We read with great interest the letter in which the authors talk about the deleterious effect of using a high fraction of inspired oxygen (FiO2) in patients with acute respiratory distress syndrome (ARDS) due pneumonia caused by SARS-CoV-2 where they advocate for the use of continuous positive airway pressure (CPAP) to reduce the high FiO2 used in high-flow nasal oxygen (HFNO) therapy.1

We agree with the authors when they talk about planning CPAP or non-invasive ventilation (NIV) as an alternative to HFNO. However, clinical practice guidelines say otherwise and they recommend HFNO and choose invasive mechanical ventilation (IMV) in cases of failed HFNO.2 Based on these recommendations,2 a Spanish multicenter registry (n=876 patients) showed a wide use of HFNO (49%) vs NIV, and CPAP (<5% in both modalities). We should mention the high rate of failure in the HFNO group (60%) with mortality rates > 30% in patients who required intubation.3

The efficacy of NIV in SARS-CoV-2-induced pneumonia was demonstrated in an Italian multicenter registry (N=110 patients) that compared the NIV-helmet to HFNO. Although the registry primary endpoint—ventilation-free days (20 days vs 18 days, P=.26)—was not achieved the NIV group had lower rates of intubation on day 28 vs the HFNO group [28% vs 51%; OR, 0.37 (0.17−0.82), P=.02]. At the same time, the NIV-helmet had better oxygenation, and less dyspnea compared to the HFNO group.4 Consistent with one of the arguments that support the use of HFNO,2 the patients’ tolerance was higher with HFNO compared to NIV.4 Similarly, in our series of 27 hypoxemic patients due to SARS-CoV-2-induced pneumonia, NIV was used as the first-line therapy in 21 patients (80.8%) basically with NIV specific ventilators in CPAP mode. CPAP failed in 10 patients (48%) with an associated mortality rate of 50%. No health personnel became infected in relation to the NIV as opposed to what has been reported in the clinical practice guidelines.2

Therefore, based on these guidelines,2 we believe that the HFNO plays a key role in the early ventilatory therapy of hypoxemic patients. Unfortunately, the rate of failure in hypoxemic patients is high,3 which may have overestimated the true efficacy of HFNO in advanced stages of ARDS. Like the authors say, a high FiO2 in HFNO added to the pulmonary damage caused by the virus worsens ARDS, and eventually leads to IMV.1 On the contrary, results from the latest clinical trials,3 and observational studies4 pave the way for the safe use of NIV in its different modes (CPAP or NIV) by applying positive end-expiratory pressure (PEEP) that recruits the damaged lung, which allows reaching a non-detrimental FiO21 while avoiding IMV and its deleterious effect.3,5