Background: Non-invasive respiratory support including high-flow nasal oxygen (HFNO), and continuous positive airway pressure (CPAP) have been used to provide therapy in selected SARS-CoV-2 patients with acute respiratory failure (ARF). The value of the ROX index, a validated benchmark for outcomes in HFNO is unknown in CPAP.
Objective: Can the ROX, a validated benchmark in HFNO be used for measuring treatment outcomes of CPAP in SARS-COV-2 ARF?
Study Design and Methods: A non-randomised prospective protocol driven observational non-intensive care unit study in 130 SARS-COV-2 patients with ARF treated with non-invasive therapy from March 2020 to January 2021. The primary end point was failure of therapy (death or escalation). Secondary outcomes included time to failure including invasive mechanical ventilation (IMV) or death, the effect of escalation to CPAP from HFNO and the utility of ROX in ARF.
Results: HFNO was better than CPAP in treating SARS-COV-2 ARF: 17/35 (48.5%) with successful HFNO therapy versus 24/95 (25.2%) with CPAP. The ROX index was more sensitive to outcomes with CPAP compared to HFNO and distinguished treatment failure early at 1, 4, 6, 12, and 24 hours with the highest sensitivity at 24 hours (ROX-24h). The AUC for the ROX-24h was 0.77 for HFNO (P<0.0001), and 0.84 for CPAP (P<0.0001). The ROX-24h cut-points predicted failure with HFNO when < 3.9 (PPV 71%, NPV 75%) and CPAP < 4.3 (PPV 75%, NPV 91%). For success, ROX-24h cut-points of 7.6 for HFNO (PPV 85%, NPV 48%) and 6.1 for CPAP (PPV 88%, NPV 62%) were observed. Escalation from HFNO to CPAP was mostly not successful.
Conclusion: ARF in SARS-COV-2 can be successfully managed by non-invasive support. The ROX index, validated for HFNO, provides a timely, low resource measure for both HFNO and CPAP avoiding delayed intubation.
Trial registration: Study approved by NHS HRAREC (20/HRA/2344;ethics 283888)