Human Reproduction vol.13 no.10 pp.2954–2957, 1998
Microscopic investigation of villi from chorionic villous
sampling
Carien Vis1, Ellen Everhardt2, Jan te Velde3 and
Niek Exalto1,4
1Department of Obstetrics and Gynaecology, Spaarne Ziekenhuis,
Haarlem, 2Department of Obstetrics and Gynaecology, Medisch
Spectrum Enschede, and 3Department of Clinical Pharmacology,
Spaarne Ziekenhuis, Haarlem, van Heythuijzenweg 1, 2003 BR
Haarlem, The Netherlands
4To
whom correspondence should be addressed
The aim of the present study was to investigate the
morphology of cytogenetically normal chorionic villi from
chorionic villous sampling (CVS) specimens. This information can serve as a reference for morphological investigation
of cytogenetically abnormal CVS specimens. We were also
interested in any relationship between chorionic villous
architecture and the outcome of pregnancy. In a reference
group (n J 94, normal karyotype and ongoing pregnancies), we observed a considerable variation in villous diameter (range 116–377 µm) and vascular density (range
0.5–6.7 vessels/villus) and a high incidence of morphological
criteria, classically mentioned in relation to chromosomal or
other abnormalities, such as: fibrinoid deposition (74.5%),
trophoblastic layer degeneration (3.2%) and abnormal
proliferation (7.4%), avascular villi (54.2%), stromal
oedema (55.3%), trophoblastic inclusions (23.4%) and fibrosis (23.4%). In the cytogenetic abnormal group (n J 10),
neither the diameter nor the vascular density of the villi
differed from the values observed in the reference group.
In the reference group, we only observed a tendency for
larger birthweights in relation to respectively larger and
more vascularized villi. It is concluded that in CVS specimens, chorionic villous architecture and morphological
criteria do not have any clinical relevance, neither do they
have any important predictive value.
Key words: chorionic villi/CVS/karyotype/vascularization
Introduction
Knowledge on the histomorphology of human chorionic villi
is based on specimens from legal and spontaneous abortions
(Rushton, 1984; Benirschke and Kaufmann, 1990; Lewis and
Benirschke, 1992).
The stem, intermediate and terminal first trimester villi
contain three major structures: (1) an outer layer of invasive
syncytiotrophoblast, (2) a middle layer of cytotrophoblast from
which the syncytiotrophoblast originates, and (3) an inner
mesenchymal centre with vessels and capillaries.
A relation between abnormal karyotype and abnormal chor2954
ionic villi was first presented by Philippe and Boué (1969) in
a classic publication. Chromosomal anomalies like triploidy
have been associated with abnormal size of the villi, trophoblastic inclusions, hydropic degeneration, and/or a reduced
number of vessels. These and other characteristics such as
fibrosis, basophilic staining under the basement membrane and
intervillous fibrinoid depositions, however, have also been
described in relation to prolonged retention in the uterus after
embryonic death. Monosomy (45 XO) is always associated
with some development of an embryo (Canki et al., 1988). The
sensitivity of histology in detecting chromosomal anomalies is
low, but can be improved by adding features of the materno–
embryonic interface (Jauniaux and Hustin, 1992). An untimely
initiation of blood flow in the intervillous space is associated
with arrest of pregnancy and eventual expulsion of the fetus
(Hustin et al., 1990; Jauniaux et al., 1994).
In our study on chorionic villous vascularization, we found
deficient vascularization in cases of embryonic death and
blighted ova (Meegdes et al., 1988). Villous vascularization
appeared unaffected by prolonged postmortem intrauterine
retention. The incidence of trophoblastic degeneration, fibrosis
and hydropic degeneration, however, increased after prolonged
retention. An unexpected finding was that human embryonic
development is possible for as long as 8 weeks, even in the
presence of agenesis or marked hypoplasia of the chorionic
vascular system. This indicates the possibility of embryonic
development in suboptimal conditions of nutrition and oxygen
exchange (Exalto, 1995). With the help of CD34 immunohistochemistry, we demonstrated that vasculogenesis of normal
chorionic villi is characterized by maturation of luminized
vessels from haemangioblastic cords and margination to peripherally located vessels at the end of organogenesis (te Velde
et al., 1997).
Recent studies have also concluded that, except for triploidy,
villous morphology is an insensitive and inaccurate indicator
of chromosomal abnormalities (Fox, 1993; Lijnschoten et al.,
1993b). This is partly due to considerable inter-observer
variation (Lijnschoten et al., 1993a).
In studies concerning chorion villous sampling (CVS),
morphological investigation of the villi is almost completely
lacking. In a prospective cytogenetic and morphological study
of CVS specimens, villi from proven chromosomal anomalies
were characterized by hydropic degeneration and/or stromal
fibrosis in four out of nine cases, including three mosaics
(Rüschoff et al., 1989). Studies concerning the morphology of
CVS specimens in relation to the outcome of pregnancy are
also lacking.
The aim of our present prospective study was systematically
to investigate the incidence of various pathologic changes in
© European Society for Human Reproduction and Embryology
Morphology of villi from CVS specimens
cytogenetically normal chorionic villi from CVS specimens.
The information can serve as a reference for morphological
investigation of cytogenetically abnormal CVS specimens. We
were also interested in any relationship between chorionic
villous morphology in cytogenetically normal CVS specimens
and the outcome of pregnancy.
Table I. Types of chromosomal anomaly in the cytogenetically abnormal
group (n 5 10)
Translocation carrier
46,
46,
46,
46,
XX, t (13q14q) mat
XX, t (13q14q) mat
XY, t (2;3) (q21;p24) mat
XY, t (2;3) (q21;p24) pat
Inversions
46, XX, inv (9) (p11;q13)
46, XY, inv (9) (p11;q13)
Materials and methods
Over a period of 1.5 years, 127 consecutive CVS specimens were
prospectively collected for combined cytogenetic and microscopic
investigation. The indications for CVS were advanced maternal age
of 36 years and more (n 5 103) and other risk factors for congenital
anomalies (n 5 24). The CVS specimens were obtained by ultrasound
guided transcervical aspiration from the chorion frondosum between
9 and 13 weeks amenorrhoea. Gestational age, expressed as menstrual
age in days, was determined more accurately on the day of CVS by
measuring the crown–rump length according to the reference values
of Robinson and Flemming (1975).
Each patient received an explanation and written information about
the study and signed an informed consent. The study protocol was
approved by the Medical Ethical Committee of both hospitals.
After sampling and preparing for cytogenetic investigation, a small
amount of chorionic tissue was fixed in 2% buffered formalin for
microscopic investigation. After being embedded in paraffin, the
specimens were serially sectioned (4 µm) and stained with haematoxylin–eosin and with periodic acid Schiff.
Microscopic sections were examined by three authors (magnification 3100, standard micrometer) without knowledge of the source
of the villi or the results of karyotyping. Only preparations containing
ù10 cross-sections through villi were accepted. After a crude examination, 10 villi per section were randomly taken for systematic microscopic examination by the first author, using a checklist.
The diameter was measured for each of 10 villi per specimen. The
mean villous diameter was defined as the mean diameter of 10 villi.
In these villi, the number of vessels was also counted. To this end, a
structure was considered a vessel only when it contained embryonic
blood cells. The vascular density was defined as the total number of
vessels per 10 villi. The presence or absence in the intervillous space
of fibrinoid depositions and or decidua was noted. For each of the
10 villi, the trophoblast layer was examined for abnormal degeneration,
abnormal proliferation and/or basophilic staining under the basement
membrane. The stromal area of the villi was examined for the
presence or absence of oedema, trophoblastic inclusions and/or fibrosis
(Rushton, 1984).
After finishing the microscopic examination of all available specimens, the clinical data were added. Only specimens from pregnancies
with a normal karyotype and not ending in an abortion before 16
weeks menstrual age served as a reference group. In cases with
abnormal karyotype, the results for various parameters were studied
against the background of the results in the reference group.
For patients in the reference group, we collected obstetric information such as date of birth, birthweight and birthweight percentiles.
For the assessment of correlation between various morphological
and clinical data, we performed a linear regression analysis. Correlation coefficients were calculated and tested for significance according
to the Pearson method
Results
A total of 127 patients entered the study. Cytogenetic investigation of the villi succeeded in all cases and revealed an
Mosaic trisomy
46, XX / 47, XX, 13
46, XX / 47, XX, 113
46, XY / 47, XY, 18
Trisomy
47, XXX
Figure 1. Distribution of all individual values of the mean villous
diameter in the reference group (n 5 94) and the cytogenetically
abnormal group (n 5 10), arranged in order of increasing
magnitude.
abnormal karyotype in 12 patients (9.4%) and a normal
karyotype in 115 (90.6%).
Morphological examination, however, was not always possible. Three specimens did not arrive at the department properly
(2.4%) and 14 contained fewer than 10 villi (11.0%). As a
result, only 110 specimens were available for morphological
examination. Three spontaneous and three legal abortions
reduced the study group to 104, leaving a reference group
(normal karyotype, ongoing pregnancy, n 5 94) and a cytogenetic abnormal group (abnormal karyotype, n 5 10) available
for analysis.
The different types of chromosomal anomaly in the cytogenetic abnormal group are listed in Table I. Cytogenetic
investigation in the reference group revealed 45 normal female
and 49 normal male karyotypes (female/male ratio 5 0.92).
The mean duration of pregnancy at the time of CVS was
74.4 days (range 63–93; SD 5.5) in the reference group
and 71.1 days (range 62–81; SD 6.4) in the cytogenetically
abnormal group.
In the reference group, there was a considerable variation
in the mean villous diameter (mean 204 µm; range 116–377,
SD 55) as well as in the mean number of vessels (mean 3.1;
range 0.5–6,7; SD 1.2). The data are presented graphically
2955
C.Vis et al.
Figure 2. Distribution of all individual values of the mean number
of vessels in the reference group (n 5 94) and the cytogenetically
abnormal group (n 5 10), arranged in order of increasing
magnitude.
Figure 3. Relationship between the mean villous diameter and the
mean number of vessels in the reference group (r 5 0.314; n 5 94;
P , 0.01).
Table II. Incidence and distribution of morphologic criteria (n 5 94; 10
villi per specimen)
Number of villi
0
1
2
3
4
5
6
7
8
9
10
Decidual cells
Fibrinoid
depositions
Abnormal
degeneration
Abnormal
proliferation
Basophilic
staining
Avascular villi
Oedema
Trophoblastic
inclusions
Fibrosis
53 17 15 7
24 27 19 8
2
5
–
2
–
4
–
1
–
2
–
–
–
–
87
7 –
–
–
–
–
–
–
–
–
91
3 –
–
–
–
–
–
–
–
–
43 11 19
8 5
2
4
1
2
–
–
43 30 10
42 33 9
72 14 5
7 2
6 2
2 –
0
2
1
2
–
–
–
–
–
–
–
–
–
–
–
–
–
–
72 15
1 –
1
–
–
–
–
–
5
and served as background for the values observed in the
cytogenetically abnormal group (Figures 1 and 2).
As we expected from reading the literature, we did not
observe any relationship between both parameters and the
2956
Figure 4. Relationship between the birthweight percentiles and the
mean villous diameter in the reference group (r 5 0.172; n 5 91;
not significant).
Figure 5. Relationship between the birthweight percentiles and the
mean number of vessels in the reference group (r 5 0.129; n 5 91;
not significant).
duration of pregnancy (Meegdes et al., 1988). The only
relationship was between the diameter and the number of
vessels (r 5 0.314; n 5 94; P , 0.01), with a tendency for
larger villi to contain more vessels (Figure 3).
The incidence and distribution of the other morphological
criteria is summarized in Table II. Decidual tissue was present
in nearly half of the specimens (41/94; 43.6%). We frequently
encountered fibrinoid depositions in the intervillous space (70/
94; 74.5%). Abnormal proliferation and degeneration of the
trophoblastic layer, however, were only seen in respectively
3.2% (3/94) and 7.4% (7/94) of cases. Basophilic staining
under the basement membrane was a common finding (51/94;
54.2%). In specimens with this phenomenon, it was usually
seen in two or more villi, suggesting a systematic distribution.
Morphological abnormalities of the stromal part of the
villi, traditionally described in combination with chromosomal
abnormalities, were also seen frequently in our reference
group. Avascular villi (51/94; 54.2%), oedema (52/94; 55.3%),
trophoblastic inclusions (22/94; 23.4%) and fibrosis (22/94;
23.4%) were commonly observed.
Morphology of villi from CVS specimens
Two patients were lost to follow up and in another the
birthweight measurement was not performed accurately. As a
result, calculated data for birthweight percentiles were only
available in 91 cases. In this group, we observed a relationship
between the birthweight percentiles and the mean villous
diameter (r 5 0.172; n 5 91; P 5 0.1) and between the
birthweight percentiles and the mean number of vessels (r 5
0.129; n 5 91; P . 0.2), with a tendency for larger birthweights
in relation to larger villi (Figure 4) and more vascularized villi
(Figure 5) respectively, without, however, reaching a level of
statistical significance. We did not observe any relation between
birthweight percentiles and all other morphological parameters
studied in the reference group.
Discussion
In this study, the histomorphology of chorionic villi was studied
in CVS specimens with normal and abnormal karyotype. In
the reference group, with a normal karyotype, we observed a
wide variation in the mean villous diameter as well as in the
mean villous vascularization. This variation was also described
in an earlier study on chorionic villous vascularization
(Meegdes et al., 1988). We did not observe a relationship
between the diameter or the vascularization and the duration
of pregnancy. There was a tendency for an increased vascular
density only in larger villi.
In the cytogenetic abnormal group, the diameter of the villi
and the vascular density did not differ from the data in the
reference group. This confirms earlier findings in the literature
concerning the difficulties in recognizing chromosomal abnormalities by histomorphological examination (Fox, 1993;
Lijnschoten et al., 1993b).
Furthermore, other ‘classical’ criteria for chromosomal
abnormalities such as avascular villi, oedema, and trophoblastic
inclusions were encountered in the reference group so commonly that they should not be described as abnormalities.
Fibrinoid depositions, mentioned in relation to lupus anticoagulant syndrome, were seen very frequently in our reference
group. Abnormal proliferation of the trophoblastic layer was
seen in only a few villi per specimen, whereas trophoblastic
degeneration, usually seen after a long retention time in cases
of missed abortion, was more frequently observed. It is
interesting that after degeneration, regeneration of a new
syncytiotrophoblast layer has been observed during organ
culture of first trimester villi (Palmer et al., 1997).
In the reference group, we only observed a relationship
between the birthweight percentiles and the mean villous
diameter and the mean number of vessels respectively. Other
parameters were not related to the outcome of pregnancy.
Therefore, it can be concluded that maternal factors are more
important than trophoblastic architecture and fetal vascular
density.
In conclusion, a large variety of morphological criteria,
‘classically’ mentioned in relation to chromosomal or other
abnormalities (Philippe and Boué, 1969; Rushton, 1984), can
be observed rather frequently during morphological examination of normal chorionic villi shortly after organogenesis. They
do not seem to have any clinical relevance, neither do they
seem to have important predictive value.
References
Benirschke, K. and Kaufmann, P. (eds) (1990) Pathology of the Human
Placenta. Springer-Verlag, Heidelberg.
Canki, N., Warburton, D. and Byrne J. (1988) Morphological characteristics
of monosomy X in spontaneous abortions. Ann. Genet., 31, 4–13.
Exalto, N. (1995) Early human nutrition. Eur. J. Obstet. Gynecol. Reprod.
Biol., 61, 3–6.
Fox, H. (1993) Histological classification of tissue from spontaneous abortions:
a valueless exercise? Histopathology, 22, 599–600.
Hustin, J., Jauniaux, E. and Schaaps J.P. (1990) Histological study of
the materno–embryonic interface in spontaneous abortion. Placenta, 11,
477–486.
Jauniaux, E. and Hustin, J. (1992) Histologic examination of first trimester
spontaneous abortions: the impact of materno–embryonic interface features.
Histopathology, 21, 409–414.
Jauniaux, E., Zaidi, J., Jurkovic, D. et al. (1994) Comparison of colour
Doppler features and pathological findings in complicated early pregnancy.
Hum. Reprod., 9, 2432–2437.
Lewis, S.H. and Benirschke, K. (1992) Placenta. In Sternberg, S.S. (ed.),
Histology for Pathologists. Raven Press, New York, pp. 835–863.
Lijnschoten, van G., Arends, J.W., Fuente de la A.A. et al. (1993a) Intraand inter-observer variation in the interpretation of histological features
suggesting chromosomal abnormality in early abortion specimens.
Histopathology, 22, 25–29.
Lijnschoten, van G., Arends, J.W., Leffers, P. et al. (1993b) The value of
histomorphological features of chorionic villi in early spontaneous abortion
for the prediction of karyotype. Histopathology, 22, 557–563.
Meegdes, B.H.L., Ingenhoes, R., Peeters, L.L. and Exalto, N. (1988) Early
pregnancy wastage: relation between chorionic vascularization and
embryonic development. Fertil. Steril., 49, 216–220.
Palmer, M.E., Watson, A.L. and Burton, G.J. (1997) Morphological analysis
of degeneration and regeneration of syncytiotrophoblast in first trimester
placental villi during organ culture. Hum. Reprod., 12, 379–382.
Philippe, E. and Boué, E. (1969) Le placenta des aberrations chromosomiques
létales. Ann. Anat. Pathol., 14, 249–266.
Robinson, H. and Flemming, J.E.E. (1975) A critical evaluation of sonar
crown–rump length measurements. Br. J. Obstet. Gynaecol., 82, 702–710.
Rüschoff, J., Köhler, A., Chudoba, I. and Steuber, E.D. (1989) Investigations
of chorionic villi after chorionic villus sampling (CVS). Hum. Genet., 81,
329–334.
Rushton, D.I. (1984) The classification and mechanisms of spontaneous
abortion. Perspect. Pediatr. Pathol., 8, 269–287.
Velde, E.A. te, Exalto, N., Hesseling, P. and Linden, H.C. van der (1997)
First trimester development of human chorionic villous vascularization
studied with CD34 immunohistochemistry. Hum. Reprod., 12, 1577–1581.
Received on February 13, 1998; accepted on July 1, 1998
2957