Highlights of the American Society for Reproductive Medicine 61st Annual Meeting

October 14-19, 2005; Montreal, Quebec

Peter Kovacs, MD


November 07, 2005

This year, the 61st Annual Meeting of the American Society for Reproductive Medicine was held together with the 51st Canadian Fertility and Andrology Society's Annual Meeting in Montreal, Quebec, Canada. This report discusses presentations that focused on single embryo transfer, preimplantation genetic diagnosis, and some topics of ovarian stimulation protocols.

Dr. Andersen (Copenhagen, Denmark)[1] discussed the correlation between the endocrinologic profile with various stimulation protocols and embryo quality. The goal during an in vitro fertilization (IVF) cycle is to create enough good-quality embryos for the fresh transfer and, if possible, for cryopreservation as well. Controlled ovarian stimulation (COH) is used to produce several eggs. After fertilization, the embryos are assessed on a daily basis, and those with symmetric blastomeres and the least amount of fragmentation are chosen for transfer. Several types of medications (eg, recombinant follicle-stimulating hormone [rFSH], urinary human menopausal gonadotropin [HMG], highly purified urinary [hpHMG]), and stimulation protocols (the short antagonist and the more widely used midluteal long gonadotropin-releasing hormone agonist [GnRHa]) are available for administering COH. The results of large randomized studies can guide the decision about medication and stimulation use.

The results of a number of studies show that when compared with natural cycles, stimulation itself does not affect the cleavage potential and the fragmentation rate of the embryos. However, these studies do not provide detailed embryology parameters. Therefore, a large multicenter, randomized study was designed to study the effects of hpHMG vs rFSH on embryo quality and embryo profile (MERIT Study). Stimulation (long protocol, fixed dose of gonadotropin) and transfer policies were standardized. Both on-site and central embryo evaluations were performed, and their outcome correlated well. More oocytes were retrieved in rFSH cycles, and there were more fertilized oocytes in this group. The cumulus appearance, embryo quality, and embryo cleavage rate were comparable. There were more embryos with < 20% fragmentation (high quality) with hpHMG.

There were significant differences in the hormonal profile during stimulation. FSH, estradiol, testosterone, and androstenedione levels were higher on the day of human chorionic gonadotropin (hCG) injection and retrieval in the hpHMG group, whereas progesterone levels were higher in the rFSH group; hCG levels were also higher from Day 6 of stimulation onward with hpHMG. Higher hCG levels were associated with more top-quality embryos and higher ongoing pregnancy rates.

It seems that in cycles with lower androstenedione (precursor of estradiol) levels, there was an increased rate of atresia (more fragmentation, fewer top-quality embryos); whereas in cycles with higher androstenedione levels (and therefore higher estradiol levels), embryo parameters were superior. In addition, higher progesterone levels on the day of hCG injection were associated with a higher incidence of hyperechogenic endometrium, which is unfavorable for implantation. The overall result of this study was a significantly higher implantation rate (41% vs 27%) and pregnancy rate (48% vs 32%) with hpHMG use as compared with rFSH. Higher progesterone levels observed during rFSH stimulation probably alter endometrial receptivity by inducing early luteinization and therefore affect implantation rates.

There are high expectations for the available stimulatory drugs during assisted reproduction: They should be effective, free of side effects, easy to administer, and preferably inexpensive. Various aromatase inhibitors have been evaluated in the past few years in different subgroups of infertile women. Aromatase inhibitors disrupt the androgen-to-estrogen conversion and thereby reduce the central negative feedback. In return, increased amounts of FSH and LH are produced, and the simultaneous growth of several follicles can be induced. As the feedback mechanisms remain intact (unlike with clomiphene citrate [CC]), one does not need to worry about the development of too many follicles. Aromatase inhibitors have a short half-life and are therefore quickly eliminated.

In a randomized trial, Al-Fadhli (Montreal, Canada) and colleagues[2] compared outcomes with 2 different doses of letrozole (2.5 mg vs 5.0 mg) in superovulation and intrauterine insemination (IUI) cycles. Demographic parameters and the length of stimulation were similar for the 2 groups. The number of follicles was higher in the 5-mg group, but the endometrial thickness did not differ between the 2 study arms. The pregnancy rate per cycle was significantly higher in the 5-mg group (26.3% vs 5.9%). There were no multiple pregnancies in either group. The authors concluded that the 5-mg dose performed significantly better.

In certain IVF stimulation protocols, oral and injectable medications are combined (eg, CC plus hMG) to reduce costs. Serafini (Sao Paulo, Brazil) and colleagues[3] assessed whether the combination of letrozole and rFSH was effective during stimulation for IVF. Study participants were randomly assigned to receive letrozole plus rFSH or rFSH only. Premature luteinization was prevented with GnRH antagonist in both groups. There were fewer oocytes and mature oocytes in the letrozole plus gonadotropin group, but the amount of gonadotropins used was about 40% less. The number of embryos transferred was similar, and pregnancy rates were comparable (37.5% vs 36.6%). These investigators observed similar outcome in the fresh cycles with the combination of letrozole and gonadotropins when compared with gonadotropins alone. Medication costs were significantly reduced in the combination group. On the other hand, there were fewer embryos for cryopreservation; therefore, cumulative pregnancy rates (fresh plus frozen) will likely be lower in the letrozole plus gonadotropin group.

Coelho (Springfield, Massachusetts, USA) and colleagues[4] used letrozole among CC-resistant women with polycystic ovarian syndrome (PCOS) undergoing ovulation induction. A daily dose of 2.5 mg was administered from Day 3 to Day 7 of the cycle. Ovulation was evident in 11/14 cycles (79%). Endometrial thickness was significantly higher when compared with that measured in previous CC cycles. Pregnancy was achieved in 1 of the cycles with letrozole. This preliminary report suggests that there might be a role for aromatase inhibitors in the management of CC-resistant women with PCOS.

Kim (Seoul, Republic of Korea) and colleagues[5] studied the effects of letrozole among women with PCOS who either responded with very thin endometrium or multiple follicles to previous CC stimulation. The number of mature follicles was smaller and the endometrium was thicker following letrozole stimulation when compared with previous CC cycles. Thus, the results of this study suggest that an aromatase inhibitor could be considered for women with PCOS who have had suboptimal response to CC.

Busso (Valencia, Spain) and colleagues[6] compared letrozole (5 mg/day) to rFSH stimulation among women with PCOS undergoing IUI. Pregnancy rates per cycles started were comparable (12.9% vs 18.5% [letrozole]). There was a trend for higher miscarriage rate (60% vs 12.5%) and therefore lower take-home baby rate (7.4% vs 11.3%) after letrozole stimulation.

The primary indication for aromatase inhibitor use is for breast cancer management. With increased use among reproductive-age women, its safety during ovulation induction needs to be evaluated. Data about maternal side effects as well as about pre- and perinatal complications need to be collected.

Biljan (Montreal, Canada) and colleagues[7] reported the results of a study of the births of 150 babies following treatment with the aromatase inhibitor letrozole. In this study, 5 mg of letrozole was given in all cycles analyzed. Perinatal outcome information over 36,000 deliveries from the same hospital was used to provide the control population. Maternal age and the proportion of multifetal deliveries (110 singleton and 20 twins) were higher in the letrozole group. Malformations were detected in 4.7% of the newborns in the letrozole group and in only 1.8% of the newborns in the control population. When the type of malformation was considered, locomotor, bone and cardiac malformations were more common following letrozole treatment.

The findings of this study are surprising, as until now no adverse perinatal effects had been reported with aromatase inhibitors. They were considered to be safe because they are used in only the early follicular phase and their half-life is short. Some points about this report need to be raised, however. The rate of malformations in the control group is lower than that usually observed. Several factors influence the rate of malformations, such as maternal age, the order of the pregnancy, gestational age at delivery, medication use during pregnancy, and maternal medical complications. One must make an effort to control for these factors during the analysis. Maternal age was higher in the study group and there were more multiple gestations. It is also known that the incidence of congenital malformations is higher following infertility treatment, so it would be better to compare the letrozole group to groups undergoing other types of infertility treatment. Still, the findings of this study should lead to more careful use of aromatase inhibitors for infertility management. Further studies need to evaluate the association between aromatase inhibitors (various doses) and congenital malformations.

Soon after IVF became widely available, it was realized that the number of multiple gestations and high-order multiple gestations had dramatically increased. According to 2002 reports, the rate of twin gestations following IVF was 32% in the United States and 24% in Europe; the rate of triplets was 7% in the United States and 2% in Europe. A multiple gestation not only increases perinatal complications but is also associated with significant maternal morbidity and mortality. A multifetal pregnancy is more likely to end before term and result in the birth of low-birthweight babies. The rate of congenital malformations is increased, and the postnatal development of the child is affected. The excess risks that mothers are exposed to include medical complications (eg, gestational diabetes, hypertension), operative delivery, and increased blood loss. For a long time, emphasis has been on improving stimulation protocols that will lead to more good-quality embryos.

In recent years, there have been more discussions about better embryo selection and the transfer of fewer, preferably single, embryos. Several presentations addressed the issue of elective single embryo transfer (SET).

Dr. Newton (London, Ontario, Canada)[8] discussed patient attitude toward the transfer of fewer embryos. According to his group's analysis, younger women were more likely to accept the transfer of 1 or 2 embryos. Those patients who considered a twin gestation a high-risk pregnancy were more likely to elect for the transfer of 1 or 2 embryos. Some patients consider twins the desired outcome of their treatment. Couples who desired twins were more likely to have 3 or 4 embryos transferred. Of interest, Newton and colleagues did not find any association between the length of infertility, costs of infertility treatment, the number of previous attempts, and the desired number of embryos transferred.

In a second assessment, the investigators presented various risk scenarios to patients with regard to multiple gestations (low risk, the actual risk, high risk) and asked them about the number of embryos they wanted to be transferred. The number of embryos that patients would have desired for transfer was reduced when the high-risk scenario was presented to them. This underlines the importance of appropriate counseling. Many patients will choose to have fewer embryos transferred when they are adequately informed about the risks associated with a multifetal gestation.

Several groups shared their experience with SET. Billay (Calgary, Alberta, Canada) and colleagues[9] compared pregnancy outcome, multiple gestation rate, and the risk of ovarian hyperstimulation syndrome (OHSS). SET was recommended to those young patients (< 36 years) who responded well to stimulation and had extra embryos for cryopreservation as well as to patients who were at risk for OHSS or who had preexisting medical problems. The control group consisted of similarly high-responding patients. More oocytes were retrieved and more embryos were available for freezing in the SET group. Although pregnancy rates were high following SET, they remained significantly lower when compared with the 2-embryo transfer group (53% vs 73.7%). On the other hand, almost half of the pregnancies were multiple gestations in this latter group (45.4% vs 2.9%). The risk of OHSS was similar.

Stillman (Rockville, Maryland, USA) and colleagues[10] discussed their experience with elective Day 5 or Day 6 (blastocyst stage) SET. Theoretically, extended embryo culture should improve the embryo selection process and should be associated with even higher implantation rates. SET was recommended to young patients with normal ovarian reserve and proven past fertility. They had to have at least 1 excellent-quality blastocyst. In this study, 112 patients agreed to SET. The pregnancy rate in this group was 51.8% (no multiples). When the cumulative pregnancy rate was evaluated following fresh and frozen embryo transfer, it was found to be 60.7%. The pregnancy rate following the transfer of 2 blastocysts during the same study period was 57%, 43% of which were multifetal. The researchers concluded that for young patients with a good prognosis who had high-quality blastocysts, SET during the blastocyst stage does not negatively influence pregnancy rates but does significantly reduce the incidence of multiple gestations.

Sparks (Iowa City, Iowa, USA) and colleagues [11] reviewed their results with single blastocyst transfer. In 2004, they implemented a new policy of single blastocyst transfer among high responders to reduce the high multiple gestation rate after the transfer of 2 blastocysts. According to their new policy, only 1 blastocyst would be transferred to women < 38 years with no previous treatment failure who had at least 1 good-quality blastocyst. Since the introduction of the new transfer policy, pregnancy rates have remained comparable (1 vs 2 blastocysts: 63% vs 57%), with no multiple gestations in the SET group. The authors also concluded that in a selected group of patients, elective SET on Day 5 or 6 does not affect pregnancy rates but minimizes the risk of multiple gestations.

It is also very important to know how well we can identify the embryo with the best implantation potential. Zech (Brussels, Belgium) and colleagues[12] evaluated whether the blastocyst transferred on Day 5 would have been chosen on the basis of morphologic criteria on Day 3 as well. According to their prospective analysis, only 41% of the embryos selected on Day 3 progressed to become the top-quality blastocyst and were transferred on Day 5. In a second arm of their study, Day 3 vs Day 5 SET outcome was compared. They observed a trend for higher ongoing pregnancy rates following Day 5 SET (34% vs 23%).

Anderson (Charlotte, North Carolina, USA) and colleagues[13] studied whether the availability of supernumerary embryos available for cryopreservation made a difference in the outcome of SET. In cycles during which 1 or more blastocysts were transferred but none were frozen, pregnancy rates were lower with SET (38% vs 59%). However, in cycles during which extra embryos were available for cryopreservation, clinical pregnancy rates were comparable following the transfer of a single or multiple blastocysts (71% vs 68%).

These retrospective and observational studies have shown that there is a role for SET. Young patients in their first or second IVF cycle who respond well to stimulation and have at least 1 good-quality blastocyst (and preferably more for cryopreservation) could elect to undergo SET without compromising their chances. Multiple pregnancy rates are dramatically reduced following SET. Ideally, the IVF program that offers SET should have an efficient cryopreservation program and should feel comfortable with culturing embryos to the blastocyst stage.

Preimplantation genetic diagnosis (PGD) is a relatively new technique that is becoming more commonly used during IVF. The distinction between preimplantation diagnosis and screening needs to be recognized. During diagnosis, assessment for a known genetic problem (gene defect, chromosomal abnormality [or HLA typing]) is conducted, and an effort is made to transfer embryos that are free of the disease. With screening, several chromosomes (those that are most frequently found in abortuses) are assessed, and randomly occurring abnormalities are screened for. The indication for genetic diagnosis is obvious. The role of screening is less clear; currently, it is offered to patients with recurrent abortion, recurrent implantation failure, and advanced maternal age. The definitions of these groups are not uniform, however, and therefore the comparison of the study results is difficult. In some of the presentations on this topic, the same problem was answered in opposite ways, and even the outcome of the same studies was interpreted differently by the experts in the field.

During PGD, either the polar body or 1 or 2 blastomeres are removed from the embryo. Fluorescent in situ hybridization (FISH) analysis is used to screen for chromosomal abnormalities and polymerase chain reaction (PCR) for single gene defects. Upon receipt of the results, the treating physician can decide whether or not to transfer a certain embryo.

Several interactive sessions and oral and poster presentations were dedicated to the discussion of the benefits of PGD. As mentioned above, advanced maternal age, recurrent abortion, and recurrent implantation failure are currently the main indications for its use.

Dr. Surrey (Englewood, Colorado, USA)[14] reviewed the clinical application of PGD. So far, results of more than 10,000 PGD cycles and 3400 pregnancies have been reported. Most reports are case series or retrospective analyses; there are very few prospective studies. Although recurrent pregnancy loss can be associated with hormonal, uterine, immunologic, and hematologic causes, the most likely etiology is genetic abnormality. A high rate of aneuploidy can be detected in abortuses. Increased incidence of genetic abnormalities was reported among couples with recurrent implantation failure. It seems logical that by screening out the abnormal embryos, implantation (and therefore pregnancy rates) could improve. On the basis of available reports, this group of women seems to benefit from PGD; the pregnancy loss rate is reduced.

A similar improvement in outcome was reported by Munne (West Orange, New Jersey, USA) and colleagues.[15]

Dr van Streitegham (Brussels, Belgium)[16] was more careful with his conclusions. He also agreed that the pregnancy loss rate is reduced following PGD among women with recurrent abortion, but according to his interpretation of data, no improvements in pregnancy and live birth rates have been reported.

Another group of women who could benefit from PGD are those of advanced age. It is known that pregnancy rates decline and miscarriage rates increase with advanced reproductive age. This seems to be associated with problematic oocytes during cell division; the number of aneuploid oocytes also increases with age. Surrey reported a trend following PGD for increased pregnancy rate for women > 35 and for a significant increase in implantation rate among women > 37 years. When more probes are used, more abnormal embryos are identified, but implantation rates can further improve as well. It is also important to point out that PGD works best when there are sufficient embryos available for analysis. The results seem to be superior when diagnosis is made on the basis of 8 or more embryos.

Although Munne and colleagues reported improved outcome with PGD for women > 35 years of age, van Streitegham did not agree. He reviewed the results of a randomized study based on 400 participants. There was a trend for increased implantation rate (OR: 1.48 [95% IR: 0.95-2.32]) but no difference in the pregnancy and ongoing pregnancy rates. Comparative studies have come to a similar conclusion; implantation rates are higher and abortion rates are lower, but there is no significant benefit in pregnancy rates.

A meta-analysis of 2 prospective studies showed a higher implantation rate (OR: 2.32 [95% CI: 1.56-3.45]) but no significant difference in the pregnancy rate when patients with poor prognosis (advanced maternal age, recurrent implantation failure) were analyzed.[17]

PGD can be used for several purposes. By screening the embryos for genetic abnormalities, the number of those replaced can be reduced. Implantation rates appear to be increased, and pregnancy loss rates are reduced. Pregnancy rates and ongoing pregnancy rates have not been reported to improve consistently. If PGD is offered to couples, they need to be informed about these results. It also would be important to carry out randomized, prospective studies; unfortunately, limited funding will prevent the performance of appropriately sized studies.

The inconsistencies between the studies also make it difficult to combine their results. Dr. Nagy (Atlanta, Georgia, USA)[18] reviewed some of the technical problems associated with PGD and the interstudy differences. The biopsy is usually performed on Day 3. It could be repeated in the blastocyst (with elective cryopreservation of the blastocyst) when the Day 3 results are inconclusive, but it is technically more challenging. In order to remove a cell from the embryo, a hole needs to be made on the zona pellucida. Laser or acid Tyrode can be used for this; the results are comparable. He described unaffected blastocyst formation potential following PGD. Others have reported lower implantation potential following the removal a blastomere, however, and further reduction with the removal of 2 blastomeres.

This takes us to one of the controversies surrounding PGD. In Europe, there is a trend for removing 2 cells, whereas they do the analysis on the basis of 1 blastomere in the United States. The benefit of having 2 cells for analysis is that the second cell can be used as a control. It is, however, possible to obtain discordant results (mosaicism), and potentially normal embryos are discarded in such cases.

Another issue concerns variations in error rates. Dr. Nagy reported a 5% true false-negative rate, 1% true false-positive rate, and a 9% false-positive rate due to mosaicism. The true misdiagnosis rate is 5.4%. A similar 5.2% error rate was reported by Munne as well. Most of the abnormal embryos that were falsely diagnosed as normal will not implant, so the clinical misdiagnosis rate is much lower (0.5%). In 7% to 10% of the embryos, results cannot be obtained, mostly due to technical problems with cell fixation or to problems during hybridization to the probe.

Cohen (Livingston, New Jersey, USA) and colleagues[19] discussed their PGD results based on more than 4000 cycles. The average age of the women was 37.5 years. Only 32% of the 33,572 embryos were normal. In 18.4% of the cycles, transfer was not performed for lack of normal embryos. When the genetic analysis of abortuses and results of prenatal genetic diagnosis were evaluated, a lower trisomy rate was observed than one would have expected on the basis of the patients' age.

Several groups studied whether embryo cleavage rates correlated with genetic abnormalities. Hamamah (Montpellier, France) and colleagues[20] studied 185 embryos for early cleavage (25 to 27 hours post insemination) and correlated the results with PGD results. There was a higher rate of early cleavage among embryos not affected genetically (49% vs 40%).

Srivastava (Fairfax, Virginia, USA) and colleagues[21] evaluated whether blastocyst formation rate was affected by genetic abnormalities in the embryo. FISH analysis was performed for 5 chromosomes in some cases and for 8 in the majority of cases. Chromosome abnormalities were observed in 42% of the blastocysts. The investigators concluded that blastocyst formation and hatching rates were not affected by the genetic status of the embryo.

PGD by now is considered a routine laboratory procedure. It clearly is associated with benefits when a known genetic problem needs to be ruled out. Its benefits are less obvious when it is used as a screening tool. Future studies will likely continue to report higher implantation rates and lower pregnancy loss rates. The net effect could be the transfer of fewer embryos and, ultimately, higher pregnancy rates. PGD appears to be safe. It does not significantly affect the embryo cleavage rate and leads to only a small decrease in the implantation potential.

Future studies will have to answer several important questions: Should 1 or 2 cells be removed? How many chromosomes should be screened? How can the misdiagnosis rate be reduced? Until such results are available to answer these questions, PGD should continue to be offered to couples with recurrent pregnancy loss, recurrent implantation failure, or advanced maternal age. The limitations of the process also need to be discussed with them.

Assisted reproduction has helped many infertile couples to start a family. With better understanding of the reproductive and embryonic processes, the treatment continuously improves. Nowadays, a singleton birth is accepted as the only good outcome of assisted reproduction. Stimulation protocols that are tailored to the needs of the patient have made the process safer and more tolerable. Laboratory improvements now allow extended embryo culture and genetic screening of the cleavage stage embryos; both techniques should result in the transfer of fewer better-selected embryos and in a lower multiple gestation rate. This should ultimately translate into a reduction in perinatal morbidity and mortality.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.