Congenital heart defects (CHDs) account for the most common congenital abnormality, occurring in one out of one hundred and ten live births
[1][2]. Fortunately, prenatal detection of CHDs by fetal echocardiography is becoming increasingly common
[3][4][5] due to advances in diagnostic imaging and improved obstetrical screening guidelines. Fetal echocardiography can evaluate fetal cardiac anatomy and the progression of the disease. Serial assessment throughout the pregnancy with late gestation fetal echocardiography can help determine the anticipated hemodynamic changes after birth and guide perinatal management. This accurate prenatal assessment with individualized perinatal management plans has improved in utero and postnatal outcomes for CHDs
[6][7][8]. Although most newborns with CHDs are stable soon after birth, some babies with specific CHDs have severe hemodynamic instability immediately after the placental separation as the fetal circulation starts transitioning to postnatal circulation. Survival in these newborns with critical CHDs requires immediate stabilization in the delivery room along with lifesaving catheter or surgical interventions within the first few hours of life. Examples of these conditions include hypoplastic left heart syndrome (HLHS) with a restrictive or intact atrial septum or d-transposition of the great arteries (d-TGA), most often with a restrictive ductus arteriosus. In babies with these CHDs, opening the atrial septum with balloon atrial septostomy (BAS) is often needed. Another example of a critical CHD requiring urgent cardiac surgical intervention is obstructed total anomalous pulmonary venous return (TAPVR). Accurate prediction of neonatal hemodynamic instability can be challenging using standard fetal echocardiography due to an inability to predict changes from prenatal to postnatal circulation. The addition of maternal hyperoxygenation (MH) testing to standard fetal echocardiography has been shown to improve the accuracy of predictions related to postnatal hemodynamics, thus providing an opportunity for necessary resource planning for anticipated lifesaving cardiac interventions
[9][10][11][12][13].
1.1. Fetal Circulation and Transition at the Time of Birth
The normal fetal blood flow pattern (Figure 1A) is characterized by “parallel circulation”, which significantly differs from “in-series” circulation seen after birth (Figure 1B). This parallel fetal circulation is due to three fetal shunts, including ductus venosus, foramen ovale, and ductus arteriosus. The ductus venosus brings the nutrient and oxygen-rich blood from the umbilical vein to the right atrium via the inferior vena cava. The foramen ovale allows for the distribution of nearly half of this systemic venous return to the left side of the fetal heart. The left ventricle (LV) output is mostly distributed to the coronary arteries and the upper part of the fetal body via three branches that originate from the aortic arch. In contrast, most of the output from the right ventricle (RV) passes through the ductus arteriosus to be distributed to the lower part of the fetal body and then back to the placenta via the umbilical arteries. This preferential flow from the RV to the ductus arteriosus is due to the high pulmonary vascular resistance resulting in only a small percentage of the RV output being sent to the branch pulmonary arteries (PAs). Therefore, in fetal circulation, the right and left ventricles are “parallel” to each other, with both ventricles handling a part of the combined cardiac output, and a very small amount going to the lungs.
Figure 1. Normal fetal circulation (
A) and postnatal circulation (
B). Removal of placenta at the time of birth and subsequent closure of the fetal shunts (ductus arteriosus, ductus venosus and foramen ovale) within a few days causes the transition of fetal circulation to postnatal circulation
[14].
At the time of birth, dramatic changes occur, in the transition from prenatal circulation to postnatal circulation. Systemic vascular resistance increases due to the placental separation, and pulmonary vascular resistance decreases due to spontaneous respiration and pulmonary vasodilation promoted by increased oxygenation. In addition, increased oxygenation leads to constriction of the ductus arteriosus. These changes cumulatively increase pulmonary blood flow, pulmonary venous return, and left atrial pressures, closing the foramen ovale. With the closure of the fetal shunts, including the foramen ovale, ductus arteriosus, and ductus venosus, the circulatory transition is completed such that the postnatal flow through the right and left heart is “in-series”
[15]. Due to the differences between fetal and postnatal circulation, accurately predicting the postnatal hemodynamic effects of CHDs is challenging using standard fetal echocardiography alone. Addition of MH testing by giving 100% oxygen to the mother via a non-rebreather mask for 10 to 15 min partially mimics these postnatal circulatory changes in the fetal circulation, allowing for more accurate prediction of hemodynamic instability following birth as described below.
1.2. Historical Perspective on Maternal Hyperoxygenation Testing
Acute MH testing has been studied for over five decades. Studies have focused on evaluating hemodynamic changes to increased circulating oxygen content in the fetal blood, mimicking the postnatal circulatory physiology. Bertolizio et al.
[16] and Frangipani et al.
[17] published early reports of the effect of MH on amniotic fluid acid–base equilibrium in 1966 and 1969, respectively. However, the cardiac and circulatory changes with MH were not easy to evaluate prior to the availability of fetal echocardiography in the late 1980s. One of the early studies evaluating the cardiac effects of short-term MH reported that abnormal E/A ratio across the mitral and tricuspid valve inflow Doppler patterns seen in growth-restricted fetuses could be improved with MH
[18]. Another early study by Soregaroli et al.
[19] published in 1993 reported increased peak flow velocities in ductus venosus after MH but no effect on fetal heart rate. This effect of MH on fetal circulation was more obvious in the third trimester compared to early gestational age
[20]. A randomized study published in 1997 by Ramner et al.
[21] reported that reduction in the pulmonary vascular impedance (measured as pulsatility index in proximal and distal right and left PAs) with MH is significant at 31–36 weeks of gestation but not at 20–26 weeks of gestation. A more detailed evaluation of the fetal echocardiographic findings in the same cohort by Rasanen et al.
[22] further characterized this pulmonary vasoreactivity. With acute MH in late gestation, increased pulmonary blood flow was suggested by a reduction in the pulsatility index (PI) in the branch Pas, and a reduction in ductus arteriosus flow was suggested by an increase in the PI of ductus arteriosus. In addition, there was a reduction in flow across the foramen ovale. All of these circulatory changes returned to baseline after MH was discontinued. Again, these changes were observed only in late gestation (31–36 weeks of gestation) fetuses and not in early pregnancy. This development of pulmonary vasoreactivity has been attributed to the smooth muscle development in the fetal PAs during late gestation
[23]. Together, these early studies suggested that acute MH could temporarily mimic the postnatal changes in fetal circulation and paved the way to study the utility of MH in guiding postnatal resource preparation in complex CHDs. Additionally, these studies showed that all changes returned to baseline after MH was discontinued, and no untoward side effects were noted in the fetus or the mother, indicating the safety of such testing prenatally.
2. Clinical Maternal Hyperoxygenation Protocol
The physiologic change of increased fetal oxygen concentration via MH is achieved by administering oxygen to a mother in late gestation for a short duration. There is no universally accepted MH protocol, but most studies have given 100% humidified oxygen to expectant mothers via a non-rebreather mask for 8 to 15 min (Table 1).
Table 1. List of prior studies evaluating the utility of acute maternal hyperoxygenation testing in predicting postnatal hemodynamics in complex congenital heart defects.
Fetal echocardiography, including Doppler blood flow analysis, is performed at baseline, and fetal echocardiographic variables of interest are re-evaluated after the acute MH for 10 to 15 min to assess the changes in blood flow and vascular impedance across various fetal cardiac structures. The fetal echocardiographic variables of interest are based on the individual CHDs. These lesion-specific changes anticipated with MH are described in detail below and summarized in Table 1 and Table 2.
Table 2. Fetal echocardiographic findings and changes with maternal hyperoxygenation as potential predictors of postnatal hemodynamic instability and need for urgent cardiac intervention with various CHDs.
Based on the MH protocols used in prior studies, 100% humidified oxygen is typically delivered to the mother using a non-rebreather mask, which provides around 60% inhaled oxygen concentration to the mother for 10 min before reevaluating select fetal echocardiographic measures. Although initial studies reported repeat assessment after 10–15 min of recovery, many later studies did not evaluate recovery phase hemodynamics in light of prior reports of complete resolution of the circulatory changes without untoward effects on the mother or the fetus. There is a learning curve with establishing MH testing in a fetal echocardiography laboratory. Consistently obtaining branch PA pulse wave Doppler (PWD) and deriving PI is very important in gathering accurate information from this test. However, repeatability for branch PA PWD can be challenging. Baseline PWD repeated during the same fetal echocardiogram can provide PI values that have more than 10% variability. Here, the following contents share a few tips and tricks to improve the repeatability of branch PA PWD, consistency in measurements to derive accurate PI values, and interpretation of MH testing: (1) Determining the site of obtaining branch PA PWD. Three specific sites for obtaining branch PA PWD have been described by Szwast et al.
[9]. It is helpful to practice one site of interrogation at first, and we have noted success at the mid-branch PA level. PI values vary depending on the site of PWD interrogation in the branch PA. Hence, obtaining the PWD at the same site is crucial after MH testing. (2) Keeping the PWD angle of interrogation to <10°. (3) Keeping the same PWD gain and PWD scale and using the same probe pre and post-MH. Since MH testing is typically performed in the late trimester, using lower-frequency transducer. (4) When fetal position changes post-MH testing such that PWD cannot be repeated at the same angle, recognizing the expected pattern change in the PWD, not just the PI value itself. Characteristic branch PA PWD signal is spiky with a quick sharp upstroke, short systolic time interval, and absent, or only a small degree of flow in diastole. With MH, the PWD pattern should become wider with more flow in both systole and diastole (
Figure 2). (5) Verifying the automatic tracings and, when needed, performing manual measurements to derive PI. Some imaging platforms, such as Philips, can derive PI values from PWD using the high Q automatic Doppler analysis. This method can be erroneous, especially given the low end-diastolic velocities in branch PA PWD. (6) Finally, establishing an internal quality improvement project to reduce variability between repeated PWD PI values to <10% is recommended.
Figure 2. Maternal hyperoxygenation (MH) testing in a third-trimester fetus with hypoplastic left heart syndrome with mitral and aortic stenosis. (A) Oblique sagittal color compare image showing a small tunnel-like patent foramen ovale (PFO) with left to right flow. (B,C) represents the baseline test. (B) Left pulmonary artery Doppler (LPA) showing a spiked pattern with pulsatility index (PI) 2.42. (C) Right sided pulmonary vein (RPV) Doppler with forward to reverse velocity time integral (F/R VTI) ratio of 6.75. The bottom panel (D,E) represents post MH testing. (D) LPA Doppler broadening with PI 1.82, 24.7% reduction from baseline. (E) RPV Doppler showing increase in F/R VTI ratio to 10.4. LA, left atrium; PDA, patent ductus arteriosus; RA, right atrium.