Twin reversed arterial perfusion (TRAP) sequence

INTRODUCTION — Twin reversed arterial perfusion (TRAP) sequence refers to a rare pregnancy complication unique to monochorionic twins of a multiple gestation in which one fetus has an absent or rudimentary nonfunctional heart ("acardiac twin") and is entirely perfused in a retrograde direction by its sibling ("pump twin") via abnormal arterio-arterial anastomoses. The acardiac twin typically has other severe congenital anomalies, such as a poorly developed or absent upper body and head. The four types (acephalus, anceps, acormus, amorphus) reflect the degree of abnormal development (table 1); none can survive ex utero.

An additional concern is that if an acardiac twin grows to a very large size relative to the pump twin, the latter is at high risk of developing heart failure from the circulatory burden of supporting its co-twin. Heart failure and other complications may lead to preterm birth or death, which commonly occurs in the absence of intervention.

This topic will discuss issues related to TRAP sequence. More common complications of monochorionic twinning, such as twin-twin transfusion syndrome (TTTS), twin anemia-polycythemia sequence (TAPS), and selective fetal growth restriction (sFGR) are reviewed separately.

●(See "Twin-twin transfusion syndrome: Screening, prevalence, pathophysiology, and diagnosis" and "Twin-twin transfusion syndrome: Management and outcome".)

●(See "Twin anemia-polycythemia sequence (TAPS)".)

●(See "Selective fetal growth restriction in monochorionic twin pregnancies".)

INCIDENCE — The incidence of TRAP sequence is estimated to be 1 in 38 (2.6 percent) monochorionic twin pregnancies and 1 in 9500 to 11,000 (0.01 percent) pregnancies, based on mathematical models [1].

Historically, TRAP sequence had been reported to occur in approximately 1 percent of monochorionic twin pregnancies and 1 in 35,000 (0.003 percent) pregnancies [2]. These historic figures are widely cited but based on data prior to 1954. In contemporary obstetrics, the incidence appears to be higher due to factors such as the use of first-trimester obstetric ultrasound examination, which can detect early twin demises, and assisted reproductive technology, which has increased the incidence of multiple gestations, including monochorionic twins.

Despite the twin name, TRAP sequence very rarely occurs in monochorionic triplet and higher-order multiple gestations. A literature review identified 11 case reports documenting the occurrence of this phenomenon [3]. Subsequent reports have also emerged, adding to the existing body of evidence [4,5]. In monochorionic triplet pregnancies, more than one pump or acardiac fetus may be present [6,7].

PATHOPHYSIOLOGY

Normal fetal circulation — In the normal fetal circulation, relatively oxygen-rich blood from the placenta flows through the umbilical vein toward the fetal liver. The ductus venosus shunts 80 percent of this blood flow into the inferior vena cava, allowing it to bypass the liver and mix with venous return from the lower extremities and kidneys before entering the right atrium (figure 1). Once the blood enters the right atrium, the two sides of the fetal heart act in parallel through intra- and extracardiac shunts (foramen ovale, ductus arteriosus) to fill the aorta and provide the systemic circulation. The distal aorta terminates in the left and right common iliac arteries, which each divide into external and internal iliac branches. The umbilical arteries, which originate from the internal iliac arteries, carry relatively deoxygenated blood back to the placenta where it is oxygenated and then flows back to the fetus via the umbilical vein.

Fetal circulation in TRAP — In TRAP sequence, the pump twin maintains the normal fetal circulatory pattern described above. Additionally, a portion of its cardiac output travels through one or more mostly arterio-arterial anastomoses with its co-twin, resulting in retrograde flow of deoxygenated blood into one or both umbilical arteries and the systemic circulation of the co-twin, thus creating "reversed" circulatory perfusion. While the arterio-arterial anastomoses between twin circulations in TRAP sequence usually exist along the placental surface, a range of vascular atypia has been described, and other abnormalities, such as direct umbilical artery-to-umbilical artery communications that bypass the chorionic plate, are possible [8,9]. Veno-venous anastomoses may also play a minor role.

Since the pump twin carries the hemodynamic burden of perfusing its own body as well as that of its acardiac co-twin, it is at risk for developing high-output heart failure, which manifests as cardiomegaly, polyhydramnios, ascites and other pathologic fluid collections, and hydrops fetalis.

PATHOGENESIS — The precise pathogenesis of TRAP sequence is unknown. One theory is that early abnormal development of arterio-arterial anastomoses allows retrograde flow of deoxygenated blood from one twin to the other and this leads to abnormal cardiac development of recipient twin. In another theory, early abnormal cardiac development of one twin results in low systemic arterial pressure and this is the abnormality allowing retrograde flow of blood from its co-twin through arterio-arterial anastomoses. Regardless of the exact cause, the net result is that the one twin (acardiac twin) depends on retrograde blood flow from the other twin (pump twin) for circulatory support beginning very early in the first trimester. Reversed circulation of deoxygenated blood from the pump twin may contribute to a wide range of gross structural abnormalities in the co-twin [10]. For example, the lower half of the acardiac twin may receive a preferential share of the pump twin's circulatory support. This is likely the reason for relatively better development of the fetal abdomen, pelvis, and lower extremities than the upper extremities and head, which often have various degrees of necrosis, maldevelopment, or absence.

Placental vascular anastomoses alone are not sufficient for the development of TRAP sequence, as nearly all monochorionic twins have some degree of inter-twin placental anastomoses, whereas TRAP remains a rare phenomenon.

DIAGNOSIS

Criteria and findings — The prenatal diagnosis of TRAP sequence should be suspected in any monochorionic multiple gestation in which one fetus has either absent cardiac activity or a rudimentary pump structure (image 1), especially in the setting of multiple severe cranial and/or limb anomalies and skin edema. In first-trimester pregnancies, the initial presentation may appear as monochorionic twins with one demise; however, on follow-up, the "demised" twin will have grown.

A TRAP diagnosis is confirmed if umbilical artery color or spectral Doppler interrogation of the acardiac fetus demonstrates blood flowing toward, rather than away from, its body (image 2) [11]. This reversed perfusion is pathognomonic and required for sonographic diagnosis of TRAP sequence [12]. The focus of Doppler interrogation should be on placental arterio-arterial anastomoses and the umbilical artery of the demised or severely anomalous fetus to limit thermal exposure of the other fetus as much as practically possible for safety. Since both twins typically have umbilical cord insertion sites relatively close together in the placenta, flow in the arterio-arterial anastomoses can be evaluated well in this region. (See "Overview of ultrasound examination in obstetrics and gynecology", section on 'Safety'.)

The spectrum of ultrasound findings in TRAP sequence include:

●Acardiac fetus – Always has abnormal morphology, which varies widely:

•May be an amorphous tissue mass, with no readily recognizable fetal parts

OR

•Heart – Absent or present as a rudimentary pump structure.

•Head – Absent or present with major defects (eg, anencephaly, holoprosencephaly, other major neuroanatomic malformations) and/or severe edema.

•Extremities – The lower extremities are more likely to develop than the upper extremities. When extremities are present, they are typically edematous.

-Lower extremities: partially or totally missing limbs, widely spaced bones or joints.

-Upper extremities: more likely to be totally missing than lower extremities; when present, often poorly developed.

•Pelvis and abdomen – Often underdeveloped.

•Viscera – Lungs, kidney, spleen, and/or liver may be present, underdeveloped, or absent.

•Umbilical cord – Contains only two vessels in 70 percent of cases [11,13]. The cord is typically very short, limiting fetal movement.

●Pump fetus – Generally has normal morphology, but:

•At increased risk for congenital malformations, particularly cardiovascular anomalies.

•May develop signs of high-output cardiac failure, which is directly proportional to the relative size of the acardiac twin [8]. Evidence of overt or developing cardiac failure can include:

-Polyhydramnios

-Cardiomegaly

-Pericardial effusions

-Pleural effusions

-Ascites

-Subcutaneous edema/anasarca

-Hydrops fetalis

-Tricuspid regurgitation

Differential diagnosis — The sonographic differential diagnosis of TRAP sequence includes the following:

●Severely anomalous co-twin – A severely anomalous twin, even with cardiac and cephalic underdevelopment and skin edema, is not related to TRAP sequence if Doppler assessment of the umbilical artery fails to demonstrate retrograde perfusion.

●Co-twin demise – A demised twin will not have limb movements, retrograde arterial flow on Doppler assessment, or continued growth on serial ultrasound examinations, whereas an acardiac twin will demonstrate some or all of these findings. In a classic vignette of an early TRAP presentation, a patient with first-trimester sonographic diagnosis of a demised twin returns for a midtrimester fetal anatomic survey of the surviving co-twin only to discover that the presumed embryonic/fetal demise has continued to grow, at which time TRAP sequence is diagnosed. To avoid this initial misdiagnosis, color Doppler study should be performed for all early cases of suspected vanishing or demised twins in early monochorionic twin gestations to evaluate for reversed perfusion to the demised twin.

●Placental mass – Rarely, an acardiac twin presenting as a severely amorphous tissue mass may be mistaken for a placental mass, such as a chorioangioma or placental or umbilical cord teratoma. Chorioangiomas are usually well-circumscribed, rounded, vascular lesions next to the chorionic surface near the cord insertion site. Color Doppler ultrasound should demonstrate the low-resistance pulsatile blood flow in the aberrant fetal vessels perfusing the lesion. A teratoma consists of a disorganized collection of various tissue types but usually does not have an axial skeleton or separate umbilical cord attachment to the placenta [14]. (See "The placental pathology report", section on 'Mass lesions of the parenchyma'.)

INITIAL POSTDIAGNOSTIC EVALUATION

●Comprehensive anatomic assessment of both fetuses.

●Determination of amnionicity (monochorionic/diamniotic or monochorionic/monoamniotic).

●Assessment of amniotic fluid volume.

●Echocardiography of the pump twin.

●Diagnostic genetic testing – Pregnancies complicated by TRAP sequence may be at increased risk of aneuploidy. Various autosomal trisomies and sex chromosome aneuploidies have been described in case reports, and were sometimes discordant between twins [15-18]. As in any pregnancy with a fetal anomaly, patients should be offered a diagnostic procedure (eg, amniocentesis, chorionic villus sampling) to obtain fetal cells for genetic testing (karyotype, microarray). As the acardiac fetus has no prognosis for ex utero survival, it is reasonable to only sample the pump twin's amniotic sac. This information may influence patient decisions regarding pregnancy continuation, acardiac twin cord occlusion, and/or other aspects of prenatal, intrapartum, or postnatal care. (See "Prenatal genetic evaluation of the fetus with anomalies or soft markers".)

●Doppler velocimetry – Pump twin umbilical arteries and veins are interrogated proximal to their fetal abdominal wall cord insertion, which is a helpful baseline in ongoing assessment. Ductus venosus Doppler studies is also a component of pump twin evaluation.

●Calculation of the ratio of acardiac twin estimated fetal weight to pump twin estimated fetal weight (acardiac:pump twin weight ratio) is an important component of TRAP sequence evaluation. Pump twin estimated fetal weight is performed using standard biometric formula, whereas accurate estimation of acardiac twin weight can be more challenging given the large potential variations in physical shapes (eg, no head, abnormal extremities, hydrops) that are not amenable to assessment by standard biometric formula. Acardiac twins resembling amorphous tissue masses can be measured in three orthogonal planes. The following formula has been proposed to estimate acardiac weight, using the longest length of the acardiac mass:

Other formulas have also been proposed for the sonographic assessment of acardiac twin weight. Since no single formula can be expected to accurately estimate weight for all acardiac twins, subjective assessment of its size is also important in clinical practice.

●Limited data support fetal magnetic resonance imaging (MRI) as a routine component of TRAP sequence assessment. One small series suggested its possible value for evaluating pump twin neuroanatomy; however, evidence is insufficient to support MRI as a useful adjunct to ultrasound and fetal echocardiography for this indication [19].

PROGNOSIS WITHOUT INTERVENTION — The acardiac twin has no possibility of ex utero survival. The pump twin is at substantial risk of perinatal mortality due to heart failure (which may lead to pregnancy loss in any trimester) and/or the sequelae of preterm birth (which may be medically indicated or naturally occurring secondary to preterm prelabor rupture of membranes [PPROM] or preterm labor related to uterine overdistention and/or other less well-defined factors).

Poor prognostic factors — For the pump twin, one or more of the following factors raise the risk of mortality to >50 percent [20-23]. Notably, interpretation of these data is limited, as one series [20] described cases from over three decades ago and another series [21] had only small numbers of expectantly managed cases.

●High-output heart failure. Signs of developing or overt heart failure can include cardiomegaly, tricuspid regurgitation, diminished contractility, and/or reversed a-wave in the ductus venosus or umbilical vein pulsations on Doppler studies and echocardiography.

●Hydrops fetalis.

●Pump twin abnormal Doppler studies showing persistent absent or reversed end-diastolic velocity in the umbilical artery (AEDV, REDV), pulsatile blood flow in the umbilical vein, and/or reversed a-wave in the ductus venosus Doppler waveform (which may be indicative of evolving or overt cardiac failure).

●Acardiac twin weight to pump twin weight ratio >0.70. In one study, when this ratio was >0.70, the risks of heart failure, preterm birth, and polyhydramnios were: 30, 90, and 40 percent, respectively [20]. By comparison, when this ratio was <0.70, the risks of cardiac failure, preterm birth, and polyhydramnios were less: 10, 75, and 30 percent, respectively.

●Polyhydramnios.

●Monoamniotic pregnancy.

PREGNANCY MANAGEMENT

Counseling and referral — Following a diagnosis of TRAP sequence, ideally patients should be referred to a fetal diagnosis and therapy center with experience managing these pregnancies for further assessment and discussion of prognosis and management strategies. For continuing pregnancies, these strategies include expectant management and/or occlusion of the cord of the acardiac twin. Cord occlusion confers a better prognosis than expectant management in pregnancies with one or more poor prognostic features. In a 2016 meta-analysis of 26 retrospective studies of TRAP sequence, pregnancies treated with cord occlusion had a higher chance of live birth of the pump twin than those managed expectantly (79 versus 65 percent; odds ratio [OR] 2.22, 95% CI 1.23-4.01), with the greatest benefit in pregnancies with poor prognostic factors (83 versus 31 percent; OR 8.58, 95% CI 1.47-49.96) [24]. Treatment of these pregnancies is only available at a limited number of institutions due to the need for highly specialized training and equipment. A list of many such centers in North America can be found on the North American Fetal Therapy Network website (). (See 'Prognosis without intervention' above and 'Cord occlusion therapy' below.)

If referral to a center providing acardiac twin cord occlusion is not possible, amnioreduction can treat polyhydramnios (if present) and limit its consequences (eg, preterm birth, preterm prelabor rupture of membranes [PPROM]). However, it is a temporizing measure that does not address the underlying TRAP pathophysiology. Referral to a fetal therapy center experienced with cord occlusion treatment is, therefore, strongly recommended. (See "Polyhydramnios: Etiology, diagnosis, and management in singleton gestations", section on 'Amnioreduction (decompression amniocentesis)'.)

As part of counseling, patients may also be offered pregnancy termination.

Fetal surveillance — No robust data are available to support any specific evidence-based surveillance strategy for monitoring pregnancies complicated by TRAP sequence. We suggest consultation or comanagement with a center experienced with TRAP evaluation and management when possible, which can be helpful when developing surveillance plans.

Based on our clinical experience, we monitor:

●Twin weights and acardiac:pump twin weight ratios every three weeks, beginning in the midtrimester.

●The pump twin for evidence of evolving or overt failure weekly beginning at 16 weeks, although for a very small, stable acardiac twin, every-other-week monitoring may be considered.

●Biophysical profile score weekly in the third trimester.

However, if poor prognostic factors are present (table 2), we may increase second-trimester monitoring to twice weekly to optimize the timing of cord occlusion therapy, if desired. In those cases in which cord occlusion is indicated but declined or is not indicated, twice weekly monitoring can help with delivery timing when a gestational age compatible with ex utero survival has been reached.

Cord occlusion therapy — Cord occlusion interrupts the vascular communication(s) between twins, thereby relieving the pump twin of the hemodynamic burden of supporting its acardiac co-twin. Because all of the blood is being supplied by the pump twin and the cord is occluded, this twin is not at increased risk of adverse effects from demise of the acardiac twin, in contrast to the sequelae of monochorionic twins with one natural demise.

Rationale — The rationale is to optimize the outcome of the pump twin since the acardiac twin has no potential for independent survival.

Candidates — We offer cord occlusion therapy for pregnancies complicated by TRAP sequence that have at least one poor prognostic factor (table 2). In the absence of any poor prognostic factors, expectant management is reasonable. Although a ratio of the estimated weight of the acardiac twin to that of the pump twin >0.70 is one identified poor prognostic factor, many centers now offer cord occlusion therapy for ratios >0.50, or when there is evidence of a rapidly growing acardiac twin. Ultimately, the decision to intervene must be individualized to the specific clinical findings of the pregnancy and must balance maternal and obstetric risks against potential pump twin benefits. Local legal regulations regarding termination may be an additional factor, although it should be reiterated that acardiac twins often lack hearts and heads and are universally acknowledged to have no chance for ex utero survival.

The minimum gestational age for offering cord occlusion is modality specific. For example, laser photocoagulation can be performed in the late first trimester whereas radiofrequency ablation (RFA) or bipolar coagulation are typically offered at 16 to 18 weeks of gestation.

While no consensus exists on a maximum gestational age for offering cord occlusion, by 26 to 28 weeks it may be reasonable to consider delivery rather than cord occlusion when there is evidence of pump twin compromise. State and institutional limitations may also restrict the gestational age at which cord occlusion therapy can be offered. Consultation with a center experienced with TRAP sequence that offers cord occlusion therapy is strongly recommended in the evaluation and management of affected pregnancies. (See 'Prognosis without intervention' above and 'Choice of technique' below.)

Choice of technique — The choice of cord occlusion technique for a specific pregnancy should be individualized based on operator experience, center resources/ procedural availability, clinical presentation, gestational age, and patient preference. Contemporary cord occlusion techniques include [25-28]:

●Radiofrequency ablation (RFA)

●Bipolar cord coagulation

●Laser photocoagulation

In the United States, RFA is the most commonly used cord occlusion technique, and the preferred technique at the institution of the author of this topic. In many treatment centers, RFA and/or bipolar coagulation are reserved for pregnancies ≥16 weeks of gestation, but preferably ≥18 weeks, due to concerns about a higher rate of procedure-related complications, such as PPROM, with earlier intervention. Because RFA involves a smaller caliber uterine entry than bipolar occlusion to conduct the procedure, RFA proponents believe that it may have a lower complication rate (and therefore greater efficacy) for achieving midtrimester cord occlusion. However, limited data have not conclusively demonstrated improved outcomes with RFA [29,30], despite being seemingly less invasive [31,32]. In a meta-analysis that included 481 cases of bipolar cord occlusion and 320 cases of RFA from 17 studies involving complicated monochorionic gestations, overall survival was not significantly different for the two techniques (RFA 76.8 percent, bipolar occlusion 79.1 percent) [30]. While PPROM occurred less frequently in the RFA group (RFA 17.7, bipolar occlusion 28.2 percent), mean gestational age at delivery was similar for both groups (RFA 34.7±1.7 weeks, bipolar occlusion 35.1±1.6 weeks). The analysis was limited by small studies and heterogeneity of fetal anatomy, operator experience, and technique.

The optimal management of TRAP sequence presenting prior to 16 to 18 weeks is unclear. Laser photocoagulation (also called intrafetal or interstitial laser) of the umbilical arteries of the acardiac fetus may be offered as early as 12 to 14 weeks. At this early gestational age, ultrasound findings may not accurately identify TRAP pregnancies at high risk of fetal demise, which has been reported to occur acutely and unpredictably in 33 to 83 percent of such pregnancies [33]. For this reason, cord occlusion of all cases before 16 weeks has been proposed to reduce the risk of first-trimester demise; however, this approach remains investigational because of limited data, substantial risk for publication bias, and potentially high procedure-related loss rates (8 percent (1 in 12) in one study [34] but 29 and 42 percent (7 in 24 and 5 in 12) in two other studies [35,36]) that may exceed the loss rate of expectant management (33 to 83 percent) [37]. While laser photocoagulation has the advantage of potentially definitive therapy as early as the late first trimester, it has not proven to be superior to expectant management until 16 to 18 weeks. An international trial is underway comparing early intervention (13 to 14 weeks) with intrafetal laser under ultrasound guidance using an 18- to 20-gauge needle with later intervention (16 to 19 weeks) by the same intrafetal ablation technique or by fetoscopic laser coagulation [38]. This trial should contribute to our understanding of intrafetal laser as a treatment for TRAP sequence and timing of therapy.

Technique

●Patient preparation – There are no robust data to support either clear benefit or harm from perioperative antibiotic prophylaxis for percutaneous cord occlusion therapy. At our center, we administer cephalexin 250 mg orally approximately one hour before therapy and continue it every six hours for a total of 72 hours. If the patient is not able to take a cephalosporin, we substitute clindamycin 900 mg orally.

Cord occlusion procedures can be performed under local or neuraxial anesthesia. For procedures performed with local anesthesia (lidocaine block), we have the patient take oral analgesia before the procedure.

●RFA – Under ultrasound guidance, RFA is performed by percutaneously guiding a disposable 17-gauge diameter needle-like device through a small (0.5 cm) incision in the maternal abdomen to an intrafetal location adjacent to the umbilical cord insertion site at the abdominal wall of the acardiac twin. This allows for a nonmobile section of intrafetal vascularity to be ablated. Grounding pads are required to protect the mother from thermal injuries.

A commonly used device utilizes deployable tines that release in a palm-tree configuration, enabling a controlled ablation sphere (image 3) that will occlude all of the umbilical vessels contained within its ablation radius. Ablation is performed until either specific impedance or temperatures are reached, depending on the device used. Cessation of blood flow from the pump twin to the acardiac twin is confirmed by Doppler ultrasound prior to the end of the procedure [27].

If the patient has cramping or contractions following the procedure, a brief course of indomethacin may be considered (50 mg oral loading dose followed by 25 mg orally every six hours for a total of 48 to 72 hours) depending on clinical concern for preterm labor.

●Bipolar coagulation – Under ultrasound guidance, bipolar coagulation is performed by passing an operative sheath (approximately 10- to 12-French) through a small skin incision in the maternal abdomen and into the uterus to reach the acardiac fetus via the Seldinger technique or direct trochar entry.

A diagnostic fetoscope or ultrasound is used to visualize the cord of the acardiac fetus and identify an appropriate site for coagulation. In cases where the acardiac twin has oligohydramnios, amnioinfusion can be performed after the entering the uterus to improve visualization of the cord. Bipolar forceps are advanced through the sheath and guided sonographically to grasp the target site. Once confirmed to be in place, power is applied until successful cord occlusion is achieved. As described above, cessation of blood flow from the pump twin to the acardiac twin is confirmed by Doppler ultrasound. Fetoscopic visualization may assist with confirmation, as well. If the patient has cramping or contractions following the procedure, a brief course of indomethacin may be considered depending on clinical concern for preterm labor.

●Laser photocoagulation – Laser photocoagulation often involves use of an ultrasound-guided 18-gauge spinal needle, although use of other caliber needles has been described. Under sonographic guidance, the needle is percutaneously guided to an intrafetal location close to the intra-abdominal portion of the umbilical artery. The stylet is removed and a laser fiber is then advanced through the needle, after which a series of brief controlled bursts are utilized to achieve sustained interruption of vascular flow.

As described above, cessation of blood flow from the pump twin to the acardiac twin is confirmed by Doppler ultrasound. If the patient has cramping or contractions following the procedure, a brief course of indomethacin may be considered depending on clinical concern for preterm labor.

Complications

●Maternal – Major maternal complications from cord occlusion are rare and include severe uterine bleeding, other maternal vascular injury, maternal thermal injury (extremely rare with use of modern RFA grounding pads), chorioamnionitis (potentially leading to maternal sepsis), abruption, and disseminated intravascular coagulation. Laparotomy may be required to manage bleeding complications unresponsive to conservative management. Among the 98 patients identified in the North American Fetal Therapy Network Registry of RFA cord occlusion therapy for TRAP, there were no maternal deaths, no patient required a blood transfusion, and most patients were hospitalized for ≤1 day after the procedure [39]. The most common obstetric complication was PPROM, which occurred in 17 pregnancies usually within 10 weeks after the procedure.

●Pump twin – Direct or indirect pump twin injury, including perioperative demise, is possible [40].

Development of aplasia cutis congenita has been reported as a complication associated with laser photocoagulation; further study is required to evaluate the incidence of this complication [41].

●Acardiac twin – Temporary cessation of flow into the acardiac twin due to vasospasm followed by re-established patency is possible, especially with RFA. This complication is not well documented and is therefore either extremely rare or theoretical. Nevertheless, it is reason to confirm cessation of acardiac twin umbilical cord blood flow with color Doppler study upon completion of cord occlusion and upon follow-up sonographic evaluation.

Follow-up — Absence of blood flow in all acardiac twin umbilical vessels should be confirmed by Doppler studies at completion of the procedure and at subsequent ultrasound examinations. Follow-up sonograms of the pump twin should evaluate for: recovery of heart function, normalization of Doppler studies (if abnormalities existed prior to therapy), resolution of hydrops (if previously present), and normalization of amniotic fluid volume (if polyhydramnios was present).

There is no consensus or evidence basis to support any particular monitoring strategy after cord occlusion therapy. We perform weekly sonograms after RFA for several weeks until sustained normalization of findings has been demonstrated. Additionally, we perform serial pump twin growth assessments approximately every three weeks throughout pregnancy.

Pediatric outcome

●Neonatal survival – Across published experiences describing the use of RFA for TRAP, pump twin survival ranged from 80 to 100 percent, with mean gestational age at birth ranging from 33 to 37 weeks [27,39,42,43]. In the North American Fetal Therapy Network review of 98 pregnancies from multiple contributing fetal therapy centers, overall pump twin survival rate was 80 percent and the median gestational age at birth was 37 weeks [39].

As previously noted, retrospective review suggests comparable efficacy for RFA and bipolar cord occlusion [29,30]. Limited published experience with laser coagulation also suggests a neonatal survival rate of approximately 80 percent [34,35,38,44,45]. In the largest series of laser for TRAP at a single center (17 cases), the live birth rate was 82 percent at a median gestational age of 37+1 weeks of gestation [44]. An accompanying literature review including data from an additional 10 studies yielded 51 total cases with an 80 percent neonatal survival rate.

●Long-term outcomes – Data on long-term outcomes are sparse. Cases of neurodevelopmental delay have been reported after cord occlusion selective feticide for TRAP [46,47]. In the largest series focused on long-term neurodevelopmental outcomes among pump twin survivors of pregnancies complicated by TRAP sequence that were treated with RFA, information was available for 27 children at median age two years five months and none had neurodevelopmental delay (defined as total developmental quotient <70 points using the Kinder Infant Development Scale, a validated questionnaire) [48]. However, some patients were as young as four months at assessment, only 44 percent were over three years of age, and long-term follow-up was not available for another five children who underwent RFA.

Timing and route of delivery — Our approach is based on our clinical experience. Cesarean birth is performed for standard obstetric indications.

●For pregnancies not undergoing cord occlusion, we plan delivery at 34+0 to 36+6 weeks of gestation, but deliver sooner if pump twin compromise or a standard obstetric indication for preterm birth develops.

We administer a course of antenatal corticosteroids to pregnancies less than 34 weeks if we are concerned about an increased risk of delivery occurring in the next seven days. (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery".)

●For pregnancies that underwent successful cord occlusion, we await spontaneous labor at term unless a standard obstetric indication for earlier delivery develops.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Multiple gestation".)

SUMMARY AND RECOMMENDATIONS

●Definition/pathophysiology – Twin reversed arterial perfusion (TRAP) sequence refers to a rare, unique complication of monochorionic twin pregnancy in which a severely anomalous twin with an absent or rudimentary heart ("acardiac twin") is perfused by its co-twin ("pump twin") via retrograde flow of deoxygenated blood through arterio-arterial anastomoses. If the acardiac twin grows to a very large size relative to the pump twin, the pump twin is at high risk of developing heart failure from the circulatory burden of supporting its co-twin. (See 'Pathophysiology' above and 'Incidence' above.)

●Diagnosis – The prenatal diagnosis of TRAP sequence should be suspected in any monochorionic multiple gestation in which one fetus has either absent cardiac activity or a rudimentary pump structure (image 1), especially in the setting of multiple severe cranial and/or limb anomalies and skin edema. The diagnosis is confirmed in such cases by color Doppler interrogation of the umbilical artery of this fetus showing blood flowing toward, rather than away from, its body (image 2). (See 'Criteria and findings' above.)

•Acardiac phenotype – The acardiac phenotype ranges from a fetus that has well-developed lower extremities, pelvis, and abdomen but a poorly developed upper body and head (with intracranial defects) to a tissue mass that is not readily recognizable as a fetus (table 1).

•Pump phenotype – The pump twin may be normal or have signs of high-output cardiac failure.

●Differential diagnosis includes a severely anomalous co-twin, co-twin demise, and placental mass. (See 'Differential diagnosis' above.)

●Prognosis – The acardiac twin cannot survive ex utero. The perinatal mortality rate of pump twins is related to presence of poor prognostic features (table 2). In fetuses with poor prognostic features, perinatal mortality is ≥50 percent with expectant management and <20 percent with intervention (cord occlusion). (See 'Prognosis without intervention' above and 'Pediatric outcome' above.)

●Pregnancy management

•Referral – Referred to a fetal therapy center experienced in TRAP evaluation and management is desirable for further evaluation and counseling regarding prognosis and management options, including cord occlusion therapy if appropriate. (See 'Counseling and referral' above and 'Cord occlusion therapy' above.)

•Postdiagnostic evaluation (See 'Initial postdiagnostic evaluation' above.)

-Comprehensive anatomic assessment of both fetuses, including amnionicity and amniotic fluid volume

-Evaluation of pump twin cardiovascular status with Doppler of the umbilical artery, umbilical vein, and ductus venosus and with echocardiography

-Invasive procedure for diagnostic genetic testing

-Acardiac:pump weight ratio calculation

•Cord occlusion – For ongoing pregnancies with TRAP sequence and at least one poor prognostic factor (table 2), we recommend cord occlusion of the acardiac twin rather than expectant management (Grade 1B). Ideally, the procedure should be performed at a center with proficiency in the evaluation and management of TRAP sequence, and in accordance with local regulations. Techniques include radiofrequency ablation (RFA), bipolar cord coagulation, and laser photocoagulation. The choice is individualized based on operator experience, center resources/ procedural offerings, clinical presentation, gestational age, and preference. The procedure is typically performed at 16 to 18 weeks of gestation, or soon thereafter. (See 'Candidates' above and 'Choice of technique' above and 'Technique' above.)

The most common complication of RFA and bipolar cord occlusion is preterm prelabor rupture of membranes (PPROM). (See 'Complications' above.)

Pump twin survival ranges from 80 to 100 percent, with mean gestational age at birth ranging from 33 to 37 weeks. (See 'Pediatric outcome' above.)

•Route and timing of delivery – Cesarean birth is performed for standard obstetric indications. (See 'Timing and route of delivery' above.)

-For pregnancies not undergoing cord occlusion, we plan delivery at 34+0 to 36+6 weeks of gestation, but sooner if pump twin compromise or a standard obstetric indication for preterm birth develops.

-For pregnancies that underwent successful cord occlusion, we await spontaneous labor at term unless a standard obstetric indication for earlier delivery develops.

ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges Marium G Holland, MD, MPH; Joan M Mastrobattista, MD; and Michael J Lucas, MD, MPH, who contributed to earlier versions of this topic review.