Vol. 87, Issue 6, 2381-2385, December 1999
SPECIAL COMMUNICATION
Validation of fluorescent-labeled microspheres for measurement of
relative blood flow in severely injured lungs
Matthias
Hübler1,
Jennifer E.
Souders2,4,
Erin D.
Shade3,
Michael P.
Hlastala3,4,
Nayak L.
Polissar5, and
Robb W.
Glenny3,4
1 Department of Anesthesiology and Intensive
Care, Carl Gustav Carus University Hospital, 01307 Dresden,
Germany; Departments of 2 Anesthesiology,
3 Physiology and Biophysics, and
4 Medicine, University of Washington School of
Medicine, Seattle, Washington 98195; and
5 The Mountain-Whisper-Light Statistical
Consulting, Seattle, Washington 98112
 |
ABSTRACT |
The aim of the study was to
validate a nonradioactive method for relative blood flow measurements
in severely injured lungs that avoids labor-intensive tissue
processing. The use of fluorescent-labeled microspheres was compared
with the standard radiolabeled-microsphere method. In seven sheep, lung
injury was established by using oleic acid. Five pairs of radio- and
fluorescent-labeled microspheres were injected before and after
established lung injury. Across all animals, 175 pieces were selected
randomly. The radioactivity of each piece was determined by using a
scintillation counter. The fluorescent dye was extracted from each
piece with a solvent without digestion or filtering. The fluorescence
was determined with an automated fluorescent spectrophotometer.
Perfusion was calculated for each piece from both the radioactivity and
fluorescence and volume normalized. Correlations between flow
determined by the two methods were in the range from 0.987 ± 0.007 (SD) to 0.991 ± 0.002 (SD) after 9 days of soaking. Thus the
fluorescent microsphere technique is a valuable tool for investigating
regional perfusion in severely injured lungs and can replace radioactivity.
oleic acid; radiolabeled microspheres; acute respiratory distress
syndrome model
| |
INTRODUCTION |
The use of fluorescent-labeled microspheres for measurement of
regional lung perfusion is a well-established method (7). When the
number of injected microspheres is sufficiently large, and when
appropriately sized microspheres are used, regional blood flow is
proportional to the number of trapped microspheres (2). Our laboratory
(7) showed that the extraction of the fluorescent dye can be simplified
with the use of an automated spectrophotometer. We further simplified
the tissue processing by avoiding digestion and filtration of the lung
pieces, while still yielding excellent correlations. However,
retrieving fluorescent microspheres from solid organs requires many
labor-intensive steps. Radiolabeled microspheres are easier to use but
pose health risks, require special precautions for use and disposal,
have limited shelf lives, and are relatively expensive. There are few
reports about the use of nonradiolabeled microspheres in injured lungs
(11). Because severely injured lung tissue is more like solid organs
than normal lung tissue, digestion and filtration of lung pieces may be necessary.
The primary purpose of this study was to validate the use of
fluorescent-labeled microspheres for estimating organ perfusion in
severely injured lungs. Fluorescent dye was extracted from dried lung
pieces, thus avoiding the time-consuming tissue digestion and
filtration procedures.
| |
METHODS |
The study was approved by the University of Washington Animal Care Committee.
Animal preparation.
The experiments were performed on seven adult sheep of both genders,
ranging in weight from 22.6 to 27.3 kg. The animals were premedicated
with 0.6 mg/kg xylazine (Xyla-Ject, Phoenix Pharmaceutical, St. Joseph,
MO), anesthetized with thiopental sodium (20 mg/kg), intubated, and
ventilated with a servo ventilator 900 C (Siemens, Solna, Sweden).
Anesthesia was maintained by using a continuous infusion of thiopental
sodium titrated to suppress hemodynamic responses to noxious stimuli.
Once anesthetic depth was adequate, the animals were paralyzed with
pancuronium (0.1 mg/kg). A tracheotomy was performed, and a tube was
inserted. Catheters were placed in a femoral and a pulmonary artery
(Swan-Ganz thermodilution catheter) to monitor arterial, pulmonary
arterial, and wedge pressures and temperature and to sample blood. A
femoral venous catheter was inserted for infusion of anesthetic drugs
and maintenance fluids. A central venous catheter was inserted into the
left subclavian vein via the left jugular external vein and used as the
injection port for the oleic acid (OA) and the microspheres. Airway
pressures, arterial and pulmonary arterial, were measured continuously
with Validyne amplifiers (Northridge, CA) and recorded on a Western Graphtec Mark12 data-management system DMS 1000 (Irvine, CA). The
end-tidal CO2 was measured with a mass spectrometer
(Perkin-Elmer medical gas analyzer MGA-1100). Cardiac outputs and blood
temperatures were measured with a cardiac output computer (Baxter
Edwards Sat-2). Arterial and venous pH, PO2,
and PCO2 were measured with a blood-gas analyzer (Radiometer ABL 330, Acid Base Laboratory, Copenhagen, Denmark) and corrected for temperature. Throughout the experiments all
animals were prone. After an initial stabilization period during which
the respiratory minute volume was adjusted to maintain arterial
PCO2 between 36 and 46 Torr, baseline
measurements were recorded. Lung injury was induced by emulsifying 15 ml of previously extracted blood with 0.14-0.23 ml OA/kg
(C18H34O2, Sigma Chemical, St.
Louis, MO) and injecting it with continuous shaking over 30-45 min. Severe lung injury was considered established when the clinical criteria of the definition of the American/European consensus conference on acute respiratory distress syndrome (ARDS) were fulfilled
(ratio of arterial PO2 to inspiratory
O2 fraction < 200, pulmonary capillary wedge
pressure < 19 Torr) (3). The animals were then studied for 120 min
after the injury was established.
Microsphere validation.
Details of the methods are described by Glenny et al. (7). Briefly,
five fluorescent polystyrene microspheres (red, orange, blue-green,
yellow-green, and crimson) of 15 µm diameter (FluoSpheres, Molecular
Probes, Eugene, OR) and five radiolabeled styrene-divinyl benzene
microspheres (141Ce, 113Sn,
103Ru, 95Nb, and 46Sc) of 15.5 µm
diameter (DuPont NEN Research Products, Boston, MA) were used. The time
points of injection were baseline (tbase), established injury (t0), 30 min
(t30), 60 min (t60), and
120 min (t120) after established injury.
Immediately before injection, the microspheres were vortexed, sonicated
for 90 s, and combined in a single syringe. Fluorescent
(1.5 × 106) and radiolabeled microspheres
[1.63 × 106 ± 0.6 × 106 (SD)]
were injected simultaneously over 30 s, followed by a saline flush. The
amount of radiolabeled microspheres given was adjusted to ensure a
minimal radioactivity of 0.02 mCi per injection. The order of
microsphere injections was randomized in each experiment.
After completion of the study, the animals were given 1,000 U/kg of
heparin (Elkins-Sinn, Cherry Hill, NJ) and 3 mg/kg of papaverine
hydrochloride (American Regent Laboratories, Shirley, NY), then
exsanguinated. A midline sternotomy was performed, and catheters were
placed in the main pulmonary artery and the left atrium. The lungs were
flushed with 50 ml/kg of dextran 2% (Sigma Chemical), removed,
inflated to 30 cmH2O, and dried with warm air for 7 days.
When dry, the lungs were first coated with a one-component polyurethane
foam (Kwik Foam, DAP, Dayton, OH), then suspended vertically in a
square box and embedded in rapidly setting urethan foam (2 lb. polyol
and isocyanate, International Sales, Seattle, WA). The foam block was
cut into slices of ~1.2 cm in thickness. With the use of a
12-mm-diameter core, the slices were sampled systematically. Cores were
obtained in a rigid X-Y grid system, with 2 cm between the
centers of adjacent cores. The height of every core was measured by
using a micrometer, and the volume was calculated. Samples
[194 ± 39 (SD)] were obtained from each sheep lung after samples
with airways occupying >50% of the core's volume were discarded.
Twenty-five lung samples were randomly selected from each animal, and
the radiolabel-determined and the fluorescent-determined perfusion was
obtained. The average volume of the samples was 1.38 ± 0.2 cm3, and the average weight was 0.069 ± 0.02. The
radioactivity was measured by using a 3 × 3.25-in. sodium well
crystal gamma counter (Minaxi gamma counting system, model 5550, Packard, Downers Grove, IL). Correction for decay time, background
counts, and spillover was performed with the matrix inversion method
(10). Each tissue piece was counted long enough to ensure a counting
error <1%. The samples were then individually soaked for 14 days in
3 ml of 2-ethoxyethyl acetate (Cellosolve, Aldrich Chemical,
Milwaukee, WI). The fluorescence was read at days 7, 9, 11, and
14 in a luminescence spectrophotometer (Perkin-Elmer LS-50B,
Beaconsfield, Buckinghamshire, UK) fitted with a flow cell and a
red-sensitive photomultiplier tube.
The volume-normalized relative blood flow at every time point was
calculated for each lung piece, with the color-specific fluorescence
and the radiolabel-specific counting performed separately
where
rel,i is the volume-normalized relative
blood flow of the piece i; xi is the
obtained fluorescence or counted radioactivity, respectively, divided
by the volume of the piece (cm3); and n is the
number of pieces. The mean normalized relative flow was therefore 1.0.
Statistics.
The Pearson correlations between the radio- and fluorescent-labeled
microsphere relative blood flows were calculated for each animal and
each color separately. Correlations were calculated for each color at
each soaking time. The mean correlations were compared among colors for
a given soaking time and among soaking times for a given color by using
the two-tailed paired t-test. A one-sample two-tailed
t-test was used to compare correlations to a hypothesized mean
of zero. P < 0.05 was considered statistically significant.
| |
RESULTS |
The criteria for ARDS were attained after a mean time of 135 ± 65 min
after the injection of OA. One animal died 10 min before completion of
the study period, and the last injection of microspheres (crimson)
could not be given. The total number of coring samples was therefore
150 for crimson and 175 for all other colors.
The mean correlations between the volume-normalized relative blood
flows determined by the radiolabeled and the fluorescent microspheres
technique were very high and increased only slightly by prolonging the
soaking time (Table 1). Statistically
significant differences were seen for orange, comparing the 7- and
9-day soaking times, and between orange and blue-green after 7 days of
soaking. However, the magnitudes of the differences are extremely
minor. Figure 1 shows a representative plot
of the radiolabeled-microsphere-determined flow vs. the
yellow-green-microsphere-determined flow after 9 days of soaking. In
one animal, the correlation for crimson was slightly lower than for
other animals: 0.94 after 7 days of soaking, increased to 0.97 after 14 days of soaking. In this animal, the range of relative flow at the time
of injection was very small compared with other animals (typical range
0.6 to 1.5). The random measurement error was approximately the same as
for other animals, leading to a smaller signal-to-noise ratio and to a
slightly smaller correlation.
View this table:
[in this window]
[in a new window]
|
Table 1.
Mean correlations between radiolabeled vs.
fluorescent-microsphere-determined volume-normalized relative flow
|
|
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 1.
Radiolabeled microsphere-determined vs. yellow-green
microsphere-determined volume-normalized relative flow after 9 days of
soaking. No. of animals = 7, no. of pieces = 175.
|
|
The accuracy of the fluorescent method is independent of the sample
density, as shown by the small and nonsignificant mean correlations
between 1) the density of the coring samples and 2) a
difference defined by the radiolabeled-microsphere relative flow minus
the fluorescent-microsphere relative flow (Table
2). Figure 2
shows a representative plot of the density vs. the
radiolabeled-microsphere-determined minus the
yellow-green-microsphere-determined flow after 9 days of soaking,
revealing no trends in the difference over the range of the density.
View this table:
[in this window]
[in a new window]
|
Table 2.
Mean correlation of radiolabeled minus fluorescent
microsphere-determined relative flow with magnitude of density after 9 days of soaking
|
|
View larger version (39K):
[in this window]
[in a new window]
|
Fig. 2.
Radiolabeled-microsphere relative flow minus yellow-green- microsphere
relative flow vs. density of core samples. No. of animals = 7, no.
of pieces = 175. Note the small range on y-axis compared with
the total range of normalized blood flow (see Fig. 1).
|
|
The accuracy of the fluorescent method is also relatively unaffected by
the magnitude of flow. Except for blue-green, all the mean correlations
between the fluorescence-radioactivity difference with magnitude of
flow after 9 days of soaking are nonsignificant and close to zero
(Table 3). Bland-Altman plots (5) of
fluorescence-radioactive difference as a function of flow after 9 days
of soaking show only a slight trend toward greater variability at
higher values of relative flow, which is typical of count data. In any
case, there is a tight distribution about the mean. Figure
3 shows a representative plot for
yellow-green.
View this table:
[in this window]
[in a new window]
|
Table 3.
Mean correlation of radiolabeled minus fluorescent
microsphere-determined relative flow vs. relative flow after 9 days of
soaking
|
|
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 3.
Radiolabeled-microsphere minus yellow-green microsphere-relative flow
vs. relative flow determined by radiolabeled microsphere method. No. of
animals = 7, no. of pieces equals 175. Note small range on
y-axis compared with the total range of normalized blood flow
on the x-axis.
|
|
| |
DISCUSSION |
The important finding of this study is that the fluorescent-labeled
microspheres technique is an accurate method to measure volume-normalized relative blood flows in severely injured lungs. The
linear correlation between perfusion estimates by use of fluorescent and radiolabeled microspheres was excellent. The five chosen colors were easily separable, because each color has a unique and narrow excitation spectrum (7).
The correlation between the two different simultaneously injected
microspheres (radioactive and fluorescent) to measure regional perfusion was not perfect. This indicates that there is a measurable, but small, error. The reasons for these errors are well known (9).
Several studies have quantified the sources of error for radiolabeled
microspheres (1, 6, 13). Glenny et al. (7) suggested that these errors
apply also to the fluorescent microsphere technique. They concluded
also that the error increases if, in the processing of the lung pieces,
there are too many steps involved, leading to a loss of microspheres.
They found the best correlations when the samples were only soaked.
Therefore, we decided to extract the fluorescent dyes only by
increasing the soaking time. We expected a longer soaking time because
of the higher density of the tissue compared to noninjured lungs.
Noninjured lung samples are usually soaked for 2 days (7). Although the
correlations seen after 7 days of soaking were already high, a minor
improvement was noted for all colors, and was significant for orange,
by increasing the soaking time to 9 days. Further increase in duration
of the soaking time showed only a negligible improvement of the correlations.
In most studies investigating pulmonary blood flow, lung pieces with an
airway content >25% are discarded, and the relative flow is weight
normalized (4, 8, 12). In severely injured lungs, a greater weight
range for pieces of the same volume is noted. This is usually due to
cellular infiltration or hemorrhage. We volume normalized the relative
blood flows to compensate for this variability. We also included pieces
up to an airway content of 50% and showed that the accuracy of the
method is independent of the density of the lung sample. The
correlation with the magnitude of density was tested after a soaking
time of 9 days because of the results mentioned above.
One important concern when investigating blood flow is that the
accuracy of the method should be independent of the magnitude of flow
for an individual piece. Whereas the mean correlation for blue-green is
significantly different from zero (Table 3), the significance level is
not small and could readily have arisen by chance when five hypothesis
tests are carried out. In addition, the mean correlation itself is
weak. Finally, even if one accepts a real relationship between the
fluorescent-isotope difference and the magnitude of flow, the impact is
likely to be negligible. The equation for the difference vs. flow
indicates a very minor correction: blue-green 
isotope = 0.033 + 0.033 × (relative flow) (Fig. 4), based
on a regression analysis pooling n = 175 pieces from seven
animals. Across a typical relative flow range of only 0-2.5, the
blue-green relative flow would differ from isotope relative flow by
0.03 to +0.05, an extremely minor correction. The magnitude of this
correction is 1-2% of the typical flow range and could be ignored
in practice. Even with the use of the equation from the animal
with the greatest slope [blue-green isotope = 0.072 + 0.072 × (relative flow)], the correction for relative flows from 0 to 2.5 would range from 0.07 to +0.11, again very minor compared with the
wide range of flows encountered.
View larger version (39K):
[in this window]
[in a new window]
|
Fig. 4.
Radiolabeled-microsphere minus blue-green-microsphere relative flow vs.
relative flow determined by radiolabeled microsphere method. No. of
animals = 7, no. of pieces = 175. Note small range on y-axis
compared with the total range of normalized blood flow on
x-axis.
|
|
In conclusion, we have shown that the fluorescent microsphere technique
is a valuable tool for investigating regional perfusion in severely
injured lungs and can replace radioactive microspheres. The optimal
duration of soaking seems to be 9 days, and, if only four colors are
used, blue-green can be omitted.
| |
ACKNOWLEDGEMENTS |
The authors thank Susan Bernard, Carmel Schimmel, Thien Tran, and
Dowon An for expert technical help.
| |
FOOTNOTES |
This work was supported by Deutsche Forschungsgemeinschaft Grant HU
818/1-1 and by National Heart, Lung, and Blood Institute Grants
HL-12174 and HL-24163.
Address for reprint requests and other correspondence: M. P. Hlastala,
Pulmonary and Critical Care Medicine, Box 356522, Univ. of Washington,
Seattle, WA 98195-6522 (E-mail: hlastala{at}u.washington.edu).
Received 11 June 1999; accepted in final form 17 August 1999.
| |
REFERENCES |
1.
Austin, R. E.,
W. W. Hauck,
G. S. Alsea,
A. E. Flynn,
D. L. Coggins,
and
J. I. E. Hoffman.
Quantitating error in blood flow measurements with radioactive microspheres.
Am. J. Physiol.
257 (Heart Circ. Physiol. 26):
H280-H288,
1989[Abstract/Free Full Text].
2.
Bassingthwaighte, J. B.,
M. A. Malone,
T. C. Moffett,
R. B. King,
S. E. Little,
J. M. Link,
and
K. A. Krohn.
Validity of microsphere deposition for regional myocardial flows.
Am. J. Physiol.
253 (Heart Circ. Physiol. 22):
H184-H193,
1987[Abstract/Free Full Text].
3.
Bernard, G. R.,
A. Arigas,
K. L. Brigham,
J. Carlet,
K. Falke,
L. Hudson,
M. Lamy,
J. R. Legall,
A. Morris,
and
R. Spragg.
Report of the American-European Consensus Conference of ARDS: definitions, mechanisms, relevant outcomes, and clinical trials coordination.
Am. J. Respir. Care. Med.
194:
818-824,
1994.
4.
Bernard, S. L.,
R. W. Glenny,
H. H. Erickson,
M. R. Fedde,
N. Polissar,
R. J. Basaraba,
and
M. P. Hlastala.
Minimal redistribution of pulmonary blood flow with exercise in racehorses.
J. Appl. Physiol.
81:
1062-1070,
1996[Abstract/Free Full Text].
5.
Bland, J. M.,
and
D. G. Altman.
Statistical methods for assessing agreement between two methods of clinical measurement.
Lancet
1:
307-310,
1986[Medline].
6.
Buckberg, G. D.,
J. C. Luck,
D. B. Payne,
J. I. E. Hoffmann,
J. P. Archie,
and
D. E. Fixler.
Some sources of error in measuring regional blood flow with radioactive microspheres.
J. Appl. Physiol.
31:
598-604,
1986[Free Full Text].
7.
Glenny, R. W.,
S. Bernard,
and
M. Brinkley.
Validation of fluorescent-labeled microspheres for measurement of regional organ perfusion.
J. Appl. Physiol.
74:
2585-2597,
1993[Abstract/Free Full Text].
8.
Mure, M.,
K. B. Domino,
H. T. Robertson,
M. P. Hlastala,
and
R. W. Glenny.
Pulmonary blood flow does not redistribute in dogs with reposition from supine to left lateral position.
Anesthesiology
89:
483-492,
1998[Medline].
9.
Prinzen, F. W.,
and
R. W. Glenny.
Developments in non-radioactive microsphere techniques for blood flow measurement.
Cardiovasc. Res.
28:
1467-1475,
1994[Medline].
10.
Schosser, R.,
K. E. Arfors,
and
K. Messmer.
MIC-II
a program for the determination of cardiac output, arterio-venous shunt and regional blood flow using the radioactive microsphere method.
Comp. Prog. Biomed. Res.
9:
19-38,
1979.
11.
Segerer, H.,
W. van Gelder,
F. W. M. Angenent,
L. J. P. M. van Woerkens,
T. Curstedt,
M. Obladen,
and
B. Lachmann.
Pulmonary distribution and efficacy of exogenous surfactant in lung-lavaged rabbits are influenced by the instillation technique.
Pediatr. Res.
34:
490-494,
1993[Medline].
12.
Walther, S. M.,
K. B. Domino,
R. W. Glenny,
N. L. Polissar,
and
M. P. Hlastala.
Pulmonary blood flow distribution has a hilar-to-peripheral gradient in awake, prone sheep.
J. Appl. Physiol.
82:
678-685,
1997[Abstract/Free Full Text].
13.
Zwissler, B.,
R. Schosser,
C. Weiss,
V. Iber,
M. Weiss,
C. Schwicker,
P. Spengler,
and
K. Messmer.
Methodologic error and spatial variability of organ flow measurements using radiolabeled microspheres.
Res. Exp. Med.
191:
47-63,
1991[Medline].
J APPL PHYSIOL 87(6):2381-2385
8570-7587/99 $5.00
Copyright © 1999 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
