Talk:Paper - An analysis of the age and size of 483 human embryos

An Analysis of the Age and Size of 483 Human Embryos

F. D. SKIDMORE

Department of Surgery, Manchester University, University Hospital of South

Manchester, Manchester, M20 8LR, England

ABSTRACT

The crown-rump length of 483 ﬁxed human embryos of Carnegie stages 6-23 was analyzed and median and predicted mean lengths were calculated. The results were compared with those of other series and confirmed that early human growth rates are different from those of macaques with which human embryo growth has previously been compared. The study indicated that it is possible to predict: 1. the median size of an embryo of given Streeter horizon or Carnegie stage; 2. the age of a fresh embryo, or one of undisputed Streeter staging, by comparison with mean ﬁgures of other authorities; and 3. the corrected age of an em» bryo of known length or known Streeter staging or both in terms of postovulation age. Since teratogens may reduce the embryonic growth rate this information is relevant in the analysis of teratogenic factors in human development.

Knowledge of the relation between the age and size of ﬁxed embryos is important to the descriptive anatomist. The rapid evolution of teratology has made this relation important and relevant to clinical work. Mechanical, physical, and chemical factors may adversely affect development of embryos and fetuses and the list of known pharmacological and other external environment factors that can cause malformations is increasing daily (Wilson, ’73; Nishimura and Tanimura, ’76).

Applied embryological studies take two forms: experimental work on animals, and retrospective and prospective analyses of developmental aberrations in human populations. Research in the former category has often been precipitated by disasters in the latter (McBride, "61). Retrospective studies in human beings require careful analysis of records to pinpoint the stage of fetal development at which a teratogenic factor may have been present. In correlating malformation and cause it is important to understand the chronological development of normal human embryos and fetuses. It has been pointed out that there is good correlation between size and morphological characteristics (Streeter, ’-42, ’-45, ’-48; Nishimura et al., ’68; ’74).

Streeter F48) made a graphic plot of human specimens that had been assigned to various stages or “developmental horizons”. These stages were assigned on the basis of speciﬁc morphological criteria of structures in major systems of the body. From the evidence in experimental primates, in particular macaques, estimates were made of the human embryonic age range assigned to a particular horizon. The correlation was approximately of two days per horizon or stage. But extrapolation of a speciﬁc growth rate should not be used for this determination because the coefficient of linear growth of nonhuman primate species is different from that of man. Recent work (Iffy et al., ’67; Nishimura et al., ’68; Jirasek, ’7l) and a consideration of the basic premise used in comparing people and monkeys make a reevaluation of the age and size of early human embryos imperative. O’Rahilly (73) studied stages 1-9 of human embryogenesis and suggested size “norms” for each stage.

Embryonic age

Descriptions and collections of early embryos have been documented (0’Rahilly,"71; Cooper and O’Rahilly, ’71). For ethical reasons it has been impossible to chart accurately the early stages of human development with the invasive methods used in experimental animals. The possible starting points used to assess age are the following.

1. Time of ovulation. Mittleschnfierz, midcycle bleeding, or the rise in body temperature after ovulation may be an indication of ovulation; 92% of conceptions result from intercourse during days 12 to 20 of the menstrual period. In women with long cycles the luteal phase is more constant than the follicular phase.

2. Postcoital staging. Correlation of in tercourse and menstrual dates enables estimates to be made of the time of fertiliza tion. Reversed uterine polarity and rapid arrival of sperm in the fallopian tubes cast doubt on estimates of a 30-hour lag between intercourse and fertilization.

3. Menstrual age. Variation in luteal and follicular phases makes this method of little value in studies of early development of human embryos.

Agenesis and developmental arrest are major avenues of congenital malformation during embryogenesis. Interruptions in the progressive differentiation of tissues and organs may be due to either mutant genes and chromosomal aberrations or external physical and chemical factors affecting the embryonic environment. Reduced size of an embryo with morphological criteria that enable it to be allocated to a Carnegie stage may be the first indication that developmental arrest has occurred. A table of mean embryonic size for each stage is therefore essential if early teratogenic influences are to be discovered and studied in human embryos.

MATERIALS AND METHODS

The present study was made on embryos in the Carnegie collection and includes the most recent acquisitions in the series. Information on the size of 483 embryos of horizons 6 to 23 was made available by courtesy of Professor R. O’Rahilly.

Histologists at the Carnegie Institute measure the crown-rump length (CRL) of specimens after two weeks of ﬁxation. This measurement is open to the following types of errors. 1. Shrinkage due to ﬁxation. A variation of not more than 5% of the fresh CRL can be attributed to this factor (Ramsey, ’64; quoted by Iffy et al., ’67). 2. Preﬁxation hydropic maceration. 3. Variable curvature in the ﬁxed state. 4. Artefactual problems when a specimen is still attached to a portion of uterine wall. The state of histological preservation was estimated and three groups were identiﬁed: 1, poor; 2, good; and 3, excellent. The number of embryos in stages 6 to 9 was small, and these stages were not included in these groups (table 1).

The size and stage of the specimens were recorded on punched cards, and the information submitted to the Department of Computer Science; Iohns Hopkins University. Polynomial regression analysis was used to compute observed and predicted values of the median length of embryos of each stage. The mean value was determined. for each group and a correlation coefficent expressed for the series. The computer also plotted a graph of size against stage with observed and predicted mean values. Standard deviations of the mean predicted values were also calculated.

Stage No.	embryos
Table I Number of embryos of each stage examined
6	9
7	3
8	5
9	3
10	11
11	19
12	29
13	43
14	53
15	35
16	54
17	38
18	44
19	36
20	30
21	25
22	22
23	24
Total -	483

RESULTS

Table 1 lists the number of embryos in each stage; tables 2 to 4 record the results for the three histological groups. An analy sis was also made of all the specimens together (table 5), which was compared with that of group 3. The results of other studies of embryonic age and length

TABLE 4

Length and stage of human embryos.

TABLE 2

Length and stage of human embryos. Group 1: histological grade -—- poor

Predicted Stage No. Length values embryos (mm) 1 (mm) 1 3:: 1 SD 10 2 1.87i 0.13 1.25: 0.13 11 2 2.42:‘: 0.32 2.39i 0.38 12 9 385$ 1.09 3.68: 0.59 13 20 4.63:1: 0.60 5.17: 0.80 14 17 6.61 i 0.67 6.80:1: 0.92 15 11 8.03: 0.90 860$ 1.12 16 16 9.351" 1.13 10.57: 1.23 17 10 12.973; 1.54 12.70i 1.32 18 11 14.831 1.92 15.001 1.38 19 14 18.431 0.86 17.47: 1.40 20 15 21.99.*.: 1.39 20.10: 1.41 21 8 23.18:: 1.80 22.895: 1.39 22 6 25.11: 1.40 25.85$ 1.33 23 5 28.20t 0.98 28.98: 1.25

Group 8: histological grade -~—— excellent

Predicted Stage No. Length values embryos (mm) 1 (mm) 1 ‘X 1 SD 10 . 1 3.301 0 2.70:: 0.42 11 10 3.401 1.01 3.35i 0.52 12 9 4.03¢ 0.66 4.231 0.63 13 14 4.81 :I: 0.74 5.351 0.73 14 14 6.46i 0.77 6.70: 0.83 15 7 8.06:: 0.92 8.281 0.94 16 11 9.53: 1.14 10.09: 1.05 17 11 12.29: 1.10 1214.11.16 18 11 14.36: 1.41 14.42: 1.28 19 5 18.361 1.52 16.92i 1.39 20 9 20.42: 1.51 19.68:.t 1.51 21 5 22.74: 0.68 22.65.14: 1.64 22 5 24.401 2.56 25.863: 1.76 23 3 29.56: 1.82 29.30.21: 1.89

1 Correlation coefficient: 0.9865.

TABLE 3

Length and stage of human embryos. Group 2: histological grade ----good

1 Correlation coefficient: 0.9770.

TABLE 5

Total series, including specimens of stages 6~9 which were not

Predicted Stage No. Length values embryos (mm) 1 (mm) 1 E t SE 10 8 2.09i 0.72 1.77:}: 0.63 11 7 3.141 0.71 267$ 0.73 12 12 3.56: 0.51 3.78.1 0.80 13 8 4.83:2: 0.79 5.09:1: 0.89 _ 14 22 6.421 0.90 6.61 i 0.97 15 17 7.561. 1.21 832$ 1.06 16 27 9.54: 1.64 10.24: 1.14 17 17 12.131 1.33 12.361 1.23 18 22 15.2121: 1.21 14.68i 1.31 19 17 18.11: 1.55 17.21:}: 1.39 20 6 21.06: 0.97 19.93i 1.49 21 12 22.90i 1.76 22.86: 1.57 22 11 25.58: 1.79 25.99¢ 1.66 23 16 28.71 i 2.46 29.32$ 1.75

graded histologically Predicted Stage Length values (mm) 1 (mm) 1 Tc: SD 6 020$ 0.05 0.31 1 0.02 7 0.301 0.09 0.41 i 0.15 8 0.97i 0.29 0.72i 0.26 9 1.44.? 0.06 1.22 i 0.39 10 2.16: 0.72 1.92%. 0.51 11 3.20: 0.91 2.821" 0.63 12 3.79: 0.78 3.92: 0.75 13 4.73: 0.70 5.23: 0.86 14 649i 0.80 6.73.11: 0.97 15 7.81:1: 1.10 8.43.4: 1.08 16 9.48: 1.41 10.33i 1.19 17 12.41i 1.36 12.43: 1.30 18 14.91 i 1.58 14.731140 19 18.275; 1.33 17.23: 1.47 20 21.33: 1.52 19.93: 1.61 21 22.96: 1.64 22.83: 1.71 22 25.181 1.60 25.931 1.80 23 28.71: 19 29.233: 1.90

1 Correlation coefficient: 0.9836.

1 Correlation coefficient: 0.9633. 100

(Olivier and Pineau, ’61; Jirasek, "7 1) were also analyzed using the same polynomial regression program (tables, 6, 7).

The correlation coefficient of the CRL of the 483 embryos was 0.96 (table 5),

human embryos.

which supports that size is a valid way of comparing the stage of development of

From the results ﬁgure 1 was drawn. Figure 2 shows the length of the embryos

(mean :1: SD). The vertical scale was partially derived from macaque development

TABLE 6 , , scales. The results of Jirasek ( 71) and Comparison of stage rind length Olivier and Pineau F51), analyzed by the 0f 9mb"!I03 same polynomial regression program used Source Stage Olivier TABLE 7 and Present series Jirasek Pineau The “S9 of h“ma"* embryas

6 0.20 0.20 ...... Olivier ,

7 0'30 0'30 M Stage Streeter Iirasek Mall Pianndiiu

8 0.97 0.50 --—

9 1.44 1.25 --- 11 24 24.5 28 24.2 25.3 10 2.16 2.00 --—-- 12 26 28 29.5 28 27.9 11 3.20 2.75 3.10 13 27.5 30 32.5 28 30.2 12 3-79 435 3-70 14 29.4 33 36.5 32 34 13 4.73 5.00 4.70 15 31.25 36.5 38 33 35.4 14 6.49 6.50 6.50 16 33.25 39.5 40.6 37 39 15 7-81 750 7.20 17 35.2 43 44 41 42.5 15 9-43 9-50 9-40 18 37.1 46 46.8 44 45.5 17 12.41 13.00 12.50 19 39 49,5 50 47,5 49 18 14-91 15.00 14.90 20 41 52 52.6 50.5 51.75 19 18.27 18.00 18.20 21 43 53,5 54 52 53 20 21.33 20.50 21.40 22 45 55 56,2 54 55 21 22.96 23.00 22.80 23 47 53 59 55 57,5 22 25.18 25.50 25.40 _ _ 23 2331 30.00 28.90 ‘ The median value is the mean of the last three columns

dotted line (table 5).

STAGE OF DEVELOPMENT

LERGTH ( mm. )

30

and is contrasted with Streeter's estimates.

Fig. 1 The length of 483 human embryos of Streeter stages 6-23. The median size for each group has been computed and plotted. The computed predicted values for each stage are indicated by the

THE SIZE OF HUMAN EMBRYOS

STAGE OF DEVELOPMENT

15

101

20 25 30

LENGTH ( mm.)

Fig. 2 The length of human embryos in this series was within 1 SD of the specimens identified by other authors as belonging to that particular stage (tables 6, 7). The outer lines show the SD

conﬁdence limits from table 5.

AGE: DAYS POST OVULATION

28 7 3 9

Ii} 11 22 1.3 14 15 I6 1? 18 19 29

21 22 23

STAGE OF DEVELOPMENT

Fig. 3 The age of embryos. The bottom line shows Streeter’s (’42, ’45, ’48} estimate based partially on evidence of macaque development. The upper line is based on human specimens (table 7}.

for the Carnegie embryos, show that the mean CRL for each stage lies within 1 SD of the Carnegie ﬁgures (tables 2-5). Table 6 compares the size of these various specimens. Table 7 compares the embryonic age in days for each stage from previous studies (Mall, ’l0, ’l8; Olivier and Pineau, ’61; Jirasek, ’7ll ﬁgure 3 compares the means of these values with Streeter’s F42, ’45, ’48) calculations; the upper line of this graph can be used to calculate the ages of embryos of given stages.

DISCUSSION

In the group 1 series, with “poor” histological preservation, the stage 10 and 11 embryos were smaller than those in group 2, which in turn were smaller than those in group 3. The smallest specimens are most prone to cell shrinkage before ﬁxation; this problem does not occur in larger embryos. There was close agreement between embryonic size in each stage as calculated and the values reported by other workers.

From this information it is possible to predict the mean length of specimens at particular stages of development, as de—ﬁned by Streeter C42, ’45, ’48l.

Nishimura et al. F68) suggested that differences in the time of ovulation, length of the preirnplantation period, and rate of early development may play a role in the observed variation in embryonic length. There is no evidence that delay in implantation is associated with signiﬁcant delay of maturation; the uterus apparently provides sufficient nutrition for the conceptus to enable normal cellular development to occur before establishment of connection be» tween the uterine and fetal circulations.

ACKNOWLEDGMENTS

The author was the holder of a British Heart Foundation Research Fellowship while this work was being carried out in Cambridge and Baltimore. I would like to thank Professor R. Harrison, F. R. S. and Doctor ]. D. Ebert for providing facilities during the tenure of the Fellowship.

LITERATURE CITED

Cooper, M. H., and R. 0’Bahilly 1971 The human heart at seven postovulatory weeks. Acta Anat., Basel, 79: 280299.

Iffy, L., T. H. Shepard, A. Iakobovits, R. J. Lernire and P. Kerner 1967 The rate of growth in young human embryos of Streeter’s horizons 13-23. Acta Anat. (Basel), 66: 178-186.

Jirasek, J. E. 1971 Development of the Genital System and Male Pseudohermaphroditisrn. Johns Hopkins, Baltimore.

Mall, F. P. 1910 The determination of the age of human embryos and fetuses. In: Human Embryology. F. Keibel, and F. P. Mall, eds. Lippincott, Philadelphia.

--—-~— 1918 On the age of human embryos. Am. ]. Anat., 23: 397-422.

McBride, W. G. 1961 Thalidomide and congenital abnormalities. Lancet, 2: 1358.

Nishimura, H., and T. Tanimura 1976 Clinical Aspects of the Teratogenicity of Drugs. Excerpta Medica, Amsterdam.

Nishimura, H., K. Takano, T. Tanimura and M. Yasuda 1968 Normal and abnormal development of human embryos. First report of the analysis of 1213 intact embryos. Teratology, 1: 281-290.

Nishimura, H., T. Tanimura, R. Semba and C. Uwabe I974 Normal development of early human embryos: observation of 90 specimens at Carnegie stages 7 to 13. Teratology, 10: 1-5.

Olivier, G., and H. Pineau 1961 Horizons de Streeter et age ernbryonnaire. Bull. Ass. Anat., 4 7: 573-576.

O’Ral1illy, B. 1971 The timing and sequence of events in human cardiogenesis. Acta Anat. (Basel). 79: 70-75.

Streeter, C. L. 1942 Developmental horizons in human embryos: age group XI, 13 to 20 somites, and age group XII 21 to 29 somites. Contrib. Ernbryol. Carnegie. Inst., 30: 211-245.

1945 Developmental horizons in human embryos. Descriptions of age group XIII, embryos about 4 or 5 mm long and age group XIV, period of indentation of the lens vesicle. Contrib. Ernbryol. Carnegie. Inst, 31: 29-63.

1948 Developmental horizons in human embryos. Descriptions of age groups XV, XVI, XVII, XVIII. Contrib. Embryol. Carnegie Inst, 32: 133-203.

Wilson, J. C. 1973 Environment and Birth Defects. Academic, New York.