THE MEASUREMENT OF THE PULMONARY DIFFUSING CAPACITY IN THE PRESENCE OF LUNG DISEASE D. V. Bates J Clin Invest. 1958;37(4):591-605. https://doi.org/10.1172/JCI103642. Research Article Find the latest version: https://jci.me/103642/pdf THE MEASUREMENT OF THE PULMONARY DIFFUSING CAPACITY IN THE PRESENCE OF LUNG DISEASE BY D. V. BATES (From McGill University Clinic, Royal Victoria Hospital, Montreal, Canada) (Submitted for publication September 26, 1957; accepted December 26, 1957) Recent work (1-4) has shown that the rate of MacIntosh, and Wright (5). When it is known, pulmonary gas diffusion is abnormal in a variety however, that there is not only unevenness of of clinical conditions, and its measurement seems distribution of inspired gas within the lung, but certain to become a necessary part of any complete imbalance between ventilation and perfusion, as in evaluation of pulmonary function. The various the case of emphysema, the situation is very much methods of measurement so far described, whether more complicated. It cannot be assumed that in using carbon monoxide or oxygen, are capable of such lungs there is any constant ratio between providing a theoretically valid figure for the dif- the surface area available for diffusion and the gas fusing capacity in normal subjects, since in them volume and ventilation of different parts of the the lung may be regarded as behaving in an ap- lung. It is immediately clear that in patients in proximately uniform manner. A number of pa- whom CO2 retention is present and in whom there pers have been published in which measurements is a gross unevenness of gas distribution within of the rate of gas diffusion in disease have been the lung, the meaning of any single figure for the presented. A steady state Dco technique devised pulmonary diffusing capacity, however deter- by Filley, MacIntosh, and Wright (5) has been mined, will be questionable. Ideally, the pul- used in a study of a variety of patients with pul- monary diffusing capacity should be the sum of monary fibrosis (3, 6). The oxygen method has the individual diffusing capacities for each alveo- been used in studies of sarcoidosis (7), emphysema lus. When the perfusion, diffusion, and ventila- (2), mitral stenosis (8), and a number of other tion are varying widely throughout the lung, the conditions. The single breath modified Krogh true diffusing capacity will be the aggregate sum method developed by Ogilvie, Forster, Blakemore, of a vast number of widely differing numbers. and Morton has been used to study selected cases It seems unlikely that such a figure can ever be of a variety of different clinical conditions (9) and arrived at, since if carried to its logical con- one study has been reported in a small group of clusion it would involve measurements made on patients using radioactive carbon monoxide (10). individual alveolae. It is reasonably certain that A steady state CO method in which the alveolar no overall number for the pulmonary diffusing CO was calculated from an assumed value of re- capacity which is found in these complex disease spiratory dead space has been used in an attempt states will approximate to this ideal and unob- to establish the value of this type of function test tainable figure. The problem, therefore, becomes in the assessment of emphysema (4), and a modi- one of attempted assessment of the meaning of fied CO steady state technique has been used to individual diffusion figures, however obtained, and investigate the causes of disability in pneumo- it is the purpose of this paper to discuss the coniosis (11). validity of steady state measurements of pulmo- Patients in whom there is little or no inequality nary diffusing capacity in lung diseases in which of gas distribution or circulation in the lung do there is known to be gross unevenness in relation not present a major problem, since in them it can to perfusion and ventilation within the lung. Al- be assumed that during a steady state, a reason- though it is not possible, for obvious reasons, to ably representative estimate of the mean alveolar calibrate any particular method against a theoreti- tension of CO can be obtained either by measuring cal ideal figure, enough data have been collected to the end tidal CO tension, or by calculating alveolar permit discussion of the significance of the results CO tension from the measured values of arterial found in a number of different diseases. In par- carbon dioxide tension, as described by Filley, ticular, it is possible to estimate the significance 591 592 D. V. BATES a8- on a closed helium circuit as previously described (13). x The arterial pCO2 has been calculated from the pH measured on a Beckman pH meter at 37.5° C., and 7- * 0O total Co CO, content measured on a Van Slyke apparatus. 0 x A small respiratory circuit was built so that an end UPTAKE 0 6 - 0 tidal sample could be taken during quiet breathing of * 0 0 00 X (ml / min.) 00* room air during the simultaneous collection of a resting X o o o x arterial blood sample. The design of this was exactly 5- similar to the circuit 0 00198 used for the measurement of the 0 STANDARDISED 00 x pulmonary diffusing capacity, but the end tidal gas was 0 passed a oio 80 x x through Cambridge differential catharometer, 4 - the calibration of which had previously shown that the XOX CO2 concentration could be read to TO 01% x approximately 0.05 x per cent. The formulae used in the calculations are 3- 0 given in the Appendix. INSPIRED *X 2A RESULTS (AT REST) The results obtained are shown in Figures 1 to 3 and may be conveniently described under vari- ous subheadings: s5 l0 20 25 30 Rest ml CO/min./mmHg DMnU In a preceding communication (12), it was out FIG. 1 (A). RELATIONSHIP BETWEEN THE CO UPTAKE pointed that in normal subjects at rest the STANDARDIZED TO A CONSTANT INSPIRED GAS PERCENTAGE calculation of a mean alveolar CO using an as- (0.1 PER CENT CO) AND THE PULMONARY DIFFUSING sumed value of respiratory dead space is liable to CAPACITY CALCULATED FROM AN END TIDAL SAMPLE be grossly in error. This fact was appreciated (DcoI1) by Krogh and Krogh (14), and is the result of Although there is a general relationship between these the of the Bohr at low tidal two observations it is clear that the diffusing capacity sensitivity equation could not be adequately calculated from the rate of uptake volumes to changes in the assumed dead space of carbon monoxide alone. All subjects studied at rest. value. If, however, the rate of CO uptake in ml. X = normal subjects; 0 = patients with normal gas per minute is grossly reduced, as it is in emphy- mixing; = patients with uneven gas mixing [helium sema for instance (15), this error becomes less mixing efficiency below 50 per cent (13) ]. important, and an estimate of gas diffusion can be made in patients of this type even when an as- of the results obtained a state using steady carbon sumed value of dead space is used, provided the monoxide method, which is undoubtedly the limitations of this type of measurement are clearly simplest method of measurement to use. recognized. In Figure 1 (a), the CO uptake, standardized METHODS to a constant inspired concentration of 0.1 per cent, The technique used for the measurement of the pul- has been plotted against the DcoI1 (diffusing ca- monary diffusing capacity has already been described pacity calculated from measured end tidal CO (12). The original circuit has been modified by the concentration). It will be realized that the nu- addition of a photoelectric relay to open the electro- merator of a fraction is magnetic valve at the end tidal sampling position. The plotted against the nu- light to the photocell is interrupted by a vane in the merator divided by the mean alveolar CO con- transparent plastic tube on the inspiratory side of the centration. It is clear from this graph that the mouthpiece. An electronic timer is thus actuated by the resting diffusing capacity cannot be judged with beginning of inspiration and opens the end tidal electro- any accuracy from the rate of uptake of CO. A magnetic valve for any chosen time period. In this way figure of 11 ml. CO per minute per mm. the volume of the end tidal sample can be varied, and Hg may is independent of tidal volume. be taken as aproximately the lower limit of nor- The subdivisions of lung volume have been measured mal for a resting Dco by this technique (5, 12). PULMONARY DIFFUSING CAPACITY IN LUNG DISEASE 593 On the basis of this figure a resting uptake of less mation of degree of abnormality. It is of par- than 2.5 ml. CO per minute at an inspired con- ticular interest that there is no evidence of greater centration of 0.1 per cent might be taken to indi- discrepancy between these two calculated values cate considerable impairment of diffusing capacity of diffusing capacity in patients known to have at rest. The rate of uptake of CO at rest is, how- uneven gas distribution in the lungs (Figure 2, ever, greatly influenced by ventilation (15) and solid circles) than there is in patients in whom thus a demonstrably abnormal diffusing capacity the index of helium mixing was normal (Figure 2, may exist in someone with a rate of uptake of CO open circles). The general agreement between as high as about 4.0 ml. CO per minute per mm.