No abstract

Acute Respiratory Distress Syndrome (ARDS) caused by COVID-19 is substantially different from ARDS caused by other diseases and its treatment is dissimilar and challenging. As many studies showed conflicting results regarding the use of Non-invasive ventilation in COVID-19-associated ARDS, no unquestionable indications by operational guidelines were reported. The aim of this study was to estimate the use and success rate of Helmet (h) Continuous Positive Airway Pressure (CPAP) in COVID-19-associated ARDS in medical regular wards patients and describe the predictive risk factors for its use and failure. In our monocentric retrospective observational study, we included patients admitted for COVID-19 in medical regular wards. hCPAP was delivered when supplemental conventional or high-flow nasal oxygen failed to achieve respiratory targets. The primary outcomes were hCPAP use and failure rate (including the need to use Bilevel (BL) PAP or oro-tracheal intubation (OTI) and death during ventilation). The secondary outcome was the rate of in-hospital death and OTI. We computed a score derived from the factors independently associated with hCPAP failure. Out of 701 patients admitted with COVID-19 symptoms, 295 were diagnosed with ARDS caused by COVID-19 and treated with hCPAP. Factors associated with the need for hCPAP use were the PaO2/FiO2 ratio < 270, IL-6 serum levels over 46 pg/mL, AST > 33 U/L, and LDH > 570 U/L; age > 78 years and neuropsychiatric conditions were associated with lower use of hCPAP. Failure of hCPAP occurred in 125 patients and was associated with male sex, polypharmacotherapy (at least three medications), platelet count < 180 × 109/L, and PaO2/FiO2 ratio < 240. The computed hCPAP-f Score, ranging from 0 to 11.5 points, had an AUC of 0.74 in predicting hCPAP failure (significantly superior to Call Score), and 0.73 for the secondary outcome (non-inferior to IL-6 serum levels). In conclusion, hCPAP was widely used in patients with COVID-19 symptoms admitted to medical regular wards and developing ARDS, with a low OTI rate. A score computed combining male sex, multi-pharmacotherapy, low platelet count, and low PaO2/FiO2 was able to predict hCPAP failure in hospitalized patients with ARDS caused by COVID-19.