Extended embryo culture in human assisted reproduction treatments

Transcription

1 Human Reproduction Vol.16, No.5 pp , 2001 Extended embryo culture in human assisted reproduction treatments M.T.Langley 1,3, D.M.Marek 1, D.K.Gardner 2, K.M.Doody 1 and K.J.Doody 1 1 Center for Assisted Reproduction, Bedford, Texas, and 2 Colorado Center for Reproductive Medicine, Englewood, Colorado, 80110, USA 3 To whom correspondence should be addressed. martinl@embryo.net In order to evaluate the niche of extended embryo culture in an IVF programme, retrospective analysis of nonselected IVF patients, who underwent ovarian stimulation from April 1998 to June 1999 in a single private practice assisted reproductive technology centre, was performed. Embryos were cultured for 48 h in S1/G1.2 medium followed by 48 to 72 h of culture in S2/G2.2 to day 5 or day 6. Only fertilized oocytes exhibiting two pronuclei from donor and non-donor IVF and intracytoplasmic sperm injection (ICSI) cases were examined to determine the relationship between embryo cell number on day 3 and subsequent rate of blastocyst formation. Results indicated that a proportional relationship existed between the number of blastomeres present in day 3 embryos and the rate of blastocyst formation. Fifty-four per cent of embryos that had six cells on day 3 formed blastocysts, while 76% of those embryos with eight cells formed blastocysts. Blastocyst development did not increase further when embryos had more than eight cells on day 3, indicating that embryos with greater cell numbers on day 3 are not always predictive of a greater likelihood of blastocyst formation. Fertilized oocytes exhibiting two pronuclei from donors produced significantly more blastocysts (67%) than those from IVF patients (52%; P < 0.01), and had a significantly higher implantation rate (54%) compared with IVF patients (30%; P < 0.01). Furthermore, blastocyst cryopreservation resulted in significantly higher implantation rates than cryopreserved cleavage stage embryos (P < 0.001). Key words: embryo culture/embryo development/implantation/pregnancy/viability Introduction Sequential culture media have been used to achieve successful embryo culture beyond the first 72 h in patients with a good response to gonadotrophins (Gardner et al., 1998a,b; Schoolcraft et al., 1999). This has enabled the selection of viable embryos, thereby reducing the necessity to transfer higher numbers of embryos to achieve a pregnancy, ultimately resulting in a decline in multiple gestations. However, there are concerns that extended embryo culture for non-selected patients may result in reduced embryo quality, increased cleavage arrest, an increase in cancelled transfers, and ultimately a decrease in ongoing pregnancy rates due to the availability of fewer embryos. It is widely held that embryonic cleavage rate during the first 72 h following insemination is an important indicator of embryo quality and suggests that embryos with a higher cleavage rate have an increased implantation rate (Shoukir et al., 1998). Therefore, many programmes attempting blastocyst transfer elect to culture embryos with the highest cell number beyond day 3 (Milki et al., 1999). Conversely, when faced with either a low number of fertilized oocytes or few rapidly-dividing embryos the decision is made to transfer at the cleavage stage on day 3. In our programme prior to April 1998, cleavage stage embryo transfers were routinely performed on day 3. After March 1998, routine blastocyst transfers were completed on day 5 or day 6. The change from day 3 to day 5 transfers included all patients irrespective of age, aetiology, number of follicles, and the number of fertilized oocytes or number of cleaving embryos. Therefore, it was possible to determine the relationship between blastocyst formation and embryo cell stage on day 3, irrespective of age. In addition, the viability of blastocysts derived from embryos of different cell number on day 3 was determined. Further, the application of extended culture to poor prognosis patients with slow cleaving embryos and/or low zygote number was evaluated. The results of extended culture and blastocyst transfer in oocyte donor cases were considered and finally the ability to cryopreserve blastocysts analysed. Materials and methods A retrospective analysis of IVF cycles from April 1998 to June 1999 was performed on patients at the Center for Assisted Reproduction, Bedford, Texas, USA. Donor and non-donor patients underwent ovarian stimulation with gonadotrophin-releasing hormone (GnRH) 902 European Society of Human Reproduction and Embryology

2 Blastocyst transfer analogue and pure FSH (Follistim ; Organon, West Orange, NJ, USA; Gonal-F ; Serono, Randolph, MA, USA) using either conventional down-regulation (Ziegler et al., 1987) or flare (Garcia et al., 1990) protocols. Upon confirmation of at least two dominant follicles (mean diameter 17 mm), human chorionic gonadotrophin (HCG) was administered with oocyte retrieval occurring h later. Criteria for donor selection included women who were healthy nonsmokers aged years. Demonstration of prior fertility was not a requirement in our donor programme. Oocyte donors were primarily anonymous to the recipient couple. Further screening included an extensive medical and social history on the donor and their family followed by infectious disease screening including hepatitis, human immunodeficiency virus and syphilis. Oocytes from donors were normally distributed to only one recipient at our programme. This analysis evaluates oocytes from 67 donors distributed between 68 recipients, one donor was split between two recipients. Following retrieval, oocytes were placed into human tubal fluid (HTF; Irvine Scientific, Santa Ana, CA, USA) 12% synthetic serum substitute (SSS; Irvine Scientific) and cultured for ~4 h. Embryo culture was performed in a humidified incubator with 5% CO 2 and 95% air at 37 C. For traditional IVF, a short 2 h coincubation of spermatozoa ( /ml) with oocytes was initiated 38 to 42 h post HCG (Gianaroli et al., 1996). After shortened coincubation with spermatozoa, oocytes were rinsed through fresh media and moved to 100 µl micro-drops of S1/G1.2 media under oil (Squibb, Princeton, NJ, USA). Sperm injection was performed on metaphase II oocytes at h post HCG administration following hyaluronidase treatment using previously described techniques (Van Steirteghem et al., 1992). Fertilization was evaluated h after IVF or intracytoplasmic sperm injection (ICSI). Beginning in April 1998, extended embryo culture to day 5 in serum-free sequential media was performed in all patients undergoing IVF. From April 1998 to February 1999 embryos were cultured for 48 h in S1 (IVF Science) medium followed by 48 to 72 h culture in S2 (IVF Science, Vero Beach, FL, USA) to day 5 or day 6. In February 1999 a media change was made and G1.2 (IVF Science) and G2.2 (IVF Science) replaced S1 and S2 respectively. Fertilized oocytes were cultured in Petri dishes containing 100 µl micro-drops of S1/G1.2 medium under oil, two embryos per drop, for 48 h. On day 3, cleavage stage embryos were transferred to fresh 100 µl microdrops of S2/G2.2 medium under oil and cultured, two embryos per drop, for an additional 48 to 72 h to day 5 or day 6. Blastocysts remaining after transfer were frozen using modified protocols (Ménézo et al., 1992). These modifications included cryopreservation of blastocyst embryos in glycerol using a two-step freeze and thaw technique. Only good or excellent quality blastocyst embryos were frozen. Blastocyst embryos exhibiting an expanded or expanding blastocele cavity with a clearly defined inner cell mass and a trophectoderm consisting of many cells forming a cohesive epithelium were frozen. According to a previously published scoring system (Gardner and Schoocraft, 1999) all cryopreserved blastocysts were graded as 3BB or better. Blastocyst embryos for freezing were exposed to 5% glycerol (Sigma) solution of phosphate buffered saline (PBS; Irvine Scientific) and human serum albumin (HSA; Irvine Scientific) for 10 min then transferred to 10% glycerol containing 0.2 mol/l sucrose (Sigma, St Louis, MO, USA) for 2 min. Blastocysts were loaded, one embryo to each 1.2 ml Nunc vial (Nalge Nunc International) containing 300 µl 10% glycerol/0.2 mol/l sucrose solution and placed in a Bio-cool III methanol bath freezer (FTS Systems, Stone Ridge, NY, USA). Temperature was reduced at a controlled rate dropping 2.0 C/min from 20 C to 7.0 C. Temperature was held at 7.0 C for 20 min for seeding. Following seeding, methanol temperature was further reduced at a rate of 0.3 C/min to 37 C. Blastocysts were then plunged into liquid nitrogen for storage. Cryopreserved blastocysts were thawed by removing each vial from the liquid nitrogen. Thawing immediately ensued by placing the vial on a countertop for s, only long enough to break the seal of the vial. After the seal was broken, each vial was thawed for 3 min in a 30 C water bath. Blastocysts were recovered from each vial and placed in 10% glycerol/0.2 mol/l sucrose solution for s and then moved to 5% glycerol/0.2 mol/l sucrose solution for 3 min. Blastocysts were then transferred to 0.2 mol/l sucrose for 2 min and then moved to PBS/HSA solution for 1 min. Thawed blastocysts were then transferred to S2/G2.2 culture media and analysed for survival, embryo stage and quality. Prior to embryo transfer, ~2 4 h after thawing, frozen embryos were again evaluated for survival, embryo stage and quality. An abdominal ultrasound (5 MHz) was utilized for all transfers to assist intrauterine placement of the embryo transfer catheter (Edwards- Wallace catheter; Marlow Technologies, Willoughby, OH, USA) (Gardner et al., 1998b). Patients were advised to transfer two or three blastocysts depending only on patient age and/or embryo quality. Donor oocyte recipients were advised to transfer no more than two blastocyst stage embryos. Criteria for the number of frozen blastocysts transferred were the same for fresh embryo transfer. Intramuscular luteal support with 50 mg progesterone (PCCA, Houston, TX, USA) in oil was initiated following retrieval. Ongoing pregnancies were confirmed with the use of an ultrasound to detect the presence of an intrauterine sac with fetal cardiac activity days after oocyte retrieval. Statistical analysis Statistical analysis was performed with Fisher s exact test or χ 2 to determine significant differences between day 3 versus day 5/6 ongoing pregnancy rates, implantation rates for frozen cycles. Statistical analysis was performed with Fisher s exact test or χ 2 to determine significant differences between embryo transfer results for day 5/6 age specific and donor group pregnancy rates, ongoing pregnancy rates, and implantation rates for fresh cycles. Analysis of variance followed by the Bonferroni procedure for multiple comparisons was performed on fresh cycles including non-donor and donor cycles to compare numbers of oocytes, pronucleate embryos and embryos transferred and cryopreserved. Independent t-tests were used to compare the mean number of frozen embryos transferred on day 3 versus day 5/6. Linear regression test was performed to determine the relationship between maternal age and blastocyst development. Values are mean SEM. Results Of 3575 non-donor pronucleate embryos from IVF patients undergoing extended culture, 1860 (52.0%) reached the blastocyst stage on day 5 or day 6 (Table I). On day 5, 42% of fertilized oocytes exhibiting two pronuclei reached the blastocyst stage, while 10.0% reached the blastocyst stage on day 6. Of 764 donor oocytes exhibiting two pronuclei, 513 (67.1%) reached the blastocyst stage on day 5 or day 6 (Table II). Of this 67.1%, 57.3% of oocytes exhibiting two pronuclei reached the blastocyst stage on day 5 while 9.8% reached the blastocyst stage on day 6. There was no significant difference between the performance of the S and G media, although there was a trend to increasing blastocyst formation with the G media (54.0% blastulation rate with G media versus 50.4% blastulation rate with S media). No differences other than rates of 903

3 M.T.Langley et al. Table I. Blastocyst development from day 3 cleavage stage embryos. Excludes all donors Age (years) Day 3 cell stage 2-cell 3-cell 4-cell 5-cell 6-cell 7-cell 8-cell 9-cell 10-cell All cells 35 No. embryos Blastocysts day 5 (%) 4 (5.5) 4 (4.3) 86 (22.1) 81 (29.7) 163 (45.3) 171 (54.1) 375 (68.8) 37 (60.7) 54 (64.3) 975 (44.4) Blastocysts day 6 (%) 1 (1.4) 11 (11.7) 54 (13.9) 29 (10.6) 46 (12.8) 36 (11.4) 49 (9.0) 10 (16.4) 3 (3.6) 239 (10.9) Total blastocysts (%) 5 (6.9) 15 (16.0) 140 (36.0) 110 (40.3) 209 (58.1) 207 (65.5) 424 (77.8) 47 (77.1) 57 (67.9) 1214* (55.3) No. embryos Blastocysts day 5 (%) 1 (2.5) 1 (1.7) 31 (12.6) 46 (30.3) 67 (38.3) 88 (53.7) 178 (65.0) 14 (77.8) 20 (71.4) 446 (38.6) Blastocysts day 6 (%) 0 (0) 2 (3.4) 32 (13.0) 10 (6.6) 19 (10.9) 14 (8.5) 22 (8.0) 0 (0) 2 (7.1) 101 (8.7) Total blastocysts (%) 1 (2.5) 3 (5.1) 63 (25.5) 56 (36.8) 86 (49.1) 102 (62.2) 200 (73.0) 14 (77.8) 22 (78.6) 547* (47.3) 39 No. embryos Blastocysts day 5 (%) 0 (0) 0 (0) 7 (14.3) 5 (20.8) 11 (31.4) 13 (54.2) 41 (66.1) 3 (100 2 ( (36.8 Blastocysts day 6 (%) 0 (0) 0 (0) 7 (14.3) 3 (12.5) 2 (5.7) 3 (12.5) 1 (1.6) 0 (0) 1 (16.7) 17 (7.6) Total blastocysts (%) 0 (0) 0 (0) 14 (28.6) 8 (33.3) 13 (37.1) 16 (66.7) 42 (67.7) 3 (100) 3 (50.0) 99* (44.4) All No. embryos ages Blastocysts day 5 (%) 5 (4.1) 5 (3.1) 124 (18.1) 132 (29.4) 241 (42.3) 272 (54.0) 594 (67.4) 54 (65.9) 76 (64.4) 1503 (42.0) Blastocysts day 6 (%) 1 (0.8) 13 (7.9) 93 (13.6) 42 (9.4) 67 (11.8) 53 (10.5) 72 (8.2) 10 (12.2) 6 (5.1) 357 (10.0) Total blastocysts (%) 6 (4.9) 18 (11.0) 217 (31.7) 174 (38.8) 308 (54.1) 325 (64.5) 666 (75.6) 64 (78.1) 82 (69.5) 1860 (52.0) *Significant difference of total blastocyst formation from all cells between age categories: P Table II. Blastocyst development from day 3 cleavage stage embryos. Includes only donors Day 3 cell stage 2-cell 3-cell 4-cell 5-cell 6-cell 7-cell 8-cell 9-cell 10-cell All cells No. embryos Blastocysts day 5 (%) 1 (7.1) 3 (10.7) 21 (25.3) 29 (29.3) 66 (51.2) 83 (65.4) 191 (83.4) 13 (72.2) 31 (83.8) 438 (57.3) Blastocysts day 6 (%) 0 (0) 1 (3.6) 8 (9.6) 15 (15.2) 16 (12.4) 17 (13.4) 14 (6.1) 2 (11.1) 2 (5.4) 75 (9.8) Total blastocysts (%) 1 (7.1) 4 (14.3) 29 (34.9) 44 (44.4) 82 (63.6) 100 (78.7) 205 (89.5) 15 (83.3) 33 (89.2) 513 (67.1) 904

4 Blastocyst transfer Table III. Extended embryo culture in poor prognosis patients. Table includes only patients with slow cleaving embryos containing less than eight cells on day 3. Excludes all donor cycles. Implantation calculation based on presence of fetal sac and cardiac activity eight 2PN eight 2PN Mean patient age ( SEM) Number retrievals Number embryo transfers Pregnancies ( HCG)/ongoing 15/11 44/31 Biochemical/miscarriage/ectopic 3/1/0 6/6/1 Pregnancy/ongoing rate per transfer (%) 51.7/ /30.7 Total embryos/mean transferred 73/ /2.29 Total ova/average ova per patient 469/15.6* 987/8.7 2PN/fertilization per ova (%) 286/61.0* 436/44.2 Number 2PN/mean number 2PN 286/ /3.86 Number blastocysts/per 2PN (%) 111/38.8* 115/26.4 Number clinical sacs (1/2/3) a 7/3/1 21/8/2 Implantation rate (%) Figure 1. Relationship between day 3 cell number and blastocyst development. The closed circles represent IVF patients, the open circles the oocyte donors. There is a significant linear trend between cell number and blastocyst development from the 2- to 8-cell stage for both groups of patients (linear regression; r 0.96, P 0.01). *P versus 8 2PN. a The number of clinical sacs is reflective of delivered pregnancies. PN pronuclear; HCG human chorionic gonadotrophin. blastocyst formation were determined between S and G media. A proportional relationship existed between number of blastomeres present on day 3 and rates of blastocyst formation in both IVF patients and oocyte donors (Table I; Figure 1). As day 3 blastomere number increased a significant increase in blastocyst formation for both IVF and donor patients was observed (Figure 1, P 0.01). Fifty-four per cent of embryos from IVF patients that had six cells on day 3 formed blastocysts, while 76% of those embryos with eight cells formed blastocysts (P 0.01). In donor cycles, 67% of embryos that had six cells on day 3 formed blastocysts, while 90% of those embryos with eight cells formed blastocysts (P 0.01). Eighty per cent of donor embryos containing six to eight cells on day 3 developed into blastocysts by day 5/6. Blastocyst development did not increase further when embryos had more than eight cells on day 3 (Table II, Figure 1). Non-donor patients undergoing IVF with slow cleaving embryos containing less than eight cells on day 3 and/or low zygote numbers were evaluated (Table III). An implantation rate of 18.6% and a pregnancy rate of 30.7% was obtained with extended culture from poor prognosis patients with slowly cleaving embryos (less than eight cells on day 3) and when less than eight two-pronucleate embryos were present. However, rates of blastocyst formation for patients with fewer than eight zygotes (26.4%) are significantly lower than for patients with eight or more zygotes (38.8%; P 0.01). These results may be attributed to poor stimulus response and/or lowered fertilization. The average number of oocytes retrieved per patient with fewer than eight zygotes was 8.7 while the average number of oocytes retrieved per patient with eight or more zygotes was 15.6 (P 0.01). In addition, fertilization rates between the two groups were significantly lower for patients Figure 2. Relationship between maternal age and percent blastocyst development. There is a significant linear decrease in blastocyst development with increasing maternal age (480 points, r , P 0.05). with fewer than eight zygotes (44.2%) than in patients with eight or more zygotes (61.0%; P 0.01). However, implantation rates between the two groups were similar. No difference was observed in patient age and aetiology of infertility between the two groups. There was a significant effect of maternal age on blastocyst development (Table I; Figure 2), with patients aged 35 years producing 55.3% blastocysts, but patients 39 years producing 44.4% blastocysts (P 0.05). It was possible in a limited number of cases to track the outcome of blastocyst transfer to cell stage on day 3 as follows: 1 to 2 cells, 3 to 4 cells, 5 to 6 cells, 7 to 8 cells, 9 to 10 cells. An implantation rate of 30% was attained in patients whose embryos on day 3 had at least five cells, compared with implantation rates of closer to 40% when embryos had 7 to 8 cells on day 3 (Table IV). Only one patient had an embryo at the 1- to 2-cell stage on day 3 transferred after extended culture, and no successful implantation resulted. Two patients, average age 29.4 years, with embryos at the 9- to 10-cell stage 905

5 M.T.Langley et al. Table IV. Transfer of blastocysts from embryos of known cell number on day 3. Includes only non-donor IVF and intracytoplasmic sperm injection (ICSI) cases Day 3 cell stage Weighted cell Patients (n) Patient age Ongoing Average no. Implantation embryo transfer stage (years) pregnancy transferred rate 3 4 4C 93.0% (17.4%) b,c % a,b,c 5 6 5C 60.0% (32.4%) % 7 8 8C 72.9% (49.3%) % a P 0.01 (3 4 versus 5 6). b P 0.01 (3 4 versus 7 8). c Significant trend between day 3 cell stage embryo transfer P Table V. Embryo transfer results for day 5/day 6 for age specific groups Age 35 Age Age 39 All ages Donors Mean patient age (years) Mean donor age (years) a No. of retrievals No. of transfers Total no. of oocytes b a Total no. of two pronuclear oocytes c a Mean no. of embryos transferred b d a Mean no. of embryos for cryopreservation b a No. of clinical sacs: (1/2/3/4) 80/54/9/0 53/16/9/0 8/3/0/0 141/73/18/0 23/23/1/0 Implantation rate (%) 35.6 b a Pregnancies (positive HCG) (%) 61.8 e f Ongoing pregnancy (%) 45.2 h g 58.2 a P 0.01 (donors versus 35). b P 0.01 ( 35 versus 35 39, 39). c P 0.05 ( 35 versus 35 39). d P 0.05 (35 39 versus 39). e P 0.01 ( 35 versus 39). f P 0.05 (donors versus 35). g P 0.05 (all ages versus donors). h Significant trend between age groups P on day 3 produced one ongoing pregnancy after blastocyst transfer and an implantation rate of 66.6%. While slight, a trend of decreasing average age was observed as day 3 cell stage increased. The number of good and poor prognosis patients were distributed equally between the groups examined. Fresh embryo transfer results were evaluated for three different age categories ( 35, 35 39, and 39 years), overall for day 5/6 and for oocyte donors (Table V). Oocyte donors were compared with the 35 years age group, the closest group in age to the donors. Oocyte donors had significantly fewer blastocysts transferred than patients 35 years (Table V; P 0.01) and had significantly more blastocysts cryopreserved (P 0.01). Implantation rates in the donor oocyte group were significantly higher than in IVF patients (P 0.01). Within the IVF patients there was a significant decrease in implantation rate with respect to increasing maternal age (P 0.01 for 35 versus 35 39, 39 years and P 0.05 for versus 39 years). Oocyte donor results were further analysed by comparing embryo transfer on day 3 versus embryo transfer on day 5/6 (Table VI). Although no significant increase in ongoing pregnancy rate was observed, transfer of fresh day 5/6 blastocysts compared with cleavage stage embryos was associated with a significant reduction in the mean number of embryos trans- 906 Table VI. Comparison of embryo transfer results for day 3 versus day 5/day 6 donor oocyte cycles (January 1997 to June 1999) Variable Transfer Day 3 Day 5 Mean patient age (years) Mean donor age (years) No. of retrievals No. of transfers Total no. of oocytes Total no. of two pronuclear oocytes Mean no. of embryo transferred * Mean no. of embryos for cryopreservation No. of clinical sacs: (1/2/3/4) 18/11/10/1 23/23/1/0 Implantation rate (%) ** Pregnancies (positive HCG) (%) Ongoing pregnancy (%) *P 0.01 (versus day 3 total). **P 0.05 (versus day 3 total). ferred (P 0.01) and a significant increase in embryo implantation (P 0.05). Most importantly, a significant reduction in the number of high order multiple gestations was observed between the two groups (P 0.01).

6 Blastocyst transfer Table VII. Comparison of frozen embryo transfer results for day 3 versus day 5 non-age specific (January 1997 to June 1999) Variable Transfer Day 3 Day 5 Mean patient age (years) No. of transfers Total no. of embryos transferred Mean no. of embryos transferred a No. of clinical sacs (1/2/3/4) 20/7/1/0 22/6/1/0 Implantation rate (%) a Pregnancies (positive HCG) per embryo a transfer (%) Ongoing pregnancy per embryo transfer b (%) a P 0.01 (versus day 3 total). b P 0.05 (versus day 3 total). The transfer of frozen blastocysts was compared with cleavage stage embryos frozen between January 1, 1997 and May 1, All cleavage stage embryos were thawed and transferred at the multi-cell stage. Blastocyst transfer resulted in a significant increase in the percentage of embryos resistant to thawing, 82.0% of all blastocysts thawed survived to be transferred while 68.5% of all multi-cell embryos thawed survived to be transferred (P 0.001). Blastocyst transfer after freezing resulted in a significant increase in embryo implantation rate, 21.9 versus 10.1% respectively (P 0.001). Thawed blastocyst transfers also resulted in a significant increase in the ongoing pregnancy rate, 33.3 compared with 18.5% for day 3 (P 0.05), and a significant decrease in mean number of embryos transferred, 2.35 compared with 3.07 for day 3 (P 0.01) (Table VII). Discussion Published results of improved pregnancy and implantation rates continue to promote extended blastocyst culture over cleavage stage embryo transfer (Marek et al., 1999). The use of commercially available sequential media allows the human assisted reproduction treatment laboratory the capability of routine extended embryo culture in vitro. Blastocyst culture permits selection of more viable embryos and synchronization of embryonic stage and uterus, culminating in higher implantation rates, thereby eliminating the need to transfer higher numbers of embryos, ultimately resulting in a reduction of high order multiple pregnancies (Gardner et al., 1998a,b). However, many clinics are concerned about the use of extended culture for all patients. Initial studies reported that development to the blastocyst stage was increased by the number of oocytes collected, the number of oocytes inseminated, the number of two-pronucleate zyogtes, and the number of embryos cleaving to at least the 8-cell stage by day 3 in culture (Jones et al., 1998; Gardner and Schoolcraft, 1999). For this reason, patients are often selected for blastocyst transfer based on the number of quality embryos available on day 3 (Milki et al., 1999). However, in a recent study by Marek et al. (1999), it has been determined that when blastocyst culture and transfer is applied to all patients, irrespective of age and cause of infertility, the number of patients who did not receive a transfer is 6.7%. Although this is higher than the 2.9% of cancelled transfers for patients having a day 3 transfer (Marek et al., 1999), it is not an unacceptable rate, when one considers that pregnancy rate was actually increased and the number of embryos transferred decreased when extended culture was employed. However, poor prognosis patients, those patients with less than eight two-pronucleate oocytes and fewer than eight cells on day 3 exhibited a higher rate of cancelled transfers (10.7%) when extended embryo culture was applied (Table III). An increase in cancelled transfers in this group can be attributed to a lower number of two-pronucleate embryos for patients not having an embryo transfer. Among those patients not receiving an embryo transfer an average of two normally fertilized oocytes and a zero percentage blastulation rate was observed. Poor prognosis patients meeting the same criteria examined between the dates of January 1997 and March 1998 (prior to initiation of extended embryo culture) also showed an increase in cancelled transfers of 8.6%. It is evident from this study that a proportional relationship exists between number of blastomeres present in day 3 embryos and blastocyst formation. Blastocyst production from day 3 cleavage stage embryos was linear with a correlation coefficient approaching 1.00 in both IVF patients and oocyte donors (r 0.96). It appears that blastocyst production from embryos containing at least eight cells begins to plateau around 80% for non-donor cases and 90% for donor cases. These data are in agreement with one previously published study (Shapiro et al., 2000) which found that the more developed embryos at 72 h were more likely to become blastocysts, but in disagreement with another (Racowsky et al., 2000). In the report by Racowsky et al. (2000) it was proposed that patients who did not have at least one 8-cell embryo on day 3 should not have blastocyst culture. The data presented here are in contrast with this proposal and suggest that, although higher blastulation rates were observed in embryos that had progressed to at least eight cells by 72 h post insemination, a significant number of embryos with initially retarded development were able to progress to the blastocyst stage by 120 to 144 h post insemination. Further, when evaluating the implantation and pregnancy rates over the same time period when blastocyst culture is applied to poor prognosis patients with slow cleaving embryos (less than eight cells) and/or low zygote number (less than eight 2PN) acceptable rates are obtained. Although Racowksy et al. (2000) used a sequential culture system similar to that of the current study, it was not identical. Racowsky et al. (2000) used a simple medium for the first 48 h of culture (IVF50), as opposed to one containing amino acids (G1.2), which may have contributed to their less than optimum results in their poor prognosis patient group (Gardner and Lane, 1997). In addition, it is plausible that the differences reported in blastocyst development between the programmes are due simply to other laboratory or clinic related factors. A decrease in the average number of embryos available for extended culture was observed with increasing maternal age. The average number of embryos available for extended culture was 7.9, 6.7, and 6.1 for patients in the age categories 35, 907

7 M.T.Langley et al , and 39 respectively (Table V). A significant linear trend of decreasing blastocyst development with increasing maternal age was observed. Such a decrease in a general population of IVF patients has previously been reported (Janny and Ménézo, 1996) using co-culture to support embryo development. Increasing maternal age was also associated with a decrease in implantation rate, a phenomenon previously reported (Schoolcraft et al., 1999). This is consistent with an increasing incidence of aneuploidies with advancing maternal age (Munne et al., 1995). Blastocyst development was not only significantly higher in the oocyte donor patients, but resulting implantation rates were also higher. As such these data indicate that oocytes from donors have a higher developmental potential both in culture and post transfer. This has been reported previously (Patton et al., 1999; Schoolcraft and Gardner, 2000) and confirms that blastocyst culture is of great benefit when using oocyte donation. This is of significance given the potential problems that older patients would face conceiving a high order multiple gestation. Importantly, it was shown that blastocysts cultured in commercially available sequential media could be cryopreserved and thawed successfully, resulting in implantation rates similar to those previously reported for co-cultured embryos (Kaufman et al., 1995). The ability to adequately cryopreserve human blastocysts has been proposed to be a weak component of a blastocyst transfer cycle. However, it is evident that this is not the case. With the advent of new vitrification procedures (Lane et al., 1999) it is envisaged that the success of blastocyst cryopreservation may be further increased. Furthermore, these data show that the transfer of frozen-thawed blastocysts gave rise to significant increases in embryo survival, implantation and pregnancy rates compared to frozen-thawed embryos transferred on day 3. In conclusion, this report demonstrates the ability of cleavage stage embryos of different cell number to develop to the blastocyst stage by day 5 or day 6, and highlights the difficulty in selecting embryos for transfer at the cleavage stage. Most importantly, these data suggest that blastocyst transfer can generate very acceptable pregnancy and implantation rates for poor prognosis patients and especially in good prognosis patients and donor oocyte IVF cycles. With the ability to identify the most viable blastocysts within a cohort (Gardner et al., 2000) it should be possible to perform single blastocyst transfers for many patients. Although the current analysis suggests many advantages of blastocyst transfer over multicell embryo transfer, prospective randomized clinical trials may further define the benefit and role of extended embryo culture in human assisted reproduction techniques. Acknowledgments The authors would like to thank Dr Michelle Lane of the Colorado Center for Reproductive Medicine for her valuable comments on the manuscript. References Garcia, J.E., Padilia, S.L., Bayati, J. and Baramki, T.A. (1990) Follicular phase gonadotropin-releasing hormone agonist and human gonadotropin: a better alternative for ovulation induction in in-vitro fertilization. Fertil. Steril., 57, Gardner, D.K. and Lane, M. (1997) Culture and selection of viable blastocysts: a feasible proposition for human IVF? Hum. Reprod. Update, 3, Gardner, D.K. and Schoolcraft, W.B. (1999) In-vitro culture of human blastocysts. In Jansen, R. and Mortimer, D. (eds), Towards Reproductive Certainty: Fertility and Genetics Beyond. Parthenon Press, Carnforth, pp Gardner, D.K., Vella, P., Lane, M. et al. (1998a) Culture and transfer of human blastocyst increases implantation rates and reduces the need for multiple embryo transfers. Fertil. Steril., 69, Gardner, D.K., Schoolcraft, W.B., Wagley, L. et al. (1998b) A prospective randomized trial of blastocyst culture and transfer in in-vitro fertilization. Hum. Reprod., 13, Gardner, D.K., Lane, M., Stevens, J. et al. (2000) Blastocyst score affects implantation and pregnancy outcome: towards a single blastocyst transfer. Fertil. Steril., 73, Gianaroli, L., Magli, M.C., Ferraretti, A.P. et al. (1996) Reducing the time of sperm-oocyte interaction in the human in-vitro fertilization improves the implantation rate. Hum. Reprod., 11, Janny, L. and Ménézo, Y.J.R. (1996) Maternal age effect on early human embryonic development and blastocyst formation. Molec. Reprod. Dev., 45, Jones, G.M., Trounson, A.O., Lolatgis, N. and Wood, C. (1998) Factors affecting the success of human blastocyst development and pregnancy following in vitro fertilization and embryo transfer. Fertil. Steril., 70, Kaufman, R.A., Ménézo, Y., Hazout, A. et al. (1995) Cocultured blastocyst cryopreservation: experience of more than 500 transfer cycles. Fertil. Steril., 64, Lane, M., Schoolcraft, W.B. and Gardner, D.K. (1999) Vitrification of mouse and human blastocysts using a novel cryoloop container-less technique. Fertil. Steril., 72, Marek, D., Langley, M., Gardner, D.K. et al. (1999) Introduction of blastocyst culture and transfer for all patients in an IVF program. Fertil. Steril., 72, Ménézo, Y., Nicollet, B., Herbaut, N. and Andre, D. (1992) Freezing cocultured human blastocysts. Fertil. Steril., 58, Milki, A.A., Fisch, J.D. and Behr, B. (1999) Two-blastocyst transfer has similar pregnancy rates and a decreased multiple gestation rate compared with three-blastocyst transfer. Fertil. Steril., 72, Munné, S., Alikani, M., Tomkin, G. et al. (1995) Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil. Steril., 64, Patton, P.E., Sadler-Fredd, K., Burry, K.A. et al. (1999) Development and integration of an extended embryo culture program. Fertil. Steril., 72, Racowsky, C., Jackson, K.J., Cekleniak, N.A. et al. (2000) The number of eight-cell embryos is a key determinant for selecting day 3 or day 5 transfer. Fertil. Steril., 73, Schoolcraft, W.B. and Gardner, D.K. (2000) Blastocyst culture and transfer increases the efficiency of oocyte donation. Fertil. Steril., 74, Schoolcraft, W.B., Gardner, D.K., Lane, M. et al. (1999) Blastocyst culture and transfer: Analysis of results and parameters affecting outcome in two IVF programs. Fertil. Steril., 72, Shapiro, B.S., Harris, D.C. and Richter, K.S. (2000) Predictive value of 72- hour blastomere cell number on blastocyst development and success of subsequent transfer based on the degree of blastocysts development. Fertil. Steril., 73, Shoukir, Y., Chardonnens, D., Campana, A. et al. (1998) The rate of development and time of transfer play different roles in influencing the viability of human blastocysts. Hum. Reprod., 13, Van Steirteghem, A.C., Liu, J., Joris, H. et al. (1992) Higher success rate by intracytoplasmic sperm injection than by subzonal insemination. A report of a second series of 300 consecutive treatment cycles. Hum. Reprod., 8, Ziegler, D., Lu, J., Cedars, M. et al. (1987) Suppression of the ovary using a gonadotropin releasing-hormone agonist prior to stimulation for oocyte retrieval. Fertil. Steril., 48, Received on September 29, 2000; accepted on January 18,