A novel device for conservation of supplemental oxygen: A randomized, crossover study of the efficiency and non-inferiority of the OXFO System

# Background Oxygen is critical to the prevention of hypoxemia-induced morbidity and mortality; yet, access remains inadequate in most low- and middle-income countries despite billions of dollars spent over decades to increase supply. We present a new method for supplemental oxygen delivery whereby the OXFO System (OXFO Corporation, Boston, MA, USA) converts pressurized oxygen to ambient pressure for patient inhalation on demand at a fixed concentration across fluctuations in respiratory rate or tidal volume. This study's primary aims were to examine the oxygen volume savings and test the non-inferiority of mean peripheral oxygen saturation (S~pO2~) with the OXFO System (OXFO Condition) compared to conventional continuous flow (Standard Condition) in a clinical population of oxygen-dependent hospitalized adults. # Methods We conducted a non-inferiority, single-visit, randomized, crossover design. A total of 20 subjects were randomized to receive either OXFO Condition or Standard Condition first. Subjects were administered oxygen for 20 minutes during each condition after oxygen was titrated to target therapeutic S~pO2~. A one-sample t-test was performed to compare the mean ratio of oxygen volume dispensed (OXFO Condition/Standard Condition) with 100%. A paired t-test for non-inferiority was used to test the hypothesis that the mean S~pO2~ for OXFO Condition is no worse than 2.8% lower (absolute difference) than the mean S~pO2~ for Standard Condition. # Results Nineteen subjects were included in the analyses. When considering the dispensing of oxygen as a ratio (OXFO Condition / Standard Condition) for each subject, the mean was 7.7% (standard deviation(SD)=5.4%; range=0-20%) and this was significantly different from 100% (*P*\