Twin anemia-polycythemia sequence (TAPS)

INTRODUCTION — 

Twin anemia-polycythemia sequence (TAPS) is a complication of monochorionic (MC) twin pregnancies characterized by highly discordant hemoglobin values between the twins. In contrast to twin-twin transfusion syndrome (TTTS), both twins have normal amniotic fluid volume in the classic, pure form of TAPS [1-4].

TAPS is caused by unbalanced slow transfusion of red blood cells through a few small (<1 mm diameter) placental arteriovenous (AV) anastomoses [1]. Unbalanced transfusion over time leads to large intertwin hemoglobin differences resulting in anemia of one twin and polycythemia of the cotwin. The diagnosis can be challenging because TAPS, TTTS, and a third disorder of MC twins, selective fetal growth restriction (sFGR), are not mutually exclusive of each other and can present together or in any combination (table 1).

This topic will focus on diagnosis and management of TAPS. Diagnosis and management of TTTS and sFGR are discussed separately.

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

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

TYPES OF TAPS

●Spontaneous – Spontaneous TAPS refers to a type of chronic TTTS, involving small caliber AV anastomoses. It has been detected in 3 to 6 percent of uncomplicated third-trimester MC diamniotic twins [5,6]. It has also been detected rarely in monozygotic, dichorionic pregnancies in which hairline anastomoses between placental masses lead to chronic exchange of red blood cells [7,8].

●Post-laser ablation – Post-laser ablation TAPS refers to a potential sequelae of use of laser ablation for treatment of TTTS [9,10]. It occurs in 2 to 13 percent of such pregnancies, usually within one month but up to 17 weeks after the procedure [9,11]. The wide range in incidence has been attributed to use of different laser ablation techniques and different definitions and criteria for TAPS. Risk factors include TTTS with few anastomoses and no arterio-arterial (AA) anastomoses before laser ablation [12].

PATHOPHYSIOLOGY

Spontaneous TAPS — All MC placentas have arteriovenous (AV), arterio-arterial (AA), and venovenous (VV) anastomoses. AV anastomoses are unidirectional while both AA and VV anastomoses are bidirectional. Normal MC placentas average eight AV anastomoses that are typically around 4 mm in diameter [13]. In spontaneous TAPS, the placenta has only three to four AV anastomoses that are very small (<1 mm diameter). These tiny unidirectional unbalanced anastomoses allow slow passage of red blood cells (5 to 15 mL per 24 hours [3]) from the donor twin to the recipient twin, gradually leading to highly discordant hemoglobin levels [9]. The donor twin becomes anemic (which can lead to hydrops fetalis), while the recipient twin becomes polycythemic (which can lead to fetal and placental thrombosis) [14,15]. The slowness of this process allows for hemodynamic compensation across the placental maternal-fetal interface, which is hypothesized to be the reason for absence of the amniotic fluid volume discordancy seen in classic TTTS [16].

AA anastomoses also occur in 10 to 20 percent of TAPS placentas and are smaller in these placentas than in placentas of uncomplicated MC twins and placentas in TTTS, both of which have more of these anastomoses [3]. The small number and size of AA anastomoses also contribute to development of TAPS because the bidirectional blood flow in normal AA anastomoses allows intertwin equilibration of blood.

Of note, in MC twins without TAPS, the smaller twin typically has a smaller portion of the shared placenta than the larger twin. In TAPS, however, the smaller twin is the donor and often has a larger share of the placenta than the larger, recipient twin [3].

Post-laser ablation TAPS — Based on placental injection studies from TTTS pregnancies treated with laser ablation, post-laser TAPS occurs when laser ablation is incomplete, leaving a few very small (<1 mm) residual anastomoses. These are typically AV anastomoses without accompanying AA anastomoses. As with spontaneous TAPS, these small, unbalanced, residual AV anastomoses allow slow passage of red blood cells, usually from the former recipient twin to the former donor twin, gradually leading to highly discordant hemoglobin levels [9]. The former recipient twin becomes anemic, while the former donor twin becomes polycythemic, a reversal of the previous net blood flow during TTTS [14,15]. In the authors' experience, the residual anastomoses can also be in opposite directions, leading the former recipient to become polycythemic and the former donor to become anemic.

Use of the Solomon technique (ie, ablating between the anastomoses to the edges of the placenta) helps to reduce the risk of post-laser TAPS [17]. (See "Twin-twin transfusion syndrome: Management and outcome", section on 'Procedure'.)

CLINICAL PRESENTATION — 

In the authors' practice, all MC twins are monitored for development of TTTS, TAPS, and sFGR using the protocol shown in the table (table 2). TAPS presents with discordance in middle cerebral artery (MCA) peak systolic velocity (PSV) measurements. Discordance in placental echogenicity and cardiomegaly in one twin may also be seen.

There is growing consensus for routine surveillance of MC twins for TAPS. The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) guidelines recommend monitoring for TAPS (in addition to TTTS) every two weeks beginning at 16 weeks [18]. The Society for Maternal-Fetal Medicine (SMFM) recommends incorporating Doppler MCA-PSV determinations into all MC twins surveillance beginning at 16 weeks of gestation [19].

DIAGNOSTIC CRITERIA — 

Prenatal sonographic diagnosis of TAPS can be challenging because TAPS, TTTS, and sFGR are not necessarily mutually exclusive of each other and can present together in any combination. Furthermore, the accuracy of prenatal sonographic diagnosis of TAPS has not been verified against definitive hematological criteria because of the risks associated with performing umbilical vein (UV) sampling to confirm discordant hemoglobin levels between the twins.

Prenatal diagnosis — TAPS in its classic form should be suspected in MC twins when the middle cerebral artery-peak systolic velocity (MCA-PSV) is >1.5 multiples of median (MoM) in one twin (suggestive of moderate to severe anemia) and <0.8 MoM in the other twin (suggestive of polycythemia) [9,15,20,21].

Supportive findings include:

●Discordant placental echogenicity is a common but nonspecific finding that supports the diagnosis: The anemic donor's region of the placenta is thickened and hyperechoic, while the plethoric recipient twin's region of the placenta has a normal appearance, with clear demarcation between the donor and recipient territories (image 1) [22].

●Cardiomegaly, tricuspid regurgitation, and/or ascites in the donor twin [23].

●Starry sky liver in the recipient twin (ie, bright echogenic dots throughout a background of diminished echogenicity of the liver parenchyma) [23]. The bright dots are due to increased brightness of the portal venule walls.

Criteria are not uniform across studies (eg, some authors use MCA-PSV <1.0 MoM to diagnose polycythemia [6]) [24]. Furthermore, prenatal sonographic diagnosis of TAPS is shifting from reliance upon the strict MCA-PSV cutoffs described above to a system based on absolute intertwin MCA-PSV discordance (delta MCA-PSV) because the latter appears to have a stronger correlation with the postnatal diagnosis of TAPS [25,26]. A Delphi process resulted in the proposal for delta MCA-PSV ≥1 MoM to diagnose TAPS [18], with some authorities proposing a delta MCA-PSV >0.5 [25] or >0.37 MoM for diagnosis [27].

The use of delta MCA-PSV rather than strict MCA-PSV cutoffs is supported by prospective data from children postnatally diagnosed with TAPS [26]. Although MCA-PSV was significantly higher in anemic fetuses than in normal fetuses (1.15 MoM versus 1.02 MoM), MCA-PSV was similar for polycythemic and normal fetuses (0.95 MoM versus 1.02 MoM) and among the 25 children with polycythemia, nine (36 percent) had an MCA PSV >1. Thus, many cases did not meet the traditional criteria for TAPS. On the other hand, the delta MCA-PSV was positively correlated with the intertwin differences in hematocrit required for postnatal diagnosis of TAPS. Although delta MCA-PSV appears to have better diagnostic accuracy than using the MCA-PSVs of each twin, the optimum cut-off needs to be determined. In a retrospective cohort of 35 twin pregnancies with a postnatal diagnosis of TAPS, a delta MCA-PSV >0.5 was 85 percent sensitive, 100 percent specific, and had a negative likelihood ratio of 0.17 for the diagnosis of TAPS [25]. Validation studies are needed.

Postnatal diagnosis — At birth, the obstetric provider should suspect TAPS when a striking color difference on the maternal side of the placenta is observed: one side (donor) is pale and the other side (recipient) is dark (picture 1) [28].

In the nursery, TAPS should be suspected if one twin is pale and the other has a ruddy, plethoric appearance and/or one twin is anemic (hematocrit <45 percent) and the other is polycythemic (hematocrit >65 percent). Postnatal diagnosis of TAPS is based on an intertwin hemoglobin difference ≥8.0 g/dL in conjunction with an intertwin reticulocyte ratio >1.7 (reticulocyte count of the donor twin divided by the reticulocyte count of the recipient twin).

If performed, placental injection will reveal few very small (1 mm) AV anastomoses with flow in a single direction, but this procedure is not required for diagnosis [6].

STAGING (DISEASE SEVERITY) — 

The stage of disease (ie, disease severity) is the basis for further follow-up and management. Staging is based on results of Doppler flow studies of the umbilical artery (UA), umbilical vein (UV), and ductus venosus (DV) or, postnatally, hemoglobin discordance between twins, as shown in the table (table 3).

DIFFERENTIAL DIAGNOSIS — 

Although there is some overlap between the findings for TAPS and TTTS (table 1), there are several characteristics that favor each diagnosis:

●MCA-PSV discordance is diagnostic of TAPS and not present in pure TTTS.

●Amniotic fluid discordance (oligohydramnios and polyhydramnios sequence) is essential in pure TTTS and not present in pure TAPS.

●Spontaneous TAPS has been diagnosed between 14 and 35 weeks of gestation, in contrast to TTTS, which is typically diagnosed in the early- to mid-second trimester [29,30].

●Postnatally, an intertwin reticulocyte ratio of >1.7 is required for diagnosis of TAPS because a large intertwin hemoglobin difference alone at birth may represent acute peripartum TTTS rather than TAPS. Accurate diagnosis is important because the anemia secondary to TAPS is typically euvolemic, whereas anemia in peripartum TTTS requires rapid correction of hypovolemia [28].

Other causes of fetal anemia should be considered, including infection (especially parvovirus B19), alloimmunization, and fetomaternal bleeding. Both fetuses tend to be anemic in cases caused by infection and there may be other findings, such as pleural or pericardial effusions. Alloimmunization is easily excluded by a negative maternal red blood cell antibody screen and significant fetomaternal bleeding can be excluded by Kleihauer-Betke test or flow cytometry. (See "Parvovirus B19 infection during pregnancy", section on 'Maternal-fetal effects' and "RhD alloimmunization in pregnancy: Overview" and "Management of non-RhD red blood cell alloantibodies during pregnancy" and "Spontaneous massive fetomaternal hemorrhage".)

PROGNOSIS — 

Spontaneous resolution of TAPS has been observed in 16 percent of cases [31]. In the remaining cases, TAPS may result in birth of two healthy neonates with isolated intertwin hemoglobin differences that may require blood transfusion or partial exchange transfusion. However, cerebral injury (leading to neurodevelopmental impairment [eg, cognitive impairment, hearing loss]) or death of one or both twins are other possible and more serious outcomes. Postnatal neurodevelopmental impairment may be suspected antenatally because of brain lesions such as intracerebral hemorrhage or ischemic lesions on fetal sonogram and/or magnetic resonance imaging (MRI) [32].

The following studies illustrate overall TAPS prognosis. The outcomes with various management approaches are reviewed below. (See 'Rationale' below.)

●Short-term outcome – In a meta-analysis of 506 pregnancies in 38 studies, spontaneous TAPS was associated with better pregnancy outcome than post-laser TAPS [33]:

•Fetal death: 5.2 versus 10.2 percent

•Neonatal death: 4.0 versus 9.2 percent

•Severe neonatal morbidity: 29.3 versus 33.3 percent

•Severe neurological morbidity: 4.0 versus 11.1 percent

•Preterm birth (spontaneous and iatrogenic): 86.3 versus 100 percent

Limitations of the observational data preclude clear conclusions regarding the better prognosis of spontaneous TAPS and the superiority of any management approach (expectant, laser, fetal transfusion, selective reduction). Also, the case reports and retrospective nature of original studies in the systematic review limit the ability to assess exact risks.

●Long-term outcome

•In a study of long-term outcomes (median 48 months of age) of 49 twin pregnancies with spontaneous TAPS [34]:

-Neurodevelopmental impairment occurred in 30 percent (22 out of 74 survivors) and was more frequent in the anemic fetus (44 versus 18 percent, odds ratio [OR] 4.1, 95% CI 1.8-9.1).

-Severe neurodevelopmental impairment was detected in 9 percent of survivors and was also more frequent in the anemic fetus than the plethoric fetus (18 versus 3 percent), although the difference did not reach statistical significance.

On multivariate analysis, independent risk factors for neurodevelopmental impairment were gestational age at birth and severe anemia.

•In a study of long-term neurodevelopmental outcomes (at 24 to 96 months of age) of 33 twin pregnancies with post-laser ablation TAPS [35]:

-Mild to moderate cognitive delay (score <85) occurred in 17 percent (8 of 47 children assessed) and severe cognitive delay (score <70) occurred in 4 percent (2 of 47 children assessed).

-Overall, severe neurodevelopmental impairment occurred in 9 percent (4 of 47 children): cerebral palsy (1 child), severe motor delay (1 child), severe cognitive delay (2 children); these four children were born very preterm (at 28, 29, 29, and 32 weeks of age), which may account for at least some of these impairments.

The small sample size and variety of tests used for neurodevelopmental evaluation in these two studies limit interpretation of the findings.

PREGNANCY MANAGEMENT

Our approach — Most aspects of prenatal care are routine; exceptions are described below. TAPS is not associated with an increased risk for chromosomal/genetic or structural abnormalities; therefore, TAPS is not a specific indication for offering fetal genetic studies.

Our management is generally similar for spontaneous and post-laser ablation TAPS [31].

●Stage 1 TAPS <32 weeks (table 3)

•Weekly Doppler assessment to assess for progression in stage.

•If no progression or resolution, deliver between 34+0 and 37+0 weeks of gestation [36]. More complicated cases are delivered in the first 10 days of this range and less complicated cases are delivered in the last 10 days of this range.

•If progression, manage according to stage and gestational age, as described below.

●Stage 2 TAPS <32 weeks (table 3)

•Administer a course of antenatal corticosteroids between 24+0 and 32+0 weeks of gestation to reduce the neonatal risks in the event of preterm birth. (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery".)

•Twice weekly Doppler assessment to assess for persistence or progression (characterized by development of fetal heart failure and secondary signs of TAPS [discordant placental echogenicity, starry liver]).

-If stage 2 TAPS resolves, deliver between 34+0 and 37+0 weeks of gestation. More complicated cases are delivered in the first 10 days of this range and less complicated cases are delivered in the last 10 days of this range.  

-If stage 2 TAPS persists for more than one to two weeks or progresses and pregnancy is <32 weeks: manage as stage 3 or 4 TAPS, as described below.

-If stage 2 TAPS persists for more than one to two weeks or progresses and pregnancy is 32 to 34 weeks, deliver.

●Stage 3 or 4 TAPS <32 weeks (table 3) – There is no consensus regarding optimal treatment; expectant management, laser surgery, in utero transfusion, selective feticide, and early delivery have been used [30,31]. We make the management decisions on a case-by-case basis based on the gestational age and the acuity of the TAPS:

•For post-laser ablation TAPS within two weeks of the procedure, termination of pregnancy should be considered because of the poor prognosis; otherwise, a repeat laser procedure is offered up to 28+0 weeks of gestation [37,38]. A repeat laser procedure can often be difficult due to bloody amniotic fluid as a result of the previous procedure. Amnioinfusion and/or amnioreduction may be used to improve visualization of fine anastomoses across the intertwin membrane.

•For spontaneous TAPS before 28+0 weeks, fetoscopic laser ablation is probably the best option. Alternatively, the anemic fetus can be transfused with red blood cells. Both intravenous (IV) and intraperitoneal (IP) transfusion are acceptable methods to correct fetal anemia. We prefer IP transfusion because the slow absorption of red blood cells that typically occurs with IP transfusion may more closely mimic fetal physiology. Sometimes a combination approach is used: IV transfusion to correct severe anemia with IP transfusion of a portion of the blood for slower absorption to limit the need for frequent, multiple transfusions. (See "Fetal transfusion of red blood cells".)

After treatment, we perform ultrasound weekly until resolution of TAPS and deliver these pregnancies between 34+0 and 37+0 weeks of gestation. If there is a suspicion of an intracranial lesion, we perform fetal MRI.

•For spontaneous or post-laser ablation TAPS between 28+0 and 32+0 weeks of gestation, either of two options is reasonable. Some centers transfuse the anemic fetus with red blood cells, while others perform fetal transfusion of the anemic fetus plus a partial exchange transfusion of the polycythemic twin to potentially reduce the complications associated with hyperviscosity. In these cases, 5 mL aliquots of blood are removed and replaced with equal volumes of sterile saline. Repeat transfusion or partial exchanges are based on subsequent MCA-PSVs.

After treatment, we perform ultrasound weekly until resolution of TAPS and deliver these pregnancies between 34+0 and 37+0 weeks of gestation. If there is a suspicion for an intracranial lesion, we perform fetal MRI.

Rationale — The management approach described above is supported, in part, by data from a series of 376 MC twin pregnancies prenatally diagnosed with stage 1 to 4 TAPS and managed in 17 fetal therapy centers (table 4) [31]. Stage-based outcomes were not reported and the optimal management was unclear because of extensive heterogeneity in the management of TAPS, both within and among fetal therapy centers. Nevertheless, these data suggest that laser surgery treated the cause of TAPS and most prolonged the overall diagnosis to birth interval in the absence of feticide, thus enabling more time for fetal maturation, but it was technically challenging and did not clearly improve or worsen perinatal outcome when compared with expectant management. The authors hypothesized that definitive treatment with laser might be the optimal intervention remote from term to improve perinatal outcome since expectant management likely results in continuous exposure to the potential detrimental effects of TAPS and risks for perinatal mortality and morbidity increase with increasing TAPS stage.

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

●Etiology/pathophysiology – Twin anemia-polycythemia sequence (TAPS) is characterized by highly discordant hemoglobin values between monochorionic (MC) twins. It may occur spontaneously or post-laser ablation for twin-twin transfusion syndrome (TTTS). It is caused by unbalanced red cell transfusion across tiny placental anastomoses in MC placentas that gradually leads to anemia in the donor twin and polycythemia in the recipient twin. Amniotic fluid volumes remain normal. (See 'Pathophysiology' above.)

●Monitoring – We monitor MC twin pregnancies for development of TTTS, TAPS, and selective fetal growth restriction (sFGR) (table 2). To identify TAPS, we perform serial middle cerebral artery (MCA) peak systolic velocity (PSV) measurements starting at 16 weeks and assess for placental discordance in echogenicity and cardiomegaly. (See 'Clinical presentation' above.)

●Diagnosis – TAPS has traditionally been diagnosed prenatally when the middle cerebral artery-peak systolic velocity (MCA-PSV) is >1.5 multiples of median (MoM) in one twin (suggestive of anemia) and <0.8 MoM in the other twin (suggestive of polycythemia. An alternative approach is to use an absolute intertwin MCA-PSV discordance (delta MCA-PSV) ≥1 MoM. Discordant placental echogenicity, cardiomegaly (donor twin), and starry sky liver (recipient twin) support the diagnosis. (See 'Prenatal diagnosis' above.)

●Diagnostic evaluation and staging – After making the diagnosis of TAPS, Doppler flow studies of the umbilical artery (UA), umbilical vein (UV), and ductus venosus (DV) should be performed to classify disease severity (table 3), which is the basis for further follow-up and management. (See 'Staging (disease severity)' above.)

●Differential diagnosis – Although there is some overlap between the findings for TAPS and TTTS (table 1), key characteristics that help in differential diagnosis are that MCA-PSV discordance is diagnostic of TAPS and not present in pure TTTS and amniotic fluid discordance (oligohydramnios and polyhydramnios sequence) is suggestive of pure TTTS and not present in TAPS. (See 'Differential diagnosis' above.)

●Prognosis – In the absence of fetal treatment, TAPS may result in delivery of two healthy neonates, need for neonatal blood transfusion or partial exchange transfusion, or death of one or both twins. Outcomes with various management approaches are shown in the table (table 4). TAPS resolves spontaneously in a minority of cases (16 percent in one large series). (See 'Prognosis' above.)

●Pregnancy management

•For patients with stable stage 1 TAPS, we suggest expectant management (Grade 2C). In the absence of progression to a more advanced stage or resolution, we deliver these pregnancies between 34+0 and 37+0 weeks of gestation. (See 'Pregnancy management' above.)

•For patients with stage 2, 3, or 4 TAPS before 32 weeks, we suggest intervention rather than expectant management (Grade 2C). The choice of intervention (pregnancy termination, laser ablation, fetal transfusion, selective feticide) is based on gestational age, disease stage, patient values and preferences, and physician expertise. If TAPS resolves, we deliver between 34+0 and 37+0 weeks of gestation. (See 'Pregnancy management' above.)

For patients with stage 2, 3, or 4 TAPS ≥32 weeks, we suggest delivery rather than intervention (Grade 2C). (See 'Pregnancy management' above.)