Oxygen administration with high nasal flow in covid-19

Both patients in the case have acute hypoxemic respiratory failure (type 1) due to covid-19. Oxygen administration with high nasal flow (‘high flow nasal oxygen’; HFNO) can be used to provide these patients with respiratory support in a non-invasive way (Figure 1). With HFNO, warm and humidified air, supplemented with a high fraction of oxygen, is administered at high flow via nasal cannulas. The potential physiological mechanisms of action of HFNO are summarized in Figure 2.1

Photo of the equipment (Airvo2, Fisher & Paykel, New Zealand) that can deliver oxygen with a high nasal flow. Via the nasal cannulas, the patient is supplied with heated and humidified air with a high oxygen fraction, with a flow of 10-60 l/min.

Due to the use of nasal goggles and the administration of warm and moist air, HFNO is often more comfortable than conventional oxygen therapy and non-invasive ventilation. In addition, HFNO can potentially prevent invasive ventilation, which has the advantage of preventing (long-term) immobilization and infectious complications. Previous research has shown that HFNO is effective in patients with hypoxemic respiratory failure without Covid-19, most of whom had community-acquired pneumonia.2 Studies on the role of HFNO in Covid-19 are still relatively scarce. Nevertheless, HFNO is frequently used in covid-19, at home and abroad.

Flowchart of the mechanisms of action of high nasal flow oxygen delivery in patients with acute hypoxemic respiratory failure.

Current Guidelines

The current guidelines state that HFNO can be used in patients with covid-19 in whom invasive ventilation is not immediately necessary.3-5 These guidelines are mainly based on observational and often retrospective studies.6-9 Recently, a few randomized studies have been published. In this article we describe the possible clinical applications of HFNO in patients with covid-19, without and with treatment restrictions, and we name the points for attention when using HFNO.

Recent studies

Through PubMed, we searched randomized or large cohort studies on the use of HFNO in covid-19 published as of January 1, 2020. We used the following search terms, including synonyms and abbreviations: ‘COVID-19’, ‘SARS-CoV-2’ , ‘high flow oxygen’, ‘high flow therapy’ and ‘optiflow’. We found four relevant RCTs and a cohort study, all conducted in patients with a comprehensive treatment regimen.10-14 The table provides an overview of these studies and the main outcomes.

Patients with covid-19 without treatment restrictions

All studies comparing HFNO to conventional oxygen therapy reported no difference in mortality.12-14 In contrast, two of the three studies suggested a positive effect of HFNO on intubation frequency.12,14 Strikingly, the largest RCT in this area – although terminated early – showed no reduction in the number of intubations with HFNO compared to conventional oxygen therapy. However, this study did show a reduction in the number of intubations among patients receiving continuous positive airway pressure therapy (CPAP) compared to conventional oxygen therapy.13

Two studies that compared HFNO with non-invasive ventilation (‘non-invasive ventilation’, NIV) also showed no difference in mortality.10,11 With regard to the effect on intubation frequency, these studies did not show a clear result: one study showed no effect, while the other showed a higher intubation frequency in the HFNO group compared to helmeted NIV. However, the outcome of the latter study may be biased because the patients with helmet NIV were allowed to switch to HFNO after 48 hours, while the primary outcome measure was not established until 28 days. In addition, the researchers had many years of experience with the application of helmet NIV, something that is often lacking in Dutch centers, especially in nursing wards. This makes the results more difficult to generalize to the Dutch situation.10

In short, compared to conventional oxygen therapy, HFNO may have a positive effect on intubation frequency. CPAP or NIV may be better than conventional oxygen therapy or HFNO in that regard. Additional studies are certainly needed to more robustly substantiate the role of HFNO in patients with covid-19, in addition to CPAP or NIV and conventional oxygen therapy.15 A search on clinicaltrials.gov shows that there are currently at least four randomized studies on the use of high nasal flow oxygen delivery in patients with covid-19 (NCT05197686, NCT04655638, NCT04381923, NCT03643939).

Patients with covid-19 and a treatment limitation

In the Netherlands, a treatment limitation is agreed for about a third of all admitted patients with covid-19.16 In these patients, admission to the ICU is no longer considered proportional on the basis of age, pre-existing vulnerable health or serious comorbidities, often captured in the term ‘frailty’. It may be a consideration to use HFNO in the ward instead of conventional oxygen therapy in these patients, with the aim of providing comfort and possibly better outcomes, such as lower mortality.

Very little research has been done on the effect of HFNO in this patient category; no randomized controlled trials have been conducted since the covid-19 outbreak. Some descriptive studies report a mortality of between 55 and 75% in this population.6,17-20

A retrospective cohort study is the only one to compare the use of HFNO with a non-rebreathing mask (NRM) in 67 patients > 75 years with covid-19 who were not admitted to ICU.21 The main outcome was that HFNO is associated with a lower mortality (17% vs. 5%). In addition, fewer patients were administered morphine or midazolam, suggesting that HFNO may provide greater comfort and better relief of dyspnea symptoms than using an NRM; this is consistent with an earlier study.22 Since there is little literature available, we conducted an observational study in the Covid-19 nursing ward of the UMCG among patients with a treatment disability who received HFNO as part of care (see box).

In conclusion, there is hardly any evidence in the literature that HFNO leads to better outcomes in patients with treatment limitations. However, it is widely used in clinical practice. Besides the fact that HFNO offers more comfort than conventional oxygen therapy, the fact that there is probably no better alternative for this population also plays a role. In them, a trial treatment with HFNO can be considered. Based on improvement or worsening of the work of breathing, oxygenation and comfort, a decision can be made to continue or stop HFNO.

Points of attention when using HFNO

When using HFNO it is important to be aware of the possible risks. It should be understood that this form of oxygen therapy is intended for critically ill respiratory insufficiency patients whose margins of further deterioration are narrow. Therefore, patients must be closely monitored and clear agreements must be made about extending care in the event of deterioration. It is recommended to monitor the oxygen saturation, respiratory rate and heart rate frequently or continuously (see figure 1), to set alarms and to monitor the patient. Certainly in a nursing ward, this is not necessarily organized. This involves both logistical and personnel challenges, such as the limited availability of devices, isolation rooms, emergency sockets with a power failure detector and experienced and trained personnel. These limitations may hinder the safe use of HFNO. Another condition is that it must be technically possible on site to increase the maximum oxygen flow stock.

A number of initial reservations about HFNO have since been disproved. It was initially thought that delaying intubation through the use of non-invasive respiratory support would lead to worse outcomes. To date, there are no indications for this.8,25 The risk of contamination of healthcare providers by aerosol spreading with HFNO also appears to be less high than initially feared.23 Although much is still unknown about the specific risk of transmission of Covid-19, HFNO seems to be not lead to an increased risk of infection with SARS-CoV-2 among healthcare providers, provided that adequate protective equipment is worn.24

Back to the patients

Patient A and B both underwent trial treatment with HFNO. Patient A initially responded well to HFNO: she tolerated the therapy well, her respiratory rate decreased and her oxygenation remained acceptable. After a few days her condition deteriorated. Despite an increase in the HFNO, the patient eventually died 14 days after admission. She used the HFNO until shortly before her death, as she found the therapy comfortable.

Patient B was admitted to the covid-19 ward. There, oxygen administration through a Venturi mask was replaced by HFNO. Respiratory rate decreased and oxygenation improved. The HFNO could be converted to oxygen delivery via a standard nasal cannula after four days. After a total hospitalization of eight days, the patient was discharged home in reasonably good condition.


In this article we have reviewed the limited literature on the application of high nasal flow oxygen (HFNO) in patients with covid-19. HFNO has been used clinically since the onset of the Covid-19 pandemic, although its effectiveness had not yet been demonstrated. The first randomized studies suggest that HFNO has no effect on survival, but may reduce the number of intubations compared to conventional oxygen therapy. Therefore, HFNO can be considered in patients with hypoxemic respiratory failure without treatment limitations. Adequate monitoring is therefore recommended in order to detect deterioration in time.

No comparative research has been conducted into the use of HFNO in patients with a treatment limitation; there are positive clinical experiences. HFNO may be considered in these patients, for whom comfort may be an important treatment goal. However, it is clear that several studies are needed to demonstrate the effectiveness of HFNO, also specifically for the Dutch situation. In this way, a better-founded estimate can be made of which patients will benefit from HFNO, and it may also be possible to predict at an early stage in whom HFNO will not be effective. In addition, research with a higher weight of evidence is absolutely needed into the use of HFNO in patients with covid-19 and a treatment limitation. Such studies may also provide a basis for care in other viral pneumonias.


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Recent News

Editor's Pick