Can Sonographic Assessment of Pulmonary Vascular Reactivity Following Maternal Hyperoxygenation Predict Neonatal Pulmonary Hypertension
Objective: Persistent pulmonary hypertension of the newborn (PPHN) occurs in 0.5 to 7 per 1000 live births and can result in significant cardiovascular instability. It occurs when there is a failure of the normal circulatory transition during the early newborn period. Recent studies have demonstrated that the fetal pulmonary vasculature responds in utero to maternal hyperoxygenation (MH). Maternal hyperoxygenation is used in an attempt to mimic the pulmonary vascular changes that occur at birth, when the neonate is required to take its first breath. The aim of this thesis work was to evaluate the fetal response to MH and to assess if fetal reactivity to MH would identify fetuses that would develop PPHN. Additionally, we aimed to ascertain the feasibility, acceptability and safety of performing the hyperoxygenation test in pregnant women.
Study Design: Forty-six women with a singleton pregnancy greater than or equal to 31 weeks’ gestational age were prospectively recruited to the study. Participants were selected and grouped as either, at risk of PPHN or as a study control. A fetal echocardiogram was performed on all subjects. A multivessel fetal Doppler examination was performed on the following vessels: the fetal pulmonary artery (PA), ductus arteriosus (DA), aortic isthmus (AoI), ductus venosus (DV), umbilical artery (UA) and middle cerebral artery (MCA). Pulsatility index (PI), resistance index (RI), peak systolic (PSV) and end diastolic velocity (EDV), time-averaged velocity (TAV), acceleration time (AT), and ejection time (ET) were measured within the distal PA. The acceleration-to-ejection time ratio (AT:ET) was used to assess pulmonary vascular resistance (PVR). Doppler measurements were taken at baseline and repeated immediately following MH for 10 minutes (O2 100% v/v inhalational gas) at a rate of 12L/min via a partial non-rebreather mask. Non-invasive cardiac output monitoring was undertaken for the duration of the hyperoxygenation exposure. Response to MH was considered where the decrease in PA PI was ≥10% from the baseline.
Postnatally, a comprehensive neonatal functional echocardiogram was performed within the first 24 hours of life to assess ejection fraction (EF), left ventricular output (LVO), and neonatal PA acceleration time (nPAAT). An additional twenty non-pregnant women were recruited to serve as controls for the evaluation of the safety of MH.
Results: The median gestational age during fetal assessment was 35 [IQR 33–37] weeks. There was a decrease in fetal PA PI and PA RI following MH of 21% [IQR 9-36] from the baseline. There was an increase in the median fetal PA AT (43ms [40-47] to 57ms [47-60], p=0.005) leading to an increase in AT:ET following MH (p=0.005), indicating a fall in PVR. No changes in the mean PIs of the UAD or MCA were observed following MH. There were no significant changes in the resistance indices of the DA. There was a significant increase in MCA blood flow, but not in MCA resistance indices. Fetuses that responded to hyperoxygenation were more likely to have a higher LVO (p<0.01) and EF (p=0.03) within the first 24 hours of life. These findings were not dependent on left ventricular size or mitral valve annular diameter but were related to an increased mitral valve inflow. There was no difference in nPAAT in the neonatal population. In the pregnant group there was a fall in maternal cardiac index (CI) during MH (p=0.009) coupled with a rise in systemic vascular resistance (SVR) with no recovery at ten minutes following the cessation of MH (p=0.02). In the non-pregnant group, there were no significant changes in any haemodynamic variable.
Conclusion: Maternal hyperoxygenation offers the opportunity to assess the reactivity of the fetal pulmonary vasculature before birth. Our findings indicate a reduction in fetal PVR with a resultant increase in fetal pulmonary blood flow, increased left atrial return and increased LVO. This was not achieved at the expense of ductal constriction. There was evidence of improved MCA peak systolic velocity parameters; this was likely due to the positive impact of improved pulmonary venous return on left ventricular preload. An increase in LVO and EF in the neonates who demonstrated a prenatal response to hyperoxia, suggests that the hyperoxygenation test can reflect functional rather than anatomical information in relation to the pulmonary arteries in utero. In our study, an appropriate response to hyperoxia in utero was also reflective of an optimal adaptation to postnatal life with rapid postnatal reduction in PVR increasing measured cardiac output. This warrants further exploration in a larger cohort to establish the ability of MH to predict the myocardial changes that occur during the transition to neonatal life.
We have demonstrated that MH is feasible and acceptable however, we have raised a concern in relation to the injudicious use of high flow oxygen in pregnancy. The haemodynamic changes observed in this study in response to MH during pregnancy could counteract any intended increase in oxygen delivery. The observed maternal effects of MH call for a re-evaluation of the role of hyperoxygenation treatment in the non-hypoxemic pregnant patient.
The Rotunda Foundation
First SupervisorProf. Fionnuala Breathnach
Second SupervisorProf. Afif El-Khuffash
Third SupervisorDr Orla Franklin
CommentsSubmitted for the Award of Doctor of Philosophy to the Royal College of Surgeons in Ireland, 2020
Published CitationMcHugh A,. Can Sonographic Assessment of Pulmonary Vascular Reactivity Following Maternal Hyperoxygenation Predict Neonatal Pulmonary Hypertension [PhD Thesis] Dublin: Royal College of Surgeons in Ireland; 2020
Degree NameDoctor of Philosophy (PhD)
Date of award30/11/2020
- Doctor of Philosophy (PhD)
- Gynaecology, Obstetrics and Perinatal Health