Since December 2019 and the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) outbreak in China, a large number of cases have been reported worldwide. Although the majority of patients show mild symptoms, coronavirus disease 2019 (COVID-19) was responsible for severe illness characterized by progressive hypoxemic respiratory failure. Some patients had rapidly developed acute respiratory distress syndrome (ARDS) and other systemic complications followed by multiple organs failure and need for invasive mechanical ventilation and intensive care unit (ICU) admission [1-3]. In patients who are intubated for moderate to severe ARDS, prone position (PP) was proposed as an adjuvant non respiratory tool allowing to improve oxygenation and reduce mortality. Its benefit was mainly reported highly hypoxemic patients [4]. Recently, awake PP has been proposed to enhance oxygenation in non-intubated patients with ARDS even those with severe COVID-19 [5] and it has been recently incorporated into guidelines and expert consensus statements [6,7] and has been identified as a research priority by the Surviving Sepsis Research Committee [8]. In this context, we aimed to assess if awake PP reduces the rate death or intubation in patients with severe COVID-19 acute hypoxemic respiratory failure who require respiratory support.

Our study confirmed that awake PP can improve severe hypoxemia and respiratory failure in COVID-19 patients with spontaneous breathing. In fact, awake PP in these patients was associated with significant reduce in need to invasive mechanical ventilation and mortality compared with SP.

This positive effect seems to be related to direct relationship on chest compliance, gas exchange and hemodynamic effects. In fact, overall compliance of the chest wall is influenced by the flexibility of its three anatomical boundaries: anterior, posterior, and abdominal. In the SP, changes in compliance are most influenced by the abdomen and anterior chest wall, whereas in the PP, the posterior chest and abdomen are key determinants [7]. In ARDS patients, lung compliance is mainly determined by the degree of ventilation opening of the lungs. Notably, specific lung compliance was similar in ARDS patients and healthy individuals, suggesting that surfactant changes or early fibrosis were not major in altering lung mechanical properties [8]. It can be seen that any change in lung compliance is mainly due to the opening of new lung units and/or the mechanical improvement of the opened units to achieve a more favorable position on the volume-pressure curve [9]. In the PP, this beneficial displacement can occur by promoting a uniform distribution of total stress and strain [10]. The most striking change observed on CT scans when changing from supine to PPs is a dorsal to ventral redistribution of density [11]. Impact of PP on improving gas exchange is also determined by several effects. The first element is the amount of tissue that is open to ventilation and perfusion during the respiratory cycle. If recruitment of the dorsal lung exceeds that of the ventral lung, oxygenation should improve with unchanged perfusion distribution. The second element is the homogeneity of inflation. In homogeneity is related to the uneven distribution of ventilation. Since perfusion remains nearly constant, more uniform ventilation results in a more uniform ventilation to perfusion ratio distribution, which is reflected in reduced venous admixture and reduced dead space. Local changes in chest wall compliance may also contribute to improved oxygenation [10]. Due to the reduced compliance of the anterior chest wall and the curvature of the diaphragm, the distribution of tidal volumes actually shifts toward the posterior paraventral region of the lungs, where ventilation is usually absent in the SP. This improvement is a result of reduced heterogeneity in shunt and ventilator perfusion, as the anatomically cone-like lung fits into its cylindrical thorax, deforming less when the patient is prone or supine [11]. This reduces atelectasis in the dorsal lung region, where shunts are preferentially distributed in ARDS [12].

Although expert clearly recommend employment of PP in severe ARDS invasively ventilated, its place in critically ill COVID-19 patients non intubated still debated. The streamlined Surviving Sepsis Campaign Guidelines on the operation of grown-ups with coronavirus disease 2019 in the ICU started that there is inadequate substantiation to issue a recommendation on use of awake PP in non-intubated grown-ups with severe COVID-19 [13]. There were several confounding factors, duration, frequency of PP and type of oxygen supplementation ways weren't formalized across all included studies.

In the same way, World Health Organisation COVID-19 Clinical Management Living Guidance suggested a tentative recommendation for awake PP of oppressively ill COVID-19 cases taking supplemental oxygen (including high-inflow nasal oxygen)[14]. Lately streamlined "British Thoracic Society/Intensive Care Society: respiratory care in patients with acute hypoxaemic respiratory failure associated with COVID-19" also suggested PP or side displacing for COVID-19 cases in the respiratory support pathway [15]. A recent expert agreement for the operation of COVID-19 affiliated acute respiratory failure concluded that awake PP considered to ameliorate oxygenation and it should be used when supplemental oxygenation is needed to maintain SpO₂ >90 [16]. Numerous studies had evaluated benefit of awake PP in ARDS COVID-19 [16-20]. Elharrar et al had conducted a prospective, single-center study of PP applied in hypoxemic ARF treated with high-flow oxygen therapy [17]. Among the tolerant cases, PaO₂ increased from 73.6 mm Hg before to 94.9 mm Hg during PP. Only three cases maintained better oxygenation 6–12 hours after resupination [18].

In COVID-19 hospitalized cases other analogous studies have reported advanced oxygenation during PP [18]. A methodical review and meta-analysis evaluated the effect, timing, and populations that might profit from awake proning regarding oxygenation, mortality, and tracheal intubation compared with SP in hypoxemic acute respiratory failure. This study concluded that a PP can ameliorate oxygenation among non-intubated patients with acute hypoxemic respiratory failure when applied for at least 4 hours. Awake proning appears safe, but generalization of this results needs more randomized controlled trials. The results of feasibility and safety studies will potentially help in the design of larger definitive trials to address this crucial knowledge gap [19]. Methodical review and meta-analysis of non-randomized trials demonstrated that PP enhanced PaO₂/FiO₂ with better SpO₂ than SP in COVID-19 cases. It also reported that the mortality rate was lower in those who entered prone ventilation [20].

Efficacy of PP combined with high-flow oxygen therapy remedy or non-invasive ventilation was lately reported in small cohorts ofnon-infectious and contagious non–COVID-19 ARDS cases [21]. Bahloul et al. [22] demonstrated in a prospective experimental study of severe, critically ill adult COVID-19 cases admitted to ICU, that the early operation of PP can ameliorate hypoxemia in severe hypoxemia COVID-19 patients.

This study presents some limitations. First, It was a single-center one. Second, feasibility of PP was not admitted for all patients and finally number of patients was small.

We supposed that awake PP could improves oxygenation during related ARDS COVID-19 non-intubated patients. Due to its relative ease of use, awake PP can reduce need for invasive mechanical and decrease mortality. Association of PP to non-invasive ventilation or high-flow nasal cannula is suggested to improve respiratory status. Randomized controlled trials are needed to confirm the beneficial effects of PP in COVID-19 patients with spontaneous breathing.