Monochorionic twin pregnancies are at increased risk of adverse outcome because of the vascular anastomoses that connect the 2 fetal circulation systems. The shared circulation is responsible for some unique complications in monochorionic twins, such as the twin-to-twin transfusion syndrome, the twin anemia polycythemia sequence, the twin reversed arterial perfusion sequence, and monoamniotic twinning. Another consequence of the shared circulation is that the well-being of one twin critically depends on that of the other. In this review, we will describe the technique of placental injection. Further, we will discuss the role of the vascular anastomoses in each of the complications described above and provide an update on their management.
All twin pregnancies are at increased risk of death due to miscarriage and preterm birth. However, monochorionic twins, which constitute 20% of all twins, face the highest risks. For a monochorionic twin, the risk of death between the first trimester and 24 weeks is about 12%, which is 6 times higher than the 2% risk of a dichorionic twin pregnancy. Although most deaths occur <24 weeks, even after viability monochorionic twins remain at significantly increased risk. Complications of the shared circulation account for this excess mortality. During intrauterine life, dichorionic twins have completely separate circulation systems, whereas about 95% of monochorionic twins have vascular anastomoses on the placental surface that connect the 2 circulations.
For Editors’ Commentary, see Contents
The nearly ever-present blood exchange is responsible for some unique complications in monochorionic twins, such as the twin-to-twin transfusion syndrome (TTTS), the twin anemia polycythemia sequence (TAPS), the twin reversed arterial perfusion (TRAP) sequence, and monoamniotic twinning. Another consequence of the shared circulation is that the well-being of one twin critically depends on that of the other. After the diagnosis of spontaneous demise of one of a monochorionic pair, the survivor has a 15% risk of death and a 25% risk of neurodevelopmental impairment because of acute exsanguination along the anastomoses into its demised cotwin. The fact that their well-being is interrelated also poses some specific problems in the management of poor growth and imminent demise of the growth-restricted twin or if 1 twin has a severe anomaly and selective reduction is considered. In this review, we will describe the technique of placental injection. Further, we will discuss the role of the vascular anastomoses in each of the complications described above and provide an update on their management.
Injection studies of the monochorionic placenta
Placental injection from complicated monochorionic twin gestations often provides exceptionally useful information that would otherwise be lost on routine pathological examination. Nevertheless, placentas from pregnancies complicated with single death and delayed delivery can no longer be evaluated because of postmortem involution of the placental part of the demised twin.
Although placental injection is time-consuming, it is basically a simple technique that can be performed by anyone with minimal surgical skills. After delivery, it is important to mark the cords of the first- and/or second-born twin with 1 and/or 2 clamps, respectively. The placenta should be stored in the refrigerator (at 4°C) until examination. The sooner the placenta is examined the better, but an interval of up to 10 days is feasible. It takes about 1 hour to inject a placenta.
Before catheterization, it is best to cut both cords at about 5 cm from their placental insertion, because the arteries are less coiled in their distal segment, which makes it easier for catheter insertion ( Figure 1 ). Of each cord, we catheterize both arteries and the vein and use a 22G (blue) catheter for the artery and a 16G (gray) for the vein (Insyte-W; Vialon, Becton Dickinson, Franklin Lakes, NJ). For placentas from pregnancies delivered <26 weeks, pediatric 26G and 24G catheters (BD Neoflon; Becton Dickinson) can be used for arterial and venous catheterization, respectively. Once inside the vessel, we fix each catheter to the cord with Vicryl 3/0 (Ethicon, Johnson & Johnson, New Brunswick, NJ) to prevent subsequent dislodgement. In >90%, there is an anastomosis that connects the 2 umbilical arteries at the base of the cord (Hyrtl anastomosis). To save time, only 1 artery may be catheterized, but we prefer to catheterize both from the start as it is more difficult to catheterize the other once injection is started.
For color injection, we always use undiluted barium sulphate (Micropaque; Guerbet, Villepinte, France), which is available in every radiology department. To have a better macroscopic view of the anastomoses, we add water-soluble color agents for each set of arteries and veins: blue (methylene blue), red (eosin), and purple (mixture of methylene blue and eosin). We inject the arteries and vein of each twin successively with a 20-mL syringe until all peripheral branches are filled and backpressure prevents further injection. We then clamp the cords again to prevent further leakage with 1 and 2 clamps to identify the first- and second-born twin, respectively. Finally, for optimal visualization, we remove the amniotic membranes from the chorionic surface and rinse the placenta under cold tap water. After injection, the number and type of anastomoses are recorded. We take a high-resolution digital photograph perpendicular to the chorionic surface to document the angioarchitecture.
Anastomoses can be of 3 types: arterioarterial, arteriovenous, and venovenous. Arterioarterial and venovenous anastomoses are superficial and bidirectional anastomoses. “Superficial” refers to the fact that they are visible on the surface of the chorionic plate, forming direct communications between the arteries and veins. “Bidirectional” means that they allow flow in both directions depending on the relative intertwin vascular pressure gradients. An arterioarterial anastomosis functions as a flexible arteriovenous connection and thus compensates for any imbalance that occurs over the unidirectional arteriovenous anastomoses ( Figure 2 ). Most monochorionic placentas typically have only 1 arterioarterial anastomosis. Venovenous anastomoses are more rare and seen in only about 25% of monochorionic placentas. The function of venovenous anastomoses is less well established. Venovenous anastomoses may be associated with decreased perinatal survival and cause demise because of sudden changes in venous return. However, this association has not been confirmed by other series. Nevertheless, in the absence of a venovenous anastomoses, each twin has its own placental territory defined by the venous vessels that drain oxygen-rich blood back to its owner. In contrast, in the presence of a venovenous anastomosis, there is no longer an individual but rather a common and most probably flexible venous drainage area ( Figure 3 ).
On the other hand, arteriovenous anastomoses are usually referred to as deep and unidirectional anastomoses. “Deep” refers to the fact that the anastomosis itself occurs at the capillary level within a shared placental lobule. “Unidirectional” means that they allow flow in 1 direction only. It receives its arterial supply from one twin and gives its venous (well-oxygenated) drainage to the other. The supplying artery and draining vein of the arteriovenous anastomosis are visible on the chorionic surface as an unpaired artery and vein that pierce the chorionic plate at close proximity of one other to supply the underlying, shared placental lobule. Because of the arteriovenous anastomoses, the monochorionic placenta actually consists of 3 parts: 2 that belong to each twin individually, and a third part that is shared and supplied by arteriovenous anastomoses.
Because of their unidirectional nature, arteriovenous anastomoses can create a transfusion imbalance, unless an oppositely directed transfusion by other superficial or deep anastomoses provides adequate compensation. About 90% of monochorionic placentas have several arteriovenous and venoarterial anastomoses in combination with an arterioarterial and/or venovenous anastomosis. In about 5% of monochorionic placentas, there are only arteriovenous anastomoses and in another 5% there are no anastomoses.
TTTS
TTTS is an antenatal sonographic diagnosis based on the presence of polyhydramnios (deepest vertical pocket ≥8 cm) with a distended bladder in the recipient and oligohydramnios (deepest vertical pocket ≤2 cm) with a small or empty bladder in the donor. Because amniotic fluid increases with gestation, most Europeans adhere to the gestational age-dependent cutoff of a deepest pocket ≥10 cm after 20 weeks to define the degree of polyhydramnios. “Twin-to-twin transfusion syndrome” is somewhat a misnomer, as in the overall majority there is no difference in hemoglobin concentration between donor and recipient twin. As such, the name “twin oligopolyhydramnios sequence” may be better suited.
About 1 in 10 monochorionic twins develop TTTS, usually between 16-26 weeks. Why certain monochorionic twins are affected and others are not remains elusive because of the lack of animal models and ethical constraints to routinely perform fetal blood sampling. Both postnatal injection studies and in vivo fetoscopic observations indicate the presence of at least 1 unidirectional arteriovenous anastomosis as an anatomical prerequisite for the development of TTTS ( Figure 4 ). Although there is no unique pattern of anastomoses, the presence of an arterioarterial anastomosis seems to be protective. As such, an arterioarterial anastomosis is present in only 1 of 5 TTTS placentas in contrast to 4 of 5 placentas not complicated by TTTS. Also, a mathematical computer model simulating TTTS demonstrated that an arterioarterial anastomosis indeed compensates more efficiently for any flow imbalance than oppositely directed venoarterial anastomoses. This is due to a much lower resistance over the direct arterioarterial anastomosis than over arteriovenous anastomoses, which occur at a capillary level.
Although the vascular anastomoses are an anatomical prerequisite, the pathogenesis of TTTS must be more complex than a simple net transfer of red blood cells, because both twins usually have similar hemoglobin values. As mentioned above, TTTS is a problem of amniotic fluid discordance with a volume-overloaded recipient and a volume-depleted donor and not a problem of hemoglobin discordance with a polycythemic recipient and anemic donor. Therefore, endocrine factors related to fluid and pressure homeostasis are likely to be involved as well. Due to the intertwin blood exchange, one twin is exposed to the endocrine environment of the other. As such, transfer of renin-angiotensin-aldosterone effectors from the donor may partly explain the recipient’s hypertensive cardiomyopathy and volume overload.
TTTS is the most important cause of death and handicap in a monochorionic twin pregnancy. Because TTTS occurs prior to viability, the prognosis is dismal without treatment. Polyhydramnios-related miscarriage or the preterm birth of 2 sick neonates is common, as is the intrauterine demise of one or both twins. Fetoscopic laser coagulation of all anastomoses along the equator is currently the best treatment option regardless of disease severity. For early detection and because TTTS is often difficult to predict, it is generally recommended that all monochorionic twins are scanned every fortnight, especially between 16-26 weeks. Laser aims to cure the disease by disconnecting the 2 fetal circulations. Successful interruption leads to a normalization of urine output, of amniotic fluid volumes, and of cardiac function in the recipient twin. After laser, there is a 50-60% survival of both twins and an 80% survival of survival of at least 1 twin. Donor twins seem to have somewhat lower survival rates (60%) than recipient twins (70%). Of surviving infants, 11% have some form of developmental impairment. Cerebral palsy is the most common impairment, affecting 5%. Both donor and recipient are at equal risk of long-term impairment. Also, in single survivors, the risk of handicap is not increased, supporting the concept that laser protects the survivor in case of intrauterine demise of its cotwin. The single most important predictor of adverse long-term outcome is an early gestational age at birth.
It is clear that outcome after laser surgery is far from perfect, because at best only 50-60% of parents will take home 2 healthy babies. Missed anastomoses are common and, depending on the thoroughness of the injection technique, are found in about 5-30% of cases. The type and size of missed anastomoses correlate with the outcome. As such, missed large arteriovenous anastomoses may lead to recurrent TTTS or double demise, unless a compensating arterioarterial anastomosis is missed as well. On the other hand, missed small arteriovenous anastomoses can result in TAPS. The latter develops usually several weeks after the procedure with mostly a recipient that becomes anemic and the former donor becoming polycythemic. TAPS is associated with missed small-caliber anastomoses (<1 mm) ( Figure 5 ) that, owing to the polyhydramnios and increased intrauterine pressure, may not be visible at the time of surgery. However, a chronic net transfusion via these hairlike anastomoses may lead to severe hemoglobin discordances several weeks later in ongoing twin pregnancies. TAPS is usually not associated with severe amniotic fluid discordance and is only picked up by serial measurement of the middle cerebral artery (MCA)-peak systolic velocities (PSV). Connecting the dots of coagulated anastomoses with laser may decrease the chance of missing these tiny anastomoses and thereby improve outcome. The benefit of drawing a line along the entire equator from one edge of the placenta to the other is currently the subject of a multicenter randomized trial ( Figure 6 ).
Fetal complications because of incomplete coagulation commonly occur. We therefore recommend placental injection of all placentas that underwent laser treatment, except those with single intrauterine demise and delayed delivery as the anastomoses can no longer be documented. Placental injection is also a means of quality control and may improve the surgical technique. As such, most missed anastomoses are located near the placental edge. Finally, placental examination provides a good learning experience for any fetal medicine specialist who seeks to embark on endoscopic laser surgery.
TTTS
TTTS is an antenatal sonographic diagnosis based on the presence of polyhydramnios (deepest vertical pocket ≥8 cm) with a distended bladder in the recipient and oligohydramnios (deepest vertical pocket ≤2 cm) with a small or empty bladder in the donor. Because amniotic fluid increases with gestation, most Europeans adhere to the gestational age-dependent cutoff of a deepest pocket ≥10 cm after 20 weeks to define the degree of polyhydramnios. “Twin-to-twin transfusion syndrome” is somewhat a misnomer, as in the overall majority there is no difference in hemoglobin concentration between donor and recipient twin. As such, the name “twin oligopolyhydramnios sequence” may be better suited.
About 1 in 10 monochorionic twins develop TTTS, usually between 16-26 weeks. Why certain monochorionic twins are affected and others are not remains elusive because of the lack of animal models and ethical constraints to routinely perform fetal blood sampling. Both postnatal injection studies and in vivo fetoscopic observations indicate the presence of at least 1 unidirectional arteriovenous anastomosis as an anatomical prerequisite for the development of TTTS ( Figure 4 ). Although there is no unique pattern of anastomoses, the presence of an arterioarterial anastomosis seems to be protective. As such, an arterioarterial anastomosis is present in only 1 of 5 TTTS placentas in contrast to 4 of 5 placentas not complicated by TTTS. Also, a mathematical computer model simulating TTTS demonstrated that an arterioarterial anastomosis indeed compensates more efficiently for any flow imbalance than oppositely directed venoarterial anastomoses. This is due to a much lower resistance over the direct arterioarterial anastomosis than over arteriovenous anastomoses, which occur at a capillary level.
Although the vascular anastomoses are an anatomical prerequisite, the pathogenesis of TTTS must be more complex than a simple net transfer of red blood cells, because both twins usually have similar hemoglobin values. As mentioned above, TTTS is a problem of amniotic fluid discordance with a volume-overloaded recipient and a volume-depleted donor and not a problem of hemoglobin discordance with a polycythemic recipient and anemic donor. Therefore, endocrine factors related to fluid and pressure homeostasis are likely to be involved as well. Due to the intertwin blood exchange, one twin is exposed to the endocrine environment of the other. As such, transfer of renin-angiotensin-aldosterone effectors from the donor may partly explain the recipient’s hypertensive cardiomyopathy and volume overload.
TTTS is the most important cause of death and handicap in a monochorionic twin pregnancy. Because TTTS occurs prior to viability, the prognosis is dismal without treatment. Polyhydramnios-related miscarriage or the preterm birth of 2 sick neonates is common, as is the intrauterine demise of one or both twins. Fetoscopic laser coagulation of all anastomoses along the equator is currently the best treatment option regardless of disease severity. For early detection and because TTTS is often difficult to predict, it is generally recommended that all monochorionic twins are scanned every fortnight, especially between 16-26 weeks. Laser aims to cure the disease by disconnecting the 2 fetal circulations. Successful interruption leads to a normalization of urine output, of amniotic fluid volumes, and of cardiac function in the recipient twin. After laser, there is a 50-60% survival of both twins and an 80% survival of survival of at least 1 twin. Donor twins seem to have somewhat lower survival rates (60%) than recipient twins (70%). Of surviving infants, 11% have some form of developmental impairment. Cerebral palsy is the most common impairment, affecting 5%. Both donor and recipient are at equal risk of long-term impairment. Also, in single survivors, the risk of handicap is not increased, supporting the concept that laser protects the survivor in case of intrauterine demise of its cotwin. The single most important predictor of adverse long-term outcome is an early gestational age at birth.
It is clear that outcome after laser surgery is far from perfect, because at best only 50-60% of parents will take home 2 healthy babies. Missed anastomoses are common and, depending on the thoroughness of the injection technique, are found in about 5-30% of cases. The type and size of missed anastomoses correlate with the outcome. As such, missed large arteriovenous anastomoses may lead to recurrent TTTS or double demise, unless a compensating arterioarterial anastomosis is missed as well. On the other hand, missed small arteriovenous anastomoses can result in TAPS. The latter develops usually several weeks after the procedure with mostly a recipient that becomes anemic and the former donor becoming polycythemic. TAPS is associated with missed small-caliber anastomoses (<1 mm) ( Figure 5 ) that, owing to the polyhydramnios and increased intrauterine pressure, may not be visible at the time of surgery. However, a chronic net transfusion via these hairlike anastomoses may lead to severe hemoglobin discordances several weeks later in ongoing twin pregnancies. TAPS is usually not associated with severe amniotic fluid discordance and is only picked up by serial measurement of the middle cerebral artery (MCA)-peak systolic velocities (PSV). Connecting the dots of coagulated anastomoses with laser may decrease the chance of missing these tiny anastomoses and thereby improve outcome. The benefit of drawing a line along the entire equator from one edge of the placenta to the other is currently the subject of a multicenter randomized trial ( Figure 6 ).
Fetal complications because of incomplete coagulation commonly occur. We therefore recommend placental injection of all placentas that underwent laser treatment, except those with single intrauterine demise and delayed delivery as the anastomoses can no longer be documented. Placental injection is also a means of quality control and may improve the surgical technique. As such, most missed anastomoses are located near the placental edge. Finally, placental examination provides a good learning experience for any fetal medicine specialist who seeks to embark on endoscopic laser surgery.
TAPS
In contrast to TTTS, which is essentially an amniotic fluid discordance, TAPS is characterized by a severe hemoglobin discordance caused by a chronic net transfusion over minuscule and usually unidirectional anastomoses. TAPS can be diagnosed antenatally based on discordant MCA-PSV measurements. One twin typically has an elevated MCA-PSV (>1.5 multiples of the median) and in the other twin the MCA-PSV is decreased (<1 multiples of the median). The anemic twin often has decreased amniotic fluid whereas the recipient twin may have increased fluid, but the discordance is not as severe as is required for the diagnosis of TTTS. TAPS can also be defined by postnatal criteria. Here, the hemoglobin discordance should be at least 8 g/dL. Also, to differentiate chronic TAPS from an acute intrapartum transfusion, there should be an increased reticulocyte count in the anemic twin and a decreased count in the polycythemic twin with a ratio of >1.7 or placental injection should demonstrate only tiny (<1 mm) anastomoses.
Spontaneous TAPS occurs in about 5% of previously uncomplicated monochorionic twin pregnancies, whereas iatrogenic TAPS after incomplete laser treatment for TTTS occurs in up to 13% of ongoing twin pregnancies. In contrast to TTTS that typically presents in the previable period between 16-26 weeks, spontaneous TAPS usually occurs >26 weeks and thus in the viable period. On the other hand, iatrogenic TAPS usually develops within 1-5 weeks after the procedure.
The placentas of spontaneous and iatrogenic TAPS show some striking similarities ( Figure 7 ). In contrast to TTTS, the anastomoses in TAPS are smaller and fewer in number, supporting the hypothesis that TAPS is a pure chronic net transfusion of red blood cells, whereas in TTTS with its larger anastomoses there is a more elaborate plasma exchange so endocrine factors may play a role as well. Similar to TTTS, an arterial anastomosis seems to protect against the development of TAPS. As such, an arterioarterial anastomosis is present in only 1 of 5 TAPS placentas in contrast to 4 of 5 placentas not complicated by TAPS. If present, the diameter of the arterioarterial anastomoses is also smaller in TAPS and always <1 mm, so it can inadequately compensate for any flow imbalance.
