High-Flow Nasal Cannula in Pediatric Respiratory Distress: What the Evidence Shows
Synthesising data from 9 RCTs, 3 systematic reviews, and the 2024 Cochrane analysis on HFNC in children — examining indications across bronchiolitis, pneumonia, and asthma, optimal flow rates, efficacy versus standard oxygen and CPAP, and evidence-based escalation thresholds. Evidence through 2025.
This is an original evidence synthesis. See References for source studies.
One-Minute Takeaway
One-Minute Takeaway
- High-flow nasal cannula in pediatric respiratory distress significantly reduces treatment failure compared with standard oxygen therapy in bronchiolitis: the landmark Franklin et al. NEJM trial (n=1,472) showed 12% escalation with HFNC versus 23% with standard therapy (risk difference −11 percentage points; 95% CI −15 to −7; p<0.001), with 61% of standard therapy failures rescued by HFNC crossover.
- Despite reducing treatment failure rates, HFNC has not been shown to reduce ICU admission rates, hospital length of stay, or duration of oxygen therapy in any bronchiolitis RCT — positioning it as an effective rescue/escalation therapy rather than a disease-modifying intervention.
- The 2024 Cochrane review of HFNC in bronchiolitis confirms that HFNC reduces treatment failure compared to standard oxygen (RR 0.50, 95% CI 0.40–0.62) but results in higher failure rates when compared to nasal CPAP (RR 1.61, 95% CI 1.06–2.42), establishing HFNC as an intermediate-level support between standard oxygen and CPAP.
- A flow rate of 2 L/kg/min is the most evidence-supported setting, generating modest positive airway pressure (2–4 cmH₂O), reducing dead space, and decreasing work of breathing. Higher flows (3 L/kg/min) have not demonstrated superior clinical outcomes and may increase the risk of pneumothorax.
- Escalation predictors validated in prospective studies include persistent tachycardia and tachypnoea despite 1–2 hours of HFNC, SpO₂/FiO₂ ratio below 238 at 4 hours, and a ROX index below 6.5–7.2 at 12 hours — offering clinicians objective thresholds to guide timely escalation to CPAP or intubation.
Why This Topic Matters
Respiratory distress accounts for the majority of non-elective paediatric hospital admissions, with bronchiolitis alone representing the leading cause of hospitalisation in the first year of life. Despite decades of research, no pharmacological intervention has been shown to modify the disease course of bronchiolitis — treatment remains fundamentally supportive. Within this supportive framework, the delivery of respiratory support has been transformed over the past decade by the widespread adoption of high-flow nasal cannula (HFNC) therapy.
HFNC delivers heated, humidified gas at flow rates that exceed the patient’s inspiratory demand (typically 2 L/kg/min in infants), generating modest positive airway pressure, washing out nasopharyngeal dead space, reducing inspiratory resistance, and allowing precise FiO₂ titration. Its simplicity, tolerance by infants, and perceived safety have driven rapid uptake in emergency departments and general paediatric wards — often preceding high-quality evidence.
The critical clinical questions are: does HFNC improve meaningful outcomes beyond standard oxygen therapy? Where does it sit relative to CPAP? When should clinicians recognise HFNC failure and escalate? And for which conditions beyond bronchiolitis is the evidence robust? This synthesis examines the data.
What Evidence Was Reviewed
This synthesis draws on the three pivotal bronchiolitis RCTs, the 2024 Cochrane systematic review, meta-analyses comparing HFNC with both standard oxygen and CPAP, prospective studies evaluating HFNC failure predictors, and observational data on non-bronchiolitis indications.
| Study | Design | n | Population | Key Finding | Quality |
|---|---|---|---|---|---|
| Franklin et al. 2018 (NEJM) | Multicentre RCT | 1472 | Bronchiolitis <12 mo, Australia/NZ | Treatment failure: HFNC 12% vs standard O₂ 23% (p<0.001); no LOS/O₂ duration difference | High |
| Kepreotes et al. 2017 (Lancet) | Multicentre RCT | 202 | Moderate bronchiolitis <24 mo | No difference in time on oxygen; lower treatment failure with HFNC (14% vs 33%) | High |
| Durand et al. 2020 (Eur Respir J) | Multicentre RCT | 268 | Moderate bronchiolitis <6 mo, France | No significant reduction in escalation with HFNC at 3 L/kg/min (14% vs 20%, p=0.21); 3 pneumothoraces in HFNC arm | High |
| Armarego et al. 2024 (Cochrane) | Systematic review | NR | Bronchiolitis HFNC vs SOT/CPAP | HFNC reduces failure vs SOT (RR 0.50); higher failure vs CPAP (RR 1.61); no LOS difference | High |
| TRAMONTANE 2 (Milési et al. 2018) | Multicentre RCT | 142 | Severe bronchiolitis <6 mo, PICU | 3 L/kg/min vs 2 L/kg/min: no significant difference in treatment failure (39% vs 36%) | High |
| Ergul et al. 2020 (Pediatr Pulmonol) | RCT pilot | 63 | Critical bronchiolitis, PICU | HFNC vs CPAP: similar treatment failure (37% vs 36%); similar PICU LOS | Moderate |
| Nascimento et al. 2024 (Sci Rep) | Prospective multicentre | 102 | Bronchiolitis, PICU | ROX index at 12 h predicts HFNC failure; cutoff 6.5–7.2 (AUC 0.72) | Moderate |
| Sitthikarnkha et al. 2025 (PLoS One) | Retrospective cohort | 116 | LRTI children, PICU | SpO₂/FiO₂ <238 at 4 h predicted HFNC failure (AUC 0.89) | Moderate |
| Lin et al. 2019 (BMJ Open Respir Res) | SR & meta-analysis | 2121 | 9 RCTs; bronchiolitis | HFNC reduces treatment failure vs SOT; no difference in LOS, O₂ duration, ICU transfer, or intubation | High |
| Omran et al. 2024 (Cureus) | Retrospective cohort | 265 | PICU, ages 1 day–14 years | 69.4% improved; 17% required intubation; pneumonia (38.5%) and bronchiolitis (13.2%) commonest indications | Moderate |
Key Findings: High-Flow Nasal Cannula in Pediatric Respiratory Distress
Bronchiolitis: HFNC Reduces Treatment Failure but Not Length of Stay
The Franklin et al. NEJM trial (2018) is the largest and highest-quality RCT of HFNC in paediatric bronchiolitis. Among 1,472 infants under 12 months treated outside ICUs across 17 Australian and New Zealand centres, HFNC at 2 L/kg/min reduced the primary outcome of treatment failure requiring escalation of care: 12% in the HFNC group versus 23% in the standard therapy group (risk difference −11 percentage points; 95% CI −15 to −7; p<0.001). Crucially, 61% of infants who failed standard therapy were rescued with HFNC crossover, demonstrating its role as effective step-up therapy.
However, neither this trial nor any subsequent RCT has demonstrated that HFNC reduces hospital length of stay, duration of oxygen therapy, or PICU admission rates. The 2024 Cochrane review of all HFNC bronchiolitis trials confirms this pattern: treatment failure is consistently reduced (RR 0.50, 95% CI 0.40–0.62), but patient-centred outcomes including length of stay and ICU transfer remain unchanged. This paradox — reduced treatment failure without reduced downstream resource use — suggests that HFNC prevents clinical deterioration within the ward setting but does not accelerate the overall recovery trajectory.
HFNC vs CPAP: An Intermediate-Level Support
A critical positioning question is whether HFNC can replace CPAP for more severe respiratory distress. The 2024 Cochrane review addresses this directly: HFNC was associated with a significantly higher treatment failure rate compared to nasal CPAP (RR 1.61, 95% CI 1.06–2.42). A PICU-based pilot RCT comparing HFNC to CPAP in critical bronchiolitis (n=63) found similar failure rates of approximately 36–37% for both modalities, though this small study lacked power to detect clinically meaningful differences.
The aggregate evidence positions HFNC as an intermediate-level respiratory support — superior to standard low-flow oxygen but inferior to CPAP for more severely affected children. This hierarchy has practical implications: HFNC is appropriate for ward-based management of moderate respiratory distress, while CPAP remains the preferred modality for severe cases typically managed in ICU or high-dependency settings. HFNC may serve as a bridge to CPAP, potentially reducing the number of children requiring CPAP by managing a proportion who would otherwise deteriorate on standard oxygen alone.
The TRAMONTANE 2 trial (Milési et al. 2018) directly compared HFNC at 3 L/kg/min versus 2 L/kg/min in 142 young infants with severe viral bronchiolitis admitted to PICUs. There was no significant difference in treatment failure (39% vs 36%) between the higher and lower flow rates. A 2025 three-arm RCT from India comparing 1, 2, and 3 L/kg/min similarly found no additional clinical benefit at the highest flow rate.
Physiological studies demonstrate that positive pharyngeal pressure generated by HFNC increases linearly with flow rate, reaching approximately 2–4 cmH₂O at 2 L/kg/min with mouth closed. Higher flows generate more pressure but may increase patient discomfort, risk of gastric distension, and — as observed in the Durand et al. French RCT (3 pneumothoraces at 3 L/kg/min) — potentially increase barotrauma risk. Current evidence supports 2 L/kg/min as the optimal balance between efficacy, safety, and tolerability for bronchiolitis in infants.
While bronchiolitis dominates the RCT evidence, HFNC is increasingly used for other paediatric respiratory conditions. In a large PICU retrospective cohort (Omran et al. 2024, n=265), pneumonia accounted for 38.5% of HFNC indications, followed by bronchiolitis (13.2%) and asthma (8.7%). Overall, 69.4% of patients improved with HFNC, while 17% required intubation. Observational data suggest HFNC may reduce the need for escalation in acute asthma exacerbations and support post-extubation respiratory management, though RCT evidence for these indications in children is limited.
An RCT of HFNC in paediatric patients with congenital heart disease demonstrated reduced desaturations, reduced need for escalation to NIV, and lower hypercapnia while preserving haemodynamic stability — leveraging HFNC’s lower PEEP generation compared with CPAP, which may better suit preload-dependent cardiac physiology. However, robust RCT data for non-bronchiolitis indications remain sparse, and current practice in these settings is largely guided by clinical experience and physiological rationale rather than high-quality trial evidence.
Defining clear, objective criteria for HFNC failure is critical for patient safety — delayed escalation may lead to adverse outcomes, while premature escalation increases ICU resource utilisation. Several approaches have been investigated:
Composite clinical criteria (Franklin/PARIS): The Franklin NEJM trial used a composite of 3 of 4 criteria: persistent tachycardia, persistent tachypnoea, FiO₂ requirement exceeding 40% on HFNC (or >2 L/min on standard O₂), and hospital early warning tool triggering medical review. This composite approach, applied at any time during treatment, provides a pragmatic, reproducible framework.
SpO₂/FiO₂ ratio at 4 hours: A 2025 Thai retrospective study (n=116 children with LRTI) found that SpO₂/FiO₂ ratio below 238 at 4 hours after HFNC initiation predicted failure with an AUC of 0.89 (95% CI 0.78–0.99) — the strongest single predictor identified in paediatric populations.
ROX index at 12 hours: A 2024 prospective multicentre study (n=102 infants with bronchiolitis) validated the ROX index (SpO₂/FiO₂ divided by respiratory rate) for predicting HFNC failure in infants. A cutoff of 6.5–7.2 at the 12-hour evaluation demonstrated an AUC of 0.72. This adapted adult-derived metric shows promise but requires further validation in larger paediatric cohorts.
HFNC equipment and consumables cost an estimated 16 times more per patient than standard nasal cannula oxygen delivery. Given that HFNC has not been shown to reduce hospital length of stay or ICU admission rates, the cost-effectiveness of routine first-line HFNC use for all bronchiolitis admissions is uncertain. In high-income settings with available resources, the reduction in treatment failure (which translates to improved ward-level management and potentially reduced emergency ICU transfers) may justify the cost for higher-acuity patients.
In low- and middle-income country (LMIC) settings, one meta-analysis demonstrated that HFNC was associated with reduced length of stay compared with standard oxygen — a finding not replicated in high-income settings and potentially reflecting that HFNC substitutes for unavailable ICU-level care in resource-limited environments. This suggests that the cost-effectiveness calculation may favour HFNC more strongly in LMICs, where ICU resources are scarce and HFNC can prevent clinical deterioration requiring transfer to distant tertiary centres.
Quality & Consistency of Evidence
What the Evidence Does Not Show
Which subgroup of children benefit most from first-line HFNC vs rescue HFNC: No study has identified reliable clinical predictors of which children will require HFNC versus those who will recover with standard oxygen alone. This limits the ability to target HFNC use to those most likely to benefit, rather than applying it broadly to all bronchiolitis admissions.
HFNC for non-bronchiolitis paediatric respiratory conditions: RCT evidence for pneumonia, acute asthma, croup, and post-extubation support in children remains minimal. Current practice in these settings is largely extrapolated from bronchiolitis data and adult studies, representing a significant evidence-practice gap.
Standardised weaning protocols: Criteria for when and how to wean HFNC (flow reduction vs FiO₂ reduction vs both) are not standardised. The Franklin trial permitted weaning of FiO₂ to 0.21 before discontinuation after 4 hours, but no RCT has compared different weaning strategies.
Validated paediatric escalation thresholds: While the ROX index and SpO₂/FiO₂ ratio show promise, no prospective interventional study has demonstrated that protocolised use of these tools to guide escalation decisions improves outcomes compared with clinical judgement alone.
Practical Implications
Based on the current evidence, the following considerations may inform the use of HFNC in paediatric respiratory distress.
The strongest evidence supports HFNC as step-up therapy when standard low-flow oxygen fails (61% rescue rate in the Franklin trial). For most bronchiolitis patients, initiating standard oxygen and escalating to HFNC if failure criteria are met is a resource-efficient, evidence-supported strategy. First-line HFNC may be reasonable for patients presenting with more severe respiratory distress at the outset.
A flow rate of 2 L/kg/min in infants is supported by the largest RCTs and dose-comparison studies. Higher flow rates (3 L/kg/min) have not shown additional clinical benefit and may increase pneumothorax risk. FiO₂ should be titrated to maintain oxygen saturation ≥92% (or ≥94% per institutional policy), with weaning of FiO₂ before flow rate reduction.
Clinicians should use a structured approach to recognise HFNC failure: persistent tachycardia and tachypnoea despite 1–2 hours of therapy, FiO₂ requirement exceeding 40%, and early warning tool activation all warrant consideration of escalation to CPAP or PICU referral. The SpO₂/FiO₂ ratio at 4 hours (<238 predicts failure) offers an objective threshold validated in a paediatric LRTI population.
The evidence consistently shows HFNC is inferior to CPAP for severe bronchiolitis. Children with critical respiratory distress — particularly those presenting directly to PICU — should receive CPAP as the initial noninvasive support. HFNC is best suited to the ward and emergency department setting for moderate disease, serving as a bridge between standard oxygen and CPAP-level support.
Evidence Grade & Bottom Line
Evidence Grade: Moderate (for bronchiolitis); Limited (for other indications)
Strong evidence supports that HFNC reduces treatment failure compared with standard oxygen in bronchiolitis. However, the absence of benefit for patient-centred outcomes (length of stay, ICU admission, oxygen duration) tempers the clinical significance of this finding. The overall grade is moderate because the intervention improves an intermediate outcome (treatment failure) without demonstrating benefit on the outcomes that most matter to families and healthcare systems. For non-bronchiolitis indications, the evidence remains limited to observational studies.
- HFNC at 2 L/kg/min reduces treatment failure by approximately 50% compared with standard oxygen in bronchiolitis (RR 0.50), and 61% of standard oxygen failures can be rescued with HFNC crossover.
- No RCT has demonstrated that HFNC reduces hospital length of stay, duration of oxygen therapy, or PICU admission rates in bronchiolitis — limiting the case for universal first-line use.
- HFNC is inferior to CPAP for severe respiratory distress (RR for failure 1.61 vs CPAP), establishing HFNC as intermediate-level support between standard oxygen and CPAP.
- SpO₂/FiO₂ ratio below 238 at 4 hours and ROX index below 6.5–7.2 at 12 hours are emerging evidence-based thresholds for predicting HFNC failure and guiding escalation decisions.
- RCT-level evidence for HFNC in paediatric pneumonia, asthma, and post-extubation support remains a critical gap requiring dedicated trials.
Article Information
For Educational Purposes Only. This is an original evidence synthesis informed by the studies listed below. It does not replace clinical judgement. Drug dosages should be verified against current prescribing information.
References
- 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(12):1121–1131. DOI: 10.1056/NEJMoa1714855
- Kepreotes E, Whitehead B, Attia J, et al. High-flow warm humidified oxygen versus standard low-flow nasal cannula oxygen for moderate bronchiolitis (HFWHO RCT). Lancet. 2017;389(10072):930–939. DOI: 10.1016/S0140-6736(17)30061-2
- Durand P, Guiddir T, Kyheng C, et al. A randomised trial of high-flow nasal cannula in infants with moderate bronchiolitis. Eur Respir J. 2020;56(1):1901926. DOI: 10.1183/13993003.01926-2019
- Armarego M, Forde H, Wills K, Beggs SA. High-flow nasal cannula therapy for infants with bronchiolitis. Cochrane Database Syst Rev. 2024;3(3):CD009609. DOI: 10.1002/14651858.CD009609.pub3
- Milési C, Pierre AF, Deho A, et al. A multicenter randomized controlled trial of a 3-L/kg/min versus 2-L/kg/min high-flow nasal cannula flow rate in young infants with severe viral bronchiolitis (TRAMONTANE 2). Intensive Care Med. 2018;44(11):1870–1878. DOI: 10.1007/s00134-018-5343-1
- Ergul AB, Calıskan E, Samsa H, et al. High-flow nasal cannula versus continuous positive airway pressure in critical bronchiolitis: a randomized controlled pilot. J Pediatr Intensive Care. 2021;10(3):240–242. DOI: 10.1055/s-0040-1713611
- Nascimento MS, Quinto DER, Zamberlan GC, et al. ROX index as a predictor of failure of high-flow nasal cannula in infants with bronchiolitis. Sci Rep. 2024;14:754. DOI: 10.1038/s41598-024-51214-4
- Sitthikarnkha P, Uppala R, Thiraratananukulchai A, et al. Predictive factors for failure of high flow nasal cannula among children with lower respiratory tract infection. PLoS One. 2025;20(7):e0328442. DOI: 10.1371/journal.pone.0328442
- Lin J, Zhang Y, Xiong L, et al. High-flow nasal cannula therapy for children with bronchiolitis: a systematic review and meta-analysis. Arch Dis Child. 2019;104(6):564–568. DOI: 10.1136/archdischild-2018-315846
- Omran AA, et al. Utilization of high-flow nasal cannulas in pediatric critical care: a single-center experience. Cureus. 2024;16(12):e76284. DOI: 10.7759/cureus.76284
- Dalziel SR, Haskell L, O’Brien S, et al. Bronchiolitis. Lancet. 2022;400(10349):392–406. DOI: 10.1016/S0140-6736(22)01016-9
- Catarossi G, Manti S, Parisi GF, et al. Heated humidified high-flow nasal cannula in children: state of the art. Children. 2022;9(10):1538. DOI: 10.3390/children9101538