Abstract

ABSTACTWith an ear oximeter we measured the arterial oxygen saturation (SaO2) of 33 patients with stable chronic obstructive pulmonary disease (COPD) who were breathing room air (a) under baseline conditions and (b) while eating lunch. Data from each patient were analyzed in two ways: The average baseline SaO2 was compared with the average meal SaO2 to ascertain the time-weighted overall effects of eating on saturation; the lowest SaO2 during the baseline period was compared with the lowest SaO2 during the meal in order to evaluate the magnitude of possible short-duration periods of desaturation, which would not be evident from examining only the average data. Although the average baseline SaO2 for the group (mean ± SD) of 87.4% ± 5.1% decreased by only 1.8% during meals to an average of 85.6% ± 6.45% (P < .001), there was a subgroup of patients (7/33 or 21%) who experienced significant (4% or more) meal desaturation. The group mean of the lowest baseline SaO2 observed was 85.8% ± 5.5% and decreased by 2.5% to a minimal meal SaO2 of 83.8% ± 7.2% (P < .001); when analyzed in this way, a similar subgroup (9/33 or 27%) of patients had significant meal-related desaturation. The magnitude and incidence of desaturation increased with decreasing baseline SaO2. No patients with a baseline SaO2 of >90% experienced significant desaturation; three of 16 patients (19%) with a baseline SaO2 of between 85% and 90% and five of nine patients with a baseline SaO2 of < 85% experienced significant desaturation, by up to 5% and 7%, respectively. The severity of obstructive disease had only a weak correlation with the magnitude of desaturation. We conclude that a subgroup of patients with severe COPD experiences significant arterial oxygen desaturation during meals and that this desaturation occurs more commonly in patients with lower baseline saturation.