Blastocyst-stage embryo transfer in patients who failed to conceive in three or more day 2 3 embryo transfer cycles: a prospective, randomized study

Transcription

1 FERTILITY AND STERILITY VOL. 81, NO. 3, MARCH 2004 Copyright 2004 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A. Blastocyst-stage embryo transfer in patients who failed to conceive in three or more day 2 3 embryo transfer cycles: a prospective, randomized study Eliahu Levitas, M.D., Eitan Lunenfeld, M.D., Iris Har-Vardi, Ph.D., Sarit Albotiano, M.S., Yael Sonin, M.S., Rinat Hackmon-Ram, M.D., and Gad Potashnik, M.D. Fertility and IVF Unit, Department of Obstetrics and Gynecology, Soroka University Medical Center, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel Objective: To compare blastocyst-stage embryo transfers (ETs) with day 2 3 ETs in patients who failed to conceive in three or more day 2 3 IVF/ET cycles. Design: Prospective, randomized. Setting: Fertility unit in a university medical center. Patient(s): Fifty-four patients with an adequate ovarian response underwent oocyte retrievals. The patients were prospectively and randomly divided into blastocyst-stage and day 2 3 ET groups. Intervention(s): Ovarian down-regulation was obtained using GnRH agonist, and controlled ovarian hyperstimulation was achieved using daily administration of gonadotropins. Main Outcome Measure(s): The rate of blastocyst formation, ET cancellations, pregnancies, implantation, multiple gestation, and live births. Result(s): The clinical pregnancy rates per oocyte retrieval were 21.7% and 12.9% per blastocyst and day 2 3 ETs, respectively. Although there was a significantly higher implantation rate for blastocyst embryos (21.2%) as compared with 48- to 72-hour embryos (6%), the multiple-pregnancy rate was not significantly different between both groups. An ET cancellation rate of 26% and 6.4% for blastocyst and day 2 3 ETs, respectively, was observed. The presence of two or more 8-cell embryos on day 3 in culture carried a high probability of obtaining blastocysts for transfer. Conclusion(s): This prospective randomized study suggests that in patients with an adequate ovarian response who failed to conceive in at least three IVF/ET cycles [1] transfer of blastocyst-stage embryos carries a significantly higher implantation rate; [2] the pregnancy rate per oocyte retrieval and ET are higher in the blastocyst-stage group, even if it did not reach statistical significance; [3] a higher ET cancellation rate was observed in the whole blastocyst-stage group; [4] the ET cancellation rate was reduced significantly if the decision to proceed to blastocyst transfer was made on day 3 after oocyte retrieval, which is a post hoc conclusion. (Fertil Steril 2004;81: by American Society for Reproductive Medicine.) Key Words: Failed IVF/ET cycles, blastocyst-stage ET Received October 28, 2002; revised and accepted August 7, Reprint requests: Eliahu Levitas, M.D., IVF Unit, Department of Obstetrics and Gynecology, Soroka University Medical Center, P.O. Box 151, Beer Sheva 84101, Israel (FAX: ; loits@ bgumail.bgu.ac.il) /04/$30.00 doi: /j.fertnstert Women with multiple IVF failures are a negatively selected group of patients and present a clinical challenge. Among this bad prognosis patient group, the observed decrease in the pregnancy rate (1) in successive IVF attempts is considered a factor against further IVF trials. The reason for such repeated IVF failures may reside at the level of the endometrium or at the level of embryo quality because of nonoptimal embryo selection for transfer. Since improved embryo culture techniques are available, it is possible to grow embryos to the blastocyst stage and to transfer fewer but higher-quality embryos at the blastocyst stage. Cruz et al. (2), in a retrospective cohort study, stated that blastocyst-stage embryo transfer (ET) is of benefit to patients with multiple IVF failures. The purpose of the present study was to examine prospectively and randomly the possibility that transferring embryos at the blastocyst stage, instead of at the cleavage stage, will have a beneficial effect 567

2 on the IVF outcome in patients with multiple failed treatment attempts. MATERIALS AND METHODS Patients and Design All patients fulfilling the entry criteria were offered enrollment, and 54 couples with primary or secondary infertility agreed to enter the study. According to the study design, each couple underwent one treatment cycle. They were characterized by failure to achieve pregnancy during at least three IVF/ET cycles, despite an acceptable ovarian response and fertilization rate in previous cycles. This prospective randomized study was approved by the Institutional Review Board at the Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel. The study included female patients aged 37 years who were being treated mainly for tubal or male factor infertility, who had evidence of a normal uterine cavity, and who had no contraindications to pregnancy. We excluded patients with poor ovarian response on previous IVF cycles (peak E 2 level below 500 pg/ml or retrieval of fewer than three oocytes) and patients with ETs from donor oocytes or frozen-thawed embryos. Blind randomization with sealed and opaque envelopes was performed immediately after informed consent was signed and before treatment initiation. The patients were randomly divided into a blastocyst transfer group and a day 2 3 ET group. The blastocysts were transferred on day 5, 6, or 7 according to the day of expanded blastocysts appearance, while day 2 or 3 transfers were performed according to scheduling issues and weekend avoidance. Randomization was performed according to a computer-generated random-number table. Ovarian Stimulation Protocols All women participating in the study were treated with GnRH agonist (GnRHa). The standard treatment regimen was according to the long protocol, and ovarian downregulation was achieved by administration of controlledrelease GnRHa 3.75 mg (Decapeptyl CR 3.75 mg, Ferring Pharmaceutical GmbH, Kiel, Germany) at the midluteal phase of the preceding cycle. Serum E 2 levels below 50 pg/ml days after GnRHa injection were used to define ovarian quiescence. In women with a reduced ovarian response to gonadotropin stimulation, the short protocol was administered, using daily injections of GnRHa 0.1 mg (Decapeptyl 0.1 mg; Ferring Pharmaceutical GmbH, Kiel, Germany) commenced at the first day of the menstrual cycle, concomitantly with daily gonadotropin injections. Controlled ovarian hyperstimulation was performed using hmg (Pergonal; Teva, Ramat-Gan, Israel) according to an individually adjusted technique monitored by serum E 2 and transvaginal ovarian ultrasound. HCG (Chorigon; Teva) 10,000 U was injected IM when serum E 2 levels were at least 500 pg/ml and at least two follicles 15 mm in diameter were observed. Transvaginal ultrasound-guided ovum retrieval was performed under general anesthesia hours after hcg administration. According to semen quality on the day of oocyte retrieval, the oocytes were inseminated or subjected to intracytoplasmic sperm injection (ICSI). Embryo transfer for the day 2 3 group of patients was carried out using embryos with the highest number of blastomeres and having the highest embryo grading score. We applied a 5-grade embryo scoring system according to the amount of embryonic fragmentations and the size and shape of blastomeres. Blastocyst transfer was performed using 5 7 day embryos cultured according to the sequential media system: the first 72 hours of culture in G1.2 medium (IVF Science, Scandinavia) followed by G2.2 medium (Scandinavia IVF Science, Gothenburg, Sweden) to day 5 7. Luteal phase was supported by five injections of hcg 1,250 U every other day starting 48 hours after oocyte retrieval or daily IM administration of 50 mg P (Gestone, Paines & Byrne Limited, West Byfleet, Surrey, UK) in patients at high risk for developing ovarian hyperstimulation syndrome (peak E 2 levels 2,000 pg/ml) or with combined luteal support, adding four injections of hcg 1,250 U every other day in P-supported cycles in which the serum E 2 and P levels dropped sharply 7 days after ET. In all patients, serum -hcg was obtained days after ET and pregnancies were confirmed by the presence of a pregnancy sac and cardiac activity on ultrasound. Statistical Analysis Categorical data were analyzed with the 2 or Fisher s exact test. Continuous data were analyzed with Student s t-test. P.05 was considered statistically significant. RESULTS The study was designed to include 100 patients who failed to conceive during at least three IVF-ET cycles. However, because of difficulties in recruiting patients for randomization, the study was ceased after 54 cases were included. Thirty-one patients were randomly assigned for day 2 3 ET, and 23 were assigned for blastocyst-stage transfer. The unequal number of patients included in the groups was due to randomization among 100 sealed envelopes. In terms of the number of embryos replaced per cycle, implantation rates, and pregnancy rates, there was no difference between the 17 day 2 ETs versus the 12 day 3 ETs, and since these findings are supported by other investigators (3, 4), we combined day 2 and 3 ETs into one group. The long protocol was applied in 93.1% of day 2 3 ET and 94.1% of blastocyst transfer patients. As shown in Table 1, the mean ( SD) age of patients was and years, duration of infertility was and years, day 3 FSH levels were and , and 568 Levitas et al. Blastocyst transfer in IVF/ET failures Vol. 81, No. 3, March 2004

3 TABLE 1 Patient characteristics for day 2 3 versus blastocyststage ETs. Characteristics Day 2 3 ET Blastocyst ET No. of patients Age (y) FSH level (day 3) (miu/ml) Duration of infertility (y) No. of previous IVF/ET cycles Male infertility (%) Tubal factor (%) Note: Values are expressed as mean SD. No statistically significant differences were noted between the day 2 3 and blastocyst transfer patients for any of the variables tested. number of previous IVF/ET attempts was and for the day 2 3 ET and blastocyst group patients, respectively. None of the above-mentioned differences reached statistical significance. The main causes of infertility were male and tubal factor infertility with or without pelvic adhesions, which were similarly distributed in both groups. All 54 women underwent oocyte retrieval, but ET occurred in only 46 patients: 29 of 31 (93.5 %) in the day 2 3 ET group compared with 17 of 23 (73.9%) in the blastocyststage ET group (P.05 vs. the day 2 3 group). No statistically significant differences were found between the groups concerning the following parameters: type of pituitary downregulation and ovarian hyperstimulation protocol, mean number of ovarian follicles and mean endometrial thickness observed by ultrasound, peak E 2 level on the hcg administration day, mean number of oocytes collected per cycle, or mean number of cycles using regular oocyte insemination or ICSI. The decision to perform ICSI or regular oocyte insemination was made according to semen parameters on the day of oocyte retrieval. Furthermore, the number of oocytes collected and fertilization rates were similar in both groups. The number of two pronuclear (2PN) zygotes obtained was and in the day 2 3 ET group and blastocyst group, respectively. The differences regarding the type of luteal phase support after ET, that is, hcg, P, or combined, were not found to be statistically significant. In particular, among the 29 day 2 3 ET patients the distribution of luteal support was hcg in 12 (41.3%) patients, with one pregnancy; P in 3 (10.3%) patients, with no pregnancy; and combined support in 14 (48.2%) patients, with three pregnancies. The luteal phase in the 17 blastocyst group patients was supported by hcg in 5 (29.4%) patients, with two pregnancies; P in 3 (17.3%) patients, with two pregnancies; and combined support in 9 (52.9%) patients, with one pregnancy. We transferred 100 embryos with a mean of 3.4 embryos per transfer and obtained four clinical pregnancies among day 2 3 ET women with a clinical pregnancy rate (PR) per oocyte retrieval of 12.9% (4/31) and a clinical PR per transfer of 13.8% (4/29). The calculated implantation rate was 6% (6/100). Three of four were twin pregnancies, one of them ended with first trimester abortion, and the other two with a live birth of twins. The singleton pregnancy ended with a live birth. Sixty-five of 151 (43%) embryos in the blastocyst group developed into blastocysts and 33 were transferred in 17 blastocyst group patients. In 11 cases, blastocyst formation occurred and transfer was performed on day 5 after ovum pickup with 2 0 (22/11) blastocysts per transfer; in five cases, transfer was performed on day 6 with (9/5) blastocysts per ET; and in one case two blastocysts were observed and transferred on day 7. Most of the remaining blastocysts were not suitable for freezing. Pregnancy was achieved in five of 17 ETs performed in the blastocyst group women, with a clinical PR per oocyte retrieval of 21.7% (5/23, P.05 vs. day 2 3 group), and a clinical PR per transfer of 29.4% (5/17, P.05 vs. day 2 3 group). There were three pregnancies among day 5 and two pregnancies among day 6 transfers. Seven gestational sacs with eight live fetuses resulted, and the calculated blastocyst implantation rate was 21.2% versus 6% in the day 2 3 ET group (P.01). Thus we achieved a 40% multiple-pregnancy rate (2/5): one woman became pregnant with twins, which ended with a live birth of twins, and one became pregnant with triplets, which, because of monozygotic twinning, ended with a first trimester abortion. Two of three singleton pregnancies ended with live births (Table 2). Embryo transfer was cancelled in two of 31 (6.4%) cases in the day 2 3 group because of arrest at the 2PN zygote stage, and in six cases among the 23 (26%) blastocyst group patients. The difference in the cancellation rates did not reach statistical significance, while the reasons for ET cancellation in the blastocyst group were 2PN zygote stage arrest in two cases and arrest of embryo cleavage in four cases. The relationship between presence of 8-cell embryos on day 3 in culture and blastocyst development is illustrated in Table 3. In 15 of 17 (88.2%) blastocyst transfer cycle patients, at least one good-quality 8-cell embryo on day 3 in culture was observed, while in eight of 17 (47%) cases, two or more 8-cell embryos were present. The mean score of the 8-cell embryos was 2.7 (top score, 5). In two of 17 (11.8%) cases, blastocysts developed despite the absence of an 8-cell embryo on day 3 in culture, while the cleavage-stage embryo score was higher for those who did versus those who did not develop blastocysts (mean score of 2.5 vs. 1.5, P.05). Three of five pregnancies in the blastocyst group occurred in cycles with just one or no (one pregnancy) 8-cell embryo on FERTILITY & STERILITY 569

4 TABLE 2 Cycle characteristics and outcome for day 2 3 versus blastocyst ETs. Characteristics Day 2 3 ET Blastocyst ET No. of retrievals No. of ETs ET cancellation rate (%) 6.4 (2/31) 26 (6/23) No. of ETs per cycle E 2 level on hcg day (pg/ml) 2,170 1,028 2, No. of follicles 15mm Endometrial thickness (mm) No. of oocytes collected No. of zygotes Fertilization rate (%) Blastocyst formation rate (%) 43 (65/151) ICSI cases (%) Clinical pregnancy rate per 12.9 (4/31) 21.7 (5/23) retrieval (%) Clinical pregnancy rate per (4/29) 29.4 (5/17) transfer (%) Implantation rate (%) 6 (6/100) 21.2 (7/33) a Multiple pregnancy rate (%) 75 (3/4) 40 (2/5) Live births rate (%) 75 (3/4) 60 (3/5) Note: Values are expressed as mean SD, and P.05 unless otherwise indicated. a P.01 versus day 2 3 ET. day 3 in culture, and the remaining two pregnancies occurred in cycles with five 8-cell embryos. The comparison of characteristics from cycles where blastocyst formation occurred with those cases that did not develop blastocysts showed a significantly higher peak E 2 level (2, vs. 1, pg/ml, P.03) and more follicles 15 mm in diameter ( vs , P.05) in favor of cycles where blastocysts developed. DISCUSSION To the best of our knowledge, this is the only prospective randomized study in patients with repeated IVF-ET failure in which randomization with treatment group assignment was performed before treatment initiation, immediately after signing informed consent, independent of the index cycle parameters such as ovarian response to stimulation, number of oocytes retrieved, number of oocytes fertilized, or number of 8-cell embryos on day 3 in culture. Although 26% of patients in the blastocyst transfer group failed to obtain a transfer, a trend toward a higher clinical PR per oocyte retrieval and a higher implantation rate among blastocyst transfer patients was observed (21.7% and 21.2% in blastocyst transfer patients vs. 12.9% and 6% in day 2 3 ET patients, respectively). We transferred fewer embryos per transfer in the blastocyst group ( for blastocyst patients and in the day 2 3 group of patients) without compromising the PR; in fact, we improved the PR and reduced the multiple-pregnancy rate among blastocysttreated patients compared with day 2 3 ET patients. Patients included in this study belong to a particular group of women who failed to conceive through more than three IVF/ET attempts despite acceptable ovarian response and fertilization and embryo cleavage rates. In the present study, we admitted for randomization into treatment groups only women aged 37 years with an acceptable ovarian reserve and normal response to gonadotropin stimulation according to previous IVF cycles. We followed the performance in the TABLE 3 Characteristics of blastocyst group embryos on day 3 in culture in cycles that developed blastocysts (ET performed) versus those that did not (ET cancelled). No. of cycles Characteristics on day 3 Present (ET performed) Blastocysts Absent (ET cancelled) Total Blastocyst development rate (%) Pregnancies (n) Zygote stage arrest 2 2 Five-cell embryos only 2 (2.5) 2 (1.5) eight-cell embryo 7 (2.6) 2 (3) eight-cell embryos 2 (2.7) eight-cell embryos 1 (2.7) eight-cell embryos 3 (3.1) eight-cell embryos 1 (2.3) eight-cell embryos 1 (2.6) Total cases Note: Values in parentheses express the mean embryo score (top score, 5). 570 Levitas et al. Blastocyst transfer in IVF/ET failures Vol. 81, No. 3, March 2004

5 two study groups (day 2 3 ET and blastocyst-stage ET) from the very early stages of the index IVF cycle, keeping the inclusion criteria as wide as possible while excluding mainly the poor responders and the frozen-thawed ET cycles. Transferring blastocyst-stage embryos instead of cleavage-stage embryos in an unselected IVF population is accepted by many investigators (5 10), particularly with the purpose of reducing multiple-pregnancy rates by limiting the number of embryos transferred while maintaining, or even improving, the PR. In a retrospective study, Shapiro et al. (11) reported a dramatic decline in implantation and PRs with repeated blastocyst transfers after even one previous failed IVF attempt. Others, such as Cruz et al. (2), have advocated transferring blastocyst-stage embryos in the failed repeated IVF patient population and reported an increase in implantation and PRs in a nonrandomized population with multiple IVF failures. The problem with blastocyst transfer is that a percentage of fertilized eggs will never reach the blastocyst stage. We have compared the cycles in which blastocysts were formed with those without blastocyst development and observed an increased number of follicles above 15 mm in diameter and a significantly higher peak E 2 level for cycles where blastocysts formed versus those lacking blastocyst formation. We assume that proper selection of cases suitable for blastocyst transfer is critical when the purpose is to reduce to a minimum the cycle cancellations. According to our data, the presence of at least one good-quality 8-cell embryo on day 3 in culture will lead to a 77.7% chance of having a blastocyst available for transfer, while the presence of two or more 8-cell embryos will increase the chances to about 100%. However, using the number of 8-cell embryos as a predictor for blastocyst development was defined post hoc, which may undermine the statistical significance of these criteria. Since cycle cancellation carries emotional and monetary costs, a 50% ET cancellation rate in the blastocyst group patients should be assumed when there is no 8-cell embryo on day 3 in culture. Another issue to be considered in relation to blastocyst transfers is monozygotic twinning. Observations of a possible increase in monozygotic twinning related to ovulation induction or blastocyst transfers were previously documented by us and others (12 14). Currently, the real frequency of such a phenomenon is not clear, but the occurrence of a case of monozygotic twinning among the pregnancies in our study is cause for concern. Collectively, the findings of this study indicate that women with multiple IVF failures may benefit from blastocyst transfer by increasing their PR and improving their implantation rate, provided an acceptable ovarian response to gonadotropins and acceptable fertilization and embryo cleavage rates are present. Better embryo selection for transfer and improved endometrial receptivity 5 or 6 instead of 2 or 3 days after oocyte retrieval may have some effect. However, one should keep in mind that aiming for blastocyst transfer without using proper case selection criteria may increase the ET cancellation rate, decrease the embryo cryopreservation rates, and increase facility cost. Therefore, based on our findings in this prospective and randomized study, it is reasonable to advise patients with multiple IVF implantation failures to undergo treatment with blastocyst-stage embryos after proper case selection criteria. References 1. Hershlag A, Kaplan EH, Loy RA, DeCherney AH, Lavy G. Heterogeneity in patient populations explain differences in in vitro fertilization programs. Fertil Steril 1991;56: Cruz JR, Dubey AN, Patel J, Peak D, Hartog B, Gindoff PR. Is blastocyst transfer useful as an alternative treatment for patients with multiple in vitro fertilization failures? Fertil Steril 1999;72: van Os HC, Alberda A, Janssen-Caspers HAB, Leerentveld RA, Scholtes MCW, Zeilmaker GH. The influence of the interval between in vitro fertilization and embryo transfer and some other variables on treatment outcome. Fertil Steril 1989;2: Edwards RG, Fishel SB, Cohen J, Fehilly CB, Purdy JM, Slater JM, et al. Factors influencing the success of in vitro fertilization for alleviating human infertility. J In Vitro Fert Embr Trans 1984;1: Milki AA, Hinckley MD, Fish JF, Dasig D, Behr B. Comparison of blastocyst transfer with day 3 embryo transfer in similar patient populations. Fertil Steril 2000;73: Alper MM, Brinsden P, Fisher R, Wikland M. To blastocyst or not to blastocyst? That is the question. Hum Reprod 2001;16: Balaban B, Urman B, Alatas C, Mercan R, Aksoy S, Isiklar A. Blastocyst-stage transfer of poor-quality embryos results in higher implantation rates. Fertil Steril 2001;75: Schoolcraft WB, Gardner DK, Lane M, Schlenker T, Hamilton F, Meldrum DR. Blastocyst culture and transfer: analysis of results and parameters affecting outcome in two in vitro fertilization programs. Fertil Steril 1999;72: Van der Auwera I, Debrock S, Spiessens C, Afschrift H, Bakelants E, Meuleman C, et al. A prospective randomized study: day 2 versus day 5 embryo transfer. Hum Reprod 2002;17: Wilson M, Hartke K, Kiehl M, Rodgers J, Brabec C, Lyles R. Integration of blastocyst transfer for all patients. Fertil Steril 2002;77: Shapiro BS, Richter KS, Harris DC, Daneshmand ST. Dramatic declines in implantation and pregnancy rates in patients who undergo repeated cycles of in vitro fertilization with blastocyst transfer after one or more failed attempts. Fertil Steril 2001;76: Sheiner E, Har-Vardi I, Potashnik G. The potential association between blastocyst transfer and monozygotic twinning. Fertil Steril 2001;75: Sheiner E, Kivilevitch Z, Levitas E, Sonin Y, Albotiano S, Har-Vardi I. Monozygotic twins following blastocyst transfer: a report of two cases. Eur J Obstet Gynecol Reprod Biol 2001;98: Derom C, Vlietnick R, Derom R, Van den Berghe H, Thiery M. Increased monozygotic twinning rate after ovulation induction. Lancet 1987;9: FERTILITY & STERILITY 571