J Appl Physiol 99: 1985-1991, 2005.
First published July 14, 2005; doi:10.1152/japplphysiol.00348.2005
8750-7587/05 $8.00
INNOVATIVE METHODOLOGY
An open-circuit method for determining lung diffusing capacity during exercise: comparison to rebreathe
Eric M. Snyder,1
Bruce D. Johnson,1 and
Kenneth C. Beck2
1Department of Internal Medicine, Division of Cardiovascular Diseases, Mayo Clinic and Foundation, Rochester, Minnesota; and 2Physiological Imaging, Department of Radiology, University of Iowa, Iowa City, Iowa
Submitted 25 March 2005
; accepted in final form 10 July 2005
 |
ABSTRACT
|
|---|
To avoid limitations associated with the use of single-breath and rebreathe methods for assessing the lung diffusing capacity for carbon monoxide (DLCO) during exercise, we developed an open-circuit technique. This method does not require rebreathing or alterations in breathing pattern and can be performed with little cognition on the part of the patient. To determine how this technique compared with the traditional rebreathe (DLCO,RB) method, we performed both the open-circuit (DLCO,OC) and the DLCO,RB methods at rest and during exercise (25, 50, and 75% of peak work) in 11 healthy subjects [mean age = 34 yr (SD 11)]. Both DLCO,OC and DLCO,RB increased linearly with cardiac output and external work. There was a good correlation between DLCO,OC and DLCO,RB for rest and exercise (mean of individual r2 = 0.88, overall r2 = 0.69, slope = 0.97). DLCO,OC and DLCO,RB were similar at rest and during exercise [e.g., rest = 27.2 (SD 5.8) vs. 29.3 (SD 5.2), and 75% peak work = 44.0 (SD 7.0) vs. 41.2 ml·min–1·mmHg–1 (SD 6.7) for DLCO,OC vs. DLCO,RB]. The coefficient of variation for repeat measurements of DLCO,OC was 7.9% at rest and averaged 3.9% during exercise. These data suggest that the DLCO,OC method is a reproducible, well-tolerated alternative for determining DLCO, particularly during exercise. The method is linearly associated with cardiac output, suggesting increased alveolar-capillary recruitment, and values were similar to the traditional rebreathe method.
carbon monoxide; gas exchange; lung surface area
THE DIFFUSING CAPACITY OF the lungs for carbon monoxide (DLCO) is reduced in a number of disease states (e.g., emphysema, heart failure, interstitial lung disease), and a low value measured at rest may be predictive of exercise intolerance and ventilatory inefficiency (increased minute ventilation/CO2 production) (1, 17). In addition, the rise in DLCO with exercise [typically paralleling the rise in cardiac output (
) in health] may be reduced in these same patient groups, suggesting a limited expansion of the alveolar-capillary bed (15, 17, 27).
There are a number of techniques that have been used to assess DLCO. These include the single-breath (DLCO,SB), the rebreathe (DLCO,RB), and the steady-state methods (DLCO,SS) (16, 23, 25). The single-breath method is clinically the most widely used, largely because of standardization efforts from a number of organizations (2). In contrast, the other two techniques have been used primarily in research settings, with the rebreathe technique gaining the most prominence (15). Each method of assessing DLCO has its advantages and disadvantages, with the single-breath and rebreathe methods having particular shortcomings for use during heavy exercise. For example, the DLCO,SB method requires a breath hold at a full inflation volume, whereas the DLCO,RB technique is accompanied by a rise in carbon dioxide, leading to dyspnea and an unpredictable change in the rebreathe volume due to a changing CO2 production/O2 uptake relationship (5, 20). Thus, during heavy exercise or in patient populations in which dyspnea is a contributor to exercise intolerance, the DLCO,RB method may further contribute to exercise limitations. The DLCO,SS allows a natural breathing pattern on the part of the subject without a buildup of carbon dioxide. However, the DLCO,SS takes more time to perform, appears to be most affected by changes in alveolar dead space (compared with the other methods), and has classically required an arterial blood sample (4, 18, 26).
Given the usefulness of assessing DLCO with exercise, it would be optimal to have a technique that allowed a natural breathing pattern, did not cause a buildup of CO2, and did not require a lengthy measurement time or a blood draw. Our laboratory has recently developed an eight-breath, open-circuit gas washin technique to measure during exercise (19). This technique takes advantage of the solubility of acetylene (C2H2) in the blood as it passes through the pulmonary circulation. The maneuver and algorithms for assessing can be adapted to measure the disappearance of carbon monoxide (CO) in the lungs [open-circuit DLCO (DLCO,OC)]. Using this technique, the subject can breathe at a normal rate and tidal volume over 8–10 breaths. We have found that this method is much more easily tolerated for use during heavy exercise. We hypothesized that the DLCO,OC technique would yield similar values as the classic DLCO,RB technique at rest and throughout exercise.
|
METHODS
|
|---|
The protocol was reviewed and approved by the Mayo Clinic Institutional Review Board, and all participants signed informed consent before study. Eleven subjects were screened, agreed to participate in the study, and had no exclusion criteria (history of cardiac- or pulmonary-related abnormalities, use of prescription medications, pregnancy, or an inability to exercise).
Protocol.
Subjects performed an initial maximal exercise test on a stationary cycle ergometer to determine workloads for the test day. On a subsequent visit, subjects performed incremental cycle ergometry at work levels that approximated 25, 50, and 75% of their maximal work intensity for 10 min per exercise stage. At each work intensity, subjects performed DLCO,OC, DLCO,RB, and a repeat DLCO,OC, each separated by 1–2 min to allow test gas to be cleared from the lungs. The DLCO measures were started after 3 min following a change in exercise intensity to allow time for steady-state or near steady-state conditions. Following completion of each series of measurements at a given workload, the workload was increased, so that the testing was completed with no pause between stages.
DLCO,RB.
For the DLCO,RB, subjects breathed into a two-way switching valve (Hans Rudolph, Kansas City, MO), which was connected to a pneumotachometer (Hans Rudolph) and a mass spectrometer (Perkin-Elmer, 1100). Custom software was used to acquire data and to perform the analyses. The inspiratory port of the switching valve was set to either room air or a 5-liter anesthesia bag (Hans Rudolph), which was filled with 1.0–3.0 liters of test gas (35% O2, 0.6% C2H2, 0.3% C18O, 9% He, and balance N2), depending on the tidal volume of the subject and the exercise intensity. Consistent bag volumes were ensured by using one of the following methods: 1) the use of a timed switching circuit, which, given a consistent flow rate from the tank, resulted in the desired volume; or 2) filling of a 3-liter syringe with the test gas to the appropriate bag volume and subsequently injecting this into the evacuated rebreathe bag. Target volumes for the bag were estimated, starting with 1.2 liters at rest and increasing to 2.5–3.0 liters during exercise, depending on subject size and measurement of tidal volumes obtained from the initial incremental exercise test. C18O was used instead of the more common C16O as the test gas, because the mass spectrometer cannot distinguish C16O from N2. At the end of a normal expiration [end-expiratory lung volume (EELV)], the subjects were switched into the rebreathing bag. The subjects were then instructed to nearly empty the bag with each breath for 10 consecutive breaths. The respiratory rate was controlled at 20 breaths/min with a metronome at rest. During exercise, the subjects were allowed to breathe at a normal respiratory rate. Following each DLCO,RB maneuver, the anesthesia bag was emptied with a suction device and refilled for the next maneuver. Gas concentrations were continuously sampled with the mass spectrometer by using custom data-acquisition software sampling at 120 s–1. From these data streams, DLCO was obtained by the Roughton-Forster method (24). Briefly, end-inspiratory and end-tidal gas concentrations for each breath were identified from the volume tracing and after shifting the gas channels to account for the time delay between flow and gas signals (0.25–0.29 s). To determine DLCO,RB, the ratio of ln(C18O/He) was plotted against time for each breath, and the slope was determined by linear regression after excluding the first two breaths, any obvious outlier breaths, and all breaths after 
15 s of rebreathing (see Fig. 1 ). DLCO,RB is determined from the slope of this regression (24). Time 0 was determined from the point at which this regression line reached zero (25). In our laboratory, the coefficient of variation on repeat testing within a subject averages 4.2%.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1. Analysis of diffusing capacity of the lungs for carbon monoxide (DLCO) via the rebreathe (DLCO,RB) technique. Given are graphical descriptions of tidal volume (top), comparison of acetylene vs. helium washin (middle; thick line is acetylene), and the disappearance of carbon monoxide (C18O) (bottom).
|
|
DLCO,OC.
The open-circuit method has been described previously for the assessment of (19). In brief, during the DLCO,OC test, subjects breathed into a non-rebreathing valve (Hans Rudolph) with the inspiratory port connected to a pneumatic switching valve with low dead space and low resistance (Hans Rudolph). A pneumotachometer (Hans Rudolph) and mass spectrometer sampling line were connected to the common port of the non-rebreathing valve. Custom data acquisition and analysis software were used to acquire data at 120 Hz and perform the analysis to determine DLCO,OC (see Fig. 2 and APPENDIX). The inspiratory ports of the switching valve were connected to room air and a large reservoir that was filled with 10–30 liters of a test gas mixture (35% O2, 0.6% C2H2, 0.3% C18O, 9% He, and balance N2). To make a measurement, an operator switched the two-way valve during a normal expiration, allowing the next inspiration to be from the gas reservoir. Subjects breathed the gas mixture for up to 10 breaths and were subsequently switched back to room air. During the maneuver, the subject's breathing pattern could be visualized by the operator on the computer screen. At rest, the respiratory rate was controlled at 20 breaths/min, while, during exercise, subjects were allowed to breathe at their usual respiratory rate and tidal volume. The data were analyzed after shifting the gas concentration channels by 0.25–0.29 s to align them with the flow signal. DLCO,OC was calculated by an iterative solution of alveolar gas exchange equations (see Fig. 3, APPENDIX, and Ref. 19).
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 2. Analysis of DLCO via the open-circuit (DLCO,OC) technique. Shown are graphical descriptions of volume (top), comparison of acetylene vs. helium washin (middle; thick line is acetylene), and washin of C18O (bottom).
|
|
The final DLCO,OC value was then corrected to account for the difference in alveolar PO2 (PAO2), as measured by average end-tidal PO2 during the maneuvers, using the correction recommended by the American Thoracic Society (11). This correction is necessary because the PAO2 decreases during the DLCO,RB maneuver but stays constant during the DLCO,OC maneuver. The end-tidal fractional O2 concentration ([O2]) fell to 16% during the DLCO,RB tests but remained at nearly 21% during the DLCO,OC maneuvers, despite both techniques using gas with 35% O2 (see Table 3). Because of these O2 differences, there is likely less competition between O2 and CO during the DLCO,RB technique than during the DLCO,OC technique, which would artificially inflate the DLCO,RB values compared with the DLCO,OC values.
.
was obtained from the same maneuvers as the DLCO,OC by using the iterative calculation technique previously described (OpCirc2) (19).
Assessment of dyspnea.
Symptoms of shortness of breath were determined in a subset of subjects (n = 5) with the use of a 0–4 scale, where 0 = no shortness of breath, 1 = very mild, 2 = mild, 3 = moderate, and 4 = severe. Symptoms were assessed immediately after each DLCO measurement, and subjects were asked how their breathing was during the maneuver.
Data analysis.
Data are expressed as means (SD). The DLCO,OC and DLCO,RB were compared by using regression analysis and paired t-tests. The replicate DLCO,OC values were averaged before comparison to DLCO,RB. Differences between replicate DLCO,OC values were tested by using paired t-tests. Changes in DLCO with exercise were tested by using ANOVA followed by paired t-tests. Significance was accepted at P < 0.05.
|
RESULTS
|
|---|
Subject characteristics.
See Table 1. All subjects were mild-to-moderately active at the time of the study. Two of the 11 subjects tested were women. Average exercise intensities are given in Table 2.
Changes in DLCO,OC and DLCO,RB with exercise.
The DLCO increased linearly with for both DLCO methods (Table 2, Fig. 3). The slopes of the increase with were similar; DLCO,OC was slightly lower than DLCO,RB at rest and slightly higher during exercise, although these differences reached statistical significance only at level 2 of exercise, where DLCO,OC was 40.6 and DLCO,RB was 37.6 ml·min–1·mmHg–1.
Comparison of DLCO,OC to DLCO,RB.
The end-tidal PO2 was lower during DLCO,RB maneuver compared with DLCO,OC, with the difference becoming greater with exercise (Table 3). The mean correction for this difference increased the DLCO,OC by 8.8 (SD 2.9), 17 (SD 3.9), 18.7 (SD 4.4), and 19% (SD 4.2) at rest and the three levels of exercise, respectively. There was a good correlation between the corrected DLCO,OC method and the DLCO,RB method [mean of individual r2 = 0.86 and slopes = 1.28 (SD 0.29), overall r2 = 0.69, overall slope = 0.97, N = 42] (Fig. 4). The test-retest variability of the corrected DLCO,OC method was <5% at all workloads (Table 2). The Bland-Altman plot showed that the difference between the methods compared with the average of the two methods was slightly positively skewed (slope = 0.17, r2 = 0.07, P = 0.11) (Fig. 5).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4. Comparison of DLCO,RB to DLCO,OC for all subjects at rest and throughout exercise. The x-axis contains DLCO,RB values, and the y-axis contains the DLCO,OC values. Solid line is linear regression, with 95% confidence limits of the regression line shown as dashed lines. Units are in ml·min–1·mmHg–1.
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 5. Bland-Altman analysis of DLCO,RB and DLCO,OC. Solid line is linear regression, with 95% confidence limits of the regression line shown as dashed lines.
|
|
Symptoms of dyspnea obtained in a subset (N = 5) of the subjects during the maneuvers were higher with the DLCO,RB technique than the DLCO,OC method at rest [0.8 (SD 0.7) vs. 0.0 (SD 0.0), P < 0.05] and with exercise, with the differences being accentuated with the higher exercise intensities [exercise 1 = 1.7 (SD 0.5) vs. 0.2 (SD 0.4), exercise 2 = 3.0 (SD 0.6) vs. 1.0 (SD 0.6), and exercise 3 = 3.7 (SD 0.5) vs. 2.3 (SD 0.6), P < 0.001].
|
DISCUSSION
|
|---|
We have shown that DLCO obtained using an open-circuit method 1) increases linearly with , 2) is reproducible at rest and during exercise, 3) is similar to established rebreathe method, and 4) can be performed during exercise with little or no alteration in the subject's breathing pattern.
Measurement of the DLCO for assessing physiological parameters of the lung has been used since the early 20th century (7). There are several techniques that allow for assessment of DLCO at rest and during exercise, including the DLCO,RB, DLCO,SB, and DLCO,SS techniques (3, 21). The major limitation of the DLCO,SB method is the requirement of a full inspiration and a breath hold, which are difficult for most subjects to perform during exercise. Limitations of the DLCO,SS method include the requirement of an arterial blood sample to estimate alveolar CO2, the nearly 2 min required to perform the maneuver (for allowance of steady-state breathing by the subject), and the large amount of CO that the subject absorbs, limiting the number of trials per session.
All methods for measuring DLCO can be affected by inhomogeneities of alveolar ventilation (
A), , and A/. The issue of how closely any laboratory estimate of DLCO matches "true" DLCO is a matter of considerable debate. For instance, it has been shown that DLCO,SB is affected by subtle differences in calculation techniques that can lead to overestimation or underestimation of DLCO in the presence of lung disease (3a, 9). Modeling analysis suggests any measurement of DLCO will be affected by lung inhomogeneities (13), although a detailed modeling comparison of DLCO,OC to DLCO,RB has not been performed. Because both DLCO,OC and DLCO,RB involve tidal breathing (as opposed to breath hold), gas exchange in each breath will be likely equally inefficient for both techniques, so it is our expectation that DLCO,OC and DLCO,RB will be nearly equally affected by A/ mismatch.
Advantages of DLCO,OC over traditional methods.
The advantages of the DLCO,OC technique include 1) minimal coaching of the subjects in breathing technique (no need to match bag volume to tidal volume or to match the timing of the switch to the rebreathe bag exactly at EELV), 2) subjects do not experience increased shortness of breath due to CO2 buildup, 3) O2 uptake remains normal and PAO2 is more stable during the maneuver compared with single-breath and rebreathing methods, and 4) subjects can breathe normally (no breath hold, which makes the maneuver easy during exercise). The maneuver is also brief, only requiring the gas washin for 8–10 breaths, allowing for multiple runs within an exercise test.
Limitations to the DLCO,OC method.
The DLCO,OC method requires a higher volume of test gas compared with either the single-breath methods or rebreathing methods, potentially adding to the cost of measurements and exposure of subjects to the noxious gas CO. The costs can be greatly reduced by using a fast-response CO analyzer rather than using the mass spectrometer that requires the more expensive isotope of CO. The increased exposure of the subjects to CO could cause an increase in carboxyhemoglobin with repeat maneuvers. However, we did not see substantial increases in end-tidal C18O during the course of this study, in which we performed more maneuvers than a typical study would require. However, we did not measure blood carboxyhemoglobin levels.
The DLCO,OC method analyzes the kinetics of gas washin and thus is computationally more complex than either the single-breath or rebreathing methods, although the computational method is very similar to that used by the "three-equation" solution of the DLCO,SB (8, 9). The technique and algorithms applied in this study result in DLCO values that are similar to those obtained by the rebreathe method.
Differences between DLCO,OC and DLCO,RB.
The DLCO,RB method is a well-accepted method that uses a 5- to 7-liter bag that is typically filled with 1–3 liters of test gas mixture. The subject is asked to breathe in and out of the bag for 10–12 breaths, and calculations usually involve the end-expiratory points fitted to a logarithmic decay. This technique requires the subject to nearly empty the rebreathe bag without a complete inspiratory collapse of the bag. To prevent forceful collapse of the bag (which could alter breathing pattern or pressurize the gas sample line, influencing the measured gas concentration values), the rebreathe bag is filled to a volume that is at least as big as the tidal breath, although some laboratories use volumes as large as the subject's inspiratory capacity (31). A recent innovation for precisely matching bag volume to the subject is to use a double-switching valve, allowing the subject to freely draw from an inspired bag on the first breath and subsequently turning into the rebreathe bag (31). Although, in theory, this method should allow more appropriate matching of bag volume, even this method could result in an uncomfortable mismatch between bag and tidal volume, if the first breath were not representative of the subject's average breathing pattern. In addition, the rise in end-tidal and alveolar PCO2 (PACO2) during rebreathing stimulates subjects to increase tidal volume toward the end of the maneuver, often causing them to reach the bag volume limits. The DLCO,OC does not require precise matching of the subject's lung volume and does not substantially alter PAO2 and PACO2, and breathing pattern does not need to change during the maneuver. In our experience, a regular breathing pattern during the maneuver produces more reliable data, however.
We anticipated that the DLCO,OC and DLCO,RB methods would yield generally similar results, particularly if breathing patterns were similar between the techniques and they were performed at similar EELVs and similar PAO2. As shown in Table 3, our subjects tended to breathe with larger tidal volumes and at higher EELVs during the DLCO,RB method, which would expose a slightly larger alveolar surface area with the DLCO,RB (12, 29, 30). The larger tidal volume was largely due to coaching of the subjects to collapse the rebreathe bag with each breath, an increasing tidal volume during the maneuver secondary to increasing PACO2, and due to our attempts to make the initial rebreathe bag volume slightly higher than the subjects' actual tidal volume. The apparently lower EELV found using DLCO,OC may, in part, be an underestimate due to the effects of inhomogeneities on the calculation of lung volumes in early breaths of the open-circuit method (which uses simple single-compartment model gas dilution equations), similar to what was described in the studies of Horsfield and Cumming (6, 14). This difference in EELV had little impact on the final DLCO,OC; substituting higher EELV values into the analysis program during calculations yielded lower DLCO,OC by <0.2 ml·min–1·mmHg–1·l–1 change in EELV.
When using the same test gas for both maneuvers, the average PAO2 is lower during the DLCO,RB maneuver due to continual oxygen uptake and depletion of O2 in the closed lung-bag system (Table 3). For this study, we corrected for this difference using standard correction methods (11), although it would be appropriate to use lower [O2] in the test gas mixture for the DLCO,OC test. It should be noted that the PAO2 is more consistent during the DLCO,OC maneuver, since inspired [O2] remains constant, compared with either the DLCO,SB or DLCO,RB, where inspired [O2] is continually falling, potentially making the DLCO,OC more appropriate for studying the effects of altered PAO2 on DLCO in future studies.
In conclusion, the present study showed that the DLCO,OC compared favorably with the DLCO,RB at rest and during exercise. We also found the DLCO,OC to be reproducible, linearly associated with with a vs. DLCO,OC slope that was similar to the vs. DLCO,RB slope, and it was better tolerated compared with DLCO,RB. DLCO,OC is, therefore, a suitable method for measuring DLCO. The technique used for assessing DLCO in clinical or research studies should reflect a balance of subject comfort, equipment availability, implementation expertise, cost, and possibly other considerations related to specific protocols.
|
APPENDIX: CALCULATION METHOD FOR DETERMINING DLCO,OC
|
|---|
The DLCO,OC technique requires a subject to breath a mixture of gas containing trace amounts of C18O, acetylene (C2H2), 10% helium and balance O2 and N2 (Fig. 3). During the maneuver, concentrations of C18O, C2H2, helium, and gas flow, all measured at the mouth, are obtained at a 120-Hz sampling rate using custom data-acquisition software running on a personal computer with a 16-bit resolution analog-to-digital board. The DLCO value is obtained by iteratively adjusting the DLCO value to minimize the sum square error between measured end-tidal gas concentrations and end-tidal values obtained from a mathematical model of gas exchange for every breath recorded.
The lung is modeled as an alveolar compartment separated from the inspired gas by a non-gas-exchanging conductive dead space. Exchange of C18O occurs in the alveolar compartment following simple laws of diffusion (see below). The volume of the dead space (VD) for each run is determined from mass balance of helium in the first few breaths of washin, as detailed in our laboratory's previous study (19):
 | (1) |
where F
He, FEHe, and FIHe are mixed-expired, end-expired, and inspired helium concentrations, respectively; and VT is the tidal volume. FHe is obtained from the ratio of volume of helium expired (obtained by simple integration of flow and helium signals) to VT for each breath. Equation 1 is evaluated and averaged only over the first three breaths to avoid aberrant values when FEHe approaches FIHe near the end of washin. To simulate "plug" flow of gas through the dead space, the computer program divides the VD into 1-ml units. At the beginning of a calculation for a washin run, gas values in all of the VD units are set to end-expiratory values found for the breath immediately before the start of washin.
The solution for DLCO begins with setting DLCO to a nominal value, typically 20 ml·min–1·mmHg–1. Mixed-venous partial pressure of C18O (P
C18O) (see Eq. 2 below) is obtained from the end-expiratory value for C18O in the breath immediately preceding the start of washin. The PC18O values could be a few percentages of the tank value, if adequate time for purging of gas from the previous maneuver has not elapsed. Gas exchange in the lung is then modeled for every 8.33-ms time point in the raw data stream as follows (19), starting with inspiration. First, the flow, volume increment, and gas concentrations for the time point are read from the data stream. The dead space elements are then advanced by the number of milliliters in the volume increment, filling the mouth-end elements with gas concentration values equal to the values in the data stream. For instance, if the volume increment is 3 ml, the gas values in the first three elements of dead space are set to gas concentrations from the data stream. At the alveolar end of the dead space, the last three dead space elements are read, and volumes of helium and C2H2 are added to the alveolar compartment using simple gas dilution equations, and the total volume of the alveolar compartment is increased by the volume increment.
To model uptake of C18O, a simple gas diffusion process is assumed:
 | (2) |
where PAC18O is partial pressures of C18O in alveolar gas; DLCO is the rate constant for diffusion for C18O in units of ml·min–1·mmHg–1, and C18O is the transfer rate of C18O across the alveolar membrane in ml/min.
The following equation dictates gas uptake in the alveolar region:
 | (3) |
where VA is the alveolar volume, FAC18O is alveolar concentration of C18O, F
C18O is mixed-venous concentration of C18O, FDS'C18O is the concentration of C18O at the alveolar end of the dead space element during inspiration, and t is time. Because of the plug flow through the dead space element, FDS'C18O equals expiratory values from the end of the previous breath early in inspiration and becomes equal to concentration of C18O of the inspired bag for inspired volumes greater than VD. The term on the left is the change in volume of C18O in the alveolus per unit time, the first term on the right is the diffusion of gas into the alveolar blood, and the second term is the contribution of changing gas volume from the dead space, either during inspiration (top term) or during expiration (bottom). The change in VA per time increment, dVA/dt has a positive value for inspiration and negative value for expiration. This equation can be solved for dFAC18O(t):
 | (4) |
Using this equation, the FAC18O is updated for each time increment in the data stream. At end expiration, the value at the mouth end of the dead space is saved. At the end of each run through the complete data stream, the mean square error between model-estimated values and actual end-tidal values is calculated for C18O, excluding the values for the first two breaths. Because the first two breaths are likely affected by dead space more than subsequent breaths, we found the DLCO,OC value was more reasonable and reproducible, if we excluded these breaths from the mean square error (note the first two breaths were included in the gas uptake calculations, however). The Powell iterative search method is used to change the DLCO value to minimize the mean square error term (22). Briefly, this method uses results from pairs of iterations to find the rate of change in mean square error per unit change in DLCO and uses this slope to find the point at which mean square error is at a minimum.
|
GRANTS
|
|---|
This study was supported by American Heart Association Grant 56051Z, National Heart, Lung, and Blood Institute Grant HL-71478, and United States Public Health Service Grant M01-RR00585.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank Kathy O'Malley for technical assistance and Renee Blumers for assistance with manuscript preparation.
|
FOOTNOTES
|
|---|
Address for reprint requests and other correspondence: B. D. Johnson, Division of Cardiovascular Disease, Gonda 5-369, Mayo Clinic and Foundation, 200 1st St. SW, Rochester, MN 55905 (e-mail: Johnson.bruce{at}mayo.edu)
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES
|
|---|
- Agostoni PG, Busstti M, Palermo P, and Guazzi M. Does lung diffusion impairment affect exercise capacity in patients with heart failure. Heart 88: 453–459, 2002.
- American Thoracic Society. Single-breath carbon monoxide diffusing capacity (transfer factor). Recommendations for a standard technique—1995 update. Am J Respir Crit Care Med 154: 265–266, 1996.
- Anderson S and Godfrey S. Transfer factor for CO during exercise in children. Thorax 26: 51–54, 1971.
- Beck KC, Offord KP, and Scanlon PD. Comparison of four methods for calculating diffusing capacity by the single breath method. Chest 105: 594–600, 1994.
- Boro A, Sharma S, and Pande J. Comparison of single breath and steady state methods for the measurement of pulmonary diffusing capacity for carbon monoxide in normal subjects, patients with bronchial asthma and chronic obstructive airway disease. Indian J Chest Dis Allied Sci 34: 1–5, 1992.
- Boutellier U and Farhi E. A fundamental problem in determining functional residual capacity or residual volume. J Appl Physiol 60: 1810–1813, 1986.
- Cumming G. Gas mixing efficiency in the human lung. Respir Physiol 2: 213–224, 1967.
- Forster RE. Diffusion of gases across the alveolar membrane. In: Handbook of Physiology. The Respiratory System. Gas Exchange. Bethesda, MD: Am. Physiol. Soc., 1987, sect. 3, vol. IV, chapt. 5, p. 71–88.
- Graham BL, Dosman JA, and Cotton DJ. A theoretical analysis of the single breath diffusing capacity for carbon monoxide. IEEE Trans Biomed Eng 27: 221–227, 198O.
- Graham BL, Mink JT, and Cotton DJ. Effect of breath-hold time on DLCO(SB) in patients with airway obstruction. J Appl Physiol 58: 1319–1325, 1985.
- Gray G, Zamel N, and Crapo RO. Effect of a simulated 3,048 meter altitude on the single-breath transfer factor. Bull Eur Physiopath Respir 22: 429–431, 1986.
- Hamer NA. Variations in components of the diffusing capacity as the lung expands. Clin Sci 24: 275–285, 1963.
- Hlastala MP. Diffusing capacity inheterogeneity. In: Handbook of Physiology. The Respiratory System. Gas Exchange. Bethesda, MD: Am. Physiol. Soc., 1987, sect. 3, vol. IV, chapt 12, p. 217–232.
- Horsfield K and Cumming G. Functional consequences of airway morphology. J Appl Physiol 24: 384–390, 1968.
- Hsia CC. Recruitment of lung diffusing capacity: update of concept and application. Chest 122: 1774–1783, 2002.
- Huang YC, O'Brien SR, and MacIntyre NR. Intrabreath diffusing capacity of the lung in healthy individuals at rest and during exercise. Chest 122: 177–185, 2002.
- Hughes JM, Lockwood DN, Jones HA, and Clark RJ. DLCO/Q and diffusion limitation at rest and on exercise in patients with interstitial fibrosis. Respir Physiol 83: 155–166, 1991.
- Jebavy P, Hurych J, and Widimsky J. Influence of pulmonary hypertension on pulmonary diffusing capacity in patients with mitral stenosis. Respiration 35: 1–7, 1978.
- Johnson BD, Beck KC, Proctor DN, Miller J, Dietz NM, and Joyner MJ. Cardiac output during exercise by the open circuit acetylene washin method: comparison with the direct Fick. J Appl Physiol 88: 1650–1658, 2000.
- Johnson BD, Seow KC, Pegelow DF, and Dempsey JA. Adaptation of the inert gas FRC technique for use in heavy exercise. J Appl Physiol 68: 802–809, 1990.
- Mohsenifar Z, Lee S, Diaz P, Criner G, Sciurba F, Ginsburg M, and Wise R. Single-breath diffusing capacity of the lung for carbon monoxide: a predictor of PaO2, maximum work rate, and walking distance in patients with emphysema. Chest 123: 1394–1400, 2003.
- Press WH, Flannery BP, Taukolsky SA, and Vettering WT. Numerical Recipes: The Art of Scientific Computing. Cambridge, MA: Cambridge University Press, 1986.
- Rose GL, Cassidy SS, and Johnson RL Jr. Diffusing capacity at different lung volumes during breath holding and rebreathing. J Appl Physiol 47: 32–36, 1979.
- Roughton FJ and Forster RE. Relative importance of diffusion and chemical reaction rates in determining rate of exchange of gases in the human lung, with special reference to true diffusing capacity of pulmonary membrane and volume of blood in the lung capillaries. J Appl Physiol 11: 290–302, 1957.
- Sackner MA, Greeneltch D, Heiman MS, Epstein S, and Atkins N. Diffusing capacity, membrane diffusing capacity, capillary blood volume, pulmonary tissue volume, and cardiac output measured by a rebreathing technique. Am Rev Respir Dis 111: 157–165, 1975.
- Savoy J, Michoud M, Robert M, Geiser J, Haab P, and Piiper J. Comparison of steady state pulmonary diffusing capacity estimates for O2 and CO in dogs. Respir Physiol 42: 43–59, 1980.
- Smith AA, Cowburn PJ, Parker ME, Denvir M, Puri S, Patel KR, and Cleland JG. Impaired pulmonary diffusion during exercise in patients with chronic heart failure. Circulation 100: 1406–1410, 1999.
- Stam H, Kreuzer FJ, and Versprille A. Effect of lung volume and positional changes on pulmonary diffusing capacity and its components. J Appl Physiol 71: 1477–1488, 1991.
- Stam H, Versprille A, and Bogaard JM. The components of the carbon monoxide diffusing capacity in man dependent on alveolar volume. Bull Eur Physiopath Respir 19: 17–22, 1983.
- Tamhane RM, Johnson RL Jr, and Hsia CC. Pulmonary membrane diffusing capacity and capillary blood volume measured during exercise from nitric oxide uptake. Chest 120: 1850–1856, 2001.
Copyright © 2005 by the American Physiological Society.
TOP
ABSTRACT
METHODS
, DLCO,RB (RBDLCO); 
, DLCO,OC (OCDLCO); 
and dashed line, regression result from Ref.