The present results in humans are in agreement with results obtained in conscious dogs, where neither acute nor chronic (six days) hypoxia affected theophylline disposition. However, these results differ from those obtained in a rabbit study, where it was shown that short term hypoxia (mean Pa02 of 55 mmHg) significantly decreased theophylline clearance. This discrepancy may be due to species differences or to protocol differences (in the rabbits, hypoxia was acute and of short duration and in addition, the rabbits had slight respiratory alkalosis). The discrepancy may also have been caused by other factors related to the patients chronic respiratory disease, which may have influenced theophylline disposition. It is unlikely, however, that our results were biased by the experimental protocol (eg, blood sampling for only six hours), as the estimated theophylline kinetic parameters are in total agreement with those reported in the literature.

It is interesting to note that in our patients, oxygen therapy slightly increased the PaC02 with no noticeable decrease in arterial pH. The theophylline apparent volume of distribution was not increased by these slight changes. However, the changes in PaC02 were inversely related to theophylline clearance values (Fig 3). Keeping in mind that the average theophylline clearance was slightly lower in subjects with a high PaC02 (Fig 2), we are tempted to speculate that, in man, primarily hypercapnia and/or the pathology associated with hypercapnia is able to influence theophylline clearance. in detail
Our results agree with those reported by Wester-field et al, and by Cusack et al, who, using a different approach, were unable to demonstrate that hypoxia could reduce the rate of biotransformation of theophylline. On the other hand, in their patients, the individual PaC02 values did not correlate with theophylline clearance. This could be due to the fact that their patients had multiple additional conditions, including cor pulmonale, congestive heart failure, pneumonia, and liver disease, all of which are capable of reducing theophylline clearance. In addition, these patients were receiving several drugs which could alter the disposition of theophylline. Thus, when numerous factors are present within the same patient, which are capable of influencing theophylline clearance, correlations between changes in arterial blood gases and the values of theophylline apparent clearance may be masked.
In summary, in patients with COLD and severe hypoxia, the biotransformation of theophylline does not appear to be affected by hypoxia, since oxygen therapy did not alter significantly the clearance of theophylline. Our results do not support the hypothesis that in man, hypoxia affects the oxidation or the demethylation of theophylline. Therefore, the decrease in theophylline clearance reported by many investigators may be due instead to other factors such as inflammation secondary to viral- or bacterial processes, liver disease, or perhaps hyper-capnia.