Ultrasonography at twelfth weeks’ gestation revealed a DCDA pregnancy (Fig. 2), i.e., two fetuses with one chorion each. The histopathological assessment determined dichorionic placenta weights of 306 g and 437 g, respectively. The examinations during pregnancy, especially ultrasound, were all unremarkable. In the 38th week of pregnancy, two healthy children with regular birth masses were born.

Fig. 2

Dizygotic (a) and monozygotic twinning and placenta (b). Dichorionic-diamniotic (occurrence of 30–34% or 32–35%) and monochorionic-diamniotic (occurrence of < 1% or 21–30%) can have one or two placentas, with cleavages reported at morula or early blastula stages. Monoamniotic monochorial twins always share one placenta (occurrence of 1%) with a late blastula cleavage. Conjoined twinning occurs when the zygote cleaves after 13 days, resulting in embryos that are abnormally fused (not shown). Figure created with BioRender.com following [46]

Standard STR profiling resulted in a perfect match between the DNA profiles of both CVS samples. With clinical certainty that the prenatal material came from two different chorions and thus embryos, the matched DNA profiles provided evidence of an identical twin split that unexpectedly occurred at the late blastula stage. For the identical prenatal DNA profiles, no exclusion of parenthood was found in 23 STRs. The combined statistical calculation yielded an index of 7.6415*1020 for the assumed parenthood and, thus, confirmed the MZ twins as the biological offspring of the parents in question.

The concerned couple had explicitly opted in advance for the implantation of a single embryo. After the gemini pregnancy was known, the reproductive fertility center abroad had provided written assurances that only one embryo had been transferred but otherwise failed to respond to any of the parents’ further inquiries and concerns about the patient’s remaining egg cells and the parallel treatment of another woman with two implants at the clinic. Thus, for the expectant patient, the insemination of foreign embryos seemed more likely than a possible late twinning of the blastula. Suspecting an embryonic mix-up, the parents urged clarifying their relationship to the unborn if possible or, if not, a pregnancy termination. By law, severe physical or mental distress of the expectant mother must be claimed to justify a pregnancy termination after the twelfth week. Therefore, an interdisciplinary counseling team was consulted in the 11th week of pregnancy, consisting of legal, medical, and ethical professionals. Their responsibility was to provide advice and support in the decision-making process of whether to continue or terminate the pregnancy, taking into account the country-specific legislation, medical risks, and ethical considerations. Further, the results on parenthood could not have been expected before the 14th /15th week of pregnancy due to the necessity to cultivate the CVS cells prior to the genetic analysis, which complicated the legal (i.e., later than week 12) and ethical terms (i.e., fully developed fetus). To this end, the presumption of an embryonic mix-up, along with the related doubts and insecurities, caused significant physical and psychological stress for the couple concerned, making a prenatal clarification of parenthood indispensable. Thus, surgical abortion would have been legally permissible, even after the 12th week of pregnancy [20].

To make an informed decision, the expectant parents must be precisely informed about the possibilities and risks of prenatal procedures. In addition, the interdisciplinary council provided and discussed the following alternatives, among others:

(1)

An abortion within the timeline up to the 12th week of pregnancy without genetic testing;

(2)

A genetic test and carrying the pregnancy to term, with the prospect of exchanging the children if they are not related; or,

(3)

A genetic test with the prospect of a legal termination after the 12th week of pregnancy if unrelated.

Concerning the first option, an abortion can be legally performed within the first twelve weeks of pregnancy [20]. However, due to the physician’s duty of care, the patient should be advised against abortion based on an unconfirmed suspicion of embryo mix-up during in vitro fertilization, particularly in view of their desire to have a child. The second option, carrying the pregnancy to term, followed by the subsequent release of the children for adoption between the concerned couples, was also considered. Child exchange followed by adoption would require that the other woman involved was pregnant with the patient’s embryo, would be informed, carry the couple’s child to term, and agree to an exchange of offspring. Option 2 seemed to be very questionable to be feasible in practice, as it would mainly depend on the involuntary willingness of surrogacies for both women (coping with mental and psychological stress to carry out a foreign child), provided excellent cooperation with the IVF center abroad, in addition to data protection issues. Therefore, the third option, the prenatal test for relatedness and possible post-twelfth-week termination, was the remaining choice for the couple, also supported by the counseling board.

For prenatal analysis, fetal material can be obtained invasively from the 10th week of gestation by chorionic villus biopsy [16], from the 15th week by amniocentesis [16], and from the 18-20th week by cordocentesis [28]. Noninvasively, fetal cell-free (cf) DNA can be examined, which is present in the maternal bloodstream from the 9th week onwards [29]. This ‘liquid biopsy’ allows clarification of the paternity of the unborn child without any severe or fatal risks. From the ninth week of pregnancy, the developing fetus has produced so many cells that numerous cf DNA fragments circulate in the mother’s bloodstream [30]. However, in twins, the method has shown an increase in the total fetal fraction in maternal material but with decreased fractions per fetus [31, 32], with possibly indeterminate or uninterpretable results, which have also been observed for non-strain single prenatal tests [30].

Regarding multiple pregnancies, likewise, a paucity of data exists for invasive procedures [33], but these biopsies are common and well-established in prenatal diagnosis of singletons [16, 17]. For cordocentesis, one option is to introduce a needle at the placental cord insertion site to obtain material from the fetus [28]. In amniocenteses, a needle is inserted through the mother’s abdominal wall to remove some amniotic fluid, in which fetal cells are present. However, both surgeries could not be considered due to the late stage of pregnancy (second trimester), leaving CVS as the remaining and recommended procedure. Despite being commonly used, this first-trimester biopsy is still associated with the small but existent procedural risk of miscarriage reported in a range of 0.03–0.3% [29, 34]. During this surgical procedure, a small piece of the placenta is removed, from which the embryo forms half of the tissue that can be further investigated for its genetic origin.

In the presented parenthood clarification, another question was raised, namely whether the chorionic villi of both embryos could and should be collected and if – due to close proximity – the same individual could be sampled twice by chance. As MZ twins share the genetic origin, their DNA profiles are expected to be identical [35]. Hence, STR analysis would not reveal whether CVS of one or both MZ fetuses were taken. By DNA sequence analysis, however, existing genome differences, e.g., rare de novo mutations between germlines [26], could be identified and potentially used for MZ twin differentiation [13, 27]. Nevertheless, the probability of genetic relatedness to only one fetus was considered very low, which could have originated by spontaneous conception before ART or due to contamination with the zygote of the other parents. After consultation with the medical physician for CVS sampling, differentiation of the unborn twins, although not easy, was possible without reasonable doubt. With an error considered minor, the chorionic villi provided were assumed to derive from two individuals with proven parentage for the twin pair.

Prior to the discussion of prenatal sampling and genetic analysis, another less obvious challenge and a potential source of mistake was also addressed: prenatal twin determination with potential pitfalls in the accuracy of chorioamnionicity [36]. Errors in twin sonographic interpretation can happen, which in the presented case could have falsely led to the DCDA designation, which reinforced the doubt of correct embryonic implantation as one factor. For example, Lu et al. described twins at week 12 initially believed to be DCDA based on a thick “intertwin membrane”. However, the second ultrasound revealed that this “membrane” was an intrauterine septum. Thus, the true intertwin membrane had only two layers of amnion, revealing monozygotic twinning. Another error could be caused by a rare occurrence of a bilobate placenta in monochorionic twins (3% of monozygotic twinning [37]). The phenomenon describes two nearly equal-sized placental lobes connected by chorion leave, mimicking dichorionicity. Taking into account a possible misinterpretation of the DCDA assignment, an MDMA or MDMC twin type would have been more likely for the affected couple with respect to the blastula split time point, which would have theoretically made the prenatal test unnecessary.

For late twinning of the blastula, it has been suggested that the number of incidents increases due to MAR. For example, ovulation induction [38], patient age [39, 40], or duration of cultivation [41,42,43,44] were shown to have a positive impact on the occurrence of multiplicity pregnancies. Others proposed that the high incidence of monozygocity in infertility-clinic patients was conditioned by hereditary factors, enhanced by excellent ovarian function [45]. However, during preliminary medical clarification, no monozygotic or dizygotic twinning in the parents’ families was observed within the last four generations. Thus, late twinning at the blastula stage was the fortunate explanation for the observed twin pregnancy reported here.