Computer-controlled, multilevel, morphometric analysis of blastomere size as biomarker of fragmentation and multinuclearity in human embryos

Christina Hnida, Elisabete Engenheiro and Søren Ziebe1

The Fertility Clinic, Rigshospitalet, Section 4071, University Hospital of Copenhagen, Blegdamsvej 9, DK-2100 Copenhagen, Denmark

1 To whom correspondence should be addressed. e-mail: sziebe{at}rh.dk


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Little is known about blastomere size at different cleavage stages and its correlation with embryo quality in human embryos. Using a computer system for multilevel embryo morphology analysis we have analysed blastomeres of human embryos and correlated mean blastomere size with embryonic fragmentation and multinuclearity. METHODS: A consecutive cohort of 232 human 2-, 3- and 4-cell embryos from patients referred for ICSI treatment were included. Sequences of digital images were taken by focusing at 5-µm intervals through the embryo. Blastomere sizes and number of nuclear structures were evaluated based on these sequences. The degree of embryonic fragmentation was evaluated by normal morphological assessment prior to transfer and correlated to the blastomere sizes. RESULTS: As a result of normal cell cleavage, mean blastomere size decreased significantly from a volume of 0.28 x 106 µm3 at the 2-cell stage to 0.15 x 106 µm3 at the 4-cell stage (P < 0.001). Mean blastomere size decreased significantly (P < 0.001) with increasing degree of embryonic fragmentation, where highly fragmented embryos showed a 43–67% reduction in blastomere volume compared with embryos with no fragmentation. Multinucleated blastomeres were significantly larger than non-multinucleated blastomeres (P < 0.001). On average, multinucleated blastomeres were 51.5, 67.8 and 73.1% larger than their non-multinucleated sibling blastomeres at the 2-, 3- and 4-cell stage, respectively. Furthermore, the average volume of non-multinucleated blastomeres originating from multinucleated embryos was significantly smaller than the average volume of the blastomeres from mononucleated embryos (P < 0.001). CONCLUSIONS: The results of this study show that the average blastomere size is significantly affected by degree of fragmentation and multinuclearity, and that computer-assisted, multilevel analysis of blastomere size may function as a biomarker for embryo quality.

Key words: blastomere size/computer-controlled morphometric analysis/embryo fragmentation/embryo quality/multinuclearity


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Selection of embryos for transfer after IVF treatment is based mainly on subjective light microscopic morphology analysis. Many studies have shown that morphological structures in the embryo can be used as biomarkers of embryonic quality (Puissant et al., 1987Go; Schulman et al., 1993Go; Giorgetti et al., 1995Go; Van Royen et al., 1999Go), and that embryo selection based on morphology assessment is important to improve implantation and pregnancy rates (Hill et al., 1989Go; Erenus et al., 1991Go; Giorgetti et al., 1995Go; Ziebe et al., 1997Go). Most of the existing scoring systems are based on combinations of several morphological parameters such as cleavage stage, embryonic fragmentation and blastomere uniformity (Puissant et al., 1987Go; Erenus et al., 1991Go; Steer et al., 1992Go; Giorgetti et al., 1995Go). A previous report indicated that embryos with unevenly sized blastomeres have increased rates of multinuclearity and lower implantation rates (Hardarson et al., 2001Go). Previous studies have demonstrated that oocyte size may be an indicator of its maturational stage and biological competence (Goyanes et al., 1990Go; Durinzi et al., 1995Go; Wolf et al., 1995Go; Combelles et al., 2002Go). However, only little is known about the actual blastomeric size during early human embryonic development.

The lack of objective and standardized methods to assess embryonic fragmentation remains a concern in relation to defining embryo quality. As fragments by definition are anucleate structures of blastomeric origin, the degree of fragmentation in the embryo may be reflected in the average blastomeric volume.

The presence of multinucleated blastomeres may indicate an uncoupling of processes controlling the karyokinesis and cytokinesis during the first cleavage divisions. The effect of this would be a duplication of the nucleus without subsequent cell cleavage, resulting in a multinucleated blastomere with the same cell volume as the previous generation (Hardy et al., 1993Go; Pickering et al., 1995Go). However, other mechanisms have also been suggested to be involved in the formation of multinucleated blastomeres. These includes errors in the chromosome migration at mitosis, incorrect packaging of the chromosomes by the nuclear membrane after mitosis, fragmentation of the nuclei or asymmetrical cytokinesis (Hardy et al., 1993Go; Pickering et al., 1995Go; Johansson et al., 2003Go). The embryo quality score, as well as blastomere characterization, is based on the subjective judgement of the operator. Furthermore, time is a limiting factor when analysing the embryo in order not to compromise its quality, which makes it difficult to investigate embryo morphology in detail.

In the present study we used a computer-controlled system for multilevel and non-invasive embryo morphology analysis. This allowed us to overcome the time limitation, as well as perform a more precise and detailed analysis of embryo morphology.

The aim of this study was to determine the blastomere size at different cleavage stages of early embryonic development, and to define deviations in mean blastomere volume as a consequence of embryonic fragmentation and multinuclearity.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Selection of embryos
This study included a consecutive cohort of 232 fertilized oocytes that had developed to the 2-, 3- or 4-cell stage 48 h after oocyte aspiration (Table I), consisting of a total of 708 blastomeres. This included the embryos from patients undergoing ICSI treatment at our clinic during a 10-week period between April and June 2002. Only embryos transferred during weekdays were included. Digital microscopy images were taken of all included embryos and all further analysis was performed on these images.


View this table:
[in this window]
[in a new window]
 
Table I. Mean blastomere sizes of embryos at different cleavage stages
 
Patients
Sixty-three ICSI patients were included in this study. The average age of the women was 32.1 years (range 24–39). Fifty-one patients were referred to ICSI due to male factor. Eleven patients had shown poor cleavage in previous treatments and one patient underwent ICSI due to preimplantation genetic diagnosis treatment. The average number of oocytes aspirated per patient was 11.2.

Patients were treated with the long protocol, using GnRH agonist for down-regulation and recombinant FSH (Gonal-F®, Serono; or Puregon®, Organon) for ovarian stimulation. HCG (Profasi®; Serono) was given 36 h before oocyte retrieval.

ICSI procedure
ICSI was performed according to routine procedures. Briefly, oocytes were aspirated 36 h after HCG injection and the ICSI procedure was performed 4–6 h later. On the following morning (18 h after insemination) the oocytes were checked for fertilization and cultured for a further 24 h. Embryo transfer was carried out 48 h after oocyte aspiration. Immediately prior to transfer, embryos were selected for transfer by evaluating cleavage stage and quality score in accordance with the normal procedures at our clinic. The selection of embryos for transfer and cryopreservation was carried out independently of this study. Subsequently, all 2- to 4-cell stage embryos were re-evaluated based on the recorded sequences.

Recording of digital images
Using the FertiMorph computer system for multilevel embryo morphology analysis (Image House Medical A/S, Copenhagen, Denmark), image sequences were recorded of all included embryos 48 h after oocyte aspiration. Each sequence consisted of 26 images (Figure 1) of the same embryo with the FertiMorph system automatically focusing in 5-µm intervals through the embryo. The automatically controlled image recording and storing took ~15 s per sequence. All recordings were performed at 400x magnification with Hoffman modulation contrast illumination.



View larger version (154K):
[in this window]
[in a new window]
 
Figure 1. Digital image sequence of a 7-cell embryo recorded in 5-µm intervals.

 
Measurements on the images are done in pixels and converted to actual physical units by knowing the distance between two adjacent pixels. The calibration is done by taking an image of a micrometre slide with the same magnification as the embryo images. A line is drawn on the micrometre slide. Knowing the outlined distance on the slide the system calculates the physical distance between two pixels.

Computer-controlled morphometric analysis of embryo morphology
Based on the digital image sequences, blastomere size and nuclear structures were analysed in a semi-automatic manner using the morphology analysis software of the FertiMorph system. All images of one sequence could be viewed in detail, enabling us to select the pictures where the different structures were in focus. For all embryos, the outer border of all morphological structures that were considered to be a blastomere and all visible nuclei were outlined on the particular digital images where the structures were in focus. Thus, the different blastomeres and nuclei structures of one embryo could be outlined on different images, representing different focus depths.

To differentiate between blastomeres and fragments we defined the property of the outlined structures in accordance with the following criteria. If at least one nuclear structure was detected it was considered a blastomere regardless of its size. If no nuclear structures could be detected it was considered a blastomere when the average diameter was ≥40 µm, and a fragment when the average diameter was <40 µm.

Morphometric values describing the size of the blastomeres and nuclear structures (area, diameter and volume) were calculated automatically. The system calculated the area of the blastomere. A radius (r) was calculated from a circle with the equivalent area and the cell volume (V) was computed from the assumption that blastomeres were spherical, therefore using the equation V = 4/3{pi}r3.

The calculation of each individual morphological structure was only based on the particular image where this structure was outlined.

The embryonic fragmentation was evaluated by normal morphology assessment, allocating each embryo to one of the following five groups: group I, 0% fragmentation; group II, 1–10% fragmentation; group III, 11–20% fragmentation; group IV, 21–50% fragmentation; or group V, >50% fragmentation. Multinuclear embryos were defined as embryos having more than one visible nuclear structure in at least one blastomere. Mononucleated embryos were defined as embryos having only blastomeres with no or one nuclear structure(s). Furthermore, the non-multinucleated blastomeres of multinucleated embryos included all mono- and all annucleated blastomeres.

Statistical analysis
In case of parametric data, one-way or two-way ANOVA was performed, followed by Tukey test for all pairwise multiple comparisons. For non-parametric data, Kruskal–Wallis ANOVA on ranks were performed, followed by Dunns method for multiple comparison. Linear regression was performed when evaluating blastomere volumes in relation to different embryonic cleavage or fragmentation stages. Differences were considered significant when P < 0.05.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Overall
In total, 708 blastomeres from 232 embryos that had developed to the 2-, 3- or 4-cell stage 48 h after oocyte aspiration were analysed. Analysis of the normal reduction in blastomere size as a consequence of cell cleavage showed a significant decrease in mean blastomere size (P < 0.001), from a cell volume of 0.278 x 106 µm3 at the 2-cell stage to 0.182 x 106 µm3 at the 3-cell stage, and to 0.149 x 106 µm3 at the 4-cell stage (Table I). The volumes correspond to diameters of 80.1, 68.7 and 64.9 µm for 2-, 3- and 4-cell embryos, respectively (Table I). Linear regression for mean blastomere volumes of 2-, 3- and 4-cell embryos gave a square of correlation coefficient of R2 = 0.925 (P = 0.176) and a slope coefficient of {alpha} = –0.065.

Fragmentation
The mean blastomere volume decreased significantly with increasing degree of fragmentation (P < 0.001) (Table II). The decrease was negative linear for all analysed cleavage stages and R2 were 0.969 (P = 0.002), 0.891 (P = 0.0169) and 0.932 (P = 0.008), with {alpha} = – 0.0583, –0.0236 and –0.0200 for 2-cell, 3-cell and 4-cell embryos, respectively. Two-cell embryos with >50% fragmentation showed a 67% reduction in mean blastomere volume (from 0.341 to 0.112 x 106 µm3), compared with 2-cell embryos with no fragmentation. The corresponding reduction was 44% (from 0.200 to 0.112 x 106 µm3) and 43% (from 0.164 to 0.094 x 106 µm3) for the 3- and 4-cell embryos, respectively (Table II).


View this table:
[in this window]
[in a new window]
 
Table II. Mean blastomere volume as function of degree of fragmentation for 2-, 3- and 4-cell embryos
 
Multinuclearity
Of all the embryos included, 55 and 60% were considered mononucleated, and 45 and 40% were multinucleated in the 2- and 4-cell embryos, respectively. Twenty-eight percent of the 3-cell embryos were mononucleated and 72% were multinucleated. Overall, 48% of all the embryos analysed showed more than one nuclear structure in at least one of its blastomeres (Table III).


View this table:
[in this window]
[in a new window]
 
Table III. Analysed embryos in relation to cleavage stage and nuclearity
 
Multinucleated blastomeres were significantly larger than their non-multinucleated sibling blastomeres (P < 0.001) (Figure 2), with an average volume of 0.312 x 106 µm3 and 0.206 x 106 µm3 for 2-cell embryos, 0.230 x 106 µm3 and 0137 x 106 µm3 for 3-cell embryos and 0.206 x 106 µm3 and 0.119 x 106 µm3 for 4-cell embryos (Table IV). Thus, on average, multinucleate blastomeres were 51.5, 67.8 and 73.1% larger than their non-multinucleated sibling blastomeres at the 2-, 3- and 4-cell stage, respectively. Furthermore, the average volume of non-multinucleated blastomeres originating from multinucleated embryos was significant smaller than the average volume of the blastomeres from mononucleated embryos (P < 0.001) (Figure 2). Mean blastomere volumes of mononucleate and anucleate blastomeres were not statistically different for 2- and 3-cell embryos. For 4-cell embryos, the mean blastomere volume of anucleate blastomeres was significantly smaller than of mononucleate blastomeres in multinucleated embryos (P = 0.037) and mononucleated embryos (P < 0.01), respectively (Table IV). The percentages of anucleate blastomeres among all non-multinucleate blastomeres in multinucleated embryos were 17.2, 40.7 and 30.5% at the 2-, 3- and 4-cell stage, respectively. In mononucleated embryos, 13.5, 35.9 and 12.7% of the blastomeres were anucleate at the 2-, 3- and 4-cell stage, respectively.



View larger version (95K):
[in this window]
[in a new window]
 
Figure 2. Volume of (A) multinucleated blastomeres and (B) non-multinucleated blastomeres from multinucleated embryos, and (C) volume of the blastomeres from mononucleated embryos.

 

View this table:
[in this window]
[in a new window]
 
Table IV. Mean blastomere volume of mononucleated and multinucleated embryos
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Data from the present study showed that the mean blastomere volume decreased significantly with increasing degree of fragmentation.

As fragments are structures of blastomeric origin they may extract considerable amounts of cytoplasm from the blastomeres, resulting in blastomeres with a quantitative deficiency of important cytoplasmic contents such as cell organelles, mRNA or proteins, as suggested by Johansson et al. (2003)Go. The question is therefore whether it is the fragments themselves or the small blastomere size that is associated with high fragmentation that compromises embryo quality. Blastomeres in highly fragmented embryos may be too small to be biologically competent.

Evaluation of fragmentation is included in all embryo scoring systems and the presence of high amounts of fragmentation correlates negatively with implantation and pregnancy rates (Puissant et al., 1987Go; Hill et al., 1989Go; Erenus et al., 1991Go; Steer et al., 1992Go; Giorgetti et al., 1995Go; Ziebe et al., 1997Go; Ebner et al., 2001Go).

However, normal evaluation of fragmentation is highly subjective, resulting in only a rough estimate (Roux et al., 1995aGo), often with huge interobserver variation. The lack of objectivity and standardized methods makes it difficult to analyse correlations between fragmentation and embryo quality in detail.

The findings of this study may indicate that blastomere size could be used as an indicator of degree of fragmentation. In the case of total fragmentation of whole blastomeres, the size of the remaining unfragmented sibling blastomere will most likely remain unchanged, and therefore will not reflect the degree of embryonic fragmentation. However, as the total cytoplasmic volume has been shown to remain constant during early embryonic development (Goyanes et al., 1990Go; Roux et al., 1995GobGo), the degree of embryonic fragmentation may be quantified as the reduction in the total blastomeric volume compared with the ooplasm of the zygote. This may allow an assessment of fragmentation, independently of the size of a particular intra-embryonic blastomere, based of the sum of their volumes.

In the present study we found that 48% of the embryos analysed were multinucleated. This high incidence of multinuclearity is in accordance with the findings of Van Royen et al. (2003)Go, who found that 33.6% of a cohort of embryos from patients undergoing IVF or ICSI treatment were multinucleated. However, the study by Van Royen et al. (2003)Go may underestimate the incidence of multinuclearity, as the assessment of the number of nuclei was done solely on the basis of traditional light microscopy.

We found that multinucleated blastomeres were significantly larger (52–73%) than their non-multinucleated sibling blastomeres. Previous studies have shown that unevenly sized blastomeres are correlated with high frequencies of multinuclearity (Hardarson et al., 2001Go). It has also been demonstrated that transfer of multinucleated embryos results in low implantation and pregnancy rates (Jackson et al., 1998Go; Pelinck et al., 1998Go). It has previously been suggested that multinucleated blastomeres can originate from an uncoupling of processes controlling karyokineses and cytokinesis, resulting in duplication of the nuclear material without a subsequent cell cleavage (Hardy et al., 1993Go; Pickering et al., 1995Go). The consequence of this would be a multinucleated blastomere without the size reduction from the cell cleavage, which thus remained the same size as the previous cell generation. These speculations are supported by our finding that a multinucleated 4-cell blastomere was approximately the same size as a non-multinucleated 2-cell blastomere. Furthermore, a very close interaction and timing of the processes controlling the karyokinesis and cytokinesis has been shown in normal cleaving cells (Straight and Field, 2000Go; Burke et al., 2002Go). The cytokinesis starts at anaphase, which coincides with the start of the reassembling of the nuclear envelopes. These nuclei are not light microscopically visible until cytokinesis is almost or entirely finished (Straight and Field, 2000Go; Burke et al., 2002Go). Thus, the appearance of nuclear membranes prior to cell cleavage may indicate an abnormal cell cleavage at all analysed cleavage stages. However, the findings that non-multinucleated blastomeres originating from multinucleated embryos on average are smaller in size than the blastomeres from mononucleated embryos indicates that such an uncoupling of karyokinesis and cytokinesis may not be the full explanation, and that incompetent cytoplasm or other mechanisms controlling cell cleavage may be involved.

Furthermore, we did not distinguish between different types of nuclear structures. Fragmented nuclei are thought to indicate apoptosis, which to some extent may reflect degradation of ceased multinucleated cells (Hardy et al., 1993Go; Hardy, 1999Go). However, fragmentation of nuclei may also be a result of other cell cleavage abnormalities, not all of them necessarily related to multinuclearity (Hardy et al., 1993Go; Hardy, 1999Go).

The percentage of anucleate blastomeres was higher among multinucleated embryos than among mononucleated embryos. This could be due to an asymmetrical karyokinesis in multinucleated embryos, while the anucleate blastomeres in mononucleated embryos reflect normal blastomeres undergoing mitosis.

The frequency of multinuclearity among 3-cell embryos was ~30% higher than in 2- and 4-cell embryos. Thus, the 3-cell embryos contribute considerably to the overall frequency of multinuclearity. In normal developing embryos the 3-cell stage is an intermediate stage which is the result of a minor but normal asynchrony in the timing of the cleavage processes (Roux et al., 1995aGo,bGo). However, major delayed or arrested division of a blastomere may result in embryos with abnormally high variations in their blastomere size, one reason being an asynchrony of the karyokinesis and cytokinesis resulting in multinucleated blastomeres (Roux et al., 1995aGo,bGo). In accordance with the normal kinetics of embryonic development, the majority of healthy embryos reach the 4-cell stage after 2 days of culture. Thus, the high frequency of multinucleated 3-cell embryos at day 2 may reflect a major delay in or arrested division of one of the two blastomeres at the 2-cell stage.

Light microscopy morphology assessment is an important non-invasive tool when selecting good quality embryos for transfer, and correlations between morphological structures such as fragmentation and cleavage stage with the embryo quality and clinical outcome are well documented (Giorgetti et al., 1995Go; Ziebe et al., 1997Go; Van Royen et al., 1999Go). However, the use of a computer system for multilevel embryo morphology analysis allowed us to achieve a more precise measurement of the blastomere size and enabled us to overcome the time factor that normally limits embryo morphology assessment.

Our criteria for distinguishing between a blastomere and a fragment is supported by the findings by Johansson et al. (2003)Go, which suggest that on day 2 cells ≥45 µm should be classified as blastomeres and <45 µm as fragments. However, the precise minimum sizes for biologically competent blastomeres at different cleavage stages are still unknown, and should be revealed at least in order to achieve a more objective assessment of fragmentation based on blastomere sizes.

Our finding that anucleate blastomeres in 4-cell embryos were significantly smaller in size than mononucleate blastomeres may indicate that the definition of a minimum size of 40 µm was too low, and that a number of these anucleated cells might be fragments rather than biologically competent blastomeres.

In conclusion, the results from this study showed that blastomere size is significantly affected by degree of embryonic fragmentation and multinuclearity. Furthermore, it indicates that computer-assisted, multilevel analysis of blastomere size may function as a biomarker for embryo quality, at least with regard to embryonic fragmentation and multinuclearity.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Burke B and Ellenberg J (2002) Remodelling the walls of the nucleus. Nat Rev 3,487–497[ISI]

Combelles CMH, Cekleniak NA, Racowsky C and Albertini DF (2002) Assessment of nuclear and cytoplasmic maturation in in-vitro matured human oocytes. Hum Reprod 17,1006–1016.[Abstract/Free Full Text]

Durinzi KL, Saniga EM and Lanzendorf SE (1995) The relationship between size and maturation in vitro in the unstimulated human oocyte. Fertil Steril 63,404–406.[ISI][Medline]

Ebner T, Yaman C, Moser M, Sommergruber M, Polz W and Tews G (2001) Embryo fragmentation in vitro and its impact on treatment pregnancy outcome. Fertil Steril 76,281–285.[CrossRef][ISI][Medline]

Erenus M, Zouves C, Rajamahendran P, Leung S, Fluker M and Gomel V (1991) The effect of embryo quality on subsequent pregnancy rates after in vitro fertilization. Fertil Steril 56,707–710.[ISI][Medline]

Giorgetti C, Terriou P, Auquier P, Hans E, Spach JL, Salzmann J and Roulier R (1995) Embryo score to predict implantation after in-vitro fertilization: based on 957 single embryo transfers. Hum Reprod 10,2427–2431.[Abstract]

Goyanes VJ, Ron-Corzo A, Costas E and Maneiro E (1990) Morphometric categorization of the human oocyte and early conceptus. Hum Reprod 5,613–618.[Abstract]

Hardarson T, Hanson C, Sjögren A and Lundin K (2001) Human embryos with uneven sized blastomeres have lower pregnancy and implantation rates: indications for aneuploidy and multinucleation. Hum Reprod 16,313–318.[Abstract/Free Full Text]

Hardy K (1999) Apoptosis in the human embryo. Rev Reprod 4,125–134[Abstract/Free Full Text]

Hardy K, Winston RML and Handyside AH (1993) Binucleate blastomeres in preimplantation human embryos in vitro: failure of cytokinesis during early cleavage. J Reprod Fertil 98,549–558.[Abstract]

Hill GA, Freeman M, Bastias MC, Rogers BJ, Herbert CM III, Osteens KG and Wentz AC (1989) The influence of oocyte maturity and embryo quality on pregnancy rate in a program for in vitro fertilization-embryo transfer. Fertil Steril 52,801–806.[ISI][Medline]

Jackson KV, Ginsburg ES, Hornstein MD, Rein MS and Clarke RN (1998) Multinucleation in normal fertilized embryos is associated with an accelerated ovulation induction response and lower implantation rates in in vitro fertilization-embryo transfer cycles. Fertil Steril 70,60–66.[CrossRef][ISI][Medline]

Johansson M, Hardarson T and Lundin K (2003) There is a cutoff limit in diameter between a blastomere and a small anucleate fragment. J Assist Reprod Genet 20,309–313.[CrossRef][ISI][Medline]

Pelinck MJ, De Vos M, Dekens M, Van der Elst J, De Sutter P and Dhont M (1998) Embryos cultured in vitro with multinucleated blastomeres have poor implantation potential in human in-vitro fertilization and intracytoplasmic sperm injection. Hum Reprod 13,960–963.[Abstract]

Pickering SJ, Taylor A, Johnson MH and Braude PR (1995) An analysis of multinucleated blastomere formation in human embryos. Hum Reprod 10,1912–1922.[Abstract]

Puissant F, Van Rysselberge M, Barlow P, Deweze J and Leroy F (1987) Embryo scoring as a prognostic tool in IVF treatment. Hum Reprod 2,705–708.[Abstract]

Roux C, Joanne C, Agnani G, Fromm M, Clavequin MC and Bresson JL (1995a) Morphometric parameters of living human in-vitro fertilization embryos; importance of asynchronous division process. Hum Reprod 10,1201–1207.[Abstract]

Roux C, Borodkine R, Joanne C, Bresson JL and Agnani G (1995b) Morphological classification of human in-vitro fertilization embryos based on the regularity of asynchronous division process. Hum Reprod Update 1,488–496.[Abstract]

Schulman A, Ben-Num I, Gethler Y, Kaneti H, Shilon M and Beyth Y (1993) Relationship between embryo morphology and implantation rate after in vitro fertilization treatment in conception cycles. Fertil Steril 60,123–126.[ISI][Medline]

Steer CV, Mills CL, Tan SL, Campbell S and Edwards RG (1992) The cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and transfer programme. Hum Reprod 7,117–119.[Abstract]

Straight AF and Field CM (2000) Microtubules, membranes and cytokinesis. Curr Biol 10,760–770[CrossRef]

Van Royen E, Mangelschots K, De Neubourg D, Valkenburg M, Van de Meerssche M, Ryckaert G, Eestermans W and Gerris J (1999) Characterization of top quality embryo, a step towards single-embryo transfer. Hum Reprod 14,2345–2349.[Abstract/Free Full Text]

Van Royen E, Mangelschots K, Vercruyssen M, De Neubourg D, Valkenburg M, Ryckaert G and Gerris J (2003) Multinucleation in cleavage stage embryos. Hum Reprod 18,1062–1069.[Abstract/Free Full Text]

Wolf JP, Bulwa S, Rodrigues D and Jouannet P (1995) Human oocyte cytometry and fertilisation rate after subzonal insemination. Zygote 3,101–109.[ISI][Medline]

Ziebe S, Petersen K, Lindenberg S, Andersen A-G, Gabrielsen A and Nyboe Andersen A (1997) Embryo morphology or cleavage stage: how to select the best embryo for transfer after in vitro fertilization. Hum Reprod 12,1545–1549.[Abstract]

Submitted on May 22, 2003; resubmitted on August 29, 2003; accepted on October 15, 2003.