Introduction

In human-assisted reproductive technology (ART), approximately 1.6–7.7 % of embryos show abnormal zygotic stages with a single pronucleus (1PN) following conventional in vitro fertilization (cIVF) or intracytoplasmic sperm injection (ICSI) [1–6]. Although such abnormal pronuclear-stage embryos are considered unsuitable for transfer to patients, they are occasionally transferred and a few attempts for normal pregnancies have been reported including embryos derived from 1PN zygotes [4, 5, 7, 8]. However, formation rate and clinical pregnancy rate of blastocysts derived from 1PN zygotes were unclear. Recently, the analyses of chromosomal status and genetic composition in 1PN zygotes and embryos have been reported. For example, when the chromosomal status of 1PN embryos were analyzed using fluorescence in situ hybridization (FISH), the rate of diploid chromosome constitution of the cleavage-stage embryos and blastocysts derived from 1PN in cIVF were significantly higher than in ICSI [6, 9, 10]. Furthermore, van der Heijden et al. reported that the genetic composition of 1PN zygotes was observed based on histon modification using immunostaining, and some of these zygotes included the maternal and paternal genome in 1PN [11]. Further, they reported that the rate of diploid chromosomes in 1PN zygotes were 86.7 % in cIVF and 30.3 % in ICSI [11]. Moreover, Liao et al. reported that the diploid rate of blastocysts derived from 1PN embryos was significantly higher than in arrested cleavage-stage embryos [12]. Although there have been several reports analyzing the chromosome constitution of embryos derived from 1PN by FISH or immunostaining, there is little embryological and clinical data for 1PN embryos. The strongest evidence of good-quality embryos is the production of live healthy babies. In this study, we retrospectively studied the clinical outcomes of 1PN embryos which were cultured until blastocyst stage in cIVF and ICSI cycles. To evaluate the 1PN embryos, all 1PN embryos were analyzed for the blastocyst formation rate and good-quality blastocyst rate; some 1PN blastocysts were transferred into patients to examine their potential for development into healthy babies. The cycles in which at least one 1PN embryo was observed (1PN groups) were compared to cycles with 2PN embryos (control groups) obtained in the same cIVF or ICSI cycles.

Materials and methods

Patients

In this study, a total of 623 patients were included who underwent cIVF/ICSI at a private clinic [Department of Infertility, Angel Bell Hospital (Aichi Prefecture, Japan)] between May 2010 and September 2014. All patients underwent cIVF/ICSI, embryo culture until blastocyst stage, and blastocyst transfer onto an endometrium prepared using hormone therapy. Cycles with split (cIVF + ICSI) were excluded. We avoided transferring blastocysts derived from 1PN zygotes as much as possible, since the safety of these embryos has not been established. Consequently, these embryos we only transferred when there were no other embryos available.

Ovarian stimulation

Ovarian stimulation, oocyte retrieval, and cIVF/ICSI procedures have been previously described by Murata et al. [13]. Most often, the ovarian stimulation protocol entailed administration of a gonadotrophin-releasing hormone (GnRH) antagonist cetrolelix (Cetrotide; Merck Serono) with subsequent addition of recombinant follicle stimulating hormone (FSH; Gonal-F®, Merck Serono)/human menopausal gonadotrophin (hMG; Menopur; Ferring Pharmaceuticals). The exogenous FSH/hMG was administered daily until the leading follicles averaged 18 mm in diameter. Measurements were made using serial ultrasound recordings. In addition to exogenous FSH/hMG administration, cetrorelix was injected daily once the leading follicle reached 16 mm in diameter or a premature luteinizing hormone (LH) surge was suspected based on serum LH monitoring. These injections were continued until the day before the human chorionic gonadotrophin (hCG; HCG Mochida, Mochida Pharmaceutical, Tokyo, Japan) was injected.

Alternatively, ovarian stimulation entailed administration of a GnRH agonist (buserelin acetate; Sprecure, Mochida Pharmaceutical) in a short (flare-up) protocol with subsequent addition of FSH/hMG. In another cycle, clomiphene citrate with/without exogenous FSH/hMG was used for follicular growth induction.

hCG (5000 IU) was administered 36 h before transvaginal oocyte retrieval or a GnRH agonist was administered twice, 35 and 36 h, before retrieval. Following retrieval, oocytes were fertilized using cIVF or ICSI.

Fertilization, embryo culture and evaluation, and cryopreservation

Fertilization was performed 3–6 h after retrieval in Universal IVF medium (ORIGIO Japan, Yokohama, Japan) with either cIVF or ICSI as appropriate for the presence or absence of male infertility factor. After cIVF or ICSI, the oocytes were cultured in individual 20-μL droplets of universal IVF medium on petridishes (60 mm; Falcon) at 37 °C under a humid atmosphere of 6 % CO2, 5 % O2, and 89 % N2. Fertilization was assessed 18–20 h after cIVF or ICSI based on the presence of two pronuclei. The zygotes were divided and cultured as 2PN or 1PN on culture dishes (4-well plastic dish; Nunc) containing 100 μL of culture medium (Single-step medium; global total medium, LifeGlobal), covered with mineral oil (light mineral oil, Irvine Scientific) and cultured until day 5/6.

Evaluation of embryos on day 3 was based on the number and symmetry of the blastomeres and the percentage fragmentation and multinucleation. Blastocysts were evaluated by two embryologists who assigned each blastocyst a grade using the grading system of Schoolcraft et al. [14]. Embryos were cryopreserved on either day 5 or 6, depending on the grade of the blastocyst morphology. Cooling and embryo warming were performed using the Cryotop methodology for human embryo vitrification described by Kuwayama et al. [15]. Equilibration, vitrification, warming, dilution, and washing solutions were provided in a vitrification kit (VT-101 and VT-102; Kitazato Corporation, Shizuoka, Japan).

Frozen-thawed blastocyst transfer cycle

All frozen-thawed blastocyst transfer (F-TBT) cycles were performed during artificial hormone replacement cycles with identical endometrial preparation protocols [13]. The proliferative phase was induced by transdermal administration of incremental doses of estradiol (E2) (Estrana; Hisamitsu Pharmaceutical, Tokyo, Japan). After confirming the endometrial thickness to be more than 8 mm using ultrasonography, daily administration of 6 mg of chlormadinone acetate (Lutoral; Shionogi & Co., Ltd., Tokyo, Japan) was initiated, and blastocyst transfer was performed on the 5th or 6th day of chlormadinone administration. Serum hCG was checked 9 days after the blastocyst transfer, and patients with hCG >50 IU/L were considered pregnant. One week later, transvaginal ultrasound was performed to confirm intrauterine pregnancy. In cases of pregnancy, transdermal E2 (2.16 mg/every 2 days) and chlormadinone (6 mg/day) were administered for the first 9 weeks of gestation. Fetal viability was then assessed after 3 weeks, and ongoing pregnancy was determined after 12 weeks of pregnancy. The implantation rate was determined by dividing the number of gestational sacs by the number of embryos transferred.

Statistical analysis

The distribution of variable data was assessed, and the normal and non-normal distributions were appropriately treated. Data were presented as means ± standard deviation. Group characteristics were compared using the Mann–Whitney U test. The χ 2 test was used to evaluate the significance of the proportion (%). Values of p < 0.05 were considered significant.

Results

Figure 1 summarized the flow chart of this study. A total of 623 cycles were treated with cIVF or ICSI (426 cIVF and 197 ICSI cycles) between May 2010 and September 2014 that included at least one 1PN embryo in our clinic. In all cycles, the incidence of 1PN formation was 5.7 % (576/10,141) in cIVF and 4.7 % (252/5325) in ICSI.

Fig. 1
figure 1

Flow chart of cIVF/ICSI cycles and embryo development in the study

Full size image

As shown in Table 1, the patient characteristics and the pronuclear status after cIVF or ICSI differed significantly with respect to the mean number of previous oocyte retrievals, 2PN rate, and the mean number of cryopreserved blastocysts of 1PN. On the other hand, the rate of 1PN was similar between the cIVF (22.1 %) and ICSI (25.1 %) cycles.

Table 1 Patient characteristics in cIVF and ICSI cycles including at least one 1PN embryo
Full size table

As shown in Table 2, after culturing the cIVF- or ICSI-derived 1PN embryos for 3 days, the development rates of the good-quality embryos did not significantly differ (19.8 vs. 17.5 %, respectively). When 1PN embryos were cultured to blastocyst, the blastocyst formation rate and good-quality blastocyst rate in the cIVF cycles (21.4 and 7.7 %) was significantly higher than those in the ICSI cycles (10.7 and 0.7 %) (p < 0.01). However, blastocyst formation rate derived from 1PN embryos were significantly lower compared to 2PN embryos (p < 0.01).

Table 2 Embryological outcomes of 2PN and 1PN zygotes in cIVF and ICSI cycles
Full size table

After culturing 1PN embryos derived from cIVF and ICSI to blastocysts, these blastocysts were cryopreserved until F-TBT was performed (Table 3). F-TBT was performed in 791 cycles including 36 cycles of 1PN (cIVF 33 and ICSI 3 cycles). However, 1PN blastocysts were only transferred when there were no other embryos available. This population was included with the patients with multiple courses of F-TBT. The implantation rates and clinical pregnancy rates of blastocysts from 1PN in cIVF cycles (33.3 and 33.3 %) were not significantly different compared with controls for cIVF (39.0 and 35.9 %) and ICSI (46.7 and 41.1 %). In this study, 33 transfer cycles with 33 blastocysts derived from 1PN embryos in cIVF resulted in nine deliveries. However, although four blastocysts (3 cycles) from 1PN in ICSI were transferred, no implantation was observed. We also examined the neonatal data from 1PN blastocysts in cIVF compared with 2PN blastocysts in cIVF and ICSI (Table 4). There were no differences in the weeks of gestation and weight of the newborns at birth. Moreover, nine newborns from F-TBT with 1PN blastocyst were delivered and presented with no major malformations.

Table 3 Clinical outcomes of 1PN and 2PN blastocysts in cIVF and ICSI cycles
Full size table
Table 4 Neonatal outcomes of 2PN and 1PN blastocysts in cIVF and ICSI cycles
Full size table

Discussion

We obtained normal healthy babies without malformation successfully from 1PN zygotes cultured until the blastocyst stage. The 1PN embryos in cIVF that did develop to blastocysts showed adequate potential for implantation and ongoing pregnancy as with 2PN blastocysts.

Several attempts to analyze human 1PN zygotes and embryos in cIVF and ICSI cycles have been reported, such as the mechanism of 1PN formation on fertilization [4, 6, 16–18]; chromosome constitution in zygote, embryo, and blastocyst stages [6, 9, 10, 12, 19–21]; and genetic composition of the pronucleus in human [11] and mouse [22]. According to these data, when the chromosomal status of 1PN embryos was analyzed using fluorescence in situ hybridization (FISH), the rate of diploid chromosome constitution in cIVF was significantly higher than that in ICSI [6, 9, 10]. Liao et al. reported that the diploid rate of the blastocysts derived from 1PN embryos (74.6 %) was significantly higher than that from the arrested cleavage-stage 1PN embryos (31.6 %) [9]. Otsu et al. also reported that all analyzed 1PN blastocysts in cIVF were diploid [20]. However, the diploid rate of ICSI 1PN blastocysts was 37.5 % [19]. van der Heijden et al. reported that 86.7 % of the 1PN zygotes included the maternal and paternal genome in cIVF and that the diploid rate of cIVF cycles was higher than that of ICSI cycles (30.3 %) using immunostaining based on histon modification [11]. Moreover, Mateo et al. reported that when 1PN embryos were cultured to blastocyst in ICSI, the blastocyst formation and good-quality blastocyst rates were only 14.8 and 7.4 %, respectively [19]. In addition, Reichman et al. reported that the implantation rate and pregnancy rates of the 1PN embryos from cIVF or ICSI were only 6.4 and 1.3 %, respectively [5]. Most of these studies have reported that the rates of diploid chromosome constitution in 1PN zygotes in cIVF were higher than ICSI, though poor embryo development and clinical outcomes were present in both cycles. Consequently, it has been proposed that the transfer of 1PN embryos should be avoided and discarded in human ART programs. However, the 1PN embryos were sometimes clinically transferred and the few reports regarding newborn outcome of 1PN zygotes were almost all small study and case reports.

In the present study, the incidence of abnormal zygotic stages with 1PN following cIVF and ICSI was similar to previous reports [4, 6]. Further, when 1PN embryos in cIVF and ICSI were cultured until day 6, the blastocyst formation rate was significantly reduced (Table 2), similar to previous reports [5]. In addition, although the rate of good-quality embryos from 1PN embryos in ICSI on day 3 did not significantly differ from 1PN embryos in cIVF, 1PN embryos in ICSI showed a significantly lower rate of development to blastocyst. This may be due to the fact that human 1PN zygotes in cIVF are formed by the enclosure of the juxtaposed male and female nuclei in a common pronuclear envelope [18]. In addition, sperm was able to enter ooplasm in human oocyte at any location [23]. Krukowska et al. demonstrated the fusion of female and male chromatin before nuclear envelope formation by injection of a sperm at the vicinity of the metaphase II spindle [22]. Thus, diploid single-nucleated zygotes are more common in cIVF than in ICSI. The diploid 1PN zygotes in ICSI may be derived from same origin in cIVF; however, the rate of haploid 1PN zygotes and embryos were higher than in cIVF [9, 11, 21]. This was because most of ICSI 1PN zygotes were parthenogenesis. In addition, the 1PN zygotes in ICSI included one developed prominent female PN and the sperm head which either decondensed and subsequently formed a small male PN of <5-μm diameter or did not decondense in cytoplasm [17]. Thus, ICSI 1PN zygotes poorly developed to blastocysts compared with cIVF 1PN. Therefore, all 1PN zygotes were cultured until the blastocyst stage and only fully developed or expanded blastocysts were cryopreserved and transferred. As a result, higher clinical pregnancy and ongoing pregnancy rates were observed in cIVF.

As our main goal of this study was the clinical and neonatal outcomes from 1PN embryos, we divided the F-TBT cycles of blastocysts into four groups (Table 3). Our data showed that most of the F-TBTs were performed on 2PN blastocysts in cIVF and ICSI because of the lower priority of the 1PN blastocysts. However, our results which in 1PN blastocysts in cIVF, and not in ICSI, showed a similar rate of clinical outcomes compared with 2PN blastocysts. In contrast, Staessen et al. [4] and Reichman et al. [5] demonstrated lower implantation and clinical pregnancy rates after the transfer of cIVF or ICSI cleavage-stage embryos on day 3. This suggests that it is difficult to select the embryos which have adequate potential for implantation on day 3.

The neonatal data of live babies had no significant difference in the birth weight and weeks of gestation between 1PN in cIVF and control groups (Table 4). Similar reports on neonatal data derived from 1PN embryos in cIVF and ICSI have been reported in the literature [4, 5, 7, 8]. Gras and Trounson [7] and Reichman et al. [5] have reported similar findings on normal birth weight, the weeks of gestation, and a birth of a healthy child. Staessen et al. [4] and Dasig et al. [8] also described the delivery of healthy babies without neonatal data. Previous studies regarding live birth babies derived from 1PN embryos have reported no major malformations.

In conclusion, we have shown that blastocysts derived from 1PN zygotes may lead to normal healthy newborns. We performed embryo transfer of only 1PN blastocysts after informed consent; 9 cycles led to pregnancy and continued to delivery, and no malformations were detected in the newborns. We have also indicated that it may be possible to expect selection of embryos with normal chromosomes and an adequate ongoing pregnancy rate after blastocyst transfer derived from 1PN zygotes in cIVF cycles. These results suggested that blastocysts derived from 1PN embryos in cIVF are available for clinical use and may lead to an increase in the chance of pregnancy in patients receiving ART with 1PN embryos that were considered unsuitable for transfer before. However, there were few cases to ensure the safety of transferring 1PN blastocysts in this report, so it will be necessary to increase the number of cases and to investigate the relationship between major malformations and 1PN.