Medical and Dental Consultantsí Association of Nigeria
Home - About us - Editorial board - Search - Ahead of print - Current issue - Archives - Submit article - Instructions - Subscribe - Advertise - Contacts - Login 
  Users Online: 5758   Home Print this page Email this page Small font sizeDefault font sizeIncrease font size

  Table of Contents 
Year : 2022  |  Volume : 25  |  Issue : 7  |  Page : 1088-1093

Role and limitations of high-flow nasal oxygen therapy in COVID-19 patients: An observational study

1 Department of Anesthesiology and Reanimation Bursa Uludag Faculty of Medicine, Bursa City Hospital, Bursa, Turkey
2 Department of Anesthesiology and Reanimation, Bursa City Hospital, Bursa, Turkey

Date of Submission01-Jul-2021
Date of Acceptance23-May-2022
Date of Web Publication20-Jul-2022

Correspondence Address:
Dr. P Kucukdemirci-Kaya
Bursa Uludag Universty Faculty of Medicine, 16059 Görükle / Bursa
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/njcp.njcp_1646_21

Rights and Permissions

Background: The management of acute respiratory failure in COVID-19 patients and the role and limitations of high-flow nasal oxygen therapy (HFNOT) remain unclear. Aim: This study aimed to investigate the effect of HFNOT, identify the characteristics of patients who will benefit from therapy, and determine monitoring strategies to decide on endotracheal intubation for patients with COVID-19. Patients and Methods: We conducted a prospective observational study of COVID-19 patients who were admitted to the intensive care unit (ICU) and required HFNOT for at least 2 days between 20 March 2020 and 20 June 2020. The exclusion criteria were a severe respiratory failure, reduced levels of consciousness, combination with other noninvasive ventilation strategies, and exhaustion. The patients were followed up until ICU discharge. The primary outcome was the proportion of patients with COVID-19 who were successfully weaned from HFNOT, whereas failure comprised intubation or death on HFNOT. Results: Thirty-five subjects (24 males, mean-age: 61.62, standard deviation: 14.9 yr.) were included in the study. A total of 20/35 (57.1%) subjects survived to discharge. C-reactive-protein (CRP) and interleukin-6 (IL-6) levels were significantly increased in the treatment failure group (CRP; effect size (r):0.35, P: 0.037, IL-6; r: 0.37, P: 0.03). Although there was a difference between repeated measures of partial-pressure-of-oxygen/fraction-of-inspired-oxygen (PaO2/FiO2:P/F) rates (partial-eta-squared (ηp2):0.79, P < 0.001), no difference was found between carbon dioxide levels (ηp2:0.29, p: 0.44). There was also no difference between ROX (ratio-of-oxygen-saturation/FiO2 to respiratory-rate) rates (Kendall's W: 0.33 P = 0.310). Conclusion: In COVID-19 patients with mild-to-moderate dyspnea and hypoxemia who are nonresponsive to conventional-oxygen-therapies, the initial approach may involve the use of HFNOT. In this study, patient monitoring could be performed with ROX and P/F ratios, and the effectiveness of the treatment could be decided by looking at these rates in the second hour. Prolongation of the period and awake prone positioning did not improve the outcome.

Keywords: Acute respiratory failure, awake prone positioning, COVID-19, high-flow nasal oxygen therapy, intensive care unit

How to cite this article:
Kucukdemirci-Kaya P, Kilic I, Kaya M, Kelebek-Girgin N. Role and limitations of high-flow nasal oxygen therapy in COVID-19 patients: An observational study. Niger J Clin Pract 2022;25:1088-93

How to cite this URL:
Kucukdemirci-Kaya P, Kilic I, Kaya M, Kelebek-Girgin N. Role and limitations of high-flow nasal oxygen therapy in COVID-19 patients: An observational study. Niger J Clin Pract [serial online] 2022 [cited 2022 Aug 15];25:1088-93. Available from:

   Background Top

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2 named COVID-19), a highly transmissible respiratory virus, has become a worldwide pandemic and as of 1 December 2020, 61.8 million cases and 1.4 million deaths have been reported.[1] Although most patients present with mild respiratory symptoms, 25%–34% of them have serious symptoms such as acute respiratory failure (ARF), acute respiratory failure syndrome (ARDS), septic shock, and multiorgan failure.[2],[3] In serious COVID-19 pneumonia, acute hypoxemic respiratory failure is frequently seen, and oxygen administration forms the basis of supportive therapy for these patients.[4] In recent years, high-flow nasal oxygen therapy (HFNOT) has become prominent in the treatment of respiratory failure.[4],[5] This noninvasive support system provides a higher concentration and flow of oxygen, resulting in decreasing anatomic dead space by preventing rebreathing and ensuring a continuous positive airway pressure (CPAP) effect.[5] In addition, the prone position, which is known to improve oxygenation by reducing shunt fraction and recruiting atelectatic lung tissue, can be applied.[4],[6] However, in COVID-19, the usage of HFNOT is highly controversial because of concerns about the benefits and risks of aerosol dispersion.[4],[7] The Surviving Sepsis Campaign COVID-19 guidelines provide a weak recommendation for the preferential use of HFNOT over other noninvasive ventilation strategies in patients refractory to conventional oxygen therapy (COT).[7],[8]

Direct cytopathic effects of the virus on pneumocytes as opposed to inflammatory injury and virus-induced decreases in surfactant levels causing atelectasis are some of the unique pathologic findings seen in patients with COVID-19.[9] Hypoxemia is the hallmark of pulmonary derangement of the disease; in fact, a case series of COVID-19 patients demonstrated the presence of significant hypoxemia with no sign of respiratory distress (''silent hypoxemia'').[9],[10] These patients without respiratory distress try to compensate for impaired gas diffusion by increasing minute ventilation with preserved lung compliance, which may lead to extreme hypocapnia.[11] Despite profound hypoxemic respiratory failure being the dominating clinical feature of COVID-19,[12] the management of hypoxemia is still controversial. However, HFNOT use and awake prone positioning recommendations were based on limited clinical data.[9]

This study aimed to investigate the effect of HFNOT, identify the characteristics of patients who will benefit from therapy, and determine monitoring strategies to decide on endotracheal intubation for patients with COVID-19.

   Materials Top

Study design

We conducted a prospective observational study that was approved by the Bursa City Hospital Ethical Committee (no: 2020-3/13)

Inclusion criteria

Adult subjects admitted to any of the adult ICUs at our facility with laboratory-confirmed COVID-19 infection and respiratory failure requiring HFNOT for at least 2 days between 20 March 2020 and 20 June 2020 were included in this study. COVID-19 was confirmed by a positive result on a reverse-transcriptase-polymerase-chain-reaction assay of a specimen collected on a nasopharyngeal swab, and respiratory failure was defined as a respiratory rate ≥30 breaths per min with oxygen saturation ≤92% despite oxygen at 15 L/min via reservoir bag and/or arterial oxygen partial pressure to fractional inspired oxygen (PaO2/FiO2) ratio <150.

Exclusion criteria

Individuals were excluded if they (1) had severe respiratory failure; (2) were unconscious or confused; (3) had received HFNOT or mechanical ventilation (invasive or non-invasive) support when they were admitted to the hospital; (4) were administered a combination of HFNOT and other noninvasive ventilation strategies; or (5) were exhausted.

HFNC protocol

Heated and humidified HFNOT was exclusively provided within the ICU. The patients wore surgical masks, and all personnel were supplied with personal protective equipment, including N95 masks and visors. HFNOT was delivered either by a Hamilton C3 ventilator (Hamilton Medical AG, Bonaduz, Switzerland) or Airvo 2 (Fisher & Paykel Healthcare, Irvine, California, USA) machine. The flow was initiated at 50–80 L/min with FiO2 0,6–1.0, titrated to aim for oxygen saturation (SpO2) ≥88%.

Data collection

Demographic data (age and gender) of the subjects, their comorbidities, acute physiology, and chronic health evaluation (APACHE II) scores, sequential organ failure assessment (SOFA) scores (while on initial HFNOT), whether they remained in the prone position for at least 12/24 h during HFNOT, lung tomography Type I (L type) or Type II (H type), acute phase reactants admitted to the ICU (ferritin, C-reactive protein, procalcitonin and interleukin-6), partial-pressure-of-oxygen/fraction-of-inspired-oxygen (PaO2/FiO2:P/F) ratios (initial (P/F1), after; 2 h (P/F2), 1 day (P/F3), and 2 days (P/F4), partial carbon dioxide pressures initial (PaCO21)), after; 2 hours (PaCO22), 1 day (PaCO23) and 2 days (PaCO24), ROX (ratio of oxygen saturation/FiO2 to respiratory rate) rates after; 2 hours (ROX1), 1 day (ROX2), and 2 days (ROX3) and outcome (patients who were successfully weaned from HFNOT and discharged from the ICU were defined as the success group and those who comprised intubation or death on HFNOT were defined as the failure group) were recorded. The use of antiviral medication, sedatives, and antipsychotics was also recorded during HFNOT.

Primary outcome

The primary outcome was the proportion of patients with a successful outcome (weaned from HFNOT and discharged from the ICU). Failure was defined as the need for intubation or death while on HFNOT.

Secondary outcomes

We assessed the following secondary outcomes:

  1. Effects of prone positioning and lung phenotypes on the outcome.
  2. Prognostic values of acute-phase reactants.
  3. Monitoring strategies determined for HFNOT.
  4. Oxygenation and carbon dioxide wash-out after initial HFNOT.

Statistical analysis

Continuous variables are expressed as the mean (standard deviation) or median (interquartile range) and n (%), depending on the normality of the distribution for continuous variables; the Kolmogorov–Smirnov test was used to test the normality of the distribution for continuous variables. Comparisons between two groups (treatment success vs failure) were analyzed by an independent t-test or the Mann–Whitney U test. A repeated-measures ANOVA with a Greenhouse–Geisser correction determined the P/F and CO2 differences between time points. Post hoc tests using the Bonferroni correction revealed which measurement differed. The Chi-square test of independence was used to determine the relationship between two nominal (sex, prone position, and lung phenotype) variables. The Friedman test was used for repeated measures of ROX. Data analysis was conducted with statistical package for social science (SPSS) statistical software (SPSS 26.0: SPSS; Chicago, II, USA), and P values of <0.05 and <0.0125 (for Bonferroni correction) were considered to be significant.

   Results Top

Patient population and characteristics

Between 20 March 2020 and 20 June 2020, 284 subjects with laboratory-confirmed COVID-19 were admitted to intensive care. HFNOT was initiated in 126 subjects, and we excluded sixty-two of them who needed intermittent CPAP treatment. Thirty-five patients who only received HFNOT as NIV for a minimum of 2 days were included in the study. The patient characteristics are shown in [Table 1].
Table 1: Overall patient characteristics and comparisons between the treatment success and treatment failure groups

Click here to view

Primary outcome: proportion of patients with a successful outcome

A total of 57.1% (n: 20) of these patients recovered after HFNOT, and they were discharged to service within 14 days, whereas 14 (40%) patients were intubated during HFNOT because of intermittent mandatory ventilation (IMV) requirements; sudden cardiac death occurred during the treatment of one patient (2.9%).

Secondary outcomes

The prone position and lung phenotypes

Thirteen subjects were placed in the prone position two or more times for a minimum of 12 h a day. A total of 5/13 (25%) subjects survived to be discharged with HFNOT. Prone positioning was not higher in the treatment success group than in the treatment failure group (5/13 vs 11/13, Odds Ratio: OR: 0.29, p: 0.173). When patients were classified according to their lung tomography [Figure 1], 45.7% (n: 16) of the patients were classified as Type I (L type), and 54.3% (n: 19) were classified as Type II (H type) [Figure 1]. There was no significant difference in the H type to L type ratio between the treatment success and failure groups (OR: 0.24, P: 0.11).
Figure 1: Type L and Type H phenotypes of two Covid-19 patients

Click here to view

Prognostic values of acute phase reactants

The levels of interleukin-6 and C-reactive protein were significantly increased in the treatment failure group on admission (P = 0.03, pP = 0.037, respectively). No significant difference was found when compared according to ferritin and procalcitonin levels (P: 0.95, P: 0.11).

Monitoring strategies for HFNOT

Both of the P/F and ROX ratios measured at different time points were significantly higher in the treatment success group than the treatment failure group (P/F; P/F(1) (initial with COT); 95% Confidence Interval (CI): 32.16–137.71, p: 0.04, P/F(2) (2 hours after HFNOT); CI: 53.40–162.63, P: 0.01, P/F(3) (1 day after HFNOT); CI: 87.06–198.63, P < 0.01, P/F(4) (2 days after HFNOT); CI: 137.73–264.47, P < 0.01) (ROX; ROX(1) (2 hours after HFNOT); CI: 98.52–171, P < 0.01, ROX(2) (1 day after HFNOT); CI: 99.7–177, P < 0.01, ROX(3) (2 days after HFNOT); r: 0.79, P < 0.01) [Figure 2] and [Figure 3].
Figure 2: P/F rates of the treatment success and failure groups

Click here to view
Figure 3: ROX rates of the treatment success and failure group

Click here to view

Oxygenation and carbon dioxide washout after initial HFNOT

According to repeated measures ANOVA (F (1.622, 55.161) = 15.313 P < 0.01) with the post hoc test, P/F(1) rates were different than all measured P/F rates after HFNOT (P/F(1) vs P/F(2) P < 0.01, P/F(1) vs P/F(3) P < 0.01, P/F(1) vs P/F(4) p: 0.01). There was no significant difference between the other P/F rates (P/F(2) vs P/F(3) P: 0.59, P/F(2) vs P/F(4) P: 0.17, P/F(3) vs P/F(4) P = 0.4). Additionally, there was no significant difference between ROX rates at different time points (x2 (2) = 2.341 P = 0.310). These two statistical results were associated with improvement in oxygenation, and adequacy of oxygenation could be decided by looking at these rates in the second hour after initial HFNOT. Arterial blood gas partial carbon dioxide (PaCO2) levels at different times (PaCO2(1): initial, PaCO2(2): 2 hours after HFNOT, PaCO2(3): 1 day after HFNOT, PaCO2(4): and 2 days after HFNOT) were not significantly different between the treatment success and failure groups (F (2.864, 97.38) = 0.902, p: 0.439). PaCO2 measurements were compared with repeated measures ANOVA, and there was no significant difference observed (with the Greenhouse Geisser correction) associated with HFNOT having a considerable effect on carbon dioxide washout in COVID-19 patients.

   Discussion Top

In the absence of an evidence base, many reports based on expert opinion and previous studies demonstrating that HFNOT is associated with more ventilator-free days and lower mortality and reintubation rates in acute hypoxemic respiratory failures as a result of various causes except COVID-19 have been published referencing the role and limitations of HFNOT in COVID-19.[13],[14] We suggest that for spontaneously breathing COVID-19 patients with mild-to-moderate dyspnea and hypoxemia, who are nonresponsive to COT, the initial approach may involve the use of HFNOT based on our clinical data.

The evidence for awake prone positioning is limited to case series data and small observational studies with heterogeneous approaches to noninvasive respiratory support. These reports have demonstrated short-term improvements in oxygen requirements (PaO2) and demand (FiO2) with no harm to patients.[15] Prone positioning with HFNOT in our subjects did not improve outcomes. This finding might be explained by prone positioning applied, which had lower P/F rates.

Gattinoni and colleagues describe two time-related phenotypes in COVID-19 and accordingly recommended different respiratory failure management.[10] There was no difference according to lung phenotypes between the treatment success and failure groups. This is the first report, to our knowledge, to describe outcomes associated with these phenotypes in HFNOT.

We analyzed disease severity with regard to inflammation markers. Similar to the findings from other studies,[16],[17] CRP and IL-6 were higher in the treatment failure group.

As with other NIV strategies, delay in initiating invasive ventilation via tracheal intubation caused by inappropriate perseverance with HFNOT may result in increased ICU mortality and worsened outcomes.[5] Therefore, clinicians need to determine the correct monitoring protocols. However, extreme hypocapnic hypoxia and respiratory alkalosis in patients with respiratory failure have previously been relatively uncommon.[11] Respiratory alkalosis shifts the oxyhemoglobin dissociation curve to the left, thereby increasing the oxygen affinity of hemoglobin, which is evident from a decrease in the P50 value and an increase in arterial oxygen saturation (SaO2). Additionally, the alveolar gas equation predicts decreased alveolar CO2 tension (PACO2). For all these mechanisms, Ottestad and SΦvik argue that it is impossible to infer from SpO2, the degree of hypoxemia and consequently the severity of respiratory failure without knowledge of the accompanying PaCO2 value.[11] A new index termed “ROX,” defined as the ratio of oxygen saturation as measured by pulse oximetry/FiO2 to respiratory rate, has been recommended for monitoring HFNOT in some reports.[4] We found that ROX ratios were as reliable as P/F ratios to predict clinical stage, and better P/F and ROX rates were associated with better outcomes.

HFNOT has emerged as a noninvasive strategy for improving oxygenation and carbon dioxide clearance relative to other NIV strategies.[7] Similar to the findings from other studies,[4],[7],[18] the P/F ratio improved after initial HFNOT in our study, but we did not find differences in the improvement of the P/F ratio with prolonged treatment time. However, we did not find a considerable effect on carbon dioxide washout, unlike previous studies.[19]

A limitation of this study was that Type L and Type H patients were identified by CT scan. Second, we did not report flow and FiO2 changes to investigate the weaning period. Future studies are needed to understand the role and limitations of HFNOT in large populations. Last, we were unable to measure esophageal pressures during HFNOT.

In summary, appropriately selected patients with COVID-19 ARF frequently responded to HFNOT. Patient monitoring could be performed with ROX and P/F ratios. It was determined that the effectiveness of the treatment could be decided by looking at these rates in the second hour, and prolongation of the period and prone position would not improve the outcome.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

   References Top

Coronavirus disease (COVID-19) Situation Report-20201201 Available from: [Last accsessed on 2020 Dec 01].  Back to cited text no. 1
Wang D, Hu B, Hu C, Zhu F, Liu X, Zhang J, et al. Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China. JAMA 2020;323:1061-9.  Back to cited text no. 2
Goyal P, Choi JJ, Pinheiro LC, Schenck EJ, Chen R, Jabri A, et al. Clinical characteristics of Covid-19 in New York City. N Engl J Med 2020;382:2372-4.  Back to cited text no. 3
Gürün Kaya A, Öz M, Erol S, Çiftçi F, Çiledağ A, Kaya A. High flow nasal cannula in COVID-19: A literature review. Tuberk Toraks 2020;68:168-74.  Back to cited text no. 4
Ashraf-Kashani N, Kumar R. High-flow nasal oxygen therapy. Br J Anaesth Education 2017;17:63-7.  Back to cited text no. 5
Nishimura M. High-flow nasal cannula oxygen therapy devices. Respiratory Care 2019;64:735-42.  Back to cited text no. 6
Agarwal A, Basmaji J, Muttalib F, Granton D, Chaudhuri D, Chetan D, et al. High-flow nasal cannula for acute hypoxemic respiratory failure in patients with COVID-19: Systematic reviews of effectiveness and its risks of aerosolization, dispersion, and infection transmission. J Can Anesth 2020;67:1217-48.  Back to cited text no. 7
Alhazzani W, Moller MH, Arabi YM, Loeb M, Ng M, Fan E. et al. Surviving Sepsis Campaign: Guidelines on the Management of Critically III Adults with Coronavirus Disease 2019 ( COVID-19). Critical Care Medicine 2020;48:e440-469. doi:10.1097/CCM.0000000000004363.  Back to cited text no. 8
Navas-Blanco JR, & Dudaryk R. Management of respiratory distress syndrome due to COVID-19 infection. BMC Anesthesiol 2020;20:177. doi: 10.1186/s12871-020-01095-7.  Back to cited text no. 9
Gattinoni L, Chiumello D, Caironi P, Busana M, Romitti F, Brazzi L, et al. COVID-19 pneumonia: Different respiratory treatments for different phenotypes? Intensive Care Med 2020;46:1099-102.  Back to cited text no. 10
Ottestad W, SΦvik S. COVID-19 patients with respiratory failure: What can we learn from aviation medicine? Br J Anaesth 2020;125:e 280-341.  Back to cited text no. 11
Sherren PB, Ostermann M, Agarwal S, Meadows CIS, Ioannou N, Camporota L. COVID-19-related organ dysfunction and management strategies on the intensive care unit: A narrative rewiew. Br J Anaesth 2020;125:912-25.  Back to cited text no. 12
Frat JP, Thille AW, Mercat A, Girault C, Ragot S, Perbet S, et al. High flow oxygen through nasal cannula in hypoxemic respiratory failure. N Engl J Med 2015;372:2185-96.  Back to cited text no. 13
Rochwerg B, Granton D, Wang DX, Helviz Y, Einav S, Frat JP, et al. High flow nasal cannula compared with conventional oxygen therapy for acute hypoxemic respiratory failure: A systematic rewiew and meta-analysis. Intensive Care Med 2019;45:563-72.  Back to cited text no. 14
Ding I, Wang I, Ma W, He H. Efficacy and safety of early prone positioning combined with HFNC or NIV in moderate to severe ARDS: A multi-centre prospective cohort study. Crit Care 2020;24:28. doi: 10.1186/s13054-020-2738-5.  Back to cited text no. 15
Abou-Arab O, Bennis Y, Gauthier P, Boudot C, Bourdenet G, Gubler B, Beyls C, et al. Association between inflammation, angiopoietins, and disease severity in critically ill COVID-19 patients: A prospective study. Br J Anaesth 2021;126:e127-30.  Back to cited text no. 16
Liu F, Li L, Xu M, Wu J, Luo D, Zhu Y, et al. Prognostic value of interleukin-6, C-reactive protein, and procalcitonin in patients with COVID-19. J Clin Virol 2020;127:104370. doi: 10.1016/j.jcv. 2020.104370.  Back to cited text no. 17
Calligaro GL, Lalla U, Audley G. The utility of high-flow nasal oxygen for severe COVID-19 pneumonia in are source-constrained settings: A multi-centre prospective observational study. E Clin Med 2020;28:100570. doi: 10.1016/j.eclinm.2020.100570.  Back to cited text no. 18
Sivieri ME, Eichenwald E, Bakri SM, Abbasi S. Effect of high frequency oscillatory high flow nasal cannula on carbon dioxide clearance in a premature infant lung model: A bench study. Pediatr Pulmonol 2019;54:436-43. doi: 10.1002/ppul.24216.  Back to cited text no. 19


  [Figure 1], [Figure 2], [Figure 3]

  [Table 1]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded24    
    Comments [Add]    

Recommend this journal