In summary, for infants with bronchiolitis treated with high-flow oxygen therapy:

Benefits in NNT

  • 11% lower risk of escalation of care
  • No benefit in terms of duration of hospitalization or oxygen therapy
    9 for preventing escalation of care

Harms in NNT

    No difference between the groups for adverse events
    No difference between the groups for adverse events

View As:

Source: Franklin D, Babl FE, Schlapbach LJ, et al. A randomized trial of high-flow oxygen therapy in infants with bronchiolitis. N Engl J Med 2018;378:1121–31.

Study Population: 1,472 infants younger than 12 months with signs of bronchiolitis with oxygen requirement

Efficacy Endpoints: Treatment failure (requiring escalation of care), admission to intensive care unit, duration of hospital stay, the duration of intensive care unit stay, duration of oxygen therapy, intubation rates

Harm Endpoints: Serious adverse events including pneumothorax, respiratory arrest, cardiac arrest, apnea, emergency intubation

Narrative: Bronchiolitis is the most common reason for hospitalization in infants worldwide.1 Current recommendations by the American Academy of Pediatrics are for supportive care including maintenance of hydration and oxygen support for hypoxemia.1 Other interventions such as the use of bronchodilators have failed to show any benefit when compared to supportive care alone. However, it has been proposed that the obstructive process of bronchiolitis that causes increased work of breathing, hypoxia, and hypercapnea might respond to the moderate positive pressure provided by high-flow oxygen therapy.2

The randomized control trial referenced here was conducted in Australia and New Zealand across multiple institutions on otherwise healthy infants (less than 12 months old) with bronchiolitis with an oxygen requirement.3 For the purposes of the study, oxygen requirement was defined as the need for supplemental oxygen to maintain oxygen levels between 92% and 98% (11 institutions used site-specific standard of 94%–98%).3 Patients were randomized to heated and humidified high-flow oxygen at a rate of 2 L/kilogram body weight/min delivered by the Optiflow system with the use of an age-appropriate Optiflow Junior cannula and the Airvo 2 high-flow system (intervention group) or supplemental oxygen through a nasal cannula, up to a maximum of 2 L/min, to maintain an oxygen-saturation level in the range of 92% to 98% (control group).3

Treatment failure was defined as the need for escalation of care based on standardized clinical criteria: persistent or worsening tachycardia, tachypnea, worsening of hypoxemia requiring >40% FiO2 in the high-flow oxygen group and >2 L/min flow rate of nasal cannula in the standard therapy group. Each hospital was allowed to use its own escalation protocol to be used as the criteria for treatment failure. Each episode of escalation of care was reviewed to ensure that it met study criteria. Escalation of care in the standard oxygen group was recommended to switch each patient to high-flow therapy.

The trial showed a 11% absolute risk reduction in the need for escalation of care in patients receiving high-flow oxygen therapy (relative risk = 0.52, 95% confidence interval = 0.40–0.66; NNT = 9). This trial did not show any significant difference between the groups for other outcomes such as duration of hospital or ICU stay and intubation rates (although a very small percentage of patients [12/1,472] required intubation).3

High-flow oxygen therapy did not result in any significant increase in the risk of adverse events, although the rate of adverse events was very low in both groups and no patients in any of the groups required emergency intubation and cardiopulmonary resuscitation. One child in each group was diagnosed with pneumothorax but none required thoracostomy.3

Similar results were found by another recent smaller trial that reported an absolute risk reduction of 9% in treatment failure rate (NNT = 11) in patients allocated to high-flow oxygen therapy but no statistically significant difference between the groups for time to oxygen weaning or length of stay. The rates of adverse events were similar between the two groups in this trial as well.4

Caveats: This is the largest randomized trial to date addressing this important research question.3 The major limitation of this trial was the absence of blinding, which was not possible due to difference between the equipment. To reduce the risk of bias, the investigators remained blinded to the trial outcome until the trial was completed.

The primary outcome of this trial was treatment failure defined as requiring escalation of care. This was a composite outcome which reflected admission to a higher level of care or changing from low-flow oxygen to high-flow oxygen therapy (control group) and may not be considered a patient-centered outcome. In addition, determining this outcome was somewhat subjective. Analyzing individual patient-centered outcomes such as length of hospital or ICU stay and intubation rate did not show any benefits from using high-flow oxygen therapy. It must be noted that according to the Australian New Zealand Clinical Trials Registry, the initial primary outcome of the trial was reduction in transfer rate from regional hospital to tertiary center. This outcome was changed after inclusion of tertiary centers since this outcome would not be applicable anymore for patients who present directly to a tertiary emergency department (ED).5

While the overall rate of treatment failure and the need for escalation of care was lower in patients allocated to high-flow oxygen therapy, when the high-flow group was divided by hospital with an on-site pediatric intensive care unit (PICU) versus no PICU, the escalation rate was significantly higher in hospitals with an on-site PICU (14% vs. 7%). Therefore, availability of an on-site PICU could be an important factor in escalation of care by treating physicians.3

It is notable to mention that 61% of the patients in the standard therapy group who experienced treatment failure were transitioned to high-flow oxygen therapy and responded positively.3 High-flow oxygen therapy may potentially have the highest overall benefit in hospitals without an intensive care unit as it may decrease the need for interfacility transfers.

Another limitation of the reported data is that 34% of all patients that had escalation of care did not meet the criteria for escalation of care based on the study criteria but met the individual hospitals escalation criteria. This can present some confounding when looking at treatment failure between the groups.3

It must be noted that the trial did not control for the effect of high-flow oxygen therapy itself as a main factor for the need for higher level of care. Assignment to high-flow oxygens above 2 L/kg might have prompted certain physicians to escalate the level of care for closer observation and higher demands for nursing care.

The trial discussed in this review did not exclusively enroll patients in the ED.3 Patient enrollment occurred both in the ED and on the pediatric wards. Therefore, a trial originated exclusively in the ED might produce different results. Er et al.6 explored the characteristics of ED patients with bronchiolitis who respond poorly to high-flow oxygen therapy. These investigators concluded that low initial oxygen saturation, respiratory acidosis, and an oxygen saturation/ fraction of inspired oxygen ratio less than 195 at the first hours of treatment were related to unresponsiveness to high-flow oxygen therapy in the pediatric ED.5

Unfortunately, this trial does not evaluate the cost effectiveness of high-flow oxygen therapy. Other published trials have suggested cost saving benefits from using high-flow oxygen therapy.4, 7, 8, 9 Kepreotes et al.4 discussed the estimated cost savings with the use of high-flow oxygen therapy and concluded that high-flow oxygen therapy might have a role as a rescue therapy to reduce the proportion of children requiring high cost intensive care. Heikkilä et al.,7 performed a cost analysis of high-flow oxygen therapy versus standard oxygen therapy and found that using high-flow oxygen therapy was associated with a $441 saving per patient due to decreases in ICU admission and hospital transfers. Finally, this trial used pulse oximetry levels of 92% to 98% (94%–98% in specific institutions) to evaluate response to therapy while the American Academy of Pediatrics recommends initiation of oxygen therapy at pulse oximetry levels of 90% or below.10

In conclusion, high-flow oxygen therapy in infants with bronchiolitis reduces the risk of treatment failure and the need for escalation of care. However, it does not offer any benefit as far as direct patient-centered outcomes are concerned. Therefore, we assign a color recommendation of yellow (unclear benefits) to this intervention. However, this trial still has clinical implications. It appears that for patients with bronchiolitis who do not respond to low-flow oxygen therapy (first line of therapy) based on criteria used in this trial or other institutional criteria, high-flow oxygen therapy should be considered as the next logical step before employing other more aggressive measures.

The original manuscript was published in Academic Emergency Medicine as part of the partnership between and AEM.

Author: Isaac Gordon, MD; Ambreen S. Khan, MD
Supervising Editor: Kabir Yadav, MD

Published/Updated: June 14, 2019

  1. The Title Bar

    The title bar is color-coded with our overall recommendation.

    • Green: Benefits outweigh risks.
    • Yellow: Unclear risk/benefit profile.
    • Red: Benefits do not outweigh risks.
    • Black: Obvious harms, no clear benefits.
  2. Tip content...