- Email: [email protected]
The effect of gonadotropin-releasing hormone agonist on embryo quality and pregnancy rate following cryopreservation*
Vol. 59, No.5, May 1993
FERTILITY AND STERILITY
Printed on acid-free paper in U.S.A.
Copyright e 1993 The American Fertility Society
The effect of gonadotropin-releasing hormone agonist on embryo quality and pregnancy rate following cryopreservation*
Abraham Benshushan, M.D.t YossefEzra, M.D. Alex Simon, M.D.
Nathan Mordel, M.D. Aby Lewin, M.D. Neri Laufer, M.D.
In Vitro Fertilization Unit, Department of Obstetrics and Gynecology, Hadassah University Hospital, Ein-Kerem, Jerusalem, Israel
Objective: To examine the effect of GnRH analogue (GnRH-a) on the quality of frozen-thawed embryos and the pregnancy rate (PR) resulting from transfer. Design: A retrospective study of two groups of women undergoing transfers of frozen-thawed embryos. In group 1 ovulation induction with hMG was begun after hypophyseal desensitization with GnRH-a for 2 weeks. In group 2 ovulation was induced with hMG only. The freezing and thawing techniques were identical for both groups. Setting: In vitro fertilization unit at a university hospitaL Patients: The study group (group 1) included 108 women who underwent 137 transfer cycles of frozen-thawed embryos, and the control group (group 2) included 44 women in 51 cycles of thawed ETs. Interventions: Ovum pick-up and ET techniques were the same for both groups. Methods of embryo freezing and thawing were identical, as were the morphological criteria for grading the embryos. Results: The morphology of embryos was similar in both groups, as was the number of embryos that had at least 50% intact blastomeres (83% ± 23% and 78% ± 30% for group 1 and group 2, respectively). The PR (16 of 137 [1L7%] and 6 of 51 [11.8%], respectively) as well as the abortion rate (30%) were similar for both groups. Conclusions: The use of GnRH-a does not affect the quality of embryos nor the pregnancy outcome. Because the yield of frozen embryos per ovum pick-up is higher in cycles stimulated by GnRH-a/hMG, the PR per pick-up cycle is thus anticipated to be higher. Fertil Steril 1993;59:1065-9 Key Words: Frozen-thawed embryos, GnRH analogue, in vitro fertilization, pregnancy rate, embryo quality
Controlled ovarian hyperstimulation with menotropins after medical hypophysectomy induced by GnRH agonist (GnRH -a) has become a routine treatment in most IVF programs. Higher pregnancy rates (PRs), a decreased incidence of cycle cancel-
Received October 19, 1992; revised and accepted January 28, 1993. * Presented at the Annual Meeting of The Israeli Society of Fertility, Tel Aviv, Israel, March 3 to 4, 1992. t Reprint requests: Abraham Benshushan, M.D., Department of Obstetrics and Gynecology, Hadassah University Hospital, Post Office Box 12000, Jerusalem 91120, Israel. Vol. 59, No.5, May 1993
lation, and an increase in the number of oocytes retrieved and in the number of patients proceeding to ET have been reported with this stimulation regimen (1, 2). In addition, this combination of medications was shown to result in a lower incidence of spontaneous abortion, suggesting better embryo quality or endometrial receptivity (3). Recently, concern has been raised regarding the possible adverse effects of GnRH -a pretreatment on the survival and quality of frozen-thawed-embryos and on the PR resulting from transfer of frozenthawed embryos (4-6). Testart et al. (4) suggested that this untoward effect may be the result of a
Benshushan et al.
Effect of GnRH-a on frozen-thawed embryos
1065
mobilization of defective follicles and oocytes that produce embryos less able to resist freezing-thawing, i.e., those demonstrating irregular blastomeres or cytoplasmic exudates. Lornage et al. (5) have reported a rise in the percentage of embryos with <50% intact blastomeres when using the leuprolide acetate-hMG protocol. An alternative explanation was suggested by Keenan et al. (6). These authors demonstrated that the development of embryos after suppression with GnRH -a was accelerated. This phenomenon interfered with the cryoprotectant qualities of propanediol, which was found to be maximally effective when the cleaving embryo is in interphase. When embryo development is accelerated, cell cycle becomes increasingly disorganized, blastomeres divide asynchronously, resulting in a reduced cryoprotective effect and diminished survival of frozen embryos. Given the important role of cryopreservation in IVF, any demonstrable adverse effect ofGnRH-a on the efficiency of this process may have major implications on its use in ovulation induction for assisted reproductive techniques. This study was therefore undertaken to evaluate the effect of GnRH -a pretreatment on the quality and the PR of frozen-thawed embryos in menotropin-stimulated cycles. MATERIALS AND METHODS Patient Population
Patients undergoing IVF with frozen-thawed embryos at the Hadassah University Hospital between 1987 and 1991 were included in this retrospective study. Group 1 consisted of 108 women who underwent 137 transfer cycles of frozen-thawed embryos. These embryos were obtained from 108 stimulation cycles using GnRH-a and hMG (Pergonal; Ikafarm, Ramat-Gan, Israel). Group 2 consisted of 44 women who had 51 transfers of frozen-thawed embryos from 44 hMG-stimulated cycles. Stimulation Protocols
Follicular stimulation in Group 1 was achieved using a combination of GnRH -a (D-Trp6 , Decapeptyl; Ferring Pharmaceuticals, Kiel, Germany) and hMG as previously described (7). Briefly, GnRH-a was begun at the midluteal phase of the preceding cycle (0.5 mg SC) and continued until the day of hCG administration. Three ampules of hMG were initiated on the 7th day of menses and were 1066
Benshushan et al.
continued in an individualized fashion. In group 2, hMG was started on cycle day 3 (3 ampules per day) and continued for 4 days. Starting cycle day 7 hMG dosage was adjusted depending on serum E 2 levels and follicular development. Ovulation was triggered with hCG (10,000 U IM, Chorigon; Ikafarm) when at least three follicles attained a mean diameter of 16 mm and serum E 2 levels were >500 pg/mL (conversion factor to SI unit, 3.671). In group 1, administration of Decapeptyl was discontinued at this time. Oocyte retrieval was performed under general anesthesia using a transvaginal ultrasound-guided technique 34 to 36 hours after hCG administration. Embryo Freezing
Embryos were frozen 2 to 3 days after in semination at the two- to eight-cell stages. The embryos were graded by the morphological criteria described previously by Bolton et al. (8): grade A, regular, spherical blastomeres with no extracellular fragmentation; grade B, regular, spherical blastomeres with some extracellular fragmentation; grade C, blastomeres slightly irregular in size and shape with considerable extracellular fragmentation; grade D, barely defined blastomeres with considerable extracellular fragmentation. Embryos were incubated for 15 minutes in phosphate-buffered saline (PBS) supplemented with 1.5 M propanediol, 20% heat-inactivated human maternal serum, and 0.4% human serum albumin (HSA). Subsequently, they were placed into a plastic cryotube (3-63401; Nunc, Roskilde, Denmark) containing the above medium supplemented with 0.1 M sucrose. Freezing was performed using a computerized freezer (Planer Biomed Kryo 10-16, Middlesex, London, United Kingdom) with manual seeding. Embryos were stored in liquid nitrogen at -196°C for 1 to 54 months until thawing. Embryo Thawing
On the day of transfer, the cryotube was placed at room temperature for 1 minute and then in warm water (35°C to 37°C) for another minute until complete thawing ofthe medium was observed. The embryos were incubated for 5 minutes in PBS medium containing PROH (1.5 M), sucrose (0.2 M), 20% human serum, and 0.4% HSA. The embryos were incubated in serially decreasing concentrations of propanediol (1.0 M, 0.5 M, 0 M) for 5 minutes at each concentration. The embryos were rinsed three times in PBS and HSA and then incubated in cleavage medium (Ham's F-10; Flow Laboratories, Rickmansworth, Herts, United Kingdom) with 20% pre-
Effect of GnRH-a on frozen-thawed embryos
Fertility and Sterility
Table 1 Distribution of the Indications for IVF in the Study Group
Table 2
Clinical Parameters During Ovulation Induction Group 1
Group 2 (n = 41)
Group 1 (n = 108) %
Indications Tubal factor Unexplained Male factor Endometriosis Ovum donation
67 16.8 9.5 0.7 5.8
68 15.7 7.8 3.9 3.9
ovulatory serum at 37°C in an atmosphere of 5% C0 2 for 0.5 to 4 hours until transfer. Immediately before transfer, the embryos were placed in a medium containing 90% human serum. Transfer of Frozen-Thawed Embryos
Endometrial preparation for ET was achieved by one of the following protocols: spontaneous cycles were used in 67.5% of group 1 and in 62.7% of group 2 transfers; cycles regulated by micronized estrogen (Estrofem; Novo Industry A/S, Copenhagen, Denmark) were used in 20.4% and 23.5% of cycles, respectively; cycles stimulated by clomiphene citrate (CC) were used in 9.5% and 7.7% of cycles, respectively; and finally, cycles stimulated by hMG were used in 2.9% and 3.9% of cycles, respectively. The selection criteria for ovulation induction agents were the following: for normal ovulatory patients we have used spontaneous cycles; for oligo-ovulatory patients either CC or menotropins were used to induce ovulation; and when thawed embryos were transferred in an ovum donation program, artificial cycles were used. Cycles were canceled when they were thought to be suboptimal, i.e., maximal E 2 levels were <200 pg/mL (conversion factor to SI unit, 3.671), P levels <3 ng/mL (conversion factor to SI unit, 3.18) on the day after LH surge, or endometrial thickness of <10 mm on the day of ET. The embryos were transferred 48 to 72 hours after the detection of a spontaneous LH surge or hCG injection (10,000 IU). Determination of the occurrence of the LH surge was established by Clearplain kits for detection of urinary LH (Unipath Limited, Bedford, United Kingdom). A maximum of four fresh or frozen-thawed embryos were usually transferred. Our policy regarding fresh ET is to transfer the morphologically best embryos immediately and to freeze all other embryos graded A to D (according to the criteria of Bolton et al. [8]). However, morphological criteria to determine embryo survival are Vol. 59, No.5, May 1993
Age HMG dose:j: Peak E 2§ No. of oocytes retrieved No. of embryos produced Oocyte fertilized (%)
33.6 32.7 2281 14.5
± ± ± ±
4* 12.4 1067 6.2
Group 2 32.4 26.2 1828 10.9
± ± ± ±
Probability
3.8 11.7 817 4.3
NSt 0.01 0.02 0.001
9.7 ± 4.0
8.2 ± 2.8
0.014
70.6
77.4
0.036
*Values are means± SD. t NS, not significant. :j: Number of hMG ampules needed for adequate response. §Values are in pg/mL, conversion factor to SI unit, 3.671.
controversial and can be misleading; therefore, we usually transfer only frozen-thawed embryos with >50% intact blastomeres. Statistical Analysis
Results are presented as means± SD. Statistical evaluation was performed using Student's t-test and X2 test whenever appropriate.
RESULTS
Oocyte retrieval was performed in 108 women treated with GnRH-a/hMG (group 1) and 44 women treated with hMG alone (group 2). The average age of the women in both groups was not significantly different (33.6 ± 4.0 versus 32.4 ± 3.8), nor were there differences between the two groups in the indications for IVF (Table 1). The mean number of hMG ampules per cycle, peak serum E 2 levels, as well as the average number of oocytes retrieved, and embryos obtained were significantly higher in group 1. However, the proportion of oocytes that fertilized was greater in the group that received hMG only (Table 2). The morphological grading of the embryos at the time of freezing was not different between the two groups (Table 3). Similarly, the duration of freezing,
Table 3
Embryo Quality at Time of Freezing Group 1
Grade A Grade B Grade C GradeD
50 69 14 4
(36.2)* (50.4) (10.5) (2.9)
Group 2 22 (43.1) 24 (47.1) 5 (9.8)
* Values in parentheses are percents.
Benshushan et al.
Effect of GnRH-a on frozen-thawed embryos
1067
the age of the embryos at the time of freezing, and the percent of embryos with >50% intact blastomeres after thawing was not different between the two groups (Table 4). Finally, the number of thawed embryos transferred was similar for both groups (2.9 ± 2.3 and 2.5 ± 1.2, respectively). The PR after transfer of frozen-thawed embryos obtained from GnRH-a/hMG stimulation (group 1) was 11.7% (16/ 137). This rate was not significantly different from the PRof 11.8% (6/51) obtained from embryos cryopreserved after hMG stimulation alone (group 2). Ofthe pregnancies achieved in group 1, 12 are either delivered or ongoing (>20 weeks' gestation), and 4 ended in spontaneous abortions. In group 2, 4 pregnancies are either delivered or ongoing and 2 ended in abortion. DISCUSSION
This work demonstrates that the use of GnRH -a pretreatment did not adversely affect embryo quality before freezing and did not negatively interfere with their ability to withstand cryopreservation as evidenced by their morphology after thawing. Moreover, the capacity of these embryos to continue and develop in the uterus was not influenced by the analogue and resulted in a PR of 11.7%, which was comparable with that achieved after the transfer of embryos derived from menotropin stimulation only (11.8%). These results are in sharp contrast to the findings of Testart et al. (4), Lornage et al. (5), and Keenan et al. (6) who found an adverse effect of GnRH-a on embryo survival after cryopreservation. Testart et al. (4) demonstrated that embryos obtained from IVF cycles using GnRH-a were significantly more susceptible to the freezing and thawing procedures and that, consequently, the proportion of embryos surviving cryopreservation was higher in cycles without GnRH-a. They have also reported a significantly lower PR after the transfer of cryopreserved embryos obtained from cycles with GnRH-a as compared with those without GnRH-a (2.8% and 8.9%, respectively). It should be noted that Testart et al. (4) have used five different ovulation induction protocols for IVF patients. To minimize the effect of the different protocols on the results, we have compared our results only to the relevant protocols that were similar to ours, namely, long GnRH-a/ hMG as compared either with hMG alone or CChMG. Lornage et al. (5) found that the survival of embryos after cryopreservation was significantly lower after a long GnRH -a protocol when compared 1068
Benshushan et al.
Table 4
Characteristics of Frozen-Thawed Embryos* Group 1
Duration of freezing (mo) 7 Age of embryos at freezing (d) 2.67 ET day from LH surge 2.57 No. of embryos transferred 2.89 Embryos with >50% blastomeres (%) 83.7
*P
=
Group 2
± 8t 7.9 ± 11.1 ± 0.48 2.62 ± 0.6 ± 0.56 2.6 ± 0.51 ± 1.29 2.47 ± 1.18 ± 23.5 78.3 ± 30
not significant for all values.
t Values are means± SD.
with that of embryos obtained from CC-hMG-stimulated cycles (52.2% and 66.6%, respectively). However, no similar effect on PR was noted (16% and 14.5%, respectively). The differences between our results and those of Testart et al. (4) and Lornage et al. (5) may stem from the different protocols used for down regulation that, in turn, may have affected oocyte and embryo quality. Both groups used a long 14-day pituitary down regulation but, in contrast to our group, started at the commencement of cycle and not in the midluteal phase. Consequently, the initial flare-up of gonadotropins in the beginning of the follicular phase may have prematurely exposed the oocytes to LH and P, resulting in a higher proportion of oocytes and embryos of reduced quality (9). We suggest that such embryos with reduced quality that were generated from cycles with premature P secretion are more vulnerable to cryopreservation. On the other hand, our results are in agreement with those of Keenan et al. (6) who used similar groups for comparison. They reported PRs of 29% and 14% per frozen-thawed ET generated from conventional cycles or down-regulated cycles, respectively. However, these rates did not differ significantly because of the small samples used for comparison. In addition, our observations obtained after the transfer of cryopreserved embryos, corroborate a previous recent report by Oehninger et al. (10) who evaluated the results of cryopreserved preembryos. They compared the performance of preembryos after ovulation induction with and without GnRH-a and found no significant differences in preembryo survival (71.6% versus 71.5%), PR per transfer (32.8% versus 24.7%) and in implantation rate (12% versus 16%, respectivelyf. Both groups employed a midluteallong GnRH -a down regulation but differed in the mode of ovulation induction: our group employed hMG only and the Norfolk group a combination of FSH/hMG. The fact that both groups employing midluteal down regulation demonstrated similar good results after cryopreservation of pre-embryos and embryos strengthens the
GnRH-a 's effect on frozen-thawed embryos
Fertility and Sterility
assumption that early follicular GnRH-a treatment may have untoward effects on oocytes. In conclusion, this work demonstrates that treatment with GnRHa is not associated with a decrease in the quality of frozen-thawed embryos and does not adversely affect PR or its outcome. Because the PR after transfer of fresh embryos has not been shown to be adversely affected by the use of GnRH -a, the extra embryos produced and subsequently cryopreserved present an opportunity to increase the PR per aspiration cycle after the transfer of frozen-thawed embryos. However, the present study is a retrospective analysis of data and is subjected to the effect of confounding factors. Therefore, only a prospective matched double-blind study will be able to clarify the subject objectively.
4.
5.
6.
7.
8.
REFERENCES 1. Neveu S, Hedon B, Bringer J, Chinchole J-M, Arnal F, Humeau C, et a!. Ovarian stimulation by a combination of gonadotropin-releasing hormone agonist and gonadotropins for in vitro fertilization. Fertil Steril 1987;4 7:639-43. 2. Frydman R, Parneix I, Bellaisch-Allart JC, Forman R, Hazout A, Testart J. LHRH agonists in IVF: different methods of utilization and comparison with previous ovulation stimulation treatments. Hum Reprod 1988;3:559-61. 3. Abdalla HI, Ahuja KK, Leonard T, Morris NN, Honour JW, Jacobs HS. Comparative trial ofluteinizing hormone-releas-
Vol. 59, No.5, May 1993
9.
10.
ing hormone analog/human menopausal gonadotropin and clomiphene citrate/human menopausal gonadotropins in an assisted conception program. Fertil Steril 1990;53:4 73-8. Testart J, Forman R, Bellaisch-Allart J, Volante M, Hazout A, Sturrb N, eta!. Embryo quality and uterine receptivity in in vitro fertilization cycles with or without agonists of gonadotrophin releasing hormone. Hum Reprod 1989;4:198201. Lornage J, Boulieu D, Mathieu C, Guerin JF, Pinatel MC, James R, eta!. Transfer of frozen-thawed human embryos in cycles stimulated by HMG. Hum Reprod 1990;5:60-5. Keenan D, Cohen J, Suzman M, Wright G, KortH, Massey J. Stimulation cycles suppressed with gonadotropin-releasing hormone analog yield accelerated embryos. Fertil Steril 1991;55:792-6. Porter RN, Smith WI, Craft IL, Abdulwahid NA, Jacobs HS. Induction of ovulation for in vitro fertilization using buserelin and gonadotropins. Lancet 1984;2:1284-5. Bolton NV, Hawes MS, Taylor TC, Parsons HJ. Development of spare human preimplantation embryos in vitro: an analysis of the correlations among gross morphology cleavage rate and development to the blastocyst. J In Vitro Fert Embryo Transf 1989;6:30-5. Mio Y, Sekijima A, Iwabe T, Onohara Y, Harada T, Terakawa N. Subtle rise in serum progesterone during the follicular phase as a predictor of the outcome of in vitro fertilization. Fertil Steril 1992;58: 159-66. Oehninger S, Toner JP, Veeck LL, Brzyski RG, Acosta AA, Muasher SJ. Performance of cryopreserved pre-embryos obtained in vitro fertilization cycles with or without a gonadotropin-releasing hormone agonist. Fertil Steril 1992;5 7: 620-5.
Benshushan et al.
Effect of GnRH-a on frozen-thawed embryos
1069
FERTILITY AND STERILITY
Printed on acid-free paper in U.S.A.
Copyright e 1993 The American Fertility Society
The effect of gonadotropin-releasing hormone agonist on embryo quality and pregnancy rate following cryopreservation*
Abraham Benshushan, M.D.t YossefEzra, M.D. Alex Simon, M.D.
Nathan Mordel, M.D. Aby Lewin, M.D. Neri Laufer, M.D.
In Vitro Fertilization Unit, Department of Obstetrics and Gynecology, Hadassah University Hospital, Ein-Kerem, Jerusalem, Israel
Objective: To examine the effect of GnRH analogue (GnRH-a) on the quality of frozen-thawed embryos and the pregnancy rate (PR) resulting from transfer. Design: A retrospective study of two groups of women undergoing transfers of frozen-thawed embryos. In group 1 ovulation induction with hMG was begun after hypophyseal desensitization with GnRH-a for 2 weeks. In group 2 ovulation was induced with hMG only. The freezing and thawing techniques were identical for both groups. Setting: In vitro fertilization unit at a university hospitaL Patients: The study group (group 1) included 108 women who underwent 137 transfer cycles of frozen-thawed embryos, and the control group (group 2) included 44 women in 51 cycles of thawed ETs. Interventions: Ovum pick-up and ET techniques were the same for both groups. Methods of embryo freezing and thawing were identical, as were the morphological criteria for grading the embryos. Results: The morphology of embryos was similar in both groups, as was the number of embryos that had at least 50% intact blastomeres (83% ± 23% and 78% ± 30% for group 1 and group 2, respectively). The PR (16 of 137 [1L7%] and 6 of 51 [11.8%], respectively) as well as the abortion rate (30%) were similar for both groups. Conclusions: The use of GnRH-a does not affect the quality of embryos nor the pregnancy outcome. Because the yield of frozen embryos per ovum pick-up is higher in cycles stimulated by GnRH-a/hMG, the PR per pick-up cycle is thus anticipated to be higher. Fertil Steril 1993;59:1065-9 Key Words: Frozen-thawed embryos, GnRH analogue, in vitro fertilization, pregnancy rate, embryo quality
Controlled ovarian hyperstimulation with menotropins after medical hypophysectomy induced by GnRH agonist (GnRH -a) has become a routine treatment in most IVF programs. Higher pregnancy rates (PRs), a decreased incidence of cycle cancel-
Received October 19, 1992; revised and accepted January 28, 1993. * Presented at the Annual Meeting of The Israeli Society of Fertility, Tel Aviv, Israel, March 3 to 4, 1992. t Reprint requests: Abraham Benshushan, M.D., Department of Obstetrics and Gynecology, Hadassah University Hospital, Post Office Box 12000, Jerusalem 91120, Israel. Vol. 59, No.5, May 1993
lation, and an increase in the number of oocytes retrieved and in the number of patients proceeding to ET have been reported with this stimulation regimen (1, 2). In addition, this combination of medications was shown to result in a lower incidence of spontaneous abortion, suggesting better embryo quality or endometrial receptivity (3). Recently, concern has been raised regarding the possible adverse effects of GnRH -a pretreatment on the survival and quality of frozen-thawed-embryos and on the PR resulting from transfer of frozenthawed embryos (4-6). Testart et al. (4) suggested that this untoward effect may be the result of a
Benshushan et al.
Effect of GnRH-a on frozen-thawed embryos
1065
mobilization of defective follicles and oocytes that produce embryos less able to resist freezing-thawing, i.e., those demonstrating irregular blastomeres or cytoplasmic exudates. Lornage et al. (5) have reported a rise in the percentage of embryos with <50% intact blastomeres when using the leuprolide acetate-hMG protocol. An alternative explanation was suggested by Keenan et al. (6). These authors demonstrated that the development of embryos after suppression with GnRH -a was accelerated. This phenomenon interfered with the cryoprotectant qualities of propanediol, which was found to be maximally effective when the cleaving embryo is in interphase. When embryo development is accelerated, cell cycle becomes increasingly disorganized, blastomeres divide asynchronously, resulting in a reduced cryoprotective effect and diminished survival of frozen embryos. Given the important role of cryopreservation in IVF, any demonstrable adverse effect ofGnRH-a on the efficiency of this process may have major implications on its use in ovulation induction for assisted reproductive techniques. This study was therefore undertaken to evaluate the effect of GnRH -a pretreatment on the quality and the PR of frozen-thawed embryos in menotropin-stimulated cycles. MATERIALS AND METHODS Patient Population
Patients undergoing IVF with frozen-thawed embryos at the Hadassah University Hospital between 1987 and 1991 were included in this retrospective study. Group 1 consisted of 108 women who underwent 137 transfer cycles of frozen-thawed embryos. These embryos were obtained from 108 stimulation cycles using GnRH-a and hMG (Pergonal; Ikafarm, Ramat-Gan, Israel). Group 2 consisted of 44 women who had 51 transfers of frozen-thawed embryos from 44 hMG-stimulated cycles. Stimulation Protocols
Follicular stimulation in Group 1 was achieved using a combination of GnRH -a (D-Trp6 , Decapeptyl; Ferring Pharmaceuticals, Kiel, Germany) and hMG as previously described (7). Briefly, GnRH-a was begun at the midluteal phase of the preceding cycle (0.5 mg SC) and continued until the day of hCG administration. Three ampules of hMG were initiated on the 7th day of menses and were 1066
Benshushan et al.
continued in an individualized fashion. In group 2, hMG was started on cycle day 3 (3 ampules per day) and continued for 4 days. Starting cycle day 7 hMG dosage was adjusted depending on serum E 2 levels and follicular development. Ovulation was triggered with hCG (10,000 U IM, Chorigon; Ikafarm) when at least three follicles attained a mean diameter of 16 mm and serum E 2 levels were >500 pg/mL (conversion factor to SI unit, 3.671). In group 1, administration of Decapeptyl was discontinued at this time. Oocyte retrieval was performed under general anesthesia using a transvaginal ultrasound-guided technique 34 to 36 hours after hCG administration. Embryo Freezing
Embryos were frozen 2 to 3 days after in semination at the two- to eight-cell stages. The embryos were graded by the morphological criteria described previously by Bolton et al. (8): grade A, regular, spherical blastomeres with no extracellular fragmentation; grade B, regular, spherical blastomeres with some extracellular fragmentation; grade C, blastomeres slightly irregular in size and shape with considerable extracellular fragmentation; grade D, barely defined blastomeres with considerable extracellular fragmentation. Embryos were incubated for 15 minutes in phosphate-buffered saline (PBS) supplemented with 1.5 M propanediol, 20% heat-inactivated human maternal serum, and 0.4% human serum albumin (HSA). Subsequently, they were placed into a plastic cryotube (3-63401; Nunc, Roskilde, Denmark) containing the above medium supplemented with 0.1 M sucrose. Freezing was performed using a computerized freezer (Planer Biomed Kryo 10-16, Middlesex, London, United Kingdom) with manual seeding. Embryos were stored in liquid nitrogen at -196°C for 1 to 54 months until thawing. Embryo Thawing
On the day of transfer, the cryotube was placed at room temperature for 1 minute and then in warm water (35°C to 37°C) for another minute until complete thawing ofthe medium was observed. The embryos were incubated for 5 minutes in PBS medium containing PROH (1.5 M), sucrose (0.2 M), 20% human serum, and 0.4% HSA. The embryos were incubated in serially decreasing concentrations of propanediol (1.0 M, 0.5 M, 0 M) for 5 minutes at each concentration. The embryos were rinsed three times in PBS and HSA and then incubated in cleavage medium (Ham's F-10; Flow Laboratories, Rickmansworth, Herts, United Kingdom) with 20% pre-
Effect of GnRH-a on frozen-thawed embryos
Fertility and Sterility
Table 1 Distribution of the Indications for IVF in the Study Group
Table 2
Clinical Parameters During Ovulation Induction Group 1
Group 2 (n = 41)
Group 1 (n = 108) %
Indications Tubal factor Unexplained Male factor Endometriosis Ovum donation
67 16.8 9.5 0.7 5.8
68 15.7 7.8 3.9 3.9
ovulatory serum at 37°C in an atmosphere of 5% C0 2 for 0.5 to 4 hours until transfer. Immediately before transfer, the embryos were placed in a medium containing 90% human serum. Transfer of Frozen-Thawed Embryos
Endometrial preparation for ET was achieved by one of the following protocols: spontaneous cycles were used in 67.5% of group 1 and in 62.7% of group 2 transfers; cycles regulated by micronized estrogen (Estrofem; Novo Industry A/S, Copenhagen, Denmark) were used in 20.4% and 23.5% of cycles, respectively; cycles stimulated by clomiphene citrate (CC) were used in 9.5% and 7.7% of cycles, respectively; and finally, cycles stimulated by hMG were used in 2.9% and 3.9% of cycles, respectively. The selection criteria for ovulation induction agents were the following: for normal ovulatory patients we have used spontaneous cycles; for oligo-ovulatory patients either CC or menotropins were used to induce ovulation; and when thawed embryos were transferred in an ovum donation program, artificial cycles were used. Cycles were canceled when they were thought to be suboptimal, i.e., maximal E 2 levels were <200 pg/mL (conversion factor to SI unit, 3.671), P levels <3 ng/mL (conversion factor to SI unit, 3.18) on the day after LH surge, or endometrial thickness of <10 mm on the day of ET. The embryos were transferred 48 to 72 hours after the detection of a spontaneous LH surge or hCG injection (10,000 IU). Determination of the occurrence of the LH surge was established by Clearplain kits for detection of urinary LH (Unipath Limited, Bedford, United Kingdom). A maximum of four fresh or frozen-thawed embryos were usually transferred. Our policy regarding fresh ET is to transfer the morphologically best embryos immediately and to freeze all other embryos graded A to D (according to the criteria of Bolton et al. [8]). However, morphological criteria to determine embryo survival are Vol. 59, No.5, May 1993
Age HMG dose:j: Peak E 2§ No. of oocytes retrieved No. of embryos produced Oocyte fertilized (%)
33.6 32.7 2281 14.5
± ± ± ±
4* 12.4 1067 6.2
Group 2 32.4 26.2 1828 10.9
± ± ± ±
Probability
3.8 11.7 817 4.3
NSt 0.01 0.02 0.001
9.7 ± 4.0
8.2 ± 2.8
0.014
70.6
77.4
0.036
*Values are means± SD. t NS, not significant. :j: Number of hMG ampules needed for adequate response. §Values are in pg/mL, conversion factor to SI unit, 3.671.
controversial and can be misleading; therefore, we usually transfer only frozen-thawed embryos with >50% intact blastomeres. Statistical Analysis
Results are presented as means± SD. Statistical evaluation was performed using Student's t-test and X2 test whenever appropriate.
RESULTS
Oocyte retrieval was performed in 108 women treated with GnRH-a/hMG (group 1) and 44 women treated with hMG alone (group 2). The average age of the women in both groups was not significantly different (33.6 ± 4.0 versus 32.4 ± 3.8), nor were there differences between the two groups in the indications for IVF (Table 1). The mean number of hMG ampules per cycle, peak serum E 2 levels, as well as the average number of oocytes retrieved, and embryos obtained were significantly higher in group 1. However, the proportion of oocytes that fertilized was greater in the group that received hMG only (Table 2). The morphological grading of the embryos at the time of freezing was not different between the two groups (Table 3). Similarly, the duration of freezing,
Table 3
Embryo Quality at Time of Freezing Group 1
Grade A Grade B Grade C GradeD
50 69 14 4
(36.2)* (50.4) (10.5) (2.9)
Group 2 22 (43.1) 24 (47.1) 5 (9.8)
* Values in parentheses are percents.
Benshushan et al.
Effect of GnRH-a on frozen-thawed embryos
1067
the age of the embryos at the time of freezing, and the percent of embryos with >50% intact blastomeres after thawing was not different between the two groups (Table 4). Finally, the number of thawed embryos transferred was similar for both groups (2.9 ± 2.3 and 2.5 ± 1.2, respectively). The PR after transfer of frozen-thawed embryos obtained from GnRH-a/hMG stimulation (group 1) was 11.7% (16/ 137). This rate was not significantly different from the PRof 11.8% (6/51) obtained from embryos cryopreserved after hMG stimulation alone (group 2). Ofthe pregnancies achieved in group 1, 12 are either delivered or ongoing (>20 weeks' gestation), and 4 ended in spontaneous abortions. In group 2, 4 pregnancies are either delivered or ongoing and 2 ended in abortion. DISCUSSION
This work demonstrates that the use of GnRH -a pretreatment did not adversely affect embryo quality before freezing and did not negatively interfere with their ability to withstand cryopreservation as evidenced by their morphology after thawing. Moreover, the capacity of these embryos to continue and develop in the uterus was not influenced by the analogue and resulted in a PR of 11.7%, which was comparable with that achieved after the transfer of embryos derived from menotropin stimulation only (11.8%). These results are in sharp contrast to the findings of Testart et al. (4), Lornage et al. (5), and Keenan et al. (6) who found an adverse effect of GnRH-a on embryo survival after cryopreservation. Testart et al. (4) demonstrated that embryos obtained from IVF cycles using GnRH-a were significantly more susceptible to the freezing and thawing procedures and that, consequently, the proportion of embryos surviving cryopreservation was higher in cycles without GnRH-a. They have also reported a significantly lower PR after the transfer of cryopreserved embryos obtained from cycles with GnRH-a as compared with those without GnRH-a (2.8% and 8.9%, respectively). It should be noted that Testart et al. (4) have used five different ovulation induction protocols for IVF patients. To minimize the effect of the different protocols on the results, we have compared our results only to the relevant protocols that were similar to ours, namely, long GnRH-a/ hMG as compared either with hMG alone or CChMG. Lornage et al. (5) found that the survival of embryos after cryopreservation was significantly lower after a long GnRH -a protocol when compared 1068
Benshushan et al.
Table 4
Characteristics of Frozen-Thawed Embryos* Group 1
Duration of freezing (mo) 7 Age of embryos at freezing (d) 2.67 ET day from LH surge 2.57 No. of embryos transferred 2.89 Embryos with >50% blastomeres (%) 83.7
*P
=
Group 2
± 8t 7.9 ± 11.1 ± 0.48 2.62 ± 0.6 ± 0.56 2.6 ± 0.51 ± 1.29 2.47 ± 1.18 ± 23.5 78.3 ± 30
not significant for all values.
t Values are means± SD.
with that of embryos obtained from CC-hMG-stimulated cycles (52.2% and 66.6%, respectively). However, no similar effect on PR was noted (16% and 14.5%, respectively). The differences between our results and those of Testart et al. (4) and Lornage et al. (5) may stem from the different protocols used for down regulation that, in turn, may have affected oocyte and embryo quality. Both groups used a long 14-day pituitary down regulation but, in contrast to our group, started at the commencement of cycle and not in the midluteal phase. Consequently, the initial flare-up of gonadotropins in the beginning of the follicular phase may have prematurely exposed the oocytes to LH and P, resulting in a higher proportion of oocytes and embryos of reduced quality (9). We suggest that such embryos with reduced quality that were generated from cycles with premature P secretion are more vulnerable to cryopreservation. On the other hand, our results are in agreement with those of Keenan et al. (6) who used similar groups for comparison. They reported PRs of 29% and 14% per frozen-thawed ET generated from conventional cycles or down-regulated cycles, respectively. However, these rates did not differ significantly because of the small samples used for comparison. In addition, our observations obtained after the transfer of cryopreserved embryos, corroborate a previous recent report by Oehninger et al. (10) who evaluated the results of cryopreserved preembryos. They compared the performance of preembryos after ovulation induction with and without GnRH-a and found no significant differences in preembryo survival (71.6% versus 71.5%), PR per transfer (32.8% versus 24.7%) and in implantation rate (12% versus 16%, respectivelyf. Both groups employed a midluteallong GnRH -a down regulation but differed in the mode of ovulation induction: our group employed hMG only and the Norfolk group a combination of FSH/hMG. The fact that both groups employing midluteal down regulation demonstrated similar good results after cryopreservation of pre-embryos and embryos strengthens the
GnRH-a 's effect on frozen-thawed embryos
Fertility and Sterility
assumption that early follicular GnRH-a treatment may have untoward effects on oocytes. In conclusion, this work demonstrates that treatment with GnRHa is not associated with a decrease in the quality of frozen-thawed embryos and does not adversely affect PR or its outcome. Because the PR after transfer of fresh embryos has not been shown to be adversely affected by the use of GnRH -a, the extra embryos produced and subsequently cryopreserved present an opportunity to increase the PR per aspiration cycle after the transfer of frozen-thawed embryos. However, the present study is a retrospective analysis of data and is subjected to the effect of confounding factors. Therefore, only a prospective matched double-blind study will be able to clarify the subject objectively.
4.
5.
6.
7.
8.
REFERENCES 1. Neveu S, Hedon B, Bringer J, Chinchole J-M, Arnal F, Humeau C, et a!. Ovarian stimulation by a combination of gonadotropin-releasing hormone agonist and gonadotropins for in vitro fertilization. Fertil Steril 1987;4 7:639-43. 2. Frydman R, Parneix I, Bellaisch-Allart JC, Forman R, Hazout A, Testart J. LHRH agonists in IVF: different methods of utilization and comparison with previous ovulation stimulation treatments. Hum Reprod 1988;3:559-61. 3. Abdalla HI, Ahuja KK, Leonard T, Morris NN, Honour JW, Jacobs HS. Comparative trial ofluteinizing hormone-releas-
Vol. 59, No.5, May 1993
9.
10.
ing hormone analog/human menopausal gonadotropin and clomiphene citrate/human menopausal gonadotropins in an assisted conception program. Fertil Steril 1990;53:4 73-8. Testart J, Forman R, Bellaisch-Allart J, Volante M, Hazout A, Sturrb N, eta!. Embryo quality and uterine receptivity in in vitro fertilization cycles with or without agonists of gonadotrophin releasing hormone. Hum Reprod 1989;4:198201. Lornage J, Boulieu D, Mathieu C, Guerin JF, Pinatel MC, James R, eta!. Transfer of frozen-thawed human embryos in cycles stimulated by HMG. Hum Reprod 1990;5:60-5. Keenan D, Cohen J, Suzman M, Wright G, KortH, Massey J. Stimulation cycles suppressed with gonadotropin-releasing hormone analog yield accelerated embryos. Fertil Steril 1991;55:792-6. Porter RN, Smith WI, Craft IL, Abdulwahid NA, Jacobs HS. Induction of ovulation for in vitro fertilization using buserelin and gonadotropins. Lancet 1984;2:1284-5. Bolton NV, Hawes MS, Taylor TC, Parsons HJ. Development of spare human preimplantation embryos in vitro: an analysis of the correlations among gross morphology cleavage rate and development to the blastocyst. J In Vitro Fert Embryo Transf 1989;6:30-5. Mio Y, Sekijima A, Iwabe T, Onohara Y, Harada T, Terakawa N. Subtle rise in serum progesterone during the follicular phase as a predictor of the outcome of in vitro fertilization. Fertil Steril 1992;58: 159-66. Oehninger S, Toner JP, Veeck LL, Brzyski RG, Acosta AA, Muasher SJ. Performance of cryopreserved pre-embryos obtained in vitro fertilization cycles with or without a gonadotropin-releasing hormone agonist. Fertil Steril 1992;5 7: 620-5.
Benshushan et al.
Effect of GnRH-a on frozen-thawed embryos
1069