Ultrasound assessment of twin reversed arterial perfusion sequence for the prediction of adverse pregnancy outcomes in the first trimester

To investigate twin reversed arterial perfusion (TRAP) sequence for the prediction of TRAP‐related adverse pregnancy outcomes at the gestational age of 11–14 weeks.

the acardiac twin is characterized by the absence of cardiac structure.
The survival of the acardiac twin depends on the blood provided by the pump twin. Most of acardiac twins die after birth. The acardiac twin severely influences the intrauterine development of the pump twin.
Studies have shown that approximately 60% of pump twins can develop heart failure, intrauterine death, miscarriage, and premature birth due to the acardiac twin presence. 1,2 Ultrasound examination is the first-line method to diagnose and monitor TRAP. One study demonstrated that larger acardiac twins were associated with a higher incidence of adverse pregnancy outcomes (30%, 40%, and 90% for heart failure, edema, premature delivery, and miscarriage, respectively). 2 As recommended in the World Health Organization guidelines on the tertiary prevention of birth defects, researchers and medical practitioners globally are committed to the early detection of TRAP. Therefore, it is important to identify the ultrasound indicators for the prediction of adverse pregnancy outcomes of pump twins with TRAP at 11-14 weeks of gestation. This will assist in devising a reasonable pregnancy management strategy to promote the survival of pump twins. Some studies reported that ultrasound indicators such as pump twin's crown-rump length (CRL) and acardiac twin's upper pole-rump length (URL) in early pregnancy were accurate in indicating the outcomes for the twins. The (CRL -URL)/CRL and URL/CRL ratios were significantly associated with adverse pregnancy outcomes of the pump twins. 1,3 However, there are no accepted criteria to predict adverse pregnancy outcomes of pump twins.
We conducted a study that aimed to explore the ultrasound indicators for the prediction of outcomes for the pump twin with TRAP at a gestational age of 11-14 weeks.

Study participants
We performed a study of pregnant women affected by TRAP sequence

Ultrasound examination
Voluson 730, Voluson E8, and iU22 color Doppler flow imaging (GE and Philips) were used for a transabdominal ultrasound examination. If nec-essary, a combination of transvaginal and ultrasound examination was performed. TRAP sequence was identified as twins with one anatomically normal retrograde perfusion and the other without functional cardiac activity in a monochorionic twin pregnancy. Monochorionic diamniotic (MCDA) is defined as a very thin membrane between the twins and the absence of "λ" sign via ultrasound detection in the monochorionic twin pregnancy. Monochorionic monoamniotic (MCMA) is defined as the absence of membrane echo between the twins.
The following data from ultrasound examination were collected: CRL of the twins, URL of the twins, nuchal translucency (NT) thickness, ductus venosus (DV) blood flow spectrum, and birth weight of the pump twin. An abnormal NT thickness was defined as NT ≥3 mm. An abnormal DV was defined as an inverted or disappearance of the a-wave in the pump twin.
We performed a routine systemic ultrasound screening of the pump twin and observed whether there were abnormal anatomical structures in the organs of the pump twin. We assessed the development of the twins, as well as the placenta and amniotic fluid of the twins. We observed the vessels, umbilical cord, and umbilical artery blood flow.
Additionally, we assessed whether there was an overlap of the twins' umbilical cords at the placental attachment site and the site of umbilical cord insertion into the placenta. Finally, we evaluated whether there was cord entanglement in the MCMA twins.

Statistical analysis methods
We analyzed relevant indices using the Man-Whitney U test, and multiple logistic regression variable data were given as median (interquartile range (IQR)). SPSS 16.0 and STATA were used for the analysis of the data. p-Values < 0.05 were considered statistically significant.

RESULTS
We prospectively recruited a total of 31 pregnant women diagnosed with TRAP from January 2015 to June 2018. The patients were 21-39 years of age; the twins' average (± standard deviation (SD)) ges-   Several studies have shown that the size of the acardiac twin was associated with the outcomes of the pump twin. 2 Moore et al. summarized the clinical outcomes of 49 cases of TRAP. 2 When the ratio of the acardiac twin body weight to the pump twin body weight was more than 70%, the risk of an adverse outcome for the pump twin increased.
Jelin et al. analyzed 18 acardiac twins whose body weights were no more than 50% of those of the pump twins; they found that only one pump twin with MCMA twins died in utero. 4 One study conducted in the Children's Hospital of Philadelphia, investigating 13 cases of TRAP, found that when the body weights of the acardiac twins were no more than 40% of those of the pump twins, no edema and heart failure were seen in the pump twins. 5 These findings were consistent with our findings. In the first trimester, it is easy to conduct the measurement of the fetal length using an ultrasound. Additionally, it can provide potential predictors to guide the clinical management of TRAP during early gestational ages. and that it may also increase the risk of skin damage and neural damage to the pump twin, as well as the death rate of the pump twin. 1,8,9 Therefore, it is pivotal to establish reasonable management strategies for TRAP.
Various studies suggested that expectant management may be a good management strategy for TRAP. One study conducted in Australia found that the survival rate of the pump twin was more than 80% after the expectant management of TRAP. 10  In some TRAP cases, the blood flow of the acardiac twins may be naturally blocked in development. Chen et al. reported that there were seven cases of naturally blocked TRAP, with three cases of intrauterine death and one case of polyhydramnion-induced premature birth. 13 In our study, we identified five TRAP cases with naturally blocked blood flow, of which four cases survived. In one case, the pump twin had polyhydramnion within 1 week after the natural blocking of the blood flow.
The pressure in the umbilical blood vessels abruptly decreased after the natural blocking of the blood flow of the acardiac twin. The blood flow of the pump twin was perfused to the acardiac twin, thereby increasing the cardiac burden on the pump twin. The pump twin consequently developed heart failure with a high cardiac output, and thereafter died of heart failure. Therefore, ultrasounds can be used in TRAP cases to monitor the development of the pump twin after the natural blood flow has been blocked.
Some studies have indicated that the resistance index of the umbilical artery and the pulsatility index of the middle cerebral artery may be indicators to assess the prognosis of TRAP. 14,15 However, these indicators for blood flow in TRAP were detected at gestational ages of greater than 20 weeks; thus, some important indicators may have eluded observation in the first trimester. Therefore, high-quality stud-ies are required in the future to explore the ability of ultrasounds to predict adverse outcomes of TRAP in the first trimester.

CONCLUSION
The (CRL -URL)/CRL and URL/CRL ratios were effective indicators for predicting adverse pregnancy outcomes for the pump twins with TRAP at 11-14 weeks of gestation in this study. After invasive management or the natural blood flow is blocked during expectant management, the pump twin with TRAP should be monitored using ultrasound.