Trial of Maternal Oxygen for Fetal Left Heart Hypoplasia

Birth defects are the leading cause of infant mortality in the United States, and congenital
heart disease is the leading cause of birth-defect related infant mortality. Despite
advances, conditions related to left heart hypoplasia (LHH) are the most severe, and
contribute significantly to this overall mortality as well as morbidity, including delayed
brain maturation.

Conditions involving left heart hypoplasia (LHH) are among the most severe, and contribute
significantly to this overall mortality. Sequelae are not limited to mortality, as recurrent
interventions and impaired neurodevelopmental outcomes are highly prevalent among survivors
with LHH. During the last 15 years, fetal cardiac intervention (FCI) has been introduced as a
novel treatment for these children. The potential to halt disease progression in utero and to
avoid or reduce the complexity of future surgery is very alluring. Current FCI is highly
invasive, and there is limited and controversial data pertaining to efficacy, risk, and
criteria for intervention. There is also a group of fetuses with LHH who require neonatal
intervention, which may include single ventricle palliation, but are not candidates for
current FCI.

The investigators intend to use a randomized control pilot study design to evaluate a novel
non-invasive FCI to improve left heart growth in the fetus with LHH. The intervention will be
chronic administration of supplemental oxygen provided to pregnant subjects (maternal
hyperoxygenation, MH) versus placebo. Currently, two FCIs are available for left-sided
disease, both of which are invasive and carry significant risk. The interventions include
aortic valvuloplasty for aortic stenosis and atrial septoplasty/stenting for hypoplastic left
heart syndrome with a restrictive or intact atrial septum. While there is tremendous
enthusiasm for these procedures, there is limited and controversial data pertaining to
efficacy, risk, and inclusion criteria for intervention.

There is also a further group of patients with LHH without frank aortic stenosis but
hypoplasia of multiple left heart structures. While these patients almost universally require
neonatal intervention which may include single ventricle palliation, they are not candidates
for the fetal interventions currently available. This is the population of fetuses that we
intend to target. The etiologies of left heart obstruction are unknown, although the common
pathway is thought to be diminished blood flow to left heart structures in utero.
Hypothesized etiologies include abnormal or restricted PFO flow, reduced flow across the
mitral valve atrial septal aneurysm, low aortic flow due to aortic valvar abnormalities,
presence of a ventricular septal defect, or subaortic narrowing. In the fetus, maternal blood
with the highest oxygen and glucose concentration is delivered via the umbilical, and travels
via the ductus venosus to the right atrium, where the majority crosses the foramen ovale and
passes to the left heart. Most of that blood is used to perfuse the developing brain. In the
normal fetal heart, about 10% of the left heart output, rather than traveling to the cerebral
circulation, passes through the aortic isthmus and joins the blood that will perfuse the
lower body. The low volume of blood flow through the fetal lungs is due to high pulmonary
vascular resistance. The fetal pulmonary circulation is very sensitive to small changes in
pulmonary arterial oxygen saturation and tension. Pulmonary vascular responses to changes in
oxygen levels are very much related to gestational age and this sensitivity increases
dramatically with advancing gestational age in the fetus.

It has been shown in both fetal sheep and human fetuses that when supplemental oxygen is
provided to gravid mothers, there is evidence of an increase in fetal pulmonary blood flow,
and this response becomes more pronounced late in gestation. This increase in pulmonary blood
flow results in greater blood flow to the left atrium, and potentially to the left ventricle
and across the aorta.

From January 2013 to March 2014 , the investigators conducted a pilot feasibility study of
CMH in mothers of fetuses with LHH defined as a combined aortic & mitral valve annuli z-score
(AMZ) less than -4.5. The investigators recruited gravidae with moderate to severe hypoplasia
of left heart structures in which growth potential still existed; therefore gravidae with
fetal hypoplastic left heart syndrome or severe aortic stenosis were excluded. Candidates
were challenged as follows: After a full fetal echocardiogram in room air, Doppler scans were
repeated after 10 minutes of maternal hyperoxygenation at 8 LPM 100% FiO2. Responders
(fetuses with greater than 10% increase in aortic/combined cardiac output) were offered
enrollment. Nine gravidae were enrolled. Chronic MH consisted of a minimum of 8 hours daily
oxygen until delivery. Enrolled gravidae were compared to retrospective controls, who were
all cared for at our institution from January 2013 to March 2014 with fetal LHH not
undergoing CMH (9 total). Rate of growth in aortic and mitral annuli was compared among
groups using longitudinal regression.

All mothers tolerated CMH, and there were no significant maternal or fetal complications.
Mean increase in percent aortic flow after acute maternal hyperoxygenation was 35.3% (range
18.1-47.9%). Mean daily CMH hours = 10 (range 6.5-14.6). Mitral annular growth was 0.19
mm/week compared to 0.14 mm/week in CMH vs. controls (p=0.33). Aortic annular growth was 0.14
mm/week compared to 0.13 mm/week in CMH vs. controls (p=0.75). More than 9 CMH hours daily
was associated with better growth of the aortic annulus in intervention fetuses (0.16 mm/week
vs. 0.08 mm/week, p=0.014). In this pilot study, CMH appears safe and feasible, although
effect estimates of annular growth were small given the studied method of delivery and dose
of oxygen.

Given these findings, the investigators determined that CMH therapy should be further
investigated as a low risk intervention for fetal LHH. The current study design is a pilot
randomized controlled trial, with a change in the route and dose of oxygen as illustrated
below.

Purpose and Objectives The investigators aim to perform a pilot randomized control trial to
investigate the novel fetal cardiac intervention of continuous supplemental oxygen (maternal
hyperoxygenation or MH) to mothers of fetuses with left heart hypoplasia (LHH). It has been
shown that with acute MH, there is evidence of an increase in fetal pulmonary blood flow
resulting in greater blood flow to the left atrium, left ventricle and across the aorta.
There is evidence that chronic MH (CMH) in fetuses with hypoplastic left heart structures may
improve growth of the left heart, although these studies did not include control patients.
CMH also has the potential to attenuate prenatal cerebral changes thought to be secondary
abnormal cerebral blood delivery in fetuses with LHH and resultant poor maturation of the
fetal brain.

For the pilot study, the investigators intend to compare CMH to placebo therapy in 12 mothers
to determine effect estimates for a larger, multicenter randomized control trial (RCT) trial.
The team has recently completed a feasibility pilot in which 9 mothers received CMH therapy.
While there were no significant maternal or fetal complications, and results were promising,
effect estimates were small and controls were only retrospective, and not randomized,
limiting the interpretability of effect.

Aim 1: To determine if chronic MH therapy in mothers of fetuses with left heart hypoplasia
results in faster growth of the fetal mitral and aortic annuli compared to fetuses of mothers
receiving placebo. Methods & approach: Recruitment will include mothers of fetuses with LHH.
Enrollees will wear a nasal cannula with 4 LPM flow 40% FiO2 oxygen or room air for 24
hours/day. Fetuses will be followed with echocardiography until delivery, when they will have
an echocardiogram at birth, and further care as clinically indicated. Rates of change in
left-sided valve annular dimensions in the CMH group vs. placebo will be compared, using
longitudinal regression.

Aim 2: To determine if chronic MH therapy in mothers of fetuses with left heart hypoplasia
results in greater neonatal brain maturity compared to fetuses of mothers receiving placebo.
Methods & approach: Recruitment, enrollment, and the intervention are identical to that
delineated above, as these will be the same patients. For this aim, tbrain maturity in
fetuses exposed to MH versus placebo will be compared using neonatal brain magnetic resonance
imaging, as defined by a well-known maturation score.

CMH for fetal LHH has the potential to have a major impact in the field of congenital heart
disease, as it is noninvasive, inexpensive, and universally available. Other institutions in
the United States and Canada have demonstrated eagerness to join in a multicenter RCT if the
pilot suggests positive results.

Inclusion Criteria:

Mothers carrying fetuses with small left sided structures likely needing neonatal
intervention, defined as the following on fetal echocardiography:

1. Sum of aortic and mitral valve z-scores (standard deviation based on gestational age)
less than -4.5

2. Flow across the atrial septum either bidirectional or left to right

3. Transverse aortic arch or isthmus z-score less than 2.0.

Exclusion Criteria:

1. Severe fetal aortic stenosis

2. Fetal mitral or aortic atresia

3. Abnormal fetal atrioventricular or ventriculoarterial relationships, including double
inlet left ventricle, double outlet right ventricle, transposition of the great
arteries

4. Multiple gestations

5. Intrauterine growth restriction

6. Persistent fetal arrhythmia

7. Very poor ultrasound images, defined by the inability to reliably measure/evaluate all
included cardiac structures

8. Major fetal extracardiac anomalies, specifically lesions that would be expected toi
ncrease mortality for the fetus/neonate or would necessitate intervention in the
neonatal period, including but not limited to congenital diaphragmatic hernia,
omphalocele, gastroschesis, meningomyelocele, lower urinary tract obstruction, and
anencephaly

9. Aneuploidy (although this is not required to be known to enroll

10. Maternal conditions that may alter fetal hemodynamic, including moderate to severe
hypertension requiring medication in pregnancy, preeclampsia, major or unrepaired
maternal congenital heart disease, obstructive sleep apnea, severe asthma (requiring
daily treatment), restrictive lung disease, severe anemia (hemoglobin less than 8
g/dL), maternal chronic renal disease (creatinine greater than 1.2 mg/dL), known
placental abnormality (complete placenta previa, accrete, or percreta), and
antiphospholipid antibody syndrome.