Nocturnal oxygenation using a pulsed-dose oxygen-conserving device compared to continuous flow

The pulsed-dose oxygen-conserving device (PDOCD) has gained wide acceptance as a tool to reduce the cost and inconvenience of portable oxygen delivery. Despite the widespread use of PDOCDs in awake and ambulating patients, few studies report their use during sleep. This study was designed to compare heart rate and oxygen saturation (measured via pulse oximetry [S(pO2)]) of sleeping patients using one brand of PDOCD versus continuous-flow oxygen.

We studied 10 home-oxygen patients who were using various continuous-flow oxygen systems and prescriptions. Baseline asleep and awake S(pO2) and heart rate were recorded while the patients used their existing home-oxygen systems (liquid oxygen or oxygen concentrator with nasal cannula) and continuous-flow oxygen prescription. Patients were then switched to a nasal cannula connected to a PDOCD. The PDOCD setting was adjusted to produce an S(pO2) equal to the patient's awake baseline on continuous-flow. This setting was then used while the patient subsequently slept. Mean values for S(pO2) and heart rate and hours of sleep were calculated by the software in the oximeter. Mean values for S(pO2) and heart rate were compared with the paired Student's t test.

There was a statistically significant but clinically unimportant S(pO2) difference between the patients who used continuous-flow oxygen and those who used the PDOCD (95.7% vs 93.2%, respectively, p = 0.043). There was no difference in heart rate (77.3 beats/min vs 77.9 beats/min, p = 0.70). The sample size was adequate to detect a difference in heart rate of 5 beats/min at a power of 80%. For the subset of patients whose PDOCD triggering sensitivity was set on sensitive (vs the default lower sensitivity) there was a statistically significant but clinically unimportant S(pO2) difference (continuous-flow 95.6% vs PDOCD 93.2%, p = 0.044). All other comparisons showed no differences, but the samples sizes were too small to make any firm conclusions. One patient experienced an 11% S(pO2) drop with the PDOCD because of an inadequate triggering sensitivity setting.

The PDOCD model we studied was able to deliver oxygen therapy (via nasal cannula) comparable to continuous-flow in 9 of 10 patients. The resting daytime S(pO2) on continuous-flow appears to be an appropriate target for setting the PDOCD to ensure adequate oxygenation, even during sleep, with the PDOCD we tested. We conclude that the PDOCD we tested is able to maintain adequate S(pO2) during sleep in selected patients. Because of differences in design, triggering-signal sensitivity, and oxygen-pulse volume, these results cannot be generalized to all patients or all oxygen-conserving devices. Further research is needed to determine the general performance of PDOCDs on larger populations of oxygen-dependent patients and patients with sleep-disordered breathing.