Is it time to add aortic isthmus evaluation to the repertoire of Doppler investigations for placental insufficiency?

During fetal life, the right and left cardiac ventricles perfuse the systemic circulation in parallel. In a non-stressed fetus, right ventricular output perfuses mainly the infradiaphragmatic area, only a small fraction of its output being directed to the pulmonary circulation, and the left ventricle is responsible for the blood supply of the upper body1-3. A characteristic of the fetal circulation is the presence of shunts. The ductus venosus and foramen ovale ensure that highly oxygenated blood enters the coronary and cerebral circulations. From a physiological point of view, the third anatomical shunt, the ductus arteriosus, is only a part of the right ventricular vascular outlet, branching from the main pulmonary artery and connecting to the descending aorta. The aortic isthmus (AoI) is a true physiological arterial shunt in the fetus. This connection between the left subclavian artery and the ductus arteriosus establishes a link between the right and left ventricles, perfusing the upper and lower body in parallel. In systole, the direction of blood flow in the AoI is dependent on the relative contributions of the right and left ventricular stroke volumes and the balance between their downstream impedances. During diastole, the aortic and pulmonary valves are closed, and thus the direction of blood flow in the AoI reflects solely the downstream impedances of the right and left ventricles (mainly placental and cerebral vascular impedances, respectively)4. These physiological properties of the AoI may help us to assess ventricular dysfunction in cases with ventricular outflow tract obstruction, to detect arteriovenous fistulae in the upper body and to evaluate fetal wellbeing during acute and chronic fetal hypoxemia4-13. The diameter of the AoI increases linearly with advancing gestation, but there is a relative linear narrowing of the AoI compared with the descending aorta throughout late gestation14, 15. In normal human pregnancy, the fetal AoI waveform may show a short retrograde blood flow component in the late third trimester, which may reflect the normal increase in placental and fetal systemic vascular resistance16. The AoI waveform can be obtained either from a sagittal view of the aortic arch or from the three vessels and trachea view17. Quantitative (velocities), semiquantitative (pulsatility index, isthmic flow index (IFI), antegrade time velocity integral (TVI)/retrograde TVI) and qualitative (presence of retrograde flow during diastole) measures have been used in its assessment4-13. Normal values for AoI Doppler parameters with advancing human gestation are available and moderate to high intra- and interobserver reproducibility in the assessment of these parameters has been reported16, 18. In placental insufficiency, fetal wellbeing can be monitored by examining hypoxemia-triggered compensatory signs in the fetal circulation. Placental vascular compromise results in decreased right ventricular cardiac output, while left ventricular cardiac output is usually maintained19. Further deterioration of oxygen supply may lead to a decrease in left cardiac output. In the presence of cardiac dysfunction, abnormal pulsatility in the precordial veins may be detected as a sign of increased systemic venous pressure. In clinical practice, calculation of ventricular cardiac output is complicated, but evaluation of systemic venous pulsatility is feasible; abnormalities have been shown to precede cardiovascular collapse and thereby have predictive value for adverse neonatal outcome in placental insufficiency20-22. Doppler assessment of the AoI also provides important information on fetal ventricular performance and cardiovascular hemodynamic status. Growth-restricted fetuses that maintain antegrade AoI net blood flow restrict their pulmonary volume blood flow and increase volume blood flow via the foramen ovale in order to ensure a highly oxygenated blood supply to the coronary and cerebral circulations9. Fetuses with retrograde AoI net blood flow fail to shift their right ventricular output from the pulmonary to the systemic circulation, and an increase in their left atrial pressure may further diminish the volume blood flow via the foramen ovale9. Consequently, even in the presence of similar umbilical venous pO2, fetuses with retrograde AoI net blood flow are predisposed to cardiac and cerebral hypoxemia, because the difference in the oxygen saturation of blood ejected from the right and the left ventricles is about 10–12%23. Indeed, experimental and human studies suggest that AoI Doppler assessment indirectly indicates changes in cerebral oxygen delivery5, 6, 10, 24. Whether diminished cerebral oxygen delivery causes significant tissue hypoxia and cerebral damage in the plastic fetal brain is another question. The poorer neurodevelopmental outcome of growth-restricted fetuses delivered at a mean of 33 weeks and with an abnormal AoI blood flow profile, however, underlines the clinical significance of this finding5, 6. On the other hand, previous studies also suggest that several other factors may influence the outcome, while a significant number of fetuses with suboptimal clinical outcome have shown normal AoI Doppler findings just prior to delivery5, 6, 13, 16, 25, 26. In addition, the tolerance for cerebral hypoxemia may vary according to gestational age. In a small population of premature (< 32 weeks) growth-restricted fetuses, retrograde AoI net blood flow was not more frequent in cases with suboptimal outcome compared with those with normal neurodevelopmental outcome at the corrected age of 1 year26. Furthermore, maternal hyperoxygenation in human pregnancy results in a less obvious change in the fetal AoI waveform at midgestation compared with later in the third trimester, and responses of the fetal pulmonary circulation to changes in fetal pO2 also vary according to gestational age27, 28. In clinical practice, an optimal tool/parameter would identify all fetuses at risk for adverse outcome without increasing the number of iatrogenic deliveries at any gestational age, but especially premature deliveries. Fetal AoI studies have focused mainly on neurodevelopmental outcome, but in this issue of the Journal, Del Rio et al.13 report on the association of AoI Doppler assessment and perinatal outcome in 51 preterm fetuses with intrauterine growth restriction (birth weight < 10th percentile and cerebroplacental ratio < 5th percentile). In their study, AoI blood flow parameters were obtained within 48 h prior to delivery from fetuses delivered between 25 and 37 gestational weeks. The indications for delivery were non-reassuring fetal testing, fetal demise and/or worsening maternal condition. Perinatal outcome was considered adverse if stillbirth, neonatal death, bronchopulmonary dysplasia, intraventricular hemorrhage and/or significant retinopathy were diagnosed, or if neonatal intensive care exceeded 14 days. About 18% of the studied fetuses died either in utero or within the first month of postnatal life. The numbers of stillbirths and postnatal deaths, respectively, were 1 + 1/41 in the antegrade AoI group (IFI I) and 5 + 2/10 in the retrograde AoI group (IFI III and V), and adverse outcome was detected in a total of 10/41 and 9/10 cases, respectively. All fetuses with intrauterine or postnatal death were delivered prior to 28 weeks, and the birth weights in the antegrade and retrograde AoI groups were 270–2080 g and 275–1050 g, while the median gestational ages at delivery were 32 weeks and 27 weeks, respectively. The authors concluded that retrograde AoI blood flow correlated significantly with adverse perinatal outcome, and pulsatility in both the AoI and the ductus venosus accounted independently for adverse outcome. From a clinical point of view, follow-up studies like that of Del Rio et al.13 are of crucial importance. Clinicians welcome identification of non-invasive sonographic parameters useful in fetal surveillance and in developing management protocols in order to optimize the timing of delivery as well as to improve short- and long-term outcomes. Often, fetal studies focus on immediate outcome measures at the time of delivery (e.g. Apgar score, pH), even though these parameters do not necessarily reflect later clinical outcome. Perinatal mortality is an extreme measure of our success in practicing maternal–fetal medicine. With advances in perinatal and neonatal care, tinier and tinier babies survive, and we must focus on both short-term and long-term morbidity. However, we have to keep in mind that several factors influence the results of studies focusing on clinical outcome. Thus, careful interpretation of the data is needed before the results can be applied to clinical practice. Gestational age at the time of delivery plays a significant role as regards to neonatal outcome. In a multicenter ultrasound study on growth-restricted fetuses < 33 weeks, gestational age was the most significant determinant of survival until 26 weeks and intact survival until 29 weeks22. In their article, Del Rio et al.13 wisely state that even though the likelihood ratio for adverse outcome was increased with a retrograde AoI blood flow profile finding, difference in gestational age was a significant confounding factor, itself having a high sensitivity and specificity as a predictor of adverse outcome and mortality. The authors did not apply statistical methods to overcome this problem, but future studies will hopefully help clinicians in their task of weighing the risks of prematurity against those of abnormal AoI Doppler findings and trying to optimize the timing of delivery in the early third trimester. Furthermore, with respect to gestational age, fetuses with placental insufficiency are monitored frequently, increasing the likelihood of iatrogenic premature delivery due to the clinician's expectations. Thus, in order to gain objective information, the criteria for inducing delivery should be clear, and, preferably, the managing clinician should be blinded to the results of the examination, or a randomized study protocol should be applied. We may, of course, ask ourselves whether this is ethically correct, particularly when, for example, Fouron et al.5, 6 have already demonstrated adverse neurodevelopmental outcome at the age of 2–4 years in premature growth-restricted fetuses with retrograde AoI net blood flow. In fact, in our field, clinical decisions are often influenced by or based on descriptive observations, and randomized clinical trials are almost non-existent. In addition to these major problems, the time interval between the Doppler assessment and delivery as well as the mode of delivery may affect results. Even in a study setting, an ideal, extremely short, time interval may not be feasible. Furthermore, longer time intervals may, in fact, better reflect the reality in clinical practice. Growth-restricted fetuses with early onset placental vascular compromise are often delivered via Cesarean section, as was found in the study by Del Rio et al.13, and thus the mode of delivery and possible acute on chronic hypoxemia seem to play no role in confounding the results. In any study, the number of emergency operations and indications for Cesarean delivery clarify the impact of the mode of delivery. Pre- and postnatal management naturally influences the clinical neonatal outcome and antenatal corticosteroid treatment plays a role in neonatal mortality and morbidity29. Del Rio et al.13 reported that, in their study, all fetuses delivered prior to 34 gestational weeks received antenatal corticosteroids. In this respect, antenatal care did not seem to differ between the groups and reflects current standard care. As regards to postnatal treatment, one can assume that, within the same institution, the treatment protocols do not vary greatly, but this aspect must be kept in mind when interpreting results from multicenter studies and when extremely small babies (< 500 g) are included in the study. Intrauterine malnutrition has been demonstrated to affect our lives up to adulthood30. The ultimate challenge for future research on intrauterine growth restriction is to determine the pathophysiological mechanisms related to placental vascular compromise and to develop intrauterine treatment. Until intrauterine treatment options are available, fetal studies focusing on short- and long-term outcomes are essential, in order to help us to identify fetuses at risk and, hopefully, to optimize their outcome. One of the major problems related to long-term clinical outcome studies is that, at the time of the outcome analysis, the tested variable used in the initial investigation is no longer considered relevant. Pioneering work by Fouron's group5-7, 10, 11, 16 has demonstrated the potential of AoI Doppler in the evaluation of fetal wellbeing during chronic hypoxemia. The findings of previous studies should encourage us to clarify the role of AoI assessment in the prediction of adverse clinical outcome throughout gestation.