Improvement in early human embryo development using new formulation sequential stage-specific culture media

Transcription

1 FERTILITY AND STERILITY VOL. 78, NO. 6, DECEMBER 2002 Copyright 2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Improvement in early human embryo development using new formulation sequential stage-specific culture media Simon Cooke, B.Sc.Agr., a Patrick Quinn, Ph.D., b Lee Kime, B.Sc., a Cheryl Ayres, M.Med.Sci., a John P. P. Tyler, Ph.D., a and Geoff L. Driscoll, M.D. a CityWest IVF, Westmead, New South Wales, Australia Objective: To determine whether altering selected components of sequential culture media can improve early development variables of human embryos. Design: Prospective, randomized, sibling oocyte split trial. Setting: Private ART center. Patient(s): Two hundred eight undergoing treatment with in vitro fertilization or microinjection. Intervention(s): Oocytes from each patient were randomly allocated to fertilization and cleavage media of a control and a trial culture formulation. Main Outcome Measure(s): Rates of fertilization, cleavage, and uncontrolled division; average embryo morphology score; blastomeres per embryo; embryo score parameter (number of blastomeres embryo morphology grade); and embryo utilization. Result(s): The trial media resulted in a higher fertilization rate, higher cleavage rate, lower rate of uncontrolled division, higher number of blastomeres per embryo, higher average embryo morphology score, a higher embryo score parameter, and higher embryo utilization rate compared to the control media. All differences were statistically significant. Conclusion(s): Improved sequential stage-specific culture media can reduce the occurrence of severe human embryo fragmentation and improve developmental variables in early IVF- and ICSI-generated embryos. (Fertil Steril 2002;78: by American Society for Reproductive Medicine.) Key Words: Sequential culture media, early embryo development, embryo morphology, in vitro development, uncontrolled division Received January 11, 2002; revised and accepted April 16, Reprint requests: Simon Cooke, B.Sc.Agr., CityWest IVF, 12 Caroline Street, Westmead, New South Wales 2145, Australia (FAX: ; lab@ivf.com.au). a CityWest IVF. b SAGE BioPharma, San Clemente, California /02/$22.00 PII S (02) The embryology laboratory plays a critical role in determining the success rate of a human IVF program. One of the most important components that influences the success of the laboratory is the quality and composition of the culture used. Human tubal fluid (HTF) (1), has been used in many IVF programs worldwide, but this culture is now being replaced by media based on more current physiologic evidence of the composition of the female reproductive tract environment (2). Glucose concentrations of 2.8 mm are sufficient to maintain sperm velocity in fertilization (3), and glucose is needed in cleavage stage embryos to satisfy such requirements as the pentose phosphate pathway. Phosphate should be removed entirely to minimize the negative effects of the combination of glucose and phosphate (1, 4). Magnesium concentration should be increased to compensate for an increase in intracellular calcium induced by in vitro culture that has been shown to be detrimental to embryo development in vitro (5, 6). Other ingredients, such as taurine, have been shown to be beneficial for in vitro culture of embryos (7, 8), and both taurine and ethylenediamine tetraacetic acid (EDTA) provide antioxidant activity. However, elevated concentrations of EDTA may be responsible for granularity in human embryos (Quinn P. Unpublished data). Inclusion of citric acid may be important, as it has been demonstrated to support early mouse preimplantation development (9) and is a component in the tricarboxylic acid cycle (10). Glutamine, and in particular, the type of glutamine, has been shown to be beneficial for 1254

2 human embryo development in vitro (11). Alanyl-glutamine is a very stable form of glutamine that does not break down and release ammonium. Ammonium buildup in has been shown to be embryotoxic (12). Suboptimal culture conditions have also been noted as a possible cause of low embryo viability after transfer (13). Affected embryo variables included delayed cell division (14), less incorporation of amino acids (15), and a reduction in oxidative metabolism and increase in lactate production (16 18). More recent studies (2) have shown that the use of sequential media that have components at the various sites in the female reproductive tract at different stages of early pregnancy has produced higher pregnancy and implantation rates. Previous studies were retrospective and involved one defined media (10), compared two different media (19, 20), or were done at different stages of embryo growth (19). There have been few prospective, comparative trials of improved formulations of a media series. In our study, trial media were formulated based on recent information regarding the physiological environment for human gametes and embryos and of other relevant mammalian species. We sought to prospectively compare a new series of sequential culture media with the standard HTF formulations that were currently in use in our clinic for human IVF culture. MATERIALS AND METHODS Trial Design The experiment was designed as a sibling oocyte split, in which oocytes were allocated to a particular by using a random-numbers series. Patients undergoing IVF had their oocytes allocated to each during oocyte retrieval, whereas patients undergoing ICSI had their oocytes allocated to the culture media after the cumulus corona complex had been removed and nuclear maturity had been assessed. Patients were excluded if they were 40 years of age at the commencement of a treatment attempt; if at least one embryo was not generated in each of the control and trial media; and if patients were having ICSI using testicular surgically retrieved sperm, because these patients have a reduced fertilization rate within our clinic. Two hundred fifty-four patients had oocytes allocated to the trial. Of these, 46 were excluded from analysis (15 because of failed fertilization and 31 because less than one embryo was generated in each media treatment), leaving 208 patients eligible for inclusion in the media trial. These 208 patients had an average age of 32.6 years (range, years) and their duration of infertility ranged from 0.7 to 6.0 years. Thirty-one of 254 patients were excluded from the trial because only one embryo was generated or the patient had few oocytes that failed to fertilize with IVF insemination (15 of 254 patients). Our study required one embryo to be generated in each media group. If any initial oocyte inclusion number had been chosen for inclusion in the trial, it would prove to be an unsatisfactory selection criterion because it would select for better-performing patients and a better outcome. The broad inclusion criteria that we used permitted a large cross-section and, therefore, generation of data that are representative of a total ART patient population rather than a highly selected population. Twenty-six patients had embryos generated but did not receive an embryo transfer; instead, all utilizable embryos were frozen because of medical concerns about overstimulated ovaries. Data from these patients were included because the study was concerned with early embryonic development and not pregnancy outcome. There were 107 ICSI cycles, mostly because of male factor infertility, and 101 IVF cycles done because of tubal abnormalities, endometriosis, or unexplained infertility. The institutional ethics committee approval was granted for this study. Culture Media Two sequential stage-specific culture media were used. The control was commercially available and consisted of a fertilization (HTF; SAGE BioPharma, Bedminster, NJ) and a cleavage (Basal XI; SAGE BioPharma). The trial was a modification of the control and consisted of a fertilization and a cleavage (SAGE BioPharma). The antibiotic content, manufacturing process, storage, utilization, and osmolarity of the media were identical. All media met standard manufacturing specifications, including a quality control test with one-cell mouse embryos in which at least 80% of zygotes cultured in the media developed to the fully expanded blastocyst stage over 96 hours of culture. Table 1 shows alterations made to the trial. Controlled Ovarian Hyperstimulation Controlled ovarian hyperstimulation was achieved by a long down-regulation protocol involving GnRH agonist and FSH. The GnRH agonist was administered as a s.c. injection (Lucrin [leuprorelin acetate]; Abbott, Kurnell, New South Wales, Australia) or a nasal spray (Syneral [nafarelin acetate solution]; Searle, Rydalmere, New South Wales, Australia). Recombinant FSH (Gonal-F; Serono, Frenchs Forrest, New South Wales, Australia, or Puregon; Organon, Lane Cove, New South Wales, Australia) was used to induce follicular growth. At least 5,000 IU of hcg (Profasi; Serono) was administered to achieve final maturation and trigger ovulation when at least two follicles were 18 mm in diameter according to transvaginal ultrasonography. FERTILITY & STERILITY 1255

3 TABLE 1 Alterations to the ingredients of the trial fertilization and cleavage. Culture stage Ingredient alteration (reference) Fertilization 2.8 m/mol Glucose (3) Fertilization and cleavage EDTA reduced to 0.01m/Mol Fertilization and cleavage Magnesium increased to 2.0 mmol (5, 6) Fertilization and cleavage Taurine added at 0.1 mmol (7, 8) Fertilization and cleavage Citric acid added at mmol (7, 9, 10) Fertilization and cleavage Alanyl-glutamine added at 1.0 mmol (11) Fertilization and cleavage Essential amino acids added (34) Cleavage Glucose reduced to 0.1 m/mol (1, 4) Cleavage Phosphate removed (1, 4) Note: EDTA ethylenediamine tetraacetic acid Cooke. Embryonic development in sequential media. Fertil Steril Thirty-eight hours later, ultrasonography-guided oocyte retrieval was performed by using a single-lumen 16-gauge ovum pick-up set (COOK, Eight Mile Plains, Queensland, Australia). All oocytes were returned to the IVF laboratory in Hepes-buffered HTF (SAGE BioPharma) at 37 C in a portable incubator (LEC-960, LEC Instruments, Scoresby, Victoria, Australia). Culture Conditions Oocytes destined for microinjection had their cumulus cells removed by 30 seconds of exposure to hyaluronidase (Sigma Type VIII Bovine Testis Enzyme; Sigma, Castle Hill, New South Wales, Australia), 80 IU/mL, in Hepesbuffered HTF, followed by washing with Hepes-buffered HTF containing human serum albumin (SAGE BioPharma), 5 mg/ml. The coronal cells were mechanically removed by using a finely drawn glass Pasteur pipette. Denuded oocytes were then assessed for their structural integrity and meiotic maturity and placed into culture in 1.0 ml of the control or trial fertilization with an oil overlay, in a 4-well Nunc dish (Nunclon; Medos, Lidcome, New South Wales, Australia) that had been pre-equilibrated at 37 C in 6% carbon dioxide in room air. After randomization using a random-numbers table, groups of three or four oocytes intended for insemination by conventional IVF were placed directly in a culture dish and inseminated with approximately 100,000 motile sperm. Oocytes for ICSI were microinjected 42 hours after the ovulatory hcg injection in a manner similar to that described by Van Steirteghem et al. (21) and then returned to culture. All oocyte embryo culture was performed in a mini incubator (MINC-1000; COOK) supplied with a humidified triple gas mixture of 6% CO 2,5%O 2, and 89% N 2 at 37 C. Fertilization was assessed 16 to 20 hours after microinjection or addition of the spermatozoa. Oocytes that fertilized normally on day 1 were washed in their respective cleavage. Five or fewer zygotes were grouped in 50- L droplets of for culture. Embryos were graded 40 to 41 hours after addition of sperm, and embryo transfer was performed transcervically 1 to 3 hours later by using a double-lumen catheter (K-JETS 7019; COOK). All patients received luteal phase supplementation with progesterone (Orion Laboratories, Welshpool, Western Australia, Australia), given as a 200-mg vaginal pessary daily for 14 days starting at oocyte retrieval, or as a 2,000-IU injection (Profasi; Serono) 2 days after oocyte retrieval, followed by three further injections at 72-hour intervals. Early Embryonic Development Fertilization rates were standardized and expressed on the basis of meiotically mature oocytes because both ICSI and IVF insemination was performed. Oocyte maturity was assessed initially at the time of cumulus cell removal in patients undergoing ICSI and during fertilization checking on day 1 in patients undergoing IVF. Oocytes that became necrotic after ICSI were excluded. The fertilization rate was calculated as follows: rate (%) [(no. of fertilized oocytes)/((no. meiotically mature) (no. necrotic after ICSI))] (100). Cleavage was defined as the presence of a two- or four-cell embryo approximately 40 to 41 hours after the addition of spermatozoa. The size and regularity of blastomeres were carefully assessed to allow determination of controlled and uncontrolled cleavage. Embryo cleavage and grading were assessed within a 2-hour window to obtain a measurement of the speed of cleavage and embryo morphology. Morphology grading was based on fragmentation and blastomere regularity (22 27). Each embryo was allocated a score from 4 to 1 according to the regularity of the blastomeres and the amount of embryo fragmentation. Grade 4 embryos had equal-sized blastomeres, were spherical in shape, and contained no extracellular fragmentation. Grade 3 embryos had 10% extracellular fragmentation. Grade 2 embryos had 10% to 50% extracellular fragmentation. Grade 1 embryos were of poor quality and had 50% fragmentation. Uncontrolled division was defined as the complete fragmentation of embryos. Unlike grade 1 embryos, which had at least some clearly defined cells, uncontrolled division embryos have no visible normal blastomere structure and masses of very small fragments, more similar to an apoptotic 1256 Cooke et al. Embryonic development in sequential media Vol. 78, No. 6, December 2002

4 event than normal cell division. Structures that had uncontrolled division were not assessed for embryo morphology because they were not considered to be dividing normally as an embryo. The average embryo score reported here is an accurate indication of the morphology of a dividing embryo. The embryo score parameter per cleaved embryo was defined as the average morphology score for embryos generated in a particular culture, multiplied by the average number of blastomeres produced in the embryos that cleaved in a controlled manner. As an example, a morphology score of 4 in a four-cell embryo assessed early on day 2 is 16: [4 4] (26). The embryo score parameter is an average score that incorporates both the morphology score and the cleavage speed of the embryos. Statistical Analysis The Wilcoxon signed-rank test (SPSS software, version 0.7; SPSS, Inc., Chicago, IL) was used to analyze fertilization rates, cleavage rates, uncontrolled division, number of blastomeres per embryo, average embryo morphology, and embryo score parameter between media and within patients. Differences in overall proportional data were analyzed by using the 2 test. The patient population was analyzed as a pooled group (IVF-generated and ICSI-generated embryos) and by individual treatments (IVF-generated embryos and ICSI-generated embryos). RESULTS Early development parameters were assessed in 2,009 oocytes from 208 treatment cycles. This included 991 oocytes from 101 treatment cycles inseminated by IVF and 1,018 oocytes from 107 treatment cycles inseminated by ICSI. Table 2 shows the results of early embryonic development as determined by pooled treatment data (IVF and ICSI) and by treatment type. When the pooled treatment cycles were assessed, all variables showed significant improvement when oocytes were cultured in the trial compared with the control. When assessed by the type of treatment used for insemination (IVF or ICSI), the trial also offered significant improvements compared with the control. The average number of blastomeres per embryo was the only early embryonic developmental variable that did not differ statistically when analyzed by insemination method; however, with the larger embryo numbers in the pooled data, this variable was significantly increased in embryos cultured in trial. The rates of cleavage and uncontrolled division differed greatly between embryos cultured in control and trial media, regardless of whether they were derived by IVF or ICSI. In the control, cleavage rates (88.8%) were significantly lower (P.0001) and rates of uncontrolled division (7.8%) were significantly higher (P.0001) than the corresponding values observed in the trial (94.5% and 2.7%, respectively). When rates of cleavage and uncontrolled division in each were compared between embryos produced by IVF or ICSI, rates of cleavage were always significantly higher and rates of uncontrolled division always significantly lower in embryos generated in trial. The increase in the early embryonic development variables in embryos generated in trial also led to a significant increase in the rate of utilization of these embryos (68.2%) compared with embryo generated in control (55.9%; ; P.005). Embryo utilization was defined as the total number of embryos transferred or frozen divided by the number of embryos that actually cleaved. A phenomenon that became apparent during the trial was the presence of intrablastomere pitting in embryos cultured in control and the absence of this pitting in embryos cultured in trial (Fig. 1). This phenomenon occurred independently of embryo fragmentation. Pregnancy was not a measured outcome in this study because the best-quality embryos from either were replaced. Embryo transfer occurred in 182 of the 208 cycles. Cryopreservation only occurred in the remaining 26 cycles. Ninety-six (53%) of the transfers resulted in a clinical pregnancy confirmed by ultrasonography. On average, 1.99 embryos per patient were replaced, and the implantation rate was 35%. Of the transferred embryos, 260 (71.6%) were derived from the trial and 103 (28.4%) from the control. As a historical reference point, in 182 patients matched for age before the start of the trial, the implantation rate was 23.9% using embryos generated from control. DISCUSSION The cleavage rate (27) and morphology (27, 28) of IVFgenerated embryos are reported to be more normal than those of ICSI-generated embryos (27). Our study confirms these findings. Most previous studies did not compare the outcome of improved media formulations from the same manufacturer or looked carefully at the cleavage or control of the cleavage in embryos. Regardless of the type of insemination, use of our trial improved all early embryonic development variables compared to the original control. The definitions that we used for cleavage and uncontrolled division provided an accurate indication of what constitutes controlled or uncontrolled division. One consequence of these definitions is that the zygote cleavage rate may appear somewhat low between the two media (88.8% vs. 94.5%). However, this reflects our requirement that embryos with controlled cleavage have clearly defined blastomeres. An absence of defined blastomeres was considered FERTILITY & STERILITY 1257

5 TABLE 2 Early embryonic development variables measured between pooled cycles (IVF and ICSI) and separately by the insemination method. Variable Insemination treatment Control Trial Value P value Fertilization rate (%) Pooled (IVF ICSI) Z IVF Generated Z ICSI Generated Z Cleavage rate (%) Pooled (IVF ICSI) Z IVF Generated Z ICSI Generated Z Rate of uncontrolled Pooled (IVF ICSI) Z division (%) IVF Generated Z ICSI Generated Z Morphology score a Pooled (IVF ICSI) Z IVF Generated Z ICSI Generated Z Embryo score parameter Pooled (IVF ICSI) Z IVF Generated Z ICSI Generated Z No. of blastomeres per embryo Pooled (IVF ICSI) Z IVF Generated Z NS ICSI Generated Z NS Embryo utilization rate (%) Pooled (IVF ICSI) IVF Generated ICSI Generated Note: Data after the plus/minus sign are SDs. NS not significant. a Of a possible 1 (worst) to 4 (best). Cooke. Embryonic development in sequential media. Fertil Steril uncontrolled division, which was significantly reduced in the trial (7.8% vs. 2.7%). Our findings suggest that the rate of uncontrolled division or complete fragmentation can be mostly attributed to the culture media and their formulation and that the occurrence of severe fragmentation or uncontrolled division is not always the expression of the genetic potential (or genotype) of the oocyte; rather, it may be significantly influenced by the environment that the oocyte is exposed to during early culture. This observation supports those of Munne et. al., (29) who found that certain culture conditions or hormonal stimulation may induce chromosomal abnormalities and partly explain differences in pregnancy rates among in vitro fertilization centers. The embryo score parameter is an important tool that combines both the speed of embryo cleavage (cell number) and how well the embryo manages to divide (morphology score). The trial produced embryos with more rapid cleavage and a better average morphology score than their sibling oocytes cultured in control. The significance of the cell number was reduced when analyzed by insemination method, which contrasts with the findings of Dumoulin et al. (30). However, their average cell number per embryo of 3.22 for IVF and 3.48 for ICSI were well below the 3.98 and 4.15 that we found at similar developmental stages. A previous study (11) reported an increase in the number of trophectoderm cells in blastocysts when glutamine was added to the culture media, but not the increase in total blastomere numbers observed by day 2 of development. Similarly, when only glucose and phosphate are altered, the average number of blastomeres reported on day 2 was only 3.2 (10), well below that rates that we report with multiple alteration to culture media. We cannot identify which changes to formulation in the trial were primarily responsible for each improvement in outcomes observed; however, it is probable that a combination of alterations has contributed to the results obtained. The improvement in embryo morphology scores and the reduction in uncontrolled division in the embryos generated in the trial indicates that culture may affect the internal processes of the oocyte during fertilization, syngamy, and early cleavage and influence the first cleavage alignment from the spindle poles. Previously, pronuclear and embryo grading systems have shown that embryos with a high development and pregnancy potential display a more ordered, predictable cleavage pattern and higher pregnancy rates with a lower incidence of aneuploidy (24, 25, 31 33). The incidence of intrablastomere pitting seen in the embryos cultured in control and absent from those 1258 Cooke et al. Embryonic development in sequential media Vol. 78, No. 6, December 2002

6 FIGURE 1 Four-cell embryos showing absence of intra-blastomere pitting when cultured in control or trial. (a), An embryo with grade 4 morphology but extensive intracytoplasmic pitting cultured in control media. (b), An embryo from the same patient, cultured in trial, having grade 4 morphology and no fragmentation or pitting. (c), An embryo of grade 3 morphology cultured in control media. (d), Embryo of grade 3 morphology from the same patient, cultured in trial. The intrablastomere pitting within a patient group occurs independently of embryo fragmentation. Bar 50 m. Cooke. Embryonic development in sequential media. Fertil Steril cultured in trial from the same patient (Fig. 1) are attributed to reduced EDTA concentration in trial. This finding confirms previously unpublished results (Quinn P). All embryos were subject to the same culture temperatures, and the media differed only in composition and not in storage or antibiotic content. This pitting of cultured human embryos has been previously observed (34) but is not linked to a specific ingredient of culture. Embryo morphology grading protocols are largely based on the volume of embryo fragmentation (22 25). Pitting is not a parameter that would be excluded or selected for by these grading protocols, since it occurs independently of fragmentation. Pitted blastomeres were rarely observed in embryos cultured in the trial. Our study shows that all aspects of early human preimplantation embryo development can be substantially improved by using chemically defined sequential stage-specific culture rather than standard HTF-based formula- FERTILITY & STERILITY 1259

7 tions. This includes a significantly higher fertilization rate, cleavage rate, average embryo morphology score, embryo score parameter, and embryo utilization rate and significantly less uncontrolled division, regardless of whether embryos were generated by IVF or ICSI. The higher rates of embryo utilization and the overall pregnancy and implantation rates obtained when fertilization and embryo development occurred in the trial have prompted us to use this formulation exclusively. References 1. Quinn P, Kerin JF, Warnes GM. Improved pregnancy rate in human in vitro fertilization with the use of a based on the composition of human tubal fluid. Fertil Steril 1985;44: Gardner DK, Lane M. Culture and selection of viable blastocysts: a feasible proposition for human IVF? Hum Reprod Update 1997;3: Quinn P. Enhanced results in mouse and human embryo culture using a modified human tubal fluid lacking glucose and phosphate. J Assist Reprod Genet 1985;12: Seshagiri PB, Bavister BD. Glucose and phosphate inhibit respiration and oxidative metabolism in cultured hamster eight-cell embryos: evidence for the Crabtree effect. Mol Reprod Dev 1991;30: Lane M, Boatman DE, Albrecht ERM, Bavister BD. Intracellular divalent cation homeostasis and developmental competence in the hamster preimplantation embryo. Mol Reprod Dev 1998;50: Lane M, Bavister BD. Calcium homeostasis in early hamster preimplantation embryos. Biol Reprod 1998;59: Pool TB, Atiee SH, Martin JE. Oocyte and embryo culture: basic concepts and recent advances. In: May J, (ed). Infertility and reproductive medicine clinics of North America. Philadelphia: Saunders, Dumoulin JCM, Evers JLH, Bras M, Pieters MHEC, Geraedts JPM. Positive effect of taurine on preimplantation development of mouse embryos in vitro. J Reprod Fert 1992;94: Brinster RL, Thomson JL. Development of eight-cell mouse embryos in vitro. Expl Cell Res 1966;42: Carillo VJ, Lane B, Pridman DD, Risch PP, Pool TB, Silverman IH, et al. Improved clinical outcomes for in vitro fertilisation with delay of embryo transfer from 48 to 72 hours after oocyte retrieval: use of glucose and phosphate free media. Fertil Steril 1998;69: Devreker F, Winston RML, Hardy K. Glutamine improves human preimplantation development in vitro. Fertil Steril 1998;69: Lane M, Gardner DK. Increase in postimplantation development of cultured mouse embryos by amino acids and induction of fetal retardation and exencephaly by ammonium ions. J Reprod Fertil 1994;102: Bavister B. Culture of preimplantation embryos: facts and artifacts. Hum Reprod Update 1985;1: Bowman P, McLaren A. Cleavage rate of mouse embryos in vivo and in vitro. J Embryol Exp Morphol 1970;54: Jung T, Fischer B, Beier HM. Quantitative aspects of protein synthesis in non cultured and cultured rabbit blastocysts. Hum Reprod 1987;2: Gardner DK, Leese HJ. Concentrations of nutrients in mouse oviduct fluid and their effects on embryo development and metabolism in vitro. J Reprod Fertil 1990;88: Gott AL, Hardy K, Winston RML, Leese HJ. The nutrition and environment of the early human embryos. Proc Nutr Sci 1990;49:Abstract 2A. 18. Leese HJ, Conaghan J, Martin KL, Hardy K. Early human embryo metabolism. Bioessays 1993;15: Gardner DK, Schoolcraft WB, Wagley L, Schlenker T, Stevens J, Hesla J. A prospective randomized trial of blastocyst culture and transfer in in-vitro fertilisation. Hum Reprod 1998;13: Van Langendonckt A, Demylle D, Wyns C, Nisolle M, Donnez J. Comparison of G1.2/G2.2 and Sydney IVF cleavage/blastocyst sequential media for the culture of human embryos: a prospective, randomized, comparative study. Fertil Steril 2001;76: Van Steirteghem AC, Liu J, Joris H, Nagy Z, Janssenswillen C, Tournaye H, et al. Higher success rate by intracytoplasmic sperm injection than by subzonal insemination. Report of a second series of 300 consecutive treatment cycles. Hum Reprod 1993;8: Almeida PA, Bolton VN. The relationship between chromosomal abnormality in the human preimplantation embryo and development in vitro. Reprod Fertil Dev 1996;8: Giorgetti C, Terriou P, Auquier P, Hans E, Spach JL, Salzmann J, et al. Embryo score to predict implantation after in-vitro fertilisation: based on 957 single embryo transfers. Hum Reprod 1995;10: Pellestor F, Dufour MC, Arnal F, Humeau C. Direct assessment of the rate of chromosomal abnormalities in grade IV human embryos produced by in-vitro fertilization procedure. Hum Reprod 1994;9: Erenus M, Zouves C, Rajamahendran P, Leung S, Fluker M, Gomel V. The effect of embryo quality on subsequent pregnancy rates after in vitro fertilization. Fertil Steril 1991;56: Steer CV, Mills CL, Tan SL, Campbell S, Edwards RG. 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. Hum Reprod 1992;7: Hsu MI, Mayer J, Aronshon M, Lanzendorf S, Muasher S, Kolm P, et al. Embryo implantation in in vitro fertilization and intracytoplasmic sperm injection: impact of cleavage status, morphology grade, and number of embryos transferred. Fertil Steril 1999;72: Frattarelli JL, Leondires MP, Miller BT, Segars JH. Intracytoplasmic sperm injection increases embryo fragmentation without affecting clinical outcome. J Assist Reprod Genet 2000;17: Munne S, Magli C, Adler A, Wright G, de Boer K, Mortimer D, et al. Treatment-related chromosome abnormalities in human embryos. Hum Reprod 1997;12: Dumoulin JC, Coonen E, Bras M, van Wissen LC, Ignoul-Vanvuchelen R, Bergers-Jansen JM, et al. Comparison of in-vitro development of embryos originating from either conventional in-vitro fertilization or intracytoplasmic sperm injection. Hum Reprod 2000;15: Tesarik J, Greco E. The probability of abnormal preimplantation development can be predicted by a single static observation on pronuclear stage morphology. Hum Reprod 1999;14: Tesarik J, Junca AM, Hazout A, Aubriot FX, Nathan C, Cohen-Bacrie P, et al. Embryos with high implantation potential after intracytoplasmic sperm injection can be recognized by a simple, non-invasive examination of pronuclear morphology. Hum Reprod 2000;15: Scott LA, Smith S. The successful use of pronuclear embryo transfers the day following oocyte retrieval. Hum Reprod 1998;13: Desai NN, Goldstein J, Rowland DY, Goldfarb JM. Morphological evaluation of human embryos and derivation of an embryo quality scoring system specific for day 3 embryos: a preliminary study. Hum Reprod 2000;15: Cooke et al. Embryonic development in sequential media Vol. 78, No. 6, December 2002