On the determination of the physiologically effective pressures of oxygen and carbon dioxide in alveolar air.

Although studies of the composition of alveolar air have been invaluable to the advance of respiratory physiology a precise definition of alveolar gas pressures and the accurate sampling of alveolar air have proved elusive goals (1). Owing to the cyclic nature of the ventilatory process the partial pressures of 02 and CO2 in the alveoli are changing continually (time factor) (9), and, owing to inequalities in intrapulmonary ventilation and circulation, the partial pressures in different parts of the lungs may differ significantly (space factor) (10, 15). Therefore, the concept that a single value represents the partial pressure of an alveolar gas requires the assumption that the spot sample of alveolar air be representative with respect to both time and space. There are cogent reasons for believing that the two principal methods for sampling alveolar air (the single complete expiration method of Haldane and Priestley and the fractional sampling technic of Sonne and Nielsen) do not guarantee that the sample always is, in fact, representative. Neither method insures that the sample contains proportional contributions of alveolar air from all portions of the lung (space error), nor that the sample obtained has not lost 02 and gained CO2 during the brief period of stasis within the alveoli (time error). For example, the partial pressures of 02 and CO2 in samples of alveolar air obtained by the Haldane-Priestley technic vary with respect to the timing of the expiratory effort (end-inspiration or end-expiration) (5, 6). And again, fractional sampling of the alveolar air by the Sonne-Nielsen technic has yielded evidence that successive samples of alveolar air taken at different stages during expiration vary appreciably with respect to gaseous composition (14). Despite these limitations many fundamental contributions to an understanding of respiratory mechanisms have come from studies of the composition of alveolar air at rest (4, 7). However, during even moderate exercise the rate of evolution of CO2 into and the escape of 02 out of the alveoli may be increased tenfold or more so that the slight delay necessary to expel the alveolar sample is sufficient to permit radical changes to develop. This failure of the direct sampling technics to provide reliable data during exercise has led us to measure alveolar gas pressures by an indirect method now to be described. The indirect measurement of alveolar gas pressures. Indirectly determined alveolar CO2 and 02 pressures are calculated from the arterial pCOz and the