- Email: [email protected]
Alveolar liquid pressures in nonedematous and kerosene-washed rabbit lungs by micropuncture
Respiration Physiology, 78 (1989) 281-296
281
Elsevier RESP 01601
Alveolar liquid pressures in nonedematous and kerosene-washed rabbit lungs by micropuncture Sundaresh Ganesan, Stephen J. La~-Fook arid Samuel SchOrch Biomedical Engineering Center. University of Kentucky. KY40506, U.S.A. and Department of Medical Physiology, University of Calgary, Alberta T2N4NI, Canada. (Accepted for publication 19 August 1989) Abstract. We studied the relationship between alveolar interfacial pressure and lung volume in kerosenefilled lungs, nonedematous air-filled lungs on lung deflation and inflation, and air-filled lungs aRer washing with kerosene or the Dow Coming oil, 0.65 cs dimethyl siloxane (DC200). We used the micropipetservonulling technique to measure alveolar liquid pressure (Pliq) in the alveolar liquid layer of isolated rabbit lungs at different airway pressures (Pair). It was not possible to measure pressure in kerosene or in DC200 by micropuncture because of its low electrical conductivity. We used the Laplace law for a spherical membrane to estimate alveolar surface tension (T). In the kerosene-filled lung, the pressure drop (Ap = Pliq- Pair) across the alveolar surfactant-kerosene interface was 1.1 cm H20 at TLC and decreased to 0.5 cm H20 at 71% TLC. These values corresponded to T values of 2.2 and 0.9 dyne/cm at TLC and 71~ TLC, which were in agreement with in vitro measurements using the captive bubble technique, in the air-filled lung on inflation,/xp values were 12.7 and 15.7 cm H20 at 48~e and 76~o TLC. Corresponding T values were 14 and 21 dyne/cm. Thus, alveolar surface tension on lung inflation is surface area dependent. In the kerosene-washed and DC200-washed lungs, Ap values were 16 and 14.5 cm H20 at TLC and decreased to 9 and 8 cm H20 at 50-56~ TLC. These values indicated a reduction of 40-60% in alveolar surface tension with lung deflation from TLC to 50% TLC. The results indicate that alveolar surface tension in both kerosene-filled and kerosene-washed air-filled lungs is surface area dependent. This is due to a surfactant-kerosene interface in the kerosene-filled lung and a surfactant-kerosene-air interface in the kerosene-washed lung.
Alveolar interracial pressure; Alveolar surface tension; Captive bubble technique; Lung inflation; Micropipet-servonulling technique
Surface forces acting on the alveolar wall may influence blood flow (Bruderman et al., 1964), the stability of alveoli (Clements et al., 1966) and lung fluid balance (Pattie, 1955; Clements, 1961; Albert et ai., 1979). Theoretically, surface forces at the curved aJrsurfactant interface would result in a pressure in the liquid layer lining alveolar surfaces that is below alveolar gas pressure. This has been confirmed by direct measurements
Correspondence address: Stephen J. Lai-Fook, Biomedical Engineering, Wenner-Gren Research Laboratory, University of Kentucky, Lexington, Kentucky 40506-0070, U.S.A. 0034-5687/89/$03.50 © 1989 Elsevier Science Pubfishers B.V. (Biomedical Division)
282
S. GANESAN et aL
of alveolar liquid pressure using the micropuncture technique in isolated edematous lungs (Lai-Fook and Beck, 1982). These measurements are generally consistent with direct and indirect estimates of surface tension in lung with normal surfactant and in lung in which surfactant is depleted (Clements et al., 1966; Faridy et al., 1966; Beck and Lal-Fook, 1983). Numerous studies using a variety of techniques have demonstrated that alveolar surface tension is high at high lung volumes but falls to low values on lung deflation to functional residual capacity (Clements, 1957; Bachofen etaL, 1970; Smith and Stamenovic, 1986; Lai-Fook and Beck, 1982). However, the alveolar surface tension-lung volume behavior during lung inflation is still controversial. Recent measurements of the pressure-volume behavior of air-filled lungs which were washed with oils of various surface tensions showed that alveolar surface tension increased to the equilibrium value of 28 dyne/cm on lung inflation to 50~ TLC and remained constant with further lung inflation to TLC (Smith and Stamenovic, 1986). This behavior conflicts with the much smaller values measured by the microdroplet technique (Bachofen et aL, 1987) and values estimated from alveolar liquid pressure measured in edematous lungs by micropuncture (Lai-Fook and Beck, 1982). A crucial assumption in the method of Smith and Stamenovic (1986) is that the oil-washed air-filled lung is depleted of surfactant and therefore has a constant surface tension. On the other hand, previous studies of alveolar liquid pressure used edematous lungs, and it may be questioned whether measurements made in edematous lungs can be extrapolated to nonedematous lungs. Also, alveolar sure'ace tension estimated from measurements of alveolar liquid pressure might not represen; the pressure in the alveolar liquid layer because the micropipet tip might be located in interstitial tissue. Accordingly, we addressed the following issues in this study. On the question as to the location of the pipet tip during a measurement of alveolar liquid pressure, we measured alveolar li,~uid pressure in kerosene-filled lungs by micropuncture. The alveolar liquid pressure measured was not the pressure in the kerosene because the micropipet-servonuiling technique cannot measure pressure in otis of low electrical conductivity. Alveolar liquid pressure was slightly less than the pressure in the kerosene as measured by a strain gauge and decreased with lung deflation. The small interracial pressures measured in the kerosene-filled lung were consistent with the surfactantkerosene interracial tensions measured in vitro using the captive bubble technique (Scht~rch etaL, 1985). This indicated that alveolar liquid pressure represented the pressure in the alveolar liquid layer and not in the interstitial tissue. In this communication we used the term 'alveolar liquid layer' to "eferto the fiquid compartment between the air-liquid interface and the alveolar epithelium. To estimate alveolar surface tension on lung inflation and in oil-washed lungs, we measured alveolar liquid pressure in nonedematous air=filled lungs on deflation and inflation, and in air-filled lungs after they were washed with oil (kerosene and 0.65 cs dimethyl siloxane [Dew Coming 200 oil]). This substantiated that an alveolar liquid layer was still present in oil-washed air-filled lungs. Values of alveolar surface tension estimated in nonedematous air-filled lungs and in oil-washed air-filled lungs showed a marked dependence on lung volume.
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
283
Methods
New Zealand white rabbits, weighing 2.7-3.6 kg, were anesthetized using sodium pentobarbital (30 mg/kg) injected into an ear vein and killed by an overdose of the anesthetic. The trachea of each rabbit was cannulated after tracheostomy. The entire thoracic cage was isolated from the rest of the rabbit with the heart and lungs intact. The diaphragm was removed to expose the pleural surface of the lung to atmospheric pressure. The lungs were degassed in a vacuum chamber. To minimize pleural leaks, the lung surface was not touched at any point throughout the experiment. Alveolar liquid pressure (Pliq) was measured at different transpulmonary pressures (Ptp, airway pressure relative to pleural pressure, which was atmospheric) under the following conditions: (1)nonedematous ak-f'dled lung, (2)kerosene-fdled lung, (3)kerosene-washed air-fdled lung, (4)air-f'dled lung after washing with Dow Coming 0.65 cs dimethyl siloxane (Dow Coming 200 oil). All measurements were done at 22 + 2 °C. Direct surface tension measurements in isolated lungs showed little change between 22 and 37 °C (SchOrch et al., 1985).
Air-filled lungs (control). The degassed lung was mounted on a stand with the diaphragmatic pleural surface superior. This orientation allowed easy manipulation of the micropipet over the exposed pleural surface for the measurement of alveolar liquid pressure by micropunture. The degassed lung was inflated with air to a Ptp of 25 cm H20 until it appeared to be uniformly inflated. The trachea was occluded and the pressure was allowed to equilibrate at the constant volume. Achievement of an equilibrated Ptp above ~21 cm H20 indicated that the lung was leak free. The volume at full capacity was denoted TLC. Alveolar liquid pressure (Pliq) was measured by the micropipet-servonulling technique using the procedure as described below. The reference zero for the Pliq measurements was obtained by immersing the pipet tip in a small pool of normal saline sprayed on the surface of the lung. Measurements of Pliq were obtained by micropuncture at the equih'brated pressure at TLC. The lung was deflated stepwise to Ptp values of ~ 15, 10 and 5 cm H20 and Pliq was remeasured at each test pressure with the lung volume constant. Ptp was measured by a water manometer. The increments in air volume withdrawn on deflation between each test Ptp were noted. After the Pliq measurements the lung was deflated to 0 cm H20 ~tp and reinflated to volumes similar to those arrived at on deflation. Pliq was remeasured at the inflation volumes. Because many ofthese lungs developed pleural leaks after several punctures, it was not always possible to obtain measurements of Pliq for both deflation and inflation lung volumes in every lung. When pleural leaks occurred, measurements of Pliq were discontinued because a constant lung volume and transpulmonary pressure could not be maintained. Therefore, in another set of lungs, we measured Pliq on inflation volumes alone. At the end of the Pliq measurements the lung was deflated to 0 cm H20 Ptp, separated from the thorax and displaced in water to determine its residual volume. The lung was weighed and dried in an oven at 70 °C to a constant weight to obtain its wet-to-dry weight ratio.
284
s. GANESAN et al.
Kerosene-filled lung. The degassed lung was placed in a saline bath and inflated with kerosene to total lung capacity (TLC) via a kerosene reservoir held at 25 cm H20 above the saline bath level. Airway pressure was measured by a kerosene-f'dled transducer (Statham P23) and monitored continuously on a strip chart recorder (Linear, Model 500). The zero reference for airway pressure was atmospheric pressure at the level of the saline bath. After the lung appeared to be uniformly expanded, the reservoir was disconnected and the airway pressure was allowed to equilibrate. The volume at the equilibrated pressure (8-12 cm H20) was denoted TLC. The pressure-volume behavior was measured by stepwise removal of volume increments and noting the corresponding airway pressures. This procedure was repeated twice using fresh kerosene. Pliq was measured by micropuncture at TLC and at several deflation Ptp values. The site of micropuncture was at the saline bath level which provided the zero reference for Pliq measurements. No measurements of Pfiq were obtained below a Ptp of 3 cm H20 (~65% TLC) because it was very difficult to puncture the very compliant pleural surface. At the end of the Pliq measurements, the lung was deflated to 0 cm H20 Ptp, separated from the thorax, weighed and displaced in water to detelmine its residual kerosene volume. The lung was air dried to a constant weight to determine its dry weight.
Kerosene-washed and DC200-washed air-filled lungs. The degassed lung was inflated with air to a Ptp of 25 cm H20, the trachea was occluded and the pressure allowed to equilibrate. The pressure-volume behavior was measured by stepwise removal ofvolume to deflation Ptp values of 15, 10, 5, 3 and 0 cm H20. After degassing the lung, the airspaces were washed with kerosene by inflating the lung with kerosene and deflating to 0 cm H20 Ptp. This procedure was repeated three times. Fresh kerosene was used for each inflation-deflation maneuver. The collapsed lung was then degassed and mounted on a stand with the diaphragmatic pleurai surface superior. The lungs were then inflated with air to a Ptp of 25 cm H20. A pressure-volume behavior of the keJ'osene-washed lung was measured on deflation as described previously for the control lung. Pliq was measured on deflation by micropuncture at TLC and at several equilibrated Ptp values which ranged from TLC to 5 cm H20 ( ~ 50~o TLC). Pliq was not measured on lung inflation because the pressure-volume behavior showed only a small hysteresis (Smith and Starnenovic, 1986). At the end of Pliq measurements the residual volume, wet weight and dry weight of the lungs were obtained as described earlier, In a separate set of lungs, the entire procedure outlined above for the kerosenewashed air-filled lungs was repeated on air-idled lungs after washing with Dew Coming 200 fluid. This fluid has a viscosity of 0.65 cs and surface tension of 16 dyne/cm.
Micropipet-servonuiling method.
The technique used to measure alveolar liquid pressures was similar to that previously reported (Lai-Fook and Beck, 1982). The servonulling pressure measuring system maintains the electrical conductivity across the tip of a glass micropipet constant. The servonulling system accomplishes this by imposing on the liquid within the micropipet a pressure equal to the pressure which is being measured. The pressure within the micropipet was measured using a strain gauge transducer (Statham P23).
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
285
Glass pipets were pulled and bevelled to tip diameters of 2-6/~m on a rotating diamond stone (Micropipette beveller; Sutter Instruments Model BV-10). The pipet was Idled with a 1 M saline solution, mounted on a micromanipulator and connected to a servonulling pressure measurement system (Instruments for Physiology and Medicine, Model 4A). A reference zero was estabfished by immersing the pipet tip in a saline pool above the lung surface. This area ofthe lung was illuminated using a fiberoptic light source (Volpi, Model Intralux 150 H) and viewed through a stereomicroscope (Zeiss, 25X). The entire experiment was conducted with the experimental apparatus placed on a vibration-free surface (Technical Manufacturing Corporation, Vibration Isolation Table, Model Micro-g series 63-500). The pipet was advanced to touch and puncture the pleura] membrane to depths within the lung parenchyma ranging from 0.1 to I mm. A stable alveolar liquid pressure was usually obtained on slow withdrawal of the pipet. This pressure was monitored on a strip-chart recorder (Linear Model 500). Readings were acceptable provided they satisfied the following conditions. 1. The pressure tracing was stable for at least 1-2 rain. 2. The pressure was reproducible, within 2.0 cm H20, on consecutive punctures, not necessarily at the same site. 3. The pressure signal was independent of small changes of the gain setting of the servo-nulling system. This indicated that the tip was immersed in a free liquid having an electrical conductivity similar to normal saline. Thus the pressure measured was in the alveolar liquid layer and not in the kerosene. By independent calibration, we determined that it was not possible to measure pressure in liquids such as kerosene and DC200 fluid with negligible electrical conductivity. 4. The reference zero measured immediately before the puncture was reproduced (within 0.5 cm H20) on removing the pipet from the lung. Preliminarymeasurements showed that the micropipet pressure measured in a calibration chamber was stable within +_0.5 cm H20 of the chamber pressure during a period of 1-2 h. This indicated that diffusion of ions and water into and out of the pipe!: did not change the electrical conductivity across the micropipet tip during the measurements of alveolar liquid pressure. In preliminary studies, we attempted to measure alveolar liquid pressure in edematous lungs which were f'dled with kerosene and in edematous lungs which were air-Idled after washing with oil. The lungs were made edematous by tracheal instillation of 20 mi saline and inflation to TLC. Stable alveolar liquid pressures were not attained in these edematous lungs.
Measurement of interracial tension in vitro. The measurement of the interracial tension of the kerosene-surfactant interface was based on the previous demonstration that the surfactant film is continuous under an oil drop if this drop is placed from the air phase onto a surfactant film (SchOrch et al., 1985). The present experimental approach was similar except that a captive air bubble (SchOrch etal., 1985) replaced the Wilhelmy-Langmuirtype surface balance of the previous demonstration. In the present investigation we chose the captive air bubble because of the greater stability of its air-saline surfactant films. An air bubble was formed below a layer of a 1~o agar gel that was molded into the lucite cover of a chamber with optically flat glass walls (fig. l). The
286
s. GANESAN et al.
A
B Fig. 1. Configurationof surfactant-keroseneinterface in the captivebubble technique. (A) Surfacetension of the large.captive bubble = 25 dyne/era; interracial tension of the kerosene drop-surfactant interface -2 dyne/cm. The diameter of the captive bubble is approximately I cm. (B) The configuration at a surface tension of 10 and an interracial tension of I dyne/cm.
chamber was filled with saline (0.9~o NaCI). A lung surfactant film was formed at the bubble air-water interface by spreading droplets of an aqueous suspension of cow lung surfactant extract (Yu et al., 1983) via a microsyringe. The needle of the syringe was advanced through holes, filled with agar gel, in the lucite cover. After achieving a fdm surface tension between 25 and 30 dyne/era at the bubble surface, a kerosene droplet was placed from the air phase onto the surfactant fdm. This kerosene drop spread to a relatively fiat lens at film surface tensions above 30 dyne/era. When the film surface tension was reduced, the contact angle of the kerosene drop increased and the drop changed from that of a flat lens to that of an inverted sessile drop. Figure I demonstrates the saline-kerosene-air bubble configuration at an air-saline film surface tension of 25 dyne/era (A) and 10 dyne/era (B). The corresponding kerosene-saline surfactant tensions were 1.8 and 1.0 dyne/era. The surface tension of the large, captive air bubble was determined by using the approximation described by Malcolm and Elliott (1980) valid for bubbles having 180 ° contact angles and whose ratio of width to height is within a given limit (SchQrch, 1982). The interracial tension ofthe surfactant-kerosene interface was determined by digitizing the photographs of the kerosene drop and by using the computer programs based on the formulas of Rotenberg et al. (1983). The simple formula of Malcolm and Elliott
287
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
(1980) could not be used for these drops since the contact angles of the kerosene were different from 180 °. The contact angles ofthe kerosene drops changed according to the air bubble surface tension from less than 20 ° to more than 140 °. We always placed a kerosene drop onto the surfactant film of the air bubble kept at a surface tension of 30 dyne/era. The bubble surface tension was then decreased to 25 dyne/cm by withdrawing air through a 26 gauge needle that reached the interior of the bubble through the ceiling of the container. Because of film continuity, the kerosene drop changed in shape accorOing to its decreasing surface tension. Additional pairs of surface and interracial tension values were generated by withdrawing air from the bubble. We also noticed that by inflating the bubble the shape of the bubble and the drop, and their corresponding interfacial tensions, were reversible.
Results The pressure-volume (PV) curves ofthe control air-fdled lung on deflation and inflation, the kerosene-filled lung, the kerosene-washed air-filled lung and the DC200-washed air-filled lungs on deflation are summarized in fig. 2. The PV curves of the control air-filled lung on inflation and the two oil-filled lungs were shifted to the right of the PV
100"
i
80'
i ~
60' 40'
.~/,e."°'°~ i
i
/
C --
/
2o'
...... '~ .....
/ ...... o ..... ,
,~ .n
O
Kerosene filled lung
oc oowa..ed,..g
o o
Kerosene washed lung
•
Control air-filled (deflation)
--
Control air-filled (inflation)
~ 0
10 Transpulmonary
20 pressure
30 (cmH20)
Fig. 2. Pressure-volume curves of the kerosene-filled lung, the control air-fiUed lung on inflation and deflation, the kerosene-washed and DC200 oil-washed lung on deflation. Points represent mean values. Bars represent standard deviation. Some bars have been omitted for clarity. Note the rightward shift in the curves of the control air-filled lung on inflation and the two oil-washed lungs, compared to the control air-filled lung on deflation.
S. GANESAN et aL
288
TABLE I
Measurements of alveolar liquid pressure in kerosene-filled lungs on lung deflation Ptp (cm H20)
N
7.9 + 2.1" 5.5 + 0.60 3.8 + 1.1
3 4 3
Pliq (cm H20)
Volume
T
( ~ TLC)
(dyne/cm)
6.8 + 0.74 4.9 + 1.0 3.3 + 0.87
100 90 71
2.2
1.1 0.9
Ptp, transpuimonary pressure = alveolar pressure relative to pleural pressure. Pliq, alveolar liquid pressure relative to pleural pressure. T, alveolar surface tension calculated using Laplace's law; *, mean ± SD. N,
number of measurements.
curve of the control air-fiUedlung on deflation, indicating an increase in alveolar surface tension. The PV curve of the kerosene-fdled lung was shifted to the left of the PV curve of the control air-fdled lung on deflation and showed a behavior similar to that of a saline-filled lung (Bachofen et al., 1970). TLC of the lungs before washing with kerosene and DC200 averaged 94.8 + 9.5 (SD) ml (N = 18). Table 1 summarizes values of alveolar liquid pressure relative to pleural pressure (Pliq) measured in alveolar liquid layer in the kerosene-f'dled lung at different transpulmonary pressures and lung volumes. Note that Pliq was on average 1.1 cm H 2 0 below the airway pressure (that is, Ptp) at 100~o TLC and decreased to 0.5 cm H20 below airway pressure on lung deflation to 7 1 ~ TLC. These values for the pressure difference across the kerosene-surfactant interface at 100 and 71 ~ TLC are consistent with values of interracial tension of 2.1 and 0.9 dyne/cm, respectively. In these calculations we used Laplace's law for a spherical interface which relates pressure difference (&P) across the interface to the surface tension T and radius of curvature (R): Ap -- 2T/R. We assumed that R varies as the cube root of the lung volume. We used a value of It of 40 tam (Tenney and Remmers, 1963). TABLE 2
Surfactant-air and surfactant-kerosene interfacial tensions measured at different degrees of compression by the captive bubble technique. The corresponding deflation lung volumes are shown Lung volume
Surface tension, dyne/era
(% TLC)
Surfactant-air interface
Surfactant-kerosene interface
30 25 10 5 1
2.0 + !.8 + 1,0 + 0.5 + 0.1 +
*Mean + SE (N = 4).
0.04* 0.04 0.04 0.05 0.05
100 95 75 60 42
289
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS TABLE 3
Measurements of alveolar liquid pressure in nonedematous air-filled lungs (control) on lung deflation and inflation Ptp (cm H20)
N
Pliq (cm H20)
Volume (% TLC)
T
(dyne/cm)
Deflation 22.5 + 2.9* 15.6 + 0.5 10.3 + 0.7 5.3 + 0.2 3.2 + 0.2
5 5 9 9 5
2.9 + 1.1 0.3 + 3.0 0.5 +_ 2.1 0.1 + 1.4 1.0 + 1.1
100 90 85 70 49
28.0 21.1 13.3 6.7 2.4
Inflation 15.0 + 1.7 13.4 + 0.7
5 5
11.3 + 0.8
4
-0,7 + 2.6 -0.9 + 3.7 - !.4 + 0.9
76 64 48
21.1 17.6 14.2
Ptp, transpulmonary pressure = alveolar pressure relative to pleural pressure. Pliq, alveolar liquid pressure relative to pleural pressure. T, alveolar surface tension calculated using Laplace's law; *, mean + SD. N, number of measurements.
TABLE 4 Measurements of alveolar liquid pressure in kerosene-washed and DC200-washed air-filled lungs on lung deflation Ptp (cm H20) Kerosene-washed 23.0 + 1.9"
N
Pliq (cm H20)
5
6.0 + 1.7
15.6 + 0.7
7
1.1 + !.4
12.1 + 0.8 8.5 + I.I
6
4
- 0 . 2 + 0.8 -0.5 + 1.8
DC200-washed 22.5 + 1.3 14.7 + 0.9 10.4 _+0.4 8.3 + 0.5
5 5 6 6
8.0 2.6 1.5 0.3
+ 2.1 + !.9 _. 1.4 + 0.8
Volume (% TLC)
T (dyne/cm)
100 90 73 57
28.0 23.0 18,2 12.2
100 94 77 53
16.0 13.1 9.0 7.2
Ptp, transpulmonary pressure = alveolar pressure relative to pleurai pressure. Pliq, alveolar liquid pressure relative to pleural pressure. T, alveolar surface tension calculated using Laplace's law; *, mean _. SD. N is number of measurements. DC200, Dew Coming 0.65-CS dimethyi siloxane.
290
s. GANESAN et al.
Table 2 shows the values of the surfactant-air and kerosene-surfactant interfacial tensions measured using the captive bubble method. Surfactant-air interracial tensions were compared to the alveolar surface tension-lung volume relationship measured in isolated perfused rabbit lungs by an improved microdroplet method (Bachofen, 1987) and the corresponding values of lung volume are shown in table 2. This comparison indicates that the surfactant-kerosen¢ interface tension is 2, 1.0 and 0.1 dyne/cm at 100 % TLC, 75 % TLC and 42 % TLC, respectively. Note the close agreement between the in vitro measurements of surfactant-kerosen¢ interracial tension and the in situ estimates based on the surfactant-kerosene interracial pressure measured in the alveolar liquid layer in kerosene-filled lungs (table 1). Table 3 summarizes values of Pliq measured in the control air-f"died lung on lung deflation and inflation. Table 4 summarizes values of Pliq measured in the kerosenewashed air-filled lung and DC200 washed air-filled lung. The difference between Pliq and alveolar air pressure (Palv = Ptp) is the interracial pressure across the air-liquid interface in these lungs and is plotted v s Ptp in fig. 3 and v s lung volume in fig. 4. The values of alveolar surface.tension calculated using values of Pliq-Palv in Laplace's law for the three sets of lungs are shown in tables 3 and 4 and plotted against lung volume in fig. 5. For these calculations we assumed values of alveolar surface tension of 28, 28 and 16 dyne/cm at TLC in the control air-filled lungs, the kerosenewashed and DC200-washed air-filled lung, respectively. These values were used previously by Smith and Stamenovic (1986). Figure 5 shows that surface tension in the air-filled lung decreased from a value of 28 dyne/cm at 100% TLC to a low value of ~ 2 dyne/cm at 50% TLC on lung deflation then increased to a value of 14 dyne/era on lung inflation from residual volume (0 cm
i
~
controldelletion w controlinflation --o- DC200washed --o--, kerosenewashed
e~ ~ ~ : ~
• E
o
1'o
2'o
Transpulmonary pressure (emH20) Fig. 3. Surfactant-air interfacial pressure vs transpulmonary pressure measured in control air-filled lungs on deflation (Q 0 ) and inflation ( i I ) , in kerosene-washed air-filled lungs ( .................) and DC200-washed lungs ( . . . . . . . . . ). The interfacial pressure was the difference between the alveolar liquid pressure and alveolar air pressure.
291
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
• ; --.o--.
D.
~---o---
controldeflalion controliNlalion 0C 200washed
kero6enew a s h e ~
0 .I-
E U
-10
_> |
o" I o.
40
60
80
100
Volume (% total lung capacity)
Fig. 4. Surfactant-air interfacial pressure versus lung volume (% TLC) measured in control air-filled lungs on deflation ( 0 O) and inflation (B---------m), in kerosene-washed air-filled lungs ( .................) and DC200-washed lungs ( . . . . . . . . . ). The interfacial pressure was the difference between the alveolar liquid pressure and alveolar air pressure.
100
,o
/aSS° 1
I
iS
//.
-
W
/
=) :s 0
/
; 6
o" /
iii
comml(deflationl control(Inftation)
---o--. DC200washed ---.¢.--- kerosenewashed
40
0
, 10
-20
"
30
Surlaca tension (dynes/cm)
Fig. 5. Lung volume (% TLC) v s alveolar surface tension for control air-filled lungs on deflation (Q Q) and on inflation (11 m), for kerosene-washed air-filled lungs ( .................) and DC200-washed lungs ( . . . . . . . . . ). Note the dependence on lung volume for all cases.
292
s. GANESANet al.
H20 Ptp) to 48% TLC. These values are in line with the results obtained by the microdroplet method (Bachofen et al., 1987). The results for lung deflation are consis-
tent with previous values estimated from alveolar liquid pressure measured in edematous rabbit lungs (Lai-Fook and Kaplowitz, 1984). Thus, a small amount of alveolar edema had only a small effect on alveolar liquid pressure. We shall return to this point in the discussion section. Note also that alveolar surface tension in the kerosene-washed lung and DC200washed lung did not remain constant with decreasing lung volumes on lung deflation. In the kerosene-washed and DC200-washed lungs, surface tension decreased from 28 and 16 dyne/cm at 100~/o TLC to 13 and 7 dyne/cm at ~ 50~o TLC, respectively. This suggests that the surfactant-kerosene-air and surfactant-DC200-ak interfaces exhibit a surface tension behavior that is surface area-dependent. This is consistent with the fact that we measured a pressure in the alveolar liquid layer of the kerosene-washed and DC200-washed lungs. Thus, washing the lung with kerosene and DC200 did not completely remove the surfactant layer in these lungs. Also, the decrease in T predicted for the DC200-washed lung was smaller than that for the kerosene-washed lung, probably due to the greater spreading of the DC200 on the alveolar surface in the presence of surfactant.
Discussion Methods. We measured alveolar liquid pressure in isolated nonedematous lungs by micropuncture and estimated alveolar surface tension using the Laplace law for a spherical membrane. There are several assumptions inherent in these estimates. One assumption is that alveolar liquid pressure was measured in the alveolar liquid layer of an alveolus and not in the interstitium in the alveolar wall or in the interstitium surrounding an extra-alveolar vessel. We believe that we were measuring pressure in the alveolar liquid layer because in the kerosene-fdled lung the alveolar liquid pressure was just below the alveolar pressure in the kerosene, and the differences between the alveolar pressure and the alveolar liquid pressure were consistent with values of the kerosenesurfactant interracial tension measured in vitro. Furthermore, previous studies show that interstitial pressures were above the pleural pressure (Fike et al., 1988). Another assumption is that alveolar liquid pressure was uniform throughout the lung. This is clearly an approximation because consecutive alveolar liquid pressures measured at sites 1-5 mm apart in some instances differed by as much as 1.5 cm H20. Since these lungs were not edematous, equilibrium in liquid pressure between adjacent alveoli might be very slow because interstitial resistance might be high. It is also possible that lung distortion caused by the pipet might have been responsible for these differences in pressure. We assumed that alveolar radius varied as the cube root of the lung volume (Klingele and Staub, 1970) on both lung deflation and inflation. Previous results show a slower decrease in alveolar radius with lung volume (Gil et al., 1979), and a marked hysteresis in the behavior of alveolar surface area and airspace curvature with lung
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
293
volume (Bachofen et al., 1987). However, our estimates of alveolar surface tension would be in error by 21 % at 50% TLC if we were to assume that R remained constant with lung deflation. Thus, our estimates of alveolar surface tension are more dependent on the interfacial pressure than on the variation of alveolar radius with lung volume. To calculate surface tension as a function oflung volume, we assumed the eqUilibrium values of alveolar surface tension of 28 dyne/cm for nonedematous air-filled lung and the kerosene-washed lung and 16 dyne/cm for the DC200-washed lung, at TLC (Smith and Stamenovic, 1986). For the kerosene-filled lung we assumed an alveolar radius of 40 J.lm at TLC (Tenney and Remmers, 1963). It is evident that the actual magnitudes of surface tension estimated were directly dependent on the above assumptions. However, these assumptions do not invalidate our conclusions concerning the reduction in surface tension with a decrease in lung volume.
Comparison with other results.
We estimated alveolar surface tension to decrease from 28 dyne/cm at TLC to 2 dyne/cm at 50% TLC on lung deflation of the nonedematous air-filled lung. These estimates of alveolar surface tension on lung deflation are consistent with the direct results using the microdroplet method (Schtlrch et ai., 1985), with studies using the pressure-volume curve (Bachofen et ai., 1970), and comparisons among PV curves of the air-filled lung and the air-filled lung after washing with oils of varying surface tensions (Smith and Stamenovic, 1986). One of the objectives of this study was to determine the relationship between alveolar surface tension and lung inflation. Our results show that alveolar surface tension increased from 13 dyne/cm at 50% TLC to 21 dyne/cm at 85% TLC on lung inflation from residual volume (0 cm H 2 0 Ptp). The micro droplet method showed that alveolar surface tension increased from 8 dyne/cm at 50% lung volume to 25 dyne/cm at 90% TLC on lung inflation (Bachofen et al., 1987). The latter result agrees with values estimated from alveolar liquid pressure measured in edematous dog lungs (Lai-Fook and Beck, 1982). The higher values estimated in this study are consistent with the greater static recoil measured on inflation (fig. 2), most likely due to the much longer equilibration times (10-15 min) at constant volume required for the micropuncture measurements. The above three results contrast with the constant value of 28 dyne/cm estimated by Smith and Stamenovic (1986). The constant surface tension was the direct result of the assumption that the oil-washed air-filled lung on lung deflation has a constant alveolar surface tension. This assumption is not supported by our results of alveolar liquid pressure in oil-washed lungs, which showed a dependence of alveolar surface tension with lung volume.
Implication of the results. The values of alveolar liquid pressure measured in nonedematous lungs in this study are in line with previous measurements in lungs with a small amount of edema (50% increase in extravascular lung water). This suggests that alveolar liquid pressure is not a strong function of interstitial edema. This may be due to the fact that interstitial edema occurs before significant alveolar edema fonnation (Heff, 1971).
294
S. GANESAN
el al.
The transmission of alveolar liquid pressure to the perivenular interstitium has recently been addressed in isolated perfused lungs (Fike et a/., 1988). There was a virtual equilibration in pressure between the alveolar liquid and perivenular interstitial spaces, which was consistent with a relatively short time constant for interstitial filling of small extra-alveolar vessels (Conhaim et a/., 1986). By contrast, previous studies indicated a difference between the perivenular interstitial pressure and the interstitial pressure in alveolar septal comers even in the presence of gross edema (Bhattacharya et aI., 1984). To the extent that alveolar liquid pressure is transmitted to the interstitium so that it is a reflection of interstitial pressure, our measurements of alveolar liquid pressure as a function of lung volume indicate that transvascular fluid flux would increase at high lung volumes. However, the exact relationship between alveolar liquid pressure and alveolar wall interstital pressure is still uncertain. Our results indicate that a small but finite interfacial pressure exists across the surfactant-kerosene interface in kerosene-filled lungs. consistent with previous in vitro measurements of surface tension (Smith and Stamenovic. 1986). This small interfacial tension would prevent kerosene from entering the interstitium and prevent the formation of interstitial fluid cuffs around large vessels as obtained in saline-inflated lungs (Conhaim et a/., 1986). In summary, alveolar liquid pressure measured in kerosene-filled and oil-washed air-filled lungs showed that an alveolar liquid layer was still present in these lungs. This indicates that a surfactant-kerosene interface in the kerosene-filled lung and a surfactant-oil-air interface in the oil-washed air-ftlled lung result in a defmite alveolar surface area dependence on surface tension. Interfacial pressures measured in kerosene-filled lungs were consistent with the surface tension of a surfactant-kerosene interface measured in in vitro studies. Parallel to the monotonically increasing lung static recoil on lung inflation, alveolar surface tension estimated from alveolar liquid pressure measurements shows a defmite increase with lung inflation, and not a constant surface tension as recently proposed (Smith and Stamenovic. 1986).
Acknowledgements. We thank Laura V. Brown for technical assistance. This research was supported by NIH grant HL 40362 and by the Medical Research Council (Canada).
References Albert, R.K., S. Lakshminarayan, 1. Hildebrandt, W. Kirk and J. Butler (1979). Increased surface tension favors pulmonary edema formation in anesthetised dogs' lungs. J. Appl. Physiol. 43: 730-738. Bachofen, H., J. Hildebrandt and M. Bachofen (1970). Pressure volume curves of air- and liquid-filled excised lungs-surface tension in situ. J. Appl. Physiol. 29: 422-43l. Bachofen, H., S. Scht1rch, M. Urbinelli and E.R. Weibel (1987). Relations among alveolar surface tensions, surface area, volume, and recoil pressure. J. Appl. Phystoi. 62: 1878-1887. Bhattacharya, J., M.A. Gropper and N.C. Staub (1984). Interstitial fluid gradient measured by micropuncture in excised dog lung. J. Appl. Physiol. 56: 271-277.
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
295
Beck, K. C. and S. J. Lai-Fook (1983). Alveolar liquid pressure in excised edematous dog lung with increased static recoil. J. Appl. Physiol. 55: 1277-1283. Bruderman, I., K. Somers, W.K. Hamilton, W.H. Tooley and J. Butler (1964). Effect of surface tension on circulation in the excised lungs of dogs. J. Appl. Physiol. 19: 707-712. Clements, J.A. Surface tension of lung extracts (1957). Proe. Soc. Exp. Bioi. Med. 95: 170-172. Clements, J.A. (1961). Pulmonary edema and permeability of alveolar membranes. Arch. Environ. Health 2: 280-283. Clements, J .A., R F. Hustead, R P. Johnson and I. Gribetz (1966). Pulmonary surface tension and alveolar stability. J. Appl. Physiol. 16: 444-450. Conhaim, RL., S.J. Lai-Fook and N.C. Staub (1986). Sequence of perivascular liquid accumulation in liquid-inflated dog lung lobes. J. Appl. Physiol. 60: 513-520. Faridy, E. E., S. Permutt and R L. Riley (1966). Effects of ventilation on surface forces in excised dogs']ungs. J. Appl. Physioi. 21: 1453-1462. Fike, e.D., S.J. Lai-Fook and RD. Bland (1988). Alveolar liquid pressure in newborn and adult rabbit lungs. J. Appl. Physiol. 64: 1629-1635. Gil, J., H. Bachofen, P. Gehr and E.R. Weibel (1979). Alveolar volume-surface area alterations in air and saline filled lungs fixed by vascular perfusion. J. Appl. Physiol. 47: 990-1001. Ileff, L.D. (1971). Extra-alveolar vessels and edema formation in excised lungs. Cire.Res. 28: 524-532. Klingele, T.G. and N.C. Staub (1970). Alveolar shape changes with volume in isolated, air filled lobes of cat lung. J. Appl. Physiol. 28: 411-414. Lai-Fook, S.J. and K.e. Beck (1982). Alveolar liquid pressure measured by micropipettes in isolated dog lung. J. Appl. Physiol. 53: 737-743. Lai-Fook, S.J. and M.R Kaplowitz (1984). Alveolar liquid pressure measured using micropipettes in isolated rabbit lungs. Respir. Physiol. 57: 61-72. Malcolm, J. D. and C. D. Elliott (1980). Interfacial tension from height and diameter of a single profile drop or captive bubble. Can. J. Chem. Eng. 58: 151-152. Pattle, RE. (1955). Properties, function and origin of the alveolar lining layer. Nature (London) 175: 1125-1126. Rotenberg, Y., L. Boruvka and A. W. Neumann (1983). Determination of surface tension and contact angle from the shape of asymmetric fluid interfaces. J. Colloid Interj. Sci. 93: 169-183. Schilrch, S. (1982). Surface tension at low lung volumes: dependence on time and alveolar size. Respir. Physioi. 48: 339-355. Schilrch, S., H. Bachofen and E.R. Weibel (1985). Alveolar surface tensions in excised rabbit lungs: effect of temperature. Respir. Physiol. 62: 31-45. Smith, J.e. and D. Stamenovic (1986). Surface forces in lungs. I. Alveolar surface tension-lung volume relationships. J. Appl. Physiol. 60: 1341-1350. Tenney, S.M. and J.E. Remmers (1963). Comparative quantitative morphology of the mammalian lung: diffusing area. Nature (London) 197: 54-56. Yu, S.-H., P.G.R. Harding, N. Smith and F. Possmayer (1983). Bovine chemical surfactant: chemical composition and physical properties. Lipids 18: 522-529.
281
Elsevier RESP 01601
Alveolar liquid pressures in nonedematous and kerosene-washed rabbit lungs by micropuncture Sundaresh Ganesan, Stephen J. La~-Fook arid Samuel SchOrch Biomedical Engineering Center. University of Kentucky. KY40506, U.S.A. and Department of Medical Physiology, University of Calgary, Alberta T2N4NI, Canada. (Accepted for publication 19 August 1989) Abstract. We studied the relationship between alveolar interfacial pressure and lung volume in kerosenefilled lungs, nonedematous air-filled lungs on lung deflation and inflation, and air-filled lungs aRer washing with kerosene or the Dow Coming oil, 0.65 cs dimethyl siloxane (DC200). We used the micropipetservonulling technique to measure alveolar liquid pressure (Pliq) in the alveolar liquid layer of isolated rabbit lungs at different airway pressures (Pair). It was not possible to measure pressure in kerosene or in DC200 by micropuncture because of its low electrical conductivity. We used the Laplace law for a spherical membrane to estimate alveolar surface tension (T). In the kerosene-filled lung, the pressure drop (Ap = Pliq- Pair) across the alveolar surfactant-kerosene interface was 1.1 cm H20 at TLC and decreased to 0.5 cm H20 at 71% TLC. These values corresponded to T values of 2.2 and 0.9 dyne/cm at TLC and 71~ TLC, which were in agreement with in vitro measurements using the captive bubble technique, in the air-filled lung on inflation,/xp values were 12.7 and 15.7 cm H20 at 48~e and 76~o TLC. Corresponding T values were 14 and 21 dyne/cm. Thus, alveolar surface tension on lung inflation is surface area dependent. In the kerosene-washed and DC200-washed lungs, Ap values were 16 and 14.5 cm H20 at TLC and decreased to 9 and 8 cm H20 at 50-56~ TLC. These values indicated a reduction of 40-60% in alveolar surface tension with lung deflation from TLC to 50% TLC. The results indicate that alveolar surface tension in both kerosene-filled and kerosene-washed air-filled lungs is surface area dependent. This is due to a surfactant-kerosene interface in the kerosene-filled lung and a surfactant-kerosene-air interface in the kerosene-washed lung.
Alveolar interracial pressure; Alveolar surface tension; Captive bubble technique; Lung inflation; Micropipet-servonulling technique
Surface forces acting on the alveolar wall may influence blood flow (Bruderman et al., 1964), the stability of alveoli (Clements et al., 1966) and lung fluid balance (Pattie, 1955; Clements, 1961; Albert et ai., 1979). Theoretically, surface forces at the curved aJrsurfactant interface would result in a pressure in the liquid layer lining alveolar surfaces that is below alveolar gas pressure. This has been confirmed by direct measurements
Correspondence address: Stephen J. Lai-Fook, Biomedical Engineering, Wenner-Gren Research Laboratory, University of Kentucky, Lexington, Kentucky 40506-0070, U.S.A. 0034-5687/89/$03.50 © 1989 Elsevier Science Pubfishers B.V. (Biomedical Division)
282
S. GANESAN et aL
of alveolar liquid pressure using the micropuncture technique in isolated edematous lungs (Lai-Fook and Beck, 1982). These measurements are generally consistent with direct and indirect estimates of surface tension in lung with normal surfactant and in lung in which surfactant is depleted (Clements et al., 1966; Faridy et al., 1966; Beck and Lal-Fook, 1983). Numerous studies using a variety of techniques have demonstrated that alveolar surface tension is high at high lung volumes but falls to low values on lung deflation to functional residual capacity (Clements, 1957; Bachofen etaL, 1970; Smith and Stamenovic, 1986; Lai-Fook and Beck, 1982). However, the alveolar surface tension-lung volume behavior during lung inflation is still controversial. Recent measurements of the pressure-volume behavior of air-filled lungs which were washed with oils of various surface tensions showed that alveolar surface tension increased to the equilibrium value of 28 dyne/cm on lung inflation to 50~ TLC and remained constant with further lung inflation to TLC (Smith and Stamenovic, 1986). This behavior conflicts with the much smaller values measured by the microdroplet technique (Bachofen et aL, 1987) and values estimated from alveolar liquid pressure measured in edematous lungs by micropuncture (Lai-Fook and Beck, 1982). A crucial assumption in the method of Smith and Stamenovic (1986) is that the oil-washed air-filled lung is depleted of surfactant and therefore has a constant surface tension. On the other hand, previous studies of alveolar liquid pressure used edematous lungs, and it may be questioned whether measurements made in edematous lungs can be extrapolated to nonedematous lungs. Also, alveolar sure'ace tension estimated from measurements of alveolar liquid pressure might not represen; the pressure in the alveolar liquid layer because the micropipet tip might be located in interstitial tissue. Accordingly, we addressed the following issues in this study. On the question as to the location of the pipet tip during a measurement of alveolar liquid pressure, we measured alveolar li,~uid pressure in kerosene-filled lungs by micropuncture. The alveolar liquid pressure measured was not the pressure in the kerosene because the micropipet-servonuiling technique cannot measure pressure in otis of low electrical conductivity. Alveolar liquid pressure was slightly less than the pressure in the kerosene as measured by a strain gauge and decreased with lung deflation. The small interracial pressures measured in the kerosene-filled lung were consistent with the surfactantkerosene interracial tensions measured in vitro using the captive bubble technique (Scht~rch etaL, 1985). This indicated that alveolar liquid pressure represented the pressure in the alveolar liquid layer and not in the interstitial tissue. In this communication we used the term 'alveolar liquid layer' to "eferto the fiquid compartment between the air-liquid interface and the alveolar epithelium. To estimate alveolar surface tension on lung inflation and in oil-washed lungs, we measured alveolar liquid pressure in nonedematous air=filled lungs on deflation and inflation, and in air-filled lungs after they were washed with oil (kerosene and 0.65 cs dimethyl siloxane [Dew Coming 200 oil]). This substantiated that an alveolar liquid layer was still present in oil-washed air-filled lungs. Values of alveolar surface tension estimated in nonedematous air-filled lungs and in oil-washed air-filled lungs showed a marked dependence on lung volume.
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
283
Methods
New Zealand white rabbits, weighing 2.7-3.6 kg, were anesthetized using sodium pentobarbital (30 mg/kg) injected into an ear vein and killed by an overdose of the anesthetic. The trachea of each rabbit was cannulated after tracheostomy. The entire thoracic cage was isolated from the rest of the rabbit with the heart and lungs intact. The diaphragm was removed to expose the pleural surface of the lung to atmospheric pressure. The lungs were degassed in a vacuum chamber. To minimize pleural leaks, the lung surface was not touched at any point throughout the experiment. Alveolar liquid pressure (Pliq) was measured at different transpulmonary pressures (Ptp, airway pressure relative to pleural pressure, which was atmospheric) under the following conditions: (1)nonedematous ak-f'dled lung, (2)kerosene-fdled lung, (3)kerosene-washed air-fdled lung, (4)air-f'dled lung after washing with Dow Coming 0.65 cs dimethyl siloxane (Dow Coming 200 oil). All measurements were done at 22 + 2 °C. Direct surface tension measurements in isolated lungs showed little change between 22 and 37 °C (SchOrch et al., 1985).
Air-filled lungs (control). The degassed lung was mounted on a stand with the diaphragmatic pleural surface superior. This orientation allowed easy manipulation of the micropipet over the exposed pleural surface for the measurement of alveolar liquid pressure by micropunture. The degassed lung was inflated with air to a Ptp of 25 cm H20 until it appeared to be uniformly inflated. The trachea was occluded and the pressure was allowed to equilibrate at the constant volume. Achievement of an equilibrated Ptp above ~21 cm H20 indicated that the lung was leak free. The volume at full capacity was denoted TLC. Alveolar liquid pressure (Pliq) was measured by the micropipet-servonulling technique using the procedure as described below. The reference zero for the Pliq measurements was obtained by immersing the pipet tip in a small pool of normal saline sprayed on the surface of the lung. Measurements of Pliq were obtained by micropuncture at the equih'brated pressure at TLC. The lung was deflated stepwise to Ptp values of ~ 15, 10 and 5 cm H20 and Pliq was remeasured at each test pressure with the lung volume constant. Ptp was measured by a water manometer. The increments in air volume withdrawn on deflation between each test Ptp were noted. After the Pliq measurements the lung was deflated to 0 cm H20 ~tp and reinflated to volumes similar to those arrived at on deflation. Pliq was remeasured at the inflation volumes. Because many ofthese lungs developed pleural leaks after several punctures, it was not always possible to obtain measurements of Pliq for both deflation and inflation lung volumes in every lung. When pleural leaks occurred, measurements of Pliq were discontinued because a constant lung volume and transpulmonary pressure could not be maintained. Therefore, in another set of lungs, we measured Pliq on inflation volumes alone. At the end of the Pliq measurements the lung was deflated to 0 cm H20 Ptp, separated from the thorax and displaced in water to determine its residual volume. The lung was weighed and dried in an oven at 70 °C to a constant weight to obtain its wet-to-dry weight ratio.
284
s. GANESAN et al.
Kerosene-filled lung. The degassed lung was placed in a saline bath and inflated with kerosene to total lung capacity (TLC) via a kerosene reservoir held at 25 cm H20 above the saline bath level. Airway pressure was measured by a kerosene-f'dled transducer (Statham P23) and monitored continuously on a strip chart recorder (Linear, Model 500). The zero reference for airway pressure was atmospheric pressure at the level of the saline bath. After the lung appeared to be uniformly expanded, the reservoir was disconnected and the airway pressure was allowed to equilibrate. The volume at the equilibrated pressure (8-12 cm H20) was denoted TLC. The pressure-volume behavior was measured by stepwise removal of volume increments and noting the corresponding airway pressures. This procedure was repeated twice using fresh kerosene. Pliq was measured by micropuncture at TLC and at several deflation Ptp values. The site of micropuncture was at the saline bath level which provided the zero reference for Pliq measurements. No measurements of Pfiq were obtained below a Ptp of 3 cm H20 (~65% TLC) because it was very difficult to puncture the very compliant pleural surface. At the end of the Pliq measurements, the lung was deflated to 0 cm H20 Ptp, separated from the thorax, weighed and displaced in water to detelmine its residual kerosene volume. The lung was air dried to a constant weight to determine its dry weight.
Kerosene-washed and DC200-washed air-filled lungs. The degassed lung was inflated with air to a Ptp of 25 cm H20, the trachea was occluded and the pressure allowed to equilibrate. The pressure-volume behavior was measured by stepwise removal ofvolume to deflation Ptp values of 15, 10, 5, 3 and 0 cm H20. After degassing the lung, the airspaces were washed with kerosene by inflating the lung with kerosene and deflating to 0 cm H20 Ptp. This procedure was repeated three times. Fresh kerosene was used for each inflation-deflation maneuver. The collapsed lung was then degassed and mounted on a stand with the diaphragmatic pleurai surface superior. The lungs were then inflated with air to a Ptp of 25 cm H20. A pressure-volume behavior of the keJ'osene-washed lung was measured on deflation as described previously for the control lung. Pliq was measured on deflation by micropuncture at TLC and at several equilibrated Ptp values which ranged from TLC to 5 cm H20 ( ~ 50~o TLC). Pliq was not measured on lung inflation because the pressure-volume behavior showed only a small hysteresis (Smith and Starnenovic, 1986). At the end of Pliq measurements the residual volume, wet weight and dry weight of the lungs were obtained as described earlier, In a separate set of lungs, the entire procedure outlined above for the kerosenewashed air-filled lungs was repeated on air-idled lungs after washing with Dew Coming 200 fluid. This fluid has a viscosity of 0.65 cs and surface tension of 16 dyne/cm.
Micropipet-servonuiling method.
The technique used to measure alveolar liquid pressures was similar to that previously reported (Lai-Fook and Beck, 1982). The servonulling pressure measuring system maintains the electrical conductivity across the tip of a glass micropipet constant. The servonulling system accomplishes this by imposing on the liquid within the micropipet a pressure equal to the pressure which is being measured. The pressure within the micropipet was measured using a strain gauge transducer (Statham P23).
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
285
Glass pipets were pulled and bevelled to tip diameters of 2-6/~m on a rotating diamond stone (Micropipette beveller; Sutter Instruments Model BV-10). The pipet was Idled with a 1 M saline solution, mounted on a micromanipulator and connected to a servonulling pressure measurement system (Instruments for Physiology and Medicine, Model 4A). A reference zero was estabfished by immersing the pipet tip in a saline pool above the lung surface. This area ofthe lung was illuminated using a fiberoptic light source (Volpi, Model Intralux 150 H) and viewed through a stereomicroscope (Zeiss, 25X). The entire experiment was conducted with the experimental apparatus placed on a vibration-free surface (Technical Manufacturing Corporation, Vibration Isolation Table, Model Micro-g series 63-500). The pipet was advanced to touch and puncture the pleura] membrane to depths within the lung parenchyma ranging from 0.1 to I mm. A stable alveolar liquid pressure was usually obtained on slow withdrawal of the pipet. This pressure was monitored on a strip-chart recorder (Linear Model 500). Readings were acceptable provided they satisfied the following conditions. 1. The pressure tracing was stable for at least 1-2 rain. 2. The pressure was reproducible, within 2.0 cm H20, on consecutive punctures, not necessarily at the same site. 3. The pressure signal was independent of small changes of the gain setting of the servo-nulling system. This indicated that the tip was immersed in a free liquid having an electrical conductivity similar to normal saline. Thus the pressure measured was in the alveolar liquid layer and not in the kerosene. By independent calibration, we determined that it was not possible to measure pressure in liquids such as kerosene and DC200 fluid with negligible electrical conductivity. 4. The reference zero measured immediately before the puncture was reproduced (within 0.5 cm H20) on removing the pipet from the lung. Preliminarymeasurements showed that the micropipet pressure measured in a calibration chamber was stable within +_0.5 cm H20 of the chamber pressure during a period of 1-2 h. This indicated that diffusion of ions and water into and out of the pipe!: did not change the electrical conductivity across the micropipet tip during the measurements of alveolar liquid pressure. In preliminary studies, we attempted to measure alveolar liquid pressure in edematous lungs which were f'dled with kerosene and in edematous lungs which were air-Idled after washing with oil. The lungs were made edematous by tracheal instillation of 20 mi saline and inflation to TLC. Stable alveolar liquid pressures were not attained in these edematous lungs.
Measurement of interracial tension in vitro. The measurement of the interracial tension of the kerosene-surfactant interface was based on the previous demonstration that the surfactant film is continuous under an oil drop if this drop is placed from the air phase onto a surfactant film (SchOrch et al., 1985). The present experimental approach was similar except that a captive air bubble (SchOrch etal., 1985) replaced the Wilhelmy-Langmuirtype surface balance of the previous demonstration. In the present investigation we chose the captive air bubble because of the greater stability of its air-saline surfactant films. An air bubble was formed below a layer of a 1~o agar gel that was molded into the lucite cover of a chamber with optically flat glass walls (fig. l). The
286
s. GANESAN et al.
A
B Fig. 1. Configurationof surfactant-keroseneinterface in the captivebubble technique. (A) Surfacetension of the large.captive bubble = 25 dyne/era; interracial tension of the kerosene drop-surfactant interface -2 dyne/cm. The diameter of the captive bubble is approximately I cm. (B) The configuration at a surface tension of 10 and an interracial tension of I dyne/cm.
chamber was filled with saline (0.9~o NaCI). A lung surfactant film was formed at the bubble air-water interface by spreading droplets of an aqueous suspension of cow lung surfactant extract (Yu et al., 1983) via a microsyringe. The needle of the syringe was advanced through holes, filled with agar gel, in the lucite cover. After achieving a fdm surface tension between 25 and 30 dyne/era at the bubble surface, a kerosene droplet was placed from the air phase onto the surfactant fdm. This kerosene drop spread to a relatively fiat lens at film surface tensions above 30 dyne/era. When the film surface tension was reduced, the contact angle of the kerosene drop increased and the drop changed from that of a flat lens to that of an inverted sessile drop. Figure I demonstrates the saline-kerosene-air bubble configuration at an air-saline film surface tension of 25 dyne/era (A) and 10 dyne/era (B). The corresponding kerosene-saline surfactant tensions were 1.8 and 1.0 dyne/era. The surface tension of the large, captive air bubble was determined by using the approximation described by Malcolm and Elliott (1980) valid for bubbles having 180 ° contact angles and whose ratio of width to height is within a given limit (SchQrch, 1982). The interracial tension ofthe surfactant-kerosene interface was determined by digitizing the photographs of the kerosene drop and by using the computer programs based on the formulas of Rotenberg et al. (1983). The simple formula of Malcolm and Elliott
287
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
(1980) could not be used for these drops since the contact angles of the kerosene were different from 180 °. The contact angles ofthe kerosene drops changed according to the air bubble surface tension from less than 20 ° to more than 140 °. We always placed a kerosene drop onto the surfactant film of the air bubble kept at a surface tension of 30 dyne/era. The bubble surface tension was then decreased to 25 dyne/cm by withdrawing air through a 26 gauge needle that reached the interior of the bubble through the ceiling of the container. Because of film continuity, the kerosene drop changed in shape accorOing to its decreasing surface tension. Additional pairs of surface and interracial tension values were generated by withdrawing air from the bubble. We also noticed that by inflating the bubble the shape of the bubble and the drop, and their corresponding interfacial tensions, were reversible.
Results The pressure-volume (PV) curves ofthe control air-fdled lung on deflation and inflation, the kerosene-filled lung, the kerosene-washed air-filled lung and the DC200-washed air-filled lungs on deflation are summarized in fig. 2. The PV curves of the control air-filled lung on inflation and the two oil-filled lungs were shifted to the right of the PV
100"
i
80'
i ~
60' 40'
.~/,e."°'°~ i
i
/
C --
/
2o'
...... '~ .....
/ ...... o ..... ,
,~ .n
O
Kerosene filled lung
oc oowa..ed,..g
o o
Kerosene washed lung
•
Control air-filled (deflation)
--
Control air-filled (inflation)
~ 0
10 Transpulmonary
20 pressure
30 (cmH20)
Fig. 2. Pressure-volume curves of the kerosene-filled lung, the control air-fiUed lung on inflation and deflation, the kerosene-washed and DC200 oil-washed lung on deflation. Points represent mean values. Bars represent standard deviation. Some bars have been omitted for clarity. Note the rightward shift in the curves of the control air-filled lung on inflation and the two oil-washed lungs, compared to the control air-filled lung on deflation.
S. GANESAN et aL
288
TABLE I
Measurements of alveolar liquid pressure in kerosene-filled lungs on lung deflation Ptp (cm H20)
N
7.9 + 2.1" 5.5 + 0.60 3.8 + 1.1
3 4 3
Pliq (cm H20)
Volume
T
( ~ TLC)
(dyne/cm)
6.8 + 0.74 4.9 + 1.0 3.3 + 0.87
100 90 71
2.2
1.1 0.9
Ptp, transpuimonary pressure = alveolar pressure relative to pleural pressure. Pliq, alveolar liquid pressure relative to pleural pressure. T, alveolar surface tension calculated using Laplace's law; *, mean ± SD. N,
number of measurements.
curve of the control air-fiUedlung on deflation, indicating an increase in alveolar surface tension. The PV curve of the kerosene-fdled lung was shifted to the left of the PV curve of the control air-fdled lung on deflation and showed a behavior similar to that of a saline-filled lung (Bachofen et al., 1970). TLC of the lungs before washing with kerosene and DC200 averaged 94.8 + 9.5 (SD) ml (N = 18). Table 1 summarizes values of alveolar liquid pressure relative to pleural pressure (Pliq) measured in alveolar liquid layer in the kerosene-f'dled lung at different transpulmonary pressures and lung volumes. Note that Pliq was on average 1.1 cm H 2 0 below the airway pressure (that is, Ptp) at 100~o TLC and decreased to 0.5 cm H20 below airway pressure on lung deflation to 7 1 ~ TLC. These values for the pressure difference across the kerosene-surfactant interface at 100 and 71 ~ TLC are consistent with values of interracial tension of 2.1 and 0.9 dyne/cm, respectively. In these calculations we used Laplace's law for a spherical interface which relates pressure difference (&P) across the interface to the surface tension T and radius of curvature (R): Ap -- 2T/R. We assumed that R varies as the cube root of the lung volume. We used a value of It of 40 tam (Tenney and Remmers, 1963). TABLE 2
Surfactant-air and surfactant-kerosene interfacial tensions measured at different degrees of compression by the captive bubble technique. The corresponding deflation lung volumes are shown Lung volume
Surface tension, dyne/era
(% TLC)
Surfactant-air interface
Surfactant-kerosene interface
30 25 10 5 1
2.0 + !.8 + 1,0 + 0.5 + 0.1 +
*Mean + SE (N = 4).
0.04* 0.04 0.04 0.05 0.05
100 95 75 60 42
289
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS TABLE 3
Measurements of alveolar liquid pressure in nonedematous air-filled lungs (control) on lung deflation and inflation Ptp (cm H20)
N
Pliq (cm H20)
Volume (% TLC)
T
(dyne/cm)
Deflation 22.5 + 2.9* 15.6 + 0.5 10.3 + 0.7 5.3 + 0.2 3.2 + 0.2
5 5 9 9 5
2.9 + 1.1 0.3 + 3.0 0.5 +_ 2.1 0.1 + 1.4 1.0 + 1.1
100 90 85 70 49
28.0 21.1 13.3 6.7 2.4
Inflation 15.0 + 1.7 13.4 + 0.7
5 5
11.3 + 0.8
4
-0,7 + 2.6 -0.9 + 3.7 - !.4 + 0.9
76 64 48
21.1 17.6 14.2
Ptp, transpulmonary pressure = alveolar pressure relative to pleural pressure. Pliq, alveolar liquid pressure relative to pleural pressure. T, alveolar surface tension calculated using Laplace's law; *, mean + SD. N, number of measurements.
TABLE 4 Measurements of alveolar liquid pressure in kerosene-washed and DC200-washed air-filled lungs on lung deflation Ptp (cm H20) Kerosene-washed 23.0 + 1.9"
N
Pliq (cm H20)
5
6.0 + 1.7
15.6 + 0.7
7
1.1 + !.4
12.1 + 0.8 8.5 + I.I
6
4
- 0 . 2 + 0.8 -0.5 + 1.8
DC200-washed 22.5 + 1.3 14.7 + 0.9 10.4 _+0.4 8.3 + 0.5
5 5 6 6
8.0 2.6 1.5 0.3
+ 2.1 + !.9 _. 1.4 + 0.8
Volume (% TLC)
T (dyne/cm)
100 90 73 57
28.0 23.0 18,2 12.2
100 94 77 53
16.0 13.1 9.0 7.2
Ptp, transpulmonary pressure = alveolar pressure relative to pleurai pressure. Pliq, alveolar liquid pressure relative to pleural pressure. T, alveolar surface tension calculated using Laplace's law; *, mean _. SD. N is number of measurements. DC200, Dew Coming 0.65-CS dimethyi siloxane.
290
s. GANESAN et al.
Table 2 shows the values of the surfactant-air and kerosene-surfactant interfacial tensions measured using the captive bubble method. Surfactant-air interracial tensions were compared to the alveolar surface tension-lung volume relationship measured in isolated perfused rabbit lungs by an improved microdroplet method (Bachofen, 1987) and the corresponding values of lung volume are shown in table 2. This comparison indicates that the surfactant-kerosen¢ interface tension is 2, 1.0 and 0.1 dyne/cm at 100 % TLC, 75 % TLC and 42 % TLC, respectively. Note the close agreement between the in vitro measurements of surfactant-kerosen¢ interracial tension and the in situ estimates based on the surfactant-kerosene interracial pressure measured in the alveolar liquid layer in kerosene-filled lungs (table 1). Table 3 summarizes values of Pliq measured in the control air-f"died lung on lung deflation and inflation. Table 4 summarizes values of Pliq measured in the kerosenewashed air-filled lung and DC200 washed air-filled lung. The difference between Pliq and alveolar air pressure (Palv = Ptp) is the interracial pressure across the air-liquid interface in these lungs and is plotted v s Ptp in fig. 3 and v s lung volume in fig. 4. The values of alveolar surface.tension calculated using values of Pliq-Palv in Laplace's law for the three sets of lungs are shown in tables 3 and 4 and plotted against lung volume in fig. 5. For these calculations we assumed values of alveolar surface tension of 28, 28 and 16 dyne/cm at TLC in the control air-filled lungs, the kerosenewashed and DC200-washed air-filled lung, respectively. These values were used previously by Smith and Stamenovic (1986). Figure 5 shows that surface tension in the air-filled lung decreased from a value of 28 dyne/cm at 100% TLC to a low value of ~ 2 dyne/cm at 50% TLC on lung deflation then increased to a value of 14 dyne/era on lung inflation from residual volume (0 cm
i
~
controldelletion w controlinflation --o- DC200washed --o--, kerosenewashed
e~ ~ ~ : ~
• E
o
1'o
2'o
Transpulmonary pressure (emH20) Fig. 3. Surfactant-air interfacial pressure vs transpulmonary pressure measured in control air-filled lungs on deflation (Q 0 ) and inflation ( i I ) , in kerosene-washed air-filled lungs ( .................) and DC200-washed lungs ( . . . . . . . . . ). The interfacial pressure was the difference between the alveolar liquid pressure and alveolar air pressure.
291
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
• ; --.o--.
D.
~---o---
controldeflalion controliNlalion 0C 200washed
kero6enew a s h e ~
0 .I-
E U
-10
_> |
o" I o.
40
60
80
100
Volume (% total lung capacity)
Fig. 4. Surfactant-air interfacial pressure versus lung volume (% TLC) measured in control air-filled lungs on deflation ( 0 O) and inflation (B---------m), in kerosene-washed air-filled lungs ( .................) and DC200-washed lungs ( . . . . . . . . . ). The interfacial pressure was the difference between the alveolar liquid pressure and alveolar air pressure.
100
,o
/aSS° 1
I
iS
//.
-
W
/
=) :s 0
/
; 6
o" /
iii
comml(deflationl control(Inftation)
---o--. DC200washed ---.¢.--- kerosenewashed
40
0
, 10
-20
"
30
Surlaca tension (dynes/cm)
Fig. 5. Lung volume (% TLC) v s alveolar surface tension for control air-filled lungs on deflation (Q Q) and on inflation (11 m), for kerosene-washed air-filled lungs ( .................) and DC200-washed lungs ( . . . . . . . . . ). Note the dependence on lung volume for all cases.
292
s. GANESANet al.
H20 Ptp) to 48% TLC. These values are in line with the results obtained by the microdroplet method (Bachofen et al., 1987). The results for lung deflation are consis-
tent with previous values estimated from alveolar liquid pressure measured in edematous rabbit lungs (Lai-Fook and Kaplowitz, 1984). Thus, a small amount of alveolar edema had only a small effect on alveolar liquid pressure. We shall return to this point in the discussion section. Note also that alveolar surface tension in the kerosene-washed lung and DC200washed lung did not remain constant with decreasing lung volumes on lung deflation. In the kerosene-washed and DC200-washed lungs, surface tension decreased from 28 and 16 dyne/cm at 100~/o TLC to 13 and 7 dyne/cm at ~ 50~o TLC, respectively. This suggests that the surfactant-kerosene-air and surfactant-DC200-ak interfaces exhibit a surface tension behavior that is surface area-dependent. This is consistent with the fact that we measured a pressure in the alveolar liquid layer of the kerosene-washed and DC200-washed lungs. Thus, washing the lung with kerosene and DC200 did not completely remove the surfactant layer in these lungs. Also, the decrease in T predicted for the DC200-washed lung was smaller than that for the kerosene-washed lung, probably due to the greater spreading of the DC200 on the alveolar surface in the presence of surfactant.
Discussion Methods. We measured alveolar liquid pressure in isolated nonedematous lungs by micropuncture and estimated alveolar surface tension using the Laplace law for a spherical membrane. There are several assumptions inherent in these estimates. One assumption is that alveolar liquid pressure was measured in the alveolar liquid layer of an alveolus and not in the interstitium in the alveolar wall or in the interstitium surrounding an extra-alveolar vessel. We believe that we were measuring pressure in the alveolar liquid layer because in the kerosene-fdled lung the alveolar liquid pressure was just below the alveolar pressure in the kerosene, and the differences between the alveolar pressure and the alveolar liquid pressure were consistent with values of the kerosenesurfactant interracial tension measured in vitro. Furthermore, previous studies show that interstitial pressures were above the pleural pressure (Fike et al., 1988). Another assumption is that alveolar liquid pressure was uniform throughout the lung. This is clearly an approximation because consecutive alveolar liquid pressures measured at sites 1-5 mm apart in some instances differed by as much as 1.5 cm H20. Since these lungs were not edematous, equilibrium in liquid pressure between adjacent alveoli might be very slow because interstitial resistance might be high. It is also possible that lung distortion caused by the pipet might have been responsible for these differences in pressure. We assumed that alveolar radius varied as the cube root of the lung volume (Klingele and Staub, 1970) on both lung deflation and inflation. Previous results show a slower decrease in alveolar radius with lung volume (Gil et al., 1979), and a marked hysteresis in the behavior of alveolar surface area and airspace curvature with lung
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
293
volume (Bachofen et al., 1987). However, our estimates of alveolar surface tension would be in error by 21 % at 50% TLC if we were to assume that R remained constant with lung deflation. Thus, our estimates of alveolar surface tension are more dependent on the interfacial pressure than on the variation of alveolar radius with lung volume. To calculate surface tension as a function oflung volume, we assumed the eqUilibrium values of alveolar surface tension of 28 dyne/cm for nonedematous air-filled lung and the kerosene-washed lung and 16 dyne/cm for the DC200-washed lung, at TLC (Smith and Stamenovic, 1986). For the kerosene-filled lung we assumed an alveolar radius of 40 J.lm at TLC (Tenney and Remmers, 1963). It is evident that the actual magnitudes of surface tension estimated were directly dependent on the above assumptions. However, these assumptions do not invalidate our conclusions concerning the reduction in surface tension with a decrease in lung volume.
Comparison with other results.
We estimated alveolar surface tension to decrease from 28 dyne/cm at TLC to 2 dyne/cm at 50% TLC on lung deflation of the nonedematous air-filled lung. These estimates of alveolar surface tension on lung deflation are consistent with the direct results using the microdroplet method (Schtlrch et ai., 1985), with studies using the pressure-volume curve (Bachofen et ai., 1970), and comparisons among PV curves of the air-filled lung and the air-filled lung after washing with oils of varying surface tensions (Smith and Stamenovic, 1986). One of the objectives of this study was to determine the relationship between alveolar surface tension and lung inflation. Our results show that alveolar surface tension increased from 13 dyne/cm at 50% TLC to 21 dyne/cm at 85% TLC on lung inflation from residual volume (0 cm H 2 0 Ptp). The micro droplet method showed that alveolar surface tension increased from 8 dyne/cm at 50% lung volume to 25 dyne/cm at 90% TLC on lung inflation (Bachofen et al., 1987). The latter result agrees with values estimated from alveolar liquid pressure measured in edematous dog lungs (Lai-Fook and Beck, 1982). The higher values estimated in this study are consistent with the greater static recoil measured on inflation (fig. 2), most likely due to the much longer equilibration times (10-15 min) at constant volume required for the micropuncture measurements. The above three results contrast with the constant value of 28 dyne/cm estimated by Smith and Stamenovic (1986). The constant surface tension was the direct result of the assumption that the oil-washed air-filled lung on lung deflation has a constant alveolar surface tension. This assumption is not supported by our results of alveolar liquid pressure in oil-washed lungs, which showed a dependence of alveolar surface tension with lung volume.
Implication of the results. The values of alveolar liquid pressure measured in nonedematous lungs in this study are in line with previous measurements in lungs with a small amount of edema (50% increase in extravascular lung water). This suggests that alveolar liquid pressure is not a strong function of interstitial edema. This may be due to the fact that interstitial edema occurs before significant alveolar edema fonnation (Heff, 1971).
294
S. GANESAN
el al.
The transmission of alveolar liquid pressure to the perivenular interstitium has recently been addressed in isolated perfused lungs (Fike et a/., 1988). There was a virtual equilibration in pressure between the alveolar liquid and perivenular interstitial spaces, which was consistent with a relatively short time constant for interstitial filling of small extra-alveolar vessels (Conhaim et a/., 1986). By contrast, previous studies indicated a difference between the perivenular interstitial pressure and the interstitial pressure in alveolar septal comers even in the presence of gross edema (Bhattacharya et aI., 1984). To the extent that alveolar liquid pressure is transmitted to the interstitium so that it is a reflection of interstitial pressure, our measurements of alveolar liquid pressure as a function of lung volume indicate that transvascular fluid flux would increase at high lung volumes. However, the exact relationship between alveolar liquid pressure and alveolar wall interstital pressure is still uncertain. Our results indicate that a small but finite interfacial pressure exists across the surfactant-kerosene interface in kerosene-filled lungs. consistent with previous in vitro measurements of surface tension (Smith and Stamenovic. 1986). This small interfacial tension would prevent kerosene from entering the interstitium and prevent the formation of interstitial fluid cuffs around large vessels as obtained in saline-inflated lungs (Conhaim et a/., 1986). In summary, alveolar liquid pressure measured in kerosene-filled and oil-washed air-filled lungs showed that an alveolar liquid layer was still present in these lungs. This indicates that a surfactant-kerosene interface in the kerosene-filled lung and a surfactant-oil-air interface in the oil-washed air-ftlled lung result in a defmite alveolar surface area dependence on surface tension. Interfacial pressures measured in kerosene-filled lungs were consistent with the surface tension of a surfactant-kerosene interface measured in in vitro studies. Parallel to the monotonically increasing lung static recoil on lung inflation, alveolar surface tension estimated from alveolar liquid pressure measurements shows a defmite increase with lung inflation, and not a constant surface tension as recently proposed (Smith and Stamenovic. 1986).
Acknowledgements. We thank Laura V. Brown for technical assistance. This research was supported by NIH grant HL 40362 and by the Medical Research Council (Canada).
References Albert, R.K., S. Lakshminarayan, 1. Hildebrandt, W. Kirk and J. Butler (1979). Increased surface tension favors pulmonary edema formation in anesthetised dogs' lungs. J. Appl. Physiol. 43: 730-738. Bachofen, H., J. Hildebrandt and M. Bachofen (1970). Pressure volume curves of air- and liquid-filled excised lungs-surface tension in situ. J. Appl. Physiol. 29: 422-43l. Bachofen, H., S. Scht1rch, M. Urbinelli and E.R. Weibel (1987). Relations among alveolar surface tensions, surface area, volume, and recoil pressure. J. Appl. Phystoi. 62: 1878-1887. Bhattacharya, J., M.A. Gropper and N.C. Staub (1984). Interstitial fluid gradient measured by micropuncture in excised dog lung. J. Appl. Physiol. 56: 271-277.
ALVEOLAR LIQUID PRESSURE IN RABBIT LUNGS
295
Beck, K. C. and S. J. Lai-Fook (1983). Alveolar liquid pressure in excised edematous dog lung with increased static recoil. J. Appl. Physiol. 55: 1277-1283. Bruderman, I., K. Somers, W.K. Hamilton, W.H. Tooley and J. Butler (1964). Effect of surface tension on circulation in the excised lungs of dogs. J. Appl. Physiol. 19: 707-712. Clements, J.A. Surface tension of lung extracts (1957). Proe. Soc. Exp. Bioi. Med. 95: 170-172. Clements, J.A. (1961). Pulmonary edema and permeability of alveolar membranes. Arch. Environ. Health 2: 280-283. Clements, J .A., R F. Hustead, R P. Johnson and I. Gribetz (1966). Pulmonary surface tension and alveolar stability. J. Appl. Physiol. 16: 444-450. Conhaim, RL., S.J. Lai-Fook and N.C. Staub (1986). Sequence of perivascular liquid accumulation in liquid-inflated dog lung lobes. J. Appl. Physiol. 60: 513-520. Faridy, E. E., S. Permutt and R L. Riley (1966). Effects of ventilation on surface forces in excised dogs']ungs. J. Appl. Physioi. 21: 1453-1462. Fike, e.D., S.J. Lai-Fook and RD. Bland (1988). Alveolar liquid pressure in newborn and adult rabbit lungs. J. Appl. Physiol. 64: 1629-1635. Gil, J., H. Bachofen, P. Gehr and E.R. Weibel (1979). Alveolar volume-surface area alterations in air and saline filled lungs fixed by vascular perfusion. J. Appl. Physiol. 47: 990-1001. Ileff, L.D. (1971). Extra-alveolar vessels and edema formation in excised lungs. Cire.Res. 28: 524-532. Klingele, T.G. and N.C. Staub (1970). Alveolar shape changes with volume in isolated, air filled lobes of cat lung. J. Appl. Physiol. 28: 411-414. Lai-Fook, S.J. and K.e. Beck (1982). Alveolar liquid pressure measured by micropipettes in isolated dog lung. J. Appl. Physiol. 53: 737-743. Lai-Fook, S.J. and M.R Kaplowitz (1984). Alveolar liquid pressure measured using micropipettes in isolated rabbit lungs. Respir. Physiol. 57: 61-72. Malcolm, J. D. and C. D. Elliott (1980). Interfacial tension from height and diameter of a single profile drop or captive bubble. Can. J. Chem. Eng. 58: 151-152. Pattle, RE. (1955). Properties, function and origin of the alveolar lining layer. Nature (London) 175: 1125-1126. Rotenberg, Y., L. Boruvka and A. W. Neumann (1983). Determination of surface tension and contact angle from the shape of asymmetric fluid interfaces. J. Colloid Interj. Sci. 93: 169-183. Schilrch, S. (1982). Surface tension at low lung volumes: dependence on time and alveolar size. Respir. Physioi. 48: 339-355. Schilrch, S., H. Bachofen and E.R. Weibel (1985). Alveolar surface tensions in excised rabbit lungs: effect of temperature. Respir. Physiol. 62: 31-45. Smith, J.e. and D. Stamenovic (1986). Surface forces in lungs. I. Alveolar surface tension-lung volume relationships. J. Appl. Physiol. 60: 1341-1350. Tenney, S.M. and J.E. Remmers (1963). Comparative quantitative morphology of the mammalian lung: diffusing area. Nature (London) 197: 54-56. Yu, S.-H., P.G.R. Harding, N. Smith and F. Possmayer (1983). Bovine chemical surfactant: chemical composition and physical properties. Lipids 18: 522-529.