| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vol. 83, Issue 6, 1976-1985, December 1997
1 Department of Biomedical Engineering and 2 Center for Pulmonary Research, Vanderbilt University, Nashville, Tennessee 37235
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
FOOTNOTES
REFERENCES
ABSTRACT 
Klaesner, Joseph W., N. Adrienne Pou, Richard E. Parker,
Charlene Finney, and Robert J. Roselli. Optical measurement of
isolated canine lung filtration coefficients at normal hematocrits. J. Appl. Physiol. 83(6):
1976-1985, 1997.
In this study, lung filtration coefficient
(Kfc) values
were measured in eight isolated canine lung preparations at normal
hematocrit values using three methods: gravimetric, blood-corrected
gravimetric, and optical. The lungs were kept in zone 3 conditions and
subjected to an average venous pressure increase of 10.24 ± 0.27 (SE) cmH2O. The resulting Kfc
(ml · min
1 · cmH2O1 · 100 g dry lung wt1) measured
with the gravimetric technique was 0.420 ± 0.017, which was
statistically different from the
Kfc measured by
the blood-corrected gravimetric method (0.273 ± 0.018) or the
product of the reflection coefficient
(
f) and
Kfc measured
optically (0.272 ± 0.018). The optical method involved the use of a
Cellco filter cartridge to separate red blood cells from plasma, which
allowed measurement of the concentration of the tracer in plasma at
normal hematocrits (34 ± 1.5). The permeability-surface area
product was measured using radioactive multiple indicator-dilution
methods before, during, and after venous pressure elevations. Results
showed that the surface area of the lung did not change significantly
during the measurement of
Kfc. These
studies suggest that
fKfc
can be measured optically at normal hematocrits, that this measurement is not influenced by blood volume changes that occur during the measurement, and that the optical
fKfc
agrees with the
Kfc obtained via
the blood-corrected gravimetric method.
Evans blue; pulmonary vascular volume
INTRODUCTION EACH year, approximately 150,000 people are affected by
acute respiratory distress syndrome (ARDS), with a 50% mortality rate, even with supportive therapy (31). There are many initial causes for
ARDS, but eventually endothelial damage occurs, causing increased protein flux and fluid filtration. Transvascular fluid filtration is
proportional to total lung mass and to the combined hydrostatic and
osmotic pressure difference across the microvascular barrier. The
constant of proportionality is defined as the filtration coefficient (Kfc) and is a
measure of the porosity of the barrier. Two phenomena are generally
responsible for an abnormally high filtration rate: an increase in the
effective pressure difference across the barrier and an increase in
Kfc.
Lung injury is normally associated with an increase in
Kfc. The
measurement of
Kfc would be of
clinical significance in the treatment of ARDS. Until recently, the
only technique available for measuring
Kfc was in
isolated lung preparations. Kfc can be
estimated in animal preparations in which lung lymph is available, but
this measurement is influenced by the unknown fraction of lung lymph
collected. Obviously, neither of these methods can be extended to
humans, since both are highly invasive.
Until recently, the "gold standard" for measurement of
Kfc was the
gravimetric method performed on an isolated lung preparation. The
method entails an isolated lung that is suspended from a weight transducer and perfused with homologous blood using an extracorporeal circuit. When the preparation reaches an isogravimetric state, the
pulmonary venous pressure is increased in a steplike fashion. The rapid
initial weight gain during the first 1-3 min is generally attributed to vascular filling, whereas fluid filtration is attributed to the slower weight gain that follows (3, 7). The
Kfc can be
computed using the constant slope or extrapolation method. Harris et
al. (11) showed that the extrapolation method was less accurate than
the constant slope method; thus the data in this study were analyzed
using the constant slope method. This method assumes the slope of the
slower rate of weight gain to be linear with respect to time and to be
the rate of fluid filtration. Kfc can be
calculated by dividing this slope by the microvascular pressure
increase and the dry lung weight (7).
Both gravimetric calculations assume that vascular blood volume (BV)
increases for only 1-3 min after the venous pressure is increased,
but Harris et al. (11) showed with the use of 51Cr-labeled red blood cells (RBC)
that vascular volumes can increase for up to 30 min after a pressure
elevation. Maron and Lane (21), with the use of indicator-dilution
methods, also showed that BV can continue to increase for >1-3
min after a venous pressure increase. They found, however, that if the
isolated lung was perfused for an extended period of time before a
pressure elevation, the BV may not increase after the first 1-3
min, even though two of six animals showed some increase. Thus the
gravimetric technique can be adversely affected by slow changes in
vascular BV and, in any event, cannot be used in vivo.
A method for estimating
Kfc in intact
animals is the measurement of pulmonary pressures and lung lymph flow.
This technique is generally used with sheep, because the caudal
mediastinal lymph node drains a large percentage of the pulmonary
interstitium. Kfc
can be calculated by dividing the "total" lung lymph flow by an
estimated transvascular pressure difference. Complications with this
method include the fact that the capillary and interstitial pressures
cannot be measured directly and the percentage of total lung lymph
collected from a single node is unknown and can vary when pressures are
changed (4). In addition, lung lymph cannulation cannot be extended to
human Kfc
measurement.
Oppenheimer et al. (22) introduced an optical technique for measurement
of Kfc.
The technique depends on measuring a
change in optical tracer concentration in lung venous blood that occurs after a step change in pressure. The concentration of the optical tracer increases after a pressure elevation, because water flows across
the pulmonary capillary barrier more readily than proteins or other
large solutes. Thus, by determining the tracer concentration change for
a given pressure step, the
Kfc can be
calculated. The main disadvantage of this technique is that RBC
strongly absorb and scatter light at the wavelengths used to measure
concentration changes of the optical tracers. Thus small hematocrit
changes, caused by the Fahraeus-Lindqvist (26) effect or respiration (20), greatly complicate optical measurements in whole blood. Because
of these problems, it was necessary for Oppenheimer's group to use
multiple lasers with different wavelengths to penetrate the blood and
correct for simultaneous changes in hematocrit and oxygen saturation
(22). Using this optical system, but observing only a single wavelength
to follow the hematocrit changes after a pressure step, Hancock et al.
(10) found that optically measured values of
Kfc were ~25%
of those calculated using weight changes. They reasoned that this
difference could be due to slow vascular volume changes that are
misinterpreted as filtration when weight analysis is used.
Harris et al. (11) used a spectrophotometer to optically measure
Kfc values for
isolated canine lungs at low flows and small hematocrits. They measured
changes in the concentration of albumin labeled with indocyanine green
caused by a pulmonary venous pressure increase while they
simultaneously corrected for fluctuations in RBC concentrations
measured at 650 nm. Lung weight changes were monitored during pressure
increases, and vascular volume changes were monitored using
radioactively labeled RBC.
Kfc values obtained via gravimetric methods were significantly larger than those
obtained via the optical techniques. After correction for BV changes,
however, the gravimetric
Kfc values were
comparable to those obtained using the optical method. Unfortunately,
artifacts introduced by RBC masked the small changes in plasma protein
concentrations, thus restricting the measurements to low hematocrits
and low flow rates.
In our preliminary work (19) it was
shown that the separation of
RBC from plasma allowed
for optical measurements of
Kfc at physiological hematocrits and flow
rates. On-line
separation of plasma and
RBC was attained by
passing blood through a polysulfone filter cartridge
before measuring the concentration of plasma albumin labeled with Evans
blue (EBA). These concentration changes were then used to calculate
Kfc
(19). However, the optical
Kfc values were not compared with
simultaneous gravimetric measurements performed on the
same isolated canine lung. In this study,
Kfc was measured in an isolated canine lung preparation
using gravimetric, blood-corrected
gravimetric, and optical methods. A commercial filter cartridge was
used to separate the plasma from the RBC, the filtering characteristics
of which were documented previously by Klaesner et al.
(18). The values from these methods were compared to
determine whether optical
Kfc values can
be obtained at higher
flow rates and physiological hematocrit levels.
A possible problem that can be associated with the
gravimetric method of measuring
Kfc is that
elevated venous pressure could alter the perfused surface area via
microvascular recruitment in the canine lung. An
increase in surface area during elevated hydrostatic
pressures would give an inflated
Kfc value. We
address this possibility
by using a multiple indicator-dilution (MID) technique
to measure the permeability-surface area product
(PS) before, during, and after the
Kfc measurement
period.
METHODS
f) and
Kfc
where
(1)
pa is arterial
plasma flow (ml/min), 
Cpv is
change in plasma concentration of nondiffusing tracer (mg/dl),
Cpv is initial plasma
concentration of nondiffusing tracer (mg/dl),
mdlw is blood-free dry lung weight
(g), Pmv is change in
microvascular pressure (cmH2O),
and f is reflection coefficient
for the optical tracer.
f was assumed
to be 1.0 for the optical tracer
EBA. Parker et al.
(24) measured
f for albumin in sheep lungs
and found it to be >0.84. Isago et al.
(17) measured microvascular reflection coefficient of
sheep by venous occlusion and found values of ~0.82.
Drake and Gabel (5) found
f to be between 0.72 and 1 in
dog lungs. These values are consistent with model predictions of 0.87 and 0.89 for the three-pore models of Harris and
Roselli (13), 0.90 for a two-pore model of Roselli et
al. (30), 0.80 for a three-pore model of Blake and
Staub (1), and 0.84 for the two-pore
model of Roselli et al. (27). Thus the
actual value of f is probably
<1.0, which implies a
slight underestimate of optical
Kfc estimated
with f = 1.0 in
Eq. 1.
The capillary pressure was estimated using the formula developed by
Garr et al. (7)
|
(2) |
W/t),
Pmv, and
Cpv is shown in Fig. 1. Figure
1C, the raw optical data, shows the
initial baseline drift caused by filtration before the pressure
elevation. The slope of the baseline drift immediately before the
change in absorbance due to the increase in venous pressure is
determined and subtracted from the raw data, resulting in the
absorbance trace in Fig. 1D. The rapid
absorbance change trails the pressure step by ~1 min because of the
transfer function of the filter system (18). Ideally,
Cpv would be the difference
between the baseline and the first plateau after the initial rapid
absorbance increase, as shown by Harris et al. (11). Unfortunately, the
transfer function of the filter tends to mask this plateau; thus
Cpv is determined by looking
for the "break point," where the initial rapid change in
absorbance slows (Fig. 1D). These
values are then used in Eq. 1 to
determine
fKfc
and in the gravimetric equation [Kfc = (W/t)/PmvMdlw]
to determine
Kfc.
It is not necessary to convert the absorbance change to concentration,
because Cpv and
Cpv are in absorbance units, and
since concentration is directly proportional to absorbance units, the
units cancel. Ideally, the Cpv
for the pressure decrease could be used for calculation of fKfc,
but the pressure step tends to be less ideal and the optical signal
much noisier, making the calculation of
Kfc difficult or impossible. The typical pressure step rise time (0 to two-thirds maximum) is ~5 s, whereas the typical pressure fall time (maximum to
one-third maximum) is ~15 s. The disparity between the rise and fall
times of the pressure changes is due to the ability to totally
interrupt the venous blood from lungs, allowing for a rapid step
increase. There is no equivalent maneuver available to allow for the
pressure to fall more quickly.
MID theory. The MID measurements of urea PS (PSU) involved the injection of several radioactive tracers with different diffusing characteristics at the inlet of the lung and the collection of samples at the outlet of the lung. The vascular, or reference, tracers used in this study were 99Tc-RBC and 125I-albumin, and the barrier-limited diffusing marker was [14C]urea. 3HOH was the flow-limited tracer and was used to determine the extravascular water volume. Forty samples were collected downstream of the lung using a revolving collection wheel that rotated at a rate of 1 sample/s. The gamma isotope activity of each sample and injectate was measured in a Packard Auto-Gamma scintillation spectrometer (model 5921), and the beta activity was measured in a liquid scintillation counter (model LS 3150T, Beckman). The radioactive counts of the isotopes in each sample were then converted to tracer concentrations and plotted with respect to time. These tracer concentration-time profiles were then used in the equations below to calculate the PSU as described by Harris and Haselton and co-workers (12, 14). The integral extraction (Ei) for [14C]urea was computed as follows
|
(3) |
|
(4) |
|
(5) |
|
(6) |
H2O
is mean transit time for labeled water and
R is mean
transit time for reference tracer.
If flow rates were altered during the experiments, the Crone-Renken
PS may have changed with flow (29).
Therefore, we calculated the product of the square root of the
diffusivity
(Dequiv)
and surface area
(D1/2S),
which is not flow dependent (28).
D1/2S
was calculated using the model proposed by Haselton et al. (14, 15),
which uses the mass balance of the tracer transport shown in
Eq. 7 to account for backdiffusion
from extravascular spaces
|
(7) |
|
(8) |
|
(9) |
|
(10) |
|
(11) |
is normalized dimension perpendicular to flow,
z is normalized dimension
parallel to flow, CDI is tracer
concentration in extravascular space, and Dequiv
is diffusivity described by the following equation
|
(12) |
The experimental protocol for the canine preparation was outlined by Harris et al. (11). Briefly, a large (>50-pound) mongrel dog, screened for microfilaria, was anesthetized with pentobarbital sodium (30 mg/kg) and maintained with halothane. A catheter was inserted into the femoral artery through which heparin was injected (600 U/kg) and allowed to circulate for 15 min to prevent clotting. The dog was then exsanguinated, and the blood was saved for the study. A thoracotomy was performed through the fifth intercostal space, the main pulmonary artery was ligated, and the heart and lungs were removed. The lower portion of the heart was removed to allow the left atrium and the main pulmonary artery to be cannulated. The lungs were then connected to the ventilation-perfusion system shown in Fig. 1, with care to ensure that air is not allowed into the vessels, and placed into the humidified chamber. The lungs were ventilated with 95% O2-5% CO2 that was kept at a constant pressure of 2 cmH2O by adjustment of the resistance of the air outflow. The lungs were held in zone 3 by setting venous pressure >3 cmH2O relative to the top of the lung. NaI detectors were positioned over the suspended lung and the perfusate to monitor radioactivity. Baseline radioactivity was monitored for 5 min at a rate of 2 samples/min before addition of any isotopes to the perfusate to obtain baseline radioactivity. RBC were radiolabeled with 20-30 µCi of 51Cr and injected into the reservoir. The labeled RBC were distributed throughout the lung by raising and lowering venous pressure several times. The increase in radioactivity due to the 51Cr-RBC was converted to pulmonary blood weight increase and subtracted from the rate of lung weight gain to correct the gravimetrically calculated Kfc (11). EBA was prepared by centrifuging enough blood (normally 100-120 ml) to obtain 65 ml of plasma and adding 13 mg of Evans blue to the resultant supernatant. This was mixed on a rocker for ~10 min to enable the Evans blue to form a tight covalent bond with the albumin (19). About 20 ml of plasma and 5 ml of EBA were set aside to make calibration standards. Once the lung perfusion and filtrate sampling systems were operating properly, the remaining 60 ml of EBA were added to the reservoir. This provided a step increase that approached a baseline plasma absorbance of ~0.2 absorbance unit. Once the optical signal was stable, the lungs were subjected to several pressure elevations. An experimental run consisted of elevating lung venous pressure by 8-15 cmH2O for 10-12 min, then lowering venous pressure back to the original value. The pressure was elevated by clamping both tubes at the resistor "Y" (Fig. 1), with the clamp on the resistor arm released when the pressure reached the appropriate level. The resistor was adjusted until the desired venous pressure was attained. Both tubes were initially clamped so that a more ideal pressure step was achieved. Samples from the reservoir and the filtrate were drawn before each pressure elevation. These samples were used to determine the initial concentration of EBA before each run. The optical responses were observed using the spectrophotometer (model 1706 UV/Vis monitor, Bio-Rad), and the pressure changes (model 1290A, Hewlett-Packard) were recorded at 1 Hz using a Kiethley-Metrabyte DAS-20 analog-to digital card. Kfc was then calculated using Eq. 2 (11). The PSU of the isolated lungs was measured using the technique outlined in MID theory. The PSU was measured before, during, and after the third venous pressure increase (Fig. 3). Thus any changes in PSU due to increases in hydrostatic pressures could be tracked.
When the experiment was completed, the lungs were weighed, homogenized, dried in a microwave oven at low power, and reweighed. Samples from the lung were counted in the Packard Auto-Gamma scintillation spectrometer to use the 51Cr-RBC to determine the blood-free dry lung weight (BFDLW) so that the PSU values of different-sized lungs could be compared by a standardized method (2, 25). Also a set of EBA concentration standards was prepared. Preparation involved serially diluting 5 ml of EBA stock with plasma. These standards were passed directly through the spectrophotometer, creating an absorbance-to-concentration conversion relation. The samples taken during the experiment were also passed directly through the spectrophotometer, and the absorbances of the samples were converted to concentrations of EBA. The concentrations were used in Eq. 2 to calculate the Kfc for each run.
RESULTS
Eight isolated canine lung preparations were used in this study. Each
lung was subjected to 5-10 venous pressure increases, with the
optical, gravimetric, and RBC-corrected gravimetric
Kfc values measured for each run (Table 1). The
constant slope technique was used to calculate the
Kfc
values for the gravimetric techniques (7). Using two-way analysis of
variance with the Student-Newman-Keuls test for multiple comparisons,
it was shown that the optically measured
fKfc
(0.268 ± 0.018 ml · min1 · cmH2O1 · 100 g dry lung wt1) was not
statistically different from the gravimetric
Kfc
corrected for blood volume changes (0.256 ± 0.018 ml · min1 · cmH2O1 · 100 g dry lung wt1). Both
values, however, were statistically different from the Kfc
measured gravimetrically (0.420 ± 0.017 ml · min1 · cmH2O1 · 100 g dry lung wt1). The
average pressure step was 10.24 ± 0.27 cmH2O, and the average hematocrit
of the blood was 34 ± 1.5.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The compliances for all the pressure elevations were calculated by
dividing the initial weight change by the pressure step and the
blood-free dry weight and are shown in Table 1. The compliance values
and the compliance values normalized by the first value were plotted
against the measurement run for all the studies in Fig.
4 to determine whether there were any
changes in compliance with respect to time. The results in Fig. 4
suggest that vascular compliance decreased slightly during the
experiment, probably because of increased fluids in the interstitium.
All the
Kfc
values and the
Kfc
values normalized by the first measured value were plotted against
experimental runs for the blood-corrected gravimetric data in Fig.
5 and the optical data in Fig.
6. The
Kfc
values tend to increase with time, probably because of breakdown in the endothelial barrier. Figure 7 shows how the
compliance and
Kfc values changed during the course of a single isolated lung study (dog FB8).
The average
PSU
values measured for the eight dogs are summarized in Table
2 and shown in Fig.
8. Figure 8 shows that most of the variance
in
PSU
values is between dogs, rather than between PSU
values measured at different pressures. The average
PSU
measured during elevated pressures (4.97 ± 2.23 ml/s) was slightly
higher than the average values measured at baseline and postfiltration lung venous pressures (4.66 ± 2.07 and 4.75 ± 1.97 ml/s), but the difference was not statistically significant. The
average PSU
values normalized by BFDLW are 0.096 ± 0.035, 0.1024 ± 0.038, and 0.098 ± 0.33 ml · s1 · g1
and are listed in Table 3. The
D1/2S
values are listed in Table 4. These values
were plotted against the
PSU
values in Fig. 9. Because the flow did not
change between the runs, there is a linear relationship between the two
measurements with an
R2
of 0.737. The
Kfc
values for all three measurement methods are plotted against
PSU/BFDLW
in Figs.
10, 11, 12.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
fKfc)
vs.
PSU/BFDLW
for optical data. Heavy line, regression without 2 high
Kfc
values.
There appears to be a linear relationship between the Kfc values and the PSU/BFDLW. Table 5 summarizes the slopes of the lines in Figs. 10, 11, 12. Two values of Kfc did not seem to group with the others, so the regression was performed on the other six values. The R2 values improved substantially without the two outlying values.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Table 6 summarizes the extravascular water content determined via MID data, as well as by postmortem methods. As expected, the extravascular water was considerably higher at the end of pressure elevations, indicating that water was forced into the interstitium or alveoli. Postmortem extravascular water content was significantly more than that found using MID techniques, because MID measures perfused extravascular water. Thus much of the water in the extravascular spaces was present in the alveolar sacs or in areas of the lung that were not perfused.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
DISCUSSION
Our
Kfc
values were compared with those obtained by other investigators using
the constant slope method of analysis for the gravimetric technique and
the optical technique. They tended to agree with measurements by other
investigators, which are summarized in Table
7 (6-9, 16, 22, 23). The
Kfc
found optically by Harris et al. (11) was based on a
f of 0.5, whereas our values
are based on a f of 1.0. Thus
our values were about twice as large as those found by Harris et al.
This can probably be attributed to the fact that Harris et al. perfused
isolated lungs with a very-low-hematocrit perfusate (~1-10%).
|
|||||||||||||||||||||||||||||||||||||||
The present studies show that the optical method of measuring
Kfc
provides values that are very comparable to those obtained by the
gravimetric method once corrected for vascular volume increases. BV
weight gain accounted for ~36% of the gravimetric
Kfc.
The MID studies indicate that
PSU
does not increase during pressure elevations, and since permeability is
not expected to change, surface area appears to remain constant during
a
Kfc
measurement. Thus vascular volume increases are probably due to
vascular relaxation, rather than capillary recruitment. One might
expect some correlation between this vascular relaxation and
compliance, but Fig. 13 shows no obvious
relationship between lung compliance and the percentage of weight gain
of the lung due to blood volume increases during the elevated venous
pressure.
Data displayed in Figs. 10, 11, 12 suggest a correlation between Kfc and PSU/BFDLW. The regression slope for the gravimetric values of Kfc was greater than the regression slopes for the blood-corrected gravimetric or optical Kfc. This could be expected because the average gravimetric Kfc value was greater than the values found with the other two methods.
Data from dogs FB3 and
FB6 seem to contradict the general
finding that the optical
Kfc
compares favorably with the blood-corrected gravimetric
Kfc.
However, a closer look at the data from one or two experimental runs
for these two studies skewed the average value obtained for the entire
dog. In dog FB3, no optical data were
collected for the last two experimental runs because of electronic difficulties. The average
Kfc
measured for dog FB3 when the last two
experimental runs are neglected for the gravimetric, blood-corrected gravimetric, and optical methods are 0.526, 0.424, and 0.447 ml · min1 · cmH2O1 · 100 g dry lung wt1,
respectively. Values for dog FB6 were
somewhat skewed by the first experimental run, which included a very
small gravimetric value of 0.05 ml · min1 · cmH2O1 · 100 g dry lung wt1. If the
first experimental run is neglected, the gravimetric, blood-corrected
gravimetric, and optical
Kfc
values are 0.322, 0.145, and 0.228 ml · min1 · cmH2O1 · 100 g dry lung wt1,
respectively. These values for
Kfc
for dogs FB3 and
FB6 are more in line with the general
findings that the optical and blood-corrected gravimetric techniques
give values that agree with each other and are less than those obtained
using the traditional gravimetric technique.
In conclusion, we believe that the optical method for measuring Kfc provides an accurate measurement of Kfc in isolated canine lung preparations. A practical advantage of this method is that it is independent of lung blood volume changes and thus does not require a separate measurement of vascular volume changes. The results were very comparable to those found by gravimetric techniques after correction for BV increases. By separating RBC from plasma with polysulfone filter fibers, optical concentration measurements can be made in plasma, without the optical artifacts associated with RBC. This allows measurements to be made in blood with physiological hematocrits and normal flow rates. Also, we have shown that the measurement technique of elevating venous pressures does not alter the PSU of the lung significantly, thus showing that lung microvascular surface area does not change during a measurement when the lung is kept in zone 3. A final advantage of the optical technique is that it is not restricted to an isolated lung preparation. The method has been extended to an intact animal preparation, and preliminary experiments are encouraging (19).
FOOTNOTES
Address for reprint requests: J. W. Klaesner, Dept. of Anesthesiology, Saint Louis University Medical School, 1402 S. Grand Ave., St. Louis, MO 63104.
Received 25 September 1996; accepted in final form 14 August 1997.
REFERENCES
| 1. | Blake, L. H., and N. C. Staub. Pulmonary vascular transport in sheep. A mathematical model. Microvasc. Res. 12: 197-220, 1976[Medline]. |
| 2. |
Collins, J. C.,
J. H. Newman,
N. E. Wickersham,
W. K. Vaughn,
J. R. Snapper,
T. R. Harris,
and
K. L. Brigham.
Relation of blood-free to blood-inclusive postmortem lung water measurements in sheep.
J. Appl. Physiol.
59:
592-596,
1985 |
| 3. |
Drake, R.,
K. A. Gaar,
and
A. E. Taylor.
Estimation of the filtration coefficient of pulmonary exchange vessels.
Am. J. Physiol.
234 ((Heart Circ. Physiol. 3):
H266-H274,
1978 |
| 4. |
Drake, R.,
M. Giesler,
G. Laine,
J. Gabel,
and
T. Hansen.
Effect of outflow pressure on lung lymph flow in unanesthetized sheep.
J. Appl. Physiol.
58:
70-76,
1985 |
| 5. |
Drake, R. E.,
and
J. C. Gabel.
Estimation of the pulmonary microvascular reflection coefficient to protein in dogs.
J. Appl. Physiol.
71:
94-98,
1991 |
| 6. |
Ehrhart, I. C.,
W. M. Granger,
and
W. F. Hofman.
Filtration coefficient obtained by stepwise pressure elevation in isolated dog lung.
J. Appl. Physiol.
56:
862-867,
1984 |
| 7. | Gaar, K. A., Jr., A. E. Taylor, L. J. Owens, and A. C. Guyton. Pulmonary capillary pressure and filtration coefficient in the isolated perfused lung. Am. J. Physiol. 213: 910-914, 1967. |
| 8. |
Hall, J. E.,
W. F. Hofman,
and
I. C. Ehrhart.
Venous occlusion pressure and vascular permeability in the dog lung after air embolization.
J. Appl. Physiol.
65:
34-40,
1988 |
| 9. |
Hancock, B. J.,
M. Hoppensack,
and
L. Oppenheimer.
Do transvascular forces in isolated lobe preparations equilibrate?
J. Appl. Physiol.
67:
628-635,
1989 |
| 10. |
Hancock, B. J.,
K. P. Landolfo,
M. Hoppensack,
and
L. Oppenheimer.
Slow phase of transvascular fluid flux reviewed.
J. Appl. Physiol.
69:
456-464,
1990 |
| 11. |
Harris, N. R.,
R. E. Parker,
N. A. Pou,
and
R. J. Roselli.
Canine pulmonary filtration coefficient calculated from optical, radioisotope, and weight measurements.
J. Appl. Physiol.
73:
2648-2661,
1992 |
| 12. | Harris, T. R., and K. L. Brigham. The exchange of small molecules as a measure of normal and abnormal lung microvascular function. Ann. NY Acad. Sci. 384: 417-434, 1982[Abstract]. |
| 13. |
Harris, T. R.,
and
R. J. Roselli.
A theoretical model of protein, fluid, and small molecule transport in the lung.
J. Appl. Physiol.
50:
1-14,
1981 |
| 14. |
Haselton, F. R.,
R. E. Parker,
R. J. Roselli,
and
T. R. Harris.
Analysis of lung multiple indicator data with an effective diffusivity model of capillary exchange.
J. Appl. Physiol.
57:
98-109,
1984 |
| 15. | Haselton, F. R., R. J. Roselli, and T. R. Harris. An effective-diffusivity model of pulmonary capillary exchange: general theory, limiting cases, and sensitivity. Math. Biosci. 70: 237-263, 1984. |
| 16. |
Hofman, W. F.,
and
I. C. Ehrhart.
Vascular reactivity and permeability to serotonin in cyclooxygenase-inhibited dog lung.
Am. J. Physiol.
255 ((Heart Circ. Physiol. 24):
H1096-H1105,
1988 |
| 17. |
Isago, T.,
L. D. Traber,
D. N. Herndon,
S. Abdi,
K. Fujioka,
and
D. L. Traber.
Determination of pulmonary microvascular reflection coefficient in sheep by venous occlusion.
J. Appl. Physiol.
69:
2311-2316,
1990 |
| 18. | Klaesner, J. W., N. A. Pou, R. L. Galloway, and R. J. Roselli. Laser system for measuring small changes in plasma tracer concentration. Biomed. Instrum. Technol. 30: 507-516, 1996[Medline]. |
| 19. | Klaesner, J. W., R. J. Roselli, S. Evans, N. A. Pou, R. E. Parker, G. Tack, and M. Parham. Optical measurements of lung microvascular filtration coefficient using polysulfone fibers. Ann. Biomed. Eng. 22: 660-673, 1994[Medline]. |
| 20. |
Lee, J. S.,
and
L. P. Lee.
Effect of hemodilution on ventilatory fluctuations of pulmonary capillary blood volume.
J. Appl. Physiol.
65:
2571-2578,
1988 |
| 21. |
Maron, M. B.,
and
S. M. Lane.
Time course of blood volume changes in an isolated lung lobe after venous pressure elevation.
J. Appl. Physiol.
77:
2720-2726,
1994 |
| 22. |
Oppenheimer, L.,
W. N. Richardson,
D. Bilan,
and
M. Hoppensack.
Colorimetric device for measurement of transvascular fluid flux in blood-perfused organs.
J. Appl. Physiol.
62:
364-372,
1987 |
| 23. |
Oppenheimer, L.,
H. W. Unruh,
C. Skoog,
and
H. S. Goldberg.
Transvascular fluid flux measured from intravascular water concentration changes.
J. Appl. Physiol.
54:
64-72,
1983 |
| 24. |
Parker, R. E.,
R. J. Roselli,
and
K. L. Brigham.
Effects of prolonged elevated microvascular pressure on lung fluid balance in sheep.
J. Appl. Physiol.
58:
869-875,
1985 |
| 25. |
Peterson, B. T.,
J. A. Brooks,
and
A. G. Zack.
Use of microwave oven for determination of postmortem water volume of lungs.
J. Appl. Physiol.
52:
1661-1663,
1982 |
| 26. |
Richardson, W. N.,
D. Bilan,
M. Hoppensack,
and
L. Oppenheimer.
Fast-phase transvascular fluid flux and the Fahraeus effect.
J. Appl. Physiol.
62:
1513-1520,
1987 |
| 27. | Roselli, R. J., S. R. Coy, and T. R. Harris. Models of lung transvascular fluid and protein transport. Ann. Biomed. Eng. 15: 127-138, 1987[Medline]. |
| 28. | Roselli, R. J., and T. R. Harris. Lung fluid and macromolecular transport. In: Respiratory Physiology: An Analytical Approach, edited by H. K. Chang, and M. Paiva. New York: Dekker, 1989, p. p.633-735. (Lung Biol. Health Dis. Ser.) |
| 29. | Roselli, R. J., T. R. Harris, and N. A. Pou. Effects of perfusion on urea extraction in zone 3 sheep lungs (Abstract). Ann. Biomed. Eng. 23: S13, 1987. |
| 30. | Roselli, R. J., R. E. Parker, and T. R. Harris. Comparison between pore model predictions and sheep lung fluid and protein transport. Microvasc. Res. 29: 320-339, 1985[Medline]. |
| 31. | Thomas, C. L. Taber's Cyclopedic Medical Dictionary. Philadelphia, PA: Davis, 1993. |
This article has been cited by other articles:
|
|
 |
|
  D. B. Pearse, P. M. Becker, and S. Permutt Effect of changing vascular volume on measurement of protein reflection coefficient in ischemic lungs Am J Physiol Heart Circ Physiol, February 1, 2001; 280(2): H918 - H924. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Klaesner, N. A. Pou, R. E. Parker, C. Finney, and R. J. Roselli Optical measurement of isolated canine lung filtration coefficients after alloxan infusion J Appl Physiol, April 1, 1998; 84(4): 1381 - 1387. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||
Palv,
cmH2O
