• 
• 

Three Part Question

In [adults with interstitial lung disease and acute respiratory failure without hypercapnia, not secondary to COVID-19 infection]
Is [high flow nasal oxygen (HFNO) better than conventional oxygen therapy (COT)]
At [reducing symptoms and all-cause mortality outcomes]?

Clinical Scenario

You are in the emergency department when a pre-alert arrives for a 70-year-old man who has become unwell over the last 3 days with shortness of breath, on a background of interstitial lung disease. He is requiring 15L of oxygen via a non-rebreather mask to maintain saturations of 94%. A rapid COVID-19 swab is negative. Your colleague mentions high flow nasal oxygen is available, but you are unsure when it is appropriate to initiate.

Search Strategy

Medline (1973 to March 2025) was searched via the PubMed interface between 1973 and 21st March 2025 using the following search criteria:
((((((high flow) AND (oxygen)) AND (interstitial lung disease)) OR (idiopathic pulmonary fibrosis)) OR (interstitial pneumonitis)) AND (acute respiratory failure)) NOT (hypercapnia)

The following additional filters were used to achieve high-quality original research results:
Clinical Study, Clinical Trial, Clinical Trial, Phase I, Clinical Trial, Phase II, Clinical Trial, Phase III, Clinical Trial, Phase IV, Comparative Study, Controlled Clinical Trial, Equivalence Trial, Evaluation Study, Meta-Analysis, Multicentre Study, Observational Study, Pragmatic Clinical Trial, Randomized Controlled Trial, Review, Systematic Review, Validation Study, from 1973 – 2025

Original research papers were selected based on the following inclusion and exclusion criteria:
Inclusion -
Adults ≥18 years old with a diagnosis of interstitial lung disease as per the American Thoracic Society and European Respiratory Society joint statement.(1)
Acute respiratory failure secondary to all causes, other than COVID-19 infection.
Use of high flow nasal oxygen (HFNO) compared with conventional oxygen therapy (COT), or no comparator.
Reported or translated to English language.

Exclusion -
Patients <18 years old.
Reported in languages other than English.
Acute respiratory failure secondary to COVID-19 infection.
Comparison to bilevel positive airway pressure (BiPAP), otherwise termed non-invasive ventilation (NIV).

Search Outcome

319 papers were initially identified through searches. After removal of duplicates and screening of titles and abstracts, 18 articles underwent full text review. Three articles were directly relevant to the 3-part question, met the inclusion criteria and of sufficient quality for reporting. One further paper was identified through review of systematic review and meta-analyses bibliographies. All included papers are displayed in table 1.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Pagliaro et al 2024 Italy	715 ILD patients from across 12 cohort studies encompassing end-stage/ acute ILD, chronic ILD and acute ILD only. 377 patients were represented from the acute ILD only setting. All ILD types were included in the studies. Idiopathic pulmonary fibrosis (IPF) was the most common subtype (185/377, 49.1%).	Systematic review with narrative synthesis, compromising 12 worldwide cohort studies (5 prospective, 7 retrospective). Comparators: COT, non-invasive ventilation (reported separately)	Mortality	No significant difference in survival between COT and HFNO was observed. Median survival time for COT was 133.0 days (95% CI 26.0 – 374.0) and HFNO 21.0 days (95% CI 13.0 – 61.0) p = 0.1323.	Small number of studies comparing HFNO use with COT. Included studies had heterogenous outcomes, making comparison challenging. Conclusions drawn from observational studies only, no available randomised control trials. Discussion appears to suffer from reporting bias, focusing on individual study positive outcomes rather than providing full narrative summary of overall picture.
			Quality of death and dying (QODD) average score – QODD is a tool used to evaluate symptom management and quality of life during the dying phase of life. There are various scales available. This paper focused on the ‘Good Death Inventory’ questionnaire: measuring family perspective through 7-point scale for 18 domains. A higher score equates to a higher quality of death.	Across all 18 domains in the QODD score, the HFNO group had a higher average score (4.58 +- 0.67) compared to the COT group (4.09 +- 0.96) suggesting higher quality of death in the HFNO group.
			Dyspnoea (modified Borg scale)	At 60 minutes, dyspnoea scores were lower within the HFNO group (2.9) compared to the COT group (4.9). However, further statistical analysis was not undertaken, so we cannot conclude if this was a statistically significant difference or not.
Sanguanwong et al 2023 Thailand	480 interstitial lung disease patients with acute respiratory failure. Adults ≥ 18 years old. All interstitial lung disease types considered. 222/480 (46.2%) included patients had IPF.	Systematic review and meta-analysis, compromising one RCT and 9 cohort studies (1 prospective, 8 retrospective) Comparators: COT, non-invasive ventilation (reported separately)	Mean PF Ratio Difference HFNO to COT (PF ratio: ratio of arterial oxygen partial pressure to fraction inspired oxygen). The PF ratio is a clinical marker of hypoxaemia, with a higher ratio representing a more favourable picture.	Using a random-effect model, HFNO was shown to have a statistically significant improvement in PF ratio compared to COT with a mean difference of 67.27 (95% CI 1.17 – 133.37), p = 0.05. The I2 statistic was 38%, suggesting low statistical heterogeneity among the included studies for this calculation.	Small number of studies included in each meta-analysis. Heterogenous outcomes and comparators within studies. Little discussion about the underlying ILD diagnosis and how this may have impacted outcomes. No consideration of additional treatment modalities alongside HFNO which may contribute to results seen.
			Mortality Risk Ratio – HFNO compared to COT	The relative risk of mortality within the HFNO group compared to the COT group was not statistically significant (RR = 1.10 (95% CI 0.83 – 1.44), p = 0.51). This result was also impacted by moderate study heterogeneity as per the I2 statistic (68%).
Koyauchi et al 2020 Japan	66 patients with acute exacerbation of interstitial lung disease (including idiopathic pulmonary fibrosis (IPF) and non-IPF). Median age was 78 (range 73-82). 31/66 (47%) had a diagnosis of IPF.	Retrospective two-centre cohort study using secondary data from medical inpatient record review between July 2013 and November 2017. Comparator: HFNO-failure (unsuccessful HFNO withdrawal)	Mortality: 30-day survival rate	48.5%	Small study numbers and retrospective in design. No comparator outside of HFNO. Lack of consideration of additional confounders in multivariate model: only LTOT and Sp02/Fi02 ratio included; may over-estimate role of this. Lack of discussion re. overall mortality poor, even in ‘successful’ HFNO wean. Very wide confidence intervals: results suffer from imprecision and underpowered.
			HFNO withdrawal rate due to improved oxygenation	39.4% withdrew successfully from HFNO
			Multivariate logistic regression analysis for survival (OR, odds ratios)	In the multivariate model for survival, long-term oxygen therapy (LTOT) was associated with reduced odds of survival (OR 0.52 (95% CI 0.09 – 3.04)) but this was not statistically significant (p = 0.47) 24h Sp02/Fi02 ratio of ≥170.9 was also included in the multivariate model. A higher ratio was significantly associated with improved survival (OR 51.3 (95% CI 6.13 – 430), p <0.001) (Where Sp02/Fi02 = pulse oximetric saturations/ fraction of inspired oxygen)
Vianello et al 2019 Italy	17 patients with acute exacerbation of idiopathic pulmonary fibrosis. Evidence of acute respiratory failure (ARF): acute, rapid deterioration in respiratory function with hypoxaemia ratio <250mmHg. Median age 67 (range 51-89)	Observational retrospective cohort single-site study via inpatient medical records review between 01 May 2013 and 30 April 2018. Comparator: conventional oxygen therapy	Mean survival time (days)	There was no significant difference in the mean survival for those on COT (21.0 days (95% CI 13.0 – 61.0)) vs. for those on HFNO (133.0 days (95% CI 26.0 – 374)) p = 0.1323.	Very small study which is retrospective in design. APACHE II score significantly different between treatment success and failure group at baseline. Limited discussion in relation to this. Wide confidence intervals of most results suggesting imprecision. Underpowered study. No clearly defined statistical plan a priori.

Comment(s)

Interstitial lung diseases (ILD) represent a heterogenous group of disorders with an unpredictable disease course, accompanied by significant unmet physical, psychological and educational needs.(6)(7) ILD patients are frequently admitted to hospital secondary to respiratory decompensation, with one dataset showing 56.2% of ILD patients had an ILD-related admission over a 5-year period.(8) The causes of acute respiratory failure (ARF) in ILD are plentiful, requiring a thorough diagnostic work-up to establish aetiology.(9) A proportion are secondary to acute exacerbations of interstitial lung disease (AEILD): a <30-day deterioration in symptoms and associated bilateral, superimposed ground-glass change on a background ILD.(10) AEILDs are associated with poor prognosis despite optimal investigative, pharmacological and non-pharmacological management.(11) In other patient groups with severe hypoxia, the role of high flow nasal oxygen (HFNO) in providing non-invasive respiratory support is becoming increasingly established. The European Respiratory Society recommends HFNO over COT in ARF – but the evidence remains limited in the ILD population.(12) Across these four studies 857 patients with ARF in ILD are included. Both Koyauchi et al and Vianello et al are included within both systematic reviews but have been reported separately due to the additional understanding of mortality and survival factors (compared to COT) that they bring. All studies considered mortality as an outcome but none demonstrated a mortality benefit with HFNO use.(2)(3)(4)(5) This was despite HFNO demonstrating an improved P/F ratio compared to COT. However, the data is limited and heterogenous in its outcomes making between-study comparison challenging. There is a small amount of evidence that ILD patients may benefit symptomatically from HFNO use over COT, including in the setting of palliative care.(2)(13) HFNO has the advantage of providing higher flow rates, helping to meet inspiratory demands that even a 15L non-rebreather mask cannot - which may explain its symptomatic benefit.(14) However, clinicians should be alert to the potential challenges of weaning HFNO in this group given the symptomatic benefit it provides.

Clinical Bottom Line

There is limited evidence to support the use of HFNO to improve mortality outcomes for ARF in ILD, but it can be of benefit in the realms of symptom management. Mortality despite its use, though, remains significant and so requires careful discussion with patients and their relatives should a trial be utilised in the setting of ARF.

References

1. Travis WD et al An Official American Thoracic Society/European Respiratory Society Statement: Update of the International Multidisciplinary Classification of the Idiopathic Interstitial Pneumonias Am J Respir Crit Care Med 2013 Sep 15;188(6):733–48
2. Pagliaro et al High-Flow Nasal Cannula System in Respiratory Failure Associated with Interstitial Lung Diseases: A Systematic Review and Narrative Synthesis Journal of Clinical Medicine 2024 May 17;13(10):2956
3. Sanguanwong et al Effect of noninvasive respiratory support on interstitial lung disease with acute respiratory failure: A systematic review and meta-analysis. Canadian Journal of Respiratory Therapy 2023 Nov 3;59
4. Koyauchi et al Pulse oximetric saturation to fraction of inspired oxygen (SpO 2 /FIO 2 ) ratio 24 hours after high-flow nasal cannula (HFNC) initiation is a good predictor of HFNC therapy in patients with acute exac Therapeutic Advances in Respiratory Disease 2020 Jan 11;14
5. Vianello et al High-flow nasal cannula oxygen therapy to treat acute respiratory failure in patients with acute exacerbation of idiopathic pulmonary fibrosis Therapeutic Advances in Respiratory Disease 2019 Jan 6;13
6. Antoniou KM et al Interstitial lung disease Eur Respir Rev 2014;23(131):40–54
7. Bramhill C et al A scoping review of the unmet needs of patients diagnosed with idiopathic pulmonary fibrosis (IPF) PLoS One 2024;19(2 February):1–22
8. Wälscher J et al Hospitalisation patterns of patients with interstitial lung disease in the light of comorbidities and medical treatment – a German claims data analysis. Respir Res 2020 Dec 26;21(1):73
9. Faverio P et al Management of Chronic Respiratory Failure in Interstitial Lung Diseases: Overview and Clinical Insights Int J Med Sci 2019;16(7):967–80
10. Leuschner G et al Acute Exacerbation in Interstitial Lung Disease Front Med 2017 Oct 23;4
11. Gannon WD et al Outcomes and Mortality Prediction Model of Critically Ill Adults With Acute Respiratory Failure and Interstitial Lung Disease. Chest 2018 Jun;153(6):1387–95
12. Oczkowski S et al ERS clinical practice guidelines: high-flow nasal cannula in acute respiratory failure Eur Respir J 2022 Apr;59(4):2101574
13. Ruangsomboon O et al High-Flow Nasal Cannula Versus Conventional Oxygen Therapy in Relieving Dyspnea in Emergency Palliative Patients With Do-Not-Intubate Status: A Randomized Crossover Study Ann Emerg Med 2020 May;75(5):615–26
14. Lodeserto FJ et al High-flow Nasal Cannula: Mechanisms of Action and Adult and Pediatric Indications Cureus 2018 Nov 26