HFNC is now recommended by French experts group for ICU management of patients infected with SARS-CoV-2 to reduce the need for invasive mechanical ventilation (6). This technology is useful in non-hypercapnic hypoxemic pneumonia. It is a simple technique to set up and monitor. The greatest risk of HFNC might be excess mortality due to delayed intubation in case of insufficient monitoring, requiring its use in ICU (7). Because our study population was not eligible for intensive care, HFNC seemed interesting to try in our non-critical care ward.
Continuous Positive Airway Pressure (CPAP) was previously used in COVID-19 patients in UK hospitals due to the rapidly increasing number of critically ill patients and intensive care shortage (8). Patients were younger than our population (median age 52) and over half of them (14/24 (58%)) did not subsequently require intubation. Another study which compared CPAP to standard oxygen therapy (including HFNC) for patients who were not suitable for ICU showed an extremely high mortality in both groups (9). A clinical trial is still ongoing to identify which respiratory support may be more beneficial for COVID-19 patients between three ventilation methods: CPAP, HFNC and standard treatment which involves oxygen delivered via a normal face mask or tubes in the nose (10).
The main limit of our study is its observational design without randomization between different oxygen delivery strategies. Few numbers prevent us to make comparison between people who died and people who survived. A previous study in the same hospital during the first wave showed that among 296 hospitalized patients, 30 of them older than 70 (median age 83, IQR 79–86), were treated with corticosteroids and required oxygen support for at least 9 L/mn, among whom 9 (30%) survived. When necessary oxygen support was 12 L/mn or more, 6/25 patients (24%) treated with corticosteroids survived (11). Our patients were older (median age 88, IQR 82–91) than people from this previous study. Further studies with careful selection of patients using the Clinical Frailty Scale (12) could help to better identify those who might benefit from treatment with HFNC.
High-flow oxygen therapy is also associated with better comfort and oxygenation than standard oxygen therapy delivered through a facemask, with a reduction in the severity of dyspnea and a decreased respiratory rate. These findings might result from the heating and humidification of inspired gas, which prevents thick secretions and subsequent atelectasis but also from low levels of PEEP (positive end-expiratory pressure) generated by a high gas flow rate and flushing of upper-airway dead space (7). This increased comfort might explain why palliative therapeutic were of little need in our study. This element of comfort is also an important help for the emotional adaptation required for the health care workers as management of death and its conditions appear as a major and central difficulty, reducing emotional fatigue, depression, anxiety and stress (13).
Indication for starting HFNC for COVID-19 patients was decided based on clinical experience. At the beginning of the wave in September 2020, we waited until oxygen standard flow reached 15 L/min to set up HFNC. We subsequently tended to propose HFNC at earlier stage when oxygen needs increased rapidly and/or when respiratory rate remained high (> 25 cycles/mn) despite oxygen support, because we hypothesized that an earlier initiation could lead to better results before respiratory exhaustion. Indeed, in the sub-group of patients initiating HFNC as soon as oxygen standard flow reached 9 L/min, survival rate increased from 32% (10/31) to 67% (10/15).