Performance of a reservoir nasal cannula (Oxymizer) during sleep in hypoxemic patients with COPD.

STUDY OBJECTIVE: To determine whether a reservoir nasal cannula (RNC) (Oxymizer) provides an arterial hemoglobin oxygen saturation as measured by pulse oximetry (SpO2) equivalent to that provided by the standard nasal cannula (SNC) during sleep in hypoxemic patients with COPD while reducing oxygen flow requirement and cost.

DESIGN: The study took place in a sleep laboratory for three nights, with the first night for acclimatization to the new sleeping environment. In a repeated-measures design, on the second and third nights, subjects used the SNC for one night and the RNC on another night. The order in which they received the two devices was counterbalanced.

SUBJECTS: The subjects were patients with COPD who had a stable PaO2 of 55 mm Hg or less or had a value of 56 to 59 mm Hg with evidence of cor pulmonale or polycythemia (or both) and an FEV1/FVC of less than 70 percent.

INTERVENTIONS: A pulse oximeter was used to measure SpO2. An arterial blood gas measurement was taken on each night while the patients with COPD were receiving oxygen therapy via the assigned device. An EEG machine was used to record measurements of electro-oculography, chin electromyography (EMG), anterior tibialis EMG and EEG.

MEASUREMENTS AND MAIN RESULTS: There was a statistically significant difference between mean SpO2 during sleep (RNC, 91 percent; SNC, 93 percent; F = 7.89; p = 0.01). Nocturnal SpO2 was less than 90 percent for 24.2 percent of the time with the RNC and for 17.5 percent of the time with the SNC (F = 5.41; p = 0.03), but there was no significant difference in the amount of time that SpO2 was less than 85 percent. Compared to the SNC, in 4 of 26 patients with COPD, the RNC performed better; in 12 patients with COPD, the RNC performed the same, and in 10 patients with COPD the RNC performed worse during sleep. Sleep parameters were not significantly different between the two devices.

CONCLUSIONS: The difference of 2 percent in mean SpO2 is within the range of SpO2 measurement error. Therefore, the two devices are equally effective when the sample is considered as a whole. Nighttime oximetry is necessary prior to prescription, since nighttime efficacy of the RNC cannot be predicted on the basis of daytime pulse oximetry.