Symposium Papers

Oxygen Therapy in the Neonatal Care Environment

Brian K Walsh RRT-NPS, Toni M Brooks RRT, and Barry M Grenier RRT-NPS

Introduction
Physiologic Effects of Oxygen Therapy: Benefits and Adverse Effects
Treatment of Hypoxia
Oxidative Stress
Retinopathy of Prematurity
Chronic Lung Disease
Long-Term Outcomes
Oxygen During Resuscitation
Oxygen Delivery Devices
Blow-By Oxygen
Oxygen Hood
Low-Flow Nasal Cannula
High-Flow Nasal Cannula
Device-Related Complications
Advances in Oxygen Therapy
Closed-Loop FIO2 Regulation
New-Generation Pulse Oximetry
Discussion
Unresolved Questions
Future of Neonatal Oxygen Therapy

The use of oxygen in the treatment of neonates with respiratory distress has been reported for more
than a century. Oxygen therapy is generally titrated to one or more measures of blood oxygenation
and administered to reverse or prevent hypoxia. Individual responses to oxygen therapy vary
greatly, depending on the particular cause of hypoxia and the degree of impairment. Despite this
focused purpose, oxygen administration in this patient population has become complex. The longer
we deliver this drug, the more we discover its beneficial and detrimental effects. New and innovative
ways to deliver and monitor this therapy have improved outcomes. Despite this vast experience
there still remain some unanswered questions regarding the use of oxygen in the neonatal envi-
ronment. Nonetheless, oxygen is a major staple in our treatment arsenal for neonates. Key words:
oxygen; neonatal; infant, newborn; retinopathy of prematurity; oxygen inhalation therapy. [Respir Care
2009;54(9):1193–1202. © 2009 Daedalus Enterprises]

Introduction
Brian K Walsh RRT-NPS, Toni M Brooks RRT, and Barry M Grenier
RRT-NPS are affiliated with the Respiratory Care Department, Chil- The use of oxygen in the treatment of neonates with
dren’s Hospital Boston, Boston, Massachusetts. respiratory distress has been reported for more than a cen-
The authors have disclosed no conflicts of interest. tury. In 1907, Budin recommended oxygen “supplied
through a funnel, the large opening of which is placed
Mr Walsh presented a version of this manuscript at the New Horizons beside the infant’s face,” for the treatment of cyanotic
Symposium, “Neonatal Respiratory Care,” at the International Respira-
tory Congress of the American Association for Respiratory Care, at the
episodes in newborns.1 In the 1930s, Hess2,3 developed an
54th International Respiratory Congress of the American Association for incubator (Fig. 1) capable of delivering approximately 40%
Respiratory Care, held December 13-16, 2008, in Anaheim, California.- oxygen for extended periods of time. By the 1940s, a

RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9 1193

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

Fig. 1. Hess bed equipped with an oxygen therapy unit (A-side view). 1: Pressure gauge. 2: Oxygen flow regulator. 3: Flow meter. 4: Glass
and metal hinged door for feeding purposes. 5: Thermometer window. 6: Metal hinged door for purposes of body care of the infant.
7: Ventilator with small and large exit openings. 8 –12:Controls for maintaining temperature in water-jacket of the incubator. (From Refer-
ence 3, with permission.)

commercially available incubator capable of providing a hyperoxia on the other— has not yet been clearly defined,5-7
high concentration of oxygen facilitated the liberal use of leading to wide variations in practice.8 Even the term “ad-
oxygen for the treatment of cyanosis, apnea, and periodic equate oxygenation” is not clear.9 Other complicating fac-
breathing in newborns.1,4 Throughout this time, oxygen tors in achieving the goals of neonatal oxygen therapy
administration was guided by the clinical observations of include patient size, tolerance of delivery devices, and
skin color, as well as the rate, regularity, and work of variability in the use of delivery devices, which suggest
breathing. It wasn’t until the 1960s and 1970s that tech- that clinicians often lack adequate knowledge in the use of
nology—micro-sampling of blood gases, transcutaneous oxygen delivery equipment,10 and the lack of training in
oxygen monitoring, and, later, pulse oximetry— became the concepts of neonatal oxygenation and equipment used
available for more precise monitoring of physiologic ef- to monitor the effects of oxygen therapy.11
fect.
The overall goal of oxygen therapy is to achieve ade- Physiologic Effects of Oxygen Therapy:
quate oxygenation using the lowest concentration of in- Benefits and Adverse Effects
spired oxygen. However, achieving this goal is compli-
cated by a number of factors. Despite over 75 years of Despite its universal acceptance as a life-saving therapy
routine oxygen administration to newborn infants, the op- for newborns, oxygen administration is associated with
timal level of oxygenation— one that avoids the detrimen- numerous physiologic effects, particularly when used to
tal effects of hypoxia on the one hand, and those caused by treat premature infants.

Treatment of Hypoxia
Correspondence: Brian K Walsh RRT-NPS, Respiratory Care Depart-
ment, Children’s Hospital Boston, 300 Longwood Avenue, MA-861, While oxygen therapy is generally titrated to some mea-
Boston MA 02115. E-mail: [Link]@[Link]. sure of arterial oxygenation in response to an abnormally

1194 RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

low level of blood oxygen, or hypoxemia, oxygen is ad-

ministered to the neonate to reverse or prevent hypoxia.
Hypoxia is defined as a deficit of oxygen at the cellular
level, and is commonly caused by one or more of the
following: the reduced availability of oxygen at the alve-
olar level, due to pulmonary disease (hypoventilation, un-
even matching of ventilation to perfusion, diffusion de-
fects); intrapulmonary shunts or “right to left” cardiac
shunts; reduced oxygen carrying capacity due to anemia or
abnormal blood hemoglobin; or impaired oxygen delivery
due to shock, heart failure, or localized decreases in per-
fusion.12,13 Left untreated, hypoxia can lead to serious and
permanent brain injury and death.12
Individual responses to oxygen therapy vary greatly,
depending on the particular cause of hypoxia and the de-
gree of impairment. Hypoxia caused by hypoventilation
and ventilation-perfusion anomalies associated with pul-
monary disease will be most responsive to oxygen therapy.
Even large increases in FIO2 will produce only small in-
creases in available oxygen if hypoxia is caused by cardiac
shunts, shock, and hemoglobin deficiency/dysfunction.12,13
Fig. 2. The proposed role of vascular endothelial growth factor
It should be stressed, however, that even small increases in (VEGF). A: It is hypothesized that normal retinal vessel develop-
oxygen availability may prevent life-threatening decom- ment is stimulated by production of VEGF (red) anterior to the
pensation in the hypoxic neonate. developing vasculature. In addition, maintenance of some retinal
vessels is dependent on VEGF. B: In the first phase of retinopathy
Oxidative Stress of prematurity, exposure to relative hyperoxia after birth interrupts
the gradient of physiologic hypoxia in the immature retina, leading
to downregulation of VEGF production, with associated vaso-
The role of oxygen and oxidative stress in the develop- obliteration and cessation of vessel growth. C: As the metabolic
ment of a number of neonatal diseases has generated much demand of the developing retina increases, the nonperfused por-
interest. Oxidative stress has been defined as an imbalance tions of the retina become hypoxic and overproduce VEGF. D: Neo-
between pro-oxidant and anti-oxidant forces in the body.14 vascularization occurs in response to overproduction of VEGF,
producing retinopathy of prematurity. If VEGF production persists,
Pro-oxidants include oxygen radicals or reactive oxygen then the retinopathy of prematurity will progress. (From Refer-
species, which can be cytotoxic because of their ability to ence 26, with permission.)
alter cellular components and function. Reactive oxygen
species are generated as a result of normal mitochondrial
respiration, but also during the reperfusion phase of hy- pathogenesis involves oxidative stress and injury, which
poxic tissue injury and in association with infection and include retinopathy of prematurity, bronchopulmonary dys-
inflammation.15,16 Oxygen is “toxic” because of the pro- plasia, necrotizing enterocholitis, and intraventricular hem-
duction of reactive oxygen species; thus oxygen adminis- orrhage.20
tration increases oxidative stress.
Antioxidant defenses include the enzymes superoxide Retinopathy of Prematurity
dismutase, catalase, and glutathione. Nonenzymatic anti-
oxidants start to cross the placenta in late gestation, and Though long recognized as a complication of oxygen
include vitamins A, C, E, and ubiquinone. Premature in- therapy, retinopathy of prematurity remains a major cause
fants are at particular risk from oxidative stress because of morbidity for premature infants.21 Retinopathy of pre-
both endogenous and passively acquired exogenous anti- maturity is a disease limited almost exclusively to prema-
oxidant defense systems do not accelerate in maturation ture infants and is characterized by abnormal vasculariza-
until late in the third trimester.15,17,18 Investigators have tion of the retina, causing a range of vision impairment,
attempted to reverse or prevent the damage associated with including blindness. Much has been described in the lit-
reactive oxygen species not only by appropriate oxygen erature regarding the role of supplemental oxygen in the
administration but also by administering antioxidants; how- development of retinopathy of prematurity.22-24 The al-
ever, this therapy has not shown to be effective.19 Saugstad tered regulation of vascular endothelial growth factor has
has suggested the term oxygen radical disease of neona- been suggested25,26 as one of the factors in the pathogen-
tology to encompass a variety of newborn diseases whose esis of retinopathy of prematurity (Fig. 2). In premature

RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9 1195

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

infants the retina is incompletely vascularized. In utero the ers of inflammation isolated from the tracheal lavage sam-
arterial oxygen pressure of the fetus is 22–24 mm Hg. ples obtained from newborns in the first 3 days of life have
After premature birth, relative hyperoxia may downregu- been shown to correlate with oxygen exposure and the
late vascular endothelial growth factor production. Admin- development of bronchopulmonary dysplasia.44 As with
istration of supplemental oxygen may lead to sustained retinopathy of prematurity, supplemental oxygen admin-
hyperoxia, setting the stage for vaso-obliteration of exist- istration is only one factor in the pathogenesis of broncho-
ing vessels and arrest of the vascularization. Over time, as pulmonary dysplasia, and the threshold for pulmonary tox-
the metabolic demands of the developing eye increase, the icity remains unclear. Observational studies have associated
immature non-perfused area of the retina becomes hypoxic the avoidance of hyperoxia with shorter duration of me-
and may overproduce vascular endothelial growth factor chanical ventilation and oxygen dependence when lower
pathologically. High levels of vascular endothelial growth oxygen saturation target ranges were maintained in pre-
factor stimulate neovascularization of the retina, which mature infants from the time of birth.8,36 Currently there
in severe cases may result in retinal fibrosis and retinal are no controlled randomized trials that have evaluated
detachment. whether a low versus high oxygen saturation strategy in
Observation studies in the 1950s demonstrated the clear the immediate postnatal period decreases the incidence of
link between the liberal use of oxygen and the develop- bronchopulmonary dysplasia.
ment of retinopathy of prematurity, or retrolental fibropla- The use of supplemental oxygen to maintain a targeted
sia, as it was then known.27-30 One study by Kinsey et al, oxygen saturation measured via pulse oximetry (SpO2)
in 1956, that was not observation, demonstrated a 17% ⬎ 93% for infants with established chronic lung disease
reduction in retinopathy of prematurity as well as a 9% has been advocated to decrease pulmonary vascular resis-
reduction in blindness when curtailing oxygen therapy to tance and airways resistance, to decrease the risk of sud-
room air within the first 48 hours.30a Of note, these studies den infant death, and to promote growth.45,46 However, 2
provoked a drastic decrease in neonatal oxygen use that large randomized studies that compared different strate-
was associated not only with a reduction in retinopathy of gies for titrating supplemental oxygen for premature in-
prematurity, but also with an increase in newborn deaths fants outside of the immediate newborn period failed to
and cerebral palsy.31,32 With the improved survival of very- confirm these clinical benefits.47,48 The Supplemental Ther-
low-birth-weight infants during the past decade, retinopa- apeutic Oxygen for Prethreshold Retinopathy of Prematu-
thy of prematurity continues to be a source of substantial rity study, which compared a low-oxygen-saturation group
morbidity. Wide intercenter variability exists in the inci- targeting oxygen saturation of 89 –94% versus a high-ox-
dence of severe (⬎ stage 3) retinopathy of prematurity in ygen-saturation group targeting 96 –99%, demonstrated no
different centers.33,34 These differences could be attributed difference in mortality or development as secondary out-
to the combination of many known and unknown factors; comes in the low-saturation group versus the higher-satu-
one explanation might be that differences in clinical prac- ration group. Further, the study showed that the infants in
tices affect the rates of retinopathy of prematurity. More the high-saturation arm had an increased incidence of pneu-
recent studies have demonstrated an association between monia and chronic lung disease exacerbations, and that
retinopathy of prematurity and high oxygen saturation,8,35,36 significantly more infants in the high-saturation group re-
and several studies suggest that fluctuations in oxygen- mained in hospital, on oxygen, and on diuretics at 50 weeks
ation level may also have a role in its development.35,37 It post-menstrual age.47 The Benefits of Oxygen Saturation
thus is possible that repeated cycles of hyperoxia and hyp- Targeting study demonstrated that extremely pre-term ox-
oxia favor the progression of retinopathy of prematuri- ygen-dependent infants treated to maintain oxygen satura-
ty.38,39 While hyperoxia clearly plays a role, other risk tion between 91–94% had growth and development that
factors include growth retardation, male sex, septicemia, was not significantly different from a group maintained
and, most significantly, low gestational age and birth with a target saturation range of 95–98%. In addition, the
weight.21 In addition, worsening retinopathy of prematu- higher-saturation infants had a significantly higher rate of
rity has been linked to the overall severity of illness of the oxygen usage at 36 weeks post-menstrual age and home
newborn and the extent of other complications.40,41 oxygen usage.48

Chronic Lung Disease Long-Term Outcomes

Oxygen administration was identified as a risk factor in In a long-term outcomes study of term infants who had
the development of neonatal chronic lung disease in early required extracorporeal membrane oxygenation for meco-
descriptions of bronchopulmonary dysplasia.42,43 In ani- nium aspiration syndrome/persistent pulmonary hyperten-
mal studies, oxygen exposure has been shown to inhibit sion of the newborn, Boykin et al evaluated lung function
lung growth and deoxyribonucleic acid synthesis.14 Mark- at 10 –15 years. That study found significant abnormalities

1196 RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

in lung function and that the most significant predictor of Oxygen Hood
long-term pulmonary outcomes was the duration of oxy-
gen use post-extracorporeal-membrane-oxygenation de- An oxygen hood (cube) is a plastic enclosure that sur-
cannulation.49 rounds the head of the neonate, to which a continuous flow
of humidified oxygen is supplied by means of an air-
Oxygen During Resuscitation entrainment device or an air-oxygen blender. Fixed oxy-
gen concentrations from 0.21 to 1.0 can be maintained
The use of 100% oxygen during neonatal resuscitation with a minimum of 7 L/min oxygen flow into the hood.
has also been challenged, on the premise that large and This minimum gas flow also ensures that exhaled carbon
abrupt increases in blood oxygen level after birth can in- dioxide is flushed out and not rebreathed. Although an
crease oxidative stress.20 Several studies have compared oxygen hood can theoretically deliver 1.0 FIO2, this device
the use of 21% to 100% oxygen during resuscitation. Three is best suited for patients who require less than 0.5 FIO2.
recent meta-analyses of these data concluded that the use Patients requiring higher FIO2 can be managed in a hood,
of room air during the resuscitation of depressed newborns but it becomes increasingly difficult to maintain higher
resulted in a significantly reduced risk of neonatal mortal- oxygen concentrations with the large neck opening and a
ity.50-52 The studies found no significant difference in the less than optimal seal around the edges.58-60 An oxygen
incidence of severe hypoxic encephalopathy between the hood is an ideal method of oxygen delivery for neonates
21% oxygen and 100% oxygen groups. Limitations to some who require higher fractional inspired oxygen concentra-
of the studies in these analyses include a lack of blinding tions but cannot tolerate more cumbersome oxygen deliv-
in some studies, and the exclusion of stillbirths.9 In one ery devices.
small recent study, the resuscitation of premature new-
borns with 50% versus 100% oxygen did not reduce the Low-Flow Nasal Cannula
incidence of bronchopulmonary dysplasia or improve other
short-term outcomes.53 The results of a recent study by Low-flow nasal cannula remains one of the most com-
Escrig et al54 indicate that extremely premature newborns mon and widely used neonatal oxygen delivery devices.
can be safely resuscitated with a low initial oxygen con- This low-flow device delivers a fractional concentration of
centration. Related to the use of oxygen in the delivery oxygen to the patient through 2 soft prongs that rest in the
room for resuscitation, limited evidence suggests that the patient’s anterior nares. The distal end of the cannula tub-
exposure of newborns to oxygen for 3 min or longer im- ing is then attached to either a 100% oxygen source flow
mediately after birth increases the risk of childhood can- meter or to an air-oxygen blender. Finer et al found that
cer.55,56 oxygen concentrations delivered to the neonate via nasal
cannula varied from 22% to 95% with a maximum flow
Oxygen Delivery Devices
rate of 2 L/min.61 The precise FIO2 actually delivered to the
Blow-By Oxygen patient is contingent upon a number of factors, but most
specifically on the set flow through the nasal cannula and
Blow-by oxygen delivery is the simplest and least cum- its relation to the patient’s inspiratory flow demand. An
bersome form of available devices to provide oxygen ther- inspiratory flow demand greater than that supplied by the
apy to the neonate, but it is also the least reliable in de- nasal cannula causes the exact FIO2 delivered to the patient
livering a specific FIO2. Blow-by oxygen can be achieved to be a blend of the nasally inhaled oxygen with entrained
numerous ways, but is most commonly done by means of room air through the nares and mouth.10,58,61 While actual
large-bore or oxygen tubing connected to a face tent or oxygen concentrations delivered to the patient are vari-
simple mask that is placed a relatively short distance from, able, a nasal cannula remains a fairly trusted and effective
and directed toward, the patient’s face. This type of oxy- method of offering oxygen therapy to the neonate.
gen delivery is ideal for patients who cannot tolerate more
cumbersome oxygen delivery devices and/or require a lesser High-Flow Nasal Cannula
amount of oxygen. There is limited evidence that suggests
that blow-by therapy can deliver low concentrations of Nasal cannula oxygen therapy is a staple and continues
oxygen (0.3– 0.4 at 10 L/min of flow) to an area large to be redefined to improve patient comfort, compliance,
enough to provide some level of oxygen therapy to the and outcomes. The concept of high flow and high humid-
neonate, assuming adequate positioning of the device.57 ity via a nasal cannula, however, is a fairly new concept
Therefore, this type of therapy should be reserved for in- and was first introduced by Vapotherm to the respiratory
fants who do not a require high inspired oxygen concen- care community in the spring of 2002, after receiving Food
tration but may require short-term or intermittent oxygen and Drug Administration 510K clearance in the fall of
therapy. 2001. Prior to high-flow nasal cannula, most clinicians

RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9 1197

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

amount of generated positive pressure varied not only with

flow rate but with cannula size; the larger cannula size
produced a mean pressure of 9.8 cm H2O at a flow rate of
2 L/min.64 Sreenan et al concluded that positive distending
pressure could be predictably applied using high-flow na-
sal cannula at flow rates of 1–2.5 L/min and that high-flow
nasal cannula is as effective as nasal continuous positive
airway pressure (CPAP) when using frequency and dura-
tion of apnea, bradycardia, and oxygen desaturations as
outcomes.63 Using the Sreenan et al equation, Campell
et al66 compared the use of high-flow nasal cannula and
nasal CPAP for preventing re-intubation in a group of
premature infants. In this study the high-flow nasal can-
nula group had a significantly higher number of re-intu-
bations, increased oxygen use, and more apnea and bra-
dycardia episodes post-extubation. In their discussion,
Campbell et al concluded that the equation for determining
CPAP was not reproducible in this patient population.

Device-Related Complications
Fig. 3. Proposed reduction in nasopharyngeal dead space that
leads to improving alveolar ventilation with high-flow nasal cannula. Device-related complications of nasally applied oxygen
therapy include skin irritation from device materials,67 na-
considered it uncomfortable to use a flow of greater than sal mucosal irritation, bleeding, and obstruction (particu-
6 L/min via nasal cannula in adults; this was primarily due larly with gas flows rates ⬎ 2 L/min),68,69 inadvertent
to the lack of adequate humidification available via nasal CPAP,63,64 intrinsic contamination (Vapotherm specific),70
cannula delivery. Little consensus existed in the neonatal subcutaneous scalp emphysema, pneumo-orbitis,71 pneu-
patient population on the parameters defining high-flow mocephalus (high-flow and low-flow cases),72-74 and dis-
nasal cannula, but for our discussion high-flow nasal can- placement leading to disruption of oxygen delivery. Of
nula is classified as a fixed-performance oxygen delivery note, many of the mucosal irritation, bleeding, and ob-
system that is capable of delivering a specific oxygen con- struction events were experienced with devices that pro-
centration at flows that meet or exceed the inspiratory flow duced a low temperature and relative humidity. Potential
demand of the patient.58 This type of oxygen delivery complications of oxygen hoods include cold stress from
device is composed of traditional nasal cannula style prongs unheated or inadequately heated gas,75 bacterial contami-
that rest in the patient’s anterior nares and allow heated, nation,76,77 and high noise level, associated with hearing
humidified oxygen to be delivered at flow rates of 1– 8 L/ impairment.78 More generally, the lack of adequately heated
min, while an air-oxygen blender allows FIO2 to be directly and humidified gas can lead to airways hyper-reactivity
manipulated.62 As higher flow rates are reached, set oxy- and mucociliary dysfunction79-82
gen flows exceed demand, thus preventing the entrainment
of room air, flushing dead space (Fig. 3), and affecting the Advances in Oxygen Therapy
delivery of higher, more precise fractional inspired oxygen
concentrations. High-flow nasal cannula use has been Closed-Loop FIO2 Regulation
adopted in many neonatal intensive care units for its ease
of use and patient tolerance. It is also used because it is Given the rapid onset and frequency of the episodes of
able to provide higher oxygen concentrations and inspira- hypoxemia and hyperoxemia that may routinely occur,
tory flows, thus providing a higher level of oxygenation maintaining SpO2 within a desired range by manual FIO2
support to the neonate than can be achieved by any of the adjustment on any device (nasal CPAP, nasal cannula, or
other devices described above. mechanical ventilator) during each episode is a difficult
Improvements in oxygenation associated with high-flow and very time-consuming task. Health-care providers re-
nasal cannula may also be related to the creation of pos- spond to high/low SpO2 alarms, but because of their re-
itive end-expiratory pressure. High-flow nasal cannula has sponsibilities under routine clinical conditions, their re-
been shown to significantly increase esophageal pres- sponse time is not always consistent and optimal. In fact,
sure63,64 and pharyngeal pressure65 in neonates. Locke et al historically, many choose to run SpO2 on the upper end of
demonstrated that in a group of premature infants, the normal because of their unfounded fear of brain injury

1198 RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

Fig. 4. Claure’s and Bancalari’s configuration of biofeedback to a computer algorithm, which then adjusted the fraction of inspired oxygen
(FIO2), as compared to conventional manual adjustments by a dedicated nurse. SpO2 ⫽ oxygen saturation measured via pulse oximetry.
(Adapted from Reference 86).

over the fear of retinopathy of prematurity. This exposes events. The introduction of new-generation oximeters, such
these infants to periods of hypoxemia and hyperoxemia as the Massimo SET, Nellcor N-395, Novametrics MARS,
that may increase the risk of retinopathy of prematuri- and Pillips Viridia 24C in the early 1990s has prompted a
ty38,39,83,84 and neonatal chronic lung disease, as mentioned number of studies comparing the performances of the new-
above.85 These limitations make the use of a system for generation pulse oximeters to conventional oximeters, as
automatic FIO2 adjustment a desirable alternative. Banca- well as comparing performances between the different
lari and Claure found that an algorithm for closed-loop brands of pulse oximeter.87
FIO2 with a mechanical control (Fig. 4) to maintain SpO2 Recent studies in the neonatal population have proven
within a target range was at least as effective as a fully that new-generation oximeters out-perform their older
dedicated nurse in maintaining SpO2 within the target range, counterparts in their ability to more accurately detect true
and it may be more effective than a nurse working under hypoxic and bradycardic events as well as reduce the num-
routine conditions.86 While this may be helpful in keeping ber of false alarms in the setting of increased patient mo-
SpO2 within range, a system not carefully alarmed could tion and low perfusion states. A study by Hay et al not
expose infants to a higher than acceptable concentration of
only concluded that new-generation oximeters were supe-
oxygen in the face of atelectasis or hypoventilation. Un-
rior in comparison with their older counterparts studied,88
fortunately, at this time there is not an available Food and
they found that even among the newer-generation systems
Drug Administration approved device to provide this type
there were differences. Additional studies have found sim-
of closed-loop control.
ilar data supporting the conclusions of Hay et al, starting
New-Generation Pulse Oximetry a technology “war” over whose “signal processing algo-
rithms” and sensors are the most promising of new-gen-
Continuous monitoring of oxygenation status by means eration pulse oximeters available.87,89-91
of pulse oximetry in the neonatal intensive care unit is The substantial advances that have been made in pulse
commonplace. Often, however, the reliability of the mea- oximetry technology allow more accurate detection of hy-
surement is questioned when interference, such as patient poxemic and bradycardic events in the neonatal popula-
motion, is detected. The number of pulse oximetry devices tion. The hope is that with these new technologies the
available to hospitals and clinicians is vast, with major number of clinician disregards for false alarms due to mo-
medical companies constantly changing and improving tion artifact will be reduced, therefore resulting in im-
technology. A new generation of motion-tolerant pulse proved clinical performance and more judicious oxygen
oximeters has been designed to resist interference and im- therapy as a result of a hypoxic or hyperoxic state. Most
prove clinical performance, most specifically by increas- recently in pulse oximetry, Masimo has developed the
ing accuracy, decreasing false alarms, and capturing true capability of measuring amounts and types of hemoglobin,

RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9 1199

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

which may lead to a more precise monitoring and control manufacturer agreements, a high-flow nasal cannula sys-
of oxygen delivery noninvasively.92 tem can cost a department approximately $18 – 80. In ad-
dition, if a neonatal unit were to switch half of its CPAP
Discussion patients over to high-flow nasal cannula, it could lose
relative-value-unit justification for its respiratory therapy
Unresolved Questions staffing model when the infants are at the same if not
higher illness-severity level than some of their typical low-
Many unresolved questions remain when discussing neo- flow nasal cannula patients.
natal oxygen therapy, but one specific question that arises
is which SpO2 ranges are most appropriate for the newborn. Future of Neonatal Oxygen Therapy
This question is complex in that the most appropriate range
may be different in different contexts. Many have con- Oxygen is a drug that is essential in the treatment and
ducted research with different SpO2 ranges, showing equiv- prevention of neonatal hypoxia. However, the excessive
ocal if not better outcomes to higher SpO2 ranges; however, use of oxygen can lead to serious and long-lasting adverse
there have not been consistent ranges among the studies. sequelae. Appropriate administration of oxygen will de-
In a recent review of resuscitation and ongoing manage- pend on controlled trials defining optimal ranges of oxy-
ment of pre-term infants, Finer discusses his recommen- genation for the newborn targets that may change with
dation that an SpO2 range of 85–93%, with alarms set ⫾1–2% different pathologies and at different stages of develop-
above and below that range, was most appropriate.17 How- ment. The types of oxygen delivery devices seem not as
ever, this does not answer the question for near-term, term important as monitoring the effects of this therapy. Im-
infants, or patients who have developed chronic lung dis- provements in oxygen therapy monitoring technology help
ease and are susceptible to pulmonary hypertension. Fur- to improve a clinician’s ability to most appropriately apply
ther studies are needed to fully answer this question. and deliver oxygen. If closed-loop FIO2 management be-
Two other unresolved questions are whether or not a comes available, it will be helpful; however, it needs to
high-flow nasal cannula can substitute for nasal CPAP and come with carefully thought-out limits. Additional im-
whether or not high-flow nasal cannula can replace the provements in humidification control and ease of use al-
high FIO2 oxygen hood. It is fairly clear that high-flow low us to recommend optimal humidification with all ox-
nasal cannula is a safe oxygen delivery device, as there ygen therapy devices. High-flow nasal cannula proves to
have been hundreds of infants studied, with few adverse be an effective high-humidity, high-FIO2 delivery device
events.62,63,66,68,70,93-96 It appears to have the same compli- that is able to improve comfort and therapeutically hydrate
cations as traditional low-flow oxygen delivery devices, the airway. High-flow nasal cannula should be considered
yet is able to provide a higher humidity content (mg/L), as an alternative but not a primary replacement for CPAP
which is probably beneficial. It has been discovered that until future studies can be conducted to prove otherwise.
high-flow nasal cannula may provide positive pressure; REFERENCES
however, it doesn’t appear to be well controlled or repli-
1. James S, Lanman JT. History of oxygen therapy and retrolental
cable.93 The real question lies in whether or not it needs to
fibroplasia. Prepared by the American Academy of Pediatrics, Com-
be well controlled or replicable? If you are using it as a mittee on Fetus and Newborn with the collaboration of special
primary CPAP device and attempting to develop treatment consultants. Pediatrics 1976;57(Suppl 2):591-642.
protocols for care, there needs to be additional randomized 2. Silverman WA. Retrolental fibroplasia: a modern parable. New
multicenter trials to better develop a flow algorithm for York: Grune & Stratton; 1980.
3. Hess J. Oxygen unit for premature and very young infants. Am J
equivocal outcomes. That being said, if you decide to use
Dis Child 1934;47:916-917.
high-flow nasal cannula as your primary CPAP delivery 4. Robertson AF. Reflections on errors in neonatology: I. The “hands-
device, it must be monitored with alarms (disconnect, tube off” years, 1920 to 1950. J Perinatol 2003;23(1):48-55.
occlusion, FIO2). If you are using it as an alternative to 5. Cole CH, Wright KW, Tarnow-Mordi W, Phelps DL. Resolving our
CPAP due to skin breakdown, mother infant bonding, to uncertainty about oxygen therapy. Pediatrics 2003;112(6 Pt 1):1415-
improve developmental care, or as a high FIO2 delivery 1419.
6. Tin W. Oxygen therapy: 50 years of uncertainty. Pediatrics 2002;
device (for example oxygen hood), there appear to be 110(3):615-616.
multiple levels of support for its use. However, adoption 7. Finer NN, Rich WD. Neonatal resuscitation: raising the bar. Curr
has been slow for multiple reasons. Initially, it was the Opin Pediatr 2004;16(2):157-162.
lack of evidence to support its use over current therapies, 8. Anderson CG, Benitz WE, Madan A. Retinopathy of prematurity
but more recently it has probably been cost. Currently, and pulse oximetry: a national survey of recent practices. J Perinatol
2004;24(3):164-168.
high-flow nasal cannula is reimbursed at the same level as 9. Higgins RD, Bancalari E, Willinger M, Raju TN. Executive sum-
a traditional nasal cannula, but with a substantially higher mary of the workshop on oxygen in neonatal therapies: controver-
cost. Depending on which system you use, as well as on sies and opportunities for research. Pediatrics 2007;119(4):790-796.

1200 RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

10. Walsh M, Engle W, Laptook A, Kazzi SN, Buchter S, Rasmussen 34. Gibson DL, Sheps SB, Uh SH, Schechter MT, McCormick AQ.
M, et al. Oxygen delivery through nasal cannulae to preterm in- Retinopathy of prematurity-induced blindness: birth weight-specific
fants: can practice be improved? Pediatrics 2005;116(4):857-861. survival and the new epidemic. Pediatrics 1990;86(3):405-412.
11. Sola A, Saldeno YP, Favareto V. Clinical practices in neonatal 35. Chow LC, Wright KW, Sola A. Can changes in clinical practice
oxygenation: where have we failed? What can we do? J Perinatol decrease the incidence of severe retinopathy of prematurity in very
2008;28(Suppl 1):S28-S34. low birth weight infants? Pediatrics 2003;111(2):339-345.
12. Guyton AC, Hall JE. Textbook of medical physiology, 10th edition. 36. Tin W, Milligan DW, Pennefather P, Hey E. Pulse oximetry, severe
Philadelphia: WB Saunders; 2005. retinopathy, and outcome at one year in babies of less than 28
13. West JB. Pulmonary pathophysiology: the essentials. Baltimore: weeks gestation. Arch Dis Child 2001;84(2):F106-F110.
Lippincott Williams & Wilkins; 2007. 37. Cunningham S, Fleck BW, Elton RA, McIntosh N. Transcutaneous
14. Saugstad OD. Bronchopulmonary dysplasia-oxidative stress and an- oxygen levels in retinopathy of prematurity. Lancet 1995;346(8988):
tioxidants. Semin Neonatol 2003;8(1):39-49. 1464-1465.
15. O’Donovan DJ, Fernandes CJ. Free radicals and diseases in pre- 38. Penn JS, Henry MM, Tolman BL. Exposure to alternating hypoxia
mature infants. Antioxidants and redox signaling 2004;6(1):169-176. and hyperoxia causes severe proliferative retinopathy in the new-
16. Saugstad OD. Oxygen for newborns: how much is too much? J born rat. Pediatric Res 1994;36(6):724-731.
Perinatol 2005;25(Suppl 2):S45-S50. 39. Saito Y, Omoto T, Cho Y, Hatsukawa Y, Fujimura M, Takeuchi T.
17. Finer N, Leone T. Oxygen saturation monitoring for the preterm The progression of retinopathy of prematurity and fluctuation in
infant: the evidence basis for current practice. Pediatr Res 2009; blood gas tension. Graefes Arch Clin Exp Ophthalmol 1993;231(3):
65(4):375-380. 151-156.
18. Baba L, McGrath JM. Oxygen free radicals: effects in the newborn 40. Gunn TR, Easdown J, Outerbridge EW, Aranda JV. Risk factors in
period. Adv Neonatal Care 2008;8(5):256-264. retrolental fibroplasia. Pediatrics 1980;65(6):1096-1100.
19. Thomas W, Speer CP. Nonventilatory strategies for prevention and 41. Palmer EA, Hardy RJ, Davis BR, Stein JA, Mowery RL, Tung B,
treatment of bronchopulmonary dysplasia: what is the evidence? et al. Operational aspects of terminating randomization in the Mul-
Neonatology 2008;94(3):150-159. ticenter Trial of Cryotherapy for Retinopathy of Prematurity. Con-
20. Saugstad OD. Oxidative stress in the newborn: a 30-year perspec- trol Clin Trials 1991;12(2):277-292.
tive. Biol Neonate 2005;88(3):228-236. 42. Bancalari E, Gerhardt T. Bronchopulmonary dysplasia. Pediatr Clin
21. Saugstad OD. Oxygen and retinopathy of prematurity. J Perinatol North Am 1986;33(1):1-23.
2006;26(Suppl 1):S46-S63. 43. Northway WH Jr, Rosan RC. Radiographic features of pulmonary
22. Gaynon MW, Stevenson DK, Sunshine P, Fleisher BE, Landers oxygen toxicity in the newborn: bronchopulmonary dysplasia. Ra-
MB. Supplemental oxygen may decrease progression of pre-thresh- diology 1968;91(1):49-58.
old disease to threshold retinopathy of prematurity. J Perinatol 1997; 44. Bourbia A, Cruz MA, Rozycki HJ. NF-kappaB in tracheal lavage
17(6):434-438. fluid from intubated premature infants: association with inflamma-
23. Phelps DL. Retinopathy of prematurity. Mead Johnson Symp Peri- tion, oxygen, and outcome. Arch Dis Child 2006;91(1):F36-F39.
nat Dev Med 1988(33):63-70. 45. Poets CF. When do infants need additional inspired oxygen? A
24. Stuart MJ, Phelps DL, Setty BN. Changes in oxygen tension and review of the current literature. Pediatric Pulmonol 1998;26(6):424-
effects on cyclooxygenase metabolites: III. Decrease of retinal pros- 428.
tacyclin in kittens exposed to hyperoxia. Pediatrics 1988;82(3):367- 46. Kotecha S, Allen J. Oxygen therapy for infants with chronic lung
372. disease. Arch Dis Child 2002;87(1):F11-F14.
25. Pierce EA, Foley ED, Smith LE. Regulation of vascular endothelial 47. Supplemental therapeutic oxygen for prethreshold retinopathy of
growth factor by oxygen in a model of retinopathy of prematurity. prematurity (STOP-ROP), a randomized, controlled trial. I: primary
Arch Ophthalmol 1996;114(10):1219-1228. outcomes. Pediatrics 2000;105(2):295-310.
26. Robbins SG, Rajaratnam VS, Penn JS. Evidence for upregulation 48. Askie LM, Henderson-Smart DJ, Irwig L, Simpson JM. Oxygen-
and redistribution of vascular endothelial growth factor (VEGF) saturation targets and outcomes in extremely preterm infants. N Engl
receptors flt-1 and flk-1 in the oxygen-injured rat retina. Growth J Med 2003;349(10):959-967.
Factors 1998;16(1):1-9. 49. Boykin AR, Quivers ES, Wagenhoffer KL, Sable CA, Chaney HR,
27. Engle MA, Baker DH, Baras I, Freemond A, Laupus WE, Norton Glass P, et al. Cardiopulmonary outcome of neonatal extracorporeal
EW. Oxygen administration and retrolental fibroplasia. AMA Am J membrane oxygenation at ages 10-15 years. Crit Care Med 2003;
Dis Child 1955;89(4):399-413. 31(9):2380-2384.
28. Lanman JT, Guy LP, Dancis J. Retrolental fibroplasia and oxygen 50. Rabi Y, Rabi D, Yee W. Room air resuscitation of the depressed
therapy. J Am Med Assoc 1954;155(3):223-226. newborn: a systematic review and meta-analysis. Resuscitation 2007;
29. Patz A, Hoeck LE, De La Cruz E. Studies on the effect of high 72(3):353-363.
oxygen administration in retrolental fibroplasia. I. Nursery obser- 51. Saugstad OD, Ramji S, Soll RF, Vento M. Resuscitation of new-
vations. Am J Ophthalmol 1952;35(9):1248-1253. born infants with 21% or 100% oxygen: an updated systematic
30. Weintraub DH, Tabankin A. Relationship of retrolental fibroplasia review and meta-analysis. Neonatology 2008;94(3):176-182.
to oxygen concentration. J Pediatr 1956;49(1):75-79. 52. Tan A, Schulze A, O’Donnell CP, Davis PG. Air versus oxygen for
30a. Kinsey VE. Retrolental fibroplasia: cooperative study of retrolental resuscitation of infants at birth. Cochrane Database Syst Rev 2005(2):
fibroplasia and the use of oxygen. AMA Arch Ophthalmol 1956; CD002273.
56(4):481-543. 53. Harling AE, Beresford MW, Vince GS, Bates M, Yoxall CW. Does
31. Avery ME. Recent increase in mortality from hyaline membrane the use of 50% oxygen at birth in preterm infants reduce lung
disease. J Pediatr 1960;57:553-559. injury? Arch Dis Child 2005;90(5):F401-F405.
32. McDonald AD. Cerebral palsy in children of very low birth weight. 54. Escrig R, Arruza L, Izquierdo I, Villar G, Saenz P, Gimeno A, et al.
Arch Dis Child 1963;38:579-588. Achievement of targeted saturation values in extremely low gestational
33. Hussain N, Clive J, Bhandari V. Current incidence of retinopathy of age neonates resuscitated with low or high oxygen concentrations: a
prematurity, 1989-1997. Pediatrics 1999;104(3):e26. prospective, randomized trial. Pediatrics 2008;121(5):875-881.

RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9 1201

 OXYGEN THERAPY IN THE NEONATAL CARE ENVIRONMENT

55. Naumburg E, Bellocco R, Cnattingius S, Jonzon A, Ekbom A. 76. Tablan O, Anderson LJ, Besser R, Bridges C, Hajjeh R. Guidelines
Supplementary oxygen and risk of childhood lymphatic leukaemia. for preventing health-care-associated pneumonia, 2003: recommen-
Acta Paediatr 2002;91(12):1328-1333. dations of CDC and the Healthcare Infection Control Practices Ad-
56. Spector LG, Klebanoff MA, Feusner JH, Georgieff MK, Ross JA. visory Committee. MMWR Recomm Rep 2004;53(RR-3):1-36.
Childhood cancer following neonatal oxygen supplementation. J Pe- 77. Kopelman AE, Holbert D. Use of oxygen cannulas in extremely
diatr 2005;147(1):27-31. low birthweight infants is associated with mucosal trauma and bleed-
57. Davies P, Cheng D, Fox A, Lee L. The efficacy of noncontact ing, and possibly with coagulase-negative staphylococcal sepsis. J
oxygen delivery methods. Pediatrics 2002;110(5):964-967. Perinatol 2003;23(2):94-97.
58. American Association for Respiratory Care. Clinical practice guide- 78. Beckham RW, Mishoe SC. Sound levels inside incubators and ox-
line. Selection of an oxygen delivery device for neonatal and pe- ygen hoods used with nebulizers and humidifiers. Respir Care 1982;
diatric patients: 2002 revision and update. Respir Care 2002;47(6): 27(1):33-40.
707-716. 79. Tepper RS. Airway reactivity in infants: a positive response to
59. Cairo JM, Susan P. Mosby’s respiratory care equipment. St Louis: methacholine and metaproterenol. J Appl Physiol 1987;62(3):1155-
Mosby; 1999. 1159.
60. Whittaker K. Comprehensive perinatology and pediatric respiratory 80. Geller DE, Morgan WJ, Cota KA, Wright AL, Taussig LM. Airway
care. Florence, KY: Delmar; 2001. responsiveness to cold, dry air in normal infants. Pediatric Pulmo-
61. Finer NN, Bates R, Tomat P. Low flow oxygen delivery via nasal nol 1988;4(2):90-97.
cannula to neonates. Pediatric Pulmonol 1996;21(1):48-51. 81. Greenspan JS, DeGiulio PA, Bhutani VK. Airway reactivity as
62. Holleman-Duray D, Kaupie D, Weiss MG. Heated humidified high- determined by a cold air challenge in infants with bronchopulmo-
flow nasal cannula: use and a neonatal early extubation protocol. J nary dysplasia. J Pediatr 1989;114(3):452-454.
Perinatol 2007;27(12):776-781. 82. Williams R, Rankin N, Smith T, Galler D, Seakins P. Relationship
63. Sreenan C, Lemke RP, Hudson-Mason A, Osiovich H. High-flow between the humidity and temperature of inspired gas and the func-
nasal cannulae in the management of apnea of prematurity: a com- tion of the airway mucosa. Crit Care Med 1996;24(11):1920-1929.
parison with conventional nasal continuous positive airway pres- 83. Phelps DL, Rosenbaum AL. Effects of marginal hypoxemia on
sure. Pediatrics 2001;107(5):1081-1083. recovery from oxygen-induced retinopathy in the kitten model. Pe-
64. Locke RG, Wolfson MR, Shaffer TH, Rubenstein SD, Greenspan diatrics 1984;73(1):1-6.
JS. Inadvertent administration of positive end-distending pressure 84. Flynn JT, Bancalari E, Snyder ES, Goldberg RN, Feuer W, Cassady
during nasal cannula flow. Pediatrics 1993;91(1):135-138. J, et al. A cohort study of transcutaneous oxygen tension and the
65. Spence KL, Murphy D, Kilian C, McGonigle R, Kilani RA. High- incidence and severity of retinopathy of prematurity. N Engl J Med
flow nasal cannula as a device to provide continuous positive air- 1992;326(16):1050-1054.
way pressure in infants. J Perinatol 2007;27(12):772-775. 85. O’Brodovich HM, Mellins RB. Bronchopulmonary dysplasia. Un-
66. Campbell DM, Shah PS, Shah V, Kelly EN. Nasal continuous pos- resolved neonatal acute lung injury. Am Rev Respir Dis 1985;
itive airway pressure from high flow cannula versus Infant Flow for 132(3):694-709.
preterm infants. J Perinatol 2006;26(9):546-549. 86. Claure N, Gerhardt T, Everett R, Musante G, Herrera C, Bancalari
67. McLaughlin AJ Jr. Allergic contact dermatitis from oxygen cannu- E. Closed-loop controlled inspired oxygen concentration for me-
las. Respir Care 1980;25(10):1024-1026. chanically ventilated very low birth weight infants with frequent
68. Woodhead DD, Lambert DK, Clark JM, Christensen RD. Compar- episodes of hypoxemia. Pediatrics 2001;107(5):1120-1124.
ing two methods of delivering high-flow gas therapy by nasal can- 87. Salyer JW. Neonatal and pediatric pulse oximetry. Respir Care
nula following endotracheal extubation: a prospective, randomized, 2003;48(4):386-396.
masked, crossover trial. J Perinatol 2006;26(8):481-485. 88. Hay WW Jr, Rodden DJ, Collins SM, Melara DL, Hale KA, Fashaw
69. Kopelman AE. Airway obstruction in two extremely low birth- LM. Reliability of conventional and new pulse oximetry in neonatal
weight infants treated with oxygen cannulas. J Perinatol 2003;23(2): patients. J Perinatol 2002;22(5):360-366.
164-165. 89. Barker SJ. “Motion-resistant” pulse oximetry: a comparison of new
70. Jhung MA, Sunenshine RH, Noble-Wang J, Coffin SE, St John K, and old models. Anesth Analg 2002;95(4):967-972.
Lewis FM, et al. A national outbreak of Ralstonia mannitolilytica 90. Sahni R, Gupta A, Ohira-Kist K, Rosen TS. Motion resistant pulse
associated with use of a contaminated oxygen-delivery device among oximetry in neonates. Arch Dis Child 2003;88(6):F505-F508.
pediatric patients. Pediatrics 2007;119(6):1061-1068. 91. Workie FA, Rais-Bahrami K, Short BL. Clinical use of new-gen-
71. O’Brien BJ, Rosenfeld JV, Elder JE. Tension pneumo-orbitus and eration pulse oximeters in the neonatal intensive care unit. Am J
pneumocephalus induced by a nasal oxygen cannula: report on two Perinatol 2005;22(7):357-360.
paediatric cases. J Paediatr Child Health 2000;36(5):511-514. 92. Barker SJ, Badal JJ. The measurement of dyshemoglobins and total
72. Frenckner B, Ehren H, Palmer K, Noren G. Pneumocephalus caused hemoglobin by pulse oximetry. Curr Opin Anaesthesiol 2008;21(6):
by a nasopharyngeal oxygen catheter. Criti Care Med 1990;18(11): 805-810.
1287-1288. 93. Lampland AL, Plumm B, Meyers PA, Worwa CT, Mammel MC.
73. Campos JM, Boechat MC, Azevedo ZM, Garrido JR, Rodrigues Observational study of humidified high-flow nasal cannula com-
SL, Pone MV. Pneumocephalus and exophthalmos secondary to pared with nasal continuous positive airway pressure. J Pediatr
acute sinusitis and nasopharyngeal oxygen catheter. Clin Pediatr 2009;154(2):177-182.
(Phila) 1994;33(2):127-128. 94. Finer NN, Mannino FL. High-flow nasal cannula: a kinder, gentler
74. Jasin LR, Kern S, Thompson S, Walter C, Rone JM, Yohannan CPAP? J Pediatr 2009;154(2):160-162.
MD. Subcutaneous scalp emphysema, pneumo-orbitis and pneumo- 95. Saslow JG, Aghai ZH, Nakhla TA, Hart JJ, Lawrysh R, Stahl GE,
cephalus in a neonate on high humidity high flow nasal cannula. J et al. Work of breathing using high-flow nasal cannula in preterm
Perinatol 2008;28(11):779-781. infants. J Perinatol 2006;26(8):476-480.
75. Scopes JW, Ahmed I. Indirect assessment of oxygen requirements 96. Shoemaker MT, Pierce MR, Yoder BA, DiGeronimo RJ. High flow
in newborn babies by monitoring deep body temperature. Arch Dis nasal cannula versus nasal CPAP for neonatal respiratory disease: a
Child 1966;41(215):25-33. retrospective study. J Perinatol 2007;27(2):85-91.

1202 RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9