A 28-year-old primigravida was referred to our institution at 15 + 6 weeks' gestation owing to suspected twin–twin transfusion syndrome (TTTS). Serial ultrasound examination showed normal amniotic fluid volume in both sacs and normal bladder filling in both fetuses, without signs of TTTS. Routine measurements of middle cerebral artery peak systolic velocity (MCA-PSV) with Doppler ultrasound were performed on a weekly or biweekly basis. MCA-PSV in Twin A (recipient) was low (< 0.8 multiples of the median (MoM)), suggesting the presence of polycythemia, whereas MCA-PSV in Twin B (donor) was often increased (> 1.5 MoM), suggesting the presence of anemia (Figure 1), indicative of twin anemia–polycythemia sequence (TAPS). After counseling, the parents opted for a conservative management approach.
At 30 weeks' gestation, the patient was referred back to her own obstetrician. At 34 + 5 weeks, a Cesarean section was performed owing to signs of fetal distress in the recipient. The first-born twin (recipient) was plethoric and weighed 2065 g and the second-born twin (donor) was pale and weighed 1805 g. Umbilical cord pH in the first and second twins was 7.21 and 7.20, respectively and Apgar scores (at 1 and 5 min) were 8/9 and 9/9, respectively. Hemoglobin levels and reticulocyte counts in recipient and donor were 25.4 and 4.3 g/dL (difference, 21.1 g/dL) and 12.3 and 66.4% (reticulocyte count ratio of 5.4), respectively, fulfilling the postnatal criteria for severe TAPS stage 5.
A partial exchange transfusion was performed in the recipient twin, who also had thrombocytopenia at birth (platelet count, 98 × 109/L). Soon after the partial exchange transfusion, the recipient developed respiratory insufficiency with apnea and was transferred to the neonatal intensive care unit for mechanical ventilation. The infant subsequently developed disseminated intravascular coagulation (DIC) and worsening thrombocytopenia, for which he received several fresh frozen plasma and platelet transfusions. Soon after admission, subclinical status epilepticus was detected using two-channel amplitude-integrated electroencephalography (aEEG). Following a loading dose of phenobarbital, the background pattern changed from a discontinuous normal voltage to an iso-electric trace without subsequent recovery during admission (Figure 2). Cranial ultrasonography performed on day 1 showed severely increased echogenicity, most marked in the temporal lobes. Magnetic resonance imaging (MRI) performed on day 3 showed intra- and extra-axial hemorrhages filling all sulci and fissures, compressing both hemispheres, and severe diffuse ischemia with complete loss of gray–white matter differentiation (Figure 3). Furthermore, there was loss of arterial and venous cerebral blood flow on both MRI and Doppler ultrasonography. Diffusion-weighted imaging showed an increased signal of the cortical and subcortical white matter and cerebellum, indicating diffuse ischemia of almost the entire brain (Figure 3). In view of the MRI and aEEG findings, the decision was made to redirect care and the infant died on day 3. Permission for autopsy was not obtained.
Macroscopic examination of the placenta revealed a plethoric appearance of the recipient's share of the placenta and a pale appearance of the donor's share. Injection with colored dye showed two minuscule unidirectional arteriovenous (AV) anastomoses (diameter 0.3 and 0.2 mm, respectively) from donor to recipient and one AV anastomosis (diameter 0.2 mm) from recipient to donor (Figure 4).
We report on a recipient twin with spontaneous TAPS who suffered massive cerebral injury due to severe polycythemia–hyperviscosity syndrome. This is the first report showing that spontaneous TAPS may lead to extensive cerebral damage, emphasizing the need for increased awareness of the importance of timely detection of the condition and for further research into possible treatment options.
TAPS is a form of chronic fetofetal transfusion, characterized by large intertwin hemoglobin differences, without signs of twin oligo–polyhydramnios sequence (TOPS). TAPS may occur spontaneously or after laser surgery for TTTS (post-laser form). The spontaneous form complicates approximately 3–5% of monochorionic pregnancies[1, 3], whereas the post-laser form occurs in 2–13% of TTTS cases[4, 5]. The pathogenesis of TAPS is based on the presence of few, minuscule AV anastomoses (diameter < 1 mm) allowing slow transfusion of blood from donor to recipient and leading gradually to discordant hemoglobin levels.
Diagnosis can be made antenatally with Doppler ultrasound, showing an increased MCA-PSV (> 1.5 MoM) in the donor (suggestive of fetal anemia) and a decreased MCA-PSV (< 0.8 MoM) in the recipient (suggestive of polycythemia), in the absence of TOPS. In the current case, the MCA-PSV in the recipient was mostly < 0.8 MoM, but Doppler values in the donor were often (but not always) > 1.5 MoM. However, the hemoglobin level at birth in the donor (4.3 g/dL) was lower than expected. The discrepancy between the observed hemoglobin level and the findings on prenatal Doppler ultrasound could be due to several factors, including imprecise Doppler measurements or worsening of the MCA-PSV measurements from 30 weeks' gestation (it is of note that MCA-PSV measurements were not performed at 32 weeks or prior to delivery at 34 weeks). Importantly, the sensitivity, specificity and predictive value of MCA-PSV measurements have not yet been evaluated in a TAPS cohort.
Postnatal diagnosis of TAPS is based on the presence of (chronic) anemia in the donor (including reticulocytosis) and polycythemia in the recipient, in association with typical placental angioarchitecture as identified by injection of colored dye. According to a recently proposed classification system, the present case fulfilled the postnatal criteria for the most severe stage of TAPS (Stage 5).
Perinatal morbidity and mortality rates in TAPS are not well known and vary according to the severity of TAPS, reflecting the heterogeneous nature of this disease. Outcome may range from dual intrauterine demise to the birth of two healthy neonates without major morbidity aside from a large discordance in hemoglobin levels.
The optimal management of TAPS is not clear, and includes expectant management, induction of labor, intrauterine blood transfusion (IUT (intravenous and/or intraperitoneal))[8, 9], selective feticide and fetoscopic laser surgery[1, 10, 11]. In the present case, treatment with IUT for the donor twin with fetal anemia would probably have resulted in a further increase of polycythemia–hyperviscosity in the recipient and thus worsened its condition. We hypothesize that an alternative approach in TAPS management would be to combine IUT with intrauterine partial exchange transfusion in the recipient twin to reduce hyperviscosity. Fetoscopic laser surgery, to coagulate the anastomoses, is a valid alternative and could have prevented the fatal outcome. However, laser surgery is technically challenging owing to the absence of polyhydramnios and the presence of only minuscule anastomoses. Moreover, laser surgery is associated with procedure-related complications such as premature rupture of the membranes, disruption of the intertwin dividing membrane and fetal demise. Whether induction of labor before 32 weeks' gestation would have prevented the fatal outcome is not clear, as the optimal timing of delivery in TAPS remains to be determined. Induction of labor should always be weighed against the disadvantages of prematurity, and the vast majority of TAPS cases delivered after 32 weeks' gestation have favorable outcomes[2, 10, 12]. In addition, the exact timing of the onset of the cerebral lesions in this case was not clear. Cerebral lesions on cranial ultrasound and neurologic symptoms were already present soon after birth (day 1), suggesting that cerebral injury was most probably of antenatal rather than postnatal origin.
Knowledge of the neonatal and pediatric morbidity in TAPS cases is limited and based mostly on case reports and small series. Neonatal morbidity appears to be mainly limited to hematological problems at birth. Donors may be severely anemic and require blood transfusion, whereas recipients may be severely polycythemic and require partial exchange transfusion. Robyr et al. reported a case of post-laser TAPS treated with several IUTs in which the recipient developed skin necrosis of the leg (hemoglobin level at birth, 28 g/dL).
The incidence of cerebral injury and neurodevelopmental impairment in TAPS has not been well studied, and to the best of our knowledge this is the first case report showing that recipient twins may also suffer severe cerebral injury. The exact cause is of this is not clear. The infant had extensive intra- and extra-axial hemorrhages with a bilateral intraventricular hemorrhage and hemorrhagic lesions in the occipitotemporal and parietal periventricular white matter. This appearance is not dissimilar to what has been described as ‘the white brain’ in the full-term infant with hypoxia–ischemia. However, the infant reported on here had good Apgar scores and only deteriorated a few hours following the partial exchange transfusion. This may have been due to extensive subarachnoid and parenchymal hemorrhages with hemorrhagic diathesis and intravascular coagulation with thrombocytopenia. The timing of these hemorrhages is not clear and may have been antenatal or immediately postnatal, worsened by suboptimal coagulation with subsequent DIC. DIC might have been caused by consumption of coagulation factors due to severe polycythemia–hyperviscosity syndrome. Polycythemia may have led to vascular sludging, obstructing the sinuses and the inflow and outflow of the brain, thus leading to cerebral sinovenous thrombosis. Sinovenous thrombosis is known to be associated with intraventricular hemorrhage and parenchymal infarction. However, although magnetic resonance venography suggested the presence of cerebral sinus venous thrombosis with loss of flow across the superior sagittal sinus and straight sinus, it is more likely that in this patient the lack of flow was due to increased intracranial pressure caused by the substantial amount of subarachnoid blood, which led to compression of the sinuses.
In conclusion, spontaneous TAPS can lead to severe cerebral injury in the recipient twin when managed conservatively. Increasing awareness and careful monitoring extending to the third trimester are crucial in order to improve the detection and treatment of future TAPS cases. More information on short- and long-term morbidity is needed, ideally through international cooperation between fetal centers using a web-based registry.