ORIGINAL ARTICLE

Use of oxygen in the resuscitation of preterm infants: Currentopinion and practice in Australia and New ZealandRowena L Clark, Kei Lui and Ju Lee Oei

Department of Newborn Care, Royal Hospital for Women, Sydney, New South Wales, Australia

Aim: The aim of this paper was to explore the opinions and practices of tertiary health-care professionals in Australia and New Zealandregarding air and oxygen blending (OB) for the resuscitation of preterm infants.Methods: Structured questionnaires were sent to the directors of 25 tertiary perinatal units, with instructions to distribute the questionnairesto 15 pertinent clinical staff.Results: Response rate was 72% (n = 271); medical-staff response was 25%. Sixteen (64%) perinatal units had OB resuscitation equipment.Among respondents, 114 (42%) had access to OB and 73 (27%) had OB for all resuscitations. Pulse oximetry was available to 160 (59%) ofrespondents. The majority (173, 64%) would initiate resuscitation with Fractional inspired oxygen (FiO2) ranging from 0.3 to 0.9 (mean 0.5), with 15%and 21% preferring air and 100% oxygen, respectively. There were large variations in managing FiO2 changes thereafter. Half of the respondentswere either unsure (39%) or not convinced (15%) that 100% oxygen during resuscitation would cause harm. Conversely, 42% suggested that OB mightimprove outcome with bronchopulmonary dysplasia and retinopathy of prematurity being the most important considerations. Most (92%) wouldadvocate for OB in the delivery suite. Set-up cost (50%) and lack of guided experience (38%) ranked highest as barriers to change.Conclusions: Two-thirds of the tertiary centres have at least some OB equipment in the delivery suite, but the ways and opinions in which OBis utilised differ widely. Most practitioners would advocate for a change. There is an urgent need for further research to achieve a consistent andmeaningful clinical management for OB resuscitation of preterm infants.

Key words: air; neonate; oxygen; premature infant; resuscitation; survey.

Since early last century, pure or 100% oxygen has been usedand recommended as the standard gas for newborn resuscita-tion. The American Heart Association (AHA) in 20051 and theInternational Liaison Committee on Resuscitation (ILCOR) in20062 advised that pure oxygen should be administered judi-ciously during newborn resuscitation, especially for asphyxiatedand premature infants who are most vulnerable to oxidativeinjury.

These recommendations are based on large studies whichshow that it is safe and probably beneficial to resuscitate term-asphyxiated infants with room air.3–8 This follows evidence dem-onstrating that hyperoxia, especially when superimposed on ahypoxic environment, results in the production of reactiveoxygen species, molecules which interfere with cell function,structure and repair, especially in the lung and brain,9,10 but alsoin the heart and kidney.11 Although not always hypoxic, pre-term infants are theoretically at a considerably higher risk ofoxidative damage because of an inherent immaturity of antioxi-dant defences9 and the pulmonary system, necessitating anincreased need for supplemental oxygen.

However, in contrast to asphyxiated term or near-terminfants, the effects of lower oxygen concentrations for theimmediate post-natal resuscitation of extremely prematureinfants have not been well studied. So far, only two small trialshave shown that resuscitating premature infants with either airor 50% oxygen causes no short-term harm (or benefits) whencompared with 100% oxygen.12,13 Despite this, the AustralianResuscitation Council (ARC) has now recommended, inrecently published guidelines,14 that air should be used to ini-tiate newborn resuscitation, without reference to the gestationalage, even though premature infants have a higher likelihood ofhaving lung disease. It has also been shown by O’Donnell et al.that only nine out of 25 Australian15 and 20 out of 40 inter-national16 perinatal centres were able to blend oxygen in adelivery-suite setting.

Key Points

1 Despite Australian Resuscitation Council guidelines now statingthat air should be used for initiation of neonatal resuscitation,one-third of tertiary neonatal units are unable to provideblended oxygen for use in preterm resuscitation.

2 There is considerable variation in attitudes to, and practices for,oxygen delivery in the premature newborns.

3 Most practitioners support the use of blended air and oxygenin the delivery suite, although half of them acknowledge thatevidence is incomplete.

Correspondence: Assistant Professor Kei Lui, Department of NewbornCare, The Royal Hospital for Women, Barker Street Randwick NSW 2031,Australia. Fax: +61 2 9382 6191; email: kei.lui@sesiahs.health.nsw.gov.au

Accepted for publication 17 June 2008.

doi:10.1111/j.1440-1754.2008.01430.x

Journal of Paediatrics and Child Health 45 (2009) 31–35© 2008 The AuthorsJournal compilation © 2008 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

31

With the very successful regionalisation of perinatal care inAustralia and New Zealand, the vast majority of very prematureinfants are delivered in tertiary centres that are also leaders inclinical practice and perinatal care. This study was thereforeconducted to explore the practices and opinions of health-careprofessionals, conversant in the resuscitation of preterm infants,regarding the use of air and blended oxygen during resuscitationin these perinatal centres.

Materials and Methods

The survey was conducted from March to July 2006. A struc-tured questionnaire with a covering letter was sent via email tothe director of each of the 25 level-3 neonatal intensive careunits (NICUs) in Australia and New Zealand with on-site deliv-eries, with the request to distribute the questionnaire amongappropriate staff of the unit via email. Identical postal question-naires with prepaid self-addressed envelopes were also sent tothe nursing unit managers. We requested the unit directors, inconjunction with the nursing managers of each unit, to sample15 staff from all professionals who attended preterm deliveriesin their service. This nested opinion-survey method invitinginterested and informed respondents across disciplines fromeach centre has been similarly used in our previous studies,17,18

with an expectation of more participation from nurses thandoctors as there are usually more nurses than doctors in anytertiary unit. To improve response rate, questionnaires wereemailed and posted again as reminders 1 month later. Unitdirectors were then surveyed via email to ask whether theirunits had the facility to blend oxygen for all, some or no babiesrequiring resuscitation. In addition, individual NICUs were sur-veyed by telephone after completion of the original survey tocompare unit with individual response.

Neonatal intensive care units and directors were identified viathe Australian and New Zealand Neonatal Network. The surveywas undertaken with the consent of the institutions involved.Individual confidentiality was maintained throughout theprocess.

The first part of the questionnaire was concerned with demo-graphic information of the respondents (age, occupation, yearsof practice and locality of hospital). The second part exploredcurrent individual practice surrounding the use of oxygen atdelivery, including blenders, pulse oximetry and resuscitationmanagement in the delivery suite. Individuals were also askedabout their views regarding acceptable pulse-oximetry readingsin a preterm infant, about their confidence with current evi-dence for blended oxygen and 100% oxygen in the deliverysuite, and their thoughts regarding potential barriers to imple-menting change. The 21-item questionnaire used a combinationof tick-box responses and rank-ordered scales, with 10 ques-tions having the option to add free text. The questionnaire waspiloted at the Royal Hospital for Women, and minor modifica-tions were made for clarity prior to distribution.

Data are shown as percentages or as mean with the range ofvalues. Analysis of continuous data was performed with t-tests,Fischer’s exact test or one-way ANOVA where appropriate. A Pvalue of <0.05 was considered significant. All statistical analyseswere performed using Graphpad Instat 3 software (GraphPadSoftware Inc., La Jolla, CA, USA).

Results

A response rate of 72% was obtained after two mailings(n = 271 of a projected total of 375, as 15 responses per unit).Not every question was answered by every individual. Thedemographics of the respondents are detailed in Table 1. Themajority of respondents (40%) were neonatal nurses with atleast 10 years of experience in a tertiary NICU and 25% weremedical staff. One-hundred fourteen respondents (43%) hadthe facility to provide blended oxygen to some (16%) or allbabies (27%) in the delivery suite. About half (140, 52%) didnot have access to oxygen-blending facility and 6% were notcertain. Telephone survey indicated that 16 (64%) of the indi-vidual NICUs had oxygen blending (OB)-resuscitation equip-ment. Nine (36%) used them for selected cases, seven (28%) forall resuscitations while nine (36%) did not have any OB facili-ties at all in their delivery suites.

How oxygen blenders would be usedduring resuscitation

Respondents with access to OB were significantly more likelyto use blended oxygen for preterm (101/114, 86%) than terminfants (92/114, 72%, P < 0.01). There were different ap-proaches to term and preterm infants (Fig. 1). Two-thirds ofrespondents (64%) would initiate resuscitation at 50% FiO2

(range 30–90%) regardless of gestational age. Fifteen per centwould commence with air and increase the oxygen concen-tration as required; 21% would start with 100% oxygenand decrease it as clinical condition allows. There was nodifference seen between either levels of experience oroccupation.

Table 1 Demographic characteristics of 271 respondents

n %

Age groups (years)

20–30 46 17

30–40 72 27

40–50 98 36

>50 54 20

Not answered 1 –

Occupation

Neonatal nurse 168 62

Midwife 35 13

Neonatalogist 33 12

Registrar 20 7

Fellow 14 5

Not answered 1 –

Years of practice

<1 18 6

1–5 62 23

5–10 42 16

>10 149 55

Not answered 0 –

Air and oxygen for preterm resuscitation R Clark et al.

Journal of Paediatrics and Child Health 45 (2009) 31–35© 2008 The Authors

Journal compilation © 2008 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

32

Over half (54%) of the respondents would reassess patientcondition and alter oxygen concentrations, if required, every30 s or less. A further 28% would reassess at 1 min intervals.Sixteen per cent stated that they would start with higher con-centrations of oxygen and then turn the oxygen down ‘depend-ing on condition’ or when the ‘patient’s condition and oximetrywas good’. A few (5%) stated that they would start with air andnot change oxygen concentration if the heart rate was good. Inresponse to reassessment, 45% would alter FiO2 by 10% incre-ments or less and 40% would alter FiO2 by 11–20% increments,with a minority of individuals choosing to use larger incre-ments. Nine per cent of respondents said that the titration ofoxygen ‘depended on the infant’s condition’ but did not specifyfurther details. Only one individual chose to titrate oxygen topulse oximetry values. Colour (41% of respondents) was themost important factor to influence changes in oxygen adminis-tration, followed by oxygen saturation (34%) and heart rate(22%).

Pulse oximetry

Delivery-suite pulse-oximetry facilities were available to 160(60%) of respondents. The median oxygen saturation consid-ered acceptable for a 5-min-old 24-week infant was 80% (range50–95%). Five (3%) respondents considered oxygen satura-tions unimportant for premature infants during resuscitation.Neither occupation nor levels of experience were important toperceptions of acceptable oximetry readings during pretermresuscitation.

Barriers to implementing change

Individuals were asked about potential difficulties when consid-ering using blended oxygen in the delivery suite. Set-up costincluding equipment was the predominant impediment (136,50%) followed by lack of guided experience including difficul-ties in monitoring (104, 38%) in instituting the use of blendedequipment.

Current evidence and beliefs

Almost half (46%) of the respondents believed that using 100%oxygen during resuscitation could cause adverse long-termeffects in preterm infants (see Fig. 2), while 14% thought thatexisting evidence was insufficient. The remaining respondentswere undecided. Conversely, 43% thought that using blendedoxygen during preterm resuscitation would improve long-termoutcome, with improvements in retinopathy of prematurity(39%), bronchopulmonary dysplasia (32%) and lung diseaseupon admission to the NICU (16%) being the most importantconsiderations. Most respondents (92%) would advocate OB intheir respective hospitals even though 58% were dissatisfiedwith current evidence supporting this practice.

Discussion

Each year, more than 1 million infants worldwide need assis-tance to complete the transition from intra- to extra-uterine

(a)

(b)

When would you use blended oxygen for term infants?

22

3

66

917

2

77

4

0

10

20

30

40

50

60

70

80

90

I will only use100% oxygen

I will only use air

Blended oxygen

for all infants

Asphyxiatedterm

infants only

% of respondents

22

2

71

5

17

2

77

4

0102030405060708090

I will only use100% oxygen

I will only use air

Blendedoxygen

for all infants

Infants less than 30

weeks only

% of respondents

When would you use blended oxygen for preterm infants?

Fig. 1 Approaches to using air, oxygen blending or 100% oxygen for

resuscitating (a) term or (b) preterm infants by (�) medical and (�) nursing

staff.

020406080

100120140160

Ch

ron

ic L

un

gdi

sea

se

Re

tino

pa

thy

of

pre

ma

turi

ty

Wo

rse

lun

gd

isea

se o

na

dm

issi

on

to

NIC

U

Intr

ave

ntr

icu

lar

ha

em

orr

ha

ge

Incr

ea

sed

ra

teo

f se

psi

s Oth

erN

um

be

r o

f re

spo

nd

en

ts

Fig. 2 Beliefs surrounding the long-term effects of using 100% oxygen

during preterm resuscitation. NICU, neonatal intensive care unit.

R Clark et al. Air and oxygen for preterm resuscitation

Journal of Paediatrics and Child Health 45 (2009) 31–35© 2008 The AuthorsJournal compilation © 2008 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

33

life.19 This makes neonatal resuscitation one of the mostcommon and important medical procedures in the world, andpremature infants are particularly likely to require immediatepost-natal support because of physiological and anatomicalinadequacy. Colour, or more accurately, ‘pinkness’, has tradi-tionally been one of the primary goals in the management of theunresponsive newborn infant and, indeed, the degree of ‘pink-ness’ constitutes a major component of the Apgar score,20 ameasure of adequacy of the resuscitative process. Therefore,pure or 100% oxygen has been used for neonatal resuscitationfor over 100 years to achieve ‘pinkness’ in the shortest possibletime frame.

Recent studies predominantly conducted in India demon-strate that using air during resuscitation decreases mortality inasphyxiated (and predominantly full-term) infants.3–8 Largemeta-analyses21–23 of over 1300 infants confirm decreased mor-tality of 30–40% in those resuscitated with air and also fasterresponse to resuscitation. Vento et al.24 showed prolonged oxi-dative stress at 4 weeks of age, highlighting that the effects ofeven brief periods of oxygen cause damage lasting far longerthan was first thought. However, the studies included in thesemeta-analyses have significant methodological limitations; ran-domisation and blinding were inadequate in five of the seventrials analysed, and the majority of patients were recruited fromdeveloping nations where ante- and perinatal care, equipmentand neonatal mortality rates are very different. Most 174 of the177 deaths occurred in developing nations, and 168 patientscrossed over from air to 100% oxygen. Only one study haslooked at limited long-term outcome,25 which showed no sig-nificant difference was seen in those infants resuscitated withair compared with those resuscitated with 100% oxygen in asurvey at 18–24 months old but perhaps a trend to increasedrates of cerebral palsy in the air group.

It is tempting to think that vulnerable preterm infants wouldbenefit far more from the judicious use of oxygen at resuscita-tion than term infants. However, there is a paucity of evidenceto guide us. In Saugstad et al.’s meta-analysis in 2005,23 sub-group meta-analysis of infants less than 37 weeks (but all morethan 1000 g) demonstrated that mortality was decreased from35% in the 100% oxygen group to 21% in the air group. Therehave been only two small studies looking solely at the use of lessthan 100% oxygen for resuscitation of preterm infants,12,13 andthese show that it is both possible and safe in most infants butwith insufficient numbers to demonstrate benefit.

In February 2006,14 the ARC recommended that air should beused to commence resuscitation regardless of gestation, but wefound that only 64% of units were able to blend oxygen forsome or all infants in the delivery suite. This is a higher figurecompared with that found in previous studies of resuscitationequipment used in Australia and New Zealand15 in 2005,perhaps a reflection of the growing uneasiness with the use of100% oxygen. In 2007, the recommendation of air to be usedregardless of gestation was emphasised in a commentary on theARC neonatal guidelines.26

The way in which blenders are used shows considerablevariation between individuals. The questionnaire may have arti-ficially widened this gap as replies were limited to tick-boxresponses. A number of practitioners chose to reassess and alteroxygen concentration ‘depending on the infant’s condition’,

highlighting that the way that individuals chose to resuscitate isaffected by previous experience and intuition. A major difficultyfor the use of blenders is the lack of guidelines. The currentARC recommendation is that ‘air should be used initially, withsupplemental oxygen reserved for those whose condition doesnot improve in the first few minutes of life’.14 There is littleguidance for how, or when, to alter oxygen concentration, or forwhich groups of babies for which blended oxygen may not besuitable.2 It has been suggested that practitioners should aimto achieve normoxia throughout the resuscitation process bytitrating oxygen according to oxygen-saturation monitoring.This is complicated by the paucity of evidence for what consti-tutes normal oxygen saturations in newly born preterm infants.Although most of the extremely preterm infants are delivered intertiary centres, many of the moderately preterm resuscitationsare conducted in levels 1 and 2 obstetric/neonatal services.More studies in guiding practices are required.

In this study, practitioners thought that oxygen saturations of80% were appropriate for a 24-week gestation infant at 5 min ofage. This assessment is in general agreement with recent datafrom a study by Kamlin et al.27 They found that those preterminfants that did not require supplemental oxygen generally hadsaturations between 75% and 90% at 5 min of age. Preterminfants took significantly longer to reach saturations greater than90% compared with term. However, the infants studied had amean gestation of 33.5 weeks, and it seems likely that the morepreterm an infant is, the longer it may take to reach saturations ofgreater than 90%. It is probable that practitioners overestimatethe ‘normal’ oxygen saturation in this extremely preterm group.The oxygen saturations defined in this and other studies28 shouldbe applied with caution to sick or extremely preterm infants,further complicating the role of oximetry during resuscitation.

Surveys are inherently limited by potential respondent-selection biases. Directors were surveyed after completion of theoriginal questionnaire. Consistently, 27% of respondents and28% of the directors reported OB availability for all deliveries.However, only 8% of the respondents in contrast to 36% of thedirectors replied that OB was available for selected deliveriesonly. This discrepancy in selected availability may have beenskewed by more respondents from centres without OB facilitiesor from an increase in recent purchases of OB equipment occur-ring soon after the initial survey.

We acknowledge that our survey was conducted only a fewmonths after publication of the recommendations of the AHA,1

ILCOR and AAP2 which may not have been translated intoroutine clinical practice in the majority of units. The targetedrespondents, however, by study design, were selected on thebasis of their being informed and interested clinicians and assuch, would have been more aware of recent practice changes,especially one suggesting that ‘excessive tissue oxygenationshould be avoided, especially in the preterm infant’.2

In conclusion, we found that one-third of tertiary perinatalcentres in Australia and New Zealand do not have the ability toblend oxygen in the delivery suite, and the ways in whichblenders are used vary considerably. Most of the practitionerswould advocate for a change in current practices despite themajority acknowledging that there was inadequate evidence toguide these changes, especially in regard to the preterm popu-lation. There is therefore an urgent need for further research

Air and oxygen for preterm resuscitation R Clark et al.

Journal of Paediatrics and Child Health 45 (2009) 31–35© 2008 The Authors

Journal compilation © 2008 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

34

into the effects of air, blended oxygen and 100% oxygen duringthe resuscitation of preterm infants. More studies are requiredto guide practice.

References

1 American Heart Association. 2005 American Heart Associationguidelines for cardiopulmonary resuscitation and emergencycardiovascular care. Part 13: Neonatal resuscitation guidelines.Circulation 2005; 112 (Suppl. I): IV 118–IV 119.

2 International Liaison Committee on Resuscitation (ILCOR). TheInternational Liaison Committee on Resuscitation (ILCOR) consensuson science with treatment recommendations for pediatric andneonatal patients: neonatal resuscitation. Pediatrics 2006; 117:e978–88.

3 Vento M, Asensi M, Sastre J, Garcia-Sala F, Pallardo FV, Vina J.Resuscitation with room air instead of 100% oxygen preventsoxidative stress in moderately asphyxiated term neonates. Pediatrics2001; 107: 642–7.

4 Ramji S, Rasaily R, Mishra PK et al. Resuscitation of asphyxiatednewborns with room air or 100% oxygen at birth: a multicentricclinical trial. Indian Pediatr. 2003; 40: 510–7.

5 Ramji S, Ahuja S, Thirupuram S, Rootwelt T, Rooth G, Saugstad OD.Resuscitation of asphyxic newborn infants with room air or 100%oxygen. Pediatr. Res. 1993; 34: 809–12.

6 Saugstad OD, Rootwelt T, Aalen O. Resuscitation of asphyxiatednewborn infants with room air or oxygen: an international controlledtrial: the Resair 2 study. Pediatrics 1998; 102: e1.

7 Vento M, Asensi M, Sastre J, García-Sala F, Viña J. Six years ofexperience with the use of room air for the resuscitation ofasphyxiated newly born term infants. Biol. Neonate 2001; 79: 261–7.

8 Bajaj N, Udani RH, Nanavati RN. Room air vs. 100 per cent oxygen forneonatal resuscitation: a controlled clinical trial. J. Trop. Pediatr. 2005;51: 206–11.

9 Asikainen TM, White CW. Pulmonary antioxidant defenses in thepreterm newborn with respiratory distress and bronchopulmonarydysplasia in evolution: implications for antioxidant therapy. Antioxid.Redox Signal. 2004; 6: 155–67.

10 Feet BA, Yu XQ, Rootwelt T, Oyasaeter S, Saugstad OD. Effects ofhypoxemia and reoxygenation with 21% oxygen or 100% oxygen innewborn piglets: extracellular hypoxanthine in cerebral cortex andfemoral muscle. Crit. Care Med. 1997; 25: 1384–91.

11 Vento M, Sastre J, Asensi MA, Vina J. Room-air resuscitation causesless damage to heart and kidney than 100% oxygen. Am. J. Respir.Crit. Care Med. 2005; 172: 1393–8.

12 Lundstrom KE, Pryds O, Greisen G. Oxygen at birth and prolongedcerebral vasoconstriction in preterm infants. Arch. Dis. Child. FetalNeonatal Ed. 1995; 73: F81–6.

13 Harling AE, Beresford MW, Vince GS, Bates M, Yoxall CW. Does theuse of 50% oxygen at birth in preterm infants reduce lung injury?Arch. Dis. Child. Fetal Neonatal Ed. 2005; 90: F401–5.

14 Australian Resuscitation Council. Neonatal Guidelines: ARC Manualof Guidelines. Section 13. February 2006. Available from: http://www.resus.org.au/public/guidelines/index.htm [accessed November2006].

15 O’Donnell CP, Davis PG, Morley CJ. Use of supplementary equipmentfor resuscitation of newborn infants at tertiary perinatal centres inAustralia and New Zealand. Acta Paediatr. 2005; 94: 1261–5.

16 O’Donnell CP, Davis PG, Morley CJ. Positive pressure ventilation atneonatal resuscitation: review of equipment and international surveyof practice. Acta Paediatr. 2004; 93: 583–8.

17 Lui K, Bajuk B, Foster K et al. Perinatal care at the borderlines ofviability: a consensus statement based on a NSW and ACT consensusworkshop. Med. J. Aust. 2006; 185: 495–500.

18 Oei J, Askie L, Tobiansky R, Lui K. Attitudes of neonatal clinicianstowards resuscitation of the extremely premature infant: anexploratory survey. J. Paediatr. Child Health 2000; 36: 357–62.

19 World Health Organization (WHO). Basic Newborn Resuscitation: APractical Guide, 1998. WHO Division of Family Health WHO/RHT/MSM/98.1. Available from www.who.int/bookorders

20 Apgar V. A proposal for a new method of evaluation of the newborninfant. Curr. Res. Anesth. Analg. 1953; 32: 260–7.

21 Rabi Y, Rabi D, Yee W. Room air resuscitation of the depressednewborn: a systematic review and meta-analysis. Resuscitation 2007;72: 353–63.

22 Tan A, Schulze A, O’Donnell CP, Davis PG. Air versus oxygen forresuscitation of infants at birth. Cochrane Database Syst. Rev. 2005;2: CD002273.

23 Saugstad OD, Ramji S, Vento M. Resuscitation of depressed newborninfants with ambient air or pure oxygen: a meta-analysis. Biol.Neonate 2005; 87: 27–34.

24 Vento M, Asensi M, Sastre J, Lloret A, García-Sala F, Viña J. Oxidativestress in asphyxiated term infants resuscitated with 100% oxygen.J. Pediatr. 2003; 142: 240–6.

25 Saugstad OD, Ramji S, Irani SF et al. Resuscitation of newborn infantswith 21% or 100% oxygen: follow-up at 18 to 24 months. Pediatrics2003; 112: 296–300.

26 Morley CJ. New Australian Neonatal Resuscitation Guidelines.J. Paediatr. Child Health 2007; 43: 6–8.

27 Kamlin CO, O’Donnell CP, Davis PG, Morley CJ. Oxygen saturation inhealthy infants immediately after birth. J. Pediatr. 2006; 148: 585–9.

28 Rabi Y, Yee W, Chen SY, Singhal N. Oxygen saturation trendsimmediately after birth. J. Pediatr. 2006; 148: 590–4.

R Clark et al. Air and oxygen for preterm resuscitation

Journal of Paediatrics and Child Health 45 (2009) 31–35© 2008 The AuthorsJournal compilation © 2008 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

35