Saturation Targets and Resuscitation in Preterm Trial

Study Design: A prospective randomized controlled trial of OX50 versus OX25 versus Ox75 will
be conducted at Parkland Hospital. This study will test the hypothesis that the OX25 target
oxygen strategy will result in successful resuscitation of preterm neonates with less O2
exposure and reduction in oxidative stress compared to the currently recommended Ox50 target
O2 strategy and novel OX75 target oxygen strategy.

Study Population: Neonates with obstetric gestational age (GA) of 23-30 weeks, born at
Parkland Hospital, Dallas, TX, who require active resuscitation, will be included. Active
resuscitation is defined as positive pressure ventilation (PPV), continuous positive airway
pressure (CPAP), or ventilation via an endotracheal tube.

Outcomes:

A. Primary outcome:

• Increase in oxidative balance ratio from baseline by 20%.

Sample size and power estimation: The primary outcome for this study is the change from
baseline (cord blood - first 2 hour after birth) in the oxidative balance ratio. From our
recently completed randomized control trial, mean ∆ oxidative balance ratio for Ox50 in
preterm neonates is 9.28 with standard deviation (SD) of 2.42.41 Based on these preliminary
data, in order to detect a 20% change in the ∆ oxidative balance ratio, 25 infants are needed
for each arm using two sided alpha of 0.05 and power of 80% and allowing 10% attrition.

B. Secondary Outcome:

- Reduction in other oxidative stress markers:

- Reduction in oxygen load in the delivery room

- Time spent in goal saturations

- Time spent with saturations above 94%

- Incidence and frequency of bradycardia during resuscitation

- Incidence of intubation in the delivery room

- Amount of ventilator support needed in the delivery room

- Incidence of neonatal morbidities such as retinopathy of prematurity (ROP),
bronchopulmonary dysplasia (BPD), necrotizing enterocolitis (NEC), intraventricular
hemorrhage (IVH), periventricular leukomalacia (PVL), neurodevelopmental impairment

- Brain abnormality score on MRI

Study procedure:

A prospective randomized controlled trial of optimal oxygen strategy will be conducted at
Parkland Hospital.

Screening procedure: A research nurse will screen and identify eligible mothers admitted to
labor and delivery suites at Parkland hospital. Investigators will obtain antenatal informed
consent from mothers admitted to the OB service. A total of 75 neonates will be enrolled in
the trial, 25 in each group.

Table: Target Oxygen Levels in Blood

Time after Birth OX25 OX50 OX75

1. min 40%-50% 60%-65% 75%-80%

2. min 50%-60% 65%-70% 80%-85%

3. min 60%-70% 70%-75% 85%-90%

4. min 70%-80% 75%-80% 90%-95%

5. min 80%-90% 80%-85% 90%-95%

6. min 85%-95% 85%-95% 90%-95%

10 min 85%-95% 85%-95% 90%-95%

Study intervention and control group: The study intervention #1 is the Ox25 where
resuscitation will be initiated at 30% O2 and target goal saturations will be the 25th
percentile saturations observed in healthy term newborns.52 The study intervention #2 is the
Ox75 where resuscitation will be initiated at 30% O2 and target goal saturations will be the
75th percentile saturations observed in healthy term newborns. The control group will be
resuscitated with the OX50 in which resuscitation will be initiated with 30% O2 and target
goal saturations will be the approximated median SpO2 observed in healthy term newborns as
per current NRP guidelines. Table 1 compare the transitional goal saturations that will be
used in OX25, OX50 and Ox75. For the first 6 minutes after birth, target ranges of different
strategies are different. After 6 min, the target ranges are similar. Target saturations are
an approximation of the percentile curves created by Dawson et al based on transitional
saturation data of 306 healthy term neonates who did not receive any intervention other than
warmth and stimulation.52

Randomization and Allocation Procedure: Assignment to the OX25, OX50 and OX75 group will use
a 1:1:1 allocation ratio. The Biostatistician will use a blocked randomization schema,
programmed using SAS Proc Plan (SAS version 9.2, SAS Institute, Cary, NC, USA). Allocation
will be concealed by serially numbered, sealed opaque envelopes that will be opened
sequentially by the resuscitation team before delivery.

Resuscitation management: The high-risk resuscitation team attends delivery of all neonates <
35 wks GA. Radical pulse oximeters (Masimo, Irvine, CA) will be set to maximal sensitivity
with 2 seconds averaging. Randomization will be done before delivery by opening a sealed,
opaque envelope. The probe will be placed on the pre-ductal hand within the first 30 seconds
of life using previously described strategies to optimize time to reliable signal as per NRP
2011 guidelines. A neonatal transducer (Invos Cerebral Oximeter Monitor , Covidien,
Mansfield, Massachusetts) will be attached to left fronto-pareital forehead to measure
continuous near infrared spectroscopy (NIRS) measurements. Apart from the randomly assigned
O2 strategy, resuscitation will follow 2011 NRP guidelines. Treatment failure will be defined
as HR< 60 bpm after 90 seconds of resuscitation. In this event, the O2 concentration will be
increased to 100% as recommended by current NRP guidelines. Once HR is stabilized, FiO2 will
be reduced to meet the goal saturations of the assigned strategy. If pulse oximetry does not
register stable values, resuscitation will be continued at the current FiO2 as long as HR >
100 bpm is maintained. If HR remains below 100 bpm, after making ventilation correcting
steps, FiO2 can be increased by 10-20% based on clinician's judgement.

Subsequent NICU management: Following NICU admission, all care decisions including ventilator
management will be at the discretion of the attending neonatologist. Target SpO2 will be
88-94% throughout the NICU stay as per unit policy. Infants with O2 requirement at 36 weeks
corrected GA will undergo an O2 challenge test to define physiologic BPD as per the NICHD
definition.

Collection of baseline variables: Baseline demographic characteristics, maternal and delivery
characteristics including maternal age, prenatal clinic attendance, antenatal steroids given,
maternal diabetes, illicit drug use, preeclampsia, prolonged rupture of membrane,
chorioamnionitis, abruption, previa and cesarean section will be collected. Baseline infant
characteristics such as gender, obstetrical GA, birth weight, intrauterine growth
restriction, breech presentation, multiple pregnancies, cord blood gas pH, base deficit will
be recorded. This data will be collected from the electronic medical records by research
nurse and Parkland neonatal intensive care unit database which collects this information on
all the infants admitted to Parkland NICU.

Outcomes and Statistical plan by each aim:

A. Specific Aim I: To determine if OX25 will decrease DR O2 load and oxidative stress
compared to OX50 and OX75. Investigators hypothesize that OX25 neonates will require less O2
in the DR, spend more time in goal SpO2 range and will have lower oxidative stress

DR O2 exposure and SpO2: SpO2 will be downloaded from pulse oximeters used during
resuscitation as per Masimo pulse oximeter guidelines using TRENDCOM software (Masimo,
Irvine, CA). Investigators have described and successfully performed acquisition of such data
in our previous study. Resuscitation interventions including changes in FiO2 will be recorded
at 30 seconds intervals by the OB circulating nurse whose sole responsibility is to record
resuscitation details in a pre-specified record sheet as per routine practice at Parkland
Hospital.

Oxidative stress and oxidative injury measurement:

Preparation and storage by research nurse: Sample will be centrifuged to separate plasma and
stored at -80 C.

A. Oxidative balance ratio: All samples will be analyzed for total hydroperoxides (TH) and
biological antioxidant capacity (BAP) using the Free Radical Analytical System (FRAS4
analyzer, H&D SRL, Parma, Italy) as described in our previous study. TH represents the total
of radical O2 metabolites produced by peroxidation of protein, lipids and amino acids,
measures oxidative damage and serves as a biomarker of overall free radical attack. BAP
measures both endogenous and exogenous antioxidative capacity of serum to reduce oxides by
inactivating, and eliminating free radicals and reactive O2 species. BAP by exploiting the
chemical principle of the well-known ferric reducing ability of plasma provides a reliable
measure of biological antioxidant potential of plasma. To estimate the global balance between
oxidative stress and antioxidant potential, the oxidative balance ratio (BAP/TH) will be
calculated.

B. Other oxidative stress markers: 8-hydroxydeoxyguanosine (8-OHdG), F2-Isoprostanes and
Isofurans and Protein Carbonyls. Along with these, investigators will also measure cytokines,
Interleukin 1 receptor antagonitst, hyaluronidase.

B. Specific Aim II: Determine if preterm neonates can be resuscitated successfully with less
ventilatory support using Ox25. Investigators hypothesize that OX25 infants will achieve
quicker heart rate stabilization and will require less ventilatory support including
intubation in the DR. Resuscitation interventions including changes in ventilator support,
need for CPR will be recorded at 30 seconds intervals by OB circulating nurse whose sole
responsibility is to record resuscitation details in a pre-specified record sheet as per the
our routine practice. The research nurse will download continuous heart rate data from Masimo
pulse oximeters and continuous NIRS data from cerebral oximeter monitor for the duration of
resuscitation. Admission parameters will also be recorded by the research nurse.

C. Specific Aim III: Determine if OX25 strategy will result in lower incidence of adverse
clinical outcomes compared to OX50 or OX75 strategy. Investigators hypothesize that OX25
neonates will have lower incidence of BPD. Investigators will also record other clinical
outcomes such as ROP, NEC, IVH, length of hospitalization, neurodevelopmental impairment and
mortality.

Neonatal morbidity during NICU stay: Short term clinical outcomes including duration and type
of respiratory support and O2 supplementation during the NICU stay, respiratory morbidities
such as respiratory distress syndrome, need for surfactant, pulmonary hypertension, and BPD
will be obtained from the NICU database. In addition, other short term morbidities such as
sepsis, Grade III/IV IVH, NEC, symptomatic patent ductus arteriosus, severe ROP, length of
hospitalization and mortality during NICU stay will also be collected from the NICU database.

Brain MRI at term equivalent age: All study infants who underwent a brain MRI at term
equivalent age as part of standard of care at parkland hospital will be assessed for brain
abnormality score

Neurodevelopmental impairment: All study infants will follow up at 18-24 months of age in the
high risk Low Birth Weight Clinic. A blinded developmental pediatrician will perform
standardized neurological examination to identify cerebral palsy. Developmental specialist
will administer the Bayley Scale of Infant Development III Instrument. Moderate to severe
neurodevelopmental impairment will be defined as Bayley III Motor Developmental Index or
Psychomotor Developmental Index score less than 70, moderate or severe cerebral palsy,
bilateral blindness, bilateral hearing loss requiring hearing aids or cochlear implants.

Inclusion Criteria:

1. Neonates of obstetric gestational age 23-30 weeks

2. Inborn at Parkland Hospital, Dallas, Tx

3. Resuscitation team present to initiate resuscitation

Exclusion Criteria:

1. Prenatally diagnosed cyanotic congenital heart disease

2. Prenatally diagnosed congenital diaphragmatic hernia

3. Prenatally diagnosed significant airway anomaly

4. Prenatally suspected hypoplasia of lungs

5. Preductal saturation cannot be measured

6. Non-viable newborns

7. Prenatally diagnosed condition due to which decision is made to limit resuscitation in
the delivery room by parents and neonatologist.