%) derived from newborn surveys for women of all ages (P = 0.398) (Hook &
Hamerton, 1977). The incidence of trisomy 21, 18, 13 (case 1, 2) (2/1120 = 0.18%) was
comparable to that (0.14%) from newborn surveys (P = 0.751) (Hook & Hamerton, 1977).
Sex chromosomal abnormalities (5/1120 = 0.45%) were also not significantly higher than
those (0.22%) from newborn surveys (P = 0.216) (Hook & Hamerton, 1977).

Table 3. Chromosomal abnormalities not confirmed on subsequent prenatal or postnatal

samples (case number: 13-17)

DISCUSSION
In our local practice, whenever an invasive prenatal test is performed for diagnosis of
61

homozygous thalassaemia, karyotyping is also performed to exclude the presence of

chromosomal abnormalities. This strategy in prenatal diagnosis is not universal over the
world. For example, in Sardinia of Italy, karyotyping is not routinely carried out for most of
the cases, despite the couples insistence (Monni et al., 1999). This is because in their social
and health system, fetal karyotyping is allowed free of charge only for women of > 35 years;
in those with previous offspring affected; after the detection of ultrasound markers for
chromosomal abnormalities and in cases of abnormal Down syndrome screening test results
(Monni et al., 1999). In our society, this strategy would not be acceptable to the thalassaemia
couples who would find an abnormal fetal karyotype unacceptable, after undergoing an
invasive procedure for the diagnosis of thalassaemia. On the other hand, routine karyotyping
(Approach 1) would imply higher and unnecessary financial costs, as most young women (<
35 years) bear babies with a normal karyotype.
We do not find a higher prevalence of chromosomal abnormalities in offspring of
thalassaemia couples in our community. The incidences of both autosomal and sex
chromosomal abnormalities were not higher than those from newborn surveys (Hook &
Hamerton, 1977). In particular, the incidence of trisomy 21, 18, 13 (0.18%), the most
common and clinically significant chromosomal abnormalities, was comparable to that
(0.14%) from newborn surveys (Hook & Hamerton, 1977). One can then argue whether it is
necessary to perform karyotyping for fetuses of thalassaemia couples when the risk of these
common aneuploidies is no higher than that of the normal population. On the other hand,
others would argue that if an invasive prenatal test has been performed, it would be unethical
not to exclude chromosomal abnormalities by karyotyping.
We then studied the feasibility of using RAT as a stand-alone test (Approach 2) to replace
additional karyotyping when invasive prenatal testing is performed for the diagnosis of
thalassaemia. One major advantage of RAT is that the result can be available within 12 days
as compared to 2 weeks or more for karyotyping. This becomes more important if
PCR-based techniques are used in the DNA analysis for thalassaemia (Clark & Thein, 2004).
As a result, the thalassaemia couple, usually very anxious, can have both the thalassaemia
and RAT results within several days. In a great majority of cases (>98.5%), the couples can
be reassured by the normal RAT results. The other advantage of RAT over karyotyping is
that it is suggested to be more cost-effective particularly when performed on a large-scale
basis (Grimshaw et al., 2003).
The potential disadvantage of using RAT as a stand-alone test is that chromosomal
abnormalities other than common aneuploidies of chromosomes 21,18,13, X and Y would be
missed. It is estimated that up to four potentially clinically significant chromosomal
62

abnormalities can be missed for every 1000 amniocenteses performed for positive Down
screening test or advanced maternal age (Leung et al., 2004a&b) when RAT is used as a
stand-alone test. Our study showed that the risk of missing clinically significant
chromosomal abnormalities using RAT as a stand-alone test is even less when an invasive
prenatal test is performed for diagnosis of thalassaemia. This is because the prevalence of
chromosomal abnormalities in offsprings of thalassaemia couples is not higher than that of
the normal population. Of the 12 cases (1.1%) of chromosomal abnormalities, only four of
them (0.36%) were clinically significant (Table 1). Only one of these four cases was not
detectable by RAT (case 4, 47,XX,+16). It is important to note that this case had major
ultrasound abnormalities. If we follow the recommendation that RAT is offered to all women,
restricting karyotyping to those cases with ultrasound fetal abnormalities (Approach 3)
(Leung et al., 2004a&b; Leung et al., 2003), this single case would also be identified. For the
other eight cases of chromosomal abnormalities with borderline clinical significance (Table
2), six of them were not detectable by RAT. For the five cases of chromosomal abnormalities
not confirmed on subsequent prenatal or postnatal samples (Table 3), three of them were not
detectable by RAT. In this regard, using RAT as a stand-alone test can have the bonus of
reducing unnecessary counselling time as well as the associated iatrogenic anxiety to the
couple.
Identification of sex chromosome aneuploidies, particularly in the absence of ultrasound fetal
abnormalities, often poses difficult counseling issues and may not be in the best interest of
the parents or the fetus, because it often presents a difficult choice regarding the continuation
of the pregnancy (Donaghue et al., 2003). It is interesting to note that if the stand-alone RAT
approach is further limited to chromosomes 21, 18, 13 for all cases, RAT for chromosomes X
and Y only when there are ultrasound fetal abnormalities consistent with Turner syndrome
(Donaghue et al., 2003), and karyotyping only performed for those cases with ultrasound
fetal abnormalities (Approach 4): all four cases of clinically significant chromosomal
abnormalities (Table 1) could be detected. Seven of the eight cases (except case 6 with
ultrasound abnormalities) of chromosomal abnormalities with borderline clinical significance
(Table 2) and four out of the five cases of chromosomal abnormalities not confirmed on
subsequent prenatal or postnatal samples (except case 13 with ultrasound abnormalities)
(Table 3) would not be identified unnecessarily.

CONCLUSIONS
RAT (21,18,13) for all + RAT (X,Y) for cases with ultrasound abnormalities consistent with
Turner syndrome + karyotyping for cases with ultrasound abnormalities seemed to be the
63

best approach for the detection of chromosomal abnormalities when invasive prenatal testing
is performed for diagnosis of thalassaemia.

ACKNOWLEDGEMENTS
We thank the staff of Mrs. Wu Chung Prenatal Diagnostic Laboratory; our nurse specialists,
HY Chan and YP Lee; TK Lau (Prince of Wales Hospital) and WL Lau (Kwong Wah
Hospital) for their help in completing the database.

64

Chapter 5.1
Anxiety Studies
A randomized controlled trial on the effect of
fast reporting by amnio-PCR on anxiety levels
in women with positive biochemical screening
for Down syndrome
Reproduced with permission from Prenat Diagn:
Leung WC, Lam YH, Wong Y, Lau ET, Tang MHY. The effect of fast
reporting by amnio-PCR on anxiety levels in women with positive
biochemical screening for Down syndrome a randomized controlled trial.
Prenat Diagn 2002;22:2569.

65

ABSTRACT
Objective
To study the effect of fast reporting by polymerase chain reaction on amniotic fluid cells
(amnio-PCR) on anxiety levels in women with positive biochemical screening for Down
syndrome.
Method
Between May 2000 and April 2001, 60 screen-positive women were randomized before
amniocentesis into either having (group A) or not having (group B) fast-reporting by
amnio-PCR. Anxiety levels were measured by the Spielberger State-Trait Anxiety Inventory
just prior to amniocentesis, three days (when PCR results were known to group A) and three
weeks (when standard karyotype results were known to both groups) afterwards.
Results
Two women were excluded because in one woman amnio-PCR showed trisomy 21 and the
other miscarried shortly after amniocentesis. The state-anxiety scores increased over the
three-week period after being informed of the positive-screen result in both groups. The traitand state-anxiety scores at all points did not differ between the two groups.
Conclusions
In contrast to the general belief, fast reporting by amnio-PCR did not alleviate anxiety in
women who are screen-positive for Down syndrome.

INTRODUCTION
Second-trimester maternal serum screening for fetal Down syndrome has become an
established practice in many countries (Cuckle, 2000). The use and acceptability of screening
with human chorionic gonadotrophin, alpha-fetoprotein and/or unconjugated oestriol has
been demonstrated by numerous large prospective intervention studies (Cuckle, 2000). We
have also demonstrated its efficacy in Hong Kong for women of all ages (Lam et al.,
1998&2000; Lam & Tang, 1998). Serum screening generates a risk of Down syndrome in the
fetus. Women being assigned a risk above an arbitrary cut-off are designated screen-positive.
Screen-positive women are offered amniocentesis for the diagnosis of Down syndrome.
Previous studies have shown that this group of screen-positive women exhibit great anxiety
(Evans et al., 1988; Abuelo et al., 1991; Marteau et al., 1992). This may be because these
women have not anticipated themselves to be at increased risk, or they may have erroneously
66

interpreted the meaning of screen positivity. Although adequate counselling of the women
has been shown to reduce their anxiety level, the effect will not be as great as when the
karyotype result turns out to be normal (Abuelo et al., 1991; Burton et al., 1985). In other
words, this group of women would be subjected to significant anxiety during the two to three
weeks of waiting period for the karyotype report. For most of the screen-positive women, this
period of anxiety is unnecessary because the odds of having a Down syndrome pregnancy is
only around 1:40. It would be useful if there were faster means to obtain the results.
Von Eggeling et al. (1993) devised a polymerase chain reaction (PCR)-based test that
allowed rapid detection of trisomy 21 in less than 24 hours. This technique can be applied to
DNA from uncultured amniotic fluid cells that have been amplified with small tandem repeat
markers using PCR technique and fluorescence labelled primers (amnio-PCR) (Pertl et al.,
1994). In a masked prospective study, including 2167 pregnant women that had
amniocentesis, Verma et al. (1998) showed that the rapid test was informative in 99.6% of
cases and there were no false-positive or false-negative diagnoses of Down syndrome.
In this study we investigated whether fast reporting by amnio-PCR would alleviate the
anxiety in women with positive biochemical screening for Down syndrome.

MATERIALS AND METHODS

Since 1997 our centre has been offering routine second-trimester maternal serum screening
for Down syndrome for women of all ages (Lam & Tang, 1998). Women aged 35 years or
older and those with other risk factors are given the options of cytogenetic diagnosis by CVS
or amniocentesis (Lam et al., 2000). Serum screening is performed between 15 and 20 weeks
of gestation by assay of alpha-fetoprotein and human chorionic gonadotrophin. Women with
an assigned risk of 1 in 250 or greater are designated screen-positive and are offered
amniocentesis. Cytogenetic diagnosis by karyotyping is performed on the cultured
amniocytes to diagnose Down syndrome. Fast reporting by amnio-PCR or FISH has not been
routinely offered. This study was performed between May 2000 and April 2001.
Screen-positive women who could speak Chinese and agreed to undergo amniocentesis were
recruited. They were randomized by randomization table into two groups: group A was
offered fast reporting by amnio-PCR and group B was not offered fast reporting. Amnio-PCR
was performed on uncultured amniocytes using multiplexed polymorphic tetranucleotide
markers D21S1411, D21S1412 and D21S1414. The report was available within 48 hours and
the patient was informed of the amnio-PCR result by one investigator (YH Lam) via
telephone as soon as it became available. They were told the accuracy of amnio-PCR in
67

diagnosing Down syndrome and that the report would be confirmed by full cytogenetic
studies. In both groups, A and B, chromosome study of the cultured amniocytes was
performed. The full cytogenetic reports were sent to the patients by mail within three weeks
of amniocentesis.
Anxiety levels were assessed by the Spielberger State-Trait Anxiety Inventory (STAI)
(Spielberger et al., 1970). The STAI consists of a 20-item state-anxiety scale that was
designed to assess the level of relatively transient situation-related stress perceived in a
particular situation and a 20-item trait-anxiety scale that was designed to measure the
relatively stable long-term resting level of anxiety in an individual. The STAI has been
translated into Chinese (Tsoi et al., 1986) and validated (Shek, 1993). Each item in the STAI
scores one to four marks. The total scores obtained on the state and trait anxiety scales
ranged from 20 to 80 marks. A slight variation in score indicates large differences in the level
of anxiety (Spielberger et al., 1970). We measured the anxiety levels in these women on three
occasions: (1) before amniocentesis, (2) three days after amniocentesis (when the amnio-PCR
report had already been given to group A), and (3) three weeks after amniocentesis (when the
full cytogenetic report had been given to groups A and B). All women were asked to answer
both the trait-anxiety questionnaire and the state-anxiety questionnaire just prior to
amniocentesis. They were requested to fill in the state-anxiety questionnaires three days and
three weeks after amniocentesis and return them by fax or mail.
Demographic data, including the womens age, marital status, parity, race, education,
occupation and total family income, were recorded. Information on whether the pregnancy
was planned or unplanned, whether there was any previous pregnancy with congenital
abnormality, gestational age at biochemical Down screening, time interval between phone up
to counselling regarding the positive biochemical screening result and time interval between
counselling to amniocentesis, and whether a second opinion was sought before the decision
for amniocentesis were also recorded. The research protocol was approved by the local ethics
committee. Statistical analysis was performed by SPSS computer software. Demographic
data of the two groups were compared using Students t-test and Chi-square test where
appropriate. The trait- and state-anxiety scores between the two groups were compared by the
MannWhitney U-test, as they were not normally distributed. A p-value of <0.05 was
considered statistically significant.

RESULTS
74 women with positive biochemical Down screening were eligible for the study and 60
68

women agreed to be recruited. 30 women were randomized into group A and 30 women were
randomized into group B. Two women in group A were excluded from the final analysis
because in one woman amnio-PCR showed trisomy 21 and the other miscarried shortly after
amniocentesis. All of the other 58 women completed the anxiety questionnaires on the three
specific occasions.
There was no significant difference between the two groups in age, marital status, parity, race,
education, occupation and total family income. The two groups did not differ in the
proportion of planned/unplanned pregnancy, previous pregnancy with congenital abnormality,
gestational age at biochemical Down screening, the time intervals between phone up to
counselling and counselling to amniocentesis, and whether a second opinion was sought
before amniocentesis. All of these factors might have affected the anxiety levels of the
studied subjects.

Table 1. Trait- and state-anxiety scores of group A (with amnio-PCR) and group B (without
amnio-PCR)

69

Figure 1. State-anxiety scores of group A (with amnio-PCR) and group B (without

amnio-PCR)
Table 1 shows the trait- and state-anxiety scores of women in groups A and B. The trait- and
state-anxiety scores at all points did not differ between the two groups. The state-anxiety
scores in both groups increased over the three-week period after being informed the positive
screen results (Figure 1).

DISCUSSION
To the best of our knowledge, this is the first study reported in the literature that documents
the effect of fast reporting on anxiety levels in women with positive biochemical screening
for Down syndrome. In contrast to the general belief, our study failed to demonstrate any
effect of fast reporting by amnio-PCR on anxiety levels in these screen positive women. We
have controlled for the various confounding variables on anxiety levels by randomization, as
reflected by the similar demographic data and trait-anxiety scores of both groups. Our study
has a 90% power to detect a clinically significant difference of five marks in state-anxiety
scores between the two groups at a two-tailed alpha of 0.05. Hence, any undetected
differences in anxiety scores are unlikely to be large.
The findings of the present study are in contrast to those reported by Tucker in 1997 (as
quoted in Pergament et al. (2000)) in his Masters Thesis, who found significant relief of
anxiety after receipt of the fast report. However, in Tuckers study, women at increased risk
70

for Down syndrome, abnormal ultrasound or maternal anxiety were not randomized but were
allowed to choose to have fast reporting by FISH (fluorescence in situ hybridization). Those
women who chose to have fast reporting had a significantly higher initial level of anxiety
than those declining FISH. This would have introduced bias in the results. The present study
was a randomized controlled trial and the two groups (with or without PCR) did not differ in
demographic data and trait- and initial state-anxiety scores, allowing a more valid
comparison.
The state-anxiety scores increased over the three-week period after being informed of the
positive-screen result in both groups. In group B (without PCR) this is understandable
because of a progressive increase in anxiety level while waiting for the karyotype report. On
the other hand, a similar increase in anxiety occurred in group A, even after being informed
of the normal PCR report. There are two possible explanations for this finding. Firstly,
although we explained to the women that amnio-PCR is highly accurate, we also emphasized
the need for confirmation by the full karyotype result. As a result, there would still be a
certain degree of anxiety while waiting for the karyotype report even after a normal PCR
report. Secondly, having been informed of the positive Down screening result, the women
might suspect that there is something wrong with the fetus despite the normal PCR and
karyotype reports, and this heightened anxiety might persist even after birth of an unaffected
baby (Marteau et al., 1992). It would be interesting to investigate the anxiety levels of these
two groups of studied subjects in the long-term.
The anxiety generated by positive screening is iatrogenic. Because the majority of these
women are having a normal pregnancy, measures should be taken to relieve their
unnecessary anxiety. Data in the present series show that fast reporting by PCR is not
effective in relieving anxiety in the short-term. Other means of reducing anxiety in these
women may include better pre-screening and pre-amniocentesis counseling together with
patient support groups. These should be explored by further studies.

ACKNOWLEDGEMENTS
We thank Professor DTL Shek for providing us with the Chinese version of the STAI and Ms.
KY Wong for collection and analysis of the data.

71

Chapter 5.2
Anxiety Studies
A prospective study on the effect of rapid
aneuploidy testing on anxiety levels and quality
of life measures in women and their partners
with positive Down screening
Reproduced with permission from Fetal Diagn Ther:
Leung WC, Lau ET, Ngai C, Lam H, Leung KY, Lee CP, Lao TT, Tang
MHY. A prospective study on the effect of rapid aneuploidy testing
(amnio-PCR) on anxiety levels and quality of life measures in women and
their partners with positive Down screening result. Fetal Diagn Ther
2008;24:165-9.

72

ABSTRACT
Objectives
To examine the effect of rapid aneuploidy testing by amnio-PCR on anxiety levels and
quality of life measures in women and their partners with positive Down screening result.
Methods
In the original design, screen-positive women were to be randomized to have amnio-PCR or
not. Of the first 60 women approached to join the study between April 2004 and April 2005,
4 declined amniocentesis, 14 agreed to be randomized, while the other 42 (75%) chose to pay
for the amnio-PCR themselves (3 excluded: 2 because of Down syndrome and 1 dropout).
The study was thus performed as a prospective observational study on the remaining 39
women of the last group. The longitudinal profile of the state-anxiety and quality of life
domain scores for this cohort were studied using the Spielberger State-Trait Anxiety
Inventory and WHO Quality of Life Measure abbreviated version (Hong Kong)
Questionnaire at 5 time points: (1) before amniocentesis, (2) 2 days after amniocentesis when
amnio-PCR report was disclosed, (3) 3 weeks after amniocentesis when karyotyping report
was disclosed, (4) 3032 weeks gestation, (5) 6 weeks after delivery.
Results
In the final cohort of 39 women and 27 partners, a significant reduction in their state-anxiety
scores was found when they received the normal amnio-PCR report. On the other hand, there
was no significant change in their quality of life domain scores.
Conclusions
There is a demand from women and their partners who had a positive Down screening result
for rapid aneuploidy testing (amnio-PCR) which can effectively alleviate their anxiety. A
rapid aneuploidy test should be made available to women in a Down screening programme.

INTRODUCTION
Down syndrome (trisomy 21) is the commonest cause of mental retardation and is also
associated with major structural anomalies of the heart and other congenital abnormalities.
Various Down screening tests in the first and second trimesters of pregnancy are now
available to our pregnant women (Lam et al., 2002). In our Prenatal Diagnostic Clinic, if the
screening test is positive (risk above an arbitrary cutoff), the couple would be contacted by
phone to attend a counselling session. Chorionic villous sampling or amniocentesis for
73

karyotyping is offered as a diagnostic test. Traditional karyotyping is not only labour

intensive, but the results are usually not available within 2 weeks which leads to great anxiety
in these couples when they are being counselled. This phenomenon had been reported for a
long time in the literature (Evans et al., 1988; Abuelo et al., 1991; Marteau et al., 1992).
Although adequate counseling can help to decrease anxiety level significantly, the effect was
not as great as when the diagnostic test result came back to be normal (Abuelo et al., 1991;
Burton et al., 1985). For most of the couples, their anxiety during the waiting period for the
diagnostic test result is unnecessary because the majority would have normal results.
Therefore the sooner they have the result, the earlier should be the relief of anxiety.
Recent advances in molecular diagnostics, using either fluorescence in situ hybridization
with chromosome-specific DNA probes or quantitative fluorescence-polymerase chain
reaction with chromosome-specific small tandem repeat markers can be applied to diagnose
trisomy 21 and other common aneuploidies within 12 days (Leung et al., 2004b). In order to
determine the effect of fast reporting by amnio-PCR on anxiety levels in women with
positive maternal serum Down screening, we had studied 60 screen-positive women
randomized before amniocentesis into either having or not having fast reporting by
amnio-PCR (Leung et al., 2002). In contrast to the general belief, our results showed that fast
reporting did not alleviate the anxiety in these women. We speculated that although we had
explained to the women that amnio-PCR was highly accurate, the need for confirmation by
karyotyping had created a certain degree of anxiety while waiting for the karyotyping report
even after a normal amnio-PCR report. To overcome this problem, we can treat the
amnio-PCR result as final, as the potential use of amnio-PCR as a stand-alone test for
indications such as positive Down screening or advanced maternal age has been suggested
(Thilaganathan et al., 2000; Thein et al., 2000; Ryall et al., 2001; Mann et al., 2001; Leung &
Lao, 2005). Indeed, a retrospective analysis from our database indicated that amnio-PCR can
be an alternative to karyotyping for women with positive Down screening in the absence of
ultrasound fetal structural abnormalities (Leung et al., 2003). We are currently treating the
amnio-PCR result regarding the diagnosis of trisomy 21 as final, i.e. to offer reassurance if
normal and the option of termination of pregnancy if abnormal.
In our present study, the primary research question is whether treating amnio-PCR, instead of
karyotyping, as the final diagnostic result could alleviate the anxiety in women with positive
Down screening.

MATERIALS AND METHODS

74

This study was designed originally as a randomized controlled trial on Chinese women who
underwent amniocenteses because of positive Down screening with either an integrated test
(nuchal translucency at 1014 weeks + maternal serum AFP and hCG at 1520 weeks) or
maternal serum screening (at 1520 weeks) alone. Counselling for screen-positive women
was given by the same two investigators (WC Leung or C Ngai) to minimize the variation in
counseling style. The screen-positive women would be randomized into two groups: group A
with amnio-PCR for trisomy 21 and group B with karyotyping. The amnio-PCR report would
be given to the women in group A by phone and by mail within 2 days after amniocentesis,
and this report was treated as a final report. Karyotyping would still be performed for group
A but a second report would not be given to the women unless the karyotyping showed
additional chromosomal abnormalities. Women in group B would be given the karyotyping
report by mail 3 weeks after amniocentesis. Written consent would be obtained. Women
could refuse to join the study. On the other hand, they could choose to have the amnio-PCR
test directly, instead of being randomized, in addition to karyotyping by paying the cost
(equivalent to GBP 70) of amnio-PCR themselves. This is because in our public health care
system, karyotyping but not amnio-PCR is funded by the government. We could not
withhold the offering of the options for ethical reasons for screen-positive women because of
the existing service in our centre at the time of our study.
We used the Spielberger State-Trait Anxiety Inventory (STAI) (Spielberger et al., 1970) to
assess the anxiety levels. STAI had also been used in our previous study (Leung et al., 2002).
It is one of the most frequently used measures of anxiety in applied psychology research. It is
a reliable and sensitive measure of anxiety. The STAI consists of a 20-item A-state scale
(designed to assess the level of relatively transient situation-related state stress perceived
in a particular situation) and a 20-item A-trait scale (designed to measure the relatively
stable long-term resting level trait of anxiety in the individual). The Chinese version of
the STAI (Tsoi et al., 1986) had been validated (Shek, 1993). Each item in the STAI is
graded from 1 to 4 marks. The scores obtained on the A-state and A-trait scales thus range
from 20 to 80 marks. The higher the scores, the greater the anxiety. A difference of 5 marks
is considered clinically significant (Spielberger et al., 1970). The anxiety levels were
assessed at the following five time points: (1) before amniocentesis, (2) 2 days after
amniocentesis when the amnio-PCR report was disclosed to the women in group A, (3) 3
weeks after amniocentesis when the karyotyping report was disclosed to the women in group
B, (4) 3032 weeks gestation, (5) 6 weeks after delivery.
In addition, the women would be asked to complete the World Health Organization Quality
of Life Measure abbreviated version (Hong Kong) Questionnaire, WHOQOL-BREF(HK)
(Chan et al., 1997), at the same five time points. This questionnaire assesses the quality of
75

life in four domains: physical health, psychological, social relationships and environment.
The response scale of all the 28 questions is on a five-point scale (1 not at all, 2 not much,
3 moderately, 4 a great deal, 5 completely). The raw domain scores are then converted
to the transformed domain scores (0100) for further analysis using a standardized
conversion table. The higher the scores, the better the quality of life.
If the women were accompanied by their partners for amniocentesis, the partners would also
be invited to join the study. They would be asked to answer the STAI and
WHOQOL-BREF(HK) questionnaires at the first three time points: (1) before amniocentesis,
(2) 2 days after amniocentesis, (3) 3 weeks after amniocentesis.
Demographic data including the womens age, marital status, socioeconomic status
(education, occupation, family income), parity, whether the pregnancy was planned, family
history or history of previous pregnancy with congenital abnormalities, gestational age at
amniocentesis, and whether any second opinion had been sought before amniocentesis,
would be recorded.
Statistical analysis was performed using a commercial statistical package (SPSS 10.0 for
Windows). The research protocol was approved by the local ethics committee.

RESULTS
Our initial sample size calculation showed that 30 women would be required in each group in
order to detect a clinically significant difference of 5 marks in state-anxiety scores between
the two groups with a power of 90% at a 2-tailed alpha of 0.05. The recruitment situation was
reviewed after the first 60 screen-positive women were invited to join the study from April
2004 to April 2005. Four of them declined the offer of amniocentesis. Out of the remaining
56 women, only 14 agreed to be randomized. The other 42 screen-positive women (75%)
chose to have the amnio-PCR test directly in addition to karyotyping, instead of being
randomized, by paying the cost (equivalent to GBP 70) of amnio-PCR themselves. From this
unexpected response, we estimated that we might need another 34 years to recruit enough
subjects for the randomized controlled trial. Therefore, we decided to abandon the
randomized controlled study design and change to a prospective observational study on the
longitudinal profile of the state-anxiety and quality of life domain scores for the cohort of 42
women who had chosen amnio-PCR. Three out of the 42 women were excluded: two of them
had fetuses affected by trisomy 21 and one woman refused to answer the questionnaires after
she had amniocentesis. The final cohort thus consisted of 39 women, 27 of whom had their
76

partners accompanying them for amniocentesis and all partners agreed to join the study. The
state-anxiety and quality of life domain scores at the various time points were compared
using the Friedman test. A p value of < 0.05 was considered statistically significant.
Table 1 shows the characteristics of the cohort including 39 screen-positive women and 27
partners. They were not different from those of the 14 women who agreed to be randomized.
The longitudinal profile of the state-anxiety and quality of life domain scores of the
screen-positive women who had chosen amnio-PCR is shown in table 2. There was a
significant reduction in their state-anxiety scores when the women received the normal
amnio-PCR report (fig. 1). Their quality of life domain scores remained static except for a
decrease in the environment domain scores after they received the normal karyotyping report.
Table 3 shows the state-anxiety and quality of life domain scores of the womens partners.
Again, there was a significant reduction in the partners state-anxiety scores when the normal
amnio-PCR report was known (fig. 1). On the other hand, there was no significant change in
the partners quality of life domain scores.

77

Table 1. Characteristics of the cohort: women (n=39) and their partners (n=27)

78

Figure 1. Mean state-anxiety scores of screen-positive women (n=39; dark grey bar) and
their partners (n=27; light grey bar)

Table 2. State-anxiety and quality of life (QOL) domain scores of screen-positive women
(n=39) who had chosen amnio-PCR

Table 3. State-anxiety and QOL domain scores of the womens partners (n=27)

79

DISCUSSION
Seventy-five percent of the screen-positive women chose to have the amnio-PCR test directly,
instead of being randomized, by paying the cost of amnio-PCR themselves. In our public
health care system, karyotyping is funded by the government, but amnio-PCR test has to be
paid by the women themselves. Their willingness to pay for the test reflected the demand to
obtain a rapid result by women with positive Down screening test (Caughey et al., 2004).
In contrast to the findings of our previous study (Leung et al., 2002) which did not show that
the normal amnio-PCR result can alleviate the anxiety of the screen-positive women, this
time we demonstrated that it can, for both the woman and her partner (fig. 1). This was
probably related to how we evaluated the amnio-PCR test during counselling. In our previous
study (Leung et al., 2002), we explained to the women that amnio-PCR was highly accurate,
but we had also emphasized the need for confirmation by karyotyping. Thereafter, we had
performed another study showing our amnio-PCR test was 100% accurate, without a
false-positive or false-negative result regarding the prenatal diagnosis of trisomy 21 (Leung
et al., 2003). As a result, in our present study, we explained to the couple that the amnio-PCR
report could be treated as the final diagnosis of trisomy 21. Unfortunately, we could not
demonstrate this phenomenon by the original randomized controlled study design. This was
the major limitation of our study. Indeed, when we looked at the results of those 14
screen-positive women who agreed to be randomized, the reduction in the state-anxiety
scores occurred when women in group A received the normal amnio-PCR report and when
women in group B received the normal karyotyping report. The subject number was however
too small for any meaningful statistical comparison. Grimshaw et al. (2003), in the NHS
Health Technology Assessment Report on evaluation of molecular tests for prenatal
diagnosis of chromosome abnormalities, had already shown that on receipt of the first test
result, whether it was the PCR or karyotyping, a significant reduction in anxiety scores was
observed (although this was not a randomized controlled trial either). The findings of our
study demonstrated this phenomenon in Chinese women and their partners. One interesting
feature in our study was that there was no further drop in the state-anxiety scores when the
couples received the karyotyping report. This implied that the couples were really treating the
amnio-PCR report as final. In the study of Grimshaw et al. (2003), a significant improvement
in quality of life scores linked to more rapid test results had also been found. This had not
been demonstrated in our study. The decrease in the environment domain scores after the
women received the normal karyotyping report might just be related to advancing pregnancy.
One possibility is that the WHOQOLBREF(HK) instrument (Chan et al., 1997) used in our
study is not sensitive enough to detect short-term changes in the various quality of life
80

domains. The other possibility is that the rapid relief of anxiety by the amnio-PCR report
prevents the occurrence of any significant deterioration in the quality of life.
We conclude from our study that there is a demand from women and their partners who had a
positive Down screening result for rapid aneuploidy testing (amnio-PCR) which can
effectively alleviate their anxiety. A rapid aneuploidy test should be made available to
women in a Down screening programme.

ACKNOWLEDGEMENTS
We thank YP Lee and Vivian Chan for recruitment of subjects, Jane Chan and MS Tam for
collection and analysis of data, and Daniel Fong for statistical advice.

81

Chapter 6
Discussion & Conclusion

82

This thesis contributes to the debate on Rapid Aneuploidy Testing vs. Traditional
Karyotyping in prenatal diagnosis.
The accuracy of rapid aneuploidy testing (fluorescence in situ hybridization or quantitative
fluorescence-polymerase chain reaction) in prenatal diagnosis of the common aneuploidies of
chromosomes 21, 18, 13, X and Y has already been demonstrated (Eiben et al., 1999a,
Estabrooks et al., 1999; Pergament et al., 2000; Witters et al., 2002; Verma et al., 1998;
Levett et al., 2001; Mann et al., 2001). This is a basic prerequisite if RAT is to be used as a
diagnostic test. The accuracy of RAT is demonstrated again by studies in this thesis (Chapter
3 (Leung et al., 2001), Chapter 4.1 (Leung et al., 2003) and Chapter 4.3 (Leung et al.,
2004a)). There were no false-positive or false-negative results regarding aneuploidies of
chromosomes 21, 18, 13, X and Y. On the other hand, Lau et al. (2009) has recently
demonstrated that 0.8% chorionic villus and 0.3% amniotic fluid samples showed discrepant
results after culturing and 40% of discrepancy involved the sex chromosomes. Discrepant
QF-PCR findings in uncultured prenatal samples likely arose from mosaicism or preferential
cell culture (Lau et al., 2009).
The major advantages of RAT include fast reporting within 24 to 48 hours and earlier relief
of anxiety. The earlier relief of anxiety by a fast and accurate report with normal result seems
logical. Surprisingly, this could not be demonstrated by our randomized controlled trial in
Chapter 5.1 (Leung et al., 2002). We speculated that although we had explained to the
women that amnio-PCR was highly accurate, the need for confirmation by karyotyping had
created a certain degree of anxiety while waiting for the karyotyping report even after a
normal amnio-PCR report. Note that this study was performed in 2000 to 2001, before we
published our study on the accuracy of amnio-PCR in 2003 (Leung et al., 2003) and at that
time, we might not be confident enough ourselves regarding whether the amnio-PCR report
could be treated as final. We tried to repeat the randomized controlled trial in 2004 to 2005.
By that time, QF-PCR has already become an existing service in our centre for women with
positive Down screening, although this is a self-financed item and they have to pay HK$1300
(PCR for chromosomes 21, 18, 13, X&Y) for the RAT. 75% of women with positive Down
screening chose to pay for the amnio-PCR themselves. As a result, Chapter 5.2 had to change
from a randomized controlled trial to a prospective observational study (Leung et al.,
2008b) a significant reduction in anxiety by the normal amnio-PCR report was
demonstrated in the women and their partners. Recently, we have studied the characteristics
of women who are willing to pay for RAT - they were younger, socioeconomically better off,
and had difficulty conceiving (Lo et al., 2010).
As mentioned in Chapter 2 (Leung et al., 2004b), a stand-alone RAT approach in prenatal
83

diagnosis would result in missing diagnosis of up to four potentially clinically significant

chromosomal abnormalities (e.g. balanced de novo translocations and marker chromosomes)
for every 1000 invasive prenatal tests (amniocentesis or chorionic villus sampling) performed.
After the publication of this paper, we have updated this meta-analysis with addition of two
more papers (Table 1) and the results remain the same (i.e. 4 per 1000 tests) (Leung & Lao,
2005).

To summarize:

Table 1. Reports of rapid aneuploidy testing (FISH or QF-PCR) versus traditional

karyotyping (Leung & Lao, 2005)
The chance of missing these potentially clinically significant chromosomal abnormalities in a
stand-alone RAT approach depends on the indication for prenatal diagnosis. We have already
noticed this phenomenon in our background study (Chapter 3) on the role of RAT in patient
management (Leung et al., 2001). The chance would be lower (2/1526 or 1 to 2 per 1000
tests) if the indication for prenatal diagnosis is positive Down screening with no ultrasound
abnormalities as described in Chapter 4.1 (Leung et al., 2003). The crucial role of a
high-standard ultrasound examination is illustrated in Chapter 4.3 (Leung et al., 2004a). The
addition of full karyotyping for cases with ultrasound abnormalities in a RAT stand-alone
approach would improve the detection rate of clinically important chromosomal
abnormalities from 82% to 95% (Leung et al., 2004a). This approach of RAT for all samples
84

+ karyotyping for cases with ultrasound abnormalities had the best performance when the
indication is thalassaemia with no ultrasound abnormalities (Chapter 4.4) (Tse et al., 2006).
The chance of missing potentially clinically significant chromosomal abnormalities could be
as low as < 1 per 1000 tests (1/1120) (Tse et al., 2006). An important characteristics in this
thalassaemia study population was their young age (30 + 4 years, mean + SD). We can see
from Chapter 4.2 (Leung et al., 2008a) that advanced maternal age as an indication for
prenatal diagnosis is an important determinant of the chance of missing potentially clinically
significant chromosomal abnormalities in a stand-alone RAT approach. The chance is highest
(63/19,517 or 3 to 4 per 1000 tests) (Leung et al., 2008a).
If RAT is to replace karyotyping for indications such as positive Down screening or
advanced maternal age when no ultrasound abnormality is detected, one has to accept the risk
that for every 1000 invasive tests performed, up to 4 potentially clinical significant
chromosomal abnormalities may be missed. Some people may argue from an ethical point of
view that since amniocentesis is an invasive procedure that carries a small risk of miscarriage,
we should maximize the information that can be obtained by performing karyotyping to
examine all the 23 pairs of chromosomes. However, even the performance of karyotyping
does not mean that the information is maximised e.g. microdeletions and common mutations
are not tested. We must also realize that Down screening is a programme designed to detect
primarily Down syndrome and therefore in principle, follow-up test with RAT alone would
have realistically fulfilled the expectations of the couples and obstetricians (targeted testing).
On the other hand, one might argue that this is a beneficence-based approach (the physician
making decisions that are best for the patient, without regard to personal gain or the interests
of others). The ethics of prenatal diagnosis should be autonomy-based (the capacity of a
rational individual to make an informed, uncoerced decision; in medicine, respect for the
autonomy of patients is considered obligatory for doctors and other health care professionals).
If the parents choose to have termination of pregnancy for minor chromosomal abnormalities,
we have the obligation to make sure that the counselling is thorough, to ensure that they have
the information about the outcomes in order to exercise their autonomy to continue or
terminate the pregnancy. If after that process, the parents still choose to terminate the
pregnancy, then their decision and autonomy must be respected. A medical attitude of we
will not look for sex chromosome abnormalities and other minor chromosomal abnormalities
as we do not consider you should have a termination could be considered paternalistic (a
figurehead that makes decisions on behalf of others for their own good, even if this is
contrary to their wishes). If both techniques are available, parents could have the autonomy
to choose the approach (RAT, or traditional karyotyping, or both) after being fully informed
of the pros and cons (Chapter 4.2) (Leung et al., 2008a). A recent conjoint analysis study
showed that women would prefer simple information on just knowing whether the fetus has
85

Down syndrome as long as the result is received 6 days sooner than karyotyping (Ryan et al.,
2005). Bui (2007) recently reported their experience in Stockholm, when Swedish women
were given this choice, 70% of them chose the RAT stand-alone approach.
While the debate surrounding RAT and traditional karyotyping is likely to continue, it is
worth mentioning that technology is continuing to develop, such as microarray-based
comparative genomic hybridisation (array CGH) (Choy et al., 2008). In contrast to RAT,
array CGH is a comprehensive, high-resolution, genomewide screening strategy for detecting
gains (duplication) or losses (deletion) of DNA segments in a single test. Compared with
traditional karyotyping, it is rapid, less labour-intensive, and readily amendable to
automation. It enables the detection of genomic changes too small to be resolved by
karyotyping, such as microdeletions & microduplications. Array CGH is not ready for
routine use yet due to the cost, but it is likely to become increasingly important and has the
potential to replace traditional karyotyping in the future. This could well be the topic for
another thesis.
One of the advantages of using RAT as a stand-alone test is cost-saving which we have not
been able to study in this thesis. Instead of adding the cost of RAT on top of that of
karyotyping, the cost of the latter can be saved by the RAT stand-alone approach. In the age
of ever-escalating cost in the provision of health care, especially in a government-funded
public medical care system, the savings can be redirected to enhance existing or fund new
programmes, thus maximising the effect of limited resources. Grimshaw et al. (2003) has
conducted a cost-effectiveness analysis on five testing policies:
1. RAT and karyotyping for all women
2. RAT as a replacement for karyotyping
3. RAT for all plus karyotyping for high-risk women (e.g. with ultrasound abnormalities)
4. Karyotyping for all plus RAT for high-risk women
5. Parental choice plus karyotyping for high-risk women
Policies 2, 3 and 5 are found to be more cost-effective than karyotyping based on the cost per
case (chromosomal abnormality) detected.

FINAL CONCLUSION
The major advantages of RAT (FISH or QF-PCR) include fast reporting within 24 to 48
hours and earlier relief of anxiety. Their accuracy in prenatal diagnosis has been
86

demonstrated. The challenge is to decide what would be the best approach - Rapid
Aneuploidy Testing or Traditional Karyotyping, or Both? The results of studies in this
thesis provide information on the pros and cons of each approach. The performance of a
stand-alone RAT approach depends on the indication for prenatal diagnosis. A high standard
ultrasound examination is essential. It is reasonable to use a stand-alone RAT approach when
prenatal tests are performed for indications such as positive Down screening with no
ultrasound abnormalities. It is important to note that if this new approach is used in prenatal
diagnosis, for every 1000 invasive tests performed, up to 4 potentially clinical significant
chromosomal abnormalities could be missed. If both techniques are available, women
should have the autonomy to choose the approach (RAT, or traditional karyotyping, or
both) after being fully informed of the pros and cons.

87

References
Abramsky L, Hall S, Levitan J, Marteau TM. What parents are told after prenatal diagnosis
of a sex chromosome abnormality: interview and questionnaire study. BMJ 2001;322:463-6.
Abuelo DN, Hopmann MR, Barsel-Bowers G, Goldstein A. Anxiety in women with low
maternal serum alpha-fetoprotein screening results. Prenat Diagn 1991;11: 3815.
Adinolfi M, Sherlock J. Prenatal detection of chromosome disorders by QF-PCR. Lancet
2001;358:1030-1.
American College of Medical Genetics. Prenatal interphase fluorescence in situ hybridization
(FISH)
policy
statement.
Am
J
Hum
Genet
1993;53:5267
and
www.faseb.org/genetics/acmg.
Beris P, Ddarbellay R, Extermann P. Prevention of -thalassaemia major and Hb Barts
hydrops fetalis syndrome. Semin Hematol 1995;32:24461.
Brock DJ, Rodeck CH, Ferguson-Smith MA. Prenatal diagnosis and screening. Edinburgh:
Churchill Livingstone; 1992.
Bryndorf T, Lundsteen C, Lamb A, Christensen B, Philip J. Rapid prenatal diagnosis of
chromosome aneuploidies by interphase fluorescence in situ hybridization: a one-year
clinical experience with high-risk and urgent fetal and postnatal samples. Acta Obstet
Gynecol Scand 2000;79: 814.
Bui TH. Prenatal cytogenetic diagnosis: gone FISHing, BAC soon! Ultrasound Obstet
Gynecol 2007;30:247-51.
Burton BK, Dillard RG, Clark EN. The psychological impact of false positive elevations of
maternal serum alpha-protein. Am J Obstet Gynecol 1985;151:7782.
Caine A, Maltby AE, Parkin CA, Waters JJ, Crolla JA, UK Association of Clinical
Cytogeneticists (ACC). Prenatal detection of Downs syndrome by rapid aneuploidy testing
for chromosomes 13, 18, and 21 by FISH or PCR without a full karyotype: a cytogenetic risk
assessment. Lancet 2005;366:123-8.

88

Caughey AB, Washington AE, Gildengorin V, Kuppermann M. Assessment of demand for

prenatal diagnostic testing using willingness to pay. Obstet Gynecol 2004;103:53945.
Chan CH, Leung KF, Lin F. Hong Kong Chinese version WHOQOL-100 and its BREF form.
Proc 1997 Qual Life Symp, Hong Kong Hospital Authority, 1997.
Chan V, Chan TK, Todd D. Alpha and beta thalassaemias in Hong Kong. Chin Med J
1997;110:4903.
Choy KW, Tsang PT, Leung TY, Wang CC, Lau TK. The application of microarray based
comparative genomic hybridization in prenatal diagnosis. Fetal and Maternal Medicine
Review 2008;19:119-33.
Cirigliano V, Lewin P, Szpiro-Tapies S, Fuster C, Adinolfi M. Assessment of new markers
for the rapid detection of aneuploidies by quantitative fluorescent PCR (QF-PCR). Ann Hum
Genet 2001;65:421-7.
Cirigliano V, Voglino G, Canadas MP, et al. Rapid prenatal diagnosis of common
chromosome aneuploidies by QFPCR. Assessment on 18,000 consecutive clinical samples.
Mol Hum Reprod 2004;10:839-46.
Clark BE, Thein SL. Molecular diagnosis of haemoglobin disorders. Clin Lab Haematol
2004;26:15976.
Cuckle H. Biochemical screening for Down syndrome. Eur J Obstet Gynecol Reprod Biol
2000;92:97101.
DAlton ME, Malone FD, Chelmow D, Ward BE, Bianchi DW. Defining the role of
fluorescence in situ hybridization on uncultured amniocytes for prenatal diagnosis of
aneuploidies. Am J Obstet Gynecol 1997;176: 76976.
Donaghue C, Roberts A, Mann K, Ogilvie CM. Development and targeted application of a
rapid QF-PCR test for sex chromosome imbalance. Prenat Diagn 2003;23:201-10.
Donaghue C, Mann K, Docherty Z, et al. Detection of mosaicism for primary trisomies in
prenatal samples by QF-PCR and karyotype analysis. Prenat Diagn 2005;25:65-72.
Dudarewicz L, Holzgreve W, Jeziorowska A, Jakubowski L, Zimmermann B. Molecular
89

methods for rapid detection of aneuploidy. J Appl Genet 2005;46:207-15.

Eiben B, Trawicki W, Hammans W, Goebel R, Pruggmayer M, Epplen JT. Rapid prenatal
diagnosis of aneuploidies in uncultured amniocytes by fluorescence in situ hybridization.
Evaluation of >3,000 cases. Fetal Diagn Ther 1999a;14:193-7.
Eiben B, Trawicki W, Hammans W, Epplen JT. False negative finding in rapid interphase
FISH analysis of uncultured amniotic cells. Prenat Diagn 1999b;19: 8923.
Estabrooks LL, Sapeta M, Lytle C. Prenatal interphase FISH using the AneuVysion probe set
in over 10,000 samples. Am J Hum Genet 1999;65:A162.
Evans MI, Bottoms SF, Carlucci T. Determinants of altered anxiety after abnormal maternal
serum alpha-fetoprotein screening. Am J Obstet Gynecol 1988:59;15014.
Evans MI, Henry GP, Miller WA, et al. International, collaborative assessment of 146,000
prenatal karyotypes: expected limitations if only chromosome-specific probes and
fluorescent in-situ hybridization are used. Hum Reprod 1999;14:1213-6.
Feldman B, Ebrahim SAD, Hazan SL, et al. Routine prenatal diagnosis of aneuploidy by
FISH studies in high-risk pregnancies. Am J Med Genet 2000;90: 2338.
Grimshaw GK, Szczepura A, Hultn M, et al. Evaluation of molecular tests for prenatal
diagnosis of chromosome abnormalities. Health Technol Assess 2003;7:1146.
Hochstenbach R, Meijer J, van de Brug J, Vossebeld-Hoff I, Jansen R, van der Luijt RB, et al.
Rapid detection of chromosomal aneuploidies in uncultured amniocytes by multiplex
ligation-dependent probe amplification (MLPA). Prenat Diagn. 2005;25:1032-9.
Homer J, Bhatt S, Huang B, Thangavelu M. Residual risk for cytogenetic abnormalities after
prenatal diagnosis by interphase fluorescence in situ hybridization (FISH). Prenat Diagn
2003;23:566-71.
Hook EB. Spontaneous deaths of fetuses with chromosomal abnormalities diagnosed
prenatally. N Engl J Med 1978;299:1036-8.
Hook EB, Hamerton JL. The frequency of chromosome abnormalities detected in
consecutive newborn studies; differences between studies; results by sex and severity of
90

phenotypic involvement. In Hook EB, Porter IH (eds), Population Cytogenetics: Studies in

Humans. Academic Press, New York, 1977, pp 6379.
Huang B, Thangavelu M, Bhatt S, J Sandlin C, Wang S. Prenatal diagnosis of 45,X and 45,X
mosaicism: the need for thorough cytogenetic and clinical evaluations. Prenat Diagn
2002;22:105-10.
Hulten MA, Dhanjal S, Pertl B. Rapid and simple prenatal diagnosis of common
chromosome disorders: advantages and disadvantages of the molecular methods FISH and
QFPCR. Reproduction 2003;126:279-97.
Jalal SM, Law ME, Carlson RO, Dewald GW. Prenatal detection of aneuploidy by directly
labeled multicolored probes and interphase fluorescence in situ hybridization. Mayo Clin
Proc 1998;73: 1327.
Kagan KO, Chitty LS, Cicero S, Eleftheriades M, Nicolaides KH. Ultrasound findings before
amniocentesis in selecting the method of analysing the sample. Prenat Diagn 2007;27:34-9.
Lam YH, Ghosh A, Tang MH, et al. Second-trimester maternal serum alpha-fetoprotein and
human chorionic gonadotrophin screening for Downs syndrome in Hong Kong. Prenat
Diagn 1998;18:5859.
Lam YH, Tang MHY. Prenatal diagnostic and counseling services in Hong Kong. Adv
Obstet Perinatol 1998;9:2930.
Lam YH, Tang MH, Lee CP, et al. Acceptability of serum screening as an alternative to
cytogenetic diagnosis of Down syndrome among 35 years or older in Hong Kong. Prenat
Diagn 2000;20:48790.
Lam YH, Lee CP, Sin SY, et al. Comparison and integration of first trimester fetal nuchal
translucency and second trimester maternal serum screening for fetal Down syndrome. Prenat
Diagn 2002;22:7305.
Lau ET, Tang L, Wong C, Lam YH, Ghosh A, Leung WC, et al. Assessing discrepant findings
between QF-PCR on uncultured prenatal samples and karyotyping on long term culture.
Prenat Diagn 2009;29:151-5.
Lee HH, Chang JG, Lin SP, et al. Rapid detection of trisomy 21 by homologous gene
91

quantitative PCR (HGQ-PCR). Hum Genet 1997;99:364-7.

Leung KY, Lee CP, Tang MHY, Lau ET, Ng LKL, Lee YP, Chan HY, Ma ESK, Chan V.
Cost-effectiveness of prenatal screening for thalassaemia in Hong Kong. Prenat Diagn
2004;24:899907.
Leung WC, Chitayat D, Seaward G et al. Role of amniotic fluid interphase fluorescence in
situ hybridization (FISH) analysis in patient management. Prenat Diagn 2001;21:327-32.
Leung WC, Lam YH, Wong Y, Lau ET, Tang MHY. The effect of fast reporting by
amnio-PCR on anxiety levels in women with positive biochemical screening for Down
syndromea randomized controlled trial. Prenat Diagn 2002;22:2569.
Leung WC, Lau ET, Lao TT, Tang MH. Can amnio-polymerase chain reaction alone replace
conventional cytogenetic study for women with positive biochemical screening for fetal
Down syndrome? Obstet Gynecol 2003;101:856-61.
Leung WC, Waters JJ, Chitty L. Prenatal diagnosis by rapid aneuploidy detection and
karyotyping: a prospective study of the role of ultrasound in 1589 second-trimester
amniocenteses. Prenat Diagn 2004a;24:790-5.
Leung WC, Lau ET, Lao TT, Tang MH. Rapid aneuploidy screening (FISH or QF-PCR): the
changing scene in prenatal diagnosis? Expert Rev Mol Diagn 2004b;4:333-7.
Leung WC, Lao TT. Rapid aneuploidy testing, traditional karyotyping, or both? Lancet
2005;366:97-8. Erratum in: Lancet 2005;366:1164.
Leung WC, Lau ET, Lau WL, Tang R, Wong SF, Lau TK, Tse KT, Wong SF, To WK, Ng LK,
Lao TT, Tang MH; Working Group on Prenatal Diagnosis and Counselling, Hospital
Authority. Rapid aneuploidy testing (knowing less) versus traditional karyotyping (knowing
more) for advanced maternal age: what would be missed, who should decide? Hong Kong
Med J 2008a;14:6-13.
Leung WC, Lau ET, Ngai C, Lam H, Leung KY, Lee CP, Lao TT, Tang MHY. A
prospective study on the effect of rapid aneuploidy testing (amnio-PCR) on anxiety levels
and quality of life measures in women and their partners with positive Down screening result.
Fetal Diagn Ther 2008b;24:165-9.

92

Levett LJ, Liddle S, Meredith R. A large-scale evaluation of amnio-PCR for the rapid
prenatal diagnosis of fetal trisomy. Ultrasound Obstet Gynecol 2001;17:115-8.
Lewin P, Kleinfinger P, Bazin A, Mossafa H, Szpiro-Tapia S. Defining the efficiency of
fluorescence in situ hybridization on uncultured amniocytes on a retrospective cohort of
27407 prenatal diagnoses. Prenat Diagn 2000;20:1-6.
Lo TK, Leung WC, Lau WL, Lam H, Lai FK, Ng LS, et al. Rapid aneuploidy test for positive
Down screening: who is willing to pay for it? J Matern Fetal Neonatal Med 2010;23:747-9.
Mann K, Fox SP, Abbs SJ, et al. Development and implementation of a new rapid
aneuploidy diagnostic service within the UK National Health Service and implications for the
future of prenatal diagnosis. Lancet 2001;358:1057-61.
Mann K, Donaghue C, Fox SP, Docherty Z, Ogilvie CM. Strategies for the rapid prenatal
diagnosis of chromosome aneuploidy. Eur J Hum Genet 2004;12:907-15.
Mansfield ES. Diagnosis of Down syndrome and other aneuploidies using quantitative
polymerase chain reaction and small tandem repeat polymorphisms. Hum Mol Genet
1993;2:4350.
Marteau TM, Cook R, Kidd J, et al. The psychological effects of false-positive results in
prenatal screening for fetal abnormality: a prospective study. Prenat Diagn 1992;12:20514.
Marteau TM. Prenatal testing: Towards realistic expectations of patients, providers and
policy makers. Ultrasound Obstet Gynecol 2002;19:56.
Mercier S, Bresson JL. Prenatal diagnosis of chromosomal aneuploidies by fluorescence in
situ hybridization on uncultured amniotic cells: experience with 630 samples. Ann Genet
1995;38:1517.
Monni G, Ibba RM, Zoppi MA, Putzolu M, Floris M, Cao A. Fetal nuchal translucency and
prenatal diagnosis of -thalassaemia. Prenat Diagn 1999;19:75860.
Morris A, Boyd E, Dhanjal S, et al. Two years prospective experience using fluorescence in
situ hybridization on uncultured amniotic fluid cells for rapid prenatal diagnosis of common
chromosomal aneuploidies. Prenat Diagn 1999;19: 54651.

93

Nicolini U, Lalatta F, Natacci F, Curcio C, Bui TH. The introduction of QF-PCR in prenatal
diagnosis of fetal aneuploidies: time for reconsideration. Hum Reprod Update
2004;10:541-8.
Ogilvie CM. Prenatal diagnosis for chromosome abnormalities: past, present and future.
Pathol Biol (Paris) 2003;51:156-60.
Ogilvie CM, Lashwood A, Chitty L, Waters JJ, Scriven PN, Flinter F. The future of prenatal
diagnosis: rapid testing or full karyotype? An audit of chromosome abnormalities and
pregnancy outcomes for women referred for Downs Syndrome testing. BJOG
2005;112:1369-75.
Pergament E, Chen PX, Thangavelu M, Fiddler M. The clinical application of interphase
FISH in prenatal diagnosis. Prenat Diagn 2000;20:215-20.
Pertl B, Yau SC, Sherlock J, Davies AF, Mathew CG, Adinolfi M. Rapid molecular method
for prenatal detection of Downs syndrome. Lancet 1994;343:1197-8.
Pertl B, Pieber D, Lercher-Hartlieb A, Orescovic I, Haeusler M, Winter R, et al. Rapid
prenatal diagnosis of aneuploidy by quantitative fluorescent PCR on fetal samples from
mothers at high risk for chromosome disorders. Mol Hum Reprod 1999;5:11769.
Rooney DE (ed.). Human cytogenetics: constitutional analysis. 3rd edition. Oxford
University Press: Oxford. 2001.
Ryall RG, Callen D, Cocciolone R, et al. Karyotypes found in the population declared at
increased risk of Down syndrome following maternal serum screening. Prenat Diagn
2001;21:553-7.
Ryan M, Diack J, Watson V, et al. Rapid prenatal diagnostic testing for Down syndrome only
or longer wait for full karyotype: the views of pregnant women. Prenat Diagn
2005;25:1206-11.
Sagi M, Meiner V, Reshef N, Dagan J, Zlotogora J. Prenatal diagnosis of sex chromosome
aneuploidy: possible reasons for high rates of pregnancy termination. Prenat Diagn
2001;21:461-5.
Shaffer LG, Bui TH. Molecular cytogenetic and rapid aneuploidy detection methods in
94

prenatal diagnosis. Am J Med Genet C Semin Med Genet 2007;145:87-98.

Shek DTL. The Chinese version of the state-trait anxiety inventory: its relationship to
different measures of psychological well-being. J Clin Psychol 1993;49:34958.
Sin SY, Ghosh A, Tang LC, Chan V. Ten years experience of antenatal mean corpuscular
volume screening and prenatal diagnosis for thalassaemia in Hong Kong. J Obstet Gynaecol
Res 2000;26:2038.
Spielberger CD, Gorsuch RL, Lushene RE. Manual for the State-Trait Anxiety Inventory
(Self-evaluation Questionnaire). Consulting Psychologists Press: Palo Alto, CA. 1970.
Stojilkovic-Mikic T, Mann K, Docherty Z, et al. Maternal cell contamination of prenatal
samples assessed by QF-PCR genotyping. Prenat Diagn 2005;25:79-83.
Tardy EP, Toth A. Cross-hybridization of the chromosome 13/21 alpha satellite DNA to
chromosome 22 or a rare polymorphism? Prenat Diagn 1997;17: 48790.
Thein AT, Abdel-Fattah SA, Kyle PM, Soothill PW. An assessment of the use of interphase
FISH with chromosome specific probes as an alternative to cytogenetics in prenatal diagnosis.
Prenat Diagn 2000;20:275-80.
Thilaganathan B, Sairam S, Ballard T, Peterson C, Meredith R. Effectiveness of prenatal
chromosomal analysis using multicolor fluorescent in situ hybridisation. BJOG
2000;107:262-6.
Tse KY, Leung WC, Leung KY, Lee CP, Ng LK, Lau ET, et al. Full karyotyping, rapid
aneuploidy diagnosis or both when invasive prenatal testing is performed for diagnosis of
thalassaemia? Mol Hum Reprod 2006;12:55-9.
Tsoi MM, Ho E, Mak KC. Becoming pregnant again after stillbirth or the birth of a
handicapped child; in Dennerstein L, Fraser I (eds): Hormone and Behaviour. Amsterdam,
Elsevier Science, 1986, pp 3106.
Tucker A. The efficiency of FISH analysis for common aneuploidies in relieving the anxiety
of patients awaiting prenatal diagnosis results. Master Thesis. Northwestern University
Graduate School; Chicago, USA. 1997.

95

UK Department of Health. Our inheritance, our future: realizing the potential of genetics in
the NHS. London: Stationery Office; 2003.
van Zwieten MC, Willems DL, Litjens LL, Schuring-Blom HG, Leschot N. How unexpected
are unexpected findings in prenatal cytogenetic diagnosis? A literature review. Eur J Obstet
Gynecol Reprod Biol 2005;120:15-21.
Verlinsky Y, Ginsberg N, Chmura M, et al. 1995. Crosshybridization of the chromosome
13/21 alpha satellite DNA probe to chromosome 22 in the prenatal screening of common
chromosomal aneuploidies by FISH. Prenat Diagn 1995;15: 8314.
Verlinsky Y, Ginsberg N, Chmura M, White M, Strom C, Kuliev A. Detection of
translocations involving the Y-chromosome in prospective prenatal screening of common
chromosomal aneuploidies by FISH. Prenat Diagn 1998;18: 3902.
Verma L, Macdonald F, Leedham P, McConachie M, Dhanjal S, Hulten M. Rapid and simple
prenatal DNA diagnosis of Downs syndrome. Lancet 1998;352:9-12.
Von Eggeling F, Freytag M, Fahsold R, Horsthemke B, Claussen U. Rapid detection of
trisomy 21 by quantitative PCR. Hum Genet 1993;91:56770.
Warburton D. De novo balanced chromosome rearrangements and extra marker
chromosomes identified at prenatal diagnosis: clinical significance and distribution of
breakpoints. Am J Hum Genet 1991;49:995-1013.
Ward BE, Gersen SL, Carelli MP, et al. Rapid prenatal diagnosis of chromosomal
aneuploidies by fluorescence in situ hybridization: clinical experience with 4,500 specimens.
Am J Hum Genet 1993;52:854-65.
Winsor EJT, Dyack S, Wood-Burgess EM, Ryan G. Risk of false-positive prenatal diagnosis
using interphase FISH testing: hybridization of alpha-satellite X probe to chromosome 19.
Prenat Diagn 1999;19: 8326.
Witters I, Devriendt K, Legius E, et al. Rapid prenatal diagnosis of trisomy 21 in 5049
consecutive uncultured amniotic fluid samples by fluorescence in situ hybridization (FISH).
Prenat Diagn 2002;22:29-33.

96

Zimmermann B, Holzgreve W, Wenzel F, et al. Novel real-time quantitative PCR test for
trisomy 21. Clin Chem 2002;48:362-3.

97

Curriculum Vitae
Wing Cheong LEUNG graduated from the Medical School of University of Hong Kong in
1989. He has completed his specialist training in Obstetrics & Gynaecology (O&G) in the
Department of O&G, Tsan Yuk Hospital & Queen Mary Hospital, University of Hong Kong
and become a specialist in O&G (Hong Kong Academy of Medicine) in 1997. Thereafter he
has further subspecialized in Maternal Fetal Medicine. He has undergone overseas training in
the Perinatal Centre, University of Toronto, Canada in 1999 and the Fetal Medicine Centre,
University College London, UK in 2002/3. He becomes an accredited subspecialist in
Maternal Fetal Medicine by the Royal College of Obstetricians & Gynaecologists (UK) in
2003 and by the Hong Kong College of Obstetricians & Gynaecologists in 2005. He is
currently the Consultant Obstetrician in Kwong Wah Hospital, Hong Kong and Honorary
Clinical Associate Professor in Department of O&G, University of Hong Kong.
In June, 2010, he becomes the Chief of Service, Department of O&G, Kwong Wah Hospital.

98

Publications
(resulted from the work of this thesis)
1. Leung WC, Chitayat D, Seaward G, et al. Role of amniotic fluid interphase fluorescence
in situ hybridization (FISH) analysis in patient management. Prenat Diagn
2001;21:327-32.
2. Leung WC, Lam YH, Wong Y, Lau ET, Tang MHY. The effect of fast reporting by
amnio-PCR on anxiety levels in women with positive biochemical screening for Down
syndrome a randomized controlled trial. Prenat Diagn 2002;22:2569.
3. Leung WC, Lau ET, Lao TT, Tang MH. Can amnio-polymerase chain reaction alone
replace conventional cytogenetic study for women with positive biochemical screening
for fetal Down syndrome? Obstet Gynecol 2003;101:856-61.
4. Leung WC, Lau ET, Lao TT, Tang MH. Rapid aneuploidy screening (FISH or QF-PCR):
the changing scene in prenatal diagnosis? Expert Rev Mol Diagn 2004;4:333-7.
5. Leung WC, Waters JJ, Chitty L. Prenatal diagnosis by rapid aneuploidy detection and
karyotyping: a prospective study of the role of ultrasound in 1589 second-trimester
amniocenteses. Prenat Diagn 2004;24:790-5.
6. Leung WC, Lau ET, Lao TT, Tang MHY. Rapid aneuploidy detection (RAD) in prenatal
diagnosis. Encyclopedia of Medical Genomics and Proteomics. Edited by J Fuchs & M
Podda. Publisher: Marcel Dekker, Inc. New York. 2005.
7. Leung WC, Lao TT. Rapid aneuploidy testing, traditional karyotyping, or both? Lancet
2005;366:97-8. Erratum in: Lancet 2005;366:1164.
8. Leung WC, Lau ET, Lao TT, Tang MHY. Rapid aneuploidy testing, traditional
karyotyping, or both, in prenatal diagnosis. HKJGOM 2005;5:33-39.
9. Tse KY, Leung WC, Leung KY, Lee CP, Ng LK, Lau ET, et al. Full karyotyping, rapid
aneuploidy diagnosis or both when invasive prenatal testing is performed for diagnosis of
thalassaemia? Mol Hum Reprod 2006;12:55-9.
10. Leung WC. The clinical application of rapid aneuploidy testing in prenatal diagnosis.
Hong Kong Paediatric Society Education Bulletin 2007 Vol. 14, No. 1, 1-4.
99

11. .
. (Chin J Obstet Gynecol)
2007;42:348-350.
12. Leung WC, Lau ET, Lau WL, Tang R, Wong SF, Lau TK, Tse KT, Wong SF, To WK, Ng
LK, Lao TT, Tang MH; Working Group on Prenatal Diagnosis and Counselling, Hospital
Authority. Rapid aneuploidy testing (knowing less) versus traditional karyotyping
(knowing more) for advanced maternal age: what would be missed, who should decide?
Hong Kong Med J 2008;14:6-13.
13. Leung WC, Lau ET, Ngai C, Lam H, Leung KY, Lee CP, Lao TT, Tang MHY. A
prospective study on the effect of rapid aneuploidy testing (amnio-PCR) on anxiety levels
and quality of life measures in women and their partners with positive Down screening
result. Fetal Diagn Ther 2008;24:165-9.
14. .
. (Chin J Practical Gynecol Obstet) 2008;24:98-101.

100