Late Breaking Abstract - High-Flow Nasal Oxygen vs. Conventional Oxygen Therapy in severe COVID-19 patients and a do-not-intubate order: a retrospective multi-cohort study

Introduction High-flow nasal oxygen (HFNO) is often used for hypoxemia in patients with COVID-19 and a do-not-intubate (DNI) order. However, its clinical effect compared to conventional oxygen therapy (COT) in this population remains largely unexplored. We hypothesize that HFNO use is associated with improved survival rates and aim to determine its clinical effect in hypoxemic COVID-19 patients with a DNI order. Methods In this sub-analysis of three multicenter cohort studies, we included patients with a DNI order, confirmed SARS-CoV-2 infection, and either HFNO or COT as ceiling oxygen support. The primary endpoint was in-hospital mortality (IHM), and secondary endpoint was hospital length of stay in days. Multivariable logistic regression was used, adjusting for pre-selected confounders (age, Clinical Frailty Score (CFS), dexamethasone treatment, and oxygen saturation, FiO2 and respiratory rate before (potential) start of HFNO). Results Between March 2020-September 2021, 116 patients received HFNO and 110 received COT. Baseline differences were age (median [IQR] 75 [69-80] (HFNO) and 80 [75-85] (COT), p<0.001) and CFS distribution (p<0.001). IHM was 64% for HFNO and 71% for COT (p=0.288). The adjusted odds ratio for IHM was 0.722 [95% CI 0.339-1.538], p=0.40. Hospital length of stay was 11 [6-18] for HFNO, and 7 [4-12] for COT (p<0.001). Conclusion When considering HFNO for frail patients with severe COVID-19, its lack of survival benefit and prolonged hospital stay should be taken into account, critically raising the need for future prospective controlled trials also assessing cost-effectiveness and patient reported outcome measures.