Interventions
After the initial screening, all patients were randomized into either of the two study groups: patients in the P5 group received 5 cmH2O PEEP during HCPAP therapy within 24 hours from PICU admission, and patients in the P10 group received 10 cmH2O PEEP. HCPAP was initially set with a gas of flow rate of 50 L/min and a fraction of inspired oxygen (FiO2) level of 0.5. Clinicians could proceed with endotracheal intubation if more than two of the following clinical signs worsened after initiation of HCPAP: thoracic or intercostal efforts with patients’ discomfort on attending clinician evaluation, increase in BSS, pH <7.10, PaO2/FiO2 <100, rising of lactate level on two consecutive blood gas analyses. If clinicians indicated to escalate the support of a patient in the P5 group before endotracheal intubation (i.e., an increase in PEEP to a level of 10 cmH2O), as well as they indicated to de-escalate from P10 to P5 (i.e., due to intolerance of the patient) this was considered a protocol violation, but the patient was still analyzed in the original group per intention-to-treat. Vital parameters (pH, respiratory rate, heart rate, PaO2/FiO2ratio, and partial pressure of carbon dioxide -PaCO2-) were recorded at the start of HCPAP, and after 1, 12, 24 and 48 hours. Outcomes were recorded until the patient was discharged from the PICU. All enrolled patients were managed according to a standard treatment protocol for severe bronchiolitis (Supplementary Table 1). A nasal polymerase chain reaction swab for the research of respiratory viruses and common bacteria was performed at admission in all patients. HCPAP was discontinued to HFNC if oxygen saturation was kept above 95% for more than 24 hours with a FIO2<0.3.