Table 1 Overview of the 26 original clinical studies including children on HFNC beyond the newborn period

Bressan 2013

[18]

Prospective observational.

27 infants with bronchiolitis in a general paediatric ward.

Age <12 months.

Max 8 L/min.

Clinical parameters (end tidal Co2, respiratory rate, heart rate, SpO2).

Feasibility of HFNC (adverse events).

Decrease in median end tidal CO2 (6–8 mmHg) and respiratory rate (13–20 per minute) in the first 3 h of HFNC and remained steady thereafter.

No adverse events.

Arora

2012

[14]

Prospective observational.

25 infants with bronchiolitis in an emergency department.

Age <12 months.

1 L/min, increasing with 0.5 L/min until clinical improvement, max 8 L/min.

Pressure in nasopharynx at varying flow rates of HFNC.

Increasing flow rates of HFNC up to 6 L/min were associated with linear increase in nasopharyngeal pressure.

Kallappa 2014

[31]

Retrospective observational.

45 infants with bronchiolitis in a general paediatric ward.

Age <15 months.

Not given.

Clinical parameters (heart rate, respiratory rate, blood gas parameters).

Adverse events.

Decrease of heart rate (median 171 to 136) and respiratory rate (median 79 to 53) and improvement in Ph (median 7.32 to 7.38) and PaCO2 (median 7.7 to 6.6 kPa), within 4 h of initiating HFNC.

No adverse events.

Hilliard

2012

[24]

Prospective interventional randomized, unblinded.

19 infants with bronchiolitis in a general paediatric ward.

Infants were randomized to head-box oxygen (n = 8) or HFNC (n = 11).

Age <12 months.

4-8 L/min.

Safety and feasibility of HFNC in infants with bronchiolitis.

SpO2 8 h after randomization and other clinical parameters at intervals up to 48 h.

Median SpO2 was higher in the HFNC group at 8 and 12 h, but similar at 24 h.

FiO2 was higher in the HFNC group at all three time points.

Mayfield 2014

[16]

Observational case control.

Cases were identified prospectively and controls identified retrospectively.

61 infants with bronchiolitis treated with HFNC in a general paediatric ward.

33 infants with bronchiolitis treated with standard low flow oxygen.

Age <12 months.

2 L/kg/min.

Max 10 L/min.

Clinical data (heart rate, respiratory rate, SpO2, LOS) admission to PICU and adverse events

Nonresponders to HFNC can be identified early.

Four times higher risk of admission to PICU in the standard treatment group than in the HFNC group.

HFNC is safe (no adverse events).

Bueno Campãna 2014

[20]

Prospective randomized unblinded controlled.

75 infants with bronchiolitis in general paediatric ward.

32 children on HFNC and 42 children inhaling hypertonic saline

Age <6 months.

Max 8 L/min.

Respiratory distress (measured by scoring system), patient comfort, LOS, admission to PICU in the two groups.

HFNC was not superior to hypertonic saline in treatment of moderate acute bronchiolitis with respect to severity and comfort scores, LOS or PICU admission rate.

Milani

[33]

Prospective observational.

36 children hospitalised with bronchiolitis in an emergency department.

18 treated with HFNC,

18 with low-flow oxygen.

Age < 12 months.

8 L/kg * respiratory rate *0.3.

Respiratory rate, respiratory effort, ability to feed and LOS in the two groups.

Improvements in respiratory rate, respiratory effort and ability to feed were faster in the HFNC group.

The HFNC group needed oxygen for 2 days less and LOS was 3 days shorter than in the low flow oxygen group.

Abboud 2012

[19]

Retrospective observational.

113 children hospitalized with bronchiolitis in PICU.

Age ≤12 months.

3-8 L/min.

Characteristics of non-responders to HFNC measured by respiratory rate, blood gas parameters and SaO2.

Nonresponders were more hypercarbic, less tachypnic and had no change in their respiratory rate after initiation of HFNC.

Milési

2013

[13]

Prospective observational.

21 infants with RSV bronchiolitis in PICU.

Age <6 months.

1-7 L/min.

Pharyngeal pressure provided by HFCNC using flow rates from 1–7 L/min and the effect of HFNC on breathing pattern and respiratory effort.

HFNC with a flow rate equal to or above 2 L/kg/min generated a clinically relevant pharyngeal pressure ≥4 cm H₂O and improved breathing pattern.

Hough

2014

[17]

Prospective observational.

13 infants with bronchiolitis in PICU.

Age <12 months.

2 and 8 L/min.

Average rate 1.7 L/kg/min.

End-expiratory lung volume, continuous distending pressure and regional ventilation distribution by measuring electrical impedance tomography.

HFNC at 8 L/min increased end-expiratory lung volume and improved respiratory rate, FiO2 and SpO2 compared with standard flow of 2 L/min.

No adverse events.

McKiernan 2010

[21]

Retrospective observational cohort.

115 infants with bronchiolitis admitted to PICU during two seasons.

57 children before introduction of HFNC and 58 children after implementation of HFNC.

Age <24 months.

7-8 L/min.

Intubation rate in PICU after introduction of HFNC.

Clinical parameters (respiratory rate, LOS).

Intubation rate decreased from 23 % (2005–2006) to 9 % (2006–2007) after introduction of HFNC in the department.

After 1 h on HFNC, respiratory rate decreased (−12 breaths/min) in infants treated with HFNC.

Median LOS decreased from 6–4 days after introduction of HFNC.

Metge

2014

[22]

Retrospective observational.

34 children with bronchiolitis in PICU.

19 children on CPAP (first season) and 15 children on HFNC (second season).

Age <12 months.

1-3 L/kg/min, max 8 L/min.

LOS and other clinical parameters in children on CPAP and HFNC during two seasons.

No difference between the groups in length of stay, respiratory rate, PaCO2, FiO2 and duration of oxygen support.

Riese

[36]

Retrospective observational.

120 infants admitted with bronchiolitis to PICU before and 170 after introduction of HFNC in a general paediatric ward.

Age < 24 months.

<6 months:

2–8 L/min.

6–18 months:

4–12 L/min.

18–24 months:

8–15 L/min.

LOS, intubation rates, 30 days readmission and median hospital charges.

LOS in PICU was reduced from 4–3 days, no difference in intubation rate or readmission, the median total hospital charges was reduced.

Pham

2014

[30]

Prospective non-randomised interventional.

14 infants with bronchiolitis.

14 infants with congenital heart disease.

Admitted to PICU.

Age <12 months.

2 L/kg/min.

Diaphragmatic electrical activity and oesophageal pressure changes as a surrogate for work of breath in infants off then on HFNC.

The electrical activity of the diaphragm and the oesophageal pressure-swings in infants with bronchiolitis were reduced.

A similar, but less prominent offload of the diaphragm was observed in the cardiac infants.

Schibler 2011

[15]

Retrospective observational cohort.

298 infants admitted to PICU, 56 % had bronchiolitis.

Age <24 months.

8 L/min at initiation.

Ventilator practice in the 5-year period after the introduction of HFNC therapy.

Intubation rate.

Intubation rate decreased from 37 % in 2005 to 7 % in 2009 in infants with bronchiolitis corresponding with an increase in the use of HFNC.

Wing

2012

[29]

Retrospective observational case control.

848 patients divided in 3 cohorts admitted to PICU with acute respiratory insufficiency.

24 % had bronchiolitis.

Cohort 1 (n = 190): HFNC not available

Cohort 2 (n = 289): HFNC available, but no guidelines.

Cohort 3 (n = 369): HFNC and guidelines available.

Age 0–18 years.

Range 2–50 L/min.

Details not given.

The need of intubation and mechanical ventilation before and after the availability of HFNC.

Intubation rate decreased from 16 to 8 % after the implementation of HFNC in PICU.

No significant change in mortality or median PICU length of stay.

Rubin

2014

[23]

Prospective observational cohort.

25 patients in ICU receiving HFNC or planned to be extubated to HFNC.

Age < 18 years.

2-8 L/min.

Effort of breathing in children on CPAP and HFNC at different flow rates by measuring the pressure-rate product (change in pleural pressure multiplied by respiratory rate).

Oesophageal pressure was used as a surrogate for pleural pressure.

Increasing flow rates (2, 5 and 8 L/min) of HFNC decreased the pressure-rate product and increased the baseline pleural pressure.

ten Brink 2013

[27]

Prospective observational.

109 children in PICU requiring respiratory support for various disease categories;

72 children on HFNC and 37 on CPAP.

HFNC: median age 6 months, CPAP: median age 5 months.

2 L/kg/min.

Level of and duration of respiratory support, and other clinical data in children on HFNC and CPAP.

No significant difference in the number of children requiring a higher level of respiratory support in the two groups. ¼ of all children on HFNC required higher level of respiratory support, these had failure of normalization of heart rate and respiratory rate and not fall in FiO2 after 2 h on HFNC.

Testa

2014

[28]

Prospective interventional randomized unblinded.

89 paediatric cardiac surgical patients in PICU.

Infants were randomized to conventional O2 therapy (n = 46) or HFNC (n = 43).

Age <18 months.

2 L/kg/min.

Clinical characteristics and need for higher respiratory support and reintubation rate.

48 h observational time.

PaCo2 did not differ between the group with HFNC and conventional O2 therapy.

PaO2 was higher in the HFNC group.

No difference in reintubation rate.

Spentzas 2009

[26]

Prospective observational.

46 neonates and children treated for respiratory distress in PICU.

Patients were switched from traditional oxygen therapy to HFNC.

Age 0–12 years.

8-12 L/min in infants.

20–30 L/min in children.

Tolerability and effectiveness of HFNC treatment using COMFORT scale and nasopharyngeal pressure.

COMFORT score and oxygen saturation improved in children after switching to HFNC.

HFNC generated a positive end expiratory pressure of 4 ± 1.99 cm H₂O; the pressure was dependent of weight and flow rate.

Kelly

2013

[34]

Retrospective observational.

498 children admitted to paediatric emergency department with respiratory distress, 46 % had bronchiolitis.

Age < 2 years.

Not given.

Clinical and patient characteristics that predicts success or failure of HFNC therapy.

Respiratory rate > 90th percentile for age, initial venous PaCO2 > 50 mmHg, and initial venous pH < 7.30 were associated with failure of HFNC therapy.

A diagnosis of acute bronchiolitis was protective with respect to intubation following HFNC.

Wraight

2015

[32]

Retrospective observational.

54 children hospitalized in PICU for various respiratory disorders.

79 % with bronchiolitis.

Median age 3.5 months.

2 L/kg/min.

Failure of HFNC therapy defined as the patient needing escalation of treatment to CPAP or intubation.

HFNC was successful in 78 % of patients and failed for 12 patients (7 needed CPAP and 5 were intubated).

The failure rate was 50 % in children with a primary diagnosis of congenital heart disease.

Long

[35]

Prospective observational.

71 children hospitalized with various respiratory distress in emergency department.

Median age 9 months.

2 L/kg/min up to 10 kg, 0.5 L/kg/min thereafter.

Failure rate, predictors of failure and adverse events.

28 (39 %) children required escalation to a higher level of respiratory support. No serious adverse events in emergency department, but one child developed air leak syndrome after transfer to ICU.

Chisti

[37]

Open randomised controlled.

Children with severe pneumonia; randomised to CPAP, HFNC, or low-flow oxygen.

<5 years of age.

2 L/kg/min, max 12 L/min.

Treatment failure after 1 h.

Oxygen therapy delivered by CPAP improved outcomes compared to low flow-oxygen, no difference between HFNC and CPAP group.

McGinley 2009

[25]

Prospective observational.

12 children with obstructive apnoea-hypopnea syndrome in a paediatric sleep disorder centre.

10 patients had undergone CPAP titration before study start.

Age 10 ± 1 year.

20 L/min.

Numbers of obstructive sleep apnoea, clinical parameters (respiratory rate, arousals).

HFNC reduced the inspiratory flow limitation and decreased respiratory rate.

HFNC decreased arousals and apnoea hypopnoea index comparable to CPAP.

Joseph

[53]

Retrospective observational.

5 children with obstructive sleep apnoea not tolerating CPAP.

Age < 18 years.

≤10 L/min.

Change in apnoea-hypopnoea index and oxygen saturation.

Treatment with HFNC improved the apnoea-hypopnoea index and increased oxygen saturation.