Intervention Review

You have free access to this content

High-flow nasal cannula therapy for infants with bronchiolitis

  1. Sean Beggs1,2,*,
  2. Zee Hame Wong3,
  3. Sheena Kaul4,
  4. Kathryn J Ogden3,
  5. Julia AE Walters2

Editorial Group: Cochrane Acute Respiratory Infections Group

Published Online: 20 JAN 2014

Assessed as up-to-date: 15 MAY 2013

DOI: 10.1002/14651858.CD009609.pub2

How to Cite

Beggs S, Wong ZH, Kaul S, Ogden KJ, Walters JAE. High-flow nasal cannula therapy for infants with bronchiolitis. Cochrane Database of Systematic Reviews 2014, Issue 1. Art. No.: CD009609. DOI: 10.1002/14651858.CD009609.pub2.

Author Information

  1. 1

    Royal Hobart Hospital, Department of Paediatrics, Hobart, Tasmania, Australia

  2. 2

    University of Tasmania, School of Medicine, Hobart, Tasmania, Australia

  3. 3

    University of Tasmania, Launceston Clinical School, Launceston, Tasmania, Australia

  4. 4

    University of Tasmania, Rural Clinical School, Burnie, Tasmania, Australia

*Sean Beggs, Department of Paediatrics, Royal Hobart Hospital, 48 Liverpool Street, Hobart, Tasmania, 7000, Australia.

Publication History

  1. Publication Status: New
  2. Published Online: 20 JAN 2014


Characteristics of included studies [ordered by study ID]
Hilliard 2012

MethodsProspective, randomised, pilot study, single centre in UK

ParticipantsInfants with a clinical diagnosis of bronchiolitis admitted to Bristol Children's Hospital Department of Paediatric Respiratory Medicine

Recruited: 21 infants (1 excluded with onset of apnoea, 1 excluded with fall in oxygen requirement)

Randomised: 19 infants, median age 3.0 months, range 0.3 to 11.3 months

Intervention group: 11 infants. Age in days: median 49 (range 8 to 334)

Control group: 8 infants. Age in days: median 125 (range 12 to 343)

Inclusion criteria: infants aged less than 12 months admitted to a general paediatric ward with a clinical diagnosis of bronchiolitis (cough, tachypnoea, chest retraction and crackles on auscultation); moderately severe disease defined as a head-box oxygen requirement of at least 35%, moderate to severe tachypnoea, increased effort of breathing and in whom feeding had been discontinued

Exclusion criteria: congenital cyanotic heart disease, repeated severe apnoeas or severe hypercapnia with acidosis on blood gas analysis of pH < 7.2

InterventionsHigh-flow nasal cannula therapy via Vapotherm 2000i device, with a 1 to 8 L/min paediatric cartridge and infant-sized nasal cannulae (VT), started at a flow of 4 L/min with 100% oxygen, at 37 °C and flow rate increased by 0.5 L every 5 minutes up to 8 L/min if tolerated to achieve target SpO2 for 24 hours. If clinically stable, FiO2 decreased in steps of 10% at 4-hourly intervals, with same target SpO2

Control: conventional therapy head-box oxygen (HBO)

OutcomesPrimary: SpO2 at 8 hours post randomisation

Heart rate, respiratory rate, blood pressure, FiO2, combined bronchiolitis severity score (maximum of 7)

Outcomes recorded: 4, 8, 12, 24, 36, 48 hours

Additional outcomes recorded:

- length of time to switch to dry oxygen

- length of time receiving oxygen therapy after randomisation

- total length of time of oxygen therapy

- time until enteral feeds were started, time to discharge

- total length of stay in hospital

Outcomes reported:

- SpO2 at 8 hours, FiO2 at 8 hours, SpO2 at 12 hours, FiO2 at 12 hours, SpO2 at 24 hours, FiO2 at 24 hours

- Time to dry O2 (hours)

- Total time in O2 (hours)

- Time to feed (hours)

- Time to discharge (hours)

- Total length of stay (hours)

NotesStudy protocol supplied by authors

To detect a difference between groups in the main outcome measure of one standard deviation, at a significance level of 5% and a power of 80%, the study required 16 infants in each group

Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDescribed as randomised; no information on method for randomisation schedule

Allocation concealment (selection bias)Low riskNumbered, sealed envelopes used for allocation

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo withdrawals noted

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported either in data form or text form

Other biasUnclear riskOne participant was changed from Vapotherm to head-box oxygen for clinical reasons, which was not expanded on. It was not specified if the infant not adhering to the allocated group was analyzed in the allocated group

Different weaning protocols may have affected the time to ceasing treatment measurements, in the direction of Vapotherm being longer

FiO2 was a titration and hence may affect SpO2. However, both data sets were recorded and reported for comparison Insufficient information to identify this as a source of bias

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskParticipants and personnel were not blinded due to the nature of the intervention. Personnel followed standardised protocols. Likelihood of lack of blinding influencing outcome was not reported and is unclear

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskOutcome was documented by unblinded nursing staff caring for participants. However, impact on bias is unclear given the objective nature of the outcome measurements

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Bajaj 2006High-flow nasal cannula therapy not an intervention arm

Balanzat 2006High-flow nasal cannula therapy not an intervention arm

Blyth 2003High-flow nasal cannula therapy not an intervention arm

Donlan 2011Review article

Figueruelo 2011Not RCT/out of age range

Hough 2011Not RCT

Kim 2011Not examining HFNC therapy versus other treatment – instead examining HFNC therapy with helium/oxygen and with oxygen alone

Martinon-Torres 2003Not RCT

Martinon-Torres 2008High-flow nasal cannula therapy not an intervention arm

Martinón-Torres 2002Review article

Milesi 2010High-flow nasal cannula therapy not an intervention arm

Smith 1993High-flow nasal cannula therapy not an intervention arm

Tie 2009High-flow nasal cannula therapy not an intervention arm

Unger 2008High-flow nasal cannula therapy not an intervention arm

Characteristics of ongoing studies [ordered by study ID]
Kepreotes 2012

Trial name or titlePilot randomised control trial of high­flow nasal prong warm, humidified oxygen (HFNP WHO) compared to standard oxygen therapy in the management of moderate bronchiolitis in infants aged < 24 months in the Emergency Department and Children's Medical Ward of a tertiary referral hospital in Australia

MethodsSingle site, open, 2-arm, pilot, randomised controlled trial of high­flow nasal prong with warm, humidified oxygen compared to standard oxygen therapy in the management of moderate bronchiolitis in infants aged < 24 months

ParticipantsInfants 0 to 24 months old

InterventionsHigh­flow, nasal prong with warmed, humidified oxygen plus supplemental oxygen if necessary

OutcomesPrimary: time to weaning off supplemental oxygen

Secondary: treatment failure, change in baseline vital observations and respiratory distress score, length of stay, adverse events, severity score, parents assessment, biomarkers and imaging

Starting dateJuly 2012


NotesACTRN12612000685819. Recruited n = 18 at 22 August 2012; anticipate completion of recruitment by end of 2013

Milner 2012

Trial name or titleHeated High Flow Oxygen Use in Infants With Bronchiolitis and Hypoxia (HHFNC) inpatients in Children's Hospitals and Clinics of Minnesota

MethodsProspective randomised study comparing clinical severity score and Pediatric Early Warning System scores of infants who receive oxygen by standard flow nasal cannula to those who receive oxygen via humidified, high­flow nasal cannula

ParticipantsChildren 3 to 18 months old with a clinical diagnosis of bronchiolitis and hypoxia, and a CSS showing moderate distress > 4

InterventionsOxygen delivered via humidified high-flow nasal cannula (HHFNC) or oxygen received by standard flow nasal cannula

OutcomesPrimary: clinical severity score

Secondary: Pediatric Early Warning System score

Starting dateAugust 2012

Contact Identifier: NCT01662544

Schibler 2013

Trial name or titleHigh Flow Nasal Cannula Treatment for Viral Bronchiolitis in Infants, a Randomised Controlled Trial

MethodsRandomised controlled trial to be conducted in New South Wales and Queensland, Australia. Randomisation on ED admission using central online randomisation allocation concealment

ParticipantsInfants with bronchiolitis less than 12 months of age

Oxygen requirement with SpO2 < 94% in room air

InterventionsHigh flow nasal cannula oxygen delivery at a rate of 2 L/kg/min for the duration of oxygen requirement

OutcomesPrimary: reduction in transfer rate from regional hospital to tertiary centre

Secondary: reduction in intubation rate, hospital length of stay

Starting date1 July 2013



Seear 2011

Trial name or titleStudy of High-flow Oxygen Therapy Against Standard Therapy in Bronchiolitis (Hi-Flo) 2 centre study

MethodsProspective, open, randomised study comparing standard ward management with high-flow nasal cannula oxygen therapy

Participants0 to 18 months old

InterventionsHigh-flow nasal cannula oxygen therapy

OutcomesPrimary: length of hospital stay

Secondary: ICU admission, work of breathing

Starting dateDecember 2011

Contact Identifier: NCT01498094. To be conducted in British Columbia Children's Hospital, Vancouver, British Columbia, Canada and Sydney Children's Hospital, New South Wales, Australia. At 23 August 2012 recruitment was not complete; possible target date January 2013

Sood 2012

Trial name or titleRandomised controlled IRB-approved prospective, multicenter study in USA

MethodsParticipants randomised into 1 of 3 groups from December 2010 through March 2011 at 2 children's hospitals

ParticipantsInfants admitted to PICU with the diagnosis of RSV bronchiolitis

Interventions1. Conventional nasal cannula (NC) with 100% oxygen and flow rate of < 2 litres/min (LPM)

2. High-flow high-humidity (HFHH) therapy with 30% oxygen, flow of 4 LPM

3. HFHH therapy with 30% oxygen, flow of 8 LPM

All other therapies identical between groups

OutcomesSerial data collected on pre/post therapy blood gases, respiratory rates (RR), validated work of breathing (WOB) scores, days on oxygen, length of hospital stay and treatment failures

Starting dateDecember 2010

Contact informationMark Rowin, Associate Professor of Pediatrics, University of Tennessee Medical College-Chattanooga

NotesData in abstract Sood 2012 represented the first year of a multi-year prospective study. 3 years of the study completed by May 2013, preliminary data analysis commencing in July 2013

Whitehall 2012

Trial name or titleTrial of high flow nasal prong (HFNP) oxygen therapy for viral bronchiolitis in children's wards of south­west Sydney, NSW, Australia

MethodsOpen, randomised controlled trial comparing high-flow nasal prong oxygen therapy with standard care on rate of intubation and requirements for intensive care in children with viral bronchiolitis

ParticipantsInfants 0 to 24 months old

InterventionsHigh-flow nasal prong therapy (HFNP) providing humidified oxygen and warmed to 37 degrees. Flow rate, set to optimise oxygen saturation, will be up to 8 L/min depending on patient comfort. Duration of HFNP will vary depending on severity of the bronchiolitis. Breaks for nasal care or mobilisation permitted. Other treatments such as drug therapy will be given according to standard protocols

Control: respiratory therapy presently available in the participating hospitals, i.e. supplemental oxygen up to 2 L/min provided by low­flow normal nasal prongs with unheated humidifier. Other treatments such as drug therapy will be given according to standard protocols

OutcomesPrimary: decrease in intubation and intensive care requirement

Secondary: reduction in transfer to children's hospital

Starting dateNovember 2011


NotesACTRN12611001016921. Not recruiting at 1 November 2011

Australian hospitals listed: Campbelltown Hospital and Mount Druitt Hospital, NSW