Abstract
The use of high flow nasal cannula (HFNC) in the treatment of bronchiolitis has markedly increased in the last decade, yet randomized controlled trials have reported little clinical benefit with early, routine use. This article provides a concise overview of the current status of HFNC therapy, discusses successful de-implementation strategies to curtail HFNC overuse, and explores future bronchiolitis and HFNC quality improvement and research considerations.
Where are we now? A Quick Summary of the Evidence
Viral bronchiolitis is the most common reason for hospitalization in infants younger than 12 months of age. Treatment is supportive, and may include addressing hydration needs and providing supplemental oxygen, including in the form of high flow nasal cannula (HFNC). Globally, bronchiolitis admission rates have been stable-to-decreasing in the past two decades . Invasive mechanical ventilation and mortality rates have been relatively stable, yet intensive care unit (ICU) admissions have increased dramatically, with estimated increases of up to 333 % between 2009 and 2019 . Noninvasive mechanical ventilation (NIV) rates have also skyrocketed, increasing by as much as 1450 % since the year 2000 , yet there is no evidence that increasing bronchiolitis severity is driving the increased use . Mirroring an increase in ICU utilization and rates of NIV, costs have also markedly increased .
Recent literature reports that over 50 % of mild-to-moderately ill bronchiolitis patients admitted in the United States, Australia and New Zealand receive HFNC . One proposed reason for the increase in ICU utilization, NIV rates, and costs is the widespread adoption of HFNC, which is reported to cost upwards 16-times more than standard LFNC . In the United States and Canada, HFNC is often subsumed under the category of NIV due to billing and coding challenges . There is wide variability in HFNC practices and current bronchiolitis guidelines in the United States, Canada, Australia/New Zealand, and the United Kingdom lack specific guidance on the use of HFNC in bronchiolitis.
HFNC utilization increased due to early observational studies that reported improved patient comfort and respiratory scores , as well as an association between HFNC use and reduced intubation rates . Three subsequent randomized controlled trials (RCTs) totalling over 2000 patients failed to identify improved patient outcomes with initial management with HFNC compared to LFNC . Specifically, there were similar rates of ICU transfer, oxygen duration, length of stay (LOS), and intubation rates. There was more “treatment failure” in the LFNC group, but the only outcome that treatment failure predicted was need to crossover to HFNC . A more recent study found not only no improvement in patient outcomes, but also reported prolonged LOS and more ICU transfers in the HFNC group . Despite these RCT findings, a recent Cochrane review reported a modest reduction in oxygen duration and LOS that favored HFNC over LFNC and found improvements in heart rate and respiratory rate in those randomized to HFNC, though the included articles were heterogenous and the authors note a high risk of bias.
Clinical application of the current literature suggests HFNC is most suitable as a rescue therapy for bronchiolitis patients who fail LFNC . Indeed, a systematic review by O’Brien et al. noted that the evidence from RCTs does not support initial treatment with HFNC in the emergency room or ward setting . Additionally, the use of HFNC strictly for work of breathing without hypoxia is not supported in the evidence.
Reducing overuse
The goal in bronchiolitis management is to improve patient outcomes, such as reducing intubation rates, ICU utilization, and LOS. A pattern has emerged in therapies directed towards bronchiolitis treatment whereby a promising intervention is rapidly adopted due to favorable early observational studies, only for RCTs and meta -analyses to find minimal benefit on the population level. Efforts then shift to deimplementation of the once promising therapy. This cyclical pattern has been observed with steroids, albuterol, and racemic epinephrine – and now with HFNC. Indeed, given the heterogeneity in HFNC practices, RCT evidence that routine HFNC does not consistently improve patient-centered outcomes, paucity of evidence-based guidelines, and marked increase in HFNC rates and associated costs, HFNC therapy in bronchiolitis is primed for efforts to standardize care and reduce overuse.
Initiation
Strategies to reduce HFNC initiation have included creating clinical pathways to standardize care , defining initiation criteria , pre-HFNC huddles , and recommending an initial trial with LFNC . One particularly successful single center QI initiative by Treasure et al. permitted HFNC initiation in afebrile patients with a clinical assessment of severe respiratory distress and at least one of the following criteria: oxygen saturation < 90 % on 3 or more liters LFNC, respiratory rate above 70 for more than 5 min, or altered mental status, persistent grunting, and/or apneic events. They also implemented pre-HFNC huddles to ensure adequate attention is paid to supportive care measures and disseminated an educational handout for families. They reported a sustained reduction in HFNC initiation from 41 % to 22 %, shortened LOS from 60 to 45 h, and a decrease in ICU transfers from 20 % to 8 % .
There is one large scale HFNC deimplementation effort entitled “High-flow Interventions to Facilitate Less Overuse” (HIFLO), spearheaded by the Value in Inpatient Pedatrics Network within the American Academy of Pediatrics . HIFLO recruited hospitals from the United States and Canada to participate in a QI collaborative focusing on either reducing HFNC initiation or reducing HFNC treatment duration, with no interventions targeting the alternate arm. The 31 hospitals in the initiation arm implemented a “Pause” protocol that delineated a series of supportive care steps that are recommended prior to commencing HFNC, including management of hydration, hunger, fever, pain, overstimulation, and consideration of a trial of LFNC. The HFNC initiation rate decreased from 44 % to 30 % (32 % relative reduction) during the intervention period, without a change in LOS. Hospitals participating in the weaning arm of the study had stable rates of HFNC initiation during the intervention period. Future HFNC deimplementation efforts should incorporate the successful strategies mentioned here and focus on efforts on identifying which patients may benefit from HFNC
Flow rates
Overuse may also occur when HFNC is dosed inappropriately. Physiologic studies report increasing positive end expiratory pressure with increasing flow rates, with maximum distending pressure seen in small infants whose mouths were closed . However, too much flow (3 Liters/kilogram/minute or more) may lead to patient discomfort, air leak, and prolonged LOS . Likewise, sub-optimal flow (1 Liter/kilogram/minute) has been associated with longer length of stay and more treatment failure . Based on physiologic studies and RCTs, the optimal flow rate is likely 1.5 – 2 Liters/kilogram/minute . While bronchiolitis pathways in the US historically set flow rates based upon patient age , that practice has been associated with increased ICU utilization when compared to dosing by weight . Accordingly, there has been a shift towards weight-based dosing , which has been the common practice in Europe and Australia. Future QI endeavors and bronchiolitis guidelines should recommend weight-based flow rates in an effort to reduce inappropriate HFNC use, ICU utilization, and LOS.
Weaning
Much like the variability in HFNC initiation practices and flow rates, there is also wide variability in HFNC weaning strategies . RCT protocols report varying methods for weaning HFNC, including in slow deccrements and rapid discontinuation once patients tolerate 21 % fraction of inspired oxygen (FiO 2 ) for 4 h . Likewise, QI strategies to reduce HFNC duration focus on rapid discontinuation and frequent weaning attempts. The QI study by Noelck et al. reduced HFNC treatment duration, LOS, and reduced sub-therapeutic HFNC use via daily LFNC or room air trials . Charvat et al. also decreased treatment duration and LOS by performing frequent HFNC “Holiday” trials (every 4 h) via a 50 % reduction in flow, transition to LFNC, or trial of room air .
In the multicenter HIFLO study , the weaning arm implemented a HFNC “Holiday” protocol that defined heart rate, FiO 2 , and saturation parameters at which a patient would be eligible for a rapid wean, with options of weaning flow by 50 %, trial LFNC, or trial room air. The 40 hospitals that participated in the Holiday arm decreased HFNC treatment duration from 51.3 to 36.9 h (28 % relative reduction) and almost no rapid deterioration following a weaning trial. Hospitals participating in the Pause arm, after adjusting for temporal trends, had no change in treatment duration. Future QI initiatives should include rapid discontinuation trials as they have proven to be safe and effective and would minimize the time patients are receiving subtherapeutic flow rates.
Challenges to deimplementation
Identified barriers to deimplementation of low value practices are largely psychological, including fear (of clinical deterioration), loss (of a popular and established intervention), and action bias (whereby clinicians prefer “to do something”) . Additional psychological barriers may include confirmation bias, availability heuristics, and the endowment effect . Gupta et al. interviewed 152 providers on their perception of HFNC use in bronchiolitis . Over half felt positively about deimplementation of HFNC, with significantly greater positivity in those knowledgeable of literature on bronchiolitis and HFNC. Noted barriers to deimplementation included discomfort with not intervening when a child was in distress, perception that HFNC helps, and differences in risk tolerance and clinical experience. Deimplementation efforts may attempt to mitigate these identified obstacles; for example, in the HIFLO study, QI interventions were designed to address psychological barriers, such as granting permission for non-conformism by offering various weaning strategies and offering supportive care measures and LFNC as substitute interventions instead of HFNC.
Future considerations
Current consensus guidelines in the US, Canada, Australia, New Zealand, and UK offer minimal guidance on use of HFNC in bronchiolitis . Future iterations should include more explicit recommendations regarding when and how to initiate and wean HFNC. Additionally, researchers should seek to identify the elusive patient population that will most benefit from HFNC and determine Achievable Benchmarks of Care (ABCs) for HFNC use in bronchiolitis to provide a motivational target of HFNC utilization to which hospitals may aspire.
Future policies may also provide recommendations on nutrition and hydration practices. There is wide variability in feeding practices and current guidelines recommend either nasogastric (NG) or peripheral intravenous supplemental hydration. Retrospective and observational literature suggests that enteral nutrition (by mouth or NG) is safe in patients receiving HFNC and may be associated with shorter LOS . Evidence also suggests that having a feeding protocol is associated with reduced time to enteral nutrition and shortened LOS . There is a current RCT underway evaluating whether bronchiolitis patients on HFNC may benefit from NG or nasoduodenal feedings , but RCT evidence comparing NG to by mouth feeds remains unstudied. More research is needed to determine the ideal approach to providing nutrition for patients receiving HFNC.
Another important consideration is how HFNC impacts patient-centered outcomes. A recent article by Lane reports that with increasing levels of medical support, parents report increasing difficulty holding, bonding with, and breast feeding their child . Overuse of HFNC, therefore, may unnecessarily contribute to negative patient (and family)-centered outcomes. Future bronchiolitis guidelines and QI efforts should seek to both reduce unnecessary HFNC use and minimize disruptions to parent-patient bonding.
Conclusions
High flow nasal cannula use in the management of acute bronchiolitis has been increasing in frequency over the last decade with associated increases in ICU utilization and total costs without convincing clinical benefit to support such widespread use. Next steps should focus on elucidating which patients actually benefit from HFNC, building evidenced-based consensus guidelines, and defining universal bronchiolitis benchmarks, with the goal of reducing HFNC overuse, ICU utilization, LOS, and hospital costs.
Directions for future research:
- –
Initiation and weaning protocol standardization
- –
Feeding protocol standardization
- –
Impact of high flow nasal cannula on patient and family experience
- –
Elucidation of which patients may benefit from high flow nasal cannula
Educational aims
The reader will be able to:
- –
Offer an overview of the current literature base of high flow nasal cannula in the treatment of bronchiolitis
- –
Critically evaluate prior quality improvement endeavors targeting high flow nasal cannula overuse, including quality projects targeting initiation and weaning
- –
Provide framework for future considerations, including creating Achievable Benchmarks, improved patient-parent bonding, and standardization of guidelines
Artificial Intelligence statement
No generative AI and AI-assisted technologies were used in the writing process
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
