Article Prediction of embryonic developmental competence by time-lapse observation and shortest-half analysis

Transcription

1 RBMOnline - Vol 17 No Reproductive BioMedicine Online; on web 30 September 2008 Article Prediction of embryonic developmental competence by time-lapse observation and shortest-half analysis Dr Arav received his bachelor degree at the Hebrew University of Jerusalem, his veterinary degree at the University of Bologna, Italy, and his PhD in the area of cryobiology and thermodynamics via a joint programme between the University of California, Berkeley and the University of Bologna. He completed his post-doctoral studies at UC Davis and holds a position at the Volcani Center as senior scientist in the field of cryobiology and reproduction. Dr Arav established IMT Ltd, Israel and has developed various issued patents. He has published over 100 papers and book chapters and received many awards. Dr Amir Arav Amir Arav 1,4, Adaya Aroyo 2, Saar Yavin 2, Zvi Roth 3 1 Institute of Animal Science, Agricultural Research Organization, the Volcani Centre, PO Box 6, Bet Dagan 50250, Israel; 2 IMT ltd Nes-Ziona, Israel; 3 Faculty of Agriculture, the Hebrew University, Rechovot, Israel 4 Correspondence: arav@agri.huji.ac.il Abstract Selecting an embryo with the highest probability of achieving a pregnancy is a major challenge. Early-cleavage embryos are considered to be of good quality; however, the exact developmental stage that predicts further development has not been defined. The aim of the study was to characterize cleavage rate and distribution of various stages of mouse preimplantation embryos using a time-lapse system. Mated mice were killed 20 h after human chorionic gonadotrophin administration and putative zygotes were recovered and cultured in an incubator-enclosed time-lapse imaging system. The shortest half analysis was used to establish the period in which at least 50% of the embryonic population cleaved within the shortest time. Analysis indicated that through embryonic development, cleavage timing becomes less uniform and the shortest half becomes longer with intervals of 2, 2.5, 3.5 and 5 h for 2-, 4-, 8-cell embryo and blastocyst stages, respectively. The shortest half for the first cleavage was closely synchronized, with 80% of embryos developing to the blastocyst stage. Moreover, slow-cleaving embryos approaching the 2-cell stage expressed inferior developmental potential in comparison to those cleaving within the shortest half. Thus, embryonic cleavage rate seems to be a biological indicator of developmental potential and may be useful for embryo selection. Keywords: cleavage timing, embryo, monitoring, shortest half, time lapse Introduction Identifying and selecting in-vitro-derived embryos with a high probability of establishing and maintaining pregnancy, and/or surviving cryopreservation, are major challenges facing the field of human assisted reproduction (Salumets et al., 2003; Van Montfoort et al., 2004) and are equally important for domestic animal embryo transfer programmes (Hansen, 2007). Although some studies have reported that blastocyst transfer improves pregnancy rates (Gardner et al., 1998, 2004; Shoukir et al., 1998; Jones and Trounson, 1999), the vast majority of transferred embryos fail to implant. Among the factors underlying poor embryo transfer outcomes is the lack of reliable criteria for selection of embryos with a high likelihood of developing to term. The predominant practice in most IVF laboratories for non-invasive evaluation is morphological grading of the cleaving embryos. Grading is based upon the degree of fragmentation, cytoplasmic appearance, and size and number of blastomeres per embryo (Cummins et al., 1986; Puissant et al., 1987; Steer et al., 1992; Gerris et al., 1999). In addition, the size, alignment and shape of pronuclei have also been utilized to predict embryo quality (Kattera and Chen, 2004). While some of these criteria have shown a positive correlation with implantation and pregnancy rates (Neuber et al., 2003), the eventual appearance of fragmented embryos emphasizes the limited value of criteria from early stages and rarely results in successful implantation (Erenus et al., 1991). Nevertheless, Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK

2 670 classical morphological criteria are not always adequate for embryo selection (Magli et al., 1998). Thus, currently, an efficient, convenient evaluation method is still not available. It is well known that cleavage-stage development is a dynamic process in which embryo morphology may change significantly over a time span of even a few hours (Gardner and Sakkas, 2003). Thus, conventional grading practices for selecting transferable embryos may not detect subtle differences between individual embryos, such as the time to progress from one cleavage division to the next (McKiernan and Bavister, 1994). Rapid cleavage, specifically as related to the time of the first mitotic division, is considered as predictive of good embryo quality and diagnostic as to whether an embryo is transferable or not (Shoukir et al., 1997; Sakkas et al., 1998). Similarly, the relationship between early-cleavage embryos and developmental competence was documented for in-vitro-derived bovine embryos (Dinnyes et al., 1999; Longergan et al., 1999). Moreover, early-cleavage embryos are more likely to survive cryopreservation with establishment of implantation and pregnancy following embryo transfer (Hasler et al., 1995; Van Montfoort et al., 2004). However, defining the temporal kinetics of early cleavage for each developmental stage (i.e. 2-cell, 8-cell and blastocyst) has shown wide variance between different studies (Shoukir et al., 1997; Sakkas et al., 1998; Fauque et al., 2008; Terriou et al., 2007). Thus, there is a real need for a better evaluation technique that will take into account measurable characteristics of early cleavage that could render embryo selection more objective and reliable. The aim of the present study was to evaluate the utility of time-lapse imaging during the culture of mice embryos to predict embryonic developmental competence. This approach facilitates non-invasive monitoring of living embryos under optimal culture conditions enabling a direct association between an individual embryo s cleavage rate or history with its future ability to undergo blastocyst formation. The data obtained allowed for implementation of the shortest-half statistical analysis, discovered by Rousseeuw and Leroy (1987) to evaluate the shortest time frame in which at least 50% of the embryonic population cleaves, which further extends the potential utility of this approach as an accurate method for selection of a cohort of competent embryos. Materials and methods Animals Mice used in this study were F1 hybrid CB6F1 females (C57BL/6J BALB/c; Harlan, Jerusalem, Israel). Three-weekold females were maintained under controlled photoperiod condition (14/10 h light/dark ratio) and fed water and Teklad pellets (Harlan, Jerusalem, Israel) ad libitum. All experimental procedures were approved by the Institutional Ethical Commission of the Hebrew University of Jerusalem. Chemicals and hormones Unless otherwise indicated, all chemicals were purchased from Sigma Chemical Co. (St Louis, MO, USA). Flushing holding medium was prepared according to Lawitts and Biggers (1993). Potassium simplex optimized medium (KSOM) was prepared according to Biggers et al. (2000) and supplemented with 0.1 mg/ ml polyvinyl alcohol (PVA), 10 µl/ml essential amino acids and 5 µl/ml non-essential amino acids purchased from Gibco BRL (Grand Island, NY, USA). Experimental design Female mice (n = 6; two replicates) were hormonally stimulated by interperitoneal administration of 5 IU equine chorionic gonadotrophin followed 48 h later by 5 IU of human chorionic gonadotrophin (HCG). Synchronized females were than paired with stud FVB/N males overnight at a 1:1 ratio. The presence of copulation plugs was considered as an indication that mating had occurred. Embryo recovery, culture and observation Mated mice were killed 20 h after HCG administration, and ampoules were transferred immediately into flushing holding medium. Putative zygotes were recovered in a 300 µg/ml drop of hyaluronidase, pooled and washed three times in fresh flushing holding medium and once in KSOM and divided into 20-µl drops of KSOM, each of which contained 20 embryos to be cultured under mineral oil. Culture dishes were placed in an EmbryoGuard TM unit (IMT Ltd, Ness Ziona, Israel; Arav, 2000) inserted in a tissue culture incubator (Forma 3110; Thermo Scientific, Ohio, USA) maintained at 37 C, 5% CO 2 in air and 95% humidity. The EmbryoGuard system is a robotic device enabling real-time monitoring of embryo dynamics without disturbing culture conditions (i.e. temperature, air composition and humidity). It comprises an inverted microscope, charged-coupled device cameras and a light source. The optical system is placed on a mobile optical table that can move along the X Y axes, and also the Z axis for focusing of all lenses. Embryos can be observed at three different magnifications: 4, 200, Automatic time-lapse photographs can be taken at specific preset intervals and are saved on a database for subsequent analysis. The camera optics enable both automatic continuous time-lapse photography as well as real-time microscopic observation of individual embryos. In this study, the unit was programmed to collect images at intervals of 30 min. Statistical analysis Differences between treatments were subjected to one-way analysis of variance. The level of significance was P < 0.05 unless otherwise mentioned; all distribution values were analysed according to confidence intervals of 0.95; and the chi-squared test was used to compare data among the experimental groups by using the JMPIN software (SAS Institute Inc, 2004). The shortest half statistical analysis (Rousseuw and Leroy, 1987) was used to evaluate the period in which at least 50% of the embryonic population cleaved within the shortest time frame. Results Cleavage timing characterization by timelapse surveillance A total of 277 putative mouse zygotes that developed to the 2-cell stage were subjected to individual continuous in-situ time-lapse surveillance. Only 75% (207/277) of the 2-cell-stage embryos

3 advanced further to the 4-cell stage. About 87% (181/207) of the 4-cell-stage embryos developed to the compaction stage and most of the latter (90%, 163/181) developed to the blastocyst stage. Developmental dynamics through subsequent embryonic stages is shown in Figure 1. The curves represent the cumulative number of embryos that completed the cleavage cycles in the course of h after HCG administration. The formation of 2-cell embryos through the first embryonic cleavage began 28 h after cell-stage HCG administration; the formation of 4-cellstage embryos (second cleavage) began after 6 h developmental plateau at the 2-cell stage, i.e. 45 h after HCG administration. The third embryonic cleavage into the 8-cell stage started 61 h after HCG administration and compact embryos were first observed 6 h later. Blastocyst formation extended over 21 h beginning at 89 h afterhcg administration. Given the distribution of cleavage timing among embryos and within each individual cleavage cycle, it was possible to define for each developmental stage the time frame in which cleavage occurred. Cleavage timing analysis by shortest half procedure Putative mouse zygotes that cleaved and developed to the 2-cell embryonic stage were subjected to individual continuous insitu time-lapse surveillance as described above. A total of 230 putative zygotes cleaved and developed to the 2-cell stage, of which 55 embryos were arrested at further developmental stages and only 175 (76%) embryos progressed and developed to the blastocyst stage (Table 1). To identify the best developmental stage that would predict further embryonic development, only data from embryos that developed to the blastocyst stage were analysed by the shortest-half analysis. This procedure enabled the determination for each developmental stage of the shortest frame of time during which at least 50% of the embryo population cleaved (i.e. densest 50% of the observation). For each developmental stage, embryos were categorized into three groups according to time of cleavage: i.e. before, within and after the shortest-half time. With respect to the first cleavage (Figure 2a), embryos that cleaved into the 2-cell stage before the shortest half time (up to 33.5 h after HCG administration) were defined as fast-cleaving embryos; those that cleaved within the shortest-half time ( h) were defined as shortest-half population and those that cleaved after the shortest-half time (35.5 h or more) were designated as slow-cleaving embryos. About 80% (93/116) of embryos that cleaved into the 2-cell stage within the shortest-half time developed to blastocyst stage. Interestingly, the proportion of shortest half 2-cell embryos that developed to blastocysts did not differ from that of fast-cleaving embryos (95%, 19/20) but was higher than that of the slow-cleaving embryos (67%, 63/94; P < 0.05). Note, that the mean cleavage timing into the 2-cell stage of embryos that further developed to the blastocyst stage was lower than that of arrested embryos (34.6 ± 1.5 versus 35.8 ± 1.9 h, respectively; P < 0.001). Moreover, the distribution of cleavage timing also differed (P < ) between these two groups: of the arrested embryos only 2% were fast cleaving, 42% cleaved within the shortest-half time and 56% were slow cleaving. Of those embryos that progressed to blastocysts, 11% were fast cleaving, 53% cleaved within the shortest-half time and only 36% were slow cleaving (Table 1). With respect to the time frame of cleavage, the analysis revealed that as early embryonic development proceeds, cleavage timing becomes less uniform and the shortest half becomes longer for each successive cell division (Figure 2). In particular, shortesthalf time was extended during successive cleavage division and was 2.5, 3.5, and 5 h for the second and third cleavages and blastocyst formation, respectively. The first cleavage into the 2-cell-stage development was closely synchronized, begun as early as 30 h after HCG administration and extended over 8 h, with 50% of embryos cleaving within a 2 h interval (i.e. shortest half ). Cleavage into the 4-cell stage (second cleavage) was initiated 49 h after HCG injection and extended over 12 h. The third cleavage into the 8-cell stage extended over 16 h, from 62.5 until 78.5 h after HCG injection and blastocyst formation extended over 22.5 h hours, from 91 until h after HCG injection. These findings also indicate that embryos that cleaved early (i.e. fast cleaving) to the 2-cell stage and those that cleaved within the shortest-half time, represented embryos with the ability to become blastocysts within 100 h from HCG administration i.e. before and within the shortest-half time (Figure 2d). Figure 1. Cleavage timing of different developmental stages. Embryonic stages were subjected to individual continuous in-situ timelapse surveillance. HCG = human chorionic gonadotrophin. 671

4 Table 1. Developmental competence of 2-cell-stage embryos grouped as fast-cleaving (<33.5 h after HCG); shortest-half ( h after HCG) and slow-cleaving (>35.5 h after HCG) embryos. Fast-cleaving Shortest-half Slow-cleaving embryos (%) embryos (%) embryos (%) Arrested embryos 1/55 (2) 23/55 (42) 31/55 (56) Embryos that formed blastocysts 19/175 (11) 93/175 (53) 63/175 (36) Total 20/230 (9) 116/230 (50) 94/230 (41) The shortest half corresponds to the period in which at least 50% of the embryonic population cleaved within the shortest time frame. Figure 2. Distribution of cleavage timing at different embryonic developmental stages. Presented are the data of embryos that developed to the blastocyst stage. Time-lapse photographs were taken at half-hour intervals. The shortest-half statistical analysis was used to evaluate the period in which at least 50% of the embryonic population cleaved within the shortest time frame (black columns). HCG = human chorionic gonadotrophin. 672 Discussion Identification and selection of in-vitro-derived embryos exhibiting the highest developmental potential remains a major challenge in the field of assisted reproduction. The present study introduces a novel approach for observation and evaluation of a cohort of developing embryos that will be valuable in assessing embryo variability and population-based screens for altered culture conditions. Timelapse imaging is shown to provide cleavage timing assessment for an individual embryo that can be related to the potential to form a blastocyst. Moreover, the shortest half analysis makes possible the evaluation of the shortest frame of time in which at least 50% of the embryo population underwent cleavage. Such information seems to be highly relevant for predicting embryo developmental competence since the proportion of embryos that cleaved into the 2-cell stage within the shortest-half time and subsequently developed to the blastocyst stage was relatively high, implying that these embryos were of good quality. Early cleavage and specifically the time of the first mitotic division are considered as indicative features of embryo quality in particular, whether the embryo is transferable (Hasler et al., 1995;

5 Shoukir et al., 1997; Sakkas et al., 1998) or competent to survive cryopreservation (Hasler et al., 1995; Van Montfoort et al., 2004). Nevertheless, the accurate time that is defined as early cleavage for each of the developmental stages (i.e.; 2-, 4-, 8-cell embryo and blastocyst) has not been defined yet. Accurate cleavage timing for each developmental division of mice embryos is reported for the first time here. The findings indicate that as early embryonic development proceeds, the cleavage timing becomes less uniform and the shortest half time becomes longer for each successive cell division. In fact the shortest half time was extended from 2.5, 3.5, and 5 h for the second and third cleavages and blastocyst formation, respectively, suggesting that an embryo s outcome can be forecasted as early as the first embryonic cleavage into the 2-cell stage. This assumption is supported by the findings that 50% of the 2-cell embryos were closely synchronized and cleaved within 2 h (i.e. the shortest half time) and that about 80% of this embryonic population developed to the blastocyst stage. Furthermore, the proportion of the shortest-half embryos that developed to blastocyst did not differ statistically from that noted for early-cleavage embryos, which is known to be associated with high developmental competence (Shoukir et al., 1997; Sakkas et al., 1998; Bos-Mikich et al., 2001). It should be noted, however, that in the present study only a small number of embryos did actually cleave early (9%). As fast-cleaving embryos are linked in some cases to chromosomal abnormalities (Magli et al., 2001; Rienzi et al., 2005), the shortest-half embryos are suggested as the main cohort of competent embryos. Although, the percentage of early-cleavage embryos that reached the blastocyst stage (95%) was higher (though not statistically different) from the shortest half embryos (80%), the total number of embryos in the shortest half group was five-fold higher then that of the early-cleavage group. Therefore, in human IVF where the number of embryos cultured is small, the selection of the embryos from the shortest half group will be advantageous. During in-vitro culture, there is a real conflict in the delicate balance between the need to maintain stable culture conditions and the need to evaluate embryos outside the incubator, so exposing them to suboptimal conditions. This conflict becomes more critical because embryo evaluation on a sequential assessment basis, which takes several development stages into account, appears to be superior to selection of the best embryos at a specific time point (Gardner and Sakkas, 2003; Neuber et al., 2003). Thus, use of the time-lapse system inserted in an incubator facilitated non-invasive continuous monitoring of individual or a cohort of embryos under optimal culture conditions. It my be claimed that using this system could adversely affect embryonic development since it is equipped with LED illumination, however, previous experiments showed that the low light intensity had no effect on mouse embryonic development (Yavin et al., 2003). In addition, the time-lapse system enables the tracking of changes such as occurrence and disappearance of fragmentation, cytoplasmic appearance and number and size of blastomeres during early stages of embryonic development. At high magnification, the system is able to detect nucleation in the embryo. During on-line observations it was found that 25% of the 2-cell-stage embryos did not continue to develop, which could be attributed to the 2-cell-stage block phenomenon (Whitten, 1957; Cole and Paul, 1965). It was also found that cell division is a well-synchronized event, and that cleavage duration to form a 2-cell-stage embryo takes approximately 8 min (Figure 3) while the development of retarded embryos takes longer, presumably due to longer cell cycle duration. The time-lapse system therefore enabled accurate evaluation of developmental and morphological features of the growing embryos. Improved ovarian stimulation protocols, as well as enhanced culture conditions for embryos, have led to both the harvesting of large numbers of oocytes and, as a result, Figure 3. Mouse embryo cleaving process as detected by the EmbryoGuard system. The entire cleavage process documented was 8.5 min long: (a) initiation of cleavage into the 2-cell stage (32:50:28 after human chorionic gonadotrophin administration), (b e) cleavage progress, (f) formation of a 2-cell stage embryo (32:58:50). 673

6 674 culturing of large quantities of embryos that have equivalent morphological scores (Fenwick et al., 2002). Decreased rate of multiple births without decreasing the overall birth rate can be achieved by applying less vigorous stimulation protocols, which in turn leads to the transfer of only a single embryo. Alternatively, enhancing the capability of selecting optimal embryos for transfer has been suggested (Coetsier and Dhont, 1998; Neuber et al., 2003). With respect to the latter, selecting the most suitable embryos among those that cleaved within the shortest half might offer an improved focus of the selection task. Practically, it is well known that the larger the number of embryos, the more difficult is the selection for the most competent embryos. However, this drawback might become an advantage when using a time-lapse observation system and the shortest-half method for selecting embryos. For example, culturing a large number of embryos for an individual patient, through the use of improved stimulation and culture conditions, might enable a more precise analysis which might increase the efficiency of the shortest half evaluation method. In light of the current findings, it is suggested that while using the time-lapse system, embryo selection should be based on the first embryonic cleavage into the 2-cell stage. Such identification might enable earlier embryo transfer and correspondingly earlier exposure of the embryos to optimal maternal environmental conditions (Emiliani et al., 2003). Supporting this assumption are the findings that extended culture has a negative effect on the embryo developmental and implantation potential (Levron et al., 2002; Kolibianakis et al., 2004) and, from this study, that only 67% of the embryos that displayed delayed cleavage into the 2-cell stage further developed to the blastocyst stage. It should be noted however, that the cleaving features reported here are mouse-specific, thus follow-up examination for other mammalian species or domestic animals should be performed. Furthermore, it should be kept in mind that data of cleavage timing based on the shortest-half analysis are considered as laboratory-dependent tools that take into account all the subjectively determined factors that could influence in-vitro embryonic development; these include culture conditions and fertilization method. In summary, early embryo parameters are becoming the key for selecting embryos with high developmental competence. Corresponding to this concept, a novel approach for embryo evaluation was proposed that combined a time-lapse system and the shortest-half analysis. It is believed that this approach offers an opportunity to collect accurate information about embryo cleavage timing and developmental dynamic that might be used for selecting competent embryos. It seems that embryos cleaved within the shortest-half time of the first cleavage and exhibiting a high morphological scoring could prove helpful during routine laboratory work, in identifying embryos of high developmental competence. It should be noted, however, that is not clear whether the shortest half can serve as independent predictor. Thus, further investigations are warranted to determine whether embryos selected by the shortest-half analysis, aside from other known predictors, are competent to achieve an ongoing pregnancy or to survive cryopreservation. References Arav A 2000 Method and apparatus for monitoring a biological sample. US patent 6,166,761. Biggers JD, McGinnis LK, Raffin M 2000 Amino acids and preimplantation development of the mouse in protein-free potassium simplex optimized medium. Biology of Reproduction 63, Bos-Mikich A, Mattos AL, Ferrari AN 2001 Early cleavage of human embryos: an effective method for predicting successful IVF/ICSI outcome. Human Reproduction 16, Coetsier T, Dhont M 1998 Avoiding multiple pregnancies in in-vitro fertilization: who s afraid of single embryo transfer? Human Reproduction 13, Cole R, Paul J 1965 Properties of cultured preimplantation mouse and rabbit embryos and cell strains derived from them. In: Wolstenholme GEW, O Connor M (eds) Preimplantation Stages of Pregnancy: a Ciba Foundation Symposium. Churchill Ltd, London, pp Cummins JM, Breen TM, Harrison KL et al A formula for scoring human embryo growth rates in in-vitro fertilization: its value in predicting pregnancy and in comparison with visual estimates of embryo quality. Journal In Vitro Fertilization Embryo Transfer 3, Dinnyés A, Lonergan P, Fair T et al Timing of the first cleavage post-insemination affects cryosurvival of in vitro-produced bovine blastocysts. Molecular Reproduction and Development 53, Emiliani S, Delbaere A, Vannin AS et al Similar delivery rates in a selected group of patients, for day 2 and day 5 embryos both cultured in sequential medium: a randomized study. Human Reproduction 18, Erenus M, Zouves C, Rajamahendran P et al The effect of embryo quality on subsequent pregnancy rates after in-vitro fertilization. Fertility and Sterility 56, Fauque P, Davy C, Jouannet P, Patrat C 2008 Case report: How to manage an accelerated embryo cohort in IVF? Reproductive BioMedicine Online 16, Fenwick J, Platteau P, Murdoch AP, Herbert M 2002 Time from insemination to first cleavage predicts developmental competence of human preimplantation embryos in vitro. Human Reproduction 17, Gardner DK, Sakkas D 2003 Assessment of embryo viability: the ability to select a single embryo for transfer a review. Placenta 24 (suppl. B), S5 S12. Gardner DK, Surrey E, Minjarez D et al Single blastocyst transfer: a prospective randomized trial. Fertility and Sterility 81, Gardner DK, Vella P, Lane M et al Culture and transfer of human blastocysts increases implantation rates and reduces the need for multiple embryo transfers. Fertility and Sterility 69, Gerris J, De Neubourg D, Mangelschots K et al Prevention of twin pregnancy after in-vitro fertilization or intracytoplasmic sperm injection based on strict embryo criteria: a prospective randomized clinical trial. Human Reproduction 14, Hansen P J 2007 To be or not to be determinants of embryonic survival following heat shock. Theriogenology 68, (Suppl. 1), S40 S48. Jones GM, Trounson AO 1999 Blastocyst stage transfer: pitfalls and benefits. The benefits of extended culture. Human Reproduction 14, Kattera S, Chen C 2004 Developmental potential of human pronuclear zygotes in relation to their pronuclear orientation. Human Reproduction 19, Kolibianakis EM, Zikopoulos K, Verpoest W et al Should we advise patients undergoing IVF to start a cycle leading to a day 3 or a day 5 transfer? Human Reproduction 19, Lawitts JA, Biggers JD 1993 Culture of preimplantation embryos. Methods in Enzymology 225,

7 Levron J, Shulman A, Bider D et al A prospective randomized study comparing day 3 with blastocyst-stage embryo transfer. Fertility and Sterility 77, Lonergan P, Khatir H, Piumi F et al Effect of time interval from insemination to first cleavage on the developmental characteristics, sex ratio and pregnancy rate after transfer of bovine embryos. Journal of Reproduction and Fertility 117, Magli MC, Gianaroli L, Ferraretti AP 2001 Chromosomal abnormalities in embryos. Molecular and Cellular Endocrinology 183 (Suppl. 1), S29 S34. Magli MC, Gianaroli L, Munne S, Ferraretti AP 1998 Incidence of chromosomal abnormalities from a morphologically normal cohort of embryos in poor-prognosis patients. Journal of Assisted Reproduction Genetics 15, McKiernan SH, Bavister BD 1994 Timing of development is a critical parameter for predicting successful embryogenesis. Human Reproduction 9, Neuber E, Rinaudo P, Trimarchi JR, Sakkas D 2003 Sequential assessment of individually cultured human embryos as an indicator of subsequent good quality blastocyst development. Human Reproduction 18, Puissant F, Van Rysselberge M, Barlow P et al Embryo scoring as a prognostic tool in IVF treatment. Human Reproduction 2, Rienzi L, Ubaldi F, Iacobelli M et al Significance of morphological attributes of the early embryo. Reproduction BioMedicine Online 10, Rousseeuw PJ, Leroy AM 1987 Robust Regression and Outlier Detection. John Wiley and Sons, New York. Sakkas D, Shoukir Y, Chardonnens D et al Early cleavage of human embryos to the two-cell stage after intracytoplasmic sperm injection as an indicator of embryo viability. Human Reproduction 13, Salumets A, Hyden-Granskog C, Makinen S et al Early cleavage predicts the viability of human embryos in elective single embryo transfer procedures. Human Reproduction 18, Shoukir Y, Chardonnens D, Campana A et al The rate of development and time of transfer play different roles in influencing the viability of human blastocysts. Human Reproduction 13, Shoukir Y, Campana A, Farley T, Sakkas D 1997 Early cleavage of in-vitro fertilized human embryos to the 2-cell stage: a novel indicator of embryo quality and viability. Human Reproduction 12, Steer CV, Mills CL, Tan SL et al The cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and embryo transfer programme. Human Reproduction 7, Terriou P, Giorgetti C, Hans E et al Relationship between even early cleavage and day 2 embryo score and assessment of their predictive value for pregnancy. Reproductive BioMedicine Online 14, Van Montfoort AP, Dumoulin JC, Kester AD, Evers JL 2004 Early cleavage is a valuable addition to existing embryo selection parameters: a study using single embryo transfers. Human Reproduction 19, Whitten WK 1957 Culture of tubal ova. Nature 179, Yavin S, Aroyo A, Arav A 2003 Affect of low intensity light on mice embryo development using the time lapse EmbryoGuard system. Fertility and Sterility 80, Declaration: Dr Amir Arav is the inventor of the EmbryoGuard time-lapse imaging system. He established IMT Ltd (Israel), the company which sells the EmbryoGuard system. The remaining authors report no financial or commercial conflicts of interest. Received 14 January 2008; refereed 7 February 2008; accepted 18 June