Long-term oxygen therapy 24 or 15 hours/day and outcomes: a multicenter, registry-based, randomized, clinical trial.

<bold>Background:</bold> Long-term oxygen therapy (LTOT) improves survival in severe hypoxemia and is recommended 24h/day based on non-randomized data. However, LTOT 24h/day is burdensome and recent observational data indicate no difference in patient-related outcomes for 24 versus 15h/day. We tested the hypothesis that LTOT prescribed 24 versus 15h/day would not reduce risk of hospitalization or death at one year. <bold>Methods:</bold> In this multicenter, registry-based, randomized, clinical trial using the Swedish Registry for Respiratory Failure, patients starting LTOT were randomized (in a 1:1 ratio) to LTOT 24 or 15h/day. Primary outcome was time to all-cause hospitalization or mortality at one year. Secondary outcomes were hospitalizations and mortality (all-causes, respiratory, or cardiac disease) at three and twelve months, in all patients and stratified according to severity of baseline hypoxemia and underlying condition (chronic obstructive pulmonary disease or other). <bold>Results:</bold> Between May 2018 and April 2022, 241 patients were randomized to LTOT 24h/day (n=117) or 15h/day (n=124). No patient was lost to follow-up. Compared to LTOT 15h/day, LTOT 24h/day did not improve the primary outcome of hospitalization or death up to one year (hazard ratio, 0.99; 90% confidence interval, 0.76–1.29; non-superiority p=0.007), or any secondary outcomes overall or in any sub-group. Self-reported adherence to the randomized treatment was high (for 24h/day: median 24.0 [interquartile range, 21.0–24.0]; for 15h/day: median 15.0 [15.0–16.0] h/day) at twelve months. <bold>Conclusion:</bold> LTOT prescribed for 24h/day compared with 15h/day did not decrease the risk of hospitalization or death up to one year.