Human Reproduction, Vol.24, No.8 pp. 1825– 1833, 2009
Advanced Access publication on May 8, 2009 doi:10.1093/humrep/dep176
ORIGINAL ARTICLE Embryology
Age determination enhanced by
embryonic foot bud and foot plate
measurements in relation to Carnegie
stages, and the influence of maternal
cigarette smoking
M.C. Lutterodt1,2,4, M. Rosendahl 1, C. Yding Andersen 1,
S.O. Skouby2,3, and A.G. Byskov 1
1
Laboratory of Reproductive Biology, The Juliane Marie Centre for Women, Children and Reproduction, Copenhagen University Hospital,
Rigshospitalet, Copenhagen, Denmark 2Department of Obstetrics & Gynaecology, Frederiksberg Hospital, Faculty of Health Sciences,
Copenhagen University, Copenhagen, Denmark 3Department of Obstetrics & Gynaecology, Herlev Hospital, Faculty of Health Sciences,
Copenhagen University, Copenhagen, Denmark
4
Correspondence address. Laboratory of Reproductive Biology, section 5712, Copenhagen University Hospital, Rigshospitalet,
Blegdamsvej 9, DK-2100 Copenhagen, Denmark. Tel: þ45-35455830; Fax: þ45-35455824; E-mail: [email protected]
background: Reliable age determination of first-trimester human embryos and fetuses is an important parameter for clinical use and
basic science. Age determination by ultrasound or morphometric parameters of embryos 4–6 weeks post conception (p.c.) have been questioned, and more accurate methods are required. Data on whether and how maternal smoking and alcohol consumption influence embryonic
and fetal foot growth is also lacking.
methods: Embryonic tissue from 102 first-trimester legal abortions (aged 35–69 days p.c.) were collected. All women answered a questionnaire concerning smoking and drinking habits, and delivered a urine sample for cotinine analysis. Embryonic age was evaluated by vaginal
ultrasound measurements and by post-termination foot length and compared with the Carnegie stages.
results: Foot bud and foot plate were defined and measured as foot length in embryos aged 35–47 days p.c. (range 0.8–2.1 mm). In
embryos and fetuses aged 41–69 days p.c., heel-toe length was measured (range 2.5– 7.5 mm). We found a significant linear correlation
between foot length and age. Morphology of the feet was compared visually with the Carnegie collection, and we found that the mean
ages of the two collections correlated well. Foot length was independent of gender, Environmental Tobacco Smoke, maternal smoking
and alcohol consumption.
conclusion: Foot length correlated linearly to embryonic and foetal age, and was unaffected by gender, ETS, maternal smoking and
alcohol consumption.
Key words: foot length / first-trimester / embryonic age / fetus / smoking
Introduction
Accurate age determination of the human embryo and fetus with regard
to normal growth patterns is a matter of interest to both basic and clinical science. Several methods have been developed and employed during
the last century, with the Carnegie staging method defined by George
L. Streeter being the most outstanding (O’Rahilly, 1972; Skidmore,
1977; O’Rahilly and Müller, 1987). Later, ultrasound determination
became the most widely used non-invasive method for measuring the
crown-rump length (CRL) (Robinson and Fleming, 1975). Almost a
century ago, a linear correlation between age and foot length was
observed in 704 human fetal specimens spanning an age of around
50 days post conception (p.c.) until birth (Streeter, 1920). There are
few studies of embryos younger than 50 days p.c. on a very limited
number of embryos (e.g. Streeter included three such very young
embryos) (Streeter, 1920; Evtouchenko et al., 1996).
In second and third trimester fetuses, several studies have demonstrated improved age determination by including the CRL and foot
& The Author 2009. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected]
1826
length, i.e. heel-toe length, and other parameters of the extremities
(Hern, 1984; Munsick, 1984; Mercer et al., 1987; Mhaskar et al.,
1989; Mandarim-de-Lacerda, 1990; de Vasconcellos et al., 1992;
Hata et al., 1996; Drey et al., 2005). However, foot length in small
embryos between week 4 and 7 p.c. (corresponding to Carnegie
stages 13– 18) is difficult to define, since the heel and toe have not
yet differentiated and appear only as a foot bud or foot plate
(O’Rahilly, 1979; O’Rahilly and Müller, 1987). However, foot bud
and foot plate measurements performed on fresh tissue could be an
easily acquired measurement to complement the Carnegie staging
method.
Accurate staging of human embryos is becoming increasingly
important in studies on the effect of environmental factors, e.g. cigarette smoking and alcohol consumption, on embryonic development
(Kramer, 1987; Cliver et al., 1995). Smoking has previously been
associated with growth restriction and lower birthweight in infants
born of mothers who smoked during the 3rd trimester or throughout
the entire pregnancy, but not of mothers who smoked during the first
or first and second trimesters (Lieberman et al., 1994). On the contrary, smoking cessation before gestational week 32 did not completely prevent reduction in crown-heel length in infants, although
cessation prevented reduction in birthweight (Lindley et al., 2000).
In a systematic review and meta-analysis, passive smoke exposure
(Environmental tobacco smoke, ETS) to non-smoking pregnant
women was found to reduce the mean birthweight of their children
by 33 g or more (Leonardi-Bee et al., 2008). Furthermore, a review
on low to moderate maternal alcohol intake during pregnancy
reports no consistent significant effect on human birthweight (Henderson et al., 2007), although general growth and skeletal development
has been shown to be impaired in offspring of rats after ethanol
exposure (Snow and Keiver, 2007). Whether smoking, ETS and
alcohol exposure during early pregnancy have any effect on the
growth of human embryos is potentially significant, but difficult to
evaluate due to the inaccessibility of current methods for embryonic
staging to non-specialists.
The aims of this study were (1) to compare mean embryonic ages
to the Carnegie collection with respect to morphology of the
embryonic foot, (2) to explore whether measurements of foot bud
and foot plate can act as extra criteria, thereby improving the Carnegie
staging system, (3) to evaluate the distribution of foot length (foot bud,
foot plate and heel-toe) relative to embryonic and fetal age determined by ultrasound CRL measurements and compare these data
to those of previous studies, (4) to investigate whether the growth
of the embryonic foot is affected by gender, ETS, maternal smoking
and/or alcohol consumption.
Materials and Methods
Embryonic tissue was obtained throughout 2006 and 2007 from 154
women undergoing legal abortions performed by the same surgeon.
Patients
Inclusion criteria were: healthy first-trimester women, age .18 years,
who were referred by a physician for elective termination of the pregnancy. Exclusion criteria were: women on permanent medication,
women who suffered from any chronic disease, or women who were
dependent on an interpreter. Most of the 154 women who fulfilled the
Lutterodt et al.
criteria were of Caucasian origin, and a few were of African origin. All
women lived in the area of Copenhagen, Denmark where the air pollution
is classified as slightly above the average for Denmark (Danish National
Environmental Research Institute, http://luft.dmu.dk). The women were
18 – 40 years old (27.5 years + 0.6 days (mean + SEM)).
To investigate whether the growth of the embryonic foot is affected by
gender, ETS, maternal smoking and/or alcohol consumption, all women
completed a questionnaire about their smoking and drinking habits
during the ongoing pregnancy. Smoking habits were divided into groups
of smokers and non-smokers. In addition all women delivered a urine
sample for analysis of cotinine content on the day of surgery. ETS was
defined as the number of hours per day of exposure to passive
smoking, and alcohol consumption, as the average daily volume (ADV,
drink/day), where 1 drink equals 1 glass of wine or 1 beer or 2 cl of
liquor. Oral and written information was provided and informed consent
was obtained from all participants, according to and approved by ‘The
Regional Committee on Biomedical Research Ethics, Copenhagen and
Frederiksberg Counties’ (H-KF (01) 258206).
Determination of embryonic age
The embryonic age was determined by vaginal ultrasound examinations,
measuring the length of the embryo from crown to rump (CRL) in a
midline sagital view of each embryo using the Robinson growth curve
(Robinson and Fleming, 1975). This particular growth curve has been
shown to provide one of the most precise estimates of embryonic age
determination when compared with known embryonic ages obtained by
IVF pregnancies (Grange et al., 2000). The gestational age, ranging from
the 7th to 12th gestational week, was converted to embryonic age in
days p.c. by a deduction of 2 weeks (equal to 5th to 10th embryonic
week p.c. or 35 –70 days p.c.). Thus, in the present study the term ‘age
in days p.c.’ was used, referring to ‘the best estimate of the day of conception’ relying on the ultrasound measurements. The inter-observer
variability of CRL measurements was kept low; the last 60 of the 154 abortions were scanned vaginally by two independent doctors showing a good
correlation, data not shown.
Pre-surgery/termination procedures
All women were given 0.4 mg misoprostol (Cytotecw Pfizer) orally the
evening before and were anaesthetized just prior to the abortion. The
abortion was performed according to the routine procedures for legal
suction abortion in the gynecology department without additional risk to
the participating women. In order to prevent severe damage to the
embryo, the normal procedure was slightly modified by using a manual
instead of a mechanical vacuum (Nauert and Freeman, 1994). The procedure was monitored with abdominal ultrasound equipment (B-K
Medical portable scanner Merlin, including a multifrequency transducer)
during the whole session, and the heart activity of the embryo was
recorded. The cervical canal was dilated and a curette (Berkely Medevices
Inc., sterile vacuum curettes rigid 8 – 12 mm) was inserted. A 50 ml syringe
was connected to the curette and a vacuum was carefully applied manually, guided by abdominal ultrasound. The collected tissues were placed
in a sterile cup and, within five minutes, transferred to a sterile 50 ml
tube with 20 ml of culture media (D-MEM, cat no. 41965-039, Gibco, Invitrogen). All handling of the tissue was performed under sterile conditions.
The tube was kept at 36 + 18C and transported to the laboratory for
further processing within 112 – 312 h.
Handling of the embryonic tissue
The embryo was identified and dissected under a stereomicroscope in a
petri dish with 0.9% NaCl. A single person measured the foot length
either with translucent graph paper placed underneath the petri dish or
1827
Foot length as a parameter for fetal age
by using the ruler incorporated in the stereomicroscope. In embryos with
developed heel and toe, the foot length was measured as the distance
between the posterior surface of the heel and the distal ending of the
foot given by the longest toe according to Streeter (1920). Younger
embryos, in which only a foot bud had developed, the length was
measured from the distal tip of the foot bud to the attachment (‘cleft’)
to the body wall. In slightly older embryos, the distal part of the foot
bud had acquired a plate-like shape, i.e. the foot plate. At the proximal
end, the foot plate was narrowed by a cleft where the heel and leg
would ultimately have developed. The foot plate length was measured
from the distal tip to the cleft (Fig. 1). Finally, a small piece of embryonic
tissue was snap-frozen in liquid nitrogen for later determination of the
chromosomal sex, which was performed using the polymerase chain
reaction with X – Y homologous primers (Nakahori et al., 1991).
Comparison of embryonic feet with those of
the Carnegie collection
The morphology of each individual foot was examined carefully and compared with the Carnegie staging system in conjunction with comments by
O’Rahilly and Müller (1987) concerning the different stages of the hind
limb development. The comparisons were conducted without knowing
the embryonic age ranges of the different Carnegie stages or the
ultrasound-determined ages of collected material. A total of 72 embryos
were staged by foot bud, foot plate and heel-toe according to the
Carnegie collection, whereas the remaining 30 fetuses were too old
(i.e. exceeded stage 23 in the Carnegie collection).
Cotinine assay
A urine sample from each patient was subject to a cotinine assay. Cotinine
is a metabolite of nicotine, which appears to be a specific and sensitive
marker for nicotine from tobacco exposure. Cotinine is detectable in
urine 20 h after smoking (Benowitz, 1999; Pickett et al., 2005). The
present study used the cotinine measurements to verify the reliability of
the pregnant women’s own report on their smoking habits and ETS.
The urine samples were stored at 2208C until assayed. Cotinine concentrations were assessed by ELISA (Cotinine, cat no. CO096D, Calbiotic,
CA, USA).
Statistical analysis
Statistical analysis was performed using SPSS 15.0 Chicago, IL, USA. A two
sided P-value , 0.05 was considered statistically significant. Data were
analyzed using parametric analyses complying with requirements for
normal distribution, individuality and homogeneity of variances. Linear
regression analysis was performed to evaluate (1) the relation between
cotinine in urine and smoking habits reported in the questionnaires (2)
the correlation between foot length and embryonic and fetal age. Multiple
regression analysis was used to (1) evaluate foot growth in relation to
environmental exposures including one interaction term: smoking
exposure and embryonic/fetal age (2) compare our data with earlier
reports in the field. The students t-test was performed to compare
‘mean age’ values of the Carnegie collection with those found in this
study. These and the regression analysis results are presented with 95%
confidence intervals. ‘Mean foot lengths’ of the present study are
presented with standard deviations.
Results
Of the 154 women whose pregnancies were terminated, embryonic
or fetal tissue was not collected in 32 cases. In addition, 20 pregnancies were excluded: 11 cases without foot identification, four cases
Figure 1 Photographs of six human embryonic feet from 36 – 50 days p.c. showing the new parameters used to measure foot length of young human
embryos in relation to the Carnegie stages.
Foot bud: measurements ($) performed from distal tip of the foot bud to the ‘cleft’ or attachment to the body wall, indicated by the dotted line (embryos 36 and 39 days
p.c.). Foot plate: measurements ($) performed from distal tip of the foot plate to the cleft (dotted line) (embryos 42 and 44 days p.c.). Heel-toe: measurements ($)
performed from distal tip of the longest toe to posterior surface of the heel (dotted line) (embryos 47 and 50 days p.c.). The translucent graph paper below each foot bud
are aligned enabling accurate comparison of embryonic foot lengths. aRefers to the Carnegie stages to which each of the embryonic or fetal feet has been correlated.
1828
Lutterodt et al.
with missing ultrasound measurements, three twin pregnancies, one
case of water pipe smoking and one embryo with macroscopic malformations. Thus, foot length was measured in 102 cases.
In Fig. 1 examples of the developing foot shows how the length
(illustrated by $) of the foot bud, foot plate and heel-toe was
measured in each case. Furthermore, the embryonic age, given in
days p.c., is shown in relation to the corresponding Carnegie stages.
The two foot buds of embryos at age 36 and 39 days p.c. are
shaped like paddles and correspond to Carnegie stages 15 and 16,
respectively. The subsequent two feet of embryos, aged 42 and
44 days p.c. (corresponding to Carnegie stages 17 and 18), have
passed the bud stage and gone through plate formation. The oldest
has developed toe rays. The last two feet (aged 47 and 50 days
p.c., corresponding to Carnegie stages 20 and 21) have visible heels
and toes, the youngest with notches between toe rays and the
oldest with webbed toes.
Table I presents the nine evaluated Carnegie stages (stages 15 –23).
Further, the number of specimens that were visually classified as a
certain Carnegie stage is shown (ranging from 2 to 20 specimens
per stage). The ‘mean embryonic age’ of the collected material was,
for each stage, calculated from the embryonic ages obtained by ultrasound measurements. The ‘mean ovulation ages’ of the Carnegie collection was calculated by the ranges shown in brackets (which were
drawn from the Carnegie collection). The mean age values in this
study showed good correlation with mean age values of the Carnegie
stages (P ¼ 0.444) (Fig. 2). The regression equation for the Carnegie
mean ages was: y ¼ 0.36x þ 1.8 (CI ¼ 0.313–0.406, R 2 ¼ 0.98) and
for our data: y¼0.44x 2 0.76 (CI ¼ 0.313–0.558, R 2 ¼ 0.91).
Further, the mean foot length calculated for each embryonic stage in
this study is given in Table I.
The increase in the lengths of the foot buds and foot plates from 0.8
to 2.1 mm between 35 and 47 days p.c. is shown in Fig. 3. The smallest heel-toe length was 2.5 mm as seen in two embryos aged 44 and
49 days p.c. The two youngest embryos with heels were 41 days p.c.
with foot lengths of 2.6 and 2.9 mm. The largest heel-toe length,
which was observed in the oldest embryo of 69 days p.c. was
7.5 mm. Linear regression analysis of the foot buds and foot plates
in relation to embryonic age was found to be highly significant i.e.
Foot length ¼ 0.07x 2 1.3 (CI ¼ 0.05 –0.11, R 2 ¼ 0.34, P , 0.001).
Similarly, the equation applied to the heel-toe length only was also significant: foot length ¼ 0.17x 2 4.8 (CI ¼ 0.15 –0.19, R 2 ¼ 0.75, P ,
0.001). In contrast, no statistical significant correlation was found
when regression analysis of the square and cubic root of the independent value, was performed. The comparison of the two regression
lines revealed a significant difference between their slopes (P , 0.003).
As the cotinine concentrations in urine correlated well with the
women’s reported smoking habits (P , 0.001), the reported
smoker/non-smoker group numbers were chosen for the analysis.
Linear regression analysis showed that both embryonic foot bud and
foot plate measurements were not influenced by maternal smoking
habits (P ¼ 0.114). Figure 4 illustrates the two regression lines for
smokers: y ¼ 0.09x 2 1.9 (CI ¼ 0.025 –0.145, R 2 ¼ 0.318), and nonsmokers: y ¼ 0.06x 2 1.0 (CI ¼ 0.024–0.102, R 2 ¼ 0.443). The corresponding regression lines for heel-toe length are presented in Fig. 4;
the regression equation for smokers: y ¼ 0.17x 2 5.0 (CI ¼ 0.139 –
0.206, R 2 ¼ 0.776) and for non-smokers: y ¼ 0.17x 2 4.7 (CI ¼
0.129–0.210, R 2 ¼ 0.724). Heel-toe length was not influenced by
maternal smoking (P ¼ 0.473). Finally, both foot bud and plate and
heel-toe length were apparently not influenced by the following
variables: gender, smoking habits, ETS, alcohol consumption and
the interaction between smoking habits and embryonic/fetal age
(P ¼ 0.078–0.909).
Having determined that foot length was independent of the above
mentioned variables, we compared our data to the two largest published sets of data on foot length and embryonic/fetal age (Fig. 5).
Bossy (n ¼ 100) (Bossy and Katz, 1964) and Streeter (n ¼ 704) (Streeter, 1920) collected their data from women with unknown social
history (e.g. drinking and smoking habits), and all their tissue was
fixed. Foot length measurements of their embryos (47 –70 days
p.c.), were compared with our data (Bossy: n ¼ 17 and Streeters:
n ¼ 75). Menstrual ages in Streeter’s data and ours were corrected
by subtracting 2 weeks, whereas Bossy apparently calculated embryonic age in days p.c. Furthermore, our data was compared with
another recent study on ‘fresh’ tissues in which no raw data was
Table I Comparison of the present study with the Carnegie collection.
Carnegie stages
No. of specimens
present study
Age in days p.c.a present
study mean (range)
Carnegie ovulation
age in days mean (range)
Foot length in mm
present study Mean + SD
.............................................................................................................................................................................................
15
2
39.0 (36–42)
36.5 (35–38)
0.9 + 0.14
16
6
38.7 (35–42)
39.5 (37–42)
1.2 + 0.16
17
12
40.6 (36–44)
41.0 (42–44)
1.5 + 0.08
18
20
43.6 (38–49)
46.0 (44–48)
2.1 + 0.34
19
5
45.4 (41–49)
49.5 (48–51)
2.8 + 0.07
20
3
45.3 (43–47)
52.0 (51–53)
2.9 + 0.12
21
3
47.0 (44–50)
53.5 (53–54)
3.2 + 0.06
22
10
51.8 (49–56)
55.0 (54–56)
3.8 + 0.33
23
11
56.9 (51–64)
58.0 (56–60)
4.7 + 0.41
The number of specimens of the present study listed in relation to the nine Carnegie stages (stage 15 –23) (O’Rahilly and Müller, 1987) covering the embryonic period of the present study.
Each specimen was correlated morphologically to one of the nine Carnegie stages. The ‘mean foot length’ presented with standard deviations and ‘mean age in days p.c.’ presented with
ranges are calculated from data obtained in this study, for each of the nine stages. Finally, the ‘mean ovulation age’ of the Carnegie collection is calculated by the range values shown in
brackets.
a
Age in days p.c. corresponds to embryonic age (determined by CRL ultrasound measurements and reduced by 2 weeks).
Foot length as a parameter for fetal age
1829
Figure 2 Morphological comparison of embryos from the Carnegie collection and our material.
The mean age values from each stage (see Table I) are plotted in relation to the Carnegie stages and to our values. The ‘mean age’ values of the two studies showed good
correlation (P ¼ 0.444).
Figure 3 Scatter plot and regression lines of the present observations of 102 embryos showing the correlation between embryonic age in days p.c.
and foot length in millimeter.
The small gap in foot measurements (between the horizontal dotted lines) from 2.1 to 2.5 mm covers the ‘Grey zone’ (vertical dotted lines) from day 41 p.c. to 47 p.c.,
and signifies the differences between the two ways of measuring when the cleft of the foot plate or the heel as end-point are used. The gradients of the regression lines
differ significantly (P , 0.003).
1830
Lutterodt et al.
Figure 4 Scatter plot showing the correlation between foot length and embryonic/fetal age for smokers and non-smokers.
The regression lines for foot bud and foot plate are shown for smokers and non-smokers from 35 to 47 days p.c. The regression lines for heel-toe length covers the period
from 41 to 69 days p.c. Embryonic and fetal foot lengths were apparently not influenced by gender, smoking habits, ETS, alcohol consumption nor the interaction between
smoking habits and embryonic/fetal age (P ¼ 0.078– 0.909).
shown (n ¼ 1099, Drey et al., 2005). Drey related age to ‘days of gestation’. The regression line in Drey et al. was corrected from ‘days of
gestation’ [Foot length ¼ 0.458x 2 30.3 (x ¼ days of gestation),
(R 2 ¼ 0.92)] to ‘days p.c.’ by subtracting 2 weeks from the gestational
age and calculating a ‘new’ intercept corresponding to days p.c. which
resulted in: foot length ¼ 0.458x 2 23.6 (x ¼ days p.c.). As Drey et al.
did not present any raw data, statistical comparison with the three
other data sets was not possible. Figure 5 shows the four regression
lines covering the period from 47–70 days p.c. Streeter’s, Bossy’s
and our sets of data, were compared with regard to linear regression;
these were significantly different (P , 0.001).
Discussion
The ‘mean ages’ of the specimens obtained for this study correlated
with those of the Carnegie collection, when based on a visual comparison of embryonic feet. This method of measuring the embryonic
foot bud, foot plate and heel-toe-lengths, therefore seems to be a
reliable way of determining embryonic age that may enhance the precision of the Carnegie staging. Further, maternal smoking did not influence the relation between foot growth and age determination in the
first-trimester.
The definition of where the ‘cleft’ appears to be located and delimits
both foot bud and foot plate is illustrated in Fig. 1. This is based on the
fact that it is not morphologically possible to recognize where the heel
will develop in the shaft between the cleft and the body wall, i.e. the
foot bud and foot plate do not represent the entire anlage of what
becomes the foot, as seen in later stages of development. Evtouchencko
et al. (1996) described a mathematical model for estimating embryonic
and fetal age with measurements of up to five limb parameters (e.g.
heel-toe length) or greatest length (GL). Obviously, this model can
only be used in embryos where GL has not been damaged, which is
often the case with suction abortion as described in this study, or in
embryos in which the extremities have developed. However, in our
case only parts of the body wall to which the limbs were attached was
usually present and used for measuring foot buds and foot plates. As a
result, we find that foot bud and foot plate are more applicable parameters than GL in young embryos obtained by suction abortion.
The data presented in Table I rely on a visual and blind comparison
between our material and that used in the Carnegie collection.
However, the material in the Carnegie collection was fixed, whereas
ours was fresh. Potentially, this introduces a bias, but we assume
that a solely visual comparison between the fresh and fixed tissue circumvents this problem. A single measurement was performed on all
feet in this study; a fact to be taken in to consideration. Further,
some of the stages rely on mean values obtained from only two specimens, although others include up to 20 specimens. However, the
comparison of embryonic ages in this study correlated well with the
Foot length as a parameter for fetal age
1831
Figure 5 The regression lines for foot length in correlation to embryonic and fetal age from 47 to 70 days p.c. in the four studies by Streeter, Bossy,
Drey and this study.
The gradient of the regression lines is less steep in this study than in the three other studies. Linear regression for Streeter, Bossy and this study showed that they were
significantly different (P , 0.001).
mean age values of the Carnegie stages (Fig. 2) in spite of the fact that
the Carnegie staging system relied on the last menstrual period,
whereas our embryonic ages were based on CRL ultrasound measurements. Despite these discrepancies, we suggest that the ‘mean foot
measurements’ in our study should be considered as an extra criterion
and an improvement of the morphogenetic staging used in the
Carnegie Collection.
The linear correlation of embryonic and fetal age to the growth of
foot bud/foot plate or heel-toe, respectively (Fig. 3) differed from
each other. This can be explained by two factors: (1) the small gap
in foot measurements from 2.1 to 2.5 mm (Fig. 3), that signifies the
differences between the two previously described ways of measuring,
in which the cleft of the foot plate or the heel are used as end-points,
and (2) the overlap (Grey zone) from embryonic day 41 –47 p.c. that
signifies the uncertainty that is either attributed to age determination
by CRL ultrasound measurements, or due to the foot growth rate differing among embryos. Because of the different ways of measuring, we
found it reasonable to analyze data separately when evaluating the
effect of environmental exposure to foot growth and when comparing
our data with other studies.
The original study by Streeter (1920), who found a linear correlation between embryonic/fetal age and foot length defined by
‘heel-toe length’ have been supported by several studies (Bossy and
Katz 1964; Hern, 1984; Munsick, 1984; Mercer et al., 1987;
Mhaskar et al., 1989; Mandarim-de-Lacerda, 1990; de Vasconcellos
et al., 1992; Drey et al., 2005). Nevertheless, the four individual
regression lines in Fig. 5 differ from each other. The study on fresh
tissue by Drey et al. showed the steepest gradient, but relied on
data that mainly span a later period (8 –24 weeks p.c.). This possibly
illustrates the fact that fetuses, later on in development, may follow
a non-linear growth curve. This would be in accordance with fetal
weight correlated to age (Hadlock et al., 1991). The differences
between Streeter’s and Bossy’s studies and our study do not harmonize with the results of Fig. 2 and Table I. Figure 2 illustrates a visual
comparison of developmental stages of the foot which corresponds
to ‘mean ages’ relying on either ‘ovulation age’ by last menstrual
period or ‘age in days p.c.’ by ultrasound measurements. Figure 5 illustrates that same-age embryos have different foot lengths. This could
be attributed to a discrepancy in the precision of heel-toe measurements, or to the inaccuracy of ultrasound age determination, or to
variations in the growth rate among embryos, or maybe all three
factors. Particularly with regard to age staging of young embryos,
this has been questioned by several authors (Nishimura et al., 1974;
Dickey and Gasser, 1993a, b; Evtouchenko et al., 1996; Harkness
et al., 1997; Grange et al., 2000). Even if the precise postovulation
age were known, as in pregnancies established resulting from in-vitro
fertilization treatment, Rossavik et al. (1988) noticed that embryos
grew at the same rate throughout the ultrasound-traceable period,
but that some embryos completed the embryonic development
sooner than others. Furthermore, the CRL of individual embryos of
identical postovulation age varied by up to 3-fold in embryos
younger than postovulation day 44 and by 55% after day 49, suggesting
that the growth rate and development of early human embryos vary
considerably (Dickey and Gasser, 1993a, b) or that ultrasound
1832
determination of embryonic age still hosts inaccuracies (Harkness
et al., 1997). These observations are likely to explain some of the
differences between the studies and call for further investigations. In
addition, the bias introduced by the differences in conservation of
tissues must be kept in mind (Fig. 5). As Streeter described, measurements were done after the specimens were fixed, which enlarged the
body and head in particular but the extremities to a lesser extent.
To our knowledge, this study illustrates for the first time that consumption of alcohol, ETS or smoking, regardless of the quantity of
cigarettes smoked a day or the cotinine levels in the mother’s urine,
have no apparent effect on growth of the foot in the first-trimester.
This corroborates with and adds to results of previous studies
(Lieberman et al., 1994; Henderson et al., 2007). These results are furthermore confirmed by Bergsjo et al. (2007) who found that growth
restriction caused by smoking apparently begins after gestational
week 17. With regard to ETS, our results do not concur with the
meta-analysis that suggests a lower birthweight in children of
mothers exposed to ETS (Leonardi-Bee et al., 2008). This raises the
question of whether ETS influences birthweight negatively without
having an effect on growth of the body length. Finally, the present
study indicates that foot growth is not influenced by gender in the firsttrimester. This is in agreement with previous studies that report that
CRL was not influenced by gender (Tezuka et al., 1998; Becker et al.,
2004). Additionally, Lieberman et al. (1994) found that birthweight
correlated to smoking habits was not gender-dependent.
In conclusion, the current study introduces new criteria to improve
the Carnegie staging system and age determination in embryos by presenting measurements of the parameters ‘foot bud’ and ‘foot plate’. A
linear correlation was found between foot bud/foot plate and
embryonic age. Similarly, a linear correlation between heel-toe
length and embryonic/fetal age was shown, which corroborated
with several previous reports. Finally, we found that the gender of
the embryo, ETS and maternal smoking and drinking habits during
the first-trimester of pregnancy apparently had no effect on the
growth of the foot in embryos and fetuses aged 35–69 days p.c.
Acknowledgement
The excellent technical assistance by Ms Kristina Sørensen and
Mrs Tiny Roed and Ms Anne Sørensen is gratefully acknowledged.
Our grateful thanks are extended to the staff at Frederiksberg Hospital, Copenhagen University, Denmark, for their collaboration and
enthusiasm enabling the collection of embryonic tissue. Finally we
wish to thank Mrs Susan Peters for her help with the English.
Funding
This work was financially supported by, The Danish Medical Research
Council (22-03-0200, AG Byskov).
References
Becker S, Ural S, Fehm T, Bienstock J. Fetal gender and sonographic
assessment of crown-rump length: implications for multifetal
pregnancy reduction. Ultrasound Obstet Gynecol 2004;24:399– 401.
Benowitz NL. Biomarkers of environmental tobacco smoke exposure.
Environ Health Perspect 1999;107:349– 355.
Lutterodt et al.
Bergsjo P, Bakketeig LS, Lindmark G. Maternal smoking does not affect
fetal size as measured in the mid-second trimester. Acta Obstet
Gynecol Scand 2007;86:156 – 160.
Bossy J, Katz JM. Croissance segmentaire d’une serie de 100 foetus
humains. Anales del desarollo 1964;12:181 – 213.
Cliver SP, Goldenberg RL, Cutter GR, Hoffman HJ, Davis RO, Nelson KG.
The effect of cigarette smoking on neonatal anthropometric
measurements. Obstet Gynecol 1995;85:625 – 630.
de Vasconcellos HA, Prates JC, de Moraes LG. A study of human foot
length growth in the early fetal period. Ann Anat 1992;174:473– 474.
Dickey RP, Gasser RF. Computer analysis of the human embryo growth
curve: differences between published ultrasound findings on living
embryos in utero and data on fixed specimens. Anat Rec 1993a;
237:400– 407.
Dickey RP, Gasser RF. Ultrasound evidence for variability in the size and
development of normal human embryos before the tenth
post-insemination week after assisted reproductive technologies. Hum
Reprod 1993b;8:331 –337.
Drey EA, Kang MS, McFarland W, Darney PD. Improving the accuracy of
fetal foot length to confirm gestational duration. Obstet Gynecol 2005;
105:773– 778.
Evtouchenko L, Studer L, Spenger C, Dreher E, Seiler RW. A
mathematical model for the estimation of human embryonic and fetal
age. Cell Transplant 1996;5:453 – 464.
Grange G, Pannier E, Goffinet F, Cabrol D, Zorn JR. Dating biometry
during the first trimester: accuracy of an every-day practice. Eur J
Obstet Gynecol Reprod Biol 2000;88:61 – 64.
Hadlock FP, Harrist RB, Martinez-Poyer J. In utero analysis of fetal growth:
a sonographic weight standard. Radiology 1991;181:129 – 133.
Harkness LM, Rodger M, Baird DT. Morphological and molecular
characteristics of living human fetuses between Carnegie stages 7 and
23: ultrasound scanning and direct measurements. Hum Reprod
Update 1997;3:25– 33.
Hata T, Senoh D, Hata K, Kitao M. Mathematical modeling of fetal foot growth:
use of the Rossavik growth model. Am J Perinatol 1996;13:155–158.
Henderson J, Gray R, Brocklehurst P. Systematic review of effects of
low-moderate prenatal alcohol exposure on pregnancy outcome.
BJOG 2007;114:243– 252.
Hern WM. Correlation of fetal age and measurements between 10 and 26
weeks of gestation. Obstet Gynecol 1984;63:26– 32.
Kramer MS. Determinants of low birth weight: methodological assessment
and meta-analysis. Bull World Health Organ 1987;65:663 – 737.
Leonardi-Bee J, Smyth A, Britton J, Coleman T. Environmental tobacco
smoke and fetal health: systematic review and meta-analysis. Arch Dis
Child Fetal Neonatal Ed 2008;93:F351 – F361.
Lieberman E, Gremy I, Lang JM, Cohen AP. Low birthweight at term and
the timing of fetal exposure to maternal smoking. Am J Public Health
1994;84:1127 – 1131.
Lindley AA, Becker S, Gray RH, Herman AA. Effect of continuing or
stopping smoking during pregnancy on infant birth weight, crown-heel
length, head circumference, ponderal index, and brain:body weight
ratio. Am J Epidemiol 2000;152:219 – 225.
Mandarim-de-Lacerda CA. Foot length growth related to crown-rump
length, gestational age and weight in human staged fresh fetuses. An
index for anatomical and medical use. Surg Radiol Anat 1990;
12:103 – 107.
Mercer BM, Sklar S, Shariatmadar A, Gillieson MS, D’Alton ME. Fetal foot
length as a predictor of gestational age. Am J Obstet Gynecol 1987;
156:350– 355.
Mhaskar R, Agarwal N, Takkar D, Buckshee K, Anandalakshmi, Deorari A.
Fetal foot length—a new parameter for assessment of gestational age.
Int J Gynaecol Obstet 1989;29:35– 38.
Foot length as a parameter for fetal age
Munsick RA. Human fetal extremity lengths in the interval from 9 to
21 menstrual weeks of pregnancy. Am J Obstet Gynecol 1984;
149:883– 887.
Nakahori Y, Hamano K, Iwaya M, Nakagome Y. Sex identification by
polymerase chain reaction using X– Y homologous primer. Am J Med
Genet 1991;39:472 – 473.
Nauert GM, Freeman TB. Low-pressure aspiration abortion for obtaining
embryonic and early gestational fetal tissue for research purposes. Cell
Transplant 1994;3:147– 151.
Nishimura H, Tanimura T, Semba R, Uwabe C. Normal development of
early human embryos: observation of 90 specimens at Carnegie
stages 7 to 13. Teratology 1974;10:1 – 5.
O’Rahilly R. Guide to the staging of human embryos. Anat Anz 1972;
130:556– 559.
O’Rahilly R. Early human development and the chief sources of
information on staged human embryos. Eur J Obstet Gynecol Reprod
Biol 1979;9:273 – 280.
O’Rahilly R, Müller F. Developmental stages in human embryos.
Washington: Carnegie Institution of Washington, Publication 637,
1987, 174 – 263.
1833
Pickett KE, Rathouz PJ, Kasza K, Wakschlag LS, Wright R. Self-reported
smoking, cotinine levels, and patterns of smoking in pregnancy.
Paediatr Perinat Epidemiol 2005;19:368 – 376.
Robinson HP, Fleming JE. A critical evaluation of sonar ’crown-rump
length’ measurements. Br J Obstet Gynaecol 1975;82:702 – 710.
Rossavik IK, Torjusen GO, Gibbons WE. Conceptual age and ultrasound
measurements of gestational sac and crown-rump length in in vitro
fertilization pregnancies. Fertil Steril 1988;49:1012 – 1017.
Skidmore FD. An analysis of the age and size of 483 human embryos.
Teratology 1977;15:97 – 102.
Snow ME, Keiver K. Prenatal ethanol exposure disrupts the histological
stages of fetal bone development. Bone 2007;41:181– 187.
Streeter G. Weight, sitting height, head size, foot length, and menstrual age
of the human embryo. Contrib Embryol 1920;143 – 170.
Tezuka N, Banzai M, Sato S, Saito H, Hiroi M. Sexual difference in early
fetal crown-rump length versus gestational age in pregnancies arising
from in vitro fertilization. Gynecol Obstet Invest 1998;45:151 – 153.
Submitted on December 9, 2008; resubmitted on April 3, 2009; accepted on
April 9, 2009