High‐flow nasal cannula oxygen therapy for children with bronchiolitis: Implementation of a national guideline

Abstract

Aim

High flow nasal cannula (HFNC) therapy is a form of respiratory support used in children with bronchiolitis. A national guideline for the use of HFNC was published in The Netherlands in 2020. We studied the implementation and use of this guideline.

Methods

We performed a multicentre observational study amongst all hospitals in the North‐West part of The Netherlands referring to the same paediatric intensive care unit (PICU). This study consisted of two parts: a comparison of local HFNC protocols to the national guideline and a survey about the use of HFNC amongst paediatricians in the participating centres.

Results

We observed considerable variations between the local protocols and the national protocol, especially regarding criteria to initiate HFNC treatment and weaning practices. Survey results showed that expectations of HFNC widely varied, while the clinical use of HFNC deviated from both the national guideline as well as local protocols, especially for weaning practices and the use of pCO₂ as a parameter for initiation and evaluation of the effect of HFNC.

Conclusion

Implementation of the national guideline for HFNC therapy in bronchiolitis was inefficacious, leading to non‐uniform clinical practice.

Abbreviations

HFNC

high‐flow nasal cannula therapy

PICU

paediatric intensive care unit

RSV

respiratory syncytial virus

Key Notes

• In The Netherlands, a national guideline regarding indications and use of high flow nasal cannula (HFNC) therapy in bronchiolitis was published in 2020.

• We report considerable variation in local HFNC protocols as well as in the clinical use of HFNC despite the implementation of this national guideline

• To overcome inefficacious guideline implementation contributing to non‐uniform clinical practice, exploration of the facilitators and barriers for guideline implementation is needed.

1. INTRODUCTION

High‐flow nasal cannula (HFNC) therapy is a system to administer humidified and heated oxygen through a nasal cannula with relatively high flow (>2 L/min). HFNC is widely used in children with respiratory syncytial virus (RSV) bronchiolitis. Evidence for the beneficial effect of HFNC for infants with bronchiolitis is limited, but it appears to improve the work of breathing and decrease heart rate and respiratory rate. ¹ However, HFNC does not prevent invasive mechanical ventilation or paediatric intensive care unit (PICU) admission ¹ and studies did not demonstrate a decrease in length of stay or duration of oxygen suppletion for HFNC as compared to standard low‐flow oxygen. ² , ³ , ⁴ , ⁵ , ⁶ Importantly, evidence or clear recommendations on when to start or how to wean HFNC are lacking. ⁷ , ⁸ No publicly available, international guidelines for the use of HFNC in bronchiolitis have been published since its introduction. Consequently, a large variety between hospitals with regard to the use of HFNC on oxygen flow rate, patient assessment, weaning and feeding practices have been reported. ⁷

To overcome these aforementioned variations in clinical practice the Dutch Society for Paediatrics, endorsed to reach consensus on indications of safe HFNC use in the general paediatric ward for children with bronchiolitis. This resulted in a national guideline that was published in 2020. ⁹ In this study, we investigate the implementation and use in clinical practice of this guideline in children with bronchiolitis amongst all hospitals in the North‐West part of The Netherlands.

2. METHODS

2.1. Study design and participants

The study consisted of two parts. First, we performed a multicentre observational study by comparing local HFNC protocols to the national guideline. Second, we performed a multicentre survey study, which was sent to all paediatricians in the participating hospitals. All ten regional hospitals in the North Western region of The Netherlands (Tergooi MC, BovenIJ Hospital, Amstelland Hospital, Noordwest Hospital Group, OLVG, Spaarne Hospital, Zaans Medical Center, Flevo Hospital, Rode Kruis Hospital and Dijklander Hospital) and the medium‐care ward of the Amsterdam University Medical Centre (Amsterdam UMC) participated in the first part of the study. Paediatricians of nine hospitals participated in the survey. All participating hospitals refer critically ill patients to the PICU of the Amsterdam UMC.

2.2. Dutch guideline on HFNC

The national guideline ‘high flow in children’, published in 2020, describes and advises on the use of HFNC for children admitted to a general paediatric ward with imminent respiratory insufficiency. ⁹ The guideline recommends reserving the use of HFNC for bronchiolitis only. In the guideline, bronchiolitis is considered a clinical diagnosis of a (viral) lower respiratory infection in young children up to 2 years old. The guideline is structured in several subheadings including indications to start HFNC, how to monitor the effect of HFNC, when to consult with a PICU, how to use HFNC during transportation of a patient and how to wean from and stop HFNC.

2.2.1. Indications to start

HFNC is considered as a safe option for respiratory support in case of imminent respiratory failure in bronchiolitis. The guideline advises to consider starting HFNC in case of increasing work of breathing under standard therapy or if the need for supplemental oxygen through nasal prongs exceeds 2 L/min with 100% oxygen.

2.2.2. Monitoring and consulting PICU

According to the guideline, the effectiveness of HFNC should be evaluated at (but not restricted to) 30, 60, 90 and 120 min after the start. Respiratory rate, heart rate, oxygen saturation, FiO₂ and work of breathing should be monitored. ‘Worried‐signs’ of parents and nurses need to be taken into account. Transcutaneous oxygen saturation levels of at least 94% should be reached. Measurement of pCO₂ is not directly advised to avoid patient distress, but can be considered in case of suspected therapy failure, based on other parameters.

Consultation with a PICU should be considered when there is no clear improvement of respiratory rate or heart rate, oxygen saturation, administered amount of oxygen or work of breathing within 60 min or if the FiO₂ of more than 0.6 is necessary to reach a percutaneous measured oxygen saturation of at least 94%.

2.2.3. Transport

The guideline does not provide a clear advice for the transportation to the PICU of children supported by HFNC therapy with imminent respiratory failure due to bronchiolitis. Invasive mechanical ventilation should be started in patients who fail HFNC therapy before the arrival of a PICU team to prevent any further delay. However, in patients who do not improve within 60 min after the start of HFNC therapy but also do not need immediate invasive respiratory therapy, support during transportation remains an individual decision made by the local team and consulting PICU.

2.2.4. Weaning strategy

The guideline advises a weaning protocol based on expert opinion. The guideline advises to first reduce FiO₂ every 2–4 h with 0.1, after the patient is stabilised for more than 12 h. At a FiO₂ level of 0.4, the flow can be reduced. Recommendations are to immediately reduce the flow to 2 L/min with a FiO₂ 1.0 on the HFNC machine instead of gradually reducing the flow. If the patient remains stable for another 4 h, the respiratory support can be changed from high flow to standard low flow.

2.3. Data collection

For the first part of the study, all hospitals provided their local HFNC protocol for bronchiolitis through an appointed contact person. Local protocols were compared with the national guideline. Items that were examined included indications for the start of HFNC, how to monitor the effect of HFNC, when to consult with a PICU, how to use of HFNC during transportation of a patient and advice on how to wean from and stop HFNC.

The second part of the study consisted of a survey that was shared with the paediatricians of the participating centres (Appendix S1). The survey contained questions concerning the use of HFNC in patients with bronchiolitis and contained questions regarding the demographic characteristics of the respondent and the included hospital, expectations regarding the use of HFNC, the indications to start HFNC, the monitoring of the effect of HFNC, reasons for consulting with a PICU, use of HFNC during patient transport and finally questions regarding weaning strategies. The survey was constructed and sent via Castor EDC. The survey was first sent out on 2 April 2024 and data extraction took place on 10 June 2024.

3. RESULTS

Ten of the eleven contacted hospitals used HFNC for children admitted with bronchiolitis and provided their local protocol.

3.1. Dutch national guideline versus local protocols

3.1.1. Indications to start

Two protocols adhered to the recommendation of the national guideline to reserve HFNC only for children with bronchiolitis. According to eight protocols HFNC could also be used for other diagnoses (Table 1). Six protocols provided similar thresholds for HFNC as stated in the national guideline, namely: severe work of breathing and/or oxygen support of more than 2 L/min with 100% oxygen. In four protocols the clinical indications of when to start HFNC were not specified.

TABLE 1.

Overview of HFNC protocols compared to national guidelines.

Hospital	1	2	3	4	5	6	7	8	9	10	Dutch Guideline
Diagnosis
Bronchiolitis	√	√	√	√	√		√	√	√	√	√
Pneumonia		√	√	√			√		√	√
Asthma		√	√	√			√			√
Post‐extubation			√	√	√		√
Muscle weakness		√			√		√		√
Mucus		√					√
Not specified						√
Indications for start
Increased wob	√		√	√	√			√		√	√
>2 L 100%	√		√	√	√			√		√	√
Apnea							√
Not specified		√				√			√
Consulting PICU
At start HFNC			√
No improvement after 60 min a	√	√ b	√	√ b			√	√	√	√	√
No improvement after 120 min a					√ b
FiO2 >0.5									√
FiO2 >0.6			√			√	√	√			√

Abbreviation: Wob, work of breathing.

^a Improvement of breathing frequency, hart rate, oxygen saturation, FiO₂ needed and work of breathing.

^b ‘Improvement’ specified as 20% reduction in heart rate and breathing frequency.

3.1.2. Monitoring and consulting PICU

Seven out of ten protocols included a monitoring advice that is similar to the National guideline. In eight protocols PICU consultation was recommended in case there is no improvement (of respiratory rate, heart rate, oxygen saturation, FiO₂ needed and work of breathing) 60 min after HFNC initiation and in five protocols this was also recommended when necessary FiO₂ exceeds more than 0.5 or 0.6. ‘Improvement’ of vitals was specified as a decrease in respiratory rate or heart rate of at least 20% in three protocols.

3.1.3. Transport

The modality of respiratory support during transportation was not mentioned in five of ten protocols. Two protocols suggested patient transportation using HFNC as the first choice of respiratory support. In three protocols a trial period on the ward with the non‐rebreathing mask before transport was suggested, following NRM support when possible or otherwise intubation before transport as recommended in the National guideline.

3.1.4. Weaning strategy

In six out of ten local protocols weaning strategies for oxygen and flow fully adhered to the national guideline. Four local protocols recommended to gradually reduce flow and oxygen, without recommendations on which parameters to consider.

3.2. Survey

3.2.1. Participants

The survey was sent out to 109 paediatricians in nine different hospitals of whom 64 (59%) completed the survey. The majority of the respondents were paediatricians with more than 10 years of experience (n = 54, 84%) and respondents estimated to have more than 20 patients admitted at their hospital yearly (n = 62, 97%). Full results are provided in Appendix S2.

3.2.2. Expectations of HFNC and indications to start

Expectations of the respondents were an improvement in patient comfort after initiating HFNC therapy (n = 63, 97%), an improvement in respiratory rate and/or work of breathing (n = 60, 94%), possibly prevention of PICU admission (n = 43, 67%), prevention of endotracheal intubation (n = 40, 63%) and decrease in the length of hospital stay (n = 9, 14%). None of the respondents started HFNC therapy as the first modality of support in children admitted for bronchiolitis, but only in case of increasing respiratory effort (rate or work of breathing) (n = 33, 51%) or mainly based on clinical assessment, without an objective scoring system (n = 26, 79%) (Appendix S3).

3.2.3. Monitoring and consulting PICU

A decrease in work of breathing (n = 63, 98%), respiratory rate (n = 55, 86%) and hypercapnia (n = 51, 80%) were considered relevant parameters for success by most respondents. Most respondents monitored HFNC treatment success 60 minutes after start therapy (n = 50, 78%). Reasons to consult the PICU were mostly no clear decrease in work of breathing (n = 50, 78%), FiO₂ >0.6 to maintain oxygen levels above 94% (n = 51, 80%) and persisting hypercapnia (n = 45, 70%).

3.2.4. Transport

Respiratory support during transport was mainly done with the patients on HFNC (n = 31, 48%).

3.2.5. Weaning strategy

Most respondents indicated that they had access to weaning recommendations in their protocol (n = 57, 89%). However, only 18 (28%) respondents replied that their local protocol corresponds with the national guideline. A vast majority (n = 58, 91%) indicated to start weaning the flow when FiO₂ was at 0.4. From them, 40 of 58 respondents switched immediately to 2 L 100%, while others used various approaches (Figure 1). Importantly, many clinicians indicated a nurse‐based approach in weaning strategies or gradually reduced flow and oxygen.

FIGURE 1.

Weaning strategies for HFNC treatment in children with bronchiolitis.

4. DISCUSSION

We studied the implementation and use of the national guideline for HFNC use in children with bronchiolitis across 11 hospitals in the North‐West region of the Netherlands. We found that the implementation of the national guideline was ineffective, as nine hospitals had local protocols that did not fully adhere to national guideline recommendations. Additionally, significant variation in HFNC use was observed in clinical practice, with deviations from both national guidelines and local protocol recommendations, particularly in treatment expectations, starting criteria and weaning practices.

Currently, no international guidelines for HFNC therapy in infants with bronchiolitis are publicly available, which may partly be due to inconsistent evidence supporting its benefit. The Dutch national guideline is a crucial first step in standardising HFNC treatment for bronchiolitis. Despite the gap in knowledge about when to start HFNC therapy, the guideline offers recommendations for clinical decision‐making. However, almost half of the local protocols lacked specified recommendations, leading to various initiation criteria being reported by clinicians. The reasons for these deviations from the guideline remain unclear but may stem from a lack of clarity in the national guideline, such as what constitutes significantly increased respiratory effort. This is supported by our survey, where most respondents initiated HFNC based on clinical assessment without using an objective scoring system.

All local protocols followed the guideline's recommendation to consult a PICU if no clinical improvement was observed within 60 min of starting HFNC treatment. Most clinicians adhered to this recommendation and evaluated HFNC treatment within that timeframe. While local protocols largely followed guideline recommendations on parameters for assessing treatment success, many clinicians also considered severe hypercapnia when deciding whether to consult a PICU. This deviation may result from challenges in assessing a young child's clinical status, especially in rapidly changing conditions influenced by external factors such as discomfort or interaction with caregivers.

More than half of the respondents expected HFNC to prevent PICU admission or the initiation of invasive mechanical ventilation (MV), aligning with a US study where 52% of healthcare providers believed HFNC could prevent PICU admission and 30%–40% of health care providers believe to decrease length of stay or duration of oxygen supplementation. ¹⁰ However, studies have not consistently shown a reduction in PICU admissions since the introduction of HFNC. As HFNC does not alter the natural course of disease in bronchiolitis ¹¹ and as studies have not yet consistently shown a decrease in PICU admissions since HFNC introduction, ¹⁰ , ¹² , ¹³ clear evidence‐based supported criteria on HFNC treatment evaluation and PICU consultation are crucial to improve guideline adherence.

Adherence to guideline recommendations for weaning could also be improved, as these were not implemented in clinical practice in at least 40% of cases. Many clinicians reported using a more gradual oxygen‐weaning approach, relying on bedside nurse assessments. Slow oxygen weaning is common in paediatrics, ¹⁴ and our findings suggest that this practice, along with low adherence to HFNC weaning protocols, may contribute to longer oxygen therapy duration and hospital stays. This aligns with other studies indicating that routine HFNC use in bronchiolitis is associated with extended treatment durations. ² , ⁸ , ¹⁴ The current study only evaluated the implementation and adherence of the national guideline as compared to the local protocols. Evaluation of the effect of these protocols on clinical outcomes including length of stay, failure rate, intensive care admission and more, is the next step but requires a different research approach which was beyond the scope of this study.

Our study highlights that successful guideline implementation is not guaranteed. Low adherence can lead to variations in clinical practice, resource waste and suboptimal patient outcomes. ¹⁵ While we did not study the specific causes for the poor implementation of the Dutch guideline, potential factors may include a lack of guideline clarity, practical concerns or user‐specific barriers such as uncertainty about the beneficial effects of HFNC. Previous research has identified textual or substantive uncertainties as major obstacles to guideline implementation. ¹⁶ In the HFNC guideline, unspecified clinical criteria for starting or stopping HFNC treatment may represent such uncertainties. Other barriers could include uncertainty around weaning practices and reliance on nursing staff for bedside assessments. ⁸ Doctors' uncertainty about the capacity of bedside nursing staff to adequately assess a patients' vital signs has previously been pointed out as an important barrier for the implementation of clinical bedside tools. ⁷ , ¹⁶ As guidelines should provide evidenced based, easily applicable recommendations that overcome such possible implementation barriers, ¹⁵ we suggest that research efforts should not conclude providing evidence for a device, but continue with studies on how to achieve successful implementation or even initiate de‐implementation if the device does not improve clinical outcomes.

We conclude that the implementation of the national guideline for HFNC support in bronchiolitis can be improved. A revised national guideline needs to include evidence‐based recommendations for the use, expectations and evaluation for HFNC therapy. As guidelines should provide evidenced based, easily applicable recommendations, successful HFNC guideline implementation requires a thorough exploration of the facilitators and barriers for HFNC in infants with bronchiolitis.

AUTHOR CONTRIBUTIONS

Nike Beckeringh: Investigation; writing – original draft; methodology; validation; formal analysis; data curation. Rosalie S. N. Linssen: Investigation; writing – original draft; methodology; validation; formal analysis; data curation. Berber Kapitein: Supervision; writing – review and editing. Job B. M. van Woensel: Supervision; writing – review and editing. Frans B. Plötz: Conceptualization; writing – review and editing; methodology; supervision.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

CONSENT

Not applicable.

ETHICS STATEMENT

Not applicable, no patient data or materials were used in this study.

Supporting information

Appendix S1.

Appendix S2.

Appendix S3.

Beckeringh N, Linssen RSN, Kapitein B, van Woensel JBM, Plötz FB. High‐flow nasal cannula oxygen therapy for children with bronchiolitis: Implementation of a national guideline. Acta Paediatr. 2025;114:1291–1297. 10.1111/apa.17566