Vol. 280, Issue 4, H1802-H1806, April 2001
Carbon monoxide wash-in method to determine gas transfer in
vascular beds: application to rat hindlimb
Carolyn S. L.
Cho1,
Allan J.
McLean1,2,3,6,
Laurent P.
Rivory4,
Paul A.
Gatenby1,2,6,
David T. A.
Hardman5, and
David G.
Le
Couteur1,2,3,4,6
1 Canberra Clinical School and Departments of
2 Medicine, 3 Physiology, and 4 Pharmacology,
University of Sydney, New South Wales 2006; 5 Department of
Vascular Surgery, Canberra Hospital, Garran 2605; and
6 John Curtin School of Medical Research, Australian
National University, Australian Capital Territory 0200, Australia
 |
ABSTRACT |
The vascular
barrier to gas transfer is an important physiological parameter;
however, no readily applicable technique exists to quantitate the
process. A simple technique to measure the permeability-surface area
(PS) product for gas transfer in vascular beds is
proposed using wash in of carbon monoxide (CO) and Crone-Renkin
analysis. Wash-in experiments were performed on the perfused
hindlimbs of male Wistar rats (n = 15) by using CO as a
surrogate marker for oxygen and technetium-99m-labeled albumin as the
vascular marker. The use of CO and erythrocyte-free perfusate and the
collection of outflow samples into tubes preloaded with erythrocytes
obviated the need for an anaerobic collection device or consideration
of Hb binding in the analysis. The PS product for CO was
determined from the early extraction as 0.013 ± 0.006 ml · s
1 · g1. Compartmental
analysis revealed that the fractional recovery of CO was 0.45 ± 0.14 and the volume of distribution was 2.31 ± 0.76 ml/g. This
technique detected a small measurable barrier to the transfer of CO
across the hindlimb vasculature and is potentially applicable to other
vascular beds in health and disease.
perfused; permeability; surface area
| |
INTRODUCTION |
THE
VASCULAR BARRIER to gas transfer is an important physiological
parameter; however, no readily applicable technique exists to
quantitate the process. Changes in the barrier to oxygen diffusion have
been described in interstitial lung disease (12) and have been shown to contribute to functional change in liver disease (8, 14). Hypoxia may also contribute to tissue injury in vascular disease. For example, in chronic venous hypertension, the
development of a pericapillary fibrin cuff is thought to impair gas
diffusion and cause tissue hypoxia (1, 11). Other diseases such as atherosclerosis and diabetes mellitus have structural alterations in vessel walls that are also considered likely to impose a
significant barrier to oxygen transfer. However, direct measurement of
the capillary barrier to gas transfer is technically complex and
requires the use of [15O]oxygen and
[18O]oxygen in multiple indicator-dilution experiments
and elaborate anaerobic collection devices (5, 13, 19).
These difficulties precluded quantification of the barrier in either
normal or pathological hindlimb vasculature. Here we describe a novel
technique for the measurement of the vascular barrier to gas transfer
in the hindlimb vasculature and report for the first time an estimate
of the permeability-surface area (PS) product for gas
transfer across the normal hindlimb vasculature.
| |
MATERIALS AND METHODS |
Animals.
Male Wistar rats (age 8-16 wk) were obtained from the John Curtin
School of Medical Research. Approval for the study was obtained from
the Australian National University Animal Experimentation Ethics Committee.
Preparation of control and test perfusates.
The control perfusate consisted of erythrocyte-free Krebs-Henseleit
bicarbonate buffer containing 4.7% BSA and 10 mM glucose equilibrated
with 95% O2-5% CO2 (Linde Gas). The test
perfusate was equilibrated with 95% CO-5% CO2 (Linde Gas)
and contained tracer quantities of [14C]sucrose (specific
activity 10.1 Ci/mmol, ICN Pharmaceuticals), [3H]water
(specific activity 100 mCi/mmol, Amersham Life Science), and
99mTc-labeled albumin (99mTc-albumin).
Isolated hindlimb perfusion.
The method of isolated hindlimb perfusion has been described
previously (20). The rats were anesthetized by an
injection of 60 mg/kg ip pentobarbitone sodium (Rhone Merieux).
Hindlimb perfusion was administered with a peristaltic pump (Extech
Equipment) via the arterial cannula at 4 ml/min in a nonrecirculating
mode. Viability was assessed by macroscopic appearance, arterial
pressure, oxygen consumption, and assays of the outflow samples for
lactate dehydrogenase, creatine kinase, and potassium.
Wash-in experiments.
After the hindlimb preparation was equilibrated with the control
perfusate for 30 min and viability was confirmed, the control perfusate
was switched to the test perfusate using a four-way valve (Cole
Parmer). This switching did not cause any fluctuations in pressure.
Outflow samples were then collected at 5-, 10-, and 30-s intervals up
to 600 s. These outflow samples were collected into Eppendorf
tubes that had been preloaded with 100-µl quantities of washed human
erythrocytes and were stored on ice; thus Hb present in the collection
tubes bound the CO present in the outflow perfusate.
Analyses of carboxyhemoglobin concentration and hematocrit were
performed using a blood gas analyzer (model 865, Chiron Diagnostics). 99mTc activity was measured using a gamma scintillation
counter (Cobra II Autogamma, Canberra Packard). After a 2-day delay to
allow the 99mTc activity to decay to negligible levels,
[14C]sucrose and [3H]water were measured in
a beta scintillation counter (model 1900, Canberra Packard) after the
addition of scintillation fluid (Starscint, Packard Instrument).
Determination of the PS product for CO.
The outflow concentrations were expressed as fractions of the inflow
dose per milliliter. The PS product was calculated according to the early extraction method of Crone and Renkin (3, 4, 18). The early extraction (E) of CO was calculated
using the formula
where Cref is the fractional outflow concentration
of the vascular marker and CCO is the fractional outflow
concentration of CO at each time point. Extraction was calculated using
both [14C]sucrose and 99mTc-albumin as
reference markers. E of CO was obtained from the plateau of
the extraction-time curve and was used to calculate the PS
product for CO across the capillary according to the equation
where Q is the flow rate of the perfusate.
Pharmacokinetic determination of volume of distribution and
fractional recovery.
A compartmental model was used to determine the volume of distribution
and recovery of each indicator as described previously (16). The outflow curves were fitted to the equation
where Ct is the concentration at time
t, R is the fractional recovery of the indicator,
t0 is the transit time within the catheters and
nonexchanging blood vessels, and k is the rate constant of
elimination, which is equal to the flow rate divided by the volume of
distribution for nonextracted indicators. For CO, which was found to
have significant extraction, the volume of distribution is equal to the
flow rate divided by the product of R and k. The
curve fitting was performed using SigmaPlot 4.01 (SPSS; Chicago, IL).
The linear superposition principle was also applied to the data. Each
time point for sucrose, water, and CO was multiplied by the ratio of
the volume of distribution of albumin to the volume of distribution of
the indicator. Likewise, each concentration point was multiplied by the
ratio of the recovery of albumin to the recovery of the indicator.
Superposition onto the albumin curve is consistent with flow-limited
distribution, whereas failure of superposition suggests the presence of
a permeability barrier (7, 16).
| |
RESULTS |
Representative data together with fitted curves from a single
hindlimb study are shown in Fig. 1. The
outflow curve for albumin appears earliest, followed by sucrose and
then the water curve. The outflow curve for CO appears last. The
sucrose and albumin curves reached a plateau; however, the CO and water
curves were still rising after the 10-min collection period.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1.
Outflow curves from a wash-in experiment in the perfused
rat hindlimb. Data have been fitted using a compartmental model (solid
lines). 99mTc-labeled albumin ( ),
[14C]sucrose ( ), [3H]water
( ), and CO ( ) are shown.
|
|
Figure 2 shows the relationship between
early extraction of CO and time for the experiment shown in Fig. 1.
This was similar with either 99mTc-albumin or
[14C]sucrose as the reference marker. With
99mTc-albumin as the reference marker, the plateau of the
extraction was 0.95 ± 0.04 at 24 ± 14 s
(n = 6), and with [14C]sucrose it was
0.94 ± 0.07 at 27 ± 13 s (n = 15).
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 2.
Relationship between time and early extraction of CO for
the experiment shown in Fig. 1. Reference marker is
99mTc-labeled albumin () or
[14C]sucrose ().
|
|
The PS product for CO, obtained using two different
reference markers, is shown in Table 1
and compared with values for oxygen obtained in other perfused
organs. With 99mTc-albumin as the vascular marker,
the PS product was 0.014 ± 0.005 ml · s1 · g1
(n = 6), and with [14C]sucrose as the
reference marker the PS product was 0.013 ± 0.006 ml · s1 · g1
(n = 15).
The derived values for the recovery and volume of distribution of
tracers are shown in Table 2. The fitted
maximum fractional concentration of CO at infinity was 0.45 ± 0.13 (n = 15), indicating that CO is sequestered within
the hindlimb.
View this table:
[in this window]
[in a new window]
|
Table 2.
Recovery and volume of distribution of CO, water, sucrose, and albumin
in perfused rat hindlimb using wash-in experiments and
compartmental analysis
|
|
After superposition, the CO curves were not perfectly superposed on the
albumin curves (Fig. 3), which is also
consistent with a small permeability barrier. In contrast, the sucrose
and water curves were superposed, which is consistent with flow-limited distribution.
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 3.
Application of the linear superposition principle to the
outflow curves shown in Fig. 1. Each time point for sucrose, water, and
CO was multiplied by the ratio of the volume of distribution of albumin
to the volume of distribution of the indicator, and each concentration
point was multiplied by the ratio of the recovery of albumin to the
recovery of the indicator. 99mTc-labeled albumin
(), [14C]sucrose (),
[3H]water (), and CO ()
are shown.
|
|
| |
DISCUSSION |
This study details the application of a novel and simple method
for the analysis of gas disposition in vascular beds. The technique
allowed the detection and estimation of the vascular barrier to gas
transfer in the perfused rat hindlimb.
CO is a recognized surrogate marker for the diffusional characteristics
of oxygen and has been widely used for this purpose in the lung
(12) and recently in the liver (16). The use
of CO in these experiments resulted in both procedural simplicity and
technical advantages over oxygen. CO avidly binds Hb; therefore, the
use of erythrocyte-free perfusate and collection of outflow samples
into tubes preloaded with erythrocytes obviated the need for
complicated anaerobic collection devices. The outflow samples were
analyzed for carboxyhemoglobin using a simple blood gas analyzer. This
is simpler than techniques required for the measurement of labeled
oxygen such as positron-detecting scintillators (5) or gas
chromatograph mass spectrometers (13, 19). In addition, the effect of Hb binding on gas transfer can be discounted when erythrocyte-free perfusate is used.
CO was found to be sequestered in the hindlimb. This greatly improves
the reliability of the Crone-Renkin method of analysis because it means
that the transfer of CO between blood and tissue is more likely to be
unidirectional at early time points. This important characteristic
improves the estimation of the PS product (3, 4,
18). Nevertheless, the early extraction of CO was high, and
because the PS product increases exponentially as
E approaches unity, the exact estimation of the
PS product becomes more prone to error. Indeed, the optimal
extraction of a solute has been suggested to be between 0.3 and 0.8 (3, 18). In the case of CO, the estimation of the
PS for gas transfer will improve in disease states that
result in impeded transfer. An additional problem with the Crone-Renkin
method is the contribution of rapidly flowing capillaries within the
heterogeneous vascular bed. This tends to result in an underestimation
of the extraction fraction at early times. This can be partly remedied
by using an extraction fraction calculated from partial integrals of
the outflow curve. This approach yielded essentially identical results when applied to the data presented here. Transit-time heterogeneity therefore did not appear to be a problem in our study but could become
important in disease states.
The values we obtained for the PS of CO were 0.014 ± 0.005 ml · s1 · g1 when
albumin was the vascular reference and 0.013 ± 0.005 ml · s1 · g1 when sucrose
was the vascular marker. Although the two PS values are
similar, albumin is the most suitable vascular marker because sucrose
enters the extravascular space to some extent (20). This
was confirmed by the observation that the volume of distribution of
sucrose was larger than that of albumin (0.22 ± 0.06 vs.
0.17 ± 0.02 ml/g, respectively). The PS product for
oxygen across capillaries has been reported to be 0.007 ml · s1 · g1 in the
perfused rabbit heart (5) and 1.17 and 2.3 ml · s1 · g1 in the in vivo
canine brain (13) and heart (19),
respectively. In contrast, the PS product for the passage of
CO across the hepatocyte cell membrane was 0.21 ± 0.11 ml · s1 · g1, which is
15-fold greater than the value we found for the hindlimb capillary
(16).
In Table 1, the derived values for permeability that were calculated
using literature values for capillary surface areas are shown. Our
value for the permeability of CO in the rat hindlimb is ~20 × 105 cm/s. Derived values for the permeability of oxygen,
calculated from reported PS products, range from 1 × 105 cm/s in the rabbit heart (5) to 780 × 105 cm/s in the dog brain (13). By
comparison, in various tissues, the permeability of water is
20-150 × 105 cm/s and the permeability of
sodium is 2.5-10 × 105 cm/s (4).
Our estimate for the permeability of CO in the perfused rat hindlimb is
greater than the range quoted for sodium and at the lower limit of the
range for water accordingly is physiologically plausible. The wide
range of values obtained for the permeability of gases in other systems
possibly reflects the diverse techniques, the complex nature of the
mathematical analyses, and the confounding effects of Hb binding.
Although the major aim of this study was to determine the PS
for CO, the data were also analyzed to determine the volume of distribution and recoveries of the indicators. We used a simple compartmental model that fitted the data well (Fig. 1). The rationale for the use of this type of model for the analysis of wash-in experiments has been described previously (16). The
fractional recovery of CO was 0.45 ± 0.13. This is most likely
secondary to the binding of CO to myoglobin. In rat myocardium, binding of CO to myoglobin has been demonstrated using NMR studies
(6).
The application of wash-in experiments in the hindlimb is novel.
Previously, Heatherington and Rowland (10) performed bolus and constant infusion experiments using the isolated rat hindlimb. However, they simply used wash-in experiments to confirm that steady-state equilibrium had been achieved before sampling the hindlimb
tissues to determine the volumes of distribution. In Table
3, the values obtained with the wash-in
method for the volumes of distribution of albumin, sucrose, and water
using this approach are compared with those found using other methods.
The results are similar to those in which constant infusion and bolus methods were used in the perfused limb. Although we used the wash-in method because of technical difficulties associated with using radiolabeled gases in bolus experiments, the results suggest that it
may be useful for other physiological studies.
View this table:
[in this window]
[in a new window]
|
Table 3.
Volumes of distribution of red blood cells, albumin, sucrose, and water
in the rat hindlimb: comparison of results obtained in wash-in
experiments with other published values
|
|
If oxygen diffusional deficits are widely represented in pathological
states as has been shown for liver cirrhosis (8, 14, 17)
and interstitial lung disease (12) and has been inferred
in peripheral vascular disease (1, 11) and aging (15), then this technique would allow these processes to
be quantified. Substantial importance would then attach to the
availability of this technique. In the area of aging research, the
association of impaired oxygen extraction with normal CO permeability
would indicate impairment in oxygen utilization rather than gas
transfer and would point to the validity of the theories on aging
implicating mitochondrial functional deficits.
In conclusion, the CO wash-in technique has been shown to be a simple
method for determining gas disposition in the perfused rat hindlimb. It
has advantages over other studies including simplicity of the
experimental design and analysis. The technique is potentially applicable to the hindlimb in disease states and to other vascular beds
and could resolve basic issues of the mechanisms related to oxygen
utilization in disease states and aging.
| |
ACKNOWLEDGEMENTS |
The authors thank Professor Michael Roberts and Dr. Wu, Department
of Medicine, University of Queensland, for assistance in establishing
the perfused hindlimb preparation and gratefully acknowledge the
technical support of the Biochemistry Department of the Canberra Hospital.
| |
FOOTNOTES |
This study was supported by a University of Sydney Research Grant, the
National Health and Medical Research Council of Australia, and the
Private Practice Fund of Canberra Hospital, Canberra, Australia.
Address for reprint requests and other correspondence: A. J. McLean, Canberra Clinical School of the Univ. of Sydney, Faculty of
Medicine, Canberra Hospital, Yamba Dr., Garran, ACT 2605, Australia (E-mail: allan.mclean{at}act.gov.au or
allanmcl{at}hotmail.com).
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 26 July 2000; accepted in final form 4 November 2000.
| |
REFERENCES |
1.
Burnand, K,
Clemenson B,
Whimster I,
and
Browse N.
The effect of sustained venous hypertension on the skin capillaries of the canine hind limb.
Br J Surg
69:
41-44,
1982[Web of Science][Medline].
2.
Chihara, E,
Morimoto T,
Shigemi K,
Natsuyama T,
and
Hashimoto S.
Vascular viscoelasticity of perfused rat hindquarters.
Am J Physiol Heart Circ Physiol
260:
H1834-H1840,
1991[Abstract/Free Full Text].
3.
Crone, C.
The permeability of capillaries in various organs as determined by the use of the "indicator diffusion" method.
Acta Physiol Scand
58:
292-305,
1963[Web of Science][Medline].
4.
Crone, C,
and
Levitt DG.
Capillary permeability to small solutes.
In: Handbook of Physiology. The Cardiovascular System. Bethesda, MD: Am. Physiol. Soc, 1984, sect. 2, vol. IV, p. 411-466.
5.
Deussen, A,
and
Bassingthwaighte JB.
Modeling [15O]oxygen tracer data for estimating oxygen consumption.
Am J Physiol Heart Circ Physiol
270:
H1115-H1130,
1996[Abstract/Free Full Text].
6.
Glabe, A,
Chung Y,
Xu D,
and
Jue T.
Carbon monoxide inhibition of regulatory pathways in myocardium.
Am J Physiol Heart Circ Physiol
274:
H2143-H2151,
1998[Abstract/Free Full Text].
7.
Goresky, CA.
A linear method for determining liver sinusoidal and extravascular volumes.
Am J Physiol
204:
626-640,
1963.
8.
Harvey, PJ,
Gready JE,
Yin ZL,
Le Couteur DG,
and
McLean AJ.
Acute oxygen supplementation restores markers of hepatocyte energy status and hypoxia in cirrhotic rat livers.
J Pharmacol Exp Ther
293:
641-645,
2000[Abstract/Free Full Text].
9.
Heatherington, AC,
and
Rowland M.
Estimation of reference spaces in the perfused rat hind limb.
Eur J Pharm Sci
2:
261-270,
1994.
10.
Heatherington, AC,
and
Rowland M.
Discrepancies in pharmacokinetic parameter estimation between bolus and infusion studies in the perfused rat hind limb.
J Pharmacokinet Biopharm
23:
441-462,
1995[Web of Science][Medline].
11.
Hopkins, NF,
and
Jamieson CW.
Diffusion barriers in venous ulceration.
J R Soc Med
78:
355-357,
1985[Web of Science][Medline].
12.
Hughes, JM,
Lockwood DN,
Jones HA,
and
Clark RJ.
DLCO/Q and diffusion limitation at rest and on exercise in patients with interstitial fibrosis.
Respir Physiol
83:
155-166,
1991[Web of Science][Medline].
13.
Kassissia, IG,
Goresky CA,
Rose CP,
Schwab AJ,
Simard A,
Huet PM,
and
Bach GG.
Tracer oxygen distribution is barrier-limited in the cerebral microcirculation.
Circ Res
77:
1201-1211,
1995[Abstract/Free Full Text].
14.
Le Couteur, DG,
Hickey H,
Harvey P,
Gready J,
and
McLean AJ.
Hepatic artery flow and propranolol metabolism in the perfused cirrhotic rat liver.
J Pharmacol Exp Ther
289:
1553-1558,
1999[Abstract/Free Full Text].
15.
Le Couteur, DG,
and
McLean AJ.
The aging liver: drug clearance and an oxygen diffusion barrier hypothesis.
Clin Pharmacokinet
34:
359-373,
1998[Web of Science][Medline].
16.
Le Couteur, DG,
Yin ZL,
Rivory LP,
and
McLean AJ.
Carbon monoxide disposition in the perfused rat liver.
Am J Physiol Gastrointest Liver Physiol
277:
G725-G730,
1999[Abstract/Free Full Text].
17.
McLean, AJ,
and
Morgan DJ.
Clinical pharmacokinetics in patients with liver disease.
Clin Pharmacokinet
21:
42-69,
1991[Web of Science][Medline].
18.
Renkin, EM.
Transport of potassium-42 from blood to tissue in isolated mammalian skeletal muscles.
Am J Physiol
197:
1205-1210,
1959.
19.
Rose, CP,
and
Goresky CA.
Limitations of tracer oxygen uptake in the canine coronary circulation.
Circ Res
56:
57-71,
1985[Abstract/Free Full Text].
20.
Wu, ZY,
Rivory LP,
and
Roberts MS.
Physiological pharmacokinetics of solutes in the isolated perfused rat hind limb: characterisation of the physiology with changing perfusate flow, protein content, and temperature using statistical moment analysis.
J Pharmacokinet Biopharm
21:
653-687,
1993[Web of Science][Medline].
Am J Physiol Heart Circ Physiol 280(4):H1802-H1806
0363-6135/01 $5.00
Copyright © 2001 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
![C<SUB><IT>t</IT></SUB><IT>=R</IT>{<IT>1−e</IT><SUP>[<IT>−k</IT>(<IT>t−t<SUB>0</SUB></IT>)]</SUP>}](/content/vol280/issue4/fulltext/H1802/img003.gif)
