The clinical outcomes of selective and spontaneous fetal reduction of twins to a singleton pregnancy in the first trimester: a retrospective study of 10 years

Multifetal pregnancy is prevalent in assisted reproduction technology (ART), which can result in a series of complications such as gestational diabetes, hypertensive disorders, premature rupture of membranes, premature birth, etc. [1‐4]. Although the incidence rate of multifetal pregnancy differs among all reproductive centers worldwide, ART has contributed to a substantial proportion of all twins, triplets, and higher-order infants born [5]. To decrease the incidence of multifetal pregnancies, single embryo transfer (SET) has received increasing attention in recent years. However, in a large number of cases, more than one embryo is still transferred into the uterus due to patient requests or the pursuit of a successful pregnancy in one transfer cycle by reproductive centers or doctors. The rate of multifetal pregnancies remains high. In 2018, the European Society of Human Reproduction and Embryology (ESHRE) found that the proportions of singleton, twin, and triplet deliveries after in vitro fertilization (IVF)/ intracytoplasmic sperm injection (ICSI) were 82.5%, 17.5%, and 0.5%, respectively [6]. To reduce the risk of complications in multifetal pregnancies, a technology known as selective fetal reduction (SEFR), was developed [7]. This is a method to reduce the number of fetuses through surgery. Moreover, the incidence of spontaneous fetal reduction (SPFR) can also reduce multifetal pregnancies to singleton pregnancies. Luo reported that the incidence of SPFR was 17.1% in dichorionic diamniotic twin pregnancies [8]. Unfortunately, its occurrence is difficult to predict based on existing research. Regardless of the method, double embryo transfer (DET) with subsequent selective and spontaneous fetal reduction can achieve singleton pregnancy status, but in either method, an inanimate fetus remains in the uterus. It is unclear whether an inanimate fetus would adversely affect another fetus or the mother. Therefore, it is unclear whether DET with subsequent SEFR or SPFR should replace SET. Research on these issues is rare. The current study collected a large amount of data and compared the clinical outcomes, including pregnancy outcomes, pregnancy complications, and newborn outcomes, of singleton pregnancies from SEFR, SPFR, and SET.

This study collected a large amount of data over 10 years. We compared the obstetric and neonatal outcomes of 5922 pregnancies, including 4667 singletons following SET, and twins following DET and subsequently reduced either spontaneously (SPFR, 865 cases) or surgically (SEFR, 390 cases). We found that SET resulted in an overall better outcome than either SPFR or SEFR, with lower risk of several pregnancy complications as well as better results for gestational age and birth-weight.

There is no dispute that twin or higher-order pregnancies can result in many complications. Thus, an increasing number of reproductive centers have begun to encourage SETs. While SETs have decreased the multifetal pregnancy rate, the rate is still higher than that in natural conception, whether in Europe, America, or China [5, 6, 9]. In our center, SET is the first choice for specific patients, as listed in the Method section. Similar to the population seen at our center, patients who do not meet these conditions comprise a large proportion of all patients. In these populations, SET is thought to decrease the incidence of complications. However, the success of ART treatment—that is, a successful pregnancy —is also an important target and should be taken into consideration in the protocol selection process. One meta-analysis reported that the clinical pregnancy rate may be lower in women who had SETs than in those who had DET, and the live birth rate may also be reduced in those with SET [10]. Furthermore, repeated treatment, advancing age and urgency to become pregnant are factors that moderate a woman's choice of SET, ignoring the risks of multifetal pregnancies [11]. In addition, some patients were seeking twins and wished to minimize physical and psychological stress by having as few IVF treatments as possible [12]. Of course, many physicians or centers did not offer SET due to their pursuit of a successful rate in one transfer cycle [11, 13]. For doctors or patients, weighing the safety and success rate of assisted reproductive technology may be a difficult task. These factors explain why DET is still popular.

For pregnant women who are not suitable or are unwilling to have a twin pregnancy, in addition to SET, singleton pregnancy can be achieved in two ways: waiting for SPFR or performing surgical SEFR if a multifetal pregnancy occurs. The SPFR rate in dichorionic diamniotic twin pregnancies was approximately 17.1%, with most cases occurring in the first trimester [8]. In all multifetal pregnancies, the rate was 38% [14]. Limited studies have reported that the risk factors for SPFR may be related to age, endometrial thickness on the day of hCG administration, and the initial number of gestational sacs [15‐17]. Despite these studies, SPFR remains largely unpredictable. Waiting for the occurrence of SPFR blindly will delay the SEFR procedure which may lead to increased risks from SEFR surgery [18, 19]. Surgical SEFR technology is more flexible, as doctors can decide when to conduct it and which fetus should be reduced [20]. In terms of the technology itself, SEFR has already been proven to have a success rate of almost 100% [21].

Many previous studies have demonstrated that SPFR and SEFR had better or comparable perinatal outcomes than those in the non-reduced multifetal pregnancy group [8, 22‐24]. However, only a few studies have focused on the outcomes of twin pregnancies reduced by SPFR/SEFR compared to original singleton pregnancies. In the current study, compared to original singleton pregnancies, SPFR led to a similar miscarriage rate but a poor live birth rate and newborn birth weight. In a study of dichorionic diamniotic twin pregnancies, Luo further reported that the live birth rate, take-home baby rate, neonatal birth weight, and other primary outcome measures in the singleton after SPFR were not inferior to those in cases of original singleton pregnancies [8]. Luo’s work included pregnancies where SPFR occurred ≤ 7 weeks of gestational age (42.1%), 8–12 weeks (47.1%) and 12–27 weeks (9.4%); by contrast, our study only included pregnancies where SPFR occurred within 12 weeks of gestational age. Although, the time of fetal reduction was divided into different groups by Luo, no further analysis of the differences in outcomes between the groups was carried out. Pinborg et al. believed that SPFR occurring at > 8 weeks of gestational age was related to a higher risk of adverse obstetric outcomes in IVF singletons [25]. It should be noted that, in all studies, for SPFR, there may be an interval between the recorded fetal reduction time and the real fetal reduction time, due to the patients’ delayed examination. Further, the singleton pregnancies in the current study were from SET, but in Luo’s and Pinborg’s studies, they may be from SET, DET, or higher embryos transfer. It has been revealed that the preterm birth rate, extremely preterm birth rate, low-birth weight rate and very low-birth weight rate of singleton pregnancies increased with the increase of embryo transfer numbers [26]. A similar conclusion was reached by Poikkeus’s study [27]. As for SEFR, Cheang et al. indicated that twin pregnancies from SEFR had a higher incidence of extreme-prematurity, prematurity, and lower birth weight than twin pregnancies without fetal reduction [28]. Similarly, SEFR in this study also reduced the length of gestational age and newborn birth weight. Meanwhile, SEFR increased the miscarriage rate and reduced the live birth rate compared to those with original singleton pregnancies. It is difficult to find literatures focusing on the differences in pregnancy complications between the original singleton group and the SPFR/SEFR group. Our research provides evidence that SPFR increases the rate of GDM, PPROM, and abnormal amniotic fluid volume. Neither SPFR nor SEFR led to a change in the risk of gestational hypertension and placental abnormalities. In general, these comparisons mean that DET with fetal reduction may have a negative influence on the remaining fetal development and maintenance of pregnancies.

SET decreases the incidence of complications from multifetal pregnancies. In terms of success, DET increases the possibility of a live birth in one transfer cycle. Considering all available embryos, SET offers a better or similar overall pregnancy rate than DET [10, 29]. The SET success rate has been rising with the development of ART, and efforts have been made to establish SET as the standard of care [30]. Although SPFR and SEFR can help achieve a singleton pregnancy, SPFR is hardly predictable and SEFR requires additional surgical operation, which means that patients suffer more pain and a greater financial burden. Moreover, medical providers should be clear that many patients have difficulty confronting the prospect of fetal reduction and they may renege on prior agreements to undergo SEFR [31]. The expanded application of SEFR will also bring about some ethical controversies [32, 33]. This is also one reason why prospective randomized controlled studies are difficult to perform.

The strength of this study is that it’s the first comprehensive analysis of the clinical outcomes of singleton pregnancies derived from SET or DET twin pregnancies with subsequent fetal reduction based on a large sample size. Nonetheless, this study has the following limitations. First, a subgroup analysis of fetal reduction time was not performed because the fetal time was concentrated in a short time period. Second, threatened miscarriage, an important pregnancy complication, was missed due to deficiencies in our follow-up work. Third, a large proportion of cases in this study were followed up by telephone, and some important information was not very precise due to recall bias. Fourth, this is a single-center retrospective study, which prevents the generalizability of our findings.

In summary, compared to an original singleton pregnancy, DET with subsequent fetal reduction by SEFR and SPFR was related to poor clinical outcomes. We recommend that DET with subsequent fetal reduction should only be considered as a rescue method for multiple pregnancy patients with potential complications, and SET is more advisable.