Sex‐discordant twins despite single embryo transfer: a report of two cases

Abstract

We report two extremely rare cases in which the patients delivered male and female infants that were dizygotic twins (DZT) despite single embryo transfer. Case 1: The patient was a 35‐year‐old woman with a 9‐year history of unexplained infertility. In an oocyte retrieval cycle, one blastocyst was transferred; at 26 weeks of gestation, she delivered a 704‐g female infant and a 420‐g male infant by cesarean section. Because both infants were of extremely low birth weight, they were placed in the neonatal intensive care unit. Congenital anomalies were not found in either infant. Case 2: The patient was a 30‐year‐old woman with a 1‐year history of infertility. Hysterosalpingogram revealed bilateral tubal occlusion. In a frozen/thawed cycle one blastocyst was transferred during her natural ovulation cycle. She achieved a pregnancy and delivered a 2,877‐g female infant and a 2,544‐g male infant at 36 weeks of gestation by cesarean section. The female infant was diagnosed with a neural tube defect. No congenital anomalies were detected in the male infant. We hypothesize that the DZTs might have been the result of concurrent embryo transfer and natural ovulation and intercourse.

Introduction

Assisted reproductive technology (ART) has resulted in pregnancies for many infertile patients; however, the procedure poses a number of problems. One shortcoming is an increase in the multiple gestation rate; therefore, the number of embryos transferred is reduced in young patients to avoid this problem [1].

In 2008, the Japan Society of Obstetrics and Gynecology recommended single embryo transfer (SET) to reduce fetal and maternal risk [2]. Since 2005, we have conducted SET to avoid multiple gestation. As a result, our multiple gestation rate has significantly decreased; however, compared with double embryo transfer (DET) [3], the pregnancy rate has not decreased. Nevertheless, even with strict adherence to SET, it is difficult to achieve infant safety because the occurrence of monozygotic twins (MZTs) is approximately 1% with early‐stage embryos transferred on day 3 [4, 5, 6]; with blastocyst stage embryo transfer, the incidence of MZTs is 3–13% [7, 8, 9, 10, 11, 12, 13, 14]. With SET, our incidence of MZT is 5/371 (1.35%); these infants were the same sex (data not shown).

However, we recently encountered two unusual cases of dizygotic twins (DZTs) despite the transfer of only one early cleavage embryo. Therefore, the purpose of this paper is to describe these rare cases and to discuss the avoidance of multiple gestation.

Case 1

The patient was a 35‐year‐old woman who had a 9‐year history of unexplained infertility. Two of four IVF procedures resulted in pregnancy; however, one of the two pregnancies terminated in an intrauterine fetal demise (IUFD) at 24 weeks of gestation, and the other terminated in a spontaneous abortion at 6 weeks of gestation. She tested positive for anti‐lipid antibody (anti‐CL, anti‐β 2GP, anti‐PE, anti‐IgG, anti‐IgM, anti‐PS, and anti‐IgM). Her husband's semen analysis was nearly normal with a sperm count of 35 × 10⁶/ml and a motility of 40%. The patient underwent five IVF cycles; ovarian stimulation was performed via recombinant follicle stimulating hormone (FSH) (rFSH; Follistim^R, Schering‐Plough, USA) 225 IU/day from cycle day 3 to 6; subsequently, from day 7 to 10, human menopausal gonadotropin (HMG) (Juji Pharma, Japan) 150 IU/day was administered. Gonadotropin‐releasing hormone (GnRH) antagonist (Cetrotide^R, AEterna Zentaris, Germany) 0.25 mg/day plus human chorionic gonadotropin (HCG) (Fuji Pharma, Japan) 50 IU/day were administered from day 8 to 10. On day 10, HCG 5,000 IU was administered. We noted 15 follicles on the right ovary and eight follicles on the left ovary; we punctured seven follicles on the right ovary and six follicles on the left ovary and retrieved six oocytes from four right ovarian follicles and two follicles from the left ovary. These were fertilized by conventional IVF (c‐IVF). Two days before oocyte retrieval, the couple had unprotected intercourse.

One blastocyst, which was graded 3BB (Gardner criteria) was transferred. The embryo did not undergo assisted hatching. We recognized a gestational sac (GS) and a fetal heart beat via ultrasound on October 1, 2007. At 5 weeks 2 days we observed one GS via ultrasound. At 7 weeks 2 days we recognized two GSs which presented each of the fetuses in their GSs and dichorionic, diamniotic (DD) twins. The patient developed hypertension, and intrauterine growth retardation was noted on ultrasound examination. She subsequently delivered a 704‐g female infant and a 420‐g male infant by cesarean section at 26 weeks of gestation on March 3, 2008. The DD membrane was visible with the naked eye. Only the male infant underwent chromosome analysis and was found to have a normal 46, XY karyotype. The two infants were transferred to the neonatal intensive care unit (NICU) and subsequently did well.

Case 2

The patient was a 29‐year‐old woman with a 1‐year history of infertility. She underwent hysterosalpingography (HSG) twice, which revealed bilateral tubal occlusion. She did not undergo laparoscopy; however, we decided to accept the HSG findings for diagnostic purposes. Her husband's semen analysis was: sperm count 66 × 10⁶/ml, motility 30–40%. No anti‐sperm antibodies were present. The endocrinologic examinations for both husband and wife were within normal limits. After HSG administration, the patient received ovarian stimulation during a fifth IVF cycle. Intracytoplasmic sperm injection (ICSI) was employed because a low fertilization rate occurred previously with c‐IVF. She underwent four embryo transfer cycles. A pregnancy was achieved in one of the four cycles; however, a spontaneous abortion occurred at 6 weeks of gestation. Ovarian stimulation was performed via the short protocol (rFSH 300 IU/day from cycle days 3 to 4; subsequently rFSH 150 IU/day was administered from day 5 to 12). A GnRH agonist (Buserecur^R, Fuji Pharma, Japan) 900 μg/day nasal spray was administered daily from day 1 to day 12 and HCG 5,000 IU was administered on day 13. We noted 13 follicles on the ovaries and retrieved eight oocytes; five of the eight oocytes were fertilized by ICSI. Four of the five oocytes developed into good quality embryos 3 days after insemination. Although one fresh day 3 embryo was transferred, a pregnancy was not achieved. The remaining three fresh day 3 embryos were subsequently cultured to the blastocyst stage; these blastocysts were vitrified in liquid nitrogen (LN2).

In this fifth treatment cycle, one of the cryopreserved blastocysts was thawed and transferred during a natural menstrual cycle without hormone‐replacement therapy.

The embryo was Gardner's grade 4 BB; assisted hatching was not conducted and a pregnancy was recognized on October 19, 2007. We observed one GS at 5 weeks 0 days gestation, and observed two GSs at 5 weeks 5 days gestation, and we diagnosed DD twins by ultrasound.

The patient delivered a 2,877‐g female infant and a 2,544‐g male infant at 36 weeks of gestation, by caesarean section, on May 29, 2008. The female infant was diagnosed with a neural tube defect. No congenital anomalies were detected in the male infant. Neither infant underwent chromosome analysis. The placenta was found to be dichorionic and DD by the naked eye at delivery.

Discussion

ART has not only significantly increased the pregnancy rate but also the multiple gestation rate; therefore, ART introduces obstetrical complications. The rate of high‐risk infants increases (i.e., intrauterine growth retardation, and fetal anomalies) and these infants require intensive care only available at specialized facilities. The incidence of multiple gestation is higher in ART than in natural pregnancy; the natural incidence is 0.4–0.45%. With ART, the incidence of multiple gestation has been reported to be 0.72–1.88% [4, 5, 6] with day 3 embryo transfer and 2.77–13.2% [7, 8, 9, 10, 11, 12, 13, 14, 15] with day 5 blastocyst transfer. With the delivery of a multiple gestation, we must be cognizant of the increased risk of abnormalities; thus, it is prudent to reduce the multiple gestation rate by transferring a limited number of embryos [6, 16, 17].

In 2008, the Japan Society of Obstetrics and Gynecology recommended SET to reduce fetal and maternal risk [2]. In an SET, twins have homologous genetics (MZT) [18]; however, in this study, we experienced two cases of twins of different sexes despite SET. A single blastocyst was transferred in both cases. In both cases, two embryologists confirmed the other's observation of an SET; furthermore, we manually performed a repeat analysis in our laboratory. We reviewed our notes, which contained the patient name, surplus embryos, and date. Therefore, we assumed that when one embryo was transferred, another embryo implanted in the patient's uterus during the transfer cycle by natural means. In Case 1, the couple engaged in unprotected intercourse 2 days before oocyte retrieval. In Case 2, the couple had no recollection of intercourse; however, both couples routinely engaged in unprotected intercourse. Furthermore, we diagnosed bilateral tubal occlusion by HSG in Case 2. We routinely use the high viscosity oil contrast medium; however, in very rare cases, if the tube contains an exceptionally narrow portion, we sometimes incorrectly diagnose an occlusion. Therefore, this case might be a false‐positive HSG. Another possibility is that after embryo transfer, the embryo divided into two cleaved cell masses via an unknown chromosomal mechanism. However, it is difficult to imagine that two embryos can be produced from triploid or tetraploid embryos. Also, it is difficult to imagine that two embryos, one with XX and the other with 48 XY chromosomes, can develop from XXY, XYY, or XXYY chromosomes. Therefore, the most likely explanation is that when a single embryo was transferred, another embryo implanted in the patient's uterus by natural means. Specifically, spontaneous ovulation occurred at the time of embryo transfer and a natural pregnancy via intercourse occurred. However, Zech et al. [19] recently described a very rare case of monochorionic, DD twins discordant in gender from a naturally conceived pregnancy; this occurred via postzygotic sex chromosome loss in a 47, XXY zygote. Therefore, in the future, we should examine the genetic zygosity of both infants and also conduct a chromosome analysis when a twin pregnancy occurs after ART.

It is not uncommon that infertility patients will achieve a pregnancy while awaiting treatment; the rate has been reported to be <10% [20]. Therefore, it is possible that cases similar to ours will occur at other ART facilities. Furthermore, as a point of caution, DZTs generally become DD twins [21]; however, other reports have described the formation of monochorionic, dizygotic twins (MC‐DZTs) [22]. Therefore, we simply cannot diagnose whether twins are monozygotic or dizygotic from the chorionic aspect only. These cases encompass recent ART controversy in regard to a blood chimera. Even if we transfer only one embryo, we must consider both phenomena, thus, limitation of embryos to transfer could not have an impact for the MZT but for DZT in the ART cycle. This case report could have occurred in ART; however, we have no detailed chromosome analysis of infants. These uncommon but not rare cases can lead to a multiple gestation if recent unprotected intercourse has occurred in this setting. Therefore, it is necessary to advise the patient to avoid intercourse during an ART cycle. We follow the recommendations of Lindsay et al. and advise patients to abstain from unprotected intercourse for 7 days before oocyte retrieval [23]. This policy should significantly reduce the possibility of the multiple gestations presented here from recurring.