A Randomized Crossover Pilot Study Comparing the Efficacy of an Auto-demand Oxygen Delivery System with That of a Conventional Demand Oxygen Delivery System in Patients with Chronic Respiratory Failure

Introduction: : When using portable oxygen, a demand oxygen delivery system (DODS), which senses the beginning of inhalation and delivers a bolus of oxygen, is often used. However, conventional DODS may not supply sufficient oxygen when reduced tidal flow fails to trigger the flow sensor. Recently, "auto-DODS," which detects the negative pressure of inhalation and switches among 3 trigger sensitivity levels (standard, high, and extra high), has been developed to improve the efficacy of oxygenation. An auto-DODS can also supply pulsed-flow oxygen when it detects apnea, whereas a conventional DODS has only standard sensitivity. This randomized, open-label, crossover pilot study compared the performance of an auto-DODS with that of a conventional DODS.

Methods: : We recruited patients with chronic obstructive pulmonary disease (COPD) or interstitial pneumonia receiving long-term oxygen therapy. Interventions were performed on 2 different days for each participant. On each day, an auto-DODS or a conventional DODS were tested at rest for 30 minutes and during the 6-minute walk test. The primary outcome was mean oxygen saturation (SpO2). Secondary outcomes were the ratios of time for each sensitivity level and pulsed-flow oxygen when using the auto-DODS, total time desaturated below SpO2 90%, percentage of time desaturated below SpO2 90%, minimum SpO2, mean and maximum pulse rate, six-minute walk distance, recovery time after 6-minute walk test, modified Borg scale, comfort, and discomfort index.

Results: : When using the auto-DODS at rest, a high or extra high sensitivity level was observed in addition to standard sensitivity in 6 of 8 participants. During the 6-minute walk test, only standard sensitivity was observed in 6 participants. Mean SpO2 differences between the auto-DODS and conventional DODS at rest and during the 6-minute walk test were -0.6 [-4.5, 3.4] and 0.0 [-2.5, 2.5] ([95% confidence interval]), respectively, neither of which were significant (P = .73 and P = .99). There were no significant differences in secondary outcomes. There were no adverse events when using the auto-DODS.

Conclusions: : This study showed that the auto-DODS did not show superiority in oxygenation either at rest or during exercise compared to a conventional DODS. The auto-DODS was shown to supply oxygen safely and detect inhalations with various trigger sensitivities.