|Table of Contents|
Table of Contents
List of references
AN ANALYSIS OF THE CONTRIBUTION OF SMALL AIRWAYS TO LOWER AIRWAY RESISTANCE
JOHN NEWTON EVANS
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
I wish to acknowledge and thank Dr. Marc J. Jaeger, chairman of the Supervisory Committee, for his guidance and inspiration throughout my graduate study.
I would also like to express my thanks to the
members of the Supervisory Committee, Dr. A. B. Otis, Dr. D. M. Travis, Dr. W. N. Stainsby and Dr. A. H. Nevis, for their suggestions and critical evaluation during my graduate training.
I would also like to recognize my fellow graduate students who provided constructive criticism and friendship through the years. I particularly wish to acknowledge the technical assistance and friendship of Ms. Linda Nichols and Ms. Patty Potter both of whom played a major role in the experiments reported herein.
I would also like to thank Dr. L. Wood for his guidance concerning techniques used in this study to measure resistance and Dr. S. Battista for his advice concerning placement of retrograde catheters in closed chest dogs. I also acknowledge Mr. J. Bolognese and
Dr. R. Marks for their statistical analysis of the data.
I would also wish to thank my parents who have provided encouragement and support throughout my education.
Most of all I wish to thank my wife, Ellen, and daughter, Melissa, for their love and understanding during my graduate training.
TABLE OF CONTENTS
Predictions of Peripheral Resistance
Based on Anatomic Data 13
Previous Measurements of Peripheral Airway
Resistance by Retrograde Catheters 16
Justification of the Retrograde
Catheter Technique 28
Oscillation Technique 34
Measurement of Flow, Volume
and Pressure 36
Placement of the Retrograde Catheter 44
Checks on Patency of the Retrograde Catheter 46
Series I 47
Series 11 48
Series 111 48
Series I 50
Series 11 57
Series 111 64
Pressure Volume Curves 71
Location of the Catheter 74
Blood Gases 74
Experimental Design 79
comparison of RP measured in vivo with
Predicted Values Based on Anatomic Data 86
The Dependence of R P and R. on Lung Volume 93
The Changes in Resistance Induced by
Opening the Chest 96
LIST OF REFERENCES 105
BIOGRAPHICAL SKETCH 109
Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
AN ANALYSIS OF THE CONTRIBUTION OF SMALL AIRWAYS
TO LOWER AIRWAY RESISTANCE
John Newton Evans
Chairman: Dr. Marc J. Jaeger Major Department: Physiology
Total airway resistance is the frictional component of the pressure drop from the alveolus to the mouth related to the flow of gas at the airway opening. This resistance can be partitioned into an upper airway resistance, which includes the nose, larynx, pharynx and trachea and a lower airway resistance which includes the bronchi and small airways.
The lower airway resistance can be partitioned into
contributions made by bronchi of various sizes. Peripheral or small airways can be defined as those which receive their blood supply from the bronchial circulation, as those upstream of the equal pressure point during forced expiration or as those peripheral to the size airway in which a pressure measuring device is located.
In this study pressure was measured with a retrograde vi
catheter wedged in bronchi which had a mean diameter of 3.0 mm. Therefore, peripheral resistance (Re) refers to the resistance of airways less than 3.0 mm. in diameter7 central resistance (R c) refers to airways peripheral to the trachea and larger than 3.0 mm.; and lower airway resistance is the sum of R and RC.
RL and R were measured in the anesthetized and paralyzed dog at transpulmonary pressures (P tp) of 15, 10 and
5 cm. H 0o, while the lungs were oscillated at 4 cycles per second. These measurements were made with the chest closed and following a sternal splitting procedure with wide retraction of the ribs. At a P of 5 cm. HZ 0 in the closed chest dog the mean RL and mean R in 13 dogs were
3.02 and 1.5 cm. HZO/L/S, respectively. Following the opening of the chest R was unchanged at 2.99 cm. H2O/L/S while was significantly (p < .02) reduced at .73 cm. H2O/L/S. A similar effect of opening the chest was seen at Ptp's of 15 and 10 cm. H O.
RL and Rp were curvilinearly related to P in both the closed and opened chest conditions. RL was maximum at a Ptp of 5 cm. H20 and was reduced at the higher pressures. was lowest at a p of 10 cm. H20.
Beta adrenergic receptor blockade with propranolol (1 mg./Kg. i.v.) did not change RL or R in the dog when vii
the chest was intact. However, it did prevent the reduction of RP upon opening the chest which had been observed in non-beta adrenergic receptor blocked animals.
Theoretical predictions of the contribution of airways less than 3.0 mm. in diameter are presented for both the human and dog lung using morphometric data and assuming laminar flow. The predictions agree with the in vivo measurements in the closed chest dog that up to 50% of the lowr pulmonary resistance is contributed by airways less than 3.0 mm. in diameter.
It was shown that the contribution of peripheral airways to lower pulmonary resistance in closed chest dog is significantly greater than that measured following the opening of the chest. This reduction of peripheral resistance is mainly mediated through an increased beta adrenergic receptor stimulation in the peripheral airway. The influence of sympathetic input on the tonus of peripheral airway smooth muscle is minimal in the closed chest dog and only becomes significant when the chest is opened. In spite of the reductions in R P due to opening of the chest there were no significant changes in RL, which means that central resistance increased in a magnitude equal to the reduction of RP. This suggests the possibility of the maintenance of a constant R L at a given lung volume by coordinated
to offset changes in R
changes in R C P*
This is a study of pulmonary resistance and in particular the contribution of airways less than 3 mm. in diameter to lower pulmonary resistance. For the first time the peripheral resistance was measured by means of a retrograde catheter with the chest closed and following a sternal splitting procedure in the same dog. In addition, measurements were made following the creation of a posumothorax and after beta adrenergic receptor blockade in an attempt to explain differences in resistance between the closed chest and opened chest condition.
Respiratory physiologists and clinicians have been concerned with the resistance to air flow in the tracheobronchial tree for many years. Its measurement can be used in the diagnosis and in the evaluation of the treatment of lung desease. Resistance is the relationship of the driving pressure needed to overcome frictional forces opposing a given flow in or out of the respiratory system.
Rohrer (25) in 1915 was one of the first researchers to consider that the total resistance of the respiratory system could be partitioned into contributions made by various
segments of the airway. He applied fluid mechanical principles to morphometric data from the human lung and made predictions concerning the distribution of resistance in the lung. He estimated that 54% of the resistance was in the upper airways, that is, the nose, pharynx, larynx and trachea. The remaining 46% was in the bronchi and smaller airways. He also presented predictions concerning the distribution of the resistances in the lower airway, ascribing 70% to airways less than I mm. in diameter.
Small or peripheral airway function has become of great importance to respiratory physiologists and clinicians in recent years. It has been suggested (1Z) that these are the airways first altered in the pathogenesis of some types of obstructive lung disease and that if subtle changes were detectable at an early stage then the disease process could be reversed with appropriate treatment. Numerous tests have been suggested to study these areas of the lung; however, at this time there is no general agreement on the test of choice (11).
The term peripheral or small airway has been defined differently by various authors and it has become an arbitrary term and perhaps even confusing. A bronchus is both small and peripheral compared to the trachea, yet is both large and central compared to a respiratory bronchiole.
The range in diameter of the bronchi is quite large as can be seen in Table 1. The more peripheral an airway the smaller its diameter, the shorter its length and the larger the total cross-sectional area of its generation because of the large degree of branching. Previous investigators have used several criteria to classify the airways as small or large: 1) large airways are those which receive their blood supply from the bronchial circulation while small airways are those which receive their blood supply from the pulmonary circulation (21); 2) large airways are those which are larger than an airway which contains a pressure measuring device while small airways are smaller than the one which contains the catheter (13); 3) during forced expiration large airways are those downstream of the equal pressure point while small airways are those upstream of the equal pressure point (18). In this paper small will be synonymous with peripheral and large will be equivalent to central. Definition 2 will be used to functionally classify the airways. In general the catheter was placed in airways 2.8 to 3.Z mm. in diameter. Therefore airways less than 3.2 mm. in diameter are classified as peripheral.
The pressure can be measured at various points in the system and pairs of these pressures can be used to measure different pressure drops, which can be used to compute
Table 1. Nomenclature and dimensions for human bronchial
tree based on data of Weibel (31).
Diameter Length Total Cross Section
Generation Name (cm) (cm) (cmz)
0 Trachea 1.8 12.0 2.54
1 Main Bronchi 1.22 4.76 2.33
z Lobar Bronchi 0.83 1.90 2.13
3 Segmental Bronchi 0.56 0.76 2.00
4 Subsegmental Bronchi 0.45 1.27 2.48
5 0.35 1.07 3.11
thru Small Bronchi
10 0.13 0.46 13.4
11 0.109 0.39 19.6
13 0.082 0.27 44.5
14 0.074 0.23 69.4
and Terminal Bronchioles
15 0.066 0.20 113.0
16 0.060 0.165 180.0
thru Respiratory Bronchioles
18 0.050 0.117 534.0
19 0.047 0.099 944.0
thru Alveolar ducts
23 0.041 0.050 11,800.0
24 Alveoli 244 pm Z38 pm 43-80 mz
resistances of different sections of the airway. The respiratory system is schematically shown in Figure 1 with examples of points indicated where pressure can be measured. Definitions of some of the symbols to be used in this paper are shown in Table 2.
Since the lung is located within the thorax it is
necessary to estimate the surrounding pressure of the lung to eliminate the elastic pressure of the chest wall. This may be accomplished in part by opening the chest in the experimental animal, which makes the surrounding pressure equal to atmospheric pressure or by measuring the pleural pressure (P pl ) in the intact subject (20).
The pressure in the trachea (P tr ) can be measured by inserting a pressure sensing device between the cartilaginous rings into the trachea. The pressure at the airway opening (P so ) can be determined by measuring the lateral pressure at that point. The measurement of bronchial pressure (P br ) poses a problem. If a catheter is passed into a small bronchus it may obstruct flow and result in erroneous values. In an early study Sonne (28) in an attempt to measure alveolar pressure pulled catheters through the parenchyma of the lung and wedged them in the periphery of the lung. Macklem and Mead (13) have refined this technique and have made lateral pressure measurements in bronchi (P br)
Figure 1. Schematic representation of the respiratory system
including the chest wall, lung and airways. Shown
are points where pressure can be measured and the
pressure drops which can be determined using pairs
of these pressures.
\,& p A Ptot 00
p 6r z A p I/
PPA > A P z L
Table 2. Definitions of symbols used in this paper.
S ymbol1 Definition
P1p pleural pressure
Pbr lateral pressure in the bronchus
P tr lateral pressure in the trachea
P a lateral pressure at the airway opening
6p. P pl-Pb; peripheral pressure drop includes a resistive, elastic and inertial component
6PC. p -rp t central pressure drop includes a resistive and inertial component
AP L' p t- pl lower pulmonary pressure drop includes an elastic, resistive and
6p ua P so-P t upper airway pressure drop includes a resistive and inertial component
6PP -P; total pressure drop includes a resistive, elastic and inertial component
TOT so pL'
i~p p resistive pressure drop in the periphery of the lung AP rssiepesr rpi h eta iwy
Ap L resistive pressure drop in the coera airways
Rp resistance in the peripheral portions of the lung includes airway and tissue
RC resistance in the central airways includes airway resistance
RL resistance in the lung includes airway and tissue resistance
by wedging the catheters in the wall of the airway. In this method the catheter does not impede flow between the carina and the point of measurement.
The difference between the bronchial pressure(Pb
and pleural pressure (P 1l) is called the peripheral pressure drop (P ). During the breathing cycle it consists of an elastic component (P el ) related to lung volume, a resistive component (P R) related to friction of air flow through the airways (R r) and to the resistance of the tissue to deformation (P R is) and an inertial component (P IN) related to the acceleration of gas.
Ap = P el+ P R+ P I
The difference between the bronchial pressure ( r
and tracheal pressure (P tr)is the central pressure drop (6P ). It consists of a resistive component (PR) related to the friction of air flow through airway (P R)f and an
The peripheral pressure drop (AP) and the central pressure drop (Ap ) can be added together to obtain the total pulmonary pressure drop (AP ). This represents the total pressure drop from the pleural space to the trachea.
AP = AP* + AP
Ap = Pel + PR + PIN
The difference between the tracheal pressure (Ptr) and the airway opening pressure (P o) is the upper airway pressure drop ( A ). It consists of resistive component (P ) retJA R
lated to the flow of air (Pfr) and an inertial component (PIN) related to the acceleration of gas.
APuA = PR + PIN
The upper airway pressure drop (AP ) can be added to the lower airway pressure drop (AP') to obtain the total pressure drop (AP To) from the pleural space to the airway opening.
6P To = APu + Ap + Ap
ToT =Pel + PR + PIN
r fr Rtiss
To obtain the resistance in each of these sections it is necessary to relate the resistive component (P R) of that section to the flow (V) of gas through the system. Therefore, one can compute total resistance (RTOT), peripheral resistance (RP), central resistance (R c), lower pulmonary resistance (RL) or upper airway resistance (RA).
RTOT = RUA +RL
RTOT = RUA + R + RC
Predictions of Peripheral Resistance Based on Anatomic Data
As previously stated, Rohrer (25) predicted that 70%
of lower airway resistance was in airways less than 1 mm. in
internal diameter. His predictions were based on anatomical data obtained from an excised human lung, which had collapsed to zero volume. According to Weibel (31), Rohrer underestimated the total number of airways less than 4 mm. in internal diameter and therefore overestimated the pressure drop peripheral to that size. Weibel's data are perhaps more relevant to the intact lung because it was obtained from human lungs which had been inflated to 75% of total lung capacity prior to fixation. These more recent measurements suggest that the peripheral resistance should be considerably less than predicted by Rohrer because of the underestimation of total cross-sectional area of the airways in the periphery.
All of the data obtained from anatomical studies of excised lungs are open to criticism when extrapolations are made to the intact lung. When the measurements on the excised lung are made the effects of nervous input and of circulating active compounds have been removed. The alteration in the tonus of the smooth muscle in a non-living preparation could alter the resting diameters and lengths observed compared to the intact animal. In addition, fixation of tissue may result in errors in size relative to the in vivo condition.
Green (7 ) has used the anatomical data of Weibel (31)
to estimate flow resistance of each generation of the human lung. Assuming laminar flow and a regular dichotomous branching pattern of the tracheobronchial tree he applied Poiseuille's law. He found that the maximal resistance occurs at the sixth generation and the resistance of each generation becomes less as one moves peripherally and the airways become smaller. The apparent discrepancy of the smaller tubes having a lower resistance is explained by the increase of total cross-sectional area of each generation and the reduction in length of the daughter compared to parent airways. According to the Poiseuille relationship the pressure drop is inversely related to the crosssectional area squared and is directly related to the length. Therefore, an increase in area with a simultaneous reduction in length must result in a reduced resistance.
Horsefield and Cunning (8 ) reported measurements made on a cast from a human lung inflated to five liters. They reported the branching in the bronchial tree to be by asymmetrical dichotomy, while in the lobules the branching is more symmetrical. They found, as did Ross (26) in the dog, that the number of terminal bronchioles arising from a bronchus of any given diameter decreased with increasing distance from the carina. They found 28 x 10 terminal bronchioles which is less than one half the number reported
by Weibel (31). Using their own data, they predicted that the structures in which the resistance to laminar flow is highest are the airways between 1 and 4 mm. in diameter.
Pedley et al. (24) have used the anatomical data of Horsefield and Cumming ( 8 ) to predict the pressure drop along the tracheo-bronchial tree and did not make the assumption of Poiseuille flow. They have taken into account the non-parabolic velocity profiles, which occur in all but the smallest airways. They found that the resistance in the larger airways calculated in this way is greater than if the Poiseuille equation is used, while there is little effect on the calculated resistance of the smaller airways.
Previous Measurements of Peripheral Airway
Resistance by Retrograde Catheters
The retrograde catheter technique (13) is used to measure lateral pressure in small bronchi. The catheter is wedged in the bronchus and passes out through the parenchyma of the lung. If pleural pressure and tracheal pressure are also measured then the peripheral (&P; central (sP ) and lower pulmonary pressure drop (AP ) can be determined. The frictional component of each of these pressure drops can be related to flow to obtain peripheral resistance
RPcentral resistance (R C) and lower pulmonary resistance
In 1967 Macklem and Mead (13) measured the resistance of central and peripheral airways by means of a retrograde catheter wedged in airways 1.5 to 2.5 mm. i.d. Using an oscillatory technique the measurements were made in opened chest, living dogs and in excised lungs from various species. In this technique the lungs are oscillated at resonant frequency with a loudspeaker driven by a sine wave generator. It is a means by which the resistive component can be separated from the total pressure and resistance calculated by relating the resistive pressure to flow. They found that peripheral resistance was too small to detect above 80% of the vital capacity, but increased at lower volumes to 15% of RL at 10% of the vital capacity. The authors stated that there did not appear to be any systematic qualitative differences between the flow resistance measured in opened chest dogs and that for excised lungs.
In 1969 Macklem et al. (15) conducted a similar series of experiments to study the contribution of a wider range of airways to lower pulmonary resistance. They found that, at 60 80% vital capacity, airways less than 3 mm. in diameter contribute 30 50% of PIL I those between 3 and 8 mm. in diameter contribute 15% to RL and those larger than 8 mm. in diameter contribute 45% to RL. At Z5 50% of vital
capacity airways less than 3 nun. in diameter contribute 14 24% of RL, those between 3 and 8 rm,. contribute 55% of R. and those larger than 8 mm. contribute 25% of R.. The fact that airways less than 3 mm. i.d. account for only approximately 20% RL at low lung volumes and for 40% at high volumes does not necessarily imply that resistance of the small airways increases at high volumes. Although there was a tendency for the resistance of airways less than 2 mm. i.d. to be higher at high lung volumes the difference was not significant. There was a sevenfold decrease from 1. 08 to 0. 14 cm. H z0/liter/second in the resistance of airways of 3 to 8 mm. i.d. when compared at low and high lung volumes. It was concluded that the 3 to 8 mm. i.d. airways are responsible for the change in RL with lung volume.
At 50% of vital capacity vagotomy reduced RL from
approximately 2.3 to 0.9 cm. H 20/liter/second, a reduction which was due mainly to a decrease in resistance of the 3 to 8 mm. i.d. airways. In all dogs studied the change in RL with lung volume was reduced following vagotomy. The data indicate that the increase in RL with decreasing lung volumes is mediated at least in part through the vagus and is a direct result of constriction of airways between 3 and
8 mm. i. d.
Woolcock et al. (35) found that upon cervical vagal stimulation the dogs showed an increase in RL due to an increase in both R Cand R.The responsiveness of individual dogs was variable; in several dogs the central airways showed a greater change, while in others the peripheral change was greater. In any given dog the response to vagal stimulation was reproducible throughout the experiment. In addition, dogs which showed only small changes in peripheral resistance upon vagal stimulation had no significant change in the static elastic properties of the lung, but dogs which showed a response in the peripheral airways had an increase in elastic recoil so that at a AP' of 30 cm. HZ0 the vital capacity was only 86% of the control value. This means that a large increase in peripheral airway tonus reduced the capacitance of these airways or impeded the filling of alveoli. In none of the dogs did the compliance become frequency dependent.
Woolcock et al. (36) in 1969 described experiments which were designed to investigate the role of the sympathetic nervous system in the control of airway caliber at various sites along the tracheobronchial tree. The interaction of beta adrenergic blockade (.25mg./Kg.) and vagal stimulation was studied in nine dogs. Treatment with propramolol resulted in a decrease in R Cand a 236% increase in R The absolute changes in RCand RPwere similar in magnitude,-
therefore no net change in RLwas observed The mean value for the ratio of RPto RL increased from 13% to Z6% when treated with propranolol. Vagal stimulation without beta blockade increased R C350% and R P700% In response to vagal stimulation following beta blockade RCincreased 600% and R p 2,*075% Section of the Ansa Subclavia nerve produced results which were similar to those obtained with propranolol, which supports the supposition that the drug's bronchial effects are due to its beta blocking action.
Wood (34) has used the retrograde catheter technique
to study the effect of physical properties of gases on lower pulmonary resistance. He found that in airways smaller than 4 mm. resistance was independent of gas density. This confirms that flow in these regions is laminar in character. The more central airways were dependent on density demonstrating a non-laminar flow condition in these regions. At 35% of vital capacity it was found that RPaccounted for 30% of the lower pulmonary resistance. It was also reported that the peripheral resistance varied little with lung volume while the central component varied significantly.
In 1973 Battista et al. ( 3 4 ) described a technique for chronically implanting retrograde catheters in the peripheral airways of dogs. In the preparation the chest was intact and the dogs were able to breathe spontaneously
during the measurement of resistance. In addition, the dogs were lightly anesthetized and could be used for repeated studies. A catheter with a 4 mm. flare was used; therefore peripheral airways would be those having a diameter of less than 4 mm. In five dogs at functional residual capacity, the mean RC was 0.65 cm. H 20/liter/second and mean RP was
1.65 cm. H 20/liter/second and the mean RL was Z.3 cm. H 20/iiter/second. This means that in these dogs the R P/RL ratio was 70%, a value equivalent to that predicted by Rohrer (25) but three times that found by Macklen and mead (13).
Recently Silvers et al. (27) partitioned airway
resistance in normal and emphysematous isolated human lungs. During passive recoil at 50% of the vital capacity in four of six lungs R p was more than 50% of RL. The reduction in the tracheal length in this particular preparation has to be considered in evaluating the data. If lungs had been studied with the trachea intact, more resistance would have been added to the central component and a smaller RP/RL ratio would have been observed.
In summary there is a discrepancy in both the measured and predicted contribution of peripheral airways to lower pulmonary resistance reported by various investigators as seen in Table 3. Some of the discrepancy may be explained
by the difference in the size of the airway in which the catheter was wedged in the various preparations. However, there are other major differences between the conditions of the various preparations. For example, Battista et al. 3 ) measured resistance in the closed chest and spontaneously breathing dog, while Macklen and Mead (13) measured the resistance in the opened chest dog while the lungs were oscillated at resonant frequency.
The variation of peripheral resistance with lung volume appears also to be another disputed point. In Macklen and Mead's (13) results it decreases with increasing lung volume; the opposite is reported by Macklen et al.(15);, while Wood
(34) reports no lung volume dependence.
It is the purpose of this study to determine the contribution of peripheral airways in the closed and opened chest dog. In order to do this measurements were made at various trarspulmonary pressures with the chest closed then at the same transpulinonary pressures following a procedure where the sternum was split and the chest widely spread. In addition, the effect of pneumothorax and beta receptor blockade on the contribution of peripheral airways was studied in an attempt to explain the difference between the closed and opened chest.
Table 3. Summary of data obtained by various investigators using
the retrograde catheter technique.
Airway Size RpL
Authors Animal (mm.) Lung Volume cm. H 2O/L/Sec. Rp /RL
Macklem and Mead (13) Dog 1.5 2.5 40% VC 0.17 1.70 0.10
Macklem et al. (15) Dog < 3.0 25-50% VC 0.35 1.90 0.18
Woolcock et al. (35) Dog .8 2.7 35% VC 0. 21 1.20 0.18
Woolcock et al. (36) Dog < 3.0 35% VC 0.19 1.95 0.10
Battista et al. ( 4) Dog < 4.0 FRC 1.70 2.59 0.66
Silvers et al. (27) Human 2.0 3.0 50% VC 0.85 1.51 0.56
Wood (34) Dog 2.0 3.0 35% VC 0.48 1.56 0.31
The experimental objectives required the development of techniques to measure lower pulmonary resistance (RL) and peripheral resistance (RP) in the dog with closed and opened chest at various lung volumes. This required the measurement of 1) lung volume; Z) flow rate of gas at the airway opening; 3) pleural pressure, 4) tracheal pressure; 5) bronchial pressure and 6) an estimate of alveolar pressure. A system diagrammed in Figure Z was constructed and is described in detail in the following pages. The intact dog is placed within a plethysmograph and all measurements are made in the supine position. The oscillatory technique ( 5), which is described in more detail later in this section, was chosen for several reasons. It is a method that generates high flow rates with relatively small volume excursions and allows for the development of a relatively high frictional component in the pressure drop. It also provides a convenient method for applying positive pressure to the lung to place it at different volumes without interfering with the measurements. Since one of the objectives was to compare 25
Figure 2. Diagram of the equipment used in the experiments.
For clarity, only the lungs are shown; however, in
the actual experiments the intact dog was placed within the plethysmograph in the supine position.
Ppi P br
d n Pt,
the opened chest results of this study to previous authors'
(13) the use of similar techniques was judged appropriate.
Justification of the Retrograde Catheter Technicue
Compared to methods in which the pressure measuring device is passed down the airway to the bronchus (14) the retrograde catheter technique offers the advantage of not obstructing the airway between the catheter tip and the airway opening. Undoubtedly, the catheter traumatizes the lung tissue through which it passes, and evidence of damage is often seen on autopsy. Air flow to areas peripheral to the catheter is altered; however, the affected region is quite small since the catheter is generally located less than 1 cm. from the surface of the lung. As can be seen in Figure 3 the areas which have been damaged by the catheter are not those which receive flow from the bronchus where the lateral pressure is measured. Rather it is an adjacent area, which is presumed to have suffered insignificant damage that is measured.
In their discussion of the technique, Macklem and Mead
(13) have established limits on the effect the catheter could have on flow. one extreme is that the catheter would have no adverse effect and, therefore, the pressure measurements are accurate. The other extreme is that there is no flow
Figure 3. Diagram of a portion of the lung with the retrograde catheter wedged in a small bronchus. The
shaded area is that portion of the lung which
receives its ventilation from the bronchus where
lateral pressure is measured.
Pleural Space Retrograde Catheter
past the catheter. In this case the bronchus into which the catheter opens would act as an extension of the catheter and the pressure would be measured at the nearest branch point toward the hilus where flow is occurring. Since the pressure is measured at a more central point than the location of the catheter tip, the pressure drop between the alveolus and catheter may be overestimated. It seems reasonable to conclude that in cases between the extremes considered, one will overestimate the peripheral pressure drop.
Errors also result when lateral pressures are used to estimate pressure differences due to frictional resistance in a tube, if the linear velocities are different at the two points of measurement. In the airways this convective acceleration should lead to an overestimation of the pressure drop along the peripheral airways on deflation but to an underestimation on inflation, resulting in an asymmetrical inspiratory and expriatory pressure flow curve. Such asymmetry was not observed in the present experiments nor in those of Macklem and Mead (13). The acceleration pressure can be calculated according to Bernoulli's equation (P ca Pv 2 /zg). where P is gas density, v is linear velocity of the gas, and g is the acceleration due to gravity. Consider the trachea where the acceleration forces should be greatest because the linear velocities are the largest. If the diameter is 20 mm.
and the flow is 1 liter/second, then
Flow = 1000 cm 3/sec.
Diameter = .2 cm.
Density = 1.162 10-6 cm. H20 sec2/cm2
Flow 1000 cm 3/sec2
Area T (.1 cm)
v = 318.3 cm/sec
-6 2 22
1.162 x 10 cm. H O-sec /cm (318.3 cm/sec)
P ca= .0588 cm.H z
In the present experiment a tracheostomy tube of 10 mm. i.d. was used, which could result in an overestimation of RL due to convective acceleration of up to 0.96 cm. H2O with expiratory flows of 1 liter/sec. Since the linear velocity of the peripheral airway is relatively small the effect of convective acceleration on the measurement of peripheral resistance should be minimal. An asymmetrical pressure flow curve was not observed for either the lower pulmonary
or peripheral pressures, indicating that the Bernoulli effect was minimal in the preparation.
If the traumatized area of the lung where the catheter is wedged has a mechanical influence on the adjacent area which is studied, then phase shifts between this part and the rest of the lung uould be expected. Macklem and mead
(13) did not observe such phase differences in their study. They estimated dynamic compliance from the pressure difference between the catheter and the pleural surface and found it to be independent of frequency and the same as the dynamic compliance measured from transpulmonary pressure. If phase differences had been induced by the catheter then dynamic compliance measured using the retrograde catheter should have been frequency dependent and different from dynamic compliance measured by transpulmonary pressure. This type of analysis is dependent upon the assumption that a single bronchus is representative of all other bronchi in the lung of a similar size. In the normal lung, which is free of frequency dependence of compliance, this is probably true and is supported by the uniformity of results obtained by Macklem and mead (13) and in the present study when the catheter is wedged in the same size airway of different lungs. The agreement between dynamic compliance determined from P L and P P also supports the assumption that the measurement is
representative of all airways that size.
In the oscillatory technique (8 ) the lungs are connected to a loudspeaker, which is powered by a sine wave generator and driven at a frequency of about 4-5 cycles per second. The volume excursions are relatively small (40 ml.) and, therefore, the changes in pressure due to the elastic component are small, but because of the high frequency both the inertial and resistive components are relatively high.
Descriptions for electrical circuits in the study of alternating current have been developed and can be used by analogy in the analysis of the respiratory system when the breathing pattern is in the form of a sine wave. In this case the impedance can be calculated
Z R + ( W L - _-C)
where z is the impedance, R is the resistance, W is Zlr times the frequency, C is the compliance and L the inertance. As frequency is increased from the normal breathing rate, to several cycles per second and WL I- approaches zero
the impedance becomes dependent upon only the resistive
component of the pressure drop. This is the resonant frequency and if the pressure drop is measured it is equal to the resistive pressure drop, which if divided by flow will give the flow resistance of the system.
During quiet breathing, inertial pressure (P IN ) is
negligible and AP is the vector sum of P e 1 and P R* Since the elastic pressure is in phase with lung volume, an increasing proportion of the volume can be subtracted from AP until the corrected signal is in phase with flow (19). At that time the remaining pressure drop is due to resistance.
In the present experiments the lungs were oscillated at 4 cycles per second and pressure-flow relationships were observed on an oscilloscope. Loops were observed indicating that some part of the pressure drop was out of phase with flow, By electrically subtracting a proportion of volume (KV L ) from each pressure drop (AP and LP ) the loops were closed (19). Since the pressure drop was in phase with flow at this point, it is the resistive pressure drop and if divided by the flow, will give the flow resistance (Rp and RL).
Volume (V L ), flow (V M ), peripheral pressure drop
(AP p ) and lower pulmonary pressure drop (6P L ) were recorded against time after loop closure. The peak to peak pressure
signals (APp and P L) were divided by the peak to peak flow signal (V M) to obtain resistance (Rp and RL). Central resistance (RC) was obtained by subtracting Rp from RL.
Measurement of Flow, Volume and Pressure
Flow rate (VM) was measured with a Fleisch pneumotachograph and a Statham PM-97 differential pressure transducer. The pneumotachograph was mounted in the wall of the plethysmograph and could be connected to the tracheostomy tube to measure flow rate. The pneumotachograph and its pressure transducer gave a linear response to flows between t 1.8 liters/second.
A whole body volume displacement plethysmograph, as seen in Figure 2, was constructed to accomodate medium sized dogs (12-18 Kg.). The change in lung volume was measured with a Krogh spirometer whose change in position was measured with Harvard apparatus rotary motion transducer. The plethysmograph had a total empty volume of 141 liters and was equipped with ports through which the catheters could be connected. The system was linear over the range used and had a flat frequency response up to 10 cycles per second.
Two Statham PM 131tc differential pressure transducers were used to sense the airway pressures during the
experiments. One sensed the pressure difference between the pleural and tracheal pressures and the other between pleural and bronchial pressures. The two transducers were matched in sensitivity and were linear over the range of
-50 to +50 cm. H 2 0.
Pleural pressure was estimated using the esophageal balloon technique (37). The balloons were 5 cm. long and had a circumference of 3.14 cm. They were molded using liquid latex and made as thin as possible without leaking. They were mounted on polyethylene catheters 45 cm. long, which had an inside diameter of 1.67 mm. and an outside diameter of 2.41 mm. Prior to use, the volume range which results in a zero pressure for each balloon catheter was determined and a minimal volume in this range was used during the experiment. The balloon was passed down the esophagus and into the stomach so that positive deflections were observed on inspiration. It was then withdrawn until the pressure was negative relative to atmospheric pressure. The response of this system to a square wave change in pressure was better than 90% in less than 10 msec.
Tracheal pressure was measured using a piezometer ring, which was constructed as part of a tracheotomy tube as seen in Figure 4. The tube was 13 cm. long and had an inside diameter of 1 cm. The lateral pressure was measured 4.5 cm.
Figure 4. Tracheostomy tube with inflatable cuff and piezometer ring for measurement of lateral pressure.
The cuffed end of the tube was placed within the
trachea of the dog.
Trocheostorny Tu6e Pie zometer Ring
Inflatable Cuf f
from the tip of the tube in the trachea. An inflatable cuff was slid over the tracheal tube, so that a leak proof seal could be made between the trachea and the tube itself.
The retrograde catheters were made from polyethylene tubing which had an inside diameter of 1.39 mm. and an outside diameter of 1.90 mm. one end was flared to a diameter of 3.0 mm. The overall length of the catheters was 75 cm. A thin suture was tied around the flared end of the catheter and was used in placement of the catheter and as a guide during the anatomical dissection at the termination of the experiment. The response of the catheter transducer system to a square wave change in pressure was better than 90% in less than 10 msec.
As seen in Figure 2, the dog lung could be connected to a Harvard ventilator or a loudspeaker by the appropriate positioning of a three way valve. An Acoustical Research model 5 loudspeaker was mounted in a sealed container and was powered by a sine wave generator. A positive pressure could be applied to the loudspeaker box and in turn to the airway opening. In this way various transpulmonary pressures could be attained while the lung was being oscillated at 4 cycles per second.
A schematic of the electronic system is shown in Figure 5.
Figure 5. Schematic of the electronics used in the experiment. The rectangles represent transducers, the
triangles represent amplifiers and the circles potentiometers. After the subtraction of YVL
the pressure drops are the resistive pressure
Transducers Amplifiers Computer Physiograph Pen Writer
~PL 6r Ap AP KV'A
AI- K2 VL APL v Lt LI\H
The pressure transducers for airway pressure measurements were powered and amplified with two Brush transducer couplers and amplifiers. The transducer used in the determination of flow was powered by a Narco Biosystems straingauge coupler and amplified with a 7070 amplifier. The transducer used to sense the motion of the Krogh spirometer was powered by a Narco Biosystems transducer coupler and amplified with a 7070 amplifier. A Heathkit analog computer was used to subtract a proportion of the volume signal from the airway pressures as required in the oscillatory technique. All four signals (V, VL, 6Pp and APL) were recorded on a Narco Biosystems Physiograph equipped with a rectilinear pen-writing system. The flow signal was also monitored as a function of the two airway pressures on a Tektronix model 5103N dual beam storage oscilloscope.
One cc. heparinized tuberculin syringes were used to obtain .5 cc arterial blood samples anaerobically from a catheter placed in the femoral artery. The samples were analyzed using a Radiometer Blood Micro System for pH, PCO2 and POZ by standard techniques.
Mongrel dogs which were maintained by the Animal Resources Department were used in these experiments and all were treated in accordance with the American Physiological Society's Guidelines for Laboratory Animal Care.
The dogs were anesthetized with sodium pentobarbital (27 mg./Kg. i.v.), intubated and prepared for surgical procedures. Catheters were placed in the femoral artery and vein for blood gas analysis and drug infusion, respectively.
A tracheostomy was performed 2 cm. below the larynx
and a cuffed tracheostomy tube inserted. A midline incision was made caudal to the sternum, the abdomen was opened and the diaphragm was exposed. The esophageal balloon was placed, as previously described, and the animal was allowed to stabilize.
Placement of the Retrograde Catheter
A polyethylene catheter, larger than the retrograde catheter was guided into the lower right lobe of the lung with a bronchoscope. The catheter was passed peripherally until resistance was encountered. A small amount of air flow was applied to the catheter, which presumably expanded the airways peripheral to the tip of the catheter and allowed its movement more peripherally. When the guide catheter was as peripheral as possible a piano wire was passed through it down to the peripheral airway, gently passed through the lung parenchyma and diaphragm and grasped through the abdominal incision. At this point one end of the wire was
visible at the trachea and the other in the abdomen. The guide catheter was removed and the wire was left in place. The retrograde catheter was then passed over the wire and guided peripherally along the path of the wire. The catheter was passed through the lung parenchyma and diaphragm and grasped in the abdomen. The guide wire was then removed. Finally the catheter was moved peripherally until the flared end became wedged in a peripheral bronchus.
The dog was than placed in the plethysmograph in the supine position and the catheters connected to the appropriate transducers. Following a period of spontaneous breathing the dog was paralyzed with succinylcholine (1.5 mg./Kg. i.v.) and ventilated with a Harvard constant volume ventilator. The paralyzed condition was maintained with supplemental infusions of succinylcholine. A quasistatic pressure-volume curve was obtained by inflating the lung to a pressure of 25 cm. HO2 and allowing it to deflate in 5 cm. H 20 steps down to functional residual capacity. During this procedure transpulmonary pressure and total pressure as well as the change in lung volume were recorded.
Following a brief psuse the lungs were reinflated to 25 cm. H O and allowed to deflate to transpulmonary pressures of 15, 10, and 5 cm. H 2 while being oscillated at
4 cycles per second. By the method previously described, the pressure-flow loops monitored on the oscilloscope were closed and the corrected pressure drops (4P and &P L recorded as a function of time along with VMand V L. Using these measurements R13, R C and RL were computed at each transpulmonary pressure.
Checks on the Patency of the Retrograde Catheter
Since there was frequent blockage of the retrograde catheter, it was imperative to have criteria to determine its patency. The first of these was the identity of quasistatic pressure volume curves determined using APand AP Since the major component on a quasi-static maneuver of these two pressure drops is P elthey should have equal amplitude unless the retrograde catheter was blocked. A similar consideration applies to the total amplitude of tiP' and iP monitored at normal breathing frequencies. Equality in their amplitudes under these conditions was also considered as evidence for patency of the catheter. During oscillations a blockage of the catheter resulted
in a reversal of the pressure-flow loop and a reduction in the amplitude of AP P, The presence of any of these three conditions could usually be alleviated by flushing the retrograde catheter with air and the experiment continued.
At the completion of all experiments the dogs were sacrificed by an intravenous injection of KCl. The lungs were removed with the catheter intact and dried for three days at a transpulmonary pressure of ZO cm. H 20. Each lung was then dissected, following the suture tied to the catheter, to determine the size airway in which the catheter was wedged.
The protocol described above was common to three series of experiments which had differences in their individual design.
Nine dogs were prepared as described above and the quasi-static pressure volume curve determined with the chest intact. Measurements of resistance (RL and R P) were made during oscillation at transpulmonary pressures of 15, 10 and 5 cm. H 20. The sternum was then split and the chest widely spread with retractors. The dog was ventilated and expired against 5 cm. H z0 positive end-expiratory pressure (PEEP) to prevent collapse of the lung. Anesthesia end muscle paralysis were maintained at a level comparable to the closed chest condition. The quasi-static pressure volume curve was repeated and resistance determined during oscillation at transpulmonary pressures of 15, 10 and 5 cm.H 0.
Six dogs which were included in Series I had pneumothoraces created between the closed and opened chest condition. After the closed chest measurements were made large bore tubes were placed between the 5th and 6th ribs and the quasi-static pressure volume curve determined. During oscillation the resistances (RP and RL) were determined at transpulmonary pressures of 15, 10 and 5 cm. H 0. After obtaining the data in the pneumothorax condition the chest was opened and measurements were made as described in Series I.
Six dogs were prepared as described in the common protocol and the quasi-static pressure volume curve determined. RL and R were measured during oscillation at transpulmonary pressures of 15, 10 and 5 cm. H 20. Following beta adrenergic receptor blockade with propranolol (lmg./Kg.i.v.)(23) the resistance measurements were repeated at transpulmonary pressures of 15, 10 and 5 cm. H O during oscillation with the chest intact. The sternum was then split and the chest wall widely retracted with maintenance of the beta adrenergic receptor blockade. The quasi-static pressure volume curve was determined. R and R were measured during oscillation at transpulmonary pressures of 15, 10 and 5 cm. H 0. Throughout the
experiment a surgical level of anesthesia and muscle paralysis were maintained. As in the previous 2 series, ventilation was performed against 5 cm. H 2 0 PEEP once the chest was opened.
Figure 6 is a plot of the lower pulmonary resistance
(RL) and the peripheral resistance as a function of trainspulmonary pressure ( p)for series I. Each point represents the mean of 13 dogs. R1L in both the closed and opened chest condition is curvilinearly related to Pt with the greatest resistance observed at the lowest pressure. Rs in the closed chest and opened chest conditions are not significantly different from each other when compared by multiple linear regression analysis (29). R pin both conditions is also curvilinearly related to P tpwith a minimal value observed at a P tpof 10 cm. H 20. RP is significantly (p <.02) reduced in the opened chest condition when compared to the values with the chest intact. At a P t of 10 cm. H 2 01 RP is reduced from 0. 99 to 0. 54 cm. H 0 /liter/second by opening the chest. This is a reduction
to 54% of the control value. At this same pressure the R to R.ratio is reduced from 50% in the closed chest to 33% in the open chest condition.
Figure 6. Mean peripheral resi-stance (solid lines)
and lower pulmonary resistance (dashed
lines) are plotted as a function of transpulmonary pressure for the 13 dogs of
series I. Values for the closed chest
condition are indicated by closed circles (0) and those for the opened chest condition are indicated by open circles (0).
Transpulmonary pressures of 15, 10 and 5 cm H ?0 in both the closed and open chest
condition correspond to 97, 75 and 45%
of the opened chest vital capacity. The
lines between the points were drawn by
n= 13 CI..d chest
o open (best :3.0
0 5 10, 15
Transpulmono-y Pressure (cml H.20)
Figure 7. The mean difference between the closed
chest and opened chest for peripheral
resistance (Rp) and lower pulmonary
resistance (RL) is plotted as a function of transpulmonary pressure. A
positive difference means the opened
chest resistance was less than the
closed chest value. one standard
error of the mean is shown above and
below the mean.
Series I n 13
<- 5 TI
5 10 15 5 10 15
Transpulmonary Pressure (cm Hp0)
Table 4. Individual data for the 13 dogs of series I. All
values are expressed in cm. H O/liter/second.
C = closed chest and 0 = opened chest.
p tp 5 cm H 2 0 10 cm H 2 0 15 cm H 2 0
RL Rp RL Rp RL R p
Dog C 0 c 0 c 0 c 1 0 c 1 0 c 0
1 4.76 2.28 3.43 1.25 2.68 1.27 1.48 1.51 2.63 1 66 1.43 1.04
2 2.22 2.13 0.70 0.25 1.76 1.93 0.90 0.44 1.61 2.35 1.14 0.78
3 5.49 9.08 2.05 0.28 4.39 7.28 1.31 0.56 4.38 5.90 1.45 0.57
4 3.85 2.15 1.19 0.95 3.41 1.58 0.82 0.70 3.25 1.64 0.88 0.68
5 1.98 1.64 0.57 0.83 1.50 1.24 0.47 0.56 1.12 1.32 0.36 0.74
6 1.35 1.53 0.95 0.29 1.27 0.64 0.97 0.38 1.28 0.70 1.06 0.41 Ln
7 3.19 5.33 2.06 1.17 1.83 4.62 1.23 0.85 1.95 4.32 1.39 1.43
8 3.67 1.59 1.31 0.51 1.54 1.23 0.61 0.31 1.25 1.55 0.65 0.76
9 2.90 4.30 1.38 0.82 2.20 1.99 1.09 0.60 2.10 1.83 1.41 0.95
10 1.44 1.51 0.83 0.24 0.87 0.97 0.47 0.24 1.15 1.27 0.68 0.31
11 3.08 3.20 0.78 0.65 2.71 2.85 0.67 0.58 2.32 2.91 0.97 0.91
12 1.86 1.69 1.35 0.88 1.66 1.25 1.30 0.48 1.94 1.47 1.49 0.82
13 3.52 2.51 2.97 1.44 2.15 1.57 1.57 0.88 2.14 1.64 1.65 1.01
3.02 2.99 1.50 0.73 2.15 2.18 0.99 0.54 2.04 2.20 1.12 0.80
s.e. 0.35 0.60] 0.24 0.11 0.26 0.50 0.10 0.04 0.26 0.39 0.10 0.07
In Figure 7 the difference in resistance between the closed and opened chest conditions is plotted as a function of P tP for RL and R P, Each point is the mean difference for the 13 dogs of series I on a paired basis and one standard error of the mean is drawn as a line above and below the point. Since the value for the open chest condition was subtracted from the closed chest condition value, a positive result indicates a reduction upon opening of the chest. The values for R L were not significantly different from zero and indicate no change in R L at any P tP when the chest is opened. The differences for R P were significantly (p< .05) different from zero, which means that RP is reduced by opening the chest at all three Ptp's tested. The differences appear to be dependent upon P tp with the largest difference observed at 5 cm. H Z 0.
Table 2 contains all individual data for the 13 dogs of series I. Included are the mean and standard error of the mean for each condition and each P tp'
The results of series II, which demonstrate the effect of a pneumothorax, were plotted in the same format as series 1. Figure 8 demonstrates the same curvilinear change Of RL and R P with P tp as was observed in series I. It may
Figure 8. Mean peripheral resistance (solid lines)
and lower pulmonary resistance (dashed
lines) are plotted as function of transpulmonary pressure for the 6 dogs of
series II. Values for the closed chest
are incicated by closed circles (0),
values for the opened chest are indicated
by open circles (o) and values in the
presence of a pneumothorax are indicated
by (X). Transpulmonary pressures of 15, 10 and 5 cm IiZO correspond to 97, 75 and
45% of the opened chest vital capacity.
The lines between the points were drawn
3.0- ClosedJ Chest
o open Chest h X Pneurnlotl,o
0 5 10 15
Transpulrnonary Pressure (Cm H20)
Figure 9. The mean difference between the closed
chest and in the presence of a pneumothorax for peripheral resistance(Rp) and lower pulmonary reistance (RL) is plotted as a function of transpulmonary pressure.
A positive difference means the resistance in the presence of a pneumothorax
was greater than with the chest closed.
One standard error of the mean is shown
above and below the mean.
Series I I
20) c .5
5 10 15 5 10 15
Transpulmonary Pressure (cm H 20)
Table 5. Individual data for 6 dogs of series II. All values
are expressed in cm. H20/liter/second. C = closed
chest, P = pneumothorax and O = opened chest.
ptp 5 mH2 010cm H 2 15 cm H 2O
RL RpRL RE L
Dog C p 10 C p 0 c p 0 C P 0 C P 0 C P 0
8 3.67 3.50 1.59 1.33 0.78 0.51 1.54 1.68 1.23 0.61 0.48 0.31 1.25 1.76 1.55 0.65 0.48 0.76
9 2.90 3.49 4.3 1.3E 1.30 0.82 2.2C 1.95 1.90 1.09 0.91 0.60 2.10 2.00 1.83 1.41 1.20 0.95
10 1.44 1.51 1.5 0.81 0.66 0.24 0.8? 1.06 0.97 0.47 0.56 0.24 1.15 1.04 1.27 0.68 0.57 0.31
11 3.08 2.74 3.20 0.7 1.48 0.65 2.71 2.37 2.85 0.67 0.99 0.58 2.32 2.25 2.91 0.97 0.73 0.91W
12 1.8611.9111.69 1.3 1.3810.88 1.66 1.59 1.25 1.30 1.1210.48,1.94 1.73 1.47 1.49 1.31 0.82
13 3.52 4.08 2.51 2.9 12.82 1.44 2.15 2.11 1.57 1.57 1.44 0.88 2.14 2.24 1.64 1.65 1.78 1.01
x 2.74 2.87 2.46 1.44 1.40 0.76 1.8211.79 1.63 0.95 0.91 0.51 1.82 1.84 1.78 1.1411.01 0.79
s.e. 0.36 0.40 0.45 0.32 0.31 0.16 0.25 0 .18 1.0.27 0.17 0.14 0.08 0.20 0.17 0.23 10.17 0.20 0.10
be seen in both Figures 8 and 9 that the production of a pneumothorax does not change RL or Rp significantly. However, the opening of the chest does result in a significant (p <.0Z) reduction of R At a Ptp of 10 cm. HzO the Rp s in the closed chest, in the presence of a pneumothorax, and opened chest conditions were 0.95, 0.91 and
0.51 cm. HzO/liter/second, respectively. This represents an Rp to R ratio in the three conditions of 53, 51 and 3Z%, respectively.
Figure 10 is a plot of the mean lower pulmonary resistance (RL) and mean peripheral resistance (R p) as a function of transpulmonary pressure (Ptp) for the 6 dogs of series III. In all three conditions RL varies curvilinearly with Ptp with a maximal value at the lowest Ptp. RL in the closed chest; closed chest with beta adrenergic blockade or open chest with beta adrenergic blockade conditions are not significantly different from each other. Rp also varied curvilinearly with lung volume with the maximal value observed at the lowest P tp There was no significant difference between R closed chest, closed chest with beta adrenergic blockade or opened chest with beta adrenergic blockade.
Figure 10. Mean peripheral resistance (solid lines)
and lower pulmonary resistance (dashed
lines) are plotted as a function of transpulmonary pressure for the 6 dogs of
series III. Values with the chest closed
are indicated by closed circles (0),
values with the chest closed with beta
adrenergic blockade are indicated by (X)
and values with the chest opened with
beta adrenergic blockade are indicated by open circles (0). The lines between
the points were drawn by eye.
Series II I-'d (Ihaf
n=60 X pen (4, oh 8Io Dkd ad,
It I I
0\ 5X1 1
Trasplmoa \ Prsue (c '0
Figure 11. The mean differences between the closed
chest without beta adrenergic block and
the opened chest with beta adrenergic blockade are plotted as a function of
transpulmonary pressure. A positive difference means the resistance with
the chest opened with a beta adrenergic
blockade is less than with the chest
closed. One standard error of the mean
is shown above and below the mean.
Series II I
5 10 15 5 10 15
Transpulmonary Pressure (cm H2)
Table 6. Individual data for 6 dogs of series III. All values
are in cm. H20/liter/second. C = closed chest, C+P = closed chest with beta adrenergic receptor blockade;
O+B = opened chest with beta adrenergic blockade.
Ptp 5 cm H20 10 cm H20 15 cm H20
RL RL RL
Dog C C+8 0+8 C C+S 0+ C C+8 0I+ C C+$ 0+8 C C+8 0+0 C C+S 0+8
14 5.22 9.24 6.4' 1.8 2.82 2.15 1.77 1.28 1.05 0.96 1.86 1.91 1.24 1.28 1.47 0.79
15 1.19 1.02 0.9C 0.42 0.46 0.33 0.95 0.96 0.91 0.42 0.45 0.35 0.97 0.97 1.28 0.49 0.62 0.59
16 2.11 2.38 2.61 1.42 1.60 1.77 1.38 1.70 0.98 0.88 1.06 0.71 1.62 1.59 1.25 1.17 1.15 0.94
17 1.72 2.44 5.13 1.12 1.43 3.85 1.30 1.59 2.00 0.80 0.85 1.39 1.19 1.14 1.54 0.75 0.69 1.05
18 3.01 3.02 2.5 2.08 1.83 1.37 1.54 1.89 1.14 1.01 1.26 0.59 1.70 1.36 0.97 0.89 0.79 0.63
19 2.46 1.74 1.34 1.11 0.86 0.67 1.34 1.49 0.88 0.63 0.68 0.43 1.15 1.26 0.90 0.57 0.66 0.48
x 2.62 3.30 3.16 1.34 1.50 1.59 1.44 1.57 1.20 0.8 0.88 0.69 1.41 1.37 1.21 0.86 0.89 0.75
s.e 0.57 1.21 1.29 0.24 0.33 0.61 0.15 0.13 0.17 0.09 0.11 0.18 0.14 0.13 0.08 0.12 0.13 0.08
Figure 11 is a plot of the differences between the closed chest and the opened chest with beta adrenergic blockade for RL and R as a function of P tp None of these values are significantly different from zero; however at a p of 5 cm. H20 there is a trend for the resistance in the opened chest with beta adrenergic blockade condition to be greater than the closed chest without beta adrenergic blockade.
Table 4 contains the individual data for the 6 dogs of Series III. Also included are the means and standard error of the means for each condition and each transpulmonary pressure.
Pressure Volume Curves
Figure 12 is a plot of lung volume as a percent
vital capacity as a function of transpulmonary (Ptp) in the closed chest, closed chest with beta blockade, opened chest, opened chest with beta blockade conditions and in the case of a pneumothorax. Vital capacity was defined as the change in lung volume during a passive deflation from a Ptp of 25 cm. H20 to 0 cm. H0 with the chest opened. Each point represents the mean for each group. There were no significant changes in lung volume at any transpulmonary presure between the various conditions.
Figure 12. The volume of the lung as a percent
of the vital capacity in the opened
chest is plotted as a function of
70- x Pneumothorox
>. Closed Chest with Beta Blockade
60 Open Chest with Beta Blockade
5 10 15 20 25
Transpulmonary Pressure (cm H 20
Transpulmonary pressure of 5, 10 and 15 cm. H2 0 corresponds to 45.3, 74.4 and 97.5 percent of vital capacity, respectively.
Lojcation of the Catheter
Table 5 contains the anatomical data obtained when the lungs were dissected after being dried at a transpulmonary pressure of Z0 cn. H 20 for 3 days. Given are the diameter, distance from the carina and generation (trachea = 0) of the airway in which the catheter was wedged. The diameter
ranged from 2.7 cm. to 4. 0 cm. with a mean of 3. 0 cm.
Table 6 contains the means and standard deviations for the pH, P C and P 02of the 19 dogs. The control values were obtained with the dogs spontaneously breathing through a cuffed tracheostomy tube prior to placement of the catheter. All other values represent the periods indicated in the table. There were no significant alterations throughout the course of the experiments in the blood gas status of the dogs.
Table 7. Diameter, generation and distance from
the carina of the airway containing the retrograde catheter for each experiment.
Dog Diameter Generation the carina
1 4.0 5 7.0
2 2.7 6 8.6
3 3.2 7 8.9
4 3.0 7 9.0
5 2.7 10 10. 0
6 2.9 9 8.7
7 3.1 7 7.7
8 3.0 9 8.9
9 3.1 9 8.8
10 2.8 8 9.0
11 Z.8 5 9.0
1z 3.1 6 6.0
13 3.6 5 7.0
14 3.3 7 8.3
15 Z.7 10 9.3
16 3.0 8 9.0
17 2.9 8 7.6
18 3.2 7 8.0
19 2. 9 9 10. 0
Table 8. Mean and standard deviation for pH, P COand P 0at the indicated
time in the experimental procedure. 2 2
PO PCO2 pH
Control 79.1 5.6 39.2 3.8 7.34 + .04
Catheter 78.9 + 6.7 39.4 + Z.4 7.34 + .0?
in Place 7 3. t _.4
Closed Chest 80.8 + 6.4 38.7 + 3.0 7.34 + .03
Closed Chest +
and Cke 77.6 + 8.2 41.8 + 3.6 7.33 + .03
and P blockade G
Pneumothorax 79.6 + 5.6 38.6 ? 2.9 7.35 + .01
Open Chest 78.5 + 6.7 38.8 Z.9 7.36 .02
Open Chest and 79.0 + 6.4 42.0 + 3.2 7.32 .03
In order to minimize animal variability, measurements of resistance were made on a paired basis, that is, each dog served as its own control. The coefficient of variation (standard deviation/mean x 100) for RL for all dogs at a transpulmonary pressure of 5 cm H 2O was 41%. This is a smaller value than that found by other authors (36). The variability was, in part, due to the fact that the dogs were of a different size, age and breeding. Since RL is dependent on lung volume any error in the volume at which the measurements were made would increase the variability in R .* The coefficient of variation for R Pat a transpulmonary pressure of 5 cm. Hz20 was 59%, a value also smaller than that of previous investigators (36). The variability in RP can be explained on the same basis as the R variation. In addition, RPis dependent on the location of the retrograde catheter, which may account for some of the variability between dogs.
Measurements of RPand R Lwithin the same dog at a
given condition were reproducible. For example, the average coefficients of variation for R L and R p in the closed chest condition at a transpulmonary pressure of 5 cm. H 2 0 for the 13 dogs of series I were 18% and 20%, respectively. This variability may be related to an error in the lung volume at which repeated measurements were made since both RP and R L are lung volume dependent. The remaining variability is either due to measurement error and/or random changes in resistances between measurements.
Since the experiment was designed to test the effect of various conditions on the contribution of peripheral resistance to lower pulmonary resistance it was necessary to maintain as many factors as possible, other than those to be tested, constant between conditions. The level of anesthesia and muscle paralysis were maintained constant throughout the experiment and should not have influenced the resistance between conditions. Since the arterial pH, p C02 and P 0 2 were unchanged in the various conditions they should not be responsible for changes of resistance between conditions. The lungs were always inflated to a transpulmonary pressure of 25 cm. H 2 0 prior to oscillation which means there was a common lung volume history in all the conditions. The lack of change in the quasi-static pressure volume curves indicates that there were no changes in
overall elastic properties of the lungs between conditions and that the lung volume for the paired transpulmonary pressures was the same. This means that the resistances for the different conditions were measured at the same lung volume. This does not necessarily imply that the geometry of the airways was the same in all conditions. It would be possible for a change in the shape of the lung to have occurred or for the distribution of the volume between the various sized airways and the alveoli to have been altered between conditions In the opened chest condition the influence of the chest wall on the shape of the lung is lost as is the influence of gradients of pleural pressure. Either of these changes could result in a change of distribution of volume, which would not be detected by a static pressure volume curve.
As previously described, the measurement of RP and RL in all conditions was done by the same technique. If errors were made in the determinations of resistance there is no reason to suspect that such errors would not be common to all conditions. That is, a change in resistance between conditions should not be explainable by any systematic error in measurement.
The lack of an asymmetrical pressure flow loop was surprising since, as was previously shown, there should have been errors in the measured pressure drop due to the Bernoulli effect. Its absence suggests an inadequacy of the measurements or that the effect is masked by the technique of closing the loop during oscillation by electrically subtracting a proportion of volume.
Figure 13 is designed to describe the effect, in theory, of the Bernoulli principle. It is a plot of pressure vs flow with an assumed resistance of 3 cm. H 2 O/liter/second. The solid line is the ideal plot. The dashed line shows the shape of the pressure flow plot for the conditions described above with the inclusion of the Bernoulli effect when the pressure drop is measured between the alveolus and a tracheal tube 1 cm. in diameter.
Since the pressure drop was measured as the peak to peak amplitude from a recording of pressure against time the calculated resistance would be independent of any Bernoulli effect. This can be seen in Figure 13 by comparing the peak to peak pressures for the ideal and the asymmetrical pressure flow plots. Moreover, since the linear velocity at the point of the bronchial pressure measurement is less than at the trachea any influence of the Bernoulli effect on R P would be smaller than that for
Figure 13. Plot of a theoretical pressure flow curve with (dashed
line) and without (solid line) the predicted influence
of the Bernoulli effect.
PZ 6cm. HO
5 i expiration
4 3 2 11 2 3 4
inspi ration /
p 6 cm. H0
P= 6 cm.HH0
R L* Again there is no reason to suspect that the Bernoulli effect would be different in any of the different conditions and therefore it can not explain changes in resistance between conditions.
It is clear that the flow resistance measured during
the oscillatory technique includes tissue viscous resistance as well as resistance to air flow. Various authors (6,9,16,17) have measured tissue resistance and arrived at values as low as 4% and as high as 40% of the pulmonary resistance. Bachofen
( 2) has shown that tissue resistance increased with increased tidal volume when flow rate and functional residual capacity were held constant. It was also observed that doubling flow rate without changing tidal volume or end expiratory volume did not appreciably increase the non-elastic work performed on the lung, so tissue resistance had to have decreased (1 These results suggest that tissue resistance does not behave as a flow resistive element but as a non-ideal elastic element exhibiting pressure volume hysteresis.
Marshall and Dubois (16) have shown that the measured tissue resistance is reduced by panting compared to that observed at normal breathing frequencies. Wood (34) has calculated the tissue resistance for a healthy human assuming a flow of 1 liter/second, a compliance of 200 ml./cm.H z 0 and resistance of 2 cm. H 2 O/liter/second for several
frequencies. At a breathing rate of 10 breaths/minute with a tidal volume of 1000 ml. he predicted that 30% of the resistive work done on the lung could be accounted for by tissue resistance. At a frequency of 100 breaths/minute and a tidal volume of 100 ml. tissue resistance accounts for only 3% of the resistive work done on the lung. When the lung is oscillated at a frequency of 4 cycles per second with a volume of 40 ml. the contribution of tissue resistance to the resistive work becomes negligible. For these reasons the resistance measured during the present experiments is considered to reflect predominantly airway resistance and not tissue resistance.
Comparison of R Measured in vivo with
Predicted Values Based on Anatomic Data
Ross (26) has conducted an extensive study of the structure of the bronchial tree in the dog. Using a plastic cast he was able to measure airways down to 0.5 mm. in diameter and up to 8 generations in number. He found that bronchial pathway lengths from the carina to terminal bronchioles, which he defined as airways 1 mm. in diameter, were quite variable and ranged from 2-14 cm. In the human lung the distribution of 2 mm. airways is from 3 to 30 cm. from the carina, a distribution similar to the dog for 1 mm.
airways. Horsefield and Cumming (8 ) found that the number of terminal bronchioles arising from a bronchus of a given cross-sectional area decreased the farther it arose from the carina. Ross (Z6) reported a similar finding in the dog lung. Irregular branching, the distribution of terminal bronchioles and the factors determining the number of terminal bronchioles arising from a bronchus of given cross-sectional area are similar to both the human and dog lung, suggesting that the dog lung may be an appropriate model for studying airway resistance.
Green (7 ) has used Weibel's (31) anatomical data for the human lung to compute the resistance in each generation assuming laminar flow. Figure 14 is a plot of the R P to RL ratio using these predicted resistances. This is an original plot and does not appear in Green's paper. It can be seen that airways 2.8 mm. and less in diameter contribute more than 50% to lower pulmonary resistance. This agrees well with the experimental results obtained in dog with the chest intact near functional residual capacity.
Ross (26) developed an equation to describe the pressure gradient across a unit length of any bronchus in terms of the relative pressure gradient across the same length of trachea for any flow rate. The differences in flow rate in the trachea and bronchus are taken into account in this
Figure 14. The resistance of the bronchi of a given generation
and those peripheral to that generation is plotted as
a percent of the total bronchial resistance. The
abscissa is a linear scale of generation number. The
corresponding diameters are also shown.
18 8.3 2.8 1.5 .82 .54 .45 .41
loo I I I I I i
0 5 10 15 20 23
% (AP/L) trachea = .75 3300 .5 A (D+Z)
A = cross-sectional area D = distance from carina
Using this equation and the anatomical data of Ross (30) the pressure drop per generation for 3 randomly selected bronchial pathways was calculated. The mean for these three pathways results is plotted as a R p /R L ratio where RL is the resistance from trachea to the eighth generation in Figure 15. Since data for the dog were available for only the first eight generations the relationship for the human lung is also plotted for only the first eight generations. in the dog 40% of the resistance in the first eight generations is predicted to be beyond the sixth generation. In the human lung 30% of the resistance of the first eight generations is predicted to be beyond the sixth generation. It is clear that there are 15 generations beyond this level, which are not included in this theoretical treatment. The added resistance of these generations would result in an increased lower pulmonary resistance and can only increase the percentage of RL contributed by the peripheral airways.
Figure 15. The resistance of a given generation and
those peripheral to it is plotted as a percent of the total resistance of the
first eight generations. The open circles
(0) are based on Green's predictions for
the human (7) and the (X) are based on
Ross' data (26) for the dog.
90 0 x
S 40 x dog
. ..... A ......
1 2 3 4 5 .6 7 8