| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
GSF-National Research Center for Environment and Health, Institute for Inhalation Biology, D-85758 Neuherberg/Munich, Germany
Submitted 18 December 2003 ; accepted in final form 8 June 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
oxygen isotope 18O; bronchoconstriction; intrapulmonary gas transport; stationary front; diffusion front theory; power moment analysis
|
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Lung volumes. Functional residual capacity (FRC) was measured by Ar rebreathing (tidal volume of 0.45 liter, respiratory rate of 40 breaths/min, 15 breaths). Inspiratory capacity was determined as the volume inhaled slowly (0.05 l/s) from FRC until an airway pressure of 25 cmH2O was reached. Total lung capacity (TLC) was calculated as the sum of inspiratory capacity and FRC.
Single-breath maneuvers. After equilibration of the lungs with test gas-free air, a standardized single-breath maneuver was performed with 0.9% Ar and 0.3% C18O2 in air. To account for interindividual differences in lung volumes, inspired and expired volumes were adjusted in a way that fixed the fractions of the individual's TLC reached at end inspiration and end expiration. For baseline measurements, the test gas mixture was inhaled at a flow rate of 0.3 l/s from FRC to an end-inspiratory lung inflation of 75% TLC. Exhalation followed without any postinspiratory pause at a flow rate of 0.1 l/s and was stopped when a lung inflation of 30% TLC was reached. This pattern was defined as a "control" maneuver. The effects of changing ventilatory patterns on phase II were studied by varying one of the following parameters: inspiratory flow rate, expiratory flow rate, tidal volume, or postinspiratory pause (see Experimental protocol below).
From the Ar expirograms, the slope of phase III was determined by least-squares regression analysis of the segment between 40 and 80% of the expired volume. This slope was used to determine series dead space volume according to the equal-area method of Fowler (4). The respective "slope" for the Fowler analysis of the C18O2 expirograms was the zero baseline.
The expirograms from both test gases were normalized to the respective inspiratory gas concentrations (Fig. 2, left), and the first derivative with respect to the expired volume (
FC18O2/V) was calculated point by point over phase II (line segment length = 0.5 ml). This derivative reflects the volumetric distribution of airway units that are completely emptied during exhalation (see DISCUSSION) and was mathematically evaluated by a power moment analysis (Fig. 2, right) (20). The analysis of phase II of C18O2 was restricted to the 98.5–1.5% interval of normalized gas concentrations, since standard deviation and particularly skewness are strongly influenced by a noisy signal at the "tails" of the derivative curve, where C18O2 concentrations are slowly changing. Similarly, for the power moment analysis of Ar expirograms, the cut-off point between phase I and phase II was 98.5% of inspired gas concentration, whereas the cut-off point between phase II and phase III was taken from the respective C18O2 expirogram recorded in parallel.
|
 |
 |
 |
 |
Respiratory resistance. Respiratory resistance was measured by airway occlusion (7), using a custom-made shutter valve that repeatedly interrupted the expiratory flow for 300 ms each during a standardized breathing maneuver. The plateau pressure during the occlusion and the expiratory flow immediately before the occlusion were determined for each of the occlusions, and the pressure-flow relationship was fitted according to Rohrer's equation: P = a x Vex + b x V
(where P is pressure and Vex is expiratory flow). Resistance was then calculated for an expiratory flow of 1 l/s from this relationship.
Induction of bronchoconstriction. Bronchoconstriction was induced by having the dogs inhale nebulized carbachol solutions during 10 breaths, using the above-mentioned standardized breathing maneuver (with expiration performed only to FRC). To target primarily the conducting airways, nebulized carbachol was inhaled as a 35 ml-bolus when 70% of the inspiration had been completed. A postinspiratory pause of 5 s was introduced to allow carbachol particles to settle and deposit on airway walls. Initially, 10 boluses of a negative control (0.9% NaCl adjusted to pH 7 with NaOH) and then 10 boluses each of subsequently increasing carbachol concentrations (0.25, 1.25, 6.25, 12.5, and 25%) were inhaled at 8-min intervals.
Carbachol inhalations were terminated prematurely in one dog who unexpectedly showed signs of bronchial hypersensitivity and impending cardiocirculatory problems. Data from this dog were excluded altogether from the analysis of bronchoconstrictory effects on the single-breath expirogram, which is therefore based on n = 5 dogs.
Experimental protocol. After the determination of lung volumes, single-breath measurements were performed in the following sequence: 1) control measurement, 2) variation of end-expiratory breath hold (2, 5, 9, and 16 s, sequentially) with "control" values for flow rates and tidal volume, 3) variation of expiratory flow rate (0.2 and 0.4 l/s) with no breath hold and inspiratory flow rate and tidal volume at control values, 4) variation of inspiratory flow rate (0.1, 0.6, and 0.9 l/s) with no breath hold and expiratory flow rate and tidal volume at control values, 5) variation of end-inspiratory lung inflation (60 and 90% of TLC) with no breath hold and flow rates at control values, and 6) control measurement.
In an effort to minimize time-related effects, all of these single-breath measurements were completed within 1 h, and the influence of variations in breathing pattern was evaluated by relating parameters to the averages from control measurements at the beginning and at the end of the single-breath series. It was found that measurements from these two control series differed by no more than 0.1% (SD 4.0%).
The latter series of control measurements was complemented by a determination of respiratory resistance and yielded baseline values for the subsequent bronchial challenge series. Isotonic saline and carbachol boluses were inhaled in the sequence described above. Three minutes after the end of each inhalation, respiratory resistance was assessed; thereafter, a standard single-breath maneuver with Ar and C18O2 was performed.
Data analysis. Signals were analog-to-digital converted and stored in a personal computer at a frequency of 500 Hz for measurement of lag and response times of C18O2 and Ar. For single-breath maneuvers, sampling frequencies ranged between 200 and 800 Hz and were adapted to a volume resolution of 0.5 ml during expiration.
Each measurement was performed in duplicate, and the means of both measurements were used for further data analyses. Statistical analysis was based on paired t-test performed by means of a commercially available statistics package (Statgraphics, Statistical Graphics, Rockville, MD). This package was also used for correlation analysis. Differences were considered statistically significant at an error probability level of 5%.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
As to be expected from the pharmacokinetics of this gas, C18O2 concentration during phase III of the expirograms was always constantly zero, and there was a clearly definable separation between phase II and phase III (Fig. 1).
Baseline values from the standard single breath maneuver were 161 ± 5.5 ml (Fowler dead space), 160 ± 5.5 ml (position), 39 ± 1.1 ml (broadness), and 0.72 ± 0.02 (skewness) for the C18O2 expirograms and 165 ± 5.8 ml (Fowler dead space), 166 ± 5.6 ml (position), 41 ± 1.1 ml (broadness), and 0.80 ± 0.03 (skewness) for the Ar expirograms. Values obtained for the Ar expirograms marginally, albeit statistically significantly (P < 0.05), exceeded those from the C18O2 expirograms.
A tight correlation was observed between dead space estimates by the Fowler technique and by power moment analysis throughout all measurement conditions, as shown in Fig. 3, left, for C18O2. Generally, close to very close correlations existed between parameters from C18O2 expirograms and the corresponding values from Ar expirograms. This is shown for Fowler dead space in Fig. 3, right. A similar correlation existed for position (position-Ar = 2.29 ml + 1.017 x position-C18O2, r2 = 0.989). Measurements of broadness and skewness generally showed a larger scatter than dead space estimates; therefore, correlations between Ar and C18O2 expirograms were less tight but still highly significant (broadness-Ar = –0.872 ml + 1.082 x broadness-C18O2, r2 = 0.782, skewness-Ar = 0.08 + 1.027 x skewness-C18O2, r2 = 0.864). None of the slopes was significantly different from 1.
|
|
|
The dynamics of changes in airway resistance and in parameters from phase II of C18O2 expirograms during bronchoconstriction are shown in Fig. 4. Baseline resistance was 2.65 ± 0.63 cmH2O·l–1·s and increased two to threefold during bronchoconstriction. Concurrently, Fowler dead space, volumetric mean, and broadness of phase II decreased, whereas skewness increased. Changes observed in Ar expirograms were generally similar to that. Occasional differences, albeit statistically significant, on average did not exceeded 10% (Fig. 4, right), with the only exception of broadness at 6.25% carbachol.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Inhaled C18O2 is transported and metabolized like ordinary C16O2. On reaching the gas-exchanging regions of the lung, it diffuses into alveolar tissues and capillary blood according to the prevailing C18O2 pressure gradients. Because the diffusing capacity of C18O2 is 
20 times higher than that of O2, it rapidly vanishes from the alveolar space (8). It is then degraded in a hydration-dehydration reaction catalyzed by carbonic anhydrase, which is abundantly present in erythrocytes and, to a lesser degree, in the alveolar lung tissues (9, 16). During the hydration and dehydration processes, continuous exchange takes place between 18O and 16O from lung tissue water, such that the restitution probability of any inhaled C18O2 is extremely small.
On the other hand, only a very small amount of inhaled C18O2 is absorbed by the epithelium of the conducting airways (25). On the basis of measurements by George and coworkers (5, 6) of gas uptake under steady-state conditions in the canine trachea, the tracheal diffusing capacity for a moderately soluble gas like carbon dioxide can be approximated to 0.04 ml C18O2·s–1·atm–1, allowing for a tracheal C18O2 uptake of 2.4 10–4 ml under the experimental conditions of our standard single-breath maneuver (partial pressure of C18O2, 2.1 Torr; time for filling and emptying of the conducting airways, 2.14 s). With the canine trachea accounting for 2% of the total surface area of conducting airways (11), C18O2 airway uptake would be 1.2 10–2 ml or 2.5% of the total amount of C18O2 actually inhaled into the conducting airways. This rough approximation will underestimate gas uptake in the more distal airways, owing to their smaller radial diffusion distances and thinner epithelia, and overestimate uptake, on the other hand, on the basis of the residence time used for the calculation; yet it actually fits very well with the experimentally observed differences between C18O2 and Ar Fowler dead space and position of phase II of 2.4 and 3.8%, respectively. Similarly, small differences in broadness and skewness of phase II between C18O2 and Ar standard expirograms (by 4.4 and 11.1%, respectively) were possibly due to the impact of gas exchange in the conducting airways. As to the differences in C18O2 uptake that might occur between the various maneuvers studied, the parameter most likely to influence C18O2 uptake velocity is bronchial blood flow, and the intervention most likely to exert a significant influence on bronchial blood flow is the induction of bronchoconstriction. According to the finding of Charan et al. (1) that bronchial blood flow increased by no more than 15% with even highest doses (2 mg/kg) of inhaled acetylcholine in anesthetized sheep, it may be estimated that the maximal expected effect of C18O2 uptake can only be in the same order of magnitude. Generally, the similarity between values obtained from Ar and from C18O2 expirograms (Fig. 3) also argues against a significant C18O2 uptake in the conducting airways under the present experimental conditions, although marginal but consistent and statistically significant differences existed: Ar dead space and position exceeded C18O2 dead space and position by 1.8 ml (P < 0.01) and 2.3 ml (P < 0.01), respectively, throughout the various breathing maneuvers. Even smaller but still significant differences were observed for broadness and skewness, as reflected in the regression intercepts. Similarly, some of the changes induced by varying the respiratory maneuver were marginally but significantly different between Ar and C18O2 (see Tables 2 and 3). Beside the above-mentioned effect of C18O2 uptake in the conducting airways, the influence of inhomogeneous alveolar ventilation on phase II of the Ar expirogram (see below) may play a role. Both effects can neither be excluded nor separated and may have contributed in particular to the differences observed during bronchoconstriction, where broadness at higher levels of carbachol inhalation tended to be smaller in C18O2 than in Ar expirograms.
The special features of C18O2 contrast with those of inert foreign test gases, which are distributed more or less inhomogeneously in the residual alveolar air. Their local concentrations vary depending on the ventilatory efficiencies of the respective compartments and influence the alveolar slope and, to a lesser degree, phase II of the expirogram. The rapid and complete absorption of C18O2 from the alveolar space extinguishes the influence of heterogeneous alveolar concentrations. Consequently, a C18O2 single-breath expirogram exhibits a phase I of constant gas concentration and a phase II with continually decreasing C18O2 concentrations, whereas the concentration during phase III is constantly zero (Fig. 1). Phase II then specifically describes the transition zone between the conducting airways and the alveolar space and reflects the differences in (functional) bronchial pathway lengths or transit times for the passage from the alveolar space to the airway opening. These particular characteristics of the C18O2 expirogram allow an exact mathematical analysis of phase II. A normalized C18O2 expirogram (i.e., inspiratory gas concentration defined as 1.0) can be viewed as the functional cumulative distribution of airway pathways to be emptied in the course of the exhalation. In other words, the decreasing concentration of C18O2 over phase II reflects the increasing number of alveolar units contributing to the exhalate. In contrast, if different alveolar compartments contributed with different gas concentrations, as is the case with conventional test gases, a particular step in concentration could not be assigned to a defined corresponding volume. Furthermore, it became evident during the analysis of Ar expirograms that, even in the healthy lung, the gradual transition between phase II and phase III introduced a fair amount of scatter and subjectivity, such that a valid analysis of phase II was largely hindered. Consequently, we chose to take the cut-off point between phase II and phase III from the corresponding C18O2 expirograms to yield equivalent conditions for the comparison of Ar and C18O2 expirograms.
The first derivative of the normalized C18O2-concentration curve with respect to the expired volume can be described as the volumetric distribution of airway pathways to be emptied. This distribution was evaluated by power moment analysis, a technique utilized for the analysis of gas and aerosol boluses (24) but so far not for single-breath expirograms. The mean of this distribution (volumetric position of phase II) is a measure of respiratory dead space or airway volume. Accordingly, a very tight correlation was observed between position and dead space volume by the Fowler technique (Fig. 3). Standard deviation characterizes the broadness of the distribution relative to its mean. It is a measure of the range of (functional) airway pathway lengths between alveolar space and airway opening. Skewness characterizes the symmetry of the distribution. A skewness of zero denotes a perfectly symmetrical shape of the distribution curve relative to its mean. In our experiments, skewness was always found to be greater than zero (with the steeper upswing at the beginning and the "tail" of the distribution following at the end of the exhalation). This indicates that there existed a number of long airway pathways or slowly or late emptying airway compartments.
Despite their dissimilar appearance, Ar and C18O2 tracings yielded very similar estimates of respiratory dead space volume (Fig. 3) and of position, broadness, and skewness of phase II (Tables 2 and 3) under the conditions and breathing maneuvers applied. Our measurements were performed in dogs with healthy lungs in which phase II was always clearly visible; therefore, the advantage of applying C18O2 as a tracer gas is less obvious than in the diseased lung. As observed by Meyer et al. (15), phase II and phase III of the C18O2 expirogram can also be separated unambiguously in emphysematous patients where the inert gas expirogram recorded in parallel does not show a clearly delimited alveolar slope (Fig. 5). Accordingly, these authors found that the diagnostic power of single-breath expirograms in emphysematous patients could be enhanced by the addition of C18O2 to the test gas mixture.
|
400 to 3,600 in the trachea, caused position to shift by no more than 7% or 11 ml and broadness to increase by 5% or 2 ml. According to the "stationary front" theory (3, 17, 21), an increased inspiratory flow primarily enhances convective forces and acts to move the front into the lung periphery. Morphological data from a symmetrical airway model for dogs of roughly similar size (28) indicate that the peripheral shift of 11 ml observed in our study corresponds to about 1.5 mm or two generations of respiratory bronchioli (20th to 22nd airway generation). Slightly larger effects were observed with variations in expiratory flow rate. Position was shifted into the lungs by 16%, with a fourfold increase in expiratory flow rate. At the same time, the symmetry of lung emptying appeared somewhat disturbed, with a rise in broadness and skewness of phase II, indicating an increase particularly of slowly emptying compartments. Comparable changes in dead space volumes have been reported from helium and sulfur hexafluoride expirograms by Meyer et al. (14). These authors interpreted the observed changes in dead space volumes as likely artificial and secondary to the changes of the slope of phase III of the expirograms, i.e., on the effects that the breathing maneuver had on alveolar inhomogeneity. Yet, their results are remarkably similar to those of the present study in which the influence of the alveolar slope was completely abolished by use of the test gas C18O2.
Increasing end-inspiratory lung volume resulted in a small increase in position and dead space volume. In contrast, dead space volume as a fraction of the actual lung volume decreased slightly but continuously with increasing lung inflation. A similar pattern was found in a previous study from our laboratory (22) and can be explained by a tendency of the stationary front to reside in similarly sized airways despite progressive lung inflation.
More pronounced effects occurred with the introduction of a postinspiratory breath hold. This maneuver does not influence airway geometry or convective forces but promotes gas transport toward the lung periphery by diffusion and cardiogenic mixing (2, 18). A continuous shift of position toward the airway opening was observed. As predicted (10) and previously found in controlled experiments (14), the decrease in dead space volume and position with time was roughly exponential. The associated decrease in volumetric broadness of phase II was considerably less than proportional to the change in dead space, indicating that the overall simultaneity of lung emptying was distinctly affected. The changes in skewness occurred in a biphasic manner. The "tail" of the distribution was reduced with a short breath hold of 2–4 s, suggesting a relative advantage of slowly emptying compartments in that situation. Further increases in breath-holding time increased the skewness and likewise the imbalance between slow and fast compartments.
Bronchoconstriction by carbachol inhalation decreased C18O2 dead space and position by almost 50% (Fig. 4). A proportional decrease in broadness indicated that the bulk of airway pathways was passed more or less as simultaneously as under baseline conditions. The increased skewness during bronchoconstriction suggested that, despite the generally preserved regularity of lung emptying, bronchoconstriction caused some compartments to empty very slowly.
In conclusion, our data confirm the notion that, despite obvious asymmetries in lung structure, filling and emptying of the lungs occur in a remarkably symmetrical fashion and are fairly robust to changes in ventilatory patterns. A comprehensive analysis of phase II of single-breath expirograms by power moment analysis yields as yet unavailable information on the dynamics of intrapulmonary gas transport. The practical benefit of this additional information must be investigated in further clinical studies.
|
FOOTNOTES |
|---|
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 |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. Moller, K. Felten, K. Sommerer, G. Scheuch, G. Meyer, P. Meyer, K. Haussinger, and W. G. Kreyling Deposition, Retention, and Translocation of Ultrafine Particles from the Central Airways and Lung Periphery Am. J. Respir. Crit. Care Med., February 15, 2008; 177(4): 426 - 432. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Tang, M. J. Turner, and A. B. Baker Systematic errors and susceptibility to noise of four methods for calculating anatomical dead space from the CO2 expirogram Br. J. Anaesth., June 1, 2007; 98(6): 828 - 834. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Tusman, M. Areta, C. Climente, R. Plit, F. Suarez-Sipmann, M. J. Rodriguez-Nieto, G. Peces-Barba, E. Turchetto, and S. H. Bohm Effect of pulmonary perfusion on the slopes of single-breath test of CO2 J Appl Physiol, August 1, 2005; 99(2): 650 - 655. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
