Iatrogenic multiple pregnancy: The risk of ART
The common denominator of most assisted reproduction techniques (ART) is ovarian (hyper)stimulation. In IVF the number of zygotes transferred in ART has been always under control. Consequently, multiple pregnancies following ART are almost exclusively physician-made, i.e., iatrogenic multiple pregnancies (IMPs). There are two exceptions to this statement. First, single embryo transfer may still be associated with an increased risk of monozygotic (MZ) twins since ART augments the rate of zygotic splitting.
Regardless of the mechanism involved in iatrogenic multiple pregnancies , ART undoubtedly increases the risk of multiple birth. A recent survey reports on a 25% twin and 5% triplets frequency following transfer of three embryos. Roughly, these reference figures represent a 20- and 50-times increased frequency for iatrogenic twins and triplets, respectively, as compared with naturally occurring multiples.
The Pregnancy
It is generally accepted that the human female is programmed for mono-ovulation, mono-fetal development, and nursing only one neonate. Consequently, pregnancies with more than one fetus overwhelm the uterine capacity to adequately nurture the fetuses. Animal and human models have repeatedly demonstrated the reciprocal relationship between birth weight and gestational age at delivery and litter size. Using singleton standards, a significant proportion of twins and all high-order multiple pregnancies (HOMPs) will be delivered preterm and will be small for gestational age. In addition to absolute growth restriction, relative (discordant) growth is common.
As a result of the limited uterine capacity, natural reduction in fetal number is frequently seen. At the early stages, the embryo may disappear (“vanishing twin syndrome”) in one of every 6-7 twin pregnancies following ART. The vanishing twin syndrome, considered by many as natural multifetal pregnancy reduction (MFPR), has recently gained special attention when Pharoah and Cooke hypothesized that single embryonic death may be implicated in cerebral palsy in the survivor.
Multiples are associated with higher frequencies of malformations of varied etiology. The yet unknown factor(s) that cause zygotic splitting has been implicated in causing structural malformations in MZs. In the subset with monochorionic (MC) placentas, also encountered in high-order multiple pregnancies, twin-twin transfusion syndrome (TTTS) may affect as many as 10-15% of the pairs and result in major morbidity of one or both twins. Later in pregnancy, single fetal demise associated with monochorionic placentas may result in severe end-organ damage in the survivor.
Finally, it has been shown that the risk of cerebral palsy (CP) is 5-6-, 23-fold increased in twins and triplets as compared with singletons. A model based on British data related to transfer of two and three embryos and on British data related to CP in multiples suggested a significantly lower estimated CP rate (2.7 / 1000 neonates) after spontaneous pregnancies as compared with transfer of three embryos (OR 6.3), two embryos (OR 3.3), and transfer of three embryos in which all triplets have been reduced to twins (OR 3.8).
Minimal mortality rates in various multifetal pregnancy reduction combinations.
MFPR 4 → 2 5 → 2 3 → 2 4 → 1 3 → 1 2 → 1
Minimal mortality 50% 60% 33% 75% 66% 50%
Considering all the risks associated with iatrogenic multiple pregnancies, one undoubtedly should prefer a singleton to a multiple pregnancy. To minimize risks, no more than a single embryo should be transferred. This policy has been implemented in recent years in many countries and a full discussion of the results is beyond the scope of this chapter. In general, the balance between the risk of multiples, the success rate of single embryo transfer, and the potential need for re-imbursement of additional cycles has been considered and the net result seem to favor single embryo transfer.
Iatrogenic multiple pregnancies following ART is usually achieved after longstanding infertility and is usually the “end-stage” procedure. At this phase of reproductive life, most couples would consider a multiple pregnancy as compensation for their efforts. No wonder that most couples will support, or even persuade the physician, to increase the chances of pregnancy by increasing the number of transferred embryos.
The epidemic of iatrogenic high-order multiple pregnancies enabled some insight into the increased maternal morbidity in these cases. The most significant morbidity found in triplets was pregnancy-induced hypertension (27-33%), HELLP syndrome (9-10.5%), anemia (27-58.1%) and postpartum hemorrhage (9-12.3%).
Since maternal morbidity clearly increases with plurality, it is expected that maternal morbidity will decrease following multifetal pregnancy reduction .
Since all pregnancies were successfully implanted triplets, this finding suggests that plurality and placental mass are probably more important to the development of preeclampsia than is successful implantation alone. Similar findings were reported on gestational diabetes.
Maternal morbidity should also be considered in the context of maternal age. ART enabled pregnancies beyond the range of reproductive years, when underlying diseases are more common and pregnancy complications are expected to be intensified.
Data from the United States national Center for Health Statistics and the Center for Disease Control (NCHS/CDC Press release, September 14, 1999) suggest that (1) between 1980-82 and 1995-97 the twin birth rate rose 63% for women between the ages 40 and 44 and nearly 1000% for women 45-49 years of age; (2) High-order multiple pregnancies birth rate rose nearly 400% for women in their thirties and more than 1000% for women in their forties. In 1997 there were more twins born to women ages 45-49 than during the whole decade of the 1980’s. Obviously, motherhood at or beyond the edge of reproductive age is a new aspect of what clinicians previously referred to as pregnancy in the "older gravida". With ART, the boundary between "old" and "young" no longer exists. Generally, the majority of the published studies have been unanimous about the special, and perhaps the super-cautious attitude required for the older mother, an approach that translates to higher rates of peripartum interventions. This is despite the fact that although some complications may occur more frequently in older mothers as a result of accumulated prior diseases, there is no direct evidence that older age per se complicates either gestation or parturition. Quite unexpectedly, Keith et al found that older age has an advantage of better perinatal outcome (mainly in terms of birth weight) of twins and triplets. It is unclear if this is a result of a better socio-economic status of older mothers or if it is related to some uterine "programming" effect.
By contrast to these sky rocketing rates, there are few series describing such “geriatric gravidas”, and therefore, the true prevalence of various complications may be underestimated. In one study, 4.5 + 1.1 cleaving embryos were transferred per cycle to 45-59 years old patients, resulting in 74 delivered pregnancies (34.9%). There were 29 (39.2%) multiple gestations, including 20 twins, 7 triplets, and two quadruplets. Two of the triplet and both of the quadruplet pregnancies underwent multifetal pregnancy reduction to twins. Antenatal complications occurred in 28 women (37.8%) including preterm labor, hypertension, diabetes, preeclampsia, HELLP syndrome, and fetal growth retardation. Cesarean section was done in 64.8%.
The age-related risk for trisomy, depending on the source of the female gametes, is of primary importance when ART is performed in the elderly. For those who conceive without donor eggs, this risk might be exceptionally high. However, in the case of a polyzyogtic multiple gestation, the risk of pregnancy loss following cytogenetic studies might be unacceptably high. Thus, the timing of these studies becomes pertinent. In countries where fetocide is permitted only before the 24th week of gestation – the only options are first trimester CVS or second trimester amniocetesis. In some countries, fetocide is not restricted to gestational age, and late fetocide is a clear option. In such instances, amniocentesis is scheduled during the 30th-32nd week, with the possibility of fetocide at 33-35 weeks. This logical scheme eliminates the risk of losing the entire pregnancy at an unsalvageable age. However, this scheme provokes two major problems. First, the patient might deliver during the time interval before the cytogenetic results. Second, legitimization of third trimester fetocide is a formidable ethical dilemma and does not imply that physicians will agree to terminate a viable fetus. These intricacies may be settled if pre-implantation diagnosis will become a useful option.
Surrogate motherhood is a good example how ART may change all we know about IMP: consider the “Angela” case, in which two embryos of unrelated couples were transferred to a surrogate uterus. The newborn twins, whose parentage was confirmed postpartum, were non-siblings who shared no common genes and, of course, shared nothing with the surrogate mother.
It goes without saying that the most common and the most risky complication of multiple pregnancies is preterm birth – for which no remedy is available. However, Irrespective of plurality, an association between preterm birth and ART has long been suspected and was related to causes such as iatrogenic preterm birth (in the so-called 'premium' pregnancies), fertility history, past obstetric performance and to underlying medical conditions of the female partner. Recent data showed that singleton as well as in multiple pregnancies resulting from IVF have increased rates of preterm birth compared with naturally conceived pregnancies. The most plausible explanation seems to be a more liberal use of elective preterm birth. In any case, the most appropriate endpoint after ART should also include preterm or term birth as measure of success.
Finally, the patient with IMP should be also considered in evolutionary terms. Innumerable studies have shown that over the millenia, evolutionary forces selected a female prototype for spontaneous twins. Black, fertile, older, taller, and heavily built women are more likely to have twins and the outcome is likely be better than in women with other characteristics. Thus, the fact that ART involves no selection (except fertility), and certainly no selection for motherhood of multiples, makes the IMP in many ways a iatrogenic contra-evolutionary phenomenon.
The physician
Three types of physicians comprise the third part of the iatrogenic multiple pregnancies triangle: those involved in ART, those caring for maternal-fetal issues, and the pediatricians. Each is charge of a different phase.
The reproduction phase.
Since there seems to be a direct relation between the number of transferred embryos and success rate of ART on one hand and the iatrogenic multiple pregnancies rate on the other hand, there seems to be an inherent conflict in the reproduction phase. An idea about the anticipated rates of iatrogenic multiple pregnancies comes from centers in which all available embryos were transferred and multifetal pregnancy reduction is not used. Before the implementation of the 2004 Italian reproduction law, the Reggio Emilia (Italy) center for reproductive medicine observed that 34.6% of the clinical pregnancies were multiples, comprising 20% twins and 14.6% high order multiple pregnancies. Interestingly, implementation of the Italian Reproductive Law from 2004, that limited the number of fertilized oocytes to 3, but obliged to transfer all embryos, did not significantly changed the incidence of multiples and somewhat improved the overall outcome. Some concern exists however regarding the age group of patients over 38 years.
Ethical, legal, religious, and technical (i.e., availability of cryopreservation) constraints that obviate selection and or disposal of surplus embryos, is the easy way for deciding on the number of embryos that should be transferred. The hard way is careful analysis of success (live birth) vs. failure (iatrogenic multiple pregnancies) rates using selected embryos. Genetical and biochemical markers would supplement morphological criteria as normal-appearing embryos may be genetically abnormal. Pre-implantation genetic studies may also replace invasive procedures during pregnancy following ART. For the time being, the first step has already been done by implementing elective single embryo transfer in several countries without significantly reducing outcomes.
Many of the recommendations have been based on embryo transfer without specifying their quality and their implantation potential. At the mean time, it has become possible to culture embryos to the blastocyst stage, selecting the fittest embryos for transfer and synchronizing the embryonic with the endometrial stages. Blastocyst transfer has been associated with a much-improved implantation rate than that of 3-days embryos. It is expected that the high “take-home baby” rate following the excellent implantation rates would lead to transfer of one or two blastocysts only, with concomitant reduction of the iatrogenic multiple pregnancies rate. However, not all embryos will become blastocysts and it is unknown which dividing embryo will become a blastocyst in vitro. Thus, physicians may not wait for the 5-day stage and will first transfer 3-days embryos and then, when blastocyst are successfully cultured, will transfer additional blastocysts, generating iatrogenic superfecundations.
To date, there are no data regarding the consequences of such protocols. Logically, mixed-stage embryo transfers will necessarily increase the chance of iatrogenic multiple pregnancies by adding the successful implantation of the 5-days to that of the 3-days embryo(s). In addition, we do not know the influence of co-implantation at different embryonic ages on the risk of zygotic splitting. We, as well as others, noticed some bizarre complex chorionicity arrangements, which have never been seen with usual IVF-ET protocols.
It is therefore reasonable to conclude that demands from infertile couples and fertility clinics to maximize success rates conflict the need to reduce the number of iatrogenic multiple pregnancies.
The pregnancy phase.
Once pregnant, the woman is not infertile anymore and there should be no difference in the management of spontaneous as compared to iatrogenic pregnancies. However, the past reproductive history continues to follow the patient albeit her pregnancy may be absolutely normal. When a iatrogenic multiple pregnancies results, the designation of ‘premium gestation’ seems appropriate and most reproduction experts may refer the patient to a clinician involved in maternal-fetal medicine (MFM) conducting high-risk pregnancy clinics.
It is not yet accepted who should treat the iatrogenic multiple pregnancies. Obviously, many subspecialties are involved, for example – the sonographer who makes the diagnosis may not be the one who will carry out the multifetal pregnancy reduction, and both may not take care of the preeclamptic patient. This complicated pregnancy follow-up is therefore never a one-man show and a well-orchestrated teamwork is encouraged. Indeed, it has been shown that special multiple pregnancy clinics do have better results.
It seems there is never a dull moment in caring for the mother with multiples, exemplified by conflicts between maternal condition and continuation of pregnancy. The lack of effective prophylactic measures against preterm labor and the risks associated with tocolysis is a good example of how the physiologic adaptation for a multiple gestation may complicate treatment with beta-mimetic drugs or with MgSO4. Thus, the risk in arresting preterm labor (to the mother) may be as significant as the risk (to the neonate) of delivering premature multiples.
“Term” in singletons is different than in twins or in high-order multiple pregnancies. Thus, it seems futile to aim for 38 week’s gestation in multiples just to conclude that this target is unattainable. It follows that a realistic gestational age based on related survival and morbidity rates should be set. For example, obstetricians should aim for 30 week’s gestation if their neonatal service provides good outcome for neonates at this age. Thus, it seems reasonable to suggest that if prematurity in multiples is not preventable, efforts should be made to prevent extreme prematurity.
Finally, a time comes when the obstetrician and the patient consider the mode of delivery. There is little doubt that a planned (daytime), elective cesarean delivery offers a simple solution in terms of required personnel and safety to mother and neonates. This seems to be intuitively true for high-order multiple pregnancies and for small twins, although there are no prospective studies to support this assumption. For twins weighing at least 1500 g each, either route of delivery seems to be appropriate, irrespective of fetal presentation. However, as mentioned above, iatrogenic multiple pregnancies are frequently considered as “premium”, high-risk pregnancies, and many will follow the dictum that “no high-risk pregnancy should end with a high-risk delivery” and opt for an elective abdominal birth.
High-order multiple pregnancies births are at much greater risk than single births. An NCHS report on the final 1996 birth statistics for the USA, found that infant mortality are 12-times higher for triplets than for singletons, triplets are 12-times more likely to die within the first year of life, the average birth weight of a triplet baby is half that of a singleton, and the gestational duration is, on the average, 7 weeks shorter. For 1995, 92% of triplets were preterm as compared with about 10% of births in single deliveries.
Delivery of a multiple pregnancy should be a carefully planned event. A minimal neonatal team for a triplet delivery may include as many as 10 persons, including physicians, assistants, and a supervisor. Obviously, chaos prevails unless teamwork is harmonized.
Logistic considerations do not end at delivery. Once at the nursery, each of the multiples must be given equal opportunity to bond with his parents and, perhaps, according to psychological view, to continue its intra-uterine contacts with its sibs. For instance, there is increasing evidence that co-bedding of twins in the neonatal intensive care unit (NICU) improves thermoregulation, feeding, and sleeping parameters. Indeed, the special and unique interaction between multiples during childhood and beyond, seems to reflect the unique relationship that exists between fetuses that grow together in utero. Thus, parents of a multiple pregnancy are more likely to experience bereavement than those with singletons. The care that parents should receive when all fetuses/babies die is not different from that when a singleton dies. When one baby of a multiple birth dies, the loss is frequently underestimated; however, the loss of parents that are left “with something” is no less painful.
One should consider the international consensus statement on the perinatal care of multiples (Appendix) where many aspects related to ART are discussed. In any case, the apocalyptic views expressed in this chapter will remain pertinent as long as demands for better pregnancy rates by couples undergoing ART will be accepted by overzealous reproduction centers without a clear definition of what should be considered successful.
Issac Blickstein MD
Kaplan Medical Center
Birth Defects
Press Release
National Birth Defects Prevention Study Shows Assisted Reproductive Technology is Associated with an Increased Risk of Certain Birth Defects
November 17, 2008
CDC Online Newsroom phone (404) 639-3286
Infants conceived with Assisted Reproductive Technology (ART, http://www.cdc.gov/ART/index.htm) are two to four times more likely to have certain types of birth defects than children conceived naturally, according to a study by the Centers for Disease Control and Prevention (CDC, 1600 Clifton Rd. Atlanta, GA 30333, USA). The report, “Assisted Reproductive Technology and Major Structural Birth Defects, United States,” was released in the journal Human Reproduction.
“Today, more than 1 percent of infants are conceived through ART and this number may continue to increase,” says Jennita Reefhuis, Ph.D., epidemiologist at CDC′s National Center on Birth Defects and Developmental Disabilities. “While the risk is low, it is still important for parents who are considering using ART to think about all of the potential risks and benefits of this technology.”
The study shows that among pregnancies resulting in a single birth, ART (which includes all fertility treatments in which both eggs and sperm are handled, such as in vitro fertilization) was associated with twice the risk of some types of heart defects, more than twice the risk of cleft lip with or without cleft palate and over four times the risk of certain gastrointestinal defects compared with babies conceived without fertility treatments. Despite these findings, the absolute risk of any individual birth defect remains low. In the United States, cleft lip with or without palate affects approximately 1 in every 950 births; doubling the risk among infants conceived by ART would result in approximately 1 in every 425 infants being affected by cleft lip with or without palate.
The study examined multiple births separately from single births because ART increases the chance of a multiple birth. Children born as part of a multiple birth are more likely to have a birth defect regardless of use of ART. The study showed use of ART did not significantly increase the risk of birth defects among multiple births.
However, ART might contribute to the risk of major birth defects by directly increasing the risk of defects among single births. It may also have an indirect impact because ART increases the likelihood of having twins, which is a risk factor for many types of birth defects. Researchers believe this suggests the need for further studies to determine risk for ART in pregnancies with multiple births.
The study examined data from 281 births conceived with ART and 14,095 conceived without infertility treatments. The National Birth Defects Prevention Study is a population-based study that currently incorporates data from birth defects research centers in Arkansas, California, Georgia, Iowa, Massachusetts, New Jersey, New York, North Carolina, Texas and Utah. These ten centers have been working on the largest study of birth defects causes ever undertaken in the United States. Information is gathered from more than 30,000 participants to look at key questions on potential causes of birth defects. While the causes of most birth defects are unknown, studies show that smoking, alcohol, and obesity increase a mother's risk of having a child with a birth defect.
Since 1981, ART has been used in the United States to help women become pregnant. It is defined as any procedure that involves surgically removing eggs from a woman’s ovaries, combining them with sperm in the laboratory, and returning them to the woman’s body or donating them to another woman. ART does not include treatments in which only sperm are handled (i.e., intrauterine—or artificial—insemination) or procedures in which a woman takes medicine only to stimulate egg production without the intention of having eggs retrieved.
The number of infants born after ART doubled in the United States from 1996 through 2004. According to data from the 2002 National Survey of Family Growth, almost 12 percent of U.S. women aged 15-44 years have reported using infertility services. In 2005, more than 134,000 ART procedures were performed and approximately 52,000 infants were born as a result of these procedures.
