Moreover, meal-related desaturation tended to be more severe in those least able to tolerate such a change, ie, these with a lower baseline SpO2. This contrasts with the present study where we found no difference in the magnitude of meal-related desaturation in patients with more severe versus less severe disease. However, the difference probably relates to the level of baseline SpO2. In the study by Brown et al , the ‘more severe’ subgroup had a mean baseline SpO2 of 83.6% whereas our ‘more severe’ COPD patients had a mean baseline SpO2 of 90.2% (AE) and 90.3% (PE). Schols et al , who also reported greater meal desaturation in hypoxemic compared with normoxemic patients with COPD, may have reconciled these differences by emphasizing the greater vulnerability of the former individuals owing to the sigmoidal shape of the oxygen dissociation curve.
There are severe possible mechanisms for meal-related desaturation in patients with COPD. The ultimate explanation is probably multifactorial, and must explain the small but significant desaturation during PE in this study and even with nasogastric feeding as shown by Brandstetter et al . As well, one must account for the greater effect noted during AE. A decrease in minute ventilation as a result of interrupted breathing related to chewing and swallowing is unlikely because previous studies have shown that desaturation during eating is not associated with an increase in partial pressure of carbon dioxide .