Three methods are in current use for measuring the diffusing capacity of the lungs for carbon monoxide (DL). Two of these are steady state methods, and they differ in the methods used to calculate the mean alveolar carbon monoxide tension. Filley, MacIntosh and Wright (1) calculate the alveolar Pco indirectly, assuming the physiological dead space for carbon dioxide to be the same as that for carbon monoxide, while Bates, Boucot and Dormer (2) use the end-tidal Pco as the mean alveolar Pco. The third method is the single breath method of Krogh and Krogh (3) as modified by Ogilvie, Forster, Blakemore and Morton (4). Each method in theory measures the total resistance to gas diffusion offered by the structures that lie between the gas in the alveoli and the hemoglobin within the red cells of the pulmonary capillaries. The results obtained by each method differ in normal subjects; and in disease states, such as emphysema, the differences may be great. While the different results obtained in similar subjects by the two steady state methods can largely be explained by the different methods of estimating the mean alveolar Pco (5), the breath-holding method consistently gives higher results than either of the steady state methods (6). Marshall (7) has recently compared the breath-holding method of measuring DL with a steady state method, using end-tidal samples in a small group of normal subjects and patients with emphysema. He found that, while in normal subjects the end-tidal sample satisfactorily reflected the mean alveolar Pco, in emphysema with impaired intrapulmonary gas mixing, the end-tidal sample overestimates the mean alveolar Pco and therefore gives falsely low results for the diffusing capacity. These results The present study was made 1) to compare the single breath and steady state methods in a large group of subjects with widely varying intrapulmonary gas mixing; and 2) since abnormalities of intrapulmonary gas mixing cannot explain the differences that occur between the two methods in normal subjects, a study was made of the variations in the type of breathing which might influence the results of the steady state method in normal subjects.
METHODSThe steady state method used was that of Bates and co-workers (2) but the apparatus was modified in a number of details. The valve assembly consisted of single inspiratory and expiratory valves set in 2.5 cm-bore metal tubing in order to reduce the dead space and allow end-tidal samples to be obtained with smaller tidal volumes than were possible with the original apparatus. An automatic end-tidal sampler, triggered by a change in pressure at the mouthpiece, snatched a 35 ml sample at the end of each breath. During the test the subject breathed 0.125 per cent carbon monoxide in air; 1.5 minutes was allowed for the subject to reach a steady state and the expired gas was collected for a further 2 minutes during which time end-tidal samples were taken. The CO concentration in the inspired, mixed expired and end-tidal samples was measured by an infrared analyzer....