Consecutive patients with hypoxemic respiratory failure, CT evidence of bilateral pneumonia and diagnosis of Coronavirus Disease 19 (COVID-19) with molecular nasopharyngeal swab consecutively admitted to the COVID Care Unit of the "Santa Maria delle Grazie" Hospital were enrolled. All the patients with clinical indication for Continuous Positive Airway Pressure (CPAP) were randomized 1:1 into two groups: Group A received a fixed Positive End Expiratory Pressure (PEEP) of 10 centimetre of water (cmH2O), Group B underwent the PEEP trial to identify the optimal PEEP (defined as the highest value that preceded the appearance of the "lung pulse" at lung ultrasound and that determined an increase in PaO2/FiO2 by at least 20%). Primary endpoint was defined as a composite in-hospital mortality+intubation, secondary endpoint was the improvement of PaO2/FiO2. As safety indicator, the incidence of pneumothorax was collected.

Consecutive COVID-19 patients affected by Acute Respiratory Syndrome (ARDS) admitted to COVID Care Unit of "Santa Maria delle Grazie" Hospital, Pozzuoli (Naples, Italy), were prospectively enrolled in the study. Inclusion criteria were: age > 18 years, diagnosis of Severe Acute Respiratory Syndrome-Coronavirus-2 (SARS-CoV-2) infection confirmed by means of a nucleic acid test by real-time reverse transcription polymerase chain reaction (RT-PCR) on specimen from nasopharyngeal swab, bilateral interstitial pneumonia at CT scan; hypoxemic respiratory failure defined as PaO2 (partial pressure of oxygen) to FiO2 (fraction of inspired oxygen) ratio < 200.

As exclusion criteria were considered: type 2 respiratory failure; hemodynamic instability requiring vasopressors and/or inotropes support; CT evidence of pneumothorax/pneumomediastinum; concomitant acute medical disease (stroke; myocardial infarction; septic shock; gastrointestinal bleeding; acute renal failure requiring dialysis) who could affect the prognosis independent from COVID - 19.

All the patients enrolled underwent clinical history, physical examination, laboratory tests, arterial blood gas analysis, lung ultrasound, chest CT. During the hospital stay, the patients were treated according to the standard and experimental protocol available; therapies administered, and collateral effects were collected.

All participants gave their written informed consent. All the patients eligible to receive CPAP were consecutively 1:1 randomized to PEEP trial (Group A) or standard treatment (Group B).

In the Group A, PEEP was increased from 5 to 7.5 and 10 cmH20 in 30-minute steps during which lung ultrasound and blood gas analyses were repeated. PEEP responsiveness was evaluated comparing ultrasound findings and arterial blood gas parameters during oxygen supplementation with helmet-CPAP at different PEEP levels (5, 7.5 and 10, respectively), maintaining the same FiO2. A PEEP-responder is defined as a subject with clinical and/or arterial blood gases improvement with helmet-CPAP without signs of lung hyperinflation. The best PEEP was defined as the PEEP value before the appearance of lung pulse and with a PaO2/FiO2 levels stable or major than 20% in comparison to the basal value. In the Standard practice group CPAP was set at 10 cmH2O for all patients and a blood gas analysis was performed after 2 hours.

Endotracheal intubation was considered in presence of impaired respiratory mechanics, worsening of respiratory acidosis, decreased mental status.

Primary outcome was defined by a composite of hospital mortality and need of endotracheal intubation. Secondary outcome was considered the effect of CPAP on PaO2/FiO2 and respiratory rate. As a safety indicator the incidence of pneumothorax was considered.

Inclusion Criteria:

• age > 18 years;
• diagnosis of SARS-CoV-2 infection confirmed by means of RT-PCR on specimen from nasopharyngeal swab;
• bilateral interstitial pneumonia at CT scan;
• hypoxemic respiratory failure defined as PaO2/FiO2 < 200 at blood gas analyses.

Exclusion Criteria:

• type 2 respiratory failure;
• hemodynamic instability requiring vasopressors and/or inotropes support;
• CT evidence of pneumothorax/pneumomediastinum;
• concomitant acute medical disease (stroke; myocardial infarction; septic shock; gastrointestinal bleeding; acute renal failure requiring dialysis)