- Email: [email protected]
Cryopreservation of zygotes and early cleaved human embryos
Vol. 49, No.2, February 1988 Printed in U.S.A.
FERTILITY AND STERILITY Copyright e
1988 The American Fertility Society
Cryopreservation of zygotes and early cleaved human embryos
Jacques Cohen, Ph.D.*H Gary W. DeVane, M.D.t Carlene W. Elsner, M.D.* Carole B. Fehilly, Ph.D.*t
Hilton I. Kort, M.D.* Joe B. Massey, M.D.* Thomas G. Turner Jr., M.S.t
Reproductive Biology Associates, Atlanta, University of Georgia, Athens, Georgia, and Sand Lake Hospital, Orlando, Florida
Zygotes and 2- to 5-cell human embryos were frozen in 1,2-propanediol and sucrose; results of the first 50 cycles (45 patients) are presented. A total of 41 zygotes (17 attempts at thawing) were thawed, resulting in six singleton clinical pregnancies (15% per embryo; 35% per cycle), of which three delivered, one aborted, and two are ongoing. Fifty-seven cleaved embryos were thawed in 33 other cycles, resulting in four singleton and one twin pregnancy (11% per embryo; 15% per cycle), of which four delivered and one is ongoing. Depending on the cell stage, 61% to 81% of embryos survived cryostorage, but 2-cell embryos did not implant. One fifth of cryoinjury was due to the formation of cracks in the zona pellucida. The incidence of implantation was not enhanced when more than one freeze/thawed embryo was replaced, most pregnancies being obtained from single embryo replacements. At least 8% more births are expected in addition to conventional in vitro fertilization methods when the current policy of replacing three fresh embryos and freezing the remainder using this technique is applied. This method will result in two to four times more pregnancies per spare embryo, compared with other cryopreservation methods using older embryos. Fertil Steril 49:283, 1988
Ferry Road, Suite 330, Atlanta, Georgia. Department of Physiology and Pharmacology, School of Veterinary Medicine, University of Georgia.
development or on morphologic features. Whether such a selection determines the outcome of in vitro fertilization (IVF) following replacement of fresh embryos has not yet been established. However, embryonic morphology may determine the outcome of the freezing process. 5 •6 Testart7 recently has shown that freeze/thawed embryos have the same chance of implantation as fresh embryos, and it was suggested that the number of embryos per replacement be reduced in order to decrease the high incidence of multiple pregnancies following IVF. So far, three methods have been developed for freezing human embryos. The first uses dimethylsulfoxide (DMSO) as cryoprotectant and is applied to the 4- to 8-cell stage. Results have been disappointing and the incidence of births, following a single egg collection, is only increased by 2%.5 •8 The second method uses glycerol as cryoprotectant, and can only be applied to expanded blasto-
Vol. 49, No. 2, February 1988
Cohen et al.
Preimplantation embryos now can be frozen and stored in liquid nitrogen from the zygote to the blastocyst stage with subsequent survival and viability.1-3 Pregnancies also have been reported following cryopreservation of ripe unfertilized oocytes.4 These reproductive options should be evaluated carefully with respect to safety and risks to the oocyte, embryo, and mother. A number of problems arise when the possibility of embryo freezing exists. Selection of embryos for cryopreservation or immediate replacement can be based on the speed of
Received July 7, 1987; revised and accepted October 7, 1987.
* Reproductive Biology Associates. t Sand Lake Hospital. :j: Reprint requests: Jacques Cohen, Ph.D., 993-D Johnson
Cryopreservation of zygotes and embryos
283
cysts. Despite a high incidence of implantation, net results are poor because of the low viability of embryos growing in vitro, between the 8-cell and the blastocyst. Only 3% more births can be expected when this method is used in conjunction with conventional IVF. 8 The efficiency of embryo freezing has increased considerably with the introduction of the third method, which uses 1,2-propanediol (PROH) and sucrose for cryoprotection, and can be applied to zygotes and very early embryos. 3•9 The initial goal of the present investigation was threefold: first, to determine whether results differ when embryos are frozen in 1,2-propanediol and sucrose on day 1 (zygote) or day 2 (3- to 5-cell) following egg collection; second, to investigate whether the incidence of implantation of fresh embryos would change when the morphologically "better" embryos would be selected for cryopreservation; third, to evaluate which of the above cryopreservation methods would produce the highest number of births, in terms of the availability of spare embryos and the age at which embryos are frozen. MATERIALS AND METHODS Patients
Couples with a long duration of infertility (range, 2 to 12 years) were treated with IVF for a variety of disorders, such as tubal disease (n = 24), endometriosis (n = 5), immune and/or cervical infertility (n = 4), male infertility (n = 9), and unexplained infertility (n = 3). The cutoff age of the female partner was 45 years. Patients were stimulated with a combination of clomiphene citrate (CC; Serophene, Serono, Rome, Italy) 100 mg daily from days 2 to 6 and follicle-stimulating hormone (FSH; Metrodin, Serono, Rome, Italy) 150 IU daily from days 1 to 5 and/ or human menopausal gonadotropin (hMG; Pergonal, Serono, Rome, Italy), usually with 150 IU daily from day 5. Patients were monitored from day 7 or 8 of their cycle and 75 or 150 IU of hMG was supplied on a day-to-day basis, depending on the number of follicles, follicular size, and serum estradiol (E 2). In more than 90% of these patients, ovulation was induced using 5000 IU of human chorionic gonadotropin (hCG; Profasi, Serono, Rome, Italy). The remainder of the patients had egg retrieval timed by an endogenous luteinizing hormone (LH) surge detected in urine, using radioimmunoassays (RIAs; LH MAlA Clone, Serono Diagnostics, Woking, England), and 5000 284
Cohen et al.
Cryopreservation of zygotes and embryos
IU hCG was given to these patients before egg collection. Menstruation in another group of patients was timed using 10 mg/day norethisterone (Norluton, Parke-Davis, Santurce, PR) for a duration of 10 to 31 days, beginning on days 15 to 18 of the menstrual cycle prior to the scheduled egg collection cycle. The last day of drug therapy was defined as day zero of the treatment cycle. 10 Follicular stimulation was achieved with similar methods as described previously. This group of patients always received hCG for ovulation induction. All patients received 3000 IU hCG on days 2 and 4, respectively, following egg collection for luteal phase support. Oocytes were collected 34 to 35 hours after hCG administration or 26 to 30 hours after onset of the LH surge. The collection was performed laparoscopically or via transvaginal ultrasound using a suction pump. Follicles were flushed repeatedly with warm culture medium and the follicular contents were checked immediately in an adjacent laboratory. Insemination was performed 3 to 12 hours after egg collection, depending on cumulus expansion. Zygote and Embryo Freezing
A maximum of three embryos were replaced 2 days after egg collection and the remainder was frozen. In a number of patients, two fresh embryos were replaced and the remainder was frozen. The policy was to freeze a minimum of two embryos per patient. Oocytes and embryos were cultured in bicarbonate-buffered Earle's medium containing 15% maternal or donor serum, kept in small droplets under paraffin oil, and maintained in 5% C0 2 , 5% 0 2 , and 90% N 2 Y The donor serum was obtained from young, fertile, healthy, normally cycling women. The medium used for gamete and embryo culture was used for cryopreservation. Petri dishes containing pronucleate or 2- to 5-cell embryos were left at room temperature for 10 minutes and exposed to serum-supplemented culture medium containing 1.5 M PROH for 10 to 15 minutes, and 1.5 M PROH with 0.1 M sucrose for 2 to 5 minutes. Shrinkage and swelling of embryos was observed and recorded on videotape using an inverted microscope. Embryos were pipetted into small glass ampules (0.3 to 0.5 ml) pulled from glass tubes (Durham Tubes, Samco, Cambridge, England, or Corning, New York, NY). The ampules were heat-sealed and cooled using either a Planer R204 (Planer, Sunbury-on-Thames, England) or FTS (FTS-Systems, Stone Ridge, NY) cell freezer. Fertility and Sterility
Embryos were cooled according to the following curve9 : -2°C/min to -6°C, hold for 15 minutes for induction of ice crystallization using cold forceps, further cooling at -0.3°C/min to -30°C and direct transfer into liquid nitrogen. Thawing and Replacement
Patients returned for their frozen embryos 2 to 10 months after storage. CC was given on patients' menstrual cycle days 2 to 6 (50 or 100 mg daily), or on days 5 to 9 in patients with irregular or prolonged cycles or when ovulatory problems were anticipated. The decision to give CC either starting on day 2 or day 5 was made randomly in order to investigate effects on implantation. The results of this selection will be evaluated at a later stage when more patients have been studied. The remainder of the patients were monitored during their natural cycle. Urine collections were obtained seven times daily at 2- to 3-hour intervals for LH determination. Serum E 2 was assayed two or three times during the follicular phase, and daily ultrasound was performed to confirm ovulation. In some of the patients who received CC and had high E2, ovulation was induced with 5000 IU hCG. Some patients did not demonstrate evidence of ovulation during the natural cycle. Their treatment was abandoned and monitoring was initiated during another cycle following CC induction. Embryos were thawed 1 or 2 days following ovulation according to their age at the time of freezing. The onset of the LH surge was defined when a urinary LH value was higher than 1.0 IU/hr, provided previous assays were consistently lower and at least three later values were higher than 1.0 IU/hr. When zygotes were frozen, the interval between ovulation and thawing was the same or 6 hours shorter than that in the retrieval cycle. Differences in timing were due to the time of the onset of the LH surge. Thawing was not performed between 1:00 and 6:00 A.M., but the interval was shortened by 6 hours. This has not affected results so far, but more data have to be collected before definite conclusions can be drawn. The interval between ovulation and thawing was 6 to 12 hours shorter than the embryonic age when 2- to 5-cell embryos were frozen. Cleaved and intact embryos normally were replaced within 3 hours after thawing; however, in some cases, further development was observed overnight. Thawed zygotes were left in culture for 20 to 28 hours until cleavage had occurred. Ampules containing the embryos were Vol. 49, No.2, February 1988
hand-stirred in a 31 °C water bath for 20 to 35 seconds until all ice crystals disappeared and their contents were emptied into Earle's medium containing 15% maternal serum, 1.5 M PROH, and 0.2 M sucrose. 2-11 PROH was removed in six steps, at room temperature, and the embryos were washed several times with culture medium and slowly warmed to 37°C. Statistics
Data were analyzed with the chi-square test in all instances unless the expected frequency was less than 5, in which case the Fisher's exact test was used. RESULTS Selection of Zygotes and Embryos for Cryopreservation
One hundred twelve IVF patients treated between January 1986 and April1987 had more than three embryos, and their spare zygotes or embryos were frozen (Table 1). The embryos were separated into two groups (A and B), based on an examination of their morphologic features under an inverted microscope. This study was performed in order to establish whether positive embryo selection on the basis of morphologic features for cryopreservation would adversely affect the incidence of implantation of fresh embryos. Group A cleaved embryos had homogenous-looking blastomeres and few or no extracellular fragments. Group B embryos had uneven or disorganized-looking blastomeres or were discolored. Many of these embryos had fragments. It was not possible to distinguish among differences in embryos from 51 patients, and their embryos were allocated randomly for cryopreservation or immediate transfer (Table 1). No differences were found in the incidence of implantation of 2-day-old fresh embryos when group Table 1 Selection of Embryos for Cryopreservation and the Effect on the Incidence of Clinical Pregnancies of Fresh Embryos in Patients Who Had Spare Embryos Frozen Incidence of clinical pregnancies of fresh embryos Stage of Positive selection Negative selection for freezing for freezing No selection embryos (group B) possible (group A) at freezing Zygotes 2- to 5-cell
Cohen et al.
11/47 (23%)
4/14 (29%)
7/28 (25%) 6/23 (26%)
Cryopreservation of zygotes and embryos
285
replacements had been unsuccessful, returned for their frozen embryos. Zygotes (n = 41) or cleaved embryos (n = 57) were thawed during 50 cycles. Several patients had a sufficient number of embryos stored to allow for more than one attempt. One twin and ten singleton pregnancies were established after 4 7 freeze/thawed embryo replacements. The incidence of clinical pregnancy per thawing attempt was 22%, but one patient with positive fetal heart beat miscarried. Seven patients delivered healthy infants, and three other pregnancies are now progressing normally beyond week 26. The expected incidence of birth per attempt is 20%, which is comparable to that of patients who had fresh embryos replaced (Table 2). Eleven babies are expected out of a group of 98 thawed (11%) and 69 replaced (16%) embryos. Fifteen of these patients still have 29 remaining embryos frozen, and it is therefore likely that the incidence of implantation will increase further for this group. Cleaved embryos were thawed in 33 cycles and zygotes during 17 other cycles: the total number of cycles was 50, but five patients had embryos thawed on two separate occasions (Table 3). All patients with zygotes had at least one intact embryo replaced after thawing, and 30 of 33 (90%) patients with cleaved embryos had a replacement. Relatively more pregnancies were established with freeze/thawed zygotes than with cleaved embryos (29% versus 18%), but the difference is not significant. A slight increase in pregnancy was observed when more than one embryo was thawed, but eight pregnancies were established when a single embryo was replaced. This compared favorably with the replacement of two or three freeze/thawed embryos (Table 3). Cleaved embryos usually were thawed early in the morning or afternoon, and re-
Table 2 Breakdown of Results When a Maximum of Two or Three Fresh Embryos Are Replaced and the Remainder Is Frozen (%)
Proportion of patients with fertilization Incidence of clinical pregnancy• Incidence of birth or ongoingb pregnancy Incidence of cryopreservation Incidence of clinical pregnancy following thawing" Incidence of birth or ongoingb pregnancy following thawing Current cumulative incidence of birth or ongoing pregnancy following IVF and cryopreservationc
148/167 45/167 36/167 70/167
(89) (27) (22) (42)
11/50 (22) 10/50 (20) 46/167 (28)
• Fetal heart beat. b Ongoing beyond week 14. c The final incidence of birth will be presumably higher since 30% of patients have remaining embryos frozen after one attempt.
A or B embryos were replaced, indicating that crude morphologic parameters have little prognostic value at that time of embryonic development. There are few morphologic differences between zygotes, and selection based on their morphologic features is not possible. Random selection of zygotes for cryopreservation did not affect the incidence of implantation after the remaining fresh zygotes cleaved and were replaced (Table 1). Implantation of Cryopreserved Zygotes and Embryos
One hundred sixty-seven patients were treated with IVF in 1986, and 89% had one or more embryos available for replacement. Seventy patients (42%) had more than three embryos; cryopreservation of zygotes and cleaved embryos was performed at random. Forty-five patients, whose fresh embryo Table 3
Incidence of Implantation and Numbers of Embryos Thawed and Replaced Incidence of clinical pregnancies per patientc No. frozen embryos Embryonic stage at freezing
Thawed"
Survived andb replaced
1 2 3 or more Total 1 2 3 Total
Zygote
2- to 5-cell
All stages
2/7 1/4 2/6 5/17 (29%) 3/10 2/6
1/9 4/19 1/5 6/33 (18%) 5/18 1111 0/1 6/30 (20%)
3/16 5/23 3/11 11/50 8/28 3/17 0/2 11/47
Oil 5/17 (29%)
• Include embryos that did not survive thawing. b Also includes embryos that partially survived the thawing process.
286
Cohen et al.
Cryopreservation of zygotes and embryos
c
(19) (22) (27) (22) (29) (18)
(0) (23)
No significant differences between groups.
Fertility and Sterility
Table 4 Survival and Implantation of Cryopreserved Zygotes and Cleaved Embryos Stage at freezing
Incidence of post-thaw survival (%)
(%)
3- to 5-cell 2-cell Zygotes
31/41 (76) 13/16 (81) 25/41 (61)
6/31 (19) 0/13 (0) 6/25 (24)
•P
=
Incidence of fetal heart beats
0.06 (Fisher's exact test).
placement was performed within 3 hours. In some cases, thawing was timed late in the evening because of the onset of the LH surge. In the latter cases, the embryos were cultured overnight and replaced in the morning for organizational purposes. Pregnancies were achieved in both groups of patients whose thawed embryos were cultured for a short time or overnight, and preliminary data did not reveal any differences. No differences in embryonic survival or implantation were found in this relatively small group of patients when the follicular stimulation was taken into account. Pregnancies were achieved regardless of whether the eggs were derived during cycles when hMG, FSH, or Norluton was supplied. Five pregnancies were obtained in the 24 patients with tubal infertility. The six other pregnancies were obtained in couples with different causes, and differences in incidence of implantation between these groups were not found. Survival and Cryoinjury
Survival of zygotes and cleaved embryos exceeded 61%, but gestational sacs were derived only from zygotes or cryopreserved 3- to 5-cell embryos. None of the frozen 2-cell embryos implanted after replacement (Table 4). Twenty of the 98 (20%) zygotes and embryos that were thawed had a crack in Table 5
the zona pellucida. The incidence of cracks was unrelated to the stage of the embryo at freezing (Table 5), but while none of the zygotes with cracked zonae pellucidae survived, three out of ten cleaved embryos with a cracked zona pellucida had intact blastomeres upon thawing. A small hole was noticed in one of these embryos, and this embryo was replaced immediately. The two other embryos were cultured for 2 more days for observation of further cleavage. Both embryos lost half of their blastomeres through a large gap in the zona pellucida, but compaction occurred and the embryos were replaced. None of these embryos implanted. The overall survival of all zygotes and cleaved embryos was 70%, but 85% of embryos with an intact zona pellucida survived cryostorage (Table 5). DISCUSSION
This study demonstrates that zygote and early cleaving embryo freezing using PROH and sucrose is an efficient and simple method for establishing a high incidence of implantation. One fifth of patients with "extra" embryos (i.e., embryos not replaced in the IVF treatment cycle) frozen and thawed had an ongoing pregnancy and this proportion may well increase, considering that 30% of these patients have more embryos frozen. An advantage of this technique is that at least 40% of the patients will have embryos for cryopreservation. It is therefore expected that one in three egg retrievals will lead to live birth when this type of cryopreservation is used in conjunction with conventional IVF. This cryopreservation method seems to be advantageous for other reasons as well. PROH and sucrose have not yet been shown to cause any chromosome damage in mammalian studies and a considerable number of mammalian and human births following cryostorage have been reported. 10 This method is less time-consuming
Observations on the Survival of Cryopreserved Zygotes and Cleaved Embryos Embryonic stage at freezing Zygote" (%)
Incidence of survival Incidence of complete survival Incidence of cracks in the zona pellucida Survival of embryos with cracked zonae pellucidae Survival when the zona pellucida is intact
25/41 25/41 10/41 0/10 25/31
3- to 5-cell"
(%)
(%)
(61) (61) (24) (0) (81)
44/57 32/57 10/57 3/10 41/47
Total
(77) (56) (18) (30) (87)
69/98 55/98 20/98 3/20 66178
(70) (56) (20) (15) (85)
• No significant differences between groups. Vol. 49, No.2, February 1988
Cohen et al.
Cryopreservation of zygotes and embryos
287
Table 6 Incidence of Embryo Cryopreservation and Implantation Following Thawing and Replacement When Embryos Are Frozen on Different Days Following Egg Retrieval Day following egg collection when cryopreservation was performed
Stage of the embryo at freezing Cryoprotective Proportion of patients whose spare embryos were frozen Proportion of egg retrievals leading to cryopreservation (I) Incidence of ongoing pregnancies or birth following thawing (II) Expected increased incidence of birth through cryopreservation based on one egg retrieval (I X 11/10) Source
1 and 2
3 and 4
5 and 6
Zygote or 2- to 5-cell PrOH 70172 (97%)
3- to 10-cell DMSO 79/162 (49%)
Expanded blastocyst Glycerol 96/270 (36%)
70/167 (42%)
79/378 (21%)
96/881 (11%)
10/50 (20%)
2/29 (7%)
7/27 (26%)
8% This article
than methods published previously, 1 •2 and the present results of survival and implantation corroborate with those of others. 3 •7 This type of cryopreservation compares favorably with freezing cleavage-stage embryos using DMSO or expanded blastocysts using glycerol. The latter method resulted in a high incidence of implantation, but few blastocysts were available because of the poor development rate in vitro beyond day 3. 6 The current results are compared in Table 6 with results presented previously. 6 •12 Although statistical comparisons are invalid, since the observations were performed in separate clinics, some tentative conclusions may be drawn from the data because of relative similarities in follicular stimulation, patient selection, and general laboratory methodologies between the two IVF clinics. Almost three times more patients have extra embryos frozen when freezing is performed 1 or 2 days, rather than 5 or 6 days, after egg collection. The incidence of implantation per patient who had embryos frozen is similar between the two groups. The efficiency of the present method is therefore larger and the incidence of births following one egg collection will increase by at least 8%. The conventional cryopreservation methods will increase results only marginally (Table 6). Several suggestions can be used in order to improve cryopreservation further. First, the number of fresh embryos for replacement could safely be reduced because frozen and fresh embryos have an equal chance of implantation. Increasing the number of oocytes through altering follicular stimulation, for instance with the use of LH-RH agonist, should be especially beneficial. The current results demonstrate that only one intact cryopreserved embryo should be replaced at a time; this is not in 288
Cohen et al.
Cryopreservation of zygotes and embryos
2% Fe hilly et a!. 8
3% Fe hilly et a!. 8
agreement with results following replacement of fresh embryos. Despite the tremendous stress put upon the embryo during cryopreservation, the capacity to implant and develop normally remains intact. This raises the question of whether results would increase if the morphologically "better" embryos were cryopreserved and the remainder replaced. It has been demonstrated previously that gross morphology of 3- to 6-day-old embryos determines their survival. 5 •6 Despite a number of observations supporting this concept, we have insufficient data available to extrapolate this usage to younger embryos. Conversely, positive selection on the basis of embryo morphology did not impair the incidence of implantation of fresh embryos, and it is now our policy to freeze only zygotes (in which morphologic differences are hard to determine) or homogenouslooking cleaved embryos. More attention may have to be paid to the synchrony between a freeze/thawed zygote or embryo and the uterus. A previous study indicated that the chance of pregnancy increased when freeze/thawed blastocysts were transferred 24 hours earlier following ovulation than their chronologie age indicated. 2 Asynchrony following in vitro culture will presumably be small in zygotes or 4-cell stage embryos, and timing the moment of thawing may require more accuracy than is the case with blastocysts. Results of cryopreservation also may improve when patients with anovulatory or irregular cycles are given small amounts of CC. Preliminary observations have revealed that the best results are obtained when CC is given 50 mg daily, days 5 to 9 of the patient's menstrual cycle, and hCG for ovulation induction should be avoided. Decreasing the incidence of cracks in the zona Fertility and Sterility
pellucida could be another factor that improves the results of cryopreservation. The present study indicates that 85% of embryos could survive, which would probably increase the incidence of implantation further. There are presumably numerous causes for cracks in the zona pellucida. Some patients may have oocytes with weaker zonae pellucidae, as was shown recentlyY The formation of cracks also could be circumvented by reducing the size or volume of the container used for storing the embryos. Cracks are rare when straws are used for cryopreservationY Fast induction of ice crystallization, achieved through cooling in a methanol bath rather than in air, also may reduce the incidence of cracks. Embryos seem to survive better when nuclei are visible, 3 •7 the only exception being the 2-cell embryo. Dramatic changes occur in embryonic metabolism at the 2-cell stage and failure to implant may be due to a similar mechanism described for 2-cell embryos of some mammalian species (the 2-cell block). Testart 7 demonstrated recently that cryopreserved embryos are more viable than fresh embryos because of uterine hostility during the treatment cycle. This result may be biased because of possible suboptimal stimulation protocol and cannot be extrapolated to the data presented here, since the incidence of implantation of fresh and frozen embryos was the same in our work. However, the incidence of ongoing clinical implantation per freeze/ thawed embryo was 11% (11/98) in our work and 9% (14/151) in the Paris clinic. In some groups of society, freezing of unfertilized eggs is ethically accepted, whereas embryo freezing is considered to be experimental. The latter technique often is regarded as hazardous because it involves manipulation of potentially new individuals. Whether these opinions are justified or not, the truth is that some IVF centers have abandoned embryo freezing programs for ethical reasons and are now investigating oocyte cryopreservation. This investigation is being done despite increasing evidence from work in laboratory animals that the metaphase II spindle is prone to cryoinjury. A second disadvantage of this development is that the technology of cryopreservation of the mammalian embryo is well established, whereas reliable methods for freezing unfertilized eggs are still being developed. It could be argued that the zygote is ethically different from the cleaving embryo, since the chromosomes of the male and female gametes are still grouped in separate nuclei: a poten-
tial new individual has therefore not been genetically established. Another advantage of freezing zygotes is of a more technical nature. It is easier to visualize pronuclei in zygotes than nuclei in blastomeres, and the cell cycle of blastomeres may be asynchronous, leading to partial damage during the freezing process.
Vol. 49, No. 2, February 1988
Cohen et al.
Acknowledgments. We thank Benjamin Brackett, Ph.D., D.V.M., William Graves, Ph.D., and Carol Keefer, Ph.D., for their collaboration. Ms. Twala Grant, Ms. Leigh Inge, and Ms. Donna Sammond are acknowledged for technical support.
REFERENCES 1. Trounson AO, Mohr L: Human pregnancy following cryopreservation, thawing and transfer of an eight-cell embryo. Nature 305:707, 1983 2. Cohen J, Simons RF, Edwards RG, Fehilly CB, Fishel SB: Pregnancies following the frozen storage of expanding human blastocysts. J In Vitro Fert Embryo Transfer 2:59, 1985 3. Testart J, Lassalle B, Belaisch-Allart J, Hazout A, Forman R, Rainhorn JD, Frydman R: High pregnancy rate after early human embryo freezing. Fertil Steril 46:268, 1986 4. Chen C: Pregnancy after human oocyte cryopreservation. Lancet 1:884, 1986 5. Freemann L, Trounson A, Kirby C: Cryopreservation of human embryos: progress on the clinical use of the technique in human in vitro fertilization. J In Vitro Fert Embryo Transfer 3:53, 1986 6. Cohen J, Simons RF, Fehilly CB, Edwards RG: Factors affecting survival and implantation of cryopreserved human embryos. J In Vitro Fert Embryo Transfer 3:46, 1986 7. Testart J: Evidence of uterine inadequacy to egg implantation in stimulated in vitro fertilization cycles. Fertil Steril 47:855, 1987 8. Fehilly CB, Cohen J, Simons RF, Fishel SB, Edwards RG: Cryopreservation of cleaving embryos and expanded blastocysts in the human: a comparative study. Fertil Steril 44:638, 1985 9. Lassalle B, Testart J, Renard JP: Human embryo features that influence the success of cryopreservation with the use of 1,2 propanediol. Fertil Steril 44:645, 1985 10. Frydman R, Forman R, Rainhorn JD, Belaisch-Allart J, Hazout A, Testart J: A new approach to follicular stimulation for in vitro fertilization: programed oocyte retrieval. Fertil Steril 46:657, 1986 11. Edwards RG, Fishel SB, Cohen J, Fehilly CB, Steptoe PC, Webster J: Factors influencing the success of in vitro fertilization for alleviating human infertility. J In Vitro Fert Embryo Transfer 1:3, 1984 12. Cohen J, Simons RF, Edwards RG, Fehilly CB, Fishel SB, Hewitt J, Steptoe PC, Webster J: Birth following the replacement of a cryopreserved hatching blastocyst. Lancet 1:647, 1985 13. Testart J: Unpublished data
Cryopreservation of zygotes and embryos
289
FERTILITY AND STERILITY Copyright e
1988 The American Fertility Society
Cryopreservation of zygotes and early cleaved human embryos
Jacques Cohen, Ph.D.*H Gary W. DeVane, M.D.t Carlene W. Elsner, M.D.* Carole B. Fehilly, Ph.D.*t
Hilton I. Kort, M.D.* Joe B. Massey, M.D.* Thomas G. Turner Jr., M.S.t
Reproductive Biology Associates, Atlanta, University of Georgia, Athens, Georgia, and Sand Lake Hospital, Orlando, Florida
Zygotes and 2- to 5-cell human embryos were frozen in 1,2-propanediol and sucrose; results of the first 50 cycles (45 patients) are presented. A total of 41 zygotes (17 attempts at thawing) were thawed, resulting in six singleton clinical pregnancies (15% per embryo; 35% per cycle), of which three delivered, one aborted, and two are ongoing. Fifty-seven cleaved embryos were thawed in 33 other cycles, resulting in four singleton and one twin pregnancy (11% per embryo; 15% per cycle), of which four delivered and one is ongoing. Depending on the cell stage, 61% to 81% of embryos survived cryostorage, but 2-cell embryos did not implant. One fifth of cryoinjury was due to the formation of cracks in the zona pellucida. The incidence of implantation was not enhanced when more than one freeze/thawed embryo was replaced, most pregnancies being obtained from single embryo replacements. At least 8% more births are expected in addition to conventional in vitro fertilization methods when the current policy of replacing three fresh embryos and freezing the remainder using this technique is applied. This method will result in two to four times more pregnancies per spare embryo, compared with other cryopreservation methods using older embryos. Fertil Steril 49:283, 1988
Ferry Road, Suite 330, Atlanta, Georgia. Department of Physiology and Pharmacology, School of Veterinary Medicine, University of Georgia.
development or on morphologic features. Whether such a selection determines the outcome of in vitro fertilization (IVF) following replacement of fresh embryos has not yet been established. However, embryonic morphology may determine the outcome of the freezing process. 5 •6 Testart7 recently has shown that freeze/thawed embryos have the same chance of implantation as fresh embryos, and it was suggested that the number of embryos per replacement be reduced in order to decrease the high incidence of multiple pregnancies following IVF. So far, three methods have been developed for freezing human embryos. The first uses dimethylsulfoxide (DMSO) as cryoprotectant and is applied to the 4- to 8-cell stage. Results have been disappointing and the incidence of births, following a single egg collection, is only increased by 2%.5 •8 The second method uses glycerol as cryoprotectant, and can only be applied to expanded blasto-
Vol. 49, No. 2, February 1988
Cohen et al.
Preimplantation embryos now can be frozen and stored in liquid nitrogen from the zygote to the blastocyst stage with subsequent survival and viability.1-3 Pregnancies also have been reported following cryopreservation of ripe unfertilized oocytes.4 These reproductive options should be evaluated carefully with respect to safety and risks to the oocyte, embryo, and mother. A number of problems arise when the possibility of embryo freezing exists. Selection of embryos for cryopreservation or immediate replacement can be based on the speed of
Received July 7, 1987; revised and accepted October 7, 1987.
* Reproductive Biology Associates. t Sand Lake Hospital. :j: Reprint requests: Jacques Cohen, Ph.D., 993-D Johnson
Cryopreservation of zygotes and embryos
283
cysts. Despite a high incidence of implantation, net results are poor because of the low viability of embryos growing in vitro, between the 8-cell and the blastocyst. Only 3% more births can be expected when this method is used in conjunction with conventional IVF. 8 The efficiency of embryo freezing has increased considerably with the introduction of the third method, which uses 1,2-propanediol (PROH) and sucrose for cryoprotection, and can be applied to zygotes and very early embryos. 3•9 The initial goal of the present investigation was threefold: first, to determine whether results differ when embryos are frozen in 1,2-propanediol and sucrose on day 1 (zygote) or day 2 (3- to 5-cell) following egg collection; second, to investigate whether the incidence of implantation of fresh embryos would change when the morphologically "better" embryos would be selected for cryopreservation; third, to evaluate which of the above cryopreservation methods would produce the highest number of births, in terms of the availability of spare embryos and the age at which embryos are frozen. MATERIALS AND METHODS Patients
Couples with a long duration of infertility (range, 2 to 12 years) were treated with IVF for a variety of disorders, such as tubal disease (n = 24), endometriosis (n = 5), immune and/or cervical infertility (n = 4), male infertility (n = 9), and unexplained infertility (n = 3). The cutoff age of the female partner was 45 years. Patients were stimulated with a combination of clomiphene citrate (CC; Serophene, Serono, Rome, Italy) 100 mg daily from days 2 to 6 and follicle-stimulating hormone (FSH; Metrodin, Serono, Rome, Italy) 150 IU daily from days 1 to 5 and/ or human menopausal gonadotropin (hMG; Pergonal, Serono, Rome, Italy), usually with 150 IU daily from day 5. Patients were monitored from day 7 or 8 of their cycle and 75 or 150 IU of hMG was supplied on a day-to-day basis, depending on the number of follicles, follicular size, and serum estradiol (E 2). In more than 90% of these patients, ovulation was induced using 5000 IU of human chorionic gonadotropin (hCG; Profasi, Serono, Rome, Italy). The remainder of the patients had egg retrieval timed by an endogenous luteinizing hormone (LH) surge detected in urine, using radioimmunoassays (RIAs; LH MAlA Clone, Serono Diagnostics, Woking, England), and 5000 284
Cohen et al.
Cryopreservation of zygotes and embryos
IU hCG was given to these patients before egg collection. Menstruation in another group of patients was timed using 10 mg/day norethisterone (Norluton, Parke-Davis, Santurce, PR) for a duration of 10 to 31 days, beginning on days 15 to 18 of the menstrual cycle prior to the scheduled egg collection cycle. The last day of drug therapy was defined as day zero of the treatment cycle. 10 Follicular stimulation was achieved with similar methods as described previously. This group of patients always received hCG for ovulation induction. All patients received 3000 IU hCG on days 2 and 4, respectively, following egg collection for luteal phase support. Oocytes were collected 34 to 35 hours after hCG administration or 26 to 30 hours after onset of the LH surge. The collection was performed laparoscopically or via transvaginal ultrasound using a suction pump. Follicles were flushed repeatedly with warm culture medium and the follicular contents were checked immediately in an adjacent laboratory. Insemination was performed 3 to 12 hours after egg collection, depending on cumulus expansion. Zygote and Embryo Freezing
A maximum of three embryos were replaced 2 days after egg collection and the remainder was frozen. In a number of patients, two fresh embryos were replaced and the remainder was frozen. The policy was to freeze a minimum of two embryos per patient. Oocytes and embryos were cultured in bicarbonate-buffered Earle's medium containing 15% maternal or donor serum, kept in small droplets under paraffin oil, and maintained in 5% C0 2 , 5% 0 2 , and 90% N 2 Y The donor serum was obtained from young, fertile, healthy, normally cycling women. The medium used for gamete and embryo culture was used for cryopreservation. Petri dishes containing pronucleate or 2- to 5-cell embryos were left at room temperature for 10 minutes and exposed to serum-supplemented culture medium containing 1.5 M PROH for 10 to 15 minutes, and 1.5 M PROH with 0.1 M sucrose for 2 to 5 minutes. Shrinkage and swelling of embryos was observed and recorded on videotape using an inverted microscope. Embryos were pipetted into small glass ampules (0.3 to 0.5 ml) pulled from glass tubes (Durham Tubes, Samco, Cambridge, England, or Corning, New York, NY). The ampules were heat-sealed and cooled using either a Planer R204 (Planer, Sunbury-on-Thames, England) or FTS (FTS-Systems, Stone Ridge, NY) cell freezer. Fertility and Sterility
Embryos were cooled according to the following curve9 : -2°C/min to -6°C, hold for 15 minutes for induction of ice crystallization using cold forceps, further cooling at -0.3°C/min to -30°C and direct transfer into liquid nitrogen. Thawing and Replacement
Patients returned for their frozen embryos 2 to 10 months after storage. CC was given on patients' menstrual cycle days 2 to 6 (50 or 100 mg daily), or on days 5 to 9 in patients with irregular or prolonged cycles or when ovulatory problems were anticipated. The decision to give CC either starting on day 2 or day 5 was made randomly in order to investigate effects on implantation. The results of this selection will be evaluated at a later stage when more patients have been studied. The remainder of the patients were monitored during their natural cycle. Urine collections were obtained seven times daily at 2- to 3-hour intervals for LH determination. Serum E 2 was assayed two or three times during the follicular phase, and daily ultrasound was performed to confirm ovulation. In some of the patients who received CC and had high E2, ovulation was induced with 5000 IU hCG. Some patients did not demonstrate evidence of ovulation during the natural cycle. Their treatment was abandoned and monitoring was initiated during another cycle following CC induction. Embryos were thawed 1 or 2 days following ovulation according to their age at the time of freezing. The onset of the LH surge was defined when a urinary LH value was higher than 1.0 IU/hr, provided previous assays were consistently lower and at least three later values were higher than 1.0 IU/hr. When zygotes were frozen, the interval between ovulation and thawing was the same or 6 hours shorter than that in the retrieval cycle. Differences in timing were due to the time of the onset of the LH surge. Thawing was not performed between 1:00 and 6:00 A.M., but the interval was shortened by 6 hours. This has not affected results so far, but more data have to be collected before definite conclusions can be drawn. The interval between ovulation and thawing was 6 to 12 hours shorter than the embryonic age when 2- to 5-cell embryos were frozen. Cleaved and intact embryos normally were replaced within 3 hours after thawing; however, in some cases, further development was observed overnight. Thawed zygotes were left in culture for 20 to 28 hours until cleavage had occurred. Ampules containing the embryos were Vol. 49, No.2, February 1988
hand-stirred in a 31 °C water bath for 20 to 35 seconds until all ice crystals disappeared and their contents were emptied into Earle's medium containing 15% maternal serum, 1.5 M PROH, and 0.2 M sucrose. 2-11 PROH was removed in six steps, at room temperature, and the embryos were washed several times with culture medium and slowly warmed to 37°C. Statistics
Data were analyzed with the chi-square test in all instances unless the expected frequency was less than 5, in which case the Fisher's exact test was used. RESULTS Selection of Zygotes and Embryos for Cryopreservation
One hundred twelve IVF patients treated between January 1986 and April1987 had more than three embryos, and their spare zygotes or embryos were frozen (Table 1). The embryos were separated into two groups (A and B), based on an examination of their morphologic features under an inverted microscope. This study was performed in order to establish whether positive embryo selection on the basis of morphologic features for cryopreservation would adversely affect the incidence of implantation of fresh embryos. Group A cleaved embryos had homogenous-looking blastomeres and few or no extracellular fragments. Group B embryos had uneven or disorganized-looking blastomeres or were discolored. Many of these embryos had fragments. It was not possible to distinguish among differences in embryos from 51 patients, and their embryos were allocated randomly for cryopreservation or immediate transfer (Table 1). No differences were found in the incidence of implantation of 2-day-old fresh embryos when group Table 1 Selection of Embryos for Cryopreservation and the Effect on the Incidence of Clinical Pregnancies of Fresh Embryos in Patients Who Had Spare Embryos Frozen Incidence of clinical pregnancies of fresh embryos Stage of Positive selection Negative selection for freezing for freezing No selection embryos (group B) possible (group A) at freezing Zygotes 2- to 5-cell
Cohen et al.
11/47 (23%)
4/14 (29%)
7/28 (25%) 6/23 (26%)
Cryopreservation of zygotes and embryos
285
replacements had been unsuccessful, returned for their frozen embryos. Zygotes (n = 41) or cleaved embryos (n = 57) were thawed during 50 cycles. Several patients had a sufficient number of embryos stored to allow for more than one attempt. One twin and ten singleton pregnancies were established after 4 7 freeze/thawed embryo replacements. The incidence of clinical pregnancy per thawing attempt was 22%, but one patient with positive fetal heart beat miscarried. Seven patients delivered healthy infants, and three other pregnancies are now progressing normally beyond week 26. The expected incidence of birth per attempt is 20%, which is comparable to that of patients who had fresh embryos replaced (Table 2). Eleven babies are expected out of a group of 98 thawed (11%) and 69 replaced (16%) embryos. Fifteen of these patients still have 29 remaining embryos frozen, and it is therefore likely that the incidence of implantation will increase further for this group. Cleaved embryos were thawed in 33 cycles and zygotes during 17 other cycles: the total number of cycles was 50, but five patients had embryos thawed on two separate occasions (Table 3). All patients with zygotes had at least one intact embryo replaced after thawing, and 30 of 33 (90%) patients with cleaved embryos had a replacement. Relatively more pregnancies were established with freeze/thawed zygotes than with cleaved embryos (29% versus 18%), but the difference is not significant. A slight increase in pregnancy was observed when more than one embryo was thawed, but eight pregnancies were established when a single embryo was replaced. This compared favorably with the replacement of two or three freeze/thawed embryos (Table 3). Cleaved embryos usually were thawed early in the morning or afternoon, and re-
Table 2 Breakdown of Results When a Maximum of Two or Three Fresh Embryos Are Replaced and the Remainder Is Frozen (%)
Proportion of patients with fertilization Incidence of clinical pregnancy• Incidence of birth or ongoingb pregnancy Incidence of cryopreservation Incidence of clinical pregnancy following thawing" Incidence of birth or ongoingb pregnancy following thawing Current cumulative incidence of birth or ongoing pregnancy following IVF and cryopreservationc
148/167 45/167 36/167 70/167
(89) (27) (22) (42)
11/50 (22) 10/50 (20) 46/167 (28)
• Fetal heart beat. b Ongoing beyond week 14. c The final incidence of birth will be presumably higher since 30% of patients have remaining embryos frozen after one attempt.
A or B embryos were replaced, indicating that crude morphologic parameters have little prognostic value at that time of embryonic development. There are few morphologic differences between zygotes, and selection based on their morphologic features is not possible. Random selection of zygotes for cryopreservation did not affect the incidence of implantation after the remaining fresh zygotes cleaved and were replaced (Table 1). Implantation of Cryopreserved Zygotes and Embryos
One hundred sixty-seven patients were treated with IVF in 1986, and 89% had one or more embryos available for replacement. Seventy patients (42%) had more than three embryos; cryopreservation of zygotes and cleaved embryos was performed at random. Forty-five patients, whose fresh embryo Table 3
Incidence of Implantation and Numbers of Embryos Thawed and Replaced Incidence of clinical pregnancies per patientc No. frozen embryos Embryonic stage at freezing
Thawed"
Survived andb replaced
1 2 3 or more Total 1 2 3 Total
Zygote
2- to 5-cell
All stages
2/7 1/4 2/6 5/17 (29%) 3/10 2/6
1/9 4/19 1/5 6/33 (18%) 5/18 1111 0/1 6/30 (20%)
3/16 5/23 3/11 11/50 8/28 3/17 0/2 11/47
Oil 5/17 (29%)
• Include embryos that did not survive thawing. b Also includes embryos that partially survived the thawing process.
286
Cohen et al.
Cryopreservation of zygotes and embryos
c
(19) (22) (27) (22) (29) (18)
(0) (23)
No significant differences between groups.
Fertility and Sterility
Table 4 Survival and Implantation of Cryopreserved Zygotes and Cleaved Embryos Stage at freezing
Incidence of post-thaw survival (%)
(%)
3- to 5-cell 2-cell Zygotes
31/41 (76) 13/16 (81) 25/41 (61)
6/31 (19) 0/13 (0) 6/25 (24)
•P
=
Incidence of fetal heart beats
0.06 (Fisher's exact test).
placement was performed within 3 hours. In some cases, thawing was timed late in the evening because of the onset of the LH surge. In the latter cases, the embryos were cultured overnight and replaced in the morning for organizational purposes. Pregnancies were achieved in both groups of patients whose thawed embryos were cultured for a short time or overnight, and preliminary data did not reveal any differences. No differences in embryonic survival or implantation were found in this relatively small group of patients when the follicular stimulation was taken into account. Pregnancies were achieved regardless of whether the eggs were derived during cycles when hMG, FSH, or Norluton was supplied. Five pregnancies were obtained in the 24 patients with tubal infertility. The six other pregnancies were obtained in couples with different causes, and differences in incidence of implantation between these groups were not found. Survival and Cryoinjury
Survival of zygotes and cleaved embryos exceeded 61%, but gestational sacs were derived only from zygotes or cryopreserved 3- to 5-cell embryos. None of the frozen 2-cell embryos implanted after replacement (Table 4). Twenty of the 98 (20%) zygotes and embryos that were thawed had a crack in Table 5
the zona pellucida. The incidence of cracks was unrelated to the stage of the embryo at freezing (Table 5), but while none of the zygotes with cracked zonae pellucidae survived, three out of ten cleaved embryos with a cracked zona pellucida had intact blastomeres upon thawing. A small hole was noticed in one of these embryos, and this embryo was replaced immediately. The two other embryos were cultured for 2 more days for observation of further cleavage. Both embryos lost half of their blastomeres through a large gap in the zona pellucida, but compaction occurred and the embryos were replaced. None of these embryos implanted. The overall survival of all zygotes and cleaved embryos was 70%, but 85% of embryos with an intact zona pellucida survived cryostorage (Table 5). DISCUSSION
This study demonstrates that zygote and early cleaving embryo freezing using PROH and sucrose is an efficient and simple method for establishing a high incidence of implantation. One fifth of patients with "extra" embryos (i.e., embryos not replaced in the IVF treatment cycle) frozen and thawed had an ongoing pregnancy and this proportion may well increase, considering that 30% of these patients have more embryos frozen. An advantage of this technique is that at least 40% of the patients will have embryos for cryopreservation. It is therefore expected that one in three egg retrievals will lead to live birth when this type of cryopreservation is used in conjunction with conventional IVF. This cryopreservation method seems to be advantageous for other reasons as well. PROH and sucrose have not yet been shown to cause any chromosome damage in mammalian studies and a considerable number of mammalian and human births following cryostorage have been reported. 10 This method is less time-consuming
Observations on the Survival of Cryopreserved Zygotes and Cleaved Embryos Embryonic stage at freezing Zygote" (%)
Incidence of survival Incidence of complete survival Incidence of cracks in the zona pellucida Survival of embryos with cracked zonae pellucidae Survival when the zona pellucida is intact
25/41 25/41 10/41 0/10 25/31
3- to 5-cell"
(%)
(%)
(61) (61) (24) (0) (81)
44/57 32/57 10/57 3/10 41/47
Total
(77) (56) (18) (30) (87)
69/98 55/98 20/98 3/20 66178
(70) (56) (20) (15) (85)
• No significant differences between groups. Vol. 49, No.2, February 1988
Cohen et al.
Cryopreservation of zygotes and embryos
287
Table 6 Incidence of Embryo Cryopreservation and Implantation Following Thawing and Replacement When Embryos Are Frozen on Different Days Following Egg Retrieval Day following egg collection when cryopreservation was performed
Stage of the embryo at freezing Cryoprotective Proportion of patients whose spare embryos were frozen Proportion of egg retrievals leading to cryopreservation (I) Incidence of ongoing pregnancies or birth following thawing (II) Expected increased incidence of birth through cryopreservation based on one egg retrieval (I X 11/10) Source
1 and 2
3 and 4
5 and 6
Zygote or 2- to 5-cell PrOH 70172 (97%)
3- to 10-cell DMSO 79/162 (49%)
Expanded blastocyst Glycerol 96/270 (36%)
70/167 (42%)
79/378 (21%)
96/881 (11%)
10/50 (20%)
2/29 (7%)
7/27 (26%)
8% This article
than methods published previously, 1 •2 and the present results of survival and implantation corroborate with those of others. 3 •7 This type of cryopreservation compares favorably with freezing cleavage-stage embryos using DMSO or expanded blastocysts using glycerol. The latter method resulted in a high incidence of implantation, but few blastocysts were available because of the poor development rate in vitro beyond day 3. 6 The current results are compared in Table 6 with results presented previously. 6 •12 Although statistical comparisons are invalid, since the observations were performed in separate clinics, some tentative conclusions may be drawn from the data because of relative similarities in follicular stimulation, patient selection, and general laboratory methodologies between the two IVF clinics. Almost three times more patients have extra embryos frozen when freezing is performed 1 or 2 days, rather than 5 or 6 days, after egg collection. The incidence of implantation per patient who had embryos frozen is similar between the two groups. The efficiency of the present method is therefore larger and the incidence of births following one egg collection will increase by at least 8%. The conventional cryopreservation methods will increase results only marginally (Table 6). Several suggestions can be used in order to improve cryopreservation further. First, the number of fresh embryos for replacement could safely be reduced because frozen and fresh embryos have an equal chance of implantation. Increasing the number of oocytes through altering follicular stimulation, for instance with the use of LH-RH agonist, should be especially beneficial. The current results demonstrate that only one intact cryopreserved embryo should be replaced at a time; this is not in 288
Cohen et al.
Cryopreservation of zygotes and embryos
2% Fe hilly et a!. 8
3% Fe hilly et a!. 8
agreement with results following replacement of fresh embryos. Despite the tremendous stress put upon the embryo during cryopreservation, the capacity to implant and develop normally remains intact. This raises the question of whether results would increase if the morphologically "better" embryos were cryopreserved and the remainder replaced. It has been demonstrated previously that gross morphology of 3- to 6-day-old embryos determines their survival. 5 •6 Despite a number of observations supporting this concept, we have insufficient data available to extrapolate this usage to younger embryos. Conversely, positive selection on the basis of embryo morphology did not impair the incidence of implantation of fresh embryos, and it is now our policy to freeze only zygotes (in which morphologic differences are hard to determine) or homogenouslooking cleaved embryos. More attention may have to be paid to the synchrony between a freeze/thawed zygote or embryo and the uterus. A previous study indicated that the chance of pregnancy increased when freeze/thawed blastocysts were transferred 24 hours earlier following ovulation than their chronologie age indicated. 2 Asynchrony following in vitro culture will presumably be small in zygotes or 4-cell stage embryos, and timing the moment of thawing may require more accuracy than is the case with blastocysts. Results of cryopreservation also may improve when patients with anovulatory or irregular cycles are given small amounts of CC. Preliminary observations have revealed that the best results are obtained when CC is given 50 mg daily, days 5 to 9 of the patient's menstrual cycle, and hCG for ovulation induction should be avoided. Decreasing the incidence of cracks in the zona Fertility and Sterility
pellucida could be another factor that improves the results of cryopreservation. The present study indicates that 85% of embryos could survive, which would probably increase the incidence of implantation further. There are presumably numerous causes for cracks in the zona pellucida. Some patients may have oocytes with weaker zonae pellucidae, as was shown recentlyY The formation of cracks also could be circumvented by reducing the size or volume of the container used for storing the embryos. Cracks are rare when straws are used for cryopreservationY Fast induction of ice crystallization, achieved through cooling in a methanol bath rather than in air, also may reduce the incidence of cracks. Embryos seem to survive better when nuclei are visible, 3 •7 the only exception being the 2-cell embryo. Dramatic changes occur in embryonic metabolism at the 2-cell stage and failure to implant may be due to a similar mechanism described for 2-cell embryos of some mammalian species (the 2-cell block). Testart 7 demonstrated recently that cryopreserved embryos are more viable than fresh embryos because of uterine hostility during the treatment cycle. This result may be biased because of possible suboptimal stimulation protocol and cannot be extrapolated to the data presented here, since the incidence of implantation of fresh and frozen embryos was the same in our work. However, the incidence of ongoing clinical implantation per freeze/ thawed embryo was 11% (11/98) in our work and 9% (14/151) in the Paris clinic. In some groups of society, freezing of unfertilized eggs is ethically accepted, whereas embryo freezing is considered to be experimental. The latter technique often is regarded as hazardous because it involves manipulation of potentially new individuals. Whether these opinions are justified or not, the truth is that some IVF centers have abandoned embryo freezing programs for ethical reasons and are now investigating oocyte cryopreservation. This investigation is being done despite increasing evidence from work in laboratory animals that the metaphase II spindle is prone to cryoinjury. A second disadvantage of this development is that the technology of cryopreservation of the mammalian embryo is well established, whereas reliable methods for freezing unfertilized eggs are still being developed. It could be argued that the zygote is ethically different from the cleaving embryo, since the chromosomes of the male and female gametes are still grouped in separate nuclei: a poten-
tial new individual has therefore not been genetically established. Another advantage of freezing zygotes is of a more technical nature. It is easier to visualize pronuclei in zygotes than nuclei in blastomeres, and the cell cycle of blastomeres may be asynchronous, leading to partial damage during the freezing process.
Vol. 49, No. 2, February 1988
Cohen et al.
Acknowledgments. We thank Benjamin Brackett, Ph.D., D.V.M., William Graves, Ph.D., and Carol Keefer, Ph.D., for their collaboration. Ms. Twala Grant, Ms. Leigh Inge, and Ms. Donna Sammond are acknowledged for technical support.
REFERENCES 1. Trounson AO, Mohr L: Human pregnancy following cryopreservation, thawing and transfer of an eight-cell embryo. Nature 305:707, 1983 2. Cohen J, Simons RF, Edwards RG, Fehilly CB, Fishel SB: Pregnancies following the frozen storage of expanding human blastocysts. J In Vitro Fert Embryo Transfer 2:59, 1985 3. Testart J, Lassalle B, Belaisch-Allart J, Hazout A, Forman R, Rainhorn JD, Frydman R: High pregnancy rate after early human embryo freezing. Fertil Steril 46:268, 1986 4. Chen C: Pregnancy after human oocyte cryopreservation. Lancet 1:884, 1986 5. Freemann L, Trounson A, Kirby C: Cryopreservation of human embryos: progress on the clinical use of the technique in human in vitro fertilization. J In Vitro Fert Embryo Transfer 3:53, 1986 6. Cohen J, Simons RF, Fehilly CB, Edwards RG: Factors affecting survival and implantation of cryopreserved human embryos. J In Vitro Fert Embryo Transfer 3:46, 1986 7. Testart J: Evidence of uterine inadequacy to egg implantation in stimulated in vitro fertilization cycles. Fertil Steril 47:855, 1987 8. Fehilly CB, Cohen J, Simons RF, Fishel SB, Edwards RG: Cryopreservation of cleaving embryos and expanded blastocysts in the human: a comparative study. Fertil Steril 44:638, 1985 9. Lassalle B, Testart J, Renard JP: Human embryo features that influence the success of cryopreservation with the use of 1,2 propanediol. Fertil Steril 44:645, 1985 10. Frydman R, Forman R, Rainhorn JD, Belaisch-Allart J, Hazout A, Testart J: A new approach to follicular stimulation for in vitro fertilization: programed oocyte retrieval. Fertil Steril 46:657, 1986 11. Edwards RG, Fishel SB, Cohen J, Fehilly CB, Steptoe PC, Webster J: Factors influencing the success of in vitro fertilization for alleviating human infertility. J In Vitro Fert Embryo Transfer 1:3, 1984 12. Cohen J, Simons RF, Edwards RG, Fehilly CB, Fishel SB, Hewitt J, Steptoe PC, Webster J: Birth following the replacement of a cryopreserved hatching blastocyst. Lancet 1:647, 1985 13. Testart J: Unpublished data
Cryopreservation of zygotes and embryos
289