Pregnancy rate, multiple pregnancy rate, and embryo quality: Clues for single blastocyst transfer from double blastocyst transfer in an unselected population

Abstract

Objective:  Minimizing multiple pregnancy is a priority in assisted reproduction. As implantation rates are critical to success and reduce multiple pregnancy, we investigated whether blastocyst grade determined implantation rate following double blastocyst transfer in unselected cases.

Materials and Methods:  We studied 69 three‐cleavage stage embryo transfers and 64 two‐blastocyst transfers. Two blastocysts, or one when two blastocysts were not available, were transfered after evaluating the grade of blastocysts. The difference in pregnancy and implantation rates to patient age, the number of retrieved oocytes and grade of blastocysts were analyzed.

Results:  Blastocyst and grade 3AA rates per fertilized egg were 50.3% and 26.0%, respectively. Following two‐blastocyst transfer, pregnancy rate per transfer, implantation rate per embryo, and multiple pregnancy rate per pregnancy were 39.1%, 26.5%, and 24.0%, respectively. Two‐blastocyst transfer achieved implantation more often than three‐cleavage‐stage embryo transfer, but did not reduce multiple pregnancy. Pregnancy, implantation, and multiple pregnancy rates did not reflect maternal age. Higher pregnancy and implantation rates per transfer were attained for with six or more oocytes retrieved or transfer of two‐blastocyst graded 3AA or higher especially when two or more blastocysts graded 3AA or higher are available, but the latter showed a high multiple pregnancy rate (38.5%).

Conclusions:  Single embryo transfer could be carried out when two or more blastocysts of grade 3AA or higher have been developed. (Reprod Med Biol 2005; 4: 153–160)

INTRODUCTION

DECREASING THE RATE of multiple pregnancy, especially high‐order pregnancies, is an urgent issue for in vitro fertilization (IVF) and other assisted reproductive technologies (ART). Multiple pregnancy is recognized as a major complication of both ART and ovulation induction therapies. The number of multiple births associated with infertility treatment has been estimated to be as high as 32% for twins and 4.7% for triplets or higher orders, ¹ compared with respective rates in the general population of 2.1% and 0.08%. ² Over many years, multiple gestations as opposed to singleton pregnancies have been linked to increased health risks for both mothers and infants. Although twins are often regarded as an acceptable or even desirable outcome after infertility treatment, even these pregnancies carry increased risks of infant mortality and morbidity compared with a singleton pregnancy. The infant mortality rate in 2000 was reported as 6.1 per 1000 live births for singleton pregnancies and 31.1 for multiple pregnancies in the USA, increasing to 28.9 per 1000 live births for twins, 63.2 for triplets, and 95.5 for quadruplets. ³ Twins showed an approximately fivefold excess in risk of fetal death and a sevenfold excess in risk of neonatal death when compared with singletons, although the risk varied between twin pregnancies; second‐born twins, twins from same‐sex or growth‐discordant pairs, and twins whose co‐twin died in utero showed further increases in risk of death. ⁴ Emotional burdens on the family are enormous as are costs, which ultimately are borne by society as a whole. Projected excess costs per family ranged from US$43 300 for twins to US$174 000 for quadruplets. ¹

Implantation rate is a critical factor for successful outcome of human ART. The pregnancy rate and multiple pregnancy rates are both functions of the implantation rate and the number of embryos transferred. Therefore, the number of embryos transferred is a critical factor in reducing the multiple pregnancy rate. When the number of embryos transferred per cycle is limited to one or two, uterine receptivity and quality of embryos transferred are the keys to the success of the procedure. However, multivariant analysis has identified many predictors of pregnancy such as maternal age, number of oocytes retrieved, developmental stage and morphology scores of the two best embryos available for transfer, and day of transfer, cleavage stage, or blastocyst stage. ⁵ Transfer of only one embryo should be limited to a selected population to maintain the overall pregnancy rate, as likelihood of implantation will differ for each embryo and patient. ⁵ , ⁶ , ⁷ Feasibility of elective single embryo transfer in cleavage‐stage embryos (eSET) initially was demonstrated retrospectively after judging the quality of embryos transferred in double embryo transfer (DET) procedures. The answer to the question of what grade of blastocyst represents a putative high competence (PHC) embryo is an unresolved question, although many studies have reported that blastocyst transfer offers a higher implantation rate than transfer of cleavage‐stage embryos. ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ Most blastocyst transfer programs are intended for selected groups of patients, so even randomized controlled studies involve self‐selection bias. ⁸

In the present study we determined whether multiple pregnancy rate could be decreased without decreasing pregnancy rate following transfer of two high‐grade blastocysts in an unselected group of patients. We also investigated which grade of blastocysts could achieve a high implantation rate potentially suitable for elective single blastocyst transfer (eSBT).

MATERIALS AND METHODS

Patients

WITH WRITTEN INFORMED consent of all patients under the approval of our institutional review board, 133 cycles of conventional IVF carried out in the Department of Obstetrics and Gynecology at Hiroshima University School of Medicine were enrolled from December 1999 to December 2003. During the infertility work‐up, we followed a standard protocol for history‐taking, physical examination, and investigations for all patients, including conventional semen analysis on at least two occasions, basal body temperature, hysterosalpingography, serum progesterone, prolactin, thyroid‐stimulating hormone in the luteal phase, postcoital test, and transvaginal ultrasonography. From December 1999 to December 2001, we transferred three embryos at cleavage stage in 69 cycles on day 2 after oocyte retrieval. The main cause of infertility was tubal factor in 19 cycles (27%), male factor in six cycles (9%), endometriosis in eight cycles (12%), and unexplained infertility in 36 cycles (52%). From December 2001 to December 2003 we transferred two embryos in 64 cycles on day 5 after oocyte retrieval. The main cause of infertility was tubal factor in 22 cycles (34%), male factor in five cycles (8%), endometriosis in nine cycles (14%), and unexplained infertility in 28 cycles (44%). Patient ages in each group are compared in Table 1. All infertility patients undergoing conventional IVF in our department were included during this period. In couples with repeated semen analyses where either sperm count in the ejaculate was below 1 × 10⁶/mL or the motility rate was below 20%, intracytoplasmic sperm injection (ICSI) was used. Cycles with ICSI (116 cases) or frozen and thawed embryos (42 cases) were not included in this series.

Table 1.

Comparison of pregnancy outcome following cleavage‐stage embryo transfer and blastocyst transfer

	Cleavage stage	Blastocyst
Number of cycles	69	64
Mean age of patients (mean ± SE)	33.5 ± 0.6	34.3 ± 0.5
Age range (years)	26–47	26–42
Number of oocytes per retrieval (mean ± SE)	9.4 ± 0.8	7.8 ± 0.6
Number of pronuclear embryos (mean ± SE)	5.8 ± 0.5	5.3 ± 0.4
Number of embryos transferred (mean ± SE)	2.6 ± 0.09	1.8 ± 0.06
Clinical pregnancy rate (%)	26.1%	39.1%
Implantation rate (%)	13.7%*	26.8%*
Multiple pregnancy rate (%)	33.3%	24.0%
Spontaneous abortion rate (%)	11.1%	24.0%

^* P < 0.05. SE, standard error of the mean.

Controlled ovarian stimulation, oocyte retrieval, and insemination

Pituitary secretion was downregulated by administering a gonadotropin‐releasing hormone (GnRH analog), buserelin (Suprecur; Aventis, Tokyo, Japan), as a nasal spray. It was administered three times daily (900 µg) beginning in the mid‐luteal phase of the cycle preceding the treatment cycle. Pituitary downregulation was confirmed by both transvaginal ultrasonography and serum estradiol measurement carried out on day 7 of the treatment cycle. Injections of human menopausal gonadotropin (150–300 IU/day; Humegon; Organon, OSS, the Netherlands, or HMG Nikken, Nikken Kagaku, Tokyo, Japan) were then begun. Follicular growth was monitored by transvaginal ultrasonography and serum estradiol concentrations. An intramuscular injection of human chorionic gonadotropin (HCG; 10 000 IU) was given when at least two follicles exceeded 17 mm in diameter. Oocytes were retrieved with an 18‐gauge single‐lumen needle (Kitazato, Shizuoka, Japan) guided by transvaginal ultrasonography under intravenous anesthesia, 34–36 h after HCG injection. After each oocyte was cultured for 4 h in human tubal fluid (HTF) (Irvine Scientific, Santa Ana, CA, USA), it was inseminated with spermatozoa selected in a ‘swim‐up’ manner over 2 h during incubation at 37°C with 5% CO₂ in air. Fertilization was determined to have occurred when two pronuclei could be identified after approximately 20 h.

Cleavage stage embryo transfer

After being graded morphologically according to Veeck, a maximum of three embryos graded 3 or higher were transferred using a Wallace catheter (Edwards‐Wallace catheter, Marlow Technologies, Willoughby, OH, USA) under ultrasonographic guidance on day 2 after oocyte retrieval.

Blastocyst transfer

Fertilized oocytes were cultured in sequential medium (BlastAssist system, Jyllinge, Denmark) at 37°C with an atmosphere of 5% CO₂, 5% O₂, and 90% N₂ for 5 days. Two blastocyst embryos, or one when two were not available, were transferred 5 days after oocyte retrieval using a Wallace catheter under ultrasonographic guidance after being graded morphologically according to Gardner. Briefly, the blastocysts were graded according to size: 1 or early blastocyst, with the blastocele presenting less than half the volume of the embryo; 2 or blastocyst, with a blastocele making half or more of the volume of the embryo; 3 or full blastocyst, with a blastocele completely filling the embryo; 4 or expanded blastocyst, with a blastocele volume larger than that of the early embryo and also thinning of the zona; 5 or hatching blastocyst, with trophectoderm beginning to herniate through the zona; and 6 or hatched blastocyst, having completely escaped from the zona. For blastocysts graded 3–6, development of the inner cell mass (ICM) and trophectoderm also were assessed. The ICM grading was as follows: A, tightly packed with many cells; B, loosely grouped with several cells; and C, with very few cells. Trophectoderm grading was as follows: A, many cells forming a tightly knit epithelium; B, few cells; and C, very few cells forming a loose epithelium. The quality of embryo was evaluated by one observer.

Only morula or lower‐grade embryos were available in six patients, and in one patient only cleavage‐stage embryos were available. All such patients chose to have the embryos transferred. No embryos underwent assisted hatching before transfer. Cryopreservation of supernumerary blastocysts on day 5 was carried out using vitrification. Only blastocysts scoring 3BB or higher on day 5 were frozen.

Luteal support and outcome of in vitro fertilization

After oocyte retrieval, micronized vaginal progesterone (450 mg/day) was used for luteal support, at least until a pregnancy test was carried out.

Clinical pregnancy was detected by demonstration of a gestational sac by transvaginal ultrasonography after a positive urine pregnancy test (HCG ≥50 IU/L). Pregnancy rate was calculated as the number of patients diagnosed with clinical pregnancy divided by the number of transfers performed. The embryo implantation rate was calculated as the number of intrauterine gestational sacs noted on ultrasonographic examination carried out at least 6 weeks after transfer divided by the total number of embryos transferred. A set of monozygotic twins was considered to have one gestational sac. Implantation rate was calculated for all patients undergoing embryo transfer, not only those who became pregnant. Miscarriage and multiple pregnancy rates were determined as occurrences of miscarriage or multiple pregnancy divided by the number of clinical pregnancies.

Statistical analysis

Differences in pregnancy and implantation rates according to number of oocytes retrieved and grade of embryos were analyzed by χ² methods. P < 0.05 was considered to be significant.

RESULTS

IN VITRO FERTILIZATION outcomes for cleavage‐stage embryo transfer and blastocyst transfer are shown in Table 1. Pregnancy rate per transfer, implantation rate per embryo, multiple pregnancy rate per pregnancy were 39.1, 26.8, and 24.0%, respectively, following double blastocyst transfer (DBT). The implantation rate in DBT was higher than that in cleavage‐stage transfer, even though fewer embryos were transferred in blastocyst transfer. The multiple pregnancy rate was not reduced by two‐blastocyst transfer.

Five hundred and two oocytes were retrieved for blastocyst transfer (Table 2). Of these, 342 oocytes were fertilized, 172 embryos developed into blastocysts, and 89 embryos were graded as 3AA or higher. The fertilization rate was 68.1%. The blastocyst rate and rate of grade 3AA or higher per fertilized egg were 50.3 and 26.0%, respectively.

Table 2.

Embryo development to blastocyst stage

	Total
Cumulative number of oocytes	502
Cumulative number of pronuclear embryos	342
fertilization rate	68.1%
Cumulative number of blastocysts	172
blastocyst rate per pronuclear embryos	50.3%
Cumulative number of blastocysts grade 3 or better	128
% per pronuclear embryo	37.4%
% per blastocyst	74.4%
Cumulative number of blastocysts grade 3AA or better	89
% per pronuclear embryo	26.0%
% per blastocyst	51.7%

The effect of the patient age on IVF outcome following blastocyst transfer is shown in Table 3. Clinical pregnancy, implantation, and multiple pregnancy rates were lower in patients over 35 years old than in those below 34, although the difference between two age groups was not significant.

Table 3.

Effect of patient age on pregnancy outcome following blastocyst transfer

Patient age (years)	Number of transfers	Total number of embryos transferred	Number of embryos transferred	Clinical pregnancy (%)	Implantation (%)	Multiple pregnancy (%)	Miscarriage (%)
Below 34 years	32	57	1.8	15 (46.9)	18 (31.6)	4 (26.7)	2 (13.3)
Over 35 years	32	56	1.6	10 (31.3)	12 (21.4)	2 (20.0)	4 (40.0)
Total	64	113	1.8	25 (39.1)	30 (26.5)	6 (24.0)	6 (24.0)

Rates did not differ between the three age groups in the same column (P > 0.05).

Effect of the number of oocytes retrieved on IVF outcome following blastocyst transfer is shown in Table 4. Significantly higher pregnancy and implantation rates per transfer were attained in cycles with six or more oocytes retrieved (56.4 vs 12.0%, 35.1 vs 10.3%, respectively; P < 0.01).

Table 4.

Effects of the number of oocytes retrieved on pregnancy outcome following blastocyst transfer

Number of oocytes retrieved (range, mean ± SD)		Mean age (years)	Number of cycles transferred	Number of embryos transferred	Clinical pregnancy (%)	Implantation (%)	Multiple pregnancy (%)	Miscarriage (%)
6 or more	(6–25, 10.5 ± 4.0)	33.8	39	1.9	22 (56.1)*	26 (35.1)*	5 (22.7)	5 (22.7)
below 6	(1–5, 7.8 ± 4.7)	34.9	25	1.6	3 (12.0)*	4 (10.3)*	1 (33.3)	1 (33.3)
Total		34.3	64	1.8	25 (39.1)	30 (26.8)	6 (24.0)	6 (24.0)

^* P < 0.05.

Effects of blastocyst quality and numbers of blastocysts transferred on IVF outcomes are summarized in Table 5. Significantly higher pregnancy and implantation rates per transfer were achieved following transfer of two blastocysts graded 3AA or higher (65.0 vs 27.3%, 45 vs 16.7%, respectively; P < 0.01). The overall multiple pregnancy rate was 24%, becoming much higher (38.5%) following transfer of two blastocysts graded 3AA or higher. Other situations were associated with no multiple pregnancies except for one set of monochorionic twins.

Table 5.

Effects of blastocyst quality and the number of blastocysts transferred on pregnancy outcome

Number of embryos transferred and embryo quality	Number of embryos transferred	Number of cycles transferred	Clinical pregnancy (%)	Implantation (%)	Multiple pregnancy (%)	Miscarriage (%)
3AA or higher	2	20	13 (65.0)*	18 (45)*	5 (38.5)	3 (23.1)
Other groups		44	12 (27.3)*	12 (16.7)*	1 (8.3)	3 (25.0)
3AA or higher and also below 3AA	2	13	7 (53.6)	7 (26.9)	0 (0.0)	1 (14.3)
3AA or higher	1	6	1 (16.7)	1 (16.7)	1 (100)	0 (0.0)
below 3AA	2	15	2 (13.3)	2 (6.7)	0 (0.0)	0 (0.0)
below 3AA	1	10	2 (20.0)	2 (20.0)	0 (0.0)	2 (100)
Total		64	25 (39.1)	30 (26.8)	6 (24.0)	6 (24.0)

^* P < 0.05 or 0.01.

DISCUSSION

IMPLANTATION RATE IS the determining factor in evaluating successful human IVF, ⁸ with pregnancy rate being a function of implantation rate and number of embryos transferred. Therefore, knowledge of the implantation potential of embryos transferred is essential for optimal embryo selection in IVF. ¹⁷ In the present study, significantly higher implantation and pregnancy rates per transfer could be attained by transfer of two blastocysts of grade 3AA or higher following two or more blastocysts of grade 3AA or higher could be developed (65.0 vs 27.3% and 45 vs 16.7%; P < 0.01, respectively). However, the high implantation rate of blastocyst transfer resulted in a high twin pregnancy rate in DBT, as also was true when transferring three cleavage‐stage embryos. The overall multiple pregnancy rate was 24%, becoming very high (38.5%), following transfer of two graded at least 3AA blastocysts. Furthermore, dichorionic diamniotic twin pregnancies occurred only in this group.

Several criteria have been proposed to identify the best embryos, including pronuclear morphology, embryo development rate, fragmentation, blastomere multinucleation, and development to blastocyst. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² In cleavage‐stage embryos, PHC have been defined as absence of multinucleated blastomeres, presence of four or five blastomeres on day 2, a minimum of seven cells on day 3, and a maximum of 20% anucleated fragments, based on retrospective assessments after judging the quality of embryos prior to double cleavage‐stage embryo transfer. ⁶ , ¹⁷ However, prior to blastocyst formation, the maternal‐embryonic genome transition is not complete. A selection of embryos prior to blastocyst formation will reflect the function of proteins and mRNA synthesized during oocyte maturation. ²³ Although selection of morphologically favorable blastocysts proved easy and reliable, PHC has not been well defined for single‐blastocyst transfer (SBT). Our data suggest that a 3AA blastocyst should be considered as PHC embryo for SBT in situations where two or more 3AA blastocysts are available. However, development to a blastocyst is not in itself an indication of an implantable normal embryo; for example, 26% of chromosomally abnormal embryos have been found able to develop to the blastocyst stage. ²⁴ In our present series a morphologically favorable blastocyst was highly implantable, but this did not assure morphologically good embryos with euploidy or an ongoing pregnancy. Some morphologically favorable blastocysts were associated with spontaneous abortion. In two such cases where chromosomal analysis was carried out, one showed aneuploidy and the other showed euploidy. An embryo scoring system better able to identify not only implantable but also euploid embryos is needed. Uterine receptivity is another critical factor to determining implantation rates following IVF, although the current study did not analyze this important factor. Improving reliability of assessment of uterine receptivity remains an ongoing challenge.

Many factors such as patient age, the duration of infertility, previous IVF treatment, and previous pregnancy, particularly a live birth, are known to influence implantation rate in IVF procedures. ²⁵ , ²⁶ In general, patient age is the most important factor for successful outcome following IVF and also for multiple pregnancy rate. ²⁵ In this series, pregnancy rate, implantation rate, and multiple pregnancy rate were lower in patients over 35 years old than in those below 34, although the difference between two age groups was not significant. The result was similar to that reported by Kuramoto et al. ⁹ Multiple pregnancy was not seen in patients over 36 years old. The rate of spontaneous abortion in patients over 35 years old also was higher than in those below 34, although the difference was not significant. Better statistical power will be needed for conclusiveness. The number of embryos available for transfer is now recognized as an important factor in addition to the patient's age. ²⁷ Success rates are higher if embryos can be selected from a large pool rather than from only one or two available embryos. ⁵ , ²⁵ , ²⁶ When more than four eggs are fertilized and available for transfer, a woman's chance of giving birth is not diminished by transferring only two embryos. ²⁶ , ²⁸ This finding was highly consistent with our result that a woman's chance of pregnancy increased when six or more eggs were retrieved. As the fertilization rate in the present series was 68%, a retrieval of over six eggs yielded a likelihood of more than four eggs fertilized. Furthermore, rates of blastocyst development and 3AA development after such retrieved were approximately 50 and 25%, respectively. These data indicate that among embryos of patients with six or more oocytes available, at least one can reach a grade 3AA blastocyst stage. These high‐quality embryos were expected to implant reliably following transfer.

The multiple pregnancy rate in the present series was not considered acceptable, although triplet or higher‐order multiple pregnancy was not recognized. What twin pregnancy rate is an acceptable one remains controversial. To reduce the twin pregnancy rate, single embryo transfer should be carried out, although transfer of only one embryo should be limited to a selected population with high‐quality embryos if the overall pregnancy rate is to be maintained. ⁵ , ⁶ , ⁷ Another critical factor for blastocyst transfer is whether or not the rate of zygotic splitting increases. ²⁹ The rate of multiple pregnancies after a single embryo transfer represents a 2.3% rate of zygotic splitting among in vitro conceptions, which is six times the reported rate of 0.4% for spontaneous conception. ³⁰ In the current series, one monochorionic, diamniotic twins out of 25 clinical pregnancies was recognized. All forms of assisted conception including ovulation induction increase the rate of zygotic splitting, although the exact reasons and detailed process of zygotic splitting remain unclear. Alterations produced by ovarian stimulation, in vivo culture conditions, extended in vitro culture, and manipulation of the zona pellucida have been mentioned as possible causes of more frequent monozygotic twinning in pregnancies after assisted reproduction than after natural conception. ³⁰ A greater overall frequency of monozygotic twins for IVF patients may be a function of the higher number of embryos transferred in IVF, rather than discrete zona manipulation. ³¹ Prolonging embryo culture in vitro to the blastocyst stage has been presented as an effective form of embryo selection that results in increased implantation rates. However, ideal culture conditions for an embryo to reach its biologic potential are not completely defined, and prolonged exposure of the embryo to laboratory conditions may not be free of risk for zygotic splitting to result in a monozyotic multiple pregnancy. ³² Improved culture conditions, including provision of free‐radical scavengers and growth factors such as granuluocyte and macrophage colony‐stimulating factor can reduce the rate of zygotic splitting with prolonged embryo culture. ²⁹ , ³³ The rate of zygotic splitting with prolonged culture would be a determining factor for choosing cleavage stage embryo transfer or blastocyst transfer. The risk of iatrogenic zygotic splitting and its potential adverse outcome at least should be explained to the infertile couple before therapy is initiated, even if a single embryo is transferred to minimize repercussions from a possible multiple pregnancy. ³⁰

In summary, a significantly high pregnancy and implantation rate per transfer could be reached following the transfer of two blastocysts graded 3AA or higher, especially when two or more 3AA or higher blastocysts are available. The DD twins were recognized in this selected group of patients when they were below 35 years old. In this situation, eSBT could be carried out without decreasing the rate of pregnancy per transfer.

In the future, prospective randomized trials comparing effectiveness of eSBT and eDBT in unselected populations are needed, as well as trials comparing effectiveness of eSET and eSBT in unselected populations. Multiple‐regression analysis should identify the factors contributing to increased IVF success rate. Importantly, quality of an IVF program is measured by implantation rate and cumulative single pregnancy rate per woman including cycles with cryopreservation and thawed embryo transfer, rather than simply by pregnancy rate per cycle.