Portal de Periódicos da CAPES

Monochorionicity in perspective

2006; Wiley; Volume: 27; Issue: 3 Linguagem: Inglês

10.1002/uog.2730

1469-0705

Isaac Blickstein,

Pregnancy and preeclampsia studies

Some of the risks related to monochorionicity are, to say the least, fascinating. It is no wonder that many studies, some of which are very sophisticated, have been carried out in relation to very complex but extremely rare situations2, 3, while the perinatal risks of so-called 'uncomplicated' monochorionic twin gestations received relatively little attention4, 5. Having said that, it seems important to place the monochorionic pregnancy in the proper perspective. First, monochorionicity is rare. In natural conceptions monozygotic twins occur in about one third of all twin pregnancies, i.e. 4 : 1000 births, and about two thirds of these are monochorionic, i.e. 2–3 : 1000 births. Although the total number of twins increases and although zygotic splitting is higher following all assisted reproduction methods, the prevalence of monochorionic twins is much lower in a case-mix of spontaneous and iatrogenic conceptions (1 : 15–20 compared with 1 : 3 in spontaneous pregnancies) because of the higher occurrence of polyzygotic pregnancies following assisted reproduction. Thus, the commonly cited prevalence of 4 : 1000 may actually be as low as 1–2 : 1000 births. Second, relatively few (10–20%) monochorionic twins exhibit the most common related syndrome, twin–twin transfusion (TTTS), giving a prevalence of 1–3 : 10 000 births in a mixed population. Other syndromes, such as twin-reversed arterial perfusion (TRAP) are even more rare. Finally, complications related to intrauterine growth restriction and prematurity are more frequent by far, and therefore carry a much higher absolute risk for short- and long-term adverse outcome compared with the seemingly more esoteric or exotic syndromes5. A single fetus is the ultimate result of a single zygote in more than 99% of natural conceptions. Rarely, a deviation from this ordinary sequence may occur, with the zygote splitting and resulting in a serious reproductive anomaly, monozygotic multiple gestation. The anomalous characteristics of this deviation are reflected in the higher incidence of malformations in monozygotic compared with dizygotic or singleton gestations, as well as in the unique anomalies that are found exclusively in monozygotic twins. Zygotic splitting implies some form of separation. However, it is the level of residual sharing that bears a direct relationship to the type of monozygosity. The minimum level of sharing is that of the genome, whereby the separated embryos have essentially the same genetic identity. This led to the designation of 'identical' twinning, although genetic as well as phenotypic differences usually exist. The second level of sharing is the mutual possession of a common placenta and chorion (while retaining individual amniotic cavities), hence the designation of monochorionic twins. The sharing of a single placenta, an anomaly in itself, is frequently associated with unequal vascular territories for each twin and eccentric insertions of the umbilical cords. Moreover, as a result of this abnormality, monochorionic embryos/fetuses are invariably connected by transplacental vascular anastomoses that lead inevitably to sharing of the circulation. Fortunately, the shared circulation is relatively balanced, but in as many as 10–20% of monochorionic twins, an unbalanced intertwin shunt may be created, leading to so-called TTTS. More rarely, one twin might be supported entirely by its co-twin via an arterioarterial shunt, leading to the so-called TRAP sequence, a situation commonly associated with one of the most bizarre anomalies, the acardiac, acephalic twin. The third level involves sharing of all placental structures, whereby less than 1% of monozygotic conceptions also share the same amnion. Before the era of ultrasound, such monoamniotic conceptions rarely survived because of fatal cord entanglement. Finally, the maximum level of sharing is seen in the various presentations of conjoined twins. Incidence is inversely related to the level of sharing: the higher the level of sharing, the lower the incidence. Whereas the incidence of monozygotic twins in spontaneous conceptions is roughly 1 : 250, the incidence of monochorionic twins is about 2/3 that of monozygotic conceptions (1 : 350–400 births), that of monochorionic monoamniotic twins is less than 1 in 2500 births, and that of conjoined twins is less than 1 : 40 000 births. The risk of perinatal morbidity and mortality is related directly to the level of sharing: the higher the level of sharing, the greater the risk of an adverse outcome. Whereas monozygotic dichorionic twins and dizygotic twins have a similar outcome, the risk of an adverse outcome is substantially increased in monochorionic twins, especially in those with an unbalanced shared circulation, and it is vastly increased in the rare monochorionic monoamniotic and in the extremely rare conjoined varieties. In clinical practice, it is possible to differentiate monochorionic from dichorionic twinning, either by antenatal sonography or by postpartum placental examination. At the same time, it is not possible to determine zygosity in nearly 50% of the cases, because like-sex dizygotic (1/2 of all dizygotic twins) and monozygotic dichorionic (1/3 of all monozygotic) twins cannot be differentiated without sophisticated genetic analyses6. However, as most pathology is related to monochorionic twinning, the role of early sonography in establishing chorionicity has become a keystone in the management of multiple gestations. The mechanism of zygotic splitting is unknown. All current speculations are based not on experimental evidence but on educated deductions7-11. Several theories have been proposed to explain the division of the fertilized oocyte. The first proposes that every embryo forms an axis, along which it develops. When more than one axis develops, the dominant axis causes the regression of the other(s). According to this theory, monozygotic twining occurs when two axes co-dominate11. The second and more popular theory suggests that early in embryonic development, cells exhibit some genetic differences that cause them to repel each other7, 8. As attractive as they may be, both theories are unable to explain why monozygotic splitting rates are almost constant in all races, both are unsuitable to explain the formation of monozygotic triplets and quadruplets, and neither is able to explain the invariably increased incidence following all methods of assisted conception. With the advent of assisted reproduction, a third theory was proposed, suggesting that a gap in the zona pellucida is created during the in-vitro process, leading to herniation of blastomeres through the gap and subsequent division of the cells. This pseudo-mechanical theory is implausible because it does not explain why ovulation induction alone is associated with a much higher rate of zygotic splitting compared with pregnancies following micromanipulation or assisted hatching, in which the zona pellucida is clearly breached12. In addition, large series of pregnancies following zona manipulation did not confirm the theory13, 14. The major limitation encountered in studies on monozygotic splitting is the lack of animal models. Indeed, the human and the nine-banded armadillo are the only known mammals that produce monozygotic gestations, although some species occasionally deliver a two-headed or six-legged offspring. Obviously, any 'new' theory to explain zygotic splitting should explain the consistent increase of monozygotic twinning with every method of assisted conception. Data from the East Flanders Prospective Twin Survey—the only population-based series with complete zygosity assessment—show a quasi dose-effect relationship between the method of assisted conception and the incidence of monozygotic twin births15, 16. Any 'new' theory to explain zygotic splitting should also explain why embryologists do not observe any physical splitting of the embryo in in-vitro fertilization (IVF) programs. Embryonic splitting was observed recently, concomitant with disintegration of the zona pellucida, when the embryo was inadvertently left to grow beyond day 5 (i.e. beyond the blastocyst stage)17. If the zygote is predestined to undergo splitting, in-vitro growth of the zygote to a later stage may enhance the potential of the zygote to split later and to result in monochorionic twins. In 2000, Behr et al.18 reported 5% monozygotic (all monochorionic diamnionic) twinning associated with human embryos obtained through standard IVF stimulation protocols, cultured in commercially available, cell-free media systems and transferred as unmanipulated blastocysts. In a more recent analysis19, the same group compared pregnancies following blastocyst transfer and those following cleavage-stage transfer. In this single-center cohort, there were 11 cases of monozygotic twins in 197 (5.6%) viable pregnancies with blastocyst transfer compared with seven of 357 (2%) viable pregnancies with cleavage-stage transfers. Interestingly, in 10 of 18 pregnancies, monozygotic twinning was observed in the setting of a higher-order multiple gestation (6/11 for blastocyst transfer and 4/7 for cleavage-stage transfer). Zona pellucida micromanipulation did not increase this splitting rate compared with cycles without zona breaching. In both studies the authors believe that this phenomenon is real and that this information should be considered when counseling patients for treatment. However, at the same time, one may speculate that the enhanced potential of the zygote to produce monochorionic twins is rather because the potential to split early (and to produce dichorionic twins) has been avoided by prolonged in-vitro culturing. Theories aside, no doubts exist that assisted reproduction has definitely changed both scientific and clinical attitudes towards multiple gestations20. As a result, twins and especially higher-order multiples are no longer considered as curios of nature. The contribution of iatrogenic conception to the incidence of monozygotic twins is estimated by assuming 2% twin maternities, of which 30% are iatrogenic. These would result in 140 spontaneous and 60 iatrogenic twin births per 10 000 births, including 47 spontaneous and four iatrogenic monozygotic twin gestations. In such a construct, iatrogenic pregnancies increase the monozygotic rate by approximately 8.5%, but the rate in the entire population increases from 0.0047% to 0.0051%. If the incidence of monochorionic twins is of importance, and assuming the same proportion of mono- to dichorionicity in iatrogenic and spontaneous conceptions, the rate in the entire population increases about 10%, from 0.0031% to 0.0034%. It follows that the net contribution of iatrogenic conceptions to the increased incidence of monochorionic twins seems insignificant. At the same time, one should remember that all the formidable complications that might occur in these iatrogenic monochorionic pregnancies are iatrogenic complications. With this in mind, it is no wonder that even small changes in the equation might be associated with significant changes in the rate of iatrogenic monochorionic twins (e.g., a specific infertility method that significantly increases the incidence of monochorionic twins). This change might be the reason why clinicians are worried about the higher frequency of monochorionic pairs following blastocyst transfers on day 5 compared with cleavage-stage transfers on day 318, 19, 21. The fact that a monochorionic pregnancy is an important clinical entity, which deserves the most vigilant follow-up, is not arguable. It is also indisputable that most of the complications associated with monochorionicity are extremely interesting and their treatment is a constant challenge to our profession. It is no wonder that some of the most sophisticated technologies were invented specifically to treat these complications. In addition, the in-utero treatment of various conditions associated with monochorionicity is at the forefront of manual and technical expertise. Yet, at the same time, the vast majority of short- and long-term morbidity related to monochorionicity at large does not come from complications of TTTS or TRAP sequence, but from 'banal' ones related to prematurity and growth restriction. Prematurity and growth restriction, however, are not specific to anomalous splitting of the zygote, but plague all multiple gestations simply because the human uterus is unable to carry multiples to the same extent that it carries singletons. Until a (real) advance is made to reduce the risks affecting all twins, irrespective of chorionicity, we shall continue to rely on the formidable remedies devised for individual cases of complicated monochorionic twin gestation.